"Vinlagr Martin Gardnrr-d .. ar, (ompul h ... a nd a to d from co er to .

It
- I "\I ST F \\ ART. M n nl:'nit} l.lf War"i(il
"' Martin . ardnf!r s pu f much with the
t wor of urreali t rt: surprising, and
innurntial ror Rt'nt'l1uion, to (omt' In colle<tion
lakes his timef'xplainina both
tht' iOf,pirational outxro", uf math("mati( all
richest pu les resuhinK in a book that is fun and
insightful.
pu
SIIASIIA . \;f'_ VOf'k
l.Inhr"f'nit \, ;au I hot' u' Ib.. Po. a- Ad .... nlu"'
column for \rlrwri 1
is cola I ('ollt'clion of Martin Garon ,'s f
and columns rem:ukable- retrospective- of
hi" a."toll i hin I, inn... tial r
traonUn
1011"-11 ."LtF'-J PAll.OS.pruJruurolm.lht'm,lUl
't-.. l nlvrnh : ulhor 011.".,. ( . a d
I \I",I.,..uli, i R J. ,,...
,.yarn'" : Martin ha turnf"d do ens of inno-
I )oun sle into milt h professors lind thou nds of
milth 0 i nlu inn nt luun ,t r . lit
I Boo' 01 "'a,,.
tion:'
- PERSI 1>1 ACOl'I S ,. n'prd l n h r n 6t ,
BN 0- 393- 02023-1
I I
9 7803
\'It \\ \\ \\ \\ ((
ISBN 0-393-02023-1 USA $35.00
CAN. $51.00
"Newton said that his many mathematical 3l:l:omplish.
ments came because he stood on the shoulders of giants.
For those of us who have tried to make mathematics acces-
sible to a wider audienl:e, there is one giant who towers
well above anyone else: Martin Gardner."
-K E I T H DE V LIN, Stanrord University, unler ror Ih .. Study of Languag ..
and Inrormat ion; aut hor or Til,. Mal" Gmr, wi nner or t he 2000
CommuniClltions Award or the Joint Polic::y Board ror Mathematics
W
hether discussing hexaflexagon!' or nllmher
Klein bottles or the c:"sSf'ncf' of "nothing," Martin
Gardner has single-handedly creat d tht: fit: Id of '"recreational
mathematics." The Colossal R()()k ()f MntheHunics collects
togethtr Gardner's most popular pieces frorn hi. legendary
"Mathematical Games" olumn, which rdn in Scientific
American for twenty-five vt:'ars. Gardner's array ofahsorhing PU7-
zles and mind-twisting paradoxes openL IllLlthernatks up to the
world at large, inspiring people to see past numbers and funnulJs
and experience." the application of mathematical principles to the
mvstf'riolls world arounu thf'm. vVith articles on topi cs
from simple algebrd to the twisting surfaces of Mobius strip.
fTom an endless game of Bulgarian to the unre<.Kh .. lhle
dream of time travel, this volumt L' omprisc, a suhstantial and
definitive monument to Gardner's influenu' on m.ltht matles,
science, and culture.
In its twelve sections, "lhe C()/(lS5{l/ &ok of Mllthemlltics
explores a wide range of Jreas, eclch startlingly illuminated hv
Gardner's incisive expertise. Beginning with seemingly simple'
topiL"s, Gardner expertlY guides us through (omplic.lted .mel
wondrous worlds: by WdY of basic we \..ontempl"te the
mesmeri7ing, often hilarious, Iingui "ti( and numerical possihili-
ties of palindromes; using simple geomt'try. hc dissc(tstht, pnncj
pies of symmetry upon which the renowned mathematical Jrtist
M C. F,cher constructs his unique, dizzvin univen;t' . Gdrdner,
like lew thinkers today. mdds J ngorous scientific skepticism
with a prot()und artisti( and imaginative impulse. I lis stunning
t'xplorati<.m nf"The <.. 'hurch of the Fourth l )imensinn," for exam-
ple, bridges the dispJr Jte worlds of religinn .md sl' ience hy hril -
liantly imJgining the spatial possihility ofGtxl\ presem:e in the
world dS a fOurth dimension, at once "t:'vervwhere and nowhere."
With hnundle s wi .. dom dnd his tradem<lrk wit, Gardner
Ie mHHlIlld (m lUI( k II(/p)
I. (''''''Illt'(( I,.t"" "(lHI 'elr)
the to nlrthf'r Fngaxt' ch(lllFnging topi\. s like proha-
hilitv and gJme theory which helve plagu d clever gclmhlers,
FJmOLS mathematicians, for centuri es. Whether dehunking
Paslal's wa Jer with hasic prohability. mc!kin \ isual music with
fra -tals. or uncoiling a "knotted doughnut" with introdu "tory
topology. Gardner continuollsi displays his fierce intelligence
and gentle humor I lis articles (onn-ont both the comrortinljl}
mum..lane-"Generalil.ed TIcktacktoe
if
and "Spn. uts and Rrussel
Sprouts" -and the quakingly ahstract-'Ilexaflt"xagons," "Nothing,"
and "Evervthing: He navigates the, e taggeringly oh clire topics
with a deft intelligenL"e and. with addendums dnd suggested read
in lists, he mforms tht"se classIc articles With new insight
Admired by s ientists and mathemati ians, writers and
readers, Gardner's vast knowledge and burning curiosity reveal
themselves on every page. The culmination of a lifelong devotion
to the wonders ofmathemati s, The ColOS5flI &ok of Mathematics
is the largest and moc;t comprehensive rnath hook ever
hIed hv Gardner and remams an mdispensahle volume for ama-
teurs and experts alike,
MARTIN GARDNER iC\ the authofofmore than seventy
ona vast rangeoftoplCs.mdudingfluisand Fallacies in the Nflm
of Science The Annotated Alice and Did Adnm and hJe Hnve
Nnvels? He wrote the "Mdthemati 'al Games" oJumn for
Scientific American for more than twentv-f1ve years and is a regular
oiumnist fi)r The Skepricnl Enquirer He lives in Hendersonvill
North ..arolina.
M C KET Of ' /C; N BY OI;:BRA MORTON HO't'T
JAC KET ART C REZA ESTAKHRIANISTON E
A THOR PHUTO(;RAPIi BY OLAN MILLS
PRINTED IN THE UJ'.o1T ED STATES ot- "MERI A 8-01
The
Colossal
Book of
Mathematics
}llso by Martin Gardner
Fads and Fallacies in the Name of Science
Mathematics, Magic, and Mystery
Great Essays in Science (ed.)
Logic Machines and Diagrams
The Scientific American Book of Mathematical
Puzzles and Diversions
The Annotated Alice
The Second Scientific American Book of
Mathematical Puzzles and Diversions
Relativity for the Million
The Annotated Snark
The Ambidextrous Universe
The Annotated Ancient Mariner
New Mathematical Diversions from Scientific
American
The Annotated Casey at the Bat
Perplexing Puzzles and Tantalizing Teasers
The Unexpected Hanging and Other
Mathematical Diversions
Never Make Fun of a Thrtle, My Son (verse)
The Sixth Book of Mathematical Games from
Scientific American
Codes, Ciphers, and Secret Writing
Space Puzzles
The Snark Puzzle Book
The Flight of Peter Fromm (novel)
Mathematical Magic Show
More Perplexing Puzzles and Tantalizing Teasers
The Encyclopedia of Impromptu Magic
Aha! Insight
Mathematical Carnival
Science: Good, Bad, and Bogus
Science Fiction Puzzle Tales
Aha! Gotcha
Wheels, Life, and Other Mathematical
Amusements
Order and Surprise
Also by Martin Gardner
The Whys of a Philosophical Scrivener
Puzzles from Other Worlds
The Magic Numbers of Dr. Matrix
Knotted Doughnuts and Other Mathematical
Entertainments
The Wreck of the Titanic Foretold?
Riddles of the Sphinx
The Annotated Innocence of Father Brown
The No-Sided Professor (short stories)
Time Travel and Other Mathematical
Bewilderments
The New Age: Notes of a Fringe Watcher
Gardner's Whys and Wherefores
Penrose Tiles to Trapdoor Ciphers
How Not to Test a Psychic
The New Ambidextrous Universe
More Annotated Alice
The Annotated Night Before Christmas
Best Remembered Poems (ed.)
Fractal Music, Hypercards, and More
The Healing Revelations of Mary Baker Eddy
Martin Gardner Presents
My Best Mathematical and Logic Puzzles
Classic Brainteasers
Famous Poems of Bygone Days (ed.)
Urantia: The Great Cult Mystery
The Universe Inside a Handkerchief
On the Wild Side
Weird Water and Fuzzy Logic
The Night Is Large
Mental Magic
The Annotated Thursday
Visitors from Oz (novel)
The Annotated Alice: The Definitive Edition
Did Adam and Eve Have Navels?
Martin Gardner
The
COLOSSAL
Classic Puzzles, Paradoxes, and Problems
w. W. Norton &- Company
New York London
Number Theory
Algebra
Geometry
Probability
Topology
Game Theory
Infinity
and other Topics
of Recreational
Mathematics
Copyright 2001 by Martin Gardner
All rights reserved
Printed in the United States of America
First Edition
The text of this book is composed in Melior, with the display set in Goudy
Sans Italic with Augustea Open.
Composition by Allentown Digital Services
Manufacturing by The Maple-Vail Book Manufacturing Group
Book design by Charlotte Staub
Production manager: Andrew Marasia
Library of Congress Cataloging-in-Publication Data
Gardner, Martin, 1914-
The colossal book of mathematics: classic puzzles, paradoxes, and
problems: number theory, algebra, geometry, probability, topology, game
theory, infinity, and other topics of recreational mathematics I Martin
Gardner.
p.cm.
Includes bibliographical references and index.
ISBN 0-39302023-1
1. Mathematical recreations. 1. Title.
QA95.G245 2001
793.7'4-dc21 2001030341
W. W. Norton & Company, Inc., 500 Fifth Avenue, New York, N.Y. 10110
www.wwnorton.com
W. W. Norton & Company Ltd., Castle House, 75/76 Wells Street, London
W1T3QT
1234567890
For Qerard Piel
Who made all this possible
Contents
Preface xi
I Arithmetic and Algebra
I. The Monkey and the Coconuts 3
2. The Calculus of Finite Differences 10
3. Palindromes: Words and Numbers 23
II Plane Geometry
4. Curves of Constant Width 35
5. Rep-Tiles 46
6. Piet Hein's Superellipse 59
1. Penrose Tiles 73
8. The Wonders of a Planiverse 94
III Solid Geometry and Higher Dimensions
9. The Helix 117
10. Packing Spheres 128
II. Spheres and Hyperspheres 137
12. The Church of the Fourth Dimension 150
13. Hypercubes 162
14. Non-Euclidean Geometry 175
IV Symmetry
15. Rotations and Reflections 189
16. The Amazing Creations of Scott Kim 198
11. The Art of M. C. Escher 212
V Topology
18. Klein Bottles and Other Surfaces 227
19. Knots 239
20. Doughnuts: Linked and Knotted 254
VI Probability
21. Probability and Ambiguity 273
22. Nontransitive Dice and Other Paradoxes 286
23. More Nontransitive Paradoxes 297
VII Infinity
24. Infinite Regress 315
25. Aleph-Null and Aleph-One 327
26. Supertasks 340
21. Fractal Music 350
28. Surreal Numbers 369
VIII Combinatorics
29. Hexaflexagons 385
30. The Soma Cube 398
31. The Game of Life 409
32. Paper Folding 423
33. Ramsey Theory 437
34. Bulgarian Solitaire and Other
Seemingly Endless Tasks 455
IX qames and Decision Theory
35. A Matchbox Game-Learning Machine 471
36. Sprouts and Brussels Sprouts 485
37. Harary's Generalized Ticktacktoe 493
38. The New Eleusis 504
X Physics
39. Time Travel 517
40. Does Time Ever Stop? 531
41. Induction and Probability 541
42. Simplicity 553
XI Logic and Philosophy
43. The Unexpected Hanging 567
44. Newcomb's Paradox 580
45. Nothing 592
46. Everything 610
XII Miscellaneous
47. Melody-Making Machines 627
48. Mathematical Zoo 640
49. Godel, Escher, Bach 660
50. Six Sensational Discoveries 674
Selected Titles by the Author on
Mathematics 695
Index 697
Preface
My long and happy relationship with Scientific American, back
in the days when Gerard Piel was publisher and Dennis Flanagan was
the editor, began in 1952 when I sold the magazine an article on the his-
tory of logic machines. These were curious devices, invented in pre-
computer centuries, for solving problems in formal logic. The article
included a heavy paper insert from which one could cut a set of win-
dow cards I had devised for solving syllogisms. I later expanded the ar-
ticle to Logic Machines and Diagrams, a book published in 1959.
My second sale to Scientific American was an article on hexa-
flexagons, reprinted here as Chapter 29. As I explain in that chapter's
addendum, it prompted Piel to suggest a regular department devoted to
recreational mathematics. Titled "Mathematical Games" (that M. G. are
also my initials was a coincidence), the column ran for a quarter of a
century. As these years went by I learned more and more math. There
is no better way to teach oneself a topic than to write about it.
Fifteen anthologies of my Scientific American columns have been
published, starting with the Scientific American Book of Mathematical
Puzzles and Diversions (1959) and ending with Last Recreations (1997).
To my surprise and delight, Robert Weil, my editor at W. W. Norton,
suggested that I select 50 of what I consider my "best" columns, mainly
in the sense of arousing the greatest reader response, to make this hefty,
and, in terms of my career, definitive book you now hold. I have not in-
cluded any of my whimsical interviews with the famous numerologist
Dr. Irving Joshua Matrix because all those columns have been gathered
in The Magic Numbers of Dr. Matrix (1985).
To each chapter I have added an addendum, often lengthy, to update
the material. I also have provided selected bibliographies for further
reading.
Martin Gardner
Hendersonville, NC
xi
I
Arithmetic
and Algebra
Chapter I The Monkey
and the Coconuts
In the October 9,1926, issue of The Saturday Evening Post ap-
peared a short story by Ben Ames Williams entitled "Coconuts." The
story concerned a building contractor who was anxious to prevent a
competitor from getting an important contract. A shrewd employee of
the contractor, knowing the competitor's passion for recreational math-
ematics, presented him with a problem so exasperating that while he
was preoccupied with solving it he forgot to enter his bid before the
deadline.
Here is the problem exactly as the clerk in Williams's story phrased it:
Five men and a monkey were shipwrecked on a desert island, and they
spent the first day gathering coconuts for food. Piled them all up to-
gether and then went to sleep for the night.
But when they were all asleep one man woke up, and he thought there
might be a row about dividing the coconuts in the morning, so he de-
cided to take his share. So he divided the coconuts into five piles. He
had one coconut left over, and he gave that to the monkey, and he hid his
pile and put the rest all back together.
By and by the next man woke up and did the same thing. And he had
one left over, and he gave it to the monkey. And all five of the men did
the same thing, one after the other; each one taking a fifth of the co-
conuts in the pile when he woke up, and each one having one left over
for the monkey. And in the morning they divided what coconuts were
left, and they came out in five equal shares. Of course each one must
have known there were coconuts missing; but each one was guilty as the
others, so they didn't say anything. How many coconuts were there in
the beginning?
Williams neglected to include the answer in his story. It is said that
the offices of The Saturday Evening Post were showered with some
3
2,000 letters during the first week after the issue appeared. George Ho-
race Lorimer, then editor-in-chief, sent Williams the following historic
wire:
FOR THE LOVE OF MIKE, HOW MANY COCONUTS? HELL POPPING AROUND HERE.
For 20 years Williams continued to receive letters requesting the an-
swer or proposing new solutions. Today the problem of the coconuts is
probably the most worked on and least often solved of all the Dio-
phantine brainteasers. (The term Diophantine is descended from Dio-
phantus of Alexandria, a Greek algebraist who was the first to analyze
extensively equations calling for solutions in rational numbers.)
Williams did not invent the coconut problem. He merely altered a
much older problem to make it more confusing. The older version is the
same except that in the morning, when the final division is made, there
is again an extra coconut for the monkey; in Williams's version the final
division comes out even. Some Diophantine equations have only one
answer (e.g., x
2
+ 2 = r); some have a finite number of answers; some
(e.g., r + r = Z3) have no answer. Both Williams's version of the co-
conut problem and its predecessor have an infinite number of answers
in whole numbers. Our task is to find the smallest positive number.
The older version can be expressed by the following six indetermi-
nate equations which represent the six successive divisions of the co-
conuts into fifths. N is the original number; F, the number each sailor
received on the final division. The 1's on the right are the coconuts
tossed to the monkey. Each letter stands for an unknown positive inte-
ger:
N=5A+ 1,
4A = 5B+ 1,
4B 5C+ 1,
4C= 5D+ 1,
4D= 5E+ 1,
4E= 5F+ 1.
It is not difficult to reduce these equations by familiar algebraic
methods to the following single Diophantine equation with two un-
knowns:
1,024N= 15,625F+ 11,529.
This equation is much too difficult to solve by trial and error, and al-
though there is a standard procedure for solving it by an ingenious use
4 ARITHMETIC AND ALGEBRA
of continued fractions, the method is long and tedious. Here we shall
be concerned only with an uncanny but beautifully simple solution
involving the concept of negative coconuts. This solution is sometimes
attributed to the University of Cambridge physicist P.A.M. Dirac
(1902-1984), but in reply to my query Professor Dirac wrote that he ob-
tained the solution from J.H.C. Whitehead, professor of mathematics
(and nephew of the famous philosopher). Professor Whitehead, an-
swering a similar query, said that he got it from someone else, and I
have not pursued the matter further.
Whoever first thought of negative coconuts may have reasoned some-
thing like this. Since N is divided six times into five piles, it is clear
that 56 (or 15,625) can be added to any answer to give the next highest
answer. In fact any multiple of 56 can be added, and similarly any mul-
tiple can be subtracted. Subtracting multiples of 56 will of course even-
tually give us an infinite number of answers in negative numbers.
These will satisfy the original equation, though not the original prob-
lem, which calls for a solution that is a positive integer.
Obviously there is no small positive value for N which meets the
conditions, but possibly there is a simple answer in negative terms. It
takes only a bit of trial and error to discover the astonishing fact that
there is indeed such a solution: -4. Let us see how neatly this works
out.
The first sailor approaches the pile of -4 coconuts, tosses a positive
coconut to the monkey (it does not matter whether the monkey is given
his coconut before or after the division into fifths), thus leaving five
negative coconuts. These he divides into five piles, a negative coconut
in each. After he has hidden one pile, four negative coconuts remain-
exactly the same number that was there at the start! The other sailors
go through the same ghostly ritual, the entire procedure ending with
each sailor in possession of two negative coconuts, and the monkey,
who fares best in this inverted operation, scurrying off happily with six
positive coconuts. To find the answer that is the lowest positive inte-
ger, we now have only to add 15,625 to -4 to obtain 15,621, the solu-
tion we are seeking.
This approach to the problem provides us immediately with a gen-
eral solution for n sailors, each of whom takes one nth of the coconuts
at each division into nths. If there are four sailors, we begin with three
negative coconuts and add 4
5
. If there are six sailors, we begin with five
negative coconuts and add 6
7
, and so on for other values of n. More for-
The Monkey and the Coconuts 5
mally, the original number of coconuts is equal to k(nn+l) - m(n - 1),
where n is the number of men, m is the number of coconuts given to the
monkey at each division, and k is an arbitrary integer called the para-
meter. When n is 5 and m is 1, we obtain the lowest positive solution
by using a parameter of 1.
Unfortunately, this diverting procedure will not apply to Williams's
modification, in which the monkey is deprived of a coconut on the
last division. I leave it to the interested reader to work out the solution
to the Williams version. It can of course be found by standard Dio-
phantine techniques, but there is a quick shortcut if you take advantage
of information gained from the version just explained. For those who
find this too difficult, here is a very simple coconut problem free of all
Diophantine difficulties.
Three sailors come upon a pile of coconuts. The first sailor takes half
of them plus half a coconut. The second sailor takes half of what is left
plus half a coconut. The third sailor also takes half of what remains
plus half a coconut. Left over is exactly one coconut which they toss to
the monkey. How many coconuts were there in the original pile? If you
will arm yourself with 20 matches, you will have ample material for a
trial and error solution.
Addendum
If the use of negative coconuts for solving the earlier version of
Ben Ames Williams's problem seems not quite legitimate, essentially
the same trick can be carried out by painting four coconuts blue. Nor-
man Anning, now retired from the mathematics department of the Uni-
versity of Michigan, hit on this colorful device as early as 1912 when
he published a solution (School Science and Mathematics, June 1912,
p. 520) to a problem about three men and a supply of apples. Anning's
application of this device to the coconut problem is as follows.
We start with 56 coconuts. This is the smallest number that can be di-
vided evenly into fifths, have one-fifth removed, and the process re-
peated six times, with no coconuts going to the monkey. Four of the 56
coconuts are now painted blue and placed aside. When the remaining
supply of coconuts is divided into fifths, there will of course be one left
over to give the monkey.
After the first sailor has taken his share, and the monkey has his co-
conut, we put the four blue coconuts back with the others to make a
6 ARITHMETIC AND ALGEBRA
pile of 55 coconuts. This clearly can be evenly divided by 5. Before
making this next division, however, we again put the four blue co-
conuts aside so that the division will leave an extra coconut for the
monkey.
This procedure-borrowing the blue coconuts only long enough to
see that an even division into fifths can be made, then putting them
aside again-is repeated at each division. After the sixth and last divi-
sion, the blue coconuts remain on the side, the property of no one.
They play no essential role in the operation, serving only to make
things clearer to us as we go along.
A good recent reference on Diophantine equations and how to solve
them is Diophantus and Diophantine Equations by Isabella Bash-
makova (The Mathematical Association of America, 1997).
There are all sorts of other ways to tackle the coconut problem. John
M. Danskin, then at the Institute for Advanced Study, Princeton, NJ, as
well as several other readers, sent ingenious methods of cracking the
problem by using a number system based on 5. Scores of readers wrote
to explain other unusual approaches, but all are a bit too involved to ex-
plain here.
Answers
The number of coconuts in Ben Ames Williams's version of the
problem is 3,121. We know from the analysis of the older version that
55 4, or 3,121, is the smallest number that will permit five even divi-
sions of the coconuts with one going to the monkey at each division.
After these five divisions have been made, there will be 1,020 coconuts
left. This number happens to be evenly divisible by 5, which permits
the sixth division in which no coconut goes to the monkey.
In this version of the problem a more general solution takes the form
of two Diophantine equations. When n, the number of men, is odd, the
equation is
Number of coconuts = (1 + nk)nn - (n - 1).
When n is even,
Number of coconuts = (n - 1 + nk)nn - (n - 1).
In both equations k is the parameter that can be any integer. In
Williams's problem the number of men is 5, an odd number, so 5 is sub-
The Monkey and the Coconuts 7
stituted for n in the first equation, and k is taken as 0 to obtain the low-
est positive answer.
A letter from Dr. J. Walter Wilson, a Los Angeles dermatologist, re-
ported an amusing coincidence involving this answer:
Sirs:
I read Ben Ames Williams's story about the coconut problem in 1926,
spent a sleepless night working on the puzzle without success, then
learned from a professor of mathematics how to use the Diophantine
equation to obtain the smallest answer, 3,121.
In 1939 I suddenly realized that the home on West 80th Street, Ingle-
wood, California, in which my family and I had been living for several
months, bore the street number 3121. Accordingly, we entertained all of
our most erudite friends one evening by a circuit of games and puzzles,
each arranged in a different room, and visited by groups of four in rota-
tion.
The coconut puzzle was presented on the front porch, with the table
placed directly under the lighted house number blazingly giving the se-
cret away, but no one caught on!
The simpler problem of the three sailors, at the end of the chapter,
has the answer: 15 coconuts. If you tried to solve this by breaking
matches in half to represent halves of coconuts, you may have con-
cluded that the problem was unanswerable. Of course no coconuts
need be split at all in order to perform the required operations.
Ben Ames Williams's story was reprinted in Clifton Fadiman's an-
thology, The Mathematical Magpie (1962), reissued in paperback by
Copernicus in 1997. David Singmaster, in his unpublished history of
famous mathematical puzzles, traces similar problems back to the Mid-
dle Ages. Versions appear in numerous puzzle books, as well as in text-
books that discuss Diophantine problems. My bibliography is limited
to periodicals in English.
Bibliography
N. Anning, "Solution to a Problem," School Science and Mathematics, June 1912,
p.520.
N. Anning, "Monkeys and Coconuts," The Mathematics Teacher, Vol. 54, Decem-
ber 1951, pp. 560-62.
J. Bowden, "The Problem ofthe Dishonest Men, the Monkeys, and the Coconuts,"
Special Topics in Theoretical Arithmetic, pp. 203-12. Privately printed for the
author by Lancaster Press, Inc., Lancaster, PA, 1936.
8 ARITHMETIC AND ALGEBRA
P. W. Brashear, "Five Sailors and a Monkey," The Mathematics Teacher, October
1967, pp. 597-99.
S. King, "More Coconuts," The College Mathematics Journal, September 1998, pp.
312-13.
R. B. Kirchner, "The Generalized Coconut Problem," The American Mathematical
Monthly, Vol. 67, June-July 1960, pp. 516-19.
R. E. Moritz, "Solution to Problem 3,242," The American Mathematical Monthly,
Vol. 35, January 1928, pp. 47-48.
T. Shin and G. Salvatore, "On Coconuts and Integrity," Crux Mathematicorum,
Vol. 4, August/September 1978, pp. 182-85.
S. Singh and D. Bhattacharya, "On Dividing Coconuts: A Linear Diophantine Prob-
lem," The College Mathematics Journal, May 1997, pp. 203-4.
The Monkey and the Coconuts 9
Chapter 2 The Calculus of
Finite Differences
The calculus of finite differences, a branch of mathematics that
is not too well known but is at times highly useful, occupies a halfway
house on the road from algebra to calculus. W. W. Sawyer, a mathe-
matician at Wesleyan University, likes to introduce it to students by
performing the following mathematical mind-reading trick.
Instead of asking someone to "think of a number" you ask him to
"think of a formula." To make the trick easy, it should be a quadratic
formula (a formula containing no powers of x greater than x
2
). Suppose
he thinks of 5x
2
+ 3x - 7. While your back is turned so that you cannot
see his calculations, ask him to substitute 0,1, and 2 for x, then tell you
the three values that result for the entire expression. The values he
gives you are -7, 1, 19. After a bit of scribbling (with practice you can
do it in your head) you tell him the original formula!
The method is simple. Jot down in a row the values given to you. In
a row beneath write the differences between adjacent pairs of num-
bers, always subtracting the number on the left from its neighbor on the
right. In a third row put the difference between the numbers above it.
The chart will look like this
-7 1 19
8 18
10
The coefficient of x
2
, in the thought-of formula, is always half the bot-
tom number of the chart. The coefficient of x is obtained by taking half
the bottom number from the first number of the middle row. And the
constant in the formula is simply the first number of the top row.
What you have done is something analogous to integration in calcu-
lus. If y is the value of the formula, then the formula expresses a func-
10
tion of y with respect to x. When x is given values in a simple arith-
metic progression (0, 1, 2, ... ), thenyassumes a series of values (-7, 1,
19, ... ). The calculus of finite differences is the study of such series.
In this case, by applying a simple technique to three terms of a series,
you were able to deduce the quadratic function that generated the three
terms.
The calculus of finite differences had its origin in Methodus Incre-
mentorum, a treatise published by the English mathematician Brook
Taylor (who discovered the "Taylor theorem" of calculus) between
1715 and 1717. The first important work in English on the subject (after
it had been developed by Leonhard Euler and others) was published in
1860 by George Boole, of symbolic-logic fame. Nineteenth-century al-
gebra textbooks often included a smattering of the calculus, then it
dropped out of favor except for its continued use by actuaries in check-
ing annuity tables and its occasional use by scientists for finding for-
mulas and interpolating values. Today, as a valuable tool in statistics
and the social sciences, it is back in fashion once more.
For the student of recreational mathematics there are elementary pro-
cedures in the calculus of finite differences that can be enormously
useful. Let us see how such a procedure can be applied to the old prob-
lem of slicing a pancake. What is the maximum number of pieces into
which a pancake can be cut by n straight cuts, each of which crosses
each of the others? The number is clearly a function of n. If the func-
tion is not too complex, the method of differences may help us to find
it by empirical techniques.
No cut at all leaves one piece, one cut produces two pieces, two cuts
yield four pieces, and so on. It is not difficult to find by trial and error
that the series begins: 1, 2, 4,7,11, ... (see Figure 2.1). Make a chart as
before, forming rows, each representing the differences of adjacent
terms in the row above:
NUMBER OF CUTS
Number of pieces
First differences
Second differences
o 1 2 3 4
1 2 4 7 11
1 2 3 4
1 1 1
If the original series is generated by a linear function, the numbers in
the row of first differences will be all alike. If the function is a qua-
dratic, identical numbers appear in the row of second differences. A
The Calculus of Finite Differences 11
2 CUTS
4 PIECES
Figure 2.1. The pancake problem
o CUTS
1 PIECE
3 CUTS
7 PIECES
1 CUT
2 PIECES
4 CUTS
11 PIECES
cubic formula (no powers higher than Xl) will have identical numbers
in the row of third differences, and so on. In other words, the number
of rows of differences is the order of the formula. If the chart required
10 rows of differences before the numbers in a row became the same,
you would know that the generating function contained powers as high
as x
10
.
Here there are only two rows, so the function must be a quadratic. Be-
cause it is a quadratic, we can obtain it quickly by the simple method
used in the mind-reading trick.
The pancake-cutting problem has a double interpretation. We can
view it as an abstract problem in pure geometry (an ideal circle cut by
ideal straight lines) or as a problem in applied geometry (a real pancake
cut by a real knife). Physics is riddled with situations of this sort that
can be viewed in both ways and that involve formulas obtainable from
empirical results by the calculus of finite differences. A famous exam-
ple of a quadratic formula is the formula for the maximum number of
electrons that can occupy each "shell" of an atom. Going outward from
the nucleus, the series runs 0, 2, 8, 18, 32, 50 .... The first row of dif-
ferences is 2, 6, 10, 14, 18 .... The second row is 4, 4, 4, 4 .... Apply-
12 ARITHMETIC AND ALGEBRA
ing the key to the mind-reading trick, we obtain the simple formula 2n2
for the maximum number of electrons in the nth shell.
What do we do if the function is of a higher order? We can make use
of a remarkable formula discovered by Isaac Newton. It applies in all
cases, regardless of the number of tiers in the chart.
Newton's formula assumes that the series begins with the value of the
function when n is O. We call this number a. The first number of the
first row of differences is b, the first number of the next row is c, and
so on. The formula for the nth number of the series is
b
cn(n - 1) dn(n - 1)(n - 2)
a + n + + --=-------'---'-----=-
2 2-3
en(n - l)(n - 2)(n - 3) ...
+--'-----'-'-------'-""""'-------'-
2-3-4
The formula is used only up to the point at which all further addi-
tions would be zero. For example, if applied to the pancake-cutting
chart, the values of 1, 1, 1 are substituted for a, b, c in the formula. (The
rest of the formula is ignored because all lower rows of the chart con-
sist of zeros; d, e, f, ... therefore have values of zero, consequently the
entire portion of the formula containing these terms adds up to zero.)
In this way we obtain the quadratic function ~ n 2 + n + 1.
Does this mean that we have now found the formula for the maxi-
mum number of pieces produced by n slices of a pancake? Unfortu-
nately the most that can be said at this point is "Probably." Why the
uncertainty? Because for any finite series of numbers there is an infin-
ity of functions that will generate them. (This is the same as saying
that for any finite number of points on a graph, an infinity of curves can
be drawn through those points.) Consider the series 0,1,2,3 .... What
is the next term? A good guess is 4. In fact, if we apply the technique
just explained, the row of first differences will be l's, and Newton's for-
mula will tell us that the nth term of the series is simply n. But the
formula
1
n + 24n(n l)(n - 2)(n - 3)
also generates a series that begins 0, 1, 2, 3 .... In this case the series
continues, not 4, 5, 6, ... but 5, 10, 21 ....
There is a striking analogy here with the way laws are discovered in
science. In fact, the method of differences can often be applied to phys-
The Calculus of Finite Differences 13
ical phenomena for the purpose of guessing a natural law. Suppose, for
example, that a physicist is investigating for the first time the way in
which bodies fall. He observes that after one second a stone drops 16
feet, after two seconds 64 feet, and so on. He charts his observations
like this:
o 16
16 48
32
64
80
32
144
112
32
256
Actual measurements would not, of course, be exact, but the num-
bers in the last row would not vary much from 32, so the physicist as-
sumes that the next row of differences consists of zeros. Applying
Newton's formula, he concludes that the total distance a stone falls in
n seconds is 16n
2
But there is nothing certain about this law. It repre-
sents no more than the simplest function that accounts for a finite se-
ries of observations: the lowest order of curve that can be drawn
through a finite series of points on a graph. True, the law is confirmed
to a greater degree as more observations are made, but there is never
certainty that more observations will not require modification of the
law.
With respect to pancake cutting, even though a pure mathematical
structure is being investigated rather than the behavior of nature, the
situation is surprisingly similar. For all we now know, a fifth slice may
not produce the sixteen pieces predicted by the formula. A single fail-
ure of this sort will explode the formula, whereas no finite number of
successes, however large, can positively establish it. "Nature," as
George P6lya has put it, "may answer Yes or No, but it whispers one an-
swer and thunders the other. Its Yes is provisional, its No is defini-
tive." P6lya is speaking of the world, not abstract mathematical
structure, but it is curious that his point applies equally well to the
guessing of functions by the method of differences. Mathematicians do
a great deal of guessing, along lines that are often similar to methods of
induction in science, and P6lya has written a fascinating work, Math-
ematics and Plausible Reasoning, about how they do it.
Some trial and error testing, with pencil and paper, shows that five
cuts of a pancake do in fact produce a maximum of sixteen pieces. This
successful prediction by the formula adds to the probability that the
formula is correct. But until it is rigorously proved (in this case it is not
hard to do) it stands only as a good bet. Why the simplest formula is so
14 ARITHMETIC AND ALGEBRA
often the best bet, both in mathematical and scientific guessing, is one
of the lively controversial questions in contemporary philosophy of
science. For one thing, no one is sure just what is meant by "simplest
formula."
Here are a few problems that are closely related to pancake cutting
and that are all approachable by way of the calculus of finite differ-
ences. First you find the best guess for a formula, then you try to prove
the formula by deductive methods. What is the maximum number of
pieces that can be produced by n simultaneous straight cuts of a flat fig-
ure shaped like a crescent moon? How many pieces of cheesecake can
be produced by n simultaneous plane cuts of a cylindrical cake? Into
how many parts can the plane be divided by intersecting circles of the
same size? Into how many regions can space be divided by intersecting
spheres?
Recreational problems involving permutations and combinations
often contain low-order formulas that can be correctly guessed by the
method of finite differences and later (one hopes) proved. With an un-
limited supply of toothpicks of n different colors, how many different
triangles can be formed on a flat surface, using three toothpicks for the
three sides of each triangle? (Reflections are considered different, but
not rotations.) How many different squares? How many different tetra-
hedrons can be produced by coloring each face a solid color and using
n different colors? (Two tetrahedrons are the same if they can be turned
and placed side by side so that corresponding sides match in color.)
How many cubes with n colors?
Of course, if a series is generated by a function other than a polyno-
mial involving powers of the variable, then other techniques in the
method of differences are called for. For example, the exponential func-
tion 2
n
produces the series 1, 2,4,8, 16 .... The row of first differences
is also 1, 2,4,8, 16, ... , so the procedure explained earlier will get us
nowhere. Sometimes a seemingly simple situation will involve a series
that evades all efforts to find a general formula. An annoying example
is the necklace problem posed in one of Henry Ernest Dudeney's puz-
zle books. A circular necklace contains n beads. Each bead is black or
white. How many different necklaces can be made with n beads? Start-
ing with no beads, the series is 0, 2, 3, 4, 6, 8, 13, 18, 30 .... (Figure 2.2
shows the 18 different varieties of necklace when n = 7.) I suspect that
two formulas are interlocked here, one for odd n, one for even, but
whether the method of differences will produce the formulas, I do not
The Calculus of Finite Differences 15
two
16 AND
Figure 2.3. Five lines make ten triangles.
n
3
+ 3n
2
+ an
6
that can be obtained by applying Newton's formula to results obtained
empirically, but it does not seem to apply to the zero case. When a
doughnut is not cut at all, clearly there is one piece, whereas the for-
mula says there should be no pieces. To make the formula applicable,
we must define "piece" as part of a doughnut produced by cutting.
Where there is ambiguity about the zero case, one must extrapolate
backward in the chart of differences and assume for the zero case a
value that produces the desired first number in the last row of differ-
ences.
To prove that the formula given for the maximum number of regions
into which a pancake (circle) can be divided by n straight cuts, consider
first the fact that each nth line crosses n - 1 lines. The n - 1 lines di-
vide the plane into n regions. When the nth line crosses these n regions,
it cuts each region into two parts, therefore every nth line adds n re-
gions to the total. At the beginning there is one piece. The first cut
adds one more piece, the second cut adds two more pieces, the third
cut adds three more, and so on up to the nth cut which adds n pieces.
Therefore the total number of regions is 1 + 1 + 2 + 3 + ... + n. The sum
of 1 + 2 + 3 + ... + n is %n(n - 1). To this we must add 1 to obtain the
final formula.
The bead problem was given by Dudeney as problem 275 in his Puz-
The Calculus of Finite Differences 17
zles and Curious Problems. John Riordan mentions the problem on
page 162, problem 37, of his Introduction to Combinatorial Analysis
(Wiley, 1958; now out of print), indicating the solution without giving
actual formulas. (He had earlier discussed the problem in "The Com-
binatorial Significance of a Theorem of P6Iya," Journal of the Society
for Industrial and Applied Mathematics, Vol. 5, No.4, December 1957,
pp. 232-34.) The problem was later treated in considerable detail, with
some surprising applications to music theory and switching theory, by
Edgar N. Gilbert and John Riordan, in "Symmetry Types of Periodic Se-
quences," Illinois Journal of Mathematics, Vol. 5, No.4, December
1961, pages 657-65. The authors give the following table for the num-
ber of different types of necklaces, with beads of two colors, for neck-
laces of 1 through 20 beads:
NUMBER OF NUMBER OF
BEADS NECKLACES
1 2
2 3
3 4
4 6
5 8
6 13
7 18
8 30
9 46
10 78
11 126
12 224
13 380
14 687
15 1,224
16 2,250
17 4,112
18 7,685
19 14,310
20 27,012
The formulas for the necklace problem do not mean, by the way, that
Dudeney was necessarily wrong in saying that a solution was not pos-
sible, since he may have meant only that it was not possible to find a
polynomial expression for the number of necklaces as a function of n
so that the number could be calculated directly from the formula with-
18 ARITHMETIC AND ALGEBRA
out requiring a tabulation of prime factors. Because the formulas in-
clude Euler's phi function, the number of necklaces has to be calculated
recursively. Dudeney's language is not precise, but it is possible that he
would not have considered recursive formulas a "solution." At any
rate, the calculus of finite differences is not in any way applicable to the
problem, and only the recursive formulas are known.
Several dozen readers (too many for a listing of names) sent correct
solutions to the problem before Golomb's formulas were printed, some
of them deriving it from Riordan, others working it out entirely for
themselves. Many pointed out that when the number of beads is a
prime (other than 2), the formula for the number of different necklaces
becomes very simple:
2
n
-
1
- 1 n-l
---- + 2-2 + 1.
n
The following letter from John F. Gummere, headmaster of William
Penn Charter School, Philadelphia, appeared in the letters department
of Scientific American in October 1961:
Sirs:
I read with great interest your article on the calculus of finite differ-
ences. It occurs to me that one of the most interesting applications of
Newton's formula is one I discovered for myself long before I had reached
the calculus. This is simply applying the method of finite differences to
series of powers. In experimenting with figures, I noticed that if you
wrote a series of squares such as 4,9,16,25,36,49 and subtracted them
from each other as you went along, you got a series that you could sim-
ilarly subtract once again and come up with a finite difference.
So then I tried cubes and fourth powers and evolved a formula to the
effect that if n is the power, you must subtract n times, and your constant
difference will be factorial n. I asked my father about this (he was for
many years director of the Strawbridge Memorial Observatory at Haver-
ford College and teacher of mathematics). In good Quaker language he
said: "Why, John, thee has discovered the calculus of finite differences."
Donald Knuth called my attention to the earliest known solution of
Dudeney's bead problem. Percy A. MacMahon solved the problem as
early as 1892. This and the problem are discussed in Section 4.9 of
Concrete Mathematics (1994), by Ronald Graham, Donald Knuth, and
Oren Patashnik.
The Calculus of Finite Differences 19
Answers
How many different triangles can be formed with n straight
lines? It takes at least three lines to make one triangle, four lines will
make four triangles, and five lines will make 10 triangles. Applying the
calculus of finite differences, one draws up the table in Figure 2.4.
NUMBER OF LINES 0 1 2 3 4 5 Figure 2.4. The answer
to the triangle problem
NUMBER OF TRIANGLES 0 0 0 1 4 10
FIRST DIFFERENCES 0 0 1 3 6
SECOND DIFFERENCES 0 1 2 3
THIRD DIFFERENCES 1 1 1
The three rows of differences indicate a cubic function. Using New-
ton's formula, the function is found to be: %n(n - l)(n - 2). This will
generate the series 0, 0, 0, 1, 4, 10, ... and therefore has a good
chance of being the formula for the maxinlum number of triangles
that can be made with n lines. But it is still just a guess, based on a
small number of pencil and paper tests. It can be verified by the fol-
lowing reasoning.
The lines must be drawn so that no two are parallel and no more than
two intersect at the same point. Each line is then sure to intersect every
other line, and every set of three lines must form one triangle. It is not
possible for the same three lines to form more than one triangle, so the
number of triangles formed in this way is the maximum. The problem
is equivalent, therefore, to the question: In how many different ways
can n lines be taken three at a time? Elementary combinatorial theory
supplies the answer: the same as the formula obtained empirically.
Solomon W. Golomb, was kind enough to send me his solution to the
necklace problem. The problem was to find a formula for the number
of different necklaces that can be formed with n beads, assuming tha!
each bead can be one of two colors and not counting rotations and re-
flections of a necklace as being different. The formula proves to be far
beyond the power of the simple method of differences.
Let the divisors of n (including 1 and n) be represented by dl' d
2
,
d
3
For each divisor we find what is called Euler's phi function for
that divisor, symbolized <I>(d). This function is the number of positive
integers, not greater than d, that have no common divisor with d. It is
assumed that 1 is such an integer, but not d. Thus <1>(8) is 4, because 8
20 ARITHMETIC AND ALGEBRA
has the following four integers that are prime to it: 1, 3, 5, 7. By con-
vention, <P(1) is taken to be 1. Euler's phi functions for 2, 3,4,5,6, 7 are
1,2,2,4,2,6, in the same order. Let a stand for the number of differ-
ent colors each bead can be. For necklaces with an odd number of
beads the formula for the number of different necklaces with n beads
is the one given at the top of Figure 2.5. When n is even, the formula is
the one at the bottom of the illustration.
1 [ .ll. n n 11]
- q,(d
1
) ad, + q,(d
2
) ali; ... + - (1 + a) a"2
2n 2
Figure 2.S. Equations for the solution of the necklace problem
The single dots are symbols for multiplication. Golomb expressed
these formulas in a more compressed, technical form, but I think the
above forms will be clearer to most readers. They are more general than
the formulas asked for because they apply to beads that may have any
specified number of colors.
The formulas answering the other questions in the chapter are:
1. Regions of a crescent moon produced by n straight cuts:
n
2
+ 3n
---+1.
2
2. Pieces of cheesecake produced by n plane cuts:
n
3
+ 5n
---+1.
6
3. Regions of the plane produced by n intersecting circles:
n
2
- n + 2.
4. Regions of the plane produced by n intersecting ellipses:
2n2 2n + 2.
5. Regions of space produced by n intersecting spheres:
- 3n + 8)
3
The Calculus of Finite Differences 21
6. Triangles formed by toothpicks of n colors:
n
3
+ 2n
3
7. Squares formed with toothpicks of n colors:
n
4
+ n
2
+ 2n
4
8. Tetrahedrons formed with sides of n colors:
n
4
+ 11n2
12
9. Cubes formed with sides of n colors:
n
6
+ 3n
4
+ 12n
3
+ 8n
2
24
Bibliography
E. L. Baker, "The Method of Differences in the Determination of Formulas," School
Science and Mathematics, April 1967, pp. 309-15.
C. Jordan, The Calculus of Finite Differences, Chelsea, 1965.
K. S. Miller, An Introduction to the Calculus of Finite Differences and Difference
Equations, Henry Holt, 1960.
W. E. Milne, Numerical Calculus, Princeton University Press, 1949.
L. M. Milne-Thomson, The Calculus of Finite Differences, Macmillan, 1951.
R. C. Read, "P6Iya's Theorem and its Progeny," in Mathematics Magazine, VoL 60,
December 1987, pp. 275-82. Eighteen references are listed.
22 ARITHMETIC AND ALGEBRA
A man} a a canal-Suez!
-Ethel Merperso:n,
in Son of Giant Sea Tortoise,
Ann AY.lQ'UUW;;',I..l
usually defined as a or
.;J'\J..L.l.I..'\J.LJ.'-" .... LJl that spell the same backward as forward. The term is also ap-
are types
long
play, perhaps because of a
aesthetic pleasure in the kind of symmetry palindromes possess. Palin-
dromes that are
'-' ..... , ........ , ................ some
3.1. seagull: a visual palindrome
An old palindrome conjecture origin (there are
enees it
integer. it and add the
is repeated with the sum to obtain a second sum, and the process COD-
a
a
68 generates a palindrome in three steps:
23
68
+ 86
154
+451
605
+506
1,111
For all two-digit numbers it is obvious that if the sum of their digits
is less than 10, the first step gives a two-digit palindrome. If their dig-
its add to 10, 11, 12, 13, 14, 15, 16, or 18, a palindrome results after 2,
1, 2, 2, 3, 4, 6, 6 steps, respectively. As Angela Dunn points out in
Mathematical Bafflers (McGraw-Hill, 1964; Dover, 1980), the excep-
tions are numbers whose two digits add to 17. Only 89 (or its reversal,
98) meets this proviso. Starting with either number does not produce
a palindrome until the 24th operation results in 8,813,200,023,188.
The conjecture was widely regarded as being true until 1967, al-
though no one had succeeded in proving it. Charles W. Trigg, a Cali-
fornia mathematician well known for his work on recreational
problems, examined the conjecture more carefully in his 1967 article
"Palindromes by Addition." He found 249 integers smaller than
10,000 that failed to generate a palindrome after 100 steps. The small-
est such number, 196, was carried to 237,310 steps in 1975 by Harry
J. Saal, at the Israel Scientific Center. No palindromic sum appeared.
Trigg believed the conjecture to be false. (The number 196 is the
square of 14, but this is probably an irrelevant fact.) Aside from the
249 exceptions, all integers less than 10,000, except 89 and its rever-
sal, produce a palindrome in fewer than 24 steps. The largest palin-
drome, 16,668,488,486,661, is generated by 6,999 (or its reversal) and
7,998 (or its reversal) in 20 steps.
The conjecture has not been established for any number system and
has been proved false only in number notations with bases that are
powers of 2. (See the paper by Heiko Harborth listed in the bibliogra-
phy.) The smallest binary counterexample is 10110 (or 22 in the deci-
mal system). After four steps the sum is 10110100, after eight steps it
is 1011101000, after 12 steps it is 101111010000. Every fourth step in-
creases by one digit each of the two sequences of underlined digits.
Brother Alfred Brousseau, in "Palindromes by Addition in Base Two,"
proved that this asymmetric pattern repeats indefinitely. He also found
other repeating asymmetric patterns for larger binary numbers.
24 ARITHMETIC AND ALGEBRA
There is a small but growing literature on the properties of palin-
dromic prime numbers and conjectures about them. Apparently there
are infinitely many such primes, although so far as I know this has not
been proved. It is not hard to show, however, that a palindromic prime,
with the exception of 11, must have an odd number of digits. Can the
reader do this before reading the simple proof in the answer section?
Norman T. Gridgeman conjectured that there is an infinity of prime
pairs of the form 30,103-30,203 and 9,931,399-9,932,399 in which all
digits are alike except the middle digits, which differ by one. But
Gridgeman's guess is far from proved.
Gustavus J. Simmons wrote two papers on palindromic powers. After
showing that the probability of a randomly selected integer being palin-
dromic approaches zero as the number of digits in the integer increases,
Simmons examined square numbers and found them much richer than
randomly chosen integers in palindromes. There are infinitely many
palindromic squares, most of which, it seems, have square roots that
also are palindromes. (The smallest nonpalindromic root is 26). Cubes
too are unusually rich in palindromes. A computer check on all cubes
less than 2.8 x 10
14
turned up a truly astonishing fact. The only palin-
dromic cube with a nonpalindromic cube root, among the cubes ex-
amined by Simmons, is 10,662,526,601. Its cube root, 2,201, had been
noticed earlier by Thigg, who reported in 1961 that it was the only non-
palindrome with a palindromic cube less than 1,953,125,000,000. It is
not yet known if 2,201 is the only integer with this property.
Simmons' computer search of palindromic fourth powers, to the
same limit as his search of cubes, failed to uncover a single palindromic
fourth power whose fourth root was not a palindrome of the general
form 10 ... 01. For powers 5 through 10 the computer found no palin-
dromes at all except the trivial case of 1. Simmons conjectured that
there are no palindromes of the form Xk where k is greater than 4.
"Repunits," numbers consisting entirely of l's, produce palindromic
squares when the number of units is one through nine, but 10 or more
units give squares that are not palindromic. It has been erroneously
stated that only primes have palindromic cubes, but this is disproved
by an infinity of integers, the smallest of which is rep unit 111. It is di-
visible by 3, yet its cube, 1,367,631, is a palindrome. The number 836
is also of special interest. It is the largest three-digit integer whose
square, 698,896, is palindromic, and 698,896 is the smallest palin-
dromic square with an even number of digits. (Note also that the num-
Palindromes: Words and Numbers 25
ber remains palindromic when turned upside down.) Such palin-
dromic squares are extremely rare. The next-larger one with an even
number of digits is 637,832,238,736, the square of 798,644.
Turning to language palindromes, we first note that no common Eng-
lish words of more than seven letters are palindromic. Examples of
seven-letter palindromes are reviver, repaper, deified, and rotator. The
word "radar" (for radio detecting and ranging) is notable because it
was coined to symbolize the reflection of radio waves. Dmitri
Borgmann, whose files contain thousands of sentence palindromes in
all major languages, asserts in his book Language on Vacation that the
largest nonhyphenated word palindrome is saippuakauppias, a
Finnish word for a soap dealer.
Among proper names in English, according to Borgmann, none is
longer than Wassamassaw, a swamp north of Charleston, sc. Legend
has it, he writes, that it is an Indian word meaning "the worst place ever
seen." Yreka Bakery has long been in business on West Miner Street in
Yreka, CA. Lon Nol, the former Cambodian premier, has a palindromic
name, as does U Nu, once premier of Burma. Revilo P. Oliver, a classics
professor at the University of Illinois, has the same first name as his fa-
ther and grandfather. It was originally devised to make the name
palindromic. If there is anyone with a longer palindromic name I do
not know of it, although Borgmann suggests such possibilities as
Norah Sara Sharon, Edna Lala Lalande, Duane Rollo Renaud, and
many others.
There are thousands of excellent sentence palindromes in English, a
few of which were discussed in a chapter on word play in my Sixth
Book of Mathematical Games from Scientific American. The interested
reader will find good collections in the Borgmann book cited above
and in the book by Howard Bergerson. Composing palindromes at night
is one way for an insomniac to pass the dark hours, as Roger Angell so
amusingly details in his article" Ainmosni" (HInsomnia" backward) in
The New Yorker. I limit myself to one palindrome that is not well
known, yet is remarkable for both its length and naturalness: "Doc note,
I dissent. A fast never prevents a fatness. I diet on cod." It won a prize
for James Michie in a palindrome contest sponsored by the New States-
man in England; results were published in the issue for May 5,1967.
Many of the winning palindromes are much longer than Michie's, but,
as is usually the case, the longer palindromes are invariably difficult to
understand.
26 ARITHMETIC AND ALGEBRA
Palindromists have employed various devices to make the unintelli-
gibility of long palindromes more plausible: presenting them as
telegrams, as one side only of a telephone conversation, and so on.
Leigh Mercer, a leading British palindromist (he is the inventor of the
famous "A man, a plan, a canal-Panama!"), has suggested a way of
writing a palindrome as long as one wishes. The sentence has the form,
" , sides reversed, is ' .' " The first blank can be any se-
quence of letters, however long, which is repeated in reverse order in
the second blank.
Good palindromes involving the names of U.S. presidents are ex-
ceptionally rare. Borgmann cites the crisp "Taft: fat!" as one of the
shortest and best. Richard Nixon's name lends itself to "No 'x' in 'Mr.
R. M. Nixon'?" although the sentence is a bit too contrived. A shorter,
capitalized version of this palindrome, NO X IN NIXON, is also in-
vertible.
The fact that "God" is "dog" backward has played a role in many sen-
tence palindromes, as well as in orthodox psychoanalysis. In Freud's
Contribution to Psychiatry A. A. Brill cites a rather farfetched analysis
by Carl Jung and others of a patient suffering from a ticlike upward
movement of his arms. The analysts decided that the tic had its origin
in an unpleasant early visual experience involving dogs. Because of the
"dog-god" reversal, and the man's religious convictions, his uncon-
scious had developed the gesture to symbolize a warding off of the evil
"dog-god." Edgar Allan Poe's frequent use of the reversal words "dim"
and "mid" is pointed out by Humbert Humbert, the narrator of
Vladimir Nabokov's novel Lolita. In the second canto of Pale Fire, in
Nabokov's novel of the same title, the poet John Shade speaks of his
dead daughter's propensity for word reversals:
... She twisted words: pot, top,
Spider, redips. And upowder" was ured wop. "
Such word reversals, as well as sentences that are different sentences
when they are spelled backward, are obviously close cousins of palin-
dromes, but the topic is too large to go into here.
Palindrome sentences in which words, not letters, are the units have
been a specialty of another British expert on word play, J. A. Lindon.
Two splendid examples, from scores that he has composed, are
"You can cage a swallow, can't you, but you can't swallow a cage, can
you?"
Palindromes: Words and Numbers 27
"Girl, bathing on Bikini, eyeing boy, finds boy eyeing bikini on
bathing girl."
Many attempts have been made to write letter-unit palindrome
poems, some quite long, but without exception they are obscure,
rhymeless, and lacking in other poetic values. Somewhat better poems
can be achieved by making each line a separate palindrome rather than
the entire poem or by using the word as the unit. Lindon has written
many poems of both types. A third type of palindrome poem, invented
by Lindon, employs lines as units. The poem is unchanged when its
lines are read forward but taken in reverse order. One is allowed, of
course, to punctuate duplicate lines differently. The following example
is one of Lindon's best:
As I was passing near the jail
I met a man, but hurried by.
His face was ghastly, grimly pale.
He had a gun. I wondered why
He had. A gun? I wondered . .. why,
His face was ghastly! Grimly pale,
I met a man, but hurried by,
As I was passing near the jail.
This longer one is also by Lindon. Both poems appear in Howard W.
Bergerson's Palindromes and Anagrams (Dover, 1973).
28
DOPPELGANGER
Entering the lonely house with my wife,
I saw him for the first time
Peering furtively from behind a bush-
Blackness that moved,
A shape amid the shadows,
A momentary glimpse of gleaming eyes
Revealed in the ragged moon.
A closer look (he seemed to turn) might have
Put him to flight forever-
I dared not
(For reasons that I failed to understand),
Though I knew I should act at once.
I puzzled over it, hiding alone,
Watching the woman as she neared the gate.
He came, and I saw him crouching
ARITHMETIC AND ALGEBRA
Night after night.
Night after night
He came, and I saw him crouching,
Watching the woman as she neared the gate.
I puzzled over it, hiding alone-
Though I knew I should act at once,
For reasons that I failed to understand
I dared not
Put him to flight forever.
A closer look (he seemed to turn) might have
Revealed in the ragged moon
A momentary glimpse of gleaming eyes,
A shape amid the shadows,
Blackness that moved.
Peering furtively from behind a bush,
I saw him, for the first time,
Entering the lonely house with my wife.
Lindon holds the record for the longest word ever worked into a
letter-unit palindrome. To understand the palindrome you must know
that Beryl has a husband who enjoys running around his yard without
any clothes on. Ned has asked him if he does this to annoy his wife. He
answers: "Named un denominationally rebel, I rile Beryl? La, no! I tan.
I'm, 0 Ned, nude, man!"
Addendum
A. Ross Eckler, editor and publisher of Word Ways, a quarterly
journal on word play that has featured dozens of articles on palin-
dromes of all types, wrote to say that the "palindromic gap" between
English and other languages is perhaps not as wide as I suggested. The
word "semitime" can be pluralized to make a 9-letter palindrome and
"kinnikinnik" is an ii-letter palindrome. Dmitri Borgmann pointed
out in Word Ways, said Eckler, that an examination of foreign dictio-
naries failed to substantiate such long palindromic words as the
Finnish soap dealer, suggesting that they are artificially created words.
Among palindromic towns and cities in the United States, Borgmann
found the 7-letter Okonoko (in West Virginia). If a state (in full or ab-
breviated form) is part of the palindrome, Borgmann offers Apollo, PA,
Palindromes: Words and Numbers 29
and Adaven, Nevada. Some U.S. towns, Eckler continued, are inten-
tional reversal pairs, such as Orestod and Dotsero, in Eagle County,
Colorado, and Colver and Revloc, in Cambria County, Pennsylvania.
Nova and Avon, he added, are Ohio towns that are an unintentional re-
versal pair.
George L. Hart III sent the following letter, which was published in
Scientific American, November 1970:
Sirs:
Apropos of your discussion of palindromes, I would like to offer an ex-
ample of what I believe to be the most complex and exquisite type of
palindrome ever invented. It was devised by the Sanskrit aestheticians,
who termed it sarvatobhadra, that is, "perfect in every direction." The
most famous example of it is found in the epic poem entitled
Sisupiilavadha.
sa - ka - ra - na - na - ra - ka - sa -
ka - ya - sa - da - da- sa - ya - ka
ra - sa - ha - va va - ha - sa - ra -
na - da - va - da - da - va - da - na.
(na da va da da va da na
ra sa ha va va ha sa ra
ka ya sa da da sa ya ka
sa ka ra na na ra ka sa)
Here hyphens indicate that the next syllable belongs to the same word.
The last four lines, which are an inversion of the first four, are not part
of the verse but are supplied so that its properties can be seen more eas-
ily. The verse is a description of an army and may be translated as fol-
lows: "[That army], which relished battle [rasahava], contained allies
who brought low the bodes and gaits of their various striving enemies
[sakarananarakasakayasadadasayaka], and in it the cries of the best of
mounts contended with musical instruments [vahasaranadavada-
davadana]."
Two readers, D. M. Gunn and Rosina Wilson, conveyed the sad news
that the Yreka Bakery no longer existed. However, in 1970 its premises
were occupied by the Yrella Gallery, and Ms. Wilson sent a Polaroid
picture of the gallery'S sign to prove it. Whether the gallery is still there,
I do not know.
Lee Sallows repaired the "near miss" palindrome in this chapter's
epigraph by adding a word: "Zeus! A man, a plan, a canal-Suez!"
30 ARITHMETIC AND ALGEBRA
Answers
Readers were asked to prove that no prime except 11 can be a
palindrome if it has an even number of digits. The proof exploits a well-
known test of divisibility by 11 (which will not be proved here): If the
difference between the sum of all digits in even positions and the sum
of all digits in odd positions is zero or a multiple of 11, the number is a
multiple of 11. When a palindrome has an even number of digits, the
digits in odd positions necessarily duplicate the digits in even posi-
tions; therefore the difference between the sums of the two sets must be
zero. The palindrome, because it has 11 as a factor, cannot be prime.
The same divisibility test applies in all number systems when the fac-
tor to be tested is the system's base plus one. This proves that no palin-
drome with an even number of digits, in any number system, can be
prime, with the possible exception of11. The number 11 is prime if the
system's base is one less than a prime, as it is in the decimal system.
Bibliography
The palindrome number conjecture
D. Ahl, "Follow-up on Palindromes," Creative Computing, May 1975, p. 18.
A. Brousseau, "Palindromes by Addition in Base Two," Mathematics Magazine,
Vol. 42, November 1969, pp. 254-56.
F. Gruenberger, "How to Handle Numbers With Thousands of Digits, and Why One
Might Not Want To," Scientific American, Vol. 205, April 1984, pp. 19-26.
H. Harborth, "On Palindromes," Mathematics Magazine, Vol. 46, March 1973,
pages 96-99.
G. Johns and J. Wiegold, "The Palindromic Problem in Base 2," Mathematical
Gazette, Vol. 78, November 1994, p. 312.
W. Koetke, "Palindromes: For Those Who Like to Start at the Beginning." Creative
Computing, January 1975, pp. 10-12.
G. Kr5ber, "On Non-Palindromic Patterns in Palindromic Processes," Mathemati-
cal Gazette, Vol. 80, November 1996, pp. 577-79.
D. J. Lanska, "The 196 Problem," Journal of Recreational Mathematics, Vol. 18, No.
2,1985-86,pp.152-53.
D. J. Lanska, "More on the 196 Problem," Journal of Recreational Mathematics, Vol.
18, No.4, 1985-86, pp. 305-6.
R. Sprague, Recreation in Mathematics, Dover, 1963, Problem 5.
C. W. Trigg, "Palindromes by Addition," Mathematics Magazine, Vol. 40, January
1967, pp. 26-28.
C. W. Trigg, "More on Palindromes by Reversal-Addition," Mathematics Magazine,
Vol. 45, September 1972, pp. 184-86.
Palindromes: Words and Numbers 31
C. W. Trigg, "Versum Sequences in the Binary System." Pacific Journal of Mathe-
matics, Vol. 47, 1973, pp. 263-75.
"The 196 Problem," Popular Computing, Vol. 3, September 1975, pp. 6-9.
Palindromic primes
1. Card, "Patterns in Primes," Journal of Recreational Mathematics, Vol. 1, April
1968, pp. 93-99.
H. Gabai and D. Coogan, "On Palindromes and Palindromic Primes," Mathematics
Magazine, Vol. 42, November 1969, pp. 252-54.
C. W. Thigg, "Special Palindromic Primes," Journal of Recreational Mathematics,
Vol. 4, July 1971, pages 169-70.
Palindromic Powers
G. J. Simmons, "Palindromic Powers," Journal of Recreational Mathematics, Vol.
3, April 1970, pp. 93-98.
G. J. Simmons, "On Palindromic Squares of Non-Palindromic Numbers," Journal
of Recreational Mathematics, Vol. 5, January 1972, pp. 11-19.
C. W. Thigg, "Palindromic Cubes," Mathematics Magazine, Vol. 34, March 1961, p.
214.
Word Palindromes
R. Angell, "Ainmosni," The New Yorker, May 31,1969, pp. 32-34.
H. W. Bergerson, Palindromes and Anagrams, Dover, 1973.
C. C. Bombaugh, Oddities and Curiosities of Words and Literature, M. Gardner
(ed.), Dover, 1961.
D. Borgmann, Language on Vacation, Scribner's, 1965.
D. Borgmann, Beyond Language, Scribner's, 1967.
S. J. Chism, From A to Zotamorf: The Dictionary of Palindromes, Word Ways, 1992.
M. Donner, I Love Me, Vol. I: S. Wordrow's Palindrome Encyclopedia, Algonquin
Books, 1996.
M. Gardner, "The Oulipo," Penrose Tiles To Trapdoor Ciphers, Chapters 6 and 7,
W. H. Freeman, 1989.
S. W. Golomb, "I Call on Professor Osseforp," Harvard Bulletin, March 1972, p. 45.
W. Irvine, "Madam, I'm Adam and Other Palindromes," Scribner's, 1987.
]. Kuhn and M. Kuhn, Rats Live on No Evil Star, Everett House, 1981.
R. Lederer, "Palindromes: The Art of Reverse English," Verbatim, Vol. 11, Winter
1985, pp. 14-15.
L. Levine. Dr. Award and Olson in Oslo: A Palindromic Novel, By author, 1980.
J. Pool, Lid Off a Daffodil: A Book of Palindromes, Holt, Rinehart, and Winston,
1985.
R. Stuart, Too Hot To Hoot, McKay, 1977.
Word Ways: The Journal of Recreational Linguistics. Vol. 1, February 1968.
"Weekend Competition," New Statesman, April 14, 1967, p. 521; May 5, 1967, p.
630; December 22,1972, p. 955; March 9, 1973, p. 355.
"Results of Competition," New York Magazine, December 22, 1969, p. 82.
"The Contest," Maclean's Magazine, December 1969, pp. 68-69.
32 ARITHMETIC AND ALGEBRA
II
Plane Geometry
Chapter 4 Curves of
Constant Width
If an enormously heavy object has to be moved from one spot
to another, it may not be practical to move it on wheels. Axles might
buckle or snap under the load. Instead the object is placed on a fiat plat-
form that in turn rests on cylindrical rollers. As the platform is pushed
forward, the rollers left behind are picked up and put down again in
front.
An object moved in this manner over a fiat, horizontal surface obvi-
0usly does not bob up and down as it rolls along. The reason is simply
that the cylindrical rollers have a circular cross section, and a circle is
a closed curve possessing what mathematicians call "constant width."
If a closed convex curve is placed between two parallel lines and the
lines are moved together until they touch the curve, the distance be-
tween the parallel lines is the curve's "width" in one direction. An el-
lipse clearly does not have the same width in all directions. A platform
riding on elliptical rollers would wobble up and down as it rolled over
them. Because a circle has the same width in all directions, it can be ro-
tated between two parallel lines without altering the distance between
the lines.
Is the circle the only closed curve of constant width? Most people
would say yes, thus providing a sterling example of how far one's math-
ematical intuition can go astray. Actually there is an infinity of such
curves. Anyone of them can be the cross section of a roller that will roll
a platform as smoothly as a circular cylinder! The failure to recognize
such curves can have and has had disastrous consequences in indus-
try. To give one example, it might be thought that the cylindrical hull
of a half-built submarine could be tested for circularity by just mea-
suring maximum widths in all directions. As will soon be made clear,
such a hull can be monstrously lopsided and still pass such a test. It is
35
B
A ~ - - - - - - - - . . . . . . . . ; J C
4.1
Reuleaux L.I. ........... "",L\J rc)tating in SQllare.
36
Cross section of drill in hole
Watts chuck and drill
4.2.
37
the compass and draw arc FG. Do the same at the other corners. The re-
sulting curve has a width, in all directions, that is the sum of the same
two radii. This of course makes it a curve of constant width. Other
symmetrical curves of constant width result if you start with a regular
pentagon (or any regular polygon with an odd number of sides) and fol-
low similar procedures.
~ ~ ~ ~ - - - - - G
,
H
Figure 4.3. Symmetrical rounded-corner curve of constant width
There are ways to draw unsymmetrical curves of constant width.
One method is to start with an irregular star polygon (it will necessar-
ily have an odd number of points) such as the seven-point star shown
in black in Figure 4.4. All of these line segments must be the same
length. Place the compass point at each corner of the star and connect
the two opposite corners with an arc. Because these arcs all have the
same radius, the resulting curve (shown in gray) will have constant
width. Its corners can be rounded off by the method used before. Ex-
tend the sides of the star a uniform distance at all points (shown with
broken lines) and then join the ends of the extended sides by arcs
drawn with the compass point at each corner of the star. The rounded-
corner curve, which is shown in black, will be another curve of con-
stant width.
Figure 4.5 demonstrates another method. Draw as many straight lines
as you please, all mutually intersecting. Each arc is drawn with the
compass point at the intersection of the two lines that bound the arc.
Start with any arc, then proceed around the curve, connecting each arc
to the preceding one. If you do it carefully, the curve will close and will
have constant width. (Proving that the curve must close and have con-
38 PLANE GEOMETRY
Figure 4.4. Star-polygon method of drawing a curve of constant width
stant width is an interesting and not difficult exercise.) The preceding
curves were made up of arcs of no more than two different circles, but
curves drawn in this way may have arcs of as many different circles as
you wish.
B
Figure4.S. Crossed-lines method Random CUIve and tangents
A curve of constant width need not consist of circular arcs. In fact,
you can draw a highly arbitrary convex curve from the top to the bot-
tom of a square and touching its left side (arc ABC in Figure 4.5), and
this curve will be the left side of a uniquely determined curve of con-
stant width. To find the missing part, rule a large number of lines, each
parallel to a tangent of arc ABC and separated from the tangent by a dis-
tance equal to the side of the square. This can be done quickly by using
both sides of a ruler. The original square must have a side equal to the
ruler's width. Place one edge of the ruler so that it is tangent to arc
Curves of Constant Width 39
ABC at one of its points, then use the ruler's opposite edge to draw a
parallel line. Do this at many points, from one end of arc ABC to the
other. The missing part of the curve is the envelope of these lines. In
this way you can obtain rough outlines of an endless variety of lopsided
curves of constant width.
It should be mentioned that the arc ABC cannot be completely arbi-
trary. Roughly speaking, its curvature must not at any point be less
than the curvature of a circle with a radius equal to the side of the
square. It cannot, for example, include straight-line segments. For a
more precise statement on this, as well as detailed proofs of many ele-
mentary theorems involving curves of constant width, the reader is re-
ferred to the excellent chapter on such curves in The Enjoyment of
Mathematics, by Hans Rademacher and Otto Toeplitz.
If you have the tools and skills for woodworking, you might enjoy
making a number of wooden rollers with cross sections that are various
curves of the same constant width. Most people are nonplused by the
sight of a large book rolling horizontally across such lopsided rollers
without bobbing up and down. A simpler way to demonstrate such
curves is to cut from cardboard two curves of constant width and nail
them to opposite ends of a wooden rod about six inches long. The
curves need not be of the same shape, and it does not matter exactly
where you put each nail as long as it is fairly close to what you guess
to be the curve's "center." Hold a large, light-weight empty box by its
ends, rest it horizontally on the attached curves and roll the box back
and forth. The rod wobbles up and down at both ends, but the box
rides as smoothly as it would on circular rollers!
The properties of curves of constant width have been extensively in-
vestigated. One startling property, not easy to prove, is that the perime-
ters of all curves with constant width n have the same length. Since a
circle is such a curve, the perimeter of any curve of constant width n
must of course be 1tn, the same as the circumference of a circle with di-
ameter n.
The three-dimensional analogue of a curve of constant width is the
solid of constant width. A sphere is not the only such solid that will ro-
tate within a cube, at all times touching all six sides of the cube; this
property is shared by all solids of constant width. The Simplest exam-
ple of a nonspherical solid of this type is generated by rotating the
Reuleaux triangle around one of its axes of symmetry (see Figure 4.6,
left). There is an infinite number of others. The solids of constant width
40 PLANE GEOMETRY
4.6. Two constant
41
pentagons, hexagons, and even octagons. In three-space, Goldberg has
shown, there are nonspherical rotors for the regular tetrahedron and oc-
tahedron, as well as the cube, but none for the regular dodecahedron
and icosahedron. Almost no work has been done on rotors in dimen-
sions higher than three.
Figure 4.1.
Least-area rotor in equilateral triangle Line rotated in deltoid curve
Closely related to the theory of rotors is a famous problem named the
Kakeya needle problem after the Japanese mathematician Soichi
Kakeya, who first posed it in 1917. The problem is as follows: What is
the plane figure of least area in which a line segment of length 1 can be
rotated 360 degrees? The rotation obviously can be made inside a cir-
cle of unit diameter, but that is far from the smallest area.
For many years mathematicians believed the answer was the deltoid
curve shown at the right of Figure 4.7, which has an area exactly half
that of a unit circle. (The deltoid is the curve traced by a point on the
circumference of a circle as it rolls around the inside of a larger circle,
when the diameter of the small circle is either one third or two thirds
that of the larger one.) If you break a toothpick to the size of the line seg-
ment shown, you will find by experiment that it can be rotated inside
the deltoid as a kind of one-dimensional rotor. Note how its end points
remain at all times on the deltoid's perimeter.
In 1927, ten years after Kakeya popped his question, the Russian
mathematician Abram Samoilovitch Besicovitch (then living in Copen-
hagen) dropped a bombshell. He proved that the problem had no an-
swer. More accurately, he showed that the answer to Kakeya's question
is that there is no minimum area. The area can be made as small as one
wants. Imagine a line segment that stretches from the earth to the moon.
We can rotate it 360 degrees within an area as small as the area of a
postage stamp. If that is too large, we can reduce it to the area of Lin-
coln's nose on a postage stamp.
42 PLANE GEOMETRY
Besicovitch's proof is too complicated to give here (see references in
bibliography), and besides, his domain of rotation is not what topolo-
gists call simply connected. For readers who would like to work on a
much easier problenl: What is the smallest convex area in which a line
segment of length 1 can be rotated 360 degrees? (A convex figure is one
in which a straight line, joining any two of its points, lies entirely on
the figure. Squares and circles are convex; Greek crosses and crescent
moons are not.)
Addendum
Although Watts was the first to acquire patents on the process
of drilling square holes with Reuleaux-triangle drills, the procedure
was apparently known earlier. Derek Beck, in London, wrote that he
had met a man who recalled having used such a drill for boring square
holes when he was an apprentice machinist in 1902 and that the prac-
tice then seemed to be standard. I have not, however, been able to learn
anything about the history of the technique prior to Watts's 1917
patents.
In 1969 England introduced a 50-pence coin with seven slightly
curved sides that form a circle of constant width, surely the first seven-
sided coin ever minted. The invariant width allows the coin to roll
smoothly down coin-operated machines.
Answers
What is the smallest convex area in which a line segment of
length 1 can be rotated 360 degrees? The answer: An equilateral trian-
gle with an altitude of 1. (The area is one third the square root of 3.)
Any figure in which the line segment can be rotated obviously must
have a width at least equal to 1. Of all convex figures with a width of
1, the equilateral triangle of altitude 1 has the smallest area. (For a
proof of this the reader is referred to Convex Figures, by I. M. Yaglom
and V. G. Boltyanskii, pp. 221-22.) It is easy to see that a line segment
of length 1 can in fact be rotated in such a triangle (see Figure 4.8).
The deltoid curve was believed to be the smallest simply connected
area solving the problem until 1963 when a smaller area was discov-
ered independently by Melvin Bloom and I. J. Schoenberg.
Curves of Constant Width 43
1 2 3
4 5 6
Figure 4.8. Answer to the needle-turning problem
Bibliography
W. Blaschke, Kreis und Kugel, Leipzig, 1916; Berlin: W. de Gruyter, 1956.
J. H. Cadwell, Topics in Recreational Mathematics, Cambridge, England: Cam-
bridge University Press, 1966, Chapter 15.
J. Casey, "Perfect and Not-So-Perfect Rollers," The Mathematics Teacher, Vol. 91,
January 1998, pp. 12-20.
G. D. Chakerian and H. Groemer, "Convex Bodies of Constant Width," Convexity
and Its Applications, P. M. Gruber and J. M. Wills (eds.), Boston: Birkhauser,
1983.
J. A. Dossey, "What?-A Roller with Corners?," Mathematics Teacher, December
1972, pp. 720-24.
J. C. Fisher, "Curves of Constant Width from a Linear Viewpoint," Mathematics
Magazine, Vol. 60, June 1987, pp. 131-40.
J. A. Flaten, "Curves of Constant Width," The Physics Teacher, Vol. 37, October
1999, pp. 418-19.
M. Goldberg, "Trammel Rotors in Regular Polygons," American Mathematical
Monthly, Vol. 64, February 1957, pp. 71-78.
M. Goldberg, "Rotors in Polygons and Polyhedra," Mathematical Tables and Other
Aids to Computation, Vol. 14, July 1960, pp. 229-39.
M. Goldberg, "N-Gon Rotors Making N + 1 Contacts with Fixed Simple Curves,"
American Mathematical Monthly, Vol. 69, June/July 1962, pp. 486-91.
M. Goldberg, "Two-lobed Rotors with Three-Lobed Stators," Journal of Mecha-
nisms, Vol. 3, 1968, pp. 55-60.
C. G. Gray, "Solids of Constant Breadth," Mathematical Gazette, December 1972,
pp.289-92.
R. Honsberger, Mathematical Gems, Washington, D.C., 1973, Chapter 5.
H. Rademacher and O. Toeplitz, The Enjoyment of Mathematics, Princeton, NJ:
Princeton University Press, 1957; Dover, 1990. See pp. 163-77, 203.
44 PLANE GEOMETRY
F. Reuleaux, The Kinematics of Machinery, New York: Macmillan, 1876; Dover
Publications, 1964, pp. 129-46.
S. G. Smith, "Drilling Square Holes," The Mathematics Teacher, Vol. 86, October
1993, pp. 579-83.
I. M. Yaglom and V. G. Boltyanskii, Convex Figures, New York: Holt, Rinehart &
Winston, 1961, Chapters 7 and 8.
On I<.akeya's needle problem:
A. S. Besicovitch, "The Kakeya Problem," American Mathematical Monthly, Vol.
70, August/September 1963, pp. 697-706.
A. A. Blank, "A Remark on the Kakeya Problem," American Mathematical Monthly,
Vol. 70, August/September 1963, pp. 706-11.
J. H. Cadwell, Topics in Recreational Mathematics, Cambridge, England: Cam-
bridge University Press, 1966. See pp. 96-99.
F. Cunningham, Jr., "The Kakeya Problem for Simply Connected and Star-shaped
Sets," American Mathematical Monthly, Vol. 78, February 1971, pp. 114-29.
I. M. Yaglom and V. G. Boltyanskii, Convex Figures, New York: Holt. Rinehart &
Winston, 1961, pp. 61-62, 226-27.
Curves of Constant Width 45
Chapter 5 Rep-Tiles
Only three regular polygons-the equilateral triangle, the
square, and the regular hexagon-can be used for tiling a floor in such
a way that identical shapes are endlessly repeated to cover the plane.
But there is an infinite number of irregular polygons that can provide
this kind of tiling. For example, a triangle of any shape whatever will
do the trick. So will any four-sided figure. The reader can try the fol-
lowing test. Draw an irregular quadrilateral (it need not even be con-
vex, which is to say that it need not have interior angles that are all less
than 180 degrees) and cut 20 or so copies from cardboard. It is a pleas-
ant task to fit them all together snugly, like a jigsaw puzzle, to cover a
plane.
There is an unusual and less familiar way to tile a plane. Note that
each trapezoid at the top of Figure 5.1 has been divided into four
smaller trapezoids that are exact replicas of the original. The four repli-
cas can, of course, be divided in the same way into four still smaller
replicas, and this can be continued to infinity. To use such a figure for
tiling we have only to proceed to infinity in the opposite direction: we
put together four figures to form a larger model, four of which will in
turn fit together to make a still larger one. The British mathematician
Augustus De Morgan summed up this sort of situation admirably in the
following jingle, the first four lines of which paraphrase an earlier jin-
gle by Jonathan Swift:
46
Great fleas have little fleas
Upon their backs to bite 'em,
And little fleas have lesser fleas,
And so ad infinitum.
The great fleas themselves, in turn,
Have greater fleas to go on;
While these again have greater still,
And greater still, and so on.
Until 1962 not much was known about polygons that have this curi-
ous property of making larger and smaller copies of themselves. In
1962 Solomon W. Golomb, who was then on the staff of the Jet Propul-
sion Laboratory of the California Institute of Technology and is now a
professor at the University of Southern California, turned his attention
to these "replicating figures"-or "rep-tiles," as he calls them. The re-
sult was three privately issued papers that lay the groundwork for a
general theory of polygon "replication." These papers, from which al-
most all that follows is extracted, contain a wealth of material of great
interest to the recreational mathematician.
In Golomb's terminology a replicating polygon of order k is one that
can be divided into k replicas congruent to one another and similar to
the original. Each of the three trapezoids in Figure 5.1, for example, has
a replicating order of 4, abbreviated as rep-4. Polygons of rep-k exist for
any k, but they seem to be scarcest when k is a prime and to be most
abundant when k is a square number.
Three trapezoids that have a replicating order of 4
Figure 5.1. The only known rep-2 polygons
Only two rep-2 polygons are known: the isosceles right triangle and
the parallelogram with sides in the ratio of 1 to the square root of 2 (see
bottom of Figure 5.1). Golomb found simple proofs that these are the
only possible rep-2 triangles and quadrilaterals, and there are no other
convex rep-2 polygons. The existence of concave rep-2 polygons ap-
pears unlikely, but so far their nonexistence has not been proved.
Rep-Tiles 47
The interior angles of the parallelogram can vary without affecting its
rep-2 property. In its rectangular form the rep-2 parallelogram is al-
most as famous in the history of art as the "golden rectangle," discussed
in the Second Scientific American Book of Mathematical Puzzles and
Diversions. Many medieval and Renaissance artists (Albrecht Diirer,
for instance) consciously used it for outlining rectangular pictures. A
trick playing card that is sometimes sold by street-corner pitchmen ex-
ploits this rectangle to make the ace of diamonds seem to diminish in
size three times (see Figure 5.2). Under cover of a hand movement the
card is secretly folded in half and turned over to show a card exactly
half the size of the preceding one. If each of the three smaller aces is a
rectangle similar to the original, it is easy to show that only a 1-by-v2
rectangle can be used for the card. The rep-2 rectangle also has less friv-
olous uses. Printers who wish to standardize the shape of the pages in
books of various sizes find that in folio, quarto, or octavo form it pro-
duces pages that are all similar rectangles. European writing paper also
has a similar shape.
A

+

y



Figure S.2. A trick diminishing card based on the rep-2 rectangle
The rep-2 rectangle belongs to the family of parallelograms shown in
the top illustration of Figure 5.3. The fact that a parallelogram with
sides of 1 and Vk is always rep-k proves that a rep-k polygon exists for
any k. It is the only known example, Golomb asserts, of a family of fig-
ures that exhibit all the replicating orders. When k is 7 (or any prime
greater than 3 that has the form 4n - 1), a parallelogram of this family
is the only known example. Rep-3 and rep-5 triangles exist. Can the
reader construct them?
A great number of rep-4 figures are known. Every triangle is rep-4
and can be divided as shown in the second illustration from the top of
48 PLANE GEOMETRY
V2 VS" 2
I l; Ill; 0Zll
V5 V6
IIlTll !lATII
The I-by-vk parallelogram is a rep-k polygon
Every triangle and parallelogram is rep-4
The Sphinx, the only known rep-4 pentagon
I
I I
U f--
The three known varieties of rep-4 hexagons
Figure S.3.
Rep-Tiles 49
Figure 5.3. Among the quadrilaterals, any parallelogram is rep-4, as
shown in the same illustration. The three trapezoids in the top illus-
tration of Figure 5.1 are the only other examples of rep-4 quadrilater-
als so far discovered.
Only one rep-4 pentagon is known: the sphinx-shaped figure in the
third illustration from the top of Figure 5.3. Golomb was the first to dis-
cover its rep-4 property. Only the outline of the sphinx is given so that
the reader can have the pleasure of seeing how quickly he can dissect
it into four smaller sphinxes. (The name "sphinx" was given to this fig-
ure by T. H. O'Beirne of Glasgow.)
There are three known varieties of rep-4 hexagons. If any rectangle is
divided into four quadrants and one quadrant is thrown away, the re-
maining figure is a rep-4 hexagon. The hexagon at the right at the bot-
tom of Figure 5.3 shows the dissection (familiar to puzzlists) when the
rectangle is a square. The other two examples of rep-4 hexagons (each
of which can be dissected in more than one way) are shown at the mid-
dle and left in the same illustration.
No other example of a standard polygon with a rep-4 property is
known. There are, however, "stellated" rep-4 polygons (a stellated poly-
gon consists of two or more polygons joined at single points), two exam-
ples of which, provided by Golomb, are shown at the top of Figure 5.4.
In the first example a pair ofidentical rectangles can be substituted for the
squares. In addition, Golomb has found three nonpolygonal figures that
are rep-4, although none is constructible in a finite number of steps. Each
of these figures, shown at the left in the bottom illustration of Figure 5.4,
is formed by adding to an equilateral triangle an endless series of smaller
triangles, each one-fourth the size of its predecessor. In each case four of
these figures will fit together to make a larger replica, as shown at the
right in the same illustration. (There is a gap in each replica because the
original cannot be drawn with an infinitely long series of triangles.)
It is a curious fact that every known rep-4 polygon of a standard type
is also rep-9. The rep-4 Nevada-shaped trapezoid of Figure 5.5 can be
dissected into nine replicas in many ways, only one of which is shown.
(Can the reader dissect each of the other rep-4 polygons, not counting
the stellated and infinite forms, into nine replicas?) The converse is
also true: All known standard rep-9 polygons are also rep-4.
Three interesting examples of stellated rep-9 polygons, discovered
and named by Golomb, are shown in Figure 5.B. None of these poly-
gons is rep-4.
50 PLANE GEOMETRY
1\vo stellated
5.4. Three eXSllDJ:ues
5.5. is
51
(A),
(B).
52
S.1. A
5.8.
53
144. It is conjectured that none of the three has a lower replicating
order.
Figure 5.9. Three rep-144 polygons
Golomb has noted that every known polygon of rep-4, including the
stellated polygons, will divide a parallelogram with a multiplicity of 2.
In other words, if any known rep-4 polygon is replicated, the pair can
be fitted together to form a parallelogram! It is conjectured, but not yet
proved, that this is true of all rep-4 polygons.
An obvious extension of Golomb's pioneer work on replication the-
ory (of which only the most elementary aspects have been detailed
here) is into three or even higher dimensions. A trivial example of a
replicating solid figure is the cube: it obviously is rep-B, rep-27, and so
on for any order that is a cubical number. Other trivial examples result
from giving plane replicating figures a finite thickness, then forming
layers of larger replicas to make a model of the original solid. Less triv-
ial examples certainly exist; a study of them might lead to significant
results.
In addition to the problems already posed, here are two unusual dis-
section puzzles closely related to what we have been considering (see
Figure 5.10). First the easier one: Can the reader divide the hexagon
(left) into two congruent stellated polygons? More difficult: Divide the
pentagon (right) into four congruent stellated polygons. In neither case
are the polygons similar to the original figure.
Addendum
I gave the conjecture that none of the three polygons shown in
Figure 5.9 has a replicating order lower than 144. Wrong for all three!
MarkA. Mandel, then 14, wrote to show how the middle polygon could
be cut into 36 replicas (see Figure 5.11). Robert Reid, writing from Peru,
found a way for 121 copies of the first hexomino to replicate and 64
54 PLANE GEOMETRY
Figure S.ID. Two dissection problems
Figure 5.11. Hexomino: rep-tile of order 36
Figure 5.12. A Rep-at polygon
copies of the third polygon to do the same. Reid also proved that the
hexomino shown in Figure 5.12 is a rep-tile of order 81.
Rep-Tiles 55
Ralph H. Hinrichs, Phoenixville, PA, discovered that if the middle
hexagon at the bottom of Figure 5.3 is dissected in a slightly different
way (the pattern within each rectangle is mirror-reflected), the entire
figure can undergo an infinite number of affine transformations (the 90-
degree exterior angle taking any acute or obtuse value) to provide an in-
finity of rep-4 hexagons. Only when the angle is 90 degrees is the figure
also rep-9, thus disproving an early guess that all rep-4 standard poly-
gons are rep-9 and vice versa.
The three nonpolygon rep-tiles shown in Figure 5.4 are now called
"self-similar fractals." Such fractals are being extensively studied. I
have not listed recent references in my bibliography, but interested
readers can consult Christoph Bandt's "Self-Similar Sets 5," in the Pro-
ceedings of the American Mathematical Society, Vol. 112, June 1991,
pages 349-62, and his list of references.
The rep-tile triangle second from the top in Figure 5.12-incidently
it is not accurately drawn-is mentioned in Plato's Timaeus. Timaeus
points out that it is half of an equilateral triangle and that he considers
it the "most beautiful" of all scalene triangles. (See the Random House
edition of Plato, edited by Benjamin Jowett, Vol. 2, p. 34.)
Sol Golomb is best known for his work on polyominoes-shapes
formed by joining n unit squares along their edges. Golomb's classic
study, Polyominoes, was reissued by Princeton University Press in
1994.
The most spectacular constructions of what Golomb calls "infin-tiles
(rep-tiles with infinitely many sides) are in the papers by Jack Giles, Jr.,
cited in the bibliography. Giles calls them "superfigures." Many of
Golomb's infin-tiles, and those of Giles, are early examples of fractals.
Golomb tells me that Giles was a parking lot attendant in Florida when
he sent his papers to Golomb, who in turn submitted them to the Jour-
nal of Combinatorial Theory.
Answers
The problem of dissecting the sphinx is shown in Figure 5.13,
top. The next two illustrations show how to construct rep-3 and rep-5
triangles. The bottom illustration gives the solution to the two dissec-
tion problems involving stellated polygons. The first of these can be
varied in an infinite number of ways; the solution shown here is one of
the simplest.
56 PLANE GEOMETRY
S.I l. Solutions to dissection problems
second solution is an old-timer. Loyd, in his column
Woman's Home Companion (October 1905) out that the fig-
a
that he a year trying to cut the mitre shape into four congruent
Rep-Tiles 57
parts, each simply connected, but was unable to do better than the so-
lution reproduced here. It can be found in many old puzzle books a n ~
tedating Loyd's time.
Bibliography
R. O. Davies, "Replicating Boots," Mathematical Gazette, Vol. 50, May 1966, p.
157.
F. M. Dekking. "Replicating Superfigures and Endomorphisms of Free Groups,"
Journal of Combinatorial Theory, Vol. 32A, 1982, pp. 315-20.
J. Doyen and M. Lenduyt, "Dissection of Polygons," Annals of Discrete Mathemat-
ics, Vol. 18, 1983, pp. 315-18.
J. Giles, Jr., "Infinite-level Replication Dissections of Plane Figures," "Construction
of Replicating Superfigures, " and "Superfigures Replicating with Polar Symme-
try," Journal of Combinatorial Theory, 26A, 1979, pp. 319-27,328-34,335-37.
J. Giles, Jr., "The Gypsy Method of Superfigure Construction," Journal of Recre-
ational Mathematics, Vol. 13, 1980/81, pp. 97-101.
M. Goldberg and B. M. Stewart, "A Dissection Problem for Sets of Polygons," Amer-
ican Mathematical Monthly, Vol. 71, December 1964, pp. 1077-95.
S. W. Golomb, "Replicating Figures in the Plane," Mathematical Gazette, Vol. 48,
December 1964, pp. 403-12.
H. D. Grossman, "Fun with Lattice Points," Scripta Mathematica, Vol. 14, June
1948, pp. 157-59.
C. D. Langford, "Uses of a Geometrical Puzzle," Mathematical Gazette, Vol. 24,
July 1940, pp. 209-11.
R. Sibson, "Comments on Note 1464," Mathematical Gazette, Vol. 24, December
1940, p. 343.
G. Valette and T. Zamfirescu, "Les Partages d'un Polygone Convexe en 4 Polygones
Semblables au Premier," Journal of Combinatorial Theory, Ser. B, Vol. 16, 1974,
pp.1-16.
58 PLANE GEOMETRY
Chapter 6
There is one art, no more, no less: to do all
things with artlessness. -Piet Hein
Piet Hein's
Superellipse
Civilized man is surrounded on all sides, indoors and out, by
a subtle, seldom-noticed conflict between two ancient ways of shaping
things: the orthogonal and the round. Cars on circular wheels, guided
by hands on circular steering wheels, move along streets that intersect
like the lines of a rectangular lattice. Buildings and houses are made up
mostly of right angles, relieved occasionally by circular domes and
windows. At rectangular or circular tables, with rectangular napkins on
our laps, we eat from circular plates and drink from glasses with cir-
cular cross sections. We light cylindrical cigarettes with matches from
rectangular packs, and we pay the rectangular bill with rectangular
credit cards, checks, or dollar bills and circular coins.
Even our games combine the orthogonal and the round. Most outdoor
sports are played with spherical balls on rectangular fields. Indoor
games, from pool to checkers, are similar combinations of the round
and the rectangular. Rectangular playing cards are held in a fanlike cir-
cular array. The very letters on this rectangular page are patchworks of
right angles and circular arcs. Wherever one looks the scene swarms
with squares and circles and their affinely stretched forms: rectangles
and ellipses. (In a sense the ellipse is more common than the circle, be-
cause every circle appears elliptical when seen from an angle.) In op
paintings and textile designs, squares, circles, rectangles, and ellipses
jangle against one another as violently as they do in daily life.
The Danish writer and inventor Piet Hein asked himself a fascinat-
ing question: What is the simplest and most pleasing closed curve that
mediates fairly between these two clashing tendencies? Originally a
scientist, Piet Hein (he is always spoken of by both names) was well
known throughout Scandinavia and English-speaking countries for his
enormously popular volumes of gracefully aphoristic poems (which
59
critics have likened to the epigrams of Martial) and for his writings on
scientific and humanistic topics. To recreational mathematicians he is
best known as the inventor of the game Hex, of the Soma cube, and of
other remarkable games and puzzles. He was a friend of Norbert
Wiener, whose last book, God and Golem, Inc., is dedicated to him.
The question Piet Hein asked himself had been suggested by a knotty
city-planning problem that first arose in 1959 in Sweden. Many years
earlier Stockholm had decided to raze and rebuild a congested section
of old houses and narrow streets in the heart of the city, and after World
War IT this enormous and costly program got under way. Two broad
new traffic arteries running north-south and east-west were cut through
the center of the city. At the intersection of these avenues a large rec-
tangular space (now called SergeI's Square) was laid out. At its center is
an oval basin with a fountain surrounded by an oval pool containing
several hundred smaller fountains. Daylight filters through the pool's
translucent bottom into an oval self-service restaurant, below street
level, surrounded by oval rings of pillars and shops. Below that there are
two more oval floors for dining and dancing, cloakrooms, and kitchen.
In planning the exact shape of this center the Swedish architects ran
into unexpected snags. The ellipse had to be rejected because its
pointed ends would interfere with smooth traffic flow around it; more-
over, it did not fit harmoniously into the rectangular space. The city
planners next tried a curve made up of eight circular arcs, but it had a
patched-together look with ugly "jumps" of curvature in eight places.
In addition, plans called for nesting different sizes of the oval shape,
and the eight-arc curve refused to nest in a pleasing way.
At this stage the architectural team in charge of the project consulted
Piet Hein. It was just the kind of problem that appealed to his combined
mathematical and artistic imagination, his sense of humor, and his
knack of thinking creatively in unexpected directions. What kind of
curve, less pointed than the ellipse, could he discover that would nest
pleasingly and fit harmoniously into the rectangular open space at the
heart of Stockholm?
To understand Piet Hein's novel answer we must first consider the el-
lipse, as he did, as a special case of a more general family of curves with
the following formula in Cartesian coordinates:
60 PLANE GEOMETRY
where a and b are unequal parameters (arbitrary constants) that repre-
sent the two semiaxes of the curve and n is any positive real number.
The vertical brackets indicate that each fraction is to be taken with re-
spect to its absolute value; that is, its value without regard to sign.
When n =2, the real values of x and ythat satisfy the equation (its "so-
lution set") determine the points on the graph that lie on an ellipse with
its center at the origin of the two coordinates. As n decreases from 2 to
1, the oval becomes more pointed at its ends ("subellipses," Piet Rein
called them). When n = 1, the figure is a parallelogram. When n is less
than 1, the four sides are concave curves that become increasingly con-
cave as n approaches O. At n = 0 they degenerate into two crossed
straight lines.
If n is allowed to increase above 2, the oval develops flatter and flat-
ter sides, becoming more and more like a rectangle; indeed, the rec-
tangle is its limit as n approaches infinity. At what point is such a
curve most pleasing to the eye? Piet Rein settled on n = 2%. With the
help of a computer, 400 coordinate pairs were calculated to 15 decimal
places and larger, precise curves were drawn in many different sizes,
all with the same height-width ratios (to conform with the proportions
of the open space at the center of Stockholm), The curves proved to be
strangely satisfying, neither too rounded nor too orthogonal, a happy
blend of elliptical and rectangular beauty. Moreover, such curves could
be nested, as shown in Figures 6.1 and 6.2, to give a strong feeling of
harmony and parallelism between the concentric ovals. Piet Rein
called all such curves with exponents above 2 "superellipses." Stock-
holm immediately accepted the 2 1/2-exponent superellipse as the
basic motif of its new center. Already the large superelliptical pool has
conferred upon Stockholm an unusual mathematical flavor, like the
big catenary curve of St. Louis's Gateway Arch, which dominates the
local skyline.
Meanwhile Piet Rein's superellipse has been enthusiastically
adopted by Bruno Mathsson, a well-known Swedish furniture designer.
Re first produced a variety of superelliptical desks, now in the offices
of many Swedish executives, and has since followed with superellip-
tical tables, chairs, and beds. (Who needs the corners?) Industries in
Denmark, Sweden, Norway, and Finland turned to Piet Rein for solu-
tions to various orthogonal-versus-circular problems, and he worked on
superelliptical furniture, dishes, coasters, lamps, silverware, textile
patterns, and so on. The tables, chairs, and beds embodied another Piet
Piet Rein's Superellipse 61
Figure 6.1. Concentric superellipses
Figure 6.2. Plan of Stockholm's underground restaurants and the pools above them
62 PLANE GEOMETRY
IS
6.3.
on
63
In the same way, one can raise the exponent in the corresponding
Cartesian formulas for spheres and ellipsoids to obtain what Piet Hein
called "superspheres" and "superellipsoids." If the exponent is 2%,
such solids can be regarded as spheres and ellipsoids that are halfway
along the road to being cubes and bricks.
The true ellipsoid, with three unequal axes, has the formula
x
2
Z2
2" + b
2
+ 2 = 1,
a c
where a, b, and c are unequal parameters representing half the length
of each axis. When the three parameters are equal, the figure is a
sphere. When only two are equal, the surface is called an "ellipsoid of
rotation" or a spheroid. It is produced by rotating an ellipse on either
of its axes. If the rotation is on the longer axis, the result is a prolate
spheroid-a kind of egg shape with circular cross sections perpendic-
ular to the axis.
It turns out that a solid model of a prolate spheroid, with homoge-
neous density, will no more balance upright on either end than a
chicken egg will, unless one applies to the egg a stratagem usually cred-
ited to Columbus. Columbus returned to Spain in 1493 after having
discovered America, thinking that the new land was India and that he
had proved the earth to be round. At Barcelona a banquet was given in
his honor. This is how Girolamo Benzoni, in his History of the New
World (Venice, 1565), tells the story (I quote from an early English trans-
lation):
Columbus, being at a party with many noble Spaniards ... one of them
undertook to say: "Mr. Christopher, even if you had not found the Indies,
we should not have been devoid of a man who would have attempted
the same thing that you did, here in our own country of Spain, as it is
full of great men clever in cosmography and literature." Columbus said
nothing in answer to these words, but having desired an egg to be
brought to him, he placed it on the table saying: "Gentlemen, I will lay
a wager with any of you, that you will not make this egg stand up as I
will, naked and without anything at all." They all tried, and no one suc-
ceeded in making it stand up. When the egg came round to the hands of
Columbus, by beating it down on the table he fixed it, having thus
crushed a little of one end; wherefore all remained confused, under-
standing what he would have said: That after the deed is done, every-
body knows how to do it.
64 PLANE GEOMETRY
The story may be true, but a suspiciously similar story had been told
15 years earlier by Giorgio Vasari in his celebrated Lives of the Most Em-
inent Painters, Sculptors and Architects (Florence, 1550). Young Fil-
ippo Brunelleschi, the Italian architect, had designed an unusually
large and heavy dome for Santa Maria del Fiore, the cathedral of Flo-
rence. City officials had asked to see his model, but he refused,
"proposing instead ... that whosoever could make an egg stand upright
on a flat piece of marble should build the cupola, since thus each man's
intellect would be discerned. Taking an egg, therefore, all those Masters
sought to make it stand upright, but not one could find a way. Where-
upon Filippo, being told to make it stand, took it graciously, and, giv-
ing one end of it a blow on the flat piece of marble, made it stand
upright. The craftsmen protested that they could have done the same;
but Filippo answered, laughing, that they could also have raised the
cupola, if they had seen the model or the design. And so it was re-
solved that he should be commissioned to carry out this work."
The story has a topper. When the great dome was finally completed
(many years later, but decades before Columbus's first voyage), it had
the shape of half an egg, flattened at the end.
What does all this have to do with supereggs? Well, Piet Hein (my
source, by the way, for the references on Columbus and Brunelleschi)
discovered that a solid model of a 2Vz-exponent superegg-indeed, a
superegg of any exponent-if not too tall for its width, balances im-
mediately on either end without any sort of skulduggery! Indeed,
dozens of chubby wooden and silver supereggs are now standing po-
litely and permanently on their ends all over Scandinavia.
Consider the silver superegg shown in Figure 6.3, which has an ex-
ponent of 2Vz and a height-width ratio of 4 : 3. It looks as if it should
topple over, but it does not. This spooky stability of the superegg (on
both ends) can be taken as symbolic of the superelliptical balance be-
tween the orthogonal and the round, which is in turn a pleasant sym-
bol for the balanced mind of individuals such as Piet Hein who
mediated so successfully between C. P. Snow's "two cultures."
Addendum I
The family of plane curves expressed by the formula I xl a In +
I yl bin = 1 was first recognized and studied by Gabriel Lame, a 19th-
century French physicist, who wrote about them in 1818. In France they
Piet Rein's Supere1Jipse 65
are called courbes de Lame; in Germany, Lamesche kurven. The curves
are algebraic when n is rational, transcendent when n is irrational.
When n = 2/3 and a = b (see Figure 6.4) the curve is an astroid. This
is the curve generated by a point on a circle that is one-fourth or three-
fourths the radius of a larger circle, when the smaller circle is rolled
around the inside of the larger one. Solomon W. Golomb called atten-
tion to the fact that if n is odd, and the absolute value signs are dropped
in the formula for Lame curves, you get a family of curves of which the
famous Witch of Agnesi (The curve studied by Maria Gaetana Agnesi)
is a member. (The witch results when n = 3.) William Hogan wrote to
say that parkway arches, designed by himself and other engineers, often
are Lame curves of exponent 2.2. In the thirties, he said, they were
called "2.2 ellipses."
Figure 6.4. Supercircle and related curves
When a superellipse (a Lame curve with exponent greater than 2) is
applied to a physical object, its exponent and parameters a and b can,
of course, be varied to suit circumstances and taste. For the Stockholm
center, Piet Hein used the parameters n = 2% and alb = 6/5. A few years
66 PLANE GEOMETRY
later Gerald Robinson, a Toronto architect, applied the superellipse to
a parking garage in a shopping center in Peterborough, a Toronto sub-
urb. The length and width were required to be in the ratio alb = 9/7.
Given this ratio, a survey indicated that an exponent slightly greater
than 2.7 produced a superellipse that seemed the most pleasing to those
polled. This suggested e as an exponent (since e = 2.718 ... ).
Readers suggested other parameters. J. D. Turner proposed mediating
between the extremes of circle and square (or rectangle and ellipse) by
picking the exponent that would give an area exactly halfway between
the two extreme areas. D. C. Mandeville found that the exponent me-
diating the areas of a circle and square is so close to pi that he won-
dered if it actually is pi. Unfortunately it is not. Norton Black, using a
computer, determined that the value is a trifle greater than 3.17. Thrner
also proposed mediating between ellipse and rectangle by choosing an
exponent that sends the curve through the midpoint of a line joining
the rectangle's corner to the corresponding point on the ellipse.
Turner and Black each suggested that the superellipse be combined
with the aesthetically pleasing "golden rectangle" by making alb the
golden ratio. Turner's vote for the most pleasing superellipse went to
the oval with parameters alb = golden ratio and n = e. Michel L. Balin-
ski and Philetus H. Holt III, in a letter published in The New York
Times in December 1968 (I failed to record the day of the month) rec-
ommended a golden superellipse with n = 21fz as the best shape for the
negotiating table in Paris. At that time the diplomats preparing to ne-
gotiate a Vietnam peace were quarreling over the shape of their table.
If no table can be agreed upon, Balinski and Holt wrote, the diplomats
should be put inside a hollow superegg and shaken until they are in
"su perelli ptic agreement."
The superegg is a special case of the more general solid shape which
one can call a superellipsoid. The superellipsoid's formula is
I
Xl
n
IYln IZl
n
- + - + - = 1.
abc
When a = b = c, the solid is a supersphere, its shape varying from
sphere to cube as the exponent varies. When a = b, the solid is super-
circular in cross section, with the formula
Piet Rein's Superellipse 67
Supereggs, with circular cross-sections, have the formula
When I wrote my column on the superellipse, I believed that any
solid superegg based on an exponent greater than 2 and less than in-
finity would balance on its end provided its height did not exceed its
width by too great a ratio. A solid superegg with an exponent of infin-
ity would, of course, be a right circular cylinder that would, in princi-
ple, stand on its flat end regardless of how much higher it was than
wide. But short of infinity it seemed intuitively clear that for each ex-
ponent there was a critical ratio beyond which the egg would be un-
stable. Indeed, I even published the following proof that this is the case:
If the center of gravity, CG, of an egg is below the center of curvature, CC,
of the egg's base at the central point of the base, the egg will balance. It
balances because any tipping of the egg will raise the CG. If the CG is
above the CC, the egg is unstable because the slightest tipping lowers the
CG. To make this clear, consider first the sphere shown at the left in Fig-
ure 6.5. Inside the sphere the CG and CC are the same point: the center
of the sphere. For any supersphere with an exponent greater than 2, as
shown second from left in the illustration, the CC is above the CG be-
cause the base is less convex. The higher the exponent, the less convex
the base and the higher the CC.
Now suppose the supersphere is stretched uniformly upward along its
vertical coordinates, transforming it into a superellipsoid of rotation, or
what Piet Hein calls a superegg. As it stretches, the CC falls and the CG
rises. Clearly there must be a point X where the CC and the CG coincide.
Before this crucial point is reached the superegg is stable, as shown third
from left in Figure 6.5. Beyond that point the superegg is unstable (right).
C. E. Gremer, a retired U.S. Navy commander, was the first of many
readers to inform me that the proof is faulty. Contrary to intuition, at the
base point of all supereggs, the center of curvature is infinitely high! If
we increase the height of a superegg while its width remains constant,
the curvature at the base point remains "flat." German mathematicians
call it a flachpunkt. The superellipse has a similar flachpunkt at its
ends. In other words, all supereggs, regardless of their height-width
ratio, are theoretically stable! As a superegg becomes taller and thinner,
there is of course a critical ratio at which the degree of tilt needed to
68 PLANE GEOMETRY
Figure 6.5. Diagrams for a false proof of superegg instability
topple it comes so close to zero that such factors as inhomogeneity of
the material, surface irregularity, vibrations, air currents, and so on
make it practically unstable. But in a mathematically ideal sense there
is no critical height-width ratio. As Piet Hein put it, in theory one can
balance any number of supereggs, each an inch wide and as tall as the
Empire State Building, on top of one another, end to end, and they will
not fall. Calculating precise "topple angles" at which a given superegg
will not regain balance is a tricky problem in calculus. Many readers
tackled it and sent their results.
Speaking of egg balancing, the reader may not know that almost any
chicken egg can be balanced on its broad end, on a smooth surface, if
one is patient and steady-handed. Nothing is gained by shaking the
egg first in an attempt to break the yolk. Even more puzzling as a par-
lor trick is the following method of balancing an egg on its pointed
end. Secretly put a tiny amount of salt on the table, balance the egg on
it, then gently blow away the excess grains before you call in viewers.
The few remaining grains which hold the egg are invisible, especially
on a white surface. For some curious reason, balancing chicken eggs le-
gitimately on their broad ends became a craze in China in 1945-at
least, so said Life in its picture story of April 9, 1945.
The world's largest superegg, made of steel and aluminum and
weighing almost a ton, was set up outside Kelvin Hall in Glasgow, Oc-
tober 1971, to honor Piet Hein's appearance there as a speaker during
an exhibition of modern homes. The superellipse has twice appeared
on Danish postage stamps: In 1970 on a blue two-kroner honoring
Bertel Thorvaldsen and in 1972 on a Christmas seal bearing portraits of
the queen and the prince consort.
Piet Hein's Superellipse 69
6.6. .... ..... .. '-' ... J . I . . . I . . I . l l J g ~ on two .J.JCI...U..I..:J'.u. stamps
70
1. Nimbi. This is a 12-counter version of Piet Hein's nim-type game.
The counters are locked-in, sliding pegs on a reversible circular
board so that after a game is played by pushing the pegs down and
turning the board over it is set for another game.
2. Anagog. Here we have a spherical cousin ofPiet Hein's Soma cube. Six
pieces of joined unit spheres are to be formed into a 20-sphere tetra-
hedron or two 10-sphere tetrahedrons or other solid and flat figures.
3. Crux. A solid cross of six projecting arms is so designed that each arm
rotates separately. One of several problems is to bring three spots of
different colors together at each intersection.
4. Twitchit. A dodecahedron has rotating faces and the problem is to
turn them until three different symbols are together at each corner.
5. Bloxbox. W. W. Rouse Ball, discussing the standard 14-15 sliding-
block puzzle in his Mathematical Recreations and Essays, wrote in
1892: "We can conceive also of a similar cubical puzzle, but we could
not work it practically except by sections." Eighty-one years later,
Piet Hein found an ingenious practical solution. Seven identical unit
cubes are inside a transparent plastic order-2 cube. When the cube is
tilted properly, gravity slides a cube (with a pleasant click) into the
hole. Each cube has three black and three white sides. Problems in-
clude forming an order-2 cube (minus one corner) with all sides one
color, or all sides checkered, or all striped, and so on.
Does the parity principle involved in flat versions apply to the
three-dimensional version? And what are the minimum required
moves to get from one pattern to another? Bloxbox opens a Pandora's
box of questions.
Scantion International, a Danish management and consulting com-
pany, adopted the superegg as its logo. In 1982 it moved its world head-
quarters to Princeton, NJ, where Scantion-Princeton, as it is called, built
a luxurious hotel and conference center hidden within the 25 acres of
Princeton's Forrestal Center. An enormous stone superegg stands on the
plaza in front of the hotel. The Schweppes Building, in Stamford, CT, just
south of Exit 25 on the Merritt Parkway, has the shape of a superellipse.
Hermann Zapf designed a typeface whose "bowls" are based on the
superellipse. He called it "Melior" because the curve meliorates be-
tween an ellipse and a rectangle. You'll find a picture of the upper- and
lower-case letters on page 284 of Douglas Hofstadter's Metamagical
Themas (Basic Books, 1985), with comments on page 291.
Piet Rein's SuperelJipse 71
6.1. "'-7J. .... u.J:\.L>J cut a u ....... 'v ................ t"' .. "'J.
J.
31-37.
55-66.
72
Chapter 7 Penrose Tiles
At the end of a 1975 Scientific American column on tiling the
plane periodically with congruent convex polygons (reprinted in my
Time Travel and Other Mathematical Bewilderments) I promised a later
column on nonperiodic tiling. This chapter reprints my fulfillment of
that promise-a 1977 column that reported for the first time a remark-
able nonperiodic tiling discovered by Roger Penrose, the noted British
mathematical physicist and cosmologist. First, let me give some defin-
itions and background.
A periodic tiling is one on which you can outline a region that tiles
the plane by translation, that is, by shifting the position of the region
without rotating or reflecting it. M. C. Escher, the Dutch artist, was fa-
mous for his many pictures of periodic tilings with shapes that resem-
ble living things. Figure 7.1 is typical. An adjacent black and white
bird constitute a fundamental region that tiles by translation. Think of
the plane as being covered with transparent paper on which each tile
is outlined. Only if the tiling is periodic can you shift the paper, with-
out rotation, to a new position where all outlines again exactly fit.
An infinity of shapes-for instance the regular hexagon-tile only pe-
riodically. An infinity of other shapes tile both periodically and non-
periodically. A checkerboard is easily converted to a nonperiodic tiling
by identical isosceles right triangles or by quadrilaterals. Simply bisect
each square as shown in Figure 7.2A, left, altering the orientations to
prevent periodicity. It is also easy to tile nonperiodically with domi-
noes.
Isoceles triangles also tile in the radial fashion shown in the center
of Figure 7.2(A). Although the tiling is highly ordered, it is obviously
not periodic. As Michael Goldberg pointed out in a 1955 paper titled
"Central Tessellations," such a tiling can be sliced in half, and then the
73
7.1. A pelLlOlalC .. "''''''"''' .... QUlV.U.
A
--
-- "- "-
- ./ -
- ...,....
/
/
B
at
74
tiles spirally. Figure 7.3 shows a striking pattern obtained in this way
from a nine-sided polygon. It was first found by Heinz Voderberg in a
complicated procedure. Goldberg's method of obtaining it makes it al-
most trivial.
Figure 1.3. A spiral tiling by Heinz Voderberg
In all known cases of nonperiodic tiling by congruent figures the fig-
ure also tiles periodically. Figure 7.2(B}, right, shows how two of the
Voderberg enneagons go together to make an octagon that tiles period-
ically in an obvious way.
Another kind of nonperiodic tiling is obtained by tiles that group to-
gether to form larger replicas of themselves. Solomon W. Golomb calls
them "reptiles." Figure 7.4 shows how a shape called the H sphinx"
tiles nonperiodically by giving rise to ever larger sphinxes. Again, two
sphinxes (with one sphinx rotated 180 degrees) tile periodically in an
obvious way.
Are there sets of tiles that tile only nonperiodically? By "only" we
mean that neither a single shape or subset nor the entire set tiles peri-
odically but that by using all of them a nonperiodic tiling is possible.
Rotating and reflecting tiles are allowed.
For many decades experts believed no such set exists, but the sup-
Penrose Tiles 75
are
six a nOllpe:nodic
77
A
Ace
78
H
Kite
Short bow tie
C
T
T
T
Dart
B
tie
and dart
should make at least 100 kites and 60 darts. The pieces need be colored
on one side only. The number of pieces of the two shapes are (like their
areas) in the golden ratio. You might suppose you need more of the
smaller darts, but it is the other way around. You need 1.618 ... as
many kites as darts. In an infinite tiling this proportion is exact. The ir-
rationality of the ratio underlies a proof by Penrose that the tiling is
nonperiodic because if it were periodic, the ratio clearly would have to
be rational.
A good plan is to draw as many darts and kites as you can on one
sheet, with a ratio of about five kites to three darts, using a thin line for
the curves. The sheet can be photocopied many times. The curves can
then be colored, say, red and green. Conway has found that it speeds
constructions and keeps patterns stabler if you make many copies of the
three larger shapes as is shown in Figure 7.6(C). As you expand a pat-
tern, you can continually replace darts and kites with aces and bow
ties. Actually an infinity of arbitrarily large pairs of shapes, made up of
darts and kites, will serve for tiling any infinite pattern.
A Penrose pattern is made by starting with darts and kites around one
vertex and then expanding radially. Each time you add a piece to an
edge, you must choose between a dart and a kite. Sometimes the choice
is forced, sometimes it is not. Sometimes either piece fits, but later you
may encounter a contradiction (a spot where no piece can be legally
added) and be forced to go back and make the other choice. It is a good
plan to go around a boundary, placing all the forced pieces first. They
cannot lead to a contradiction. You can then experiment with unforced
pieces. It is always possible to continue forever. The more you play
with the pieces, the more you will become aware of "forcing rules"
that increase efficiency. For example, a dart forces two kites in its con-
cavity, creating the ubiquitous ace.
There are many ways to prove that the number of Penrose tilings is
uncountable, just as the number of points on a line is. These proofs rest
on a surprising phenomenon discovered by Penrose. Conway calls it
"inflation" and "deflation." Figure 7.7 shows the beginning of inflation.
Imagine that every dart is cut in half and then all short edges of the orig-
inal pieces are glued together. The result: a new tiling (shown in heavy
black lines) by larger darts and kites.
Inflation can be continued to infinity, with each new "generation" of
pieces larger than the last. Note that the second-generation kite, al-
though it is the same size and shape as a first-generation ace, is formed
Penrose Tiles 79
1.1.
80
Not all possible sequences of the four symbols can be produced this
way, but those that label different patterns can be shown to correspond
in number with the number of points on a line.
We have omitted the colored curves on our pictures of tilings be-
cause they make it difficult to see the tiles. If you work with colored
tiles, however, you will be struck by the beautiful designs created by
these curves. Penrose and Conway independently proved that when-
ever a curve closes, it has a pentagonal symmetry, and the entire region
within the curve has a fivefold symmetry. At the most a pattern can
have two curves of each color that do not close. In most patterns all
curves close.
Although it is possible to construct Penrose patterns with a high de-
gree of symmetry (an infinity of patterns have bilateral symmetry), most
patterns, like the universe, are a mystifying mixture of order and un-
expected deviations from order. As the patterns expand, they seem to
be always striving to repeat themselves but never quite managing it. G.
K. Chesterton once suggested that an extraterrestrial being, observing
how many features of a human body are duplicated on the left and the
right, would reasonably deduce that we have a heart on each side. The
world, he said, "looks just a little more mathematical and regular than
it is; its exactitude is obvious, but its inexactitude is hidden; its wild-
ness lies in wait." Everywhere there is a "silent swerving from accuracy
by an inch that is the uncanny element in everything . . . a sort of se-
cret treason in the universe." The passage is a nice description of Pen-
rose's planar worlds.
There is something even more surprising about Penrose universes. In
a curious finite sense, given by the "local isomorphism theorem," all
Penrose patterns are alike. Penrose was able to show that every finite
region in any pattern is contained somewhere inside every other pat-
tern. Moreover, it appears infinitely many times in every pattern.
To understand how crazy this situation is, imagine you are living on
an infinite plane tessellated by one tiling of the uncountable infinity of
Penrose tilings. You can examine your pattern, piece by piece, in ever
expanding areas. No matter how much of it you explore you can never
determine which tiling you are on. It is no help to travel far out and ex-
amine disconnected regions, because all the regions belong to one large
finite region that is exactly duplicated infinitely many times on all pat-
terns. Of course, this is trivially true of any periodic tessellation, but
Penrose Tiles 81
Penrose universes are not periodic. They differ from one another in in-
finitely many ways, and yet it is only at the unobtainable limit that one
can be distinguished from another.
Suppose you have explored a circular region of diameter d. Call it the
"town" where you live. Suddenly you are transported to a randomly
chosen parallel Penrose world. How far are you from a circular region
that exactly matches the streets of your home town? Conway answers
with a truly remarkable theorem. The distance from the perimeter of the
home town to the perimeter of the duplicate town is never more than
d times half of the cube of the golden ratio, or 2.11 + times d. (This is
an upper bound, not an average.) If you walk in the right direction,
you need not go more than that distance to find yourself inside an exact
copy of your home town. The theorem also applies to the universe in
which you live. Every large circular pattern (there is an infinity of dif-
ferent ones) can be reached by walking a distance in some direction
that is certainly less than about twice the diameter of the pattern and
more likely about the same distance as the diameter.
The theorem is quite unexpected. Consider an analogous isomor-
phism exhibited by a sequence of unpatterned digits such as pi. If you
pick a finite sequence of 10 digits and then start from a random spot in
pi, you are pretty sure to encounter the same sequence if you move far
enough along pi, but the distance you must go has no known upper
bound, and the expected distance is enormously longer than 10 digits.
The longer the finite sequence is, the farther you can expect to walk to
find it again. On a Penrose pattern you are always very close to a du-
plicate of home.
There are just seven ways that darts and kites will fit around a ver-
tex. Let us consider first, using Conway's nomenclature, the two ways
with pentagonal symmetry.
The sun (shown in white in Figure 7.8) does not force the placing of
any other piece around it. If you add pieces so that pentagonal sym-
metry is always preserved, however, you will be forced to construct the
beautiful pattern shown. It is uniquely determined to infinity.
The star, shown in white in Figure 7.9, forces the 10 light gray kites
around it. Enlarge this pattern, always preserving the fivefold symmetry,
and you will create another flowery design that is infinite and unique.
The star and sun patterns are the only Penrose universes with perfect
pentagonal symmetry, and there is a lovely sense in which they are
equivalent. Inflate or deflate either of the patterns and you get the other.
82 PLANE GEOMETRY
1.8.
1.9.
It no more
ure
83
Deuce Jack Queen
1.10. queen
84
7.11.
85
...... , ... , ................ '"' ...... to
Buzzsaw Starfish Asterlx
1.12.
86
87
areas.
1.14. A n011pe:flOCllc
88
about kites and darts can be translated into a theorem about the Penrose
rhombuses or any other pair of Penrose tiles and vice versa. Conway
prefers to work with darts and kites, but other mathematicians prefer
working with the simpler rhombuses. Robert Ammann has found a be-
wildering variety of other sets of nonperiodic tiles. One set, consisting
of two convex pentagons and a convex hexagon, forces non periodicity
without any edge markings. He found several pairs, each a hexagon
with five interior angles of 90 degrees and one of 270 degrees.
Figure 1.1 s. The Pythagorean E
pentagram
A ~ - - - - - - ~ ~ - - - - ~ - - - - - - - - ~
Are there pairs of tiles not related to the golden ratio that force non-
periodicity? Is there a pair of similar tiles that force non periodicity? Is
there a pair of convex tiles that will force nonperiodicity without edge
markings?
Of course, the major unsolved problem is whether there is a single
shape that will tile the plane only nonperiodically. Most experts think
not, but no one is anywhere near proving it. It has not even been shown
that if such a tile exists, it must be nonconvex.
Addendum
For much more on Penrose tiles, see Chapter 2 in my Penrose
Tiles to Trapdoor Ciphers (Mathematical Association of America 1989;
paperback, 1997) in which I discuss solid forms of the tiles and their as-
tonishing application to what are called quasicrystals. Until Penrose's
discovery, crystals based on fivefold symmetry were believed to be im-
Penrose Tiles 89
possible to construct. My bibliography gives only a small sampling of
books and papers about quasicrystals that have been published in re-
cent years.
In 1993 John Horton Conway made a significant breakthrough when
he discovered a convex solid called a biprism that tiles space only ape-
riodically. (Aperiodic has replaced nonperiodic as a term for a tile that
tiles only in a nonperiodic way.) A few years earlier Peter Schmitt, at
the University of Vienna, found a nonconvex solid that fills space ape-
riodically, though in a trivial fashion. Conway's subtler solid is de-
scribed and pictured in Keith Devlin's The Language of Mathematics
(W. H. Freeman, 1998, pp. 219-20; paperback, 2000). Doris Schatt-
schneider kindly supplied me with the pattern shown in Figure 7.16 for
constructing the Conway solid.
To assemble, score on all interior
lines, then cut around the outline of
the pattern. Tabs labeled u are to be
folded up, those with d are to be
folded down. Two prisms are then
assembled, one on each side of the
common rhombus face.
Figure 1.16. Conway's biprism. The central rhombus (which is inside the model when as-
sembled) has sides of length 2 and short diagonal length Yz. The small angle of the
rhombus is arcos(3/4) "" 41.4 degrees. Two triangular prisms are built on this common
rhombus face. The diagonal of the prism parallelogram face has length Y(33/4) = 2.87.
When assembled, the vertices of the rhombus that is a common face of the two prisms
are the poles of 2 twofold rotation raxes.
As Devlin explains, Conway's biprism fills space in layers. Every
layer is periodic, but each adjacent layer must be rotated by an irra-
tional angle that forces aperiodicity. Unfortunately neither of the two
aperiodic solids leads to the construction of a flat tile that covers the
plane aperiodically. Finding such a tile or proving it nonexistent is the
top unsolved problem in tiling theory.
In 1997 Penrose sued England's Kimberly-Clark Company for putting
his tiling pattern on their quilted toilet paper without his permission.
90 PLANE GEOMETRY
See Time, May 5, 1997, page 26, for the story and a picture of the
paper's pattern. I don't know the outcome of the lawsuit.
For a long time Penrose tiles were unavailable for purchase. Happily
they are now on the market in a variety of forms that can be obtained
from Kadon Enterprises, 1227 Lorrene Drive, Pasadena, MD 21122.
Bibliography
Nonperiodic tiling
E.P.M. Beenker, "Algebraic Theory of Nonperiodic Tilings of the Plane by Two
Simple Building Blocks: A Square and a Rhombus," Eindhoven University of
Technology, the Netherlands, Department of Mathematics and Computer Sci-
ence, T. H. Report 82-WSK-04, September 1982.
R Berger "The Undecidability of the Domino Problem," Memoirs of the American
Mathematical Society, No. 66, 1966.
N. G. de Bruijn. "Algebraic Theory of Penrose's Nonperiodic THings of the Plane,"
Koninklijke Nederlandse Akademie van Wetenschappen Proceedings Ser. A,
Vol. 84,1981,pp. 39-66.
M. Goldberg, "Central Tessellations," Scripta Mathematica, Vol. 21, 1955, pp.
253-60.
B. Griinbaum and G. C. Shephard, "Some Problems on Plane THings," The Mathe-
matical Gardner, D. A Klamer (ed.), Prindle, Weber & Schmidt, 1981.
B. Griinbaum and G. C. Shephard, Tilings and Patterns, W. H. Freeman, 1986.
R K. Guy, "The Penrose Pieces," Bulletin of the London Mathematical Society,
Vol. 8, 1976, pp. 9-10.
I. Hargittai, "Quasicrystal Structure in Bad Ragaz," The Mathematical Intelligencer,
44,1992,pp.58-59.
M. Naylor, "Nonperiodic Tiling," The Mathematics Teacher, 92, 1999, pp. 34-39.
K. Miyazaki, On Some Periodical and Non-Periodical Honeycombs, A Kobe Uni-
versity monograph, 1977.
R Penrose, "The Role of Aesthetics in Pure and Applied Mathematical Research,"
Bulletin of the Institute of Mathematics and Its Applications, Vol. 10, 1974, pp.
266-71.
R Penrose, "Pentaplexity: A Class of Nonperiodic THings of the Plane," Eureka, 39,
1978, pp. 16-22. Reprinted in The Mathematical Intelligencer, Vol. 2, 1979, pp.
32-37 and in Geometrical Combinatorics, F. C. Holroyd and R J. Wilson, (eds.),
Pitman, 1984.
R Penrose, "Remarks on Tiling," The Mathematics of Long-Range Aperiodic Order,
RV. Moody (ed.), Kluwer, 1997, pp. 467-97.
I. Peterson, "From Here to Infinity," Islands of Truth, W. H. Freeman, 1990, pp.
86-95.
I. Peterson, "Clusters and Decagons," Science News, 150, 1996, pp. 232-33.
Penrose Tiles 91
C. Radin, Miles of Tiles, American Mathematical Society, 1999.
R. M. Robinson, "Undecidability and Nonperiodicity for THings of the Plane," In-
ventiones Mathematicae, Vol. 12, 1971, pp. 177-209.
T. Sibley and S. Wagon, "Rhombic Penrose Tiling Can Be 3-Colored," American
Mathematical Monthly, vol. 107, March 2000, pp. 251-53.
S. Wagon, "Penrose Tiles," Mathematica in Action, W. H. Freeman, 1993, pp.
108-17.
H. Wang, "Games, Logic and Computers," Scientific American, November 1965, pp.
98-106.
,Quasierystals
M. W. Browne, "Puzzling Crystals Plunge Scientists into Uncertainty," The New
York Times, July 30, 1985, p. Cl ff.
M. W. Browne, "Potential Seen in New Quasicrystals," The New York Times, Sep-
tember 5, 1989.
M. W. Browne, " 'Impossible' Form of Matter Takes Spotlight in Study of Solids,"
The New York Times, September 5, 1989.
M. W. Browne, "New Data Help Explain Crystals That Defy Nature," The New York
Times, November 24, 1998.
L. Danzer, "Three-Dimensional Analogs of the Planar Penrose Tiling and Qua-
sicrystals," Discrete Mathematics, 76,1989, pp. 1-7.
N. G. de Bruijn, "A Riffle Shuffle Card Trick and Its Relation to Quasicrystal The-
ory," Nieuw Archiefvoor Wiskunde, Vol. 5, November 1987.
A. Goldman et al., "Quasicrystalline Materials," American Scientist, 8, 1996, pp.
230-41.
I. Hargittai (ed.), Fivefold Symmetry, World Scientific, 1991.
J. Horgan, "Quasicrystals: Rules of the Game," Science, 247, 1990, pp. 1020-22.
C. Jano, Quasicrystals: A Primer, Oxford Science, 1994.
M. V. Jaric (ed.), Introduction to Quasicrystals, Vol. 1, Aperiodicityand Order, Aca-
demic Press, 1988 and Vol. 2, Introduction to the Mathematics of Quasicrystals,
Academic Press, 1989; M. V. Jaric and D. Gratias (eds.), Vol. 3, Extended Icosa-
hedral Structures, Academic Press, 1990.
M. La Brecque, "Opening the Door to Forbidden Symmetries," Mosaic, Vol. 18,
1987-1988,pp.2-23.
D. Levine, "Quasicrystals: A New Class of Ordered Structure," Ph.D. thesis, Uni-
versity of Pennsylvania, 1986.
R. Mosseri and J. F. Sadoc, "Pauling's Model Not Universally Accepted," Nature,
Vol. 319,1986, p. 104. Reply to Pauling's 1985 paper.
D. R. Nelson, "Quasicrystals," SCientifiC American, August 1986, pp. 42-51.
D. R. Nelson and B. L Halperin, "Pentagonal and Icosahedral Order in Rapidly
Cooled Metals," Science, Vol. 229,1985, pp. 235-38.
J. Patera, Quasicrystals and Discrete Geometry, American Mathematical Society,
1998.
L. Pauling, "Apparent Icosahedral Symmetry is Due to Directed Multiple Twinning
of Cubic Crystals," Nature, VoL 317. 1985. p. 512.
92 PLANE GEOMETRY
L. Pauling, "Icosahedral Symmetry," Science, Vol. 239, 1988, pp. 963-64. This let-
ter from Linus Pauling continues his attack on quasicrystals, with a reply by Paul
Steinhardt.
1. Peterson, "The Fivefold Way for Crystals," Science News, Vol. 127, 1985, pp.
188-89. Reprinted in Peterson's collection The Mathematical Tourist, W. H.
Freeman, 1988, pp. 200-212.
1. Peterson, "Tiling to Infinity," Science News, Vol. 134, July 16, 1988, p. 42.
I. Peterson, "Shadows and Symmetries," Science News, 140, 1991, pp. 408-19.
M. Senechal, Quasicrystals and Geometry, Cambridge University Press, 1995.
M. Senechal and J. Taylor, "Quasicrystal: The View From Les Houches," Mathe-
maticallntelligencer, 12, 1990, pp. 54-63.
P. J. Steinhardt, "Quasicrystals," American Scientist, November/December 1986,
pp. 586-597. Its bibliography lists more than 40 technical papers.
P. Steinhardt, "Crazy Crystals," New Scientist, January 25,1997, pp. 32-35.
P. Steinhardt and S. Ostlund (eds.), The Physics of Quasicrystals, World Scientific,
1987.
P. W. Stepheus and A. I. Goldman, "The Structure of Quasicrystals," Scientific
American, April 1991, pp. 44ff.
D. P. Vincenzo and P. J. Steinhardt (eds.), Quasicrystals: The State of the Art, World
Scientific, 1991.
H. C. Von Baeyer, "Impossible Crystals," Discover, February 1990, pp. 69-78.
D. Voss, "Smash the System," New Scientist, 27, 1999, pp. 42-46.
Penrose Tiles 93
Chapter 8
Planiversal scientists are not a very
common breed.
-Alexander Keewatin Dewdney
The Wonders
of a Planiverse
Jls far as anyone knows the only existing universe is the one
we live in, with its three dimensions of space and one of time. It is not
hard to imagine, as many science-fiction writers have, that intelligent
organisms could live in a four-dimensional space, but two dimensions
offer such limited degrees of freedom that it has long been assumed in-
telligent 2-space life forms could not exist. Two notable attempts have
nonetheless been made to describe such organisms.
In 1884 Edwin Abbott Abbott, a London clergyman, published his
satirical novel Flatland. Unfortunately the book leaves the reader al-
most entirely in the dark about Flatland's physical laws and the tech-
nology developed by its inhabitants, but the situation was greatly
improved in 1907 when Charles Howard Hinton published An Episode
of Flatland. Although written in a flat style and with cardboard char-
acters, Hinton's story provided the first glimpses of the possible science
and technology of the two-dimensional world. His eccentric book is,
alas, long out of print, but you can read about it in the chapter "Flat-
lands" in my book The Unexpected Hanging and Other Mathematical
Diversions (Simon & Schuster, 1969).
In "Flatlands" I wrote: "It is amusing to speculate on two-
dimensional physics and the kinds of simple mechanical devices that
would be feasible in a flat world." This remark caught the attention of
Alexander Keewatin Dewdney, a computer scientist at the University
of Western Ontario. Some of his early speculations on the subject were
set down in 1978 in a university report and in 1979 in "Exploring the
Planiverse," an article in Journal of Recreational Mathematics (Vol. 12,
No.1, pp. 16-20; September). Later in 1979 Dewdney also privately
published "Two-dimensional Science and Technology," a 97-page tour
de force. It is hard to believe, but Dewdney actually lays the ground-
94
work for what he calls a planiverse: a possible two-dimensional world.
Complete with its own laws of chemistry, physics, astronomy, and bi-
ology, the planiverse is closely analogous to our own universe (which
he calls the steriverse) and is apparently free of contradictions. I should
add that this remarkable achievement is an amusing hobby for a math-
ematician whose serious contributions have appeared in some 30 pa-
pers in technical journals.
Dewdney's planiverse resembles Hinton's in having an earth that he
calls (as Hinton did) Astria. Astria is a dislike planet that rotates in pla-
nar space. The Astrians, walking upright on the rim of the planet, can
distinguish east and west and up and down. Naturally there is no north
or south. The "axis" of Astria is a point at the center of the circular
planet. You can think of such a flat planet as being truly two-
dimensional or you can give it a very slight thickness and imagine it as
moving between two frictionless planes.
As in our world, gravity in a planiverse is a force between objects that
varies directly with the product of their masses, but it varies inversely
with the linear distance between them, not with the square of that dis-
tance. On the assumption that forces such as light and gravity in a
planiverse move in straight lines, it is easy to see that the intensity of
such forces must vary inversely with linear distance. The familiar text-
book figure demonstrating that in our world the intensity of light varies
inversely with the square of distance is shown at the top of Figure 8.1.
The obvious planar analogue is shown at the bottom of the illustra-
tion.
To keep his whimsical project from" degenerating into idle specula-
tion" Dewdney adopts two basic principles. The "principle of similarity"
states that the planiverse must be as much like the steriverse as possi-
ble: a motion not influenced by outside forces follows a straight line,
the flat analogue of a sphere is a circle, and so on. The "principle of
modification" states that in those cases where one is forced to choose
between conflicting hypotheses, each one equally similar to a steriver-
sal theory, the more fundamental one must be chosen and the other
must be modified. To determine which hypothesis is more fundamen-
tal Dewdney relies on the hierarchy in which physics is more funda-
mental than chemistry, chemistry is more fundamental than biology,
and so on.
To illustrate the interplay between levels of theory Dewdney con-
siders the evolution of the planiversal hoist in Figure 8.2. The engineer
The Wonders of a Planiverse 95
FigureS.I.
ARC
LIGHT
NINE
SQUARE
INCHES
I ').
I I ..... "
I I '
I
1,.1 I I
I 1'",,,, l
I I I "''I
I.. : 1 I
I '-I.. I I
I 1"'..1. 1
I I r"' ....
I I I
.... '" I I
"' .... I

tEt=--------THREE FEET---------:.t

THREE
INCHES
who designed it first gave it arms thinner than those in the illustration,
but when a metallurgist pointed out that planar materials fracture more
easily than their 3-space counterparts, the engineer made the arms
thicker. Later a theoretical chemist, invoking the principles of similar-
ity and modification at a deeper level, calculated that the planiversal
molecular forces are much stronger than had been suspected, and so
the engineer went back to thinner arms.
The principle of similarity leads Dewdney to posit that the planiverse
Figure 8.2.
96 PLANE GEOMETRY
is a three-dimensional continuum of space-time containing matter
composed of molecules, atoms, and fundamental particles. Energy is
propagated by waves, and it is quantized. Light exists in all its wave-
lengths and is refracted by planar lenses, making possible planiversal
eyes, planiversal telescopes, and planiversal microscopes. The plani-
verse shares with the steriverse such basic precepts as causality; the
first and second laws of thermodynamics; and laws concerning inertia,
work, friction, magnetism, and elasticity.
Dewdney assumes that his planiverse began with a big bang and is
currently expanding. An elementary calculation based on the inverse-
linear gravity law-shows that regardless of the amount of mass in the
planiverse the expansion must eventually halt, so that a contracting
phase will begin. The Astrian night sky will of course be a semicircle
along which are scattered twinkling points of light. If the stars have
proper motions, they will continually be occulting one another. If As-
tria has a sister planet, it will over a period of time occult every star in
the sky.
We can assume that Astria revolves around a sun and rotates, thereby
creating day and night. In a planiverse, Dewdney discovered, the only
stable orbit that continually retraces the same path is a perfect circle.
Other stable orbits roughly elliptical in shape are possible, but the axis
of the ellipse rotates in such a way that the orbit never exactly closes.
Whether planiversal gravity would allow a moon to have a stable orbit
around Astria remains to be determined. The difficulty is due to the
sun's gravity, and resolving the question calls for work on the planar
analogue of what our astronomers know as the three-body problem.
Dewdney analyzes in detail the nature of Astrian weather, using
analogies to our seasons, winds, clouds, and rain. An Astrian river
would be indistinguishable from a lake except that it might have faster
currents. One peculiar feature of Astrian geology is that water cannot
flow around a rock as it does on the earth. As a result rainwater steadily
accumulates behind any rock on a slope, tending to push the rock
downhill: the gentler the slope is, the more water accumulates and the
stronger the push is. Dewdney concludes that given periodic rainfall
the Astrian surface would be unusually flat and uniform. Another con-
sequence of the inability of water to move sideways on Astria is that it
would become trapped in pockets within the soil, tending to create
large areas of treacherous quicksand in the hollows of the planet. One
hopes, Dewdney writes, that rainfall is infrequent on Astria. Wind too
The Wonders of a Planiverse 97
would have much severer effects on Astria than on the earth because
like rain it cannot "go around" objects.
Dewdney devotes many pages to constructing a plausible chemistry
for his planiverse, modeling it as much as possible on three-
dimensional matter and the laws of quantum mechanics. Figure 8.3
shows Dewdney's periodic table for the first 16 planiversal elements.
Because the first two are so much like their counterparts in our world,
they are called hydrogen and helium. The next 10 have composite
names to suggest the steriversal elements they most resemble; for ex-
ample, lithrogen combines the properties of lithium and nitrogen. The
next four are named after Hinton, Abbott, and the young lovers in Hin-
ton's novel, Harold Wall and Laura Cartwright.
ATOMIC NAME SYMBOL SHELL STRUCTURE VALENCE
NUMBER ls 2s 2p 3s 3p 3d 4s 4p ...
1 HYDROGEN H 1 1
2 HELIUM He 2 2
3 LITROGEN Lt 2 1 1
4 BEROXYGEN Bx 2 2 2
5 FLUORON FI 2 2 1 3
6 NEOCARBON Nc 2 2 2 4
7 SODAUNUM Sa 2 2 2 1 1
8 MAGNILICON Me 2 2 2 2 2
9 ALUPHORUS Ap 2 2 2 2 1 3
10 SULFICON Sp 2 2 2 2 2 4
11 CHLOPHORUS Cp 2 2 2 2 2 1 5
12
ARGOFUR At 2 2 2 2 2 2 B
13 HINTONIUM Hn 2 2 2 2 2 2 1 1
14
ABBOOEN Ab 2 2 2 2 2 2 2 2
15 HAROLDIUM Wa 2 2 2 2 2 2 2 1 3
16 LAURANIUM La 2 2 2 2 2 2 2 2 4
FigureS.3.
In the flat world atoms combine naturally to form molecules, but of
course only bonding that can be diagrammed by a planar graph is al-
lowed. (This result follows by analogy from the fact that intersecting
bonds do not exist in steriversal chemistry.) As in our world, two
asymmetric molecules can be mirror images of each other, so that nei-
ther one can be "turned over" to become identical with the other. There
are striking parallels between planiversal chemistry and the behavior
of steriversal monolayers on crystal surfaces (see J. G. Dash, "Two-
dimensional Matter," Scientific American, May 1973). In our world
molecules can form 230 distinct crystallographic groups, but in the
98 PLANE GEOMETRY
pI ani verse they can form only 17. I am obliged to pass over Dewdney's
speculations about the diffusion of molecules, electrical and magnetic
laws, analogues of Maxwell's equations, and other subjects too techni-
cal to summarize here.
Dewdney assumes that animals on Astria are composed of cells that
cluster to form bones, muscles, and connective tissues similar to those
found in steriversal biology. He has little difficulty showing how these
bones and muscles can be structured to move appendages in such a
way that the animals can crawl, walk, fly, and swim. Indeed, some of
these movements are easier in a planiverse than in our world. For ex-
ample, a steriversal animal with two legs has considerable difficulty
balancing while walking, whereas in the planiverse if an animal has
both legs on the ground, there is no way it can fall over. Moreover, a fly-
ing planiversal animal cannot have wings and does not need them to
fly; if the body of the animal is aerodynamically shaped, it can act as a
wing (since air can go around it only in the plane). The flying animal
could be propelled by a flapping tail.
Calculations also show that Astrian animals probably have much
lower metabolic rates than terrestrial animals because relatively little
heat is lost through the perimeter of their body. Furthermore, animal
bones can be thinner on Astria than they are on the earth, because they
have less weight to support. Of course, no Astrian animal can have an
open tube extending from its mouth to its anus, because if it did, it
would be cut in two.
In the appendix to his book The Structure and Evolution of the Uni-
verse (Harper, 1959) G. J. Whitrow argues that intelligence could not
evolve in 2-space because of the severe restrictions two dimensions
impose on nerve connections. "In three or more dimensions," he
writes, "any number of [nerve] cells can be connected with [one an-
other] in pairs without intersection of the joins, but in two dimensions
the maximum number of cells for which this is possible is only four."
Dewdney easily demolishes this argument, pointing out that if nerve
cells are allowed to fire nerve impulses through "crossover points,"
they can form flat networks as complex as any in the steriverse.
Planiversal minds would operate more slowly than steriversal ones,
however, because in the two-dimensional networks the pulses would
encounter more interruptions. (There are comparable results in the the-
ory of two-dimensional automatons.)
Dewdney sketches in detail the anatomy of an Astrian female fish
The Wonders of a Planiverse 99
100
life-giving plants and animals, it is clear that very little of the Astrian
surface can be disturbed without inviting the biological destruction of
the planet. For example, here on earth we may build a modest highway
through the middle of several acres of rich farmland and destroy no
more than a small percentage of it. A corresponding highway on Astria
will destroy all the 'acreage' it passes over .... Similarly, extensive
cities would quickly use up the Astrian countryside. It would seem
that the only alternative for the Astrian technological society is to go
underground." A typical subterranean house with a living room, two
bedrooms, and a storage room is shown in Figure 8.5. Collapsible chairs
and tables are stored in recesses in the floors to make the rooms easier
to walk through.
BEDROOMS
Figure B.S.
The many simple three-dimensional mechanical elements that have
obvious analogues on Astria include rods, levers, inclined planes,
springs, hinges, ropes, and cables (see Figure 8.6, top). Wheels can be
rolled along the ground, but there is no way to turn them on a fixed
axle. Screws are impossible. Ropes cannot be knotted; but by the same
token, they never tangle. Tubes and pipes must have partitions, to keep
their sides in place, and the partitions have to be opened (but never all
of them at once) to allow anything to pass through. It is remarkable that
in spite of these severe constraints many flat mechanical devices can be
built that will work. A faucet designed by Dewdney is shown in Figure
8.6, bottom. To operate it the handle is lifted. This action pulls the
valve away from the wall of the spout, allowing the water to flow out.
When the handle is released, the spring pushes the valve back.
The device shown in Figure 8.7 serves to open and close a door (or
a wall). Pulling down the lever at the right forces the wedge at the bot-
tom to the left, thereby allowing the door to swing upward (carrying the
wedge and the levers with it) on a hinge at the top. The door is opened
The Wonders of a Planiverse 101
HINGE 2 J ~ ~ - - ,
SPRING
-'VVVVVv-
ROD
PLANE
I ~
~
Figure 6 (top)
f'cgure 8.6.
from the left by pushing up on the other lever. The door can be lowered
from either side and the wedge can be moved back to stabilize the wall
by moving a lever in the appropriate direction. This device and the
faucet are both mechanisms with permanent planiversal hinges: circu-
lar knobs that rotate inside hollows but cannot be removed from them.
Figure B.B depicts a planiversal steam engine whose operation par-
allels that of a steriversal engine. Steam under pressure is admitted
into the cylinder of the engine through a sliding valve that forms one
of its walls (top). The steam pressure causes a piston to move to the
right until steam can escape into a reservoir chamber above it. The sub-
sequent loss of pressure allows the compound leaf spring at the right
of the cylinder to drive the piston back to the left (bottom). The sliding
valve is closed as the steam escapes into the reservoir, but as the pis-
ton moves back it reopens, pulled to the right by a spring-loaded arm.
Figure B.9 depicts Dewdney's ingenious mechanism for unlocking a
door with a key. This planiversallock consists of three slotted tumblers
(a) that line up when a key is inserted (b) so that their lower halves
move as a unit when the key is pushed (e). The pushing of the key is
transmitted through a lever arm to the master latch, which pushes
down on a slave latch until the door is free to swing to the right (d). The
bar on the lever arm and the lip on the slave latch make the lock diffi-
cult to pick. Simple and compound leaf springs serve to return all the
102 PLANE GEOMETRY
Figure B.7.
parts of the lock except the lever arm to their original positions when
the door is opened and the key is removed. When the door closes, it
strikes the bar on the lever arm, thereby returning that piece to its orig-
inal position as well. This flat lock could actually be employed in the
steriverse; one simply inserts a key without twisting it.
"It is amusing to think," writes Dewdney, "that the rather exotic de-
sign pressures created by the planiversal environment could cause us
to think about mechanisms in such a different way that entirely novel
solutions to old problems arise. The resulting designs, if steriversally
practical, are invariably space-saving."
Thousands of challenging planiversal problems remain unsolved. Is
there a way, Dewdney wonders, to design a two-dimensional windup
motor with flat springs or rubber bands that would store energy? What
The Wonders of a Planiverse 103
8.8.
a
8.9.
104
d
SPRING-
LOADED
ARM
PLANE
1
ments) about what might be observed. "The experimentalist's loss,"
observes Dewdney, "is the theoretician's gain."
A marvelous exhibit could be put on of working models of planiver-
sal machines, cut out of cardboard or sheet metal, and displayed on a
surface that slopes to simulate planiversal gravity. One can also imag-
ine beautiful cardboard exhibits of planiversallandscapes, cities, and
houses. Dewdney has opened up a new game that demands knowledge
of both science and mathematics: the exploration of a vast fantasy
world about which at present almost nothing is known.
It occurs to me that Astrians would be able to play two-dimensional
board games but that such games would be as awkward for them as
three-dimensional board games are for us. I imagine them, then, play-
ing a variety of linear games on the analogue of our 8 x 8 chessboard.
Several games of this type are shown in Figure 8.10. Part (a) shows the
start of a checkers game. Pieces move forward only, one cell at a time,
and jumps are compulsory. The linear game is equivalent to a game of
regular checkers with play confined to the main diagonal of a standard
board. It is easy to see how the second player wins in rational play and
how in misere, or "giveaway," checkers the first player wins just as
easily. Linear checkers games become progressively harder to analyze
as longer boards are introduced. For example, which player wins stan-
dard linear checkers on the ll-cell board when each player starts with
checkers on the first four cells at his end of the board?
2 3 4 5 6 7 8
a
I r=J1r=J 1r=J I
b
a
c
E
d
e
Figure 8. I O.
106 PLANE GEOMETRY
Part (b) in the illustration shows an amusing Astrian analogue of
chess. On a linear board a bishop is meaningless and a queen is the
same as a rook, so the pieces are limited to kings, knights, and rooks.
The only rule modification needed is that a knight moves two cells in
either direction and can jump an intervening piece of either color. If the
game is played rationally, will either White or Black win or will the
game end in a draw? The question is surprisingly tricky to answer.
Linear go, played on the same board, is by no means trivial. The ver-
sion I shall describe, called pinch, was invented in 1970 by James
Marston Henle, a mathematician at Smith College.
In the game of pinch players take turns placing black and white
stones on the cells of the linear board, and whenever the stones of one
player surround the stones of the other, the surrounded stones are re-
moved. For example, both sets of white stones shown in part (c) of Fig-
ure 8.10 are surrounded. Pinch is played according to the following two
rules.
Rule 1: No stone can be placed on a cell where it is surrounded un-
less that move serves to surround a set of enemy stones. Hence in the
situation shown in part (d) of the illustration, White cannot play on
cells 1, 3, or 8, but he can play on cell 6 because this move serves to sur-
round cell 5.
Rule 2: A stone cannot be placed on a cell from which a stone was re-
moved on the last play if the purpose of the move is to surround some-
thing. A player must wait at least one turn before making such a move.
For example, in part (e) of the illustration assume that Black plays on
cell 3 and removes the white stones on cells 4 and 5. White cannot play
on cell 4 (to surround cell 3) for his next move, but he may do so for any
later move. He can play on cell 5, however, because even though a
stone was just removed from that cell, the move does not serve to sur-
round anything. This rule is designed to decrease the number of stale-
mates, as is the similar rule in go.
Two-cell pinch is a trivial win for the second player. The three- and
four-cell games are easy wins for the first player if he takes the center
in the three-cell game and one of the two central cells in the four-cell
one. The five-cell game is won by the second player and the six- and
seven-cell games are won by the first player. The eight-cell game jumps
to such a high level of complexity that it becomes very exciting to play.
Fortunes often change rapidly, and in most situations the winning
player has only one winning move.
The Wonders of a P]aniverse 107
Addendum
My column on the planiverse generated enormous interest.
Dewdney received some thousand letters offering suggestions about
flatland science and technology. In 1979 he privately printed Two-
Dimensional Science and Technology, a monograph discussing these
new results. Two years later he edited another monograph, A Sympo-
sium of Two-Dimensional Science and Technology. It contained papers
by noted scientists, mathematicians, and laymen, grouped under the
categories of physics, chemistry, astronomy, biology, and technology.
Newsweek covered these monographs in a two-page article, "Life in
Two Dimensions" (January 18, 1980), and a similar article, "Scientific
Dreamers' Worldwide Cult," ran in Canada's Maclean's magazine (Jan-
uary 11,1982). Omni (March 1983), in an article on "Flatland Redux,"
included a photograph of Dewdney shaking hands with an Astrian.
In 1984 Dewdney pulled it all together in a marvelous work, half
nonfiction and half fantasy, titled The Planiverse and published by Po-
seidon Press, an imprint of Simon & Schuster. That same year he took
over the mathematics column in Scientific American, shifting its em-
phasis to computer recreations. Several collections of his columns have
been published by W. H. Freeman: The Armchair Universe (1987), The
Turing Omnibus (1989), and The Magic Machine (1990).
An active branch of physics is now devoted to planar phenomena. It
involves research on the properties of surfaces covered by a film one
molecule thick, and a variety of two-dimensional electrostatic and elec-
tronic effects. Exploring possible flatlands also relates to a philosoph-
ical fad called "possible worlds." Extreme proponents of this
movement actually argue that if a universe is logically possible-that
is, free of logical contradictions-it is just as "real" as the universe in
which we flourish.
In Childhood's End Arthur Clarke describes a giant planet where in-
tense gravity has forced life to evolve almost flat forms with a vertical
thickness of one centimeter.
The following letter from J. Richard Gott III, an astrophysicist at
Princeton University, was published in Scientific American (October
1980):
I was interested in Martin Gardner's article on the physics of Flatland,
because for some years I have given the students in my general relativ-
108 PLANE GEOMETRY
ity class the problem of deriving the theory of general relativity for Flat-
land. The results are surprising. One does not obtain the Flatland ana-
logue of Newtonian theory (masses with gravitational fields falling off
like llr) as the weak-field limit. General relativity in Flatland predicts no
gravitational waves and no action at a distance. A planet in Flatland
would produce no gravitational effects beyond its own radius. In our
four-dimensional space-time the energy momentum tensor has 10 in-
dependent components, whereas the Riemann curvature tensor has 20
independent components. Thus it is possible to find solutions to the
vacuum field equations GJ,LV == 0 (where all components of the energy
momentum tensor are zero) that have a nonzero curvature. Black-hole
solutions and the gravitational-field solution external to a planet are ex-
amples. This allows gravitational waves and action at a distance. Flat-
land has a three-dimensional space-time where the energy momentum
tensor has six independent components and the Riemann curvature ten-
sor also has only six independent components. In the vacuum where all
components of the energy momentum tensor are zero all the compo-
nents of the Riemann curvature tensor must also be zero. No action at a
distance or gravity waves are allowed.
Electromagnetism in Flatland, on the other hand, behaves just as one
would expect. The electromagnetic field tensor in four-dimensional
space-time has six independent components that can be expressed as
vector E and B fields with three components each. The electromagnetic
field tensor in a three-dimensional space-time (Flatland) has three in-
dependent components: a vector E field with two components and a
scalar B field. Electromagnetic radiation exists, and charges have electric
fields that fall off like llr.
Two more letters, published in the same issue, follow. John S. Harris,
of Brigham Young University's English Department, wrote:
As I examined Alexander Keewatin Dewdney's planiversal devices in
Martin Gardner's article on science and technology in a two-dimensional
universe, I was struck with the similarity of the mechanisms to the lock-
work of the Mauser military pistol of 1895. This remarkable automatic
pistol (which had many later variants) had no pivot pins or screws in its
functional parts. Its entire operation was through sliding cam surfaces
and two-dimensional sockets (called hinges by Dewdney). Indeed, the
lockwork of a great many firearms, particularly those of the 19th century,
follows essentially planiversal principles. For examples see the cutaway
drawings in Book of Pistols and Revolvers by W.H.B. Smith.
Gardner suggests an exhibit of machines cut from cardboard, and that
The Wonders of a Planiverse 109
is exactly how the firearms genius John Browning worked. He would
sketch the parts of a gun on paper or cardboard, cut out the individual
parts with scissors (he often carried a small pair in his vest pocket), and
then would say to his brother Ed, "Make me a part like this." Ed would
ask, "How thick, John?" John would show a dimension with his thumb
and forefinger, and Ed would measure the distance with calipers and
make the part. The result is that virtually every part of the 100 or so
Browning designs is essentially a two-dimensional shape with an added
thickness.
This planiversality of Browning designs is the reason for the obsoles-
cence of most of them. Dewdney says in his enthusiasm for the plani-
verse that "such devices are invariably space-saving." They are also
expensive to manufacture. The Browning designs had to be manufac-
tured by profiling machines: cam-following vertical milling machines. In
cost of manufacture such designs cannot compete with designs that can
be produced by automatic screw-cutting lathes, by broaching machines,
by stamping, or by investment casting. Thus although the Browning de-
signs have a marvelous aesthetic appeal, and although they function
with delightful smoothness, they have nearly all gone out of produc-
tion. They simply got too expensive to make.
Stefan Drobot, a mathematician at Ohio State University, had this to
say:
In Martin Gardner's article he and the authors he quotes seem to have
overlooked the following aspect of a "planiverse": any communication
by means of a wave process, acoustic or electromagnetic, would in such
a universe be impossible. This is a consequence of the Huygens princi-
ple, which expresses a mathematical property of the (fundamental) so-
lutions of the wave equation. More specifically, a sharp impulse-type
signal (represented by a "delta function") originating from some point is
propagated in a space of three spatial dimensions in a manner essentially
different from that in which it is propagated in a space of two spatial di-
mensions. In three-dimensional space the signal is propagated as a
sharp-edged spherical wave without any trail. This property makes it
possible to communicate by a wave process because two signals follow-
ing each other in a short time can be distinguished.
In a space with two spatial dimensions, on the other hand, the funda-
mental solution of the wave equation represents a wave that, although it
too has a sharp edge, has a trail of theoretically infinite length. An ob-
server at a fixed distance from the source of the signal would perceive the
oncoming front (sound, light, etc.) and then would keep perceiving it, al-
110 PLANE GEOMETRY
though the intensity would decrease in time. This fact would make com-
munication by any wave process impossible because it would not allow
two signals following each other to be distinguished. More practically
such communication would take much more time. This letter could not
be read in the planiverse, although it is (almost) two-dimensional.
My linear checkers and chess prompted many interesting letters. Abe
Schwartz assured me that on the ll-cell checker field Black also wins if
the game is give-away. I. Richard Lapidus suggested modifying linear
chess by interchanging knight and rook (the game is a draw), by adding
more cells, by adding pawns that capture by moving forward one space,
or by combinations of the three modifications. If the board is long
enough, he suggested duplicating the pieces-two knights, two rooks-
and adding several pawns, allowing a pawn a two-cell start option as in
standard chess. Peter Stampolis proposed sustituting for the knight two
pieces called "kops" because they combine features of knight and bishop
moves. One kop moves only on white cells, the other moves on black.
Of course many other board games lend themselves to linear forms,
for example, Reversi (also called Othello), or John Conway's Phutball,
described in the two-volume Winning Ways written by Elwyn
Berlekamp, Richard Guy, and John Conway.
Burr puzzles are wooden take-apart puzzles often referred to as Chi-
nese puzzles. Dewdney's Scientific American column (October 1985)
describes a clever planar version of a burr puzzle designed by Jeffrey
Carter. It comes apart only after a proper sequence of pushes and pulls.
A graduate student in physics, whose name I failed to record, had a
letter in Science News, December 8,1984, protesting the notion that re-
search in planar physics was important because it led to better under-
standing of three-dimensional physics. On the contrary, he wrote,
planar research is of great interest for its own sake. He cited the quan-
tum Hall effect and the whole field of microelectronics which is based
on two-dimensional research and has many technological applications.
Answers
In ll-celllinear checkers (beginning with Black on cells 1,2,3,
and 4 and White on cells 8, 9, 10, and 11) the first two moves are forced:
Black to 5 and White to 7. To avoid losing, Black then goes to 4, and
White must respond by moving to 8. Black is then forced to 3 and
The Wonders of a Planiverse 111
White to 9. At this point Black loses with a move to 2 but wins with a
move to 6. In the latter case White jumps to 5, and then Black jumps to
6 for an easy end -game victory.
On the eight-cell linear chessboard White can win in at most six
moves. Of White's four opening moves, RxR is an instant stalemate and
the shortest possible game. R-5 is a quick loss for White if Black plays
RxR. Here White must respond with N-4, and then Black mates on his
second move with RxN. This game is one of the two "fool's mates," or
shortest possible wins. The R-4 opening allows Black to mate on his
second or third move if he responds with N-5.
White's only winning opening is N-4. Here Black has three possible
replies:
1. RxN. In this case White wins in two moves with RxR.
2. R-5. White wins with K-2. If Black plays R-6, White mates with NxR.
If Black takes the knight, White takes the rook, Black moves N-5, and
White mates by taking Black's knight.
3. N-5. This move delays Black's defeat the longest. In order to win
White must check with NxR, forcing Black's king to 7. White moves
his rook to 4. If Black plays KxN, White's king goes to 2, Black's K-7
is forced, and White's RxN wins. If Black plays N-3 (check), White
moves the king to 2. Black can move only the knight. If he plays N-
1, White mates with N-8. If Black plays N-5, White's N-8 forces
Black's KxN, and then White mates with RxN.
The first player also has the win in eight-cell pinch (linear go) by
opening on the second cell from an end, a move that also wins the six-
and seven-cell games. Assume that the first player plays on cell 2. His
unique winning responses to his opponent's plays on 3,4,5,6,7, and
8 are respectively 5,7,7,7,5, and 6. I leave the rest of the game to the
reader. It is not known whether there are other winning opening moves.
James Henle, the inventor of pinch, assures me that the second player
wins the nine-cell game. He has not tried to analyze boards with more
than nine cells.
Bibliography
D. Burger, Sphereland: A Fantasy about Curved Spaces and an Expanding Uni-
verse, T. Y. Crowell, 1965.
112 PLANE GEOMETRY
A. K. Dewdney, The Planiverse, Poseidon, 1984; Copernicus, 2000.
A. K. Dewdney, 200 Percent of Nothing: An Eye Opening Tour Through the Twists
and Turns of Math Abuse and Innumeracy, Wiley, 1994.
A. K. Dewdney, Introductory Computer Science: Bits of Theory, Bytes of Practice,
W. H. Freeman, 1996.
A. K. Dewdney, "The Planiverse Project: Then and Now," Mathematical Intelli-
gencer, Vol. 22, No.1, 2000, pp. 46-48.
C. H. Hinton, An Episode of Flatland, Swan Sonnenschein & Co., 1907.
R. Jackiw, "Lower Dimensional Gravity," Nuclear Physics, Vol. B252, 1985, pp.
343-56.
I. Peterson, "Shadows From a Higher Dimension," Science News, Vol. 126, No-
vember 3,1984, pp. 284-85.
A. Square and E. A. Abbott, Flatland: A Romance of Many Dimensions, Dover Pub-
lications, Inc., 1952. Several other paperback editions are currently in print.
I. Stewart, Flatterland, Perseus Publishing, 2001.
P. J. Stewart, "Allegory Through the Computer Class: Sufism in Dewdney's Plani-
verse," Sufi, Issue 9, Spring 1991, pp. 26-30.
The Wonders of a Planiverse 113
lilili
Solid Qeometry
and Higher
Dimensions
Chapter 9
Rosy's instant acceptance of our model at
first amazed me. I had feared that her sharp,
stubborn mind, caught in her self-made an-
tihelical trap, might dig up irrelevant results
that would foster uncertainty about the cor-
rectness of the double helix. Nonetheless,
like almost everyone else, she saw the ap-
peal of the base pairs and accepted the fact
that the structure was too pretty not to be
true. -James D. Watson, The Double Helix
The Helix
A straight sword will fit snugly into a straight scabbard. The
same is true of a sword that curves in the arc of a circle: it can be
plunged smoothly into a scabbard of the same curvature. Mathemati-
cians sometimes describe this property of straight lines and circles by
calling them "self-congruent" curves; any segment of such a curve can
be slid along the curve, from one end to the other, and it will always
"fit. "
Is it possible to design a sword and its scabbard that are not either
straight or curved in a circular arc? Most people, after giving this care-
ful consideration, will answer no, but they are wrong. There is a third
curve that is self-congruent: the circular helix. This is a curve that coils
around a circular cylinder in such a way that it crosses the "elements"
of the cylinder at a constant angle. Figure 9.1 makes this clear. The el-
ements are the vertical lines that parallel the cylinder's axis; A is the
constant angle with which the helix crosses every element. Because of
the constant curvature of the helix a helical sword would screw its
way easily in and out of a helical scabbard.
Actually the straight line and the circle can be regarded as limiting
cases of the circular helix. Compress the curve until the coils are very
close together and you get a tightly wound helix resembling a Slinky
toy; if angle A increases to 90 degrees, the helix collapses into a circle.
On the other hand, if you stretch the helix until angle A becomes zero,
the helix is transformed into a straight line. If parallel rays of light
shine perpendicularly on a wall, a circular helix held before the wall
with its axis parallel to the rays will cast on the wall a shadow that is
117
9.1. .... ...... ''-'''''''''L.LL on
118 AND
ral around cones), including bedsprings and spiral ramps such as the
inverted conical ramp in Frank Lloyd Wright's Guggenheim Museum in
New York City.
Not so in nature! Helical structures abound in living forms, from the
simplest virus to parts of the human body, and in almost every case the
genetic code carries information that tells each helix precisely "which
way to go." The genetic code itself is carried by a double-stranded he-
lical molecule of DNA, its two right-handed helices twining around
each other like the two snakes on the staff of Hermes. Moreover, since
Linus Pauling's pioneer work on the helical structure of protein mole-
cules, there has been increasing evidence that every giant protein mol-
ecule found in nature has a "backbone" that coils in a right-handed
helix. In both nucleic acid and protein, the molecule's backbone is a
chain made up of units each one of which is an asymmetric structure
of the same handedness. Each unit, so to speak, gives an additional
twist to the chain, in the same direction, like the steps of a helical stair-
case.
Larger helical structures in animals that have bilateral symmetry usu-
ally come in mirror-image pairs, one on each side of the body. The
horns of rams, goats, antelopes, and other mammals are spectacular
examples (see Figure 9.2). The cochlea of the human ear is a conical
helix that is left-handed in the left ear and right-handed in the right. A
curious exception is the tooth of the narwhal, a small whale that flour-
ishes in arctic waters. This whimsical creature is born with two teeth
in its upper jaw. Both teeth remain permanently buried in the jaw of the
female narwhal, and so does the right tooth of the male. But the male's
left tooth grows straight forward, like a javelin, to the ridiculous length
of eight or nine feet-more than half the animal's length from snout to
tail! Around this giant tooth are helical grooves that spiral forward in
a counterclockwise direction (see Figure 9.3). On the rare occasions
when both teeth grow into tusks, one would expect the right tooth to
spiral clockwise. But no, it too is always left-handed. Zoologists dis-
agree on how this could come about. Sir D' Arcy Thompson, in his book
On Growth and Form, defends his own theory that the whale swims
with a slight screw motion to the right. The inertia of its huge tusk
would produce a torque at the base of the tooth that might cause it to
rotate counterclockwise as it grows (see J. T. Bonner, "The Horn of the
Unicorn," Scientific American, March 1951).
Whenever a single helix is prominent in the structure of any living
The Helix 119
Figure 9.2. Helical horns of the Pamir sheep have opposite handedness.
Figure 9.3. Helical grooves of the narwhal tooth are always left-handed.
plant or animal, the species usually confines itself to a helix of a spe-
cific handedness. This is true of countless forms of helical bacteria as
well as of the spermatozoa of all higher animals. The human umbilical
cord is a triple helix of one vein and two arteries that invariably coil to
the left. The most striking instances are provided by the conical helices
of the shells of snails and other mollusks. Not all spiral shells have a
handedness. The chambered nautilus, for instance, coils on one plane;
like a spiral nebula, it can be sliced into identical left and right halves.
But there are thousands of beautiful molluscan shells that are either
left- or right-handed (see Figure 9.4). Some species are always left-
handed and some are always right-handed. Some go one way in one lo-
cality and the other way in another. Occasional "sports" that twist the
wrong way are prized by shell collectors.
A puzzling type of helical fossil known as the devil's corkscrew (Dae-
monelix) is found in Nebraska and Wyoming. These huge spirals, six
feet or more in length, are sometimes right-handed and sometimes left-
handed. Geologists argued for decades over whether they are fossils of
extinct plants or helical burrows made by ancestors of the beaver. The
beaver theory finally prevailed after remains of small prehistoric
beavers were found inside some of the corkscrews.
In the plant world helices are common in the structure of stalks,
stems, tendrils, seeds, flowers, cones, leaves-even in the spiral
arrangement of leaves and branches around a stalk. The number of
120 SOLID GEOMETRY AND HIGHER DIMENSIONS
Figure 9.4. Three molluscan shells that are right-handed conical helices
turns made along a helical path, as you move from one leaf to the leaf
directly above it, tends to be a number in the familiar Fibonacci series:
1, 2, 3, 5, 8, 13, ... (Each number is the sum of the preceding two num-
bers.) A large literature in the field known as "phyllotaxy" (leaf arrange-
ment) deals with the surprising appearance of the Fibonacci numbers
in botanical phenomena of this sort.
The helical stalks of climbing plants are usually right-handed, but
thousands of species of twining plants go the other way. The honey-
suckle, for instance, is always left-handed; the bindweed (a family that
includes the morning glory) is always right-handed. When the two
plants tangle with each other, the result is a passionate, violent embrace
that has long fascinated English poets. "The blue bindweed," wrote
Ben Jonson in 1617, "doth itself enfold with honeysuckle." And Shake-
speare, in A Midsummer Night's Dream, has Queen Titania speak of her
intention to embrace Bottom the Weaver (who has been transformed
into a donkey) by saying: "Sleep thou, and I will wind thee in my
arms'! ... So doth the woodbine the sweet honeysuckle/Gently en-
twist." In Shakespeare's day "woodbine" was a common term for
bindweed. Because it later carne to be applied exclusively to honey-
suckle many commentators reduced the passage to absurdity by su p-
posing that Titania was speaking of honeysuckle twined with
honeysuckle. Awareness of the opposite handedness of bindweed and
honeysuckle heightens, of course, the meaning of Titania's metaphor.
The Helix 121
More recently, a charming sQng called "Misalliance," celebrating the
love of the honeysuckle for the bindweed, has been written by the
British poet and entertainer Michael Flanders and set to music by his
friend Donald Swann. With Flanders' kind permission the entire song
is reproduced on the opposite page. (Readers who would like to learn
the tune can hear it sung by Flanders and Swann on the Angel record-
ing of At the Drop of a Hat, their hilarious two-man revue that made
such a hit in London and New York.) Note that Flanders' honeysuckle
is right-handed and his bindweed, left-handed. It is a matter of con-
vention whether a given helix is called left- or right-handed. If you
look at the point of a right-handed wood screw, you will see the helix
moving toward you counterclockwise, so that it can just as legitimately
be called left-handed. Flanders simply adopts the convention oppo-
site to the one taken here.
The entwining of two circular helices of opposite handedness is also
involved in a remarkable optical-illusion toy that was sold in this coun-
try in the 1930s. It is easily made by twisting together a portion of two
wire coils of opposite handedness (see Figure 9.6). The wires must be
soldered to each other at several points to make a rigid structure. The
illusion is produced by pinching the wire between thumb and forefin-
ger of each hand at the left and right edges of the central overlap. When
the hands are moved apart, the fingers and thumbs slide along the wire,
causing it to rotate and create a barber's-pole illusion of opposite hand-
edness on each side. This is continuously repeated. The wire seems to
be coming miraculously out of the inexhaustible meshed portion. Since
the neutrino and antineutrino are now known to travel with screw mo-
tions of opposite handedness, I like to think of this toy as demonstrat-
ing. the endless production of neutrinos and their mirror-image
particles.
On the science toy market at the time I write (2000) is a device called
"Spinfinity." It consists of two aluminum wire helices that are inter-
twined. When rotated you see the two helices moving in opposite di-
rections.
The helical character of the neutrino's path results from the fusion of
its forward motion (at the speed of light) with its "spin." Helical paths
of a similar sort are traced by many inanimate objects and living things:
a point on the propeller of a moving ship or plane, a squirrel running
up or down a tree, Mexican free-tailed bats gyrating counterclockwise
when they emerge from caves at Carlsbad, NM. Conically helical paths
122 SOLID GEOMETRY AND HIGHER DIMENSIONS
Figure 9.S.
MISALLIANCE
The fragrant Honeysuckle spirals clockwise to the sun
And many other creepers do the same.
But some climb counterclockwise, the Bindweed does, for one,
Or Convolvulus, to give her proper name.
Rooted on either side a door, one of each species grew,
And raced toward the window ledge above.
Each corkscrewed to the lintel in the only way it knew,
Where they stopped, touched tendrils, smiled and fell in love.
Said the right-handed Honeysuckle
To the left-handed Bindweed:
"Oh, let us get married,
If our parents don't mind. We'd
Be loving and inseparable.
Inextricably entwined, we'd
Live happily ever after, JJ
Said the Honeysuckle to the Bindweed.
To the Honeysuckle's parents it came as a shock.
"The Bindweeds," they cried, "are inferior stock.
They're uncultivated, of breeding bereft.
We twine to the right and they twine to the left!"
Said the countercockwise Bindweed
To the clockwise Honeysuckle:
"We'd better start saving-
Many a mickle maks a muckle-
Then run away for a honeymoon
And hope that our luck'll
Take a turn for the better, JJ
Said the Bindweed to the Honeysuckle.
A bee who was passing remarked to them then:
"I've said it before, and I'll say it again:
Consider your offshoots, if offshoots there be.
They'll never receive any blessing from me."
Poor little sucker, how will it learn
When it is climbing, which way to turn?
Right-left-what a disgrace!
Or it may go straight up and fall flat on its face!
Said the right-hand-thread Honeysuckle
To the left-hand-thread Bindweed:
"It seems that against us all fate has combined.
Oh my darling, oh my darling,
Oh my darling Columbine
Thou art lost and gone forever,
We shall never intertwine. JJ
Together they found them the very next day
They had pulled up their roots and just shriveled away,
Deprived of that freedom for which we must fight,
To veer to the left or to veer to the right!
MICHAEL FLANDERS
Figure 9.6. Helical toy that suggests the production of neutrinos
are taken by whirlpools, water going down a drain, tornadoes, and
thousands of other natural phenomena.
Writers have found helical motions useful on the metaphorical level.
The progress of science is often likened to an inverted conical spiral:
the circles growing larger and larger as science probes further into the
unknown, always building upward on the circles of the past. The same
spiral, a dark, bottomless whirlpool into which an individual or hu-
manity is sliding, has also been used as a symbol of pessimism and de-
spair. This is the metaphor that closes Norman Mailer's book
Advertisements for Myself. "Am I already on the way out?" he asks.
Time for Mailer is a conical helix of water flushing down a cosmic
drain, spinning him off "into the spiral of star-lit empty waters."
And now for a simple helix puzzle. A rotating barber's pole consists
of a cylinder on which red, white, and blue helices are painted. The
cylinder is four feet high. The red stripe cuts the cylinder's elements
(vertical lines) with a constant angle of 60 degrees. How long is the red
stripe?
The problem may seem to lack sufficient information for determin-
ing the stripe's length; actually it is absurdly easy when approached
properly.
Addendum
In my Second Scientific American Book of Mathematical Puz-
zles and Diversions (1961) I introduced the following brainteaser in-
volving two helices of the same handedness:
THE TWIDDLED BOLTS
Two icl,entical bolts are placed together so that their helical grooves in-
termesh (Figure 9.7). If you move the bolts around each other as you
would twiddle your thumbs, holding each bolt firmly by the head so that
it does not rotate and twiddling them in the direction shown, will the
heads (a) move inward, (b) move outward, or (c) remain the same dis-
tance from each other? The problem should be solved without resorting
to actual test.
124 SOLID GEOMETRY AND HIGHER DIMENSIONS
Figure 9.7. Twiddled bolts
Two similar helices are also involved in a curious patent by someone
named Socrates Scholfield of Providence, RI. I cannot now recall how
I came across it. A picture of the device, from the patent's first page, is
shown in Figure 9.B. It is to be used in classrooms for demonstrating the
nature of God. One helix represents good; the other, evil. As you can
see, they are holessly intertwined. The use of the device is given in de-
tail in five pages of the patent.
I know nothing about Mr. Scholfield except that in 1907 he pub-
lished a 59-page booklet titled The Doctrine of Mechanicalism. I tried
to check it in the New York Public Library many years ago, but the li-
brary has lost its copy. I once showed Scholfield's patent to the philoso-
pher Charles Hartshorne, one of my teachers at the University of
Chicago. I expected him to find the device amusing. To my surprise,
Hartshorne solemnly read the patent's pages and pronounced them ad-
mirable.
The Slinky toy furnishes an interesting problem in physics. If you
stand on a chair, holding one end of Slinky so that the helix hangs
straight down, then drop the toy, what happens? Believe it or not, the
lower end of Slinky doesn't move until the helix has come together,
then it falls with the expected rate. Throughout the experiment the
toy's center of gravity descends with the usual acceleration.
Jlnswers
If a right triangle is wrapped around any type of cylinder, the
base of the triangle going around the base of the cylinder, the triangle's
hypotenuse will trace a helix on the cylinder. Think of the red stripe of
the barber's pole as the hypotenuse of a right triangle, then "unwrap"
the triangle froIfL the cylinder. The triangle will have angles of 30 and
60 degrees. The hypotenuse of such a triangle must be twice the alti-
The Helix 125
1,087,186.
Ffs.2.
WITNESSES.


Figure 9.S.
126
S. SOHOLFIELD.
ILLUSTRATIVE EDUOATIONAL DEVIOE.
!l'I'LIOATIOH rlUD JUl. II, nOB.
Patented Feb. 17,1914.
fiG .1.

FiG.4.
FiG.5.
INVENTOR.
SOLID GEOMETRY AND HIGHER DIMENSIONS
tude. (This is easily seen if you place two such triangles together to
form an equilateral triangle.) In this case the altitude is four feet, so that
the hypotenuse (red stripe) is eight feet.
The interesting part of this problem is that the length of the stripe is
independent not only of the diameter of the cylinder but also of the
shape of its cross section. The cross section can be an irregular closed
curve of any shape whatever; the answer to the problem remains the
same.
The twiddled bolts move neither inward nor outward. They behave
like someone walking up a down escalator, always staying in the same
place.
Bibliography
T. A. Cook, The Curves of Life, New York: Henry Holt, 1914; Dover, 1979.
H.S.M. Coxeter, Introduction to Geometry, New York: John Wiley and Sons, 1961;
Second edition, 1989.
F. E. Graves, "Nuts and Bolts," Scientific American, June 1984, pp. 136-44.
J. B. Pettigrew, Design in Nature, Vol. II, London: Longmans, Green, and Co., 1908.
J. D. Watson, The Double Helix, New York: Atheneum, 1968; Norton, 1980.
The Helix 127
Chapter 10 Packing Spheres
Spheres of identical size can be piled and packed together in
many different ways, some of which have fascinating recreational fea-
tures. These features can be understood without models, but if the
reader can obtain a supply of 30 or more spheres, he will find them an
excellent aid to understanding. Ping-pong balls are perhaps the best for
this purpose. They can be coated with rubber cement, allowed to dry,
then stuck together to make rigid models.
First let us make a brief two-dimensional foray. If we arrange spheres
in square formation (see Figure 10.1, right), the number of balls in-
volved will of course be a square number. If we form a triangle (see Fig-
ure 10.1, left), the number of balls is a triangular number. These are the
simplest examples of what the ancients called "figurate numbers."
They were intensively studied by early mathematicians (a famous trea-
tise on them was written by Blaise Pascal), and although little attention
is paid them today, they still provide intuitive insights into many as-
pects of elementary number theory.
2
3
4
5
15
25
Figure 10.1. The basis of triangular numbers (left) and of square numbers (right)
128
For example, it takes only a glance at Figure 10.1, left, to see that the
sum of any number of consecutive positive integers, beginning with 1,
is a triangular number. A glance at Figure 10.1, right, shows that square
numbers are formed by the addition of consecutive odd integers, be-
ginning with 1. Figure 10.2 makes immediately evident an interesting
theorem known to the ancient pythagoreans: Every square number is
the sum of two consecutive triangular numbers. The algebraic proof is
simple. A triangular number with n units to a side is the sum of 1 + 2
+ 3 + ... n, and can be expressed by the formula ~ n ( n + 1). The pre-
ceding triangular number has the formula n (n - 1). If we add the two
formulas and simplify, the result is n
2
Are there numbers that are si-
multaneously square and triangular? Yes, there are infinitely many of
them. The smallest (not counting 1, which belongs to any figurate se-
ries) is 36; then the series continues: 1225,41616, 1413721,48024900,
.... It is not so easy to devise a formula for the nth term of this series.
~ 2 1
~ 2 8
49
F'lgUre 10.2. Square and triangular numbers are related.
Three-dimensional analogies of the plane-figurate numbers are ob-
tained by piling spheres in pyramids. Three-sided pyramids, the base
and sides of which are equilateral triangles, are models of what are
called the tetrahedral numbers. They form the series 1,4, 10, 20, 35, 56,
84, ... and can be represented by the formula !n(n + l)(n + 2), where n
is the number of balls along an edge. Four-sided pyramids, with square
bases and equilateral triangles for sides (Le., half of a regular octahe-
dron), represent the (square) pyramidal numbers 1, 5, 14, 30, 55, 91,
140, .... They have the formula !n(n + 1)(2n + 1). Just as a square can
be divided by a straight line into two consecutive triangles, so can a
Packing Spheres 129
PYTamids. (If you
has to be kept
or
Many
into two consecutive tetrahedral
a of a PYTamidal numberp the bottom
rolling apart. This can by ..., ...................... L"""
wood u..J.u .... .A.JOi.
t-'Tn,O' of pyra-
answer are ................. ,..., ...
................. " ~ , ............ now that we have a very large box
t
say a crate for a piano,
we wish to with as many golf as we can.
we a OJ" ................... "-''--'- as
circumferences Figure 10.3. The
10.3. In ne:x:agolnal ClOS8-1::lacking,
lows labeled B.
130
go in .. U..l .......... ", in in
AND
1. We place each ball on a hollow A that is directly above a ball in the
first layer. If we continue in this way, placing the balls of each layer
directly over those in the next layer but one, we produce a structure
called hexagonal close-packing.
2. We place each ball in a hollow B, directly above a hollow in the first
layer. If we follow this procedure for each layer (each ball will be di-
rectly above a ball in the third layer beneath it), the result is known
as cubic close-packing. Both the square and the tetrahedral pyramids
have a packing structure of this type, though on a square pyramid the
layers run parallel to the sides rather than to the base.
In forming the layers of a close-packing we can switch back and forth
whenever we please from hexagonal to cubic packing to produce vari-
ous hybrid forms of close-packing. In all these forms-cubic, hexago-
nal, and hybrid-each ball touches 12 other balls that surround it, and
the density of the packing (the ratio of the volume of the spheres to the
total space) is 1t/\118 = .74048 +, or almost 75 percent.
Is this the largest density obtainable? No denser packing is known,
but in an article published in 1958 (on the relation of close-packing to
froth) H.S.M. Coxeter of the University of Toronto made the startling
suggestion that perhaps the densest packing has not yet been found. It
is true that no more than 12 balls can be placed so that all of them
touch a central sphere, but a thirteenth ball can almost be added. The
large leeway here in the spacing ofthe 12 balls, in contrast to the com-
plete absence of leeway in the close-packing of circles on a plane, sug-
gests that there might be some form of irregular packing that would be
denser than .74. No one has yet proved that no denser packing is pos-
sible, or even that 12 point-contacts for each sphere are necessary for
densest packing. As a result of Coxeter's conjecture, George D. Scott of
the University of Toronto made some experiments in random packing
by pouring large numbers of steel balls into spherical flasks, then
weighing them to obtain the density. He found that stable random-
packings had a density that varied from about .59 to .63. So if there is
a packing denser than. 74, it will have to be carefully constructed on a
pattern that no one has yet thought of.
Assuming that close-packing is the closest packing, readers may like
to test their packing prowess on this exceedingly tricky little problem.
The interior of a rectangular box is 10 inches on each side and 5 inches
Packing Spheres 131
deep. What is the largest number of steel spheres 1 inch in diameter
that can be packed in this space?
If close-packed circles on a plane expand uniformly until they fill the
interstices between them, the result is the familiar hexagonal tiling of
bathroom floors. (This explains why the pattern is so common in na-
ture: the honeycomb of bees, a froth of bubbles between two flat sur-
faces almost in contact, pigments in the retina, the surface of certain
diatoms and so on.) What happens when closely packed spheres ex-
pand uniformly in a closed vessel or are subjected to uniform pressure
from without? Each sphere becomes a polyhedron, its faces corre-
sponding to planes that were tangent to its points of contact with other
spheres. Cubic close-packing transforms each sphere into a rhombic do-
decahedron (see Figure 10.4, top), the 12 sides of which are congruent
rhombi. Hexagonal close-packing turns each ball into a trapezo-
rhombic dodecahedron (see Figure 10.4, bottom), six faces of which are
rhombic and six, trapezoidal. If this figure is sliced in half along the
gray plane and one half is rotated 60 degrees, it becomes a rhombic do-
decahedron.
In 1727 the English physiologist Stephen Hales wrote in his book
Vegetable Staticks that he had poured some fresh peas into a pot, com-
pressed them, and had obtained "pretty regular dodecahedrons." The
experiment became known as the "peas of Buff on" (because the Comte
de Buffon later wrote about a similar experiment), and most biologists
accepted it without question until Edwin B. Matzke, a botanist at Co-
lumbia University, repeated the experiment. Because of the irregular
sizes and shapes of peas, their nonuniform consistency and the random
packing that results when peas are poured into a container, the shapes
of the peas after compression are too random to be identifiable. In ex-
periments reported in 1939 Matzke compressed lead shot and found
that if the spheres had been cubic close-packed, rhombic dodecahe-
drons were formed; but if they had been randomly packed, irregular 14-
faced bodies predominated. These results have important bearing,
Matzke has pointed out, on the study of such structures as foam and liv-
ing cells in undifferentiated tissues.
The problem of closest packing suggests the opposite question: What
is the loosest packing; that is, what rigid structure will have the lowest
possible density? For the structure to be rigid, each sphere must touch
at least four others, and the contact points must not be all in one hemi-
sphere or all on one equator of the sphere. In his Geometry and the
132 SOLID GEOMETRY AND HIGHER DIMENSIONS
133
- 2
AND
10.6 (top) is a "look-see"
four circles packed centers on a square lattice equal
area of area a Figure 10.6 (bottom) is a
between on a .................... ,...
tice is equal the area of the triangle minus half the
10.6. Two look-see proofs
................................ a 150-page '4proor' of Kepler's conjecture. It was widely hailed
as valid, but soon were finding
them, consensus
false and beyond
In 1998 Thomas Hales, at the University of Michigan and one of
Hsiang's harshest proof
cause it
on lengthy computer calculations, the verdict is still
whether Kepler's conjecture been verified.
Incidently,
more
pacldng of any identical convex solids.
The smallest number of oranges that will form two tetrahedral
pyramids of one ~ ..... ~
1
680. This is a tetrahedral number that can be split into two smaller
tetrahedral numbers: 120 and 560. The edges of the three pyramids are
8, 14, and 15.
A box 10 inches square and 5 inches deep can be close-packed with
1 inch-diameter steel balls in a surprising variety of ways, each ac-
commodating a different number of balls. The maximum number, 594,
is obtained as follows: Turn the box on its side and form the first layer
by making a row of five, then a row of four, then of five, and so on. It
is possible to make 11 rows (6 rows of five each, 5 rows of four each),
accommodating 50 balls and leaving a space of more than .3 inch to
spare. The second layer also will take 11 rows, alternating four and
five balls to a row, but this time the layer begins and ends with four-ball
rows, so that the number of balls in the layer is only 49. (The last row
of four balls will project .28+ inch beyond the edge of the first layer, but
because this is less than .3 inch, there is space for it.) Twelve layers
(with a total height of 9.98+ inches) can be placed in the box, alternat-
ing layers of 50 balls with layers of 49, to make a grand total of 594
balls.
Bibliography
Kepler's Conjecture
B. Cipra, "Gaps in a Sphere Packing Proof," Science, Vol. 259, 1993, p. 895.
T. C. Hales, "The Status of the Kepler Conjecture," Mathematical Intelligencer, Vol.
16, No. 3, 1994,pp.47-58.
T. C. Hales, "Cannonballs and Honeycombs," Notices of the American Mathemat-
ical Society, Vol. 47, April 2000, pp. 440-49.
J. Horgan, "Mathematicians Collide Over a Claim About Packing Spheres," Scien-
tific American, December 1994, pp. 32-34.
D. Mackenzie, "The Proof is in the Packing," American Scientist, VoL 86, Novem-
ber/December, 1998, pp. 524-25.
S. Singh, "Mathematics 'Proves' What the Grocer Always Knew," The New York
Times, August 25, 1998, p. F3.
1. Stewart, "Has the Sphere Packing Problem Been Solved?" New Scientist, Vol. 134,
May 2,1992, p. 16.
H. Wu, "On the Sphere Packing Problem and the Proof of Kepler's Conjecture," In-
ternational Journal of Mathematics, Vol. 4, No.5, 1998, pp. 739-831.
H. Wu, "A Rejoinder to Hales's Article," Mathematical Intelligencer, Vol. 17, No.1,
1995, pp. 35-42.
136 SOLID GEOMETRY AND HIGHER DIMENSIONS
Chapter II Spheres and
Hyperspheres
A circle is the locus of all points on the plane at a given dis-
tance from a fixed point on the plane. Let's extend this to Euclidean
spaces of all dimensions and call the general n-sphere the locus of all
points in n-space at a given distance from a fixed point in n-space. In
a space of one dimension (a line) the i-sphere consists of two points at
a given distance on each side of a center point. The 2-sphere is the cir-
cle, the 3-sphere is what is commonly called a sphere. Beyond that are
the hyperspheres of 4, 5, 6, ... dimensions.
Imagine a rod of unit length with one end attached to a fixed point.
If the rod is allowed to rotate only on a plane, its free end will trace a
unit circle. If the rod is allowed to rotate in 3-space, the free end traces
a unit sphere. Assume now that space has a fourth coordinate, at right
angles to the other three, and that the rod is allowed to rotate in 4-
space. The free end then generates a unit 4-sphere. Hyperspheres are
impossible to visualize; nevertheless, their properties can be studied by
a simple extension of analytic geometry to more than three coordinates.
A circle's Cartesian formula is a
2
+ b
2
= r, where r is the radius. The
sphere's formula is a
2
+ b
2
+ c
2
= r. The 4-sphere's formula is a
2
+ b
2
+
c
2
+ cP = r, and so on up the ladder of Euclidean hyperspaces.
The "surface" of an n-sphere has a dimensionality of n -1. A circle's
"surface" is a line of one dimension, a sphere's surface is two-
dimensional, and a 4-sphere's surface is three-dimensional. Is it possi-
ble that 3-space is actually the hypersurface of a vast 4-sphere? Could
such forces as gravity and electromagnetism be transmitted by the vi-
brations of such a hypersurface? Many late-19th-century mathemati-
cians and physicists, both eccentric and orthodox, took such
suggestions seriously. Einstein himself proposed the surface of a 4-
sphere as a model of the cosmos, unbounded and yet finite. Just as
137
Flatlanders on a sphere could travel the straightest possible line in any
direction and eventually return to their starting point, so (Einstein sug-
gested) if a spaceship left the earth and traveled far enough in anyone
direction, it would eventually return to the earth. If a Flatlander started
to paint the surface of the sphere on which he lived, extending the
paint outward in ever widening circles, he would reach a halfway point
at which the circles would begin to diminish, with himself on the in-
side, and eventually he would paint himself into a spot. Similarly, in
Einstein's cosmos, if terrestrial astronauts began to map the universe in
ever-expanding spheres, they would eventually map themselves into a
small globular space on the opposite side of the hypersphere.
Many other properties of hyperspheres are just what one would ex-
pect by analogy with lower-order spheres. A circle rotates around a
central point, a sphere rotates around a central line, a 4-sphere rotates
around a central plane. In general the axis of a rotating n-sphere is a
space of n - 2. (The 4-sphere is capable, however, of a peculiar double
rotation that has no analogue in 2- or 3-space: it can spin simultane-
ously around two fixed planes that are perpendicular to each other.)
The projection of a circle on a line is a line segment, but every point on
the segment, with the exception of its end points, corresponds to two
points on the circle. Project a sphere on a plane and you get a disk, with
every point inside the circumference corresponding to two points on
the sphere's surface. Project a 4-sphere on our 3-space and you get a
solid ball with every internal point corresponding to two points on the
4-sphere's hypersurface. This too generalizes up the ladder of spaces.
The same is true of cross sections. Cut a circle with a line and the
cross section is a l-sphere, or a pair of points. Slice a sphere with a
plane and the cross section is a circle. Slice a 4-sphere with a 3-space
hyperplane and the cross section is a 3-sphere. (You can't divide a 4-
sphere into two pieces with a 2-plane. A hyperapple, sliced down the
middle by a 2-plane, remains in one piece.) Imagine a 4-sphere moving
slowly through our space. We see it first as a point and then as a tiny
sphere that slowly grows in size to its maximum cross section, then
slowly diminishes and disappears.
A sphere of any dimension, made of sufficiently flexible material,
can be turned inside out through the next-highest space. Just as we can
twist a thin rubber ring until the outside rim becomes the inside, so a
hypercreature could seize one of our tennis balls and turn it inside out
through his space. He could do this all at once or he could start at one
138 SOLID GEOMETRY AND HIGHER DIMENSIONS
spot on the ball, turn a tiny portion first, then gradually enlarge it until
the entire ball had its inside outside.
One of the most elegant of the formulas that generalize easily to
spheres of all dimensions is the formula for the radii of the maximum
number of mutually touching n-spheres. On the plane, no more than
four circles can be placed so that each circle touches all the others,
with every pair touching at a different point. There are two possible sit-
uations (aside from degenerate cases in which one circle has an infinite
radius and so becomes a straight line): either three circles surround a
smaller one (Figure 11.1, left) or three circles are inside a larger one
(Figure 11.1, right). Frederick Soddy, the British chemist who received
a Nobel prize in 1921 for his discovery of isotopes, put it this way in
the first stanza of The Kiss Precise, a poem that appeared in Nature
(Vol. 137, June 20, 1936, p. 1021):
4
Figure 11.1. Find the radius of the fourth circle.
For pairs of lips to kiss maybe
Involves no trigonometry.
'Tis not so when four circles kiss
Each one the other three.
To bring this off the four must be
As three in one or one in three.
If one in three, beyond a doubt
Each gets three kisses from without.
If three in one, then is that one
Thrice kissed internally.
Spheres and Hyperspheres 139
Soddy's next stanza gives the simple formula. His term "bend" is
what is usually called the circle's curvature, the reciprocal of the ra-
dius. (Thus a circle of radius 4 has a curvature or "bend" of 1/4.) If a cir-
cle is touched on the inside, as it is in the case of the large circle
enclosing the other three, it is said to have a concave bend, the value
of which is preceded by a minus sign. As Soddy phrased all this:
Four circles to the kissing come.
The smaller are the benter.
The bend is just the inverse of
The distance from the center.
Though their intrigue left Euclid dumb
There's now no need for rule of thumb.
Since zero bend's a dead straight line
And concave bends have minus sign,
The sum of the squares of all four bends
Is half the square of their sum.
Letting a, b, c, d stand for the four reciprocals, Soddy's formula is
2(a
2
+ b
2
+ c
2
+ d
2
) = (a + b + c + dJ2. The reader should have little dif-
ficulty computing the radii of the fourth kissing circle in each illustra-
tion. In the poem's third and last stanza this formula is extended to five
mutually kissing spheres:
To spy out spherical affairs
An oscular surveyor
Might find the task laborious,
The sphere is much the gayer,
And now besides the pair of pairs
A fifth sphere in the kissing shares.
Yet, signs and zero as before,
For each to kiss the other four
The square of the sum of all five bends
Is thrice the sum of their squares.
The editors of Nature reported in the issue for January 9, 1937 (Vol.
139, p. 62), that they had received several fourth stanzas generalizing
Soddy's formula to n-space, but they published only the following, by
Thorold Gosset, an English barrister and amateur mathematician:
140
And let us not confine our cares
To simple circles, planes and spheres,
But rise to hyper flats and bends
SOLID GEOMETRY AND HIGHER DIMENSIONS
Where kissing multiple appears.
In n-ic space the kissing pairs
Are hyperspheres, and Truth declares-
As n + 2 such osculate
Each with an n + 1-fold mate.
The square of the sum of all the bends
Is n times the sum of their squares.
In simple prose, for n-space the maximum number of mutually
touching spheres is n + 2, and n times the sum of the squares of all
bends is equal to the square of the sum of all bends. It later developed
that the formula for four kissing circles had been known to Rene
Descartes, but Soddy rediscovered it and seems to have been the first
to extend it to spheres.
Note that the general formula even applies to the three mutually
touching two-point "spheres" of i-space: two touching line segments
"inside" a third segment that is simply the sum of the other two. The
formula is a great boon to recreational mathematicians. Puzzles about
mutually kissing circles or spheres yield readily to it. Here is a pretty
problem. Three mutually kissing spherical grapefruits, each with a ra-
dius of three inches, rest on a flat counter. A spherical orange is also on
the counter under the three grapefruits and touching each of them.
What is the radius of the orange?
Problems about the packing of unit spheres do not generalize easily
as one goes up the dimensional ladder; indeed, they become increas-
ingly difficult. Consider, for instance, the problem of determining the
largest number of unit spheres that can touch a unit sphere. For circles
the number is six (see Figure 11.2). For spheres it is 12, but this was not
proved until 1874. The difficulty lies in the fact that when 12 spheres
are arranged around a thirteenth, with their centers at the corners of an
imaginary icosahedron (see Figure 11.3), there is space between every
pair. The waste space is slightly more than needed to accommodate a
thirteenth sphere if only the 12 could be shifted around and properly
packed. If the reader will coat 14 ping-pong balls with rubber cement,
he will find it easy to stick 12 around one of them, and it will not be at
all clear whether or not the thirteenth can be added without undue
distortions. An equivalent question (can the reader see why?) is: Can 13
paper circles, each covering a 60-degree arc of a great circle on a sphere,
be pasted on that sphere without overlapping?
H.S.M. Coxeter, writing on "The Problem of Packing a Number of
Spheres and Hyperspheres 141
, 1.2.
a
a
142 AND
Equal Nonoverlapping Circles on a Sphere" (in Transactions of the
New York Academy of Sciences, Vol. 24, January 1962, pp. 320-31),
tells the story of what may be the first recorded discussion of the prob-
lem of the 13 spheres. David Gregory, an Oxford astronomer and friend
of Isaac Newton, recorded in his notebook in 1694 that he and Newton
had argued about just this question. They had been discussing how
stars of various magnitudes are distributed in the sky and this had led
to the question of whether or not one unit sphere could touch 13 oth-
ers. Gregory believed they could. Newton disagreed. As Coxeter writes,
"180 years were to elapse before R. Hoppe proved that Newton was
right." Simpler proofs have since been published, the latest in 1956 by
John Leech, a British mathematician.
How many unit hyperspheres in 4-space can touch a unit hyper-
sphere? It is not yet known if the answer is 24 or 25. Nor is it known
for any higher space. For spaces 4 through 8 the densest possible pack-
ings are known only if the centers of the spheres form a regular lattice.
These packings give 24, 40, 72, 126, and 240 for the "kissing number"
of spheres that touch another.
Why the difficulty with 9-space? A consideration of some paradoxes
involving hypercubes and hyperspheres may cast a bit of dim light on
the curious turns that take place in 9-space. Into a unit square one can
pack, from corner to diagonally opposite corner, a line with a length of
Vz. Into a unit cube one can similarly pack a line of vi The distance
between opposite corners of an n-cube is vIi, and since square roots in-
crease without limit, it follows that a rod of any size will pack into a
unit n-cube if n is large enough. A fishing pole 10 feet long will fit di-
agonally in the one-foot 100-cube. This also applies to objects of higher
dimension. A cube will accommodate a square larger than its square
face. A 4-cube will take a 3-cube larger than its cubical hyperface. A 5-
cube will take larger squares and cubes than any cube of lower dimen-
sion with an edge of the same length. An elephant or the entire Empire
State Building will pack easily into an n-cube with edges the same
length as those of a sugar cube if n is sufficiently large.
The situation with respect to an n-sphere is quite different. No mat-
ter how large n becomes, an n-sphere can never contain a rod longer
than twice its radius. And something very queer happens to its n-vol-
ume as n increases. The area of the unit circle is, of course, n. The vol-
ume of the unit sphere is 4.1+. The unit 4-sphere's hypervolume is
4.9+. In 5-space the volume is still larger, 5.2+, then in 6-space it de-
Spheres and Hyperspheres 143
creases to 5.1+, and thereafter steadily declines. Indeed, as n ap-
proaches infinity the hypervolume of a unit n-sphere approaches zero!
This leads to many unearthly results. David Singmaster, writing "On
Round Pegs in Square Holes and Square Pegs in Round Holes" (Math-
ematics Magazine, Vol. 37, November 1964, pp. 335-37), decided that
a round peg fits better in a square hole than vice versa because the ratio
of the area of a circle to a circumscribing square (re/4) is larger than the
ratio of a square inscribed in a circle (2/re). Similarly, one can show that
a ball fits better in a cube than a cube fits in a ball, although the differ-
ence between ratios is a bit smaller. Singmaster found that the differ-
ence continues to decrease through 8-space and then reverses: in
9-space the ratio of n-ball to n-cube is smaller than the ratio of n-cube
to n-ball. In other words, an n-ball fits better in an n-cube than an n-
cube fits in an n-ball if and only if n is 8 or less.
The same 9-space turn occurs in an unpublished paradox discov-
ered by Leo Moser. Four unit circles will pack into a square of side 4
(see Figure 11.4). In the center we can fit a smaller circle of radius Vz
- 1. Similarly, eight unit spheres will pack into the corners of a cube of
side 4 (see Figure 11.5). The largest sphere that will fit into the center
has a radius of v'3 - 1. This generalizes in the obvious way: In a 4-cube
of side 4 we can pack 16 unit 4-spheres and a central4-sphere of radius
V4 -1, which equals 1, so that the central sphere now is the same size
as the others. In general, in the corners of an n-cube of side 4 we can
pack 2
n
unit n-spheres and presumably another sphere of radius vIi -
1 will fit at the center. But see what happens when we come to 9-space:
the central hypersphere has a radius of y'g - 1 = 2, which is equal to
half the hypercube's edge. The central sphere cannot be larger than
this in any higher n-cube because it now fills the hypercube. No longer
is the central hypersphere inside the spheres that surround the center
of every hyperface, yet there is space at 2
9
= 512 corners to take 512 unit
9-spheres!
A related unpublished paradox, also discovered by Moser, concerns
n-dimensional chessboards. All the black squares of a chessboard are
enclosed with circumscribed circles (see Figure 11.6). Assume that
each cell is of side 2 and area 4. Each circle has a radius of Vz and an
area of 2re. The area in each white cell that is left white (is not enclosed
by a circle) is 8 - 2re = 1.71+. In the analogous situation for a cubical
chessboard, the black cubical cells of edge 2 are surrounded by spheres.
The volume of each black cell is 8 and the volume of each sphere,
144 SOLID GEOMETRY AND HIGHER DIMENSIONS
11.4. Four "" ..... " .... ..,.,. ........................ """ one of radius -1
".S.
one with a 1.
Spheres 145
11.6. Leo Moser's hVlper'chlessbo,ard problem
146 AND
such packing lattices are closely related to the construction of error-
correcting codes-efficient ways of transmitting information with a
minimum number of errors. The definitive treatise on the topic, by
John Horton Conway and N.J.A. Sloane (1988), runs 663 pages and has
a bibliography of 1,500 entries!
The densest packing of unit spheres, either lattice-based or irregu-
lar, is known only for the circle, regarded as a two-dimensional
"sphere." Densest lattice packings are known only for n-dimensional
spheres when n = 2 through 8. Beyond n 8 there are only conjectures
about such packings. In some cases irregular packings have been dis-
covered that are denser than lattice packings. The most famous of
higher space packings is a very dense lattice in 24 dimensions, known
as the Leech lattice after the British Mathematician John Leech who
discovered it. Each sphere in this remarkable structure touches
196,560 others!
Answers
The first problem was to determine the sizes of two circles,
each of which touches three mutually tangent circles with radii of one,
two, and three units. Using the formula given in the chapter,
2 1 + ~ + ~ + -
(
1 1 1 )
4 9 x
2 (
1 1 1)2
1+-+ + ~
2 3 x'
where x is the radius of the fourth circle, one obtains a value of 6/23 for
the radius of the smaller circle, 6 for the larger one.
The second problem concerned three grapefruits with three-inch
radii and an orange, all resting on a counter and mutually touching.
What size is the orange? The plane on which they rest is considered a
fifth sphere of infinite radius that touches the other four. Since it has
zero curvature it drops out of the formula relating the reciprocals of the
radii of five mutually touching spheres. Letting x be the radius of the
orange, we write the equation,
which gives x a value of one inch.
The problem can, of course, be solved in other ways. When it ap-
peared as problem 43 in the Pi Mu Epsilon Journal, November 1952,
Spheres and Hyperspheres 147
Leon Bankoff solved it this way, with R the radius of each large sphere
and r the radius of the small sphere:
"The small sphere, radius r, touches the table at a point equidistant
from the contacts of each of the large spheres with the table. Hence it
lies on the circumcenter of an equilateral triangle, the side of which is
2R. Then (R + r) is the hypotenuse of a right triangle, the altitude of
which is (R - r) and the base of which is 2Rv3/3. So
(R + r)2 = (R - r)2 + 4R2/3, or r = R/3."
The answer to Leo Moser's paradox of the hypercubic chessboard in
four-dimensional space is that no portion of a white cell remains un-
enclosed by the hyperspheres surrounding each black cell. The radius
of each hypersphere is \14, or 2. Since the hypercubic cells have edges
of length 2, we see at once that each of the eight hyperspheres around
a white cell will extend all the way to the center of that cell. The eight
hyperspheres intersect one another, leaving no portion of the white
cell unenclosed.
Bibliography
The Kiss Precise
W. S. Brown, "The Kiss Precise," American Mathematical Monthly, Vol. 76, June
1969, pp. 661-63.
D. Pedoe, "On a Theorem in Geometry," American Mathematical Monthly, Vol. 76,
June/July 1967, pp. 627----40.
Maximum Spheres that Touch a Sphere
H.S.M. Coxeter, "An Upper Bound for the Number of Equal Nonoverlapping
Spheres That Can Touch Another of the Same Size," Proceedings of Symposia
in Pure Mathematics, Vol. 7, 1963, pp. 53-71.
J. Leech, "The Problem of the Thirteen Spheres," The Mathematical Gazette, VoL
240, February 1956, pp. 22-23.
A. M. Odlyzko and N.J.A. Sloane, "New Bounds on the Number of Unit Spheres
That Can Touch a Unit Sphere in n Dimensions," Journal of Combinatorial The-
DIY, Ser. A, Vol. 26, March 1979, pp. 210-14.
J. C. Prepp, "Kepler's Spheres and Rubik's Cube," Mathematics Magazine, Vol. 61,
October 1988, pp. 231-39.
I. Stewart, "The Kissing Number," Scientific American, February 1992, pp. 112-15.
A. J. Wasserman, "The Thirteen Spheres Problem," Eureka, Spring 1978, pp. 4649.
148 SOLID GEOMETRY AND HIGHER DIMENSIONS
N-space Packing
J. H. Conway and N.J.A. Sloane, Leech Lattice, Sphere Packings, and Related Top-
ics, Springer-Verlag, 1984.
J. H. Conway and N.J.A. Sloane, Sphere Packings, Lattices, and Groups, Springer-
Verlag, 1988.
I. Peterson, "Curves For a Tighter Fit," Science News, Vol. 137, May 19, 1990, pp.
316-17.
N.J.A. Sloane, "The Packing of Spheres," Scientific American, January 1984, pp.
116-25.
Spheres and Hyperspheres 149
Chapter 12 The Church of
the Fourth Dimension
"Could 1 but rotate my arm out of the limits
set to it," one of the Utopians had said to
him, "1 could thrust it into a thousand
dimensions." -H. G. Wells. Men Like Gods
Alexander Pope once described London as a "dear, droll, dis-
tracting town." Who would disagree? Even with respect to recreational
mathematics, I have yet to make an imaginary visit to London without
coming on something quite extra-ordinary. Once, for instance, I was
reading the London Times in my hotel room a few blocks from Pic-
cadilly Circus when a small advertisement caught my eye:
Weary of the world of three dimensions? Come worship with us Sunday
at the Church of the Fourth Dimension. Services promptly at 11 A.M., in
Plato's grotto. Reverend Arthur Slade, Minister.
An address was given. I tore out the advertisement, and on the follow-
ing Sunday morning rode the Underground to a station within walking
distance of the church. There was a damp chill in the air and a light
mist was drifting in from the sea. I turned the last corner, completely
unprepared for the strange edifice that loomed ahead of me. Four enor-
mous cubes were stacked in one column, with four cantilevered cubes
jutting in four directions from the exposed faces of the third cube from
the ground. I recognized the structure at once as an unfolded hyper-
cube. Just as the six square faces of a cube can be cut along seven lines
and unfolded to make a two-dimensional Latin cross (a popular floor
plan for medieval churches), so the eight cubical hyperfaces of a four-
dimensional cube can be cut along seventeen squares and "unfolded"
to form a three-dimensional Latin cross.
A smiling young woman standing inside the portal directed me to a
stairway. It spiraled down into a basement auditorium that I can only
describe as a motion-picture theater combined with a limestone cavern.
The front wall was a solid expanse of white. Formations of translucent
pink stalactites glowed brightly on the ceiling, flooding the grotto with
150
a rosy light. Huge stalagmites surrounded the room at the sides and
back. Electronic organ music, like the score of a science-fiction film,
surged into the room from all directions. I touched one of the stalag-
mites. It vibrated beneath my fingers like the cold key of a stone xylo-
phone.
The strange music continued for 10 minutes or more after I had taken
a seat, then slowly softened as the overhead light began to dim. At the
same time I became aware of a source of bluish light at the rear of the
grotto. It grew more intense, casting sharp shadows of the heads of
the congregation on the lower part of the white wall ahead. I turned
around and saw an almost blinding point oflight that appeared to come
from an enormous distance.
The music faded into silence as the grotto became completely dark
except for the brilliantly illuminated front wall. The shadow of the
minister rose before us. After announcing the text as Ephesians, Chap-
ter 3, verses 17 and 18, he began to read in low, resonant tones that
seemed to come directly from the shadow's head: " ... that ye, being
rooted and grounded in love, may be able to comprehend with all saints
what is the breadth, and length, and depth, and height. ... "
It was too dark for note-taking, but the following paragraphs sum-
marize accurately, I think, the burden of Slade's remarkable sermon.
Our cosmos-the world we see, hear, feel-is the three-dimensional
"surface" of a vast, four-dimensional sea. The ability to visualize, to
comprehend intuitively, this "wholly other" world of higher space is
given in each century only to a few chosen seers. For the rest of us, we
must approach hyperspace indirectly, by way of analogy. Imagine a
Flatland, a shadow world of two dimensions like the shadows on the
wall of Plato's famous cave (Republic, Chapter 7). But shadows do not
have material substance, so it is best to think of Flatland as possessing
an infinitesimal thickness equal to the diameter of one of its funda-
mental particles. Imagine these particles floating on the smooth surface
of a liquid. They dance in obedience to two-dimensional laws. The in-
habitants of Flatland, who are made up of these particles, cannot con-
ceive of a third direction perpendicular to the two they know.
We, however, who live in 3-space can see every particle of Flatland.
We see inside its houses, inside the bodies of every Flatlander. We can
touch every particle of their world without passing our finger through
their space. If we lift a Flatlander out of a locked room, it seems to him
a miracle.
The Church of the Fourth Dimension 151
In an analogous way, Slade continued, our world of 3-space floats on
the quiet surface of a gigantic hyperocean; perhaps, as Einstein once
suggested, on an immense hypersphere. The four-dimensional thick-
ness of our world is approximately the diameter of a fundamental par-
ticle. The laws of our world are the "surface tensions" of the hypersea.
The surface of this sea is uniform, otherwise our laws would not be uni-
form. A slight curvature of the sea's surface accounts for the slight,
constant curvature of our space-time. Time exists also in hyperspace.
If time is regarded as our fourth coordinate, then the hyperworld is a
world of five dimensions. Electromagnetic waves are vibrations on the
surface of the hypersea. Only in this way, Slade emphasized, can sci-
ence escape the paradox of an empty space capable of transmitting en-
ergy.
What lies outside the sea's surface? The wholly other world of God!
No longer is theology embarrassed by the contradiction between God's
immanence and transcendence. Hyperspace touches every point of 3-
space. God is closer to us than our breathing. He can see every portion
of our world, touch every particle without moving a finger through our
space. Yet the Kingdom of God is completely "outside" 3-space, in a di-
rection in which we cannot even point.
The cosmos was created billions of years ago when God poured
(Slade paused to say that he spoke metaphorically) on the surface of the
hypersea an enormous quantity of hyperparticles with asymmetric
three-dimensional cross sections. Some of these particles fell into 3-
space in right-handed form to become neutrons, the others in left-
handed form to become antineutrons. Pairs of opposite parity
annihilated each other in a great primeval explosion, but a slight pre-
ponderance of hyperparticles happened to fall as neutrons and this ex-
cess remained. Most of these neutrons split into protons and electrons
to form hydrogen. So began the evolution of our "one-sided" material
world. The explosion caused a spreading of particles. To maintain this
expanding universe in a reasonably steady state, God renews its mat-
ter at intervals by dipping his fingers into his supply of hyperparticles
and flicking them toward the sea. Those which fall as antineutrons are
annihilated, those which fall as neutrons remain. Whenever an an-
tiparticle is created in the laboratory, we witness an actual "turning
over" of an asymmetric particle in the same way that one can reverse
in 3-space an asymmetric two-dimensional pattern of cardboard. Thus
152 SOLID GEOMETRY AND HIGHER DIMENSIONS
the production of antiparticles provides an empirical proof of the real-
ity of 4-space.
Slade brought his sermon to a close by reading from the recently dis-
covered Gnostic Gospel of Thomas: "If those who lead you say to you:
Behold the kingdom is in heaven, then the birds will precede you. If
they say to you that it is in the sea, then the fish will precede you. But
the kingdom is within you and it is outside of you."
Again the unearthly organ music. The blue light vanished, plunging
the cavern into total blackness. Slowly the pink stalactites overhead
began to glow, and I blinked my eyes, dazzled to find myself back in 3-
space.
Slade, a tall man with iron-gray hair and a small dark mustache, was
standing at the grotto's entrance to greet the members of his congrega-
tion. As we shook hands I introduced myself and mentioned my Sci-
entific American column. "Of course!" he exclaimed. "I have some of
your books. Are you in a hurry? If you wait a bit, we'll have a chance
to chat."
After the last handshake Slade led me to a second spiral stairway of
opposite handedness from the one on which I had descended earlier.
It carried us to the pastor's study in the top cube of the church. Elabo-
rate models, 3-space projections of various types of hyperstructures,
were on display around the room. On one wall hung a large reproduc-
tion of Salvador Dali's painting "Corpus Hypercubus." In the picture,
above a flat surface of checkered squares, floats a three-dimensional
cross of eight cubes; an unfolded hypercube identical in structure with
the church in which I was standing.
"Tell me, Slade," I said, after we were seated, "is this doctrine of
yours new or are you continuing a long tradition?"
"It's by no means new," he replied, "though I can claim to have es-
tablished the first church in which hyperfaith serves as the corner-
stone. Plato, of course, had no conception of a geometrical fourth
dimension, though his cave analogy clearly implies it. In fact, every
form of Platonic dualism that divides existence into the natural and su-
pernatural is clearly a nonmathematical way of speaking about higher
space. Henry More, the 17th-century Cambridge Platonist, was the first
to regard the spiritual world as having four spatial dimensions. Then
along came Immanuel Kant, with his recognition of our space and time
as subjective lenses, so to speak, through which we view only a thin
The Church of the Fourth Dimension 153
slice of transcendent reality. After that it is easy to see how the concept
of higher space provided a much needed link between modern science
and traditional religions."
"You say 'religions,' " I put in. "Does that mean your church is not
Christian? "
"Only in the sense that we find essential truth in all the great world
faiths. I should add that in recent decades the Continental Protestant
theologians have finally discovered 4-space. When Karl Barth talks
about the 'vertical' or 'perpendicular' dimension, he clearly means it in
a four-dimensional sense. And of course in the theology of Karl Heim
there is a full, explicit recognition of the role of higher space."
"Yes," I said. "I recently read an interesting book called Physicist
and Christian, by William G. Pollard (executive director of the Oak
Ridge Institute of Nuclear Studies and an Episcopal clergyman). He
draws heavily on Heim's concept of hyperspace."
Slade scribbled the book's title on a note pad. "I must look it up. I
wonder if Pollard realizes that a number of late-19th-century Protes-
tants wrote books about the fourth dimension. A. T. Schofield's Another
World, for example (it appeared in 1888) and Arthur Willink's The
World of the Unseen (subtitled "An Essay on the Relation of Higher
Space to Things Eternal"; published in 1893). Of course modern oc-
cultists and spiritualists have had a field day with the notion. Peter D.
Ouspensky, for instance, has a lot to say about it in his books, although
most of his opinions derive from the speculations of Charles Howard
Hinton, an American mathematician. Whately Carington, the English
parapsychologist, wrote an unusual book in 1920-he published it
under the byline of W. Whately Smith-on A Theory of the Mecha-
nism of Surviva1. "
"Survival after death?"
Slade nodded. "I can't go along with Carington's belief in such things
as table tipping being accomplished by an invisible four-dimensional
lever, or clairvoyance as perception from a point in higher space, but I
regard his basic hypothesis as sound. Our bodies are simply three-
dimensional cross sections of our higher four-dimensional selves. Ob-
viously a man is subject to all the laws of this world, but at the same
time his experiences are permanently recorded-stored as information,
so to speak-in the 4-space portion of his higher self. When his 3-space
body ceases to function, the permanent record remains until it can be
154 SOLID GEOMETRY AND HIGHER DIMENSIONS
attached to a new body for a new cycle of life in some other 3-space
continuum. "
"I like that," I said. HIt explains the complete dependence of mind on
body in this world, at the same time permitting an unbroken continu-
ity between this life and the next. Isn't this close to what William James
struggled to say in his little book on immortality?"
"Precisely. James, unfortunately, was no mathematician, so he had to
express his meaning in nongeometrical metaphors."
"What about the so-called demonstrations of the fourth dimension by
certain mediums," I asked. "Wasn't there a professor of astrophysics in
Leipzig who wrote a book about them?"
I thought I detected an embarrassed note in Slade's laugh. "Yes, that
was poor Johann Karl Friedrich Zollner. His book Transcendental
Physics was translated into English in 1881, but even the English copies
are now quite rare. Zollner did some good work in spectrum analysis,
but he was supremely ignorant of conjuring methods. As a consequence
he was badly taken in, I'm afraid, by Henry Slade, the American
medium."
"Slade?" I said with surprise.
"Yes, I'm ashamed to say we're related. He was my great-uncle. When
he died, he left a dozen fat notebooks in which he had recorded his
methods. Those notebooks were acquired by the English side of my
family and handed down to me."
"This excites me greatly," I said. "Can you demonstrate any of the
tricks?"
The request seemed to please him. Conjuring, he explained, was one
of his hobbies, and he thought that the mathematical angles of several
of Henry's tricks would be of interest to my readers.
From a drawer in his desk Slade took a strip of leather, cut as shown
at the left in Figure 12.1, to make three parallel strips. He handed me a
ballpoint pen with the request that I mark the leather in some way to
prevent later substitution. I initialed a corner as shown. We sat on op-
posite sides of a small table. Slade held the leather under the table for
a few moments, then brought it into view again. It was braided exactly
as shown at the right in the illustration! Such braiding would be easy
to accomplish if one could move the strips through hyperspace. In 3-
space it seemed impossible.
Slade's second trick was even more astonishing. He had me examine
The Church of the Fourth Dimension 155
Figure 12.1. Slade's leather strip
-braided in hyperspace?
a rubber band of the wide, flat type shown at the left in Figure 12.2.
This was placed in a matchbox, and the box was securely sealed at
both ends with cellophane tape. Slade started to place it under the
table, then remembered he had forgotten to have me mark the box for
later identification. I drew a heavy X on the upper surface.
Figure 12.2. Slade's rubber band -knotted in hyperspace?
"If you like," he said, "you yourself may hold the box under the
table."
I did as directed. Slade reached down, taking the box by its other end.
There was a sound of movement and I could feel that the box seemed
to be vibrating slightly.
Slade released his grip. "Please open the box."
First I inspected the box carefully. The tape was still in place. My
mark was on the cover. I slit the tape with my thumbnail and pushed
open the drawer. The elastic band-mira bile dictu-was tied in a sim-
ple knot as shown at the right in Figure 12.2.
"Even if you managed somehow to open the box and switch bands,"
I said, "how the devil could you get a rubber band like this?"
Slade chuckled. "My great-uncle was a clever rascal."
156 SOLID GEOMETRY AND HIGHER DIMENSIONS
I was unable to persuade Slade to tell me how either trick was done.
The reader is invited to think about them before he reads this chapter's
answer section.
We talked of many other things. When I finally left the Church of the
Fourth Dimension, a heavy fog was swirling through the wet streets of
London. I was back in Plato's cave. The shadowy forms of moving cars,
their headlights forming flat elliptical blobs of light, made me think of
some familiar lines from the Rubaiyat of a great Persian mathemati-
cian:
We are no other than a moving row
Of magic shadow-shapes that come and go
Round with the sun-illumined lantern held
In midnight by the Master of the Show.
Addendum
Although I spoke in the first paragraph of this chapter of an
"imaginary visit" to London, when the chapter first appeared in Sci-
entific American several readers wrote to ask for the address of Slade's
church. The Reverend Slade is purely fictional, but Henry Slade the
medium was one of the most colorful and successful mountebanks in
the history of American spiritualism. See my article on Slade in The
Encyclopedia of the Paranormal, edited by Gordon Stein, Prometheus
Books, 1996, and my remarks about Slade's fourth-dimensional trick-
ery in The New Ambidextrous Universe (W. H. Freeman, 1990).
At the time I wrote about the Church of the Fourth Dimension no em-
inent physicist had ever contended that there might actually be spaces
"out there," higher than our familiar 3-space. (The use of a fourth di-
mension in relativity theory was no more than a way of handling time
in the theory's equations.) Now, however, particle physicists are in a eu-
phoric state over a theory of superstrings in which fundamental parti-
cles are not modeled as geometrical points, but as extremely tiny closed
loops, of great tensile strength, that vibrate in higher spaces. These
higher spaces are "compacted"-curled up into tight little structures
too small to be visible or even to be detected by today's atom smashers.
Some physicists regard these higher spaces as mere artifices of the
mathematics, but others believe they are just as real as the three spaces
we know and love. (On superstrings, see my New Ambidextrous Uni-
The Church of the Fourth Dimension 157
verse.) It is the first time physicists have seriously entertained the no-
tion of higher spaces that are physically real. This may have stimu-
lated the publication of recent books on the fourth and higher
dimensions. I particularly recommend Thomas F. Banchoff's Beyond
the Third Dimension, if for no other reason than for its wondrous com-
puter graphics.
In recent years the theory of superstrings has been extended to mem-
brane or brane theory in which the strings are attached to vibrating
surfaces that live in higher space dimensions. Some theorists are spec-
ulating that our universe is an enormous brane floating in hyperspace
along with countless other "island universes," each with its own set of
laws. A possibility looms that these other universes could be detected
by their gravity seeping into our universe and generating the elusive in-
visible dark matter which physicists suspect furnishes 90 percent of the
matter in our universe. For these wild speculations see George John-
son's mind-boggling article, "Physicists Finally Find a Way to Test Su-
perstring Theory," in The New York Times, April 4, 2000.
Membrane theory is also called M-theory, the M standing for mem-
brane, mystery, magic, marvel, and matrix. See "Magic, Mystery, and
Matrix," a lecture by Edward Witten, the world's top expert on super-
strings, published in Notices of the AMS (American Mathematical So-
ciety), Vol. 45, November 1998, and his 1998 videotape M-Theory,
available from the AMS.
Answers
Slade's method of braiding the leather strip is familiar to Boy
Scouts in England and to all those who make a hobby of leathercraft.
Many readers wrote to tell me of books in which this type of braiding
is described: George Russell Shaw, Knots, Useful and Ornamental (p.
86); Constantine A. Belash, Braiding and Knotting (p. 94); Clifford Pyle,
Leather Craft as a Hobby (p. 82); Clifford W. Ashley, The Ashley Book
of Knots (p. 486); and others. For a full mathematical analysis, see
J.A.H. Shepperd, "Braids Which Can Be Plaited with Their Threads
Tied Together at Each End," Proceedings of the Royal Society, A, Vol.
265,1962, pages 229-44.
There are several ways to go about making the braid. Figure 12.3 was
drawn by reader George T. Rab of Dayton, OH. By repeating this pro-
cedure one can extend the braid to any multiple of six crossings. An-
158 SOLID GEOMETRY AND HIGHER DIMENSIONS
"
159
this when he started to put the box under the table, then "remembered"
that I had not yet initialed it. The prepared box could have been stuck
to the underside of the table with magician's wax. It would require
only a moment to press the unprepared box against another dab of wax,
then take the prepared one. In this way the switch occurred before I
marked the box. The vibrations I felt when Slade and I held the box
under the table were probably produced by one of Slade's fingers press-
ing firmly against the box and sliding across it.
Fitch Cheney, mathematician and magician, wrote to tell about a sec-
ond and simpler way to create a knotted elastic band. Obtain a hollow
rubber torus-they are often sold as teething rings for babies-and cut
as shown by the dotted line in Figure 12.5. The result is a wide endless
band tied in a single knot. The band can be trimmed, of course, to nar-
rower width.
Figure 12.S. A second way
to produce a knotted
rubber band
It was Stover, by the way, who first suggested to me the problem of
tying a knot in an elastic band. He had been shown such a knotted
band by magician Winston Freer. Freer said he knew three ways of
doing it.
Bibliography
T. F. Banchoff, Beyond the Third Dimension, New York: W. H. Freeman, 1990.
M. Gardner, "Group Theory and Braids," New Mathematical Diversions from Sci-
entific American, Chapter 2, New York: Simon and Schuster, 1966. Reissued by
the University of Chicago Press, 1983.
K. Heim, Christian Faith and Natural Science, New York: Harper, 1953; paper-
back,1957.
C. H. Hinton, The Fourth Dimension, London: Swan Sonnenschein, 1904; Allen &
Unwin, 1951.
160 SOLID GEOMETRY AND HIGHER DIMENSIONS
H. P. Manning, Geometry of Four Dimensions, New York: Macmillan, 1914; Dover
Publications (paperback), 1956.
H. P. Manning, The Fourth Dimension Simply Explained, New York: Munn, 1910;
New York: Dover Publications (paperback), 1960.
E. H. Neville, The Fourth Dimension, Cambridge, England: Cambridge University
Press, 1921.
C. Pickover, Surfing Through Hyperspace, England: Oxford University Press, 1999.
R. Rucker (ed.), Speculations on the Fourth Dimension, New York: Dover, 1980.
R. Rucker, The Fourth Dimension: Toward a Geometry of Higher Reality, Boston:
Houghton-Mifflin, 1984.
J.A.H. Shepperd, "Braids Which Can Be Plaited with Their Threads Tied Together
at Each End," Proceedings of the Royal Society, A, Vol. 265, 1962, pp. 229-44.
D.M.Y. Sommerville, An Introduction to the Geometry of N Dimensions, London:
Methuen, 1929; New York: Dover Publications (paperback), 1958.
The Church of the Fourth Dimension 161
Chapter 13
The children were vanishing.
They went in fragments, like thick smoke in a
wind, or like movement in a distorting mirror.
Hand in hand they went, in a direction Para-
dine could not understand . ...
-LEwIs PADGETI, from "Mimsy Were the Borogoves"
Hypercubes
The direction that Paradine, a professor of philosophy, could
not understand is a direction perpendicular to each of the three coor-
dinates of space. It extends into 4-space in the same way a chess piece
extends upward into 3-space with its axis at right angles to the x and y
coordinates of the chessboard. In Padgett's great science fiction story,
Paradine's children find a wire model of a tesseract (a hypercube of four
dimensions) with colored beads that slide along the wires in curious
ways. It is a toy abacus that had been dropped into our world by a 4-
space scientist tinkering with a time machine. The abacus teaches the
children how to think four-dimensionally. With the aid of some cryp-
tic advice in Lewis Carroll's Jabberwocky they finally walk out of 3-
space altogether.
Is it possible for the human brain to visualize four-dimensional struc-
tures? The 19th-century German physicist Hermann von Helmholtz ar-
gued that it is, provided the brain is given proper input data.
Unfortunately our experience is confined to 3-space and there is not the
slightest scientific evidence that 4-space actually exists. (Euclidean 4-
space must not be confused with the non-Euclidean four-dimensional
space-time of relativity theory, in which time is handled as a fourth co-
ordinate.) Nevertheless, it is conceivable that with the right kind of
mathematical training a person might develop the ability to visualize
a tesseract. "A man who devoted his life to it," wrote Henri Poincare,
"could perhaps succeed in picturing to himself a fourth dimension."
Charles Howard Hinton, an eccentric American mathematician who
once taught at Princeton University and who wrote a popular book
called The Fourth Dimension, devised a system of using colored blocks
for making 3-space models of sections of a tesseract. Hinton believed
that by playing many years with this "toy" (it may have suggested the
162
toy in Padgett's story), he had acquired a dim intuitive grasp of 4-space.
"I do not like to speak positively," he wrote, "for I might occasion a loss
of time on the part of others, if, as may very well be, I am mistaken. But
for my own part, I think there are indications of such an intuition .... "
Hinton's colored blocks are too complicated to explain here (the
fullest account of them is in his 1910 book, A New Era a/Thought). Per-
haps, however, by examining some of the simpler properties of the
tesseract we can take a few wobbly first steps toward the power of vi-
sualization Hinton believed he had begun to achieve.
Let us begin with a point and move it a distance of one unit in a
straight line, as shown in Figure 13.1(a). All the points on this unit line
can be identified by numbering them from 0 at one end to 1 at the
other. Now move the unit line a distance of one unit in a direction per-
pendicular to the line (b). This generates a unit square. Label one cor-
ner 0, then number the points from 0 to 1 along each of the two lines
that meet at the zero corner. With these x and y coordinates we can now
label every point on the square with an ordered pair of numbers. It is
just as easy to visualize the next step. Shift the square a unit distance
in a direction at right angles to both the x and the yaxes (e). The result
is a unit cube. With x, y, z coordinates along three edges that meet at a
corner, we can label every point in the cube with an ordered triplet of
numbers.
Although our visual powers boggle at the next step, there is no logi-
cal reason why we cannot assume that the cube is shifted a unit dis-
tance in a direction perpendicular to all three of its axes (d). The space
generated by such a shift is a 4-space unit hypercube-a tesseract-
with four mutually perpendicular edges meeting at every corner. By
choosing a set of such edges as w, x, y, z axes, one might label every
point in the hypercube with an ordered quadruplet of numbers. Ana-
lytic geometers can work with these ordered quadruplets in the same
way they work with ordered pairs and triplets to solve problems in
plane and solid geometry. In this fashion Euclidean geometry can be ex-
tended to higher spaces with dimensions represented by any positive
integer. Each space is Euclidean but each is topologically distinct: a
square cannot be continuously deformed to a straight line, a cube de-
formed to a square, a hypercube to a cube, and so on.
Accurate studies of figures in 4-space can be made only on the basis
of an axiomatic system for 4-space, or by working analytically with the
w, x, y, z equations of the four-coordinate system. But the tesseract is
Hypercubes 163
a
....... _----...
o
C
~ 1
;;
o
""",;
;"
.,;
"
1 ;;
.,'
;'"
... ;
b
0 1
, ,
I I
1 I I 1
I I
I I
I I
1-
-
d
Figure 13.1. Steps toward generating a hypercube
such a simple 4-space structure that we can guess many of its proper-
ties by intuitive, analogical reasoning. A unit line has two end points.
When it is moved to generate a square, its ends have starting and stop-
ping positions and therefore the number of corners on the square is
twice the number of points on the line, or four. The two moving points
generate two lines, but the unit line has a start and a stop position and
so we must add two more lines to obtain four as the number of lines
bounding the square.
In similar fashion, when the square is moved to generate a cube, its
four corners have start and stop positions and therefore we multiply
four by two to arrive at eight corners on the cube. In moving, each of the
four points generates a line, but to those four lines we must add the
square's four lines at its start and the four lines at its stop, making 4 +
4 + 4 = 12 edges on the cube. The four lines of the moving square gen-
erate four new faces, to which the start and stop faces are added, mak-
ing 4 + 1 + 1 = 6 faces on the cube's surface.
Now suppose the cube is pushed a unit distance in the direction of
a fourth axis at right angles to the other three, a direction in which we
cannot point because we are trapped in 3-space. Again each corner of
the cube has start and stop positions, so that the resulting tesseract has
164 SOLID GEOMETRY AND HIGHER DIMENSIONS
2 x 8 = 16 corners. Each point generates a line, but to these eight lines
we must add the start and stop positions of the cube's 12 edges to make
8 + 12 + 12 = 32 unit lines on the hypercube. Each of the cube's 12 edges
generates a square, but to those 12 squares we must add the cube's six
squares before the push and the six after the push, making 12 + 6 + 6 =
24 squares on the tesseract's hypersurface.
It is a mistake to suppose the tesseract is bounded by its 24 squares.
They form only a skeleton of the hypercube, just as the edges of a cube
form its skeleton. A cube is bounded by square faces and a hypercube
is bounded by cubical faces. When the cube is pushed, each of its
squares moves a unit distance in an unimaginable direction at right
angles to its face, thereby generating another cube. To the six cubes
generated by the six moving squares we must add the cube before it is
pushed and the same cube after it is pushed, making eight in all. These
eight cubes form the hypercube's hypersurface.
The chart in Figure 13.2 gives the number of elements in "cubes" of
spaces one through four. There is a simple, surprising trick by which
this chart can be extended downward to higher n-cubes. Think of the
nth line as an expansion of the binomial (2x + l)n. For example, the line
segment of 1-space has two points and one line. Write this as 2x+ 1 and
multiply it by itself:
2x + 1
2x + 1
4r + 2x
2x + 1
4x
2
+ 4x + 1
n-SPACE POINTS
0 1
1 2
2 4
3 8
4 16
LINES
0
1
4
12
32
SQUARES CUBES TESSERACTS
0 0 0
0 0 0
1 0 0
6 1 0
24 8 1
Figure 13.2. Elements of structures analogous to the cube in various dimensions
Hypercubes 165
Note that the answer correspond the chart's
Indeed, each line of the chart, written as a polynomial
+ 1, gives the next are of a
space as a fourth-power
and multiply by 2x+ 1:
16x4 + + +
+1
2x + 1
+ 16x
2
+
+ 24x2 + Bx + 1
+ 80.0 + 4ox2 + + 1
one
............... LI"' .... on the chart equals twice the number
it number diagonally
If you ............ '-, ....................... ofa on
a plane, you can the
comes from a I'J\J.L.L.L'- close to the cube and the cube is held a cer-
the projection 1
example, a fly cannot along all edges of a cube in a
_..., ......... _ over an edge twice J nor can it do this on
the
166
GEOMETRY AND
Figure 13.4 is the analogous projection in 3-space of the edges of a
tesseract; more accurately, it is a plane projection of a three-
dimensional model that is in turn a projection of the hypercube. All the
elements of the tesseract given by the chart are easily identified in the
model, although six of the eight cubes suffer perspective distortions just
as four of the cube's square faces are distorted in its projection on the
plane. The eight cubes are the large cube, the small interior cube, and
the six hexahedrons surrounding the small cube. (Readers should also
try to find the eight cubes in Figure 13.1(d)-a projection of the tesser-
act, from a different angle, into another 3-space model.) Here again the
topological properties of both models are the same'as those of the edges
of the tesseract. In this case a fly can walk along all the edges without
traversing any edge twice. (In general the fly can do this only on hy-
percubes in even spaces, because only in even spaces do an even num-
ber of edges meet at each vertex.)
Figure 13.4. Projection of the
tesseract in 3-space
Many properties of unit hypercubes can be expressed in simple for-
mulas that apply to hypercubes of all dimensions. For example, the di-
agonal of a unit square has a length of Vz. The longest diagonals on the
unit cube have a length of vi In general a diagonal from corner to op-
posite corner on a unit cube in n-space is Vn.
A square of side x has an area of x
2
and a perimeter of 4x. What size
square has an area equal to its perimeter? The equation x
2
= 4x gives x
a value of 4. The unique answer is therefore a square of side 4. What
size cube has a volume equal to its surface area? After the reader has
answered this easy question he should have no difficulty answering
two more: (1) What size hypercube has a hypervolume (measured by
unit hypercubes) equal to the volume (measured by unit cubes) of its
hypersurface? (2) What is the formula for the edge of an n-cube whose
n-volume is equal to the (n - l)-volume of its "surface"?
Hypercubes 167
Puzzle books often ask questions about cubes that are easily asked
about the tesseract but not so easily answered. Consider the longest
line that will fit inside a unit square. It is obviously the diagonal, with
a length of Yz. What is the largest square that will fit inside a unit
cube? If the reader succeeds in answering this rather tricky question,
and if he learns his way around in 4-space, he might try the more dif-
ficult problem of finding the largest cube that can be fitted into a unit
tesseract.
An interesting combinatorial problem involving the tesseract is best
approached, as usual, by first considering the analogous problems for
the square and cube. Cut open one corner of a square (see top drawing
in Figure 13.5) and its four lines can be unfolded as shown to form a
one-dimensional figure. Each line rotates around a point until all are in
the same i-space. To unfold a cube, think of it as formed of squares
joined at their edges; cut seven edges and the squares can be unfolded
(bottom drawing) until they all lie in 2-space to form a hexomino (six
unit squares joined at their edges). In this case each square rotates
around an edge. By cutting different edges one can unfold the cube to
make different hexomino shapes. Assuming that an asymmetric hex-
omino and its mirror image are the same, how many different hexomi-
noes can be formed by unfolding a cube?
The eight cubes that form the exterior surface of the tesseract can be
cut and unfolded in similar fashion. It is impossible to visualize how
a 4-space person might "see" (with three-dimensional retinas?) the hol-
low tesseract. Nevertheless, the eight cubes that bound it are true sur-
faces in the sense that the hyperperson can touch any point inside any
cube with the point of a hyperpin without the pin's passing through
any other point in the cube, just as we, with a pin, can touch any point
inside any square face of a cube without the pin's going through any
other point on that face. Points are "inside" a cube only to us. To a hy-
perperson every point in each cubical "face" of a tesseract is directly
exposed to his vision as he turns the tesseract in his hyperfingers.
Even harder to imagine is the fact that a cube in 4-space will rotate
around any of its faces. The eight cubes that bound the tesseract are
joined at their faces. Indeed, each of the 24 squares in the tesseract is a
joining spot for two cubes, as can easily be verified by studying the 3-
space models. If 17 of these 24 squares are cut, separating the pair of
cubes attached at that spot, and if these cuts are made at the right
places, the eight cubes will be free to rotate around the seven uncut
168 SOLID GEOMETRY AND HIGHER DIMENSIONS
o
A
B C 0
E
F
13.5. Unfolding a square (top) and a cube (bottom)
their faces),
(Figure 13 owned
to
;Ch.:>LJO ... JLUUW ....... polycube analogous
Observe how Dali has the contrast between 2-space and 3-
by a ..................... L"\.'u.L
cross an un-
" O T " ~ " " ' T Dali symbolizes the belief that the
death of Christ was a metahistorical event, taking place in a region
our limited use
of the "wholly other" world has long been a favorite
as P. as well as of ........ "I'J' .............
a more mundane level the unfolded
story "-And
Hypercubes 169
MS'trol)Olltan Museum of Chester 1955
11.6. 1954
170 AND
to Flatlanders as a square. There are some remarkable adventures inside
the tesseract and some unearthly views through its windows before
the house, jarred by. another earthquake, falls out of our space alto-
gether.
The notion that part of our universe might fall out of 3-space is not
so crazy as it sounds. The eminent American physicist J. A. Wheeler
has a perfectly respectable" dropout" theory to explain the enormous
energies that emanate from quasi-stellar radio sources, or quasars.
When a giant star undergoes gravitational collapse, perhaps a central
mass is formed of such incredible density that it puckers space-time
into a blister. If the curvature is great enough, the blister could pinch
together at its neck and the mass could fall out of space-time, releas-
ing energy as it vanishes.
But back to hypercubes and one final question. How many different
order-8 polycubes can be produced by unfolding a hollow hypercube
into 3-space?
Addendum
Hiram Barton, a consulting engineer of Etchingham, Sussex,
England, had the following grim comments to make about Hinton's col-
ored cubes:
Dear Mr. Gardner:
A shudder ran down my spine when I read your reference to Hinton's
cubes. I nearly got hooked on them myself in the nineteen-twenties.
Please believe me when I say that they are completely mind-destroying.
The only person I ever met who had worked with them seriously was
Francis Sedlak, a Czech neo-Hegelian philosopher (he wrote a book
called The Creation of Heaven and Earth) who lived in an Oneida-like
community near Stroud, in Gloucestershire.
As you must know, the technique consists essentially in the sequen-
tial visualizing of the adjoint internal faces of the poly-colored unit
cubes making up the large cube. It is not difficult to acquire considerable
facility in this, but the process is one of autohypnosis and, after a while,
the sequences begin to parade themselves through one's mind of their
own accord. This is pleasurable, in a way, and it was not until I went to
see Sedlak in 1929 that I realized the dangers of setting up an au-
tonomous process in one's own brain. For the record, the way out is to
establish consciously a countersystem differing from the first in that the
Hypercubes 171
core cube shows different colored faces, but withdrawal is slow and I
wouldn't recommend anyone to play around with the cubes at all.
The problem of pushing a larger cube through a hole in a smaller
cube is known as Prince Rupert's problem. The question seems to have
first been asked by Prince Rupert (1619-1682), a nephew of England's
King Charles. If you hold a cube so one corner points directly toward
you, you will see a regular hexagon. The largest square that will go
into a cube has a face that can be inscribed within this hexagon. Note
that two of the interior square's sides can be drawn on the outside of the
cube. The other two edges are within the cube.
Apparently I was the first to pose the analogous problem of deter-
mining the largest cube that would go inside a hypercube. Over the
years I received many letters from readers who claimed to have an-
swered this question. Most of the letters were too technical for me to
understand, and wildy different results were claimed.
Richard Guy and Richard Nowakowski, in their feature on unsolved
problems (American Mathematical Monthly, Vol. 104, December 1997,
pp. 967-69) report on a lengthy proof by Kay R. Pechenick DeVicci of
Moorestown, NJ, that the desired cube's edge is 1.007434775 .... It is
the square root of 1.014924 ... , the smaller root of 4.0 - 28r - 7x
2
+
16x + 16. Her paper, which includes a generalization to the largest m-
cube in an n-cube, has not been published. Guy and his two collabora-
tors on Unsolved Problems in Geometry (1991), Section B4, refer to
DeVicci's work as having solved the problem for the 3-cube in a 4-cube,
but they refer to the more general problem of n dimensions as unsolved.
When I began editing columns for this anthology I searched my files
for all the letters I had received on this problem. One writer believed
the answer was simply the unit cube of side 1, which turned out to be
quite close to the truth.
In going more carefully through my correspondence on this prob-
lem I was astonished to find that no less than four readers had obtained
the same answer as Da Vicci-a cube with an edge of 1.00743 .... I here
list them alphabetically along with the year of their letter: Hermann
Baer, Post Gilboa, Israel (1974); Eugen I. Bosch, Washington, D.C.
(1966); G. de Josselin de Jong, Delft, The Netherlands (1971); and Kay
R. Pechenick, Lafayette Hill, PA (1983).
The problem of finding all the ways a hypercube can be "unfolded"
to make distinct polycubes (unit cubes joined at their faces) was solved
172 SOLID GEOMETRY AND HIGHER DIMENSIONS
by Peter Turney in his 1984 paper (see bibliography). Using graph the-
ory, Turney found 261 unfoldings. His method extends easily to hy-
percubes of any number of dimensions.
In the Journal of Recreational Mathematics (Vol. 15, No.2. 1982-83,
p. 146) Harry Nelson showed how the 11 polyominoes that fold into a
cube would fit inside a 9 x 9 square and also inside a 7 x 11 rectangle.
It is not known if they will fit inside a smaller rectangle.
Answers
A tesseract of side x has a hypervolume of .0. The volume of its
hypersurface is 8x3. If the two magnitudes are equal, the equation gives
x a value of 8. In general an n-space "cube" with an n-volume equal to
the (n - l)-volume of its "surface" is an n-cube of side 2n.
The largest square that can be fitted inside a unit cube is the square
shown in Figure 13.7. Each corner of the square is a distance of % from
a corner of the cube. The square has an area of exactly 9/8 and a side
that is three-fourths of the square root of 2. Readers familiar with the
old problem of pushing the largest possible cube through a square hole
in a smaller cube will recognize this square as the cross section of the
limiting size of the square hole. In other words, a cube of side not quite
three-fourths of the square root of 2 can be pushed through a square
hole in a unit cube.
Figure 13.7. Packing a square in a cube
Figure 13.8 shows the 11 different hexominoes that fold into a cube.
They form a frustrating set, because they will not fit together to make
any of the rectangles that contain 66 unit squares, but perhaps there are
some interesting patterns they will form.
Hypercubes 173
,
F;.pre 13.8. The 11 hexominoes that fold into cubes
Bibliography
H.S.M. Coxeter, Regular Polytopes, Dover, 3e, 1973.
M. Gardner, "And He Built Another Crooked House," Puzzles From Other Worlds,
Vintage, 1984.
K. Heim, Christian Faith and Natural Science, Harper Torchbook, 1957.
C. H. Hinton, A New Era of Thought, Swan Sonnenschein, 1888.
C. H. Hinton, The Fourth Dimension, Allen & Unwin, 1904.
R. A. Jacobson, "Paths of Minimal Length Within Hypercubes," The American
Mathematical Monthly, Vol. 73, October 1966, pp. 868-72.
H. P. Manning, Geometry of Four Dimensions, Dover, 1956.
H. P. Manning, The Fourth Dimension Simply Explained, Dover, 1960.
E. H. Neville, The Fourth Dimension, Cambridge University Press, 1921.
R. Rucker (ed), Speculations on the Fourth Dimension: Selected Writings of C. H.
Hinton, Dover, 1980.
R. Rucker, Infinity and the Mind, Birkiiuser, 1982.
R. Rucker, The Fourth Dimension, Houghton Mifflin, 1984.
D.M.Y. Sommerville, An Introduction to the Geometry of N Dimensions, Dover,
1958.
W. S. Tevis, Jr., "The Ifth of Oofth," Galaxy Science Fiction, April, 1957, pp. 59-69.
A wild and funny tale about a cube that distorts space and
time.
P. Thmey, "Unfolding the Tesseract," Journal of Recreational Mathematics, Vol. 17,
No.1, 1984-85, pp. 1-16.
174 SOLID GEOMETRY AND HIGHER DIMENSIONS
Chapter 14
"Lines that are parallel
meet at Infinity!"
Euclid repeatedly,
heatedly,
urged.
Until he died,
and so reached that vicinity:
in it he
found that the damned things
diverged.
-Piet Hein, Grooks VI
Non-Euclidean
Qeometry
Euclid's Elements is dull, long-winded, and does not make ex-
plicit the fact that two circles can intersect, that a circle has an outside
and an inside, that triangles can be turned over, and other assumptions
essential to his system. By modern standards Bertrand Russell could
call Euclid's fourth proposition a "tissue of nonsense" and declare it a
scandal that the Elements was still used as a textbook.
On the other hand, Euclid's geometry was the first major effort to or-
ganize the subject as an axiomatic system, and it seems hardly fair to
find fault with him for not anticipating all the repairs made when
David Hilbert and others formalized the system. There is no more strik-
ing evidence of Euclid's genius than his realization that his notorious
fifth postulate was not a theorem but an axiom that had to be accepted
without proof.
Euclid's way of stating the postulate was rather cumbersome, and it
was recognized early that it could be given the following simpler form:
Through a point on a plane, not on a given straight line, only one line
is parallel to the given line. Because this is not quite as intuitively ob-
vious as Euclid's other axioms mathematicians tried for 2,000 years to
remove the postulate by making it a theorem that could be established
on the basis of Euclid's other axioms. Hundreds of proofs were at-
tempted. Some eminent mathematicians thought they had succeeded,
but it always turned out that somewhere in their proof an assumption
had been made that either was equivalent to the parallel postulate or re-
quired the postulate.
175
For example, it is easy to prove the parallel postulate if you assume
that the sum of the angles of every triangle equals two right angles. Un-
fortunately you cannot prove this assumption without using the paral-
lel postulate. An early false proof, attributed to Thales of Miletus, rests
on the existence of a rectangle, that is, a quadrilateral with four right an-
gles. You cannot prove, however, that rectangles exist without using the
parallel postulate! In the 17th century John Wallis, a renowned English
mathematician, believed he had proved the postulate. Alas, he failed to
realize that his assumption that two triangles can be similar but not
congruent cannot be proved without the parallel postulate. Long lists
can be made of other assumptions, all so intuitively obvious that they
hardly seem worth asserting, and all equivalent to the parallel postu-
late in the sense that they do not hold unless the postulate holds.
In the early 19th century trying to prove the postulate became some-
thing of a mania. In Hungary, Farkas Bolyai spent much of his life at the
task, and in his youth he discussed it often with his German friend
Karl Friedrich Gauss. Farkas' son Janos became so obsessed by the
problem that his father was moved to write in a letter: "For God's sake,
I beseech you, give it up. Fear it no less than sensual passions because
it too may take all your time and deprive you of your health, peace of
mind and happiness in life."
Janos did not give it up, and soon he became persuaded not only
that the postulate was independent of the other axioms but also that a
consistent geometry could be created by assuming that through the
point an infinity of lines were parallel to the given line. "Out of noth-
ing I have created a new universe," he proudly wrote to his father in
1823.
Farkas at once urged his son to let him publish these sensational
claims in an appendix to a book he was then completing. "If you have
really succeeded, it is right that no time be lost in making it public, for
two reasons: first, because ideas pass easily from one to another who
can anticipate its publication; and secondly, there is some truth in this,
that many things have an epoch in which they are found at the same
time in several places, just as the violets appear on every side in spring.
Also every scientific struggle is just a serious war, in which I cannot say
when peace will arrive. Thus we ought to conquer when we are able,
since the advantage is always to the first comer."
Janos' brief masterpiece did appear in his father's book, but as it hap-
pened the publication of the book was delayed until 1832. The Russ-
176 SOLID GEOMETRY AND HIGHER DIMENSIONS
ian mathematician Nikolai Ivanovitch Lobachevski had beat him to it
by disclosing details of the same strange geometry (later called by Felix
Klein hyperbolic geometry) in a paper of 1829. What is worse, when
Farkas sent the appendix to his old friend Gauss, the Prince of Mathe-
maticians replied that if he praised the work, he would only be prais-
ing himself, inasmuch as he had worked it all out many years earlier
but had published nothing. In other letters he gave his reason. He did
not want to arouse an Houtcry" among the "Boeotians," by which he
meant his conservative colleagues. (In ancient Athens the Boeotians
were considered unusually stupid.)
Crushed by Gauss's response, Janos even suspected that his father
might have leaked his marvelous discovery to Gauss. When he later
learned of Lobachevski's earlier paper, he lost interest in the topic and
published nothing more. "The nature of real truth of course cannot but
be one and the same in Marcos-Vasarhely as in Kamchatka and on the
moon," he wrote, resigned to having published too late to win the
honor for which he had so passionately hoped.
In some ways the story of the Italian Jesuit Giralamo Saccheri is even
sadder than that of Bolyai. As early as 1733, in a Latin book called Eu-
clid Cleared of All Blemish, Saccheri actually constructed both types of
non-Euclidean geometry (we shall come to the second type below)
without knowing it! Or so it seems. At any rate Saccheri refused to be-
lieve either geometry was consistent, but he came so close to accepting
them that some historians think he pretended to disbelieve them just
to get his book published. "To have claimed that a non-Euclidean sys-
tem was as 'true' as Euclid's," writes Eric Temple Bell (in a chapter on
Saccheri in The Magic of Numbers), "would have been a foolhardy in-
vitation to repression and discipline. The Copernicus of Geometry
therefore resorted to subterfuge. Taking a long chance, Saccheri de-
nounced his own work, hoping by this pious betrayal to slip his heresy
past the censors."
I cannot resist adding two anecdotes about the Bolyais. Janos was a
cavalry officer (mathematics had always been strictly a recreation)
known for his swordsmanship, his skill on the violin, and his hot tem-
per. He is said to have once challenged 13 officers to duels, provided
that after each victory he would be allowed to play to the loser a piece
on his violin. The elder Bolyai is reported to have been buried at his
own request under an apple tree, with no monument, to commemorate
history's three most famous apples: the apple of Eve, the golden apple
Non-Euclidean Geometry 177
Paris gave Venus as a beauty-contest prize, and the falling apple that in-
spired Isaac Newton.
Before the 19th century had ended it became clear that the parallel
postulate not only was independent of the others but also that it could
be altered in two opposite ways. If it was replaced (as Gauss, Bolyai,
and Lobachevski had proposed) by assuming an infinite number of "ul-
traparallel" lines through the point, the result would be a new geome-
try just as elegant and as "true" as Euclid's. All Euclid's other postulates
remain valid; a "straight" line is still a geodesic, or shortest line. In this
hyperbolic space all triangles have an angle sum less than 180 degrees,
and the sum decreases as triangles get larger. All similar polygons are
congruent. The circumference of any circle is greater than pi times the
diameter. The measure of curvature of the hyperbolic plane is negative
(in contrast to the zero curvature of the Euclidean plane) and every-
where the same. Like Euclidean geometry, hyperbolic geometry gener-
alizes to 3-space and all higher dimensions.
The second type of non-Euclidean geometry, which Klein names "el-
liptic," was later developed simultaneously by the German mathe-
matician Georg Friedrich Bernhard Riemann and the Swiss
mathematician Ludwig SchHifli. It replaces the parallel postulate with
the assumption that through the point no line can be drawn parallel to
the given line. In this geometry the angle sum of a triangle is always
more than 180 degrees, and the circumference of a circle is always less
than pi times the diameter. Every geodesic is finite and closed. The
lines in every pair of geodesics cross.
To prove consistency for the two new geometries various Euclidean
models of each geometry were found showing that if Euclidean geom-
etry is consistent, so are the other two. Moreover, Euclidean geometry
has been "arithmetized," proving that if arithmetic is consistent, so too
is Euclid's geometry. We now know, thanks to Kurt G6del, that the con-
sistency of arithmetic is not provable in arithmetic, and although there
are consistency proofs for arithmetic (such as the famous proof by Ger-
hard Gentzen in 1936), no such proof has yet been found that can be
considered entirely constructive by an intuitionist (see A. Calder, "Con-
structive Mathematics," Scientific American, October 1979). God ex-
ists, someone once said, because mathematics is consistent, and the
Devil exists because we are not able to prove it.
The various metaproofs of arithmetic's consistency, as Paul C. Rosen-
bloom has put it, may not have eliminated the Devil, but they have re-
178 SOLID GEOMETRY AND HIGHER DIMENSIONS
179
though they are unaware of any change because all their measuring in-
struments Similarly get smaller. At the boundary their size would be-
come zero, but they can never reach the boundary. If they proceed
toward it with uniform velocity, their speed (to us) steadily decreases,
although to them it seems constant. Thus their universe, which we see
as being finite, is to them infinite. Hyperbolic light follows geodesics,
but because its velocity is proportional to its distance from the bound-
ary it takes paths that we see as circular arcs meeting the boundary at
right angles.
In this hyperbolic world a triangle has a maximum finite area, as is
shown in Figure 14.2, although its three "straight" sides go to infinity
in hyperbolic length and its three angles are zero. You must not think
of Escher's mosaic as being laid out on a sphere. It is a circle enclosing
an infinity of fish-Coxeter calls it a "miraculous draught"-that get
progressively smaller as they near the circumference. In the hyperbolic
plane, of which the picture is only a model, the fish are all identical in
size and shape. It is important to remember that the creatures of a hy-
perbolic world would not change in shape as they moved about, light
would not change in speed, and the universe would be infinite in all
directions.
figure 14.2.
The curved white lines in Escher's woodcut do not, as many people
have supposed, model hyperbolic geodesics. The lines are called
equidistant curves or hypercycles. Each line has a constant perpendic-
ular distance (measured hyperbolically) from the hyperbolic straight
line that joins the arc's ends. Note that along each white curve fish of
the same color swim head to tail. If you consider all the points where
180 SOLID GEOMETRY AND HIGHER DIMENSIONS
four fins meet, these points are the vertexes of a regular tiling of the hy-
perbolic plane by equilateral triangles with angles of 45 degrees. The
centers of the triangles are the points where three left fins meet and
three mouths touch three tails. The 45-degree angles make it possible
for eight triangles to surround each vertex, where in a Euclidean tiling
by equilateral triangles only six triangles can surround each vertex.
Escher and Coxeter had corresponded from the time they met in
1954, and Escher's interest in tilings of the hyperbolic plane had been
aroused by the illustrations in a 1957 paper on crystal symmetry that
Coxeter had written and sent to him. In a lovely article titled "The
Non-Euclidean Symmetry of Escher's Picture 'Circle Limit III' " (in the
journal Leonardo, Vol. 12, 1979, pp. 19-25) Coxeter shows that each
white arc meets the boundary at an angle of almost 80 degrees. (The
precise value is 27/4 + 25/4 arc secants.) Coxeter considers Circle Limit
III the most mathematically sophisticated of all Escher's pictures. It
even anticipated a discovery Coxeter did not make until five years after
the woodcut was finished!
Elliptic geometry is roughly modeled by the surface of a sphere. Here
Euclidean straight lines become great circles. Clearly no two can be
parallel, and it is easy to see that triangles formed by arcs of great cir-
cles must have angles that add up to more than two right angles. The
hyperbolic plane is similarly modeled by the saddle-shaped surface of
a pseudosphere, generated by rotating a tractrix about its asymptote.
It is a misuse of the word "crank" to apply it to mathematicians who
erred in thinking, before the independence of the parallel postulate
was established, that they had proved the postulate. The same cannot
be said of those amateurs of later decades who could not understand
the proofs of the postulate's independence or who were too egotistical
to try. Augustus De Morgan, in his classic compendium of eccentric
mathematics, A Budget of Paradoxes, introduces us to Britain's most in-
defatigable 19th-century parallel-postulate prover, General Perronet
Thompson. Thompson kept issuing revisions of his many proofs (one
was based on the equiangular spiral), and although De Morgan did his
best to dissuade him from his futile efforts, he was unsuccessful.
Thompson also wanted to replace the tempered scale of the piano with
an octave of 40 notes.
The funniest of the American parallel-postulate provers was the Very
Reverend Jeremiah Joseph Callahan, then president of Duquesne Uni-
versity in Pittsburgh. In 1931, when Father Callahan announced he
Non-Euclidean Geometry 181
had trisected the angle, Time gave the story sober treatment and ran his
photograph. The following year Callahan published his major work,
Euclid or Einstein: A Proof of the Parallel Theory and a Critique of
MetageometIY(Devon-Adair, 1932), a 31D-page treatise in which he as-
cended to heights of argumentem ad hominem. Einstein is "fuddled,"
he "has not a logical mind," he is in a "mental fog," he is a "careless
thinker." "His thought staggers, and reels, and stumbles, and falls, like
a blind man rushing into unknown territory." "Sometimes one feels
like laughing," Callahan wrote, "and sometimes one feels a little irri-
tated .... But there is no use expecting Einstein to reason."
What Callahan found so irritating was Einstein's adoption of a gen-
eralized non-Euclidean geometry, formulated by Riemann, in which
the curvature of physical space varies from point to point depending on
the influence of matter. One of the great revolutions brought about by
relativity theory was the discovery that an enormous overall simplifi-
cation of physics is obtained by assuming physical space to have this
kind of non-Euclidean structure.
It is now commonplace (how astonished, and I think delighted, Kant
would have been by the notion!) to recognize that all geometric systems
are equally "true" in the abstract but that the structure of physical space
must be determined empirically. Gauss himself thought of triangulat-
ing three mountain peaks to see if their angles added up to two right an-
gles. It is said he actually made such a test, with inconclusive results.
Although experiments can prove physical space is non-Euclidean, it is
a curious fact that there is no way to prove it is Euclidean! Zero curva-
ture is a limiting case, midway between elliptic and hyperbolic curva-
tures. Since all measurement is subject to error, the deviation from zero
could always be too slight for detection.
Poincare held the opinion that if optical experiments seemed to show
physical space was non-Euclidean, it would be best to preserve the
simpler Euclidean geometry of space and assume that light rays do not
follow geodesics. Many mathematicians and physicists, including Rus-
sell, agreed with Poincare until relativity theory changed their mind.
Alfred North Whitehead was among the few whose mind was never
changed. He even wrote a book on relativity, now forgotten, in which
he argued for preserving a Euclidean universe (or at least one of con-
stant curvature) and modifying the physical laws as necessary. (For a
discussion of Whitehead's controversy with Einstein, see Robert M.
182 SOLID GEOMETRY AND HIGHER DIMENSIONS
Palter's Whitehead's Philosophy of Science, University of Chicago
Press, 1960.)
Physicists are no longer disturbed by the notion that physical space
has a generalized non-Euclidean structure. Callahan was not merely
disturbed; he was also convinced that all non-Euclidean geometries
are self-contradictory. Einstein, poor fellow, did not know how easy it
is to prove the parallel postulate. If you are curious about how Callahan
did it, and about his elementary error, see D. R. Ward's paper in The
Mathematical Gazette (Vol. 17, May 1933, pp. 101-4).
Like their cousins who trisect the angle, square the circle, and find
simple proofs of Fermat's last theorem, the parallel-postulate provers
are a determined breed. A more recent example is William 1. Fischer
of Munich, who in 1959 published a 100-page Critique of Non-
Euclidean Geometry. Ian Stewart exposed its errors in the British jour-
nal Manifold (No. 12, Summer 1972, pp. 14-21). Stewart quotes from
a letter in which Fischer accuses establishment mathematicians of sup-
pressing his great work and orthodox journals of refusing to review it:
"The university library at Cambridge refused even to put my booklet on
file .... I had to write to the vice-chancellor to overcome this boycott."
There are, of course, no sharp criteria for distinguishing crank math-
ematics from good mathematics, but then neither are there sharp crite-
ria for distinguishing day from night, life from non-life, and where the
ocean ends and the shore begins. Without words for parts of continu-
ums we could not think or talk at all. If you, dear reader, have a way to
prove the parallel postulate, don't tell me about it!
Addendum
Imagine a small circle around the north pole of the earth. If it
keeps expanding, it reaches a maximum size at the equator, after which
it starts to contract until it finally becomes a point at the south pole. In
similar fashion, an expanding sphere in four-dimensional elliptical
space reaches a maximum size, then contracts to a point.
In addition to the three geometries described in this chapter, there is
what Bolyai called "absolute geometry" in which theorems are true in
all three. It is astonishing that the first 28 theorems of Euclid's Ele-
ments are in this category, along with other novel theorems that Bolyai
showed to be independent of the parallel postulate.
Non-Euclidean Geometry 183
I was surprised to see in a 1984 issue of Speculations in Science and
Technology (VoL 7, pp. 207-16), a defense of Father Callahan's proof of
the parallel postulate! The authors are Richard Hazelett, vice president
of the Hazelett Strip-Casting Corporation, Colchester, VT, and Dean E.
Thrner, who teaches at the University of North Colorado, in Greeley.
Hazelett is a mechanical engineer with master's degrees from the Uni-
versity of Texas and Boston University. Taylor, an ordained minister in
the Disciples of Christ Church, has a doctorate from the University of
Texas.
It is easy to understand why both men do not accept Einstein's gen-
eral theory of relativity. Indeed, they have edited a book of papers at-
tacking Einstein. Titled The Einstein Myth and the Ives Papers, it was
published in 1979 by Devin-Adair.
In an earlier column on geometrical fallacies, reprinted in Wheels,
Life, and Other Mathematical Amusements (1983), I discussed in detail
Father Callihan's false "proof" of the parallel postulate. For other false
proofs, see Underwood Dudley's book cited in the bibliography.
In the Encyclopaedia Britannica's tenth edition Bertrand Russell con-
tributed an article on "Geometry, Non-Euclidean." It was revised for the
eleventh edition (1910) with Whitehead's name added as coauthor.
They argued that at present, on grounds of simplicity, if an experiment
ever contradicted Euclidean geometry it would be preferable to ques-
tion the experiment rather than give up Euclidean geometry. Of course
they wrote this before relativity theory proved otherwise. Poincare's ob-
jection to non-Euclidean geometry was unqualified. Russell changed
his mind in the light of relativity theory, but if Whitehead ever changed
his mind I've been unable to find evidence for it.
Bibliography
R. BonoIa, Non-Euclidean Geometry, Open Court, 1912; Dover, 1955.
E. Breitenberger, "Gauss's Geodesy and the Axiom of Parallels," Archive for History
of Exact Sciences, Vol. 31, 1984, pp. 273-89.
H.S.M. Coxeter, Regular Compound Tessellations of the Hyperbolic Plane, Pro-
ceedings of the Royal Society, A, VoL 278, 1964, pp. 147-67.
H.S.M. Coxeter, Non-Euclidean Geometry, 5e, University of Toronto Press, 1965.
H. S.M. Coxeter, The Non-Euclidean Symmetry of Escher's Picture 'Circle Limit III. '
Leonardo, Vol. 12, 1979, pp. 19-25.
U. Dudley, Euclid's Fifth Postulate, Mathematical Cranks, Mathematical Associa-
tion of America, 1992, pp. 137-58.
184 SOLID GEOMETRY AND HIGHER DIMENSIONS
S. Gindikin, The Wonderland of Poincaria, Quantum, November/December 1992,
pp.21-28.
S. H. Gould, The Origin of Euclid's Axioms, Mathematical Gazette, Vol. 46, De-
cember, 1962, pp. 269-90.
M. J. Greenberg, Euclidean and Non-Euclidean Geometries: Development and His-
tory, 3e, W. H. Freeman, 1994.
S. Kulczycki, Non-Euclidean Geometry, Macmillan, 1961.
W. W. Maiers, Introduction to Non-Euclidean Geometry, Mathematics Teacher, No-
vember 1964, pp. 457-61.
R. M. Palter, Whitehead's Philosophy of Science, University of Chicago Press, 1960.
Reviewed by Adolf Griinbaum in The Philosophy Review, Vol. 71, April 1962.
D. Pedoe, Non-Euclidean Geometry, New Scientist, No. 219, January 26,1981, pp.
206-7.
J. F. Rigby, Some Geometrical Aspects of a Maximal Three-Coloured Triangle-Free
Graph, Journal of Combinatorial Theory, Series B, Vol. 34, June 1983, pp.
313-22.
D.M.Y. Sommerville, The Elements of Non-Euclidean Geometry, Dover, 1958.
J. W. Withers, Euclid's Parallel Postulate: Its Nature, Validity, and Place in Geo-
metrical Systems, Open Court, 1905.
H. E. Wolfe, Non-Euclidean Geometry, Henry Holt, 1945.
Non-Euclidean Geometry 185
N
Symmetry
Chapter 15 Rotations
and Reflections
A geometric figure is said to be symmetrical if it remains un-
changed after a "symmetry operation" has been performed on it. The
larger the number of such operations, the richer the symmetry. For ex-
ample, the capital letter A is unchanged when reflected in a mirror
placed vertically beside it. It is said to have vertical symmetry. The
capital B lacks this symmetry but has horizontal symmetry: it is un-
changed in a mirror held horizontally above or below it. S is neither
horizontally nor vertically symmetrical but remains the same if rotated
180 degrees (twofold symmetry). All three of these symmetries are pos-
sessed by H, I, 0, and X. X is richer in symmetry than H or I because,
if its arms cross at right angles, it is also unchanged by quarter-turns
(fourfold symmetry). 0, in circular form, is the richest letter of all. It is
unchanged by any type of rotation or reflection.
Because the earth is a sphere toward the center of which all objects
are drawn by gravity, living forms have found it efficient to evolve
shapes that possess strong vertical symmetry combined with an obvi-
ous lack of horizontal or rotational symmetry. In making objects for his
use man has followed a similar pattern. Look around and you will be
struck by the number of things you see that are essentially unchanged
in a vertical mirror: chairs, tables, lamps, dishes, automobiles, air-
planes, office buildings-the list is endless. It is this prevalence of ver-
tical symmetry that makes it so difficult to tell when a photograph has
been reversed, unless the scene is familiar or contains such obvious
clues as reversed printing or cars driving on the wrong side of the road.
On the other hand, an upside-down photograph of almost anything is
instantly recognizable as inverted.
The same is true of works of graphic art. They lose little, if anything,
by reflection, but unless they are completely nonrepresentational no
189
careless museum director is likely to hang one upside down. Of course,
abstract paintings are often inverted by accident. The New York Times
Magazine (October 5, 1958) inadvertently both reversed and inverted a
picture of an abstraction by Piet Mondrian, but only readers who knew
the painting could possibly have noticed it. In 1961, at the New York
Museum of Modern Art, Matisse's painting, Le Bateau, hung upside
down for 47 days before anyone noticed the error.
So accustomed are we to vertical symmetry, so unaccustomed to see-
ing things upside down, that it is extremely difficult to imagine what
most scenes, pictures, or objects would look like inverted. Landscape
artists have been known to check the colors of a scene by the undigni-
fied technique of bending over and viewing the landscape through their
legs. Its upside-down contours are so unfamiliar that colors can be seen
uncontaminated, so to speak, by association with familiar shapes.
Thoreau liked to view scenes this way and refers to such a view of a
pond in Chapter 9 of Walden. Many philosophers and writers have
found symbolic meaning in this vision of a topsy-turvy landscape; it
was one of the favorite themes of G. K. Chesterton. His best mystery sto-
ries (in my opinion) concern the poet-artist Gabriel Gale (in The Poet
and the Lunatics), who periodically stands on his hands so that he can
"see the landscape as it really is: with the stars like flowers, and the
clouds like hills, and all men hanging on the mercy of God. "
The mind's inability to imagine things upside down is essential to
the surprise produced by those ingenious pictures that turn into some-
thing entirely different when rotated 180 degrees. Nineteenth-century
political cartoonists were fond of this device. When a reader inverted
a drawing of a famous public figure, he would see a pig or jackass or
something equally insulting. The device is less popular today, although
Life for September 18, 1950, reproduced a remarkable Italian poster on
which the face of Garibaldi became the face of Stalin when viewed up-
side down. Children's magazines sometimes reproduce such upside-
down pictures, and now and then they are used as advertising
gimmicks. The back cover of Life for November 23,1953, depicted an
Indian brave inspecting a stalk of corn. Thousands of readers probably
failed to notice that when this picture was inverted it became the face
of a man, his mouth watering at the sight of an open can of corn.
I know of only four books that are collections of upside-down draw-
ings. Peter Newell, a popular illustrator of children's books who died
in 1924, published two books of color plates of scenes that undergo
190 SYMMETRY
amusing transformations when inverted:
Topsys and Turvys Number 2 (1894). In 1946 a London publisher is-
a faces drawn by Rex
an ............... " ...... 0..;, ......... in 1
richly symmetrical title
15.1.}
(Its title page reproduced in
on page
invertible book
The technique upside-down drawing was carried to unbelievable
heights a
he a six-panel color cornie for the Sunday of
One took the panels in order, reading the cap-
one the page upside down and
reverse to
achieve continuity by means of two chief characters called Little Lady
Old Man
o ; ; ; . . , - , ~ ........... mad surpasses all understanding. A collection of 25 of
his comics was published by G. W. Dillingham in 1905 under the
of
cause it
a Au..uL '-AU 'U U. "."..1
and Reflections 191
Figure 1S.2. A typical upside-down cartoon by Gustave Verbeek
century Swiss painter Matthaus Merian that becomes a man's profile
when the picture is given a quarter-turn counterclockwise. The rab-
bit-duck in Figure 15.3 is the best-known example of a quarter-turn
picture. Psychologists have long used it for various sorts of testing.
Harvard philosopher Morton White once reproduced a rabbit-duck
drawing in a magazine article to symbolize the fact that two historians
can survey the same set of historical facts but see them in two essen-
tially different ways.
Our lifelong conditioning in the way we see things is responsible for
a variety of startling upside-down optical illusions. All astronomers
know the necessity of viewing photographs of the moon's surface so
that sunlight appears to illuminate the craters from above rather than
below. We are so unaccustomed to seeing things illuminated from
below that when such a photograph of the moon is inverted, the craters
instantly appear to be circular mesas rising above the surface. One of
the most amusing illusions of this same general type is shown in Fig-
ure 15.4. The missing slice of pie is found by turning the picture upside
down. Here again the explanation surely lies in the fact that we almost
always see plates and pies from above and almost never from below.
Upside-down faces could not be designed, of course, if it were not for
192 SYMMETRY
Figure 15.3. A quarter-turn clockwise
makes the duck into a rabbit.
Figure 15.4. Where is the missing slice?
the fact that our eyes are not too far from midway between the top of
the head and the chin. School children often amuse themselves by
turning a history book upside down and penciling a nose and mouth on
the forehead of some famous person.
When this is done on an actual face, using eyebrow pencil and lip-
stick, the effect becomes even more grotesque. It was a popular party
pastime of the late 19th century. The following account is from an old
book entitled What Shall We Do Tonight?
The severed head always causes a sensation and should not be suddenly
exposed to the nervous .... A large table, covered with a cloth suffi-
ciently long to reach to the floor all around and completely hide all be-
neath, is placed in the center of the room .... A boy with soft silky hair,
rather long, being selected to represent the head, must lie upon his back
under the table entirely concealed, excepting that portion of his face
above the bridge of his nose. The rest is under the tablecloth.
His hair must now be carefully combed down, to represent whiskers,
and a face must be painted ... upon the cheeks and forehead; the false
eyebrows, nose and mouth, with mustache, must be strongly marked
with black water color, or India ink, and the real eyebrows covered with
a little powder or flour. The face should also be powdered to a deathlike
pallor ....
The horror of this illusion may be intensified by having a subdued
light in the room in which the exhibition has been arranged. This
conceals in a great degree any slight defects in the "making-up" of the
head ....
Needless to add, the horror is heightened when the "head" suddenly
opens its eyes, blinks, stares from side to side, wrinkles its cheeks (fore-
head).
The physicist Robert W. Wood (author of How to Tell the Birds from
the Flowers) invented a funny variation of the severed head. The face
is viewed upside down as before, but now it is the forehead, eyes, and
Rotations and Reflections 193
nose that are covered, leaving only the mouth and chin exposed. Eyes
and nose are drawn on the chin to produce a weird little pinheaded
creature with a huge, flexible mouth. The stunt was a favorite of Paul
Winchell, the television ventriloquist. He wore a small dummy's body
on his head to make a figure that he called Ozwald, while television
camera techniques inverted the screen to bring Ozwald right side up.
In 1961 an Ozwald kit was marketed for children, complete with the
dummy's body and a special mirror with which to view one's own face
upside down.
It is possible to print or even write in longhand certain words in
such a way that they possess twofold symmetry. The Zoological Soci-
ety of San Diego, for instance, publishes a magazine called ZOONOOZ,
the name of which is the same upside down. The longest sentence of
this type that I have come across is said to be a sign by a swimming pool
designed to read the same when viewed by athletes practicing hand-
stands: NOW NO SWIMS ON MON. (See Figure 15.5.)
----
. " . ~ . " .. -
I' f ~
, I'/., ...r-
1 \ ~ __
I.... -
sketch reproduced by courtesy of the artist, John McClellan
Figure 15.5. An invertible sign
It is easy to form numbers that are the same upside down. As many
have noticed, 1961 is such a number. It was the first year with twofold
symmetry since 1881, the last until 6009, and the twenty-third since the
year 1. Altogether there are 38 such years between A.D. 1 and A.D. 10000
(according to a calculation made by John Pomeroy), with the longest in-
terval between 1961 and 6009. J. F. Bowers, writing in the Mathemati-
194
SYMMETRY
cal Gazette for December 1961, explains his clever method of calculat-
ing that by A.D. 1000000 exactly 198 invertible years will have passed.
The January 1961 issue of Mad featured an upside-down cover with the
year's numerals in the center and a line predicting that the year would
be a mad one.
Some numbers, for example 7734 (when the 4 is written so that it is
open at the top), become words when inverted; others can be written
to become words when reflected. With these quaint possibilities in
mind, the reader may enjoy tackling the following easy problems:
1. Oliver Lee, age 44, who lives at 312 Main Street, asked the city to
give his car a license plate bearing the number 337-31770. Why?
2. Prove the sum in Figure 15.6 to be correct.
3 4 I 4 Figure 15.6. Is the sum correct?
340
74813
43374813
3. Circle six digits in the group below that will add up to exactly 21.
1
3
5
9
1
3
5
9
1
3
5
9
4. A basket contains more than half a dozen eggs. Each egg is either
white or brown. Let x be the number of white eggs, and y be the num-
ber of brown. The sum of x and y, turned upside down, is the product
of x and y. How many eggs are in the basket?
Addendum
George Carlson, art editor of John Martin's Book, a monthly
magazine for children that flourished in the 1920s, contributed some
dozen upside-down pictures to the magazine. Many other examples
can be found in several books on optical illusions and related pictures
by England's Keith Kay. In 1980 Lothrop published The Turn About,
Think About, Look About Book, by Beau Gardner. It features pictures
to be viewed both upside down and after quarter-turns.
Rotations and Reflections 195
90 to lanlosca}:)e 90
to
196
three swans on a lake. Their reflections in the water are three elephants.
The landscape is reproduced in color in Richard Gregory's book on
mirrors, Mirrors in Mind (1997).
Landscapes that turn into faces when rotated 90 degrees were popu-
lar among German painters of the Renaissance. Two examples are re-
produced in Figure 15.7.
Answers
1. The number 337-31770 upside down spells "Ollie Lee."
2. Hold the sum to a mirror.
3. Turn the picture upside down, circle three 6's and three 1 's to
make a total of 21.
4. The basket has nine white eggs and nine brown eggs. When the
sum, 18, is inverted, it becomes 81, the product. Had it not been spec-
ified that the basket contained more than six eggs, three white and
three brown would have been another answer.
Rotations and Reflections 197
Chapter 16 The Amazing
Creations of Scott Kim
Scott Kim's Inversions, published in 1981 by Byte Publications,
is one of the most astonishing and delightful books ever printed. Over
the years Kim has developed the magical ability to take just about any
word or short phrase and letter it in such a way that it exhibits some
kind of striking geometrical symmetry. Consider Kim's lettering of my
name in Figure 16.1. Turn it upside down and presto! It remains exactly
the same!
Students of curious wordplay have long recognized that short words
can be formed to display various types of geometrical symmetry. On the
Rue Mozart in Paris a clothing shop called "New Man" has a large sign
lettered "NeW MaN" with the e and the a identical except for their ori-
entation. As a result the entire sign has upside-down symmetry. The
names VISTA (the magazine of the United Nations Association).
ZOONOOZ (the magazine of the San Diego zoo) and NISSIN (a Japan-
ese manufacturer of camera flash equipment) are all cleverly designed
so that they have upside-down symmetry.
BOO HOO, DIOXIDE, EXCEEDED, and DICK COHEN DIED 10 DEC
1883 all have mirror symmetry about a horizontal axis. If you hold
them upside down in front of a mirror, they appear unchanged. One
day in a supermarket my sister was puzzled by the name on a box of
crackers, "spep oop," until she realized that a box of "doo dads" was on
the shelf upside down. Wallace Lee, a magician in North Carolina, liked
198
to amuse friends by asking if they had ever eaten any "ittaybeds," a
word he printed on a piece of paper like this:
1 ...... aLlbeds
After everyone said no, he would add:
"Of course, they taste much better upside down."
Many short words in conventional typefaces turn into other words
when they are inverted. MOM turns into WOW and "up" becomes the
abbreviation "dn." SWIMS remains the same. Other words have mirror
symmetry about a vertical axis, such as "bid" (and "pig" if the g is
drawn as a mirror image of the p). Here is an amusing way to write
"minimum" so that it is the same when it is rotated 180 degrees:
o 0
o 0
It is Kim who has carried this curious art of symmetrical calligraphy
to heights not previously known to be possible. By ingeniously dis-
torting letters, yet never so violently that one cannot recognize a word
or phrase, Kim has produced incredibly fantastic patterns. His book is
a collection of such wonders, interspersed with provocative observa-
tions on the nature of symmetry, its philosophical aspects, and its em-
bodiment in art and music as well as in wordplay.
Kim is no stranger to my Scientific American columns. He is of Ko-
rean descent, born in the U.S., who in 1981 was doing graduate work
in computer science at Stanford University. He was in his teens when
he began to create highly original problems in recreational mathemat-
ics. Some that have been published in Scientific American include his
"lost-king tours" (April 1977), the problem of placing chess knights on
the corners of a hypercube (February 1978), his solution to "boxing a
box" (February 1979), and his beautifully symmetrical "m-pire map"
given in Chapter 6 of my Last Recreations. In addition to a remarkable
ability to think geometrically (not only in two and three dimensions but
also in 4-space and higher spaces) Kim is a classical pianist who for
years could not decide between pursuing studies in mathematics or in
music. In the early 1980s he was intensely interested in the use of com-
The Amazing Creations of Scott Kim 199
puters for designing typefaces, a field pioneered by his friend and men-
tor at Stanford, the computer scientist Donald E. Knuth.
For several years Kim's talent for lettering words to give them unex-
pected symmetries was confined to amusing friends and designing fam-
ily Christmas cards. He would meet a stranger at a party, learn his or her
name, then vanish for a little while and return with the name neatly
drawn so that it would be the same upside down. His 1977 Christmas
card, with upside-down symmetry, is shown in Figure 16.2. (Lester
and Pearl are his father and mother; Grant and Gail are his brother and
sister.) The following year he found a way to make "Merry Christmas,
1978," mirror-symmetrical about a horizontal axis, and in 1979 he
made the mirror axis vertical. (See Figures 16.3 and 16.4.)
Figure '6.2.
For a wedding anniversary of his parents Kim designed a cake with
chocolate and vanilla frosting in the pattern shown in Figure 16.5.
("Lester" is in black, "Pearl" is upside down in white.) This is Kim's
"figure and ground" technique. You will find another example of it in
Godel, Escher, Bach: An Eternal Golden Braid, the Pulitzer-prize-
winning book by Kim's good friend Douglas R. Hofstadter. Speaking of
200 SYMMETRY
Figure 16.3.
Figure 16.S
Figure 16.6.
o o
Kurt Gade!, J. S. Bach, and M. C. Escher, Figure 16.6 shows how Kim
has given each name a lovely mirror symmetry. In Figure 16.7 Kim has
lettered the entire alphabet in such a way that the total pattern has
left-right symmetry.
Kim's magic calligraphy came to the attention of Scot Morris, an ed-
itor at Omni. Morris devoted a page of his popular column on games to
Kim's work in Omni's September 1979 issue, and he announced a
reader's contest for similar patterns. Kim was hired to judge the thou-
sands of entries that came in. You will find the beautiful prizewinners
in Omni's April 1980 issue and close runners-up in Morris' columns for
May and November of the same year.
All the patterns in Kim's book are his own. A small selection of a few
The Amazing Creations of Scott Kim 201
}l3CDEf\
GltB
1 M r
pq
mr
1JJW
r_e'6.7. ~
Figure 16.8.
lIfJlQi
202
m!<bE _Dltdll

(ommu N+N
nimtia) V16IY1ON
infinityinfinUy
infinityinfinUy
infinityinfinUy
SYMMETRY
more is given in Figure 16.8 to convey some notion of the amazing va-
riety of visual tricks Kim has up his sleeve.
I turn now to two unusual mathematical problems originated by Kim,
both of which are still only partly solved. In 1975, when Kim was in
high school, he thought of the following generalization of the old prob-
lem of placing eight queens on a chessboard so that no queen attacks
another. Let us ask, said Kim, for the maximum number of queens that
can be put on the board so that each queen attacks exactly n other
queens. As in chess, we assume that a queen cannot attack through an-
other queen.
When n is 0, we have the classic problem. Kim was able to prove that
when n is 1, 10 queens is the maximum number. (A proof is in Journal
a/Recreational Mathematics, Vol. 13, No.1, 1980-81, p. 61.) A pleasing
solution is shown in Figure 16.9 top. The middle illustration shows a
maximal solution of 14 queens when n is 2, a pattern Kim described in
a letter as being "so horribly asymmetric that it has no right to exist."
There are only conjectures for the maximum when n is 3 or 4. Kim's best
result of 16 queens for n = 3 has the ridiculously simple solution shown
in Figure 16.9, bottom, but there is no known proof that 16 is maxi-
mum. For n = 4 Kim's best result is 20 queens. Can you place 20 queens
on a chessboard so each queen attacks exactly four other queens?
The problem can of course be generalized to finite boards of any size,
but Kim has a simple proof based on graph theory that on no finite
board, however large, can n have a value greater than 4. For n = 1 Kim
has shown that the maximum number of queens cannot exceed the
largest integer less than or equal to 4k13, where k is the number of
squares along an edge of the board. For n = 2 he has a more difficult
proof that the maximum number of queens cannot exceed 2k - 2, and
that this maximum is obtainable on all even-order boards.
Kim's problem concerning polycube snakes has not previously been
published, and he and I would welcome any light that readers can
throw on it. First we must define a snake. It is a single connected chain
of identical unit cubes joined at their faces in such a way that each cube
(except for a cube at the end of a chain) is attached face to face to ex-
actly two other cubes. The snake may twist in any possible direction,
provided no internal cube abuts the face of any cube other than its two
immediate neighbors. The snake may, however, twist so that any num-
ber of its cubes touch along edges or at corners. A polycube snake may
be finite in length, having two end cubes that are each fastened to only
The Amazing Creations of Scott Kim 203

figure 16.9.

:00

II




n - 1








n=2

f--
OJ r--
I

n-3
one cube, or it may be finite and closed so that it has no ends. A snake
may also have just one end and be infinite in length, or it may be infi-
nite and endless in both directions.
We now ask a deceptively simple question. What is the smallest
number of snakes needed to fill all space? We can put it another way.
Imagine space to be completely packed with an infinite number of unit
cubes. What is the smallest number of snakes into which it can be dis-
sected by cutting along the planes that define the cubes?
If we consider the two-dimensional analogue of the problem (snakes
made of unit squares), it is easy to see that the answer is two. We sim-
ply intertwine two spirals of infinite one-ended flat snakes, one gray,
one white, as in Figure 16.10.
204
SYMMETRY
16.10.
how fill
O.L.J..C:LI. ..... 'C:)O is not so easily answered. a
four infinitely long one-ended snakes (it is convenient of
as a interlocked
helical shapes that fill all space.
plain in a limited space; you will have to take my
with two! "A solution with only two snakes," he wrote a letter,
a yin-yang symbol:
space left one snake
beauty of such an entwining, and the possibility a .............. , ..............
large enough me searching for a solution."
can course
cubes any number of dimensions. Kim has a
space of n dimensions the minimum number of snakes that completely
fill it
I once
John Horton Conway, the Cambridge mathematician now
When I Kim had not yet shown that
Conway
stared into three-space for a minute or two, then exclainled,
vious!"
notob-
I no idea through I can only
that if the impossibility of filling three-space with two snakes is not
obvious to Conway or to Kim, it probably is not obvious to anyone
The ...,. ............ ..., ..... ..> of Kim 205
16.1 I.
see a
David Morice published this two-stanza "poem" in Wordways (No-
vember 1987, p. 235).
DICK HID
CODEBOOK+
DOBIE KICKED
HOBQ-OH HECK-I DECIDED
I EXCEEDED ID-I BOXED
HICK-ODD DODO-EH KID
DEBBIE CHIDED-HOCK CHECKBOOK
ED-BOB BEDDED CHOICE CHICK
HO HQ-HE ECHOED-OH OH
DOBIE ICED HOODED IBEX
I COOKED OXHIDE COD
EDIE HEEDED COOKBOOK +
ED
DECKED
BOB
To read the second stanza, hold the poem upside down in front of a
mirror.
Donald Knuth, Ronald Graham, and Oren Patashnik, in their mar-
velous book Concrete Mathematics (the word is a blend of Continuous
and Discrete mathematics), published by Addison-Wesley in 1989, in-
troduce their readers to the "umop-apisdn" function. Rotate the word
180 degrees to see what it means in English.
One conjecture about the origin of the expression "Mind your ps and
qs" is that printers often confused the two letters when they were in
lower case. A more plausible theory is that British tavern owners had
to mind their pints and quarts.
In his autobiography Arrow in the Blue, Arthur Koestler recalls meet-
ing many science cranks when he was a science editor in Berlin. One
was a man who had invented a new alphabet. Each letter had fourfold
rotational symmetry. This, he proclaimed, made it possible for four
people, seated on the four sides of a table, to simultaneously read a
book or newspaper at the table's center.
Have you heard about the dyslexic atheist who didn't believe in dog?
Or D.A.M.N., an organization of National Mothers Against Dyslexia?
I could easily write another chapter about the amazing Scott Kim. He
received his Ph.D. in Computers and Graphic Design, at Stanford Uni-
The Amazing Creations of Scott Kim 207
versity, working under Donald Knuth. At a curious gathering of mathe-
maticians, puzzle buffs, and magicians in Atlanta in 1995, Kim demon-
strated how your fingers can model the skeleton of a tetrahedron and a
cube, and how they can form a trefoil knot of either handedness. He also
played an endless octave on a piano, each chord rising up the scale yet
never going out of hearing range, and proved he could whistle one tune
and hum another at the same time. During the Atlanta gathering, he
and his friends Karl Schaffer and Erik Stern of the Dr. Schaffer & Mr.
Stern Dance Ensemble presented a dance performance titled "Dances for
the Mind's Eye." Choreographed by the three performers, the perfor-
mance was based throughout on mathematical symmetries.
Among books illustrated by Kim are my Aha! Gotcha (W. H. Free-
man, 1982) and Han Vardi's Illustrated Computational Recreations in
Mathematica (Addison-Wesley, 1991). Together with Ms. Robin Samel-
son, Kim produced Letterform and illusion, a computer disk with an ac-
companying 48-page book of programs designed for use with Claris's
MacPaint. In 1994 Random House published Kim's Puzzle Workout, a
collection of 42 brilliant puzzles reprinted from his puzzle column in
New Media Magazine. It is the only book of puzzles known to me in
which every single puzzle is totally original with the author.
Scott Kim's queens problem brought many letters from readers who
sent variant solutions for n = 2,3, and 4 on the standard chessboard, as
well as proofs for maximum results and unusual ways to vary the prob-
lem. The most surprising letters came from Jeffrey Spencer, Kjell
Rosquist, and William Rex Marshall. Spencer and Rosquist, writing in
1981, each independently bettered by one Kim's 20-queen solution for
n = 4 on the chessboard. Figure 16.12 shows how each placed 21
queens. It is not unique. Writing in 1989 from Dunedin, New Zealand,
Marshall sent 36 other solutions!
Marshall also went two better than Kim's chessboard pattern for n =
3. He sent nine ways that 18 queens can each attack three others on the
chessboard. The solution shown in Figure 16.13 is of special interest be-
cause only three queens are not on the perimeter. Marshall found a sim-
ple pigeonhole proof that for the order-8 board, n = 4, 21 queens is
indeed maximum. His similar proof shows 18 maximum for n = 3. More
generally, he showed that for n = 4, with k the order of the board, the
maximum is 3k- 3 for k greater than 5. When n = 3, Marshall proved that
the maximum number of queens is the largest even number less than or
208
SYMMETRY
66.12.

41
I t
I
i

i

........



... L_ .... _ 16.13
- 4)/5. n = 2, found
p lies to all boards than 2, not just to boards of even
Perhaps it is worth noting that when n = 4, no queen can occupy a
corner
.................................. sent a n ..... L.LL .............
sider the topmost queen in the leftmost occupied row. At the most it
can
a
independently obtained the same results, as
obtained by William They were published in a paper ti-
Math-
264-71.
In 1996 I received a second letter from William Marshall. He sent me
the provided complete
=1
K N=l N=2 N=3 N=4
3 0 4 2 0
4 5 2 4 0
5 0 1 31 0
6 2 1 304 (307) 1
7 138 (149) 5 2 3
8 47 (49) 2 9 40
9 1 15 755 655
3
Figun 16.14.
of Scott 209
16.15.
n 2, k= 5 n= 2, k=6 n=4. k=6
n=1,k=9
n 2, k= 8
n=3. k=9
210
16.18.
of 211
Chapter 17
What I give form to in daylight is only one
percent of what I've seen in darkness.
-M. C. ESCHER
The Art of
M. C. Escher
There is an obvious but superficial sense in which certain
kinds of art can be called mathematical art. Op art, for instance, is
"mathematical," but in a way that is certainly not new. Hard-edged,
rhythmic, decorative patterns are as ancient as art itself, and even the
modern movement toward abstraction in painting began with the geo-
metric forms of the cubists. When the French Dadaist painter Hans Arp
tossed colored paper squares in the air and glued them where they fell,
he linked the rectangles of cubism to the globs of paint slung by the
later "action" painters. In a broad sense even abstract expressionist art
is mathematical, since randomness is a mathematical concept.
This, however, expands the term "mathematical art" until it becomes
meaningless. There is another and more useful sense of the term that
refers not to techniques and patterns but to a picture's subject matter.
A representational artist who knows something about mathematics can
build a composition around a mathematical theme in the same way
that Renaissance painters did with religious themes or Russian painters
do with political themes. No living artist has been more successful
with this type of "mathematical art" than Maurits C. Escher of the
Netherlands.
"I often feel closer to mathematicians than to my fellow artists," Es-
cher has written, and he has been quoted as saying, "All my works are
games. Serious games." His lithographs, woodcuts, wood engravings,
and mezzotints can be found hanging on the walls of mathematicians
and scientists in all parts of the world. There is an eerie, surrealist as-
pect to some of his work, but his pictures are less the dreamlike fan-
tasies of a Salvador Dali or a Rene Magritte than they are subtle
philosophical and mathematical observations intended to evoke what
the poet Howard Nemerov, writing about Escher, called the "mystery,
212
absurdity, and sometimes terror" of the world. Many of his pictures
concern mathematical structures that have been discussed in books on
recreational mathematics, but before we examine some of them, a word
about Escher himself.
He was born in Leeuwarden in Holland in 1898, and as a young man
he studied at the School of Architecture and Ornamental Design in
Haarlem. For 10 years he lived in Rome. After leaving Italy in 1934 he
spent two years in Switzerland and five in Brussels, then settled in the
Dutch town of Baarn where he and his wife lived until his death in
1972. Although he had a successful exhibit in 1954 at the Whyte
Gallery in Washington, he was much better known in Europe than here.
A large collection of his work is now owned by the National Gallery of
Art in Washington, D.C.
Among crystallographers Escher is best known for his scores of in-
genious tessellations of the plane. Designs in the Alhambra reveal how
expert the Spanish Moors were in carving the plane into periodic rep-
etitions of congruent shapes, but the Mohammedan religion forbade
them to use the shapes of living things. By slicing the plane into jigsaw
patterns of birds, fish, reptiles, mammals, and human figures, Escher
has been able to incorporate many of his tessellations into a variety of
startling pictures.
In Reptiles, the lithograph shown in Figure 17.1, a little monster
crawls out of the hexagonal tiling to begin a brief cycle of 3-space life
that reaches its summit on the dodecahedron; then the reptile crawls
back again into the lifeless plane. In Day and Night, the woodcut in Fig-
ure 17.2, the scenes at the left and the right are not only mirror images
but also almost "negatives" of each other. As the eye moves up the cen-
ter, rectangular fields flow into interlocking shapes of birds, the black
birds flying into daylight, the white birds flying into night. In the cir-
cular wood-cut Heaven and Hell (Figure 17.3) angels and devils fit to-
gether, the similar shapes becoming smaller farther from the center and
finally fading into an infinity of figures, too tiny to be seen, on the rim.
Good, Escher may be telling us, is a necessary background for evil, and
vice versa. This remarkable tessellation is based on a well-known Eu-
clidean model, devised by Henri Poincare, of the non-Euclidean hy-
perbolic plane; the interested reader will find it explained in H.S.M.
Coxeter's Introduction to Geometry (Wiley, 1961), pages 282-90.
If the reader thinks that patterns of this kind are easy to invent, let
him try it! "While drawing I sometimes feel as if I were a spiritualist
The Art of M. C. Escher 213
Ill.
11.2.
214
2000 Cordon Art B.V.-Baam-Holland. All
lGVllit:l:i. 1l.th()griapJl. 1943
Mickelson
and wooaC:llt. 1938
reserved.
' ~ J . Heaven
2000 Cordon Art B.V.-Baarn-Holland. All
\ ..1> ...1\ ....."" ....., .... 1960
reserved.
215
the International Union of Crystallography, reproduces 41 of Escher's
tessellations, many in full color.
Figures 17.4 and 17.5 illustrate another category of Escher's work, a
play with the laws of perspective to produce what have been called
"impossible figures." In the lithograph Belvedere, observe the sketch of
the cube on a sheet lying on the checked floor. The small circles mark
two spots where one edge crosses another. In the skeletal model held
by the seated boy, however, the crossings occur in a way that is not re-
alizable in 3-space. The belvedere itself is made up of impossible struc-
tures. The youth near the top of the ladder is outside the belvedere but
the base of the ladder is inside. Perhaps the man in the dungeon has
lost his mind trying to make sense of the contradictory structures in his
world.
The lithograph Ascending and Descending derives from a perplexing
impossible figure that first appeared in an article, "Impossible Objects:
A Special Type of Visual Illusion," by L. S. Penrose, a British geneticist,
and his son, the mathematician Roger Penrose (British Journal of Psy-
chology, February 1958). The monks of an unknown sect are engaged
in a daily ritual of perpetually marching around the impossible stair-
way on the roof of their monastery, the outside monks climbing, the in-
side monks descending. "Both directions," comments Escher, "though
not without meaning, are equally useless. Two refractory individuals
refuse to take part in this 'spiritual exercise.' They think they know bet-
ter than their comrades, but sooner or later they will admit the error of
their nonconformity."
Many Escher pictures reflect an emotional response of wonder to the
forms of regular and semiregular solids. "In the midst of our often
chaotic society," Escher has written, "they symbolize in an unrivaled
manner man's longing for harmony and order, but at the same time
their perfection awes us with a sense of our own helplessness. Regular
polyhedrons have an absolutely nonhuman character. They are not in-
ventions of the human mind, for they existed as crystals in the earth's
crust long before mankind appeared on the scene. And in regard to the
spherical shape--is the universe not made up of spheres?"
The lithograph Order and Chaos (Figure 17.6) features the "small
stellated dodecahedron," one of the four "Kepler-Poinsot polyhedrons"
that, together with the five Platonic solids, make up the nine possible
"regular polyhedrons." It was first discovered by Johannes Kepler, who
called it "urchin" and drew a picture of it in his Harmonices mundi
216 SYMMETRY
11.4.
The
2000 Cordon Art D.V.-Baam-Holland. All

reserved.
217
11.5.
218
2000 Gordon Art B.V.-Ba.a.rn-Holland. All
.. .. .......,,'-'JOi.i .......... ,LA..1960
reserved.
2000 Cordon Art B.V.-Baam-Holland. All
L .. 1950
reserved.
219
220
III Hand with
1935
All any person can possibly know about the world is derived from
what enters his skull through various sense organs; there is a sense in
which one never experiences anything except what lies within the cir-
cle of his own sensations and ideas. Out of this "phenomenology" he
constructs what he believes to be the external world, including those
other people who appear to have minds in egocentric predicaments
like his own. Strictly speaking, however, there is no way he can prove
that anything exists except himself and his shifting sensations and
thoughts. Escher is seen staring at his own reflection in the sphere.
The glass mirrors his surroundings, compressing them inside one per-
fect circle. No matter how he moves or twists his head, the point mid-
way between his eyes remains exactly at the center of the circle. "He
cannot get away from that central point," says Escher. "The ego re-
mains immovably the focus of his world."
Escher's fascination with the playthings of topology is expressed in
a number of his pictures. At the top of the woodcut Knots (Figure 17.8)
we see the two mirror-image forms of the trefoil knot. The knot at top
left is made with two long flat strips that intersect at right angles. This
double strip was given a t w ~ s t before being joined to itself. Is it a sin-
gle one-sided band that runs twice around the knot, intersecting itself,
or does it consist of two distinct but intersecting Mobius bands? The
large knot below the smaller two has the structure of a four-sided tube
that has been given a quarter-twist so that an ant walking inside, on one
of the central paths, would make four complete circuits through the
knot before it returned to its starting point.
The wood engraving Three Spheres (Figure 17.9), a copy of which is
owned by New York's Museum of Modern Art, appears at first to be a
sphere undergoing progressive topological squashing. Look more care-
fully, however, and you will see that it is something quite different. Can
the reader guess what Escher, with great verisimilitude, is depicting here?
Addendum
When Escher died in 1972, at the age of 73, he was just begin-
ning to become world-famous; not only among mathematicians and
scientists (who were the first to appreciate him), but also with the pub-
lic at large, especially with the young counterculture. The Escher cult
is still growing. You see his pictures everywhere: on the covers of math-
ematical textbooks, on albums of rock music, on psychedelic posters
The Art of M. C. Escher 221
2000 Cordon Art B.V.-Baam-HoHand. All rights reserved.
11.8. Knots, woodcut, 1965
glow under black light, even on T-shirts. When I
an my
can ran one on cover) J I purchased from Es-
only one print, a woodcut. For a mere I
pictures that now
fame?
222 SYMMETRY
223
Bibliography
This highly selective list of references is limited to books and
articles in English. For an extensive bibliography, including foreign
references and motion pictures, see Visions of Symmetry, the definitive
work on Escher by Doris Schattschneider.
H.S.M. Coxeter, "Angels and Devils," The Mathematical Gardner, D. Klarner (ed.),
Wadsworth,1981.
H.S.M. Coxeter et al. (eds.), M. C. Escher: Art and Science, Elsevier, 1986.
H.S.M. Coxeter, "The Trigonometry of Escher's Woodcut 'Circle Limit 1lI,' " The
Mathematical Intelligencer, Vol. 18, No.4, 1996, pp. 42-46.
B. Ernst. The Magic Mirror ofM. C. Escher, Random House, 1976; Ballantine, 1976.
M. C. Escher, Escher on Escher, Abrams, 1989. (Translated by Karin Ford.)
L. Glasser, "Teaching Symmetry," The Journal of Chemical Education, Vol. 44, Sep-
tember 1967, pp. 502-11.
J. L. Locher (ed.), The World of M. C. Escher, Abrams, 1971.
J. L. Locher (ed.), M. C. Escher: His Life and Complete Graphic Work, Abrams,
1982. Reviewed by H.S.M. Coxeter in The Mathematical Intelligencer, Vol. 7, No.
1,1985,pp.59-69.
C. H. MacGillavry, Symmetry Aspects of M. C. Escher's Periodic Drawings, A. Oost-
hoek's Uitgeversmaatschappij NV, 1965. Reprinted as Fantasy and Symmetry-
The Periodic Drawings of M. C. Escher, Abrams, 1976
H. Nemerov, "The Miraculous Transformations of Maurits Cornelis Escher,"
Artists's Proof, Vol. 3, Fall/Winter 1963-64, pp. 32-39.
E. R. Ranucci, "Master of Tesselations: M. C. Escher, 1898-1972," Mathematics
Teacher, Vol. 67, April 1974, pp. 299-306.
J. C. Rush, "On the Appeal ofM. C. Escher's Pictures," Leonardo, Vol. 12, 1979, pp.
48-50.
D. Schattschneider, "The P61ya Escher Connection," Mathematics Magazine, VoL
60, December 1987, pp. 293-98.
D. Schattschneider, Visions of Symmetry, W. H. Freeman, 1990.
D. Schattschneider, "Escher's Metaphors," Scientific American, November 1994,
pp.66-71.
D. Schattschneider and W. Walker, M. C. Escher: Kaleidocyc1es, Ballantine, 1977;
revised edition, Pomegranate Artbooks, 1987.
M. Severin, "The Dimensional Experiments of M. C. Escher," Studio, February
1951, pp. 50-53.
J. L. Teeters, "How to Draw Tesselations of the Escher Type," Mathematics Teacher,
VoL 67, April 1974, pp. 307-10.
M. L. Teuber, "Sources of Ambiguity in the Prints of Maurits C. Escher," Scientific
American, Vol. 231, July 1974, pp. 90-104. See also the correspondence on this
article in Vol. 232, January 1975, pp. 8-9.
K. Wilkie, "Escher: The Journey to Infinity," Holland Herald, Vol. 9, No.1, 1974,
pp.20-43.
224 SYMMETRY
V
Topology
Chapter 18 Klein Bottles
and Other Surfaces
Three jolly sailors from Blaydon-on-Tyne
They went to sea in a bottle by Klein.
Since the sea was entirely inside the hull
The scenery seen was exceedingly dull.
-Frederick Winsor,
The Space Child's Mother Goose
To a topologist a square sheet of paper is a model of a two-
sided surface with a single edge. Crumple it into a ball and it is still
two-sided and one-edged. Imagine that the sheet is made of rubber.
You can stretch it into a triangle or circle, into any shape you please,
but you cannot change its two-sidedness and one-edgedness. They are
topological properties of the surface, properties that remain the same
regardless of how you bend, twist, stretch, or compress the sheet.
Two other important topological invariants of a surface are its chro-
matic number and Betti number. The chromatic number is the maxi-
mum number of regions that can be drawn on the surface in such a way
that each region has a border in common with every other region. If
each region is given a different color, each color will border on every
other color. The chromatic number of the square sheet is 4. In other
words, it is impossible to place more than four differently colored re-
gions on the square so that any pair has a boundary in common. The
term "chromatic number" also designates the minimum number of col-
ors sufficient to color any finite map on a given surface. It is now
known that 4 is the chromatic number, in this map-coloring sense, for
the square, tube, and sphere, and for all other surfaces considered in
this chapter, the chromatic number is the same under both definitions.
The Betti number, named after Enrico Betti, a 19th-century Italian
physicist, is the maximum number of cuts that can be made without di-
viding the surface into two separate pieces. If the surface has edges,
each cut must be a "crosscut": one that goes from a point on an edge to
another point on an edge. If the surface is closed (has no edges), each
cut must be a "loop cut": a cut in the form of a simple closed curve.
Clearly the Betti number of the square sheet is O. A crosscut is certain
to produce two disconnected pieces.
227
le.l. a
IB.2. Torus sUIltace .......... " ............... a SQlllare
229
230
shows how the bottle is traditionally depicted. Imagine the lower end of
a tube stretched out, bent up, and plunged through the tube's side, then
joined to the tube's upper mouth. In an actual model made, say, of glass
there would be a hole where the tube intersects the side. You must dis-
regard this defect and think of the hole as being covered by a continua-
tion of the bottle's surface. There is no hole, only an intersection of
surfaces. This self-intersection is necessary because the model is in 3-
space. If we conceive of the surface as being embedded in 4-space, the
self-intersection can be eliminated entirely. The Klein bottle is one-sided,
no-edged, and has a Betti number of 2 and a chromatic number of 6.
Daniel Pedoe, a mathematician at the University of Minnesota, is the
author of The Gentle Art of Mathematics. It is a delightful book, but on
page 84 Professor Pedoe slips into a careless bit of dogmatism. He de-
scribes the Klein bottle as a surface that is a challenge to the glass
blower, but one "which cannot be made with paper." Now, it is true that
at the time he wrote this apparently no one had tried to make a paper
Klein bottle, but that was before Stephen Barr, a science-fiction writer
and an amateur mathematician of Woodstock, NY, turned his attention
to the problem. Barr quickly discovered dozens of ways to make paper
Klein bottles. Here I will describe a variation of my own that is made
from a paper tube. The tube can be a sealed envelope with its left and
right edges cut open.
The steps are given in Figure 18.4. First, make a tube by folding the
square in half and joining the right edges with a strip of tape as shown
(Step 1). Cut a slot about a quarter of the distance from the top of the
tube (Step 2), cutting only through the thickness of paper nearest you.
This corresponds to the "hole" in the glass model. Fold the model in
half along the broken line A. Push the lower end of the tube up through
the slot (Step 3) and join the edges all the way around the top of the
model (Step 4) as indicated by the arrows. It is not difficult to see that
this flat, square model is topologically identical with the glass bottle
shown in Figure 18.3. In one way it is superior: there is no actual hole.
True, you have a slot where the surface self-intersects, but it is easy to
imagine that the edges of the slot are joined so that the surface is every-
where edgeless and continuous.
Moreover, it is easy to cut this paper model and demonstrate many
of the bottle's astonishing properties. Its Betti number of 2 is demon-
strated by cutting the two loops formed by the two pairs of taped edges.
If you cut the bottle in half vertically, you get two Mobius bands, one
Klein Bottles and Other Surfaces 231
232
F
a
233
along the solid black lines shown in Step 1. Fold the square along the
diagonal A-A /, inserting slot C into slot B (Steps 2 and 3). You must
think of the line where the slots interlock as an abstract line of self-
intersection. Fold up the two bottom triangular flaps E and F, one on
each side (Step 4), and tape the edges as indicated.
The model is now what topologists call a cross-cap, a self-
intersecting Mobius strip with an edge that can be stretched into a cir-
cle without further self-intersection. This edge is provided by the edges
of cut D, originally made along the square's diagonal. Note that unlike
the usual model of a Mobius strip, this one is symmetrical: neither
right- nor left-handed. When the edge of the cross-cap is closed by tap-
ing it (Step 5), the model becomes a projective plane. You might expect
it to have a Betti number of 2, like the Klein bottle, but it does not. It
has a Betti number of 1. No matter how you loop-cut it, the cut pro-
duces either two pieces or a piece topologically equivalent to a square
sheet that cannot be cut again without making two pieces. If you re-
move a disk from anywhere on the surface of the projective plane, the
model reverts to a cross-cap.
Figure lB.7 summarizes all that has been said. The square diagrams
in the first column show how the edges join in each model. Sides ofthe
same color join each to each, with the direction of their arrows coin-
ciding. Corners labeled with the same letter are comers that come to-
gether. Broken lines are sides that remain edges in the finished model.
Next to the chromatic number of each model is shown one way the sur-
face can be mapped to accommodate the maximum number of colors.
It is instructive to color each sheet as shown, coloring the regions on
both sides of the paper (as though the paper were cloth through which
the colors soaked), because you must think of the sheet as having zero
thickness. An inspection of the final model will show that each region
does indeed border on every other one.
Addendum
Although I was not the first to model Klein bottles with paper
(credit for this goes to Stephen Barr-see bibliography), my contribu-
tion was to show how easily a model can be made from an envelope
and cut in half to make two Mobius strips of opposite handedness.
A simple way to demonstrate the one-sided property of a Klein bot-
tle is to punch a hole in a model and insert a piece of string. No matter
234 TOPOLOGY
BETTI
SURFACE CHROMATIC NUMBER SIDES EDGES NUMBER
A B
r-----....
~
I I
SQUARE
I I
(OR DISK)
I I
4 2 1 0
I I
I I
.... _____ ...l
C 0
A A
TUBE
[ I _ ~ ~ ]
4
~
2 2 1
B B
A B
SPHERE
D
4
B C
~
2 0 0
A B
[ ~ ~ ! ]
~
MOBIUS
6 3 4 5 3
STRIP
1 1 1
-
6 1
B A
A A
0 ~
7
1
TORUS 7 56 2 0 2
3
4
A A
A A
D ~
KLEIN
6 2 3
BOTTLE
1 0 2
4 6
A A
A B
PROJECTIVE [I ~ ) t[
1
5
6 21
3 t-- 1 0 1
PLANE
-E--
4
6
B A
Figure IB.7. Topological invariants of seven basic surfaces
Klein Bottles and Other Surfaces
235
where you make the hole, you can always tie the ends of the string to-
gether.
The Klein bottle continues to intrigue limerick writers. Here are three
I have encountered:
Topologists try hard to floor us
With a nonorientable torus.
The bottle of Klein
They say is divine
But it is so exceedingly porous.
-Anonymous
A geometrician named Klein
Thought the Mobius band asinine.
"Though its outside is in,
Still it's ugly as sin;
It ain't round like that bottle o'minef"
-M. M. H. Coffee and J. J. Zeltmacher, Jr.
An anti-strong-drinker named Klein
Invented a bottle for wine.
"There's no stopper," he cried,
"And it has no inside,
So the grapes have to stay on the vine!"
-James Albert Lindon
I confess that I have made use of a Klein bottle in two works of fic-
tion. Professor Slapenarski falls into a Klein bottle and disappears at
the end of my mathematically flawed story "The Island of Five Colors"
(you'll find it in Clifton Fadiman's anthology Fantasia Mathematica),
and again in my novel Visitors from Oz (1999) where it is used as a de-
vice for transporting Dorothy, Scarecrow, and Tin Woodman from Oz
(now in a parallel world) to Central Park in Manhattan.
If you would like to own a glass Klein bottle, a firm called Acme,
6270 Colby Street, Oakland, CA 94618, has five handsome glass Klein
bottles for sale, of various sizes, shapes, and prices. The firm can also
be reached on the Internet at www.kleinbottle.com.
Answers
The torus-cutting problem is solved by first ruling three paral-
leI lines on the unfolded square (see Figure 18.8), When the square is
236 TOPOLOGY
Figure 18.8. Solution to the torus-
cutting problem
folded into a torus, as explained, the lines make two closed loops. Cut-
ting these loops produces two interlocked bands, each two-sided with
two half-twists.
How does one find a loop cut on the Klein bottle that will change the
surface to a single Mobius strip? On both left and right sides of the nar-
row rectangular model described you will note that the paper is creased
along a fold that forms a figure-eight loop. Cutting only the left loop
transforms the model into a Mobius band; cutting only the right loop
produces an identical band of opposite handedness.
What happens if both loops are cut? The result is a two-sided, two-
edged band with four half-twists. Because of the slot the band is cut
apart at one point, so that you must imagine the slot is not there. This
self-intersecting band is mirror-symmetrical, neither right- nor left-
handed. You can free the band of self-intersection by sliding it carefully
out of the slot and taping the slot together. The handedness of the re-
sulting band (that is, the direction of the helices formed by its edges)
depends on whether you slide it out to the right or the left. This and the
previous cutting problems are based on paper models that were in-
vented by Stephen Barr and are described in his Experiments in Topol-
ogy (Crowell, 1964).
Bibliography
M. A. Armstrong, Basic Topology, Springer-Verlag, 1983.
B. H. Arnold, Intuitive Concepts in Elementary Topology, Prentice-Hall, 1962.
S. Barr, Experiments in Topology, Crowell, 1964.
W. G. Chinn and N. E. Steenrod, First Concepts o/Topology. Mathematical Associ-
ation of America, 1966.
Klein Bottles and Other Surfaces 237
F. H. Crowe, Introduction to Topology, Saunders, 1989.
G. K. Francis, A Topological Picture Book, Springer-Verlag, 1987.
D. W. Hall and G. L. Spencer, Elementary Topology, Wiley, 1955.
W. Lietzmann, Visual Topology, Chatto and Windus, 1965.
M. J. Mansfield, Introduction to Topology, Van Nostrand, 1963; Krieger, 1972.
W. S. Massey, Algebraic Topology: An Introduction, Harcourt, Brace, 1967.
B. Mendelsohn, Introduction to Topology, Blackie, 1963.
E. M. Patterson, Topology, Oliver and Boyd, 1956.
V. V. Prasolov, Intuitive Topology, Providence, Rl: American Mathematical Society,
1995.
I. Stewart, "The Topological Dressmaker," Scientific American, July 1993, pp.
110-12.
L Stewart, "Glass Klein Bottles," Scientific American, March 1998, pp. 100-1. See
also the note at the end of Stewart's December 1998 column.
I. Stewart, "Tangling With Topology," Scientific American, April 1999, pp. 123-24.
A. W. Tucker and H. S. Bailey, "Topology," Scientific American, January 1950, pp.
18-24.
R. L. Wilder, "Topology: Its Nature and Significance," Mathematics Teacher, Octo-
ber 1962, pp. 462-75.
238 TOPOLOGY
Chapter 19
"A knot!" said Alice, always ready to make
herself useful, and looking anxiously about
her. "Oh, do let me help to undo itl"
-Alice in Wonderland, Chapter 3
Knots
To a topologist knots are closed curves embedded in three-
dimensional space. It is useful to model them with rope or cord and to
diagram them as projections on a plane. If it is possible to manipulate
a closed curve-of course, it must not be allowed to pass through it-
self-so that it can be projected on a plane as a curve with no crossing
points then the knot is called trivial. In ordinary discourse one would
say the curve is not knotted. "Links" are two or more closed curves that
cannot be separated without passing one through another.
The study of knots and links is now a flourishing branch of topology
that interlocks with algebra, geometry, group theory, matrix theory,
number theory, and other branches of mathematics. Some idea of its
depth and richness can be had from reading Lee Neuwirth's excellent
article "The Theory of Knots" in Scientific American (June 1979). Here
we shall be concerned only with some recreational aspects of knat the-
ory: puzzles and curiosities that to be understood require no more than
the most elementary knowledge of the topic.
Let's begin with a question that is trivial but that can catch even
mathematicians off guard. Tie an overhand knot in a piece of rope as is
shown in Figure 19.1. If you think of the ends of the rope as being
joined, you have tied what knot theorists call a trefoil knot. It is the sim-
plest of all knots in the sense that it can be diagrammed with a mini-
mum of three crossings. (No knot can have fewer crossings except the
trivial knot that has none.) Imagine that end A of the rope is passed
through the loop from behind and the ends are pulled. Obviously the
knot will dissolve. Now suppose the end is passed twice through the
loop as is indicated by the broken line. Will the knot dissolve when the
ends of the rope are pulled?
Most people guess that it will form another knot. Actually the knot
239
Figure 19.1.
dissolves as before. The end must go three times through the loop to
produce another knot. If you try it, you will see that the new trefoil cre-
ated in this way is not the same as the original. It is a mirror image. The
trefoil is the simplest knot that cannot be changed to its mirror image
by manipulating the rope.
The next simplest knot, the only one with a minimum of four cross-
ings, is the figure eight at the right in Figure 19.1. In this form it is eas-
ily changed to its mirror image. Just turn it over. A knot that can be
manipulated to make its mirror image is called amphicheiral because
like a rubber glove it can be made to display either handedness. After
the figure eight the next highest amphicheiral knot has six crossings,
and it is the only 6-knot of that type. Amphicheiral knots become pro-
gressively scarcer as crossing numbers increase.
A second important way to divide knots into two classes is to dis-
tinguish between alternating and nonalternating knots. An alternating
knot is one that can be diagrammed so that if you follow its curve in ei-
ther direction, you alternately go over and under at the crossings. Al-
ternating knots have many remarkable properties not possessed by
nonalternating knots.
Still another important division is into prime and composite knots.
A prime knot is one that cannot be manipulated to make two or more
separated knots. For example, the square knot and the granny knot are
not prime because each can be changed to two side-by-side trefoils.
The square knot is the "product" of two trefoils of opposite handed-
240 TOPOLOGY
ness. The granny is the product of two trefoils of the same handed-
ness, and therefore (unlike the square knot) it is not amphicheiral. Both
knots are alternating. As an easy exercise, see if you can sketch a square
knot with six (the minimum) alternating crossings.
All prime knots of seven or fewer crossings are alternating. Among
the 8-knots only the three in Figure 19.2 are nonalternating. No matter
how long you manipulate a rope model of one of these knots, you will
never get it to lie flat in the form of an alternating diagram. The knot at
top right is a bowline. The bottom knot is a torus knot as explained
below.
Figure 19.2.
A fourth basic binary division of knots is into the invertible and non-
invertible. Imagine an arrow painted on a knotted rope to give a direc-
tion to the curve. If it is possible to manipulate the rope so that the
structure remains the same but the arrow points the other way, the knot
is invertible. Until the mid-1960s one of the most vexing unsolved
problems in knot theory was whether noninvertible knots exist. All
knots of seven or fewer crossings, and all but one 8-knot and four 9-
knots had earlier been found invertible by manipulating rope models.
It was in 1963 that Hale F. Trotter, now at Princeton University, an-
nounced in the title of a surprising paper "Non-invertible Knots Exist"
(Topology, Vol. 2, No.4, December 1963, pp. 275-80).
Trotter described an infinite family of pretzel knots that will not in-
vert. A pretzel knot is one that can be drawn, without any crossings, on
the surface of a pretzel (a two-hole torus). It can be drawn as shown in
Knots 241
Figure 19.3 as a two-strand braid that goes around two "holes," or it can
be modeled by the edge of a sheet of paper with three twisted strips. If
the braid surrounds just one hole, it is called a torus knot because it can
be drawn without crossings on the surface of a doughnut.
Figure 19.3.
Trotter found an elegant proof that all pretzel knots are noninvertible
if the crossing numbers for the three twisted strips are distinct odd in-
tegers with absolute values greater than 1. Positive integers indicate
braids that twist one way and negative integers indicate an opposite
twist. Later Trotter's student Richard L. Parris showed in his unpub-
lished Ph.D. thesis that the absolute values can be ignored provided the
signed values are distinct and that these conditions are necessary as
well as sufficient for noninvertible pretzels. Thus the simplest nonin-
vertible pretzel is the one shown. Its crossing numbers of 3, -3, and 5
make it an ll-knot.
It is now known that the simplest noninvertible knot is the am-
phicheiral 8-knot in Figure 19.4. It was first proved noninvertible by
Aldo Kawauchi in Proceedings o/the Japan Academy (Vol. 55, Series
A, No. 10, December 1979, pp. 399-402). According to Richard Hartley,
in "Identifying Non-invertible Knots" (Topology, Vol. 22, No.2, 1983,
pp. 137-45), this is the only noninvertible knot of eight crossings, and
there are only two such knots of nine crossings and 33 of 10. All 36 of
these knots had earlier been declared noninvertible by John Horton
Conway, but only on the empirical grounds that he had not been able
to invert them. The noninvertible knots among the more than 550 knots
with 11 crossings had not yet been identified.
In 1967 Conway published the first classification of all prime knots
with 11 or fewer crossings. (A few minor errors were corrected in a later
printing.) You will find clear diagrams for all prime knots through 10
crossings, and all links through nine crossings, in Dale Rolfsen's valu-
242 TOPOLOGY
Figure 19.4.
able 1990 book Knots and Links. There are no knots with 1 or 2 cross-
ings, one with 3, one with 4, two with 5, three with 6, seven with 7, 21
with 8 crossings, 49 with 9, 165 with 10, and 552 with 11, for a total of
801 prime knots with 11 or fewer crossings. At the time I write, the clas-
sification has been extended through 16 crossings.
There are many strange ways to label the crossings of a knot, then de-
rive an algebraic expression that is an invariant for all possible dia-
grams of that knot. One of the earliest of such techniques produces
what is called a knot's Alexander polynomial, named after the Ameri-
can mathematician James W. Alexander who discovered it in 1928.
Conway later found a beautiful new way to compute a "Conway poly-
nomial" that is equivalent to the Alexander one.
For the unknotted knot with no crossings the Alexander polynomial
is 1. The expression for the trefoil knot of three crossings is x
2
- X + 1,
regardless of its handedness. The figure-eight knot of four crossings
has the polynomial x
2
- 3x + 1. The square knot, a product of two tre-
foils, has an Alexander polynomial of (x
2
- x+ 1)2, the square of the tre-
foil's expression. Unfortunately, a granny knot has the same
polynomial. If two knot diagrams give different polynomials, they are
sure to be different knots, but the converse is not true. Two knots may
Knots 243
have the same polynomial yet not be the same. Finding a way to give
any knot an expression that applies to all diagrams of that knot, and
only that knot, is the major unsolved problem in knot theory.
Although there are tests for deciding whether any given knot is triv-
ial, the methods are complex and tedious. For this reason many prob-
lems that are easy to state are not easy to resolve except by working
empirically with rope models. For instance, is it possible to twist an
elastic band around a cube so that each face of the cube has an
under-over crossing as shown in Figure 19.5. To put it another way, can
you tie a cord around a cube in this manner so that if you slip the cord
off the cube, the cord will be unknotted?
Figure 19.5.
Note that on each face the crossing must take one of the four forms
depicted in the illustration. This makes 4
6
= 4,096 ways to wrap the
cord. The wrapping can be diagrammed as a 12-knot, with six pairs of
crossings, each pair of which can have one of four patterns. The prob-
lem was first posed by Horace W. Hinkle in Journal of Recreational
Mathematics in 1978. In a later issue (Vol. 12, No.1, 1979-80, pp.
60-62) Karl Scherer showed how symmetry considerations reduce the
number of essentially different wrappings to 128. Scherer tested each
wrapping empirically and found that in every case the cord is knotted.
This has yet to be confirmed by others, and no one has so far found a
simpler way to attack the problem. The impossibility of getting the de-
sired wrapping with an unknotted cord seems odd, because it is easy
to twist a rubber band around a cube to put the under-over crossings on
just two or four faces (all other faces being straight crossings), and
seemingly impossible to do it on just one face, three faces, or five faces.
One would therefore expect six to be possible, but apparently it is not.
It may also be impossible to get the pattern even if two, three, or four
rubber bands are used.
244 TOPOLOGY
Figure 19.6 depicts a delightful knot-and-link puzzle that was sent to
me by its inventor, Majunath M. Regde, then a mathematics student in
India. The rope's ends are tied to a piece of furniture, say a chair. Note
that the two trefoil knots form a granny. The task is to manipulate the
rope and ring so that the ring is moved to the upper knot as is indicated
by the broken line. All else must remain identical.
Figure 19.6.
It is easy to do if you have the right insight. Of course, the rope must
not be untied from the chair, nor are you allowed to open a knot and
pass the chair through it. It will help if you think of the ends of the rope
as being permanently fastened to a wall.
The trick of dissolving or creating knots by passing a person through
a loop was actually used by fake mediums in the days when it was
fashionable to relate psychic phenomena to the fourth dimension.
Knots in closed curves are possible only in 3-space. In 4-space all knots
dissolve. If you could toss an unknotted loop of rope to a creature in 4-
space, it could tie any knot in the loop and toss it back to you with the
knot permanently formed. There was a popular theory among physi-
cists who believed in spiritualism that mediums had the power to move
objects in and out of higher spaces. Some mediums, such as the Amer-
ican mountebank Renry Slade, exploited this theory by pretending to
put knots into closed loops of cord. Johann Karl F. Zollner, an Austrian
physicist, devoted an entire book to Slade and hyperspace. Its English
translation, Transcendental Physics (Arno Press, 1976), is worth read-
ing as striking testimony to the ease with which an intelligent physicist
can be gulled by a clever conjurer.
Knots 245
In another instance, psychic investigators William Cox and John
Richards exhibited a stop-action film that purported to show two
leather rings becoming linked and unlinked inside a fish tank. "Later
examination showed no evidence that the rings were severed in any
way," wrote National Enquirer when it reported this "miracle" on Oc-
tober 27,1981. I was then reminded of an old conjuring stage joke. The
performer announces that he has magically transported a rabbit from
one opaque box to another. Then before opening either box he says
that he will magically transport the rabbit back again.
It is easy, by the way, to fabricate two linked "rubber bands." Just
draw them linked on the surface of a baby's hollow rubber teething
ring and carefully cut them out. Two linked wood rings, each of a dif-
ferent wood, can be carved if you insert one ring into a notch cut into
a tree, then wait many years until the tree grows around and through
it. Because the trefoil is a torus knot, it too is easily cut from a teething
ring.
The trick I am about to describe was too crude for Slade, but less
clever mediums occasionally resorted to it. You will find it explained,
along with other knot-tying swindles, in Chapter 2 of Hereward Car-
rington's The Physical Phenomena of Spiritualism, Fraudulent and
Genuine (H. B. Turner & Co., Boston, 1907). One end of a very long
piece of rope is tied to the wrist of one guest and the other end is tied
to the wrist of another guest. After the seance, when the lights are
turned on, several knots are in the rope. How do they get there?
The two guests stand side by side when the lights go out. In the dark
the medium (or an accomplice) makes a few large coils of rope, then
passes them carefully over the head and body of one of the guests. The
coils lie flat on the floor until later, when the medium casually asks that
guest to step a few feet to one side. This frees the coils from the person,
allowing the medium to pull them into a sequence of tight knots at the
center of the rope. Stepping to one side seems so irrelevant to the phe-
nomenon that no one remembers it. Ask the guest himself a few weeks
later whether he changed his position, and he will vigorously and hon-
estly deny it.
Roger Penrose, the British mathematician and physicist, once
showed me an unusual trick involving the mysterious appearance of a
knot. Penrose invented it when he was in grade school. It is based on
what in crocheting, sewing, and embroidery is called a chain stitch.
Begin the chain by trying a trefoil knot at one end of a long piece of
246 TOPOLOGY
heavy cord or thin rope and hold it with your left hand as in step 1 in
Figure 19.7. With your right thumb and finger take the cord at A and
pull down a loop as in step 2. Reach through the loop, take the cord at
B, and pull down another loop (step 3). Again reach forward through
the lowest loop, take the cord at D, and pull down another loop (step
4). Continue in this way until you have formed as long a chain as pos-
sible.
Figure 19.7.
With your right hand holding the lower end of the chain, pull the
chain taut. Ask someone to select any link he likes and then pinch the
link between his thumb and forefinger. Pull on both ends of the cord.
All links dissolve, as expected, but when he separates his finger and
thumb, there is a tight knot at precisely the spot he pinched!
Joel Langer, a mathematician at Case Western Reserve University,
made a remarkable discovery. He found a way of constructing what he
calls "jump knots" out of stainless-steel wire. The wire is knotted and
then its ends are bonded. When it is manipulated properly, it can be
pressed flat to form a braided ring. Release pressure on the ring; tension
in the wire causes it to spring suddenly into a symmetrical three-
dimensional shape. It is now a frustrating puzzle to collapse the wire
back to its ring form.
In 1981 Langer and his associate Sharon O'Neil formed a company
they called Why Knots. It made and sold three handsome jump knots:
the Figure Eight, the Chinese Button Knot, and the Mathematician's
Loop. When you slide one of these wire knots out of its square enve-
lope, it pops into an elegant hanging ornament. The figure eight is the
easiest to put back into its envelope. The Chinese button knot (so called
because it is a form widely used in China for buttons on nightclothes)
is more difficult. The mathematician's loop is the most difficult.
Knots 247
These shapes make it easier to understand how the 18th-century
physicists could have developed a theory, respectable in its day, that
molecules are different kinds of knots into which vortex rings of ether
(today read "space-time") get themselves tied. Indeed, it was just such
speculation that led the Scottish physicist Peter Guthrie Tait to study
topology and conduct the world's first systematic investigation of knot
theory.
Addendum
Enormous advances in knot theory have been made since this
chapter was written in 1983, and knot theory is now one of the most ex-
citing and active branches of mathematics. Dozens of new polynomials
for classifying knots have been discovered. One is called the Homfly
after the last initials of its six independent discoverers. The most sig-
nificant new expression is the Jones polynomial found in 1984 by the
New Zealand mathematician Vaughan F. R. Jones, now at the Univer-
sity of California, Berkeley. It has since been improved and generalized
by Louis Kauffman and others. Although these new polynomials are
surprisingly simple and powerful, no one has yet come up with an al-
gebraic technique for distinguishing all knots. Knots with different
polynomials are different, but it is still possible that two distinct knots
will have the same expression.
The Alexander polynomial does not decide between mirror-image
knots, and as we have seen, it does not distinguish the square knot
from the granny. The Jones polynomial provides both distinctions. So
far, it is not clear just why the Jones and the other new polynomials
work. "They're magic" is how Joan Birman, a knot expert at Barnard
College, put it.
The most amazing development in recent knot theory was the dis-
covery that the best way to understand the Jones polynomial was in
terms of statistical mechanics and quantum theory! Sir Michael Atiyah
now retired from the University of Edinburgh, was the first to see these
connections, then Edward Witten, at the Institute for Advanced Study
in Princeton, did the pioneer work in developing the connections. Knot
theory now has surprising applications to superstrings, a theory that ex-
plains basic particles by treating them as tiny loops, and to quantum
field theory. There is intense interaction between today's physicists
248 TOPOLOGY
and topologists. Discoveries in physics are leading to new discoveries
in topology, and vice versa. No one can predict where it will all lead.
Another unexpected application of knot theory is in broadening our
understanding of the structure and properties of large molecules such
as polymers, and especially the behavior of DNA molecules. DNA
strands can become horribly knotted and linked, unable to replicate
until they are untied or unlinked by enzymes called topoisom-erases.
To straighten out a DNA strand, enzymes have to slice them so they can
pass through themselves or another strand, then splice the ends to-
gether again. The number of times this must occur to undo a knot or
linkage of DNA determines the speed with which the DNA unknots or
unlinks.
There is a delightful three-color test for deciding if a knot diagram
represents a knot. Draw the diagram, then see if you can color its "arcs"
(portions of the line between two crossings) with three colors so that ei-
ther all three colors meet at each crossing or there is only one color at
each crossing, and provided at least one crossing shows all three col-
ors. If you can do this, the line is knotted. If you can't, the line mayor
may not be knotted. The three-coloring can also be used to prove that
two knots are different.
In 1908 the German mathematician Heinrich Tietze conjectured that
two knots are identical if and only if their complements-the topolog-
ical structure of the space in which they are embedded-are identical.
His conjecture was proved in 1988 by two American mathematicians,
Cameron M. Gordon and John E. Luecke. A knot's complement is a
structure in 3-space, in contrast to the knot which is one-dimensional.
Its topological structure is more complicated than the knot's, but of
course it contains complete information about the knot. The theorem
fails for links. Two links that are not the same can have identical com-
plements.
Associated with each knot's complement is a group. Like the poly-
nomials, which can be extracted from the group, two knots can have the
same group yet not be the same knots. An anonymous poet summed up
the situation this way in the British periodical Manifold (Summer
1972):
Knots
A knot and
another
249
knot may
not be the
same knot, though
the knot group of
the knot and the
other knot's
knot group
differ not; BUT
if the knot group
ofa knot
is the knot group
of the not
knotted
knot,
the knot is
not
knotted.
The American philosopher Charles Peirce, in a section on knots in
his New Elements of Mathematics (Volume 2, Chapter 4), shows how
the Borromean rings (three rings linked in such a way that although
they can't be separated, no two rings are linked) can be cut from a
three-hole torus. Peirce also shows how to cut the figure-eight knot
and the bowline knot from a two-hole torus.
Richard Parris called attention to the fact that not all of the 4,096
ways to wrap string around the cube, in the problem I posed, are knots.
Most of them are links of two, three, or four separate loops.
Conway has proved that if you draw a complete graph for seven
points located anywhere in space-that is, draw a line connecting each
pair of points-the lines will form at least one knot.
For the most recent results in knot theory there is now a Journal of
Knot Theory and Its Ramifications.
Answers
Figure 19.8 shows how a square knot can be changed to an al-
ternating knot of six crossings. Simply flip dotted arc a over to make
arc b.
Figure 19.9 shows one way to solve the ring-and-granny-knot puzzle.
First make the lower knot small, then slide it (carrying the ring with it)
250 TOPOLOGY
FigUI'e 19.8.
a
FigUI'e 19.9.
-,
'\
,
It,
I
I
/
--""
b c
up and through the higher knot (a). Open it. Two trefoil knots are now
side by side (b). Make the ringless knot small, then slide it through and
down the other knot. Open it up and you have finished (e).
Bibliography
Books
C. C. Adams, The Knot Book. W. H. Freeman, 1994.
M. Atiyah. The Geometry and Physics of Knots, Cambridge University Press, 1990.
R. H. Crowell and R. H. Fox, Introduction to Knot Theory, Blaisdell, 1963; Springer-
Verlag, 1977.
D. W. Fanner and T. B. Stanford, Knots and Surfaces, American Mathematical So-
ciety, 1998.
L. Kauffman, On Knots, Princeton University Press, 1987.
L. Kauffman, Knots and Physics, World Scientific, 1991.
T. Kohno, New Developments in the Theory of Knots, World Scientific, 1990.
C. Livingston, Knot Theory, Washington, DC: Mathematical Association of Amer-
ica, 1993.
K. Murasugi, Knot Theory and Its Applications, Birkhauser, 1999.
V. V. Pracolov, Knots, Links, Braids, and 3-Manifolds, Providence, RI: American
Mathematical Society, 1997.
D. Rolfsen, Knots and Links, Publish or Perish, 1976, Second edition, 1990.
J. c. Thrner and P. van de Griend, The History and Science of Knots, World Scien-
tific, 1996.
Knots 251
D.J.A. Walsh, Complexity Knots, Colourings and Counting, Cambridge University
Press, 1993.
Papers
Out of hundreds of papers on knot theory published since 1980,
I have selected only a few that have appeared since 1990.
C. C. Adams, "THings of Space by Knotted Tiles," Mathematical Intelligencer, Vol.
17, No.2, 1995, pp. 41-51.
P. Andersson "The Color Invariant for Knots and Links," American Mathematical
Monthly, Vol. 102, May 1995, pp. 442-48.
M. Atiyah, "Geometry and Physics," The Mathematical Gazette, March 1996, pp.
78-82.
J. S. Birman "Recent Developments in Braid and Link Theory," The Mathematical
Intelligencer, Vol. 13 1991, pp. 57-60.
B. Cipra, "Knotty Problems-and Real-World Solutions," Science, Vol. 255, Janu-
ary 24,1992, pp. 403-4.
B. Cipra, "From Knot to Unknot," What's Happening in the Mathematical Sci-
ences, Vol 2, 1994, pp. 8-13.
O. T. Dasbach and S. Hougardy, "Does the Jones Polynomial Detect Unknotted-
ness?" Experimental Mathematics, Vol. 6, 1997, pp. 51-56.
S. Harris and G. Quenell, "Knot Labelings and Knots Without Labelings." Mathe-
matical Intelligencer, Vol. 21, No.2, 1999, pp. 51-56.
B. Hayes, "Square Knots," American Scientist, November/December, 1997, pp.
506-10.
J. Haste, M. Thistlewaite, and J. Weeks, "The First 1,701,936 Knots," Mathemati-
cal Intelligencer, Vol. 20, No.4, 1998, pp. 33-48.
V. F. R. Jones, "Knot Theory and Statistical Mechanics," Scientific American, No-
vember 1990, pp. 98-103.
R. Matthews, "Knots Landing," New Scientist, February 1, 1997, pp. 42-43.
W. Menasco and 1. Rudolph, "How Hard is it to Untie a Knot?" American Scien-
tist, Vol. 83, January/February 1995, pp. 38-50.
I. Peterson, "Knotty Views," Science News, Vol. 141, March 21,1992, pp. 186-7.
A. Sosinsky, "Braids and Knots," Quantum, January/February 1995, pp. 11-15.
A. Sosinsky, "Knots, Links, and Their Polynomials," Quantum, July/August 1995,
pp.9-14.
I. Stewart, "Knots, Links and Videotape," Scientific American, January 1994, pp.
152-54.
D. W. Sumners, "Untangling DNA," The MathematicalIntelligencer, Vol. 12, 1990,
pp.71-80.
D. W. Sumners, "Lifting the Curtain: Using Topology to Probe the Hidden Action
of Enzymes," Notices of the American Mathematical Society, Vol. 42, May 1995,
pp.528-31.
O. Viro, "Tied Into Knot Theory," Quantum, May/June 1998, pp. 18-20.
252 TOPOLOGY
Knot Theory and Quantum Mechanics
M. Atiyah, The Geometry and Physics of Knots, Cambridge University Press, 1990.
M. L. Ge and C. N. Yang (eds), Braid Groups, Knot Theory and Statistical Mechan-
ics, World Scientific, 1989.
V. F. R. Jones, "Knot Theory and Statical Mechanics," Scientific American, No-
vember 1990, pp. 98-103.
E. Witten, "Quantum Field Theory and the Jones Polynomial," Communications in
Mathematical Physics, Vol. 1 2 ~ , No.3, 1989, pp. 351-99.
Knots 253
-ANON.
20.1. torus
254
with one handle, a cube with one hole through it, and so on. Think of
these surfaces as a thin membrane that can be stretched or compressed
as much as one wishes. Each can be deformed until it becomes a per-
fect toroidal surface. In what follows, "torus" will mean any surface
topologically equivalent to a torus.
A common misunderstanding about topology is the belief that a rub-
ber model of a surface can always be deformed in three-dimensional
space to make any topologically equivalent model. This often is not the
case. A Mobius strip, for example, has a handedness in 3-space that
cannot be altered by twisting and stretching. Handedness is an extrin-
sic property it acquires only when embedded in 3-space. Intrinsically
it has no handedness. A 4-space creature could pick up a left-handed
strip, turn it over in 4-space and drop it back in our space as a right-
handed modeL
A similar dichotomy applies to knots in closed curves. Tie a single
overhand (or trefoil) knot in a piece of rope and join the ends. The sur-
face of the rope is equivalent to a knotted torus. It has a handedness,
and no amount of fiddling with the rope can change the parity. Intrin-
sically the rope is not even knotted. A 4-space creature could take from
us an unknotted closed piece of rope and, without cutting it, return it
to us as knotted in either left or right form. All the properties of knots
are extrinsic properties of toruses (or, if you prefer, one-dimensional
curves that may be thought of as toruses whose meridians have shrunk
to points) that are embedded in 3-space.
It is not always easy to decide intuitively if a given surface in 3-space
can be elastically deformed to a different but topologically equivalent
surface. A striking instance, discussed nearly 50 years ago (see Albert
W. Tucker and Herbert S. Bailey, Jr., "Topology," Scientific American,
January 1950), concerns a rubber torus with a hole in its surface. Can
it be turned inside out to make a torus of identical shape? The answer
is yes. It is hard to do with a rubber model (such as an inner tube), but
a model made of wool reverses readily. Stephen Barr, in his Second
Miscellany of Puzzles (Macmillan, 1969), recommends making it from
a square piece of cloth. Fold the cloth in half and sew together oppo-
site edges to make a tube. Now sew the ends of the tube together to
make a torus that is square shaped when flattened. For ease in revers-
ing, the surface hole is a slot cut in the outer layer of cloth (shown by
the broken line in Figure 20.2).
After the cloth torus is turned inside out, it is exactly the same shape
Doughnuts: Linked and Knotted 255
2
3
)
J
I
\.)
20.2. torus
256
b
20.3. Torus and
257
Figure 20.4. Three varieties of a two-hole torus
Figure 20.5. R. H. Bing's proof
is moved over the cube's surface, as indicated by the arrows, dragging
the tube along with it. It goes left to the base of the other tube, climbs
that tube's side, moves to the right across the top of the cube, circles its
top hole counterclockwise, continues left around the other hole, over
the cube's front edge, down the front face, around the lower edge to the
cube's bottom face, and then across that face to the position it formerly
occupied. It is easy to see that the tube attached to this hole has been
untied. Naturally the procedure is reversible. If you had a sufficiently
pliable doughnut surface with two holes, you could manipulate it until
one hole became a knot tied around the other.
Topologists worried for decades about whether two separate knots
side by side on a closed rope could cancel each other; that is, could the
rope be manipulated until both knots dissolved? No pair of canceling
258 TOPOLOGY
knots had been found, but proving the impossibility of such a pair was
another matter. It was not even possible to show that two trefoil knots
of opposite handedness could not cancel. Proofs of the general case
were not found until the early 1950s. One way of proving it is ex-
plained by Ralph H. Fox in "A Quick Trip through Knot Theory," in
Topology of 3-Manifolds and Related Topics, edited by M. K. Fort, Jr.
(Prentice-Hall, 1963). It is a reductio ad absurdum proof that unfortu-
nately involves the sophisticated concept of an infinity of knots on a
closed curve and certain assumptions about infinite sets that must be
carefully specified to make the proof rigorous.
When John Horton Conway, a former University of Cambridge math-
ematician now at Princeton University, was in high school, he hit on a
simpler proof that completely avoids infinite sets of knots. Later he
learned that essentially the same proof had been formulated earlier,
but I have not been able to determine by whom. Here is Conway's ver-
sion as he explained it years ago in a letter. It is a marvelous example
of how a knotted torus can play an unexpected role in proving a fun-
damental theorem of modern knot theory.
Conway's proof, like the one for the infinite knots, is a reductio ad ab-
surdum. We begin by imagining that a closed string passes through the
opposite walls of a room (see Figure 20.6). Since we shall be concerned
only with what happens inside the room, we can forget about the string
outside and regard it as being attached to the side walls. On the string
are knots A and B. Each is assumed to be genuine in the sense that it
cannot be removed by manipulating the string if it is the only knot on
the string. It also is assumed that the two knots will cancel each other
when both are on the same closed curve. The proof applies to pairs of
knots of any kind whatever, but here we show the knots as simple tre-
foils of opposite parity. If the knots can cancel, it means that the string
can be manipulated until it stretches straight from wall to wall. Think
of the string as being elastic to provide all the needed slack for such an
operation. In the center figure we introduce an elastic torus around the
string. Note that the tube "swallows" knot A but "circumnavigates"
knot B (Conway's terminology). Any parallel drawn on this tube, on the
section between the walls, obviously must be knotted in the same way
as knot B. Indeed, it can be shown that any line on the tube's surface,
stretching from wall to wall and never crossing itself at any spot on the
tube's surface, will be knotted like knot B.
"Now," writes Conway, "comes the crunch." Perform on the string
Doughnuts: Linked and Knotted 259
A B
Figure 20.6. John Horton Conway's proof
the operation that we assumed would dissolve both knots. This can be
done without breaking the tube. Because the string is never allowed to
pass through itself during the deformation, we can always push the
tube's wall aside if it gets in the way. The third drawing in Figure 20.6
shows the final result. The string is unknotted. The tube may have
reached a horribly complicated shape impossible to draw. Consider a
vertical plane passing through the straight string and cutting the
twisted tube. We can suppose that the tube's cross section will look
something like what is shown with the possibility of various "islands,"
but there will necessarily be two lines, XYand MN, from wall to wall
that do not cross themselves at any point on the vertical plane. Each
line will be unknotted. Moreover, each line also is a curve that does not
cross itself on the tube's surface. As we have seen, all such lines were
(before the deformation) knotted like knot B. The deformation has
therefore removed a knot equivalent to knot B from each of these two
260 TOPOLOGY
20.1. "" LJ--J.'-LJ.J.O toruses
261
There are doughnuts and doughnuts
with knots and with no knots
and many a doughnut
so nuts that we know not.
Here are three more toroidally knotty questions.
1. How many closed curves can be drawn on a torus, each a trefoil knot
of the same handedness, so that no two curves cross each other at any
point?
2. If two closed curves are drawn on a torus so that each forms a trefoil
knot but the knots are of opposite parity, what is the minimum num-
ber of points at which the two curves will intersect each other?
3. Show how to cut a solid two-hole doughnut with one slice of a knife
so that the result is a solid outside-knotted torus. The "slice" is not,
of course, planar. More technically, show how to remove from a two-
hole doughnut a section topologically equivalent to a disk so that
what remains is a solid knotted torus. (This amusing result was dis-
covered by John Stallings in 1957 and communicated to me by James
Stasheff.)
Addendum
In studying the properties of topological surfaces, one must al-
ways keep in mind the distinction between intrinsic properties, inde-
pendent of the space in which the surface is embedded, and properties
that arise from the embedding. The "complement" of a surface con-
sists of all the points in the embedding space that are not in the surface.
For example, a torus with no knot, one with an outside knot, and one
with an inside knot all have identical intrinsic properties. No two have
topologically identical complements; hence, no two are equivalent in
their extrinsic topological properties.
John Stillwell, a mathematician at Monash University, Australia, sent
several fascinating letters in which he showed how an unknotted torus
with any number of holes-such toruses are equivalent to the surfaces
<;>f spheres with handles-could be turned inside out through a surface
hole. He was not sure whether a knotted torus, even with only one
hole, could be turned inside out through a hole in its surface.
Many beautiful, counterintuitive problems involving links and knots
in toruses have been published. See Rolfsen's book, cited in the bibli-
262 TOPOLOGY
20.8. can
ence
torus A?
on
are
two
263
2. Two topologists were discussing at lunch the two linked surfaces
shown at the left in Figure 20.10, which one of them had drawn on
a paper napkin. You must not think of these objects as solids, like
ropes or solid rubber rings. They are the surfaces of toruses, one sur-
face of genus 1 (one hole), the other of genus 2 (two holes).
Figure 20.10.
Thinking in the mode of "rubber-sheet geometry," assume that the
surfaces in the illustration can be stretched or shrunk in any desired
way provided there is no tearing or sticking together of separate parts.
Can the two-hole torus be deformed so that one hole becomes un-
linked as is shown at the right in the illustration?
Topologist X offers the following impossibility proof. Paint a ring
on each torus as is shown by the black lines. At the left the rings are
linked. At the right they are unlinked.
"You will agree," says X, "that it is impossible by continuous de-
formation to unlink two linked rings embedded in three-dimensional
space. It therefore follows that the transformation is impossible."
"But it doesn't follow at all," says Y.
Who is right? I am indebted to Herbert Taylor for discovering and
sending this mystifying problem.
3. Figure 20.11 depicts a familiar linkage known as the Borromean
rings. No two rings are linked, yet all three are linked. It is impossi-
ble to separate them without cutting at least one torus. Show how it
is possible for any number of toruses to form a linked structure even
though no two of them are linked.
4. Are the two structures shown in Figure 20.12 topologically the same?
That is, can one be transformed to the other by a continuous defor-
mation?
264 TOPOLOGY
figure 20.1 J.
an ... JLJ. ............ .LJl ................ can re-
versed through a hole to produce an externally knotted (see Fig-
13). to cover almost
to
its original size.
2D.13. to
in reversing the unknotted through a hole, the deformation
appears the same
......... A7'l"!."'V ........... that initially it is a parallel circling the torus's elongated hole,
whereas after the
no
Doughnuts: Linked Knotted 265
is now closed at both ends. As indicated by the arrow, the hole is now
surrounded by the knotted tube.
Piet Hein's two-hole torus, with an internal knot passing through an
external one, is easily shown to be the same as a two-haler with only
an external knot. Simply slide one end of the inside knot around the
outside knot (in the manner explained earlier) and back to its starting
point. This unties the internal knot. Piet Hein's two-haler, with the ex-
ternal knot going through a hole, can be unknotted by the deformation
shown in Figure 20.14.
5 6 7
Figure 20.14. Unknotting a two-hole torus
Answers to the first three toroidal questions are as follows:
1. An infinity of noncrossing closed curves, each knotted with the same
handedness, can be drawn on a torus (see Figure 20.15, top). If a
torus surface is cut along any of these curves, the result is a two-
sided, knotted band.
2. Two closed curves on a torus, knotted with opposite handedness,
will intersect each other at least 12 times.
3. A rotating slice through a solid two-hole doughnut is used to produce
a solid that is topologically equivalent to a solid, knotted torus (see
Figure 20.15, bottom). Think of a short blade as moving downward
and rotating one and a half turns as it descends. If the blade does not
turn at all, two solid toruses result. A half-turn produces one solid,
266 TOPOLOGY
8
b
J,,, ...u,u .... <:> ...... curves on a torus .. s..J..I., ......... ,;::O.1J. a
1.
267
The torus with the mouth can, however, swallow the other one so
that the eaten torus is inside in the first sense explained above. Fig-
ure 20.16 shows how it is done. In the process it is necessary for the
cannibal torus to turn inside out.
A good way to understand what happens is to imagine that torus
A is shrunk until it becomes a stripe of paint that circles B. Tum A
inside out through its mouth. The painted stripe goes inside, but in
Figure 20.16. How one torus eats another
268
1. Mouth starts
to open.
2. It lengthens to
an enonnous grin.
3. Grin widens until
torus becomes
two attached bands.
4. Horizontal band
enlarges, vertical
band shrinks.
5. Vertical band
widens and creeps
around victim.
6. Mouth closes.
TOPOLOGY
doing so it ends up circling B's hole. Expand the stripe back to a
torus and you have the final picture of the sequence.
2. Figure 20.17 shows how a continuous deformation of the two-hole
torus will unlink one of its holes from the single-hole torus. The ar-
gument for the impossibility of this task fails because if a ring is
painted around one hole (as is shown by the black line), the ring be-
comes distorted in such a way that after the hole is unlinked the
painted ring remains linked through the one-hole torus.
A B c
D E F
Figure 20.11.
For a mind-boggling selection of similar problems involving linked
toruses, see Herbert Taylor's article "Bicycle Tubes Inside Out," in
The Mathematical Gardner (Wadsworth, 1981), edited by David
Klarner.
3. Figure 20.18 shows a circular chain that obviously can be enlarged
to include any number of links. No two toruses are linked. If one
link is cut and removed, all the others are free of one another.
4. The two forms are topologically identical. To prove this, imagine the
linked form deformed to a sphere with two "handles" as shown in
Figure 20.19.
Now imagine the base of one handle being moved over the surface,
by shrinking the surface in front and stretching it in back, along the
path shown by the dotted line. This links the two handles. The struc-
ture is now easily altered to correspond with the linked form of the
toroids.
Doughnuts: Linked and Knotted 269
20.18.
20.19.
VI
Probability
Chapter 21 Probability
and Ambiguity
Charles Sanders Peirce once observed that in no other branch
of mathematics is it so easy for experts to blunder as in probability the-
ory. History bears this out. Leibniz thought it just as easy to throw 12
with a pair of dice as to throw 11. Jean Ie Rond d' Alembert, the great
18th-century French mathematician, could not see that the results of
tossing a coin three times are the same as tossing three coins at once,
and he believed (as many amateur gamblers persist in believing) that
after a long run of heads, a tail is more likely.
Today, probability theory provides clear, unequivocal answers to
simple questions of this sort, but only when the experimental proce-
dure involved is precisely defined. A failure to do this is a common
source of confusion in many recreational problems dealing with
chance. A classic example is the problem of the broken stick. If a stick
is broken at random into three pieces, what is the probability that the
pieces can be put together in a triangle? This cannot be answered with-
out additional information about the exact method of breaking to be
used.
One method is to select, independently and at random, two points
from the points that range uniformly along the stick, then break the
stick at these two points. If this is the procedure to be followed, the an-
swer is 1/4, and there is a neat way of demonstrating it with a geomet-
rical diagram. We draw an equilateral triangle, then connect the
midpoints of the sides to form a smaller shaded equilateral triangle in
the center (see Figure 21.1). If we take any point in the large triangle
and draw perpendiculars to the three sides, the sum of these three lines
will be constant and equal to the altitude of the large triangle. When
this point, like point A, is inside the shaded triangle, no one of the
three perpendiculars will be longer than the sum of the other two.
273
prC)baltJillty is 1/4
274
represent the smaller piece. In order for this line to be smaller than the
sum of the other two perpendiculars, the point where the lines meet
cannot be inside the small triangle at the top of the diagram. It must
range uniformly over the lower three triangles. The shaded triangle
continues to represent favorable points, but now it is only one-third the
area under consideration. The chances, therefore, are 1/3 that when
we break the larger piece, the three pieces will form a triangle. Since
our chance of picking the larger piece is 1/2, the answer to the original
question is the product of 1/2 and 1/3, or 1/6.
Geometrical diagrams of this sort must be used with caution because
they too can be fraught with ambiguity. For example, consider this
problem discussed by Joseph Bertrand, a famous French mathemati-
cian. What is the probability that a chord drawn at random inside a cir-
cle will be longer than the side of an equilateral triangle inscribed in the
circle?
We can answer as follows. The chord must start at some point on the
circumference. We call this point A, then draw a tangent to the circle
at A, as shown in the top illustration of Figure 21.2. The other end of
the chord will range uniformly over the circumference, generating an
infinite series of equally probable chords, samples of which are shown
on the illustration as broken lines. It is clear that only those chords that
cut across the triangle are longer than the side of the triangle. Since the
angle of the triangle at A is 60 degrees, and since all possible chords lie
within a 180-degree range, the chances of drawing a chord larger than
the side of the triangle must be 60/180, or 113.
Now let us approach the same problem a bit differently. The chord
we draw must be perpendicular to one of the circle's diameters. We
draw the diameter, then add the triangle as shown in the illustration at
bottom left of Figure 21.2. All chords perpendicular to this diameter
will pass through a point that ranges uniformly a] ong the diameter.
Samples of these chords are again shown as broken lines. It is not hard
to prove that the distance from the center of the circle to A is half the
radius. Let B mark the midpoint on the other side of the diameter. It is
now easy to see that only those chords crossing the diameter between
A and B will be longer than the side of the triangle. Since AB is half the
diameter, we obtain an answer to our problem of 1/2.
Here is a third approach. The midpoint of the chord will range uni-
formly over the entire space within the circle. A study of the illustra-
tion at bottom right of Figure 21.2 will convince you that only chords
Probability and Ambiguity 275
on it,
we draw the chord on which the fly is the midpoint. The probability
that this chord is longer than the side of the triangle is 1/4.
Each of these procedures is a legitimate method of obtaining a "ran-
dom chord." The problem as originally stated, therefore, is ambiguous.
It has no answer until the meaning of "draw a chord at random" is
made precise by a description of the procedure to be followed. Appar-
ently nothing resembling any of the three procedures is actually
adopted by most people when they are asked to draw a random chord.
In an interesting unpublished paper entitled "The Human Organism as
a Random Mechanism" Oliver L. Lacey, professor of psychology at the
University of Alabama, reports on a test which showed the probability
to be much better than 1/2 that a subject would draw a chord longer
than the side of the inscribed triangle.
Another example of ambiguity arises from a failure to specify the
randomizing procedure. Readers were told that Mr. Smith had two chil-
dren, at least one of whom was a boy, and were asked to calculate the
probability that both were boys. Many readers correctly pointed out
that the answer depends on the procedure by which the information "at
least one is a boy" is obtained. If from all families with two children,
at least one of whom is a boy, a family is chosen at random, then the an-
swer is 1/3. But there is another procedure that leads to exactly the
same statement of the problem. From families with two children, one
family is selected at random. If both children are boys, the informant
says "at least one is a boy. If both are girls, he says "at least one is a girl."
And if both sexes are represented, he picks a child at random and says
"at least one is a ... ," naming the child picked. When this procedure
is followed, the probability that both children are of the same sex is
clearly 1/2. (This is easy to see because the informant makes a state-
ment in each of the four cases-BB, BG, GB, GG-and in half of these
cases both children are of the same sex.)
The following wonderfully confusing little problem involving three
prisoners and a warden is even more difficult to state unambiguously.
Three men-A, B, and C-were in separate cells under sentence of
death when the governor decided to pardon one of them. He wrote
their names on three slips of paper, shook the slips in a hat, drew out
one of them, and telephoned the warden, requesting that the name of
the lucky man be kept secret for several days. Rumor of this reached
Probability and Ambiguity 277
prisoner A. When the warden made his morning rounds, A tried to
persuade the warden to tell him who had been pardoned. The warden
refused.
"Then tell me," said A, "the name of one of the others who will be
executed. If B is to be pardoned, give me C's name. If C is to be par-
doned, give me B's name. And if I'm to be pardoned, flip a coin to de-
cide whether to name B or C."
"But if you see me flip the coin," replied the wary warden, "you'll
know that you're the one pardoned. And if you see that I don't flip a
coin, you'll know it's either you or the person I don't name."
"Then don't tell me now," said A. "Tell me tomorrow morning."
The warden, who knew nothing about probability theory, thought it
over that night and decided that if he followed the procedure suggested
by A, it would give A no help whatever in estimating his survival
chances. So next morning he told A that B was going to be executed.
After the warden left, A smiled to himself at the warden's stupidity.
There were now only two equally probable elements in what mathe-
maticians like to call the "sample space" of the problem. Either C
would be pardoned or himself, so by all the laws of conditional prob-
ability, his chances of survival had gone up from 1/3 to 1/2.
The warden did not know that A could communicate with C, in an
adjacent cell, by tapping in code on a water pipe. This A proceeded to
do, explaining to C exactly what he had said to the warden and what
the warden had said to him. C was equally overjoyed with the news be-
cause he figured, by the same reasoning used by A, that his own sur-
vival chances had also risen to 1/2.
Did the two men reason correctly? If not, how should each have cal-
culated his chances of being pardoned?
Addendum
In giving the second version of the broken stick problem I could
hardly have picked a better illustration of the ease with which experts
can blunder on probability computations, and the dangers of relying on
a geometrical diagram. My solution was taken from William A. Whit-
worth's DCC Exercises in Choice and Chance, Problem 677; the same
answer will be found in many other older textbooks on probability. It
is entirely wrong!
278 PROBABILITY
In the first version of the problem, in which the two breaking points
are simultaneously chosen, the representative point on the diagram
ranges uniformly over the large triangle, permitting a comparison of
areas to obtain a correct answer. In the second version, in which the
stick is broken, then the larger piece is broken, Whitworth assumed that
the point on the diagram ranged uniformly over the three lower trian-
gles. It doesn't. There are more points within the central triangle than
in the other two.
Let the length of the stick be 1 and x be the length of the smallest
piece after the first break. To obtain pieces that will form a triangle,
the larger segment must be broken within a length equal to 1 - x.
Therefore the probability of obtaining a triangle is xll - x. We now
have to average all values of x, from 0 to 1/2, to obtain a value for this
expression. It proves to be - 1 + 2 log 2, or .386. Since the probability
is 1/2 that the larger piece will be picked for breaking, we multiply
.386 by 1/2 to obtain .193, the answer to the problem. This is a trifle
larger than 1/6, the answer obtained by following Whitworth's rea-
soning.
A large number of readers sent very clear analyses of the problem. In
the above summary, I followed a solution sent by Mitchell P. Marcus of
Binghamton, NY. Similar solutions were received from Edward Adams,
Howard Grossman, Robert C. James, Gerald R. Lynch, G. Bach and R.
Sharp, David Knaff, Norman Geschwind, and Raymond M. Redheffer.
Professor Redheffer, at the University of California, is coauthor (with
Ivan S. Sokolnikoff1 of Mathematics of Physics and Modern Engineer-
ing (McGraw-Hill, 1958), in which will be found (p. 636) a full discus-
sion of the problem. See also Ingenious Mathematical Problems and
Methods by L. A. Graham (Dover, 1959, Problem 32) for other methods
of solving the problem's first version.
Frederick R. Kling, John Ross, and Norman Cliff, all with the Educa-
tional Testing Service, Princeton, NJ, also sent a correct solution of the
problem's second version. At the close of their letter they asked which
of the following three hypotheses was most probable:
1. Mr. Gardner honestly blundered.
2. Mr. Gardner deliberately blundered in order to test his readers.
3. Mr. Gardner is guilty of what is known in the mathematical world
as keeping up with the d' Alemberts.
The answer: number three.
Probability and Ambiguity 279
Answers
The answer to the problem of the three prisoners is that A's
chances of being pardoned are 1/3, and that Cs chances are 2/3.
Regardless of who is pardoned, the warden can give A the name of a
man, other than A, who will die. The warden's statement therefore has
no influence on A's survival chances; they continue to be 1/3.
The situation is analogous to the following card game. Two black
cards (representing death) and a red card (the pardon) are shuffled and
dealt to three men: A, B, C (the prisoners). If a fourth person (the war-
den) peeks at all three cards, then turns over a black card belonging to
either B or C, what is the probability that A's card is red? There is a
temptation to suppose it is 1/2 because only two cards remain face-
down, one of which is red. But since a black card can always be shown
for B or C, turning it over provides no information of value in betting
on the color of A's card.
This is easy to understand if we exaggerate the situation by letting
death be represented by the ace of spades in a full deck. The deck is
spread, and A draws a card. His chance of avoiding death is 51/52.
Suppose now that someone peeks at the cards, then turns face up 50
cards that do not include the ace of spades. Only two face-down cards
are left, one of which must be the ace of spades, but this obviously
does not lower A's chances to 112. It doesn't because it is always pos-
sible, if one looks at the faces of the 51 cards, to find 50 that do not in-
clude the ace of spades. Finding them and turning them face up,
therefore, has no effect on A's chances. Of course if 50 cards are turned
over at random, and none prove to be the ace of spades, then the chance
that A drew the death card does rise to 1/2.
What about prisoner C? Since either A or C must die, their respective
probabilities for survival must add up to 1. A's chances to live are 1/3;
therefore Cs chances must be 2/3. This can be confirmed by consider-
ing the four possible elements in our sample space, and their respective
initial probabilities:
1. C is pardoned, warden names B (probability 1/3).
2. B is pardoned, warden names C (probability 1/3).
3. A is pardoned, warden names B (probability 1/6).
4. A is pardoned, warden names C (probability 1/6).
280 PROBABILITY
Only cases 1 and 3 apply when it becomes known that B will die. The
chances that it is case 1 are 1/3, or twice the chances (1/6) that it is case
3, so Cs survival chances are two to one, or 213. In the card-game model
this means that there is a probability of 2/3 that Cs card is red.
This problem of the three prisoners brought a flood of mail, pro and
con; happily, all objections proved groundless. Sheila Bishop of East
Haven, CT, sent the following well-thought-out analysis:
SIRS:
I was first led to the conclusion that A's reasoning was incorrect by the
following paradoxical situation. Suppose the original conversation be-
tween A and the warden had taken place in the same way, but now sup-
pose that just as the warden was approaching A's cell to tell him that B
would be executed, the warden fell down a manhole or was in some
other way prevented from delivering the message.
A could then reason as follows: "Suppose he was about to tell me that
B would be executed. Then my chance of survival would be 1/2. If, on
the other hand, he was going to tell me that C would be executed, then
my chances would still be 1/2. Now I know as a certain fact that he
would have told me one of those two things; therefore, either way, my
survival chances are bound to be 1/2." Following this line of thought
shows that A could have figured his chances to be 1/2 without ever ask-
ing the warden anything!
After a couple of hours I finally arrived at this conclusion: Consider a
large number of trios of prisoners all in this same situation, and in each
group let A be the one who talks to the warden. If there are 3n trios al-
together, then in n of them A will be pardoned, in n B will be pardoned,
and in n C will be pardoned. There will be 3nl2 cases in which the war-
den will say, "B will be executed." In n of these cases C will go free and
in nl2 cases A will go free; Cs chances are twice as good as A's. Hence
A's and C's chances of survival and 1/3 and 2/3 respectively ....
Lester R. Ford, Jr., and David N. Walker, both with the Arizona office
of General Analysis Corporation, felt that the warden has been unjustly
maligned:
SIRS:
We are writing to you on behalf of the warden, who is a political ap-
pointee and therefore unwilling to enter into controversial matters in his
own behalf.
You characterize him in a slurring manner as "The warden, who knew
Probability and Ambiguity 281
nothing about probability theory, ... " and I feel that a grave injustice is
being done. Not only are you incorrect (and possibly libelous), but I can
personally assure you that his hobby for many years has been mathe-
matics, and in particular, probability theory. His decision to answer A's
question, while based on a humanitarian attempt to brighten the last
hours of a condemned man (for, as we all now know, it was C who re-
ceived the pardon), was a decision completely compatible with his in-
structions from the governor.
The only point on which he is open to criticism (and on this he has
already been reprimanded by the governor) is that he was unable to pre-
vent A from communicating with C, thereby permitting Cto more accu-
rately estimate his chances of survival. Here too, no great damage was
done, since C failed to make proper use of the information.
If you do not publish both a retraction and an apology, we shall feel
impelled to terminate our subscription.
Addendum
The problem of the two boys, as I said, must be very carefully
stated to avoid ambiguity that prevents a precise answer. In my Aha,
Gotcha I avoided ambiguity by imagining a lady who owned two par-
rots--one white, one black. A visitor asks the owner, "Is one bird a
male?" The owner answers yes. The probability both parrots are male
is 1/3. Had the visitor asked, "Is the dark bird a male?'" a yes answer
would have raised the probability that both birds are male to 1/2.
Richard E. Bedient, a mathematician at Hamilton College, described
the prisoner's paradox in a poem that appeared in The American Math-
ematical Monthly, Vol. 101, March 1994, page 249:
282
THE PRISONER'S PARADOX REVISITED
Awaiting the dawn sat three prisoners wary,
A trio of brigands named Tom, Dick and Mary.
Sunrise would signal the death knoll of two,
Just one would survive, the question was who.
Young Mary sat thinking and finally spoke.
To the jailer she said, "You may think this a joke"
But it seems that my odds of surviving 'til tea,
Are clearly enough just one out of three.
But one of my cohorts must certainly go,
Without question, that's something I already know.
PROBABILITY
Telling the name of one who is lost,
Can't possibly help me. What could it cost?"
The shriveled old jailer himself was no dummy,
He thought, But why not?" and pointed to Tommy.
"Now it's just Dick and [" Mary chortled with glee,
((One in two are my chances, and not one in three!"
Imagine the jailer's chagrin, that old elf,
She'd tricked him, or had she? Decide for yourself.
When I introduced the three prisoners paradox in my October 1959
column, I received a raft of letters from mathematicians who believed
my solution was invalid. The number of such letters, however, was
small compared to the thousands of letters Marilyn vos Savant received
when she gave a version of the problem in her popular Parade column
for September 9,1990.
Ms. Savant's version of the paradox was based on a then-popular
television show called Let's Make a Deal, hosted by Monty Hall. Imag-
ine three doors, Marilyn wrote, to three rooms. Behind one door is a
prize car. Behind each of the other two doors is a goat. A guest on the
show is given a chance to win the prize by selecting the door with the
car. If she chooses at random, clearly the probability she will select the
prize door is 1/3. Now suppose, that after the guest's selection is voiced,
Monty Hall, who knows what is behind each door, opens one door to
disclose a goat. Two closed doors remain. One might reason that be-
cause the car is now behind one of just two doors, the probability the
guest had chosen the correct door has risen to 1/2. Not so! As Marilyn
correctly stated, it remains 1/3. Because Monty can always open a door
with a goat, his opening such a door conveys no new information that
alters the 1/3 probability.
Now comes an even more counterintuitive result. If the guest
switches her choice from her initial selection to the other closed door,
her chances of winning rise to 2/3. This should be obvious if one grants
that the probability remains 1/3 for the first selection. The car must be
behind one of the two doors, therefore the probabilities for each door
must add to 1, or certainty. If one door has a probability of 1/3 being
correct, the other door must have a 2/3 probability.
Marilyn was flooded with letters from irate readers, many accusing
her of being ignorant of elementary probability theory and many from
professional mathematicians. So awesome was the mail, and so con-
Probability and Ambiguity 283
troversial, that The New York Times, on July 21, 1991, ran a front page,
lengthy feature about the flap. The story, written by John Tierney, was
titled "Behind Monty Hall's Doors: Puzzle, Debate and Answer?" (See
also letters about the feature in The Times, August 11,1991.)
The red-faced mathematicians, who were later forced to confess they
were wrong, were in good company. Paul Erdos, one of the world's
greatest mathematicians, was among those unable to believe that
switching doors doubled the probability of success. Two recent bi-
ographies of the late Erdos reveal that he could not accept Marilyn's
analysis until his friend Ron Graham, of Bell Labs, patiently explained
it to him.
The Monty Hall problem, as it came to be known, generated many ar-
ticles in mathematical journals. I list some of them in this chapter's bib-
liography.
Bibliography
The Three Prisoners Paradox
D. H. Brown, "The Problem of the Three Prisoners," Mathematics Teacher, Febru-
ary 1966, pp. 181-82.
R. Falk, "A Closer Look at the Probabilities ofthe Notorious Three Prisoners." Cog-
nition, Vol. 43,1992, pp. 197-223.
S. Ichikawa and H. Takeichi, "Erroneous Beliefs in Estimating Posterior Probabil-
ity," Bahaviormetrika, No. 27,1990, pp. 59-73.
S. Shimojo and S. Ichikano, "Intuitive Reasoning About Probability: Theoretical
and Experiential Analysis of Three Prisoners," Cognition, Vol. 32, 1989, pp.
1-24.
N. Starr, "A Paradox in Probability Theory," Mathematics Teacher, February 1973,
pp.166-68.
The Monty Hall Problem
E. Barbeau, "Fallicies, Flaws, and Flimflams," The College Mathematics Journal,
Vol. 26, May 1995, pp. 132-84.
A. H. Bohl, M. J. Liberatore, and R. L. Nydick, "A Tale of Two Goats ... and a Car,"
Journal of Recreational Mathematics, Vol. 27, 1995, pp. 1-9.
J. P. Georges and T. V. Craine, "Generalizing Monty's Dilemma," Quantum,
March/April1995, pp. 17-21.
L. Gillman, "The Car and the Goats." American Mathematical Monthly, Vol. 99,
1992, pp. 3-7.
A. Lo Bello, "Ask Marilyn: The Mathematical Controversy in Parade Magazine,"
Mathematical Gazette, Vol, 75, October 1991, pp. 271-72.
284 PROBABILITY
J. Paradis, P. Viader, and L. Bibiloni, "A Mathematical Excursion: From the Three-
door Problem to a Cantortype Set." American Mathematical Monthly, VoL 106,
March 1999, pp. 241-51.
J. M. Shaughnessy and T. Dick, "Monte's Dilemma: Should You Stick or Switch?"
Mathematics Teacher, April 1991, pp. 252-56.
M. vos Savant, "Ask Marilyn," Parade, Septembr 9,1990; December 2,1990; Feb-
ruary 17, 1991; July 7, 1991; September 8, 1991; October 13 1991; January 5,
1992; January 26,1992.
Probability and Ambiguity 285
Chapter 22 Nontransitive Dice
and Other Paradoxes
Probability theory abounds in paradoxes that wrench common
sense and trap the unwary. In this chapter we consider a startling new
paradox involving the relation called transitivity and a group of para-
doxes stemming from the careless application of what is called the
principle of indifference.
Transitivity is a binary relation such that if it holds between A and B
and between Band C; it must also hold between A and C. A common
example is the relation "heavier than." If A is heavier than Band B is
heavier than C; then A is heavier than C. The three sets of four dice
shown "unfolded" in Figure 22.1 were designed by Bradley Efron, a sta-
tistician at Stanford University, to dramatize some discoveries about a
general class of probability paradoxes that violate transitivity. With any
of these sets of dice you can operate a betting game so contrary to in-
tuition that experienced gamblers will find it almost impossible to com-
prehend even after they have completely analyzed it.
The four dice at the top of the illustration are numbered in the sim-
plest way that provides the winner with the maximum advantage.
Allow someone to pick any die from this set. You then select a die
from the remaining three. Both dice are tossed and the person who gets
the highest number wins. Surely, it seems, if your opponent is allowed
the first choice of a die before each contest, the game must either be fair
or favor your opponent. If at least two dice have equal and maximum
probabilities of winning, the game is fair because if he picks one such
die, you can pick the other; if one die is better than the other three, your
opponent can always choose that die and win more than half of the
contests. This reasoning is completely wrong. The incredible truth is
that regardless of which die he picks you can always pick a die that has
a 2/3 probability of winning, or two-to-one odds in your favor!
286

10 4
__________________
Figure 22.1. Nontransitive dice
The paradox (insofar as it violates common sense) arises from the
mistaken assumption that the relation "more likely to win" must be
transitive between pairs of dice. This is not the case with any of the
three sets of dice. In each set the relation "more likely to win" is indi-
cated by an arrow that points to the losing die. Die A beats B, B beats
C, Cbeats D-and D beats A! In the first set the probability of winning
with the indicated die of each pair is 2/3. This is easily verified by list-
ing the 36 possible throws of each pair, then checking the 24 cases in
which one die bears the highest number.
The other two sets of four dice, also designed by Efron, have the
same nontransitive property but fewer numbers are repeated in order
to make an analysis of the dice more difficult. In the second set the
probability of winning with the indicated die is also 2/3. Because ties
are possible with the third set it must be agreed that ties will be broken
by rolling again. With this procedure the winning probability for each
of the four pairings in the third set is 11/17, or .647.
It has been proved, Efron writes, that 2/3 is the greatest possible ad-
vantage that can be achieved with four dice. For three sets of numbers
the maximum advantage is .618, but this cannot be obtained with dice
because the sets must have more than six numbers. If more than four
sets are used (numbers to be randomly selected within each set), the
possible advantage approaches a limit of 3/4 as the number of sets in-
creases.
A fundamental principle in calculating probabilities such as dice
Nontransitive Dice and Other Paradoxes 287
throws is one that goes back to the beginnings of classical probability
theory in the 18th century. It was formerly called "the principle of in-
sufficient reason" but is now known as "the principle of indifference/'
a crisper phrase coined by John Maynard Keynes in A Treatise on Prob-
ability. (Keynes is best known as an economist, but his book on proba-
bility has become a classic. It had a major influence on the inductive
logic of Rudolf Carnap.) The principle is usually stated as follows: If
you have no grounds whatever for believing that anyone of n mutually
exclusive events is more likely to occur than any other, a probability of
1/ n is assigned to each.
For example, you examine a die carefully and find nothing that fa-
vors one side over another, such as concealed loads, noncubical shape,
beveling of certain edges, stickiness of certain sides, and so on. You as-
sume that there are six equally probable ways the cube can fall; there-
fore you assign a probability of 1/6 to each. If you toss a penny, or play
the Mexican game of betting on which of two sugar cubes a fly will
alight on first, your ignorance of any possible bias prompts you to as-
sign a probability of 1/2 to each of the two outcomes. In none of these
samples do you feel obligated to make statistical, empirical tests. The
probabilities are assigned a priori. They are based on symmetrical fea-
tures in the structures and forces involved. The die is a regular solid,
the probability of the penny's balancing on its edge is virtually zero,
there is no reason for a fly to prefer one sugar cube to another, and so
on. Ultimately, of course, your analysis rests on empirical grounds,
since only experience tells you, say, that a weighted die face would af-
fect the odds, whereas a face colored red (with the others blue) would
not.
Some form of the principle of indifference is indispensable in prob-
ability theory, but it must be carefully qualified and applied with ex-
treme caution to avoid pitfalls. In many cases the traps spring from a
difficulty in deciding which are the equally probable cases. Suppose,
for instance, you shuffle a packet of four cards-two red, two black-
and deal them face down in a row. Two cards are picked at random, say
by placing a penny on each. What is the probability that those two
cards are the same color?
One person reasons: "There are three equally probable cases. Either
both cards are black, both are red, or they are different colors. In two
cases the cards match, therefore the matching probability is 2/3."
"No," another person counters, "there are four equally probable
288 PROBABILITY
cases. Either both cards are black, both are red, card x is black and y is
red, or x is red and y is black. More simply, the cards either match or
they do not. In each way of putting it the matching probability clearly
is 1/2."
The fact is that both people are wrong. (The correct probability will
be given in the Answer Section. Can the reader calculate it?) Here the
errors arise from a failure to identify correctly the equally probable
cases. There are, however, more confusing paradoxes-actually fallac-
ies-in which the principle of indifference seems intuitively to be ap-
plicable, whereas it actually leads straight to a logical contradiction.
Cases such as these result when there are no positive reasons for be-
lieving n events to be equally probable and the assumption of equiprob-
ability is therefore based entirely, or almost entirely, on ignorance.
For example, someone tells you: "There is a cube in the next room
whose size has been selected by a randomizing device. The cube's edge
is not less than one foot or more than three feet." How would you esti-
mate the probability that the cube's edge is between one and two feet
as compared with the probability that it is between two and three feet?
In your total ignorance of additional information, is it not reasonable to
invoke the principle of indifference and regard each probability as 1/2?
It is not. If the cube's edge ranges between one and two feet, its vol-
ume ranges between 1
3
, or one, cubic foot and 2
3
, or eight, cubic feet.
But in the range of edges from two to three feet, the volume ranges be-
tween 2
3
(eight) and 3
3
(27) cubic feet-a range almost three times the
other range. If the principle of indifference applies to the two ranges of
edges, it is violated by the equivalent ranges of volume. You were not
told how the cube's "size" was randomized, and since "size" is am-
biguous (it could mean either the cube's edge or its volume) you have
no clues to guide your guessing. If the cube's edge was picked at ran-
dom, the principle of indifference does indeed apply. It is also applic-
able if you are told that the cube's volume was picked at random, but
of course you then have to assign a probability of 1/2 to each of the two
ranges from one to 14 and from 14 to 27 cubic feet, and to the corre-
sponding ranges for the cube's edge. If the principle applies to the edge,
it cannot apply to the volume without contradiction, and vice versa.
Since you do not know how the size was selected, any application of
the principle is meaningless.
Carnap, in attacking an uncritical use of the principle in Harold Jef-
freys' Theory of Probability, gives the following example of its misuse.
Nontransitive Dice and Other Paradoxes 289
You know that every ball in an urn is blue, red, or yellow, but you
know nothing about how many balls of each color are in the urn. What
is the probability that the first ball taken from the urn will be blue? Ap-
plying the principle of indifference, you say it is 1/2. The probability
that it is not blue must also be 1/2. If it is not blue, it must be red or yel-
low, and because you know nothing about the number of red or yellow
balls, those colors are equally probable. Therefore you assign to red a
probability of 1/4. On the other hand, if you begin by asking for the
probability that the first ball will be red, you must give red a
ity of 1/2 and blue a probability of 1/4, which contradicts your previ-
ous estimates.
It is easy to prove along similar lines that there is life on Mars. What
is the probability that there is simple plant life on Mars? Since argu-
ments on both sides are about equally cogent, we answer 1/2. What is
the probability that there is simple animal life on Mars? Again, 1/2.
Now we seem forced to assert that the probability of there being "either
plant or animal life" on Mars is 1/2 + 1/2 = 1, or certainty, which is ab-
surd. The philosopher Charles Sanders Peirce gave a similar argument
that seems to show that the hair of inhabitants on Saturn had to be
ther of two different colors. It is easy to invent others.
In the history of metaphysics the most notorious misuse of the prin-
ciple surely was by Blaise Pascal, who did pioneer work on probabil-
ity theory, in a famous argument that became known as "Pascal's
wager." A few passages from the original and somewhat lengthy argu-
ment (in Pascal's Pensees, Thought 233) are worth quoting:
"God is, or he is not." To which side shall we incline? Reason can de-
termine nothing about it. There is an infinite gulf fixed between us. A
game is playing at the extremity of this infinite distance in which heads
or tails may turn up. What will you wager? There is no reason for back-
ing either one or the other, you cannot reasonably argue in favor of ei-
ther ....
Yes, but you must wager .... Which will you choose? ... Let us weigh
the gain and the loss in choosing "heads" that God is .... If you gain, you
gain all. If you lose, you lose nothing. Wager, then, unhesitatingly that
he is.
Lord Byron, in a letter, rephrased Pascal's argument effectively: "In-
disputably, the firm believers in the Gospel have a great advantage over
all others, for this simple reason-that, if true, they will have their re-
290 PROBABILITY
ward hereafter; and if there be no hereafter, they can be but with the in-
fidel in his eternal sleep, having had the assistance of an exalted hope
through life, without subsequent disappointment, since (at the worst
for them) out of nothing nothing can arise, not even sorrow." Similar
passages can be found in many contemporary books of religious apolo-
getics.
Pascal was not the first to insist in this fashion that faith in Christian
orthodoxy was the best bet. The argument was clearly stated by the
4th-century African priest Arnobius the Elder, and non-Christian forms
of it go back to Plato. This is not the place, however, to go into the cu-
rious history of defenses and criticisms of the wager. I content myself
with mentioning Denis Diderot's observation that the wager applies
with equal force to other major faiths such as Islam. The mathemati-
cally interesting aspect of all of this is that Pascal likens the outcome
of his bet to the toss of a coin. In other words, he explicitly invokes the
principle of indifference to a situation in which its application is math-
ematically senseless.
The most subtle modern reformulation of Pascal's wager is by
William James, in his famous essay The Will to Believe, in which he ar-
gues that philosophical theism is a better gamble than atheism. In a still
more watered-down form it is even used occasionally by humanists to
defend optimism against pessimism at a time when the extinction of
the human race seems as likely in the near future as its survival.
"While there is a chance of the world getting through its troubles,"
says the narrator of H. G. Wells's little read novel Apropos of Dolores,
"I hold that a reasonable man has to behave as though he was sure of
it. If at the end your cheerfulness is not justified, at any rate you will
have been cheerful."
Addendum
The following letter, from S. D. Turner, contains some surpris-
ing information:
Your bit about the two black and two red cards reminds me of an exer-
cise I did years ago, which might be called N-Card Monte. A few cards,
half red, half black, or nearly so, are shown face up by the pitchman,
then shuffled and dealt face down. The sucker is induced to bet he can
pick two of the same color.
The odds will always be against him. But because the sucker will
Nontransitive Dice and Other Paradoxes 291
make erroneous calculations (like the 2/3 and 1/2 in your 2:2 example),
or for other reasons, he will bet. The pitchman can make a plausible
spiel to aid this: "Now, folks, you don't need to pick two blacks, and you
don't need to pick two reds. If you draw either pair you win!"
The probability of getting two of the same color, where there are R reds
and B blacks, is:
R2 + B2 - (R + B)
P=
(R + B) (R + B-1)
This yields the figures in the table [see Figure 22.2], one in lowest-
terms fractions, the other in decimaL Only below and to the left of the
stairstep line does the sucker get an even break or better. But no pitch-
man would bother with odds more favorable to the sucker than the 1/3
probability for 2:2, or possibly the 2/5 for 3:3.
Surprisingly, the two top diagonal lines are identicaL That is, if you
are using equal reds and blacks, odds are not changed if a card is re-
moved before the two are selected! In your example of 2:2, the proba-
Red Cards
1 2 3 4 5 6 7 8 13 26
- - -
---
-"'-
2
1 1
3 3
3
1 2
~ 2 5
4
~
7
3
~ 5 15 7
5
2
it S ~ a 3
6
~ 4 1
i5
-f, -f,
7 7 ~
7
3 11
i
2Z
fi
6
..2..
4 18 35 -rn 13
8
Z
~
a1 11
19
fi
7
L
9 55 33 39 1!' 15
13
12
Ui
25
'E 26
g
1II
51
0
~
2 .333 .333
m
3 .500 .400 .400
4 .600 .466 .429 .429
5 .667 .524 .465 .444 .444
6 .714 .572 .500 .467 .455 .455
7 .750 .611 .533 .491 .470 .462 .462
8 .778 .645 .564 .515 .488 .472 .466 .466
13
.460
26 .490
Figure 22.2. Probability of drawing two cards of the same color
292 PROBABILITY
bility is 1/3 and it is also 1/3 when starting with 2:1 (as is evident be-
cause the one card not selected can be anyone of the three). The gener-
ality of this can be shown thus: If B = Rand B R 1 are substituted into
the above equation, the result in each case is R - 1/ 2R 1.
Some readers sent detailed explanations of why the arguments be-
hind the fallacies that I described were wrong, apparently not realizing
that these fallacies were intended to be howlers based on the misuse of
the principle of indifference. Several readers correctly pointed out that
although Pascal did invoke the principle of indifference by referring to
a coin flip in his famous wager, the principle is not essential to his ar-
gument. Pascal posits an infinite gain for winning a bet in which the
loss (granting his assumptions) would always be finite regardless of
the odds.
Efron's nontransitive dice aroused almost as much interest among
magicians as among mathematicians. It was quickly perceived that the
basic idea generalized to k sets of n-sided dice, such as dice in the
shapes of regular octahedrons, dodecahedrons, icosahedrons, or cylin-
ders with n flat sides. The game also can be modeled by k sets of n-
sided tops, spinners with n numbers on each dial, and packets of n
playing cards.
Karl Fulves, in his magic magazine The Pallbearers Review (January
1971) proposed using playing cards to model Efron's dice. He suggested
the following four packets: 2,3,4, 10,], Q; 1,2,8,9,9, 10; 6, 6, 7, 7,8,
8; and 4,5, 5, 6, Q, K. Suits are irrelevant. First player selects a packet,
shuffles it, and draws a card. Second player does the same with another
packet. If the chosen cards have the same value, they are replaced and
two more cards drawn. Ace is low, and high card wins. This is based
on Efron's third set of dice where the winning probability, if the second
player chooses properly, is 11/17. To avoid giving away the cyclic se-
quence of packets, each could be placed in a container (box, cup, tray,
etc.) with the containers secretly marked. Before each play, the con-
tainers would be randomly mixed by the first player while the second
player turned his back. Containers with numbered balls or counters
could of course be substituted for cards.
In the same issue of The Pallbearers Review cited above, Columbia
University physicist Shirley Quimby proposed a set of four dice with
the following faces:
Nontransitive Dice and Other Paradoxes 293
3,4,5,20,21,22
1,2,16,17,18,19
10,11,12,13,14,15
6,7,8,9,23,24
Note that numbers 1 through 24 are used just once each in this ele-
gant arrangement. The dice give the second player a winning proba-
bility of 2/3. If modeled with 24 numbered cards, the first player would
select one of the four packets, shuffle, then draw a card. The second
player would do likewise, and high card wins.
R.C.H. Cheng, writing from Bath University, England, proposed a
novel variation using a single die. On each face are numbers 1 through
6, each numeral a different color. Assume that the colors are the rain-
bow colors red, orange, yellow, green, blue, and purple. The chart
below shows how the numerals are colored on each face.
FACE RED ORANGE YELLOW GREEN BLUE PURPLE
A 1 2 3 4 5 6
B 6 1 2 3 4 5
C 5 6 1 2 3 4
D 4 5 6 1 2 3
E 3 4 5 6 1 2
F 2 3 4 5 6 1
The game is played as follows: The first player selects a color, then
the second player selects another color. The die is rolled and the per-
son whose color has the highest value wins. It is easy to see from the
chart that if the second player picks the adjacent color on the right-the
sequence is cyclic, with red to the "right" of purple-the second player
wins five out of six times. In other words, the odds are 5 to 1 in his
favor!
To avoid giving away the sequence of colors, the second player
should occasionally choose the second color to the right, where his
winning odds are 4 to 2, or the color third to the right where the odds
are even. Perhaps he should even, on rare occasions, take the fourth or
fifth color to the right where odds against him are 4 to 2 and 5 to 1 re-
spectively. Mel Stover has suggested putting the numbers and colors on
a six-sided log instead of a cube.
294 PROBABILITY
This, too, models nicely with 36 cards, formed in six piles, each
bearing a colored numeral. The chart's pattern is obvious, and easily ap-
plied to n
2
cards, each with numbers 1 through n, and using n differ-
ent colors. In presenting it as a betting game you should freely display
the faces of each packet to show that all six numbers and all six colors
are represented. Each packet is shuffled and placed face down. The
first player is "generously" allowed first choice of a color and to select
any packet. The color with the highest value in that packet is the win-
ner. In the general case, as Cheng pointed out in his 1971 letter, the sec-
ond player can always choose a pile that gives him a probability of
winning equal to (n -l)/n.
A simpler version of this game uses 16 playing cards. The four pack-
ets are
AS, JH, QC, KD
KS, AH, JC, QD
QS, KH, AC, JD
JS, QH, KC, AD
Ace here is high, and the cyclic sequence of suits is spades, hearts,
clubs, diamonds. The second player wins with 3 to 1 odds by choos-
ing the next adjacent suit, and even odds if he goes to the next suit but
one.
These betting games are all variants of nontransitive voting para-
doxes, about which there is extensive literature.
Answers
The probability that two randomly selected cards, from a set of
two red and two black cards, are the same color is 1/3. If you list the 24
equally probable permutations ofthe four cards, then pick any two po-
sitions (for example, second and fourth cards), you will find eight cases
in which the two cards match in color. One way to see that this proba-
bility of 8/24 or 1/3 is correct is to consider one of the two chosen
cards. Assume that it is red. Of the remaining three cards only one is
red, and so the probability that the second chosen card will be red is
1/3. Of course, the same argument applies if the first card is black.
Most people guess that the odds are even, when actually they are 2:1 in
favor of the cards' having different colors.
Nontransitive Dice and Other Paradoxes 295
Bibliography
R. Carnap, "Statistical and Inductive Probability," The Structure of Scientific
Thought, E. H. Madden (ed.), Houghton Mifflin, 1960.
J. M. Keynes, A Treatise on Probability, Macmillan, 1921; Harper & Row (paper-
back),1962.
On Pascal's Wager:
J. Cargile. "Pascal's Wager," Philosophy: The Journal of the Royal Institute of Phi-
losophy, Vol. 41, July 1966, pp. 250-57.
I. Hacking, The Emergence of Probability: A Philosophical Study of Early Ideas
About Probability, Induction, and Statistical Inference, Cambridge University
Press, 1976.
M.B. Turner, "Deciding for God-the Bayesian Support of Pascal's Wager," Philos-
ophy and Phenomenological Research, Vol. 29, September 1968, pp. 84-90.
On Nontransitive Dice
G. Finnell, "A Dice Paradox," Epilogue, July 1971, pp. 2-3. This is another magic
magazine published by Karl Fulves.
R. P. Savage, "The Paradox of Nontransitive Dice," American Mathematical
Monthly, VoL 101, May 1994, pp. 429-36.
296 PROBABILITY
Chapter 23 More
Nontransitive Paradoxes
I have just so much logic, as to be able to see . ..
that for me to be too good for you, and for you to
be too good for me, cannot be true at once, both
ways.
-ELIZABETH BARRETT, in a letter to Robert Browning.
When a relation R that is true with respect to xRy and yRz also
holds for xRz, the relation is said to be transitive. For example, "less
than" is transitive among all real numbers. If 2 is less than n, and the
square root of 3 is less than 2, we can be certain that the square root of
3 is less than n. Equality also is transitive: if a = band b = c, then a = c.
In everyday life such relations as "earlier than," "heavier than," "taller
than," "inside," and hundreds of others are transitive.
It is easy to think of relations that are not transitive. If A is the father
of Band B is the father of C, it is never true that A is the father of C. If
A loves Band B loves C, it does not follow that A loves C. Familiar
games abound in transitive rules (if poker hand A beats B and B beats
C, then A beats C), but some games have nontransitive (or intransitive)
rules. Consider the children's game in which, on the count of three, one
either makes a fist to symbolize "rock," extends two fingers for "scis-
sors," or all fingers for "paper." Rock breaks scissors, scissors cut paper,
and paper covers rock. In this game the winning relation is not transi-
tive.
Occasionally in mathematics, particularly in probability theory and
decision theory, one comes on a relation that one expects to be transi-
tive but that actually is not. If the nontransitivity is so counterintuitive
as to boggle the mind, we have what is called a nontransitive paradox.
The oldest and best-known paradox of this type is a voting paradox
sometimes called the Arrow paradox after Kenneth J. Arrow because of
its crucial role in Arrow's "impossibility theorem," for which he shared
a Nobel prize in economics in 1972. In Social Choice and Individual
Values, Arrow specified five conditions that almost everyone agrees are
essential for any democracy in which social decisions are based on in-
dividual preferences expressed by voting. Arrow proved that the five
297
conditions are logically inconsistent. It is not possible to devise a vot-
ing system that will not, in certain instances, violate at least one of the
five essential conditions. In short, a perfect democratic voting system
is in principle impossible.
As Paul A. Samuelson has put it: "The search of the great minds of
recorded history for the perfect democracy, it turns out, is the search for
a chimera, for a logical self-contradiction .... Now scholars allover the
world-in mathematics, politics, philosophy, and economics-are try-
ing to salvage what can be salvaged from Arrow's devastating discov-
ery that is to mathematical politics what Kurt Godel's 1931
impossibility-of-proving-consistency theorem is to mathematical
logic."
Let us approach the voting paradox by first considering a funda-
mental defect of our present system for electing officials. It frequently
puts in office a man who is cordially disliked by a majority of voters but
who has an enthusiastic minority following. Suppose 40 percent of the
voters are enthusiastic supporters of candidate A. The opposition is
split between 30 percent for Band 30 percent for C. A is elected even
though 60 percent of the voters dislike him.
One popular suggestion for avoiding such consequences of the split
vote is to allow voters to rank all candidates in their order of preference.
Unfortunately, this too can produce irrational decisions. The matrix in
Figure 23.1 (left) displays the notorious voting paradox in its simplest
form. The top row shows that a third of the voters prefer candidates A,
B, and C in the order ABC. The middle row shows that another third
A
31
B
C
RANK ORDER
2
B
c
A
3
c
A
B
D E F
A 8 1 6
3 5 7
c 4 9 2
-
Figure 23.1. The voting paradox (left) and the tournament paradox based on the magic
square (right)
298 PROBABILITY
rank them BCA, and the bottom row shows that the remaining third
rank them CAB. Examine the matrix carefully and you will find that
when candidates are ranked in pairs, non transitivity rears its head.
Two-thirds of the voters prefer A to B, two-thirds prefer B to C, and two-
thirds prefer Cto A. If A ran against B, A would win. If B ran against C,
B would win. If C ran against A, C would win. Substitute proposals for
candidates and you see how easily a party in power can rig a decision
simply by its choice of which paired proposals to put up first for a
vote.
The paradox was recognized by the Marquis de Condorcet and oth-
ers in the late 18th century and is known in France as the Condorcet ef-
fect. Lewis Carroll, who wrote several pamphlets on voting,
rediscovered it. Most of the early advocates of proportional represen-
tation were totally unaware of this Achilles' heel; indeed, the paradox
was not fully recognized by political theorists until the mid-1940s,
when Duncan Black, a Welsh economist, rediscovered it in connection
with his monumental work on committee decision making. The ex-
perts are nowhere near agreement on which of Arrow's five conditions
should be abandoned in the search for the best voting system. One sur-
prising way out, recommended by many decision theorists, is that
when a deadlock arises, a "dictator" is chosen by lot to break it. Some-
thing close to this solution actually obtains in certain democracies,
England for instance, where a constitutional monarch (selected by
chance in the sense that lineage guarantees no special biases) has a
carefully limited power to break deadlocks under certain extreme con-
ditions.
The voting paradox can arise in any situation in which a decision
must be made between two alternatives from a set of three or more.
Suppose that A, B, and C are three men who have simultaneously pro-
posed marriage to a girl. The rows of the matrix for the voting paradox
can be used to show how she ranks them with respect to whatever
three traits she considers most important, say intelligence, physical at-
tractiveness, and income. Taken by pairs, the poor girl finds that she
prefers A to B, B to C, and C to A. It is easy to see how similar conflicts
can arise with respect to one's choice of a job, where to spend a vaca-
tion, and so on.
Paul R. Halmos once suggested a delightful interpretation of the ma-
trix. Let A, B, and C stand for apple pie, blueberry pie, and cherry pie.
A certain restaurant offers only two of them at any given meal. The
More Nontransitive Paradoxes 299
rows show how a customer ranks the pies with respect to three prop-
erties, say taste, freshness, and size of slice. It is perfectly rational, says
Halmos, for the customer to prefer apple pie to blueberry, blueberry to
cherry, and cherry to apple. In his Adventures of a Mathematician
(Scribner, 1976), Stanislaw Ulam speaks of having discovered the non-
transitivity of such preferences when he was eight or nine and of later
realizing that it prevented one from ranking great mathematicians in a
linear order of relative merit.
Experts differ on how often nontransitive orderings such as this one
arise in daily life, but some recent studies in psychology and economics
indicate that they are commoner than one might suppose. There are even
reports of experiments with rats showing that under certain conditions
the pairwise choices of individual rats are nontransitive. (See Warren S.
McCulloch, "A Heterarchy of Values Determined by the Topology of Ner-
vous Nets," Bulletin of Mathematical Biophysics 7,1945, pp. 89-93.)
Similar paradoxes arise in round-robin tournaments between teams.
Assume that nine tennis players are ranked in ability by the numbers
1 through 9, with the best player given the number 9 and the worst
given the number 1. The matrix in Figure 23.1 (right) is the familiar
order-3 magic square. Let rows A, B, and C indicate how the nine play-
ers are divided into three teams with each row comprising a team. In
round-robin tournaments between teams, where each member of one
team plays once against each member of the others, assume that the
stronger player always wins. It turns out that team A defeats B, B de-
feats C, and C defeats A, in each case by five games to four. It is im-
possible to say which team is the strongest. The same non transitivity
holds if columns D, E, and F of the matrix are the teams.
Many paradoxes of this type were jointly investigated by Leo Moser
and J. W. Moon. Some of the Moser-Moon paradoxes underlie striking
and little-known sucker bets. For example, let each row (or each col-
umn) of an order-3 magic-square matrix be a set of playing cards, say
the ace, 6, and 8 of hearts for set A, the 3, 5, and 7 of spades for set B,
and the 2, 4, and 9 of clubs for C (see Figure 23.2). Each set is random-
ized and placed face down on a table. The sucker is allowed to draw a
card from any set, then you draw a card from a different set. The high
card wins. It is easy to prove that no matter what set the sucker draws
from, you can pick a set that gives you winning odds of five to four. Set
A beats B, Bbeats C, and Cbeats A. The victim may even be allowed to
decide each time whether the high or the low card wins. If you play low
300 PROBABILITY
A
c
Figure 23.2. Nontransitive sucker bet based on magic square: A ~ B ~ C ~ A
card wins, simply pick the winning pile with respect to a nontransitive
circle that goes the other way. A good way to play the game is to use
sets of cards from three decks with backs of different colors. The packet
of nine cards is shuffled each time, then separated by the backs into the
three sets. The swindle is, of course, isomorphic with the tennis-
tournament paradox.
Nontransitivity prevails in many other simple gambling games. In
some cases, such as the top designed by Andrew Lenard (see Figure
23.3), the nontransitivity is easy to understand. The lower part of the
top is fixed but the upper disk rotates. Each of two players chooses a
different arrow, the top is spun (in either direction), and the person
whose arrow points to the section with the highest number wins. A
beats B, B beats C, and C beats A, in each case with odds of two to one.
More Nontransitive Paradoxes 301
23.3. Nontransitive
A B
2 2 4
3 4 5 6
C D
3 5
4 5 2 6
A-B-C-D-A
23.4. Nontransitive
covered (appropriately) by a mathematician named Walter Penney, was
given as a problem in the Journal of Recreational Mathematics (October
1969, p. 241). It is not well known, and most mathematicians simply
cannot believe it when they first hear of it. It is certainly one of the finest
of all sucker bets. It can be played with a penny, or as a side bet on the
reds and blacks of a roulette wheel, or in any situation in which two al-
ternatives are randomized with equal odds. We shall assume that a
penny is used. If it is flipped three times, there are eight equally proba-
ble outcomes: HHH, HHT, HTH, HTT, THH, THT, TTH, and TTT. One
player selects one of these triplets, and the other player selects a differ-
ent one. The penny is then flipped repeatedly until one of the chosen
triplets appears as a run and wins the game. For example, if the chosen
triplets are HHT and THT and the flips are THHHT, the last three flips
show that HHThas won. In brief, the first triplet to appear as a run wins.
One is inclined to assume that one triplet is as likely to appear first
as any other, but it takes only a moment to realize that this is not the
case even with doublets. Consider the doublets HH, HT, TH, and TT.
HH and HT are equally likely to appear first because, after the first H
appears, it is just as likely to be followed by an H as by a T. The same
reasoning shows that TT and TH are equal. Because of symmetry, HH
= TT and HT = TH. TH beats HH with odds of three to one, however,
and HTbeats TTwith the same probability. Consider HT and TT. TTis
always preceded by HT except when TT appears on the first two flips.
This happens in the long run only once in four times. and so the prob-
ability that HT beats TT is 3/4. Figure 23.5 shows the probability that
B, the second player, will win for all pairs of doublets.
HT TH TT
114 112
HT
112 3/4
TH 3/4 112
TT 112 114 112
figure 23.S. Probabilities of B winning
More Nontransitive Paradoxes 303
When we turn to triplets, the situation becomes much more surpris-
ing. Since it does not matter which side of a coin is designated heads,
we know that HHH = TTT, TTH HHT, HTH = THT, and so on. When
we examine the probabilities for unequal pairs, however, we discover
that the game is not transitive. No matter what triplet the first player
takes, the second player can select a better one. Figure 23.6 gives the
probability that B, the second player, defeats A for all possible pairings.
To find B's best response to a triplet chosen by A, find A's triplet at the
top, go down the column until you reach a probability (shown in gray),
then move left along the row to B's triplet on the left.
A
B
HHT HTH HTT THH THT TTH TTT
HHH
112 2/5 215 118 5/12 3/10 112
HHT
112 213 2/3 1/4 5/8 112 7110
HTH 3/5 113 112 112 112 3/8 7/12
HTT
3/5 1/3 112 112 112 3/4 7/8
THH 7/8 3/4 112 112 l/2 1/3 3/5
THT 7/12 3/8 112 112 l/2 l/3 3/5
TTH 7/10 112 5/8 1/4 213 213 112
TTT
112 3/10 5112 1/8 2/5 2/5 112
Figure 23.6. Probabilities of B winning in a triplet game
Note that B's probability of winning is, at the worst, 2/3 (or odds of
two to one) and can go as high as 7/8 (or odds of seven to one). The
seven-to-one odds are easy to comprehend. Consider THH and HHH. If
HHH first appears anywhere except at the start, it must be preceded by
304
PROBABILITY
a T, which means that THHhas appeared earlier. HHHwins, therefore,
only when it appears on the first three flips. Clearly this happens only
once in eight flips.
Barry Wolk of the University of Manitoba has discovered a curious
rule for determining the best triplet. Let X be the first triplet chosen.
Convert it to a binary number by changing each H to zero and each T
to 1. Divide the number by 2, round down the quotient to the nearest
integer, multiply by 5, and add 4. Express the result in binary, then con-
vert the last three digits back to Hand T.
Nontransitivity holds for all higher n-tuplets. A chart supplied by
Wolk gives the winning probabilities for B in all possible pairings of
quadruplets (see Figure 23.7). Like the preceding two charts and charts
for all higher n-tuplets, the matrix is symmetric about the center. The
upper right quadrant is the lower left quadrant upside down, and the
same holds for the upper left and lower right quadrants. The probabil-
ities for B's best responses to A are shown in gray.
In studying these figures, Wolk discovered another kind of anomaly
as surprising as nontransitivity. It has to do with what are called wait-
ing times. The waiting time for an n-tuplet is the average number of
tosses, in the long run, until the specified n-tuplet appears. The longer
you wait for a bus, the shorter becomes the expected waiting time. Pen-
nies, however, have no memory, so that the waiting time for an n-tuplet
is independent of all previous flips. The waiting time for Hand Tis 2.
For doublets the waiting time is 4 for HT and TH, and 6 for HH and TT.
For triplets the waiting times are 8 for HHT, HHT, THH, and TTH; 10
for HTH and THT; and 14 for HHH and TTT. None of this contradicts
what we know about which triplet of a pair is likely to show first. With
quadruplets, however, contradictions arise with six pairs. For example,
THTHhas a waiting time of 20 and HTHHhas a waiting time of 18. Yet,
THTH is more likely to turn up before HTHH with a probability of
9/14, or well over one-half. In other words, an event that is less frequent
in the long run is likely to happen before a more frequent event. There
is no logical contradiction involved here, but it does show that "aver-
age waiting time" has peculiar properties.
There are many ways to calculate the probability that one n-tuplet
will precede another. You can do it by summing infinite series, by
drawing tree diagrams, by recursive techniques that produce sets of
linear equations, and so on. One of the strangest and most efficient
techniques was devised by John Horton Conway of Princeton Univer-
More Nontransitive Paradoxes 305
HHHT HHTH HHIT HTHH HTHT HITH HITT THHH THHT THTH THIT ITHH ITHT ITTH TTTT
2/5 3/8 3/8 3/8 1/4 3/8 7122 112
112
517 112 5112 9/16 9116 5/14 112 7/16 5/8
213 3/4
112 5/8
HTHT 7112 3/8 217 112 9/14 7/16 7/16 7/16 5/8
HITH 7/11 317 112 1/3 112 9116 5112 7112 7112 7/16 5/8
HITT 7/11 317 112 113 112 9/16 5/12 7/12 7/1
7112 7/12 5112 9/16 112 112 1/3
7116 7/12 7112 5/12 9116 112 112 112 113
7/16 7/16 7116 9114 112 7116 7/16
/16 7116 7/16 112 5/14 7112 7112
112 9116 5112 9/16 9/16 9/16 1/8
1TTT 112 7122 3/8 114 3/8 3/8 3/8
Figure 23.7. Probabilities of B winning in a quadruplet game
sity. I have no idea why it works. It just cranks out the answer as if by
magic, like so many of Conway's other algorithms.
The key to Conway's procedure is the calculation of four binary num-
bers that Conway calls leading numbers. Let A stand for the 7-tuplet
HHTHHHT and B for THHTHHH. We want to determine the probabil-
ity of B beating A. To do this, write A above A, B above B, A above B,
and B above A (see Figure 23.8). Above the top tuplet of each pair a bi-
nary number is constructed as follows. Consider the first pair, AA. Look
at the first letter of the top tuplet and ask yourself if the seven letters,
beginning with this first one, correspond exactly to the first seven let-
ters of the tuplet below it. Obviously they do, and so we put a 1 above
the first letter. Next, look at the second letter of the top tuplet and ask
306
PROBABILITY
if the six letters, starting with this one, correspond to the first six let-
ters of the tuplet below. Clearly they do not, and so we put zero above
the second letter. Do the five letters starting with the third letter of the
top tuplet correspond to the first five letters of the lower tuplet? No, and
so this letter also gets zero. The fourth letter gets another zero. When we
check the fifth letter of the top A, we see that HHT does correspond to
the first three letters of the lower A, and so the fifth letter gets a 1. Let-
ters six and seven each get zero. The "A leading A number," or AA, is
1000100, in which each 1 corresponds to a yes answer; each zero, to a
no. Translating 1000100 from binary to decimal gives us 68 as the lead-
ing number for AA.
Figure 23.8 shows the results of this procedure in calculating lead-
ing numbers AA, BB, AB, and BA. Whenever an n-tuplet is compared
with itself, the first digit of the leading number must, of course, be 1.
When compared with a different tuplet, the first digit mayor may not
be 1.
1000100 = 68
A=HHTHHHT
A=HHTHHHT
1000000 =64
B= THHTHHH
B= THHTHHH
0000001 = 1
A=HHTHHHT
B=THHTHHH
0100011=35
B= THHTHHH
A=HHTHHHT
AA -AB:BB-BA
68 - 1 :64 - 35
67:29
Figure 23.8. John Horton Conway's algorithm for calculating odds of B's n-tuplet beating
A's n-tuplet
The odds in favor of B beating A are given by the ratio AA-AB:BB-
BA. In this case 68-1:64-35 = 67:29. As an exercise, the reader can try
calculating the odds in favor of THH beating HHH. The four leading
numbers will be AA = 7, BB = 4, AB = 0, and BA = 3. Plugging these into
the formula, AA-AB:BB-BA gives odds of 7-0:4-3, or seven to one,
as expected. The algorithm works just as well on tuplets of unequal
lengths, provided the smaller tuplet is not contained within the larger
one. If, for example, A = HH and B = HHT, A obviously wins with a
probability of 1.
More Nontransitive Paradoxes 307
I conclude with a problem by David 1. Silverman, who was the first
to introduce the Penney paradox in the problems department that he
then edited for the Journal of Recreational Mathematics (Vol. 2, Octo-
ber 1969, p. 241). The reader should have little difficulty solving it by
Conway's algorithm. TTHH has a waiting time of 16 and HHH has a
waiting time of 14. Which of these tuplets is most likely to appear first
and with what probability?
Addendum
Numerous readers discovered that Barry Wolk's rule for picking
the best triplet B to beat triplet A is equivalent to putting in front of A
the complement of its next-to-last symbol, then discarding the last sym-
bol. More than half these correspondents found that the method also
works for quadruplets except for the two in which Hand T alternate
throughout. In such cases the symbol put in front of A is the same as its
next to last one.
Since October 1974, when this chapter first appeared in Scientific
American, many papers have been published that prove Conway's al-
gorithm and give procedures for picking the best n-tuplet for all values
of n. Two important early articles are cited in the bibliography. The
paper by Guibas and Odlyzko gives 26 references.
Readers David Sachs and Bryce Hurst each noted that Conway's
"leading number," when an n-tuplet is compared with itself, automat-
ically gives that tuplet's waiting time. Simply double the leading num-
ber.
Ancient Chinese philosophers (I am told) divided matter into five
categories that form a nontransitive cycle: wood gives birth to fire, fire
to earth, earth to metal, metal to water, and water to wood. Rudy
Rucker's science-fiction story "Spacetime Donuts" (Unearth, Summer
1978) is based on a much more bizarre nontransitive theory. If you
move down the scale of size, to several steps below electrons, you get
back to the galaxies of the same universe we now occupy. Go up the
scale several stages beyond our galactic clusters, and you are back to the
elementary particles-not larger ones, but the very same particles that
make our stars. The word "matter" loses all meaning.
The following letter was published in Scientific American Uanuary
1975):
308 PROBABILITY
Sirs:
Martin Gardner's article on the paradoxical situations that arise from
non transitive relations may have helped me win a bet in Rome on the
outcome of the Ali v. Foreman world heavyweight boxing title match in
ZaIre on October 30.
Ali, though slower than in former years, and a 4-1 betting underdog,
may have had a psychological and motivational advantage for that par-
ticular fight. But in addition, Gardner's mathematics might be relevant.
Even though Foreman beat Frazier, who beat Ali, Ali could still beat
Foreman because there may be a nontransitive relation between the
three.
I ranked the three fighters against the criteria of speed, power, and
technique (including psychological technique) as reported in the press,
and spotted a nontransitive relation worth betting on:
Speed
Power
Technique
ALI
2
3
1
FRAZIER
1
2
3
FOREMAN
3
1
2
Foreman's power and technique beat Frazier, but Ali's technique and
speed beat Foreman. It was worth the bet. The future implications are,
however, that Frazier can still beat Ali!
ANTHONY PIEL
Vaud, Switzerland
David Silverman (Journal of Recreational Mathematics 2, October
1969, p. 241) proposed a two-person game that he called "blind
Penney-ante." It is based on the nontransitive triplets in a run of fair
coin tosses. Each player simultaneously chooses a triplet without
knowing his opponent's choice. The triplet that shows up first wins.
What is a player's best strategy? This is not an easy problem. A full so-
lution, based on an 8 x 8 game matrix, is given in The College Mathe-
matics Journal as the answer to Problem 299 (January 1987, pp. 74-76).
Answers
Which pattern of heads and tails, TTHH or HHH, is more likely
to appear first as a run when a penny is repeatedly flipped? Applying
More Nontransitive Paradoxes 309
John Horton Conway's algorithm, we find that TTHH is more likely to
precede HHHwith a probability of 7/12, or odds of seven to five. Some
quadruplets beat some triplets with even greater odds. For example,
THHH precedes HHH with a probability of 7/8, or odds of seven to
one. This is easy to see. HHH must be preceded by a T unless it is the
first triplet of the series. Of course, the probability of that is 1/8.
The waiting time for TTHH and for THHH is 16, compared with a
waiting time of 14 for HHH. Both cases of the quadruplet versus the
triplet, therefore, exhibit the paradox of a less likely event occurring be-
fore a more likely event with a probability exceeding 1/2.
Bibliography
Nontransitiue Voting and Tournament Paradoxes
K. J. Arrow, Social Choice and Individual Values, Wiley, 1951; second edition,
1970.
D. Black, The Theory of Committees and Elections, Cambridge University Press,
1958; Kluwer Academic, 1986.
S. J. Brams, Paradoxes in Politics, Free Press, 1976.
R Farquharson, Theory of Voting, Yale University Press, 1969.
P. C. Fishburn, "Paradoxes of Voting," The American Political Science Review, Vol.
68, June 1974, pp. 539-46.
D. Khahr, "A Computer Simulation of the Paradox of Voting," American Political
Science Review, Vol. 50, June 1966, pp. 384-96.
RD. Luce and H. Raiffa, Games and Decision, Wiley, 1957.
A. F. MacKay, Arrow's Theorem: The Paradox of Social Choice, Yale University
Press, 1980.
J. W. Moon, Topics on Tournaments, Holt, Rinehart, and Winston, 1968.
W. H. Riker, "Voting and the Summation of Preferences: An Interpretive Biblio-
graphical Review of Selected Developments During the Last Decade." The Amer-
ican Political Science Review 55, December 1961, pp. 900-11.
W. H. Riker, "Arrow's Theorem and Some Examples of the Paradox of Voting,"
Mathematical Applications in Political Science, Q.M. Claunch (ed.), Southern
Methodist University, 1965.
Z. Usiskin, "Max-Min Probabilities in the Voting Paradox," The Annals of Mathe-
matical Statistics, Vol. 35. June 1964, pp. 857-62.
Nontransitiue Betting qames
M. Gardner, "Lucifer at Las Vegas," Science Fiction Puzzle Tales, Clarkson Potter,
1981, Problem 27.
R Honsberger, "Sheep Fleecing with Walter Funkenbusch," Mathematical Gems
III, Washington, DC: Mathematical Association of America, 1985.
310 PROBABILITY
J. C. Frauenthal and A. B. Miller, "A Coin Tossing Game," Mathematics Magazine
53, September 1980, pp. 239-43.
R. L. Tenney and C. C. Foster, "Nontransitive Dominance," Mathematics Magazine
49, May 1976, pp. 115-20.
Conway's Algorithm
S. Collings, "Coin Sequence Probabilities and Paradoxes," Bulletin of the Institute
of Mathematics and its Applications 18, November/December 1982, pp. 227-32.
R. L. Graham, D. E. Knuth, and O. Patashnik, Concrete Mathematics, 2e, Addison-
Wesley, 1994, pp. 403-10.
L. J. Guibas and A. M. Odlyzko, "String Overlaps, Pattern Matching, and Nontran-
sitive Games," Journal of Combinatorial Theory 30, March 1981, pp. 183-208.
More Nontransitive Paradoxes 311
VJ[I
Infinity

Chapter 24 Infinite Regress
Chairman of a meeting of the Society of Lo-
gicians: "Before we put the motion: 'That the
motion be now put,' should we not first put
the motion: That the motion: "That the mo-
tion be now put" be now put'?"
From an old issue of Punch
The infinite regress, along which thought is compelled to march
backward in a never ending chain of identical steps, has always aroused
mixed emotions. Witness the varied reactions of critics to the central
symbol of Broadway's most talked-about 1964 play, Edward Albee's Tiny
Alice. The principal stage setting-the library of an enormous castle
owned by Alice, the world's richest woman-was dominated by a scale
model of the castle. Inside it lived Tiny Alice. When lights went on and
off in the large castle, corresponding lights went on and off in the small
one. A fire erupted simultaneously in castle and model. Within the
model there was a smaller model in which a tinier Alice perhaps lived,
and so on down, like a set of nested Chinese boxes. ("Hell to clean,"
commented the butler, whose name was Butler.) Was the castle itself,
into which the play's audience peered, a model in a still larger model,
and that in turn ... ? A similar infinite nesting is the basis ofE. Nesbit's
short story, "The Town in the Library in the Town in the Library" (in her
Nine Unlikely Tales); perhaps this was the source of Albee's idea.
For many of the play's spectators the endless regress of castles stirred
up feelings of anxiety and despair: Existence is a mysterious, impene-
trable, ultimately meaningless labyrinth; the regress is an endless cor-
ridor that leads nowhere. For theological students, who were said to
flock to the play, the regress deepens an awareness of what Rudolf Otto,
the German theologican, called the mysterium tremendum: the ulti-
mate mystery, which one must approach with awe, fascination, hu-
mility, and a sense of "creaturehood." For the mathematician and the
logician the regress has lost most of its terrors; indeed, as we shall soon
see, it is a powerful, practical tool even in recreational mathematics.
First, however, let us glance at some of the roles it has played in West-
ern thought and letters.
315
Aristotle, taking a cue from Plato's Parmenides, used the regress in
his famous "third man" criticism of Plato's doctrine of ideas. If all men
are alike because they have something in common with Man, the ideal
and eternal archetype, how (asked Aristotle) can we explain the fact
that one man and Man are alike without assuming another archetype?
And will not the same reasoning demand a third, fourth, and fifth ar-
chetype, and so on into the regress of more and more ideal worlds?
A similar aversion to the infinite regress underlies Aristotle's argu-
ment, elaborated by hundreds of later philosophers, that the cosmos
must have a first cause. William Paley, an 18th-century English the-
ologian, put it this way: "A chain composed of an infinite number of
links can no more support itself than a chain composed of a finite num-
ber of links." A finite chain does indeed require support, mathemati-
cians were quick to point out, but in an infinite chain eve.lY link hangs
securely on the one above. The question of what supports the entire se-
ries no more arises than the question of what kind of number precedes
the infinite regress of negative integers.
Agrippa, an ancient Greek skeptic, argued that nothing can be
proved, even in mathematics, because every proof must be proved valid
and its proof must in turn be proved, and so on. The argument is re-
peated by Lewis Carroll in his paper "What the Tortoise Said to
Achilles" (Mind, April 1895). After finishing their famous race, which
involved an infinite regress of smaller and smaller distances, the Tor-
toise traps his fellow athlete in a more disturbing regress. He refuses to
accept a simple deduction involving a triangle until Achilles has writ-
ten down an infinite series of hypothetical assumptions, each necessary
to make the preceding argument valid.
F. H. Bradley, the English idealist, argued (not very convincingly)
that our mind cannot grasp any type of logical relation. We cannot say,
for example, that castle A is smaller than castle B and leave it at that,
because "smaller than" is a relation to which both castles are related.
Call these new relations c and d. Now we have to relate c and d to the
two castles and to "smaller than." This demands four more relations,
they in turn call for eight more, and so on, until the shaken reader col-
lapses into the arms of Bradley's Absolute.
In recent philosophy the two most revolutionary uses of the regress
have been made by the mathematicians Alfred Tarski and Kurt Gade!.
Tarski avoids certain troublesome paradoxes in semantics by defining
truth in terms of an endless regress of "metalanguages," each capable
316 INFINITY
of discussing the truth and falsity of statements on the next lower level
but not on its own level. As Bertrand Russell once explained it: "The
man who says 'I am telling a lie of order n' is telling a lie, but a lie of
order n + 1." In a closely related argument Godel was able to show that
there is no single, all-inclusive mathematics but only an infinite regress
of richer and richer systems.
The endless hierarchy of gods implied by so many mythologies and
by the child's inevitable question "Who made God?" has appealed to
many thinkers. William James closed his Varieties of Religious Experi-
ence by suggesting that existence includes a collection of many gods,
of different degrees of inclusiveness, "with no absolute unity realized
in it at all. Thus would a sort of polytheism return upon us .... " The
notion turns up in unlikely places. Benjamin Franklin, in a quaint lit-
tle work called Articles of Belief and Acts of Religion, wrote: "For I be-
lieve that man is not the most perfect being but one, but rather that
there are many degrees of beings superior to him." Our prayers, said
Franklin, should be directed only to the god of our solar system, the
deity closest to us. Many writers have viewed life as a board game in
which we are the pieces moved by higher intelligences who in turn are
the pieces in a vaster game. The prophet in Lord Dunsany's story "The
South Wind" observes the gods striding through the stars, but as he
worships them he sees the outstretched hand of a player "enormous
over Their heads."
Graphic artists have long enjoyed the infinite regress. The striking
cover of the April 1965, issue of Scientific American showed the mag-
azine cover reflected in the pupil of an eye. The cover of the Novem-
ber 1964, Punch showed a magician pulling a rabbit out of a hat. The
rabbit in turn is pulling a smaller rabbit out of a smaller hat, and this
endless series of rabbits and hats moves up and off the edge of the
page. It is not a bad picture of contemporary particle physics. It is now
known that protons and neutrons are made of smaller units called
quarks, and if superstring theory is correct, all particles are made of ex-
tremely tiny loops that vibrate at different frequencies. Does the uni-
verse have, as some physicists believe, infinite levels of structure?
The play within the play, the puppet show within the puppet show,
the story within the story have amused countless writers. Luigi Piran-
delIo's Six Characters in Search of an Author is perhaps the best-
known stage example. The protagonist in Miguel de Unamuno's novel
Mist, anticipating his death later in the plot, visits Unamuno to protest
Infinite Regress 317
and troubles the author with the thought that he too is only the figment
of a higher imagination. Philip Quarles, in Aldous Huxley's Point
Counter Point, is writing a novel suspiciously like Point Counter Point.
Edouard, in Andre Gide's The Counterfeiters, is writing The Counter-
feiters. Norman Mailer's story "The Notebook" tells of an argument be-
tween the writer and his girl friend. As they argue he jots in his
notebook an idea for a story that has just come to him. It is, of course,
a story about a writer who is arguing with his girl friend when he gets
an idea ....
J. E. Littlewood, in A Mathematician's Miscellany, recalls the fol-
lowing entry, which won a newspaper prize in Britain for the best piece
on the topic: "What would you most like to read on opening the morn-
ing paper?"
OUR SECOND COMPETITION
The First Prize in the second of this year's competitions goes to Mr.
Arthur Robinson, whose witty entry was easily the best of those we re-
ceived. His choice of what he would like to read on opening his paper
was headed "Our Second Competition" and was as follows: "The First
Prize in the second ofthis year's competitions goes to Mr. Arthur Robin-
son, whose witty entry was easily the best of those we received. His
choice of what he would like to read on opening his paper was headed
'Our Second Competition,' but owing to paper restrictions we cannot
print all of it."
One way to escape the torturing implications of the endless regress
is by the topological trick of joining the two ends to make a circle, not
necessarily vicious, like the circle of weary soldiers who rest them-
selves in a bog by each sitting on the lap of the man behind. Albert Ein-
stein did exactly this when he tried to abolish the endless regress of
distance by bending three-dimensional space around to form the hy-
persurface of a four-dimensional sphere. One can do the same thing
with time. There are Eastern religions that view history as an endless
recurrence of the same events. In the purest sense one does not even
think of cycles following one another, because there is no outside time
by which the cycles can be counted; the same cycle, the same time go
around and around. In a similar vein, there is a sketch by the Dutch
artist Maurits C. Escher of two hands, each holding a pencil and sketch-
ing the other (see Figure 24.1). In Through the Looking Glass Alice
dreams of the Red King, but the King is himself asleep and, as Twee-
318 INFINITY
24.1. Maurits C ......... "" .............
IrrniAnrln Hands
319
of old-Asgool, Trodath, Skun, and Rhoog-he sees the shadowy forms
of three larger gods farther up the slope. He leads his disciples up the
mountain only to observe, years later, two larger gods seated at the
summit, from which they point and mock at the gods below. Shaun
takes his followers still higher. Then one night he perceives across the
plain an enormous, solitary god looking angrily toward the mountain.
Down the mountain and across the plain goes Shaun. While he is carv-
ing on rock the story of how his search has ended at last with the dis-
covery of the ultimate god, he sees in the far distance the dim forms of
four higher deities. As the reader can guess, they are Asgool, Trodath,
Skun, and Rhoog.
No branch of mathematics is immune to the infinite regress. Numbers
on both sides of zero gallop off to infinity. In modular arithmetics they
go around and around. Every infinite series is an infinite regress. The
regress underlies the technique of mathematical induction. Georg Can-
tor's transfinite numbers form an endless hierarchy of richer infinities.
A beautiful modern example of how the regress enters into a mathe-
matical proof is related to the difficult problem of dividing a square into
other squares no two of which are alike. The question arises: Is it pos-
sible similarly to cut a cube into a finite number of smaller cubes no
two of which are alike? Were it not for the deductive power of the
regress, mathematicians might still be searching in vain for ways to do
this. The proof of impossibility follows.
Assume that it is possible to "cube the cube." The bottom face of
such a dissected cube, as it rests on a table, will necessarily be a
"squared square." Consider the smallest square in this pattern. It can-
not be a corner square, because a larger square on one side keeps any
larger square from bordering the other side (see Figure 24.2{a)). Simi-
larly, the smallest square cannot be elsewhere on the border, between
corners, because larger squares on two sides prevent a third larger
square from touching the third side (Figure 24.2{b)). The smallest
square must therefore be somewhere in the pattern's interior. This in
turn requires that the smallest cube touching the table be surrounded
by cubes larger than itself. This is possible (Figure 24.2{ c)), but it means
that four walls must rise above all four sides of the small cube-pre-
venting a larger cube from resting on top of it. Therefore on this small-
est cube there must rest a set of smaller cubes, the bottoms of which
will form another pattern of squares.
The same argument is now repeated. In the new pattern of squares
320 INFINITY
b
cannot
the smallest cube) on
will form another pattern of squares. Clearly the argument leads an
endless regress of smaller like the endless hierarchy of fleas in
solvable.
constructions one, involving an infinite
curve a area one
inch?
equilateral
pathological curves are infinite in number. Start
24.3{a)) on the central third of
a
321
cwve
322
24.4.
curve
323
friend once said. the universe seems to be made of nothing, yet some-
how it manages to exist.
Some top mathematical physicists, Stanislaw Ulam and David Bohm
for example, defended and defend the notion that matter has infinite
layers of structure in both directions. One of H. G. Wells's fantasies
portrayed our universe as a molecule in a ring worn by a gigantic hand.
I described the snowflake curve before it became known as the old-
est and simplest of Benoit Mandelbrofs famous fractals. For more on
fractals, see this volume's Chapter 27 and Chapter 8 of my Penrose
Tiles to Trapdoor Ciphers.
Answers
The cross-stitch curve has, like its analogue the snowflake, an
infinite length. It bounds an area twice that of the original square. The
drawing at the left in Figure 24.5 shows its appearance after the third
construction. After many more steps it resembles (when viewed at a
distance) the drawing at the right. Although the stitches seem to run di-
agonally' actually every line segment in the figure is vertical or hori-
zontal. Similar constructions of pathological curves can be based on
any regular polygon, but beyond the square the figure is muddied by
overlapping, so that certain conventions must be adopted in defining
what is meant by the enclosed area.
Samuel P. King, Jr., of Honolulu, supplied a good analysis of curves
of this type, including a variant of the cross-stitch discovered by his fa-
ther. Instead of erecting four squares outwardly each time, they are
Figure 24.S. Solution to cross-stitch curve problem
324 INFINITY
erected inwardly from sides of each
infinite but encloses zero area.
J. L. Borges, "Avatars of the Tortoise," Labyrinths: Selected Stories
ings, New York: New Directions, 1964.
curve an
J. L. Borges. "Partial Magic in the Quixote," Labyrinths: Selected and Other
Writings, New New Directions, 1964.
W. King, "Snowflake Curves," The Mathematics Teacher, 57. April,
1964, pp. 219-22.
A Dutch
Infinite Regress 325
H. P. Owen, "Infinity in Theology and Metaphysics," The Encyclopedia of Philos-
ophy. Vol. 4, New York: Crowell Collier, 1967, pp. 190-93.
J. Passmore, "The Infinite Regress," Philosophical Reasoning, New York: Charles
Scribner's Sons, 1961.
J. E. Schneider, "A Generalization ofthe Von Koch Curve," Mathematics Magazine,
Vol. 38, No.3, May, 1965, pp. 144-47.
J. Thomson, "Infinity in Mathematics and Logic," The Encyclopedia of Philosophy,
Vol. 4, New York: Crowell Collier, 1967, pp. 183-90.
326
INFINITY
Chapter 25
A graduate student at Trinity
Computed the square of infinity.
But it gave him the fidgets
To put down the digits,
Aleph-Null
and Aleph-One
So he dropped math and took up divinity
-ANONYMOUS
In 1963 Paul J. Cohen, then a 29-year-old mathematician at
Stanford University, found a surprising answer to one of the great prob-
lems of modern set theory: Is there an order of infinity higher than the
number of integers but lower than the number of points on a line? To
make clear exactly what Cohen proved, something must first be said
about those two lowest known levels of infinity.
It was Georg Ferdinand Ludwig Philipp Cantor who first discovered
that beyond the infinity of the integers-an infinity to which he gave
the name aleph-null-there are not only higher infinities but also an in-
finite number of them. Leading mathematicians were sharply divided
in their reactions. Henri Poincare called Cantorism a disease from
which mathematics would have to recover, and Hermann Weyl spoke
of Cantor's hierarchy of alephs as "fog on fog."
On the other hand, David Hilbert said, "From the paradise created for
us by Cantor, no one will drive us out," and Bertrand Russell once
praised Cantor's achievement as "probably the greatest of which the age
can boast." Today only mathematicians of the intuitionist school and a
few philosophers are still uneasy about the alephs. Most mathemati-
cians long ago lost their fear of them, and the proofs by which Cantor
established his "terrible dynasties" (as they have been called by the
world-renowned Argentine writer Jorge Luis Borges) are now univer-
sally honored as being among the most brilliant and beautiful in the
history of mathematics.
Any infinite set of things that can be counted 1, 2, 3, ... has the car-
dinalnumber alepho (aleph-null), the bottom rung of Cantor's aleph
ladder. Of course, it is not possible actually to count such a set; one
merely shows how it can be put into one-to-one correspondence with
the counting numbers. Consider, for example, the infinite set of primes.
327
It is easily put in one-to-one correspondence with the positive inte-
gers:
1
L
2
2
L
3
3
L
5
4
L
7
5 6 ...
J- J-
11 13 ...
The set of primes is therefore an aleph-null set. It is said to be "count-
able" or "denumerable." Here we encounter a basic paradox of all infi-
nite sets. Unlike finite sets, they can be put in one-to-one
correspondence with a part of themselves or, more technically, with
one of their "proper subsets." Although the primes are only a small por-
tion of the positive integers, as a completed set they have the same
aleph number. Similarly, the integers are only a small portion of the ra-
tional numbers (the integers plus all integral fractions), but the rationals
form an aleph-null set too.
There are all kinds of ways in which this can be proved by arranging
the rationals in a countable order. The most familiar way is to attach
them, as fractions, to an infinite square array of lattice points and then
count the points by following a zigzag path, or a spiral path if the lat-
tice includes the negative rationals. Here is another method of ordering
and counting the positive rationals that was proposed by the American
logician Charles Sanders Peirce.
Start with the fractions 0/1 and 1/0. (The second fraction is mean-
ingless, but that can be ignored.) Sum the two numerators and then the
two denominators to get the new fraction 1/1, and place it between the
previous pair: 0/1, 1/1, 1/0. Repeat this procedure with each pair of ad-
jacent fractions to obtain two new fractions that go between them:
0 1 1 2 1
1 2 1 1 0
The five fractions grow, by the same procedure, to nine:
0 1 1 2 1 3 2 3 1
-
-
- -
1 3 2 3 1 2 1 1 0
In this continued series every rational number will appear once and
only once, and always in its simplest fractional form. There is no need,
as there is in other methods of ordering the rationals, to eliminate frac-
tions, such as 10/20, that are equivalent to simpler fractions also on the
list, because no reducible fraction ever appears. If at each step you fill
328 INFINITY
the cracks, so to speak, from left to right, you can count the fractions
simply by taking them in their order of appearance.
This series, as Peirce said, has many curious properties. At each new
step the digits above the lines, taken from left to right, begin by re-
peating the top digits of the previous step: 01, 011, 0112, and so on.
And at each step the digits below the lines are the same as those above
the lines but in reverse order. As a consequence, any two fractions
equally distant from the central 111 are reciprocals of each other. Note
also that for any adjacent pair, alb, eld, we can write such equalities as
be - ad = 1, and eld - alb = llbd. The series is closely related to what
are called Farey numbers (after the English geologist John Farey. who
first analyzed them), about which there is now a considerable litera-
ture.
It is easy to show that there is a set with a higher infinite number of
elements than aleph-null. To explain one of the best of such proofs, a
deck of cards is useful. First consider a finite set of three objects, say a
key, a watch, and a ring. Each subset of this set is symbolized by a row
of three cards (see Figure 25.1.), a face-up card (white) indicates that the
object above it is in the subset, a face-down card (black) indicates that
it is not. The first subset consists of the original set itself. The next
three rows indicate subsets that contain only two of the objects. They
are followed by the three subsets of single objects and finally by the
empty (or null) subset that contains none of the objects. For any set of
n elements the number of subsets is 2n. (It is easy to see why. Each el-
ement can be either included or not, so for one element there are two
subsets, for two elements there are 2 x 2 = 4 subsets, for three elements
there are 2 x 2 x 2 = 8 subsets, and so on.) Note that this formula applies
even to the empty set, since 2 = 1 and the empty set has the empty set
as its sole subset.
This procedure is applied to an infinite but countable (aleph-null) set
of elements at the left in Figure 25.2. Can the subsets of this infinite set
be put into one-to-one correspondence with the counting integers? As-
sume that they can. Symbolize each subset with a row of cards, as be-
fore, only now each row continues endlessly to the right. Imagine these
infinite rows listed in any order whatever and numbered 1., 2, 3, ...
from the top down.
If we continue forming such rows, will the list eventually catch all
the subsets? No-because there is an infinite number of ways to pro-
duce a subset that cannot be on the list. The simplest way is to consider
Aleph-Null and Aleph-One 329
2
3
4
5
6
7
8
25.1. bUI)seiLS a set
330
CD
@ @
CD
@ @

0110
0 0
2
D II
2 1
3
II 10
3 .0
4
I
4 .0
5
0111
5 1
Figure 2S.2. A countable infinity has an uncountable infinity of subsets (left) that corre-
spond to the real numbers (right).
nal number 2 raised to the power of aleph-nulL This proof shows that
such a set cannot be matched one-to-one with the counting integers. It
is a higher aleph, an "uncountable" infinity.
Cantor's famous diagonal proof, in the form just given. conceals a
startling bonus. It proves that the set of real numbers (the rationals plus
the irrationals) is also uncountable. Consider a line segment, its ends
numbered 0 and 1. Every rational fraction from 0 to 1 corresponds to a
point on this line. Between any two rational points there is an infinity
of other rational points; nevertheless, even after all rational points are
identified, there remains an infinity of unidentified points-points that
correspond to the unrepeating decimal fractions attached to such alge-
braic irrationals as the square root of 2 and to such transcendental ir-
rationals as pi and e. Every point on the line segment, rational or
irrational, can be represented by an endless decimal fraction. But these
fractions need not be decimal; they can also be written in binary nota-
tion. Thus every point on the line segment can be represented by an
endless pattern of l's and O's, and every possible endless pattern of l's
and O's corresponds to exactly one point on the line segment. See Ad-
dendum.
Now, suppose each face-up card at the left in Figure 25.2 is replaced
by 1 and each face-down card is replaced by 0, as shown at the right in
the illustration. Put a binary point in front of each row and we have an
Aleph-Null and Aleph-One 331
infinite list of binary fractions between 0 and 1. But the diagonal set of
symbols, after each 1 is changed to 0 and each 0 to 1, is not on the list,
and the real numbers and points on the line are uncountable. By care-
ful dealing with the duplications Cantor showed that the three sets-
the subsets of aleph-null, the real numbers, and the totality of points on
a line segment-have the same number of elements. Cantor called this
cardinal number C, the "power of the continuum." He believed it was
also Xl (aleph-one), the first infinity greater than aleph-null.
By a variety of simple, elegant proofs Cantor showed that C was the
number of such infinite sets as the transcendental irrationals (the alge-
braic irrationals, he proved, form a countable set), the number of points
on a line of infinite length, the number of points on any plane figure or
on the infinite plane, and the number of points in any solid figure or in
all of 3-space. Going into higher dimensions does not increase the num-
ber of points. The points on a line segment one inch long can be
matched one-to-one with the points in any higher-dimensional solid, or
with the points in the entire space of any higher dimension.
The distinction between aleph-null and aleph-one (we accept, for
the moment, Cantor's identification of aleph-one with C) is important
in geometry whenever infinite sets of figures are encountered. Imagine
an infinite plane tessellated with hexagons. Is the total number of ver-
tices aleph-one or aleph-null? The answer is aleph-null; they are eas-
ily counted along a spiral path (see Figure 25.3). On the other hand, the
number of different circles of one-inch radius that can be placed on a
sheet of typewriter paper is aleph-one because inside any small square
near the center of the sheet there are aleph-one points, each the center
of a different circle with a one-inch radius.
Consider in turn each of the five symbols J. B. Rhine uses on his
"ESP" test cards (see Figure 25.4). Can it be drawn an aleph-one num-
ber of times on a sheet of paper, assuming that the symbol is drawn
with ideal lines of no thickness and that there is no overlap or inter-
section of any lines? (The drawn symbols need not be the same size, but
all must be similar in shape.) It turns out that all except one can be
drawn an aleph-one number of times. Can the reader show which sym-
bol is the exception?
Richard Schlegel, a physicist, attempted to relate the two alephs to
cosmology by calling attention to a seeming contradiction in the steady-
state theory. According to that theory, the number of atoms in the cos-
mos at the present time is aleph-null. (The cosmos is regarded as
332 INFINITY
25.3. counts
25.4. Five "ESp
li
blings, they would have grown to an aleph-one set. But the cosmos
cannot contain an aleph-one set of atoms. Any collection of distinct
physical entities (as opposed to the ideal entities of mathematics) is
countable and therefore, at the most, aleph-nulL
In his paper, "The Problem of Infinite Matter in Steady-State Cos-
mology," Schlegel found the way out. Instead of regarding the past as
a completed aleph-null set of finite time intervals (to be sure, ideal in-
stants in time form an aleph-one continuum, but Schlegel is concerned
with those finite time intervals during which doublings of atoms
occur), we can view both the past and the future as infinite in the in-
ferior sense of "becoming" rather than completed. Whatever date is
suggested for the origin of the universe (remember, we are dealing with
the steady-state model, not with a Big Bang or oscillating theory), we
can always set an earlier date. In a sense there is a "beginning," but we
can push it as far back as we please. There is also an "end," but we can
push it as far forward as we please. As we go back in time, continually
halving the number of atoms, we never halve them more than a finite
number of times, with the result that their number never shrinks to
less than aleph-nulL As we go forward in time, doubling the number of
atoms, we never double more than a finite number of times: therefore
the set of atoms never grows larger than aleph-nulL In either direction
the leap is never made to a completed aleph-null set of time intervals.
As a result the set of atoms never leaps to aleph-one and the disturbing
contradiction does not arise.
Cantor was convinced that his endless hierarchy of alephs, each ob-
tained by raising 2 to the power of the preceding aleph, represented all
the alephs there are. There are none in between. Nor is there an Ulti-
mate Aleph, such as certain Hegelian philosophers of the time identi-
fied with the Absolute. The endless hierarchy of infinities itself, Cantor
argued, is a better symbol of the Absolute.
All his life Cantor tried to prove that there is no aleph between aleph-
null and C, the power of the continuum, but he never found a proof. In
1938 Kurt Godel showed that Cantor's conjecture, which became
known as the Continuum Hypothesis, could be assumed to be true,
and that this could not conflict with the axioms of set theory.
What Cohen proved in 1963 was that the opposite could also be as-
sumed. One can posit that C is not aleph-one; that there is at least one
aleph between aleph-null and C, even though no one has the slightest
notion of how to specify a set (for example, a certain subset of the tran-
334 INFINITY
scendental numbers) that would have such a cardinal number. This
too is consistent with set theory. Cantor's hypothesis is undecidable.
Like the parallel postulate of Euclidean geometry, it is an independent
axiom that can be affirmed or denied. Just as the two assumptions about
Euclid's parallel axiom divided geometry into Euclidean and non-
Euclidean, so the two assumptions about Cantor's hypothesis now di-
vide the theory of infinite sets into Cantorian and non-Cantorian. It is
even worse than that. The non-Cantorian side opens up the possibility
of an infinity of systems of set theory, all as consistent as standard the-
ory now is and all differing with respect to assumptions about the
power of the continuum.
Of course Cohen did no more than show that the continuum hy-
pothesis was undecidable within standard set theory, even when the
theory is strengthened by the axiom of choice. Many mathematicians
hope and believe that some day a "self-evident" axiom, not equivalent
to an affirmation or denial of the continuum hypothesis, will be found,
and that when this axiom is added to set theory, the continuum hy-
pothesis will be decided. (By "self-evident" they mean an axiom which
all mathematicians will agree is "true.") Indeed, G6del expected and
Cohen expects this to happen and are convinced that the continuum
hypothesis is in fact false, in contrast to Cantor, who believed and
hoped it was true. So far, however, these remain only pious Platonic
hopes. What is undeniable is that set theory has been struck a gigantic
cleaver blow, and exactly what will come of the pieces no one can say.
Addendum
In giving a binary version of Cantor's famous diagonal proof
that the real numbers are uncountable, I deliberately avoided compli-
cating it by considering the fact that every integral fraction between 0
and 1 can be represented as an infinite binary fraction in two ways. For
example, % is .01 followed by aleph-null zeroes and also .001 followed
by aleph-null ones. This raises the possibility that the list of real binary
fractions might be ordered in such a way that complementing the di-
agonal would produce a number on the list. The constructed number
would, of course, have a pattern not on the list, but could not this be a
pattern which expressed, in a different way, an integral fraction on the
list?
The answer is no. The proof assumes that all possible infinite binary
Aleph-Null and Aleph-One 335
patterns are listed, therefore every integral fraction appears twice on the
list, once in each of its two binary forms. It follows that the constructed
diagonal number cannot match either form of any integral fraction on
the list.
In every base notation there are two ways to express an integral frac-
tion by an aleph-null string of digits. Thus in decimal notation V4 =
.2500000 ... = .2499999 .... Although it is not necessary for the validity
of the diagonal proof in decimal notation, it is customary to avoid am-
biguity by specifying that each integral fraction be listed only in the
form that terminates with an endless sequence of nines, then the diag-
onal number is constructed by changing each digit on the diagonal to
a different digit other than nine or zero.
Until I discussed Cantor's diagonal proof in Scientific American, I
had not realized how strongly the opposition to this proof has per-
sisted; not so much among mathematicians as among engineers and
scientists. I received many letters attacking the proof. William Dil-
worth, an electrical engineer, sent me a clipping from the LaGrange
Citizen, LaGrange, IL, January 20,1966, in which he is interviewed at
some length about his rejection of Can tori an "numerology." Dilworth
first delivered his attack on the diagonal proof at the International Con-
ference on General Semantics, New York, 1963.
One of the most distinguished of modern scientists to reject Canto-
rian set theory was the physicist P. W. Bridgman. He published a paper
about it in 1934, and in his Reflections of a Physicist (Philosophical Li-
brary, 1955) he devotes pages 99-104 to an uncompromising attack on
transfinite numbers and the diagonal proof. "I personally cannot see an
iota of appeal in this proof," he writes, "but it appears to me to be a per-
fect nonsequitur-my mind will not do the things that it is obviously
expected to do if it is indeed a proof."
The heart of Bridgman's attack is a point of view widely held by
philosophers of the pragmatic and operationalist schools. Infinite num-
bers, it is argued, do not "exist" apart from human behavior. Indeed, all
numbers are merely names for something that a person does, rather
than names of "things." Because one can count 25 apples but cannot
count an infinity of apples, "it does not make sense to speak of infinite
numbers as 'existing' in the Platonic sense, and still less does it make
sense to speak of infinite numbers of different orders of infinity, as
does Cantor."
"An infinite number," declares Bridgman, "is a certain aspect of what
336 INFINITY
one does when he embarks on carrying out a process . . . an infinite
number is an aspect of a program of action."
The answer to this is that Cantor did specify precisely what one must
"do" to define a transfinite number. The fact that one cannot carry out an
infinite procedure no more diminishes the reality or usefulness of Can-
tor's alephs than the fact that one cannot fully compute the value of pi di-
minishes the reality or usefulness of pi. It is not, as Bridgman maintained,
a question of whether one accepts or rejects the Platonic notion of num-
bers as "things." For an enlightened pragmatist, who wishes to ground all
abstractions in human behavior, Cantorian set theory should be no less
meaningful or potentially useful than any other precisely defined ab-
stract system such as, say, group theory or a non-Euclidean geometry.
Several readers attacked Schlegel's claim that Cantor's alephs expose
a contradiction in a steady-state theory of the universe. They focused
on his argument that after a countable infinity of atom doublings the
cosmos would contain an uncountable infinity of atoms. For details on
this objection see Rudy Rucker's Infinity and the Mind (The Mathe-
matical Association of America, 1982), pages 241-42.
For an account of crank objections to Cantor's alephs, see "Cantor's
Diagonal Process," in Underwood Dudley's Mathematical Cranks (The
Mathematical Association of America, 1992).
Answers
Which of the five ESP symbols cannot be drawn an aleph-one
number of times on a sheet of paper, assuming ideal lines that do not
overlap or intersect, and replicas that may vary in size but must be
similar in the strict geometric sense?
Only the plus symbol cannot be aleph-one replicated. Figure 25.5
shows how each of the other four can be drawn an aleph-one number
of times. In each case points on line segment AB form an aleph-one
continuum. Clearly a set of nested or side-by-side figures can be drawn
so that a different replica passes through each of these points, thus
putting the continuum of points into one-to-one correspondence with
a set of nonintersecting replicas. There is no comparable way to place
replicas of the plus symbol so that they fit snugly against each other.
The centers of any pair of crosses must be a finite distance apart (al-
though this distance can be made as small as one pleases), forming a
countable (aleph-null) set of points. The reader may enjoy devising a
Aleph-Null and Aleph-One 337
H++H++ts
A S
Figure 25.5. Proof for "ESP" -symbol problem
formal proof that aleph-one plus symbols cannot be drawn on a page.
The problem is similar to one involving alphabet letters that can be
found in Leo Zippin 's Uses of Infinity (Random House, 1962), page 57.
So far as I know, no one has yet specified precisely what conditions
must be met for a linear figure to be aleph-one replicable. Some figures
are aleph-one replicable by translation or rotation, some by shrinkage,
some by translation plus shrinkage, some by rotation plus shrinkage. I
rashly reported in my column that all figures topologically equivalent
to a line segment or a simple closed curve were aleph-one replicable,
but Robert Mack, then a high school student in Concord, MA, found a
simple counterexample. Consider two unit squares, joined like a verti-
cal domino, then eliminate two unit segments so that the remaining
segments form the numeral 5. It is not aleph-one replicable.
Bibliography
Farey numbers
A. H. Beiler, Recreations in the Theory of Numbers. Dover, 1964, Chapter 16.
G. S. Cunningham, "Farey Sequences." Enrichment Mathematics for High School,
National Council of Teachers of Mathematics. 1963. Chapter 1.
338 INFINITY
R. Honsberger, Ingenuity in Mathematics, The Mathematical Association of Amer-
ica New Mathematical Library, 1970, Chapter 5.
Transfinite numbers
J. Breuer, Introduction to the Theory of Sets, Prentice-Hall, 1958.
E. V. Huntington, The Continuum, Dover, 1955.
M. M. Zuckerman, Sets and Transfinite Numbers, Macmillan, 1974.
Cohen's proof
P. J. Cohen, Set Theory and the Continuum Hypothesis, W. A. Benjamin, 1966.
P. J. Cohen and R. Hersh, "Non-Cantorian Set Theory," Scientific American, De-
cember 1967, pp. 104-16.
D. S. Scott, "A Proof of the Independence of the Continuum Hypothesis." Mathe-
matical Systems Theory, Vol. 1, 1967, pp. 89-111.
R. M. Smullyan, "The Continuum Problem," The Encyclopedia of Philosophy,
Macmillan, 1967.
R. M. Smullyan, "The Continuum Hypothesis," The Mathematical Sciences, MIT
Press, 1969, pp. 252-71.
Alephs and cosmology
R. Schlegel, "The Problem ofInfinite Matter in Steady-State Cosmology," Philoso-
phy of Science, Vol. 32, January 1965, pp. 21-31.
R. Schlegel. Completeness in Science, Appleton-Century-Crofts, 1967, pp. 138-49.
S. Ulam, "Combinatorial Analysis in Infinite Sets and Some Physical Theories,"
SIAM Review, Vol. 6, October 1964, pp. 343-55.
Aleph-Null and Aleph-One 339
Chapter 26
Points
Have no parts or joints.
How then can they combine
To form a line?
-J. A. LINDON
Supertasks
Every finite set of n elements has 2
n
subsets if one includes the
original set and the null, or empty, set. For example, a set of three ele-
ments, ABC, has 2
3
= 8 subsets: ABC, AB, BC, AC, A, B, C, and the null
set. As the philosopher Charles Sanders Peirce once observed (Col-
lected Papers Vol. 4, p. 181), the null set "has obvious logical pecu-
liarities." You can't make any false statement about its members
because it has no members. Put another way, if you say anything logi-
cally contradictory about its members, you state a truth, because the so-
lution set for the contradictory statement is the null set. Put
colloquially, you are saying something true about nothing.
In modern set theory it is convenient to think of the null set as an
"existing set" even though it has no members. It can also be said to have
2
n
subsets because 2 = 1, and the null set has one subset, namely itself.
And it is a subset of every set. If set A is included in set B, it means that
every member of set A is a member of set B. Therefore, if the null set is
to be treated as a legitimate set, all its members (namely none) must be
in set B. To prove it by contradiction, assume the null set is not in-
cluded in set B. Then there must be at least one member of the null set
that is not a member of B, but this is impossible because the null set has
no members.
The n elements of any finite set obviously cannot be put into one-to-
one correspondence with its subsets because there are always more
than n subsets. Is this also true of infinite sets? The answer is yes, and
the general proof is one of the most beautiful in all set theory.
It is an indirect proof, a reductio ad absurdum. Assume that all ele-
ments of N, a set with any number or members, finite or infinite, are
matched one-to-one with all of N's subsets. Each matching defines a
coloring of the elements:
340
1. An element is paired with a subset that includes that element. Let us
call all such elements blue.
2. An element is paired with a subset that does not include that ele-
ment. We call all such elements red.
The red elements form a subset of our initial set N. Can this subset
be matched to a blue element? No, because every blue element is in its
matching subset, therefore the red subset would have to include a blue
element. Can the red subset be paired with a red element? No, because
the red element would then be n c l u d ~ d in its subset and would there-
fore be blue. Since the red subset cannot be matched to either a red or
blue element of N, we have constructed a subset of Nthat is not paired
with any element of N. No set, even if infinite, can be put into one-to-
one correspondence with its subsets. If n is a transfinite number, then
2n-by definition it is the number of subsets of n-must be a higher
order of infinity than n.
Georg Cantor, the founder of set theory, used the term aleph-null for
the lowest transfinite number. It is the cardinal number of the set of all
integers, and for that reason is often called a "countable infinity." Any
set that can be matched one-to-one with the counting numbers, such as
the set of integral fractions, is said to be a countable or aleph-null set.
Cantor showed that when 2 is raised to the power of aleph-null-giving
the number of subsets of the integers-the result is equal to the cardi-
nal number of the set of all real numbers (rational or irrational), called
the "power of the continuum," or C. It is the cardinal number of all
points on a line. The line may be a segment of any finite length, a ray
with a beginning but no end, or a line going to infinity in both direc-
tions. Figure 26.1 shows three intuitively obvious geometrical proofs
that all three kinds of line have the same number of points. The slant
p
P A A P B
Figure 26.1. The number of points on a line segment AB is the same as on a longer line
segment (left), a ray (center), and a line (right).
Supertasks 341
lines projected from point P indicate how all points on the line segment
AB can be put into one-to-one correspondence with all points on the
longer segment, on a ray, and on an endless line.
The red-blue proof outlined above (Cantor published it in 1890) of
course generates an infinite hierarchy of transfinite numbers. The lad-
der starts with the set of counting numbers, aleph-null, next comes C,
then all the subsets of C, then all the subsets of all the subsets of C, and
so on. The ladder can also be expressed like this:
1 h 11 C 2
c 22c 2
22C
a ep -nu , , , , , ....
Cantor called C "aleph-one" because he believed that no transfinite
number existed between aleph-null and C. And he called the next num-
ber aleph-two, the next aleph-three, and so on. For many years he tried
unsuccessfully to prove that C was the next higher transfinite number
after aleph-null, a conjecture that came to be called the Continuum
Hypothesis. We now know, thanks to proofs by Kurt G6del and Paul
Cohen, that the conjecture is undecidable within standard set theory,
even when strengthened by the axiom of choice. We can assume with-
out contradiction that Cantor's alephs catch all transfinite numbers, or
we can assume, also without contradiction, a non-Cantorian set theory
in which there is an infinity of transfinite numbers between any two
adjacent entries in Cantor's ladder. (See the previous chapter for a brief,
informal account of this.)
Cantor also tried to prove that the number of points on a square is the
next higher transfinite cardinal after C. In 1877 he astounded himself
by finding an ingenious way to match all the points of a square to all
the points of a line segment. Imagine a square one mile on a side, and
a line segment one inch long (see Figure 26.2). On the line segment
every point from 0 to 1 is labeled with an infinite decimal fraction:
The point corresponding to the fractional part of pi is .14159 ... , the
point corresponding to 1/3 is .33333 ... , and so on. Every point is rep-
resented by a unique string of aleph-null digits, and every possible
aleph-null string of digits represents a unique point on the line seg-
ment. (A slight difficulty arises from the fact that a fraction such as
.5000 ... is the same as .4999 ... , but it is easily overcome by dodges
we need not go into here.)
Now consider the square mile. Using a Cartesian coordinate system,
every point on the square has unique x and y coordinates, each of
which can be represented by an endless decimal fraction. The illustra-
342 INFINITY
.73205 ... - ---<

. 14159 ... .1743125095 ...

Figure 26.2. Points in square mile and on line segment
tion shows a point whose x coordinate is the fractional part of pi and
whose y coordinate is the fractional part of the square root of 3, or
.73205 .... Starting with the x coordinate, alternate the digits of the two
numbers: .lZ4g,lZ5Q9.Q. .... The result is an endless decimal labeling a
unique point on the line segment. Clearly this can be done with every
point on the square. It is equally obvious that the mapping procedure
can be reversed: we can select any point on the line segment and, by
taking alternate digits of its infinite decimal, can split it into two end-
less decimals that as coordinates label a unique point on the square.
(Here we must recognize and overcome the subtle fact that, for exam-
ple, the following three distinct points on the segment-.449999 ... ,
.459090 ... , and .540909 ... -all map the same point [V2, V2] in the
square.) In this way the points of any square can be put into one-to-one
correspondence with the points on any line segment; therefore the two
sets are equivalent and each has the cardinal number C.
The proof extends easily to a cube (by interlacing three coordinates)
or to a hypercube of n dimensions (by interlacing n coordinates). Other
proofs show that C also numbers the points in an infinite space of any
finite number of dimensions, even an infinite space of aleph-null di-
mensions.
Cantor hoped that his transfinite numbers would distinguish the dif-
ferent orders of space but, as we have seen, he himself proved that this
was not the case. Mathematicians later showed that it is the topologi-
cal way the points of space go together that distinguishes one space
from another. The matchings in the previous paragraphs are not con-
tinuous; that is, points close together on, for instance, the line are not
necessarily close together on the square, and vice versa. Put another
way, you cannot continuously deform a line to make it a square, or a
square to make it a cube, or a cube to make a hypercube, and so on.
Supertasks 343
Is there a set in mathematics that corresponds to 2
c
? Of course we
know it is the number of all subsets of the real numbers, but does it
apply to any familiar set in mathematics? Yes, it is the set of all real
functions of x, even the set of all real one-valued functions. This is the
same as the nurnber of all possible permutations of the points on a line.
Geometrically it is all the curves (including discontinuous ones) that
can be drawn on a plane or even a small finite portion of a plane the
size, say, of a postage stamp. As for 2 to the power of 2
c
, no one has yet
found a set, aside from the subsets of 2
c
, equal to it. Only aleph-null,
C, and 2
c
seem to have an application outside the higher reaches of set
theory. As George Gamow once said, "We find ourselves here in a po-
sition exactly opposite to that of ... the Hottentot who had many sons
but could not count beyond three." There is an endless ladder of trans-
finite numbers. but most mathematicians have only three "sons" to
count with them. This has not prevented philosophers from trying to
find metaphysical interpretations for the transfinite numbers. Cantor
himself, a deeply religious man, wrote at length on such matters. In the
United States, Josiah Royce was the philosopher who made the most
extensive use of Cantor's alephs, particularly in his work The World
and the Individual.
The fact that there is no highest or final integer is involved in a vari-
ety of bewildering paradoxes. Known as supertasks, they have been
much debated by philosophers of science since they were first sug-
gested by the mathematician Hermann Weyl. For instance, imagine a
lamp (called the Thomson lamp after James F. Thomson, who first
wrote about it) that is turned off and on by a push-button switch. Start-
ing at zero time, the lamp is on for 1/2 minute, then it is off for 1/4
minute, then on for 1/8 minute, and so on. Since the sum of this halv-
ing series, 1/2 + 1/4 + 1/8 + ... +, is 1, at the end of one minute the
switch will have been moved aleph-null times. Will the lamp be on or
off?
Everyone agrees that a Thomson lamp cannot be constructed. Is such
a lamp logically conceivable or is it nonsense to discuss it in the ab-
stract? One of Zeno's celebrated paradoxes concerns a constant-speed
runner who goes half of a certain distance in 1/2 minute, a fourth of the
distance in the next 114 minute, an eighth of the distance in the next
1/8 minute, and so on. At the end of one minute he has had no diffi-
culty reaching the last point of the distance. Why, then, cannot we say
that at the end of one minute the switch of the Thomson lamp has
344 INFINITY
made its last move? The answer is that the lamp must then be on or off
and this is the same as saying that there is a last integer that is either
even or odd. Since the integers have no last digit, the lamp's operation
seems logically absurd.
Another supertask concerns an "infinity machine" that calculates
and prints the value of pi. Each digit is printed in half the time it takes
to print the preceding one. Moreover, the digits are printed on an ide-
alized tape of finite length, each digit having half the width of the one
before it. Both the time and the width series converge to the same limit,
so that in theory one might expect the pi machine, in a finite time, to
print all the digits of pi on a piece of tape. But pi has no final digit to
print, and so again the supertask seems self-contradictory.
One final example: Max Black (1909-1988) imagined a machine that
transfers a marble from tray A to tray B in one minute and then rests for
a minute as a second machine returns the marble to A. In the next half-
minute the first machine moves the marble back to B; then it rests for
a half-minute as the other machine returns it to A. This continues, in a
halving time series, until the machines' movements become, as Black
put it, a "grey blur." At the end of four minutes each machine has made
aleph-null transfers. Where is the marble? Once more, the fact that
there is no last integer to be odd or even seems to rule out the possi-
bility, even in principle, of such a supertask. (The basic articles on su-
pertasks, by Thomson, Black, and others, are reprinted in Wesley C.
Salmon's 1970 paperback anthology Zeno's Paradoxes.)
One is tempted to say that the basic difference between supertasks
and Zeno's runner is that the runner moves continuously whereas the
supertasks are performed in discrete steps that form an aleph-null set.
The situation is more complicated than that. Adolph Griinbaum, in
Modern Science and Zeno's Paradoxes, argues convincingly that Zeno's
runner could also complete his run by what Griinbaum calls a "stac-
cato" motion of aleph-null steps. The staccato runner goes the first half
of his distance in 1/4 minute, rests 1/4 minute, goes half of the re-
maining distance in 1/8 minute, rests 1/8 minute, and so on. When he
is running, he moves twice as fast as his "legato" counterpart, but his
overall average speed is the same, and it is always less than the veloc-
ity of light. Since the pauses of the staccato runner converge to zero, at
the end of one minute he too will have reached his final point just as
an ideal bouncing ball comes to rest after an infinity of discrete
bounces. Griinbaum finds no logical objection to the staccato run, even
Supertasks 345
though it cannot be carried out in practice. His attitudes toward the su-
pertasks are complex and controversial. He regards infinity machines
of certain designs as being logically impossible and yet in most cases,
with suitable qualifications, he defends them as logically consistent
variants of the staccato run.
These questions are related to an old argument to the effect that Can-
tor was mistaken in his claim that aleph-null and C are different orders
of infinity. The proof is displayed in Figure 26.3. The left side is an end-
less list of integers in serial order. Each is matched with a number on
the right that is formed by reversing the order of the digits and putting
a decimal in front of them. Since the list on the left can go to infinity,
it should eventually include every possible sequence of digits. If it
does, the numbers on the right will also catch every possible sequence
and therefore will represent all real numbers between 0 and 1. The real
numbers form a set of size C. Since this set can be put in one-to-one cor-
respondence with the integers, an aleph-null set, the two sets appear to
be equivalent.
INTEGERS
1
2
3
10
11
12
100
101
1234
DECIMAL FRACTIONS
.1
.2
.3
.01
.11
.21
.001
.101
.4321
Figure 26.3. Fallacious proof concerning two alephs
I would be ashamed to give this proof were it not for the fact that
every year or so I receive it from a correspondent who has rediscovered
it and convinced himself that he has demolished Cantorian set theory.
Readers should have little difficulty seeing what is wrong.
346 INFINITY
Addendum
Among physicists, no one objected more violently to Cantorian
set theory than Percy W. Bridgman. In Reflections of a Physicist (1955)
he says he "cannot see an iota of appeal" in Cantor's proof that the real
numbers form a set of higher infinity than the integers. Nor can he find
paradox in any of Zeno's arguments because he is unable to think of a
line as a set of points (see the Clerihew by Lindon that I used as an epi-
graph) or a time interval as a set of instants.
"A point is a curious thing," he wrote in The Way Things Are (1959),
"and I do not believe that its nature is appreciated, even by many math-
ematicians. A line is not composed of points in any real sense .... We
do not construct the line out of points, but, given the line, we may con-
struct points on it. 'All the points on the line' has the same sort of
meaning that the 'entire line' has .... We create the points on a line just
as we create the numbers, and we identify the points by the numerical
values of the coordinates."
Merwin J. Lyng, in The Mathematics Teacher (April 1968, p. 393),
gives an amusing variation of Black's moving-marble supertask. A box
has a hole at each end: Inside the box a rabbit sticks his head out of hole
A, then a minute later out of hole B, then a half-minute later out of hole
A, and so on. His students concluded that after two minutes the head
is sticking out of both holes, "but practically the problem is not possi-
ble unless we split hares."
For what it is worth, I agree with those who believe that paradoxes
such as the staccato run can be stated without contradiction in the lan-
guage of set theory, but as soon as any element is added to the task that
involves a highest integer, you add something not permitted, therefore
you add only nonsense. There is nothing wrong in the abstract about an
ideal bouncing ball coming to rest, or a staccato moving point reaching a
goal, but nothing meaningful is added if you assume that at each bounce
the ball changes color, alternating red and blue; then ask what color it is
when it stops bouncing, or if the staccato runner opens and shuts his
mouth at each step and you ask if it is open or closed at the finish.
A number of readers called my attention to errors in this chapter, as
I first wrote it as a column, but I wish particularly to thank Leonard
Gillman, of the University of Texas at Austin for reviewing the column
and suggesting numerous revisions that have greatly simplified and
improved the text.
Supertasks 347
Answers
The fundamental error in the false proof that the counting num-
bers can be matched one-to-one with the real numbers is that, no mat-
ter how long the list of integers on the left (and their mirror reversals
on the right), no number with aleph-null digits will ever appear on
each side. As a consequence no irrational decimal fraction will be listed
on the right. The mirror reversals of the counting numbers, with a dec-
imal point in front of each, form no more than a subset of the integral
fractions between 0 and 1. Not even 1/3 appears in this subset because
its decimal form requires aleph-null digits. In brief, all that is proved
is the well-known fact that the counting numbers can be matched one-
to-one with a subset of integral fractions.
The false proof reminds me of a quatrain I once perpetrated:
Pi vs e
Pi goes on and on and on ...
And e is just as cursed.
I wonder: Which is larger
When their digits are reversed?
Bibliography
P. Benacerraf, "Tasks, Super-Tasks, and Modern Eleatics," Journal of Philosophy,
Vol. 59, 1962,pp. 765-84.
J. Benardete, Infinity, an Essay in Metaphysics, Clarendon Press, 1964.
P. Clark and S. Read, "Hypertasks," Synthese, Vol. 61, 1984, pp. 387-90.
J. Earman and J. D. Norton, "Infinite Pains: The Trouble with Supertasks," Benac-
erraf and His Critics, A. Morton and S. P. Stich (eds.), Blackwell, 1966, pp.
231-61.
C. D. Gruender, "The Achilles Paradox and Transfinite Numbers," British Journal
for the Philosophy of Science, Vol. 17, November 1966, pp. 219-31.
A. Griinbaum, Modern Science and Zeno's Paradoxes, Wesleyan University Press,
1967.
A. Griinbaum, "Are 'Infinity Machines' Paradoxical?" Science, Vol. 159, January 26,
1968,pp.396-406.
A. Griinbaum, "Can an Infinitude of Operations Be Performed in a Finite Time?"
British Journal for the Philosophy of Science, Vol. 20, October 1969, pp. 203-18.
C. W. Kilmister, "Zeno, Aristotle, Weyl and Shuard: Two-and-a-half millenia of
worries over number," The Mathematical Gazette, Vol. 64, October 1980, pp.
149-58.
J. P. Laraudogoittia, "A Beautiful Supertask," Mind, Vol. 105, 1996, pp. 81-83.
348 INFINITY
J. P. Laraudogoittia, "Classical Particle Dynamics, Indeterminism, and a Super-
task," British Journal for the Philosophy of Science, Vol. 48, 1997, pp. 49-54.
J. P. Laroudogoittia, "Some Relativistic and Higher Order Supertasks," Philosophy
of Science, Vol. 65, 1998, pp. 502-17.
W. C. Salmon (ed.), Zeno's Paradoxes, Babbs-Merrill, 1970.
J. P. Van Bendegem, "Ross' Paradox is an Impossible Super-Task," British Journal
for the Philosophy of Science, Vol. 45, 1994, pp. 743-48.
Supertasks 349
Chapter 27 Fractal Music
For when there are no words [accompanying
music] it is very difficult to recognize the
meaning of the harmony and rhythm, or to
see that any worthy object is imitated by
them. -PLATO, Laws, Book II
Plato and Aristotle agreed that in some fashion all the fine arts,
including music, "imitate" nature, and from their day until the late
18th century "imitation" was a central concept in western aesthetics.
It is obvious how representational painting and sculpture "represent,"
and how fiction and the stage copy life, but in what sense does music
imitate?
By the mid-18th century philosophers and critics were still arguing
over exactly how the arts imitate and whether the term is relevant to
music. The rhythms of music may be said to imitate such natural
rhythms as heartbeats, walking, running, flapping wings, waving fins,
water waves, the periodic motions of heavenly bodies, and so on, but
this does not explain why we enjoy music more than, say, the sound of .
cicadas or the ticking of clocks. Musical pleasure derives mainly from
tone patterns, and nature, though noisy, is singularly devoid of tones.
Occasionally wind blows over some object to produce a tone, cats
howl, birds warble, bowstrings twang. A Greek legend tells how Her-
mes invented the lyre: he found a turtle shell with tendons attached to
it that produced musical tones when they were plucked.
Above all, human beings sing. Musical instruments may be said to
imitate song, but what does singing imitate? A sad, happy, angry, or
serene song somehow resembles sadness, joy, anger, or serenity, but if
a melody has no words and invokes no special mood, what does it
copy? It is easy to understand Plato's mystification.
There is one exception: the kind of imitation that plays a role in
"program music." A lyre is severely limited in the natural sounds it can
copy, but such limitations do not apply to symphonic or electronic
music. Program music has no difficulty featuring the sounds of thun-
der, wind, rain, fire, ocean waves, and brook murmurings; bird calls
350
(cuckoos and crowing cocks have been particularly popular), frog
croaks, the gaits of animals (the thundering hoofbeats in Wagner's Ride
of the Valkyries), the flights of bumblebees; the rolling of trains, the
clang of hammers; the battle sounds of marching soldiers, clashing
armies, roaring cannons, and exploding bombs. Slaughter on Tenth Av-
enue includes a pistol shot and the wail of a police-car siren. In Bach's
Saint Matthew Passion we hear the earthquake and the ripping of the
temple veil. In the Alpine Symphony by Richard Strauss, cowbells are
imitated by the shaking of cowbells. Strauss insisted he could tell that
a certain female character in Felix Mottl's Don Juan had red hair, and
he once said that someday music would be able to distinguish the clat-
tering of spoons from that of forks.
Such imitative noises are surely a trivial aspect of music even when
it accompanies opera, ballet, or the cinema; besides, such sounds play
no role whatsoever in "absolute music," music not intended to "mean"
anything. A Platonist might argue that abstract music imitates emo-
tions, or beauty, or the divine harmony of God or the gods, but on more
mundane levels music is the least imitative of the arts. Even nonobjec-
tive paintings resemble certain patterns of nature, but nonobjective
music resembles nothing except itself.
Since the turn of the century most critics have agreed that "imitation"
has been given so many meanings (almost all are found in Plato) that it
has become a useless synonym for "resemblance." When it is made
precise with reference to literature or the visual arts, its meaning is ob-
vious and trivial. When it is applied to music, its meaning is too fuzzy
to be helpful. In this chapter we take a look at a surprising discovery by
Richard F. Voss, a physicist from Minnesota who joined the Thomas J.
Watson Research Center of the International Business Machines Cor-
poration after obtaining his Ph.D. at the University of California at
Berkeley under the guidance of John Clarke. This work is not likely to
restore "imitation" to the lexicon of musical criticism, but it does sug-
gest a curious way in which good music may mirror a subtle statistical
property of the world.
The key concepts behind Voss's discovery are what mathematicians
and physicists call the spectral density (or power spectrum) of a fluc-
tuating quantity, and its" autocorrelation." These deep notions are tech-
nical and hard to understand. Benoit Mandelbrot, who is also at the
Watson Research Center, and whose work makes extensive use of spec-
tral densities and autocorrelation functions, has suggested a way of
Fractal Music 351
avoiding them here. Let the tape of a sound be played faster or slower
than normal. One expects the character of the sound to change consid-
erably. A violin, for example, no longer sounds like a violin. There is a
special class of sounds, however, that behave quite differently. If you
playa recording of such a sound at a different speed, you have only to
adjust the volume to make it sound exactly as before. Mandelbrot calls
such sounds "scaling noises."
By far the simplest example of a scaling noise is what in electronics
and information theory is called white noise (or "Johnson noise"). To
be white is to be colorless. White noise is a colorless hiss that is just as
dull whether you play it faster or slower. Its autocorrelation function,
which measures how its fluctuations at any moment are related to pre-
vious fluctuations, is zero except at the origin, where of course it must
be 1. The most commonly encountered white noise is the thermal noise
produced by the random motions of electrons through an electrical re-
sistance. It causes most of the static in a radio or amplifier and the
"snow" on radar and television screens when there is no input.
With randomizers such as dice or spinners it is easy to generate white
noise that can then be used for composing a random "white tune," one
with no correlation between any two notes. Our scale will be one oc-
tave of seven white keys on a piano: do, re, me, fa, so, la, ti. Fa is our
middle frequency. Now construct a spinner such as the one shown at
the left in Figure 27.1. Divide the circle into seven sectors and label
them with the notes. It matters not at all what arc lengths are assigned
to these sectors; they can be completely arbitrary. On the spinner
shown, some order has been imposed by giving fa the longest arc (the
highest probability of being chosen) and assigning decreasing proba-
Figure 21.1. Spinners for white music (left) and brown music (right)
352 INFINITY
bilities to pairs of notes that are equal distances above and below fa.
This has the effect of clustering the tones around fa.
To produce a "white melody" simply spin the spinner as often as you
like, recording each chosen note. Since no tone is related in any way to
the sequence of notes that precedes it, the result is a totally uncorre-
lated sequence. If you like, you can divide the circle into more parts
and let the spinner select notes that range over the entire piano key-
board, black keys as well as white.
To make your white melody more sophisticated, use another spinner,
its circle divided into four parts (with any proportions you like), and
labeled 1, 1/2, 1/4, and 1/8 so that you can assign a full, a half, a quar-
ter, or an eighth of a beat to each tone. After the composition is com-
pleted, tap it out on the piano. The music will sound just like what it
is: random music of the dull kind that a two-year-old or a monkey
might produce by hitting keys with one finger. Similar white music
can be based on random number tables or the digits in an irrational
number.
A more complicated kind of scaling noise is one that is sometimes
called Brownian noise because it is characteristic of Brownian motion,
the random movements of small particles suspended in a liquid and
buffeted by the thermal agitation of molecules. Each particle executes
a three-dimensional "random walk," the positions in which form a
highly correlated sequence. The particle, so to speak, always "remem-
bers" where it has been.
When tones fluctuate in this fashion, let us follow Voss and call it
Brownian music or brown music. We can produce it easily with a spin-
ner and a circle divided into seven parts as before, but now we label the
regions, as shown at the right in Figure 27.1, to represent intervals be-
tween successive tones. These step sizes and their probabilities can be
whatever we like. On the spinner shown, plus means a step up the
scale of one, two, or three notes and minus means a step down of the
same intervals.
Start the melody on the piano's middle C, then use the spinner to
generate a linear random walk up and down the keyboard. The tune
will wander here and there, and it will eventually wander off the key-
board. If we treat the ends of the keyboard as "absorbing barriers," the
tune ends when we encounter one of them. We need not go into the
ways in which we can treat the barriers as reflecting barriers, allowing
the tune to bounce back, or as elastic barriers. To make the barriers
Fractal Music 353
elastic we must add rules so that the farther the tone gets from middle
C, the greater is the likelihood it will step back toward C, like a marble
wobbling from side to side as it rolls down a curved trough.
As before, we can make our brown music more sophisticated by vary-
ing the tone durations. If we like, we can do this in a brown way by
using another spinner to give not the duration but the increase or de-
crease of the duration-another random walk but one along a different
street. The result is a tune that sounds quite different from a white tune
because it is strongly correlated, but a tune that still has little aesthetic
appeal. It simply wanders up and down like a drunk weaving through
an alley, never producing anything that resembles good music.
If we want to mediate between the extremes of white and brown, we
can do it in two essentially different ways. The way chosen by previ-
ous composers of "stochastic music" is to adopt transition rules. These
are rules that select each note on the basis of the last three or four. For
example, one can analyze Bach's music and determine how often a cer-
tain note follows, say, a certain triplet of preceding notes. The random
selection of each note is then weighted with probabilities derived from
a statistical analysis of all Bach quadruplets. If there are certain transi-
tions that never appear in Bach's music, we add rejection rules to pre-
vent the undesirable transitions. The result is stochastic music that
resembles Bach but only superficially. It sounds Bachlike in the short
run but random in the long run. Consider the melody over periods of
four or five notes and the tones are strongly correlated. Compare a run
of five notes with another five-note run later on and you are back to
white noise. One run has no correlation with the other. Almost all sto-
chastic music produced so far has been of this sort. It sounds musical
if you listen to any small part but random and uninteresting when you
try to grasp the pattern as a whole.
Voss's insight was to compromise between white and brown input by
selecting a scaling noise exactly halfway between. In spectral termi-
nology it is called llfnoise. (White noise has a spectral density of llf;
brownian noise has a spectral density of llf. In "one-over-/' noise the
exponent of fis 1 or very close to 1.) Tunes based on llfnoise are mod-
erately correlated, not just over short runs but throughout runs of any
size. It turns out that almost every listener agrees that such music is
much more pleasing than white or brown music.
In electronics llf noise is well known but poorly understood. It is
sometimes called flicker noise. Mandelbrot, whose book The Fractal
354 INFINITY
Geometry of Nature (W. H. Freeman, 1982) has already become a mod-
ern classic, was the first to recognize how widespread 11fnoise is in na-
ture, outside of physics, and how often one encounters other scaling
fluctuations. For example, he discovered that the record of the annual
flood levels of the Nile is a 1lffluctuation. He also investigated how the
curves that graph such fluctuations are related to "fractals," a term he
invented. A scaling fractal can be defined roughly as any geometrical
pattern (other than Euclidean lines, planes, and surfaces) with the re-
markable property that no matter how closely you inspect it, it always
looks the same, just as a slowed or speeded scaling noise always sounds
the same. Mandelbrot coined the term fractal because he assigns to
each of the curves a fractional dimension greater than its topological di-
mension.
Among the fractals that exhibit strong regularity the best-known are
the Peano curves that completely fill a finite region and the beautiful
snowflake curve discovered by the Swedish mathematician Helge von
Koch in 1904. The Koch snowflake appears in Figure 27.2 as the bound-
ary of the dark "sea" that surrounds the central motif. (For details on
the snowflake's construction, and a discussion of fractals in general, see
Chapter 3 of my Penrose Tiles to Trapdoor Ciphers [W. H. Freeman,
1989]).
The most interesting part of Figure 27.2 is the fractal curve that forms
the central design. It was discovered by Mandelbrot and published for
the first time as the cover of Scientific American's April 1978 issue. If
you trace the boundary between the black and white regions from the
tip of the point of the star at the lower left to the tip of the point of the
star at the lower right, you will find this boundary to be a single curve.
It is the third stage in the construction of a new Peano curve. At the
limit this lovely curve will completely fill a region bounded by the tra-
ditional snowflake! Thus Mandelbrot's curve brings together two path-
breaking fractals: the oldest of them all, Giuseppe Peano's 1890 curve,
and Koch's later snowflake!
The secret of the curve's construction is the use of line segments of
two unequal lengths and oriented in 12 different directions. The curve
is much less regular than previous Peano curves and therefore closer to
the modeling of natural phenomena, the central theme of Mandelbrot's
book. Such natural forms as the gnarled branches of a tree or the shapes
of flickering flames can be seen in the pattern.
At the left in Figure 27.3 is the first step of the construction. A
Fractal Music 355
A B
21.3. two in curve
356
357
vertical cross section of the topography has a profile that models a ran-
dom walk. The white patches, representing water or snow in the hol-
lows below a certain altitude, were added to enhance the relief.
The profile at the top of the mountain range is a scaling fractal. This
means that if you enlarge any small portion of it, it will have the same
statistical character as the line you now see. If it were a true fractal, this
property would continue forever as smaller and smaller segments are
enlarged, but of course such a curve can neither be drawn nor appear
in nature. A coastline, for example, may be self-similar when viewed
from a height of several miles down to several feet, but below that the
fractal property is lost. Even the Brownian motion of a particle is lim-
ited by the size of its microsteps.
Since mountain ranges approximate random walks, one can create
"mountain music" by photographing a mountain range and translating
its fluctuating heights to tones that fluctuate in time. Villa Lobos actu-
ally did this using mountain skylines around Rio de Janeiro. If we view
nature statically, frozen in time, we can find thousands of natural
curves that can be used in this way to produce stochastic music. Such
music is usually too brown, too correlated, however, to be interesting.
Like natural white noise, natural brown noise may do well enough,
perhaps, for the patterns of abstract art but not so well as a basis for
music.
When we analyze the dynamic world, made up of quantities con-
stantly changing in time, we find a wealth of fractal like fluctuations
that have 11 f spectral densities. In his book Mandelbrot cites a few:
variations in sunspots, the wobbling of the earth's axis, undersea cur-
rents, membrane currents in the nervous system of animals, the fluc-
tuating levels of rivers, and so on. Uncertainties in time measured by
an atomic clock are llf: the error is 10-
12
regardless of whether one is
measuring an error on a second, minute, or hour. Scientists tend to
overlook llf noises because there are no good theories to account for
them, but there is scarcely an aspect of nature in which they cannot be
found.
T. Musha, a physicist at the Tokyo Institute of Technology, discovered
that traffic flow past a certain spot on a Japanese expressway exhibited
llf fluctuation. In a more startling experiment, Musha rotated a radar
beam emanating from a coastal location to get a maximum variety of
landscape on the radar screen. When he rotated the beam once, varia-
tions in the distances of all objects scanned by the beam produced a
358 INFINITY
Brownian spectrum. But when he rotated it twice and then subtracted
one curve from the other the resulting curve-representing all the
changes of the scene-was close to 11 f.
We are now approaching an understanding of Voss's daring conjec-
ture. The changing landscape of the world (or, to put it another way, the
changing content of our total experience) seems to cluster around llf
noise. It is certainly not entirely uncorrelated, like white noise, nor is
it as strongly correlated as brown noise. From the cradle to the grave
our brain is processing the fluctuating data that comes to it from its sen-
sors. If we measure this noise at the peripheries of the nervous system
(under the skin of the fingers), it tends, Mandelbrot says, to be white.
The closer one gets to the brain, however, the closer the electrical fluc-
tuations approach llf. The nervous system seems to act like a complex
filtering device, screening out irrelevant elements and processing only
the patterns of change that are useful for intelligent behavior.
On the canvas of a painting, colors and shapes are static, reflecting
the world's static patterns. Is it possible, Mandelbrot asked himself
many years ago, that even completely nonobjective art, when it is pleas-
ing, reflects fractal patterns of nature? He is fond of abstract art and
maintains that there is a sharp distinction between such art that has a
fractal base and such art that does not, and that the former type is
widely considered the more beautiful. Perhaps this is why photogra-
phers with a keen sense of aesthetics find it easy to take pictures, par-
ticularly photomicrographs, of natural patterns that are almost
indistinguishable from abstract expressionist art.
Motion can be added to visual art, of course, in the form of the mo-
tion picture, the stage, kinetic art, and dance, but in music we have
meaningless, nonrepresentational tones that fluctuate to create a pat-
tern that can be appreciated only over a period of time. Is it possible,
Voss asked himself, that the pleasures of music are partly related to
scaling noise of llf spectral density? That is, is this music "imitating"
the llf quality of our flickering experience?
That mayor may not be true, but there is no doubt that music of al-
most every variety does exhibit llf fluctuations in its changes of pitch
as well as in the changing loudness of its tones. Voss found this to be
true of classical music, jazz, and rock. He suspects it is true of all music.
He was therefore not surprised that when he used a llf flicker noise
from a transistor to generate a random tune, it turned out to be more
pleasing than tunes based on white and brown noise sources.
Fractal Music 359
Figure 27.5, supplied by Voss, shows typical patterns of white, llf,
and brown when noise values (vertical) are plotted against time (hori-
zontal). These patterns were obtained by a computer program that sim-
ulates the generation of the three kinds of sequences by tossing dice.
The white noise is based on the sum obtained by repeated tosses of 10
dice. These sums range from 10 to 60, but the probabilities naturally
force a clustering around the median. The Brownian noise was gener-
50
40
30
20
WhIte noise
50
40
30
20
1/f Noise
60
40
20
o
Brownian noIse
o
128 256 384 512
Time
640 768
Figure 27.S. Typical patterns of white, 1lf, and Brownian noise
360
896 1024
INFINITY
ated by tossing a single die and going up one step on the scale if the
number was even and down a step if the number was odd.
The llfnoise was also generated by simulating the tossing of 10 dice.
Although llf noise is extremely common in nature, it was assumed
until recently that it is unusually cumbersome to simulate llfnoise by
randomizers or computers. Previous composers of stochastic music
probably did not even know about llf noise, but if they did, they would
have had considerable difficulty generating it. As this article was being
prepared Voss was asked if he could devise a simple procedure by
which readers could produce their own llf tunes. He gave some
thought to the problem and to his surprise hit on a clever way of sim-
plifying existing llf computer algorithms that does the trick beauti-
fully.
The method is best explained by considering a sequence of eight
notes chosen from a scale of 16 tones. We use three dice of three col-
ors: red, green, and blue. Their possible sums range from 3 to 18. Select
16 adjacent notes on a piano, black keys as well as white if you like,
and number them 3 through 18.
Write down the first eight numbers, 0 through 7, in binary notation,
and assign a die color to each column as is shown in Figure 27.6. The
first note of our tune is obtained by tossing all three dice and picking
the tone that corresponds to the sum. Note that in going from 000 to 001
only the red digit changes. Leave the green and blue dice undisturbed,
still showing the numbers of the previous toss. Pick up only the red die
\I)
~ '0 :;,
1ii
- \I)
(!J ex:
t t t
0 000
1 001
2 o 1 0
3 o 1 1
4 1 0 0
5 101
6 1 1 0
7 1 1 1
Figure 27.6. Binary chart for Voss's llf dice algorithm
Fractal Music 361
and toss it. The new sum of all three dice gives the second note of your
tune. In the next transition, from 001 to 010, both the red and green dig-
its change. Pick up the red and green dice, leaving the blue one undis-
turbed, and toss the pair. The sum of all three dice gives the third tone.
The fourth note is found by shaking only the red die; the fifth, by shak-
ing all three. The procedure, in short, is to shake only those dice that
correspond to digit changes.
It is not hard to see how this algorithm produces a sequence halfway
between white and brown. The least significant digits, those to the
right, change often. The more significant digits, those to the left, are
more stable. As a result, dice corresponding to them make a constant
contribution to the sum over long periods of time. The resulting se-
quence is not precisely llfbut is so close to it that it is impossible to
distinguish melodies formed in this way from tunes generated by nat-
uralllfnoise. Four dice can be used the same way for a llf sequence
of 16 notes chosen from a scale of 21 tones. With 10 dice you can gen-
erate a melody of 210, or 1,024, notes from a scale of 55 tones. Similar
algorithms can of course be implemented with generalized dice (octa-
hedrons, dodecahedrons, and so on), spinners, or even tossed coins.
With the same dice simulation program Voss has supplied three typ-
ical melodies based on white, brown, and llf noise. The computer
printouts of the melodies are shown in 27.7-27.9. In each case Voss var-
ied both the melody and the tone duration with the same kind of noise.
Above each tune are shown the noise patterns that were used.
Over a period of two years, tunes of the three kinds were played at
various universities and research laboratories, for many hundreds of
people. Most listeners found the white music too random, the brown
too correlated, and the llf "just about right." Indeed, it takes only a
glance at the music itself to see how the 11 f property mediates between
the two extremes. Voss's earlier llf music was based on natural llf
noise, usually electronic, even though one of his best compositions de-
rives from the record of the annual flood levels of the Nile. He has
made no attempt to impose constant rhythms. When he applied llf
noise to a pentatonic (five-tone) scale and also varied the rhythm with
llf noise, the music strongly resembled Oriental music. He has not
tried to improve his llfmusic by adding transition or rejection rules.
It is his belief that stochastic music with such rules will be greatly im-
proved if the underlying choices are based on 11 f noise rather than the
white noise so far used.
362 INFINITY
35
30
.c
~ 25
c:
o
~
:s
o
20
15
o 16
Figure 21.1. White music
10
.c
.g 5
Q.
o
5
-5
o 16
Figure 27.8. Brown music
Fractal Music
32 48
32 48
64
Time
64
Time
80
80
96 112 128
96 112 128
363
35
30
.s::.
!3 25
a::
20
15
35
30
c
0
;;
25
m
..
::I
20
Q
15
0 16
Figure 21.9. llf music
32 48 64
Time
80 96 112 128
Note that tlfmusic is halfway between white and brown in a fractal
sense, not in the manner of music that has transition rules added to
white music. As we have seen, such music reverts to white when we
compare widely separated parts. But tlf music has the fractal self-
similarity of a coastline or a mountain range. Analyze the fluctuations
on a small scale, from note to note, and it is tlf. The same is true if you
break a long tune into to-note sections and compare them. The tune
never forgets where it has been. There is always some correlation with
its entire past.
It is commonplace in musical criticism to say that we enjoy good
music because it offers a mixture of order and surprise. How could it
be otherwise? Surprise would not be surprise if there were not suffi-
cient order for us to anticipate what is likely to come next. If we guess
too accurately, say in listening to a tune that is no more than walking
up and down the keyboard in one-step intervals, there is no surprise at
all. Good music, like a person's life or the pageant of history, is a won-
drous mixture of expectation and unanticipated turns. There is nothing
new about this insight, but what Voss has done is to suggest a mathe-
matical measure for the mixture.
I cannot resist mentioning three curious ways of transforming a
364 INFINITY
melody to a different one with the same l/t spectral density for both
tone patterns and durations. One is to write the melody backward, an-
other is to turn it upside down, and the third is to do both. These trans-
formations are easily accomplished on a player piano by reversing
and/or inverting the paper roll. If a record or tape is played backward,
unpleasant effects result from a reversal of the dying-away quality of
tones. (Piano music sounds like organ music.) Reversal or inversion
naturally destroys the composer's transition patterns, and that is prob-
ably what makes the music sound so much worse than it does when it
is played normally. Since Voss composed his tunes without regard for
short-range transition rules, however, the tunes all sound the same
when they are played in either direction.
Canons for two voices were sometimes deliberately written, partic-
ularly in the 15th century, so that one melody is the other backward,
and composers often reversed short sequences for contrapuntal ef-
fects in longer works. Figure 27.10 shows a famous canon that Mozart
wrote as a joke. In this instance the second melody is almost the same
as the one you see taken backward and upside down. Thus if the sheet
is placed flat on a table, with one singer on one side and the other
singer on the other, the singers can read from the same sheet as they
harmonize!
No one pretends, of course, that stochastic l/t music, even with
added transition and rejection rules, can compete with the music of
good composers. We know that certain frequency ratios, such as the
three-to-two ratio of a perfect fifth, are more pleasing than others, either
when the two tones are played simultaneously or in sequence. But just
what composers do when they weave their beautiful patterns of mean-
ingless sounds remains a mystery that even they do not understand.
It is here that Plato and Aristotle seem to disagree. Plato viewed all
the fine arts with suspicion. They are, he said (or at least his Socrates
said), imitations of imitations. Each time something is copied some-
thing is lost. A picture of a bed is not as good as a real bed, and a real
bed is not as good as the universal, perfect idea of bedness. Plato was
less concerned with the sheer delight of art than with its effects on
character, and for that reason his Republic and Laws recommend strong
state censorship of all the fine arts.
Aristotle, on the other hand, recognized that the fine arts are of value
to a state primarily because they give pleasure and that this pleasure
springs from the fact that artists do much more than make poor copies.
Fractal Music 365
Allegro
.I
" t r r I r I I t I rtf r I r r' r J r r I r , , t I r r r, r , , r .t
" 0 J S j (f f a Ire 0 aBi! J r I rr G tEO ' D J J j .,
"r r rei au Ell! , u Q e en, Ell! , r , , , I, i 9 ' ,t
,t , ~ " r r j I j , ~ fir r f r , r r I r J r r I r f , r I r f f .,
" J ~ ~ f 1 f rr uri r , , r ,e e 0 t t 0 I CW" P' r '. tID .,
" n J Q J t ~ r () !t f!j J I r _, J J I r l fIiJ .,
" fU) J I r r I EI!J r I, ; Utt, 0 r r f) r r,' 'f F r u .,
,. ffil tJ J J I J ~ ~ r I r u _ I J ~ , fir r r I r if "
lZI!IOft o.lll'nv
Figure 21.'0. Mozart's palindromic and invertible canon
They said, "You have a blue guitar,
You do not play things as they are. "
The man replied, ((Things as they are
Are changed upon the blue guitar. "
Wallace Stevens intended his blue guitar to stand for all the arts, but
music, more than any other art and regardless of what imitative as-
pects it may have, involves the making of something utterly new. You
may occasionally encounter natural scenes that remind you of a paint-
ing, or episodes in life that make you think of a novel or a play. You will
never come on anything in nature that sounds like a symphony. As to
366
INFINITY
whether mathematicians will someday write computer programs that
will create good music-even a simple, memorable tune-time alone
will tell.
Addendum
Irving Godt, who teaches music history at the Indiana Univer-
sity of Pennsylvania, straightened me out on the so-called Mozart
canon with the following letter. It appeared in Scientific American (July
1978):
A few musical errors slipped past Martin Gardner's critical eye when he
took up "Mozart's palindromic and invertible canon" in his report on
fractal curves and "one-over-!, fluctuations.
Mozart scholars now agree that the canon is almost certainly not by
Mozart, even though publishers have issued it under his name. For more
than 40 years the compilers of the authoritative Kochel catalogue of
Mozart's compositions have relegated it to the appendix of doubtful at-
tributions, where along with three other pieces of a similar character, it
bears the catalogue number K. Anh. C 10.16. We have no evidence that
the piece goes back any further than the last century.
The piece is not for two singers but for two violins. Singers cannot pro-
duce the simultaneous notes of the chords in the second measure (and
elsewhere), and the ranges of the parts are quite impractical. To perform
the piece the two players begin from opposite ends of the sheet of music
and arrive at a result that falls far below the standard of Mozart's au-
thentic canons and other jeux d
1
esprit. The two parts combine for long
stretches of parallel octaves, they rarely achieve even the most rudi-
mentary rhythmic or directional independence, and their harmony con-
sists of little more than the most elementary writing in parallel thirds.
This little counterfeit is not nearly as interesting as Mr. Gardner's
columns.
John G. Fletcher wrote to suggest that because Ilfmusic lies between
white and brown music it should be called tan music. The term "pink"
has also been suggested, and actually used by some writers. Fate mag-
azine (October 1978) ran a full-page advertisement for an LP record
album produced by "Master Wilburn Burchette," of Spring Valley, CA,
titled Mind Storm. The ad calls it "fantastic new deep-hypnotic music
that uses a phenomenon known in acoustical science as 'pink sound'
to open the mind to thrilling psychic revelations! This astonishing new
Fractal Music 367
music acts something like a crystal ball reflecting back the images pro-
jected by the mind .... Your spirit will soar as this incredible record
album carries you to new heights of psychic awareness!"
Frank Greenberg called my attention to some "mountain music"
composed by Sergei Prokofiev for Sergei Eisenstein's film Alexander
Nevsky in 1938. "Eisenstein provided Prokofiev with still shots of in-
dividual scenes of the movie as it was being filmed. Prokofiev then
took these scenes and used the silhouette of the landscape and human
figures as a pattern for the position of the notes on the staff. He then or-
chestrated around these notes."
Bibliography
T. Johnson, "Self-Similar Melodies," Two-Eighteen Press, 1996.
S. K. Langer, "On Significance in Music," Philosophy in a New Key, Harvard Uni-
versity Press, 1957.
T. Musha, "11/Fluctuations in Biological Systems," Sixth International Symposium
on Noise in Physical Systems," National Bureau of Standards, 1981, pp. 143-46.
A. T. Scarpelli, "11/Random Tones: Making Music with Fractals," Personal Com-
puting, Vol. 3, 1979, pp. 17-27.
M. Schroeder, "Noises: White, Pink, Brown, and Black," Fractals, Chaos, Power
Laws, W. H. Freeman, 1991, Chapter 5.
D. E. Thomsen, "Making Music Fractally," Science News, Vol. 117,1980, pp. 187ff.
B. J. West and M. Shies inger, "The Noise in Natural Phenomena," American Sci-
entist, Vol. 78, 1978, pp. 40-45.
R. F. Voss and J. Clarke, "11/Noise in Music and Speech," Nature, Vol. 258, 1975,
pp.317-18.
R. F. Voss and J. Clarke, "11/Noise in Music," The Journal o/the Acoustical Soci-
etyo/ America, Vol. 63, 1978, pp. 258-63.
A. van der Ziel., Noise: Sources, Characterization, Measurement, Prentice-Hall,
1970.
368 INFINITY
Chapter 28
Some said "John, print it";
others said, "Not so."
Some said "It might do good";
others said, "No."
-JOHN BUNYAN, Apology for His Book
Surreal Numbers
John Horton Conway, the almost legendary mathematician of
the University of Cambridge (now at Princeton University)' quotes the
above lines at the end of the preface to his book, On Numbers and
Games (Academic Press, 1976), or ONAG, as he and his friends call it.
It is hard to imagine a mathematician who would say not so or no. The
book is vintage Conway: profound, pathbreaking, disturbing, original,
dazzling, witty, and splattered with outrageous Carrollian wordplay.
Mathematicians, from logicians and set theorists to the humblest ama-
teurs, will be kept busy for decades rediscovering what Conway has left
out or forgotten and exploring the strange new territories opened by his
work.
The sketch of Conway reproduced below could be titled "John
'Horned' (Horton) Conway." The infinitely regressing, interlocking
horns form, at the limit, what topologists call a "wild" structure; this
one is termed the Alexander horned sphere. Although it is equivalent
to the simply connected surface of a ball, it bounds a region that is not
simply connected. A loop of elastic cord circling the base of a horn can-
not be removed from the structure even in an infinity of steps.
Conway is the inventor of the computer game Life, discussed in de-
tail in Chapter 31 of this text. By carefully choosing a few ridiculously
simple transition rules Conway created a cellular automaton structure
of extraordinary depth and variety. Soon after, by invoking the sim-
plest possible distinction-a binary division between two sets-and
adding a few simple rules, definitions, and conventions, he constructed
a rich field of numbers and an equally rich associated structure of two-
person games.
The story of how Conway's numbers are created on successive
"days," starting with the zeroth day, is told in Donald E. Knuth's nov-
369
John "Horned" (Horton) Conway,
as sketched by a colleague on
computer-printout paper
eleUe Surreal Numbers (Addison-Wesley, 1974). Because I discuss
Knuth's book in chapter 48, I shall say no more about it here except to
remind readers that the construction of the numbers is based on one
rule: If we are given a left set L and a right set R and no member of Lis
equal to or greater than any member of R, then there is a number {L I R}
that is the "simplest number" (in a sense defined by Conway) in be-
tween.
By starting with literally nothing at all (the empty set) on the left
370 INFINITY
and the right, { I }, one obtains a definition of zero. Everything else fol-
lows by the technique of plugging newly created numbers back into the
left-right arrangement. The expression {o I o} is not a number, but {o I },
with the null set on the right, defines 1, { I o} defines -1, and so on.
Proceeding inductively, Conway is able to define all integers, all in-
tegral fractions, all irrationals, all of Georg Cantor's transfinite numbers,
a set of infinitesimals (they are the reciprocals of Cantor's numbers,
not the infinitesimals of nonstandard analysis), and infinite classes of
weird numbers never before seen by man, such as
V(ro + 1)
1t
ro
where co is omega, Cantor's first infinite ordinal.
Conway's games are constructed in a similar but more general way.
The fundamental rule is: If Land R are any two sets of games, there is
a game {L I R}. Some games correspond to numbers and some do not, but
all of them (like the numbers) rest on nothing. "We remind the reader
again," Conway writes, "that since ultimately we are reduced to ques-
tions about members of the empty set, no one of our inductions will re-
quire a 'basis.' "
In a "game" in Conway's system two players, Left and Right, alternate
moves. (Left and Right designate players, such as Black and White or
Arthur and Bertha, not who goes first or second.) Every game begins
with a first position, or state. At this state and at each subsequent state
a player has a choice of "options," or moves. Each choice completely
determines the next state. In standard play the first person unable to
make a legal move loses. This is a reasonable convention, Conway
writes, "since we normally consider ourselves as losing when we can-
not find any good move, we should obviously lose when we cannot
find any move at all!" In "misere" play, which is usually much more
difficult to analyze, the person who cannot move is the winner. Every
game can be diagrammed as a rooted tree, its branches signifying each
player's options at each successive state. On Conway's trees Left's op-
tions go up and to the left and Right's go up and to the right.
Games may be "impartial," as in Nim, which means that any legal
move can be made by the player whose turn it is to move. If a game is
not impartial, as in chess (where each player must move only his own
pieces), Conway calls it a partizan game. His net thus catches both an
enormous variety of familiar games, from Nim to chess, and an infin-
Surreal Numbers 371
ity of games never before imagined. Although his theory applies to
games with an infinity of states or to games with an infinity of options
or to both, he is concerned mainly with games that end after a finite
number of moves. "Left and Right," he explains, "are both busy men,
with heavy political responsibilities."
Conway illustrates the lower levels of his theory with positions taken
from a partizan domino-placing game. (Conway calls it Domineering.)
The board is a rectangular checkerboard of arbitrary size and shape.
Players alternately place a domino to cover two adjacent squares, but
Left must place his pieces vertically and Right must place his horizon-
tally. The first player who is unable to move loses.
An isolated empty square
D
allows no move by either player. "No move allowed" corresponds to
the empty set, so that in Conway's notation this simplest of all games
is assigned the value { I } = 0, the simplest of all numbers. Conway calls
it Endgame. Its tree diagram, shown at the right of the square, is merely
the root node with no branches. Because neither side can move, the sec-
ond player, regardless of whether he is Left or Right, is the winner. HI
courteously offer you the first move in this game," writes Conway.
Since you cannot move, he wins.
A vertical strip of two (or three) cells
offers no move to Right but allows one move for Left. Left's move leads
to a position of value 0, so that the value of this region is {o I } = 1. It is
the simplest of all positive games, and it corresponds to the Simplest
positive number. Positive games are wins for Left regardless of who
starts. The region's tree diagram is shown at the right above.
A horizontal strip of two (or three) cells
ITJ/
allows one move for Right but no move for Left. The value of the region
is { I o} = -1. It is the simplest of all negative games and corresponds to
372 INFINITY
the simplest negative number. Negative games are wins for Right re-
gardless of who starts.
A vertical strip of four (or five) cells
has a value of 2. Right has no moves. Left can, if he likes, take the two
middle cells in order to leave a zero position, but his "best" move is to
take two end cells, because that leaves him an additional move. If this
region is the entire board, then of course either play wins, but if it is an
isolated region in a larger board or in one of many boards in a "com-
pound game," it may be important to make the move that maximizes
the number of additional moves it leaves for the player. For this reason
the tree shows only Left's best line of play. The value of the game is
{l,o I } = {11 } = 2. A horizontal strip of four cells has a value of -2. If
only one player can move in a region and he can fit n of his dominoes
into it but no more, then clearly the region has the value +n if the player
is Left and -n if the player is Right.
Things get more interesting if both players can move in a region, be-
cause then one player may have ways of blocking his opponent. Con-
sider the following region:
Left can place a domino that blocks any move by Right, thus leaving a
zero position and winning. Right cannot similarly block Left because
Right's only move leaves a position of value 1. In Conway's notation the
value of this position is to, 111} = {Oil}, an expression that defines
1/2. The position therefore counts as half a move in favor of Left. By
turning the L region on its side one finds that the position is { - 11 O,l}
= { - 11 O} = -1/2, or half a move in favor of Right.
More complicated fractions arise in Conway's theory. For example,
Surreal Numbers 373
has a value of {1/211} = 314 of a move for Left, since 3/4 is the simplest
number between 1/2 and 1, the values of the best options for Left and
Right. In a game called partizan Hackenbush, Conway gives an exam-
ple of a position in which Left is exactly 5/64 move ahead!
The values of some game positions are not numbers at all. The sim-
plest example is illustrated in Domineering by this region:
EbV
Both Left and Right have opening moves only, so that the first person
to play wins regardless of whether he is Left or Right. Since each player
can reduce the value to 0, the value of the position is {o I O}. This is not
a number. Conway symbolizes it with * and calls it "star." Another ex-
ample would be a Nim heap containing a single counter. It is the sim-
plest "fuzzy" game. Fuzzy values correspond to positions in which
either player can win if he moves first.
The value of a compound game is simply the sum of the values of its
component games. This statement applies also to the value of a position
in a game in progress that has been divided by play into a set of sub-
games. For example, Figure 28.1 shows a position in a game of Domi-
neering played on a standard chessboard. The values of the isolated
regions are indicated. The position seems to be well balanced, but the
regions have a sum of 1 %, which means that Left is one and one-fourth
of a move ahead and therefore can win regardless of who moves next.
This outcome would be tedious to decide by drawing a complete tree,
but Conway's theory gives it quickly and automatically.
A game is not considered "solved" until its outcome (assuming that
both players make their best moves) is known (that is, whether the
game's value is zero, positive, negative, or fuzzy) and a successful strat-
egy is found for the player who has the win. This stricture applies only
to games that must end, but such games may offer infinite options, as
in the game Conway calls "My Dad Has More Money than Yours." Play-
ers alternately name a sum of money for just two moves and the high-
est sum wins. Although the tree, Conway admits, is complicated, the
outcome is clearly a second-player win.
Are these not trivial beginnings? Yes, but they provide a secure foun-
dation on which Conway, by plugging newly created games back into
his left-right scheme, carefully builds a vast and fantastic edifice. I
374 INFINITY
28.1. A UOJJllIlleeIulg IJ ..... ., .. L, ..... ' ..... in a win
I a few un-
usual the light
games we assume standard play in that the first person who unable
to move ... ""u ............
on
paper. L has black paint, R has white. They alternately color a region
be same
....... V ........... JL,c:;. a
all regions bordering a black one as ....,' ..H ........................ 'L ............ ...., ........ , ............ .
A region acquiring both tints drops out of the map as being an un-
..................... ,...., ....... one.
Colon the dual 28.2)'
defining what he calls "explosive nodes" and marking with light-
ning bolts. Of course, can played on white paper with pen-
distinct connected maps of one re-
gions 9 are first-player wins are second-player wins. The
2. same as that
neighboring regions must be the same color. It too is unsolved. Conway
Numbers 375
28.2. A for or Snort
........... 1-.1'1-' ........... it
can a
so.
A
.. II
IVRT-f:lrnn game same game
376
reader is familiar with Nim and knows how to determine the winning
strategy. If not, he can consult any number of books, including Con-
way's book or my Scientific American Book of Mathematical Puzzles 8'
Diversions [Simon and Schuster, 1959]). Corresponding to the Nim
heaps (or rows) of counters are the vacant cells between pennies, start-
ing with the vacancy at the extreme right and including only alternate
vacancies. In the illustration the Nim heaps are indicated by brackets
and one arrow. The heaps are 3, 4, 0, and 5, so that the game is equiv-
alent to playing Nim with rows of three, four, and five counters.
Rational play is exactly as in Nim: Move to reduce the Nim sum of
the heaps to 0, a game with a second-player win. The one trivial dif-
ference is that here a heap can increase in size. If, however, you have
the win and your opponent makes such a move, you immediately re-
store the heap to its previous size by moving the coin that is just to the
right of the heap.
If in the illustrated position it is your move, you are sure to win if you
make the move indicated by the curved arrow. If your opponent re-
sponds by moving counter A two cells to the left, the move raises the
empty heap to 2. Your response is to move B two cells to the left, thus
returning the heap to 0.
4. Silver Dollar Game with the Dollar. This is the same as the pre-
ceding game except that one of the coins (anyone) is a silver dollar and
the cell farthest to the left is a money bag (see Figure 28.3(B)). A coin
farthest to the left can move into the bag. When the dollar is bagged, the
game ends and the next player wins by taking the bag.
This game too is Nim disguised. Count the bag as being empty if the
coin at its right is a penny and full if it is the dollar, and play Nim as
before. If you have the win, your opponent will be forced to drop the
dollar in the bag. If the winner is deemed to be the player who bags the
dollar, count the bag as being full if the coin at its right is the coin just
to the left of the dollar and count it as being empty otherwise. The po-
sition shown corresponds to Nim with heaps of 4,3,0, and 2. The first
player wins in both versions only if he makes the move indicated by the
curved arrow.
5. Rims. The initial position of this pleasant way of playing a variant
of Nim consists of two or more groups of spots. The move is to draw a
simple closed loop through any positive number of spots in one group.
The loop must not cross itself or cross or touch any other loop. A game
is shown in Figure 28.4.
Surreal Numbers 377
3 5 7
28.4. A posiiticm in Rims
378
7. Cutcake. This is a partizan game invented by Conway. It is played
with a set of rectangular cakes, each scored into unit squares like a
waffle. Left's move is to break a piece of cake into two parts along any
horizontal lattice line, Right's is to break a piece along any vertical
line. The game has a surprisingly simple theory.
Figure 28.5 shows a 4 x 7 cake. In Conway's notation its value is 0,
which means it is a second-player win regardless of who goes first. It
looks as if the vertical breaker, who has twice as many opening moves
as his opponent, would have the advantage, but he does not if he goes
first. Assume that the vertical breaker goes first and breaks along the line
indicated by the arrow. What is the second player's winning response?
figure 28.5. A format for Cutcake, with the first move being a vertical break at the arrow
1 have given only a few examples of Conway's exotic nomenclature.
Games can be short, small, all small, tame, restive, restless, divine, ex-
traverted, and introverted. There are ups, downs, remote stars, semi-
stars, and superstars. There are atomic weights and sets with such
names as On, No, Ug, and Oz. Conway has a temperature theory, with
thermographs on which hot positions are cooled by pouring cold water
on them. He has a Mach principle for the small world: the atomic
weight of a short all-small game is at least 1 if and only if the game ex-
ceeds the remote stars!
Conway's theorem 99 conveys some notion of the book's whimsical
flavor. It tells us (I paraphrase to remove a slight error that Conway
discovered too late to correct) that any short all-small game of atomic
weight zero is dominated by some superstar. Only a feeling of incom-
pleteness, Conway adds, prompts him to give a final theorem. Theorem
100 is: "This is the last theorem in this book."
Surreal Numbers 379
Addendum
Since this chapter first ran as a Scientific American column in
1976, Conway has written other books and papers, and articles about
him have appeared, some of which are listed in the bibliography. Es-
pecially notable is the two-volume Winning Ways, on which he col-
laborated with Elwyn Berlekamp and Richard Guy. It has already
become a classic of recreational mathematics and has made important
technical contributions to game theory and combinatorics. For Con-
way's work on Penrose tiling, see Chapters 1 and 2 of my Perrose Tiles
to Trapdoor Ciphers. For his work on sporadic groups and knot theory,
see my Last Recreations, Chapters 9 and 5 respectively.
Conway spoke on his game theories at a conference on recreational
mathematics at Miami University in September 1976. His lecture,
reprinted as "A Gamut of Game Theories" in Mathematics Magazine
(see bibliography), concludes:
The theories can be applied to hundreds and thousands of games-really
lovely little things; you can invent more and more and more of them. It's
especially delightful when you find a game that somebody's already
considered and possibly not made much headway with, and you find
you can just turn on one of these automatic theories and work out the
value of something and say, "Ah! Right is 47/64ths of a move ahead, and
so she wins."
Martin Kruskal, now a mathematician at Rutgers University, became
fascinated by Conway's surreal numbers. For many years he has been
working on their clarification and elaboration, and their potential ap-
plications to other fields of mathematics. For an introduction to this ex-
citing work, see the last two entries, by Polly Shulman and Robert
Matthews, in the chapter's bibliography under the subhead of "Surreal
Numbers." Kruskal is working on what he calls "surreal calculus," in
which Conway's numbers take care of the infinitesimals so essential in
the calculus of Newton and Leibniz.
Answers
The problem was to find the winning response to a first-player
move in Cutcake. The cake is a 4 x 7 rectangle. If the first player breaks
380 INFINITY
the cake vertically into a 4 x 4 square and a 4 x 3 rectangle, the unique
winning reply is to break the 4 x 3 piece into two 2 x 3 rectangles.
Bibliography
On Qames
E. Berlekamp, J. Conway, and R. Guy, Winning Ways (two volumes), Academic
Press, 1982.
J. H. Conway, "All Numbers Great and Small," University of Calgary Mathematical
Research Paper No. 149, 1972.
J. H. Conway, "All Games Bright and Beautiful," University of Calgary Mathemat-
ical Research Paper No. 295, 1975. Reprinted in The American Mathematical
Monthly, Vol. 84, 1977, pp. 417-34.
J. H. Conway, On Numbers and Games, Academic Press, 1976.
J. H. Conway, "A Gamut of Game Theories," Mathematics Magazine, Vol. 51, 1978,
pp.5-12.
J. H. Conway and R. Guy, The Book of Numbers, Springer-Verlag, 1996.
S. W. Golomb, "A Mathematical Investigation of Games of 'Take Away,' " Journal
of Combinatorial Theory, Vol. 1, 1966, pp. 443-58.
R. K. Guy and C. A. B. Smith, "The G-Values of Various Games," Proceedings of the
Cambridge Philosophical Society, Vol. 52, Part 3,1956, pp. 514-26.
On Surreal Numbers
H. Gonshor, An Introduction to the Theory of Surreal Numbers, London Mathe-
matical Society Lecture Note Series 110, 1986.
D. E. Knuth, Surreal Numbers, Addison-Wesley, 1974.
R. Matthews, "The Man Who Played God with Infinity," New Scientist, September
2,1995,pp.36-40.
P. Shulman, "Infinity Plus One and Other Surreal Numbers," Discover, December
1995, pp. 97-105.
About Conway
D. Albers, "John Horton Conway-Talking a Good Game," Mathematical Horizons,
Spring 1994, pp. 6-9.
M. Alpert, "Profile: Not Just Fun and Games," Scientific American, April 1 999, pp.
40-42.
M. Browne, "Intellectual Duel," The New York Times, August 30,1988.
R. K. Guy, "John Horton Conway: Mathematical Magus," Two- Year College Mathe-
matics Journal, Vol. 13, 1982, pp. 290-99.
G. Kolata, "At Home in the Elusive World of Mathematics," The New York Times,
October 12, 1993.
Surreal Numbers 381
C. Moseley, "The Things that Glitter," Princeton Alumni Weekly, December 23,
1992, pp. 10-13.
C. Seife, "Mathematician," The Sciences, May/June 1994, pp. 12-15.
G. Taubes, "John Horton Conway: A Mathematical Madness in Cambridge," Dis-
cover, August 1984, pp. 40-50.
M. Thompson and M. Niemann, "Mathematical Mania," Science Today, Spring
1987, pp. 8-9.
VIII
Combinatorics
Chapter 29 Hexaflexagons
Flexagons are paper polygons, folded from straight or crooked
strips of paper, which have the fascinating property of changing their
faces when they are "flexed." Had it not been for the trivial circum-
stance that British and American notebook paper are not the same size,
flexagons might still be undiscovered, and a number of top-flight math-
ematicians would have been denied the pleasure of analyzing their cu-
rious structures.
It all began in the fall of 1939. Arthur H. Stone, then a 23-year-old
graduate student from England, in residence at Princeton University on
a mathematics fellowship, had just trimmed an inch from his American
notebook sheets to make them fit his English binder. For amusement he
began to fold the trimmed-off strips of paper in various ways, and one
of the figures he made turned out to be particularly intriguing. He had
folded the strip diagonally at three places and joined the ends so that
it made a hexagon (see Figure 29.1). When he pinched two adjacent tri-
angles together and pushed the opposite corner of the hexagon toward
the center, the hexagon would open out again, like a budding flower,
and show a completely new face. If, for instance, the top and bottom
faces of the original hexagon were painted different colors, the new
face would come up blank and one of the colored faces would disap-
pear!
This structure, the first flexagon to be discovered, has three faces.
Stone did some thinking about it overnight, and on the following day
confirmed his belief (arrived at by pure cerebration) that a more com-
plicated hexagonal model could be folded with six faces instead of
only three. At this point Stone found the structure so interesting that he
showed his paper models to friends in the graduate schooL Soon
"flexagons" were appearing in profusion at the lunch and dinner tables.
385
a
/
!\ !\ !\ :\ /
" " 1\ 1\ I
I \ I \ I \ 1\ J
'\ '\ I \ I \ I
/' '\ /' '\ ,/ '\ /' '\ ,/
/ \ / \/ \/ \/
Figure 29.1. Trihexaflexagon is con-
structed by cutting a strip of paper so
that it may be marked off in 10 equi-
lateral triangles (A).
The strip is folded backward along
the line ab and turned over (B).
It is then folded backward again along
the line cd and the next to the last
triangle is placed on top of the first
(C).
The last triangle is now folded back-
ward and glued to the other side of
the first (D).
The figure may be flexed but it is not
meant to be cut out. Fairly stiff
paper at least an inch and a half
wide is recommended.
,0
~ - _ ~ : - - __ -71
/
b
A
A "Flexagon Committee" was organized to probe further into the mys-
teries of flexigation. The other members besides Stone were Bryant
Tuckerman, a graduate student of mathematics; Richard P. Feynman, a
graduate student in physics; and John W. Tukey, a young mathematics
instructor.
The models were named hexaflexagons-"hexa" for their hexagonal
form and "flexagon" for their ability to flex. Stone's first model is a tri-
hexaflexagon ("tri" for the three different faces that can be brought into
view); his elegant second structure is a hexahexaflexagon (for its six
faces).
To make a hexahexaflexagon you start with a strip of paper which is
divided into 19 equilateral triangles (see Figure 29.2). You number the
386 COMBINATORICS
A
b
D
is
in 19 (A). on one
other, 4, 5, and 6. A similar pattern
may also be used. hexagon is then folded as shown.
show six different faces.
are 5, 6,
the strip so that the same underside numbers face each other-4 on 4,
so on.
folded
hexagon (D); finally the blank triangle is turned under pasted to the
other side of the strip. All
a ............. "'''",''"'_
you have made the folds on one visible face
of the hexagon are numbered i j and on the other face all are num-
line between them, and push in the
387
Figure 29.4. Diagram of a Tuckerman traverse on a hexahexaflexagon
straight edges) it was possible to produce a tetrahexaflexagon (four
faces) or a pentahexaflexagon. There are three different hexahexa-
flexagons-one folded from a straight strip, one from a chain bent into
a hexagon, and one from a form that somewhat resembles a three-leaf
clover. The decahexaflexagon (10 faces) has 82 different variations, all
folded from weirdly bent strips. Flexagons can be formed with any de-
sired number of faces, but beyond 10 the number of different species for
each increases at an alarming rate. All even-numbered flexagons, by the
way, are made of strips with two distinct sides, but those with an odd
number of faces have only a single side, like a Mobius surface.
A complete mathematical theory of flexigation was worked out in
1940 by Tukey and Feynman. It shows, among other things, exactly
how to construct a flexagon of any desired size or species. The theory
has never been published, though portions of it have since been redis-
covered by other mathematicians. Among the flexigators is Tucker-
man's father, the distinguished physicist Louis B. Tuckernmn, who was
formerly with the National Bureau of Standards. Tuckerman senior de-
vised a simple but efficient tree diagram for the theory.
Pearl Harbor called a halt to the committee's flexigation program,
and war work soon scattered the four charter members to the winds.
Stone became a lecturer in mathematics at the University of Manches-
ter in England, and is now at the University of Rochester, NY. Feynman
became a famous theoretical physicist at the California Institute of
Hexaflexagons 389
Technology. Tukey, a professor of mathematics at Princeton, has made
brilliant contributions to topology and to statistical theory which have
brought him worldwide recognition. Tuckerman became a mathemati-
cian at IBM's research center in Yorktown Heights, NY.
One of these days the committee hopes to get together on a paper or
two which will be the definitive exposition of flexagon theory. Until
then the rest of us are free to flex our flexagons and see how much of
the theory we can discover for ourselves.
Addendum
In constructing flexagons from paper strips it is a good plan to
crease all the fold lines back and forth before folding the model. As a
result, the flexagon flexes much more efficiently. Some readers made
more durable models by cutting triangles from poster board or metal
and joining them with small pieces of tape, or by gluing them to one
long piece of tape, leaving spaces between them for flexing. Louis Tuck-
erman keeps on hand a steel strip of such size that by wrapping paper
tape of a certain width around it he can quickly produce a folded strip
of the type shown in Figure 29.2(B J. This saves considerable time in
making flexagons from straight chains of triangles.
Readers passed on to me a large variety of ways in which flexagon
faces could be decorated to make interesting puzzles or display strik-
ing visual effects. Each face of the hexahexa, for example, appears in at
least two different forms, owing to a rotation of the component trian-
gles relative to each other. Thus if we divide each face as shown in Fig-
ure 29.5, using different colors for the A, B, and C sections, the same
face may appear with the A sections in the center as shown, or with the
B or C sections in the center. Figure 29.6 shows how a geometrical pat-
tern may be drawn on one face so as to appear in three different con-
figurati ons.
Of the 18 possible faces that can result from a rotation of the trian-
gles, three are impossible to achieve with a hexahexa of the type made
from a straight strip. This suggested to one correspondent the plan of
pasting parts of three different pictures on each face so that by flexing
the model properly, each picture could presumably be brought together
at the center while the other two would be fragmented around the rim.
On the three inner hexagons that cannot be brought together, he pasted
the parts of three pictures of comely, undraped young ladies to make
390 COMBINATORICS
advertise their service award banquet in 1955. A handsome hexahexa,
designed to display a variety of colored snowflake patterns, was used
by Scientific American for their 1956 Christmas card.
For readers who would like to construct and analyze flexagons other
than the two described in the chapter, here is a quick run-down on
some low-order varieties.
1. The unahexa. A strip of three triangles can be folded flat and the op-
posite ends can be joined to make a Mobius strip with a triangular
edge. Since it has one side only, made up of six triangles, one might
call it a unahexaflexagon, though of course it isn't six-sided and it
doesn't flex.
2. The duahexa. Simply a hexagon cut from a sheet of paper. It has two
faces but doesn't flex.
3. The trihexa. This has only the one form described in this chapter.
4. The tetrahexa. This likewise has only one form. It is folded from the
crooked strip shown in Figure 29.7(A).
5. The pentahexa. One form only. Folded from the strip in Figure 29.7
(B).
6. The hexahexa. There are three varieties, each with unique properties.
One of them is described in this chapter. The other two are folded
from the strips pictured in Figure 29.7(C).
7. The heptahexa. This can be folded from the three strips shown in Fig-
ure 29.7(D). The first strip can be folded in two different ways, mak-
ing four varieties in all. The third form, folded from the overlapping
figure-eight strip, is the first of what Louis Thckerman calls the
"street flexagons." Its faces can be numbered so that a Tuckerman tra-
verse will bring uppermost the seven faces in serial order, like pass-
ing the street numbers on a row of houses.
The octahexa has 12 distinct varieties, the enneahexa has 27, and the
decahexa, 82. The exact number of varieties of each order can be fig-
ured in more than one way depending on how you define a distinct va-
riety. For example, all flexagons have an asymmetric structure which
can be right-handed or left-handed, but mirror-image forms should
hardly be classified as different varieties. For details on the number of
nonequivalent hexaflexagons of each order, consult the paper by Oak-
ley and Wisner listed in the bibliography.
Straight chains of triangles produce only hexaflexagons with orders
that are multiples of three. One variety of a 12-faced hexa is particularly
392 COMBINATORICS
D
29.1. Crooked strips fOT folding hexaflexagons.
pasting.
easy to fold. Start with a straight chain twice as long as the one used for
the hexahexa. it form shown in Figure 29.2(B). The strip
now same length as the one for
rolled exactly as if you were making a hexahexa. a dodec-
ahexaflexagon.
393
In experimenting with higher-order flexagons, a handy rule to bear in
mind is that the sum of the number of leaves (thicknesses of paper) in
two adjacent triangular sections always equals the number of faces. It
is interesting to note also that if each face of a flexagon is given a num-
ber or symbol, and the symbol is marked on each triangular component,
the order of symbols on the unfolded strip always exhibits a threefold
symmetry. For example. the strip for the hexahexa in Figure 29.2 bears
the following top and bottom pattern of digits:
123123 123123 123123
445566 445566 445566
A triple division similar to this is characteristic of all hexahexa-
flexagons. although on models of odd order one of the three divisions
is always inverted.
Of the hundreds of letters received about flexagons, the following
two were the most amusing. They appeared in the March and May is-
sues of Scientific American, 1957.
SIRS:
I was quite taken with the article entitled "Flexagons" in your De-
cember issue. It took us only six or seven hours to paste the hexahexa-
flexagon together in the proper configuration. Since then it has been a
source of continuing wonder.
But we have a problem. This morning one of our fellows was sitting
flexing the hexahexaflexagon idly when the tip of his necktie became
caught in one of the folds. With each successive flex. more of his tie van-
ished into the flexagon. With the sixth flexing he disappeared entirely.
We have been flexing the thing madly, and can find no trace of him,
but we have located a sixteenth configuration of the hexahexaflexagon.
Here is our question: Does his widow draw workmen's compensation
for the duration of his absence, or can we have him declared legally
dead immediately? We await your advice.
Allen B. Du Mont Laboratories, Inc.
Clifton, N.J.
SIRS:
NEIL UPTEGROVE
The letter in the March issue of your magazine complaining of the
disappearance of a fellow from the Allen B. Du Mont Laboratories
"down" a hexahexaflexagon, has solved a mystery for us.
394 COMBINATORICS
One day, while idly flexing our latest hexahexaflexagon, we were con-
founded to find that it was producing a strip of multicolored material.
Further flexing of the hexahexaflexagon finally disgorged a gum-
chewing stranger.
Unfortunately he was in a weak state and, owing to an apparent loss
of memory, unable to give any account of how he came to be with us. His
health has now been restored on our national diet of porridge, haggis and
whisky, and he has become quite a pet around the department, answer-
ing to the name of Eccles.
Our problem is, should we now return him and, if so, by what
method? Unfortunately Eccles now cringes at the very sight of a hexa-
hexaflexagon and absolutely refuses to "flex."
The Royal College of Science and Technology
Glasgow, Scotland
ROBERT M. HILL
Although this chapter first ran as an article in Scientific American
(December 1956), it became the first of my monthly columns to be
headed "Mathematical Games." I had learned about hexaflexagons from
Royal Vale Heath, a New York City broker who was also an amateur ma-
gician and mathematician. His little book Mathemagic (1933) was one
of the earliest books about magic tricks that operate with mathematical
principles.
I cannot now recall how Heath discovered flexagons, but he was able
to steer me to the four Princeton University inventors. To gather mater-
ial for my article, I made a trip to Princeton where I interviewed Tukey
and Tuckerman. (Stone was then in England; Feynman was at Cal Tech.)
After my piece appeared, people all over Manhattan were flexing
flexagons. Gerry Piel, publisher of Scientific American, called me into
his office to ask if there was enough similar material to make a regular
column. I assured him there was and immediately made the rounds of
Manhattan's used bookstores to buy as many books on recreational
math (there were not many) as I could find. Once the column got un-
derway, I began to receive fresh ideas from mathematicians and writing
the column became an easy and enjoyable task that lasted more than a
quarter-century.
Although flexagons, as far as anyone knows, are completely free of
applications (except of course as playthings), mathematicians continue
to be intrigued by their whimsical properties. In The Second "Scientific
American" Book of Mathematical Puzzles and Diversions I introduced
Hexaflexagons 395
their square-shaped cousins, the tetraflexagons. Both types have been
discussed in many subsequent articles, but no one has yet written a de-
finitive work on flexagon theory. Frank Bernhart, who taught mathe-
matics at the Rochester Institute of Technology, knows more about
flexagons than anyone. Some day, let us hope, he will find a publisher
for a monograph about them.
Bibliography
D. A. Engel, "Hybrid Flexagons," Journal of Recreational Mathematics, Vol. 2, Jan-
uary 1969, pp. 35-41.
D. A. Engel, "Hexaflexagon + HHFG = Slipagon," Journal of Recreational Mathe-
matics, No.3, 1995, pp. 161-66.
M. Gilpin, "Symmetries of the Trihexaflexagon," Mathematics Magazine, Vol. 49,
September 1976, pp. 189-92.
P. Hilton, J. Pedersen, and H. Walker, "The Faces of the Tri-Hexaflexagon," Math-
ematics Magazine, Vol. 70, October 1997, pp. 243-51.
P. Jackson, Flexagons, British Origami SOciety, 1978.
D. Johnson, Mathemagic with Flexagons, Activity Resources, 1974.
M. Jones, The Mysterious Flexagons, Crown, 1966. A book for children printed on
heavy paper so the child can cut strips for folding.
M. Joseph, "Hexahexaflexagrams," Mathematics Teacher, Vol. 44 April 1951, pp.
247-48. Tells how to make a straight chain hexahexaflexagon.
P. Liebeck, "The Construction of Flexagons," Mathematical Gazette, VoL 48, De-
cember 1964, pp. 397-402.
J. Madachy, Mathematics on Vacation, Scribner's, 1966, pp. 62-84.
F. G. Maunsell, "The Flexagon and the Hexahexaflexagram," Mathematical Gazette,
Vol. 38, 1954, pp. 213-14. Describes the trihexa and hexahexa. See also Vol. 41,
1957, pp. 55-56, for a note on this article by Joan Crampin. It describes hexa-
flexagons folded from straight strips (orders 3, 6, 9, 12,. .. ).
T. B. McLean, "V-Flexing the Hexaflexagon," American Mathematical Monthly,
VoL 86, June/July 1979, pp. 457-66.
S. Morris, "Hexplay," Omni, October 1984, pp. 89ff.
C. O. Oakley and R. J. Wisner, "Flexagons," American Mathematical Monthly, Vol.
64, March 1957, pp. 143-54.
T. O'Reilly, "Classifying and Counting Hexaflexagrams," Journal of Recreational
Mathematics, Vol. 8, No.3, 1975-76, pp. 182-87.
A. Panov, "To Flexland With Mr. Flexman," Quantum, March/April 1992, pp.
64-65.
A. Panov and A. Keltnin, "Flexland Revisited," Quantum, July/August, 1993, pp.
64-65.
W. R. Ransom, "A Six-Sided Hexagon," School Science and Mathematics, Vol. 52,
1952, p. 94. Tells how to make a trihexaflexagon.
396 COMBINA TOR ICS
W. Ransom, "Protean Shapes with Flexagons," Recreational Mathematics Maga-
zine, No. 13, February 1963, pp. 35-37.
S. Scott, "How to Construct Hexaflexagons," Recreational Mathematics Magazine,
No. 12, December 1962, pp. 43-49.
V. Silva, "Flexagons," Eureka, No. 52, March 1993, pp. 30-44.
R. F. Wheeler, "The Flexagon Family," Mathematical Gazette, Vol. 42 February
1958, pp. 1-6. A fairly complete analysis of straight and crooked strip forms.
Hexaflexagons 397
Chapter 30 The Soma Cube
", .. no time, no leisure . .. not a moment to
sit down and think-or if ever by some un-
lucky chance such a crevice of time should
yawn in the solid substance of their dis-
tractions, there is always soma, delicious
soma ... " -Aldous Huxley, Brave New World
The Chinese puzzle game called tangrams, believed to be thou-
sands of years old, employs a square of thin material that is dissected
into seven pieces. The game is to rearrange those pieces to form other
figures. From time to time efforts have been made to devise a suitable
analog in three dimensions. None, in my opinion, has been as suc-
cessful as the Soma cube, invented by Piet Hein, the Danish writer
whose mathematical games, Hex and Tac Tix, are discussed in the first
Scientific American Book of Mathematical Puzzles. (In Denmark, Piet
Hein was best known for his books of epigrammatic poems written
under the pseudonym Kumbel.)
Piet Hein conceived of the Soma cube during a lecture on quantum
physics by Werner Heisenberg. While the noted German physicist was
speaking of a space sliced into cubes, Piet Hein's supple imagination
caught a fleeting glimpse of the following curious geometrical theo-
rem. If you take all the irregular shapes that can be formed by combin-
ing no more than four cubes, all the same size and joined at their faces,
these shapes can be put together to form a larger cube.
Let us make this clearer. The simplest irregular shape-"irregular" in
the sense that it has a concavity or corner nook in it somewhere-is
produced by joining three cubes as shown in Figure 30.1, piece 1. It is
the only such shape possible with three cubes. (Of course no irregular
shape is possible with one or two cubes.) Turning to four cubes, we find
that there are six different ways to form irregular shapes by joining the
cubes face to face. These are pieces 2 to 7 in the illustration. To iden-
tify the seven pieces Piet Hein labels them with numerals. No two
shapes are alike, although 5 and 6 are mirror images of each other. Piet
Hein points out that two cubes can be joined only along a single coor-
dinate, three cubes can add a second coordinate perpendicular to the
398
:2 3
4
6 7
30.1. seven Soma
30.2. A two
WEll
PYRAMID
WAll
DOG
SKYSCRAPER
30.3. One 12 cannot up
ferent solutions (not counting rotations and reflections) been
ulated by of of Malaya, in Singapore, but
the determined. A
to adopt on this as well as
more irregular shapes (5, and 7) in place first, because the other
a Piece 1 in
solving the cube, try your hand at the more seven-
piece structures a time-consuming trial
and effOf con-
CASTlE
CHAIR
STEAMER
401
An 1m J;)OSSlble A means of laDeung the form
402
of three cubes. This gives us a total of eight white cubes and 19 black,
quite an astounding disparity.
The next step is to examine each of the seven components, testing it
in all possible orientations to ascertain the maximum number of black
cubes it can possess if placed within the checkerboard structure. The
chart in Figure 30.5 displays this maximum number for each piece. As
you see, the total is 18 black to nine white, just one short of the 19-8
split demanded. If we shift the top black block to the top of one of the
columns of white blocks, then the black-white ratio changes to the re-
quired 18-9, and the structure becomes possible to build.
SOMA MAXIMUM MINIMUM
Figure 30.5. Table for the
PIECE BLACK CUBES WHITE CUBES
impossibility proof
I. 2
1. 3
:1 3
2 2
~ ) . 3
I>. 3
7 2 :2
18 9
I must confess that one of the structures in Figure 30.3 is impossible
to make. It should take the average reader many days, however, to dis-
cover which one it is. Methods for building the other figures will not
be given in the answer section (it is only a matter of time until you suc-
ceed with anyone of them), but I shall identify the figure that cannot
be made.
The number of pleasing structures that can be built with the seven
Soma pieces seems to be as unlimited as the number of plane figures
that can be made with the seven tangram shapes. It is interesting to note
that if piece 1 is put aside, the remaining six pieces will form a shape
exactly like 1 but twice as high.
Addendum
When I wrote the column about Soma, I supposed that few
readers would go to the trouble of actually making a set. I was wrong.
Thousands of readers sent sketches of new Soma figures and many
complained that their leisure time had been obliterated since they were
bitten by the Soma bug. Teachers made Soma sets for their classes. Psy-
The Soma Cube 403
Cryslol
lO.6.
ures were
404
Wafl
Modern
Tower
405
sible to form siInilar models of each of the other seven pieces. Katsanis
also found that the eight pieces can be divided into two sets of four,
each set making a 2 x 2 x 4 rectangular solid. These two solids can
then be put together in different ways to make double-sized models of
six of the eight pieces.
In a previous column I described the 12 pentominoes: flat shapes
formed by connecting unit squares in all possible ways. Mrs. R. M.
Robinson, wife of a mathematics professor at the University of Califor-
nia in Berkeley, discovered that if the pentominoes are given a third di-
mension, one unit thick, the twelve pieces will form a 3 x 4 x 5
rectangular solid. This was independently discovered by several others,
including Charles W. Stephenson, M.D., of South Hero, VT. Dr.
Stephenson also found ways of putting together the three-dimensional
pentominoes to make rectangular solids of 2 x 5 x 6 and 2 x 3 x 10.
The next step in complexity is to the 29 pieces formed by putting five
cubes together in all possible ways. Katsanis, in the same letter men-
tioned above, suggested this and called the pieces "pentacubes." Six
pairs of pentacubes are mirror-image forms. If we use only one of each
pair, the number of penta cubes drops to 23. Both 29 and 23 are primes,
therefore no rectangular solids are possible with either set. Katsanis
proposed a triplication problem: choose one of the 29 pieces, then use
27 of the remaining 28 to form a model of the selected piece, three
times as high.
A handsome set of pentacubes was shipped to me in 1960 by David
Klamer of Napa, CA. I dumped them out of the wooden box in which
they were packed and have not yet succeeded in putting them back in.
Klarner has spent considerable time developing unusual pentacube fig-
ures, and I have spent considerable time trying to build some of them.
He writes that there are 166 hexacubes (pieces formed by joining six-
unit cubes), of which he was kind enough not to send a set.
The seven Soma pieces are a subset of what are now called poly-
cubes-polyhedrons formed by joining unit cubes by their faces. Since
I introduced the Soma cube in a 1958 column it has been made and
sold by numerous toy companies around the world. Here Parker Broth-
ers sold the cube along with an instruction booklet written by Piet
Hein. The firm also distributed three issues of Soma Addict, a newslet-
ter edited by game agent Thomas Atwater.
Several computer programs verified that there are 240 ways, not
counting rotations and reflections, to make the Soma cube. John Con-
406 COMBINATORICS
way produced what he called the Somap. You'll find a picture of it in
Winning Ways, Vol. 2, by Elwyn Berlekamp, Conway, and Richard Guy,
pages 802-3. This amazing graph shows how you can start with any of
239 solutions to the cube, then transform it to any other solution by
moving no more than two or three pieces. There is one solution unob-
tainable in this way.
J. Edward Hanrahan wrote to me about a Soma task he invented. The
challenge is to form a 4 x 4 x 2 structure so that its five "holes" on the
top layer have the shape of each of the 12 pentominoes. The problem
is solvable for each pentomino except the straight one which obviously
can't fit into the structure.
Puzzle collector Jerry Slocum owns dozens of different dissections of
a 3
3
cube into seven or fewer pieces, most of them marketed after my
column on the Soma cube appeared. In the chapter on polycubes in
Knotted Doughnuts and Other Mathematical Entertainments, I describe
the Diabolical cube, sold in Victorian England. The earliest dissection
of a 3
3
cube known, its six polycube pieces form the cube in 13 ways.
I also describe the Miksinski cube, another six-piece dissection-one
that has only two solutions.
Many later 3
3
dissections limit the number of solutions by coloring
or decorating the unit cube faces in various ways. For example, the
unit cubes are either black or white and the task is to make a cube that
is checkered throughout or just on its six faces. Other dissections put
one through six spots on the unit cubes so that the assembled cube
will resemble a die. The unit cubes can be given different colors. The
task is to form a cube with a specified pattern of colors on each face.
Another marketed cube had digits 1 through 9 on the unit cubes, and
the problem was to make a cube with each face a magic square. A puz-
zle company in Madison, WI, advertised a game using a set of nine
polycubes. Players selected any seven at random by rolling a pair of
dice, then tried to make a 3
3
cube with the pieces. In 1969 six different
dissections of the cube, each made with five, six, or seven polycubes,
were sold under the name Impuzzibles.
It's easy to cut a 3
3
cube into six or fewer polycubes that will make a
cube in just one way, but with seven unmarked pieces, as I said earlier,
it is not so easy. 1Wo readers sent such dissections which they believed
solved this problem, but I was unable to verify their claims. Rhoma, a
slant version of Soma produced by a shear distortion that changes each
unit cube as well as the large cube to a rhomboid shape, went on sale
The Soma Cube 407
here. A more radically squashed Soma was sold in Japan. With such
distortions the solution becomes unique.
John Brewer, of Lawrence, KS, in a little magazine he used to publish
called Hedge Apple and Devil's Claw (Autumn 1995 issue), introduced
the useful device of giving each Soma piece a different color. Solutions
could then be represented by showing three sides of the cube with its
unit cubes properly colored. He sent me a complete Somap using such
pictures for each solution. His article also tells of his failed effort to lo-
cate Marguerite Wilson, the first to publish a complete set of solutions
to the Soma cube.
Alan Guth, the M.l. T. physicist famous for his conjecture that a mo-
ment after the big bang the universe rapidly inflated, was quoted as fol-
lows in Discover (December 1997):
My all-time favorite puzzle was a game called Soma, which I think was
first marketed when I was in college. A set consisted of seven odd-
shaped pieces that could be put together to make a cube, or a large vari-
ety of other three-dimensional shapes. Playing with two sets at once
was even more fun. The instruction book claims that there are a certain
number of different ways of making the cube, and I remember writing a
computer program to verify this number. They were right, but I recall
that they counted each of the 24 different orientations of the cube as a
different "way" of making it.
Answer
The only structure in Figure 30.3 that is impossible to construct
with the seven Soma pieces is the skyscraper.
Bibliography
J. Brunrell et aI, "The Computerized Soma Cube," Symmetry Unifying Human Un-
derstanding, 1. Hargittai (ed.), Pergamon, 1987.
G. S. Carson, "Soma Cubes," Mathematics Teacher, November 1973, pp. 583-92.
S. Farm, Soma Cubes, Pentacubes Puzzles Ltd, Auckland, Australia, 1979. This 15-
page booklet depicts 111 Soma structures and their solutions.
D. A. Macdonald and Y. Giirsel, "Solving Soma Cube and Polyomino Problems
Using a Microcomputer," Byte, November 1989, pp. 26-41.
D. Spector, "Soma: A Unique Object for Mathematical Study," in Mathematics
Teacher, May 1982, pp. 404-7.
M. Wilson, Soma Puzzle Solutions, Creative Publications, 1973.
408 COMBINATORlCS
Chapter 31 Theqame
of Life
Most of the work of John Horton Conway, a distinguished
Cambridge mathematician now at Princeton University, has been in
pure mathematics. For instance, in 1967 he discovered a new group-
some call it "Conway's constellation"-that includes all but two of the
then-known sporadic groups. (They are called "sporadic" because they
fail to fit any classification scheme.) It is a breakthrough that has had
exciting repercussions in both group theory and number theory. It ties
in closely with an earlier discovery by John Leech of an extremely
dense packing of unit spheres in a space of 24 dimensions where each
sphere touches 196,560 others. As Conway has remarked, "There is a
lot of room up there."
In addition to such serious work Conway also enjoys recreational
mathematics. Although he is highly productive in this field, he sel-
dom publishes his discoveries. In this chapter we consider Conway's
most famous brainchild, a fantastic solitaire pastime he calls "Life."
Because of its analogies with the rise, fall, and alterations of a soci-
ety of living organisms, it belongs to a growing class of what are
called "simulation games." To play Life without a computer you
need a fairly large checkerboard and a plentiful supply of flat coun-
ters of two colors. (Small checkers or poker chips do nicely.) An Ori-
ental "Go" board can be used if you can find flat counters small
enough to fit within its cells. (Go stones are awkward to use because
they are not flat.) It is possible to work with pencil and graph paper
but it is much easier, particularly for beginners, to use counters and
a board.
The basic idea is to start with a simple configuration of counters (or-
ganisms), one to a cell, then observe how it changes as you apply Con-
way's "genetic laws" for births, deaths, and survivals. Conway chose
409
his rules carefully, after a long period of experimentation, to meet three
desiderata:
1. There should be no initial pattern for which there is a simple proof
that the population can grow without limit.
2. There should be initial patterns that apparently do grow without
limit.
3. There should be simple initial patterns that grow and change for a
considerable period of time before coming to an end in three possi-
ble ways: Fading away completely (from overcrowding or from be-
coming too sparse), settling into a stable configuration that remains
unchanged thereafter, or entering an oscillating phase in which they
repeat an endless cycle of two or more periods.
In brief, the rules should be such as to make the behavior of the pop-
ulation both interesting and unpredictable.
Conway's genetic laws are delightfully simple. First note that each
cell of the checkerboard (assumed to be an infinite plane) has eight
neighboring cells, four adjacent orthogonally, four adjacent diagonally.
The rules are:
1. Survivals. Every counter with two or three neighboring counters sur-
vives for the next generation.
2. Deaths. Each counter with four or more neighbors dies (is removed)
from overpopulation. Every counter with one neighbor or none dies
from isolation.
3. Births. Each empty cell adjacent to exactly three neighbors-no more,
no fewer-is a birth celL A counter is placed on it at the next move.
It is important to understand that all births and deaths occur simul-
taneously. Together they constitute a single generation or, as we shall
usually call it, a "tick" in the complete "life history" of the initial con-
figuration. Conway recommends the following procedure for making
the moves:
1. Start with a pattern consisting of black counters.
2. Locate all counters that will die. Identify them by putting a black
counter on top of each.
3. Locate all vacant cells where births will occur. Put a white counter
on each birth cell.
4. After the pattern has been checked and double-checked to make sure
410 COMBINATORICS
no mistakes have been made, remove all the dead counters (piles of
two) and replace all newborn white organisms with black counters.
You will now have the first generation in the life history of your ini-
tial pattern. The same procedure is repeated to produce subsequent
generations. It should be clear why counters of two colors are needed.
Because births and deaths occur simultaneously, newborn counters
play no role in causing other deaths or births. It is essential, therefore,
to be able to distinguish them from live counters of the previous gen-
eration while you check the pattern to be sure no errors have been
made. Mistakes are very easy to make, particularly when first playing
the game. After playing it for a while you will gradually make fewer
mistakes, but even experienced players must exercise great care in
checking every new generation before removing the dead counters and
replacing newborn white counters with black.
You will find the population constantly undergoing unusual, some-
times beautiful, and always unexpected change. In a few cases the so-
ciety eventually dies out (all counters vanishing), although this may not
happen until after a great many generations. Most starting patterns ei-
ther reach stable figures-Conway calls them "stilllifes"-that cannot
change or patterns that oscillate forever. Patterns with no initial sym-
metry tend to become symmetrical. Once this happens the symmetry
cannot be lost, although it may increase in richness.
Conway originally conjectured that no pattern can grow without
limit. Put another way, any configuration with a finite number of coun-
ters cannot grow beyond a finite upper limit to the number of counters
on the field. Conway offered a prize of $50 to the first person who
could prove or disprove the conjecture before the end of 1970. One
way to disprove it would be to discover patterns that keep adding coun-
ters to the field: A "gun" (a configuration that repeatedly shoots out
moving objects such as the "glider," to be explained below) or a "puffer
train" (a configuration that moves but leaves behind a trail of "smoke").
Let us see what happens to a variety of simple patterns.
A single organism or any pair of counters, wherever placed, will ob-
viously vanish on the first tick.
A beginning pattern of three counters also dies immediately unless
at least one counter has two neighbors. Figure 31.1 shows the five con-
nected triplets that do not fade on the first tick.
(Their orientation is of course irrelevant.) The first three (a, b, c) van-
The Game of Life 411
a b c d e
31.1.
412
a b c d e
o
2
3
4
5
6
7
8
9
31.2. tetrominoes
413
414
One of the remarkable of
counter glider shown 31
slightly and
) hence the name.
glider has righted itself and moved one cell diagonally the
right We earlier that the speed of a
cause highest speed at which any can occur
on the board. No pattern can itself rapidly enough to move at
light.
same orientation after four ticks, and has traveled one cell diagonally,
one says that it one"fourth the speed of light.
l'.S. The "glider"
any reader find a relatively simple figure that travels at such a
speed? Remember, the by dividing the number of
four the same traveling
unit squares horizontally or vertically, its speed will be half that of
light. Figures that move across by self-replication are ex-
which he calls "spacesbips" (the glider is a "featherweight
the others have more I will their patterns in the An-
swer Section.
by
collaborators. The stable honey farm (a in Figure 31.6) results after 14
ticks from a row of seven a 5 x 5 block one
IUv ...................ULV ..... a
The "figme 8
U
(b in .6), an oscilla"
tor found by Norton, both resembles an 8 and has a period The
415
31.6. pa11erns, one
screen. "
416
Addendum
My 1970 column on Conway's "Life" met with such an instant
enthusiastic response among computer hackers around the world that
their mania for exploring HLife" forms was estimated to have cost the
nation millions of dollars in illicit computer time. One computer ex-
pert, whom I shall leave nameless, installed a secret switch under his
desk. If one of his bosses entered the room he would press the button
and switch his computer screen from its "Life" program to one of the
company's projects.
The troublesome R pentomino becomes a 2-tick oscillator after 1,103
ticks. Six. gliders have been produced and are traveling outward. The
debris left at the center (see Figure 31.7) consists of four blinkers, one
ship, one boat, one loaf, four beehives, and eight blocks. This was first
established at Case Western Reserve University by Gary Filipski and
Brad Morgan and later confirmed by scores of "Life" hackers here and
abroad.
The fate of the 5-5-5-5-5-5-5 was first independently found by
Robert T. Wainwright and a group of hackers at Honeywell's Computer
Control Division, later by many others. The pattern stabilizes as a 2-tick
oscillator after 323 ticks with four traffic lights, eight blinkers, eight
loaves, eight beehives, and four blocks. Figure 31.8 reproduces a print-
out of the final steady state. Because symmetry cannot be lost in the his-
tory of any life form, the vertical and horizontal axes of the original
symmetry are preserved in the final state. The maximum population
(492 bits) is reached in generation 283, and the final population is 192.
When Conway visited me in 1970 and demonstrated his game of Life
on my Go board, neither he nor I had any inkling of how famous the
game would become, and what surprises about it would soon be dis-
covered. It is far and away the most played, most studied of all cellu-
lar automata recreations.
Bill Gosper, when he worked in M.l. T. 's artificial intelligence labo-
ratory run by Marvin Minsky, won Conway's $50 prize by discovering
the glider gun. I can still recall my excitement when I received Gosper's
telegram describing the gun. This is a Life pattern that shoots off a con-
stant stream of gliders. The gun soon provided the base for proving
that Life could function as a Turing machine capable, in principle, of
performing all the calculations possible by a digital computer. The
gun's reality was at once verified by Robert Wainwright who later
The Game of Life 417
3 1 . ~ R pentomino's original (black) and final (open dots) state. gliders are out
418 COMB INA TORICS
31.8. state 5-5-5-5-5-5-5 row
419
31.9.
420
P.
J.
1971.
192-93.
32-41.
two escorts
421
P. Macaluso, "Life After Death," Byte, Vol. 6, July 1981, pp. 326-33.
J. K. Millen, "One-Dimensional Life," Byte, Vol. 3, December 1978, pp. 68-74.
J. Millium, J. Reardon, and P. Smart, "Life with Your Computer," Byte, Vol. 3, De-
cember 1978, pp. 45-50.
M. Miyamoto, "An Equilibrium State for a One-Dimensional Life Game," Journal
of Ma them a tics of Kyoto University, Vol. 19, 1979, pp. 525-40.
M. D. Niemiec, "Life Algorithms," Byte, Vol. 4, January 1979, pp. 90-97.
W. Pardner, "Life is Simple with APL," PC Tech Journal, Vol. 2, September 1984,
pp.129-47.
H. A. Peelle, "The Game of Life," Recreational Computing, Vol. 7, May/June 1979,
pp.16-27.
D. T. Piele, "Population Explosion: An Activity Lesson," Mathematics Teacher,
October 1974, pp. 496-502.
W. Poundstone, The Recursive Universe, Morrow, 1984.
L. E. Schulman and P. E. Seiden, "Statistical Mechanics of a Dynamical System
Based on Conway's Game of Life," IBM Research Report RC 6802, October 1977.
R. Soderstrom, "Life Can Be Easy," Byte, Vol. 4, April 1979, pp. 166-69.
R. Wainwright (ed.), Lifeline, a newsletter on Life. Issues 1 through 11, March 1971
through September 1973.
M. Zeleny, "Self-Organization of Living Systems," International Journal of General
Systems, Vol. 4, 1977, pp. 13-28.
422 COMBINATORICS
Chapter 32 Paper Folding
The easiest way to refold a road map is dif-
ferently. -JONES'S Rule ofthe Road
One of the most unusual and frustrating unsolved problems in
modern combinatorial theory, proposed many years ago by Stanislaw
M. Ulam, is the problem of determining the number of different ways
to fold a rectangular "map." The map is precreased along vertical and
horizontal lines to form a matrix of identical rectangles. The folds are
confined to the creases, and the final result must be a packet with any
rectangle on top and all the others under it. Since there are various
ways to define what is meant by a "different" fold, we make the defin-
ition precise by assuming that the cells of the unfolded map are num-
bered consecutively, left to right and top to bottom. We wish to know
how many permutations of these n cells, reading from the top of the
packet down, can be achieved by folding. Cells are numbered the same
on both sides, so that it does not matter which side of a cell is "up" in
the final packet. Either end of the packet can be its "top," and as a re-
suit every fold will produce two permutations, one the reverse of the
other. The shape of each rectangle is irrelevant because no fold can ro-
tate a cell 90 degrees. We can therefore assume without altering the
problem that all the cells are identical squares.
The simplest case is the 1 x n rectangle, or a single strip of n squares.
It is often referred to as the problem of folding a strip of stamps along
their perforated edges until all the stamps are under one stamp. Even
this special case is still unsolved in the sense that no nonrecursive for-
mula has been found for the number of possible permutations of n
stamps. Recursive procedures (procedures that allow calculating the
number of folds for n stamps provided that the number for n 1 stamps
is known) are nonetheless known. The total number of permutations of
n objects is n! (that is, factorial n, or n x (n - 1) x (n - 2) ... x 1). All n!
permutations can be folded with a strip of two or three stamps, but for
423
1 234 234 1 341 2 4 1 2 3
1 243 2431 431 2
1 342 2 1 3 4 421 3
1 432 2 1 4 3 321 4 432 1
32.1. 16 ways to
424
2 x 3 map (lettering from left to right and from the top down) so that the
map can be folded into a packet that spells, from the top down, an ana-
gram of the original word. Each cell should be labeled the same on
both sides to make it easier to identify in the packet. For example, it is
not hard to fold ILL-FED to spell FILLED and SQUIRE to spell RISQUE.
On the other hand, OSBERG (an anagram for the last name of the Ar-
gentine writer Jorge Luis Borges that appears on page 361 of Vladimir
Nabokov's novel Ada) cannot be folded to BORGES, nor can BORGES
be folded to OSBERG. Can the reader give a simple proof of both im-
possibilities?
The 2 x 4 rectangle is the basis of two map-fold puzzles by Henry
Ernest Dudeney (see page 130 of his 536 Puzzles &- Curious Problems,
Scribner's, 1967). Dudeney asserts there are 40 ways to fold this rec-
tangle into a packet with cell No.1 on top, and although he speaks
tantalizingly of a "little law" he discovered for identifying certain pos-
sible folds, he offers no hint as to its nature. I have no notion how
many of the 8! = 40,320 permutations can be folded.
When one considers the 3 x 3, the smallest nontrivial square, the
problem becomes fantastically complex. As far as I know, the number
of possible folds (of the 9! = 362,880 permutations) has not been cal-
culated, although many paperfold puzzles have exploited this square.
One was an advertising premium, printed in 1942 by a company in
Mt. Vernon, NY, that is shown in Figure 32.2.
On one side of the sheet are pictures of Hitler and Mussolini. In the
I
I
I
I
I
I
I
I
I
I
!
,
I
1
- - ~ - - - - - - - - - - - - - - - - - - - ~ - - - - - - - - - - - - - - - - ~
~ I ~ I ~
:::=.;r.----r.----r---i-----------------t----
h
----------
~ - ~ ~ ~ ~ i
I
!
I
I
I
,
,
!
I
I
Figure 32.2. A map-fold problem from World War II
Paper Folding 425
top left corner is a window with open spaces die-cut between the bars.
The illustration incorrectly shows a picture of Tojo and a second win-
dow below him. Those two cells should have been blank. Tojo and the
second window are on the underside of the two cells. The task is to im-
prison two of the three dictators by folding the square along its grid
lines into a 1 x 1 packet so that on each side of the packet is a window
through which you see the face of a dictator. The fold is not difficult,
though it does require a final tuck.
A much tougher puzzle using a square of the same size is the creation
of Robert Edward Neale, a Protestant minister, retired professor of psy-
chiatry and religion from the Union Theological Seminary, and the au-
thor of the influential book In Praise of Play (Harper & Row, 1969).
Neale is a man of many avocations. One of them is origami, the Orien-
tal art of paper folding, a field in which he is recognized as one of the
country's most creative experts. Magic is another of Neale's side inter-
ests; his famous trick of the bunny in the top hat, done with a folded
dollar bill, is a favorite among magicians. The hat is held upside down.
When its sides are squeezed, a rabbit's head pops up.
Figure 32.3 shows Neale's hitherto unpublished Beelzebub puzzle.
Start by cutting a square from a sheet of paper or thin cardboard, crease
it to make nine cells, then letter the cells (the same letter on opposite
sides of each cell) as indicated. First try to fold the square into a packet
that spells (from the top down) these eight pseudonyms of the fallen
angel who, in Milton's Paradise Lost, is second in rank to Satan him-
self: Bel Zeebub, Bub Blezee, Ube Blezbe, Bub Zelbee, Bub Beelze, Zee
Bubble, Buz Lebeeb, Zel Beebub. If you can master these names, you are
ready to tackle the really fiendish one: Beelzebub, the true name of
"the prince of the devils" (Matthew 12:24). Its extremely difficult fold
will be explained in the Answer Section. No one who succeeds in fold-
L Z
B B
B U
426
E
E
E
Figure 32.3. Robert Edward Neale's
Beelzebub puzzle
C OMBINATORICS
ing all nine names will wonder why the general map-folding problem
is still unsolved.
Neale has invented a variety of remarkable paper-fold puzzles, but
there is space for only two more. One is in effect a nonrectangular
"map" with a crosscut at the center (see Figure 32.4). The numbers
may represent six colors: all the i-cells are one color, the 2-cells are a
second color, and so on. Here opposite sides of each cell are different.
After numbering or coloring as shown at the top in the illustration,
turn the sheet over (turn it sideways, exchanging left and right sides)
and then number or color the back as shown at the bottom. The sheet
must now be folded to form a curious species of tetraflexagon.
To fold the tetraflexagon, position the sheet as shown at the top in the
I
I
I
G
I
6 I
I
I
I
I
I
3 ~ 6
I
V I
I
I
I
--------
I
I
V
I
6 ~ 3
I
I
I
I
I
I
I
I
6
I
G I
I
I
I
I
I
4
I
3 I
I
I
I
I
I
1 2 9
I
9
I
I
I
I
9 9 2 1
I
I
3
I
4
I
I
I
Figure 32.4. Front (top) and back (bottom) of the unfolded tetraflexagon
Paper Folding 427
illustration. (It helps if you first press the creases so that the solid lines
are what origamians call "mountain folds" and the dotted lines are
"valley folds.") Reach underneath and seize from below the two free
corners of the i-cells, holding the corner of the upper cell between the
tip of your left thumb and index finger and the corner ofthe lower cell
between the tip of your right thumb and finger. A beautiful maneuver
can now be executed, one that is easy to do when you get the knack
even though it is difficult to describe. Pull the corners simultaneously
down and away from each other, turning each i-cell over so that it be-
comes a 5-cell as you look down at the sheet. The remaining cells will
come together to form two opentop boxes with a 6-cell at the bottom of
each box (see Figure 32.5).
Figure 32.S. First step in folding the
tetraflexagon
Shift your grip to the two inside corners of the 5-cells-corners di-
agonally opposite the corners you were holding. Push down on these
corners, at the same time pulling them apart. The boxes will collapse
so that the sheet becomes a flat 2 x 2 tetraflexagon with four i-cells on
top and four 2-cells on the underside (see Figure 32.6). If the collaps-
ing is not properly done, you will find a 4-cell in place of a i-cell,
and/ or a 3-cell in place of a 2-cell. In either case simply tuck the wrong
square out of sight, replacing it with the correct one.
The tetraflexagon is flexed by folding it in half (the two sides going
back), then opening it at the center crease to discover a new "face," all
Figure 32.6. Final step in folding the tetraflexagon
1
1
428 COMBINATORICS
c d
32.1. problem
429
Samuel Randlett's The Best of Origami (E. P. Dutton, 1963). Some of his
dollar-bill folds (including the jumping frog) are in Folding Money:
Volume II, edited by Randlett (Magic, Inc., 1968).
Addendum
The "little law" that Henry Ernest Dudeney hinted about in
connection with his map-fold problem has probably been rediscov-
ered. Mark B. Wells of the Los Alamos Scientific Laboratory used a
computer to confirm that the 2 x 3 map has 10 folds for each cell on top.
The program also found that the order-3 square has 152 folds for each
cell on top. In his 1971 paper Lunnon proved that for any rectangular
map every cyclic permutation of every possible fold is also a possible
fold. Thus it is necessary to determine only the folds for one cell on top
because the cyclic permutations of these folds give all the other folds.
For example, since 123654789 is a possible fold, so also are 236547891,
365478912, and so on. It is a strange law because the folds for cyclic
permutations differ wildly. It is not yet known whether the law ap-
plies to all polyomino-shaped maps or to maps with equilateral trian-
gles as cells.
In his 1971 paper Lunnon used an ingenious diagram based on two
perpendicular slices through the center of the final packet. He was able
to write a simple backtrack program for x x y maps, extend the problem
to higher dimensions, and discover several remarkable theorems. For
example, the edges of one cross section always diagram x linear maps
of y cells each, and the edges of the other cross section diagram y lin-
ear maps of x cells each.
The 2 x 3, 2 x 4, 2 x 5, 2 x 6, and 3 x 4 maps have respectively 60, 320,
1,980, 10,512, and 15,552 folds. The order-3 square has 1,368 folds, the
order-4 has 300,608, the order-5 has 186,086,600. In all cases the num-
ber of folds is the same for each cell on top, as required by cyclic law.
The order-2 cube, folded through the fourth dimension, has 96 folds.
The order-3 cube has 85,109,616. Many other results are tabulated by
Lunnon in his 1971 paper, but a non recursive formula for even planar
maps remains elusive.
The linear map-fold function, as Lunnon calls it, is the limit ap-
proached by the ratio between adjacent values of the number of possi-
ble folds for a 1 x n strip. It is very close to 3.5. In his unpublished 1981
430 COMBINATORICS
paper Lunnon narrows the upper and lower bounds to 3.3868 and
3.9821.
In 1981 Harmony Books in the United States, and Pan Books in Eng-
land, brought out a large paperback book called Folding Frenzy. It con-
tains six 3 x 3 squares, with red and green patterns on both sides, and
five pages partially die-cut. Without removing any pages, there are nine
puzzles to solve by folding the squares. The puzzles are credited to Je-
remy Cox.
In describing one of his map-fold puzzles (Modern Puzzles, No. 214)
Dudeney mentioned a curious property of map folds that is not at all
obvious until you think about it carefully. It applies not only to rec-
tangular maps, but also to maps in the shape of any polyomino; that is,
a shape formed by joining unit squares at their edges. Assume that any
such map is red on one side, white on the other. No matter how it is
folded into a 1 x 1 packet, the colors on the tops of each cell will alter-
nate regardless of which side ofthe packet is up. If the cells of the map
are colored like a checkerboard, with each cell the same color on both
sides, the final packet (after any sort of folding) will have leaves that al-
ternate colors. If the checkerboard coloring is such that each cell is red
on one side, white on the other, all cells in the folded packet will have
their red sides facing one way, their white sides facing the other way.
It occurred to me in 1971 that the parity principles involved here
could be the basis for a variety of magic tricks. One appeared under the
title "Paradox Papers" in Karl Fulves' magic periodical, The Pallbear-
ers Review. It goes like this: Fold a sheet of paper twice in each direc-
tion so that the creases make 16 cells. It is a good plan to fold the paper
each way along every crease to make refolding easier later on.
Assume in your mind that the cells are checkerboard colored black
and red, with red at the top left corner. Five red playing cards are taken
from a deck and someone selects one of them. With a red pencil jot the
names of the five cards in five cells, using abbreviations such as 4D and
QH. Tell your audience that you are taking cells at random, but actually
you must put the name of the chosen card on one of the "black" cells,
and the other four names on "red" cells.
Have another card chosen, this time from a set of five black cards.
Turn the sheet over, side for side, and jot the names of the five black
cards on cells, again apparently at random. Use a black pencil. Put the
chosen card on a "red" cell, the others on "black" cells.
Paper Folding 431
Ask someone to fold the sheet any way he likes to make a 1 x 1
packet. With a pair of shears, trim around the four sides of the packet.
Deal the 16 pieces on the table. Five names will be seen, all the same
color except for one-the chosen card of the other color. Turn over the
16 pieces. The same will be true of the other sides.
Gene Nielsen, in the May 1972 issue of the same journal, suggested
the following variant. Pencil X's and O's on all the cells, alternating
them checkerboard fashion. Thrn over the sheet horizontally, and put
exactly the same pattern on the other side. Spectators will not realize
that each cell has an X on one side, an 0 on the other. Someone folds
the sheet randomly into a packet. Pretend you are using PK to influence
the folding so that it will produce a startling result. Trim the sides of the
packet and spread the pieces on the table. All X's face one way, all O's
face the other way.
Swami, a magic periodical published in Calcutta by Sam Dalal,
printed my "Paper Fold Prediction" in its July 1973 issue. Start by
numbering the cells of a 3 x 3 sheet from 1 through 9, taking the cells
in the usual way from left to right and top down. Put the digits on one
side of the paper only. After someone folds the sheet randomly, trim the
sides of the packet and spread the pieces. Add all the numbers show-
ing. Reverse the pieces and add the digits on the other sides. The two
sums will be different. Explain that by randomly folding the sheet, the
nine digits are randomly split into two sets. Clearly there is no way to
know in advance what either sum will be when the pieces are spread.
Repeat the same procedure, but this time use a 4 x 4 square with
cells numbered 1 through 16. The sheet is randomly folded and the
edges are trimmed. Before spreading the pieces, hold them to your fore-
head and announce that the sum will be 68. Put down the packet, ei-
ther side uppermost, and spread the pieces. The numbers showing will
total 68. Discard the pieces before anyone discovers that the sum on the
reverse sides also is 68.
The trick works because if the original square has an odd number of
cells, the sums on the two sides will not be equal. (On the 3 x 3 they
will be 20 and 25.) However, if the square has an even number of cells,
the sum is a constant equal to (n
2
+ n)/4 where n is the highest number.
You can now repeat the trick with a 5 x 5 square, but instead of pre-
dicting a sum, predict that the difference between the sums on the two
sides of the pieces will be 13.
The principle applies to cells numbered with other sequences. For
432 COMBINATORICS
example, hand a wall calendar to someone and ask him to tear out the
page for the month of his birth. He then cuts from the page any 4 x 4
square of numbers. The sheet is folded, the packet is trimmed, the
pieces are spread, and the visible numbers are added. The sum will be
equal to four times the sum of the sheet's lowest and highest numbers.
You can predict this as soon as you see the square that has been cut, or
you can divine the number later by ESP.
Some other suggestions. Allow a spectator to write any digit he likes
in each cell of a sheet of any size, writing left to right and top down. As
he writes, keep a running total in your head by subtracting the second
number from the first, adding the third, subtracting the fourth, and so
on. The running total is likely to fluctuate between plus and minus.
The number you end with, whether plus or minus, will be the differ-
ence between the two sums after the sheet is folded, trimmed, and the
pieces are spread.
Magic squares lend themselves to prediction tricks of a similar na-
ture. For example, suppose a 4 x 4 map bears the numbers of a magic
square. After folding, trim only on two opposite sides of the packet.
This will produce four strips. Have someone select one ofthe four. The
other three are destroyed. You can predict the sum of the numbers on
the selected strip because it will be the magic square's constant. Of
course you do not tell the audience that the numbers form a magic
square.
Many of these tricks adapt easily to nons quare sheets, such as a 3 x
4. See my Mental Magic (Sterling, 1999), page 54, for a trick using a 3
x 4 sheet bearing the letters A through G.
In 1984 C. MaIorana, of Washington, DC, printed what he called "The
Delayed Justice Puzzle." It is the same as the 3 x 3 fold puzzle I de-
scribed except that the faces of the three dictators are replaced by three
faces of Nixon. "01' Tricky Dick escaped justice once," the instructions
read, "but this time I think we'll get him! Try and fold the paper (only
along the dashed lines of course) so that 01' Tricky is behind the bars
on each side of the packet."
Answers
A simple proof that on the 2 x 3 rectangle OSBERG cannot be
folded to spell BORGES (or vice versa) is to note that in each case the
fold requires that two pairs of cells touching only at their corners would
Paper Folding 433
have to be brought together in the final packet. It is evident that no fold
can put a pair of such cells together.
The square puzzle with the faces and prison windows is solved from
the starting position shown. Fold the top row back and down, the left
column toward you and right, the bottom row back and up. Fold the
right packet of three cells back and tuck it into the pocket. A face is now
behind bars on each side of the final packet. The central face of the
square cannot be put behind bars because its cell is diagonally adjacent
to each of the window cells.
Space prevents my giving solutions for the eight pseudonyms of
Beelzebub, but Beelzebub itself can be obtained as follows. Starting
with the layout shown, fold the bottom row toward you and up to cover
BBE. Fold the left column toward you and right to cover ZU. Fold the
top row toward you and down, but reverse the crease between L and Z
so that LZ goes between Band B on the left and the upper E goes on top
of the lower E. You now have a rectangle of two squares. On the left,
from the top down, the cells are BLZBUB, on the right EEE. The final
move is difficult. Fold the right panel (EEE) toward you and left. The
three E's are tucked so that the middle E goes between Z and B, and the
other two Ks together go between Band L. Once you grasp what is re-
quired it is easier to combine this awkward move with the previous
one. The result is a tightly locked packet that spells Beelzebub. The so-
lution is unique. If the cells of the original "map" are numbered 1
through 9, the final packet is 463129785.
To find the 5-face of the tetraflexagon, start with face 1 on the top
and 2 on the bottom. Mountain-fold in half vertically, left and right
panels going back, so that if you were to open the flexagon at the cen-
ter crease you would see the 4-face. Instead of opening it, however,
move left the lower inside square packet (with 4 and 3 on its outsides)
and move right the upper square packet (also with 4 and 3 on its out-
sides). Insert your fingers and open the flexagon into a cubical tube
open at the top and bottom. Collapse the tube the other way. This cre-
ates a new tetraflexagon structure that can be flexed to show faces 1,
3, and 5.
A similar maneuver creates a structure that shows faces 2, 4, and 6.
Go back to the original structure that shows faces 1, 2, 3, and 4 and re-
peat the same moves as before except that you begin with the 2-face up-
permost and the 1-face on the underside.
Figure 32.8 shows how to separate the sheep from the goats:
434 COMBINATORICS
1
1.
2.
5
Goats"
hold the square out of sight under a table for just a few moments and
produce the change almost as if by magic.
Bibliography
S. Baker, "Postage Stamps," "Computer Solutions to Three Topological-
Combinatorial Problems," 1976, unpublished.
J. E. Koehler, "Folding a Strip of Stamps," Journal of Combinatorial Theory, Vol. 5,
September 1968, pp. 135-52.
W. F. Lunnon, "A Map-Folding Problem," Mathematics of Computation, Vol. 22,
January 1968, pp. 193-99.
W. F. Lunnon, "Multi-Dimensional Stamp Folding," The Computer Journal, Vol. 14,
February 1971, pp. 75-79.
W. F. Lunnon, "The Map-Folding Problem Revisited," unpublished.
W. F. Lunnon, "Bounds for the Map-Folding Function," Cardiff, August, 1981, un-
published.
R. Neale, "The Devil's Fold," Games, November/December 1979, p. 33.
A. Sainte-Lague, Avec des nombres et des lignes (With Numbers and Lines), Paris:
Libraire Vuibert, 1937; third edition 1946. The linear (1 x n) postage stamp prob-
lem is discussed in Part 2, Chapter 2, pages 147-62. I am indebted to Victor
Meally for telling me of this earliest known analysis of the problem.
J. Touchard, "Contributions a l'etude du probleme des timbres poste," Canadian
Journal of Mathematics, Vol. 2, 1950, pp. 385-98.
436 COMBINATORICS
Chapter 33 Ramsey Theory
Prove that at a gathering of any six people,
some three of them are either mutual ac-
quaintances or complete strangers to each
other. -PROBLEM E 1321, The American
Mathematical Monthly, June-July, 1958
This chapter was originally written in 1977 to honor the ap-
pearance of The Journal of Gmph Theory, a periodical devoted to one
of the fastest growing branches of modern mathematics. Frank Harary,
the founding editor, is the author of the world's most widely used in-
troduction to the subject. The current managing editor is Fan Chung.
Graph theory studies sets of points joined by lines. Two articles in the
first issue of the new journal dealt with Ramsey graph theory, a topic
that has a large overlap with recreational mathematics. Although a few
papers on Ramsey theory, by the Hungarian mathematician Paul Erdos
and others, appeared in the 1930s, it was not until the late 1950s that
work began in earnest on the search for what are now called Ramsey
numbers. One of the great stimulants to this search was the innocent-
seeming puzzle quoted above. It was making the mathematical-folklore
rounds as a graph problem at least as early as 1950, and at Harary's sug-
gestion it was included in the William Lowell Putnam Mathematical
Competition of 1953.
It is easy to transform this puzzle into a graph problem. Six points
represent the six people. Join every pair of points with a line, using a
red pencil, say, to indicate two people who know each other and a blue
pencil for two strangers. The problem now is to prove that no matter
how the lines are colored, you cannot avoid producing either a red tri-
angle (joining three mutual acquaintances) or a blue triangle (joining
three strangers).
Ramsey theory, which deals with such problems, is named for an
extraordinary University of Cambridge mathematician, Frank Plump-
ton Ramsey. Ramsey was only 26 when he died in 1930, a few days
after an abdominal operation for jaundice. His father, A. S. Ramsey,
was president of Magdalene College, Cambridge, and his younger
437
brother Michael was archbishop of Canterbury from 1961 to 1974.
Economists know him for his remarkable contributions to economic
theory. Logicians know him for his simplification of Bertrand Russell's
ramified theory of types (it is said that Ramsey Ramseyfied the ramified
theory) and for his division of logic paradoxes into logical and seman-
tical classes. Philosophers of science know him for his subjective in-
terpretation of probability in terms of beliefs and for his invention of
the "Ramsey sentence," a symbolic device that greatly clarifies the na-
ture of the "theoretical language" of science.
In 1928 Ramsey read to the London Mathematical Society a now
classic paper, "On a Problem of Formal Logic." (It is reprinted in The
Foundations of Mathematics, a posthumous collection of Ramsey's es-
says edited by his friend R. B. Braithwaite.) In this paper Ramsey
proved a deep result about sets that is now known as Ramsey's theo-
rem. He proved it first for infinite sets, observing this to be easier than
his next proof, for finite sets. Like so many theorems about sets, it
turned out to have a large variety of unexpected applications to com-
binatorial problems. The theorem in its full generality is too compli-
cated to explain here, but for our purposes it will be sufficient to see
how it applies to graph-coloring theory.
When all pairs of n points are joined by lines, the graph is called a
complete graph on n points and is symbolized by Kn' Since we are con-
cerned only with topological properties, it does not matter how the
points are placed or the lines are drawn. Figure 33.1 shows the usual
ways of depicting complete graphs on two through six points. The lines
identify every subset of n that has exactly two members.
Suppose we arbitrarily color the lines of a Kn graph red or blue. We
might color the lines all red or all blue or any mixture in between. This
is called a two-coloring of the graph. The coloring is of course a simple
way to divide all the two-member subsets of n into two mutually ex-
clusive classes. Similarly, a three-coloring of the lines divides them
into three classes. In general, an r-coloring divides the pairs of points
into r mutually exclusive classes.
A "subgraph" of a complete graph is any kind of graph contained in
the complete graph in the sense that all the points and lines of the sub-
graph are in the larger graph. It is easy to see that any complete graph
is a subgraph of any complete graph on more points. Many simple
graphs have names. Figure 33.2 shows four families: paths, cycles,
stars, and wheels. Note that the wheel on four points is another way of
438 COMBINATORICS
K2---------------
Ks
Figure 33.1. Complete graphs on two through six points
!points Paths Cycles Stars
Wheels
2
/
3
/\
4
N D Y
.Q1
5
M
0
X
[gJ
6
MI 0
* *
Figure 33.2. Four important families of simple graphs
Ramsey Theory 439
drawing K
4
. It is often called a tetrahedron because it is a planar pro-
jection of the tetrahedron's skeleton.
Consider now the following problem involving six pencils of differ-
ent colors. To each color we assign any kind of graph we like. For ex-
ample:
1. Red: a pentagon (five-point cycle).
2. Orange: a tetrahedron.
3. Yellow: a seven-point star.
4. Green: a 13-point path.
5. Blue: an eight-point wheel.
6. Purple: a bow tie (two triangles sharing just one point).
We now ask a curious question. Are there complete graphs that, if
their lines are arbitrarily six-colored, are certain to contain as a sub-
graph at least one of the six graphs listed above? In other words. no mat-
ter how we color one of these complete graphs with the six pencils are
we certain to get either a red pentagon or an orange tetrahedron or a yel-
low seven-point star, and so on? Ramsey's theorem proves that beyond
a certain size all complete graphs have this property. Let's call the
smallest graph of this infinite set the Ramsey graph for the specified set
of subgraphs. Its number of points is called the Ramsey number for
that set of subgraphs.
Every Ramsey graph provides both a game and a puzzle. For our ex-
ample, the game is as follows. Two players take turns picking up any
one of the six pencils and coloring a line of the Ramsey graph. The first
person to complete the coloring of one of the specified subgraphs is the
loser. Since it is a Ramsey graph, the game cannot be a draw. Moreover,
it is the smallest complete graph on which a draw is not possible.
The related puzzle involves a complete graph with one fewer point
than the Ramsey graph. This obviously is the largest complete graph on
which the game can be a draw. Such a graph is called the critical col-
oring for the specified set of subgraphs. The puzzle consists in finding
a coloring for the critical graph in which none of the subgraphs ap-
pears.
I have no idea what the Ramsey number is for the six subgraphs
given. Its complete graph would be so large (containing hundreds of
points) that playing a game on it would be out of the question, and the
associated puzzle is far too difficult to be within the range of a feasible
computer search. Nevertheless, Ramsey games and puzzles with
440 COMBINATORICS
smaller complete graphs and with pencils of just two colors can be
quite entertaining.
The best-known Ramsey game is called Sim after mathematician
Gustavus Simmons who was the first to propose it. Sim is played on the
complete graph with six points (K
6
), which models the problem about
the party of six people. It is not hard to prove that 6 is the Ramsey
number for the following two subgraphs:
1. Red: triangle (K
3
).
2. Blue: triangle (K
3
).
In "classical" Ramsey theory it is customary to use solitary numbers
for complete graphs, and so we can express the above result with this
compact notation: R(3,3) = 6. The R stands for Ramsey number, the
first 3 for a triangle of one color (say, red) and the second 3 for a trian-
gle of another color (say, blue). In other words, the smallest complete
graph that forces a "monochromatic" (all red or all blue) triangle when
the graph is two-colored is 6. Thus if two players alternately color the
K6 red and blue, one player is certain to lose by completing a triangle
of his color. The corresponding and easy puzzle is to two-color the crit-
ical graph, K
5
, so that no monochromatic triangle appears.
It turns out that when K6 is two-colored, at least two monochromatic
triangles are forced. (If there are exactly two and they are of opposite
color, they form a bow tie.) This raises an interesting question. If a com-
plete graph on n points is two-colored, how many monochromatic tri-
angles are forced? A. W. Goodman was the first to answer this in a 1959
paper, "On Sets of Acquaintances and Strangers at Any Party." Good-
man's formula is best broken into three parts: If n has the form 2u, the
number of forced monochromatic triangles is lhu(u -l)(u - 2). If n is 4u
+ 1, the number is %2u(u - 1)(4u + 1). If n is 4u + 3, it is %2u(u + 1)(4u
- 1). Thus for complete graphs of 6 through 12 points the numbers of
forced one-color triangles are 2, 4, 8, 12, 20, 28, and 40.
Random two-coloring will usually produce more monochromatic tri-
angles than the number forced. When the coloring of a Ramsey graph
contains exactly the forced number of triangles and no more, it is called
extremal. Is there always an extremal coloring in which the forced tri-
angles are all the same color? (Such colorings have been called blue-
empty, meaning that the number of blue triangles is reduced to zero.)
In 1961 Leopold Sauve showed that the answer is no for all odd n, ex-
cept for n = 7. This suggests a new class of puzzles. For example, draw
Ramsey Theory 441
the complete graph on seven points. Can you two-color it so that there
are no blue triangles and no more than four red triangles? It is not easy.
Very little is known about "classical" Ramsey numbers. They are the
number of points in the smallest complete graph that forces a given set
of smaller complete graphs. There is no known practical procedure for
finding classical Ramsey numbers. An algorithm is known: One simply
explores all possible colorings of complete graphs, going up the ladder
until the Ramsey graph is found. This task grows so exponentially in
difficulty and at such a rapid rate, however, that it quickly becomes
computationally infeasible. Even less is known about who wins-the
first player or the second-if a Ramsey game is played rationally. Sim
has been solved (it is a second-player win), but almost nothing is
known about Ramsey games involving larger complete graphs.
So far we have considered only the kind of Ramsey game that Harary
calls an avoidance game. As he has pointed out, at least three other
kinds of game are possible. For example, in an "achievement" game
(along the lines of Sim) the first player to complete a monochromatic
triangle wins. In the other two games the play continues until all the
lines are colored, and then either the player who has the most triangles
of his color or the player who has the fewest wins. These last two games
are the most difficult to analyze, and the achievement game is the eas-
iest. In what follows "Ramsey game" denotes the avoidance game.
Apart from R(3,3) = 6, the basis of Sim, only the following seven
other nontrivial classical Ramsey numbers are known for two-
colorings:
1. R(3,4) = 9. If Kg is two-colored, it forces a red triangle (K
3
) or a blue
tetrahedron (K
4
). No one knows who wins if this is played as a Ram-
sey game, or who wins on any R(n,n) game with n larger than 3.
2. R(3,5) = 14.
3. R(4,4) = 18. If K
18
is two-colored, a monochromatic tetrahedron (K
4
)
is forced. This is not a bad Ramsey game, although the difficulty of
identifying tetrahedrons makes it hard to play. The graph and its col-
oring correspond to the fact that at a party of 18 people there is either
a set of four acquaintances or four total strangers.
4. R(3,6) = 18. At the same party there is either a set of three acquain-
tances or six total strangers. In coloring terms, if a complete graph of
18 points is two-colored it forces either a red triangle or a blue com-
plete graph of six points.
442 COMBINATORICS
5. R(3,7) = 23.
6. R(3,8) = 28.
7. R(3,9) = 36.
8. R(4,5) = 25.
9. R(6,7) = 298.
As of March 1996, here are known bounds for eight other Ramsey
numbers:
R(3,10) = 40-43.
R(4,6) = 35-41.
R(4,7) = 49-61.
R(5,5) = 43-49.
R(5,6) = 58-87.
R(5,7) = 80-143.
R(6,6) = 102-165.
R(7,7) = 205-540.
What is the smallest number of people that must include either a set
of five acquaintances or a set of five strangers? This is equivalent to ask-
ing for the smallest complete graph that cannot be two-colored without
producing a monochromatic complete graph with five points, which is
the same as asking for the Ramsey number of R(5,5). So great is the
jump in complexity from R(4,4) to R(5,5) that Stefan Burr, a leading ex-
pert on Ramsey theory who now teaches computer science at City Col-
lege of the CUNY in New York City, thinks it possible that the number
will never be known. Even R(4,5), he believes, is so difficult to analyze
that it is possible its value also may never be found.
Only one other classical Ramsey number is known, and it is for three
colors. R(3) = 3 is trivial because if you one-color a triangle, you are sure
to get a one-color triangle. We have seen that R(3,3) equals 6. R(3,3,3)
equals 17. This means that if K17 is three-colored, it forces a mono-
chromatic triangle. Actually it forces more than one, but the exact num-
ber is not known.
R( 3,3,3) = 17 was first proved in 1955. The Ramsey game for this graph
uses pencils of three different colors. Players alternately color a line,
using any color they want to, until a player loses by completing a mono-
chromatic triangle. Who wins if both players make their best possible
moves? No one knows. The corresponding Ramsey puzzle is to three-
color K
16
, the critical graph, so that no monochromatic triangle appears.
Ramsey Theory 443
What about R(3,3,3,3), the minimum complete graph that forces a
one-color triangle when it is four-colored? It is unknown, although an
upper bound of 64 was proved by Jon Folkman, a brilliant combinato-
rialist who committed suicide in 1964 at the age of 31, following an op-
eration for a massive brain tumor. The best lower bound, 51, was
established by Fan Chung, who gave the proof in her Ph.D. thesis.
Classical Ramsey theory generalizes in many fascinating ways. We
have already considered the most obvious way: the seeking of what
are called generalized Ramsey numbers for r-colorings of complete
graphs that force graphs other than complete ones. Vaclav Chvatal and
Harary were the pioneers in this territory, and Burr has been mining it
for many years. Consider the problem of finding Ramsey numbers for
minimum complete graphs that force a monochromatic star of n points.
Harary and Chvatal were the first to solve it for two-coloring. In 1973
Burr and J. A. Roberts solved it for any number of colors.
Another generalized Ramsey problem is to find Ramsey numbers for
two-colorings of Kn that force a specified number of monochromatic
"disjoint" triangles. (Triangles are disjoint if they have no common
point.) In 1975 Burr, Erdos, and J. H. Spencer showed the number to be
5d, where d is the number of disjoint triangles and is greater than two.
The problem is unsolved for more than two colors.
The general case of wheels is not even solved for two colors. The
Ramsey number for the wheel of four points, the tetrahedron, is, as we
have seen, 18. The wheel of five points (a wheel with a hub and four
spokes) was shown to have a Ramsey number of 15 by Tim Moon, a
Nigerian mathematician. The six-point wheel is unsolved, although its
Ramsey number is known to have bounds of 17 to 20 inclusive, and
there are rumors of an unpublished proof that the value is 17.
Figure 33.3, a valuable chart supplied by Burr and published for the
first time in my 1977 column, lists the 113 graphs with no more than
six lines and no isolated points, all of which have known generalized
Ramsey numbers. Note that some of these graphs are not connected. In
such cases the entire pattern, either all red or all blue, is forced by the
complete graph with the Ramsey number indicated.
Every item on Burr's chart is the basis of a Ramsey game and puzzle,
although it turns out that the puzzles-finding critical colorings for
the critical graphs-are much easier than finding critical colorings for
classical Ramsey numbers. Note that the chart gives six variations of
Sim. A two-coloring of K6 not only forces a monochromatic triangle but
444 COMBINATORICS
also forces a square, a four-point star (sometimes called a "claw"), a
five-point path, a pair of disjoint paths of two and three points (both the
same color), a square with a tail and the simple "tree" that is 15 on
Burr's chart. The triangle with a tail (8), the five-point star (12), the
Latin cross (27), and the fish (51) might be worth looking into as Ram-
sey games on K
7

Ronald L. Graham of Bell Laboratories, one of the nation's top com-
binatorialists, has made many significant contributions to generalized
Ramsey theory. It would be hard to find a creative mathematician who
less resembles the motion-picture stereotype. In his youth Graham and
two friends were professional trampoline performers who worked for
a circus under the name of the Bouncing Baers. He is also one of the
country's best jugglers and former president of the International Jug-
gler's Association. The ceiling of his office is covered with a large net
that he can lower and attach to his waist, so that when he is practicing
with six or seven balls, any missed ball obligingly rolls back to him.
In 1968 Graham found an ingenious solution for a problem of the
Ramsey type posed by Erdos and Andras Hajnal. What is the smallest
graph of any kind, not containing K
6
, that forces a monochromatic tri-
angle when it is two-colored? Graham's unique solution is the eight-
point graph shown in Figure 33.4. The proof is a straightforward
reductio ad absurdum. It begins with the assumption that a two-
coloring that avoids monochromatic triangles is possible and then
shows that this forces such a triangle. At least two lines from the top
point must be, say, gray, and the graph's symmetry allows us to make
the two outside lines gray with no loss of generality. The end points of
these two lines must then be joined by a colored line (shown dotted) to
prevent the formation of a gray triangle. Readers may enjoy trying to
complete the argument.
What about similar problems when the excluded subgraph is a com-
plete graph other than K6? The question is meaningless for K3 because
K3 is itself a triangle. Ks is unsolved. The best-known solution is a 16-
point graph discovered by two Bulgarian mathematicians. K4 is even
further from being solved. Folkman, in a paper published posthu-
mously, proved that such a Ramsey graph exists, but his construction
used more than 2 iii 2
901
points. This is such a monstrous number
that there is no way to express it without using a special arrow nota-
tion. The notation is introduced by Donald E. Knuth in his article
"Mathematics and Computer Science: Coping with Finiteness" in Sci-
Ramsey Theory 445
>I::..
f
>I::..
0)
~
w
w
~
CIl
......
S
'0
ro
()Q
.,
~
'0
~
Cf.I
0'
.,
:;:;
~
......
C)
~
....
~
CD
()Q
CD
~
CD
.,
e..
......
N
CD
~
:;0
S
Cf.I
CD
'<
g
S
0'"
CD
.,
CJ
......
a
Cf.I
~
~
~
0
~ ~
""l
a
::tJ
.....
CJ
CIl
Ramsey
Graph Number
1
..-.
2
2
..........
3
3
~
6
........
4 5 ........
5 5
6
~
6
7
0
6
8
t>-
7
9
0
10
20
0
9 39
--- 21
..-.
8 40
........
22
---
7 41

23 8 42 ........
24
~ I
7 43
25 8
44
26 e < 7
45
27
-+
7 46
28 I 8
47
~ I
9 58 !> - .. 11 77
AA
9
96>--- I
10
...........-
~
~ 9
97 : : : :
.-.
9 59 11 78
11
.........
IIII
11 60
~
11
79 >--<
9 98 A. .. 10
...-.-
~ 80 ~ A
99 ~ ~ 11
----
10 61 11 9
.-.

62 -<>- 81 t;:.. 10
-
11 8
100 t>-Il
11
-
-Ei I I
10 63
~ D .
10 82
0---------
9 101
OII
11
I III
12 64
A
11 83
0-10 102 ~ III
13
I III!
14 65
01
10 84
r>-I
9 103. 12
......... -
I><l
9 66
I>-<
11
85 !>------I
10 104 >---II 12
-
:J:j
~
CI:l
~
;;i
('!)
~
.r::.
.r::.
'J
10
t8l
11
--
---
12
~
13
14,A ......
15 >--
'16 <C
17
0-
18
t><:
19 t>--
18
6
7
6
7
6
9
6
9
9
29 >-<
8 48
IZr
30 .f::.. ------
8 49
0>
31
~
10 50
~
32
Dl
8 51
<
33
t>-I
8 52
ISr
34
LII
10 53
0
I
9
54 0-.
35
--
36
-EiI
8
55 <>---
37 ...!.... 9
S6 <I:-
38 >---1
9 57
IT:
10 67
<l4
11 86
tXt
9
105 ~ 1 1
11
9
68- - - -9
87
<1:1
9 106 :.:--..-: 12
10
69 > - 9
88
<)-1
9
107 A 11
11
70 I . . 9
89' :....: 11 -- 7 108
--
11
--
10
71 --<::
9
90 > -1 10
109 . : . ~ ~ 14
8
72 - ~
9
91 1 I 11
110
AlII
13
11
73 --+-
9 92
~ I
11 111
HI!
13
8 74
--i<1O
93 41
to 112
InII
15
11 75
>-t
9
94 >-< I
11 113
lIIIll
17
8 76
~ 1
9
95 :-:....: to
Figure 33.4. Graham's solution
to a problem by Erdos
ence (December 17, 1976). The number has since been lowered to a
more respectable size.
Imagine the universe tightly packed with spheres the size of elec-
trons. The total number of such spheres is inconceivably smaller than
the number occurring in Folkman's graph.
Folkman's graph dramatically illustrates how enormously difficult a
Ramsey problem can be even when the problem's statement mentions
no graph with more than four points. But, as Al Jolson liked to say, you
ain't heard nothin' yet. Graham has found an even more mind-boggling
example.
Consider a cube with lines joining every pair of corners. The result
is a complete graph on eight points, except now we have added a Eu-
clidean geometric structure. Imagine the lines of this spatial K8 arbi-
trarily colored red and blue. Can it be done in such a way that no
monochromatic K4 results that lies on a plane? The answer is yes, and
it is not hard to do.
Let us generalize to n-dimensional cubes. A hypercube has 2
n
cor-
ners. On the four-dimensional hypercube, it also is possible to two-
color the lines of the complete graph of 2
4
, or 16, points so that no
one-color complete planar graph of four points results. The same can
be done with the 2
5
hypercube of 32 points. This suggests the follow-
ing Euclidean Ramsey problem: What is the smallest dimension of a
hypercube such that if the lines joining all pairs of corners are
448 COMBINATORICS
two-colored, a planar K4 of one color will be forced? Ramsey's theorem
guarantees that the question has an answer only if the forced K4 is not
confined to a plane.
The existence of an answer when the forced monochromatic K4 is
planar was first proved by Graham and Bruce L. Rothschild in a far-
reaching generalization of Ramsey's theorem that they found in 1970.
Finding the actual number, however, is something else. Graham has es-
tablished an upper bound, but it is a bound so vast that it holds the
record for the largest number ever used in a serious mathematical proof.
To convey at least a vague notion of the size of Graham's number we
must first attempt to explain Knuth's arrow notation. The number writ-
ten 3 i 3 is 3 x 3 x 3 = 3
3
= 27. The number 3 i i 3 denotes the expres-
sion 3 i (3 i 3). Since 3 i 3 equals 27, we can write 3 i i 3 as 3 i 27 or
3
27
As a slanting tower of exponents it is
The tower is only three levels high, but written as an ordinary number
it is 7,625,597,484,987. This is a big leap from 27, but it is still such a
small number that we can actually print it.
When the huge number 3 iii 3 = 3 i i (3 i i 3) = 3 i i 3
27
is writ-
ten as a tower of 3's, it reaches a height of 7,625,597,484,987 levels.
Both the tower and the number it represents are now too big to be
printed without special notation.
Consider 3 iii i 3 = 3 iii (3 iii 3). Inside the parentheses is
the gigantic number obtained by the preceding calculation. It is no
longer possible to indicate in any simple way the height of the tower of
3's that expresses 3 iii i 3. The height is another universe away
from 3 iii 3. If we break 3 iii i 3 down to a series of the double-
arrow operations, it is 3 i i (3 i i (3 i i ... i i (3 i i 3) ... )), where
the number of steps to be iterated is 3 iii 3. As Knuth says, the dots
"suppress a lot of detail." 3 iii i 3 is unimaginably larger than 3 i i
i 3, but it is still small as finite numbers go, since most finite numbers
are very much larger.
We are now in a position to indicate Graham's number. It is repre-
septed in Figure 33.5. At the top is 3 iii i 3. This gives the number
of arrows in the number just below it. That number in turn gives the
number of arrows below it. This continues for 2
6
, or 64, layers. It is the
bottom number that Graham has proved to be an upper bound for the
hypercube problem.
Ramsey Theory 449
3 t t t t
3
A
3 t t
. . . . .
t t 3
r
.A
'\
t t t t
64 Layers
3
. . . . . . . . . . .
3
.
( ~ ________________ A ~ ________________ \
3 t t ...................... .
t t 3
Figure 33.S. Graham's upper bound for the solution to a Euclidean Ramsey problem
Now hold on to your hat. Ramsey-theory experts believe the actual
Ramsey number for this problem is probably 6. As Stanislaw M. Ulam
said many times in his lectures, "The infinite we shall do right away.
The finite may take a little longer."
Addendum
Since this chapter first appeared in Scientific American in
1977, such rapid progress has been made in Ramsey theory that a bib-
liography of papers would run to more than a thousand titles and any
effort to summarize here even the main results would be impossible.
Fortunately, two excellent books are now available as introductions to
the general field (of which Ramsey graph theory is only a part): Ronald
Graham's Rudiments of Ramsey Theory and Ramsey Theory, a book
Graham coauthored with two colleagues.
Ramsey theory now includes, among many other things, problems in-
volving the partitioning of lines in any graph or the partitioning of
points in any space. Euclidean Ramsey theory, pioneered in the 1970s
by Graham, Erdos, and others, concerns the k-coloring of all points in
a given Euclidean space and determining what patterns are forced. For
example, no matter how the points on the plane are two-colored, the
coloring will force the vertices of a monochromatic triangle of any spec-
ified size and shape except the equilateral triangle. (The plane can be
colored with stripes of two alternating colors of such widths that no
450 COMBINATORICS
equilateral triangle, say of side 1, can have all its corners the same
color.)
If all the points in Euclidean three-dimensional space are two-
colored, will it force an equilateral triangle of any size? Yes. Consider
the four points of any regular tetrahedron. No matter how the points are
two-colored, at least three must be the same color, and of course those
three will form a monochromatic equilateral triangle. For other theo-
rems of this type, much harder to prove, see the 1973 paper by Erdos,
Graham, and others.
During the past decade Frank Harary and his associates have been an-
alyzing Ramsey games. Only a small fraction of their results have been
published, but Harary is planning a large book on what he calls
achievement and avoidance games of the Ramsey type.
In Stefan Burr's chart (Figure 33.3) showing all graphs with six or
fewer lines, you will see that in the interval of integers 2 through 18,
only 4 and 16 are missing as generalized Ramsey numbers. There are
three graphs of seven lines each that have generalized Ramsey numbers
of 16, but there is no graph that has a generalized Ramsey number of 4.
Is every positive integer except 4 (ignoring 1, which is meaningless) a
generalized Ramsey number? In 1970 Harary showed that the answer
is yes. The impossibility of 4 is easily seen by inspecting Burr's chart.
The tetrahedron (complete graph for four points) cannot be the small-
est graph that forces a subgraph because every possible subgraph of a
tetrahedron has a Ramsey number higher or lower than 4.
What is the smallest complete graph which, if two-colored, will force
a monochromatic complete graph for five points? In other words, what
is the generalized Ramsey number R(5,5)? In 1975 Harary offered $100
for the first solution, but the prize remains unclaimed.
"Ramsey theory is only in its infancy," writes Harary in his tribute to
Ramsey in the special 1983 issue of The Journal of Graph Theory cited
in the bibliography and adds: "There is no way that Frank Ramsey
could have foreseen the theory his work has inspired." D. H. Mellor, in
another tribute to Ramsey in the same issue, has this memorable sen-
tence: "Ramsey's enduring fame in mathematics ... rests on a theorem
he didn't need, proved in the course of trying to do something we now
know can't be done!"
The discovery that Ramsey Number R(4,5) equals 25 was first an-
nounced in 1993 by the Rochester Democrat and Chronicle because one
Ramsey Theory 451
of the discoverers, Stanislaw Radziszowski, was a professor at the
Rochester Institute of Technology. Someone using the initials B.V.B.
reported this to Ann Landers who published the letter in her syndi-
cated advice column on June 22, 1993. B.V.B. put it this way:
Two professors, one from Rochester, the other from Australia, have
worked for three years, used 110 computers, and communicated 10,000
miles by electronic mail, and finally have learned the answer to a ques-
tion that has baffled scientists for 63 years. The question is this: If you
are having a party and want to invite at least four people who know
each other and five who don't, how many people should you invite?
The answer is 25. Mathematicians and scientists in countries world-
wide have sent messages of congratulations.
I don't want to take anything away from this spectacular achievement,
but it seems to me that the time and money spent on this project could
have been better used had they put it toward finding ways to get food to
the millions of starving children in war-torn countries around the world.
Miss Landers replied as follows:
Dear B.V.B.: There has to be more to this "discovery" than you re-
counted. The principle must be one that can be applied to solve impor-
tant scientific problems. If anyone in my reading audience can provide
an explanation in language a lay person can understand, I will print it.
Meanwhile I am "Baffled in Chicago."
I sent Miss Landers this letter:
You asked (June 22) for an explanation ofthe proof that 25 people are re-
quired at a party if at least four are mutually acquainted, and at least five
are mutual strangers. This is a theorem in what is called Ramsey theory.
It can be modeled by points on paper; each point representing a person.
Every point is joined to every other point by a line.
If a line is colored red it means that its two points are persons who
know each other. If a line is blue, it means its two persons are strangers.
We now can ask: what is the smallest number of points needed so that
no matter how the lines are bicolored, there is sure to be either four
points mutually joined by red lines, or five mutually joined by blue
lines? The answer, 25; was not proved until recently. For the problem's
background, see the chapter on Ramsey theory in my book Penrose Tiles
to Trapdoor Ciphers (W. H. Freeman, 1989).
As far as I know, the letter was not printed. Perhaps Miss Landers re-
mained baffled as to what earthly use this result could possibly have.
452 COMBINATORICS
Radziszowski and his associate Brendon McKay published their
proof in a paper titled "R(4,5) = 25" in The Journal of Graph Theory
(Vol. 19, No.3, 1995, pp. 309-21).
Answers
Figure 33.6 (note the color symmetry) shows how to two-color
the complete graph on seven points so that it is blue-empty (gray lines)
and contains four (the minimum) red triangles (black lines). If you en-
joyed working on this problem, you might like to tackle Ka by two-
coloring the complete graph on eight points so that it is blue-empty and
has eight (the minimum) red triangles.
Garry Lorden, in his 1962 paper, "Blue-Empty Chromatic Graphs,"
showed that problems of this type are uninteresting. If the number of
points n is even, Goodman had earlier shown that the graph can be
made blue-empty extremal simply by coloring red two complete sub-
graphs of n/2 points. If n is odd, K7 is (as I said earlier) the only com-
plete graph that can be made blue-empty extremal by two-coloring.
When two-coloring complete graphs where n is odd and not 7, how do
you create a blue-empty graph that is not extremal but has the smallest
number of red triangles? Lorden showed that this is easily done by par-
titioning the graph into two complete red subgraphs, one of (n + 1)/2
points, the other of (n - 1)/2 points. Some extra red lines may be re-
quired, but their addition is trivial.
Ramsey Theory
Figure 33.6. Solution to a
Ramsey graph puzzle
453
Bibliography
M. Albertson, "People Who Know People," Mathematics Magazine, Vol. 67, Octo-
ber 1994, pp. 278-81.
S. Burr, "Generalized Ramsey Theory for Graphs-A Survey," Graphs and Combi-
natorics, R. Bari and F. Harary (eds.), Springer-Verlag, 1974, pp. 58-75.
B. Cipra, "A Visit to Asymptopia," Science, Vol. 267, February 17, 1995, pp.
264-65.
P. Erdos, et aI., "Euclidean Ramsey Theorems 1," Journal of Combinatorial Theory,
Ser. A, Vol. 14, 1973, pp. 341-63.
A. W. Goodman, "On Sets of Acquaintances and Strangers at Any Party," The
American Mathematical Monthly, 66, 1959, pp. 778-83.
R. Graham, "On Edgewise 2-colored Graphs with Monochromatic Triangles and
Containing No Complete Hexagon," Journal of Combinatorial Theory, Vol. 4,
1968, p. 300.
R. Graham, B. Rothschild, and J. Spencer, Ramsey Theory, Wiley, 1980.
R. Graham, Rudiments of Ramsey Theory, The American Mathematical Society,
1981.
R. Graham and J. Spencer, "Ramsey Theory," Scientific American, July 1990, pp.
112-17.
F. Harary, Graph Theory, Addison-Wesley, 1969.
G. Lorden, "Blue-Empty Chromatic Graphs," The American Mathematical Monthly,
69,1962,pp.114-20.
J. Nesetril and V. Rodl (eds.), Mathematics of Ramsey Theory, Springer, 1990.
I. Peterson, "Party Numbers," Science News, Vol. 144, July 17,1993, pp. 46-47.
W. Slany, "Graph Ramsey Games," DBAl Technical Report, Technische Universi-
tat Wien, Vienna, Austria, November 5, 1999, pp. 1-45. The paper lists 69 ref-
erences.
The Journal of Graph Theory, 7, Spring 1983. This issue, dedicated to Frank Ram-
sey, contains 19 papers about Ramsey and Ramsey theory.
454 COMBINATORICS
Chapter 34
With useless endeavor,
Forever, forever,
Is Sisyphus rolling
His stone up the mountain!
Bulgarian Solitaire
and Other Seemingly
Endless Tasks
-Henry Wadsworth Longfellow,
The Masque of Pandora
Suppose you have a basket containing 100 eggs and also a sup-
ply of egg cartons. Your task is to put all the eggs into the cartons. A
step (or move) consists of putting one egg into a carton or taking one egg
from a carton and returning it to the basket. Your procedure is this:
After each two successive packings of an egg you move an egg from a
carton back to the basket. Although this is clearly an inefficient way to
pack the eggs, it is obvious that eventually all of them will get packed.
Now assume the basket can hold any finite number of eggs. The task
is unbounded if you are allowed to start with as many eggs as you like.
Once the initial number of eggs is specified, however, a finite upper
bound is set on the number of steps needed to complete the job.
If the rules allow transferring any number of eggs back to the basket
any time you like, the situation changes radically. There is no longer an
upper bound on the steps needed to finish the job even if the basket ini-
tially holds as few as two eggs. Depending on the rules, the task of
packing a finite number of eggs can be one that must end, one that can-
not end, or one that you can choose to make either finite or infinite in
duration.
We now consider several entertaining mathematical tasks with the
following characteristic. It seems intuitively true that you should be
able to delay completing the task forever, when actually there is no
way to avoid finishing it in a finite number of moves.
Our first example is from a paper by the philosopher-writer-Iogician
Raymond M. Smullyan. Imagine you have an infinite supply of pool
balls, each bearing a positive integer, and for every integer there is an
infinite number of balls. You also have a box that contains a finite
455
quantity of numbered balls. Your goal is to empty the box. Each step
consists of removing a ball and replacing it with any finite number of
balls of lower rank. The 1 balls are the only exceptions. Since no ball
has a rank lower than 1, there are no replacements for a 1 ball.
It is easy to empty the box in a finite number of steps. Simply replace
each ball higher than 1 with a 1 ball until only 1 balls remain, then take
out the 1 balls one at a time. The rules allow you, however, to replace
a ball with a rank above 1 with any finite number of balls of lower
rank. For instance, you may remove a ball of rank 1,000 and replace it
with a billion balls of rank 999, with 10 billion of rank 998, with a bil-
lion billion of rank 987, and so on. In this way the number of balls in
the box may increase beyond imagining at each step. Can you not pro-
long the emptying of the box forever? Incredible as it may seem at first,
there is no way to avoid completing the task.
Note that the number of steps needed to empty the box is unbounded
in a much stronger way than it is in the egg game. Not only is there no
bound on the number of eggs you begin with but also each time you re-
move a ball with a rank above 1 there is no bound to the number of
balls you may use to replace it. To borrow a phrase from John Horton
Conway, the procedure is "unboundedly unbounded." At every stage of
the game, as long as the box contains a single ball other than a 1 ball,
it is impossible to predict how many steps it will take to empty the box
of all but 1 balls. (If all the balls are of rank 1, the box will of course
empty in as many steps as there are 1 balls.) Nevertheless, no matter
how clever you are in replacing balls, the box eventually must empty
after a finite number of moves. Of course, we have to assume that al-
though you need not be immortal, you will live long enough to finish
the task.
Smullyan presents this surprising result in a paper, "Trees and Ball
Games," in Annals of the New York Academy of Sciences (Vol. 321,
1979, pp. 86-90). Several proofs are given, including a simple argument
by induction. I cannot improve on Smullyan's phrasing:
If all balls in the box are of rank 1, then we obviously have a losing
game. Suppose the highest rank of any ball in the box is 2. Then we
have at the outset a finite number of 2's and a finite number of l's. We
can't keep throwing away l's forever; hence we must sooner or later
throw out one of our 2's. Then we have one less 2 in the box (but possi-
bly many more l's than we started with). Again, we can't keep throwing
out l's forever, and so we must sooner or later throw out another 2. We
456 COMBINATORICS
see that after a finite number of steps we must throwaway our last 2, and
then we are back to the situation in which we have only l's. We already
know this to be a losing situation. This proves that the process must ter-
minate if the highest rank present is 2. Now, what if the highest rank is
3? We can't keep throwing away just balls of rank::; 2 forever (we just
proved that!); hence we must sooner or later throw out a 3. Then again
we must sooner or later throw out another 3, and so we must eventually
throw out our last 3. This then reduces the problem to the preceding case
when the highest rank present is 2, which we have already solved.
Smullyan also proves that the game ends by modeling it with a tree
graph. A "tree" is a set of line segments each of which joins two points,
and in such a way that every point is connected by a unique path of seg-
ments leading to a point called the tree's root. The first step of a ball
game, filling the box, is modeled by representing each ball as a point,
numbered like the ball and joined by a line to the tree's root. When a
ball is replaced by other balls of lower rank, its number is erased and
the new balls indicated by a higher level of numbered points are joined
to the spot where the ball was removed. In this way the tree grows
steadily upward, its "end points" (points that are not the root and are
attached to just one segment) always representing the balls in the box
at that stage of the game.
Smullyan proves that if this tree ever becomes infinite (has an infin-
ity of points), it must have at least one infinite branch stretching up-
ward forever. This, however, is clearly impossible because the numbers
along any branch steadily decrease and therefore must eventually ter-
minate in 1. Since the tree is finite, the game it models must end. As in
the ball version, there is no way to predict how many steps are needed
to complete the tree. At that stage, when the game becomes bounded,
all the end points are labeled 1. The number of these 1 points may, of
course, exceed the number of electrons in the universe, or any larger
number. Nevertheless, the game is not Sisyphean. It is certain to end
after a finite number of moves.
Smullyan's basic theorem, which he was the first to model as a ball
game, derives from theorems involving the ordering of sets that go back
to Georg Cantor's work on the transfinite ordinal numbers. It is closely
related to a deep theorem about infinite sets of finite trees that was
first proved by Joseph B. Kruskal and later in a simpler way by C. St. J.
A. Nash-Williams. More recently Nachum Dershowitz and Zohar
Manna have used similar arguments to show that certain computer
Bulgarian Solitaire and Other Seemingly Endless Tasks 457
programs, which involve "unboundedly unbounded" operations, must
eventually come to a halt.
A special case of Smullyan's ball game is modeled by numbering a fi-
nite tree upward from the root as in Figure 34.1, left. We are allowed to
chop off any end point, along with its attached segment, then add to the
tree as many new branches as we like, and wherever we like, provided
all the new points are of lower rank than the one removed. For exam-
ple, the figure at the right in the illustration shows a possible new
growth after a 4 point has been chopped off. In spite of the fact that after
each chop the tree may grow billions on billions of new branches, after
a finite number of chops the tree will be chopped down. Unlike the
more general ball game, we cannot remove any point we like, only the
end points, but because each removed point is replaced by points of
lower rank, Smullyan's ball theorem applies. The tree may grow in-
conceivably bushier after each chop, but there is a sense in which it al-
ways gets closer to the ground until eventually it vanishes.
A more complicated way of chopping down a tree was proposed by
Laurie Kirby and Jeff Paris in The Bulletin of the London Mathemati-
cal Society (Vol. 14, Part 4, No. 49, July 1982, pp. 285-93). They call
their tree graph a hydra. Its end points are the hydra'S heads, and Her-
cules wants to destroy the monster by total decapitation. When a head
is severed, its attached segment goes with it. Unfortunately after the
first chop the hydra acquires one or more new heads by growing a new
branch from a point (call it k) that is one step below the lost segment.
This new branch is an exact replica of the part of the hydra that extends
up from k. The figure at the top right in Figure 34.2 shows the hydra
after Hercules has chopped off the head indicated by the sword in the
figure at the top left.
The situation for Hercules becomes increasingly desperate because
when he makes his second chop, two replicas grow just below the sev-
ered segment (Figure 34.2, bottom left). And three replicas grow after
the third chop (Figure 34.2, bottom right), and so on. In general, n repli-
cas sprout at each nth chop. There is no way of labeling the hydra'S
points to make this growth correspond to Smullyan's ball game; nev-
ertheless, Kirby and Paris are able to show, utilizing an argument based
on a remarkable number theorem found by the British logician R. L.
Goodstein, that no matter what sequence Hercules follows in cutting off
heads, the hydra is eventually reduced to a set of heads (there may be
millions of them even if the starting form of the beast is simple) that are
458 COMBINATORICS
(')
o
(')
Figure 34.1.
all joined directly to the root. They are then eliminated one by one
until the hydra expires from lack of heads.
A useful way to approach the hydra game is to think of the tree as
modeling a set of nested boxes. Each box contains all the boxes reached
by moving upward on the tree, and it is labeled with the maximum
number of levels of nesting that it contains. Thus in the first figure of
the hydra the root is a box of rank 4. Immediately above it on the left
Bulgarian Solitaire and Other Seemingly Endless Tasks 459
a
b
Figure 34.2.
is a 3 box and on the right is a 2 box, and so on. All end points are
empty boxes of rank o. Each time a 0 box (hydra head) is removed the
box immediately below gets duplicated (along with all its contents), but
each of the duplicates as well as the original box now contains one
fewer empty box. Eventually you are forced to start reducing ranks of
boxes, like the ranks of balls in the ball game. An inductive argument
similar to Smullyan's will show that ultimately all boxes become
empty, after which they are removed one at a time.
lowe this approach to Dershowitz, who pointed out that it is not
even necessary for the hydra to limit its growth to a consecutively in-
creasing number of new branches. After each chop as many finite du-
plicates as you like may be allowed to sprout. It may take Hercules
much longer to slay the monster, but there is no way he can perma-
nently avoid doing so if he keeps hacking away. Note that the hydra
never gets taller as it widens. Some of the more complicated growth
programs considered by Dershowitz and Manna graph as trees that can
grow taller as well as wider, and such trees are even harder to prove ter-
minating.
Our next example of a task that looks as if it could go on forever when
it really cannot is known as the 18-point problem. You begin with a
line segment. Place a point anywhere you like on it. Now place a second
point so that each of the two points is within a different half of the line
segment. (The halves are taken to be "closed intervals," which means
460 COMBINATORICS
1/6115 114 1/3 1/2
34.3.
461
Warmus, a Warsaw mathematician, did not publish his shorter proof
until six years later in the same journal (Vol. 8, No.3, August 1976, pp.
260-63). He gives a 17-point solution, and he adds that there are 768
patterns for such a solution, or 1,536 if you count their reversals as
being different.
Our last example of a task that ends suddenly in a counterintuitive
way is one you will enjoy modeling with a deck of playing cards. Its ori-
gin is unknown, but Graham, who told me about it, says that European
mathematicians call it Bulgarian solitaire for reasons he has not been
able to discover. Partial sums of the series 1 + 2 + 3 + ... are known as
triangular numbers because they correspond to triangular arrays such
as the 10 bowling pins or the 15 pool balls. The task involves any tri-
angular number of playing cards. The largest number you can get from
a standard deck is 45, the sum of the first nine counting numbers.
Form a pile of 45 cards, then divide it into as many piles as you like,
with an arbitrary number of cards in each pile. You may leave it as a
single pile of 45, or cut it into two, three, or more piles, cutting any-
where you want, including 44 cuts to make 45 piles of one card each.
Now keep repeating the following procedure. Take one card from each
pile and place all the removed cards on the table to make a new pile.
The piles need not be in a row. Just put them anywhere. Repeat the pro-
cedure to form another pile, and keep doing it.
As the structure of the piles keeps changing in irregular ways it
seems unlikely you will reach a state where there will be just one pile
with one card, one pile with two cards, one with three, and so on to one
with nine cards. If you should reach this improbable state, without get-
ting trapped in loops that keep returning the game to a previous state,
the game must end, because now the state cannot change. Repeating the
procedure leaves the cards in exactly the same consecutive state as be-
fore. It turns out, surprisingly, that regardless of the initial state of the
game, you are sure to reach the consecutive state in a finite number of
moves.
Bulgarian solitare is a way of modeling some problems in partition
theory that are far from trivial. The partitions of a counting number n
are all the ways a positive integer can be expressed as the sum of pos-
itive integers without regard to their order. For example, the triangular
number 3 has three partitions: 1 + 2, 1 + 1, and 3. When you divide a
packet of cards into an arbitrary number of piles, any number to a pile,
you are forming a partition of the packet. Bulgarian solitaire is a way of
462 COMBINATORICS
changing one partition to another by subtracting 1 from each number
in the partition, then adding a number equal to the number of sub-
tracted l's. It is not obvious this procedure always gives rise to a chain
of partitions, without duplicates, that ends with the consecutive parti-
tion. I am told it was first proved in 1981 by J0rgen Brandt, a Danish
mathematician, but I do not know his proof or whether it has been
published.
Bulgarian solitaire for any triangular number of cards can be dia-
grammed as a tree with the consecutive partition labeling its root and
all other partitions represented by the tree's points. The picture at the
left in Figure 34.4 shows the simple tree for the three-card game. In the
picture at the right of the illustration is the less trivial tree for the 11
partitions of six cards. The theorem that any game ends with the con-
secutive partition is equivalent to the theorem that all the partitions of
a triangular number will graph as a connected tree, with each partition
one step above its successor in the game and the consecutive partition
at the tree's root.
1,1,2,2 (WORST CASE)
1,1,1
3
1,2
Figure 34.4.
Note that the highest point on the six-card tree is six steps from the
root. This partition, 1,1,2,2, is the "worst" starting case. It is easy to see
that the game must end in no more than six steps from any starting par-
tition. It has been conjectured that any game must end in no more than
k(k - 1) steps, where k is any positive integer in the formula for trian-
gular numbers lIzk(k + 1). The computer scientist Donald E. Knuth
Bulgarian Solitaire and Other Seemingly Endless Tasks 463
asked his Stanford University students to test the conjecture by com-
puter. They confirmed it for k = 10 or less, so that the conjecture is al-
most certainly true, but so far a proof has been elusive.
Figure 34.5 shows the tree for Bulgarian solitaire with 10 cards (k =
4). There are now three worst cases at the top, each 12 steps from the
root. Note also that the tree has 14 end points. We can call them Eden
partitions because unless you start with them they never arise in a
game. They are all those partitions whose number of parts exceeds the
highest number of parts by 2 or more.
The picture at the left in Figure 34.6 shows the standard way of using
dots to diagram partition 1,1,2,3,3, at the tree's top. If this pattern is ro-
A
1.2.2.2.3 1.1.1.2.2.3 1.1.2,3,3
B
1,2.2,5
C 1,1,1,1,1,1,2,2
1,1,4,4
D
3,3,4
E 1,1,1,1,3,3
2,2,3,3
1,1,1,1,1,1.4
1,1,8
F 1,1,1,1.1,1,1,1,1,1
1,1,2,2,4
F
1,1,3,5
C
2,4,4
1,1,1,1,2,2,2 D
1,3,3,3
C 1,1,1,1,2,4
2,2,2,4
Figure 34.S. 1,2,3,4
1,1.1,1,1,1.1,1.2
1,0
1,1,1,1,1.1,1,3
1,1,1,1,1,2,3
464 COMBINATORICS
1.
1.
2
3
3
Figure 34.6.
2
3
5
tated and mirror-reflected, it becomes the pattern at the right in the il-
lustration.lts rows now give the partition 2,3,5. Each partition is called
the conjugate of the other. The relation is obviously symmetrical. A
partition unchanged by conjugation is said to be self-conjugate. On the
10 tree there are just two such partitions, the root and 1,1,1,2,5. When
the remaining partitions are paired as conjugates, an amazing pattern
appears along the trunk. The partitions pair as is shown by the letters.
This symmetry holds along the main trunk of all Bulgarian trees so far
investigated.
If the symmetry holds for all such trees, we have a simple way to de-
termine the worst case at the top. It is the conjugate of the partition (there
is always only one) just above the root. An even faster way to find the
trunk's top is to prefix 1 to the root and diminish its last number by 1.
The Bulgarian operation can be diagrammed by removing the left-
most column of its flush-left dot pattern, turning the column 90 de-
grees, and adding it as a new row. Only diagrams of the 1, 2, 3,4, ...
form are unaltered by this. If you could show that no sequence of oper-
ations on any partition other than the consecutive one would return a
diagram to its original state, you would have proved that all Bulgarian
games graph as trees and therefore must end when their root is reached.
If the game is played with 55 cards (k = 10), there are 451,276 ways
to partition them, so that drawing a tree would be difficult. Even the 15-
card tree, with 176 points, calls for computer aid. How are these num-
bers calculated? Well, it is a long and fascinating story. Let us say
partitions are ordered, so that 3, for example, would have four ordered
partitions (usually called "compositions"): 1 + 2, 2 + 1, 1 + 1 + 1, and
3. It turns out that the formula for the total number of compositions is
simply 2n-l. But when the partitions are unordered, as they are in the
solitaire card game, the situation is unbelievably disheveled. Although
there are many recursive procedures for counting unordered partitions,
using at each step the number of known partitions for all smaller num-
bers, an exact asymptotic formula was finally obtained. The big break-
through was made by the British mathematician G. H. Hardy, working
Bulgarian Solitaire and Other Seemingly Endless Tasks 465
with his Indian friend Srinivasa Ramanujan. Their not quite exact for-
mula was perfected by Hans A. Rademacher in 1937. The Hardy-
Ramanujan-Rademacher formula is a horribly shaggy infinite series
that involves (among other things) pi, square roots, complex roots, and
derivatives of hyperbolic functions! George E. Andrews, in his standard
textbook on partition theory, calls it an "unbelievable identity" and
"one of the crowning achievements" in the history of his subject.
The sequence of partitions for n = 1, n = 2, n = 3, n = 4, n = 5, and n
= 6 is 1,2,3,5,7,11, and so you might expect the next partition to be the
next prime, 13. Alas, it is 15. Maybe all partitions are odd. No, the next
partition is 22. One of the deep unsolved problems in partition theory
is whether, as n increases, the even and odd partitions approach equal-
ity in number.
If you think partition theory is little more than a mathematical pas-
time, let me close by saying that a way of diagramming sets of parti-
tions, using number arrays known as the Young tableaux, has become
enormously useful in particle physics. But that's another ball game.
Addendum
Many readers sent proofs of the conjecture that Bulgarian soli-
taire must end in k(k - 1) steps, and the proof was later given in several
articles listed in the bibliography. Ethan Akin and Morton Davis began
their 1983 paper as follows:
Blast Martin Gardner! There you are, minding your own business, and
Scientific American comes along like a virus. All else forgotten, you
must struggle with infection by one of his fascinating problems. In the
August 1983 issue he introduced us to Bulgarian solitaire.
Bibliography
E. Akin and M. Davis, "Bulgarian Solitaire," American Mathematical Monthly, Vol.
92, April 1985, pp. 237-50.
G. E. Andrews, Number Theory: The Theory of Partitions, Addison-Wesley Pub-
lishing Co., 1976.
T. Bending, "Bulgarian Solitaire," Eureka, No. 50, April 1990, pp. 12-19.
J. Brandt, "Cycles of Partitions," Proceedings of the American Mathematical Soci-
ety, Vol. 85, July 1982, pp. 483-86.
M. Carver, "Hercules Hammers Hydra Herd," Discover, November 1987, pp. 94-95,
104.
466 COMBINATORICS
N. Dershowitz and Z. Manna, "Proving Termination with Multiset Orderings,"
Communications of the ACM, Vol. 22, No.8, August 1979, pp. 465-76.
G. Etienne, "Tableaux de Young et Solitaire Bulgare," The Journal of Combinator-
ial Theory, Ser. A, Vol. 58, November 1991, pp. 181-97.
R. W. Hamming, "The Tennis Ball Paradox," Mathematics Magazine, Vol. 62, Oc-
tober 1989, pp. 268-73.
K. Igusa, "Solution of the Bulgarian Solitaire Conjecture," Mathematics Magazine,
Vol. 58, November 1985, pp. 259-71.
L. Kirby and J. Paris, "Accessible Independence Results for Peano Arithmetic,"
The Bulletin of the London Mathematical Society, Vol. 14, No. 49, Part 4, July
1983, pp. 285-93.
C. Si. J. A. Nash-Williams, "On Well-quasi-ordering Finite Trees," Proceedings of
the Cambridge Philosophical Society, Vol. 59, Part 4, October 1963, pp. 833-35.
A. Nicholson, "Bulgarian Solitaire," Mathematics Teacher, Vol. 86, January 1993,
pp.84-86.
J. Propp, "Some Variants of Ferrier Diagrams," Journal of Combinatorial Theory,
Ser. A, Vol. 52, September 1989, pp. 98-128.
R. M. Smullyan, "Trees and Ball Games," Annals of the New York Academy of Sci-
ences, Vol. 321, 1979, pp. 86-90.
Bulgarian Solitaire and Other Seemingly Endless Tasks 467
IX
Qamesand
Decision Theory
Chapter 35 A Matchbox
Qame-Learning Machine
I knew little of chess, but as only a few
pieces were on the board, it was obvious that
the game was near its close . ... [Moxon's]
face was ghastly white, and his eyes glittered
like diamonds. Of bis antagonist I had only
a back view, but that was sufficient; I should
not have cared to see his face.
The quotation is from Ambrose Bierce's classic robot story,
"Moxon's Master" (reprinted in Groff Conklin's science-fiction anthol-
ogy, Thinking Machines). The inventor Moxon has constructed a chess-
playing robot. Moxon wins a game. The robot strangles him.
Bierce's story reflects a fear that someday computers will develop a
will of their own. Let it not be thought that this wOITies only those who
do not understand computers. Before his death Norbert Wiener antici-
pated with increasing apprehension the day when complex govern-
ment decisions would be turned over to sophisticated game-theory
machines. Before we know it, Wiener warned, the machines may shove
us over the brink into a suicidal war.
The greatest threat of unpredictable behavior comes from the learn-
ing machines: computers that improve with experience. Such machines
do not do what they have been told to do but what they have learned
to do. They quickly reach a point at which the programmer no longer
knows what kinds of circuits his machine contains. Inside most of
these computers are randomizing devices. If the device is based on the
random decay of atoms in a sample radioactive material, the machine's
behavior is not (most physicists believe) predictable even in principle.
Much of the research on learning machines has to do with comput-
ers that steadily improve their ability to play games. Some of the work
is secret-war is a game. The first significant machine of this type was
an IBM 704 computer programed by Arthur L. Samuel of the IBM re-
search department at Poughkeepsie, NY. In 1959 Samuel set up the
computer so that it not only played a fair game of checkers but also was
capable of looking over its past games and modifying its strategy in the
light of this experience. At first Samuel found it easy to beat his ma-
471
chine. Instead of strangling him, the machine improved rapidly, soon
reaching the point at which it could clobber its inventor in every game.
So far as I know no similar program has yet been designed for chess, al-
though there have been many ingenious programs for nonlearning
chess machines.
In 1960 the Russian chess grandmaster Mikhail Botvinnik was
quoted as saying that the day would come when a computer would
play master chess. "This is of course nonsense," wrote the American
chess expert Edward Lasker in an article on chess machines in the Fall
1961 issue of a magazine called The American Chess Quarterly. But it
was Lasker who was talking nonsense. A chess computer has three
enormous advantages over a human opponent: (1) it never makes a
careless mistake; (2) it can analyze moves ahead at a speed much faster
than a human player can; (3) it can improve its skill without limit.
There is every reason to expect that a chess-learning machine, after
playing thousands of games with experts, will someday develop the
skill of a master. It is even possible to program a chess machine to play
continuously and furiously against itself. Its speed would enable it to
acquire in a short time an experience far beyond that of any human
player.
It is not necessary for the reader who would like to experiment with
game-learning machines to buy an electronic computer. It is only nec-
essary to obtain a supply of empty matchboxes and colored beads. This
method of building a simple learning machine is the happy invention
of Donald Michie, a biologist at the University of Edinburgh. Writing on
"Trial and Error" in Penguin Science Survey 1961, VoL 2, Michie de-
scribes a ticktacktoe learning machine called MENACE (Matchbox Ed-
ucable Naughts And Crosses Engine) that he constructed with 300
matchboxes.
MENACE is delightfully simple in operation. On each box is pasted
a drawing of a possible ticktacktoe position. The machine always
makes the first move, so only patterns that confront the machine on odd
moves are required. Inside each box are small glass beads of various
colors, each color indicating a possible machine play. A V-shaped card-
board fence is glued to the bottom of each box, so that when one shakes
the box and tilts it, the beads roll into the V. Chance determines the
color of the bead that rolls into the V's corner. First-move boxes contain
four beads of each color, third-move boxes contain three beads of each
472 GAMES AND DECISION THEORY
color, fifth-move boxes have two beads of each color, seventh-move
boxes have single beads of each color.
The robot's move is determined by shaking and tilting a box, open-
ing the drawer, and noting the color of the "apical" bead (the bead in
the V's apex). Boxes involved in a game are left open until the game
ends. If the machine wins, it is rewarded by adding three beads of the
apical color to each open box. If the game is a draw, the reward is one
bead per box. If the machine loses, it is punished by extracting the api-
cal bead from each open box. This system of reward and punishment
closely parallels the way in which animals and even humans are taught
and disciplined. It is obvious that the more games MENACE plays, the
more it will tend to adopt winning lines of play and shun losing lines.
This makes it a legitimate learning machine, although of an extremely
simple sort. It does not make (as does Samuel's checker machine) any
self-analysis of past plays that causes it to devise new strategies.
Michie's first tournament with :MENACE consisted of 220 games over
a two-day period. At first the machine was easily trounced. After 17
games the machine had abandoned all openings except the comer
opening. After the twentieth game it was drawing consistently, so
Michie began trying unsound variations in the hope of trapping it in a
defeat. This paid off until the machine learned to cope with all such
variations. When Michie withdrew from the contest after losing eight
out of ten games, MENACE had become a master player.
Since few readers are likely to attempt building a learning machine
that requires 300 matchboxes, I have designed hexapawn, a much sim-
pler game that requires only 24 boxes. The game is easily analyzed-
indeed, it is trivial-but the reader is urged not to analyze it. It is much
more fun to build the machine, then learn to play the game while the
machine is also learning.
Hexapawn is played on a 3 x 3 board, with three chess pawns on each
side as shown in Figure 35.1. Dimes and pennies can be used instead
of actual chess pieces. Only two types of move are allowed: (1) A pawn
may advance straight forward one square to an empty square; (2) a
pawn may capture an enemy pawn by moving one square diagonally,
left or right, to a square occupied by the enemy. The captured piece is
removed from the board. These are the same as pawn moves in chess,
except that no double move, en passant capture, or promotion of pawns
is permitted.
A Matchbox Game-Learning Machine 473
35.1. The game of nex:apl:lWD
won in any of three ways:
1. a row.
By capturing all enemy pieces.
a position in which the move.
Players moves, one piece at a time. A draw clearly
impossible, but it is not inunediately apparent first or
Robot) need 24 empty
matchboxes and a supply colored beads. Small candies that Come in
colors-M&M's or Skittles-work nicely.
one
move. Patterns positions open to HER on
have a choice between a center or an end open- the
end is considered right
would
play. Patterns marked eleven positions that can confront
on fourth (its second) move. Patterns are the
eleven last) move. (I
included the
otherwise 19 would suffice.)
box place a single bead to ... .A-LY. ........ ..LL arrow
now move
can therefore make all possible moves
moves. The robot has no strategy. In an idiot.
as follows.
on
close your eyes, open the drawer, remove one bead. Close the
box, place
.......................... "'" arrow move ac-
it to move Continue this procedure
..... . A ~ ~ ends. If the robot wins, replace play
474 AND
6
6
arrows rePlresent
475
GAMES
lS.l. ..... .................... ........,1='. curve HER's 50 a ........................ a
476 AND
35.4. Matchbox AU ............................... ,;;, can not
LA ............. U ...... u' ..... .,.... ... ...., ........ ...... "' .... A'I"\ ma-
.......................... 'V ... , at
477
Westminster College. Salt Lake City, UT, made good use of HER as a
concession at a college carnival. The machine was designed to learn by
rewards only, so that customers would always have a chance (though
a decreasing one) of winning. and prizes to winners were increased in
value as HER became more proficient.
Several readers built two matchbox machines to be pitted against
each other. John Chambers, Toronto, called his pair THEM (Two-way
Hexapawn Educable Machines). Kenneth W. Wiszowaty. science
teacher at Phillip Rogers Elementary School, Chicago, sent me a report
by his seventh-grade pupil, Andrea Weiland, on her two machines
which played against each other until one of them learned to win every
time. John House, Waterville, OH, called his second machine RAT (Re-
lentless Autolearning Tyrant), and reported that after 18 games RAT
conceded that HER would win all subsequent games.
Peter J. Sandiford, director of operations research for Trans-Canada
Air Lines, Montreal, called his machines Mark I and Mark II. As ex-
pected, it took 18 games for Mark I to learn how to win every time and
Mark II to learn how to fight the longest delaying action. Sandiford
then devised a devilish plan. He arranged for two students, a boy and
a girl from a local high school mathematics club, who knew nothing
about the game, to play hexapawn against each other after reading a
handout describing the rules. "Each contestant was alone in a room,"
writes Sandiford, "and indicated his moves to a referee. Unknown to
the players the referees reported to a third room containing the jelly-
bean computers and scorekeepers. The players thought they were play-
ing each other by remote control, so to speak, whereas they were in fact
playing independently against the computers. They played alternately
black and white in successive games. With much confusion and muf-
tled hilarity we in the middle tried to operate the computers, keep the
games in phase, and keep the score."
The students were asked to make running comments on their own
moves and those of their opponent. Some sample remarks:
"It's the safest thing to do without being captured; it's almost sure to
win."
"He took me, but I took him too. If he does what I expect, he'll take
my pawn, but in the next move I'll block him."
"Am I stupid!"
"Good move! I think I'm beat."
478 GAMES AND DECISION THEORY
"I don't think he's really thinking. By now he shouldn't make any
more careless mistakes."
"Good game. She's getting wise to my action now."
"Now that he's thinking, there's more competition."
"Very surprising move ... couldn't he see I'd win if he moved for-
ward?"
"My opponent played well. I guess I just got the knack of it first."
When the students were later brought face to face with the machines
they had been playing, they could hardly believe, writes Sandiford,
that they had not been competing against a real person.
Richard L. Sites, at M.LT., wrote a FORTRAN program for an IBM
1620 so that it would learn to play Octapawn, a 4 x 4 version of hexa-
pawn that begins with four white pawns on the first row and four black
pawns on the fourth row. He reports that the first player has a sure win
with a corner opening. At the time of his writing, his program had not
yet explored center openings.
Judy Gomberg, Maplewood, NJ, after playing against a matchbox ma-
chine that she built, reported that she learned hexapawn faster than her
machine because "every time it lost I took out a candy and ate it."
Robert A. Ellis, at the computing laboratory, Ballistics Research Lab-
oratories, Aberdeen Proving Ground, MD, told me about a program he
wrote for a digital computer which applied the matchbox-learning tech-
nique to a ticktacktoe-learning machine. The machine first plays a stu-
pid game, choosing moves at random, and is easily trounced by human
opponents. Then the machine is allowed to play 2,000 games against it-
self (which it does in two or three minutes), learning as it goes. After
that, the machine plays an excellent strategy against human opponents.
My defense of Botvinnik's remark that computers will some day
play master chess brought a number of irate letters from chess players.
One grandmaster assured me that Botvinnik was speaking with tongue
in cheek. The interested reader can judge for himself by reading a
translation of Botvinnik's speech (which originally appeared in Kom-
somolskaya Pravda, January 3,1961) in The Best in Chess, edited by
L A. Horowitz and Jack Straley Battell (New York: Dutton, 1965, pp.
63-69). "The time will corne," Botvinnik concludes, "when mechani-
cal chessplayers will be awarded the title of International Grandmas-
ter ... and it will be necessary to promote two world championships,
one for humans, one for robots. The latter tournament, naturally, will
A Matchbox Game-Learning Machine 479
not be between machines, but between their makers and program op-
erators."
An excellent science-fiction story about just such a tournament, Fritz
Leiber's "The 64-Square Madhouse," appeared in If, May 1962, and
was reprinted in Leiber's A Pail of Air (New York: Ballantine, 1964).
Lord Dunsany, by the way, has twice given memorable descriptions of
chess games played against computers. In his short story "The Three
Sailors' Gambit" (in The Last Book of Wonder) the machine is a magic
crystal. In his novel The Last Revolution (a 1951 novel about the com-
puter revolution that has never, unaccountably, been published in the
United States) it is a learning computer. The description of the narra-
tor's first game with the computer, in the second chapter, is surely one
of the funniest accounts of a chess game ever written.
The hostile reaction of master chess players to the suggestion that
computers will someday play master chess is easy to understand; it
has been well analyzed by Paul Armer in a Rand report (p-2114-2, June
1962) on Attitudes Toward Intelligent Machines. The reaction of chess
players is particularly amusing. One can make out a good case against
computers writing top-quality music or poetry, or painting great art, but
chess is not essentially different from ticktacktoe except in its enor-
mous complexity, and learning to play it well is precisely the sort of
thing computers can be expected to do best.
Master checker-playing machines will undoubtedly come first.
Checkers is now so thoroughly explored that games between champi-
ons almost always end in draws, and in order to add interest to such
games, the first three moves are now chosen by chance. Richard Bell-
man, writing "On the Application of Dynamic Programming to the De-
termination of Optimal Play in Chess and Checkers," Proceedings of the
National Academy of Sciences (Vol. 53, February 1965, pp. 244-47),
says that "it seems safe to predict that within ten years checkers will be
a completely decidable game."
Chess is, of course, of a different order of complexity. One suspects
it will be a long time before one can (so goes an old joke in modern
dress) play the first move of a chess game against a computer and have
the computer print, after a period of furious calculation, "I resign." In
1958 some responsible mathematicians predicted that within 10 years
computers would be playing master chess, but this proved to be wildly
overoptimistic. Tigran Petrosian, when he became world chess cham-
pion, was quoted in The New York TImes (May 24,1963) as expressing
480 GAMES AND DECISION THEORY
doubts that computers would play master chess within the next 15 or
20 years.
Hexapawn can be extended simply by making the board wider but
keeping it three rows deep. John R. Brown, in his paper "Extenda-
pawn-An Inductive Analysis" (Mathematics Magazine, Vol. 38, No-
vember 1965, pp. 286-99) gives a complete analysis of this game. If n
is the number of columns, the game is a win for the first player if the
final digit of n is 1, 4, 5, 7, or 8. Otherwise the second player has the
win.
My hexapawn game was marketed in two forms. In 1969 IBM
adapted it to a board that used a spinner to choose one of four colors
and buttons to place on game positions to reward or punish the prim-
itive "learning machine." Titled "Hexapawn: A Game You Play to
Lose," it was distributed by IBM to high school science and mathe-
matics classes, and to the general public. In 1970 Gabriella, a Farm-
ingdale, NY, firm, produced and sold a similar product they called the
Gabriella Computer Kit. See Science News (October 26, 1970) for an ad-
vertisement. The game was also described in the first chapter of We
Build Our Own Computers, a projects handbook by A. B. Bolt and oth-
ers (Cambridge University Press, 1966).
Computers that learn from experience may offer the most promising
way to simulate the human mind. Based on what are called neural net-
works, they rely heavily on the parallel processing of data. The neural
network literature is so vast that I made no attempt to cite references.
Our brain is, of course, an enormously complicated, parallel-processing
learning machine, the operation of which is still only dimly under-
stood.
Computer chess has made rapid progress since this chapter was pub-
lished as a Scientific American column in 1962. Programs now play on
the master level, and on occasion have defeated grand masters. The
best programs do not learn from experience, but owe their skill to the
fantastic speed with which they explore long sequences of moves. It is
only a matter of time before chess programs achieve grand master rat-
Ings.
A proof of first player win on the 3 x 3 Go field is given in the 1972
paper by Edward Thorp and William Walden. As far as I know, my 5 x
5 minichess has not been solved in the sense that it is not known
whether the first or second player has the win when the game is played
rationally, or whether the game is a draw. It is possible that many chess
A Matchbox Game-Learning Machine 481
computer programs would be able to crack it if someone only bothered
to give them the game as a problem.
Another curious form of minichess was created by a young woman
named Chi Chi Hackenberg, whose invention was featured in Eye (No-
vember 1968, pp. 93-94). It uses a field of 4 x 8, or half the standard
chessboard.
At the start, all eight white pieces are on the first row, in their usual
formation. The second row has five white pawns, three pawns missing
on the king's bishop and knight columns and the queen's knight col-
umn. The 13 black pieces are in mirror-image formation on the other
two rows, making six vacant cells on the board. There are two new
rules. White is not allowed to move a pawn on its opening move, and
pawns are permitted to move vertically backward either one step or to
capture diagonally backward.
Hackenberg says that she thinks White's best opening is to take the
rook's pawn with its king's knight, threatening checkmate with KB to
N2. However, Black can mount a counteroffensive by taking the white
queen's pawn with its queen's bishop's pawn for the first in a series of
checks that include a capture of White's queen. This seems so disas-
trous for White that the opening suggested by Hackenberg may not be
the best after all. Or can White, after the checks subside, mount a viable
counterattack? As far as I know, this miniversion of chess also has not
been exhaustively analyzed.
Richard Bellman's prediction that checkers would be solved by the
end of 1975-that is, it would become known if the first or second
player had the win or the game is a draw-proved wide of the mark. As
I write now (in 2000), checkers is far from decided, although checker
programs today play on the same level as the best players.
Mikhail Botvinnik accurately predicted that computers would soon
reach grand master level and that tournaments would be held between
computer programs. Such contests are now annual events. In 1997
Deep Blue, IBM's chess program, defeated Gary Kasparov, the world's
chess champion, in a six-game contest in Manhattan. Kasparov won
only one game, tied three, and lost two.
There is now an enormous literature on the designing of computers
that learn from experience and on the construction of better and better
chess programs. I have made no attempt to introduce this literature in
my brief bibliography.
482 GAMES AND DECISION THEORY
A B
c D
Bibliography
W. R. Ashby, "Can a Mechanical Chess-player Outplay Its Designer?" British Jour-
nal for the Philosophy of Science, Vol. 3, No. 44, (1952).
A. Bernstein and M. de V. Roberts, "Computer v. Chess-player," Scientific Ameri-
can, June 1958, pp. 96-105.
A. Newell, J. C. Shaw, and H. A. Simon, "Chess-playing Programs and the Problem
of Complexity," IBM Journal of Research and Development, Vol. 2, No.4, Octo-
ber 1958, pp. 320-35.
A. Newell and H. A. Simon, "Computer Simulation of Human Thinking," Science,
Vol. 134, 1961,pp.2011-17.
N. Nilsson, Learning Machines, New York: McGraw-Hill, 1965.
A. L. Samuel, "Some Studies in Machine Learning, Using the Game of Checkers."
IBMJoumal of Research and Development, VoL 3, No.3, July 1959, pp. 210-29.
C. E. Shannon, "Game Playing Machines," Journal of the Franklin Institute, Vol.
260, No.6, December 1955, pp. 447-53.
J. M. Smith and D. Michie, "Machines That Play Games," The New Scientist, No.
260, November 9,1961, pp. 367-69.
484 GAMES AND DECISION THEORY
Chapter 36
I made sprouts fontaneously. ...
-JAMES JOYCE, Finnegans Wake, p. 542
Sprouts and
Brussels Sprouts
1171
.n friend of mine, a classics student at Cambridge, introduced
me recently to a game called 'Sprouts' which became a craze at Cam-
bridge last term. The game has a curious topological flavor."
So began a letter I received in 1967 from David Hartshorne, then a
mathematics student at the University of Leeds. Soon other British
readers were writing to me about this amusing pencil-and-paper game
that had sprouted suddenly on the Cambridge grounds.
I am pleased to report that I successfully traced the origin of this
game to its source: the joint creative efforts of John Horton Conway,
then a teacher of mathematics at Sidney Sussex College, Cambridge,
and Michael Stewart Paterson, then a graduate student working at Cam-
bridge on abstract computer programming theory.
The game begins with n spots on a sheet of paper. Even with as few
as three spots, Sprouts is more difficult to analyze than ticktacktoe, so
that it is best for beginners to play with no more than three or four ini-
tial spots. A move consists of drawing a line that joins one spot to an-
other or to itself and then placing a new spot anywhere along the line.
These restrictions must be observed:
1. The line may have any shape but it must not cross itself, cross a pre-
viously drawn line, or pass through a previously made spot.
2. No spot may have more than three lines emanating from it.
Players take turns drawing curves. In normal Sprouts, the recom-
mended form of play, the winner is the last person able to play. As in
Nim and other games of the "take-away" type, the game can also be
played in "misere" form, a French term that applies to a variety of card
games in the whist family in which one tries to avoid taking tricks. In
misere Sprouts the first person unable to play is the winner.
485
The typical three-spot normal game shown in Figure 36.1 was won
on the seventh move by the first player. It is easy to see how the game
got its name, for it sprouts into fantastic patterns as the game pro-
gresses. The most delightful feature is that it is not merely a combina-
torial game, as so many connect-the-dots games are, but one that
actually exploits the topological properties of the plane. In more tech-
nicallanguage, it makes use of the Jordan-curve theorem, which asserts
that simple closed curves divide the plane into outside and inside re-
gions.
One might guess at first that a Sprouts game could keep sprouting for-
START




7
Figure 36.1. A typical game of three-spot Sprouts
486 GAMES AND DECISION THEORY
ever, but Conway offers a simple proof that it must end in at most 3n -
1 moves. Each spot has three "lives"-the three lines that may meet at
that point. A spot that acquires three lines is called a "dead spot" be-
cause no more lines can be drawn to it. A game that begins with n spots
has a starting life of 3n. Each move kills two lives, at the beginning and
at the end of the curve, but adds a new spot with a life of 1. Each move
therefore decreases the total life of the game by 1. A game obviously
cannot continue when only one life remains, since it requires at least
two lives to make a move. Accordingly no game can last beyond 3n -
1 moves. It is also easy to show that every game must last at least 2n
moves. The three-spot game starts with nine lives, must end on or be-
fore the eighth move, and must last at least six moves.
The one-spot game is trivial. The first player has only one possible
move: connecting the spot to itself. The second player wins in the nor-
mal game (loses in miserel by joining the two spots, either inside or out-
side the closed curve. These two second moves are equivalent, as far as
playing the game is concerned, because before they are made there is
nothing to distinguish the inside from the outside of the closed curve.
Think of the game as being played on the surface of a sphere. If we
puncture the surface by a hole inside a closed curve, we can stretch the
surface into a plane so that all points previously outside the curve be-
come inside, and vice versa. This topological equivalence of inside and
outside is important to bear in mind because it greatly simplifies the
analysis of games that begin with more than two spots.
With two initial spots, Sprouts immediately takes on interest. The
first player seems to have a choice of five opening moves (see Figure
36.2), but the second and third openings are equivalent for reasons of
symmetry. The same holds true of the fourth and fifth, and in light of
the inside-outside equivalence just explained, all four of these moves
can be considered identical. Only two topologically distinct moves,
therefore, require exploring. It is not difficult to diagram a complete
START
A B

A B
-
A B
\)'
A B
'()
Figure 36.2. Initial spots (A and B) and first player's possible opening moves in two-spot
game
Sprouts and Brussels Sprouts 487
tree chart of all possible moves, inspection of which shows that in both
normal and misere forms of the two-spot game the second player can
always win.
Conway found that the first player can always win the normal three-
spot game and the second player can always win the misere version.
Denis P. Mollison has shown that the first player has the win in normal
four- and five-spot games. In response to a 10-shilling bet made with
Conway that he could not complete his analysis within a month, Mol-
lison produced a 49-page proof that the second player wins the normal
form of the six-spot game. The second player wins the misere four-spot
game.
Although no strategy for perfect play has been formulated, one can
often see toward the end of a game how to draw closed curves that will
divide the plane into regions in such a way as to lead to a win. It is the
possibility of this kind of planning that makes Sprouts an intellectual
challenge and enables a player to improve his skill at the game. But
Sprouts is filled with unexpected growth patterns, and there seems to
be no general strategy that one can adopt to make sure of winning.
Sprouts was invented on the afternoon of Thesday, February 21,
1967, when Conway and Paterson had finished having tea in the math-
ematics department's common room and were doodling on paper in an
effort to devise a new pencil-and-paper game. Conway had been work-
ing on a game invented by Paterson that originally involved the fold-
ing of attached stamps, and Paterson had put it into pencil-and-paper
form. They were thinking of various ways of modifying the rules when
Paterson remarked, "Why not put a new dot on the line?"
"As soon as this rule was adopted," Conway has written me, "all the
other rules were discarded, the starting position was simplified to just
n points, and Sprouts sprouted." The importance of adding the new
spot was so great that all parties concerned agree that credit for the
game should be on a basis of 3/5 to Paterson and 2/5 to Conway. "And
there are complicated rules," Conway adds, "by which we intend to
share any monies which might accrue from the game."
"The day after Sprouts sprouted," Conway continues, "it seemed that
everyone was playing it. At coffee or tea times there were little groups
of people peering over ridiculous to fantastic Sprout positions. Some
people were already attacking Sprouts on toruses, Klein bottles, and the
like, while at least one man was thinking of higher-dimensional ver-
sions. The secretarial staff was not immune; one found the remains of
488 GAMES AND DECISION THEORY
Sprout games in the most unlikely places. Whenever r try to acquaint
somebody new to the game nowadays, it seems he's already heard of it
by some devious route. Even my three- and four-year-old daughters
play it, though r can usually beat them."
The name "Sprouts" was given the game by Conway. An alternative
name, "measles," was proposed by a graduate student because the game
is catching and it breaks out in spots, but Sprouts was the name by
which it quickly became known. Conway later invented a superficially
similar game that he calls "Brussels Sprouts" to suggest that it is a joke.
r shall describe this game but leave to the reader the fun of discovering
why it is a joke before the explanation is given in the answer section.
Brussels Sprouts begins with n crosses instead of spots. A move con-
sists of extending any arm of any cross into a curve that ends at the free
arm of any other cross or the same cross; then a crossbar is drawn any-
where along the curve to create a new cross. Two arms ofthe new cross
will, of course, be dead, since no arm may be used twice. As in Sprouts,
no curve may cross itself or cross a previously drawn curve, nor may
it go through a previously made cross. As in Sprouts, the winner of the
normal game is the last person to play and the winner of the misere
game is the first person who cannot play.
After playing Sprouts, Brussels Sprouts seems at first to be a more
complicated and more sophisticated version. Since each move kills
two crossarms and adds two live crossarms, presumably a game might
never end. Nevertheless, all games do end and there is a concealed
joke that the reader will discover if he succeeds in analyzing the game.
To make the rules clear, a typical normal game of two-cross Brussels
Sprouts is shown that ends with victory for the second player on the
eighth move (see Figure 36.3).
A letter from Conway reports several important breakthroughs in
Sproutology. They involve a concept he calls the "order of moribun-
dity" of a terminal position, and the classification of "zero order" po-
sitions into five basic types: louse, beetle, cockroach, earwig, and
scorpion (see Figure 36.4). The larger insects and arachnids can be in-
fested with lice, sometimes in nested form, and Conway draws one
pattern he says is "merely an inside-out earwig inside an inside-out
louse." Certain patterns, he points out, are much lousier than others.
And there is the FTOZOM (fundamental theorem of zero-order mori-
bundity), which is quite deep. Sproutology is sprouting so rapidly that
r shall have to postpone my next report on it for some time.
Sprouts and Brussels Sprouts 489
START
+
+
Figure 36.3. Typical game of two-cross Brussels Sprouts
Figure 36.4.
Louse Beetle Cockroach Earwig Scorpion
Addendum
Sprouts made an instant hit with Scientific American readers,
many of whom suggested generalizations and variations of the game.
Ralph J. Ryan III proposed replacing each spot with a tiny arrow, ex-
tending from one side of the line and allowing new lines to be drawn
490 GAMES AND DECISION THEORY
only to the arrow's point. Gilbert W. Kessler combined spots and cross-
bars in a game he called "succotash." George P. Richardson investi-
gated Sprouts on the torus and other surfaces. Eric L. Gans considered
a generalization of Brussels Sprouts (called "Belgian Sprouts") in
which spots are replaced by "stars"-n crossbars crossing at the same
point. Vladimir Y gnetovich suggested the rule that a player, on each
turn, has a choice of adding one, two, or no spots to his line.
Several readers questioned the assertion that every game of normal
Sprouts must last at least 2n moves. They sent what they believed to be
counterexamples, but in each case failed to notice that every isolated
spot permits two additional moves.
Since this chapter appeared in Scientific American in 1967, David
Applegate, Guy Jacobson, and Danny Sleator wrote the first computer
program for analyzing Sprouts (see bibliography). They found that in
normal Sprouts the first player wins when n (the number of spots) is 3,
4, 5, 9, 10, and 11. Beyond n = 11 their program w&s unable to cope
with sprouting complexity. They did not go beyond n = 9 for misere
Sprouts. The program proved a first player win when n = 1,5, and 6.
The authors conjecture that in normal Sprouts the first player wins
if n = 3, 4, or 5 (modulo 6) and wins in misere Sprouts if n = a or 1
(modulo 5). They add that Sleator believes both conjectures, but Ap-
plegate disbelieves both. Details about their program can be obtained
from Applegate or Jacobson at Bell Labs, in Murray Hill, NJ, or Sleator
at Carnegie Mellon University.
Answers
Why is the game of Brussels Sprouts, which appears to be a
more sophisticated version of Sprouts, considered a joke by its inven-
tor, John Horton Conway? The answer is that it is impossible to play
Brussels Sprouts either well or poorly because every game must end in
exactly 5n 2 moves, where n is the number of initial crosses. If played
in standard form (the last to play is the winner), the game is always won
by the first player if it starts with an odd number of crosses, by the sec-
ond player if it starts with an even number. (The reverse is true, of
course, in misere play.) After introducing someone to Sprouts, which
is a genuine contest, one can switch to the fake game of Brussels
Sprouts, make bets, and always know in advance who will win every
Sprouts and Brussels Sprouts 491
game. I leave to the reader the task of proving that each game must end
in 5n - 2 moves.
Bibliography
P. Anthony, Macroscopes, Avon, 1969. In this novel by a popular science-fiction
writer, Sprouts is discussed in the first chapter, and games are described and il-
lustrated in several later chapters.
D. Applegate, G. Jacobson, and D. Sleator. "Computer Analysis of Sprouts," The
Mathematician and Pied Puzzler, E. Berlekamp and T. Rodgers (eds.). A.K. Pe-
ters, 1999, pp. 199-201.
M. Baxter. "Unfair Games," Eureka, No. 50, April 1990, pp. 60-65.
E. Berkekamp, J. Conway, and R. Guy, "Sprouts and Brussels Sprouts," Winning
Ways, Vol. 2, Academic Press, 1982, pp. 564-70.
M. Cooper, "Graph Theory and the Game of Sprouts," American Mathematical
Monthly, Vol. 100, May 1993, pp. 478-82.
H. D' Alarcao and T. E. Moore, "Euler's Formula and a Game of Conway's," Journal
o/Recreational Mathematics, Vol. 9, No.4, 1976-1977, pp. 249-51. The authors
prove that Brussels Sprouts must end after 5n - 2 moves.
T. K. Lam, "Connected Sprouts," American Mathematical Monthly, VoL 101, Feb-
ruary 1997, pp. 116-19.
492 GAMES AND DECISION THEORY
Chapter 37 Harary's
Qeneralized
Ticktacktoe
The world's simplest, oldest, and most popular pencil-and-
paper game is still ticktacktoe, and combinatorial mathematicians,
often with the aid of computers, continue to explore unusual varia-
tions and generalizations of it. In one variant that goes back to ancient
times the two players are each given three counters, and they take turns
first placing them on the 3 x 3 board and then moving them from cell
to cell until one player gets his three counters in a row. Moving-counter
ticktacktoe is the basis for a number of modern commercial games,
such as John Scarne's Teeko and a game called Touche, in which con-
cealed magnets cause counters to flip over and become opponent
pieces.
Standard ticktacktoe can obviously be generalized to larger fields.
For example, the old Japanese game of Go-Moku ("five stones") is es-
sentially five-in-a-row ticktacktoe played on a Go board. Another way
to generalize the game is to play it on "boards" of three or more di-
menSIOns.
In March, 1977, Frank Harary devised a delightful new way to gen-
eralize ticktacktoe. Harary was then a mathematician at the University
of Michigan. He is now the Distinguished Professor of Computer Sci-
ence at New Mexico State University, in Las Cruces. He has been called
Mr. Graph Theory because of his tireless, pioneering work in this
rapidly growing field that is partly combinatorial and partly topologi-
cal. Harary is the founder of the Journal of Combinatorial Theory and
the Journal of Graph Theory, and the author of Graph Theory, consid-
ered the world over to be the definitive textbook on the subject. His pa-
pers on graph theory, written alone or in collaboration with others,
number more than 500. Harary ticktacktoe, as I originally called his
generalization of the game, opens up numerous fascinating areas of
493
recreational mathematics. Acting on his emphatic request, I now call it
animal ticktacktoe for reasons we shall see below.
We begin by observing that standard ticktacktoe can be viewed as a
two-color geometric-graph game of the type Harary calls an achieve-
ment game. Replace the nine cells of the ticktacktoe board with nine
points joined by lines, as is shown in Figure 37.1. The players are each
assigned a color, and they take turns coloring points on the graph. The
first player to complete a straight line of three points in his color wins.
This game is clearly isomorphic with standard ticktacktoe. It is well
known to end in a draw if both players make the best possible moves.
Figure 31.1. Ticktacktoe as a two-coloring game
Let us now ask: What is the smallest square on which the first player
can force a win by coloring a straight (non-diagonal) three-point path?
It is easy to show that it is a square of side four. Harary calls this side
length the board number b of the game. It is closely related to the Ram-
sey number of generalized Ramsey graph theory, a number that plays
an important part in the Ramsey games. (Ramsey theory is a field in
which Harary has made notable contributions. It was in a 1972 survey
paper on Ramsey theory that Harary first proposed making a general
study of games played on graphs by coloring the graph edges.) Once we
have determined the value of b we can ask a second question. In how
few moves can the first player win? A little doodling shows that on a
board of side four the first player can force a win in only three moves.
Harary calls this the move number m of the game.
In ticktacktoe a player wins by taking cells that form a straight, order-
3 polyomino that is either edge- or corner-connected. (The corner-
494 GAMES AND DECISION THEORY
1-Ce1l animal b m 4-Cell animals b m
1 1
loser
b m
2 2 7 8
b m
4 3 5 4
3 3 3 5
EI
4 4
Elly
37.2. one u.u.U'-l.,J;::i.lo..I..
J7.3. 12 CLJ..L.I..L.LI.CLLa of five
b m
Loser
Loser
Loser
Loser
loser
Loser
b m
Loser
Loser
Loser
7 10
7 9
6 6
AND
If the animal chosen as the objective of a game is a loser, the second
player can always force a draw, but he can never force a win. The clever
proof of this fact is well known and applies to most ticktacktoelike
games. Assume that the second player has a winning strategy. The first
player can "steal" the strategy by first making an irrelevant opening
move (which can never be a liability) and thereafter playing the win-
ning strategy. This finding contradicts the assumption that the second
player has a winning strategy, and so that assumption must be false.
Hence the second player can never force a win. If the animal is a win-
ner and b is known, we next seek m, the minimum number of moves
in which the game can be won.
For the i-cell animal (the monomino), which is trivially a winner, b
and m are both equal to 1. When, as in this case, m is equal to the num-
ber of cells in the animal, Harary calls the game economical, because a
player can win it without having to take any cell that is not part of the
animal. The game in which the objective is the only 2-cell animal (the
domino) is almost as trivial. It is also economical, with band m both
equal to 2. The games played with the two 3-cell animals (the tromi-
noes) are slightly more difficult to analyze, but the reader can easily
demonstrate that both are economical: for the L-shaped 3-cell animal b
and m are both equal to 3, and for the straight 3-cell animal b equals 4
and m equals 3. This last game is identical with standard ticktacktoe ex-
cept that corner-connected, or diagonal, rows of three cells are not
counted as wins.
It is when we turn to the 4-cell animals (the tetrominoes) that the pro-
ject really becomes interesting. Harary has given each of the five order-
4 animals names, as is shown in Figure 37.2. Readers may enjoy
proving that the band m numbers given in the illustration are correct.
Note that Fatty (the square tetromino) has no such numbers and so is
labeled a loser. It was Andreas R. Blass, then one of Harary's colleagues
at Michigan, who proved that the first player cannot force Fatty on a
field of any size, even on the infinite lattice. Blass's result was the first
surprise of the investigation into animal ticktacktoe. From this finding
it follows at once that any larger animal containing a 2 x 2 square also
is a loser: the second player simply plays to prevent Fatty'S formation.
More generally, any animal that contains a loser of a lower order is it-
self a loser. Harary calls a loser that contains no loser of lower order a
basic loser. Fatty is the smallest basic loser.
The proof that Fatty is a minimal loser is so simple and elegant that
Rarary's Generalized Ticktacktoe 497
498 AND
shown in the illustration. The animals that can
with each
the proof
___ '............. '-P the pattern. In every case
draw the loser on the associ ..
...................... "-'1. to
player can always ............... "'" ...........
the strategy already de-
12
the straight animal of five
was by
the game is limited to an
cells in a horizontal or vertical line (eliminating wins diagonal
can a When diagonal wins
are game is believed to be a as
far as I know that has not yet been proved even for fields larger than the
Go board.
a
force the animal on a
a one I ..................... ..., ......
a
any, is no larger than 15 and its
is the outstanding un-
can
can
determine its board and move
107 are con ..
a basic loser. Therefore since higher-order con-
tain an order-7 said with confidence that there are no
student Geoffrey are, by coincidence,
exactly a dozen winners-half of them econornical-and a dozen basic
losers.
or can a
paper game or a board game. If both players know the full analysis, then
Harary's Generalized 499
depending on the animal chosen either the first player will win or the
second player will force a draw. As in ticktacktoe between inexpert
players, however, if this knowledge is lacking, the game can be enter-
taining. If the animal chosen as the objective of the game is a winner,
the game is best played on a board of side b or b - 1. (Remember that a
square of side b - 1 is the largest board on which the first player can-
not force a win.)
All the variations and generalizations of animal ticktacktoe that have
been considered so far are, as Harary once put it, "Ramseyish." For ex-
ample, one can play the misere, or reverse, form of any game-in
Harary's terminology an avoidance game-in which a player wins by
forcing his opponent to color the chosen animal.
Avoidance games are unusually difficult to analyze. The second
player trivially wins if the animal to be avoided is the monomino. If the
domino is to be avoided, the second player obviously wins on the 2 x
2 square and almost as obviously on the 2 x 3 rectangle.
On a square board of any size the first player can be forced to com-
plete the L-shaped 3-cell animal. Obviously the length of the square's
side must be at least 3 for the game to be meaningful. If the length of
the side is odd, the second player will win if he follows each of his op-
ponent's moves by taking the cell symmetrically opposite the move
with respect to the center of the board. If the first player avoids taking
the center, he will be forced to take it on his last move and so will lose.
If he takes it earlier in the game without losing, the second player
should follow with any safe move. If the first player then bikes the cell
that is symmetrically opposite the second player's move with respect
to the center, the second player should again make a harmless move,
and so on; otherwise he should revert to his former strategy. If the
length of the square's side is even, this type of symmetrical play leads
to a draw, but the second player can still win by applying more com-
plicated tactics.
On square boards the straight 3-cell animal cannot be forced on the
first player. The proof of this fact is a bit difficult, even for the 3 x 3
square, but as a result no larger animal containing the straight 3-cell
species can be forced on any square board. (The situation is analogous
to that of basic losers in animal-achievement games.) Hence among the
4-cell animals only Fatty and Tippy remain as possible nondraws. Fatty
can be shown to be a draw on any square board, but Tippy can be
500 GAMES AND DECISION THEORY
forced on the first player on all square boards of odd side. The complete
analysis of all animal-avoidance games is still in the early stages and
appears to present difficult problems.
Harary has proposed many other nontrivial variants of the basic an-
imal games. For example, the objective of a game can be two or more
different animals. In this case the first player can try to form one ani-
mal and the second player, the other, or both players can try to form ei-
ther one. In addition, achievement and avoidance can be combined in
the same game, and nonrectangular boards can be used. It is possible
to include three or more players in any game, but this twist introduces
coalition play and leads to enormous complexities. The rules can also
be revised to accept corner-connected animals or animals that are both
edge- and corner-connected. At the limit, of course, one could make
any pattern whatsoever the objective of a ticktacktoelike game, but such
broad generalizations usually lead to games that are too complicated to
be interesting.
Another way of generalizing these games is to play them with polyi-
amonds (identical edge-joined equilateral triangles) or polyhexes (iden-
tical edge-joined regular hexagons) respectively on a regular triangular
field or a regular hexagonal field. One could also investigate games
played with these animals on less regular fields. An initial investigation
of triangular forms, by Harary and Heiko Harborth, is listed in the bib-
liography.
The games played with square animals can obviously be extended to
boards of three or more dimensions. For example, the 3-space analogue
of the polyomino is the polycube: n unit cubes joined along faces.
Given a polycube, one could seek b and m numbers based on the small-
est cubical lattice within which the first player can force a win and try
to find all the polycubes that are basic losers. This generalization is al-
most totally unexplored, but see the bibliography for a paper on the
topic by Harary and Michael Weisbach.
As I have mentioned, Blass, now at Pennsylvania University, is one
ofHarary's main collaborators. The others include Exoo, A. Kabell, and
Heiko Harborth, who is investigating games with the triangular and
hexagonal cousins of the square animals. Harary is still planning a book
on achievement and avoidance games in which all these generaliza-
tions of ticktacktoe and many other closely related games will be
explored.
Harary's Generalized Ticktacktoe 501
Addendum
In giving the proof that a second player cannot have the win in
most ticktacktoelike games, I said that if the first player always wins on
a board of a certain size, he also wins on any larger board. This is true
of the square boards with which Harary was concerned, but is not nec-
essarily true when such games are played on arbitrary graphs. A. K.
Austin and C. J. Knight, mathematicians at the University of Sheffield,
in England, sent the following counterexample.
Consider the graph at the left of Figure 37.5, on which three-in-a-row
wins. The first player wins by taking A. The second player has a choice
of taking a point in either the small or the large triangle. Whichever he
chooses, the first player takes a corner point in the other triangle. The
opponent must block the threatened win, then a play in the remaining
corner of the same triangle forces a win.
__ - - - ~ ~ - - - - __ A
~ - - - - - - 4 _ - - - - - - ~ A
Figure 31.5. First player wins on graph at left, but second player can force a draw on en-
larged graph at right.
Now enlarge the "board" by adding two points as shown on the right
in Figure 37.5. The second player can draw by playing at B. If the first
player does not start with A, the second player draws by taking A.
Bibliography
M. Gardner, "Ticktacktoe Games," Wheels, Life, and Other Mathematical Amuse-
ments, W. H. Freeman, 1983, Chapter 5.
S. W. Golomb, Polyominoes, Scribner's, 1965; Revised edition, Princeton Univer-
sity Press, 1994.
F. Harary, Graph Theory, Addison-Wesley, 1969.
F. Harary, "Achieving the Skinny Animal," Eureka, No. 42, Summer 1982, pp.
8-14; Errata, No. 43, Easter, 1983, pp. 5-6.
502 GAMES AND DECISION THEORY
F. Harary, "Is Snaky a Winner?" Combinatorics, Vol. 2, April 1993, pp. 79-82.
Harary offered $100 to the first person who could prove on or before December
31,1995, whether Snaky is a winner. The question remains open.
F. Harary, Achievement and Avoidance Games, in preparation.
F. Harary and H. Harborth, "Extremal Animals," Journal of Combinatorics, Infor-
mation, and System Sciences, 1, 1976, pp. 1-8.
F. Harary and H. Harborth, "Achievement and Avoidance Games with Thiangular
Animals," Journal of Recreational Mathematics, 18, 1985-86, pp. 110-15.
F. Harary and M. Weisbach, "Polycube Achievement Games," Journal of Recre-
ational Mathematics, 15, 1982-83, pp. 241-46.
Rarary's Generalized Ticktacktoe 503
Chapter 38 The New Eleusis
I shall always consider the best guesser
the best Prophet. -CICERO, De Divinatione
Don't never prophesy-onless ye know.
-JAMES RUSSELL LOWELL, The Biglow Papers
In June 1959, I had the privilege of introducing in Scientific
American a remarkable simulation game called Eleusis. The game,
which is played with an ordinary deck of cards, is named for the an-
cient Eleusinian mysteries, religious rites in which initiates learned a
cult's secret rules. Hundreds of ingenious simulation games have been
developed for modeling various aspects of life, but Eleusis is of special
interest to mathematicians and scientists because it provides a model
of induction, the process at the very heart of the scientific method.
Since then Eleusis has evolved into a game so much more exciting to
play than the original version that I feel I owe it to readers to bring them
up to date. I will begin, however, with some history.
Eleusis was invented in 1956 by Robert Abbott of New York, who at
the time was an undergraduate at the University of Colorado. He had
been studying that sudden insight into the solution of a problem that
psychologists sometimes call the "Aha" reaction. Great turning points
in science often hinge on these mysterious intuitive leaps. Eleusis
turned out to be a fascinating simulation of this facet of science, even
though Abbott did not invent it with this in mind. In 1963 Abbott's
complete rules for the game appeared in his book, Abbott's New Card
Games (hardcover, Stein & Day; paperback, Funk & Wagnalls).
Martin D. Kruskal, a distinguished mathematical physicist at Prince-
ton University, became interested in the game and made several im-
portant improvements. In 1962 he published his rules in a monograph
titled Delphi: A Game of Inductive Reasoning. Many college profes-
sors around the country used Eleusis and Delphi to explain scientific
method to students and to model the Aha process. Artificial intelli-
gence scientists wrote computer programs for the game. At the System
Development Corporation in Santa Monica, research was done on Eleu-
504
sis under the direction of J. Robert Newman. Litton Industries based a
full-page advertisement on Eleusis. Descriptions of the game appeared
in European books and periodicals. Abbott began receiving letters from
allover the world with suggestions on how to make Eleusis a more
playable game.
In 1973 Abbott discussed the game with John Jaworski, a young
British mathematician who had been working on a computer version of
Eleusis for teaching induction. Then Abbott embarked on a three-year
program to reshape Eleusis, incorporating all the good suggestions he
could. The new game is not only more exciting, its metaphorical level
has been broadened as well. With the introduction of the roles of
Prophet and False Prophet the game now simulates the search for any
kind of truth. Here, then, based on a communication from Abbott, are
the rules of New Eleusis as it is now played by aficionados.
At least four players are required. As many as eight can play, but be-
yond that the game becomes too long and chaotic.
Two standard decks, shuffled together are used. (Occasionally a
round will continue long enough to require a third deck.) A full game
consists of one or more rounds (hands of play) with a different player
dealing each round. The dealer may be called by such titles as God, Na-
ture, Tao, Brahma, the Oracle (as in Delphi), or just Dealer.
The dealer's first task is to make up a "secret rule." This is simply a
rule that defines what cards can be legally played during a player's
turn. In order to do well, players must figure out what the rule is. The
faster a player discovers the rule, the higher his score will be.
One of the cleverest features of Eleusis is the scoring (described
below), which makes it advantageous to the dealer to invent a rule that
is neither too easy to guess nor too hard. Without this feature dealers
would be tempted to formulate such complex rules that no one would
guess them, and the game would become dull and frustrating.
An example of a rule that is too simple is: "Playa card of a color dif-
ferent from the color of the last card played." The alternation of colors
would be immediately obvious. A better rule is: "Play so that primes
and nonprimes alternate." For mathematicians, however, this might be
too simple. For anyone else it might be too difficult. An example of a
rule that is too complicated is: "Multiply the values of the last 3 cards
played and divide by 4. If the remainder is 0, playa red card or a card
with a value higher than 6. If the remainder is 1, playa black card or a
picture card. If the remainder is 2, play an even card or a card with a
The New Eleusis 505
value lower than 6. If the remainder is 3, play an odd card or a 10." No
one will guess such a rule, and the dealer's score will be low.
Here are three examples of good rules for games with inexperienced
players:
1. If the last legally played card was odd, playa black card. Otherwise
playa red one.
2. If the last legally played card was black, playa card of equal or higher
value. If the last card played was red, playa card of equal or lower
value. (The values of the jack, queen, king, and ace are respectively
11, 12, 13, and 1.)
3. The card played must be either of the same suit or the same value as
the last card legally played.
The secret rules must deal only with the sequence of legally played
cards. Of course, advanced players may use rules that refer to the en-
tire pattern of legal and illegal cards on the table, but such rules are
much harder to guess and are not allowed in standard play. Under no
circumstances should the secret rule depend on circumstances external
to the cards. Examples of such improper rules are those that depend on
the sex of the last player, the time of day, whether God scratches his (or
her) ear, and so on.
The secret rule must be written down in unambiguous language, on
a sheet of paper that is put aside for future confirmation. As Kruskal
proposed, the dealer may give a truthful hint before the play begins. For
example, he may say "Suits are irrelevant to the rule," or "The rule de-
pends on the two previously played cards."
After the secret rule has been recorded, the dealer shuffles the dou-
ble deck and deals 14 cards to each player and none to himself. He
places a single card called the "starter" at the extreme left of the play-
ing surface, as is indicated in Figure 38.1. To determine who plays first
the dealer counts clockwise around the circle of players, starting with
the player on his left and excluding himself. He counts until he reaches
the number on the starter card. The player indicated at that number be-
gins the play that then continues clockwise around the circle.
A play consists of placing one or more cards on the table. To playa
single card the player takes a card from his hand and shows it to every-
one. If according to the rule the card is playable, the dealer says "Right."
The card is then placed to the right of the starter card, on the "main line"
of correctly played cards extending horizontally to the right.
506 GAMES AND DECISION THEORY
Starter card

D DO LJ
Sidelines [Js De 0:
(incorrect
cards)
O
f! .,. = black card ;
. = red card
Figure 38.1. A typical round of Eleusis at an early stage
String played
incorrectly
(the 9. is wrong)
If the card fails to meet the rule, the dealer says "Wrong." In this
case the card is placed directly below the last card played. Vertical
columns of incorr!ct cards are called "sidelines." (Kruskal introduced
both the layout and the terminology of the main line and sidelines.)
Thus consecutive incorrect plays extend the same sideline downward.
If a player displays a wrong card, the dealer gives him two more cards
as a penalty, thereby increasing his hand.
If a player thinks he has discovered the secret rule, he may playa
"string" of 2, 3, or 4 cards at once. To playa string he overlaps the
cards slightly to preserve their order and shows them to everyone. If all
the cards in the string conform to the rule, the dealer says "Right."
Then all the cards are placed on the main line with no overlapping, as
if they were correctly played single cards.
If one or more cards in a string are wrong, the dealer declares the en-
tire string wrong. He does not indicate which cards do not conform to
the rule. The wrong cards are left overlapping to keep their identity as
a string and the entire string goes below the last card played. The player
is then dealt twice as many cards as there are in the string.
The layout shown in Figure 38.1 demonstrates all the rules of Eleu-
sis mentioned so far. The dealer's secret rule for this layout is the first
of the three given above.
Players improve their score by getting rid of as many cards as possi-
ble, and of course they can do this best if they guess the secret rule. At
The New Eleusis 507
the start of a round there is little information to go on, and plays are
necessaril y random. As the round continues and more and more infor-
mation is added to the layout, the rule becomes steadily easier to guess.
It may happen that a player thinks he knows the secret rule but finds
he has no card that can be legally played. He then has the option of de-
claring "No play." In this case he shows his hand to everyone. If the
dealer declares him right and his hand contains four cards or less, the
cards are returned to the deck and the round ends. If he is right and has
five or more cards, then his cards are put back into the deck, and he is
dealt a fresh hand with four fewer cards than he previously held.
If the player is wrong in declaring no play, the dealer takes one of his
correct cards and puts it on the main line. The player keeps the rest of
his hand and, as a penalty, is dealt five more cards. A player who thinks
he has no correct play but has not figured out the secret rule should re-
alize that the odds are against his using the no play option successfully.
He would do better to playa card at random.
When a player thinks he knows the secret rule, he has the opportu-
nity to prove it and increase his score. He does so by declaring himself
a Prophet. The Prophet immediately takes over the dealer's duties, call-
ing plays right or wrong and dealing penalty cards when the others
play. He can declare himself a Prophet only if all the following condi-
tions prevail:
1. He has just played (correctly or incorrectly), and the next player has
not played.
2. There is not already a Prophet.
3. At least two other players besides himself and the dealer are still in
the round.
4. He has not been a Prophet before in this round.
When a player declares himself a Prophet, he puts a marker on the
last card he played. A chess king or queen may be used. The Prophet
keeps his hand but plays no more cards unless he is overthrown. The
play continues to pass clockwise around the player'S circle, skipping
the Prophet.
Each time a player plays a card or string, the Prophet calls the play
right or wrong. The dealer then either validates or invalidates the
Prophet's statement by saying "Correct" or "Incorrect." If the Prophet
is correct, the card or string is placed on the layout-on the main line
508 GAMES AND DECISION THEORY
if right or on a sideline if wrong-and the Prophet gives the player
whatever penalty cards are required.
If the dealer says "Incorrect," the Prophet is instantly overthrown. He
is declared a False Prophet. The dealer removes the False Prophet's
marker and gives him five cards to add to his hand. He is not flllowed
to become a Prophet again during the same round, although any other
player may do so. The religious symbolism is obvious, but as Abbott
points out, there is also an amusing analogy here with science: "The
Prophet is the scientist who publishes. The False Prophet is the scien-
tist who publishes too early." It is the fun of becoming a Prophet and
of overthrowing a False Prophet that is the most exciting feature of
New Eleusis.
After a Prophet's downfall the dealer takes over his former duties. He
completes the play that overthrew the Prophet, placing the card or
string in its proper place on the layout. If the play is wrong, however,
no penalty cards are given. The purpose of this exemption is to en-
courage players to make unusual plays-even deliberately wrong
ones-in the hope of overthrowing the Prophet. In Karl Popper's lan-
guage, it encourages scientists to think of ways of "falsifying" a col-
league's doubtful theory.
If there is a Prophet and a player believes he has no card to play,
things get a bit complicated. This seldom happens, and so you can skip
this part of the now and refer to it only when the need arises.
There are four possibilities once the player declares no play:
1. Prophet says, "Right"; dealer says, "Correct." The Prophet simply
follows the procedure described earlier.
2. Prophet says, "Right"; dealer says, "Incorrect." The Prophet is im-
mediately overthrown. The dealer takes over and handles everything
as usual, except that the player is not given any penalty cards.
3. Prophet says, "Wrong"; dealer says, "Incorrect." In other words, the
player is right. The Prophet is overthrown, and the dealer handles the
playas usual.
4. Prophet says, "Wrong"; says, "Correct." In this case the
Prophet now must pick one correct card from the player's hand and
put it on the main line. If he does this correctly, he deals the player
the five penalty cards and the game goes on. It is possible, however,
for the Prophet to make a mistake at this point and pick an incorrect
The New Eleusis 509
card. If that happens, the Prophet is overthrown. The wrong card
goes back into the player's hand and the dealer takes over with the
usual procedure, except that the player is not given penalty cards.
After 30 cards have been played and there is no Prophet in the game,
players are expelled from the round when they make a wrong play,
that is, if they playa wrong card or make a wrong declaration of no
play. An expelled player is given the usual penalty cards for his final
play and then drops out of the round, retaining his hand for scoring.
If there is a Prophet, expulsions are delayed until at least 20 cards
have been laid down after the Prophet's marker. Chess pawns are used
as markers so that it is obvious when expulsion is possible. As long as
there is no Prophet, a white pawn goes on every tenth card placed on
the layout. If there is a Prophet, a black pawn goes on every tenth card
laid down after the Prophet's marker. When a Prophet is overthrown,
the black pawns and the Prophet's marker are removed.
A round can therefore go in and out of the phase when expulsions are
possible. For example, if there are 35 cards on the layout and no
Prophet, Smith is expelled when he plays incorrectly. Next Jones plays
correctly and declares herself a Prophet. If Brown then plays incor-
rectly, she is not expelled because 20 cards have not yet been laid down
after the Prophet's marker.
A round can end in two ways: (1) when a player runs out of cards or
(2) when all players (excluding a Prophet, if there is one) have been ex-
pelled.
The scoring in Eleusis is as follows:
1. The greatest number of cards held by anyone (including the Prophet)
is called the "high count." Each player (including the Prophet) sub-
tracts the number of cards in his hand from the high count. The dif-
ference is his score. If he has no cards, he gets a bonus of four points.
2. The Prophet, if there is one, also gets a bonus. It is the number of
main-line cards that follow his marker plus twice the number of side-
line cards that follow his marker, that is, a point for each correct card
since he became a Prophet and two points for each wrong card.
3. The dealer's score equals the highest score of any player. There is one
exception: If there is a Prophet, count the number of cards (right and
wrong) that precede the Prophet's marker and double this number; if
the result is smaller than the highest score, the dealer's score is that
smaller number.
510 GAMES AND DECISION THEORY
If there is time for another round, a new dealer is chosen. In princi-
ple the game ends after every player has been dealer, but this could take
most of a day. To end the game before everyone has dealt, each player
adds his scores for all the rounds played plus 10 more points if he has
not been a dealer. This compensates for the fact that dealers tend to
have higher-than-average scores.
The layout in Figure 38.2 shows the end of a round with five play-
ers. Smith was the dealer. The round ended when Jones got rid of her
cards. Brown was the Prophet and ended with 9 cards. Robinson was
expelled when he incorrectly played the 10 of spades; he had 14 cards.
Adams had 17 cards at the end of the game.
The high count is 17. Therefore Adams' score is 17 minus 17, or O.
Robinson's score is 17 minus 14, or 3. Jones receives 17 minus 0, or 17,
plus a 4-point bonus for having no cards, so that her score is 21. Brown
gets 17 minus 9, or 8, plus the Prophet's bonus of 34 (12 main-line and
11 sideline cards following his marker), making a total score of 42.
This is the highest score for the round. Twice the number of cards pre-
ceding the Prophet's marker is 50. Smith, the dealer, receives 42 be-
cause it is the smaller of the two numbers 42 and 50.
Readers are invited to look over this layout and see if they can guess
the secret rule. The play has been standard, and so that the rule is con-
fined strictly to the main-line sequence.
Some miscellaneous advice from Abbott should help inexperienced
Eleusis players. Since layouts tend to be large, the best way to play the
game is on the floor. Of course a large table can be used as well as
miniature cards on a smaller table. If necessary, the main line can be
broken on the right and continued below on the left.
Remember that in Eleusis the dealer maximizes his score by choos-
ing a rule that is neither too easy nor too difficult. Naturally this de-
pends both on how shrewdly the dealer estimates the ability of the
players and how accurately he evaluates the complexity of his rule.
Both estimates require considerable experience. Beginning players tend
to underestimate the complexity of their rules.
For example, the rule used in the first layout is simple. Compare it
with: "Playa red card, then a black card, then an odd card, then an even
card, and repeat cyclically." This rule seems to be simpler, but in prac-
tice the shift from the red-black variable to the even-odd variable
makes it difficult to discover. Abbott points out that in general restric-
tive rules that allow only about a fourth of the cards to be acceptable on
The New Eleusis 511
OODOOOOOOrruDO
DODD : D
D D
o D
Figure 38.2. Layout at the end of a round of Eleusis includes
a main line, several sidelines, and various markers. White
chess pawns are placed on every tenth card played in the
round and black pawns are placed on every tenth card
played after a Prophet's marker.
Prophet's mark., 1J]
(king or queen)
o
any given play are easier to guess than less restrictive rules that allow
half or more of the cards to be acceptable.
I shall not belabor the ways in which the game models a search for
truth (scientific, mathematical, or metaphysical) since I discussed them
in my first column on the game. I shall add only the fantasy that God
or Nature may be playing thousands, perhaps a countless number, of si-
multaneous Eleusis games with intelligences on planets in the uni-
verse, always maximizing his or her pleasure by a choice of rules that
the lesser minds will find not too easy and not too hard to discover if
given enough time. The supply of cards is infinite, and when a player
is expelled, there are always others to take his place.
Prophets and False Prophets come and go, and who knows when
one round will end and another begin? Searching for any kind of truth
is an exhilarating game. It is worth remembering that there would be no
game at all unless the rules were hidden.
Addendum
Two unusual and excellent induction games have been in-
vented since Eleusis and Delphi, both with strong analogies to scien-
512 GAMES AND DECISION THEORY
D D D D ~ D D [ J D D D [ ]
o D Black pawn D
D D
D D
D
BI3Ckpawntl]
o
tific method. For Sid Sackson's board game Patterns, see Chapter 4 of
my Mathematical Circus (Knopf, 1979). Pensari, using 32 special cards,
was the brainchild of Robert Katz. His Pensari Guide Book (1986),
which accompanies the cards, runs to 42 pages. Full-page advertise-
ments for Pensari appeared in Science News in a number of 1987 is-
sues.
Years ago, while reading The Life and Letters of Thomas H. Huxley,
edited by his son Leonard (Vol. 1, Appleton, 1901, p. 262), I came
across the following delightful paragraph. It is from a letter Huxley
sent to Charles Kingsley in 1863.
This universe is, I conceive, like to a great game being played out, and
we poor mortals are allowed to take a hand. By great good fortune the
wiser among us have made out some few of the rules of the game, as at
present played. We call them "Laws of Nature," and honour them be-
cause we find that if we obey them we win something for our pains. The
cards are our theories and hypotheses, the tricks our experimental veri-
fications. But what sane man would endeavour to solve this problem:
given the rules of a game and the winnings, to find whether the cards are
made of pasteboard or goldleaf? Yet the problem of the metaphysicians
is to my mind no saner.
In recent years Robert Abbott has been inventing bizarre mazes.
Many of them can be found in his book Mad Mazes, published in 1990
The New Eleusis 513
by Bob Adams. "A Genius for Games," an article about Abbott by David
Buxbaum, appeared in the Mensa Bulletin, May 1979.
Answers
The problem was to guess the secret rule that determined the
final layout for a round in the card game Eleusis. The rule was: "If the
last card is lower than the preceding legally played card, playa card
higher than the last card, otherwise play a lower one. The first card
played is correct unless it is equal to the starter card."
Bibliography
R. Abbott, The New Eleusis, Privately published, 1977.
T. Dietterich, master's thesis, "The Methodology of Knowledge Layers for Inducing
Descriptions of Sequentially Ordered Events," Department of Computer Sci-
ence, University of Illinois at Urbana, 1980.
M. D. Kruskal, Delphi: A Game of Inductive Reasoning, Plasma Physics Laboratory,
Princeton University, 1962. A monograph of 16 pages.
H. C. Romesburg, "Simulating Scientific Inquiry with the Card Game Eleusis," Sci-
ence Education, 3, 1979, pp. 599-608.
S. Sackson, "Eleusis: The Game with the Secret Rule," Games, May/June 1978, pp.
18-19.
R. W. Schmittberger, New Rules for Classic Games, Wiley, 1992.
514 GAMES AND DECISION THEORY
X
Physics
Chapter 39 Time Travel
"It's against reason," said Filby.
"What reason?" said the Time Traveller.
-B.G. WELLS, The Time Machine
H. G. Wells's short novel The Time Machine, an undisputed
masterpiece of science fiction, was not the first story about a time ma-
chine. That distinction belongs to "The Clock That Went Backward," a
pioneering but mediocre yarn by Edward Page Mitchell, an editor of the
New York Sun. It was published anonymously in the Sun on Septem-
ber 18, 1881, seven years before young Wells (he was only 22) wrote the
first version of his famous story.
Mitchell's tale was so quickly forgotten that science-fiction buffs did
not even know of its existence until Sam Moskowitz reprinted it in his
anthology of Mitchell's stories, The Crystal Man (1973). Nor did anyone
pay much attention to Wells's fantasy when it was serialized in 1888 in
The Science Schools Journal under the horrendous title "The Chronic
Argonauts." Wells himself was so ashamed of this clumsily written
tale that he broke it off after three installments and later destroyed all
the copies he could find. A completely rewritten version, "The Time
Traveller's Story," was serialized in The New Review beginning in 1894.
When it came out as a book in 1895, it brought Wells instant recogni-
tion.
One of the many remarkable aspects of Wells's novella is the intro-
duction in which the Time Traveller (his name is not revealed, but in
Wells's first version he is called Dr. Nebo-gipfel) explains the theory be-
hind his invention. Time is a fourth dimension. An instantaneous cube
cannot exist. The cube we see is at each instant a cross section of a
"fixed and unalterable" four-dimensional cube having length, breadth,
thickness, and duration. "There is no difference between Time and any
of the three dimensions of Space," says the Time Traveller, "except
that our consciousness moves along it." If we could view a person from
outside our space-time (the way human history is viewed by the Eter-
517
nals in Isaac Asimov's The End of Eternity or by the Tralfamadorians in
Kurt Vonnegut's Slaughterhouse-Five), we would see that person's past,
present, and future all at once, just as in 3-space we see all parts of a
wavy line that traces on a time chart the one-dimensional spatial move-
ments of mercury in a barometer.
Reading these remarks today, one might suppose that Wells had been
familiar with Hermann Minkowski's great work of tidying up Einstein's
special theory of relativity. The line along which our consciousness
crawls is, of course, our "world line": the line that traces our move-
ments in 3-space on a four-dimensional Minkowski space-time graph.
(My World Line is the title of George Gamow's autobiography.) But
Wells's story appeared in its final form 10 years before Einstein pub-
lished his first paper on relativity!
When Wells wrote his story, he regarded the Time Traveller's theories
as little more than metaphysical hanky-panky designed to make his fan-
tasy more plausible. A few decades later physicists were taking such
hanky-panky with the utmost seriousness. The notion of an absolute
cosmic time, with absolute simultaneity between distant events, was
swept out of physics by Einstein's equations. Virtually all physicists
now agree that if an astronaut were to travel to a distant star and back,
moving at a velocity close to that of light, he could in theory travel thou-
sands of years into the earth's future. Kurt Godel constructed a rotating
cosmological model in which one can, in principle, travel to any point
in the world's past as well as future, although travel to the past is ruled
out as physically impossible. In 1965 Richard P. Feynman received a
Nobel prize for his space-time approach to quantum mechanics in which
antiparticles are viewed as particles momentarily moving into the past.
Hundreds of science-fiction stories and in recent years many movies,
have been based on time travel, many of them raising questions about
time and causality that are as profound as they are sometimes funny. To
give the most hackneyed example, suppose you traveled back to last
month and shot yourself through the head. Not only do you know be-
fore making the trip that nothing like this happened but, assuming that
somehow you could murder your earlier self, how could you exist next
month to make the trip? Fredric Brown's "First Time Machine" opens
with Dr. Grainger exhibiting his machine to three friends. One of them
uses the device to go back 60 years and kill his hated grandfather when
the man was a youth. The story ends 60 years later with Dr. Grainger
showing his time machine to two friends. }
/
518 PIYSICS
It must not be thought that logical contradictions arise only when
people travel in time. The transportation of anything can lead to para-
dox. There is a hint of this in Wells's story. When the Time Traveller
sends a small model of his machine into the past or the future (he does
not know which), his guests raise two objections. If the time machine
went into the future, why do they not see it now, moving along its
world line? If it went into the past, why did they not see it there before
the Time Traveller brought it into the room?
One of the guests suggests that perhaps the model moves so fast in
time it becomes invisible, like the spokes of a rotating wheel. But what
if a time-traveling object stops moving? If you have no memory of a
cube on the table Monday, how could you send it back to Monday's
table on Tuesday? And if on Tuesday you go into the future, put the
cube on the table Wednesday, then return to Tuesday, what happens on
Wednesday if on Tuesday you destroy the cube?
Objects carried back and forth in time are sources of endless confu-
sion in certain science-fiction tales. Sam Mines once summarized the
plot of his own story. "Find the Sculptor," as follows: "A scientist builds
a time machine, goes 500 years into the future. He finds a statue of him-
self commemorating the first time traveler. He brings it back to his own
time and it is subsequently set up in his honor. You see the catch here?
It had to be set up in his own time so that it would be there waiting for
him when he went into the future to find it. He had to go into the future
to bring it back so it could be set up in his own time. Somewhere a
piece of the cycle is missing. When was the statue made?"
A splendid example of how paradox arises, even when nothing more
than messages go back in time, is provided by the conjecture that
tachyons, particles moving faster than light, might actually exist. Rela-
tivity theory leaves no escape from the fact that anything moving faster
than light w o u ~ d move backward in time. This is what inspired A. H.
Reginald Buller, a Canadian botanist, to write his often quoted limerick:
There was a young lady named Bright
Who traveled much faster than light.
She started one day
In the relative way,
And returned on the previous night.
Tachyons, if they exist, clearly cannot be used for communication.
G. A. Benford, D. L. Book, and W. A. Newcomb have chided physicists
TIme Travel 519
who are searching for tachyons for overlooking this. In "The Tachyonic
Antitelephone," they point out that certain methods of looking for
tachyons are based on interactions that make possible, in theory, com-
munication by tachyons. Suppose physicist Jones on the earth is in
communication by tachyonic antitelephone with physicist Alpha in
another galaxy. They make the following agreement. When Alpha re-
ceives a message from Jones, he will reply immediately. Jones promises
to send a message to Alpha at three o'clock earth time, if and only ifhe
has not received a message from Alpha by one o'clock. Do you see the
difficulty? Both messages go back in time. If Jones sends his message at
three, Alpha's reply could reach him before one. "Then," as the au-
thors put it, "the exchange of messages will take place if and only if it
does not take place ... a genuine ... causal contradiction." Large sums
of money have already gone down the drain, the authors believe, in ef-
forts to detect tachyons by methods that imply tachyonic communica-
tion and are therefore doomed to failure.
Time dilation in relativity theory, time travel in G6del's cosmos, and
reversed time in Feynman's way of viewing antiparticles are so care- I
fully hedged by other laws that contradictions cannot arise. In most
time-travel stories the paradoxes are skirted by leaving out any incident
that would generate a paradox. In some stories, however, logical con-
tradictions explicitly arise. When they do, the author may leave them
paradoxical to bend the reader's mind or may try to escape from para-
dox by making clever assumptions.
Before discussing ways of avoiding the paradoxes, brief mention
should be made of what might be called pseudo-time-travel stories in
which there is no possibility of contradiction. There can be no paradox,
for example, if one simply observes the past but does not interact with
it. The electronic machine in Eric Temple Bell's "Before the Dawn,"
which extracts motion pictures of the past from imprints left by light
on ancient rocks, is as free of possible paradox as watching a video
tape of an old television show. And paradox cannot arise if a person
travels into the future by going into suspended animation, like Rip van
Winkle, or Woody Allen in his motion picture Sleeper, or the sleepers
in such novels as Edward Bellamy's Looking Backward or Wells's When
the Sleeper Wakes. No paradox can arise if one dreams of the past (as
in Mark Twain's A Connecticut Yankee at King Arthur's Court or in the
1986 motion picture Peggy Sue Got Married), or goes forward in a rein-
carnation, or lives for a while in a galaxy where change is so slow in re-
520 PHYSICS
lation to earth time that when he returns, centuries on the earth have
gone by. But when someone actually travels to the past or the future, in-
teracts with it, and returns, enormous difficulties arise.
In certain restricted situations paradox can be avoided by invoking
Minkowski's "block universe," in which all history is frozen, as it were,
by one monstrous space-time graph on which all world lines are eter-
nal and unalterable. From this deterministic point of view one can
allow certain kinds of time travel in either direction, although one must
pay a heavy price for it. Hans Reichenbach, in a muddled discussion in
The Philosophy of Space and Time (Dover, 1957, pp. 140-2), puts it this
way: Is it possible for a person's world line to "loop" in the sense that
it returns him to a spot in space-time, a spot very close to where he
once had been and where some kind of interaction, such as speech, oc-
curs between the two meeting selves? Reichenbach argues that this
cannot be ruled out on logical grounds; it can only be ruled out on the
ground that we would have to give up two axioms that are strongly con-
firmed by experience: (1) A person is a unique individual who main-
tains his identity as he ages, and (2) a person's world line is linearly
ordered so that what he considers "now" is always a unique spot along
the line. (Reichenbach does not mention it, but we would also have to
abandon any notion of free will.) If we are willing to give up these
things, says Reichenbach, we can imagine without paradox certain
kinds of loops in a person's world line.
Reichenbach's example of a consistent loop is as follows. One day
you meet a man who looks exactly like you but who is older. He tells
you he is your older self who has traveled back in time. You think him
insane and walk on. Years later you discover how to go back in time.
You visit your younger self. You are compelled to tell him exactly what
your older duplicate had told you when you were younger. Of course,
he thinks that you are insane. You separate. Each of you leads a normal
life until the day comes when your older self makes the trip back in
time.
Hilary Putnam, in "It Ain't Necessarily So," argues in similar fashion
that such world-line loops need not be contradictory. He draws a Feyn-
man graph (see Figure 39.1) on which particle pair-production and
pair-annihilation are replaced by person pair-production and pair-
annihilation. The zigzag line is the world line of time traveler Smith.
At time t2 he goes back to t
1
, converses with his younger self, then con-
tinues to lead a normal life. How would this be observed by someone
Time Travel 521
t2 ------
tl ------- --
~ ~ - - - - - - ~ ~ - - - - - - - - - - - - - - - - - - - - - - - -
SPACE
Figure 39.1. Feynman graph for a time traveler to the past
whose world line is normal? Simply put a ruler at the bottom of the
chart, its edge parallel to the space axis, and move it slowly upward. At
to you see young Smith. At tl an older Smith suddenly materializes
out of thin air in the same room along with an anti-Smith, who is seated
in his time machine and living backward. (If he is smoking, you see his
cigarette butt lengthen into a whole cigarette, and so on.) Perhaps the
two forward Smiths converse. Finally, at t
z
, young Smith, backward
Smith, and the backward-moving time machine vanish. The older
Smith and his older time machine continue on their way. The fact that
we can draw a space-time diagram of these events, Putnam insists, is
proof that they are logically consistent.
It is true that they are consistent, but note that Putnam's scenario, like
Reichenbach's, involves such weak interaction between the Smiths that
it evades the deeper contradictions that arise in time-travel fiction.
What happens if the older Smith kills the younger Smith? Will Putnam
kindly supply a Feynman graph?
There is only one good way out, and science-fiction scribblers have
been using it for more than half a century. According to Sam
Moskowitz, the device was first explicitly employed to resolve time-
travel paradoxes by David R. Daniels in "Branches of Time," a tale that
522 PHYSICS
appeared in Wonder Stories in 1934. The basic idea is as simple as it is
fantastic. Persons can travel to any point in the future oftheir universe,
with no complications, but the moment they enter the past, the uni-
verse splits into two parallel worlds, each with its own time track.
Along one track rolls the world as if no looping had occurred. Along the
other track spins the newly created universe, its history permanently
altered. When I say "newly created," I speak, of course, from the stand-
point of the time traveler's consciousness. For an observer in, say, a fifth
dimension the traveler's world line simply switches from one space-
time continuum to another on a graph that depicts all the universes
branching like a tree in a metauniverse.
Forking time paths appear in many plays, novels, and short stories by
non-science-fiction writers. J. B. Priestley uses it in his popular play
Dangerous Corner, as Lord Dunsany had done earlier in his play If.
Mark Twain discusses it in The Mysterious Stranger. Jorge Luis Borges
plays with it in his "Garden of Forking Paths." But it was the science-
fiction writers who sharpened and elaborated the concept.
Let's see how it works. Suppose you go back to the time of Napoleon
in Universe 1 and assassinate him. The world forks. You are now in
Universe 2. If you like, you can return to the present of Universe 2, a
universe in which Napoleon had been mysteriously murdered. How
much would this world differ from the old one? Would you find a du-
plicate of yourself there? Maybe. Maybe not. Some stories assume that
the slightest alteration of the past would introduce new causal chains
that would have a multiplying effect and produce vast historical
changes. Other tales assume that history is dominated by such power-
ful overall forces that even major alterations of the past would damp
out and the future would soon be very much the same.
In Ray Bradbury's "A Sound of Thunder," Eckels travels back to an
ancient geological epoch under elaborate precautions to prevent any se-
rious alteration of the past. For example, he wears an oxygen mask to
prevent his microbes from contaminating animal life. But Eckels vio-
lates a prohibition and accidentally steps on a living butterfly. When he
returns to the present, he notices subtle changes in the office of the firm
that arranged his trip. He is killed for having illegally altered the future.
Hundreds of other stories by fantasy and science-fiction writers have
played variations on this theme. One of the saddest is Lord Dunsany's
"Lost" (in The Fourth Book of Jorkens, 1948). A man travels to his past,
by way of an Oriental charm, to right some old mistakes. Of course, this
Time Travel 523
alters history. When he gets back to the present, he is missing his wife
and home. "Lost! Lost!" he cries. "Don't go back down the years trying
to alter anything. Don't even wish to .... And, mind you, the whole
length of the Milky Way is more easily traveled than time, amongst
whose terrible ages I am lost."
It is easy to see that in such a metacosmos of branching time paths,
it is not possible to generate paradox. The future is no problem. If you
travel to next week, you merely vanish for a week and reappear in the
future a week younger than you would have been. But if you go back
and murder yourself in your crib, the universe obligingly splits. Uni-
verse 1 goes on as before, with you vanishing from it when you grow
up and make the trip back. Perhaps this happens repeatedly, each cycle
creating two new worlds. Perhaps it happens only once. Who knows?
In any case, Universe 2 with you and the dead baby in it rolls on. You
are not annihilated by your deed, because now you are an alien from
Universe 1 living in Universe 2.
In such a metacosmos it is easy (as many science-fiction writers have
done) to fabricate duplicates of yourself. You can go back a year in Uni-
verse 1, live for a year with yourself in Universe 2, then again go back
a year to visit two replicas of yourself in Universe 3. Clearly, by re-
peating such loops you can create as many replicas of yourself as you
please. They are genuine replicas, not pseudo-replicas as in the sce-
narios by Reichenbach and Putnam. Each has his independent world
line. History might become extremely chaotic, but there is one type of
event that can never occur: a logically contradictory one.
This vision of a metacosmos containing branching worlds may seem
crazy, but respectable physicists have taken it quite seriously. In Hugh
Everett Ill's Ph.D. thesis" 'Relative State' Formulation of Quantum Me-
chanics" (Reviews of Modern Physics 29, July 1957, pp. 454-62) he
outlines a metatheory in which the universe at every micromicroinstant
branches into countless parallel worlds, each a possible combination of
micro events that could occur as a result of microlevel uncertainty. The
paper is followed by John A. Wheeler's favorable assessment in which
he points out that classical physicists were almost as uncomfortable at
first with the radical notions of general relativity.
"If there are infinite universes," wrote Fredric Brown in What Mad
Universe, "then all possible combinations must exist. Then, some-
where, everything must be true . ... There is a universe in which Huck-
leberry Finn is a real person, doing the exact things Mark Twain
524 PHYSICS
described him as doing. There are, in fact, an infinite number of uni-
verses in which a Huckleberry Finn is doing every possible variation of
what Mark Twain might have described him as doing .... And infinite
universes in which the states of existence are such that we would have
no words or thoughts to describe them or to imagine them."
What if the universe never forks? Suppose there is only one world,
this one, in which all world lines are linearly ordered and objects pre-
serve their identity, come what may. Brown considers this possibility
in his story "Experiment." Professor Johnson holds a brass cube in his
hand. It is six minutes to three o'clock. At exactly three, he tells his col-
leagues, he will place the cube on his time machine's platform and
send it five minutes into the past.
"Therefore," he remarks, "the cube should, at five minutes before
three, vanish from my hand and appear on the platform, five minutes
before I place it there."
"How can you place it there, then?" asked one of his colleagues.
"It will, as my hand approaches, vanish from the platform and appear
in my hand to be placed there."
At five minutes to three the cube vanishes from Professor Johnson'S
hand and appears on the platforin, having been sent back five minutes
in time by his future action of placing the cube on the platform at three.
"See? Five minutes before I shall place it there, it is there!"
"But," says a frowning colleague, "what if, now that it has already ap-
peared five minutes before you place it there, you should change your
mind about doing so and not place it there at three o'clock? Wouldn't
there be a paradox of some sort involved?"
Professor Johnson thinks this is an interesting idea. To see what hap-
pens, he does not put the cube on the platform at three.
There is no paradox. The cube remains. But the entire universe, in-
cluding Professor Johnson, his colleagues, and the time machine, dis-
appears.
Addendum
J. A. Lindon, a British writer of comic verse, sent me his sequel
to the limerick about Miss Bright:
When they questioned her, answered Miss Bright,
"I was there when I got home that night;
Time Travel 525
So I slept with myself,
Like two shoes on a shelf,
Put-up relatives shouldn't be tight!"
Many readers called attention to two difficulties that could arise from
time travel in either direction. If travelers stay at the same spot in space-
time, relative to the universe, the earth would no longer be where it
was. They might find themselves in empty space, or inside something
solid. In the latter case, would the solid body prevent them from arriv-
ing? Would one or the other be shoved aside? Would there be an ex-
plosion?
The second difficulty is thermodynamic. After the time traveler de-
parts, the universe will have lost a bit of mass-energy. When he arrives,
the universe gains back the same amount. During the interval between
leaving and arriving, the universe would seem to be violating the law
of mass-energy conservation.
I mentioned briefly what is now called the "many-worlds interpre-
tation" of QM (quantum mechanics). The best reference is a 1973 col-
lection of papers on the topic, edited by Bryce DeWitt and Neill
Graham. Assuming that the universe constantly splits into billions of
parallel worlds, the interpretation provides an escape from the inde-
terminism of the Copenhagen interpretation of QM, as well as from the
many paradoxes that plague it.
Some physicists who favor the many-worlds interpretation have ar-
gued that the countless duplicate selves and parallel worlds produced
by the forking paths are not "real," but only artifacts of the theory. In
this interpretation of the many-worlds interpretation, the theory col-
lapses into no more than a bizarre way of saying the same things that
are said in the Copenhagen interpretation. Everett himself, in his orig-
inal 1957 thesis, added in proof this famous footnote:
In reply to a preprint of this article some correspondents have raised the
question of the "transition from possible to actual," arguing that in "re-
ality" there is-as our experience testifies-no such splitting of observer
states, so that only one branch can ever actually exist. Since this point
may occur to other readers the following is offered in explanation.
The whole issue of the transition from "possible" to "actual" is taken
care of in the theory in a very simple way-there is no such transition,
nor is such a transition necessary for the theory to be in accord with our
experience. From the viewpoint of the theory all elements of a super-
526 PHYSICS
position (all "branches") are "actual," none any more "real" than the
rest. It is unnecessary to suppose that all but one are somehow de-
stroyed, since all the separate elements of a superposition individually
obey the wave equation with complete indifference to the presence or
absence ("actuality" or not) of any other elements. This total lack of ef-
fect of one branch on another also implies that no observer will ever be
aware of any "splitting" process.
Arguments that the world picture presented by this theory is contra-
dicted by experience, because we are unaware of any branching process,
are like the criticism of the Copernican theory that the mobility of the
earth as a real physical fact is incompatible with the common sense in-
terpretation of nature because we feel no such motion. In both cases the
argument fails when it is shown that the theory itself predicts that our
experience will be what it in fact is. (In the Copernican case the addition
of Newtonian physics was required to be able to show that the earth's in-
habitants would be unaware of any motion of the earth.)
The many-worlds interpretation has been called a beautiful theory no-
body can believe. Nevertheless, a number of top physicists have indeed
accepted-some still do-its outrageous multiplicity of logically possi-
ble worlds. Here is DeWitt defending it in "Quantum Mechanics and Re-
ality," a 1970 article reprinted in the collection he edited with Graham:
The obstacle to taking such a lofty view of things, of course, is that it
forces us to believe in the reality of all the simultaneous worlds ... in
each of which the measurement has yielded a different outcome. Nev-
ertheless, this is precisely what [the inventors ofthe theory] would have
us believe .... This universe is constantly splitting into a stupendous
number of branches, all resulting from the measurement like interac-
tions between its myriads of components. Moreover, every quantum
transition taking place on every star, in every galaxy, in every remote cor-
ner of the universe is splitting our local world on earth into myriads of
copies of itself.
I still recall vividly the shock I experienced on first encountering this
multiworld concept. The idea of 10
100
+ slightly imperfect copies of one-
self all constantly splitting into further copies, which ultimately become
unrecognizable, is not easy to reconcile with commonsense.
Although John Wheeler originally supported the many-worlds inter-
pretation, he has since abandoned it. I quote from the first chapter of
his Frontiers of Time (Center for Theoretical Physics, 1978):
Time Travel 527
Imaginative Everett's thesis is, and instructive, we agree. We once sub-
scribed to it. In retrospect, however, it looks like the wrong track. First,
this formulation of quantum mechanics denigrates the quantum. It de-
nies from the start that the quantum character of nature is any clue to the
plan of physics. Take this Hamiltonian for the world, that Hamiltonian,
or any other Hamiltonian, this formulation says. I am a principle too
lordly to care which, or why there should be any Hamiltonian at all. You
give me whatever world you please, and in return I give you back many
worlds. Don't look to me for help in understanding this universe.
Second, its infinitely many unobservable worlds make a heavy load of
metaphysical baggage. They would seem to defy MendeIeev's demand of
any proper scientific theory, that it should "expose itself to destruction."
Wigner, Weizsacker, and Wheeler have made objections in more detail,
but also in quite contrasting terms, to the relative-state or many-worlds
interpretation of quantum mechanics. It is hard to name anyone who
conceives of it as a way to uphold determinism.
In the paper titled "Rotating Cylinders and the Possibility of Global
Causality Violation," physicist Frank Tipler raised the theoretical pos-
sibility of constructing a machine that would enable one to go forward
or backward in time. (Tipler is one of the enthusiasts for the many-
worlds interpretation, the coauthor of a controversial book (The An-
thropic Cosmological Principle, Oxford University Press, 1986) and
sole author of an even more controversial book, The Physics of Im-
morality (1994)). Taking off from G6del's rotating cosmos and from re-
cent work on the space-time pathologies surrounding black holes,
Tipler imagines a massive cylinder, infinitely long, and rotating so
rapidly that its surface moves faster than half the speed of light. Space-
time near the cylinder would be so distorted that, according to Tipler's
calculations, astronauts could orbit the cylinder, going with or against
its spin, and travel into their past or future.
Ti.pler speculated on the possibility that such a machine could be
built with a cylinder of finite length and mass, but later concluded that
such a device was impossible to construct with any known forms of
matter and force. Such doubts did not inhibit Poul Anderson from
using Tipler's cylinder for time travel in his novel The Avatar, nor did
it stop Robert Forward from writing "How to Build a Time Machine"
(Omni, May 1980). "We already know the theory," Omni editors com-
mented above Forward's backward article, "All that's needed is some
advanced engineering."
528 PHYSICS
I can understand the attraction of the many-world's interpretation of
quantum mechanics provided it is taken as no more than another way
of talking about quantum mechanics. If, however, the countless other
universes are assumed to be as real as the one we are in,-real in the
sense that the moon is real and unicorns are not-then the price one
has to pay is horrendous. It is true that the many-world's interpretation
simplifies concepts, but the multiplicity of other worlds is the greatest
violation of Occam's razor in the history of theoretical physics. I ap-
plaud Wheeler for abandoning his early defense of the theory and ad-
mire Roger Penrose and other quantum experts for having the courage
to reject the realistic interpretation of the many-world's interpretation
as wild and temporary fashionable nonsense.
In 1995 England issued a stamp honoring H.G. Wells's The Time Ma-
chine.
I close with two pearls of wisdom from the stand-up comic "Profes-
sor" Irwin Corey: "The past is behind us" and "the future lies ahead."
Bibliography
B. Ash (ed.), "Time and the Nth Dimension" and "Lost and Parallel Worlds," The
Visual Encyclopedia of Science Fiction, Harmony, 1977.
G. A. Benford, D. L. Bock, and W. A. Newcomb, "The Tachyonic Antitelephone,"
Physical Review D, Vol. 2, July 15, 1970, pp. 263-65.
G. Benford, "Time Again," Fantasy and Science Fiction, July 1993, pp. 96-106.
S. K. Blau, "Would a Topology Change Allow Ms. Bright to Travel Backward in
Time?" American Journal of Physics, Vol. 66, March 1998, pp. 179-85.
G. Conklin (ed.), Science Fiction Adventures in Dimension, Vanguard, 1953.
D. Deutsch and M. Lockwood, "The Quantum Physics of Time Travel," Scientific
American, March 1994. (See also the letters section in September 1994.)
B. S. DeWitt and N. Graham (eds.), The Many-Worlds Interpretation of Quantum
Mechanics, Princeton University Press, 1973.
J. Earman, "On Going Backward in Time," Philosophy of Science, Vol. 34, Sep-
tember 1967, pp. 211-22.
J. Earman, Bangs, Crunches, Whimpers, and Shrieks, Oxford University Press,
1995.
G. Feinberg, "Particles That Go Faster Than Light," Scientific American, February
1970, pp. 69-77.
D. Freedman, "Cosmic Time Travel," Discover, June 1989, pp. 58-63.
J. Richard Gatt III. "Will We Travel Back (or Forward) in Time?" Time, April 10,
2000.
J. C. Graves and J. E. Roper, "Measuring Measuring Rods," Philosophy of Science
32, January 1965, pp. 39-56.
Time Travel 529
P. Horwich, "On Some Alleged Paradoxes of Time Travel," Journal of Philosophy,
Vol. 72, 1975, pp. 432-44.
S. J. Jakiel and R. Levinthal, "The Laws of Time Travel," Extrapolation, Vol. 21,
1980, pp. 130-38.
D. Lewis, "The Paradoxes of Time Travel," American Philosophical Quarterly, VoL
13, 1976,pp. 145-52.
L. Marder, TIme and the Space-Traveler, Allen & Unwin, 1971.
H. Moravec, "Time Travel and Computing," Extropy, No.9, Summer 1992, pp.
22-28.
P. J. Nahin, TIme Machines, American Institute of Physics, 1993.
P. Nicholls (ed.), "Time TraveL" 'Time Paradoxes," "Alternate Worlds," and "Par-
allel Worlds," The Science Fiction Encyclopedia, Doubleday, 1979.
B. Parker, Cosmic Time Travel, Plenum, 1991.
P. Parsons, "A Warped View of Time Travel," Science, Vol. 274, October 11, 1996,
pp.202-3.
J. B. Priestley, "Time and Fiction in Drama," Man and Time, Doubleday, 1964.
H. Putnam, "It Ain't Necessarily So," The Journal of Philosophy, VoL 59, October
25,1962, pp. 658-71; reprinted in Putnam's Philosophical Papers, Vol. 1, Cam-
bridge University Press, 1975.
I. Redmount, "Wormholes, Time Travel, and Quantum Gravity," New Scientist,
April 28, 1990, pp. 57-61.
L. S. Schulman, "Tachyon Paradoxes," American Journal of Physics, Vol. 39, May
1971, pp. 481-84.
J. J. C. Smart, "Is Time Travel Possible?" The Journal of Philosophy, Vol. 60, 1963,
pp.237-41.
N.J.J. Smith, "Bananas Enough for Time TraveL" British Journal of the Philosophy
of Science, Vol. 48,1997, pp. 363-89.
P.V.D. Stern (ed.), Travelers in Time, Doubleday, 1947.
I. Stewart, "The Real Physics of Time Travel," Analog Science Fiction and Fact, Jan-
uary 1994, pp. 106-29.
F. J. Tipler, "Rotating Cylinders and the Possibility of Global Causality Violation,"
Physical Review D, Vol. 9, April 15, 1974, pp. 2203-6.
530 PHYSICS
Chapter 40
It is impossible to meditate on time and the
mystery of the creative passage of nature with-
out an overwhelming emotion at the limitations
of human intelligence.
-ALFRED NORTH WHITEHEAD, The Concept of Nature
Does Time
Ever Stop?
There has been a great deal of interest among physicists as to
whether there are events on the elementary-particle level that cannot be
time-reversed, that is, events for which imagining a reversal in the di-
rection of motion of all the particles involved is imagining an event that
cannot happen in nature. Richard Feynman has suggested an approach
to quantum mechanics in which antiparticles are viewed as particles
momentarily traveling backward in time. Cosmologists have specu-
lated about two universes for which all the events in one are reversed
relative to the direction of time in the other: in each universe intelligent
organisms would live normally from past to future, but if the organisms
in one universe could in some way observe events in the other (which
many physicists consider an impossibility), they would find those
events going in the opposite direction. It has even been conjectured that
if our universe stops expanding and starts to contract, there will be a
time reversal, but it is far from clear what that would mean.
In this chapter I shall consider two bizarre questions about time. In-
deed, these questions are of so little concern to scientists that only
philosophers and writers of fantasy and science fiction have had much
to say about them: Is it meaningful to speak of time stopping? Is it
meaningful to speak of altering the past?
Neither question should be confused with the familiar subject of
time's relativity. Newton believed the universe was pervaded by a sin-
gle absolute time that could be symbolized by an imaginary clock off
somewhere in space (perhaps outside the cosmos). By means of this
clock the rates of all the events in the universe could be measured. The
notion works well within a single inertial frame of reference such as the
surface of the earth, but it does not work for inertial systems moving in
relation to each other at high speeds. According to the theory of rela-
531
tivity, if a spaceship were to travel from our solar system to another
solar system with a velocity close to that of light, events would proceed
much slower on the spaceship than they would on the earth. In a sense,
then, such a spaceship is traveling through time into the future. Pas-
sengers on the spaceship might experience a round-trip voyage as tak-
ing only a few years, but they would return to find that centuries of
earth-years had elapsed.
The notion that different parts of the universe can change at different
rates of time is much older than the theory of relativity. In the Scholas-
tic theology of the Middle Ages angels were considered to be nonmate-
rial intelligences living by a time different from that of earthly creatures;
God himself was thought to be entirely outside of time. In the first act
of Lord Byron's play, Cain, A Mystery, the fallen angel Lucifer says:
With us acts are exempt from time
and we
Can crowd eternity into an hour
Or stretch an hour into eternity
We breathe not by a mortal
measurement-
But that's a mystery.
In the 20th century hundreds of science-fiction stories played with
the relativity of time in different inertial systems, but the view that
time can speed up or slow down in different parts of our universe is
central to many older tales. A popular medieval legend tells of a monk
who is entranced for a minute or two by the song of a magical bird.
When the bird stops singing, the monk discovers that several hundred
years have passed. In a Moslem legend Mohammed is carried by a mare
into the seventh heaven. After a long visit the prophet returns to the
earth just in time to catch a jar of water the horse had kicked over be-
fore starting its ascent.
Washington Irving's "Rip Van Winkle" is this country's best-known
story about someone who sleeps for what seems to him to be a normal
time while two decades of earth-years rush by. King Arthur's daughter
Gyneth slept for 500 years under a spell cast by Merlin. Every culture
has similar sleeper legends. H. G. Wells used the device in When the
Sleeper Wakes, and it is a common practice in science fiction to put as-
tronauts into a cryogenic sleep so they can survive interstellar voyages
that are longer than their normal life span. In Wells's short story "The
532 PHYSICS
New Accelerator" a scientist discovers a way to speed up a person's bi-
ological time so that the world seems to come almost to a halt. This de-
vice too is frequently encountered in later science fiction.
The issue under consideration here, however, is not how time can
vary but whether time can be said to stop entirely. It is clearly mean-
ingful to speak of all motion ceasing in one part of the universe,
whether or not such a part exists. In the theory of relativity the speed
of light is an unattainable limit for any object with mass. If a spaceship
could attain the speed of light (which the theory of relativity rules out
because the mass of the ship would increase to infinity), then time on
the spaceship would stop in the sense that all change on it would cease.
In earth time it might take 100 years for the spaceship to reach a desti-
nation, but to astronauts on the spaceship the destination would be
reached instantaneously. One can also imagine a piece of matter or
even a human being reduced to such a low temperature (by some as yet
unknown means) that even all subatomic motions would be halted.
For that piece of matter, then, one could say that time had stopped. Ac-
tually it is hard to understand why the piece of matter would not
vanish.
The idea of time stopping creates no problems for writers of fantasy
who are not constrained by the real world. For example, in 1. Frank
Baum's "The Capture of Father Time," one of the stories in his Ameri-
can Fairy Tales, a small boy lassoes Time, and for a while everything
except the movements of the boy and Father Time stops completely. In
Chapter 22 of James Branch Cabell's Jurgen: A Comedy of Justice, out-
side time sleeps while Jurgen enjoys a pleasurable stay in Cocaigne
with Queen Analtis. Later in the novel Jurgen stares into the eyes of the
God of his grandmother and is absolutely motionless for 37 days. In
Jorge Luis Borges' story "The Secret Miracle" a writer is executed by a
firing squad. Between the command to fire and the writer's death God
stops all time outside the writer's brain, giving him a year to complete
his masterpiece.
Many similar examples from legend and literature show that the no-
tion of time stopping in some part of the universe is not logically in-
consistent. But what about the idea of time stopping throughout the
universe? Does the notion that everything stops moving for a while
and then starts again have any meaning?
If it is assumed that there is an outside observer-perhaps a god-
watching the universe from a region of hypertime, then of course the
Does Time Ever Stop? 533
notion of time stopping does have meaning, just as imagining a god in
hyperspace gives meaning to the notion of everything in the universe
turning upside down. The history of our universe may be like a three-
dimensional motion picture a god is enjoying. When the god turns off
the projector to do something else, a few millenniums may go by before
he comes back and turns it on again. (After all, what are a few millen-
niums to a god?) For all we can know a billion centuries of hypertime
may have elapsed between my typing the first and the second word of
this sentence.
Suppose, however, all outside observers are ruled out and "universe"
is taken to mean "everything there is." Is there still a way to give a
meaning to the idea of all change stopping for a while? Although most
philosophers and scientists would say there is not, a few have argued
for the other side. For example, in "Time without Change" Sydney S.
Shoemaker, a philosopher at Cornell University, makes an unusual ar-
gument in support of the possibility of change stopping.
Shoemaker is concerned not with the real world but with possible
worlds designed to prove that the notion of time stopping everywhere
can be given a reasonable meaning. He proposes several worlds of this
kind, all of them based on the same idea. I shall describe only one such
world here, in a slightly dramatized form.
Imagine a universe divided into regions A, B, and C. In normal times
inhabitants of each region can observe the inhabitants of the other two
and communicate with them. Every now and then, however, a myste-
rious purple glow permeates one of the regions. The glow always lasts
for a week and is invariably followed by a year in which all change in
the region ceases. In other words, for one year absolutely nothing hap-
pens there. Shoemaker calls the phenomenon a local freeze. Since no
events take place,light cannot leave the region, and so the region seems
to vanish for a year. When it returns to view, its inhabitants are unaware
of any passage of time, but they learn from their neighbors that a year,
as measured by clocks in the other two regions, has elapsed. To the in-
habitants of the region that experienced the local freeze it seems that in-
stantaneous changes have taken place in the other two regions. As
Shoemaker puts it: "People and objects will appear to have moved in
a discontinuous manner or to have vanished into thin air or to have ma-
terialized out of thin air; saplings will appear to have grown instanta-
neously into mature trees, and so on."
In the history of each of the three regions local freezes, invariably pre-
534 PHYSICS
ceded by a week of purple light, have happened thousands of times.
Now suppose that suddenly, for the first time in history, purple light
appears simultaneously in regions A, B, and C and lasts for a week.
Would it not be reasonable, Shoemaker asks, for scientists in the three
regions to conclude that change had ceased for a year throughout the
entire universe even though no minds were aware of it?
Shoemaker considers several objections to his thesis and counters all
of them ingeniously. Interested readers can consult his paper and then
read a technical analysis of it in the fifth chapter of G. Schlesinger's
Confirmation and Confirmability Schlesinger agrees with Shoemaker
that an empirical, logically consistent meaning can be found for the
sentence "A period of time t has passed during which absolutely noth-
ing happened." Note that similar arguments about possible worlds can
provide meanings for such notions as everything in a universe turning
upside down, mirror-reversing, doubling in size, and so on.
The question of whether the past can be changed is even stranger
than that of whether time can stop. Writers have often speculated about
what might have happened if the past had taken a different turn. J. B.
Priestley's play Dangerous Corner dealt with this question, and there
have been innumerable "what if" stories in both science fiction and
other kinds of literature. In all time-travel stories where someone enters
the past, the past is necessarily altered. The only way the logical con-
tradictions created by such a premise can be resolved is by positing a
universe that splits into separate branches the instant the past is en-
tered. In other words, while time in the old branch "gurgles on" (a
phrase from Emily Dickinson) time in the new branch gurgles on in a
different way toward a different future. When I speak of altering the
past, however, I mean altering it throughout a single universe with no
forking time paths. (Pseudoalterations of the past, such as the rewriting
of history satirized by George Orwell in 1984, obviously do not qualify.)
Given this context, can an event, once it has happened, ever be made
not to have happened?
The question is older than Aristotle, who in his Ethics (Book 6)
writes: "It is to be noted that nothing that is past is an object of choice,
for example, no one chooses to have sacked Troy; for no one deliberates
about the past, but about what is future and capable of being otherwise,
while what is past is not capable of not having taken place; hence
Agathon is right in saying: 'For this alone is lacking even to God, to
make undone things that have once been done.' "
Does Time Ever Stop? 535
Thomas Aquinas believed God to be outside of time and thus capa-
ble of seeing all his creation's past and future in one blinding instant.
(Even though human beings have genuine power of choice, God knows
how each one will choose; it is in this way that Aquinas sought to har-
monize predestination and free will.) For Aquinas it was not possible
for God to do absolutely impossible things, namely those that involve
logical contradiction. For example, God could not make a creature that
was both a human being and a horse (that is, a complete human being
and a complete horse, rather than a mythical combination of parts such
as a centaur), because that would involve the contradiction of assum-
ing a creature to be simultaneously rational and nonrationaL
Similarly, God cannot alter the past. That would be the same as as-
serting that the sack of Troy both took place and did not take place.
Aquinas agreed with Aristotle that the past must forever be what it
was, and it was this view that became the official position of medieval
Scholasticism. It is not so much that God's omnipotence is limited by
the law of contradiction but rather that the law is part of God's nature.
"It is best to say," Aquinas wrote, "that what involves contradiction
cannot be done rather than that God cannot do it." Modern philoso-
phers would say it this way. God can't make a four-sided triangle, not
because he can't make objects with four sides but because a triangle is
defined as a three-sided polygon. The phrase H four-sided triangle" is
therefore a nonsense phrase, one without meaning.
Edwyn Bevan, in a discussion of time in his book Symbolism and Be-
lief, finds it odd that Aquinas would deny God the ability to alter the
past and at the same time allow God to alter the future. In the 10th
question of Summa Theologica (la. 10, article 5.3), Aquinas wrote:
"God can cause an angel not to exist in the future, even if he cannot
cause it not to exist while it exists, or not to have existed when it al-
ready has." For Aquinas to have suggested that for God the past is un-
alterable and the future is not unalterable, Bevan reasons, is surely to
place God in some kind of time, thus contradicting the assertion that
God is outside of time.
I know of no scientist or secular philosopher who has seriously be-
lieved the past could be altered, but a small minority of theologians
have maintained that it could be. The greatest of them was Peter
Damian, the zealous Italian reformer of the Roman Catholic church in
the 11th century. In On Divine Omnipotence, his most controversial
treatise, Damian argued that God is in no way bound by the law of con-
536 PHYSICS
tradiction, that his omnipotence gives him the power to do all contra-
dictory things including changing the past. Although Damian, who
started out as a hermit monk, argued his extreme views skillfully, he re-
garded all reasoning as superfluous, useful only for supporting revealed
theology. It appears that he, like Lewis Carroll's White Queen, would
have defended everyone's right to believe six impossible things before
breakfast. (Damian was also a great promoter of self-flagellation as a
form of penance, a practice that became such a fad during his lifetime
that some monks flogged themselves to death.)
One of my favorite Lord Dunsany stories is the best example I know
of from the literature of fantasy that illustrates Damian's belief in the
possibility of altering the past. It is titled "The King That Was Not," and
you will find it in Dunsany's early book of wonder tales Time and the
Gods. It begins as follows: "The land of Runazar hath no King nor ever
had one; and this is the law of the land of Runazar that, seeing that it
hath never had a King, it shall not have one for ever. Therefore in
Runazar the priests hold sway, who tell the people that never in
Runazar hath there been a King."
The start of the second paragraph is surprising: "Althazar, King of
Runazar .... " The story goes on to recount how Althazar ordered his
sculptors to carve marble statues of the gods. His command was
obeyed, but when the great statues were undraped, their faces were
very much like the face of the king. Althazar was pleased and rewarded
his sculptors handsomely with gold, but up in Pegana (Dunsany's
Mount Olympus) the gods were outraged. One of them, Mung, leaned
forward to make his sign against Althazar, but the other gods stopped
him: "Slay him not, for it is not enough that Althazar shall die, who
hath made the faces of the gods to be like the faces of men, but he must
not even have ever been."
"Spake we of Althazar, a King?"
asked one of the gods.
"Nay, we spake not."
"Dreamed we of one Althazar?"
"Nay, we dreamed not."
Below Pegana, in the royal palace, Althazar suddenly passed out of
the memory of the gods and so "became no longer a thing that was or
had ever been." When the priests and the people entered the throne
room, they found only a robe and a crown. "The gods have cast away
Does Time Ever Stop? 537
the fragment of a garment," said the priests, "and lo! from the fingers
of the gods hath slipped one little ring."
Addendum
When I wrote about time stopping I was using a colloquial ex-
pression to mean that change ceases. Because there are no moving
"clocks" of any sort for measuring time, one can say in a loose sense
that time stops. Of course time does not move or stop any more than
length can extend or not extend. It is the universe that moves. You can
refute the notion that time "flows" like a river simply by asking: "At
what rate does it flow?" Shoemaker wanted to show in his paper not
that time stops, then starts again, but that all change can stop and some
sort of transcendental hypertime still persists. Change requires time,
but perhaps, Shoemaker argued, time does not require change in our
universe.
Harold A. Segal, in a letter in The New York Times (January 11,1987)
quoted a marvelous passage from Shakespeare's As You Like It (Act III,
Scene 2) in which Rosalind explains how time can amble, trot, gallop,
or stand still for different persons in different circumstances. It trots for
the "young maid" between her engagement and marriage. It ambles for
a priest who knows no Latin because he is free from the burden of
"wasteful learning." It ambles for the rich man in good health who
"lives merrily because he feels no pain." It gallops for the thief who
awaits his hanging. For whom does it stand still? "With lawyers in the
vacation; for they sleep between term and term, and then they perceive
not how Time moves."
Isaac Asimov, in an editorial in Asimov's Science-Fiction Magazine
(June 1986) explained why it would not be possible for a person to
walk about and observe a world in which all change had stopped. To
move, she would have to push aside molecules, and this would inject
time into the outside world. She would be as frozen as the universe,
even though dancing atoms in her brain might continue to let her think.
Asimov could have added that she would not even be able to see the
world because sight depends on photons speeding from the world into
one's eyes.
Two readers, Edward Adams and Henry Lambert, independently
wrote to say that the god Koschei, in lurgen, could alter the past. At the
538 PHYSICS
end of the novel he eliminates all of Jurgen's adventures as never hav-
ing happened. However, Jurgen recalls that Horvendile (the name Ca-
bell often used for himself) once told him that he (Horvendile) and
Koschei were one and the same!
Edward Fredkin is a computer scientist who likes to think of the uni-
verse as a vast cellular automaton run by an inconceivably complex al-
gorithm that tells the universe how to jump constantly from one state
to the next. Whoever or whatever is running the program could, of
course, shut it down at any time, then later start it running again. We
who are part of the program would have no awareness of such gaps in
time.
On the unalterability of the past, readers reminded me of the stanza
in Omar's Rubaiyat about the moving finger that having writ moves
on, and all our piety and wit cannot call it back to cancel half a line. Or
as Ogden Nash once put it:
One thing about the past,
It's likely to last.
I touched only briefly on the many science-fiction stories and novels
that deal with time slowing down or halting. For references on some of
the major tales see the section HWhen time stands still" on page 153 of
The Visual Encyclopedia of Science Fiction. The most startling possi-
bility, seriously advanced by some physicists, is that the universe
comes to a complete stop billions of times every microsecond, then
starts up again. Like a cellular automaton it jumps from state to state.
Between the jumps, nothing changes. The universe simply does not
exist. Time is quantized. An electron doesn't move smoothly from here
to there. It moves in tiny jumps, occupying no space in between.
The fundamental unit of quantized time has been called the
"chronon." Between chronons one can imagine one or more parallel
universes operating within our space, but totally unknown to us. Think
of a film with two unrelated motion pictures running on alternate
frames. Between the frames of our universe, who knows what other ex-
otic worlds are unrolling in the intervals between our chronons? Both
motion pictures and cellular automata are deterministic, but in this vi-
sion of parallel universes running in the same space, there is no need to
assume determinism. Chance and free will could still play creative roles
in making the future of each universe unpredictable in principle.
Does Time Ever Stop? 539
Bibliography
E. Bevan, Symbolism and Belief, Kennikar Press, 1968.
E. C. Brewer, "Seven Sleepers" and "Sleepers," The Reader's Handbook, rev. ed.,
Lippincott, 1898.
C. D. Broad, "Time," Encyclopedia of Religion and Ethics, James Hastings (ed.),
Scribner's, 1922.
K. Bulmer, "Time and Nth Dimensions," The Visual Encyclopedia of Science Fic-
tion, Harmony Books, 1977, pp. 145-54.
M. Gardner, The New Ambidextrous Universe, W. H. Freeman, 1990.
P. Nicholls (ed.), "Time Travel," The Science Fiction Encyclopedia. Doubleday,
1979.
P. Nicholls (ed.), "Time Travel and Other Universes," The Science in Science Fic-
tion, Knopf, 1983, Chapter 5.
G. Schlesinger, Confirmation and Confirmability, Oxford University Press, 1974.
S. S. Shoemaker, "Time Without Change," The Journal of Philosophy, Vol. 66, 1969,
pp.363-81.
P.V.D. Stern (ed.), Travelers in Time: Strange Tales of Man's Journeying into the Past
and the Future. Doubleday, 1947.
540 PHYSICS
Chapter 41
The universe, so far as known to us,
is so constituted, that whatever is
true in anyone case, is true in all
cases of a certain description; the only
difficulty is, to find what description.
-JOHN STUART MILL, A System of Logic
Induction and
Probability
'magine that we are living on an intricately patterned carpet. It
mayor may not extend to infinity in all directions. Some parts of the
pattern appear to be random, like an abstract expressionist painting;
other parts are rigidly geometrical. A portion of the carpet may seem to-
tally irregular, but when the same portion is viewed in a larger context,
it becomes part of a subtle symmetry.
The task of describing the pattern is made difficult by the fact that the
carpet is protected by a thick plastic sheet with a translucence that
varies from place to place. In certain places we can see through the
sheet and perceive the pattern; in others the sheet is opaque. The plas-
tic sheet also varies in hardness. Here and there we can scrape it down
so that the pattern is more clearly visible. In other places the sheet re-
sists all efforts to make it less opaque. Light passing through the sheet
is often refracted in bizarre ways, so that as more of the sheet is re-
moved, the pattern is radically transformed. Everywhere there is a mys-
terious mixing of order and disorder. Faint lattices with beautiful
symmetries appear to cover the entire rug, but how far they extend is
anyone's guess. No one knows how thick the plastic sheet is. At no
place has anyone scraped deep enough to reach the carpet's surface, if
there is one.
Already the metaphor has been pushed too far. For one thing, the pat-
terns of the real world, as distinct from this imaginary one, are con-
stantly changing, like a carpet that is rolling up at one end while it is
unraveling at the other end. Nevertheless, in a crude way the carpet can
introduce some of the difficulties philosophers of science encounter in
trying to understand why science works.
Induction is the procedure by which carpetologists, after examining
parts of the carpet, try to guess what the unexamined parts look like.
541
Suppose the carpet is covered with billions oftiny triangles. Whenever
a blue triangle is found, it has a small red dot in one corner. After find-
ing thousands of blue triangles, all with red dots, the carpetologists
conjecture that all blue triangles have red dots. Each new blue triangle
with a red dot is a confirming instance of the law. Provided that no
counterexample is found, the more confirming instances there are, the
stronger is the carpetologists' belief that the law is true.
The leap from "some" blue triangles to "all" is, of course, a logical
fallacy. There is no way to be absolutely certain, as one can be in work-
ing inside a deductive system, what any unexamined portion of the car-
pet looks like. On the other hand, induction obviously works, and
philosophers justify it in other ways. John Stuart Mill did so by posit-
ing, in effect, that the carpet's pattern has regularities. He knew this rea-
soning was circular, since it is only by induction that carpetologists
have learned that the carpet is patterned. Mill did not regard the circle
as vicious, however, and many contemporary philosophers (R. B.
Braithwaite and Max Black, to name two) agree. Bertrand Russell, in his
last major work, tried to replace Mill's vague "nature is uniform" with
something more precise. He proposed a set of five posits about the
structure of the world that he believed were sufficient to justify induc-
tion.
Hans Reichenbach advanced the most familiar of several pragmatic
justifications. If there is any way to guess what unexamined parts of the
carpet look like, Reichenbach argued, it has to be by induction. If in-
duction does not work, nothing else will, and so science might as well
use the only tool it has. "This answer is not fallacious," wrote Russell,
"but I cannot say that I find it very satisfying."
Rudolf Carnap agreed. His opinion was that all these ways of justi-
fying induction are correct but trivial. If "justify" is meant in the sense
that a mathematical theorem is justified, then David Hume was right:
There is no justification. But if "justify" is taken in any of several
weaker senses, then, of course, induction can be defended. A more in-
teresting task, Carnap insisted, is to see whether it is possible to con-
struct an inductive logic.
It was Carnap's great hope that such a logic could be constructed. He
foresaw a future in which a scientist could express in a formalized lan-
guage a certain hypothesis together with all the relevant evidence. Then
by applying inductive logic, he could assign a probability value, called
542 PHYSICS
the degree of confirmation, to the hypothesis. There would be nothing
final about that value. It would go up or down or stay the same as new
evidence accumulated. Scientists already think in terms of such a logic,
Carnap maintained, but only in a vague, informal way. As the tools of
science become more powerful, however, and as our knowledge of
probability becomes more precise, perhaps eventually we can create a
calculus of induction that will be of practical value in the endless
search for scientific laws.
In Carnap's Logical Foundations o/Probability (University of Chicago
Press, 1950) and also in his later writings, he tried to establish a base
for such a logic. How successful he was is a matter of dispute. Some
philosophers of science share his vision Uohn G. Kemeny for one) and
have taken up the task where Carnap left off. Others, notably Karl Pop-
per and Thomas S. Kuhn, regard the entire project as having been mis-
conceived.
Carl G. Hempel, one of Carnap's admirers, has argued sensibly that
before we try to assign quantitative values to confirmations, we should
first make sure we know in a qualitative way what is meant by "con-
firming instance." It is here that we run into the worst kinds of diffi-
culty.
Consider Hempel's notorious paradox of the raven. Let us approach
it by way of 100 playing cards. Some of them have a picture of a raven
drawn on the back. The hypothesis is: "All raven cards are black." You
shuffle the deck and deal the cards face up. After turning 50 cards
without finding a counterinstance, the hypothesis certainly becomes
plausible. As more and more raven cards prove to be black, the degree
of confirmation approaches certainty and may finally reach it.
Now consider another way of stating the same hypothesis: "All non-
black cards are not ravens." This statement is logically equivalent to the
original one. If you test the new statement on another shuffled deck of
100 cards, holding them face up and turning them as you deal, clearly
each time you deal a nonblack card and it proves to have no raven on
the back, you confirm the guess that all nonblack cards are not ravens.
Since this is logically equivalent to "All raven cards are black," you
confirm that also. Indeed, if you deal all the cards without finding a red
card with a raven, you will have completely confirmed the hypothesis
that all raven cards are black.
Unfortunately, when this procedure is applied to the real world, it
Induction and Probability 543
seems not to work. "All ravens are black" is logically the same as "All
nonblack objects are not ravens." We look around and see a yellow ob-
ject. Is it a raven? No, it is a buttercup. The flower surely confirms (al-
beit weakly) that all nonblack objects are not ravens, but it is hard to see
how it has much relevance to HAll ravens are black." If it does, it also
confirms that all ravens are white or any color except yellow. To make
things worse, "All ravens are black" is logically equivalent to "Any ob-
ject is either black or not a raven." And that is confirmed by any black
object whatever (raven or not) as well as by any nonraven (black or
not). All of which seems absurd.
Nelson Goodman's "grue" paradox is equally notorious. An object is
"grue" if it is green until, say, January 1, 2500, and blue thereafter. Is the
law "All emeralds are grue" confirmed by observations of green emer-
alds? A prophet announces that the world will exist until January 1,
2500, when it will disappear with a bang. Every day the world lasts
seems to confirm the prediction, yet no one supposes that it becomes
more probable.
To make matters still worse, there are situations in which confirma-
tions make a hypothesis less likely. Suppose you turn the cards of a
shuffled deck looking for confirmations of the guess that no card has
green spots. The first 10 cards are ordinary playing cards, then sud-
denly you find a card with blue spots. It is the eleventh confirming in-
stance, but now your confidence in the guess is severely shaken. Paul
Berent has pointed out several similar examples. A man 99 feet tall is
discovered. He is a confirming instance of "All men are less than 100
feet tall," yet his discovery greatly weakens the hypothesis. Finding a
normal-size man in an unlikely place (such as Saturn's moon Titan) is
another example of a confirming instance that would weaken the same
hypothesis.
Confirmations may even falsify a hypothesis. Ten cards with all val-
ues from the ace through the 10 are shuffled and dealt face down in a
row. The guess is that no card with value n is in the nth position from
the left. You turn the first nine cards. Each card confirms the hypothe-
sis. But if none of the turned cards is the 10, the nine cards taken to-
gether refute the hypothesis.
Here is another example. Two piles of three cards each are on the
table. One pile consists of the jack, queen, and king of hearts, the other
consists of the jack, queen, and king of clubs. Each has been shuffled.
544 PHYSICS
Smith draws a card from the heart pile, Jones takes a card from the
club pile. The hypothesis is that the pair of selected cards consists of a
king and queen. The probability of this is 2/9. Smith looks at his card
and sees that it is a king. Without naming it, he announces that his card
has confirmed the hypothesis. Why? Because knowing that his card is
a king raises the probability of the hypothesis being true from 2/9 to 3/9
or 1/3. Jones now sees that he (Jones) has drawn a king, so he can make
the same statement Smith made. Each card, taken in isolation, is a con-
firming instance. Yet, both cards taken together falsify the hypothesis.
Carnap was aware of such difficulties. He distinguished sharply be-
tween "degree of confirmation," a probability value based on the total
relevant evidence, and what he called "relevance confirmation," which
has to do with how new observations alter a confirmation estimate.
Relevance confirmation cannot be given simple probability values. It is
enormously complex, swarming with counterintuitive arguments. In
Chapter 6 of Carnap's Logical Foundations he analyzes a group of
closely related paradoxes of confirmation relevance that are easily mod-
eled with cards.
For example, it is possible that data will confirm each of two hy-
potheses but disconfirm the two taken together. Consider a set of 10
cards, half with blue backs and half with green ones. The green-backed
cards (with the hearts and spades designated Hand S) are QH, 10H, 9H,
KS, QS. The blue-backed cards are KH, /H, lOS, 9S, 8S. The 10 cards are
shuffled and dealt face down in a row.
Hypothesis A is that the property of being a face card (a king, a queen,
or a jack) is more strongly associated with green backs than with blue.
An investigation shows that this is true. Of the five cards with green
backs, three are face cards versus only two face cards with blue backs.
Hypothesis B is that the property of being a red card (hearts or dia-
monds) is also more strongly associated with green backs than with
blue. A second investigation confirms this. Three green-backed cards
are red, but there are only two red cards with blue backs. Intuitively
one assumes that the property of being both red and a face card is more
strongly associated with green backs than blue, but that is not the case.
Only one red face card has a green back, whereas two red face cards
have blue backs!
It is easy to think of ways, fanciful or realistic, in which similar sit-
uations can arise. A woman wants to marry a man who is both rich and
Induction and Probability 545
kind. Some of the bachelors she knows have hair and some are bald.
Being a statistician, she does some sampling. Project A establishes that
3/5 of the men with hair are rich but only 2/5 of the bald men are rich.
Project B discloses that 3/5 of the men with hair are kind but only 2/5
of the bald men are kind. The woman might hastily conclude that she
should marry a man with hair, but if the distribution of the attributes
corresponds to that of the face cards and red cards mentioned in the
preceding example, her chances of getting a rich, kind man are twice
as great if she sets her sights on a bald man.
Another research project shows that 3/5 of a group of patients taking
a certain pill are immune to colds for five years, compared with only
2/5 in the control group who were given a placebo. A second project
shows that 3/5 of a group receiving the pill were immune to tooth cav-
ities for five years, compared with 2/5 who got the placebo. The com-
bined statistics could show that twice as many among those who got
the placebo are free for five years from both colds and cavities, com-
pared with those who got the pill.
A striking instance of how a hypothesis can be confirmed by two in-
dependent studies, yet disconfirmed by the total results, is provided by
the following game. It can be modeled with cards, but to vary the equip-
ment let's use 41 poker chips and four hats (see Figure 41.1). On table
A is a black hat containing five colored chips and six white chips. Be-
side it is a gray hat containing three colored chips and four white chips.
On table B is another pair of black and gray hats. In the black hat there
are six colored chips and three white chips. In the gray hat there are
nine colored chips and five white chips. The contents of the four hats
are shown by the charts in the illustration.
You approach table A with the desire to draw a colored chip. Should
you take a chip from the black hat or from the gray one? In the black hat
five of the eleven chips are colored. so that the probability of getting a
colored chip is 5/11. This is greater than 3/7. which is the probability
of getting a colored chip if you take a chip from the gray hat. Clearly
your best bet is to take a chip from the black hat.
The black hat is also your best choice on table B. Six of its nine chips
are colored, giving a probability of 6/9, or 2/3, that you will get a col-
ored chip. This exceeds the probability of 9/14 that you will get a col-
ored chip if you choose to take a chip from the gray hat.
Now suppose that the chips from both black hats are combined in
546 PHYSICS
A

o
41.1. E. H. ~ l I n D ~ ; o n
reverse.

o
o
o
o
o
o
o
o
o
o
547
(See "Sex Bias in Graduate Admissions: Data from Berkeley," by P. J.
Bickel, E. A. Hammel, and J. W. O'Connell.)
Blyth has invented another paradox that is even harder to believe
than Simpson's. It can be modeled with three sets of cards or three un-
fair dice that are weighted to give the required probability distribu-
tions to their faces. We shall model it with the three spinners shown in
Figure 41.2, because they are easy to construct by anyone who wants
to verify the paradox empirically.
.56 .51
\
Figure 41.2. The three spinners for Colin R. Blyth's paradox
Spinner A, with an undivided dial, is the simplest. No matter where
the arrow stops, it gives a value of 3. Spinner B gives values of 2, 4, or
6 with the respective probability distributions of .56, .22, and .22. Spin-
ner C gives values of 1 or 5 with the probabilities of .51 and .49.
You pick a spinner; a friend picks another. Each of you flicks his
arrow, and the highest number wins. If you can later change spinners
on the basis of experience, which spinner should you choose? When
the spins are compared in pairs, we find that A beats B with a proba-
bility of 1 x .56 = .56. A beats C with a probability of 1 x .51 = .51. B
beats C with a probability of (1 x .22) + (.22 x .51) + (.56 x .51) = .6178.
Clearly A, which beats both of the others with a probability of more
than 1/2, is the best choice. C is the worst because it is beaten with a
probability of more than 1/2 by both of the others.
Now for the crunch. Suppose you play the game with two others and
you have the first choice. The three spinners are flicked, and the high
number wins. Calculating the probabilities reveals an extraordinary
fact. A is the worst choice; C is the best! A wins with a probability of
.56 x .51 = .2856, or less than 1/3. B wins with a probability of (.44 x
548 PHYSICS
.51) + (.22 x .49) = .3322, or almost 1/3. C wins with a probability of .49
x .78 = .3822, or more than 1/3.
Consider the havoc this can wreak in statistical testing. Assume that
drugs for a certain illness are rated in effectiveness with numbers 1
through 6. Drug A is uniformly effective at a value of 3 (spinner A).
Studies show that drug C varies in effectiveness. Fifty-one percent of
the time it has value 1, and 49 percent of the time it has value 5 (spin-
ner C). If drugs A and C are the only two on the market and a doctor
wants to maximize a patient's chance of recovery, he clearly chooses
drug A.
What happens when drug B, with values and a probability distribu-
tion corresponding to spinner B, becomes available? The bewildered
doctor, if he considers all three drugs, finds C preferable to A.
Blyth has an even more mind-blowing way of dramatizing the para-
dox. Every night, a statistician eats at a restaurant that offers apple pie
and cherry pie. He rates his satisfaction with each kind of pie in values
1 through 6. The apple pie is uniformly 3 (spinner A); the cherry varies
in the manner of spinner C. Naturally, the statistician always takes
apple.
Occasionally the restaurant has blueberry pie. Its satisfaction varies
in the manner of spinner B.
Waitress: Shall I bring your apple pie?
Statistician: No. Seeing that today you also have blueberry, I'll take
the cherry.
The waitress would consider that a joke. Actually, the statistician is
rationally maximizing his expectation of satisfaction. (An error. See
Addendum.) Is there any paradox that points up more spectacularly the
kinds of difficulty Carnap's followers must overcome in their efforts to
advance his program?
Addendum
Many readers quite properly chided me for carelessness when
I described Colin R. Blyth's paradox of the man and the three pies. It
was I (not Blyth) who said that the man's decision was to maximize his
"expectation of satisfaction." What he is maximizing is, in Blyth's
words, "his best chance" of getting the most satisfying pie. It is a sub-
Induction and Probability 549
tIe but important difference. Both the dining statistician and the doc-
tor have a choice between two intents: maximizing their average of sat-
isfaction in the long run or maximizing their chance of getting the best
pie or drug on a particular occasion.
To put it another way, Blyth's pie eater is minimizing his regret: the
probability that he will see a better pie on the next table. His doctor
counterpart, as Paul Chernick suggested, could be trying to avoid a
malpractice suit that might result if a dissatisfied patient went to an-
other doctor and got more effective treatment. "Is the case of a scientist
closer to that of a player in the spinner game," asked George Mavrodes,
"or is it closer to that of the statistical pie eater? ... I do not know the
answer to that question."
John F. Hamilton, Jr., revised the dialogue between the waitress and
the statistician as follows:
Waitress: Which pie will be better tonight, A or B?
Statistician: The odds are on A.
Waitress: What about A and C?
Statistician: Again, A will probably win.
Waitress: I see, you mean A will probably be the best of all.
Statistician: No, actually Chas the greatest chance of being the best.
Waitress: OK, cut the funny stuff. Which pie do you want to order,
AorC?
Statistician: Neither. I'll have a slice of B, please.
The paradoxes of confirmation are not, of course, paradoxes in the
sense of being contradictions, but paradoxes in the wider sense of being
counterintuitive results that make nonsense of earlier naYve attempts,
by John Stuart Mill and others, to define the meaning of "confirming in-
stance." Philosophers who discuss the paradoxes are not ignorant of
statistical theory. It is precisely because statistical theory demands so
many careful distinctions that the task of formulating an inductive
logic is so difficult.
Richard C. Jeffrey, in The Logic of Decision Making (University of
Chicago Press, 2 ed., 1983), formulates an amusing variant of Good-
man's "grue" paradox. We define a "goy" as a girl born before 2000
or a boy born after that date, and a "birl" as a boy born before 2000
or a girl born after that date. Until now, no goy has had a penis, and
all birls have. Hence by induction, (A) the first goy born after 2000
will have no penis, and (B) the first birl born after that date will.
550 PHYSICS
However, the first goy born after 2000 will be a boy, which contra-
dicts A. Similarly, the first birl born after 2000 will be a girl, which
contradicts B.
England's famous recent philosopher of science, Karl Popper, spent
most of his life arguing against Rudolf Carnap that the so-called straight
rule-that the more confirming evidence is found for a conjecture, the
more the conjecture is probably true-has no role in deciding the worth
of a theory. Instead, he insisted that science advances by having its
conjectures pass tests for falsification. Carnap's commonsense view
was that both confirmation and disconfirmation are essential to the
scientific method.
Indeed, the falsification of a conjecture always adds confirmation to
its opposite, and confirmation of a conjecture always tends to discon-
firm its opposite. Consider the conjecture: There are life forms on Mars.
If a Martian probe finds life forms it will confirm the conjecture and si-
multaneously falsify the conjecture that there are no life forms on Mars.
And if the probe finds no life on Mars it will confirm the second con-
jecture and falsify the first. Moreover, falsification can be as fallible
and uncertain as a confirmation. Popper got a lot of contorted debate
out of his attack on what he called "induction." In my opinion his at-
tack was extreme and is destined to fade from the history of the phi-
losophy of science.
Bibliography
Confirmation Theory
P. Berent, "Disconfirmation by Positive Instances," Philosophy of Science, No. 39,
December 1972, p. 522.
W. C. Salmon, "Confirmation," Scientific American, May 1973, pp. 75-83.
G. Schlesinger, Confirmation and Confirmability, Oxford University Press, 1974.
R. Swinburne, An Introduction to Confirmation Theory, Methuen, 1973.
Simpson's Paradox
E. F. Beckenbach, "Baseball Statistics," Mathematics Teacher, Vol. 72, May 1979,
pp.351-52.
P.J. Bickel, E. A. Hammel, and J. W. O'Connell, "Sex Bias in Graduate Admissions:
Data from Berkeley," Science, Vol. 187, February 1985, pp. 398-404.
C. R. Blyth, "On Simpson's Paradox and the Sure-Thing Principle," Journal of the
American Statistical Association, Vol. 67, June 1972, pp. 364-81.
Induction and Probability 551
R. Falk and M. Bar-Hillel, "Magic Possibilities of the Weighted Average," Mathe-
matics Magazine, Vol. 53, March 1980, pp. 106-7.
T. R. Knapp, "Instances of Simpson's Paradox," The College Mathematics Journal,
Vol. 16, June 1985, pp. 209-11.
C. Wagner, "Simpson's Paradox in Real Life," American Statistician, Vol. 36, Feb-
ruary 1982, pp. 46-48.
Cirue
F. Jackson, "Grue," Journal of Philosophy, Vol. 72,1975, pp. 113-31.
D. Stalker, Grue!, Open Court, 1994. The book's bibliography lists 316 references.
R. Swinburne, "Grue," Analysis, Vol. 28, 1968, pp. 123-28.
T. E. Wilerson, "A Gruesome Note," Mind, Vol. 82, 1973, pp. 276-77.
552 PHYSICS
Chapter 42
Fulfilling absolute decree
In casual simplicity.
-EMILY DICKINSON
Simplicity
Ms. Dickinson's lines are about a small brown stone in a road,
but if we view the stone as part of the universe, fulfilling nature's laws,
all sorts of intricate and mysterious events are taking place within it on
the microlevel. The concept of "simplicity," in both science and math-
ematics, raises a host of deep, complicated, still unanswered questions.
Are the basic laws of nature few in number, or many, or perhaps infi-
nite, as Stanislaw Vlam believed and others believe? Are the laws
themselves simple or complex? Exactly what do we mean when we
say one law or mathematical theorem is simpler than another? Is there
any objective way to measure the simplicity of a law or a theory or a
theorem?
Most biologists, especially those who are doing research on the brain,
are impressed by the enormous complexity of living organisms. In con-
trast, although quantum theory has become more complicated with the
discovery of hundreds of unexpected particles and interactions, most
physicists retain a strong faith in the ultimate simplicity of basic phys-
icallaws.
This was especially true of Albert Einstein. "Our experience," he
wrote, "justifies us in believing that nature is the realization of the sim-
plest conceivable mathematical ideas." When he chose the tensor equa-
tions for his theory of gravitation, he picked the simplest set that would
do the job, then published them with complete confidence that (as he
once said to the mathematician John G. Kemeny) "God would not have
passed up an opportunity to make nature that simple." It has even been
argued that Einstein's great achievements were intellectual expressions
of a psychological compulsion that Henry David Thoreau, in Walden,
expressed as follows:
"Simplicity, simplicity, simplicity! I say, let your affairs be as two or
553
three, and not a hundred or a thousand; instead of a million count half
a dozen, and keep your accounts on your thumb nail."
In Peter Michelmore's biography of Einstein, he tells us that "Ein-
stein's bedroom was monkish. There were no pictures on the wall, no
carpet on the floor .... He shaved roughly with bar soap. He often went
barefoot around the house. Only once every few months he would
allow Elsa [his wife] to lop off swatches of his hair .... Most days he
did not find underwear necessary. He also dispensed with pajamas
and, later, with socks. 'What use are socks?' he asked. 'They only pro-
duce holes.' Elsa put her foot down when she saw him chopping off the
sleeve of a new shirt from the elbow down. He explained that cuffs had
to be buttoned or studded and washed frequently-all a waste of time."
"Every possession," Einstein said, "is a stone around the leg." The
statement could have come straight out of Walden.
Yet nature seems to have a great many stones around her legs. Basic
laws are simple only in first approximations; they become increasingly
complex as they are refined to explain new observations. The guiding
motto of the scientist, Alfred North Whitehead wrote, should be: "Seek
simplicity and distrust it." Galileo picked the simplest workable equa-
tion for falling bodies, but it did not take into account the altitude of the
body and had to be modified by the slightly more complicated equa-
tions of Newton. Newton too had great faith in simplicity. "Nature is
pleased with simplicity," he wrote, echoing a passage in Aristotle, "and
affects not the pomp of superfluous causes." Yet Newton's equations in
turn were modified by Einstein, and there are physicists, such as Robert
Dicke, who believe that Einstein's gravitational equations must be mod-
ified by still more complicated formulas.
It is dangerous to argue that because many basic laws are simple, the
undiscovered laws also will be simple. Simple first approximations
are obviously the easiest to discover first. Because the "aim of science
is to seek the simplest explanations of complex facts," to quote White-
head again (Chapter 7 of The Concept of Nature), we are apt to "fall into
the error" of thinking that nature is fundamentally simple "because
simplicity is the goal of our quest."
This we can say. Science sometimes simplifies things by producing
theories that reduce to the same law phenomena previously considered
unrelated-for example, the equivalence of inertia and gravity in gen-
eral relativity. Science equally often discovers that behind apparently
simple phenomena, such as the structure of matter, there lurks unsus-
554 PHYSICS
pected complexity. Johannes Kepler struggled for years to defend the
circular orbits of planets because the circle was the simplest closed
curve. When he finally convinced himself that the orbits were ellipses,
he wrote of the ellipse as "dung" he had to introduce to rid astronomy
of vaster amounts of dung. It is a perceptive statement because it sug-
gests that the introduction of more complexity on one level of a theory
can introduce greater overall simplicity.
Nevertheless, at each step along the road simplicity seems to enter
into a scientist's work in some mysterious way that makes the simplest
workable hypothesis the best bet. "Simplest" is used here in a strictly
objective sense, independent of human observation, even though no
one yet knows how to define it. Naturally there are all sorts of ways one
theory can be simpler than another in a pragmatic sense, but these
ways are not relevant to the big question we are asking. As philosopher
Nelson Goodman has put it, "If you want to go somewhere quickly,
and several alternate routes are equally likely to be open, no one asks
why you take the shortest." In other words, if two theories are not
equivalent-lead to different predictions-and a scientist considers
them equally likely to be true, he will test first the theory that he con-
siders the "simplest" to test.
In this pragmatic sense, simplicity depends on a variety of factors:
the kind of apparatus available, the extent of funding, the available
time, the knowledge ofthe scientist and his assistants, and so on. More-
over, a theory may seem simple to one scientist because he understands
the mathematics and complicated to another scientist less familiar with
the math. A theory may have a simple mathematical form but predict
complex phenomena that are difficult to test, or it may be a compli-
cated theory that predicts simple results. As Charles Peirce pointed
out, circumstances may be such that it is more economical to test first
the least plausible of several hypotheses.
These subjective and pragmatic factors obviously play roles in re-
search, but they fail to touch the heart of the mystery. The deep ques-
tion is: Why, other things being equal, is the simplest hypothesis
usually the most likely to be on the right track-that is, most likely to
be confirmed by future research?
Consider the following "simple" instance of a scientific investigation.
A physicist, searching for a functional relationship between two vari-
ables, records his observations as spots on a graph. Not only will he
draw the simplest curve that comes close to the spots but also he allows
Simplicity 555
y
y y
x
42.1. a J..H..ICl'i:l.1UJ..Q .1.l.1..1..1 ... ,L ... U ...... curve most
556
their power to predict new observations, although the predictions are
different. Both theories cannot be true. Both may be false. Each de-
mands a different test and each test costs $1 million. If simplicity en-
ters into the probability of a theory's being right, there is an obvious
advantage in being able to measure simplicity so that the simplest the-
ory can be tested first.
No one today knows how to measure this kind of simplicity or even
how to define it. Something in the situation must be minimized, but
what? It is no good to count the terms in a theory's mathematical for-
mulation, because the number depends on the notation. The same for-
mula may have 10 terms in one notation and three in another. Einstein's
famous E = mc
2
looks simple only because each term is a shorthand
symbol for concepts that can be written with formulas involving other
concepts. This happens also in pure mathematics. The only way to ex-
press pi with integers is as the limit of an infinite series, but by writing
1t the entire series is squeezed into one symbol.
Minimizing the powers of terms also is misleading. For one thing, a
linear equation such as x = 2y graphs as a straight line only when the
coordinates are Cartesian. With polar coordinates it graphs as a spiral.
For another thing, minimizing powers is no help when equations are
not polynomials. Even when they are polynomial, should one say that
an equation such as W= 13x+ 23y+ 132z is "simpler" than x= y2?
In comparing the simplest geometric figures the notion of simplicity
is annoyingly vague. In one of Johnny Hart's B.C. comic strips a cave-
man invents a square wagon wheel. Because it has too many corners
and therefore too many bumps, he goes back to his drawing board and
invents a "simpler" wheel in the shape of a triangle. Comers and
bumps have been minimized, but the inventor is still further from the
simplest wheel, the circle, which has no corners. Or should the circle
be called the most complicated wheel because it is a "polygon" with an
infinity of corners? The truth is that an equilateral triangle is simpler
than a square in that it has fewer sides and corners. On the other hand,
the square is simpler in that the formula for its area as a function of its
side has fewer terms.
One of the most tempting of many proposed ways to measure the
simplicity of a hypothesis is to count its number of primitive concepts.
This, alas, is another blind alley. One can artificially reduce concepts
by combining them. Nelson Goodman brings this out clearly in his fa-
mous "grue" paradox about which dozens of technical papers have
Simplicity 557
been written. Consider a simple law: All emeralds are green. We now
introduce the concept "grue." It is the property of being green if ob-
served, say, before January 1, 2001, and being blue if observed there-
after. We state a second law: All emeralds are grue.
Both laws have the same number of concepts. Both have the same
"empirical content" (they explain all observations). Both have equal
predictive power. A single instance of a wrong color, when an emerald
is examined at any future time, can falsify either hypothesis. Everyone
prefers the first law because "green" is simpler than "grue"-it does not
demand new theories to explain the sudden change of color of emer-
alds on January 1, 2001. Although Goodman has done more work than
anyone on this narrow aspect of simplicity, he is still far from final re-
sults, to say nothing of the more difficult problem of measuring the
overall simplicity of a law or theory. The concept of simplicity in sci-
ence is far from simple! It may turn out that there is no single measure
of simplicity but many different kinds, all of which enter into the com-
plex final evaluation of a law or theory.
Surprisingly, even in pure mathematics similar difficulties arise.
Mathematicians usually search for theorems in a manner not much dif-
ferent from the way physicists search for laws. They make empirical
tests. In pencil doodling with convex quadrilaterals-a way of experi-
menting with physical models-a geometer may find that when he
draws squares outwardly on a quadrilateral's sides and joins the cen-
ters of opposite squares, the two lines are equal and intersect at 90 de-
grees (see Figure 42.2). He tries it with quadrilaterals of different
shapes, always getting the same results. Now he sniffs a theorem. Like
Figure 42.2. A "simple" geometrical theorem
558 PHYSICS
a physicist, he picks the simplest hypothesis. He does not, for example,
test first the conjecture that the two lines have a ratio of one to 1.00007
and intersect with angles of 89 and 91 degrees, even though this con-
jecture may equally well fit his crude measurements. He tests first the
simpler guess that the lines are always perpendicular and equal. His
"test," unlike the physicist's, is a search for a deductive proof that will
establish the hypothesis with certainty.
Combinatorial theory is rich in similar instances where the simplest
guess is usually the best bet. As in the physical world, however, there
are surprises. Consider the following problem discovered by Leo Moser.
Two or more spots are placed anywhere on a circle's circumference.
Every pair is joined by a straight line. Given n spots, what is the max-
imum number of regions into which the circle can be divided? Figure
42.3 gives the answers for two, three, and four spots. The reader is
asked to search for the answers for five and six spots and, if possible,
find the general formula.
Figure 42.3. A combinatorial problem
Addendum
The beautiful theorem about the squares on the sides of an ar-
bitrary convex quadrilateral is known as Von Aubel's theorem. Many
readers, disappointed that I did not provide a proof, sent excellent
Simplicity 559
proofs of their own. I lack space for any here, but you will find a sim-
ple vector proof in "Von Aubel's Quadrilateral Theorem," by Paul J.
Kelly, Mathematics Magazine, January 1966, pages 35-37. A different
proof based on symmetry operations is in Geometric Transformations,
by I. M. Yaglom (Random House, 1962, pp. 95-96, problem 24b).
As Kelly points out, the theorem can be generalized in three ways
that make it even more beautiful.
1. The quadrilateral need not be convex. The lines joining the centers
of opposite squares may not intersect, but they remain equal and
perpendicular.
2. Any three or even all four of the quadrilateral's corners may be
collinear. In the first case the quadrilateral degenerates into a trian-
gle with a "vertex" on one side; in the second case, into a straight line
with two "vertices" on it.
3. One side of the quadrilateral may have zero length. This brings two
corners together at a single point which may be treated as the center
of a square of zero size.
The second and third generalizations were discovered by a reader, W.
Nelson Goodwin, Jr., who drew the four examples shown in Figure
42.4. Note that the theorem continues to hold if opposite sides of a
quadrilateral shrink to zero. The resulting line may be regarded as one
of the lines connecting midpoints of opposite squares of zero size, and
of course it equals and is perpendicular to a line joining midpoints of
two squares drawn on opposite sides of the original line.
Answers
Leo Moser's circle-and-spots problem is an amusing example of
how easily an empirical induction can go wrong in experimenting with
pure mathematics. For one, two, three, four, and five spots on the cir-
cle, the maximum number of regions into which the circle can be di-
vided by joining all pairs of spots with straight lines is 1, 2,4,8, 16 ....
One might conclude that this simple doubling series continues and
that the maximum number of regions for n spots is 2
n
-
1
Unfortunately
this formula fails for all subsequent numbers of spots. Figure 42.5
shows how six spots give a maximum of 31, not 32, regions. The cor-
rect formula is:
560 PHYSICS
-
,
-
Figure 42.4. Curious generalizations of Von Aubel's theorem
A parenthetical expression (t) is the number of ways m objects can
be combined, taken k at a time. (It equals m!/k!(m - k)!) Moser has
pointed out that the formula gives the sum of the rows of numbers at
the left of the diagonal line drawn on Pascal's triangle, as shown in the
illustration.
Written out in full, the formula is:
Simplicity 561
SPOTS
1
2
3 2
4 3 3
5 4 6
6 5 10 10
7 6 15 20
Figure 42.5. Solution to Leo Moser's spot problem
n
4
- 6n
3
+ 23n
2
18n + 24
24
4
REGIONS
1
2
4
8
16
31
57
When the positive integers are plugged into n, the formula generates
the sequence: 1,2,4,8, 16, 31, 57,99, 163,256,386,562 .... Theprob-
lem is a delightful illustration of Whitehead's advice to seek simplic-
ity but distrust it.
I have been unable to determine where or when Moser first pub-
lished this problem, but in a letter he says he thinks it was in Mathe-
matics Magazine about 1950. It has since appeared in numerous books
and periodicals, with varying methods of solution. A partial list of ref-
erences is given in the bibliography for this chapter.
Almost no progress, if any, has been made in finding a way to mea-
sure the simplicity of a theory in a manner that would be useful to sci-
entists. One recent suggestion, based on a technique for defining
random numbers, is to translate a theory into a string of binary digits.
562 PHYSICS
The theory's simplicity is then defined by the length of the shortest
computer program that will print the string. This isn't much help.
Quite apart from the difficulty of finding the shortest algorithm for a
long binary expression, how is the string formed in the first place?
How, for example, can you express superstring theory by a string ofbi-
nary digits?
Scientists all agree that somehow the simpler of two theories, each
with the same explanatory and predictive power, has the better chance
of being fruitful, but no one knows why. Maybe it's because the ultimate
laws of nature are simple, but who can be positive there really are ul-
timate laws? Some physicists suspect there may be infinite levels of
complexity. At each level the laws may get progressively simpler until
suddenly the experimenters open a trap door and another complicated
subbasement is discovered.
Bibliography
S. Barker, "On Simplicity in Empirical Theories," Philosophy of Science, Vol. 28,
1961, pp. 162-71.
M. Bunge, The Myth of Simplicity, Prentice-Hall, 1963.
K. Friedman, "Empirical Simplicity and Testability," British Journal for the Phi-
losophy of Science, Vol. 28, 1972, pp. 25-33.
N. Goodman, "The Test of Simplicity," Science, Vol. 128, October 31, 1958, pp.
1064-69.
N. Goodman "Science and Simplicity," Philosophy of Science Today, S. Mor-
genbesser (ed.), Basic Books, 1967.
C. G. Hempel, "Simplicity," Philosophy of Natural Science, Prentice-Hall, 1966, pp.
40-45.
M. Hesse "Simplicity," The Encyclopedia of Philosophy, Macmillan Company and
the Free Press, 1967.
M. Hesse, "Ramifications of 'Grue,' " The British Journal for the Philosophy of Sci-
ence, Vol. 20, May 1969, pp. 13-25.
W. Jefferson and J. Berger, "Occam's Razor and Bayesian Analysis," American Sci-
entist, Vol. 80, January/February 1992, pp. 64-72.
J. G. Kemeny, "Two Measures of Simplicity," Journal of Philosophy, Vol. 52, 1955,
pp.722-33.
K. Popper, The Logic of Scientific Discovery, Basic Books, 1959, Chapter 7.
H. R. Post, "Simplicity in Scientific Theories," British Journal of the Philosophy of
Science, Vol. 11, May 1960, pp. 32-41.
W. V. Quine and J. S. Ullian, "Hypothesis," The Web of Belief, Random House,
1970, Chapter V.
Simplicity 563
G. Schlesinger, "The Principle of Minimal Assumption," British Journal of the Phi-
losophy of Science, Vol. 11, May 1960, pp. 55-59. See also Discussion, Vol. 11,
February 1961, pp. 328-31.
G. Schlesinger, "Induction and Parsimony," American Philosophical Quarterly,
Vol. 8, April 1971.
E. Sober, Simplicity, Clarendon Press, 1975.
R. Wright, "The Razor's Edge," The Sciences, July/August, 1987, pp. 2-5.
564 PHYSICS
XI
Logic and
Philosophy
Chapter 43 The Unexpected
Hanging
]I new and powerful paradox has come to light." This is the
opening sentence of a mind-twisting article by Michael Scriven that ap-
peared in the July 1951 issue of the British philosophical journal Mind.
Scriven, who bears the title of "professor of the logic of science" at the
University of Indiana, is a man whose opinions on such matters are not
to be taken lightly. That the paradox is indeed powerful has been amply
confirmed by the fact that more than 20 articles about it have appeared
in learned journals. The authors, many of whom are distinguished
philosophers, disagree sharply in their attempts to resolve the para-
dox. Since no consensus has been reached, the paradox is still very
much a controversial topic.
No one knows who first thought of it. According to the Harvard Uni-
versity logician W. V. Quine, who wrote one ofthe articles, the paradox
was first circulated by word of mouth in the early 1940s. It often took
the form of a puzzle about a man condemned to be hanged.
The man was sentenced on Saturday. "The hanging will take place at
noon," said the judge to the prisoner, "on one ofthe seven days of next
week. But you will not know which day it is until you are so informed
on the morning of the day of the hanging."
The judge was known to be a man who always kept his word. The
prisoner, accompanied by his lawyer, went back to his cell. As soon as
the two men were alone the lawyer broke into a grin. "Don't you see?"
he exclaimed. "The judge's sentence cannot possibly be carried out."
(see Figure 43.1.)
"I don't see," said the prisoner.
"Let me explain. They obviously can't hang you next Saturday. Sat-
urday is the last day of the week. On Friday afternoon you would still
be alive and you would know with absolute certainty that the hanging
567
--
\
Figure 43.1. The prisoner
eliminates all possible days.
would be on Saturday. You would know this before you were told so on
Saturday morning. That would violate the judge's decree."
"True," said the prisoner.
"Saturday, then is positively ruled out," continued the lawyer. "This
leaves Friday as the last day they can hang you. But they can't hang you
on Friday because by Thursday afternoon only two days would remain:
Friday and Saturday. Since Saturday is not a possible day, the hanging
would have to be on Friday. Your knowledge of that fact would violate
the judge's decree again. So Friday is out. This leaves Thursday as the
last possible day. But Thursday is out because if you're alive Wednes-
day afternoon, you'll know that Thursday is to be the day."
"I get it," said the prisoner, who was beginning to feel much better.
"In exactly the same way I can rule out Wednesday, Tuesday, and Mon-
day. That leaves only tomorrow. But they can't hang me tomorrow be-
cause I know it today!"
In brief, the judge's decree seems to be self-refuting. There is nothing
logically contradictory in the two statements that make up his decree;
nevertheless, it cannot be carried out in practice. That is how the para-
dox appeared to Donald John O'Connor, a philosopher at the University
of Exeter, who was the first to discuss the paradox in print (Mind, July
1948). O'Connor's version of the paradox concerned a military com-
mander who announced that there would be a Class A blackout during
the following week. He then defined a Class A blackout as one that the
568 LOGIC AND PHILOSOPHY
participants could not know would take place until after 6 P.M. on the
day it was to occur.
"It is easy to see," wrote O'Connor, "that it follows from the an-
nouncement of this definition that the exercise cannot take place at
all." That is to say, it cannot take place without violating the definition.
Similar views were expressed by the authors of the next two articles (L.
Jonathan Cohen in Mind for January 1950, and Peter Alexander in Mind
for October 1950), and even by George Gamow and Marvin Stern when
they later included the paradox (in a man-to-be-hanged form) in their
book Puzzle Math (New York: Viking, 1958).
Now, if this were all there was to the paradox, one could agree with
O'Connor that it is "rather frivolous." But, as Scriven was the first to
point out, it is by no means frivolous, and for a reason that completely
escaped the first three authors. To make this clear, let us return to the
man in the cell. He is convinced, by what appears to be unimpeachable
logic, that he cannot be hanged without contradicting the conditions
specified in his sentence. Then on Thursday morning, to his great sur-
prise, the hangman arrives. Clearly he did not expect him. What is
more surprising, the judge's decree is now seen to be perfectly correct.
The sentence can be carried out exactly as stated. "I think this flavour
of logic refuted by the world makes the paradox rather fascinating,"
writes Scriven. "The logician goes pathetically through the motions
that have always worked the spell before, but somehow the monster,
Reality, has missed the point and advances stilL"
In order to grasp more clearly the very real and profound linguistic
difficulties involved here, it would be wise to restate the paradox in
two other equivalent forms. By doing this we can eliminate various ir-
relevant factors that are often raised and that cloud the issue, such as
the possibility of the judge's changing his mind, of the prisoner's dying
before the hanging can take place, and so on.
The first variation of the paradox, taken from Scriven's article, can be
called the paradox of the unexpected egg (see Figure 43.2).
Imagine that you have before you 10 boxes labeled from 1 to 10.
While your back is turned, a friend conceals an egg in one of the boxes.
You turn around. "I want you to open these boxes one at a time," he
tells you, "in serial order. Inside one of them I guarantee that you will
find an unexpected egg. By 'unexpected' I mean that you will not be
able to deduce which box it is in before you open the box and see it."
Assuming that your friend is absolutely trustworthy in all his state-
The Unexpected Hanging 569
Figure 43.2. The paradox of
the unexpected egg
ments, can his prediction be fulfilled? Apparently not. He obviously
will not put the egg in box 10, because after you have found the first
nine boxes empty you will be able to deduce with certainty that the egg
is in the only remaining box. This would contradict your friend's state-
ment. Box 10 is out. Now consider the situation that would arise if he
were so foolish as to put the egg in box 9. You find the first eight boxes
empty. Only 9 and 10 remain. The egg cannot be in box 10. Ergo it
must be in 9. You open 9. Sure enough, there it is. Clearly it is an ex-
pected egg, and so your friend is again proved wrong. Box 9 is out. But
now you have started on your inexorable slide into unreality. Box 8 can
be ruled out by precisely the same logical argument, and similarly
boxes 7, 6, 5, 4, 3, 2, and 1. Confident that all 10 boxes are empty, you
start to open them. What have we here in box 5? A totally unexpected
egg! Your friend's prediction is fulfilled after all. Where did your rea-
soning go wrong?
To sharpen the paradox still more, we can consider it in a third form,
one that can be called the paradox of the unexpected spade (see Figure
43.3). Imagine that you are sitting at a card table opposite a friend who
shows you that he holds in his hand the 13 spades. He shuffles them,
fans them with the faces toward him, and deals a single card face down
on the table. You are asked to name slowly the 13 spades, starting with
the ace and ending with the king. Each time you fail to name the card
on the table he will say no. When you name the card correctly, he will
say yes.
"I'll wager a thousand dollars against a dime," he says, "that you
570 LOGIC AND PHILOSOPHY
Figure 43.3. The paradox of
the unexpected spade
will not be able to deduce the name of this card before I respond with
'Yes.' "
Assuming that your friend will do his best not to lose his money, is
it possible that he placed the king of spades on the table? Obviously
not. After you have named the first 12 spades, only the king will re-
main. You will be able to deduce the card's identity with complete
confidence. Can it be the queen? No, because after you have named the
jack only the king and queen remain. It cannot be the king, so it must
be the queen. Again, your correct deduction would win you $1,000.
The same reasoning rules out all the remaining cards. Regardless of
what card it is, you should be able to deduce its name in advance. The
logic seems airtight. Yet it is equally obvious, as you stare at the back
of the card, that you have not the foggiest notion which spade it is!
Even if the paradox is simplified by reducing it to two days, two
boxes, two cards, something highly peculiar continues to trouble the
situation. Suppose your friend holds only the ace and deuce of spades.
It is true that you will be able to collect your bet if the card is the deuce.
Once you have named the ace and it has been eliminated you will be
able to say: "I deduce that it's the deuce." This deduction rests, of
course, on the truth of the statement "The card before me is either the
The Unexpected Hanging 571
ace or the deuce of spaces." (It is assumed by everybody, in all three
paradoxes, that the man will be hanged, that there is an egg in a box,
that the cards are the cards designated.) This is as strong a deduction
as mortal man can ever make about a fact of nature. You have, therefore,
the strongest possible claim to the $1,000.
Suppose, however, your friend puts down the ace of spades. Cannot
you deduce at the outset that the card is the ace? Surely he would not
risk his $1,000 by putting down the deuce. Therefore it must be the ace.
You state your conviction that it is. He says yes. Can you legitimately
claim to have won the bet?
Curiously, you cannot, and here we touch on the heart of the mystery.
Your previous deduction rested only on the premise that the card was
either the ace or the deuce. The card is not the ace; therefore, it is the
deuce. But now your deduction rests on the same premise as before
plus an additional one, namely, on the assumption that your friend
spoke truly; to say the same thing in pragmatic terms, on the assump
tion that he will do all he can to avoid paying you $1,000. But if it is
possible for you to deduce that the card is the ace, he will lose his
money just as surely as if he put down the deuce. Since he loses it ei-
ther way, he has no rational basis for picking one card rather than the
other. Once you realize this, your deduction that the card is the ace
takes on an extremely shaky character. It is true that you would be wise
to bet that it is the ace, because it probably is, but to win the bet you
have to do more than that: you have to prove that you have deduced the
card with iron logic. This you cannot do.
You are, in fact, caught up in a vicious circle of contradictions. First
you assume that his prediction will be fulfilled. On this basis you de
duce that the card on the table is the ace. But if it is the ace, his pre-
diction is falsified. If his prediction cannot be trusted, you are left
without a rational basis for deducting the name of the card. And if you
cannot deduce the name of the card, his prediction will certainly be
confirmed. Now you are right back where you started. The whole cir- .
cle begins again. In this respect the situation is analogous to the vicious
circularity involved in a famous card paradox first proposed by the
English mathematician P.E.B. Jourdain in 1913 (see Figure 43.4). Since
this sort of reasoning gets you no further than a dog gets in chasing its
tail, you have no logical way of determining the name of the card on the
table. Of course, you may guess correctly. Knowing your friend, you
may decide that it is highly probable he put down the ace. But no self-
572 LOGIC AND PHILOSOPHY
SENTENCE
ON THE OTHER SIDE
OF THIS CARD
ISTRUE
THE OTHER SIDE
/'"

IS FALSE
Figure 43.4. P.E.B. Jourdain's
card paradox
respecting logician would agree that you have "deduced" the card with
anything close to the logical certitude involved when you deduced that
it was the deuce.
The flimsiness of your reasoning is perhaps seen more clearly if you
return to the 10 boxes. At the start you "deduce" that the egg is in box
1, but box 1 is empty. You then "deduce" it to be in box 2, but box 2 is
empty also. Then you "deduce" box 3, and so on. (It is almost as if the
egg, just before you look into each box in which you are positive it
must be, were cleverly transported by secret trap doors to a box with a
higher number!) Finally you find the "expected" egg in box 8. Can you
maintain that the egg is truly "expected" in the sense that your deduc-
tion is above reproach? Obviously you cannot, because your seven pre-
vious "deductions" were based on exactly the same line of reasoning,
and each proved to be false. The plain fact is that the egg can be in any
box, including the last one.
Even after having opened nine empty boxes, the question of whether
you can "deduce" that there is an egg in the last box has no unam-
biguous answer. If you accept only the premise that one of the boxes
contains an egg, then of course an egg in box 10 can be deduced. In that
case, it is an expected egg and the assertion that it would not be is
proved false. If you also assume that your friend spoke truly when he
said the egg would be unexpected, then nothing can be deduced, for
the first premise leads to an expected egg in box 10 and the second to
an unexpected egg. Since nothing can be deduced, an egg in box 10
will be unexpected and both premises will be vindicated, but this vin-
dication cannot come until the last box is opened and an egg is found
there.
The Unexpected Hanging 573
We can sum up this resolution of the paradox, in its hanging form, as
follows. The judge speaks truly and the condemned man reasons
falsely. The very first step in his chain of reasoning-that he cannot be
hanged on the last day-is faulty. Even on the evening of the next-to-
last day, as explained in the previous paragraph with reference to the
egg in the last box-he has no basis for a deduction. This is the main
point of QUine's 1953 paper. In Quine'S closing words, the condemned
man should reason: "We must distinguish four cases: first, that I shall
be hanged tomorrow noon and I know it now (but I do not); second, that
I shall be unhanged tomorrow noon and know it now (but I do not);
third, that I shall be unhanged tomorrow noon and do not know it now;
and fourth, that I shall be hanged tomorrow noon and do not know it
now. The latter two alternatives are the open possibilities, and the last
of all would fulfill the decree. Rather than charging the judge with self-
contradiction, therefore, let me suspend judgment and hope for the
best."
The Scottish mathematician Thomas H. O'Beirne, in an article with
the somewhat paradoxical title "Can the Unexpected Never Happen?"
(The New Scientist, May 25, 1961), has given what seems to me an ex-
cellent analysis of this paradox. As O'Beirne makes clear, the key to re-
solving the paradox lies in recognizing that a statement about a future
event can be known to be a true prediction by one person but not
known to be true by another until after the event. It is easy to think of
simple examples. Someone hands you a box and says: "Open it and you
will find an egg inside." He knows that his prediction is sound, but you
do not know it until you open the box.
The same is true in the paradox. The judge, the man who puts the egg
in the box, the friend with the 13 spades-each knows that his predic-
tion is sound. But the prediction cannot be used to support a chain of
arguments that results eventually in discrediting the prediction itself.
It is this roundabout self-reference that, like the sentence on the face of
jourdain's card, tosses the monkey wrench into all attempts to prove
the prediction unsound.
We can reduce the paradox to its essence by taking a cue from
Scriven. Suppose a man says to his wife: "My dear, I'm going to sur-
prise you on your birthday tomorrow by giving you a completely un-
expected gift. You have no way of guessing what it is. It is that gold
bracelet you saw last week in Tiffany's window."
What is the poor wife to make of this? She knows her husband to be
574 LOGIC AND PHILOSOPHY
truthful. He always keeps his promises. But if he does give her the gold
bracelet, it will not be a surprise. This would falsify his prediction.
And if his prediction is unsound, what can she deduce? Perhaps he
will keep his word about giving her the bracelet but violate his word
that the gift will be unexpected. On the other hand, he may keep his
word about the surprise but violate it about the bracelet and give her
stead, say, a new vacuum cleaner. Because of the
ter of her husband's statement, she has no rational basis for choosing
between these alternatives; therefore, she has no rational basis for
pecting the gold bracelet. It is easy to guess what happens. On her birth-
day she is surprised to receive a logically unexpected bracelet.
He knew all along that he could and would keep his word. She could
not know this until after the event. A statement that yesterday appeared
to be nonsense, that plunged her into an endless whirlpool of logical
contradictions, has today suddenly been made perfectly true and non-
contradictory by the appearance of the gold bracelet. Here in the
est possible form is the queer verbal magic that gives to all the
paradoxes we have discussed their bewildering, charm.
Addendum
A great many trenchant and sometimes bewildering letters were
received from readers offering their views on how the paradox of the
unexpected hanging could be resolved. Several went on to expand their
views in articles that are listed in the bibliography for this chapter.
Lennart Ekbom, a mathematics teacher at Ostermalms College, in
Stockholm, pinned down what may be the origin of the paradox. In
1943 or 1944, he wrote, the Swedish Broadcasting Company an-
nounced that a civil-defense exercise would be held the following
week, and to test the efficiency of civil-defense units, no one would be
able to predict, even on the morning of the day of the exercise, when it
would take place. Ekbom realized that this involved a logical paradox,
which he discussed with some students of mathematics and philoso-
phy at Stockholm University. In 1947 one of these students visited
Princeton, where he heard Kurt G6del, the famous mathematician,
mention a variant of the paradox. Ekbom adds that he originally
lieved the paradox to be older than the Swedish an-
nouncement, but in view of Quine's statement that he first heard of the
paradox in the early forties, perhaps this was its origin.
The Unexpected Hanging 575
Papers on the unexpected hanging paradox continue to proliferate. I
have greatly lengthened the chapter's bibliography to catch all the ref-
erences in English that have come to my attention since 1969. It is
amazing that philosophers continue to be bemused by this paradox
and are still unable to agree on how best to resolve it.
The paradox has many variants. Douglas Hofstadter suggested that
instead of an unexpected egg in a box there be an unexpected deadly
snake. Roy Sorensen, in his 1982 paper, offered three versions in which
a time sequence is not involved: one concerning what are called
Moorean sentences, another about moves on a 3 x 3 game matrix, and
a third based on a chain of sacrificial virgins.
Bibliography
The following bibliography has been updated by references
taken from an extensive bibliography, much more complete than mine,
that has been prepared by Timothy Chow. You can download his entire
bibliography from his website: http://front.math.ucdavis.edu/search/
author:Chow+and+ Timothy.
P. Alexander, "Pragmatic Paradoxes," Mind, Vol. 59, October 1950, pp. 536-38.
A. K. Austin, "On the Unexpected Examination," Mind, VoL 78, January 1969, p.
137.
A. K. Austin, "The Unexpected Examination," Analysis, VoL 39, January 1979, pp.
63-64.
A. J. Ayer, "On a Supposed Antimony," Mind, VoL 82, January 1973, pp. 125-26.
R. Binkley, "The Surprise Examination in Modal Logic," Journal of Philosophy, Vol.
65, 1968,pp. 127-36.
J. Bosch, "The Examination Paradox and Formal Prediction," Logique et Analyse,
VoL 59, September/December 1972.
L. Bovens, "The Backward Induction Argument for the Finite Iterated Prisoner's
Dilemma and the Surprise Examination Paradox," Analysis, Vol. 57, 1997, pp.
179-86.
J. Cargile, "The Surprise Test Paradox," Journal of Philosophy, Vol. 64, September
1967, pp. 550-63.
J. Case, "Paradoxes Involving Conflicts of Interest," American Mathematical
Monthly, VoL 107, January 2000, pp. 33-43.
T. S. Chaplain, "QUine's Judge," Philosophical Studies, Vol. 29, May 1976, pp.
349-52.
J. M. Chapman and R. J. Butler, "On Quine's 'So-called Paradox,' " Mind, Vol. 74,
July 1965, pp. 424-25.
576 LOGIC AND PHILOSOPHY
T. Y. Chow, "The Surprise Examination or Unexpected Hanging Paradox," Ameri-
can Mathematical Monthly, Vol. 105, January 1998, pp. 41-51.
D. Clark, "How Unexpected is the Unexpected Hanging?" Mathematics Magazine,
Vol. 67, February 1994, pp. 55-58.
R. Clark, "Pragmatic Paradox and Rationality," Canadian Journal of Philosophy,
Vol. 24, 1994,pp.229-42.
L. J. Cohen, "Mr. O'Connor's 'Pragmatic Paradoxes,' " Mind, Vol. 59, January 1950,
pp.85-87.
F. B. Ebersole, "The Definition of 'Pragmatic Paradox,' " Mind, Vol. 62, January
1953, pp. 80-85.
M. Edman, "The Prediction Paradox," Theoria, Vol. 40, 1974, pp. 166-75.
K. G. Ferguson, "Equivocation in the Surprise Examination Paradox," Southern
Journal of Philosophy, Vol. 29, 1991, pp. 291-302.
F. B. Fitch, "A Goedelized Formulation of the Prediction Paradox," American Philo-
sophical Quarterly, Vol. 1, 1964, pp. 161-64.
J. T. Fraser, "Note Relating to a Paradox of the Temporal Order," The Voices of
Time, J. T. Fraser (ed.), New York: Braziller, 1966, pp. 524-26, 679.
J. S. Fulda, "The Paradox of the Surprise Test," Mathematical Gazette, Vol. 75, De-
cember 1991, pp. 419-23.
M. Gardner, "The Erasing of Philbert the Fudger," Science Fiction Puzzle Tales,
New York: Clarkson Potter, 1981, Chapter 20.
M. Gardner, "Again, How's That Again?," Riddles of the Sphinx, New York: Math-
ematical Association of America, 1987, Chapter 27.
1. Goldstein, "Inescapable Surprises and Aquirable Intentions," Analysis, Vol. 53,
1993, pp. 93-99.
N. Hall, "How to Set a Surprise Exam," Mind, Vol. 108, 1999, pp. 647-703.
J. Y. Halpern and Y. Moses, "Taken by Surprise: The Paradox of the Surprise Test
Revisited," Journal of Philosophical Logic, Vol. 15, 1986, pp. 281-304.
C. Harrison, "The Unexpected Examination in View of Kripne's Semantics for
Modal Logic," Philosophical Logic, J. W. Davis et al. (eds.), Holland: Reidel,
1969.
J. M. Holtzman, "An Undecidable Aspect of the Unexpected Hanging Problem,"
Philosophia, Vol. 72, March 1987, pp. 195-98.
J. M. Holtzman, "A Note on Schrodinger's Cat and the Unexpected Examination
Paradox," British Journal of the Philosophy of Science, Vol. 39, 1988, pp.
397-401.
C. Hudson, A. Solomon, and R. Walker, "A Further Examination of the 'Surprise
Examination Paradox,' " Eureka, October 1969, pp. 23-24.
B. Jangeling and T. Koetsier, "A Reappraisal of the Hanging Paradox," Philosophia,
Vol. 22,1993,pp. 199-311.
C. Janoway, "Knowing About Surprises: A Supposed Antinomy Revisited," Mind,
Vol. 98,1989,pp. 391-409.
D. Kaplan and R. Montague, "A Paradox Regained," Notre Dame Journal of Formal
Logic, Vol. 1, July 1960, pp. 79-90.
The Unexpected Hanging 577
J. Kiefer and J. Ellison, "The Prediction Paradox Again," Mind, Vol. 74, July 1965,
pp.426-27.
P. L. Kirkham, "On Paradoxes and the Surprise Examination," Philosophia, Vol. 21,
1991, pp. 31-51.
I. Kvart, "The Paradox of the Surprise Examination," Logique et Analyse, Vol. 11,
1976, pp. 66-72.
A. Lyon, "The Prediction Paradox," Mind, Vol. 68, October 1959, pp. 510-17.
A. Margalit and M. Bar-Hillel, "Expecting the Unexpected," Philosophia, Vol. 18,
October 1982, pp. 263-89.
J. McLelland, "Epistemic Logic and the Paradox of the Surprise Examination Para-
dox," International Logic Review, Vol. 3, January 1971, pp. 69-85.
J. McLelland and C. Chihara, "The Surprise Examination Paradox," Journal of
Philosophical Logic, Vol. 4, February 1975, pp. 71-89.
B. Medlin, "The Unexpected Examination," American Philosophical Quarterly,
Vol. 1, January 1964, pp. 1-7.
B. Meltzer, "The Third Possibility," Mind, Vol. 73, July 1964, pp. 430-33.
B. Meltzer and I. J. Good, "'IWo Forms of the Prediction Paradox," British Journal
for the Philosophy of Science, Vol. 16, May 1965, pp. 50-51.
G. C. Nerlich, "Unexpected Examinations and Unprovable Statements," Mind, Vol.
70, October 1961, pp. 503-13.
T. H. O'Beirne, "Can the Unexpected Never Happen?," The New Scientist, Vol. 15,
May 25, 1961, pp. 464-65; letters and replies, June 8, 1961, pp. 597-98.
D. J. O'Connor, "Pragmatic Paradoxes," Mind, Vol. 57, July 1948, pp. 358-59.
D. J. O'Connor, "Pragmatic Paradoxes and Fugitive Propositions," Mind, Vol. 60,
October 1951, pp. 536-38.
D. Olin, "The Prediction Paradox Resolved," Philosophical Studies, Vol. 44, Sep-
tember 1983, pp. 225-33.
W. Poundstone, "The Unexpected Hanging," Labyrinths of Reason, New York: Dou-
bleday, 1988, Chapter 6.
W. V. Quine, "On a So-called Paradox," Mind, Vol. 62, January 1953, pp. 65-67.
J. Schoenberg, "A Note on the Logical Fallacy in the Paradox of the Unexpected Ex-
amination," Mind, Vol. 75, January 1966, pp. 125-27.
M. Scriven, "Paradoxical Announcements," Mind, Vol. 60, July 1951, pp. 403-7.
S. C. Shapiro, "A Procedural Solution to the Unexpected Hanging and Sorites Para-
doxes," Mind, Vol. 107, 1998, pp. 751-61.
R. A. Sharpe, "The Unexpected Examination," Mind, VoL 74, April 1965, p. 255.
R. Shaw, "The Paradox of the Unexpected Examination," Mind, Vol. 67, July 1958,
pp.382-84.
B. H. Slater, "The Examiner Examined," AnalysiS, Vol. 35, December 1974, pp.
48-50.
J. W. Smith, "The Surprise Examination on the Paradox of the Heap," Philosophi-
cal Papers, Vol. 13, May 1984, pp. 43-56.
R. Smullyan, "Surprised?," Forever Undecided, New York: Knopf, 1987, Chapter 2.
R. A. Sorensen, "Recalcitrant Variations of the Prediction Paradox," Australasian
Journal of Philosophy, Vol. 60, December 1982, pp. 355-62.
578 LOGIC AND PHILOSOPHY
R. A. Sorensen, "Conditional Blindspots and the Knowledge Squeeze: A Solution
to the Prediction Paradox," Australasian Journal of Philosophy, Vol. 62, June
1984, pp. 126-35.
R. A. Sorensen, "The Bottle Imp and the Prediction Paradox," Philosophia, Vol. 15,
January 1986, pp. 421-24.
R. A. Sorensen, "A Strengthened Prediction Paradox," Philosophical Quarterly,
Vol. 36, October 1986, pp. 504-13.
R. A. Sorensen, Blindspots, London: Oxford University Press, 1988, Chapters 7-9.
R. A. Sorenson, "The Earliest Unexpected Class Inspection," Analysis, Vol. 53,
1993, p. 252.
R. A. Sorenson, "Infinite 'Backward' Induction Arguments," Pacific Philosophical
Quarterly, Vol. 80, 1999, pp. 278-83.
M. Stack, "The Surprise Examination Paradox," Dialogue, Vol. 26, June 1977.
I. Stewart, "Paradox Lost," Manifold, Autumn 1971, pp. 19-21.
R. Weintraub, "Practical Solutions to the Surprise Examination Paradox," Ratio
(New Series), Vol. 8, 1995, pp. 161-69.
P. Weiss, "The Prediction Paradox," Mind, Vol. 61, April 1952, pp. 265-69.
P. Y. Windt, "The Liar and the Prediction Paradox," American Philosophical Quar-
terly, Vol. 10, January 1973, pp. 65-68.
D. R. Woodall, "The Paradox of the Surprise Examination," Eureka, No. 30, Octo-
ber 1967, pp. 31-32.
C. Wright and A. Sudbury, "The Paradox of the Unexpected Examination," Aus-
tralasian Journal of Philosophy, May 1977, pp. 41-58.
J. A. Wright, "The Surprise Exam: Prediction on Last Day Uncertain," Mind, Vol.
76, January 1967, pp. 115-17.
The Unexpected Hanging 579
Chapter 44 Newcomb's Paradox
A common opinion prevails that the juice has
ages ago been pressed out of the free-will con-
troversy, and that no new champion can do
more than warm up stale arguments which
every one has heard. This is a radical mis-
take. I know of no subject less worn out, or in
which inventive genius has a better chance of
breaking open new ground. -WILLIAM JAMES
One of the perennial problems of philosophy is how to ex-
plain (or explain away) the nature of free will. If the concept is expli-
cated within a framework of determinism, the will ceases to be free in
any commonly understood sense, and it is hard to see how fatalism can
be avoided. Che sara, sara. Why work hard for a better future for your-
self or for others if what you do must always be what you do do? And
how can you blame anyone for anything ifhe could not have done oth-
erwise?
On the other hand, attempts to explicate will in a framework of in-
determinism seem equally futile. If an action is not caused by the pre-
vious states of oneself and the world, it is hard to see how to keep the
action from being haphazard. The notion that decisions are made by
some kind of randomizer in the mind does not provide much support
for what is meant by free will either.
Philosophers have never agreed on how to avoid the horns of this
dilemma. Even within a particular school there have been sharp dis-
agreements. William James and John Dewey, America's two leading
pragmatists, are a case in point. Although Dewey was a valiant de-
fender of democratic freedoms, his metaphysics regarded human be-
havior as completely determined by what James called the total "push
of the past." Free will for Dewey was as illusory as it is in the psy-
chology of B. F. Skinner. In contrast, James was a thoroughgoing inde-
terminist. He believed that minds had the power to inject genuine
novelty into history-that not even God himself could know the future
except partially. "That," he wrote, "is what gives the palpitating real-
ity to our moral life and makes it tingle ... with so strange and elabo-
rate an excitement."
580
A third approach, pursued in depth by Immanuel Kant, accepts both
sides of the controversy as being equally true but incommensurable
ways of viewing human behavior. For Kant the situation is something
like that pictured in one of Piet Hein's "grooks":
A bit beyond perception's reach
I sometimes believe I see
That Life is two locked boxes, each
Containing the other's key.
Free will is neither fate nor chance. In some unfathomable way it par-
takes of both. Each is the key to the other. It is not a contradictory con-
cept, like a square triangle, but a paradox that our experience forces on
us and whose resolution transcends human thought. That was how
Niels Bohr saw it. He found the situation similar to his "principle of
complementarity" in quantum mechanics. It is a viewpoint that Ein-
stein, a Spinozist, found distasteful, but many other physicists, J.
Robert Oppenheimer for one, found enormously attractive.
What has free will to do with mathematical games? The answer is
that in recent decades philosophers of science have been wrestling
with a variety of queer "prediction paradoxes" related to the problem
of wilL Some of them are best regarded as a game situation. One draws
a payoff matrix and tries to determine a player's best strategy, only to
find oneself trapped in a maze of bewildering ambiguities about time
and causality.
A marvelous example of such a paradox came to light in 1970 in the
paper "Newcomb's Problem and Two Principles of Choice" by Robert
Nozick, a philosopher at Harvard University. The paradox is so pro-
found, so amusing, so mind-bending, with thinkers so evenly divided
into warring camps, that it bids fair to produce a literature vaster than
that dealing with the prediction paradox of the unexpected hanging.
Newcomb's paradox is named after its originator, William A. New-
comb, a theoretical physicist at the University of California's Lawrence
Livermore Laboratory. (His great-grandfather was the brother of Simon
Newcomb, the astronomer.) Newcomb thought of the problem in 1960
while meditating on a famous paradox of game theory called the pris-
oner's dilemma. A few years later Newcomb's problem reached Nozick
by way of their mutual friend Martin David Kruskal, a Princeton Uni-
versity mathematician. "It is not clear that I am entitled to present this
paper," Nozick writes. "It is a beautiful problem. I wish it were mine."
Newcomb's Paradox 581
Although Nozick could not resolve it, he decided to write it up anyway.
His paper appears in Essays in Honor of Carl G. Hempel, edited by
Nicholas Rescher and published by Humanities Press in 1970. What
follows is largely a paraphrase of Nozick's paper.
Two closed boxes, B1 and B2, are on a table. B1 contains $1,000. B2
contains either nothing or $1 million. You do not know which. You
have an irrevocable choice between two actions:
1. Take what is in both boxes.
2. Take only what is in B2.
At some time before the test a superior Being has made a prediction
about what you will decide. It is not necessary to assume determinism.
You only need be persuaded that the Being's predictions are "almost
certainly" correct. If you like, you can think of the Being as God, but the
paradox is just as strong if you regard the Being as a superior intelli-
gence from another planet or a supercomputer capable of probing your
brain and making highly accurate predictions about your decisions. If
the Being expects you to choose both boxes, he has left B2 empty. If he
expects you to take only B2, he has put $1 million in it. (If he expects
you to randomize your choice by, say, flipping a coin, he has left B2
empty.) In all cases B1 contains $1,000. You understand the situation
fully. the Being knows you understand, you know that he knows, and
so on.
What should you do? Clearly it is not to your advantage to flip a
coin, so that you must decide on your own. The paradox lies in the dis-
turbing fact that a strong argument can be made for either decision.
Both arguments cannot be right. The problem is to explain why one is
wrong.
Let us look first at the argument for taking only B2. You believe the
Being is an excellent predictor. If you take both boxes, the Being almost
certainly will have anticipated your action and have left B2 empty.
You will get only the $1,000 in B1. On the other hand, if you take only
B2, the Being, expecting that, almost certainly will have placed $1 mil-
lion in it. Clearly it is to your advantage to take only B2.
Convincing? Yes, but the Being made his prediction, say a week ago,
and then left. Either he put the $1 million in B2, or he did not. "If the
money is already there, it will stay there whatever you choose. It is not
going to disappear. If it is not already there, it is not going to suddenly
appear if you choose only what is in the second box." It is assumed that
582 LOGIC AND PHILOSOPHY
no "backward causality" is operating; that is, your present actions can-
not influence what the Being did last week. So why not take both boxes
and get everything that is there? If B2 is filled, you get $1,001,000. If it
is empty, you get at least $1,000. If you are so foolish as to take only B2,
you know you cannot get more than $1 million, and there is even a
slight possibility of getting nothing. Clearly it is to your advantage to
take both boxes!
"I have put this problem to a large number of people, both friends
and students in class," writes Nozick. "To almost everyone it is per-
fectly clear and obvious what should be done. The difficulty is that
these people seem to divide almost evenly on the problem, with large
numbers thinking that the opposing half is just being silly.
"Given two such compelling opposing arguments, it will not do to
rest content with one's belief that one knows what to do. Nor will it do
to just repeat one of the arguments, loudly and slowly. One must also
disarm the opposing argument; explain away its force while showing
it due respect."
Nozick sharpens the "pull" of the two arguments as follows. Suppose
the experiment had been done many times before. In every case the
Being predicted correctly. Those who took both boxes always got only
$1,000; those who took only B2 got $1 million. You have no reason to
suppose your case will be different. If a friend were observing the
scene, it would be completely rational for him to bet, giving high odds,
that if you take both boxes you will get only $1,000. Indeed, if there is
a time delay after your choice of both boxes, you know it would be ra-
tional for you yourself to bet, offering high odds, that you will get only
$1,000. Knowing this, would you not be a fool to take both boxes?
Alas, the other argument makes you out to be just as big a fool if you
do not. Assume that Bl is transparent. You see the $1,000 inside. You
cannot see into B2, but the far side is transparent and your friend is sit-
ting opposite. He knows whether the box is empty or contains $1 mil-
lion. Although he says nothing, you realize that, whatever the state of
B2 is, he wants you to take both boxes. He wants you to because, re-
gardless of the state of B2, you are sure to come out ahead by $1,000.
Why not take advantage of the fact that the Being played first and can-
not alter his move?
Nozick, an expert on decision theory, approaches the paradox by
considering analogous game situations in which, as here, there is a
conflict between two respected principles of choice: the "expected-
Newcomb's Paradox 583
utility principle" and the "dominance principle." To see how the prin-
ciples apply, consider the payoff matrix for Newcomb's game (see Fig-
ure 44.1). The argument for taking only B2 derives from the principle
that you should choose so as to maximize the expected utility (value to
you) of the outcome. Game theory calculates the expected utility of
each action by multiplying each ofits mutually exclusive outcomes by
the probability of the outcome, given the action. We have assumed that
the Being predicts with near certainty, but let us be conservative and
make the probability a mere .9. The expected utility of taking both
boxes is
(.1 x $1,001,000) + (.9 x $1,000) = $101,000.
The expected utility of taking only B2 is
(,9 x $1,000,000) + (.1 x $0) = $900,000.
Guided by this principle, your best strategy is to take only the second
box.
BEING
MOVE 1 MOVE 2
(PREDICTS YOU (PREDICTS YOU
TAKE ONLY BOX 2) TAKE BOTH BOXES)
MOVE 1
(TAKE ONLY BOX 2) $1,000,000 $0
YOU
MOVE 2
(TAKE BOTH BOXES) $1,001,000 $1,000
Figure 44.1. Payoff matrix for Newcomb's paradox
The dominance principle, however, is just as intuitively sound. Sup-
pose the world divided into n different states. For each state k mutually
exclusive actions are open to you. If in at least one state you are better
off choosing a, and in all other states either a is the best choice or the
choices are equal, then the dominance principle asserts that you should
choose a. Look again at the payoff matrix. The states are the outcomes
of the Being's two moves. Taking both boxes is strongly dominant. For
each state it gives you $1,000 more than you would get by taking only
the second box.
That is as far as we can go into Nozick's analysis, but interested read-
584 LOGIC AND PHILOSOPHY
ers should look it up for its mind-boggling conflict situations related to
Newcomb's problem. Nozick finally arrives at the following tentative
conclusions:
If you believe in absolute determinism and that the Being has in truth
predicted your behavior with unswerving accuracy, you should
"choose" (whatever that can mean!) to take only B2. For example, sup-
pose the Being is God and you are a devout Calvinist, convinced that
God knows every detail of your future. Or assume that the Being has a
time-traveling device he can launch into the future and use to bring
back a motion picture of what you did on that future occasion when
you made your choice. Believing that, you should take only B2, firmly
persuaded that your feeling of having made a genuine choice is sheer
illusion.
Nozick reminds us, however, that Newcomb's paradox does not as-
sume that the Being has perfect predictive power. If you believe that
you possess a tiny bit of free will (or alternatively that the Being is
sometimes wrong, say once in every 20 billion cases), then this may be
one of the times the Being has erred. Your wisest decision is to take
both boxes.
Nozick is not happy with this conclusion. "Could the difference be-
tween one in n and none in n, for arbitrarily large finite n, make this dif-
ference? And how exactly does the fact that the predictor is certain to
have been correct dissolve the force of the dominance argument?" Both
questions are left unanswered. Nozick hoped that publishing the prob-
lem might "call forth a solution which will enable me to stop returning,
periodically, to it."
One such solution, "to restore [Nozick's] peace of mind," was at-
tempted by Maya Bar-Hillel and Avishai Margalit, of Hebrew University
in Jerusalem, in their paper "Newcomb's Paradox Revisited." They
adopt the same game-theory approach taken by Nozick, but they come
to an opposite conclusion. Even though the Being is not a perfect pre-
dictor, they recommend taking only the second box. You must, they
argue, resign yourself to the fact that your best strategy is to behave as
iithe Being has made a correct prediction, even though you know there
is a slight chance he has erred. You know he has played before you, but
you cannot do better than to playas if he is going to play after you. "For
you cannot outwit the Being except by knowing what he predicted,
but you cannot know, or even meaningfully guess, at what he predicted
before actually making your final choice."
Newcomb's Paradox 585
It may seem to you, Bar-Hillel and Margalit write, that backward
causality is operating-that somehow your choice makes the $1 million
more likely to be in the second box-but this is pure flim-flam. You
choose only B2 "because it is inductively known to correlate remark-
ably with the existence of this sum in the box, and though we do not
assume a causal relationship, there is no better alternative strategy than
to behave as if the relationship was, in fact, causal."
For those who argue for taking only B2 on the grounds that causality
is independent of the direction of time-that your decision actually
"causes" the second box to be either empty or filled with $1 million-
Newcomb proposed the following variant of his paradox. Both boxes
are transparent. Bl contains the usual $1,000. B2 contains a piece of
paper with a fairly large integer written on it. You do not know whether
the number is prime or composite. If it proves to be prime (you must
not test it, of course, until after you have made your choice), then you
get $1 million. The Being has chosen a prime number if he predicts you
will take only B2 but has picked a composite number if he predicts you
will take both boxes.
Obviously you cannot by an act of will make the large number
change from prime to composite, or vice versa. The nature of the num-
ber is fixed for eternity. So why not take both boxes? If it is prime, you
get $1,001,000. If it is not, you get at least $1,000. (Instead of a number,
B2 could contain any statement of a decidable mathematical fact that
you do not investigate until after your choice.)
It is easy to think of other variations. For example, there are 100 lit-
tle boxes, each holding a $10 bill. If the Being expects you to take all
of them, he has put nothing else in them. But if he expects you to take
only one box-perhaps you pick it at random-he has added to that box
a large diamond. There have been thousands of previous tests, half of
them involving you as a player. Each time, with possibly a few excep-
tions, the player who took a single box got the diamond, and the player
who took all the boxes got only the money. Acting pragmatically, on the
basis of past experience, you should take only one box. But then how
can you refute the logic of the argument that says you have everything
to gain and nothing to lose if the next time you play you take all the
boxes?
These variants add nothing essentially new. With reference to the
original version, Nozick halfheartedly recommends taking both boxes.
Bar-Hillel and Margalit strongly urge you to "join the millionaire's
586 LOGIC AND PHILOSOPHY
club" by taking only B2. That is also the view of Kruskal and Newcomb.
But has either side really done more than just repeat its case "loudly
and slowly"? Can it be that Newcomb's paradox validates free will by
invalidating the possibility, in principle, of a predictor capable of
guessing a person's choice between two equally rational actions with
better than 50 percent accuracy?
Addendum
Scientific American received such a flood of letters about New-
comb's paradox that I asked Robert Nozick if he would be willing to
look them over and write a guest column about them. To my delight, he
agreed. I packed a large carton with the correspondence and took it
along on a visit to Cambridge where I had the pleasure of lunching
with Nozick and giving him the carton. You will find his trenchant
discussion of the letters in my Knotted Doughnuts and Other Mathe-
matical Entertainments. Nozick concluded that the paradox remains
unsolved.
In 1974 Basic Books published Nozick's controversial defense of po-
liticallibertarianism, Anarchy, State, and Utopia. It was followed by
Philosophical Explanations and other books that have elevated Nozick
into the ranks of major U.S. philosophers.
As my annotated bibliography indicates, philosophers are still far
from agreement on how to resolve Newcomb's paradox. For whatever
they are worth, here are my own tentative opinions.
My sympathies are with those who say the predictor cannot exist.
Even if strict determinism in some sense holds for every event in the
history of the universe, I believe that certain events are in principle un-
predictable when predictions are allowed to interact causally with the
event being predicted. We have here, I am persuaded, something anal-
ogous to the resolution of semantic paradoxes. Contradictions arise
whenever a language is allowed to talk about the truth or falsity of its
own statements, or when sets are allowed to be members of themselves.
We can escape the semantic paradoxes by permitting talk about the
truth of a sentence only in a metalanguage. "This sentence is false"
simply is not a sentence. The notorious paradox of the barber who
shaves every person and only those persons who do not shave them-
selves, and who himself belongs to the set of persons, is a barber who
cannot exist. It is not logically inconsistent to suppose that the future
Newcomb's Paradox 587
is totally determined, whether or not an omniscient God exists, but as
soon as we permit a superbeing to make predictions that interact with
the event being predicted, we encounter contradictions that render the
existence of such a superpredictor impossible.
Consider the simplest case. A superbeing knows that when you go to
bed next Thursday you will take off your shoes. If the superbeing keeps
this knowledge from you, there is no problem; but if the superbeing in-
forms you of the prediction, you can falsify it easily by going to bed
with your shoes on. I agree with those who say that Newcomb's prob-
lem in no way settles the question of whether the future is completely
determined, but I do maintain that it brings us face to face with the eter-
nal, and to me unanswerable, problem of defining what is meant by free
choice.
Although I don't believe it, the state of the world a hundred years
from now may be determined in every detail by the state of the world
now. Innumerable future events obviously can be predicted with al-
most certain accuracy, but other events are the outcome of such complex
causes that even if determinism is true it seems likely there is no pos-
sible way they could be predicted by any technique faster than allow-
ing the universe itself to unroll to see what happens. (We leave aside the
notion of a God outside of time who sees the past and future simulta-
neously, whatever that means.) All this is by the way. The main point
is that when a prediction interacts with the predicted event, whether
human wills are involved or not, logical contradictions can arise. A fa-
miliar example is the supercomputer asked to predict if a certain event
will occur in the next three minutes. If the prediction is no, it turns on
a green light. If yes, it turns on a red light. The computer is now asked
to predict whether the green light will go on. By making the event part
of the prediction, the computer is rendered logically impotent.
It is my view that Newcomb's predictor, even if accurate only 51 per-
cent of the time, forces a logical contradiction that makes such a pre-
dictor, like Bertrand Russell's barber, impossible. We can avoid
contradictions arising from two different "shoulds" (should you take
one or two boxes?) by stating the contradiction as follows. One flawless
argument implies that the best way to maximize your reward is to take
only the closed box. Another flawless argument implies that the best
way to maximize your reward is to take both boxes. Because the two
conclusions are contradictory, the predictor cannot exist. Faced with a
Newcomb decision, I would share the suspicions of Max Black and
588 LOGIC AND PHILOSOPHY
others that I was either the victim of a hoax or of a badly controlled ex-
periment that had yielded false data about the predictor's accuracy. On
this assumption, I would take both boxes.
But, you may ask, how would I decide if I made what I would regard
as a counterfactual posit that the predictor was what it was claimed to
be? I suppose if I could persuade myself that the predictor existed I
might take only the closed box even though it would be logically irra-
tional. But I cannot so persuade myself. It is as if someone asked me to
put 91 eggs in 13 boxes, so each box held seven eggs, and then added
that an experiment had proved that 91 is prime. On that assumption,
one or more eggs would be left over. I would be given a million dollars
for each leftover egg, and 10 cents ifthere were none. Unable to believe
that 91 is a prime, I would proceed to put seven eggs in each box, take
my 10 cents and not worry about having made a bad decision.
Bibliography
M. Bar-Hillel and A. Margalit, "Newcomb's Paradox Revisited," The British Jour-
nal for the Philosophy of Science, Vol. 23, 1972, pp. 295-304. Defends the prag-
matic decision. "We hope to convince the reader to take just the one covered box,
and join the millionaire's club!"
T. M. Benditt and D. J. Ross "Newcomb's 'Paradox,' " The British Journal for the
Philosophy of Science, Vol. 27, 1976, pp.161-64. Attack's Schlesinger's reason-
ing and concludes that "there is only one rational choice ... namely choose box
II alone."
M. Black, "Newcomb's Problem Demystified," The Prevalence of Humbug and
Other Essays, Cornell University Press, 1983, Chapter 8. Argues that the exis-
tence ofthe predictor is so impossible that anyone faced with a Newcomb deci-
sion "would be well advised to suspect fraud, and to play safe by taking both
boxes."
S. J. Brams, "Newcomb's Problem and the Prisoner's Dilemma," The Journal of
Conflict Resolution, Vol. 19, 1975, pp. 596-619. Argues the close relationship of
Newcomb's problem to the prisoner's dilemma. Applying metagame theory,
Brams maintains that if you believe empirical evidence indicates the predic-
tor's accuracy is greater than .5005, you should take only one box; if less than
.5005, you should be indifferent as to your choice.
S. J. Brams, "A Paradox of Prediction," Paradoxes in Politics, Free Press, 1976,
Chapter 8. Enlarges on the previous entry.
S. J. Brams "Newcomb's Problem," Superior Beings, Springer-Verlag, 1983, pp.
46-54. Summarizes his earlier views.
J. Broome, "An Economic Newcomb Problem," Analysis, Vol. 49, October 1989, pp.
220-22.
Newcomb's Paradox 589
R. Campbell and L. Sowden (eds.), Paradoxes of Rationality and Cooperation: Pris-
oner's Dilemma and Newcomb's Problem, University of British Columbia Press,
1985.
J. Cargile, "Newcomb's Paradox," The British Journal for the Philosophy of Science,
Vol. 26, 1975, pp. 234-39. Contends that there is "no basis for determining one
course of action as the right one without additional information."
E. Eells, "Newcomb's Paradox and the Principle of Maximizing Conditional Ex-
pected Utility," Ph.D. dissertation, University of California, Berkeley, June 1980.
E. Eells, Rational Decision and Causality, Cambridge University Press, 1982.
E. Eells, "Newcomb's Many Solutions," Theory and Decision, Vol. 16, 1984, pp.
59-105.
E. Eells, "Levi's 'The Wrong Box,' " The Journal of Philosophy, Vol. 82, 1985, pp.
91-106.
A. Gibbard and W. Harper, "Counterfactuals and Two Kinds of Expected Utility,"
Foundations and Applications of Decision Theory, Reidel, 1978. Reprinted in Ifs,
Reidel, 1981.
T. Horgan, "Counterfactuals and Newcomb's Problem," The Journal of Philosophy,
Vol. 78, 1981, pp. 331-56.
N. Howard, Paradoxes of Rationality: Theory of Metagames and Political Behavior,
MIT Press, 1971, pp. 168-84.
J. Leslie, "Ensuring Two Bird Deaths With One Throw," Mind, Vol. 100, January
1991, pp. 73-86.
D. Lewis, "Prisoner's Dilemma Is a Newcomb Problem," Philosophy and Public Af-
fairs, Vol. 8, 1979, pp. 235-40.
I. Levi "Newcomb's Many Problems," Theory and Decision, Vol. 6, 1975, pp.
161-75.
1. Levi, "A Note on Newcombmania," The Journal of Philosophy, Vol. 79, 1982, pp.
337-42.
1. Levi "The Wrong Box," The Journal of Philosophy, Vol. 80, 1983, pp. 534-42.
D. Locke, "How to Make a Newcomb Choice," Analysis, January 1978, pp. 17-23.
Takes the view that the situation is not sharply enough defined to make decision
theory applicable. It is best to take both boxes,
confident that whether Box One is full or empty I cannot suffer from that choice,
hopeful that I might at least demonstrate freedom of choice in the presence of a
Newcomb Predictor, but resigned to receive a mere thousand where less ratio-
nal mortals, who from the evidence of Scientific American (March 1974, p. 102)
outnumber the rational by the order of 5 to 2, stand to gain a million. Perhaps it
is precisely because I am a rational man that the Predictor is able to predict my
choice, and leave Box One empty. Perhaps, in this respect at least, I would be
better off were I less rational than I am. The penalties of philosophy are no less
than its pleasures.
R. Nozick, "Newcomb's Problem and Two Principles of Choice," Essays in Honor
of Carl G. Hempel, N. Rescher (ed.), Humanities Press, 1969.
590 LOGIC AND PHILOSOPHY
W. Poundstone, Labyrinths of Reason, Doubleday, 1988, Chapter 12.
R. Richter, "Rationality Revisited," The Australasian Journal of Philosophy, Vol. 62,
1984,pp.392-403.
G. Schlesinger, "The Unpredictability of Free Choices," The Bl'itish Journal for the
Philosophy of Science, Vol. 25, 1974, pp. 209-21. Argues that no amount of in-
ductive evidence can support the belief that the predictor is capable of better
than chance in its predictions and that the paradox brings out the fundamental
unpredictability of human choices and the reality of free will. It is best to take
both boxes.
G. Schlesinger, "Unpredictability: A Reply to Cargile and to Benditt and Ross," The
British Journal of the Philosophy of Science, Vol. 27, 1976, pp. 267-74. The au-
thor defends his earlier position.
R. A. Sorensen, "Newcomb's Problem: Recalculations for the One-Boxer," Theory
and Decision, Vol. 15, 1983, pp. 399-404.
R. A. Sorensen, "The Iterated Versions of Newcomb's Problem and the Prisoner's
Dilemma," Synthese, VoL 63,1985, pp. 157-66.
R. A, Sorenson, Blind Spots, Oxford, 1988, pp. 371-77.
F. A. Wolf, "Newcomb's Paradox," Taking the Quantum Leap, Harper and Row,
1981. The funniest attempt at resolution yet, by a physicist deep into the para-
normal and a defender of the solipsistic view that the world "out there" really
isn't there until it is created by your observation. Since the money is not in the
closed box until you perceive it, you have to put it there by choosing and open-
ing the box. "It is your act of observation that resolves the paradox. Choosing
both boxes creates box R empty. Choosing box R creates it one million dollars
fuller .... your choices create the alternate possibilities as realities."
Newcomb's Paradox 591
Chapter 45
Nobody seems to know how to deal with it.
(He would, of course.) -Po 1. Heath
Nothing
Our topic is nothing. By definition nothing does not exist, but
the concepts we have of it certainly exist as concepts. In mathematics,
science, philosophy, and everyday life it turns out to be enormously
useful to have words and symbols for such concepts.
The closest a mathematician can get to nothing is by way of the null
(or empty) set. It is not the same thing as nothing because it has what-
ever kind of existence a set has, although it is unlike all other sets. It is
the only set that has no members and the only set that is a subset of
every other set. From a basket of three apples you can take one apple,
two apples, three apples, or no apples. To an empty basket you can, if
you like, add nothing.
The null set denotes, even though it doesn't denote anything. For ex-
ample, it denotes such things as the set of all square circles, the set of
all even primes other than 2, and the set of all readers of this book who
are chimpanzees. In general it denotes the set of all x's that satisfy any
statement about x that is false for all values of x. Anything you say
about a member of the null set is true, because it lacks a single mem-
ber for which a statement can be false.
The null set is symbolized by 0. It must not be confused with 0, the
symbol for zero. Zero is (usually) a number that denotes the number of
members of 0. The null set denotes nothing, but 0 denotes the number
of members of such sets, for example, the set of apples in an empty bas-
ket. The set of these nonexisting apples is 0, but the number of apples
is O.
A way to construct the counting numbers, discovered by the great
German logician Gottlob Frege and rediscovered by Bertrand Russell,
is to start with the null set and apply a few simple rules and axioms.
Zero is defined as the cardinal number of elements in all sets that are
592
.............. , ..........".1901
593
large part psychological. John von Neumann, for example, shortened
Frege's procedure by one step. He preferred to define 0 as the null set,
1 as the set whose sole member is the null set, 2 as the set whose mem-
bers are the null set and 1, and so on.
John Horton Conway then at the University of Cambridge hit on a re-
markable new way to construct numbers that also starts with the null
set. He first described his technique in a photocopied typescript of 13
pages, "All Numbers, Great and Small." It begins: "We wish to con-
struct all numbers. Let us see how those who were good at construct-
ing numbers have approached the problem in the past." It ends with 10
open questions, of which the last is: "Is the whole structure of any
use?"
Conway explained his new system to Donald E. Knuth, a computer
scientist at Stanford University, when they happened to meet at lunch
one day in 1972. Knuth was immediately fascinated by its possibilities
and its revolutionary content. In 1973 during a week of relaxation in
Oslo, Knuth wrote an introduction to Conway's method in the form of
a novelette. It was issued in paperback in 1974 by Addison-Wesley,
which also publishes Knuth's well-known series titled The Art of Com-
puter Programming. I believe it is the only time a major mathematical
discovery has been published first in a work of fiction. A later book by
Conway, On Numbers and Games, opens with an account of his num-
ber construction, then goes on to apply the theory to the construction
and analysis of two-person games.
Knuth's novelette, Surreal Numbers, is subtitled How Two Ex-
Students Turned On to Pure Mathematics and Found Total Happiness.
The book's primary aim, Knuth explains in a postscript, is not so much
to teach Conway's theory as "to teach how one might go about devel-
oping such a theory." He continues: "Therefore, as the two characters
in this book gradually explore and build up Conway's number system,
I have recorded their false starts and frustrations as well as their good
ideas. I wanted to give a reasonably faithful portrayal of the important
principles, techniques, joys, passions, and philosophy of mathemat-
ics, so I wrote the story as I was actually doing the research myself."
Knuth's two ex-mathematics students, Alice and Bill (A and B), have
fled from the "system" to a haven on the coast of the Indian Ocean.
There they unearth a half-buried black rock carved with ancient He-
brew writing. Bill, who knows Hebrew, manages to translate the open-
ing sentence: "In the beginning everything was void, and J.H.W.H.
594 LOGIC AND PHILOSOPHY
Conway began to create numbers." JHWH is a transliteration of how the
ancient Hebrews wrote the name Jehovah. "Conway" also appears with-
out vowels, but it was the most common English name Bill could think
of that fitted the consonants.
Translation of the "Conway stone" continues: "Conway said, 'Let
there be two rules which bring forth all numbers large and small. This
shall be the first rule: Every number corresponds to two sets of previ-
ously created numbers, such that no member of the left set is greater
than or equal to any member of the right set. And the second rule shall
be this: One number is less than or equal to another number if and
only if no member of the first number's left set is greater than or equal
to the second number, and no member of the second number's right set
is less than or equal to the first number.' And Conway examined these
two rules he had made, and behold! they were very good."
The stone's text goes on to explain how on the zero day Conway cre-
ated zero. He did it by placing the null set on the left and also on the
right. In symbolic notation 0 = {0 I 0}, where the vertical line divides
the left and right sets. No member of the left 0 is equal to or greater than
a member of the right 0 because 0 has no members, so that Conway's
first rule is satisfied. Applying the second rule, it is easy to show that
o is less than or equal to o.
On the next day, the stone reveals, Conway created the first two
nonzero integers, 1 and -1. The method is simply to combine the null
set with 0 in the two possible ways: 1 = {O I 0} and -1 = {0 I O}. It
checks out. Minus 1 is less than but not equal to 0, and 0 is less than
but not equal to 1. Now, of course, 1 and -1 and all subsequently cre-
ated numbers can be plugged back into the left-right formula, and in
this way all the integers are constructed. With 0 and 1 forming the left
set and 0 on the right, 2 is created. With 0, 1, and 2 on the left and 0
on the right, 3 is created, and so on.
At this point readers might enjoy exploring a bit on their own. Jill C.
Knuth's illustration for the front cover of Surreal Numbers shows some
huge boulders shaped to symbolize {O Ii}. What number does this de-
fine? And can the reader prove that {-1 I 1} = O?
"Be fruitful and multiply," Conway tells the integers. By combining
them, first into finite sets, then into infinite sets, the "copulation" of
left-right sets continues, aided by no more than Conway's ridiculously
simple rules. Out pour all the rest of the real numbers: first the integral
fractions, then the irrationals. At the end of aleph-null days a big bang
Nothing 595
occurs and the universe springs into being. That, however, is not all.
Taken to infinity, Conway's construction produces all of Georg Can-
tor's transfinite numbers, all infinitesimal numbers (they are recipro-
cals of infinite numbers), and infinite sets of queer new quantities such
as the roots of transfinites and infinitesimals!
It is an astonishing feat of legerdemain. An empty hat rests on a table
made of a few axioms of standard set theory. Conway waves two sim-
ple rules in the air, then reaches into almost nothing and pulls out an
infinitely rich tapestry of numbers that form a real and closed field.
Every real number is surrounded by a host of new numbers that lie
closer to it than any other "real" value does. The system is truly "sur-
real."
"Man, that empty set sure gets around!" exclaims Bill. "I think I'll
write a book called Properties of the Empty Set. "This notion that noth-
ing has properties is, of course, commonplace in philosophy, science,
and ordinary language. Lewis Carroll's Alice may think it nonsense
when the March Hare offers her nonexistent wine or when the White
King admires her ability to see nobody on the road and wonders why
nobody did not arrive ahead of the March Hare because nobody goes
faster than the hare. It is easy, however, to think of instances in which
nothing actually does enter human experience in a positive way.
Consider holes. An old riddle asks how much dirt is in a rectangular
hole of certain dimensions. Although the hole has all the properties of
a rectangular parallelepiped (corners, edges, faces with areas, volume,
and so on), the answer is that there is no dirt in the hole. The various
holes of our body are certainly essential to our health, sensory aware-
ness, and pleasure. In Dorothy and the Wizard in Oz, the braided man,
who lives on Pyramid Mountain in the earth's interior, tells Dorothy
how he got there. He had been a manufacturer of holes for Swiss
cheese, doughnuts, buttons, porous plasters, and other things. One day
he decided to store a vast quantity of adjustable postholes by placing
them end to end in the ground, making a deep vertical shaft into which
he accidentally tumbled.
The mathematical theory behind Sam Loyd's sliding-block puzzle
(15 unit cubes inside a 4 x 4 box) is best explained by regarding the hole
as a moving cube. It is analogous to what happens when a gold atom
diffuses through lead. Bubbles of nothing in liquids, from the size of a
molecule on up, can move around, rotate, collide, and rebound just
like things. Negative currents are the result of free electrons jostling one
596 LOGIC AND PHILOSOPHY
another along a conductor, but holes caused by an absence of free elec-
trons can do the same thing, producing a positive "hole current" that
goes the other way.
Lao-tzu writes in Chapter 11 of Tao Te Ching:
Thirty spokes share the wheel's hub;
It is the center hole that makes it usefu1.
Shape clay into a vessel;
It is the space within that makes it usefu1.
Cut doors and windows for a room;
It is the holes which make it usefu1.
Therefore profit comes from what is there;
Usefulness from what is not there.
Osborne Reynolds, a British engineer who died in 1912, invented an
elaborate theory in which matter consists of microparticles of nothing
moving through the ether the way bubbles move through liquids. His
two books about the theory, On an Inversion of Ideas as to the Structure
of the Universe and The Sub-Mechanics of the Universe, both pub-
lished by the Cambridge University Press, were taken so seriously that
W. W. Rouse Ball, writing in early editions of his Mathematical Recre-
ations and Essays, called the theory "more plausible than the electron
hypothesis. "
Reynolds' inverted idea is less crazy than it sounds. P.A.M. Dirac, in
his famous theory that predicted the existence of antiparticles, viewed
the positron (antielectron) as a hole in a continuum of negative charge.
When an electron and positron collide, the electron falls into the
positron hole, causing both particles to vanish.
The old concept of a "stagnant ether" has been abandoned by physi-
cists, but in its place is not nothing. The "new ether" consists of the
metric field responsible for the basic forces of nature, perhaps also for
the particles. John Archibald Wheeler proposes a substratum, called su-
perspace, of infinitely many dimensions. Occasionally a portion of it
twists in such a peculiar way that it explodes, creating a universe of
three spatial dimensions, changing in time, with its own set of laws and
within which the field gets tied into little knots that we call "matter."
On the microlevel, quantum fluctuations give space a foamlike struc-
ture in which the microholes provide space with additional properties.
There is still a difference between something and nothing, but it is
purely geometrical and there is nothing behind the geometry.
Nothing 597
Empty space is like a straight line of zero curvature. Bend the line,
add little bumps that ripple back and forth, and you have a universe
dancing with matter and energy. Outside the utmost fringes of our ex-
panding cosmos are (perhaps) vast regions unpenetrated by light and
gravity. Beyond those regions may be other universes. Shall we say
that these empty regions contain nothing, or are they still saturated
with a metric of zero curvature?
Greek and medieval thinkers argued about the difference between
being and nonbeing, whether there is one world or many, whether a
perfect vacuum can properly be said to "exist," whether God formed
the world from pure nothing or first created a substratum of matter that
was what st. Augustine called prope nihil, or close to nothing. Exactly
the same questions were and are debated by philosophers and theolo-
gians of the East. When the god or gods of an Eastern religion created
the world from a great Void, did they shape nothing or something that
was almost nothing? The questions may seem quaint, but change the
terminology a bit and they are equivalent to present controversies.
A special Christmas 1901 issue of the British philosophical magazine
Mind consisted entirely of joke papers. Its frontispiece (reproduced
here as Figure 45.1) was titled "Portrait of Its Immanence the Absolute."
The Absolute was a favorite term of Hegelian philosophers.
"Turn the eye of faith," the instructions read, "fondly but firmly on
the centre of the page, wink the other, and gaze fixedly until you see It."
The Absolute's picture was a totally blank rectangle, protected by
covering tissue. Above the picture were the words, "This side up."
There are endless examples from the arts-some jokes, some not-of
nothing admired as something. In 1951 Ad Reinhardt, a respected
American abstractionist who died in 1967, began painting all-blue and
all-red canvases. A few years later he moved to the ultimate-black. His
all-black 5 x 5-foot pictures were exhibited in 1963 in leading galleries
in New York, Paris, Los Angeles, and London. (See Figure 45.2) Al-
though one critic called him a charlatan (Ralph F. Colin, "Fakes and
Frauds in the Art World," Art in America, April 1963), more eminent
critics (Hilton Kramer, The Nation, June 22,1963, and Harold Rosen-
berg, The New Yorker, June 15, 1963) admired his black art. An "ulti-
mate statement of esthetic purity," was how Kramer put it (The New
York Times, October 17, 1976) in praising an exhibit ofthe black paint-
ings at the Pace Gallery.
In 1965 Reinhardt had three simultaneous shows at top Manhattan
598 LOGIC AND PHILOSOPHY
galleries: one of all-blacks, one of all-reds, one of all-blues. Prices
ranged from $1,500 to $12,000.
Since black is the absence of light, Reinhardt's black canvases come
as close as possible to pictures of nothing, certainly much closer than
the all-white canvases of Robert Rauschenberg and others. A New
Yorker cartoon (September 23, 1944) by R. Taylor showed two ladies at
an art exhibit, standing in front of an all-white canvas and reading from
the catalogue: "During the Barcelona period he became enamored of the
possibilities inherent in virgin space. With a courage born of the most
profound respect for the enigma of the imponderable, he produced, at
this time, a series of canvases in which there exists solely an expanse
of pregnant white."
The Museum of Modem Art
Figure 45.2. Ad Reinhardt's Abstract Painting, 1960-61 (Oil, 60- x 60-)
I know of no piece of "minimal sculpture" that is reduced to the ab-
solute minimum of nothing, though I expect to read any day now that
a great museum has purchased such a work for many thousands of dol-
lars. Henry Moore certainly exploited the aesthetics of holes (see Fig-
ure 45.3). In 1950 Ray Bradbury received the first annual award of The
Elves', Gnomes', and Little Men's Science-Fiction Chowder and March-
ing Society at a meeting in San Francisco. The award was an invisible
little man standing on the brass plate of a polished walnut pedestal.
This was not entirely nothing, says my informant, Donald Baker Moore,
Nothing 599
because there were two black shoe prints on the brass plate to indicate
that the little man was actually there.
There have been many plays in which principal characters say noth-
ing. Has anyone ever produced a play or motion picture that consists,
from beginning to end, of an empty stage or screen? Some of Andy
Warhol's early films come close to it, and I wouldn't be surprised to
learn that the limit was actually attained by some early avant-garde
playwright.
John Cage's 4'33#is a piano composition that calls for four minutes
and thirty-three seconds of total silence as the player sits frozen on the
piano stool. The duration of the silence is 273 seconds. This corre-
sponds, Cage has explained, to -273 degrees centigrade, or absolute
zero, the temperature at which all molecular motion quietly stops. I
have not heard 4'33"performed, but friends who have tell me it is
Cage's finest composition.
There are many outstanding instances of nothing in print: Chapters
18 and 19 of the final volume of Tristram Shandy, for example. Elbert
Hubbard's Essay on Silence, containing only blank pages, was bound in
brown suede and gold-stamped. I recall as a boy seeing a similar book
titled What I Know about Women, and a Protestant fundamentalist tract
called What Must You Do to Be Lost? Poeme Collectif, by Robert Filliou,
issued in Belgium in 1968, consists of sixteen blank pages.
In 1972 the Honolulu Zoo distributed a definitive monograph called
Snakes of Hawaii: An authoritative, illustrated and complete guide to
exotic species indigenous to the 50th State, by V. Ralph Knight, Jr., B.S.
A correspondent, Larry E. Morse, informs me that this entire mono-
graph is reprinted (without credit) in The Nothing Book. This volume
of blank pages was published in 1974, by Harmony House, in regular
and deluxe editions. It sold so well that in 1975 an even more expen-
sive ($5) deluxe edition was printed on fine French marble design
paper and bound in leather. According to The Village Voice (December
30, 1974), Harmony House was threatened with legal action by a Euro-
pean author whose blank-paged book had been published a few years
before The Nothing Book. He believed his copyright had been in-
fringed, but nothing ever came of it.
Howard Lyons, a Toronto correspondent, points out that the null set
has long been a favorite topic of song writers: "I ain't got nobody," "No-
body loves me," "I've got plenty of nothing," "Nobody lied when they
600 LOGIC AND PHILOSOPHY
said that I cried over you," "There ain't no sweet gal that's worth the
salt of my tears," and hundreds of other lines.
Events can occur in which nothing is as startling as a thunderclap.
An old joke tells of a man who slept in a lighthouse under a foghorn
that boomed regularly every 10 minutes. One night at 3:20 A.M., when
the mechanism failed, the man leaped out of bed shouting, "What was
that?" As a prank all the members of a large orchestra once stopped
playing suddenly in the middle of a strident symphony, causing the
conductor to fall off the podium. One afternoon in a rural section of
North Dakota, where the wind blew constantly, there was a sudden
cessation of wind. All the chickens fell over. A Japanese correspon-
dent tells me that the weather bureau in Japan issues a "no-wind warn-
ing" because an absence of wind can create damaging smog.
There are many examples that are not jokes. An absence of water
can cause death. The loss of a loved one, of money, or of a reputation
can push someone to suicide. The law recognizes innumerable occa-
sions on which a failure to act is a crime. Grave consequences will fol-
low when a man on a railroad track, in front of an approaching train
and unable to decide whether to jump to the left or to the right, makes
no decision. In the story "Silver Blaze," Sherlock Holmes based a fa-
mous deduction on the "curious incident" of a dog that "did nothing
in the night-time."
Moments of escape from the omnipresent sound of canned music
are becoming increasingly hard to obtain. Unlike cigar smoke, writes
Edmund Morris in a fine essay, "Oases of Silence in a Desert of Din"
(The New York Times, May 25, 1975), noise can't be fanned away. There
is an old joke about a jukebox that offers, for a quarter, to provide three
minutes of no music. Drive to the top of Pike's Peak, says Morris,
"whose panorama of Colorado inspired Katharine Lee Bates to write
'America the Beautiful: and your ears will be assailed by the twang and
boom of four giant speakers-N, S, E, and W-spraying cowboy tunes
into the crystal air." Even the Sistine Chapel is now wired for sound.
"At first," continues Morris, "there is something discomforting, al-
most frightening, about real silence .... You are startled by the appar-
ent loudness of ordinary noises. . . . Gradually your ears become
attuned to a delicate web of sounds, inaudible elsewhere, which George
Eliot called 'that roar which lies on the other side of silence.' " Morris
provides a list of a few Silent Places around the globe where one can
Nothing 601
escape not only from Muzak but from all the aural pollution that is the
by-product of modern technology.
These are all examples of little pockets in which there is an absence
of something. What about that monstrous dichotomy between all
being-everything there is-and nothing? From the earliest times the
most eminent thinkers have meditated on this ultimate split. It seems
unlikely that the universe is going to vanish (although I myself once
wrote a story, "Oom," about how God, weary of existing, abolished
everything, including himself), but the fact that we ourselves will soon
vanish is real enough. In medieval times the fear of death was mixed
with a fear of eternal suffering, but since the fading of hell this fear has
been replaced by what Soren Kierkegaard called an "anguish" or
"dread" over the possibility of becoming nothing.
This brings us abruptly to what Paul Edwards has called the "super-
ultimate question." "Why," asked Leibniz, Schelling, Schopenhauer,
and a hundred other philosophers, "should something exist rather than
nothing?"
Obviously it is a curious question, not like any other. Large numbers
of people, perhaps the majority, live out their lives without ever con-
sidering it. If someone asks them the question, they may fail to under-
stand it and believe the questioner is crazy. Among those who
understand the question, there are varied responses. Thinkers of a mys-
tical turn of mind, the late Martin Heidegger for instance, consider it
the deepest, most fundamental of all metaphysical questions and look
with contempt on all philosophers who are not equally disturbed by it.
Those of a positivistic, pragmatic turn of mind consider it trivial. Since
everyone agrees there is no way to answer it empirically or rationally,
it is a question without cognitive content, as meaningless as asking if
the number 2 is red or green. Indeed, a famous paper by Rudolf Carnap
on the meaning of questions heaps scorn on a passage in which Hei-
degger pontificates about being and nothingness.
A third group of philosophers, including Milton K. Munitz, who
wrote an entire book titled The Mystery of Existence, regards the ques-
tion as being meaningful but insists that its significance lies solely in
our inability to answer it. It mayor may not have an answer, argues
Munitz, but in any case the answer lies totally outside the limits of sci-
ence and philosophy.
Whatever their metaphysics, those who have puzzled most over the
superultimate question have left much eloquent testimony about those
602 LOGIC AND PHILOSOPHY
unexpected moments, fortunately short-lived, in which one is sud-
denly caught up in an overwhelming awareness of the utter mystery of
why anything is. That is the terrifying emotion at the heart of Jean-
Paul Sartre's great philosophical novel Nausea. Its red-haired protago-
nist, Antoine Roquentin, is haunted by the superultimate mystery. "A
circle is not absurd," he reflects. "It is clearly explained by the rotation
of a straight segment around one of its extremities. But neither does a
circle exist." Things that do exist, such as stones and trees and himself,
exist without any reason. They are just insanely there, bloated, ob-
scene, gelatinous, unable not to exist. When the mood is on him,
Roquentin calls it "the nausea." William James had earlier called it an
"ontological wonder sickness." The monotonous days come and go,
all cities look alike, nothing happens that means anything.
G. K. Chesterton is as good an example as any of the theist who,
stunned by the absurdity of being, reacts in opposite fashion. Not that
shifting to God the responsibility for the world's existence answers the
superultimate question; far from it! One immediately wonders why
God exists rather than nothing. But although none of the awe is less-
ened by hanging the universe on a transcendent peg, the shift can give
rise to feelings of gratitude and hope that relieve the anxiety. Chester-
ton's existential novel Manalive is a splendid complement to Sartre's
Nausea. Its protagonist, Innocent Smith, is so exhilarated by the priv-
ilege of existing that he goes about inventing whimsical ways of shock-
ing himself into realizing that both he and the world are not nothing.
Let P. L. Heath, who had the first word in this article, also have the
last. "If nothing whatsoever existed," he writes at the end of his article
on nothing in The Encyclopedia of Philosophy, "there would be no
problem and no answer, and the anxieties even of existential philoso-
phers would be permanently laid to rest. Since they are not, there is ev-
idently nothing to wony about. But that itself should be enough to keep
an existentialist happy. Unless the solution be, as some have suspected,
that it is not nothing that has been worrying them, but they who have
been worrying it."
Addendum
When this chapter first appeared in Scientific American (Feb-
ruary 1975), it prompted many delightful letters on aspects ofthe topic
I had not known about or had failed to mention.
Nothing 603
Hester Elliott was the first of several readers who were reminded, by
my story of the lighthouse keeper, of what some New Yorkers used to
call the "Bowery EI phenomenon." After the old elevated on Third Av-
enue was torn down, police began receiving phone calls from people
who lived near the El. They were waking at regular intervals during the
night, hearing strange noises, and having strong feelings of foreboding.
"The schedules of the absent trains," as Ms. Elliott put it, "reappeared
in the form of patterned calls on the police blotters." This is discussed,
she said, by Karl Pribram in his book Languages of the Brain as an ex-
ample of how our brain, even during sleep, keeps scanning the flow of
events in the light of past expectations. It is aroused by any sharp de-
viation from the accustomed pattern.
Psychologist Robert B. Glassman also referred in a letter to the EI ex-
ample, and gave others. The human brain, he wrote, has the happy fa-
cility of forgetting, of pushing out of consciousness whatever seems
irrelevant at the moment. But the irrelevant background is still per-
ceived subliminally, and changes in this background bring it back into
consciousness. Russian psychologists, he said, have found that if a
human or animal listens long enough to the repeated sound of the same
tone, they soon learn to ignore it. But if the same tone is then sounded
in a different way, even if sounded more softly or more briefly, there is
instant arousal.
Vernon Rowland, a professor of psychology at Case Western Reserve,
elaborated similar points. His letter, which follows, was printed in Sci-
entific American, April 1975:
Sirs:
I enjoyed Martin Gardner's essay on "nothing." John Horton Conway's
rule and Gardner's analysis of "nothing" are, like all human activity, ex-
pressions of the nervous system, the study of which helps in under-
standing the origins and evolution of "nothing."
The brain is marvelously tuned to detect change as well as constancies
in the environment. Sharp change between constancies is a perceptually
or intellectually recognizable boundary. "Nothing" is "knowable" with
clarity only if it is well demarcated from the "non-nothing." Even if it is
vaguely bounded, nothingness cannot be treated as an absolute. This is
an example ofthe illogicality of absolutes, because "nothing" cannot be
in awareness except as it is related to (contrasted with) non-nothing.
One can observe in the brains of perceiving animals, even animals as
primitive as the frog, special neurons responding specifically to spatial
604 LOGIC AND PHILOSOPHY
boundaries and to temporal boundaries. In the latter, for instance, neu-
rons called "off" neurons, go into action when "something," say light,
becomes "nothing" (darkness). "Nothing" is therefore positively signaled
and is thereby endowed with existence. The late Polish neuropsycholo-
gist Jerzy Konorski pointed out the possibility that closing the eyes may
activate off neurons, giving rise to "seeing" darkness and recognizing it
as being different from not seeing at all.
I and others have used temporal nothingness as a food signal for cats
by simply imposing 10 seconds of silence in an otherwise c<?ntinuously
clicking environment. Their brains show the learning of the significance
of this silence in ways very similar to those for the inverse: 10 seconds
of clicking presented on a continuous background of silence. "Nothing"
and "something" can be treated in the same way as psychologists deal
with other forms of figure-ground or stimulus-context reversal.
The nothingness of which we become aware by specific brain signals
can be known only by discriminating it from other brain signals that re-
veal the boundaries and constancies of existing objects. This requires an
act of attention. There is another form of "nothing" that is based on an
attentional shift from one sense modality to another (as in the example
of listening to music) or to a failure of the attentional mechanism. In cer-
tain forms of strokes the person "forgets" one part of his body and acts
as if it simply does not exist, for example a man who shaves only one
half of his face.
Animate systems obtain and conserve life-supporting energy by evolv-
ing mechanisms to offset or counter perturbations in their energy supply.
Detecting absences ("nothings") in the energy domain had to be acquired
early or survival could not have gone beyond the stage of actually living
in the energy supply (protozoa in nutritious pools) rather than near it
(animals that can leave the water and return).
If this pragmatic view of the biopsychological origins of "nothing"
and "absence" is insufficient for trivializing the Leibnizian question
("Why should something exist rather than nothing?"), I would argue
that the philosopher faces the necessity of showing that the statement
"Nothing [in the absolute sense] exists" is not a self-contradiction.
The reference to my story "Gom" reminded Ms. Elliott of the fol-
lowing paragraph from Jorge Luis Borges' essay on John Donne's
Biathanatos (a work which argues that Jesus committed suicide), in
Other Inquisitions, 1937-1952:
As I reread this essay, I think of the tragic Philipp Batz, who is called
Philipp Mainlander in the history of philosophy. Like me, he was an im-
Nothing 605
passioned reader of Schopenhauer, under whose influence (and perhaps
under the influence of the Gnostics) he imagined that we are fragments
of a God who destroyed Himself at the beginning of time, because He did
not wish to exist. Universal history is the obscure agony of those frag-
ments. Mainlander was born in 1841; in 1876 he published his book
Philosophy of the Redemption. That same year he killed himself.
Is Mainlander one of Borges' invented characters? No, he actually ex-
isted. You can read about him and his strange two-volume work in The
Encyclopedia of Philosophy, Vol. 6, page 119.
Several readers informed me of the amusing controversy among
graph theorists over whether the "null-graph" is useful. This is the
graph that has no points or edges. The classic reference is a paper by
Frank Harary and Ronald C. Read, "Is the Null-Graph a Pointless Con-
cept?" (The paper was given at the Graphs and Combinatorial Confer-
ence at George Washington University in 1973 and appears in the
conference lecture notes published by Springer-Verlag.)
"Note that it is not a question of whether the null-graph 'exists,' " the
authors write. "It is simply a question of whether there is any point in
it." The authors survey the literature, give pros and cons, and finally
reach no conclusion. Figure 45.3, reproduced from their paper, shows
what the null-graph looks like.
Figure 45.3. The null-graph
606 LOGIC AND PHILOSOPHY
Wesley Salmon, the philosopher of science, sent a splendid ontolog-
ical argument for the existence of the null set:
I have just finished reading, with much pleasure, your column on "noth-
ing." It reminded me of a remark made by a brilliant young philosopher
at the University of Toronto, Bas van Fraassen, who, in a lecture on phi-
losophy of mathematics, asked why there might not be a sort of onto-
logical proof for the existence of the null set. It would begin, "By the null
set we understand that set than which none emptier can be conceived
... " Van Fraassen is editor in chief of the Journal of Philosophical Logic.
I sent him the completion of the argument:
"The fool hath said in his heart that there is no null set. But if that
were so, then the set of all such sets would be empty, and hence, it
would be the null set. Q.E.D."
I still do not know why he did not publish this profound result.
Frederick Mosteller, a theoretical statistician at Harvard, made the
following comments on the superultimate question:
Ever since I was about fourteen years old I have been severely bothered
by this question, and by and large not willing to talk to other people
about it because the first few times I tried I got rather unexpected re-
sponses, mainly rather negative putdowns. It shook me up when it first
occurred to me, and has bothered me again and again. I could not un-
derstand why it wasn't in the newspapers once a week. I suppose, in a
sense, all references to creation are a reflection of this same issue, but it
is the simplicity of the question that seems to me so scary.
When I was older I tried it once or twice on physicists and again did
not get much of a response-probably talked to the wrong ones. I did
mention it to John Tukey once, and he offered a rather good remark. He
said something like this: contemplating the question at this time doesn't
seem to be producing much information-that is, we aren't making
much progress with it-and so it is hard to spend time on it. Perhaps it
is not yet a profitable question.
It seems so much more reasonable to me that there should be nothing
than something that I have secretly concluded for myself that quite pos-
sibly physicists will ultimately prove that, were there a system contain-
ing nothing, it would automatically create a physical universe. (Of
course, I know they can't quite do this.)
When my column on nothing was reprinted in my Mathematical
Magic Show (1977) an editor at Knopf asked me if it was necessary to
Nothing 607
obtain MOMA's (Museum of Modern Art) permission to reprint Ad
Reinhardt's all-black painting (Figure 45.2). I convinced her it was not
necessary. And as I anticipated, somewhere along the production line
I was asked for the missing art of Figure 45.3. This picture of the null
graph, by the way, is reproduced (without credit to the artist) as Figure
257 in the Dictionary of Mathematics, edited by E. J. Borowski and J. M.
Borwein (London: Collins, 1989).
Meditating on the recent flurry of interest in what is called the "an-
thropic principle" I suddenly realized that I could answer the supe-
rultimate question: Why is there something rather than nothing?
Because if there wasn't anything we wouldn't be here to ask the ques-
tion. I think this points up the essential absurdity of the weak anthropic
principle. It's not wrong, but it contributes nothing significant to any
philosophical or scientific question. The Danish poet Piet Hein, in one
of his "grook" verses, says it this way:
The universe may
Be as great as they say,
But it wouldn't be missed
If it didn't exist.
Lakenan Barnes, an attorney in Missouri, reminded me that Joshua
was the son of Nun (Joshua 1:1), that "love" in tennis means nothing,
that the doughnut's hole is the "dough naught." He also passed along
a quatrain of his that had appeared in the 8t. Louis Post-Dispatch (July
7,1967):
In the world of math
That Man has wrought,
The greatest gain
Was the thought of naught.
Some readers were mystified by the chapter's epigraph. It is the sec-
ond sentence of Heath's article on nothing in the Encyclopedia of Phi-
losophy. Like Lewis Carroll, in the second Alice book, Heath is taking
Nobody to be the name of a person. Here is the sentence in the context
of Heath's playful opening paragraph:
NOTHING is an awe-inspiring yet essentially undigested concept, highly
esteemed by writers of a mystical or existentialist tendency. but by most
others regarded with anxiety, nausea, or panic. Nobody seems to know
how to deal with it (he would, of course), and plain persons generally are
608 LOGIC AND PHILOSOPHY
reported to have little difficulty in saying, seeing, hearing, and doing
nothing. Philosophers, however, have never felt easy on the matter. Ever
since Parmenides laid it down that it is impossible to speak of what is
not, broke his own rule in the act of stating it, and deduced himself into
a world where all that ever happened was nothing, the impression has
persisted that the narrow path between sense and nonsense on this subject
is a difficult one to tread and that altogether the less said of it the better.
Will Shortz, The New York Times crossword puzzle editor, presented
an April Fools' crossword in Games magazine, March/April1979, page
31. The solution was to leave all the cells blank because "nothing" was
the answer to each of the puzzle's 49 definitions of words.
One of Gary Larson's Far Side cartoons showed a man equipped for
mountain climbing but who was about to descend into a huge hole in
the ground. A reporter asks him why he is doing this. He replies, "Be-
cause it's not there."
Ray Smullyan on Discover's puzzle page (November 1996) gave this
riddle which readers should have little difficulty answering: "What is
it that is larger than the universe, the dead eat it, and if the living eat
it, they die?"
Bibliography
R. M. Adams, Nil: Episodes in the Literary Conquest of Void During the Nineteenth
Century, Oxford University Press, 1966.
J. H. Conway, All Numbers Great and Small, University of Calgary Mathematical
Research Paper No. 149, February 1972.
J. H. Conway, On Numbers and Games, Academic Press, 1976.
P. Edwards, "Why," The Encyclopedia of Philosophy, MacmillanFree Press, 1967.
M. Gardner, "Oom," The Journal of Science Fiction, Fall 1951. Reprinted in M.
Gardner, The No-Sided Professor and Other Tales, Prometheus, 1987.
M. Gardner, The Annotated Snark, Simon and Schuster, 1962.
M. Gardner, "Conway's Surreal Numbers," Penrose Tiles to Trapdoor Ciphers, W.
H. Freeman, 1989, Chapter 4.
P. 1. Heath, "Nothing," The Encyclopedia of Philosophy, MacmillanFree Press,
1967.
R. Kaplan, The Nothing That Is: A Natural History of Zero, Oxford University
Press, 1999.
D. E. Knuth, Surreal Numbers, Addison-Wesley, 1974.
B. Rotman, Signifying Nothing: The Semiotics of Zero, Stanford, 1993.
C. Seife, Zero: The Biography of a Dangerous Idea, Viking, 2000.
I. Stewart, "Zero, Zilch, and Zip," New Scientist, April 25, 1998, pp. 41-44.
Nothing 609
-WILLARD VAN ORMAN
IsH
46.1. A
610
are ..................... ,-' ...... ' ......... ' ............. ....,
TYPEWRITERS {e}
46.2. A Venn sets
611
How far can we extend the universal set without losing our ability to
reason about it? It depends on our concern. If we expand the universe
of Figure 46.1 to include all concepts, the intersection set is no longer
empty because it is easy to imagine a person growing feathers. The
proofs of Euclid are valid only if the universe of discourse is confined
to points in a Euclidean plane or in 3-space. If we reason that a dozen
eggs can be equally divided only between one, two, three, four, six, or
12 people, we are reasoning about a universal set that ranges over the
integers. John Venn (who invented the Venn diagram) likened the uni-
verse of discourse to our field of vision. It is what we are looking at. We
ignore everything behind our head.
Nevertheless, we can extend the universe of discourse amazingly far.
We certainly can include abstractions such as the number 2, pi, com-
plex numbers, perfect geometric figures, even things we cannot visu-
alize such as hypercubes and non-Euclidean spaces. We can include
universals such as redness and cowness. We can include things from
the past or in the future and things real or imaginary, and can still rea-
son effectively about them. Every dinosaur had a mother. If it rains
next week in Chicago, the old Water Tower will get wet. If Sherlock
Holmes had actually fallen off that cliff at Reichenbach Falls, he would
have been killed.
Suppose we extend our universe to include every entity that can be
defined without logical contradiction. Every statement we can make
about that universe, if it is not contradictory, is (in a sense) true. The
contradictory objects and statements are not allowed to "exist" or be
"true" for the simple reason that contradiction introduces meaning-
lessness. When a philosopher such ~ Leibniz talks about "all possible
worlds," he means worlds that can be talked about. You can talk about
a world in which humans and typewriters have feathers. You cannot
say anything sensible about a square triangle or an odd integer that is
a multiple of 2.
Is it possible to expand our universe of discourse to the ultimate and
call it the set of all possible sets? No, this is a step we cannot take with-
out contradiction. Georg Cantor proved that the cardinal number of
any set (the number of its elements) is always lower than the cardinal
number of the set of all its subsets. This is obvious for any finite set (if
it has n elements, it must have 2
n
subsets), but Cantor was able to show
that it also applies to infinite sets. When we try to apply this theorem
to everything, however, we get into deep trouble. The set of all sets
612 LOGIC AND PHILOSOPHY
must have the highest aleph (infinite number) for its cardinality; oth-
erwise it would not be everything. On the other hand, it cannot have
the highest aleph because the cardinality of its subsets is higher.
When Bertrand Russell first came across Cantor's proof that there is
no highest aleph, and hence no "set of all sets," he did not believe it.
He wrote in 1901 that Cantor had been "guilty of a very subtle fallacy,
which I hope to explain in some future work," and that it was "obvi-
ous" there had to be a greatest aleph because "if everything has been
taken, there is nothing left to add." When this essay was reprinted in
Mysticism and Logic 16 years later, Russell added a footnote apologiz-
ing for his mistake. ("Obvious" is obviously a dangerous word to use in
writing about everything.) It was Russell's meditation on his error that
led him to discover his famous paradox about the set of all sets that are
not members of themselves.
To sum up, when the mathematician tries to make the final jump
from lots of things to everything, he finds he cannot make it. "Every-
thing" is self-contradictory and therefore does not exist!
The fact that tp.e set of all sets cannot be defined in standard
(Zermelo-Fraenkel) set theory, however, does not inhibit philosophers
and theologians from talking about everything, although their syn-
onyms for it vary: being, ens, what is, existence, the absolute, God, re-
ality, the Tao, Brahman, dh arm a-kaya, and so on. It must, of course,
include everything that was, is, and will be, everything that can be
imagined and everything totally beyond human comprehension. Noth-
ing is also part of everything. When the universe gets this broad, it is
difficult to think of anything meaningful (not contradictory) that does
not in some sense exist. The logician Raymond Smullyan, in one of his
several hundred marvelous unpublished essays, retells an incident he
found in Oscar Mandel's book Chi Po and the Sorcerer: A Chinese Tale
for Children and Philosophers. The sorcerer Bu Fu is giving a painting
lesson to Chi Po. "No, no!" says Bu Fu. "You have merely painted what
is. Anybody can paint what is! The real secret is to paint what isn't!"
Chi Po, puzzled, replies: "But what is there that isn't?"
This is a good place to come down from the heights and consider a
smaller, tidier universe, the universe of contemporary cosmology. Mod-
ern cosmology started with Einstein's model of a closed but unbounded
universe. If there is sufficient mass in the cosmos, our 3-space curves
back on itself like the surface of a sphere. (Indeed, it becomes the 3-
space hypersurface of a 4-space hypersphere.) We now know that the
Everything 613
universe is expanding from a primordial fireball, but there does not
seem to be enough mass for it to be closed. The steady-state theory gen-
erated much discussion and stimulated much valuable scientific work,
but it now seems to have been eliminated as a viable theory by such
discoveries as that ofthe universal background radiation (which has no
reasonable explanation except that it is radiation left over from the pri-
mordial fireball, or "big bang").
The large unanswered question is whether there is enough mass hid-
den somewhere in the cosmos (in black holes?) to halt the expansion
and start the universe shrinking. If that is destined to happen, the con-
traction will become runaway collapse, and theorists see no way to
prevent the universe from entering the "singularity" at the core of a
black hole, that dreadful spot where matter is crushed out of existence
and no known laws of physics apply. Will the universe disappear like
the fabled Poof Bird, which flies backward in ever decreasing circles
until-poof!-it vanishes into its own anus? Will everything go through
the black hole to emerge from a white hole in some completely differ-
ent spacetime? Or will it manage to avoid the singularity and give rise
to another fireball? If reprocessing is possible, we have a model of an
oscillating universe that periodically explodes, expands, contracts, and
explodes again.
Among physicists who have been building models of the universe
John Archibald Wheeler has gone further than anyone in the direction
of everything. In Wheeler's wild vision our universe is one of an infin-
ity of universes that can be regarded as embedded in a strange kind of
space called superspace.
In order to understand (dimly) what Wheeler means by superspace
let us start with a simplified universe consisting of a line segment oc-
cupied by two particles, one black and one gray (see Figure 46.3, top).
The line is one-dimensional, but the particles move back and forth (we
allow them to pass through each other) to create a space-time of two di-
mensions: one of space and one of time.
There are many ways to graph the life histories of the two particles.
One way is to represent them as wavy lines, called world lines in rel-
ativity theory, on a two-dimensional space-time graph (see Figure
46.3). Where was the black particle at time k? Find k on the time axis,
move horizontally to the black particle's world line, then move down
to read off the particle's position on the space axis.
To see how beautifully the two world lines record the history of our
614 LOGIC AND PHILOSOPHY
A one-dlmel"lsionol universe with MO pcrlicles
SPACE
a mrO-DBLITl to
figuration space. Let us see how to do this with our two particles. Our
configuration space again is two-dimensional, but now both coordi-
nates are spatial. One coordinate is assigned to the black particle and
the other, to the gray particle (see Figure 46.4). The positions of both
particles can be represented by a single point called the configuration
point. As the point moves, its coordinate values change on both axes.
One axis locates one particle; the other axis, the other particle. The tra-
jectory traced by the moving point corresponds to the changing pattern
of the system of particles; conversely, the history of the system deter-
mines a unique trajectory. It is not a space-time graph. (Time enters
later as an added parameter.) The line cannot form branches because
that would split each particle in two. It may, however, intersect itself.
If a system is periodic, the line will be a closed curve. To transform the
graph into a spacetime graph we can, if we like, add a time coordinate
and allow the point to trace a curve in three dimensions.
The technique generalizes to a system of N particles in a space with
any number of dimensions. Suppose we have 100 particles in our lit-
~
u
~
~
o 0
...
co:
9
o
u
-3
-6
-6 -3 o
Figure 46.4. A configuration-space graph of the history of two particles in a one-
dimensional universe
616 LOGIC AND PHILOSOPHY
tIe line-segment cosmos. Each particle has one degree of freedom, so
our configuration point must move in a space of 100 dimensions. If our
universe is a system of Nparticles on a plane, each particle has two de-
grees of freedom, so our configuration space must be a hyperspace of
2N dimensions. In 3-space a particle has three degrees of freedom, so
the configuration space must have 3N dimensions. In general the hy-
perspace has an order equal to the total degrees of freedom in the sys-
tem. Add another coordinate for time and the space becomes a
space-time graph.
Unfortunately the position of a configuration point at any instant
does not enable us to reconstruct the system's past or predict its future.
Josiah Willard Gibbs, working on the thermodynamics of molecules,
found a slightly more complicated space in which he could graph a sys-
tem of molecules so that the record was completely deterministic. This
is done by assigning six coordinates to each molecule: three to deter-
mine position and three to specify momentums. The movement of a
single phase point in what Gibbs called a "phase space" of 6N dimen-
sions will record the life history of N particles. Now, however, the po-
sition of the phase point provides enough information to reconstruct (in
principle) the entire previous history of the system and to predict its fu-
ture. As before, the trajectory cannot branch, but now it also cannot in-
tersect itself. An intersection would mean that a state could be reached
from two different states and could lead to two different states, but
both possibilities are ruled out by the assumption that position and
momentums (which include a vector direction) fully determine the
next state. The curve may still loop, however, indicating that the sys-
tem is periodic.
Our universe, with its non-Euclidean space-time and its quantum
uncertainties, cannot be graphed in anything as simple as phase space,
but Wheeler has found a way to do it in superspace. Like configuration
space, superspace is timeless, but it has an infinity of dimensions. A
single point in superspace has an infinite set of coordinates that spec-
ify completely the structure of our non-Euclidean 3-space: its size, the
location of every particle, and the structure of every field (including the
curvature of space itself) at every point. As the superpoint moves, its
changing coordinate numbers describe how our universe changes, not
failing to take into account the role of observers' frames of reference in
relativity and the probability parameters of quantum mechanics. The
motion of the superpoint gives the entire history of our universe.
Everything 617
At the same time (whatever that means!) that the present drama of
our cosmos is being acted on the stage of superspace countless other su-
perpoints, representing other 3-space universes, are going through their
cycles. Superpoints close to one another describe universes that most
resemble one another, like the parallel worlds that H. G. Wells intro-
duced into science fiction with his Men Like Gods. These parallel uni-
verses, cut off from one another because they occupy different slices of
superspace, are continually bursting into space-time through a singu-
larity, flourishing for a moment of eternity, then vanishing back through
a singularity into the pure and timeless "pregeometry" from whence
they came.
Whenever such a cosmos explodes into being, random factors gen-
erate a specific combination of logically consistent (Leibniz called them
compossible) particles, constants, and laws. The resulting structure
has to be tuned exceedingly fine to allow life. Alter the fine-structure
constant a trifle either way and a sun such as ours becomes impossible.
Why are we here? Because random factors generated a cosmic structure
that allowed us to evolve. An infinity of other universes, not so finely
tuned, are living and dying without there being anyone in them capa-
ble of observing them.
These "meaningless" universes, meaningless because they contain
no participator-observers, do not even "exist" except in the weak sense
of being logically possible. Bishop Berkeley said that to exist is to be
perceived, and Charles Sanders Peirce maintained that existence is a
matter of degree. Taking cues from both philosophers, Wheeler argues
that only when a universe develops a kind of self-reference, with the
universe and its observers reinforcing one another, does it exist in a
strong sense. "All the choir of heaven and furniture of earth have no
substance without a mind" was how Berkeley put it.
As far as I can tell, Wheeler does not take Berkeley's final step: the
grounding of material reality in God's perception. Indeed, the fact that
a tree seems to exist in a strong sense, even when no one is looking at
it, is the key to Berkeley'S way of proving God's existence. Imagine a
god experimenting with billions of cosmic models until he finds one
that permits life. Would not these universes be "out there," observed by
the deity? There would be no need for flimsy creatures like ourselves,
observing and participating, to confer existence on these models.
Wheeler seems anxious to avoid this view. He argues that quantum
mechanics requires participator-observers in the universe regardless of
618 LOGIC AND PHILOSOPHY
whether there is an outside observer. In one of his metaphors, a uni-
verse without internal observers is like a motor without electricity. The
cosmos "runs" only when it is "guaranteed to produce somewhere, and
for some little length of time in its history-to-be, life, consciousness,
and observership." Internal observers and the universe are both essen-
tial to the existence of each other, even if the observers exist only in a
potential sense. This raises unusual questions. How strongly does a
universe exist before the first forms of life evolve? Does it exist in full
strength from the moment of the big bang, or does its existence get
stronger as life gets more complex? And how strong is the existence of
a galaxy, far removed from the Milky Way, in which there may be no
participator-observers? Does it exist only when it is observed by life in
another galaxy? Or is the universe so interconnected that the observa-
tion of a minute portion of it supports the existence of all the rest?
There is a famous passage in which William James imagines a thou-
sand beans flung onto a table. They fall randomly, but our eyes trace
geometrical figures in the chaos. Existence, wrote James, may be no
more than the order which our consciousness singles out of a disor-
dered sea of random possibilities. This seems close to Wheeler's vi-
sion. Reality is not something out there, but a process in which our
consciousness is an essential part. We are not what we are because the
world is what it is but the other way around. The world is what it is be-
cause we are what we are.
When relativity theory first won the day, many scientists and
philosophers with a religious turn of mind argued that the new theory
supported such a view. The phenomena of nature, said James Jeans, are
"determined by us and our experience rather than by a mechanical
universe outside us and independent of us." The physical world, wrote
Arthur Stanley Eddington, "is entirely abstract and without 'actuality'
apart from its linkage to consciousness." Most physicists today would
deny that relativity supports this brand of idealism. Einstein himself
vigorously opposed it. The fact that measurements of length, time, and
mass depend on the observer's frame of reference in no way dilutes the
actuality of a space-time structure independent of all observers.
Nor is it diluted by quantum mechanics. What bearing does the sta-
tistical nature of quantum laws have on the independent existence of
a structure to which those laws apply whenever it is observed? The fact
that observations alter state functions of a system of particles does not
entail that there is nothing "out there" to be altered. Einstein may have
Everything 619
thought that quantum mechanics implies this curious reduction of
physics to psychology, but there are not many quantum experts today
who agree.
In any case, belief in an external world, independent of human exis-
tence but partly knowable by us, is certainly the simplest view and the
one held today by the vast majority of scientists and philosophers. As
I have suggested, to deny this commonsense attitude adds nothing of
value to a theistic or pantheistic faith. Why adopt an eccentric termi-
nology if there is no need for it?
But this is not the place for debating these age-old questions. Let me
turn to a strange little book called Eureka: A Prose Poem, written by
Edgar Allan Poe shortly before his death. Poe was convinced that it was
his masterpiece. "What I have propounded will (in good time) revolu-
tionize the world of Physical & Metaphysical Science," he wrote to a
friend. "I say this calmly-but I say it." In another letter he wrote, "It
is no use to reason with me now; I must die. I have no desire to live
since I have done Eureka. I could accomplish nothing more." (I quote
from excellent notes in The Science Fiction of Edgar Allan Poe, edited
by Harold Beaver, Penguin Books, 1976.)
Poe wanted his publisher, George P. Putnam, to print 50,000 copies.
Putnam advanced Poe $14 for his "pamphlet," and printed 500 copies.
Reviews were mostly unfavorable. The book seems to have been taken
seriously only in France, where it had been translated by Baudelaire.
Now suddenly, in the light of current cosmological speculation, Poe's
prose poem is seen to contain a vast vision that is essentially a theist's
version of Wheeler's cosmology! As Beaver points out, the "I" in Poe's
"Dreamland" has become the universe itself:
By a route obscure and lonely,
Haunted by ill angels only,
Where an Eidolon, named NIGHT,
On a black throne reigns upright,
I have reached these lands but newly
From an ultimate dim Th ule-
From a wild weird clime that lieth, sublime,
Out of SPACE-out of TIME.
A universe begins, said Poe, when God creates a "primordial parti-
cle" out of nothing. From it matter is "irradiated" spherically in all di-
rections, in the form of an "inexpressibly great yet limited number of
620 LOGIC AND PHILOSOPHY
unimaginably yet not infinitely minute atoms." As the universe ex-
pands, gravity slowly gains the upper hand and the matter condenses
to form stars and planets. Eventually gravity halts the expansion and
the universe begins to contract until it returns again to nothingness.
The final "globe of globes will instantaneously disappear" (how Poe
would have exulted in today's black holes!) and the God of our universe
will remain "all in all."
In Poe's vision each universe is being observed by its own deity, the
way your eye watched the two particles dance in our created world of
i-space. But there are other deities whose eyes watch other universes.
These universes are "unspeakably distant" from one another. No com-
munication between them is possible. Each of them, said Poe, has "a
new and perhaps totally different series of conditions." By introducing
gods Poe implies that these conditions are not randomly selected. The
fine-structure constant is what it is in our universe because our deity
wanted it that way. In Poe's superspace the cyclical birth and death of
an infinity of universes is a process that goes on "for ever, and for ever,
and for ever; a novel Universe swelling into existence, and then sub-
siding into nothingness at every throb of the Heart Divine."
Did Poe mean by "Heart Divine" the God of our universe or a higher
deity whose eye watches all the lesser gods from some abode in super-
superspace? Behind Brahma the creator, goes Hindu mythology, is
Brahman the inscrutable, so transcendent that all we can say about
Brahman is Neti neti (not that, not that). And is Brahman being ob-
served by a supersupersupereye? And can we posit a final order of su-
perspace, with its Ultimate Eye, or is that ruled out by the contradiction
in standard set theory of the concept of a greatest aleph?
This is the great question asked in the final stanza of the Hymn of
Creation in the Rig Veda. The "He" of the stanza is the impersonal One
who is above all gods:
Whether the world was made or was self-made,
He knows with full assurance, He alone,
Who in the highest heaven guards and watches;
He knows indeed, but then, perhaps, He knows not!
It is here that we seem to touch-or perhaps we are still infinitely far
from touching-the hem of Everything. Let C. S. Lewis (I quote from
Chapter 2 of his Studies in Words) make the final comment: " 'Every-
thing' is a subject on which there is not much to be said."
Everything 621
Addendum
Charles Peirce once complained that universes are not as plen-
tiful as blackberries, but today there are many top cosmologists who
think otherwise. The notion that our universe is only one of an enor-
mous number, perhaps an infinite number, of other universes is now a
popular conjecture. Andrei Linde at Stanford University is the leading
proponent of such a view. His "multiverse" should not be confused
with the countless universes assumed in the many-worlds interpreta-
tion of quantum mechanics. In that theory the other universes are all
part of a single metaverse that is sprouting countless branches as its
monstrously complicated tree evolves. The many worlds of Linde and
his sympathizers are modern versions of Poe's beautiful vision-uni-
verses unrelated to ours, each with its unique set of laws, particles,
and physical constants.
There is no need to invoke intelligent design to explain why our uni-
verse is so finely tuned that it can evolve stars, planets, life, and such
strange creatures as you and me. Naturally we could exist, as the an-
thropic principle maintains, only in one of the myriad of universes
that had the necessary laws and constants. David Lewis and a few other
eccentric philosophers plunge even deeper into fantasy. They argue
that every logically possible universe-one free of contradictions-ex-
ists and is just as real as the one we are in! Leibniz considered what he
called logically "compossible" worlds that are in the mind of God, but
God chose only one, the best possible, to actually exist.
Defenders of Linde's multi verse , the many world's interpretation of
quantum mechanics, and Lewis's logically possible worlds tend to be
atheists or agnostics. The total ensemble of universes is all there is. No
creator occupies a transcendent realm. On the other hand, cosmological
and philosophical theists invoke Occam's razor. Is it not much simpler,
they maintain, to posit a wholly other deity who, for reasons we cannot
know, created the one universe in which we find ourselves? It is a uni-
verse destined either to last forever, or expand and die of the cold, or
eventually to contract, as Poe imagined, to be annihilated in a Big Crunch.
Bibliography
D. Deutsch, The Fabric of Reality: The Science of Parallel Universes and Its Impli-
cations, Penguin, 1997.
622 LOGIC AND PHILOSOPHY
J. Leslie, Universes, Oxford University Press, 1989.
D. Lewis, On the Plurality of Worlds, Blackwell, 1985.
A. Linde, Particle Physics and Inflationary Cosmology, Harvard Academic, 1990.
A. Linde, "The Self-Reproducing Inflationary Universe," Scientific American, No-
vember 1994, pp. 48-55.
A. N. Prior, "Existence," The Encyclopedia of Philosophy, MacmillanFree Press,
1967.
M. Rees, Before the Beginning: Our Universe and Others, Addison-Wesley, 1997.
L. Smolin, The Life of the Cosmos, Oxford University Press, 1997.
W. V. O. Quine "On What There Is," From a Logical Point of VIew, Harvard Uni-
versity Press, 1953; revised 1961.
Everything 623
XII
Miscellaneous
Chapter 47 Melody-Making
Machines
Mathematics and music! The most glaring
possible opposites of human thought! Yet
connected, mutually sustained!
-HERMANN VON HELMHOLTZ,
Popular Scientific Lectures
There is a trivial sense in which any work of art is a combina-
tion of a finite number of discrete elements. Not only that, the precise
combination of the elements can be expressed by a sequence of digits
or, if you will, by one enormous number.
Consider a poem. Assign distinct numbers to each letter of the al-
phabet, to each punctuation symbol, and so on. A certain digit, say
zero, can be used to separate the numbers. It is obvious that one long
string of digits can express the poem. If the books of a vast library con-
tain every possible combination of words and punctuation marks, as
they do in Jorge Luis Borges's famous story "The Library of Babel,"
then somewhere in the collection is every poem ever written or that can
be written. Imagine those poems coded as digital sequences and in-
dexed. If one had enough time, billions on billions of years, one could
locate any specified great poem. Are there algorithms by which one
could find a great poem not yet written?
Consider a painting. Rule the canvas into a matrix of minute cells.
The precise color of each cell is easily coded by a number. Scanning the
cells yields a chain of numbers that expresses the painting. Since num-
bers do not decay, a painting can be re-created as long as the number
sequence is preserved. Future computers will be able to reproduce a
painting more like the original than the original itself, since after a few
decades the original will have physically deteriorated to some extent.
If a vast art museum contains every combination of colored cells for
matrixes not exceeding a certain size, somewhere in that monstrous
museum will hang every picture ever painted or that can be painted.
Are there algorithms by which a computer could search a list of the mu-
seum's code numbers and identify a sequence for a great painting not
yet painted?
627
Consider a symphony. It is a fantastically complex blend of dis-
creteness and continuity; a violin or a slide trombone can move up
and down the scale continuously, but a piano cannot produce quarter-
tones. We know, however, from Fourier analysis that the entire sound
of a symphony, from beginning to end, can be represented by a single
curve on an oscilloscope. "This curve," wrote Sir James Jeans in Sci-
ence and Music (Dover, 1968), "is the symphony-neither more nor
less, and the symphony will sound noble or tawdry, musical or harsh,
refined or vulgar, according to the quality of this curve." On a long-
playing record a symphony is actually represented by one long space
curve.
Because curves can be coded to any desired precision by numbers, a
symphony, like a painting or a poem, can be quantized and expressed
by a number chain. A vast library recording all combinations of sym-
phonic sounds, would contain every symphony ever written or that
could be written. Are there algorithms by which a computer could scan
the number sequences of such a library and pick out a great symphony
not yet written?
Such procedures would, of course, be so stupendously complex that
man may never come close to formulating them, but that is not the
point. Do they exist in principle? Is it worthwhile to look for bits and
pieces of them? Consider one of the humblest of such aesthetic tasks,
the search for rules that govern the invention of a simple melody. Is
there a procedure by which a person or a computer can compose a
pleasing tune, using no more than a set of combinatorial rules?
If we restrict the tune to a finite length and a finite number of pure
tones and rhythms, the number of possible melodies is finite. John Stu-
art Mill, in his autobiography, recalls that as a young man he was once
"seriously tormented by the thought of the exhaustibility of musical
compositions." Suppose our tune is made up of just 10 notes chosen
from the set of eight notes in a single octave. The number of melodies
is the same as the number of 10-letter words that can be formed with
eight distinct letters, allowing duplications. It is 8
10
= 1,073,741,824,
and this without even considering varying rhythms which create, in ef-
fect, varying melodies. Many of these tunes will be dull (mi, mi, mi, mi,
mi, mi, mi, mi, mi, mi for instance), but some will be extremely pleas-
ing. Are there rules by which a computer or a person could pick out the
pleasing combinations?
Attempts to formulate such rules and embody them in a mechanical
628 MISCELLANEOUS
device for composing tunes have a curious history that began in 1650
when Athanasius Kircher. a German Jesuit, published in Rome his
Musurgia universalis sive ars magna Gonsoni et dissoni (see Figure
47.1). Kircher was an ardent disciple of Ram6n Lull, the Spanish me-
dieval mystic whose Ars magna derived from the crazy notion that sig-
nificant new knowledge could be obtained in almost every field simply
by exploring all combinations of a small number of basic elements. It
was natural that Kircher, who later wrote a 500-page elaboration of
Lull's "great art," would view musical composition as a combinatorial
problem. In his music book, he describes a Lullian technique of creat-
ing polyphony by sliding columns alongside one another, as with
Napier's bones, and reading off rows to obtain various permutations
and combinations. Like all of Kircher's huge tomes, the book is a fan-
tastic mix of valuable information and total nonsense, illustrated with
elaborate engravings of vocal cords, bones in the ears of various ani-
mals, birds and their songs, musical instruments, mechanical details of
music boxes, water-operated organ pipes with animated figures of an-
imals and people, and hundreds of other curious things.
The Lullian device described by Kircher was actually built, circa
1670, for the diarist Samuel Pepys, who owned a copy of Kircher's
music book and much admired it. The original machine, called
"musarithmica mirifica," is in the Pepys Museum at Pepys's alma
mater, Magdalene College, Cambridge.
During the early 18th century many German music scholars became
interested in mechanical methods of composition. Lorenz Christoph
Mizler wrote a book in 1739 describing a system that produced figured
bass for baroque ensemble music. In 1757 Bach's pupil, Johann Philipp
Kirnberger, published in Berlin his Ever-ready Composer of Polonaises
and Minuets, using a die for randomizing certain choices. In 1783 an-
other book by Kirnberger extended his methods to symphonies and
other forms of music.
Toward the close of the 18th century the practice of generating
melodies with the aid of tables and randomizers such as dice or teeto-
turns became a popular pastime. Maximilian Stadler, an Austrian com-
poser, published in 1779 a set of musical bars and tables for producing
minuets and trios with the help of dice. At about the same time, a Lon-
don music publisher, Welcker, issued a "tabular system whereby any
person, without the least knowledge of music, may compose ten thou-
sand minuets in the most pleasing and correct manner." Similar anony-
Melody-Making Machines 629
mous works were falsely attributed to well-known composers such as
C.P.E. Bach (son of Johann Sebastian Bach) and Joseph Haydn. The
"Haydn" work, "Gioco fHarmonico" ("Philharmonic Joke")' Naples,
1790, was discovered by the Glasgow mathematician Thomas H.
Q'Beirne to be a plagiarism. Its bars and tables are identical with
Stadler's.
The most popular work explaining how a pair of dice can be used "to
compose without the least knowledge of music" as many German
waltzes as one pleases was first published in Amsterdam and in Berlin
in 1792, a year after Mozart's death. The work was attributed to Mozart.
Most Mozart scholars say it is spurious, although Mozart was fond of
mathematical puzzles and did leave handwritten notes showing his
interest in musical permutations. (The same pamphlet was issued in
Bonn a year later, with a similar work, also attributed to Mozart, for
dice composition of country dances. The contredanses pamphlet was
reprinted in 1957 by Heuwekemeijer in Amsterdam.)
Mozart's Musikalisches Wiirfelspiel, as the waltz pamphlet is usually
called, has been reprinted many times in many languages. In 1806 it ap-
peared in London as Mozart's Musical Game, Fitted in an Elegant Box,
Showing by Easy System How to Compose an Unlimited Number of
Waltzes, Rondos, Hornpipes and Reels. In New York in 1941 the
garian composer and concert pianist Alexander Laszlo brought it out
under the title The Dice Composer, orchestrating the music so that it
could be played by chamber groups and orchestras. The system popped
up again in West Germany in 1956 in a score published by B. Schott.
Photocopies of Schott's charts and musical bars appear in the
tion booklet for The Melody Dicer, issued early in 1974 by Carousel
Publishing Corporation, Brighton, MA. This boxed set also includes a
pair of dice and blank sheets of music paper.
The "Mozart" system consists of a set of short measures numbered 1
through 176. The two dice are thrown 16 times. With the aid of a chart
listing 11 numbers in each of eight columns, the first eight throws de-
termine the first eight bars of the waltz. A second chart is used for the
second eight throws that complete the 16-bar piece. The charts are con-
structed so that the waltz opens with the tonic or keynote, modulates
to the dominant, then finds its way back to the tonic on its final note.
Because all bars listed in the eighth column of each chart are alike, the
11 choices (sums 2 through 12 on the dice) are available for only 14
bars. This allows the system to produce 1114 waltzes, all with a distinct
Atelody-AtakingAtachines 631
Mozartean flavor. The number is so large that any waltz you generate
with the dice and actually play is almost certainly a waltz never heard
before. If you fail to preserve it, it will be a waltz that will probably
never be heard again.
The first commercial recording of "Mozart" dice pieces was made by
O'Beirne. Both the randomizing of the bars and the actual playing of the
melodies was done by Solidac, a small and slow experimental com-
puter designed and built between 1959 and 1964 by the Glasgow firm
of Barr and Stroud, where O'Beirne was then chief mathematician. It
was the first computer built in Scotland. O'Beirne programmed Solidac
to play the pieces in clarinetlike tones, and a long-playing recording of
selected waltzes and contredanses was issued by Barr and Stroud in
1967. (This recording is no longer available.) O'Beirne is the author of
an excellent book on mathematical recreations, Puzzles and Paradoxes
(Oxford University Press, 1965). He has been of invaluable help in the
preparation of this account.
Other methods of producing tunes mechanically were invented in
the early 19th century. Antonio Calegari, an Italian composer, used two
dice for composing pieces for the pianoforte and harp. His book on the
system was published in Venice in 1801, and later in a French transla-
tion. The Melographicon, an anonymous and undated book issued in
London about 1805, is subtitled: "A new musical work, by which an in-
terminable number of melodies may be produced, and young people
who have a taste for poetry enabled to set their verses to music for the
voice and pianoforte, without the necessity of a scientific knowledge of
the art." The book has four parts, each providing music for poetry with
a certain meter and rhyme scheme. Dice are not used. One simply se-
lects any bar from group A, any from group B, and so on to the last let-
ter of the alphabet for that section.
A photograph of a boxed dice game appears in Plate 42 of The Oxford
Companion to Music, but without mention of date, inventor, or place
of publication. Apparently, it uses 32 dice, their sides marked to indi-
cate tones, intervals, chords, modulations, and so on. There also are
ivory men whose purpose, the caption reads, is "difficult to fathom."
In 1822 a machine called the Kaleidacousticon was advertised in a
Boston music magazine, The Euterpiad. By shuffling cards it could
compose 214 million waltzes. The Componium, a pipe organ that
played its own compositions, was invented by M. Winkel of Amster-
632 MISCELLANEOUS
dam and created a sensation when it was exhibited in Paris in 1824.
Listeners could not believe that the machine actually constructed the
melodies it played. Scientists from the French Academy investigated.
"When this instrument has received a varied theme," their report
stated, "which the inventor has had time to fix by a process of his own,
it decomposes the variations of itself, and reproduces their different
parts in all the orders of possible permutation .... None of the airs
which it varies lasts above a minute; could it be supposed that but one
of these airs was played without interruption, yet, the principle of vari-
ability which it possesses, it might, without ever resuming precisely the
same combination, continue to play ... during so immense a series of
ages that, though figures might be brought to express them, common
language could not."
The report, endorsed by physicist Jean Baptiste Biot, appeared in a
British musical journal (The Harmonicon 2, 1824, pp. 40-41). Winkel's
machine inspired a Vienna inventor, Baron J. Giuliani, to build a sim-
ilar device, the construction of which is given in detail on pages
198-200 of the same volume.
In 1865, a composing system called the Quadrille Melodist, invented
by J. Clinton, was advertised in The Euterpiad. By shuffling a set of
composing cards, a pianist at a quadrille party could "keep the
evening's pleasure going by means of a modest provision of
428,000,000 quadrilles."
Joseph Schillinger, a Columbia University teacher who died in 1943,
published his mathematical system of musical composition in a book-
let, Kaleidophone, in 1940. George Gershwin is said to have used the
system in writing Porgy and Bess. In 1940 Reitor Villa-Lobos, using the
system, translated a silhouette of New York City'S skyline into a piano
composition (see Figure 47.2). The Schillinger System of Musical Com-
position is a two-volume work by L. Dowling and A. Shaw, published
by Carl Fischer in 1941. A footnote on page 673 of Schillinger's eccen-
tric opus The Mathematical Basis of the Arts (Philosophical Library,
1948) says that he left plans for music-composing machines, protected
by patents, but nothing is said about their construction.
In the 1950s, information theory was applied to musical composition
by J. R. Pierce and others. In a pioneering article "Information Theory
and Melody," chemist Richard C. Pinkerton included a graph which he
called the "banal tune-maker." By flipping a coin to determine paths
Melody-Making Machines 633
.. 'itT ~ O R K SET TO 1\1'"
N E ~ ~ uSlC
- .,.-. "t;
'8 t!J at' r riP r "r'ac t!t C!t'ffu u'a IJIJ'U'r I riC!! r I r r
,trtw-rtr(tr 'clltlftar lE1rc!t lr
fl
frrlb'Prr l'EJl tJ "f I II
nm SKYLJNE HAS rrs OWN MUSICAL PAlTElIN'l'BANSLATED
FROM SlLHOOElTE TO MUSIC NOTES WITH nm HELP OF
THE SCHILLINGER SYSTEM OF MUSICAL COMPOSITION
Figure 47.2. New York skyline translated into music by Villa-Lobos, who used the
Schillinger system
along the network, one can compose simple nursery tunes. Most of
them are monotonous, but hardly more so, Pinkerton reminds us, than
"A Tisket, a Tasket."
During the 1960s and early 1970s the proliferation of computers and
the development of sophisticated electronic tone synthesizers opened
a new era in machine composition of music. It is now possible to write
computer programs that go far beyond the crude devices of earlier days.
Suppose one wishes to compose a melody in imitation of one by
Chopin. A computer analysis is made of all Chopin melodies so that the
computer has in its memory a set of "transition probabilities." These
give the probability that any set of one, two, three, or more notes in a
Chopin melody is followed by any other note. Of course, one must also
take into account the type of melody one wishes to compose, the
rhythms, the position of each note within the melody, the overall pat-
tern, and other things. In brief, the computer makes random choices
within a specified general structure, but these choices are subject to
rules and weighted by Chopin's transition preferences. The result is a
634
MISCELLANEOUS
"Markoff chain" melody, undistinguished but nevertheless sounding
curiously like Chopin. The computer can quickly dash off several hun-
dred such pieces, from which the most pleasing may be selected.
There is now a rapidly growing literature on computer composition,
not only of music in traditional styles but also music that takes full ad-
vantage of the computer's ability to synthesize weird sounds that re-
semble none of the sounds made by familiar instruments. Microtones,
strange timbres, unbelievably complex rhythms, and harmonics are no
problem. The computer is a universal musical instrument. In principle,
it can produce any kind of sound the human ear is capable of hearing.
Moreover, a computer can be programmed to play one of its own com-
positions at the same time it is composing it.
How can we sum up? Computers certainly can compose mediocre
music, frigid and forgettable, even though the music has the flavor of a
great composer. No one, however, has yet found an algorithm for pro-
ducing even a simple melody that will be as pleasing to most people of
a culture as one of their traditional popular songs. We simply do not
know what magic takes place inside the brain of a composer when he
creates a superior tune. We do not even know to what extent a tune's
merit is bound up with cultural conditioning or even with hereditary
traits. About all that can be said is that a good melody is a mixture of
predictable patterns and elements of surprise. What the proportions
are and how the mixture is achieved, however, still eludes everybody,
incl uding composers.
O'Beirne has called my attention to how closely some systems of
musical composition resemble the buzz-phrase generator (see Figure
47.3). This is a give-away of Honeywell Incorporated. Pick at random
any four-digit number, such as 8,751, then read off phrase 8 of module
A, phrase 7 of module B, and so on. The result is a SIMP (Simplified In-
tegrated Modular Prose) sentence. "Add a few more four-digit num-
bers," the instructions say, "to make a SIMP paragraph. After you have
mastered the basic technique, you can realize the full potential of SIMP
by arranging the modules in DACB order, BACD order, or ADCB order.
In these advanced configurations, some additional commas may be
required. "
SIMP sounds very much like authentic technical prose, but on closer
inspection one discovers that something is lacking. Computer-
generated melodies are perhaps less inane, closer to the random ab-
stract art of a kaleidoscope, but still something essential (nobody knows
Melody-Making Machines 635
SIMP TABLE A
1. In particular,
2. On the other hand,
3. However,
4. Similarly,
5. As a resultant implication,
6. In this regard,
7. Based on integral subsystem
considerations,
8. For example,
9. Thus,
o. In respect to specific goals,
SIMP TABLE B
1. a large portion of the interface
coordination communication
2. a constant flow of effective information
3. the characterization of specific criteria
4. initiation of critical subsystem
development
5. the fully integrated test program
6. the product configuration baseline
7. any associated supporting element
8. the incorporation of additional mission
constraints
9. the independent functional principle
O. a primary interrelationship between
subsystem and! or subsystem technologies
Figure 47.3. Honeywell's buzz-phrase generator for writing Simplified Integrated Modu-
lar Prose (SIMP)
what) is missing. Indeed, a good simple tune is much harder to com-
pose than an orchestral piece in the extreme avant-garde manner, so
loaded with randomness and dissonance that one hesitates to say, as
Mark Twain (or was it Bill Nye1) said of Wagner's music: It is better than
it sounds.
When a computer generates a melody that becomes as popular as
(think of the title of your favorite song), you will know that a colossal
breakthrough has been made. Will it ever occur? If so, when? Experts
disagree on the answers as much as they do on if and when a computer
will write a great poem, paint a great picture, or play grand-master
chess.
Addendum
Carousel, the company that brought out Mozart's The Melody
Dicer, later issued a similar set called The Scott Joplin Melody Dicer.
Using the same system of dice and cards, one can compose endless
rags of the Scott Joplin variety.
In 1977 a curious 284-page book titled The Directory of Tunes and
Musical Themes, by Denys Parsons, was published in England by the
mathematician G. Spencer Brown. Parsons discovered that almost
every melody can be identified by a ridiculously simple method. Put
636 MISCELLANEOUS
SIMP TABLE C
1. must utilize and be functionally
interwoven with
2. maximizes the probability of project
success and minimizes the cost
time required for
3. adds explicit performance limits to
4. necessitates that urgent
consideration be applied to
5. requires considerable systems
analysis and trade off studies to
arrive at
6. is further compounded, when
taking into account
7. presents extremely interesting
challenges to
8. recognizes the importance of other
systems and the necessity for
9. effects a significant implementation of
O. adds overriding performance
constraints to
SIMP TABLE D
1. the sophisticated hardware
2. the anticipated fourth-generation
equipment
3. the subsystem compatibility and
testing
4. the structural design, based on
system engineering concepts
5. the preliminary qualification
limit
6. the evolution of specifications
over a given time period
7. the philosophy of commonality
and standardization
8. the greater fight-worthiness
concept
9. any discrete configuration mode
down an asterisk for the first note. If the second note is higher, write
down U for "up." If lower, write D for "down." If the same, use R for
"repeat." Continue with succeeding notes until you have a sequence of
up to 16 letters. This is almost always sufficient to identify the melody.
For example * UDDUUUU is enough to key "White Christmas." The
book lists alphabetically in two sections, popular and classical, some
15,000 different sequences followed by the title of the work, its com-
poser, and the date.
The idea behind the buzz-phrase generator-random selection of
words and phrases to create prose or poetry-is an old idea. Rational
Recreations, a four-volume work by W. Hooper (the fourth edition was
published in London in 1794), has a section in Volume 2 on how to use
dice for composing Latin verse. The technique surely is much older.
Similar randomizing is used in modern computer programs that gen-
erate "poems" and various kinds of imitation prose.
First-grade teachers, who call it "stringing," use the technique for
teaching reading. Children are given a simple pattern sentence with
blanks into which they insert words. The New York Times Book Review
Melody-Making Machines 637
Gune 4, 1978), printed Randolph Hogan's list of buzzwords that en-
able anyone to write impressive literary criticism. (Someone should
do a similar list, perhaps already has, for art critics.) Mad Magazine
(October 1974) featured Frank Jacobs's 12 columns of buzzwords and
phrases for writing impeachment newspaper stories. Tom Koch (Mad
Magazine, March 1982) gave a similar technique for stand-up comics.
Jacobs returned in Mad (September 1982) with another 12 columns of
words and phrases for writing the lyrics of country-western songs.
Donald Knuth, after reading this chapter, told me that Glenn Miller
also studied under Schillinger, who is mentioned in the book The
Glenn Miller Story. Knuth also informed me that circa 1960 Fred
Brooks, a computer scientist at the University of North Carolina, Chapel
Hill, published some computer-generated hymn tunes based on data
from Methodist hymnals.
The best imitations so far of great music are the computer composi-
tions of David Cope, professor of music at the University of California,
Santa Cruz. His program EMI (Experiments in Musical Intelligence),
pronounced Emmy, has created very convincing music in the styles of
Mozart, Bach, Beethoven, Brahms, Chopin, Scott Joplin, and others.
Douglas Hofstadter has expressed amazement at how authentic the
music sounds. After his "Mozart's 42nd Symphony" was produced by
the college orchestra at Santa Cruz in 1997, Cope told a reporter:
"There's no expert in the world who could, without knowing its source,
say for certain that it's not Mozart .... "
Bibliography
D. Cope, Compositions and Music Style, Oxford University Press, 1991.
M. Gardner, "The Ars Magna of Ram6n Lull," Logic Machines and Diagrams,
McGraw-Hill, 1958; Revised edition University of Chicago Press, 1982.
J. Godwin, Athanasius Kircher. Thames and Hudson, 1979.
L. A. Hiller, Jr., "Computer Music," Scientific American, December 1959, pp.
109-120.
L. A. Hiller, Jr., and L. M. Isaacson, Experimental Music, McGraw-Hill, 1959.
R. Holmes, "Requiem for the Soul," New Scientist, August 9,1997, pp. 23-27. On
David Cope's music.
K. Jones, "Dicing with Mozart," New Scientist, December 14, 1991, pp. 26-29.
T. H. O'Beirne, "From Mozart to the Bagpipe with a Small Computer," Bulletin of
the Institute of Mathematics and its Applications 7, January 1971, pp. 3-8.
C. A. Pickover, "Music Beyond Imagination, Part III," Mazes of the Mind, St. Mar-
tin's, 1992.
638 MISCELLANEOUS
R. C. Pinkerton, "Information Theory and Melody," Scientific American, February
1956, pp. 77-86.
J. Reichardt (ed.), Cybernetic Serendipity: The Computer and the Arts, Frederick A.
Praeger, 1969.
F. Schillinger, Joseph Schillinger: A Memoir, De Capo, 1976.
P. A. Scholes, "Composition Systems and Mechanisms," The Oxford Companion to
Music, J. 0. Ward (ed.), Oxford University Press, 1970.
H. Von Foerster and J. W. Beauchamp (eds.), Music by Computers, Wiley, 1969.
Melody-Making Machines 639
Chapter 48 Mathematical
Zoo
There has never been a zoo designed to display animals with
features of special interest to recreational mathematicians, yet such a
zoo could be both entertaining and instructive. It would be divided, as
I visualize it, into two main wings, one for live animals, the other for
pictures, replicas, and animated cartoons of imaginary creatures. Pa-
trons of the "mathzoo" would be kept informed of new acquisitions by
a newsletter called ZOONOOZ (with the permission of the Zoological
Society of San Diego, which issues a periodical of that name), a title
that is both palindromic and the same upside down.
A room of the live-animal wing would contain microscopes through
which one could observe organisms too tiny to be seen otherwise. Con-
sider the astonishing geometrical symmetries of radiolaria, the one-
celled organisms that flourish in the sea. Their intricate silica skeletons
are the nearest counterparts in the biological world to the patterns of
snow crystals. In his Monograph of the Challenger Radiolaria, the Ger-
man biologist Ernst Haeckel described thousands of radiolaria species
that he discovered on the Challenger expedition of 1872-76. The book
contains 140 plates of drawings that have never been excelled in dis-
playing the geometric details of these intricate, beautiful forms.
Figure 48.1, which originally appeared in Haeckel's book, is of spe-
cial interest to mathematicians. The first radiolarian is basically spher-
ical, but its six clawlike extensions mark the corners of a regular
octahedron. The second skeleton has the same solid at its center. The
third is a regular icosahedron of 20 faces. The fifth is the 12-sided do-
decahedron. Other plates in Haeckel's book show radiolaria that ap-
proximate cubical and tetrahedral forms.
It is well known that there are just five Platonic solids, three of which
have faces that are equilateral triangles. Not so widely known is that
640
Figure 48.1. Radiolaria skeletons in Ernst Haeckel's Monograph of the Challenger Radi-
olaria
there are an infinite number of semi-regular solids also with sides that
are equilateral triangles. They are called "deltahedra" because their
faces resemble the Greek letter delta. Only eight deltahedra are convex:
those with 4, 6, 8, 10, 12, 14, 16, and 20 faces. The missing 18-sided
convex deltahedron is mysterious. One can almost prove it should
exist, and it is not so easy to show why it cannot. It is hard to believe,
but the proof that there are only eight convex deltahedra was not
known until B. 1. van der Waerden and Hans Freudenthal published it
in 1947. If concavity is allowed, a deltahedron can have any number of
faces of eight or greater.
The four-faced deltahedron is the regular tetrahedron, the simplest of
the Platonic solids. The six-faced deltahedron consists of two tetrahe-
dra sharing one face. Note the fourth radiolarian in Haeckel's picture.
Mathematical Zoo 641
It is a lO-faced deltahedron, or rather one that is inflated slightly toward
a sphere. It may surprise you to learn that there are two topologically
distinct eight-sided deltahedra. One is the familiar regular octahedron.
Can you construct a model of the other one (it is not convex)?
Surfaces of radiolaria are often covered with what seems to be a net-
work of regular hexagons. The regularity is particularly striking in
Aulonia hexagona, shown in Figure 48.2. Such networks are called
"regular maps" if each cell has the same number of edges and each
vertex has the same number of edges joined to it. Imagine a regular
tetrahedron, octahedron, or icosahedron inflated like a balloon but pre-
serving its edges as lines on the resulting sphere. The tetrahedron will
form a regular map of triangles with three edges at each vertex, the oc-
tahedron will form a map of triangles with four edges at each vertex,
and the icosahedron will form a map of triangles with five edges at
each vertex. Inflating a cube produces a regular map of four-sided cells
with three edges at each vertex. Inflating a dodecahedron produces a
regular map of pentagons with three edges at each vertex.
Figure 48.2. The radiolarian Aulonia hexagona
Aulonia hexagona raises an interesting question. Is it possible to
cover a sphere with a regular map of hexagons, three edges at each ver-
tex? Only the topological properties of the map concern us. The hexa-
gons need not be regular or even convex. They may have any size or
shape, and their edges may twist and curve any way you like provided
they do not intersect themselves or one another and provided three of
them meet at each vertex.
The answer is no, and it is not hard to prove impossibility with a fa-
642 MISCELLANEOUS
mous formula that Leonhard Euler discovered for the skeletons of all
simply connected (no "holes") polyhedra. The formula is F + C - E = 2,
where the letters stand for faces, corners, and edges. Since all such
polyhedra can be inflated to spheres, the formula applies also to maps
on the sphere. In Chapter 13 of Enjoyment of Mathematics, by Hans
Rademacher and Otto Toeplitz, you will find it explained how Euler's
formula can be used in proving that no more than five regular maps can
be drawn on a sphere and that therefore no more than five regular con-
vex solids can exist. As a second problem, can you use Euler's formula
to show that a regular map of hexagons is impossible on a sphere?
D' Arcy Wentworth Thompson, whose classic work On Growth and
Form contains an excellent section on radiolaria, liked to tell about a
biologist who claimed to have seen a spherical radiolarian covered
with a perfect map of hexagons. But, said Thompson, Euler had proved
this impossible. "That," replied the biologist, "proves the superiority of
God over mathematics."
"Euler's proof happened to be correct," writes Warren S. McCulloch
in an essay where I found this anecdote, "and the observations, inac-
curate. Had both been right, far from proving God's superiority to logic,
they would have impugned his wit by catching him in a contradic-
tion." If you look carefully at the picture of Aulonia hexagona you will
see cells with more or fewer than six sides.
Under electron microscopes in our zoo's micro room would be the
many viruses that have been found to crystallize into macromolecules
shaped like regular icosahedra: the measles virus, the herpes, the triola
iridescent, and many others (see R. W. Horne's article cited in the bib-
liography). Viruses may also have dodecahedral shapes, but as far as I
know this remains unsettled. Another modern discovery is that some
viruses, such as the one that causes mumps, are helical. It had formerly
been thought that helical structures were restricted to plants and to
parts of animals: hair, the umbilical cord, the cochlea of the human ear,
the DNA molecule, and so on. A section of our zoo would feature such
spectacular helical structures as molluscan seashells, the twisted horns
of certain sheep, goats, antelopes, and other mammals, and such cu-
riosities as "devil's corkscrews"-the huge fossil burrows of extinct
beavers.
In the macro world of fishes, reptiles, birds, insects, mammals, and
human beings the most striking geometrical aspect of the body is its
overall bilateral symmetry. It is easy to understand why this symmetry
Mathematical Zoo 643
evolved. On the earth surface gravity creates a marked difference be-
tween up and down, and locomotion creates a marked difference be-
tween front and back. But for any moving, upright creature the left and
right sides of its surroundings-in the sea, on the land, or in the air-
are fundamentally the same. Because an animal needs to see, hear,
smell, and manipulate the world equally well on both sides, there is an
obvious survival value in having nearly identical right and left sides.
Animals with bilateral symmetry are of no interest for our mathzoo-
you can see them at any zoo-but it would be amusing to assemble an
exhibit of the most outrageous violations of bilateral symmetry. For ex-
ample, an aviary would feature the crossbill, a small red bird in the
finch family that has its upper and lower beaks crossed in either of the
two mirror-image ways. The bird uses its crossed bill for prying open
evergreen cones in the same way a cook uses a plierlike device to pry
off the lid of a jar or can. A medieval legend has it that the bill became
twisted as the bird was trying vainly to pull the nails from the cross
when Jesus was crucified; in the effort the bird's plumage became
stained with blood. In the same aviary would be some wry-billed
plovers from New Zealand. The entire bill of this funny bird is twisted
to the right. The bill is used for turning over stones to find food. As you
would expect, foraging wry-billed plovers search mainly on the right.
An aquarium in our mathzoo would exhibit similar instances of pre-
posterous asymmetry among marine life: the male fiddler crab, for ex-
ample, with its enormous left (or right) claw. Flatfish are even more
grotesque examples. The young are bilaterally symmetric, but as they
grow older one eye slowly migrates over the top of the head to the
other side. The poor fish, looking like a face by Picasso, sinks to the bot-
tom, where it lies in the ooze on its eyeless side. The eyes on top turn
independently so that they can look in different directions at the same
time.
Another tank would contain specimens of the hagfish. This absurd
fish looks like an eel, has four hearts, teeth on its tongue, and repro-
duces by a technique that is still a mystery. When its single nostril is
clogged, it sneezes. The hagfish is in our zoo because of its amazing
ability to tie itself into an overhand knot of i ~ h e r handedness. By slid-
ing the knot from tail to head it scrapes sliriie from its body. The knot
trick is also used for getting leverage when the hagfish tears food from
a large dead fish and also for escaping a predator's grasp (see David
Jensen's article listed in the bibliography).
644 MISCELLANEOUS
Knots are, of course, studied by mathematicians as a branch of
ogy. Another exhibit in our aquarium would be beakers filled with Leu-
cothrix mucor, a marine bacterium shaped like a long filament. A
magnifying glass in front of each beaker would help visitors see the
flimsy filaments. They reproduce by tying themselves into knots--over-
hands, figure-eights, even more complicated knots-that get tighter and
tighter until they pinch the filament into two or more parts (see Thomas
D. Brock's paper listed in the bibliography). Do higher animals ever tie
parts of themselves into knots? Fold your arms and think about it.
The most popular of our aquarium exhibits would probably be a tank
containing specimens of AnabJeps, a small Central
ican carp sometimes called the stargazer. It looks as if it has four eyes.
Each of its two bulging eyes is divided into upper and lower parts by
an opaque band. There is one lens but separate corneas and irises. This
little BEM (bug-eyed monster) swims with the band at water level. The
two upper "eyes" see above water while the two lower ones see below.
The AnabJeps is in our zoo because of its asymmetric sex life. The
young are born alive, which means that the male must fertilize the eggs
inside the female. The female opening is on either the left side or the
right. The male organ also is either on the left or the right. This makes
it impossible for two fish of the same handedness to mate. Fortunately
both males and females are equally left- or right-sexed, and so the
species is in no danger of extinction.
In a larger tank one would hope to see some narwhals, although until
now they have not survived in captivity. This small whale, from north-
polar seas, has been called the sea unicorn because the male has a sin-
gle Hhorn" that projects straight forward from its upper jaw and is about
half the whale's body length. Both sexes are born with two small side-
by-side teeth. The teeth stay small on the female, but the male's left
tooth grows into an ivory tusk, straight as a javelin and seven to 10 feet
long. This ridiculous tooth, the longest in the world, has a helical
groove that spirals around it like a stripe on a barber pole. Nobody
knows what function the tusk serves. It is not used for stabbing enemies
or punching holes in ice, but during the mating season narwhals have
been seen fencing with each other, so that its main purpose may be a
role in sexual ritual (see John Tyler Bonner's article in the bibliogra-
phy). Incidentally, the narwhal is also unusual in having a name start-
ing with the letter n. It is easy to think of mammalian names beginning
with any letter of the alphabet except n.
Mathematical Zoo 645
Among snakes, s p ~ c i e s that sidewind across the desert sands are
mathematically interesting because of their highly asymmetric tracks:
sets of parallel line segments that slant either right or left at angles of
about 60 degrees from the line of travel. Many species of snakes are ca-
pable of sidewinding, notably the sidewinder itself, a small rattlesnake
of Mexico and the U.S. Southwest, and the African desert viper. Exactly
how sidewinding works is rather complicated, but you will find it
clearly explained in Carl Gans's article.
The insect room of our mathzoo would certainly display the nests of
bees and social wasps. They exhibit a hexagonal tessellation even more
regular than the surfaces of radiolaria. A large literature, going back to
ancient Greece and still growing, attempts to explain the factors that
playa role in producing this pattern. D' Arcy Thompson, in his book
cited earlier, has a good summary of this literature. In times before Dar-
win bees were usually regarded as being endowed by the Creator with
the ability to design nests so that the cells use the least amount of wax
to hold a maximum amount of honey. Even Darwin marveled at the
bee's ability to construct a honeycomb, calling that ability "the most
wonderful of known instincts," and "absolutely perfect in economizing
labor and wax."
Actual honeycombs are not as perfect as early writers implied, and
there are ways of tessellating space with polyhedral cells that allow an
even greater economy of wax. Moreover, it seems likely that the hon-
eycomb pattern is less the result of evolution finding a way to con-
serve wax than an accidental product of how bees use their bodies and
the way they form dense clusters when they work. Surface tension in
the semiliquid wax may also playa role. The matter is still far from set-
tled. The best discussion I know is a paper by the Hungarian mathe-
matician L. Fejes T6th.
No actual animal propels itself across the ground by rolling like a
disk or a sphere, but our insect room would be incomplete without an
exhibit of a remarkable insect that transports its food by rolling near-
perfect spheres. I refer to the dung beetle, the sacred scarab of ancient
Egypt. These sometimes beautiful insects (in the Tropics they have
bright metallic colors) use their flat, sharp-edged heads as shovels to
dig a supply of fresh ordure that their legs then fashion into spheres. By
pushing with its hind legs and walking backward the dung beetle will
roll the little ball to its burrow where it will be consumed as food. No
one has described the process with more literary skill and humor than
646 MISCELLANEOUS
the French entomologist, Jean Henri Fabre, in his essay on "The Sacred
Beetle."
Our zoo's imaginary wing would lack the excitement of living crea-
tures but would make up for it in wild fantasy. In Flaubert's Temptation
of st. Anthony, for example, there is a beast called the Nasnas that is half
of an animal bisected by its plane of symmetry. Jorge Luis Borges, in his
delightful Book of Imaginary Beings, refers to an earlier invention of
such a creature by the Arabs. L. Frank Baum's fantasy, Dot and Tot of
Merryland, tells of a valley inhabited by wind-up animals. The toys are
kept wound by a Mr. Split, whose left half is bright red and whose right
half is white. He can unhook his two sides, each of which hops about
on one leg so that he gets twice as much winding done. Conversing
with a half of Mr. Split is difficult because Mr. Left Split speaks only the
left halves of words and Mr. Right Split speaks only the right halves.
A variety of mythical "palindromic" beasts violate front and back
asymmetry by having identical ends. Borges writes of the fabled am-
phisbaena (from the Greek for "go both ways"), a snake with a head at
each end. Dante puts the snake in the seventh circle of Hell, and in Mil-
ton's Paradise Lost some of Satan's devils are turned into amphisbae-
nas. Alexander Pope writes in his Dunciad:
Thus Amphisbaena (I have read)
At either end assails;
None knows which leads,
or which is led,
For both Heads are but Tails.
The fable is not without foundation. There are actual snakes called
amphisbaenas that crawl both ways and have such tiny eyes that it is
hard to distinguish one end from the other. If a flatworm's head is cut
off, another grows at the base of the severed head, so palindromic ani-
mals actually can exist. In Baum's John Dough and the Cherub one
meets Duo, a dog with a head and forelegs at both ends (see Figure
48.3). The animal anticipates the Pushmi-Pullyu (it has a two-horned
head at each end) that flourishes in the African jungle of Hugh Lofting's
Dr. Dolittle books.
Rectangular parallelepipeds are never the parts of real animals, but
in Baum's Patchwork Girl of Oz there is a block-headed, thick-skinned,
dark blue creature called the Woozy (see Figure 48.4). The animal's
head, body, legs, and tail are shaped like blocks. It is friendly as long
Mathematical Zoo 647
48.3. Duo, a }-' ...........-"-'.U-i.
L. Frank Baum's Woozy
as no one says Woozy so angry that
eyes dart fire. Nobody, knows what ..L ......... " ~ . c . . " ' " .. t!_.o.rY'r'"
that is makes it so furious. Borges reminds us the
Paul Bunyan
slopes
lumberjacks hard-boil them and use them for dice. In Stanley G.
Martian Odyssey," a
silica bricks that
648
Baum also imagined spherical creatures. The Roly-Rogues, in Queen
Zixi of Ix, are round like a ball and attack enemies by rolling at them. In
John Dough and the Cherub, one of the main characters is Para Bruin, a
large rubber bear that likes to roll into a rubber ball and bounce around.
Borges, writing about animals in the form of spheres, tells us that
Plato, in the Laws, conjectures that the earth, planets, and stars are
alive. The notion that the earth is a living, breathing organism was later
defended by such mystics as Giordano Bruno, Kepler, the German psy-
chologist Gustav Theodor Fechner, and Rudolf Steiner (who broke
away from theosophy to found his rival cult of anthroposophy). The
same notion is basic to the plot of one of Conan Doyle's stories about
Professor George Edward Challenger of Lost World fame. When Profes-
sor Challenger drills a deep hole through the earth's epidermis, in a
story called "When the Earth Screamed," the planet howls with pain.
Rotating wheels and propellers are common mechanisms for trans-
porting man-made vehicles across ground, and through the sea and the
air, but until a few decades ago it was assumed that evolution had been
unable to exploit rotational devices for propulsion. Biologists were
amazed to discover that the flagella of bacteria actually spin like pro-
pellers (see the article by Howard C. Berg).
The imaginary wing of our zoo would display two of Baum's crea-
tures that use the wheel for propulsion. In Ozma of Oz Dorothy has an
unpleasant encounter with the Wheelers, a race of fierce, four-legged
humanoids that have wheels instead of feet (see Figure 48.5). In The
Scarecrow of Oz we read about the Ork, a huge bird with a propeller at
the tip ofits tail (see Figure 48.6). The propeller can spin both ways, en-
abling the bird to fly backward as well as forward.
I know of only two imaginary beasts that bend themselves into
wheels and roll across the ground. From time to time, in most parts of
the world, people have claimed to have seen "hoop snakes" that bite
their tails to form a hoop and then go rolling across the terrain. Some
snakes, such as the American milk snake, travel by gathering their body
into large vertical loops and pushing forward so rapidly that they cre-
ate an optical illusion of a rolling ring. These animals may be the ori-
gin of hoop-snake fables.
The Dutch artist M. C. Escher made several pictures featuring his
curl-up, the beast shown in Figure 48.7. This unlikely animal moves
slowly on six humanlike feet, but when it wants to go faster it curls up
and rolls like a wheel.
Mathematical Zoo 649
Figure 48.S. A Wheeler
Figure 48.6. The Ork
Most animals, particularly the earthworm, may be thought of as being
basically toroidal-a shape topologically equivalent to a doughnut.
There must be many science-fiction animals shaped like toruses, but I
can recall only the undulating silver ringfish, floating on the canals of
Ray Bradbury's Martian Chronicles, that closes like an eye's iris around
food particles.
Topologists know that any torus can be turned inside out through a
hole in its surface. There is no parallel in earth zoology, but there is a
spherical organism called volvox that actually does turn inside out
through a hole. It is a strange freshwater-pond colony of hundreds of
flagellated cells bound together in a spherical jellylike mass that rotates
650 MISCELLANEOUS
Collection Gemeente-museum-the
48.1. The
it wants to.
an CllJ..LJIUCll Ull::t!=i.LI.-lUU by M. c ....... "' ......&.1......... can Toil like a wheel when
by photosyn-
or an moves
to decide whether it is a colony or a single
Young volvox the mother sphere, but the cells
At
the inside of the mother,
the sphere. When the infant reaches a certain it breaks away
from the mother the hole! Flagella quickly
eventually
dies splitting open and allowing her offspring to escape, one of the
earliest examples on of nonaccidental death
could have considered volvox earlier, but I kept it for now to in-
troduce the ta-ta, a of animal capable
so
and described the ta-ta in his only book, Bogey Beasts, a rare col-
lection of verses
651
THE TA-TA
There is a cosy Kitchen
Inside his roomy head
Also a tiny bedroom
In which he goes to bed.
So when his walk is ended
And he no more would roam
Inside out he turns himself
To find himself at Home.
He cleared away his brain stuff
Got pots and pans galore!
Sofas, chairs, and tables,
And carpets for the floor.
He found his brains were useless,
As many others would
If they but tried to use them
A great unlikelihood.
He pays no rent, no taxes
No use has he for pelf
Infested not with servants
He plays with work himself
And when his chores are ended
And he would walk about,
Outside in he turns himself
To get himself turned out.
Addendum
I was mistaken in saying that no animal propels itself across the
ground by rolling like a disk or sphere. Brier Lielst, Philip Schultz,
and a geologist with the appropriate name of Paul Pushcar, were among
many who informed me of a National Geographic television special on
March 6, 1978, about the Namib desert of Africa. It showed a small
spider that lives in burrows in the sides of sand dunes. When attacked
by a wasp, it extends its legs like the spokes of a wheel and escapes by
rolling down the dune.
Peter G. Trei of Belgium sent me a copy of a note in Journal of Mam-
malogy (February 1975) by Richard R. Tenaza, an American zoologist.
652 MISCELLANEOUS
.......................... describes how on Siberut, an island of
by pangolins, a species of scaly anteater,
ball and roll rapidly
a slope (see name
a Malay word meaning "to roll. JJ
48.8. An uncurled pangolin
Roy L. Caldwell, a zoologist, wrote to me about an inch-long crus-
in its mouth-paris, they are usually
path taken usually a straight line, the animal can actually climb
a five-degree slope, and it can a speed of about six cm/sec."
me a wood
pill-bugs) that, a
and roll away. Hay said his children call them roly-polys. While work-
car, with a often
.......... JL .......................... vehicles, released a tiny
they move so slowly that my work is finished before they constitute a
in a that
had this interesting comment on an earlier article, "Why An-
imals Run on Not on (see the bibliography):
653
The concept of animals developing wheels for locomotion is not so far-
fetched. A wheel with a diameter of one foot has a circumference a lit-
tle over three feet. If it were mounted on a bone-bearing joint, with
flexible veins and arteries, and a continuous series of circumferential
pads (as on a dog's paw), the wheel could be wound back one turn by its
internal muscles, then placed on the ground and rotated forward two full
turns, traveling about six and a quarter feet. While one wheel (or pair of
wheels in a four-legged animal) is driving the creature along the ground,
the other would be lifted up and rotated back in preparation for its next
turn at propulsion. At a speed often m.p.h., the creature would be trav-
eling about 15 feet per second-not an impossible pace.
Matthew Hodgart, writing from England, reminded me that the
human animal is capable of moving by repeated somersaults, cart-
wheels, forward and back flips, and that two persons can grab each
other's feet and roIl like a hoop. Hodgart quoted these lines from An-
drew Marvell's poem "To His Coy Mistress":
Let us roll all our strength, and all
Our sweetness, up into one ball:
And tear our pleasures with rough strife,
Through the iron gates of Life.
"I don't quite know what's going on here," Hodgart adds.
Chandler Davis supplemented my list of imaginary creatures that
roll by calling attention to such an animal in George MacDonald's fan-
tasy The Princess and Curdie. Ian F. Rennie thought I should have men-
tioned the Wumpetty-Dumps, found in The Log of the Ark, by Kenneth
Walker and Geoffrey Boumphrey.
Is the narwhal the only animal with a name starting with N? Garth
Slade cited the numbat, a small marsupial that lives in Western Aus-
tralia. An article in Word Ways (May 1973) gave two other examples:
the nutria, a web-footed South American aquatic rodent (now also
flourishing on the Gulf Coast and the coasts of the Pacific northwest),
and the nilgai, an antelope in India that is commonly called a "blue
bull" because of its bluish-gray color. It is curious that the best one can
do with common names are the colloquial nag and nanny-goat.
Arthur C. Statter sent his reasons for thinking that the drawing by
Haeckel which is reproduced in Figure 48.1 (I picked it up from D' Arcy
Thompson's On Growth and Form), was one of many drawings that
Haeckel deliberately faked. The forms shown, Statter says, simply do
654 MISCELLANEOUS
not exist. I have not tried to investigate this, and would welcome opin-
ions from radiolaria experts.
Rufus P. Isaacs, commenting on the impossibility of tessellating a
sphere with hexagons, sent a proof of a surprising theorem he discov-
ered many years earlier. If a sphere is tessellated with hexagons and
pentagons, there must be exactly 12 pentagons, no more and no fewer.
A soccerball is tessellated with 20 hexagonal "faces" and 12 that are
pentagons. In 1989 chemists succeeded in creating the world's tiniest
soccerball-a carbon molecule with 60 atoms at the vertices of a spher-
ical structure exactly like that of a soccerball (see Figure 48.9). It is
called a buckyball, or more technically, a buckminsterfullerene, after its
resemblance to Buckminster Fuller's famous geodesic domes. It be-
longs to a class of highly symmetrical molecules called fullerenes.
figure 48.9. The "buckyball" molecule-the world's smallest soccerball
The buckyball is known to geometers as a truncated icosahedron be-
cause it can be constructed by slicing off the 12 corners of a regular
icosahedron. It is still not clear what properties this third form of car-
bon (the other two are graphite and diamond) might have. Because of
the molecule's near spherical shape, it might provide a marvelous lu-
bricant. (See "Buckyball: The Magic Molecule," Edward Edelson, Pop-
ular Science, August 1991, p. 52ff.)
It also can be shown that if a sphere is tessellated with hexagons and
triangles, there must be an even number of hexagons and exactly four
triangles. These results suggest the following general question: What
are the integral values of k such that the sphere can be tessellated with
hexagons and exactly k polygons of side n? As far as I know, this ques-
Mathematical Zoo 655
convex
48.10. Two views
convex
656
LJ>c'L .. LL.. cases
dS(:sptabledron .. an .ii"i
w
8U::lBO u\J .. l.a.,u,IJ ........' ......... that is u..L.L.I...lLV.:J1.
"deceptahedron." It is a pity that teachers are not as familiar with the
eight convex deltahedra as they are with the five Platonic solids be-
cause constructing models and proving the set unique by way of Euler's
formula are splendid classroom challenges.
As early as 1993 more than 1,500 papers were published on bucky-
balls, and a thousand more have appeared since then. In 1996 the Nobel
Prize in chemistry went to three discoverers of buckyballs: Robert Curl,
Harold Kyoto, and Richard Smalley.
Answers
Figure 48.11 shows the answer to the first problem: an eight-
sided deltahedron (all faces equilateral triangles) that is not a regular
octahedron. The regular octahedron has four edges meeting at each
corner. On this solid two corners are meeting spots for three edges, two
for four edges, and two for five edges.
Figure 48.11. The "other" eight-sided
deltahedron
The second problem was to use Euler's formula, F + C - E = 2, to show
that no sphere can be covered with a "regular map" of hexagons, each
vertex the meeting point of three edges. Assume such a map exists.
Each hexagon has six edges and six comers. Therefore if the hexagons
did not share corners and edges, there would be six times as many
edges as faces. Each comer is shared, however, by three faces; therefore,
the number of corners in such a map must be 6F/3. Similarly, each
edge is shared by two faces; therefore, the number of corners in such a
map must be 6F/2. Substituting these values in Euler's formula gives
the equation F + 6F/3 - 6F/2 = 2, which simplifies to F + 2F - 3F= 2, or
0= 2. This contradiction proves the original assumption to be false.
What happens when the above argument is applied to the regular
Mathematical Zoo 657
maps formed by the edges of the five Platonic solids? In each case we
get a formula that gives F a unique value: 4, 8, and 20 for the tetrahe-
dron, octahedron, and icosahedron respectively, 6 for the cube, and 12
for the dodecahedron. Since a regular polyhedron cannot have faces
with more than six edges, we have proved that no more than five reg-
ular solids can exist.
Euler's formula also underlies an elementary proof that there are ex-
actly eight convex deltahedra. See the paper by Beck, Bleicher, and
Crowe cited in the bibliography.
Bibliography
J. T. Bonner, "Volvox: A Colony of Cells," Scientific American, May 1950, pp.
52-55.
J. T. Bonner, "The Horn of the Unicorn," Scientific American, March 1951, pp.
42-46.
J. L. Borges, The Book of Imaginary Beings, Dutton, 1969.
T. D. Brock, "Knots in Leucothrix Mucor," Science, Vol. 144, 1964, pp. 870-71.
C. Gans, "How Snakes Move," Scientific American, June 1970, pp. 82-96.
M. Gardner, The New Ambidextrous Universe, W. H. Freeman, 1989.
R. W. Horne, "The Structure of Viruses," Scientific American, January 1963, pp.
48-52.
D. Jensen, "The Hagfish," Scientific American, February 1966, pp. 82-90.
A. C. Neville, Animal Asymmetry, London: Edwin Arnold, 1976.
On The Bee Honeycomb
D. F. Siemens, Jr., "The Mathematics of the Honeycomb," Mathematics Teacher,
April 1965, pp. 334-37.
L. Fejes Toth, "What the Bees Know and What They Don't Know." Bulletin of the
American Mathematical Society, Vol. 70, 1964, pp. 468-86.
On Animal Wheels
H. C. Berg, "How Bacteria Swim," Scientific American, August 1975, pp. 36-44.
J. Diamond, "Why Animals Run On Legs, Not On Wheels," Discover, September
1983, pp. 64-67.
L. James, "Bacteria's Motors Work in Forward, Reverse, and Twiddle," Smithson-
ian, September 1983, pp. 127-34.
W. Marston, "Roll with it: The Life and Times of a Somersaulting Crustacean," The
Sciences, January/February 1994, p. 12.
On Deltahedra
A. Beck and D. Crowe, "Deltahedra," Excursions into Mathematics, Worth, 1969,
pp.21-26.
658 MISCELLANEOUS
J. D. Bernal, "The Structure of Liquids." Scientific American, August 1960, pp.
124-30. The author maintains that molecules in liquids form deltahedral struc-
tures.
H. M. Cundy, "Deltahedra," Mathenwtical Gazette, Vol. 36,1952, pp. 263-66.
W. E. McGowan, "A Recursive Approach to the Construction of the Deltahedra,"
Mathematics Teacher, Vol. 71, 1978, pp. 204-10.
C. W. Trigg, "An Infinite Class of Deltahedra," Mathematics Magazine, Vol. 51,
1978, pp. 55-58.
On Buckyballs
H. Aldersey-Williams, The Most Beautiful Molecule: The Discovery of the Bucky-
ball, Wiley, 1995.
F. Chung and S. Sternberg, "Mathematics and the Buckyball," American Scientist,
Vol. 81, January/February 1993, pp. 56-71.
R. Curl and R. Smalley, "Fullerines," Scientific American, October 1991, pp. 54-63.
R. Smalley, "Great Balls of Carbon," The Sciences, Vol. 31, March/April 1991, pp.
22-28.
C. Wu, "Buckyballs Can Come From Outer Space," Science News, Volume 157,
March 25, 2000, p. 490.
Mathematical Zoo 659
Chapter 49
This sentence no verb.
-DOUGLAS R. HOFSTADTER,
Codel, Escher, Bach: an
Eternal Colden Braid
C;odel, Escher, Bach
Every few decades an unknown author brings out a book of
such depth, clarity, range, wit, beauty, and originality that it is recog-
nized at once as a major literary event. Godel, Escher, Bach: an Eternal
Golden Braid, a hefty (777 pages) volume published by Basic Books
(1979), is such a work. The author (and the illustrator and typesetter)
is Douglas R. Hofstadter, then a young computer scientist at Indiana
University who is the son of the well-known physicist Robert Hofs-
tadter.
What can Kurt G6del, M. C. Escher, and Johann Sebastian Bach have
in common? The answer is symbolized by the objects shown in the
photograph that is Figure 49.1, and in the photograph on the book's
jacket. In each photograph two wood blocks floating in space are illu-
minated so that their shadows on the three walls meeting at the comer
of a room form the initials of the three surnames G6del, Escher, and
Bach. More precisely, the upper block casts "GEB" (G6del, Escher,
Bach) the heading of the book's first half, and the lower block casts
"EGB" (Eternal Golden Braid) the heading of its second half. The letters
G, E, and B may be thought of as the labels for three strands that are
braided by repeatedly switching a pair of letters. Six steps are required
to complete a cycle from GEB (through EGB) back to GEB.
Dr. Hofstadter (his Ph.D. is in physics from the University of Oregon)
calls such a block a "trip-let," a shortened form of "three letters." The
idea came to him, he explains, "in a flash." Intending to write a pam-
phlet about G6del's theorem, his thoughts gradually expanded to in-
clude Bach and Escher until finally he realized that the works of these
men were "only shadows cast in different directions by some central
solid essence." He "tried to reconstruct the central object, and came up
with this book."
660
661
words, and so on. (For this description of trip-lets I am indebted to
Hofstadter's friend Scott Kim, who worked closely with him on many
aspects of the book.)
What reality does Hofstadter see behind the work of his three giants?
One aspect of that reality is the formal structure of mathematics: a
structure that, as Godel's famous undecidability proof shows, has infi-
nitely many levels, none of which are capable of capturing all truth in
one consistent system. Hofstadter puts it crisply: "Provability is a
weaker notion than truth." In any formal system, rich enough to con-
tain arithmetic, true statements can be made that cannot be proved
within the system. To prove them one must jump to a richer system, in
which again true statements can be made that cannot be proved, and so
on. The process goes on forever.
Is the universe Godelian in the sense that there is no end to the dis-
covery of its laws? Perhaps. It may be that no matter how deeply sci-
ence probes there will always be laws uncaptured by the theories, an
endless sequence of wheels within wheels. Hofstadter argues elo-
quently for a kind of Platonism in which science, at any stage of its his-
tory, is like the shadow projections on the wall of Plato's cave. The
ultimate reality is always out of reach. It is the Tao about which noth-
ing can be said. "In a way," Hofstadter writes at the end of his preface,
"this book is a statement of my religion."
For laymen I know of no better explanation than this book presents
of what Godel achieved and of the implications of his revolutionary
discovery. That discovery concerns in particular recursion, self-
reference, and endless regress, and Hofstadter finds those three themes
vividly mirrored in the art of Escher, the most mathematical of graphic
artists, and in the music of Bach, the most mathematical of the great
composers. The book's own structure is as saturated with complex
counterpoint as a Bach composition or James Joyce's Ulysses. The first
half of the book serves as a prelude to the second, just as a Bach pre-
lude introduces a fugue. Moreover, each chapter is preceded by a kind
of prelude, which early in the book takes the form of a "Dialogue" be-
tween Achilles and the Tortoise. Other characters enter later: the Sloth,
the Anteater, the Crab, and finally Alan Turing, Charles Babbage, and
the author himself. Each Dialogue is patterned on a composition by
Bach, and in several instances the mapping is strict. For example, if the
composition has n voices, so does the corresponding Dialogue. If the
composition has a theme that is turned upside down or played back-
662 MISCELLANEOUS
ward, so does the Dialogue. Each Dialogue states in a comic way, with
incredible wordplay (puns, acrostics, acronyms, anagrams, and more),
the themes that will be more soberly explored in the chapter that fol-
lows.
There are two main reasons for Achilles and the Tortoise having been
chosen to lead off the Dialogues. First, they play the major roles in
Zeno's paradox (the topic of the book's first Dialogue), in which
Achilles must catch the Tortoise by escaping from an infinite regress.
Second, they are the speakers in an equally ingenious but less familiar
paradox devised by Lewis Carroll. In Carroll's paradox, which Hofs-
tadter reprints as his second Dialogue, Achilles wishes to prove Z, a
theorem of Euclid's, from premises A and B. The Tortoise, however,
will not accept the theorem until Achilles postulates a rule of inference
C, which explicitly states that Z follows from A and B. Achilles adds
the rule to his proof, thinking the discussion is over. The Tortoise then,
however, jumps to a higher level, demanding another rule of inference
D, which states that Z follows from A, B, and C, and so it goes. The re-
sulting endless regress seems to invalidate all reasoning in much the
same way that Zeno's paradox seems to invalidate all motion. "Plenty
of blank leaves, I see!" exclaims Carroll's Tortoise, glancing at Achilles'
notebook. "We shall need them ALL!" The warrior shudders.
One of Hofstadter's Dialogues, "Contracrostipunctus," is an acrostic
(complete with punctuation marks) asserting that if the words in it are
taken backward, they provide a second-order acrostic spelling out "J. S.
Bach." Another Dialogue, "Crab Canon," which is illustrated with an
Escher periodic tessellation of crabs, is based on Bach's "Crab Canon"
in his Musical Offering. As the Tortoise discusses Bach, his sentences
are interspersed with those of Achilles, who is discussing Escher. The
Tortoise and Achilles use the same sentences in reverse order. The Crab
enters briefly at the crossing point to knot the halves of their discourse
together, halves that interweave in time in the same way that the posi-
tive and negative crabs of Escher's tessellation interweave in space.
The initials A, T, and C (for Achilles, Tortoise, and Crab) correspond
to the initials of adenine, thymine, and cytosine, three of the four nu-
cleotides of DNA, the molecule with the extraordinary ability to repli-
cate itself. Just as Achilles pairs with the Tortoise, so adenine pairs
with thymine along the DNA double helix. Cytosine pairs with gua-
nine. The fact that the initial G can be taken to stand for "gene"
prompted Hofstadter to do a "little surgery on the Crab's speech" so that
Code1, Escher, Bach 663
it would reflect this coincidence. The striking parallel between the
tenets of mathematical logic and the "central dogma" of molecular bi-
ology is dramatized in a chart Hofstadter calls the "Central Dogmap."
The letter G also stands for G6del's sentence: the sentence at the
heart of his proof that asserts its own unprovability. To Hofstadter the
sentence provides an example of what he calls a Strange Loop, exem-
plifying the self-reference that is one of the book's central themes. (A
framework in which a Strange Loop can be realized is called a Tangled
Hierarchy, and the letters of "sloth" tum out to stand for "Strange
Loops, or Tangled Hierarchies. ") Dozens of examples of Strange Loops
are discussed, from Bach's endlessly rising canon (which modulates to
higher and higher keys until it loops back to the original key) to the
looping flow of water in Escher's Waterfall and the looping staircase of
his Ascending and Descending (see Figure 17.5). One of the most amus-
ing models of G is a record player X that self-destructs when a record
titled "I Cannot Be Played on Record Player X" is played on it.
A particularly striking example of a two-step Strange Loop is Es-
cher's drawing of two hands (see Figure 24.1), each one sketching the
other. We who see the picture can escape the paradox by "jumping out
of the system" to view it from a metalevel, just as we can escape the tra-
ditional paradoxes of logic by jumping into a metalanguage. We too,
however, have Strange Loops, because the human mind has the ability
to reflect on itself, that is, the firing of neurons creates thoughts about
neurons. From a broader perspective the human brain is at a level of the
universe where matter has acquired the awesome ability to contem-
plate itself.
By the end of Godel, Escher, Bach Hofstadter has introduced his
readers to modern mathematical logic, non-Euclidean geometries, com-
putability theory, isomorphisms, Henkin sentences (which assert their
own provability), Peano postulates (the pun on "piano" is not over-
looked), Feynman diagrams for particles that travel backward in time,
Fermat's last theorem (with a pun on "fermata"), transfinite numbers,
Goldbach's conjecture (which is cleverly linked with Bach's Goldberg
Variations), Turing machines, computer chess, computer music, com-
puter languages (Terry Winograd, an expert on the computer simulation
of natural language, appears in one Dialogue under the anagrammed
name of Dr. Tony Earrwig), molecular biology, the "mind" of an anthill
called Aunt Hillary, artificial intelligence, consciousness, free will,
664 MISCELLANEOUS
holism v. reductionism, and a kind of sentence philosophers call a
counterfactual.
Counterfactuals are statements based on hypotheses that are con-
trary to fact, for example, "If Lewis Carroll were alive today, he would
greatly enjoy Hofstadter's book." These statements pose difficult prob-
lems in the semantics of science, and there is now a great deal of liter-
ature about them. For Hofstadter they are instances of what he calls
slipping, progressing from an event to something that is almost a copy
of it. The Dialogue that precedes a chapter on counterfactuals and arti-
ficial intelligence concerns a Subjunc-TV set that enables an observer
to get an "instant replay" of any event in a football game and see how
the action would have looked if certain parameters were altered, that
is, if the ball were spherical, if it were raining, if the game were on the
moon, if it were played in four-dimensional space, and so on.
The book's discussion of artificial intelligence is also enormously
stimulating. Does the human brain obey formal rules of logic? Hofs-
tadter sees the brain as a Tangled Hierarchy: a multilevel system with
an intricately interwoven and deep self-referential structure. It follows
logical rules only on its molecular substrate, the "formal, hidden, hard-
ware level" where it operates with eerie silence and efficiency. No com-
puter, he believes, will ever do all a human brain can do until it
somehow reproduces that hardware, but he has little patience with the
celebrated argument of the Anglican philosopher J. R. Lucas that
Godel's work proves a human brain can think in ways that are in prin-
ciple impossible for a computer.
Only a glimpse can be given here of the recreational aspects of this
monstrously complicated book. In "The Magnificrab, Indeed" (a pun on
Bach's Magnificat in D), the Dialogue that introduces a discussion of
deep theorems of Alonzo Church, Turing, Alfred Tarski, and others,
appears a whimsical Indian mathematician named Mr. Najunamar. Na-
junamar has proved three theorems: he can color a map of India with
no fewer than 1,729 colors; he knows that every even prime is the sum
of two odd numbers, and he has established that there is no solution to
an + fI! = en when n is zero. All three are indeed true.
Some readers will recognize 1,729 as the number of the taxi in which
G. H. Hardy rode to visit the Indian mathematician Srinivasa Ramanu-
jan ("Najunamar" spelled backward) in a British hospital. Hardy re-
marked to Ramanujan that 1,729 was a rather dull number. Ramanujan
Code], Escher, Bach 665
rejoined instantly that on the contrary it was the smallest positive in-
teger that is the sum of two different pairs of cubes. Hardy then asked
his friend if he knew the smallest such number for fourth powers. Ra-
manujan did not know the number, although he guessed that it would
turn out to be fairly big. Hofstadter supplies the answer: 635,318,657,
or 134
4
+ 133
4
or 158
4
+ 59
4
. He also wonders if his readers can find the
smallest number that can be expressed as the sum of two squares in two
different ways, but he hides the answer. Can you determine it before I
supply it in the answer section?
To explain the meaning ofthe term Hformal system" Hofstadter opens
his book with a simple example that uses only the symbols M, I, and U.
These symbols can be arranged in strings called theorems according to
the following rules:
1.1f the lastletter of a theorem is I, U can be added to the theorem.
2. To any theorem Mx, x can be added. (For example, MUM can be
transformed into MUMUM, and MU can be transformed into MUU.)
3. If III is in a theorem, it can be replaced by U, but the converse oper-
ation is not acceptable. (For example, MIll can be transformed into
MU, and UMIIIMU can be transformed into UMUMU.)
4. If UU is in a theorem, it can be dropped. (For example, UUU can be
transformed into U, and MUUUIII can be transformed into MUIII.)
There is only one Haxiom" in the system: In forming theorems one
must begin with MI. Every string that can be made by applying the
rules, in any order, is a theorem of the system. Thus MUIIU is a theo-
rem because it can be generated from MI in six steps. If you play with
the M, I, and U system, constructing theorems at random, you will soon
discover that all theorems begin with M and that M can occur nowhere
else.
Now for a puzzle: Is MU a theorem? I shall say no more about MU
here except that it plays many other roles in the book, in particular
serving as the first two letters of "Mumon," the name of a Zen monk
who appears in a delightful chapter on Zen koans.
Even as simple a system as that of M, I, and U enables Hofstadter to
introduce a profound question. If from all the possible strings in the
system we subtract all the strings that are theorems, we are left with all
the strings that are not theorems. Hence the "figure" (the set of theo-
rems) and the "ground" between the theorems (the set of nontheorems)
666 MISCELLANEOUS
seem to carry equivalent information. Do they really? Is the system like
an Escher tessellation in which the spaces between animals of one kind
are animals of another kind, so that reproducing the shapes of either set
automatically defines the other? (Or so that a black zebra with white
stripes is the same as a white zebra with black stripes?) In this con-
nection Hofstadter reproduces a remarkable tessellation by Kim in
which the word "FIGURE" is periodically repeated in black so that the
white ground between the black letters forms the same shapes (see Fig-
ure 49.2). The same concept is playfully illustrated in the Dialogue
"Sonata for Unaccompanied Achilles" (modeled on Bach's sonatas for
unaccompanied violin), in which we hear only Achilles' end of a tele-
From Inversions (W. H. Freeman & Co., 1989)
Figure 49.2. Scott Kim's FIGURE FIGURE figure
Codel, Escher, Bach 667
phone conversation with the Tortoise about "figure" and "ground."
From Achilles' half of the conversation we can reconstruct the Tor-
toise's.
Other figure-ground examples are provided by the counting num-
bers. For example, given all the primes, we can determine all the non-
primes simply by removing the primes from the set of positive integers.
Is the same true of all formal systems? Can we always take all the the-
orems from the set of all possible statements in the system and find that
what is left-the set of nontheorems-is another complementary formal
system? An unexpected discovery of modern set theory is that this is
not always the case. To put it more technically, there are recursively
enumerable sets that are not recursive. Thus does Hofstadter lead his
readers from trivial beginnings into some of the deepest areas of mod-
ern mathematics.
The book closes with the wild Dialogue "Six-Part Ricercar," which is
simultaneously patterned after Bach's six-part ricercar and the story of
how Bach came to write his Musical Offering. (A ricercar is a compli-
cated kind of fugue.) In this Dialogue the computer pioneers Turing and
Babbage improvise at the keyboard of a flexible computer called a
"smart-stupid," which can be as smart or as stupid as the programmer
wants. (The computer's name is a play on "pianoforte," which means
"soft-loud.") Turing produces on his computer screen a simulation of
Babbage. Babbage, however, is seen looking at the screen of his own
smart-stupid, on which he has conjured up a simulation of Turing.
Each man insists he is real and the other is no more than a program. An
effort is made to resolve the debate by playing the Turing Game, which
was proposed by Turing as a way to distinguish a human being from a
computer program by asking shrewd questions. The conversation in
this scene parodies the conversation Turing gives in his classic paper
on the topic.
At this point Hofstadter himself walks into the scene and convinces
Turing, Babbage, and all the others that they are creatures of his own
imagination. He, however, is as unreal as any of the other characters of
the Dialogue, because he too is imagined by the author. The situation
resembles a painting by Rene Magritte titled The Two Mysteries, in
which a small picture of a tobacco pipe is displayed with a caption that
says (to translate from the French) "This is not a pipe." (see Figure
49.3.) Floating above the fake pipe is a presumably genuine larger pipe,
but of course it too is painted on the canvas.
668 MISCELLANEOUS
2001 C. Herscovici, Brussels! Artists New York
669
1979 by Acidic Books (Hofstadter's publisher being Basic Books). Here
is Hofstadter's comment: "A formidable hodgepodge, turgid, and con-
fused-yet remarkably similar to the present work. Contains some ex-
cellent examples of indirect self-reference. Of particular interest is a
reference in its well-annotated bibliography to an isomorphic, but
imaginary, book."
Addendum
Hofstadter's GEB took off like a rocket, staying long on the best
seller lists, and winning the 1980 Pulitzer Prize for general nonfiction.
Vintage Books paid $200,000 for paperback rights-the largest sum it
had ever paid for nonfiction rights, and the largest sum Basic Books
ever received for such a work.
Reviews in 1979 were lavish in their praise. Especially noteworthy
were reviews by Brian Hayes (The New York Times Book Review, April
29), Walter Kerrick (Village Voice, November 19), and Edward Roth-
stein (New York Review of Books, December 6). Other reviews ran in
Commonweal, Technology Review, Psychology Today, American Sci-
entist, Yale Review, American Scholar, and New Republic.
In an amusing review in the Journal of Recreational Mathematics
(14, 1981-82, pp. 52-54), Leon Bankoff observed that GEB has exactly
777 pages, and by using the cipher A = 1, B = 2, C = 3, and so on, one
discovers that G = E + B.
Hofstadter became my successor in writing the Mathematical Games
column in Scientific American, after he changed the department's name
to Metamagical Themas, an anagram of its former title. His columns
were reprinted in Metamagical Themas: Questing For the Essence of
Mind and Pattern (Basic Books, 1985), a work of 852 pages. A few years
earlier, Hofstadter and Daniel C. Dennett had edited a marvelous an-
thology, The Mind's I: Fantasies and Reflections on Self and Soul (Basic
Books, 1981). At present Hofstadter is professor of cognitive science
and computer science and technology at the Indiana University in
Bloomington.
I must confess that I could never have written my review of GEB had
I not had on hand a 33-page analysis of the book written by Scott Kim
titled Strange Loop Gazette. Kim has since obtained his doctorate under
Donald Knuth, in the computer science department of Stanford Uni-
versity. Kim's beautiful book Inversions, containing scores of names
670 MISCELLANEOUS
and phrases drawn in such a way that they magically remain the same
when inverted or mirror-reflected (or turn into another word or phrase)
has been reissued by W. H. Freeman. (A book of similar inversions by
Hofstadter, titled Ambigrammi, was published in Italy in 1987 but has
yet to have a U.S. edition.) Kim is now working on a book about how
to use one's fingers to model such things as the skeleton of a cube or
tetrahedron, or to entwine the fingers to produce such topological
structures as a trefoil knot.
Answers
I showed how MUIIU could be generated from MI in six steps.
Several readers lowered this to five, and one reader, Raymond Aaron,
did it in four: MIto MIl (rule 2), to MIlIl (rule 2), to MIIlIlIlI (rule 2), and
finally to MUIIU (rule 3).
I did not give a proof that MU is not a theorem. Here is how Hofs-
tadter handled it. Every theorem begins with M, which occurs nowhere
else. The number of 1's in a theorem is not a multiple of 3 because this
is true for the axiom MI, and every permissible operation preserves
this property. Therefore MU, whose number (zero) ofI's is a multiple of
3, cannot be obtained by the permissible operations.
Several readers wrote programs for determining Ramanujan num-
bers that solve the Diophantine equation An + Bn = en + lY'. When n is
3, William J. Butler, Jr.'s program found 4,724 solutions for values less
than 1010, of which the largest is
1,956
3
+ 1,360
3
= 2,088
3
+ 964
3

Of the 4,724 solutions, 26 are triples, the smallest being
414
3
+ 255
3
= 423
3
+ 228
3
= 436
3
+ 167
3

The number of primitive solutions (no common factor of the four
numbers) is infinite, but the number of triples, Butler conjectures,
could be finite because their density declines rapidly as the numbers
grow In SIze.
Hofstadter disclosed in a letter that the phrase "formidable hodge-
podge" in his joke review of GEB (quoted in my final paragraph) was
taken from a reviewer's comment when Indiana University considered
publishing the book. (The book was also rejected, incidentally, by an
editor then at W. H. Freeman.) Because W. V. Quine, the Harvard
Codel, Escher, Bach 671
philosopher, was one of the two reviewers, the chances are fifty percent
that the phrase was Quine's. "Turgid and confused" (the phrase also ap-
pears on p. 3 of GEB) is from a comment on Bach's style by one of his
pupils. In GEB's second printing Hofstadter added the following to his
hoax review: "Professor Gebstadter's Shandean digressions include
some excellent examples of indirect self-reference." The first four
words, with the change of name, are from Brian Hayes's The New York
Times review of GEB.
Answers
The first problem was to find the smallest positive integer that
can be expressed as the sum of two squares in two different ways. The
number is 50, which equals 52 + 52 or 12 + 7
2
. If zero squares are al-
lowed, however, the number is 25, which equals 52 + 0
2
or 3
2
+ 4
2
. If the
two squares must be nonzero and different, the solution is 65, which
equals 8
2
+ 12 or 7
2
+ 4
2

The second problem was to determine whether or not MU is a theo-
rem in the M, I, and U formal system. A simple proof of why MU is not
a theorem can be found on pages 260 and 261 of COdel, Escher, Bach:
an Eternal Golden Braid.
Bibliography
J. Gleick, "Exploring the Labyrinth of the Human Mind," The New York Times
Magazine, August 21,1983. A cover story on Douglas Hofstadter.
G. Johnson, "The Copycat Project," Machinery of the Mind: Inside the New Science
of Artificial Intelligence, Times Books, 1986, Chapter 15. The chapter is about
Hofstadter's approach to artificial intelligence.
Reviews of qodel, Escher, Bach
H. Gardner, "Strange Loops of the Mind," Psychology Today; March 1980, pp.
72-86.
B. Hayes, "This Head is False," The New York Times Book Review, April 29, 1978,
pp. 13, 18.
E. Rothstein, "The Dream of Mind and Machines," New York Review of Books, De-
cember 6, 1979, pp. 34-39.
Other books by Douglas Hofstadter
The Mind's I: Fantasies and Reflections on Self and Soul. With Daniel C. Dennett,
Basic, 1981.
672 MISCELLANEOUS
Metamagical Themas, Basic, 1985. A collection of Hofstadter's Scientific American
columns.
Ambigrams, Italy: Hopeful Monster, 1987.
Fluid Concepts and Creative Analogies: Computer Models of the Fundamental
Mechanisms of Thought, Basic, 1995.
Le Ton Beau de Marot: The Spark and Sparkle of Creative Translation, Basic, 1997.
Codel, Escher, Bach 673
Chapter 50 Six Sensational
Discoveries
As a public service, I shall comment briefly on six major dis-
coveries of 1974 that for one reason or another were inadequately re-
ported to both the scientific community and the public at large. The
most sensational of that year's discoveries in pure mathematics was
surely the finding of a counterexample to the notorious four-color-map
conjecture. That theorem is that four colors are both necessary and suf-
ficient for coloring all planar maps so that no two regions with a com-
mon boundary are the same color. It is easy to construct maps that
require only four colors, and topologists long ago proved that five col-
ors are enough to color any map. Closing the gap, however, had eluded
the greatest minds in mathematics. Most mathematicians have believed
that the four-color theorem is true and that eventually it would be es-
tablished. A few suggested it might be Godel-undecidable. H.S.M. Cox-
eter, a geometer at the University of Toronto, stood almost alone in
believing that the conjecture is false.
Coxeter's insight was vindicated. In November 1974 William Mc-
Gregor, a graph theorist of Wappingers Falls, NY., constructed a map of
110 regions that cannot be colored with fewer than five colors (see Fig-
ure 50.1). McGregor's technical report appeared in 1978 in the Journal
of Combinatorial Theory, Series B.
In number theory the most exciting discovery of 1974 was that when
the transcendental number e is raised to the power of 1t times v'i63, the
result is an integer. The Indian mathematician Srinivasa Ramanujan
had conjectured that e to the power of 1tv'i63 is integral in a note in
the Quarterly Journal of Pure and Applied Mathematics (vol. 45, 1913-
1914, p. 350). Working by hand, he found the value to be
262,537,412,640,768,743.999,999,999,999, .... The calculations were
tedious, and he was unable to verify the next decimal digit. Modern
674
~
I
I
I
I
I
I
I
IT
I
I
I
Figure 50.1. The four-color-map theorem is exploded.
computers extended the 9's much farther; indeed, a French program of
1972 went as far as two million 9's. Unfortunately, no one was able to
prove that the sequence of 9's continues forever (which, of course,
would make the number integral) or whether the number is irrational
or an integral fraction.
In May 1974 John Brillo of the University of Arizona found an inge-
nious way of applying Euler's constant to the calculation and managed
to prove that the number exactly equals 262,537,412,640,768,744. How
the prime number 163 manages to convert the expression to an integer
is not yet fully understood.
There were rumors late in 1974 that 1t would soon be calculated to six
million decimal places. This may seem impressive to laymen, but it is
a mere computer hiccup compared with the achievement of a special-
purpose chess-playing computer built in 1973 by the Artificial Intelli-
gence Laboratory at the Massachusetts Institute of Technology. Richard
Pinkleaf, who designed the computer with the help of ex-world-chess-
Six Sensational Discoveries 675
champion Mikhail Botvinnik of the former U.S.S.R., calls his machine
MacHic because it so often plays as if it were intoxicated.
Unlike most chess-playing programs, MacHic is a learning machine
that profits from mistakes, keeping a record of all games in its memory
and thus steadily improving. Early in 1974 Pinkleaf started MacHic
playing against itself, taking both sides and completing a game on an
average of every 1.5 seconds. The machine ran steadily for about seven
months.
At the end of the run, MacHic announced an extraordinary result. It
had established, with a high degree of probability, that pawn to king's
rook 4 is a win for White. This was quite unexpected because such an
opening move has traditionally been regarded as poor. MacHic could
not, of course, make an exhaustive analysis of all possible replies. In
constructing a "game tree" for the opening, however, MacHic extended
every branch of the tree to a position that any chess master would un-
hesitatingly judge to be so hopeless for Black that Black should at once
resign.
Pinkleaf has been under enormous pressure from world chess lead-
ers to destroy MacHic and suppress all records of its analysis. The Rus-
sians are particularly concerned. I am told by one reliable source that
a meeting between Kissinger and Brezhnev will take place in June, at
which the impact on world chess of MacHic's discovery will be dis-
cussed.
Bobby Fischer reportedly said that he had developed an impregnable
defense against P-KR4 at the age of 11. He has offered to play it against
MacHic, provided that arrangements can be made for the computer to
play silently and provided that he (Fischer) is guaranteed a win-or-Iose
payment of $25 million.
The reaction of chess grand masters to MacHic's discovery was mild
compared with the shock waves generated among leading physicists by
the discovery that the special theory of relativity contains a logical
flaw. The crucial "thought experiment" is easily described. Imagine a
meterstick traveling through space like a rocket, on a straight line col-
inear with the stick. A plate with a circular hole one meter in diameter
is parallel to the stick's path and moving perpendicularly to it (see Fig-
ure 50.2), We idealize the experiment by assuming that both the plate
and the meterstick have zero thickness. The two objects are on a pre-
cise collision course. At the same instant, the center of the meter stick
and the center of the hole will coincide.
676 MISCELLANEOUS
IIII
677
courtesy N.Y. Public
50.3. Lost
678
in channels inside walls, a ventilating shaft to the roof, and suspended
weights to make sure the entrance door closed. Until now, however, the
first valve flush toilet has always been credited to Sir John Harington,
a godson of Queen Elizabeth. Harington described it amusingly in his
book The Metamorphosis of Ajax (1596) a cloacal satire that got him
banished from the court. Although his "Ajax" actually was built at Kel-
ston near Bath, it was not until 200 years later that it came into general
use.
The first English patent for a valve flush toilet was granted in 1775
to Alexander Cummings, a watchmaker. Modern mechanisms, in which
a ball float and automatic cutoff stopper limit the amount of water re-
leased with each flush, date from the early 19th-century patents of
Thomas Crap per, a British manufacturer of plumbing fixtures who died
in 1910. (See L. Wright, R. Paul, and K. Paul, Clean and Decent: The
Fascinating History of the Bathroom and Water Closet, 1960 and W.
Reyburn, Flushed with Pride: The Story of Thomas Crapper, Prentice-
Hall, 1971.)
Although hundreds of books on parapsychology spewed forth from
reputable publishing houses in 1974, not one reported the most sensa-
tional psi discovery of the century: a simple motor that runs on psi en-
ergy. It was constructed in 1973 by Robert Ripoff, the noted Prague
parapsychologist and founder of the International Institute for the In-
vestigation of Mammalian Auras. When Henrietta Birdbrain, an Amer-
ican expert on Kirlian photography, visited Prague in early 1974, Dr.
Ripoff taught her how to make his psychic motor. Ms. Birdbrain
demonstrated the device many times in her lectures, but as far as I am
aware the only published report on it appeared in the Boston monthly
newspaper East West Journal (May 1974, p. 21).
Readers are urged to construct and test a model of the motor. The first
step is to cut a three-by-seven-inch rectangle from a good grade of bond
paper. Make a tiny slot in the paper at the spot shown (see Figure 50.4).
The slot must be three-eighths inch long and exactly in the center of the
strip, one-eighth inch from the top edge. Bend the paper into a cylin-
der, overlapping the ends five-sixteenths inch, and glue the ends to-
gether. Cut a second slot in the center of the overlap, directly opposite
the preceding one. It must be the same size and the same distance from
the top.
From a file card or a piece of pasteboard of similar weight, cut a strip
three-eighths inch by three inches. Insert a fine, sharp-pointed needle
Six Sensational Discoveries 679
STEP 1
1/8 INCH -'i 3/8 INCH
3 INCHES
7 INCHES
STEP 2
STEP 3
3/8 INCH
3 INCHES
5/16 INCH OVERLAP
STEP 4
STEP 5
motor is
50.5. How to motor.
The motor underwent extensive investigation at numerous parapsy-
chology laboratories around the world. Russian experts were convinced
the energy that turns the motor is the same as the psychokinetic energy
that enabled the Israeli psychic Uri Geller to bend silverware, the Rus-
sian "sensitive" Ninel Kulagina to levitate table-tennis balls, and the
Brooklyn psychic Dean Kraft to make pieces of candy leap out of bowls
and pens crawl across rugs. When Kulagina held both hands near the
motor, the cylinder flew straight up in the air for several meters. A
book on the Ripoff rotor (as it is called in Prague), with papers by 12 of
the world's leading parapsychologists, was edited by Ms. Birdbrain.
James Randi, the magician, contends that by using trickery, he can
make the motor spin rapidly in either direction. Of course, that does
not explain why the motor operates so efficiently for thousands ofpeo-
pIe who know nothing about conjuring.
Addendum
The foregoing chapter, when it ran in the April 1975 issue of
Scientific American, was intended as an April Fools' joke. It was so
crammed with preposterous ideas and outlandish names that I never
dreamed anyone would take it seriously, yet it produced more than a
thousand letters from readers who did not recognize the column as a
hoax.
The map was designed by correspondent William McGregor (his real
name), who gave me permission to print it. Hundreds of readers sent
me copies of the map, colored with four colors. Some said they had
worked on it for days before they found a way to do it. The four-color
map theorem is no longer a conjecture. It was proved in 1976 by Wolf-
gang Haken and Kenneth Appel, with the aid of a long-running com-
puter program. (See "The Solution of the Four-Color-Map Problem,"
Scientific American, October 1977, pp. 108-21.) Whether a simple, el-
egant proof not requiring a computer will ever be found, is still an open
question.
When Norman K. Roth published an article, "Map Coloring," in
Mathematics Teacher (December 1975), many readers informed him
that Scientific American had published a map disproving the four-color
theorem. A letter from Roth in the following May issue pointed out
that my column was "an apparently successful April Fools' article."
In 1977 the Vancouver Sun reported a British mathematician's claim
682 MISCELLANEOUS
(it turned out to be invalid) that he had proved the four-color map the-
orem. On January 17, 1977, the newspaper ran a letter from a lady in
Port Moody, which said in part:
To set the record straight, I would like to bring to your notice the fact that
the theorem has already been disproved by William McGregor, a graph
theorist ... in November 1975. He constructed a map ofl10 regions that
cannot be colored with less than five colors ....
Artificial Intelligence (Vol. 10, 1978, p. 116) reported that a com-
puter program had managed to color McGregor's map with four colors,
with only two backtracks in the tree search. Apparently the author did
not realize the map was a joke.
The following letter, signed by "Ivan GuffvanofflII," who claimed to
be a mathematician at the University of Wisconsin, was a bit frighten-
ing to the staff of Scientific American, until they realized that it, too,
was a joke:
This is to inform you that my lawyer will soon be contacting you for a
damage case of $25 million.
In the mathematics section of your April 1975 issue, Martin Gardner
wrote that the four-color problem had been solved. I have been working
on this problem for 25 years. I had prepared a paper to be submitted to
the American Mathematical Monthly. The paper was over 300 pages in
length. In it I had proved that the answer to the four-color problem was
no and that it would take five colors instead of four. Upon reading Gard-
ner's article that someone else would publish the solution before I could,
I destroyed my paper. Last week I read in Time magazine that Gardner's
article was a farce. I did not read Gardner's entire article, only the part
on the four-color problem, so I was not aware of the farce. Now that I
have destroyed my article, it will not be possible to reproduce all 300
pages, since the work has extended over such a long time. I therefore be-
lieve that damages are due me.
I believe that Gardner's article was the most unprofessional article I
have ever seen in your's or any other journal. This kind of activity is
below the dignity of what I thought your magazine stood for. I am not
only suing you but I am cancelling my membership, and I will ask all my
friends to cancel theirs.
In Italy the noted mathematician Beniamino Segre published a seri-
ous research note (Rendiconti 59, 1975, pp. 411-12) in which he re-
produced McGregor's map, showing how it could be four-colored. "It is
Six Sensational Discoveries 683
shown the falsehood," his summary reads, "of a presumed counterex-
ample for the four-color conjecture."
Manifold, a journal published by mathematics students at the Uni-
versity of Warwick, ran the following lines in its Autumn 1975 issue.
They are to be sung to the tune of "Oh Mr. Porter, what shall I do?"
it Oh Mr. Gardner,
What have you done?
You've started up a rumour
You should never have begun!
A four-colour hoax can't
Be undone so quick . ..
Oh Mr. Gardner, what
A bloody silly trick!"
When e is raised to the power of the product of 1t and the square
root of 163, the result is the 18-digit number I gave, minus
.000,000,000,000,75 .... John Brillo, to whom I attributed this hoax, is
a play on the name of the distinguished number theorist John Brillhart.
The reference to Ramanujan's paper is legitimate. In it the Indian math-
ematician discusses a family of remarkable near-integer numbers to
which this one belongs, but of course he knew that none were integral.
Indeed, as many readers pointed out, it is not hard to prove that they
are transcendental.
The value of the number to 39 significant decimal digits was given
by D. H. Lehmer in Mathematical Tables and Aids to Computation
(Vol. 1, January 1943, pp. 30-31). The digit following the run of 9's is
2. See also "What Is the Most Amazing Approximate Integer in the
Universe?" by I. J. Good, in the Pi Mu Epsilon Journal (Vol. 5, Fall 1972,
pp. 314-15).
The description of Richard Pinkleaf's chess-playing program,
MacHic, is a play on the chess program MacHack, written by Richard
Greenblatt of the Massachusetts Institute of Technology. The relativity
paradox that I hung on Humbert Pringle (a play on the name of Herbert
Dingle, a British physicist who maintained that relativity theory is dis-
proved by the famous twin paradox) is well known. It appears as a
problem on page 99 of the paperback edition of Spacetime Physics, by
Edwin F. Taylor and John A. Wheeler (W. H. Freeman and Company,
1966), and the solution is given on page 25 of the answer section. The
paradox is discussed at greater length by George Gamow in Mr. Tomp-
684 MISCELLANEOUS
kins in Wonderland (Macmillan, 1947), W. Rindler in American Jour-
nal of Physics, (Vol. 29, 1961, p. 365 ff.), R. Shaw (ibid., Vol. 30, 1962,
p. 72 ff.), and P. T. Landsberg in The Mathematical Gazette, (Vol. 47,
1964, p. 197 ff).
A stationary outside observer will see the meterstick just make it
through the hole. If the plate and the stick have thickness, the stick
must, of course, be a trifle shorter than the hole to prevent an end from
catching. To an observer on the plate the stick will appear Lorentz-
contracted, but it will also appear rotated, so that it seems to approach
the hole on a slant. The stick's back end actually seems to go through
the hole before its front end, so that it gets through with the same clear-
ance as before. To an observer on the stick the plate will appear
Lorentz-contracted, its hole becoming elliptical, but the plate also ap-
pears to be rotated. In this case the hole first goes over the front end of
the slanted stick, again with the same close fit. "Contractions" and "ro-
tations" are ways of speaking in a Euclidean language. In a four-
dimensional, non-Euclidean language of space-time the objects retain
their shapes and orientations. Having at one time written a book on rel-
ativity, I was abashed to receive more than 100 letters from physicists
pointing out the stupid "blunder" I had made.
Those who enjoyed explaining the paradox may wish to consider
how to escape from the following variant. Assume that the meterstick
is sliding at high speed along the surface of an enormous flat plate of
metal toward a hole slightly larger than the stick. We idealize the
thought experiment by assuming that there is no friction and that the
stick and the plate are extremely thin. When the stick is over the hole,
gravity (or some other force) pulls it down and through. For an ob-
server on the stick the sheet slides under it, and the hole is Lorentz-
contracted enough to prevent the stick from dropping through. In this
case the stick and the plate cannot rotate relative to each other. How
does the stick get through? (Please, no letters! I know the answer.)
The Leonardo da Vinci drawing was done by Anthony Ravielli, a
graphic artist well known for his superb illustrations in books on
sports, science, and mathematics. It was an earlier version of the sketch
that suggested to me the idea of a hoax column. Many years ago a friend
of Ravielli's had jokingly made a bet with a writer that Leonardo had in-
vented the first valve flush toilet. The friend persuaded Ravielli to do
a Leonardo drawing in brown ink on faded paper. It was smuggled into
the New York Public Library, stamped with a catalogue file number,
Six Sensational Discoveries 685
and placed in an official library envelope. Confronted with this evi-
dence, the writer paid off the bet.
Augusto Macaroni is a play on Augusto Marinoni, a da Vinci spe-
cialist at the Catholic University of Milan, and Ram6n Paz y Bicuspid
is a play on Ram6n Paz y Remolar, the man who actually found the two
missing da Vinci notebooks. My data on the history of the water closet
are accurate, including the reference to Thomas Crapper. The book by
Wallace Reyburn Flushed with Pride: The Story of Thomas Crapper
does exist.
For many years I assumed that Reyburn's book was the funniest
plumbing hoax since H. L. Mencken wrote his fake history of the bath-
tub. I thought this for two reasons: (1) The book implies that the slang
words "crap" and "crapper" derived from Mr. Crapper's name, but
"crap" and "crapping case" are both listed in The Slang Dictionary,
published in London in 1873. (2) Reyburn wrote a later book titled
Bust-up: The Uplifting Tale of Otto Titzling and the Development of the
Bra. It turns out, though, that both Thomas Crapper and Otto Titzling
were real people, and neither of Reyburn's books is entirely a hoax.
The Ripoff Rotor is a modification of a psychic motor described in
Hugo Gernsback's lurid magazine Science and Invention (November
1923, p. 651). Prizes were awarded in March 1924 to readers who gave
the best explanations of why the cylinder turned. The motion can be
caused by any of three forces: slight air currents in the room, convec-
tion currents produced by heat from the hand, and currents from
breathing. The three forces combine in unpredictable ways. If a person
who believes he or she has psychokinetic powers is willing the motor
to turn, it may turn in the direction willed, or it may go the other way.
Nandor Fodor, in his Encyclopedia of Psychic Science (Citadel,
1966), under the heading "Fluid Motor," credits the paper-cylinder de-
vice to one Count de Thome lin , but he doesn't say who the Count was
or when he invented the motor.
I had no individual in mind when I mentioned Ms. Henrietta Bird-
brain, but there is an East West newspaper in Boston, and the reader
who bothers to check the issue cited, will find a sober report by Stan-
ley Krippner on a psychic motor that was demonstrated to him in
Prague by Robert Pavalita. (On Pavalita, see S. Ostrander and L.
Schroeder, Psychic Discoveries behind the Iron Curtain, Prentice-Hall,
1970, Chapter 28.)
Hereward Carrington, in The Story of Psychic Science (1931, p. 138),
686 MISCELLANEOUS
mentions half a dozen early French psychic motors. Opposite page 138
is a photograph of Paul Joire's "sthenometer," a straw balanced on a
needle and under a bell jar. Incidently, Dr. Joseph B. Rhine for years
tried to find evidence that a person could turn an arrow delicately bal-
anced on a needle in a vacuum, but without success. His negative re-
sults were never published. Surely one of the great scandals of
parapsychology is its claim that the mind can affect heavy objects like
falling dice-and if Uri Geller is taken seriously, can bend spoons and
keys-but as yet no one has been able, under controlled conditions, to
rotate a featherweight pointer.
After my hoax column appeared, several psychic motors went on
sale. The most colorful was a cardboard device called the Mind Ma-
chine, sold by Unicorn Products in Santa Fe. "As with the Ouiji Board,"
the instructions said, "results are not always immediate, and often vary
from time to time .... Yet for countless people the phenomenon is a re-
ality. At parties, skeptics have been known to chop the base apart, ex-
pecting to find a hidden motor. How does the Mind Machine work?
Theories abound-but no one really knows." A similar cardboard de-
vice called a "psionic generator," marketed by Monarch Manufacturing
Company, in Roseville, MI, was advertised in Fate (July 1975, p. 91).
Professor L. E. van der Tweel, at the University of Amsterdam, sent
the photograph reproduced in Figure 50.6. It shows the noted neuro-
physiologist and philosopher Warren McCulloch demonstrating a
paper psychic motor to several colleagues. The picture was taken in
1953.
Some readers who believed in psychokinesis took the Ripoff Rotor
seriously, but most readers did not. Mark J. Hagmann reported his dis-
covery that the motor's rotation relative to the room was an inverse
function of the contents of the liquor bottle he used to support the nee-
dle. The rotation's speed increased as the level of booze went down.
I've selected the following four letters as the funniest:
Dear Mr. Gardner:
As a citizen deeply concerned about the energy crisis and full public
disclosure, I express my gratitude for your release of the psi-engine data.
I took drawings of a larger model, coupled to an electric generator, to
Ralph Nadir and obtained his tentative endorsement that, used as a
power generator, the Ripoff system would have a minimum environ-
mental impact.
Six Sensational Discoveries 687
50.6. Warren J.yJ..\..'\....JI...L.U.\../v.lJ. ClelTIo:nstratrng motor
average be used which is room temperature plus 100. To operate with
maximum power output, the Ripoff machine must "feel" this tempera-
ture. This is accomplished by filling the bottle with water heated to 100
above room temperature-but in no case above boiling.
The machine, thus adjusted and filled, will work well with only a
watchful look from the experimenter. He does not need to use his hands.
Another finding of great importance was his measurement of aura
power from various types of printed material. Using a torsion balance
wire tipped with a psi-magnetically oriented crystal of mezzanine sul-
phate, he found that purely scientific material was more potent than the
religious material used by Ripoff. Four volumes of Tom Swift or the
equivalent weight in AAAS Science magazines will do nicely. Greater
power levels can be sustained if these are also warmed to the psi-
temperature.
Finally, in a personal (telepathic) communication with Dr. Ripoff, Dr.
Feckless learned that the northern orientation favored by Dr. Ripoff was
empirically determined. Analysis showed that one of Europe's greatest
intellectual centers was due north of the Ripoff lab. Considering a great
focus of intellectual power to be the critical variable, the most favorable
orientation for U.S. experiments would therefore be toward Washington
D.C.
I hope these findings will be helpful and will move us closer to the era
of psi-power.
Dear Mr. Gardner:
Sincerely yours,
George N. Chatham
Your psi energy motor. I've got it! I know how it works. It really does
work! And I know how it works. Even Eureka!
But first let me tell you, in a scholarly fashion, how I came to this un-
expected revelation of the psi energy motor's operational mechanism
through your lucid instructions in the April Scientific American. Let
me also reassure you that, although I know several people who teach
psychology, I know very little of the subject myself, other than infre-
quent casual reading about their feeling bumps on other people's heads
to tell fortunes. The psychologists I do know, during our Friday-
afternoon beer drinking sessions, do not talk of parapsychology; here in
Seattle we call parasols umbrellas. The rain, you know. As for my knowl-
edge of energy sources, I know almost nothing of physics, aside from an
occasional Ex-Lax, as needed.
The experiment, though. I was going to tell you about the experiment.
I came home on a wet Friday afternoon to find my wife out on some do-
Six Sensational Discoveries 689
gooding mission so I picked up the April issue of Scientific American
and chanced on your startling article. I immediately went to work build-
ing the motor you described. I used the recommended good grade of
bond paper. It was stationery from the college where I teach political sci-
ence. I cut the three by seven inch piece from the center, lest the letter-
head portion might influence the psychic energy. The cardboard portion
of the assembly was easily cut from a student grade-change request card
on my desk. Finding the needle was more difficult, since my wife was
out and I had to hunt down where she hides her sewing equipment. I
found it. I found the needles, too, scattered in the bottom of her thread-
filled case. I can report that the motor runs with a slight smear of blood
on the paper and a band-aid on the finger.
With the psychic energy motor assembled on my desk, I set out in
search of the rest of the equipment necessary to make it operate. There
was no hard, smooth, plastic bottle in the house. A glass catsup bottle
could not function because of its metal cap. A water glass was too great
in diameter. At that point I spied a half-full scotch bottle in the bar. It had
a long enough neck to allow the motor to hang perfectly. But it was half
full. I poured that into the too-big water glass. I do not have a copy of I
Ching around the house, and I found the family Bible had so many pho-
tographs, pressed flowers, decades-old dance programs, baby footprints,
and grade-school report-cards that the family Bible could not have pro-
vided a flat surface on which to set the bottle.
North and south, for the alignment of the book, was not difficult to as-
certain because the house faces west and the desk in the study is along
that wall. But a book. I first tried the required text in an introductory
American government course. The results were inconclusive; nothing
but turbulent action and indecision from the psi motor. I then tried a
tract from the John Birch Society, and the motor rotated to the right.
Wondering if that had any political significance, I tried an anthology of
the works of Lenin, and sure enough, it turned to the left. I then placed
the motor assembly on Merkin Swiver's Collected Speeches of President
Ford, and noted that the motor revolved in a quite mediocre fashion,
again to the right. It did not turn at all with a football helmet over it.
Barry Goldwater's Conscience of a Conservative made the motor spin so
rapidly I could feel the draft. When placed on N. Comium's biography
of Henry M. "Stoop" Jackson, the motor rose slowly accelerating mar-
velously toward the north on what I assumed was a great-circle course.
I then tried a copy of M. P. Chment's long study of ex-President Nixon.
The motor kept falling off the bottle to the South, toward California, and
kept rolling into hiding under the desk. At last, I was beginning to gain
insight into the motive power of the psychic energy motor. But by now
690 MISCELLANEOUS
the glass into which I had poured the scotch was empty, and I had to go
open a new bottle.
On my return to the study, I placed the refilled glass on the desk and
I pried the phlebitic motor out from under the desk. Then, finding a
monograph on the functioning of Congress, I placed the bottle and the
psi motor on that legislative tome. You will not believe it. I did only after
I had closed one eye to focus more carefully. The motor was turning
both directions-right and left-at the same time and was doing so in a
most indecisive fashion.
At that instant, in a flash of intuition, I learned the secret of the mo-
tive power of the psychic motor: hot air.
Soon thereafter, however, other psychokinetic manifestations oc-
curred. The floor rose up rapidly, and with a thump, hit the side of the
chair in which I was sitting. From that position, most curiously, the
books on their shelves did not fall from their obviously unaccustomed
position on the ceiling, even though the room, too, was rotating.
Some time later my wife returned. I attempted to show her how the
psychic energy motor would work, but it was damaged beyond repair by
the floor rising to hit it during those psychokinetic events. The nearly
finished glass of scotch, quite unaccountably, still maintained its posi-
tion on the desk.
Dear Sirs:
Yours in learning,
Char les W. Harrington
This evening I was sitting in my garage, which is Vi63 meters long
and has a door at each end, waiting for my friend Wade Wykert to drive
his psychic energy powered automobile through it at his usual speed of
0.995 times the velocity of light. His automobile is also Vi63 meters
long, but it is Lorentz-contracted by a factor of 10, so I have no difficulty
in closing the entrance door before I have to open the exit door to let him
out of the garage. My friend, however, thinks that my garage is Lorentz-
contracted to one-tenth the length of his automobile and that, as he
passes through both doors have to be open at the same time. So far, I
have managed to open and close the doors at appropriate times so that
his belief in the special theory of relativity has not been destroyed.
As he passed by he threw out to me a copy of the April 1975 issue of
Scientific American open to Martin Gardner's Mathematical Games Sec-
tion. I hoped that he had not read the part about Humbert Pringle's
Gedanken-experiment disproving special relativity. At best this knowl-
edge would make him nervous next time he drove through, and might
result in the splintering of my garage doors.
Six Sensational Discoveries 691
I noticed that Wade had succeeded in coloring William McGregor's
110-region map with only four colors, thus unvindicating H. S. M. Cox-
eter's foursighted insight. A handwritten note indicated a desire to have
his achievement publicized before 1978, when McGregor's report is
scheduled to appear.
I then read with great interest the rest of Martin Gardner's article on in-
adequately reported major discoveries of 1974, noticing that I was sitting
in an attitude rather similar to that of Leonardo (the figure in the draw-
ing must surely be Leonardo himself) in the reproduction on page 127.
I was surprised that Gardner was not aware of the analysis made by
the chess-playing computer built at the editorial offices of Ms. and
named MacHaec. (One of the editors is a Latin scholar as well as a fem-
inist.) MacHaec's analysis indicates that the pawn to king's rook 4 open-
ing leads to a sure win for black. Clearly a match between MacHic and
Mac Haec is indicated, and there have been rumors that Bobby Fischer
might put up the $25 million prize money.
Martin Gardner's point that such discoveries as these are frequently
not brought to the public notice is well taken. There is, however, a sat-
isfactory alternative for the authors of such world-shaking communica-
tions. They should be submitted to either the International Journal of
Hypercritical Obfuscation or the Archives of Information Pollution. I am
the editor of both these journals and I will guarantee that articles sub-
mitted will be dealt with expeditiously.
Dear Sir,
Sincerely yours,
John L. Howarth
Professor of Physics
Having completed reading your article in the April issue of Scientific
American I feel that it is my duty to inform you that the reason the psi
motor runs was discovered approximately one year ago. You see there is
an aura of glacomoridine which can envelop the person's hands.
According to the papers I have received over the past five years, it
seems that some humans produce more of it than others. It now appears
certain that glacomoridine is a hormone. However, there was much dis-
cussion on whether or not it was produced internally or not. This prob-
lem was cleared by R.E. Coleman et al. After many arduous months of
searching for this elusive organ, they discovered what appeared to be a
satellite on the pituitary gland. Analyzation proved that this was what
they were looking for. Because of its white appearance it was decided to
name it the glacier gland.
But this doesn't really explain how the psi motor is able to work. It re-
692 MISCELLANEOUS
mained for Richard Douglas of Washington and Jefferson College to pro-
vide a mechanism which would solve this problem. It appears that
glacomoridine enters the blood stream after being secreted by the glac-
ier gland. (There would seem to be a conscious control of the gland by
the individual.) After being circulated it is then collected in some way
(still unknown) in the sweat glands of the hands. From there it is re-
leased along with any perspiration. However, instead of diffusing into air
the physical properties of glacomoridine enable it to remain close to the
hand. The glacomoridine begins to "flow" distally. As can be imagined
this movement would cause the psi motor to move. Your term psi motor
can still be used, because studies have shown that the level of glaco-
moridine actually increases when thoughts are directed to moving the
motor. More concrete evidence of the increase is demonstrated by ob-
serving an EEG taken during the tests. It was shown that both the alpha
and delta waves were dominant.
One problem which wasn't adequately answered until a month ago
dealt with why some people were able to rotate the psi motor in the re-
verse direction. The answer to this is also credited to Richard Douglas.
A recessive gene, which when homozygous, is responsible for the flow
of glacomoridine in the opposite direction than is normal.
Sincerely yours.
William Hughes
And now for the sad story of Martin von Strasser Caidin, prolific au-
thor and pilot who died of cancer at age 69 in 1997. He wrote some 140
books, most of them nonfiction, about such topics as aviation, space
flight, and UFOs, but more than 30 were science-fiction novels. His
novel Cyborg was the basis of the TV series The Six Million Dollar Man
and its spinoff The Bionic Woman. Marooned became a movie with
Gregory Peck and Gene Hackman as stranded astronauts. Other novels
include The Messiah Stone, Exit Earth, Prison Ship, Three Corners to
Nowhere, Zoboa, and Ghosts in the Air.
Why do I mention Caidin here? Because in his declining years he
convinced himself that his mind could rotate psychic motors. His home
contained a "target room" filled with little paper, cardboard, and foil
"umbrellas" suspended on needles and toothpicks. The room had no
ventilation. Caidin would sit for hours in an adjoining room, staring
through a window and trying to control the movements of the little ro-
tors with his brain. The devices behaved erratically; sometimes not
moving, sometimes rotating one way, sometimes the other way. The
Six Sensational Discoveries 693
motions were surely caused by heat convection currents, or perhaps by
air moving under doors, but Caidin believed he had proved the reality
of a mysterious force totally unknown to science, a force responsible for
what he called telekinesis. He was convinced that his discovery would
lift him into the ranks of Tesla and Einstein! He couldn't understand
why physicists ignored him. Here are a few references:
M. Caidin, "Fiction That Ain't," New Destinies, Summer 1988, pp. 211-22.
M. Caidin, "Telekinesis Demonstrated," Fate, January 1994, pp. 52-68. Includes
photos of Caidin's "rotors."
W. Cox, "Mind Moving Matter," Florida Today, January 8,1989, pp.1D-2D.
W. Robkin, "Martin Caidin," Star log, October 1985, pp. 63-65.
In concluding this long addendum, let me pass along some sage ad-
vice supplied by my brother Jim. Every responsible science writer
should constantly keep in mind the following four words: "Accuracy
above all."
694 MISCELLANEOUS
Selected Titles by
the Author on
Mathematics
The "Mathematical Games" columns in this book are arranged
by subject. The entire collection of "Mathematical Games" columns
that appeared in Scientific American has been brought together by Mar-
tin Gardner in chronological order in the series of books listed below.
In 2001 all of them were in print or being reissued. If you enjoyed the
selection here, you will want to look at these books by Martin Gardner
as well.
Martin Qardner's IIMathematical Qames" books:
The Scientific American Book of Mathematical Puzzles and Games, Simon and
Schuster, 1959 (later retitled The First Scientific American Book of Mathemati-
cal Puzzles and Games), republished as Hexaj1exagons and Other Mathematical
Diversions: The Scientific American Book of Mathematical Puzzles and Games,
University of Chicago Press, 1988.
The Second Scientific American Book of Mathematical Puzzles &- Diversions,
Simon and Schuster, 1961, republished by University of Chicago Press, 1987.
New Mathematical Diversions from Scientific American, Simon and Schuster,
1966; revised edition, New Mathematical Diversions, The Mathematical Asso-
ciation of America, 1995.
The Unexpected Hanging and Other Mathematical Diversions, Simon and Schus-
ter, 1969, republished by University of Chicago Press, 1991.
The Numerology of Dr. Matrix, Simon and Schuster, 1967; expanded and repub-
lished as The Incredible Dr. Matrix, Scribner, 1979. See below.
Martin Gardner's Sixth Book of Mathematical Games from Scientific American,
W. H. Freeman and Company, 1971, revision in press by The Mathematical As-
sociation of America, 2001.
Mathematical Carnival, Knopf, 1975, revised with a foreword by John H. Conway,
The Mathematical Association of America, 1992.
695
Mathematical Magic Show, Knopf, 1977, revised with a foreword by Ronald L.
Graham, The Mathematical Association of America, 1990.
Mathematical Circus, Knopf, 1979, revised with a foreword by Donald E. Knuth,
The Mathematical Association of America, 1992.
The Incredible Dr. Matrix, Scribner, 1979 (see The Numerology of Dr. Matrix,
above), revised and expanded as The Magic Numbers of Dr. Matrix, Prometheus
Books, 1985.
Wheels, Life, and Other Mathematical Amusements, W. H. Freeman and Company,
1983.
Knotted Doughnuts and Other Mathematical Entertainments, W. H. Freeman and
Company, 1986.
Time Travel and Other Mathematical Bewilderments, W. H. Freeman and Com-
pany, 1988.
Penrose Tiles to Trapdoor Ciphers, W. H. Freeman and Company, 1989.
Fractal Music, Hypercards, and More . .. : Mathematical Recreations from Scien-
tific American, W. H. Freeman and Company, 1992.
Last Recreations: Hydras, Eggs, and Other Mathematical Mystifications, Coperni-
cus Books, Springer Verlag, 1997.
696 Selected Titles by the Author on Mathematics
Index
Page numbers in italics refer to illustrations.
Abbott, Edwin Abbott, 94, 98
Abbott, Robert, 504-5, 511, 513-14
Abbott's New Card Games (R. Abbott), 504
Absolute, 334, 598
Abstract Painting (Reinhardt), 599, 608
ace, 78,79,83-84
Adams, Edward, 279, 538-39
Adventures of a Mathematician (Ulam),
300
Advertisements for Myself (Mailer), 124
Agnesi, Maria Gaetana, 66
Agrippa, 316
Aha! Gotcha (Gardner), 208, 282
Aha reaction, 504
"Ainmosni" (Angell), 26
Akin, Ethan, 466
Albee, Edward, 315
Alembert, Jean Ie Rond d', 273
aleph (infinite number), 613
aleph ladder, 327
aleph-null, 327-37, 341, 344, 346
aleph-one false proof, 346, 346, 348
digits, 348
sets, 344, 345, 346
aleph-one, 332-35, 337-38, 341-44, 346
aleph-null false proof, 346, 346, 348
Alexander, James w., 243
Alexander, Peter, 569
Alexander horned sphere, 369
Alexander Nevsky, 368
Alexander polynomial, 243, 248
algebra, 4, 10, 11
arithmetic and, 1-32
irrationals in, 332
knots and, 239
algorithms:
for creative arts, 627-28
Ilfnoise and, 360-62, 361
for n-tuplets, 305-8, 308, 310
for Ramsey numbers, 442
Alhambra, 213
Ali, Muhammad, 309
Alice in Wonderland (Carroll), 239, 596
Allen, Woody, 520
"All Numbers, Great and Small" (Conway),
594
Alpine Symphony (Strauss), 351
Ambigrammi (Hofstadter), 671
ambiguity, probability and, 273-85
American Fairy Tales (Baum), 533
"America the Beautiful" (Bates), 601
Ammann, Robert, 76, 89
amphisbaena, 647
Amusements in Mathematics (Dudeney),
134
Anableps, 645
Anagog, 71
Anarchy, State, and Utopia (Nozick), 587
Anderson, Poul, 528
"-And He Built a Crooked House" (Hein-
lein), 169-71
Angell, Roger, 26
angles, right, 59
animals, 495-501, 495, 496, 498, 499
bilateral symmetry in, 643-44
helices in, 119-20, 120, 121
toroidal, 650
animal ticktacktoe, 493-503, 495, 496, 498,
499
Anning, Norman, 6
Another World (Schofield), 154
Anthropic Cosmological Principle, The
(Tipler), 528
antiparticles, 597
Apology for His Book (Bunyan), 369
Appel, Kenneth, 682
Applegate, David, 491
April Fools' jokes, 609, 682
Apropos of Dolores (H. Wells), 291
Aquinas, Thomas, 536
arcs, 36, 39-40
circular, 59, 60
Aristotle, 316, 350, 365, 535-36, 554
arithmetic, 178-79
algebra and, 1-32
modular, 320
Armchair Universe, The (Dewdney), 108
Armer, Paul, 480
697
Arnobius the Elder, 291
Arp, Hans, 212
Arrow, KennethJ., 297
Arrow in the Blue (Koestler), 207
Arrow paradox, 297-300
Ars magna (Lull), 629
art:
absrract, 190, 212, 359,398
algorithms for, 627-28
cubism, 212
fractals and, 359
mathematical, 212
nothing in, 598-600, 599
numbers in, 627
Op,212
quarter-turn pictures, 191-92, 193,
195-97, 196
representational, 212
symmerry in, 189-90
upside-down pictures, 19a-92, 191, 192,
195,196-97
Articles of Belief and Acts of Religion
(Franklin),317
artificial intelligence, 504, 664, 665
Art of Computer Programming, The
(Knuth), 594
Ascending and Descending (Escher), 77,
216,218,664
Asimov, Isaac, 518, 538
Asterix, 86-87, 86
Asrria,95,97,99, 100,101, 106-7,108
As You Like It (Shakespeare), 538
Athanasius Kircher (Godwin), 630
"A Tisket, a Tasket," 634
Atiyah, Michael, 248
atoms, 13
gold,596
infinity and, 332-34
in steady-state theory, 337
At the Drop of a Hat (Flanders and Swann),
122
Attitudes Toward Intelligent Machines
(Armer),480
Atwater, Thomas, 406
Augustine, Saint, 598
Aulonia hexagona, 642-43, 642
Austin, A. K., 502
autocorrelation, 351
Avatar, The (Anderson), 528
Babbage, Charles, 662, 668
Bach, Clive, 84
Bach, C.p.E., 631
Bach, G., 279
Bach, Johann Sebastian, 201, 201, 351, 354,
629,631,638,660,662,663,664,
665,667,668,669,672
bacteria, flagella of, 649
Baer, Hermann, 172
Bailey, Herbert S., Jr., 255
698
Balinski, Michel L . 67
Ball, W. W. Rouse, 71,597
Banchoff, Thomas F., 158
Bandt, Christoph, 56
Bankoff, Leon, 148
barber's pole helix problem, 124, 126-27
Bar-Hillel, Maya, 585-87
Barnes, Lakenan, 608
Barr, Stephen, 231, 233-34, 237, 255
Barth, Karl, 154
Barton, Hiram, 171-72
Bashmakova, Isabella, 7
Bateau, Le (Matisse), 190
Bates, Katharine Lee, 601
Batman, 85, 86-87
BatteIl, Jack Srraley, 479
Batz, Philipp, 605-6
Baudelaire, Charles, 620
Baum,L.Frank,533, 596,647-49, 648
B.C., 557
Bean, R. J., 257
Beaver, Harold, 620
Beck, A., 658
Beck, Derek, 43
Bedient, Richard E., 282-83
bee hives, 646
Beelzebub puzzle, 426-27, 426, 434
"Before the Dawn" (Bell), 520
"Behind Monty Hall's Doors: Puzzle, Debate
and Answer'?" (TIerney), 284
being, nonbeing vs., 598
Belash, Constantine A., 158
Bell, Eric Temple, 177, 520
Bellamy, Edward, 520
Bellman, Richard, 480, 482
Belvedere (Escher), 216, 217
Benford, G. A., 519-20
Benzoni, Girolamo, 64
Berent, Paul, 544
Berg, Howard C., 649
Berger, Robert, 76
Bergerson, Howard W., 26, 28-29
Berkeley, George Bishop, 618
Berlekamp, Elwyn R., 111, 380, 407, 461
Bernhart, Frank, 396
Berrrand, Joseph, 275
Besicovitch, Abram Samoilovitch, 42-43
Best in Chess, The (Horowitz and Battell),
479
Best of Origami, The (Randlett), 430
Betti, Enrico, 227
Betti numbers, 227-29, 231, 235
Bevan, Edwyn, 536
Beyond the Third Dimension (Banchoff),
158
Biathanatos (Donne), 605
"Bicycle Tubes Inside Out" (Taylor), 269
Bierce, Ambrose, 471
Big Bang theory, 334,613-14
Big Crunch, 622
Index
Biglow Papers, The (Lowell), 504
bilateral symmetry, in animals, 643-44
bindweed, 121-22, 123
Bing, R. H., 257-58, 258, 265-66, 265
bingo cards, nontransitive, 302, 302
Biot, Jean Baptiste, 633
biprisms, 90, 90
Birman, Joan, 248
Bishop, Sheila, 281
Black, Duncan, 299
Black,11ax, 345, 347, 542, 588
Black, Norton, 67
black holes, 528, 614
Blass, Andreas R., 497,501
"block universe," 521
Bloom, 11elvin, 43
Bloxbox,71
"Blue-Empty Chromatic Graphs" (Lorden),
453
Blyth, Colin R., 547-49, 548
Blyth's paradoxes, 548-50, 548
board games, in a planiverse, 111-12
Bogey Beasts (Sime), 651-52
Bohm, David, 324
Bohr, Niels, 581
Bolt, A. B., 481
Boltyanskii, V. G., 43
Bolyai, Farkas, 176-78
Bolyai, Janos, 176-77, 178, 183
Bonner, John Tyler, 119, 645, 651
Book, D. L., 519-20
Book of Imaginary Beings (Borges), 647
Book of Pistols and Revolvers (W. H.
Smith), 109
Boole, George, 11
Borges, Jorge Luis, 327,523,533,605-6,
627,647,648,649
OSBERG problem and, 425, 433-34
Borgmann, Dmitri, 26-27, 29
Borowski, E. J., 608
Borromean rings, 250, 264, 265
Borwein, J. 11., 608
Bosch, Eugen I., 172
Botvinnik, 11ikhail, 479-80, 482, 676
Boumphrey, GeoftIey, 654
Bouncing Baers, 445
Bourne, S. R., 414
Bowers, J. F., 194-95
Bowery El phenomenon, 604
bowline knot, 241,241,250
boy probability problem, 277
Bradbury, Ray, 523,599
Bradley, F. H., 316
Brahma, 621
Braiding and Knotting (Belash), 158
"Braids Which Can Be Plaited with Their
Threads Tied Together at Each End"
(Shepperd), 158
Braithwaite, R. B., 438, 542
"Branches of Time" (Daniels), 522-23
Index
Brandt, J0rgen, 463
brane theory, 158
Brave New World (A. Huxley), 398
Brewer, John, 408
Bridgman, Percy W., 336-37, 347
Brill, A. A., 27
Brillhart, John, 684
Brock, Thomas D., 645
broken stick problem, 273-75, 274, 278-79
Brooks, Fred, 638
Brousseau, Alfred, 24
Brown, Fredric, 518, 524-25
Brown, G. Spencer, 636
Brown, John R., 481
Brownian:
landscape, 357-58, 357
motion, 358
spectrum, 359
Browning, Ed, 110
Browning, Elizabeth Barrett, 297
Browning, John, 110
brown music, 362-64, 363, 367
randomizers and, 352, 353-54
brown noise, 353, 354, 357, 358, 359-60,
360,362
Brunelleschi, Filippo, 65
Bruno, Giordano, 649
Brussels Sprouts, 489, 490, 491-92
"Buckyball" (Edelson), 655
buckyballs, 655,655,657
Budget of Paradoxes, A (De 11organ), 181
Buffon, Comte de, 132
Bulgarian solitaire, 462-66, 463, 464, 465
Buller, A. H. Reginald, 519
Bunyan, John, 369
Burr, Stefan, 443, 444-45, 451
graph chart of, 446-47, 451
burr puzzle, 111
Bust-up (Reyburn), 686
Butler, William J., Jr., 671
Buxbaum, David, 514
buzz-phrase generator, 635, 636,637
buzzsaw, 85, 86
buzzwords, 638
Byron, George Gordon, Lord, 290-91, 532
C, see aleph-one
Cabell, James Branch, 533, 538-39
Cage, John, 600
Caidin, 11artin von Strasser, 693
Cain, A Mystery (Byron), 532
calculus, of finite differences, 10-22
Calder, A., 178
Caldwell, Roy L., 653
Calegari, Antonio, 632
Callahan, Jeremiah Joseph, 181-84
calligraphy, symmetrical, 198-99, 198,
200-203,200,201,202
cannon ball puzzle, 130, 133-34
canons, 365,366, 367
699
"Can the Unexpected Never Happen?"
(0'Beirne},574
Cantor, Georg:
diagonal proof of, 331-32, 331, 335-37
transfinite numbers of, 320, 327, 334-37,
341-44,346,371,457,596,612
"Cantor's Diagonal Process" (Dudley), 337
"Capture of Father Time, The" (Baum), 533
card color probability problem. 288-89,
291-93,292,295
cardinal:
number of sets, 612-13
transfinite numbers, 342
Carington, Whately, 154
Carlson, George, 195
Carnap. Rudolf, 288, 289-90, 542-43. 545,
549,550,602
Carrington, Hereward, 246, 686-87
Carroll, Lewis, 665
Achilles and Tortoise paradox of, 316,
663
Alice books of, 118, 162,239,318-19,
537,596,608
on non-Euclidean geometry, 179
on voting paradox, 299
Carter, Jeffrey, 111
Cartesian:
coordinates, 60, 64, 342. 557
formula. 137-38
cartwheel pattern. 84,85
Caswell, Robert L., 483
causality:
backward, 583, 586
time and, 586
"Central Tessellations" (Goldberg), 73-74
Chambers, John, 478
chance, 539, 581
Chatham, George N., 687-89
checkerboards, 73
checkers:
computers and, 480, 482
mini, 477, 477, 483, 483
in a planiverse, 106,106,111-12
Cheney, Fitch, 160
Cheng, KC.H., 294-95
Chernick, Paul, 550
chess, 371
complexity of, 480-81
computer, 479-82,664,675-76,684
machines, 472
mini, 477,477,482
inaplaniverse, 106, 107, 111, 112
chessboards, n-dimensional, 144-46, 146,
148
chess problems, 199
queen problem, 203, 204, 208-9, 209, 210
Chesterton, G. K., 81, 190, 603
Childhood's End (A. Clarke), 108
Chinese Button Knot, 247
Chi Po and the Sorcerer (0. Mande!), 613
700
Chopin, Frederic-Fran;ois, 634-35, 638
chord, 275-77
Chow, Timothy. 576
Christianity, 154, 169, 290-91
chromatic numbers, 227-29, 231, 233, 234,
235
"Chronic Argonauts, The" (H. Wells), 517
chronon, 539
Chung,Fan,437,444
Church, Alonzo, 665
Church of the Fourth Dimension, 150-61
Chvatal, Vaclav, 444
Cicero, 504
circle-and-spots problem, Moser's, 559,
560-62,562
Circle Limit III (Escher), 179-81, 179
c ~ c l e s , 35,59,63,117, 178,275-76
Cartesian formula of. 137
curvature of, 140
definition of, 137
packing of, 141, 142
in torus. 263
as 2-sphere, 137
Claris.208
Clarke, Arthur. 108
Clarke, John. 351
Cliff, Norman, 279
Clinton, J., 633
"Clock That Went Backward, The"
(Mitchell),517
closed curves, 239, 245, 262, 267
convex, 35
infinite, 321-22
knots in, 255, 259
close packing, 130-32, 130
coconut problem. 3-9
"Coconuts" (Williams), 3-4, 8
codes, error-correcting, 147
Codex Madrid I (Leonardo), 678
Coffee, M.M.H., 236
Cohen, L. Jonathan, 569
Cohen, Paul J., 327. 334-35. 342
Col. 375-76, 376
Colin, Ralph F., 598
Collected Papers (Peirce), 340
Columbus, Christopher, 64
"Combinatorial Significance of a Theorem
of P6lya, The" (Riordan), 18
combinatorial theory, 423, 445, 559, 559
in music composition, 628
problems in, 438
complement, of surfaces, 262
complementarity, principle of, 581
complexity, simplicity vs., 553-56, 562
Componium, 632-33
compossible worlds, 618, 622
computability theory, 664
computers:
building of, 481
checkers playing, 480, 482
Index
chess playing, 479-82, 664, 675-76, 684
Eleusis programs for, 504
languages for, 664
as learning machines, 471-72, 480-82
music composed by, 628, 634-36, 664
paintings and, 627
Sprouts analyzed by, 491
Conan Doyle, Arthur, 601, 649
Concept of Nature, The (A. Whitehead),
532,554
Concrete Mathematics (R. Graham, Knuth,
and Patashnik), 19, 207
Condorcet, Marquis de, 299
Condorcet effect, 299
configuration point, 616, 617
configuration space, 615-16, 616, 617
graphs,616-17,616
confirmation:
degree of, 543, 545
and inductive reasoning, 543-46
relevance, 545
Confirmation and Confirmability
(Schlesinger), 535
conical helices, 118-19
Conklin, Groff, 471
consciousness, 664
"Constructive Mathematics" (Calder), 178
Continuum Hypothesis, 334, 342
convex:
figures, 43
quadrilaterals, arbitrary, 558-59, 558,
559-60,561
solids, 90
Convex Figures (Yaglom and Bolt yanskii) , 43
Conway, John Horton, 205,370,456
games and game theory and, 111, 369-82,
406-7,409-20,485,487-89,491
knotted torus proof of, 259-61, 260
knot theory of, 242, 243
number construction theory of, 594-96
n-tuplet algorithm of, 305-8, 308, 310
Penrose tile study of, 77-90, 90
on sphere packing, 147
temperature theory of, 379
Conway polynomial, 243
Conway's constellation, 409
Cope, David, 638
Copernican theory, 527
Corey, "Professor" Irwin, 529
"Corpus Hypercubus" (Dali), 153, 169, 170
cosmology, 332-34,613-14
countability, 328
countable set, 341
counterfactuals, 665
Counterfeiters, The (Gide), 318
counting numbers, 341, 348, 592-94
Cox, Jeremy, 431
Cox, William, 246
Coxeter, H.S.M., 131, 141-43, 179-81, 213,
220,263,674,692
Index
"Crab Canon" (J. S. Bach), 663
Crapper, Thomas, 679, 686
Creation of Heaven and Earth, The (Sed-
lak), 171
Critique of Non-Euclidean Geometry (Fis-
cher), 183
crossbill, 644
cross-cap, 234
crosscut, 227-28, 230
cross-stitch curve problem, 323,323,
324-25,324
Crowe, D., 658
Crux, 71
crystallography, 213, 215
Crystal Man, The (Mitchell), 517
crystal symmetry, 181
cubes, 42, 208,150,235, 596,640,671
palindromes in, 25
points in, 343
as replicating figures, 54
topological properties of, 228
as two-hole torus, 257-58, 258
2-space projection of, 166, 166
unfolding of, 168, 169
and unknotted cord problem, 244, 244,
250
cubic close-packing, 131
cubing a cube problem, 320-21, 321
cubism, 212
Cummings, Alexander, 679
Cundy, H. M., 220
Curl, Robert, 657
curl-up, 649, 651
curved triangle (Reuleaux triangle), 36-37,
36,40-41,43
Curve of the Snowflake, The (Walter), 322
curves, 118
astroid,66
closed, see closed curves
concave, 61
deltoid, 42, 43
infinite closed, 321-22
Lame, 66
pathological, 321, 324
Peano, 355-57
snowflake, see snowflake curves
2
C
and, 344
curves of constant width, 35-45
arbitrary arcs and tangents, 39-40, 40
crossed line method of drawing, 38-39, 39
noncircular, 36-43
star polygon method of drawing, 38, 39
symmetrical, 37-38, 38
theorems on, 40
unsymmetrical, 38-40, 39
in woodworking, 40
Cutcake, 379,379, 380-81
cycloid, 118
cylinders, 68, 117,118
cytosine, 663
701
Dadaism, 212
Dalal, Sam, 432
Dali, Salvador, 153, 169,170,196-97,
212
Damian, Peter, 536-37
dance, mathematical, 208, 211
"Dances for the Mind's Eye," 208
Dangerous Corner (Priestley), 523, 535
Daniels, David R, 522-23
Danskin, John M., 7
Dante Alighieri, 647
darts,77-79,78,80,85,87,88-89
Darwin, Charles, 646
Dash, J. G., 98
Davis, Morton, 466
Day and Night (Escher), 213, 214
DCC Exercises in Choice and Chance (Whit-
worth), 278
decahexaflexagons, 389, 392
decapods,85-87
"deceptahedron," 656-67,656
decimal fractions, 331
decision theory, 583
deductive reasoning, 15, 542, 559
Deep Blue, 482
deflation, 79,80,82,88
"Delayed Justice Puzzle, The" (MaIorana),
433
Delphi, 504, 512
Delphi (M. Kruskal), 504
deltahedra, 640-41
convex,656-57,656
De Morgan, Augustus, 46-47,181,611
Dennett, Daniel C., 670
denumerability, 328
Dershowitz, Nachum, 457-58, 460
Descartes, Rene, 141
determinism, 580, 582, 585, 587, 588
deuce,83,84
DeVicci, Kay R Pechenick, 172
devil's corkscrews, 120, 643
Devlin, Keith, 90
Dewdney, Alexander Keewatin, 94-106,
108-11
Dewey, John, 580
DeWitt, Bryce, 526, 527
Diabolical cube, 407
"Diamond Lens, The" (F. O'Brien), 319
dice:
music composed with, 629-32, 636
nontransitive, 286-88, 287, 293-94
Dice Composer, The (Laszlo), 631
Dicke, Robert, 554
Dickinson, Emily, 535, 553
Dictionary of Mathematics (Borowski and
Borwein),608
Diderot, Denis, 291
Dilworth, William, 336
Dim, 378
702
dimensions:
infinite, 597
packing in, 134
Dingle, Herbert, 684
Diophantine equations, 4, 7,8,671
Diophantus and Diophantine Equations
(Bashrnakova), 7
Diophantus of Alexandria, 4
Dirac, P.A.M., 5, 597
Directory of Tunes and Musical Themes,
The (Parsons), 636-37
disks, 223
Divinatione, De (Cicero), 504
DNA,119,663
knot theory and, 249
dodecahedrons,42,640, 642,643,658
packed spheres and, 132, 133
small stellated, 216-20
dodecahexaflexagons, 392-93
dominance principle, 584
Domineering, 372-74,375
dominoes, 73, 76,497,498
Don Juan (Mott!), 351
Donne, John, 605
"Doppelganger" (Lindon), 28-29
Dorothy and the Wizard in Oz (Baum), 596
Dot and Tot of Merryland (Bawn), 647
double helix, 119
Double Helix, The (J. Watson), 117
doughnuts, 254-70, 608
problem, 16-17
Dowling, L., 633
Drawing Hands (Escher), 318, 319
Dr. Dolittle books (Lofting), 647
Drobot, Stefan, 110-11
Dr. Schaffer & Mr. Stern Dance Ensemble,
208,211
duahexaflexagons,392
Dudeney, Henry Ernest, 15-16, 17-19, 134,
425,430,431
Dudley, Underwood. 184, 337
Dunciad (Pope), 647
dung beetle, 646-47
Dunn, Angela, 24

537-38,651
Duo (palindromic dog), 647, 648
e,674-75,684
earthworms, 650
Eckler, A. Ross, 29-30
Eddington, Arthur Stanley, 619
Edelson, Edward, 655
edges, as topological property, 227-30, 233,
234,235
Edwards, Paul, 602
Efron, Bradley, 286-87, 293
eggs, 64-65, 69
egg transfer game, 455-56
Index
egocentric predicament, 220-21
eight-sided deltahedron problem, 642, 657,
657
8-space. 144
18-point problem, 460-62, 461
Einstein, Albert, 581, 619-20
non-Euclidean geometry and, 182, 183,
184
relativity theory of, 184, 518
on simplicity, 553-54
universe model of. 137-38, 152, 318,613
Einstein, Elsa, 554
Einstein Myth and the Ives Papers, The
(Hazelett and D. Turner), 184
Eisenstein, Sergei, 368
Ekbom, Lennart, 575
electromagnetism, 108, 137, 152
electrons, 13-14, 597
free. 597
Elements (Euclid). 175, 183
Eleusinian mysteries, 504
Eleusis, 504-12, 514
dealer's role in, 505, 507, 509, 510, 511
expulsions in, 510
False Prophet in, 505, 509, 512
induction in, 504, 505

Prophetin,505,508-10,511,512
scoring of, 505
secret rule problem, 511, 512-13, 514
secret rules in, 505, 507
Eliot, George, 601
Elliott, Hester, 604, 605-6
ellipses, 35,59,60-61,63, 555
ellipsoid of rotation, 64
ellipsoids. 64
elliptic curvature, 182
Ellis, Robert A., 479
Elves', Gnomes', and Little Men's Science-
Fiction Chowder and Marching Soci-
ety, 599-600
EM! (Experiments in Musical Intelligence).
638
empty sets, see null sets
Encyclopedia of Philosophy, The, 603, 606,
608-9
Encyclopedia of Psychic Science (Fodor),
686
Encyclopedia of the Paranormal, The
(Stein), 157
Endgame, 372
endless tasks, 455-67
End of Eternity, The (Asimovl. 518
Enjoyment of Mathematics, The
(Rademacher and Toeplitz), 40, 643
enneagons, 74, 75
enneahexaflexagons, 392
Ephesians, Epistle to the, 151
Episode of Flatland, An (Hinton), 94
Index
equations:
Diophantine, 4,7,8
indeterminate, 4
Erdos, Paul, 284,437,444,445,448,450-51
error-correcting codes, 147
Escher, George, 223
Escher, M. c., 73, 74, 77, 87, 179-81,179,
201,201,212-14,318,319,649,651,
660,662,663,664,667
ESp,433
symbols problem, 332, 333, 337-38, 338
Essay on Silence (Hubbard), 600
Essays in Honor of Carl C. Hempel
(Rescher), 582
ether, stagnant, 597
Ethics (Aristotle), 535-36
Euclid, 175, 183, 335, 663
fifth postulate of, see parallel postulate
fourth proposition of, 175
proofs of, 612
Euclid Cleared of All Blemish (Saccheri),
177
Euclidean geometry, 163, 175, 179, 182,
184,335,355,448
dimensions in, 137, 178
4-space in, 162, 169
hyperspaces in, 137
language of, 685
non-Euclidean hyperbolic plane model
of, 213
plane, 612
spaces in, 134,450-51
tiling in, 181
Euclidean Ramsey:
problem, 448-50, 450
theory, 450-51
Euclid or Einstein (Callahan). 182
Euler, Leonhard, 11,219,643,657-58
Euler's formula, 219, 643, 657-58
Euler's phi function, 19, 20-21
Eureka (Poe). 620-21, 622
Everett, Hugh, III, 524, 526-27, 528
Ever-ready Composer of Polonaises and
Minuets (Kirnberger), 629
existentialism, 602, 608

expectation of satisfaction, 549-50
expected-utility principle, 583-84
"Experiment" (F. Brown), 525
"Exploring the Planiverse" (Dewdney), 94
"Extendapawn-An Inductive Analysis" U.
Brown),481
Fabre, Jean Henri, 647
faces, upside-down, 192-94
Fadiman, Clifton, 8, 170, 236
"Fakes and Frauds in the Art World"
(Colin), 598
falsification, 551
703
Fantasia Mathematica (Fadiman), 170, 236
Farey, John, 329
Farey numbers, 329
Far Side, 609
fate, 581
Fechner, Gustav Theodor, 649
Fermat's last theorem, 664
Feynman, Richard P., 386, 389-90, 395,
518,520,531
Feynman diagrams, 664
Feynman graph, on time travel, 521-22, 522
Fibonacci numbers, in botany, 121
fifty pence coins, English, 43
figurate numbers, 128-29
Filipski, Gary, 417
Filliou, Robert, 600
finite differences, calculation of, 13-15
finite sets, 340, 438, 595
subsets of, 340
finite trees, 457-58
Finnegans Wake (Joyce), 319,485,669
"First Time Machine" (F. Brown), 518
Fischer, William L., 183
fivefold symmetry, 81, 82
5-space, 143
536 Puzzles & Curious Problems (Dudeney),
425
flachpunkt, 68
flagella, of bacteria, 649
Flanders, Michael, 122, 123
Flatland, 94, 108-9, 138, 151, 171
Flatland (E. Abbott), 94
"Flatland Redux," 108
Flaubert, Gustave, 647
Fletcher, John G., 367
flexagons, 385-97
higher-order, 394
"street," 392
"flow-snake," 357
Flushed with Pride (Reyburn), 686
Fodor, Nandor, 686
Folding Frenzy (J. Cox), 431
Folding Money: Volume II (Randlett), 430
Folkman, Jon, 444, 445
Ramsey graph of, 445-48
fool's kite, 78, 79-80
Ford, Lester R., Jr., 281
Foreman, George, 309
Fort, M. K., Jr., 259
FORTRAN,479
Forward, Robert, 528
Foundations of Mathematics, The (M. Ram-
sey),438
4 '33"(Cage), 600
four-color-map theorem, 674,675,682-84
4-cube, 143, 172
see also tesseracts
four-dimensional space, 94
fourfold symmetry, 189
Fourier analysis, 628
704
4-space, 137, 153, 154, 162-64, 199,255,
318,613
Klein bottles in, 231
knots in, 249
packing in, 143
as symbolic, 169
4-sphere, 137-38, 143
Cartesian formula of, 137
Fourth Book of lorkins, The (Dunsany), 523
fourth dimension, 245,448,517,518
Church of the, 150-61
in relativity theory, 162
see also 4-space
Fourth Dimension, The (Hinton), 162
fourth powers, palindromic, 25
Fox, Ralph H., 259
fractal curves, 355-57,356
Fractal Geometry of Nature, The (Mandel-
brot), 354-55
fractals, 80, 324, 355-59
in art, 359
in music, 350-68
in nature, 357-58
self-similar, 56
fractal scaling, 355
fractions:
decimal, 331
infinite binary, 335
integral, 328, 335-36, 341,595
irrational decimal, 348
in Peirce's positive rationals proof,
328-29
rational, 331
Franklin, Benjamin, 317
Frazier, Joe, 309
Frederick the Great, 669
Fredkin, Edward, 539
Freer, Winston, 160
free will, 521, 536, 539, 580-81, 588, 664
Frege, Gottlob, 592, 594
Freudenthal, Hans, 640
Freud's Contribution to Psychiatry (Brill),
27
Frontiers of Time (Wheeler), 527-28
FTOZOM (fundamental theorem of zero-
order moribundity), in Sprouts, 489
fugues, 668
Fuller, Buckminster, 655
fullerenes, 655
Fulves, Karl, 293, 431
Gabriella Computer Kit, 481
Galileo, 554
gambling, nontransitivity and, 297,
300-310
games:
achievement, 451, 494, 501
avoidance,442,451,501
construction of, 371-80, 594
impartial, 371
Index
infinite outcomes for, 374
misere, 371,485
nontransitivity in, see nontransitivity in
games
partizan, 371-72
in a planiverse, 106-7, 111-12

simulation, 409, 504
take-away, 485
transitivity in, 297
see also specific games
gametheory,469-514,581,585
of Conway, 372-75
machines. 471
Gamow, George, 344, 518, 569, 684-85
"Gamut of Game Theories, A" (Conway),
380
Gans, Carl, 646
Gans. Eric L., 491
Gardner, Beau. 195
Gardner, Jim, 694
Garibaldi, Giuseppe, 190
Gateway Arch, 61
Gauss, Karl Friedrich, 176-77, 178, 182
Geller, Uri, 682, 687
genetic laws, 409-10
"Genius for Games, A" (Buxbaum), 514
Gentle Art of Mathematics, The (Pedoe),
231
Gentzen, Gerhard, 178
geodesic domes, 655
geodesics, 178, 180
Geometric Transformations (Yaglom), 560
geometry, 597
absolute, 183
aleph-null vs. aleph-one in, 332, 333
applied,12
elliptic, 178, 181
hyperbolic, 177-78
knots and, 239
plane, 33-113
projective, 233
pure, 12
rubber-sheet, 264
simplicity in, 558-59, 558
solid, 115-85
2
c
in, 344
see also Euclidean geometry; non-
Euclidean geometry
Geometry and the Imagination (Hilbert),
132-33
Gernsback, Hugo, 686
Gershwin, George, 633
Geschwind, Norman, 279
Gibbs, Josiah Willard, 617
Gide, Andre, 318
Gilbert, Edgar N., 18
Giles, Jack, Jr., 56
Gillman, Leonard, 347
Gillygoo, 648
Index
"Gioco filarmonico" ("Philharmonic Joke"),
631
Giuliani, Baron J., 633
Glassman, Robert B., 604
Glenn Miller Story, The, 638
gliders, 415, 415, 418
Gnostics, 606
Go, 409,481,493
mini, 477
in a planiverse, 106,107,111
God and Golem, Inc. (Wiener), 60
Godel, Escher, Bach (Hofstadter), 200-201,
660-73
Godel, Kurt, 201, 201, 575,660,662,664,
665,669
on Cantor's Continuum Hypothesis, 334,
335,342
consistency proof of, 178, 298
infinite regress and, 316-17
rotating cosmos of, 518, 520, 528
Godt, Irving, 367
Godwin, Joscelyn, 630
Goethe, Johann Wolfgang von, 88
gold, lead and, 596-97
Goldberg, Michael, 41-42,73-74,75
Goldberg Variations U. S. Bach), 664
gold bracelet paradox, 574-75
golden ratio, 77-78, 84, 88-89
"golden rectangle," 48, 67
Golomb, Solomon W., 19, 20-21, 47-52, 54,
56,66,75,402,495
Gomberg, Judy, 479
Go-Moku, 493
Good, I. J., 684
Goodman, A. W., 441, 453
Goodman, Nelson, 544, 555, 557-58
Goodstein, R. L., 458
Goodwin, W. Nelson, Jr., 560
Gordon, Cameron M., 249
Gosper, William, 357,417
Gosset, Thorold, 140-41
Gott, J. Richard, III, 108-9
"goy"I"birl" paradox, 550-51
Graham, L. A., 279
Graham, Neill, 526, 527
Graham, Ronald L., 19,207,284,445-51,
448,450,461,462
graph-coloring theory, 438-45, 439
graph theory, 437-54, 493, 606, 674
Graph Theory (Harary), 493
gravity, 137, 554
in a planiverse, 95, 109
symmetry and, 189
Greenberg, Frank, 368
Greenblatt, Richard, 684
Gregory, David, 143
Gregory, Richard, 197
Gremer, C. E., 68
Gridgeman, Norman T., 25
Grooks(Hein), 72, 72, 175,261-62,581,608
705
Grossman, Howard, 279
group theory, 215, 337,409
knots and, 239
"grue" paradox, 544, 550, 557-58
Grlinbaum, Adolph, 345
Guggenheim Museum, 119
Gummere, John F., 19
Gunn, D. M., 30
Guth, Alan, 408
Guy, M.J.T., 414
Guy, Richard K., 111, 172, 380, 401, 407
Hackenberg, Chi Chi, 482
Hackenbush, partizan, 374
Haeckel, Ernst, 640, 654
hagfish,644
Hagmann, Mark J., 687
Hajnal, Andras. 445
Hakala, Eunice, 256
Haken, Wolfgang, 682
Hales, Stephen, 132
Hales, Thomas, 135
Hall, Monty, 283-84
Halmos, Paul R., 299-300
Hamilton, John F., Jr., 550
Hand with Reflecting Globe (Escher),
220-21,220
Hanrahan, J. Edward, 407
Harary, Frank, 437, 444,451,493-95,
497-502,606
Harary ticktacktoe, 493-503, 495, 496, 498,
499
Harborth, Heiko, 24, 501
Hardy, G. H., 465-66, 665-66
Hardy-Ramanujan-Rademacher formula,
466
Harington, John, 679
Harmonices mundi (Harmony of the World)
(Kepler),216-18
Harrington, Charles W . 689-91
Harris, John S., 109-10
Hart, George L., III, 30
Hart, Johnny, 557
Hartley, Richard, 242
Hartshorne, David, 485
Hasse, Henry, 319
Hay, Thomas H., 653
Haydn, Joseph, 631
Hayes, Brian, 670, 672
Hayes, Peter, 209
Hazelett, Richard, 184
Heath, P. L., 592, 603, 608
Heath, Royal Vale, 395
Heaven and Hell (Escher), 213, 215
Heesch, Heinrich, 133, 134
Hegde, Majunath M., 245
Hegel, G.WF., 598
Hegelian philosophers, 334
Heidegger, Martin, 602
Heim, Karl, 154, 169
706
Hein, Piet, 59-66, 68-72, 398-99, 404, 406
Grooks of, 72, 72, 175, 261-62. 581, 608
Heinlein, Robert A., 169-71
Heisenberg, Werner, 398-99
helical toy, 122, 124
helical viruses, 643
helix, helices, 117-27:
in animals, 119-20, 120, 121
barber's pole problem, 124, 126-27
circular, 117-18, 118
conical, 118-19
double, 119
handedness of, 118-22, 120, 123-25, 123,
125, 126
as metaphors, 124
Helmholtz, Hermann von, 162, 627
Hempel, Carl G., 543
Henkin sentences, 664
Henle, James Marston, 107, 112
heptahexaflexagons, 392,393
HER (Hexapawn Educable Robot), 474-76,
475, 476,478
Hermes, 350
"Heterarchy of Values Determined by the
Topology of Nervous Nets, A" (Mc-
Culloch), 300
"He Who Shrank" (Hasse), 319
Hex, 60, 70, 398
hexacubes, 406
hexaflexagons, 386
hexagonal close-packing, 130-32, 130
hexagons,42,52,54,56,385,642,643,655
convex,89
regular, 46, 73
rep-4,50
hexahexaflexagons,386-88, 380 389,
390-95,391,393
hexahexafrustragon, 391
hexapawn, 473-76,474, 475, 476,478-79,
481
"Hexapawn: A Game You Play to Lose,"
481
hexominoes, 53,54-55,414
problem, 168, 173, 174
Hight, Stuart c., 476-77
Hilbert, David, 132-33, 175, 327
Hill, Robert M., 394-95
HIM (Hexapawn Instructable Matchboxes),
476
Hinduism, 621
Hinkle, Horace W, 244
Hinrichs, Ralph H., 56
Hinton, Charles Howard, 94, 95, 98, 154,
162,171-72 I
History of the New World (Benzoni), 64
Hitler, Adolf, 425
Hodgart, Matthew, 654
HofUnan, Dean, 209
Hofstadter, Douglas R., 71, 200, 576, 638,
660-73
Index
Hofstadter, Robert, 660
Hogan, Randolph, 638
Hogan, William, 66
holes, 596-97, 599
holism, 665
Holt, Philetus H., III, 67
Homfly polynomial, 248
honeycombs, 646
honeysuckle, 121-22. 123
Honeywell Incorporated, 635, 636
Honolulu Zoo, 600
Hooper, W., 637
hoop snakes, 649
Hoppe, R., 143
horizontal symmetry, 189
of words, 198
Home, R. W., 643
"Hom of the Unicorn, The" (Bonner), 119
Horowitz, I. A., 479
House. John, 478
Howarth, John L . 691-92
"How to Build a Time Machine" (Forward),
528
How to Tell the Birds from the Flowers
(Wood),193-94
Hsiang, Wu-Yi, 135
Hubbard, Elbert, 600
Hubley Toys, 70-71
Hughes, William, 692-93
"Human Organism as a Random Mecha-
nism, The" (Lacey), 277
Hume, David, 542
Hurst, Bryce, 308
Huxley, Aldous, 318, 398
Huxley, Leonard, 513
Huxley, Thomas H . 513
Huygens principle. 110
hybrid close-packing. 131
hydra game, 458-60, 460
Hymn of Creation (Rig Veda). 621
hyperbolic:
curvature, 182
geometry, 177-78
plane, 179-80, 181, 213
triangles, 180, 180
hyperchessboard, 144-46, 146, 148
hypercubes, 162-74,448-49
chess problems with, 199
elements of, 165-66, 166
generation of, 163-65, 164
into order 8-polycube problem, 171,
172-73
packing of, 143
points in, 343
properties of, 167
hyperspace, 152,534,617
hyperspheres, 134. 137-49, 152,613
packing of, 143
see also n-spheres
hypertime, 533, 538
Index
hypervolume, of spheres, 143-44
hypotheses:
confirmation and, 543-46
least plausible. 555
icosahedrons,42,640,642,643,655,658
"Identifying Non-invertible Knots" (Hart-
ley),242
Illustrated Computational Recreations in
Mathematica (Vardi), 208
impossible figures, 216
"Impossible Objects" (Penrose and Pen-
rose),216
indeterminate equations, 4
indeterminism, 580
indifference, principle of. 286. 288-90,
293
induction. inductive reasoning. 504, 506,
541-52,586
inertia, 531, 554
Inferno (Dante), 647
infinite, infinity, 61, 313-82
binary fractions, 335
cartwheel pattern, 84, 87
and higher infinities. 327-29
knots, 259
numbers, 320,336-37,596,613
regress, 315-26, 325
supertasksin, 340-49
infinite sets, 438, 457, 595, 596, 612
subsets of, 340-41
infinitesimal numbers, 371, 596
infinite subsets:
diagonal proof of, 331-32, 331
three element proof of, 329-31, 330
Infinity and the Mind (Rucker), 337
infinity machine supertask, 345, 346
infin-tiles, 56
inflation, 79-80, 80, 82. 88
information theory, 16, 352
in music, 633-34
"Information Theory and Melody" (Pinker-
ton), 633-34
Ingenious Mathematical Problems and
Methods (L. Graham), 279
In Praise of Play (Neale), 426
integers, 6, 7, 25, 328, 329, 341, 344, 345,
346,347,348,593,674
nonzero, 595
positive, 4, 23
integral fractions, 328, 335-36, 341, 595
intersection set, 610, 611
Introduction to Combinatorial Analysis (Ri-
ordan),18
Introduction to Geometry (Coxeter), 213,
263
Introduction to Higher Mathematics (C.
Reid),134
inversion, 190
Inversions (Kim), 198, 670-71
707
Inversion und ihre Anwendungen, Die
(Schmidt), 263
irrational(s) :
decimal fractions, 348
numbers, 331, 341, 595
transcendant, 332
"Irregularities in the Distributions of Finite
Sequences" (R. Graham), 461
Irving, Washington, 532
Isaacs, Rufus P., 655
"Island of Five Colors, The" (Gardner), 236
isomorphisms, 664
"Is the Null-Graph a Pointless Concept?"
(Harary and Read), 606
"It Ain't Necessarily So" (H. Putnam),
521-22
Jabberwocky (Carroll), 162
jack, 83, 84
Jacobson, Guy, 491
James, Robert C., 279
James, William, 155, 291, 317, 580, 603, 619
Jaworski, John, 505
Jeans, James, 619, 628
Jeffrey, Richard C., 550-51
Jeffreys, Harold, 289
Jensen, David, 644
Jin, Koh Chor, 210-11
John Dough and the Cherub (Baum), 647,
649
Johnson, George, 158
Joire, Paul, 687
J olson, AI, 448
Jones, Vaughan F. R., 248
Jones polynomial, 248
Jonson, Ben, 121
Jordan-curve theorem, 486
Joshua, Book of, 608
Josselin de Jong, G. de, 172
Jourdain, P.E.B., 572-73, 573,574
Jowett, Benjamin, 56
Joyce, James, 319, 485, 662, 669
Jung, Carl, 27
Jurgen (Cabell), 533, 538-39
Kabell, A., 501
Kadon Enterprises, 91
Kakeya, Soichi, 42
Kakeya needle problem, 42-43, 44
Kaleidacousticon, 632
Kant, Immanuel, 153-54, 182, 581
Kasparov, Gary, 482
Katsanis, Theodore, 405-6
Katz, Robert, 513
Kauffman, Louis, 248
Kawauchi, Akio, 242
Kay, Keith, 195
Kayles, 378
Kelly, Paul J., 560
Kelvin Hall, 69
708
Kemeny, John G., 543, 553
Kepler, Johannes, 135, 216-18, 220, 555,
649
Kepler-Poinsot polyhedrons, 216, 220
Kerrick, Walter, 670
Kessler, Gilbert W., 491
Keynes, John Maynard, 288
Kierkegaard, Soren, 602
Kim, Scott, 198-211,662,667,667,670-71
Kimberly-Clark Company, 90-91
Kindersley, David, 72
kinematics, 615
king, 84
King, Samuel E, Jr., 324
Kingsley, Charles, 513
"King That Was Not, The" (Dunsany),
537-38
Kirby, Laurie, 458
Kircher, Athanasius, 629, 630
Kirlian photography, 679
Kirnberger, Johann Philipp, 629
Kiss Precise, The (Soddy), 139-41
kites, 77-79, 78,80,87,88-89
Klarner, David, 269, 406
Klein, Felix, 177, 178,230
Klein bottles, 230-37, 230,235
glass, 230-31, 236
limericks on, 233, 236
loop problem, 233, 237
paper, 231-36,232, 233
Sprouts on, 488
Kling, Frederick R., 279
Knaff, David, 279
Knight, C. J., 502
Knight, V. Ralph, Jr., 600
knot-and-link puzzle, 245, 245, 250-51,251
knots, 239-53
algebra and, 239
alternating and nonalternating, 240, 241
amphicheiral, 240, 241,242,243
bowline, 241, 241, 250
canceling, 258-59
chain stitch trick, 246-47, 247
classification of, 243-44, 248
in closed curves, 255, 259
composite, 240-41
crossings in, 239-44, 249
figure-eight, 240, 240, 243, 247, 250
in 4-space, 249
geometry and, 239
granny,240-41,243,248,25Q-51,251
group theory and, 239
handedness of, 255,259,262
infinite, 259
inside, 256-57, 257, 261
invertible and noninvertible, 241-42
jump, 247
matrix theory and, 239
mirror images and, 240, 248
number theory and, 239
Index
outside, 256-57,257,261
overhand, 239, 255
poetry about, 249-50
pretzel,241-42
prime, 240-41, 242-43
in spiritualism, 245-46
square, 240-41,243,248,250,251
in 3-space, 249
topology of, 239-53, 645
torus, 241-42, 241,242, 246, 255,

266-67,267
trefoil, 208, 221, 239-41, 240, 243, 246-47,

in two-hole torus, 257-58, 258
Knots, Useful and Ornamental (G. Shaw),
158
Knots and Links (Rolfsen), 242-43
Knots (Escher), 221, 222
Knotted Doughnuts and Other Mathemati-
cal Entertainments (Gardner), 70,
407,587
knotted torus proof, Conway's, 259-61, 260
knot theory, 239,248-50, 380
Knuth, Donald E., 19, 72, 76, 200, 208, 302,
302,369-70,445-48,449,463-64,
594-96,638
Knuth, Jill c., 595
Koch, Helge von, 355
Koch, Tom, 638
Koch snowflake, 355-57, 356
Koehler, John E., 424
Koestler, Arthur, 207
Konorski, Jerzy, 605
Kramer, Hilton, 598
Krippner, Stanley, 686
Kruskal, Joseph B., 457
Kruskal, Martin David, 380, 504, 506, 581,
587
Kuhn, Thomas S., 543
Kyoto, Harold, 657
Lacey, Oliver 1., 277
Lambert, Henry, 538-39
Lame, Gabriel, 65, 66
Landers, Ann, 452
Landsberg, P. T., 685
landscapes:
inversion and, 190
quarter-turn, 196, 197
Langer, Joel, 247
Language of Mathematics, The (Devlin), 90
Language on Vacation (Borgmann), 26
Languages of the Brain (Pribram), 604
Lao-tzu, 597
Lapidus, I. Richard, 111
Larson, Gary, 609
Last Book of Wonder, The (Dunsany), 480
Last Recreations (Gardner), 199, 263, 380
Last Revolution, The (Dunsany), 480
Index
Laszlo, Alexander, 631
Latin cross, 150
lattice packing, 134, 135, 135
four-dimensional,145-46
of n-dimensional spheres, 147
Laves, Fritz, 133,134
Laws (Plato), 350, 365, 649
lead, gold atoms and, 596-97
learning machines, 471-84
computers as, 471-72, 480-82
matchbox game, 472-79
Leather Craft os a Hobby (Pyle), 158
leather strip braiding problem, 155, 156,
158-59, 159
Lee, Wallace, 198-99
Leech,John, 143, 147,409
Leech lattice, 147
left-right symmetry, 201, 202
Lehmer, D. H., 684
Leiber, Fritz, 480
Leibniz, Gottfried Wilhelm von, 105, 273,
380,602,605,612,618,622
Lenard, Andrew, 301, 302
length,619
Leonardo da Vinci, 677-79, 678, 685-86
"Leonardo on Bearings and Gears" (Reti),
678
Let's Make a Deal, 283
Letterform and Illusion (Kim and R.
Samuelson), 208
Leucothrix mucor, 645
Lewis, C. S., 621
Lewis, David, 622
libertarianism, 587
"Library of Babel" (Borges), 627
Lielst, Brier, 652
Life, 409-22, 412, 413, 414, 415, 416, 418,
419,420,421
problems, 414, 415, 420, 421, 422
Life and Letters of Thomas H. Huxley, The
(L. Huxley), 513
"Life in Two Dimensions," 108
light, in a planiverse, 95, 96
limericks:
on Klein bottles, 233, 236
on time travel, 519, 525-26
Linde, Andrei, 622
Lindon, James Albert, 27-29, 236, 340, 347,
525-26
linear functions, 11
linear map-fold function, 430-31
lines, 117
points in, 341-42, 341, 343
as sets of points, 347
ultraparallel, 178
zero curvature of, 598
Littlewood, J. E., 318
Litton Industries, 505
Lives of the Most Eminent Painters, Sculp-
tors and Architects (Vasari), 65
709
Lobachevski, Nikolai Ivanovitch, 177, 178
local freeze, 534-35
local isomorphism theorem, 81
Lofting, Hugh, 647
logic, 613, 665
inductive, 541-52
mathematical. 664
paradoxes, 438
and philosophy, 565-623
symbolic, 11
logical fallacies, 542
Logical Foundations of Probability (Car-
nap), 543, 545
Logic of Decision Making (Jeffrey), 550-51
Log of the Ark, The (K. Walker and
Boumphrey), 654
Lolita (Nabokov), 27
long bow ties, 78, 84, 85
Longfellow, Henry Wadsworth, 455
Lon Nol, 26
loop cuts, 227-30, 232-34
loose packing, 132-33, 134
Lorden, Garry, 453
Lorimer, George Horace, 4
"Lost" (Dunsany), 523-24
Lost World, The (Conan Doyle), 649
Lowell, James Russell, 504
Loyd,Sam,57-58,596
Lucas, Edouard, 134
Lucas, J. R., 665
Luecke, John E., 249
Lull, Ram6n, 629
Lunnon, W. F., 424, 430-31
Lynch, Gerald R., 279
Lyng, Merwin J., 347
Lyons, Howard, 600
McCulloch, Warren S., 300, 643, 687, 688
MacDonald, George, 654
MacGillavry, Caroline H., 215-16
McGovern, William, 656
McGregor, William, 674, 682-84, 692
MacHack, 684
Mack, Robert, 338
McKay, Brendon,453
McLaughlin, Sean T., 146
MacMahon, Percy A., 19
MacPaint, 208
macromolecules, 643
Madden, Mary Ann, 23
Mad Mazes (R. Abbott), 513-14
"Magic, Mystery, and Matrix" (Witten), 158
Magic Machine, The (Dewdney), 108
Magic of Numbers, The (Bell), 177
magic squares, 433
Magnificat in D (J. S. Bach), 665
Magritte, Rene, 212, 668-69, 669
Mailer, Norman, 124, 318
MainHinder, Philipp, 605-6
Malorana, C., 432
710
ManaJive (Chesterton), 603
Mandel, Mark A., 54
Mandel, Oscar, 613
Mandelbrot, Benoit, 324, 351-52, 354-57,
356, 358-59
Mandeville, D. C., 67
Manna,Zohar, 457-58, 460
"many-worlds" interpretation, in quantum
mechanics, 526-29
"Map Coloring" (Roth), 682
map-fold:
function, linear, 430-31
puzzles,423-26,424,425,430-31
"Mapping a 3-Sphere onto a Homotopy 3-
Sphere" (Bing), 257-58
map theorem, four-color, 674, 675, 682-84
marble transfer machine supertask, 345
Marcus, Mitchell P .. 279
Margalit, Avishai, 585-87
Marinoni, Augusto, 686
Mark 1, 478
MarkII,478
Markoff chain, 635
Marshall, William Rex, 208-10
"Martian Odyssey, A" (Weinbaum), 648
Marvell, Andrew, 654
Masque of Pandora, The (Longfellow), 455
mass, 619
mass-energy, 526
Mathematical Bafflers (Dunn), 24
Mathematical Basis of the Arts, The
(Schillinger),633
Mathematical Circus (Gardner), 513
Mathematical Cranks (Dudley), 337
Mathematical Gardner, The (Klarner),
269
Mathematical Magic Show (Gardner), 607
Mathematical Magpie, The (Fadiman), 8
Mathematical Models (Cundy and Rollett),
220
Mathematical Recreations and Essays
(Ball), 71, 597
mathematical zoo, 640-59
imaginary creatures in, 647-52, 654
real animals in, 640-47, 652-54
Mathematician's Loop, 247
Mathematician's Miscellany, A (Little-
wood),318
mathematics:
combinatorial. 493
nontransitivity in, 297-98
nothing in, 592-97, 606-7
pure,409,558,674
"Mathematics and Computer Science"
(Knuth), 445-48
Mathematics and Plausible Reasoning
(P61ya),14
Mathematics of Physics and Modern Engi-
neering (Redheffer and Sokolnikoft),
279
Index
mathematic zoo, imaginary creatures in,
647-52
Mathsson, Bruno, 61
Matisse, Henry, 190
matrix theory, knots and, 239
matter, 554, 597
Matthew. Gospel of, 426
Matthews, Robert. 380
Matzke, Edwin B., 132
Mavrodes, George, 550
Maze, The (Sandoz), 196
mechanics, in a planiverse, 101-6, 102,
103,104-5
Melior typeface, 71
Mellor, D. H., 451
Melody Dicer, The (Mozart), 631, 636
Melographicon (anon.), 632
membrane theory, 158
MENACE (Matchbox Educable Naughts and
Crosses Engine), 472-73
Mencken, H. L., 686
Men Like Gods (H. Wells), 150
Mental Magic (Gardner), 433
Mercer, Leigh, 27
meridians, 254, 256, 265
Merlin, 532
Merperson, Ethel, 23
metacosmos, branching worlds in, 524-25
metahistory, 169
metalanguages, 317,588
Metamagical Themas (Hofstadter), 71,670
metaverse, 622
Methodist hymnals, 638
Methodus Incrementorum (Taylor), 11
metric field, 597
Metropolitan Museum of Art, 169
Micheli, Romano, 630
Michelmore, Peter, 554
Michie, Donald, 472-73
Michie, James, 26
Midsummer Night's Dream, A (Shake-
speare), 121
Miksinski cube, 407
milk snake, 649
Mill, John Stuart, 541, 542, 550, 628
Miller, Glenn, 638
Milton, John, 647
"Mimsy Were the Borogroves" (Padgett),
162-63
Mind Machine, 687
Mind's I, The (Hofstadter and Dennett), 670
Mind Storm, 367-68
Minkowski, Hermann, 41, 518, 521
Minkowski space-time graph, 518, 521
Minsky, Marvin, 417
mirror-image knots, 240, 248
mirror reversals, 348
Mirrors in Mind (R. Gregory), 197
mirror symmetry, of words, 198, 201, 202,
206-7
Index
"Misalliance" (Flanders), 122, 123
misere Sprouts, 485, 487, 488, 489, 491
Mist (Unamuno), 317-18
Mitchell, Edward Page, 517
Mizler, Lorenz Christoph, 629
Mobius, A. F., 229
Mobius strip. 221. 228, 229, 230-34, 235,
236,255,389,392
Modern Science and Zeno's Paradoxes
(Griinbaum), 345
modular arithmetics, infinite regress in, 320
Mohammed, 532
molecular biology, 664
molecules, 96, 249
thermodynamics of, 617
Mollison, Denis P., 488
Mondrian, Piet, 190
Monograph of the Challenger Radiolaria
(Haeckel),640
Monty Hall problem, 283-84
Moon, J. W, 300-301
Moon, Tim, 444
Moore, Donald Baker, 599-600
Moore, Henry, 599
Moorean sentences, 576
Morgan, Brad, 417
Morice, David, 207
Morris, Edmund, 601-2
Morris, Scot, 201
Morse, Larry E., 600
Moser, Leo, 144,148,300-301,559,
560-62,562
Moser-Moon paradoxes, 300-301, 301
Moskowitz, Sam, 517, 522
Mottl, Felix, 351
moving-counter ticktacktoe, 493
moving-marble supertask, 347
"Moxon's Master" (Bierce), 471
Mozart, Wolfgang Amadeus, 365, 366, 367,
631-32,636,638
"Mozart's 42nd Symphony" (Cope), 638
Mozart's Musical Game, 631-32
Mr. Tompkins in Wonderland (Gamow),
684-85
M-theory, 158
multiverse, 622
Munitz, Milton K, 602
musarithmica mirifica, 629
Museum of Modern Art (MOMA), 190,608
Musha, T., 358-59
music:
"absolute," 351
brown, 352,353-54,362-64,363,367
classical, 350-51
combinatorial rules in, 628
computer, 628,634-36,664
finite nature of, 628
fractal, 350-68
information theory in, 633-34
machine-created, 627-39
711
music (continued):
"mountain," 358, 368
nature imitated by. 350-51
nothing in, 600-601
numbers in, 628
IIf, 362-65, 364, 367-68
pink, 367-68
randomizers in, 352-53, 352, 629-32, 636
reversal or inversion in, 365, 366, 367
stochastic, 354,358,361, 365
theory of, 18
white, white music, 352-53, 352, 362-64,
363,367
see also specific types of music
Musical Offering U. S. Bach), 663, 668
Musurgia universaJis sive ars magna con-
soni et dissoni (Kircher), 629, 630
MU theorem problem, 666, 672
mutually kissing spherical grapefruit and
orange problem, 141, 147-48
"My Dad Has More Money Than Yours"
game, 374
mysterium tremendum, 315
Mystery of Existence, The (Munitz), 602
MystiCism and Logic (Russell), 613
My World Line (Gamow), 518
Nabokov, Vladimir, 27,425
Naimark, George, 196
Nannosquilla decemspinosa, 653
Napier's bones, 629
Napoleon I, Emperor of France, 523
narwhals, 119, 120,645,654
Nash, Ogden, 539
Nash-Williams, C.St.J.A., 457
Nasnas,647
National Gallery of Art, 213
nature:
fractals in, 357-58
laws of, 553,554
music as imitation of, 350-51
simplicity in, 556
Nausea (Sartre), 603
nautilus, chambered, 120
Navy, U.S., 68
n-ball, ratio of n-cube to, 144
N-Card Monte, 291-93
n-cubes, 143
packing into n-spheres of, 144, 145,
146-47
ratio of n-ball to, 144
Neale, Robert Edward, 426-27, 426,
429-30,429
necklace problem, 15-16,16,17-19,20-21
Nemerov, Howard, 212-13
Nesbit, E., 315
neural networks, 481
neutrons, 323
Neuwirth, Lee, 239
"New Accelerator, The" (H. Wells), 532-33
712
New Ambidextrous Universe, The (Gard-
ner), 157-58
Newcomb. Simon, 581
Newcomb, William A., 519-20. 581, 587, 588
Newcomb's paradox, 580-91, 584
"Newcomb's Paradox Revisited" (Bar-Hillel
and Margalit), 585
"Newcomb's Problem and 'IWo Principles of
Choice" (Nozick), 581-82
New Elements of Mathematics (Peirce), 250
New Eleusis, 505-12, 507, 512, 513
Newell, Peter, 190-91
New Era of Thought, A (Hinton), 163
Newman, J. Robert, 505
Newton, Isaac, 13-14, 143, 178, 380,531.
554
Newtonian physics, 527
Newtonian theory. 108
Newton's formula, 13-14, 16-17, 20
New York Public Library, 685-86
Nicholas, Albert, 391
Nielsen, Gene, 432
Nile. 355, 362
Nim, 371,374,376-78,477,485
Nimbi, 70, 71
NIMBLE (Nim Box Logic Engine), 477
9-space, 144
packing in, 134, 143
Nine Unlikely Tales (Nesbit), 315
1974, six major discoveries in, 674-94
1984 (Orwell). 535
Nixon, Richard M., 27,433
noise:
brown, 353, 354, 357, 358, 359-60, 360,
362
Johnson, 352
IIf, 354-55, 359-63, 360
randomizers and. 360, 361-62, 361
scaling, 352-55
thermal, 352
white, 352. 354, 358-60, 360, 362
nonbeing, being vs., 598
non-Euclidean geometry, 175-85,213, 335,
337,664
language in, 685
space-time in, 617
"Non-Euclidean Symmetry of Escher's Pic-
ture 'Circle Limit ill,' The" (Coxeter),
181
"Non-invertible Knots Exist" (Trotter), 241
nonperiodic tiling, 73-77, 74, 76, 77-91,
77, 87,87,88
nontransitivity in games, 297
bingo cards, 302, 302

top, 301, 302
Norton, Simon, 414, 415
"Notebook, The" (Mailer), 318
nothing, 592-609, 610
in art, 598-600, 599
Index
in books, 600
as event, 601-2
in film, 600
in mathematics, 592-97, 606-7
in music, 600-601
in philosophy, 593, 598, 607, 608-9
psychology and, 594, 601, 604
something vs., 602-3
Nothing Book, The, 600
nothingness, temporal, 601, 604-5
Nowakowski, Richard, 172
Nozick, Robert, 581-87
n-space, 137
Soddy's formula in, 140-41
n-spheres, 137-41, 143, 147
axis of, 138
cross sections of, 138
mutually touching, 139-40, 139
packing into n-cubes of, 144, 145, 146-47
projection of, 138
volume of, 144
n-tuplets, 661
nontransitivity and, 305-10, 306, 308
nucleic acid, 119
null graph, 606, 606, 608
null sets, 340, 592-95, 596, 600, 607, 610,
612
numbers:
in art, 627
Betti, 227-29, 231, 235
chromatic, 227-29, 231, 233, 234, 235
construction of, 369-71
counting, 341, 348, 592-94
Farey, 329
Fibonacci, 121
figurate, 128-29
infinite, 320, 330, 336-37, 596, 613
infinitesimal, 371, 596
inversion problems, 195, 197
irrational, 331, 341, 595
in music, 628
palindromic primes, 25, 31
poetry about, 627
prime, see prime numbers
Ramsey, see Ramsey numbers
rational, 328, 331, 341
real, 331, 335, 341, 344, 346, 348
square, see square numbers
surreal, 369-82
transfinite, 320, 327, 334-37, 341-44,
346,371,457,596,612
triangular, 134
twofold symmetry of, 194-95
number theory, 239,409, 674-75,684
n-volume, of spheres, 143-44
Oakley, C. 0., 392
"Oases of Silence in a Desert of Din" (E.
Morris), 601-2
O'Beirne, Thomas H., 50, 574, 631, 632, 635
Index
O'Brien, Dennis T., 16
O'Brien, Fitz-James, 319
Occam's razor, 529, 622
O'Connor, Donald John, 568-69
octagons, 42
rep-16,53
octahedrons, 42,640, 642,657,658
octahexaflexagons, 392
Octapawn, 479
Odlyzko, A. M., 308
jOHO! (Whistler), 191, 191
Oliver, Revilo P., 26
Omar Khayyam, 539
On an Inversion of Ideas as to the Structure
of the Universe (Reynolds), 597
"On a Problem of Formal Logic" (M. Ram-
sey),438
On Divine Omnipotence (Damian), 536-37
1 If:
dice algorithm, 360-62, 361
music, 362-65, 364, 367-68
noise, 354-55, 359-63, 360
spectral densities, 358-59
l-space,141
l-sphere, 137, 138
one-valued functions, 344
One Hundred Problems in Elemental}'
Mathematics (Steinhaus), 461
O'Neil, Sharon, 247
On Growth and Form (D. Thompson), 119,
643,654
On Numbers and Games (Conway), 369,
594
"On Round Pegs in Square Holes and
Square Pegs in Round Holes"
(Singmaster), 144
"On Sets of Aquaintances and Strangers at
Any Party" (A. Goodman), 441
"On the Application of Dynamic Program-
ming to the Determination of Opti-
mal Play in Chess and Checkers"
(Bellman), 480
"On What There Is" (Quine), 610
"Oom" (Gardner), 602, 605
operationalist philosophers, 336
Oppenheimer, J. Robert, 581
optical illusions, 195
upside-down, 192, 193
orange pyramid puzzle, 130,135
orbits:
planetary, 555
in a planiverse, 97
Order and Chaos (Escher), 216-20, 219
order of moribundity, in Sprouts, 489
origami, 426, 429
Ork,649
orthogonal shapes, 59
Orwell, George, 535
OSBERG-BORGES problem, 425, 433-34
oscillating theory, 334
713
oscilloscope, 628
Ostrander, S., 686
Othello, 111
Other Inquisitions (Borges), 605-6
Otto, Rudolf, 315
Ouspensky, Peter D., 154, 169
ovals, 60, 61
Oxford Companion to Music, The (Owens),
632
Ozma of Oz (Baum), 649
Pace Gallery, 598
packing:
of circles, 141, 142
close, 130-32, 130
in 4-space, 134, 143
of hypercubes, 143
of hyperspheres, 143
loose, 132-33, 134
of n-spheres into n-cubes, 144, 145,
146-47
of unit spheres, 141-44, 142
Padgett, Lewis, 162-63
Pail of Air, A (Leiber), 480
Pale Fire (Nabokov), 27
Paley, William, 316
palindromes, 23-32
defined,23
imaginary beasts as, 647, 648
language, 26-30
mathematical, 23-26, 31
musical, 366, 367
poems, 28-29
sentence, 26-28
two-digit, 24
visual,23
word,206
Palindromes and Anagrams (Bergerson),
28-29
"Palindromes by Addition in Base Two"
(Brousseau),24
"Palindromes by Addition" (Trigg), 24
palindromic:
gap, 29
powers, 25
prime numbers, 25, 31
sums, 23-24
Palter, Robert M., 182-83
Pamir sheep, 120
Panama, 653
pancake-cutting problem, 11-15, 12, 17
pangolin, 653,653
paper:
folding of, 423-36
topological properties of, 227
"Paper Fold Prediction" (Gardner), 432-33
Para Bruin, 649
Paradise Lost (Milton), 647
paradoxes:
infinity and, 344
714
logical, 438
nontransitive, 286-311
prediction, 581, 585
of the raven, 543-44
relativity, 684-85
semantic, 588-89
in time travel, 520-24
see also specific paradoxes
"Paradox Papers" (Gardner), 431-32
parallelograms, 47-48, 47, 54, 61
rep-4, 49, 50
parallel postulate, 175-78, 181-82, 183,
184,335
parallels, 254, 256
parapsychology, 154,679
Paris, Jeff, 458
parity, 262
Parker Brothers, 406
Parmenides, 609
Parmenides (Plato), 316
Parris, Richard L., 242, 250
Parsons, Denys, 636-37
participator-observers, 618-19
particle physics, 157,317
partition theory, 462-66
partizan Hackenbush, 374
Pascal, Blaise, 128, 290, 291, 293, 613
Pascal's triangle, 561
Pascal's wager, 290
past:
altering of, 531, 535-39
in universe of discourse, 612
Patashnik, Oren, 19, 207
Patchwork Girl ofOz (Baum), 647-48,
648
Paterson, Michael Stewart, 485, 488
Patterns, 513
Pauling, Linus, 119
Pavalita, Robert, 686
Paz y Remolar, Ram6n, 686
Peano:
curves, 355-57, 356
postulates, 664
Peano, Giuseppe, 355
Pedoe, Daniel, 231
Peirce, Charles Sanders, 250, 273, 290, 340,
555,618,622
positive rationals proof of, 328-29
Penney, Walter, 303, 308
Penrose,L. S., 77, 216
Penrose, Roger, 73, 77, 79-81, 88, 216,
246-47
Penrose rhombuses, 88-89, 88
Penrose staircase, 77
Penrose tiles, 73, 77-91, 380
Penrose Tiles to Trapdoor Ciphers (Gard-
ner), 89-90, 263, 324,355, 380,452
Penrose universes, 81-86
Pensari, 513
Pensari Guide Book (Katz), 513
Index
Pensees (Pascal), 290
pentacubes, 406
pentagonalsynunehy,81,82
pentagons,42,54, 642, 655
convex,89
rep-4,50
pentagram, 88, 218-19
pentahexaflexagons, 389, 392, 393
pentominoes, 406, 412,414,418
Pepys, Samuel, 629
Pepys Museum, 629
perimeters, 40, 41
periodic:
tessellation, 74,81
tiling, 73
periodic table, in a planiverse, 98-99, 98
Perry, Ralph Barton, 220
Petrosian, Tigran, 480-81
phase space, 617
phenomenology, 221
phi, 78
Philosophical Explanations (Nozick), 587
philosophy:
logic and, 565-623
nothing in, 593, 598,607,608-9
Philosophy of Space and Time, The (Re-
ichenbach), 521
Phutball, 111
phyllotaxy, 121
Physical Phenomena of Spiritualism,
Fraudulent and Genuine (Carring-
ton), 246
Physicist and Christian (Pollard), 154
physicists, 347
n-space and, 157-58
"Physicists Finally Find a Way to Test Su-
perstring Theory" (Johnson), 158
physics, 12, 13-14,355, 515-64, 614
knot theory and, 248-49
mathematical,324
Newtonian, 527
planar, 111
simplicity and, 553
theoretical,529
Physics of Immorality, The (Tipler), 528
pi, 67,82, 143, 178,331, 337, 345,557,
675
pictures:
quarter-turn, 191-92, 193, 195-97, 196
upside-down, 190-92, 191, 192, 195,
196-97
Piel, Anthony, 309
Piel, Gerry, 395
Pierce, J. R, 633
Pike's Peak, 601
Pinkerton, Richard C., 633-34
Pirandello, Luigi, 317
planiverse, 94-113
Planiverse, The (Dewdney), 108
plants, helices in, 120-22, 123
Index
Plato, Platonism, 56, 143, 151, 153, 157,
291,316,335,336,337,350-51,365,
649,662,669
Platonic dualism, 143
Platonic solids, 216, 218, 640, 657, 658
Poe, Edgar Allan, 27, 620-21, 622
Poet and the Lunatics, The (Chesterton),
190
poetry:
fantasy animals in, 651-52
knots and, 249-50
mathematical, 348, 608, 684
metaphysical, 620-21, 622
numbers in, 627
palindromic, 28-29
probability theory and, 282-83
science in, 608
synunetrical, 206-7
see also limericks
Poincare, Henri, 162, 179, 182, 184,213,
327
Point Counter Point (A. Huxley), 318
poker, 297
political cartoons, 190
Pollard, William G., 154
P6lya, George, 14
polycubes, 169, 171, 172,406,407
snakes problem, 203-5, 205, 209-10, 210
polygonal tiles, 87
polygons, 41,74-75, 75, 76, 219
irregular, 46
regular, 46
replicating, 47-48
"stellated," 50, 51, 52, 54, 56, 57
polyhedrons,41,219,406,646,658
regular, 216
polyhexes, 501
polyiamonds, 501
polymers, 249
polynomials, 18
Alexander, 243, 248
in knot classification, 243-44, 248
polyominoes, 56,402,431,494-95
Polyominoes (Golomb), 56
polyphony, 629
Pomeroy, John, 194
Po of Bird, 614
Pope, Alexander, 150, 647
Popper, Karl, 509, 543
Popular Scientific Lectures (Helmholtz),
627
"Portrait of Its Immanence the Absolute,"
593,598
positrons, 597
power spectrum, 351
pragmatic philosophers, 336
predestination, 536
predictors, 585, 587, 588-89
Pribram, Karl, 604
Priestley, J. B., 523, 535
715
Prim, 378
prime factors, 19
Prime Nim, 378
prime numbers:
infinite set of, 327-28
palindromic, 25, 31
Prince Rupert's problem, 172
Princess and Curdie, The (MacDonald), 654
prisoner's dilemma paradox, 581
"Prisoner's Paradox Revisited, The" (Bedi-
ent), 282-83
prison windows problem, 425-26, 425, 434
probability, 271-311, 545-51
ambiguity and, 273-85
nontransitive paradoxes, 286-96
"Problem of Chess Queens, A" (P. Hayes),
209
"Problem of Infinite Matter in Steady-State
Cosmology. The" (Schlegel), 334
"Problem of Packing a Number of Equal
Nonoverlapping Circles on a Sphere.
The" (Coxeter), 141-43
projection, of n-spheres, 138
projective geometry, 233
projective plane, 233, 235
Prokofiev, Sergei, 368
prolate spheroid, 64
propellers, 649
pseudosphere, 181
psionic generator, 687
Psychic Discoveries behind the Iron Curtain
(Ostrander and Schroeder), 686
psychic motor, 679-82, 680, 686-94, 688
psychology, nothing and, 594,601,604
Pushcar, Paul, 652
Pushmi-Pullyu, 647
Putnam, George P., 620
Putnam, Hilary, 521-22, 524
Puzzle Math (Gamow and M. Stern), 569
Puzzles and Curious Problems (Dudeney),
17-18
Puzzles and Paradoxes (O'Beirne), 632
Puzzle Workout (Kim), 208
Pyle, Clifford, 158
pyramids, 129-30
square, 129-30, 131
tetrahedral, 130, 131, 135
Pythagoreans,88,129
quadracubes, 405
quadratic formulas, 10-11, 12
quadratic functions, 11-12, 13
quadrilaterals, 73
arbitrary convex, 558-59, 558, 559-60,
561
irregular, 46
rep-4,50
Quadrille Melodist, 633
quantum field theory, 248
quantum fluctuations, 597
716
quantum Hall effect, 111
quantum mechanics, 531, 581, 617, 620
Copenhagen interpretation of, 526
"many-worlds interpretation" of, 526-29
religion and, 618-21
space-time in, 518
"Quantum Mechanics and Reality" (De-
Witt), 527
quantum uncertainties, 617
quarks, 323
quarter-turn pictures, 191-92, 193,195-97,
196
quasars (quasi-stellar radio sources), 171
quasicrystals, 89-90
queen, 83, 84
Queen Zixi of Ix (Baum), 649
"Quick Trip through Knot Theory, A" (Fox),
259
Quimby, Shirley, 293-94
Quine, Willard Van Orman, 567, 574, 575,
610,671-72
Rab, George T., 158-59, 159
Rademacher, Hans A., 40, 466, 643
radii, of mutually touching n-spheres,
139-40,139
radiolaria, 640-42,641,642,643,654-55
radius problem, 139, 147
Radziszowski, Stanislaw, 452-53
Ramanujan, Srinivasa, 466, 665-66, 674,
684
ramified theory of types, 438
Ramsey:
games,441,442,443,444,451,494,495
problems, 448
sentence, 438
theory,437-54,494,500
Ramsey, A. S., 437
Ramsey, FrankPlumpton, 437, 451
Ramsey, Michael, 438
Ramsey graph, 440-42, 445-48
games, 495
puzzle, 442, 453, 453
theory, 450
Ramsey numbers, 437, 440, 450, 451-52, 494
algorithms for, 442
Burr's chart for, 444-45, 446-47
classical, 443
generalized, 444-45,451
nontrivial classical, 442-43
Ramsey Theory (R. Graham, Rothschild,
and Spencer), 450
Randi, James, 682
Randlett, Samuel, 430
random chord problem, 275-77, 276
randomizers, randomizing:
in music composition, 629-32, 636
noise and, 360, 361-62, 361
procedure, ambiguity and, 277
in white and brown music, 352-53, 352
Index
Rational Recreations (Hooper), 637
RAT (Relentless Autolearning Tyrant), 478
Rauschenberg, Robert, 599
Ravielli, Anthony, 685
Rayles, 378
Read. Ronald C., 606
real functions, 344
real numbers, 331, 335, 341,344,346,348
rectangles, 59, 60, 61
"golden," 48, 67
imaginary creatures, 647-48
rep-2, 48, 48
recursive formulas, 19
Redheffer, Raymond M., 279
reductionism, 665
reflection, symmetry and, 189-90
Reflections of a Physicist (Bridgman),
336-37,347
Regener, Eric, 84
Regular Polytopes (Coxeter), 220
Reichenbach, Hans, 521-22, 524, 542
Reid, Constance, 134
Reid, Robert. 54-55
Reinhardt, Ad. 598-99, 599, 608
" 'Relative State' Formulation of Quantum
Mechanics" (Everett), 524
relativity, theory of, 162, 184, 518, 519, 524,
554,614,619,676-77,677,684-85
and time, 531-32
relevance confirmation, 545
religion, 598, 620
andGod,602,606,613
infinite numbers and, 344
multiple universes and, 618-21
see also Christianity
Rennie, Ian F., 654
rep-2 polygons, 47-48, 47
rep-3 figures, 48, 56, 57
rep-4 figures, 48-50, 49, 51, 52, 54
rep-5 figures, 48, 56, 57
rep-9 figures, 50, 51, 52, 56
rep-16 figures, 53, 53
rep-36 figures, 52-53, 53, 54, 55
rep-81 figures, 55, 55
rep-144 polygons, 53-54, 54
replicating figures, 47-58
replicating polygons, 47-58
rep-tiles, 46-58, 75
Reptiles (Escher), 213, 214, 218
Republic (Plato), 151, 365
repunits, 25
Rescher, Nicholas, 582
Reti, Ladislao, 678
Reuleaux, Franz, 36
Reuleaux triangle, 36-37, 36,40-41,43
Reuleaux-triangle drills, 36-37,37,43
Reversi,l11
Reyburn, Wallace, 686
Reynolds, Osborne, 597
Rhine, Joseph B., 332, 687
Index
Rhoma, 407-8
rhombuses, 77-78
Penrose,88-89,88
ricercar, 668, 669
Richards, John, 246
Richardson, George P., 491
Ride of the Valkyries (Wagner), 351
Riemann, Georg Friedrich Bernhard, 178,
182
Riemann curvature tensor, 109
right angles, 59
right triangles, isosceles, 47, 47, 73
Rig Veda, 621
Rims, 377-78, 378
Rindler, W., 685
Riordan, John, 18, 19
"Rip Van Winkle" (Irving), 520, 532
Roberts, J. A., 444
Robinson, Gerald, 67
Robinson, Mrs. R. M., 406
Robinson, Raphael M., 76, 77
robots, 471, 473, 474, 479
rock, paper, scissors, 297
Rogers, Robert G., 653-54
Rolfsen, Dale, 242-43, 262-63
Rollett, A. P., 220
Roly-Rogues, 649
Rosenberg, Harold, 598
Rosenbloom, Paul C., 178
Rosquist, Kjell, 208
Ross, John, 279
"Rotating Cylinders and the Possibility of
Global Causality Violation" (Tipler),
528
rotational symmetry, 189
rotations and reflections, 189-97
Roth, Norman K, 682
Rothschild, Bruce L., 449
Rothstein, Edward, 670
rotors, 41-42, 42
Rowland, Vernon, 604-5
RoyalCopenhagen,72
Royce, Josiah, 344
Rubaiyat (Omar Khayyam), 539
rubber bands:
knotting problem, 155-56, 156,159-60,
159, 160
linked,246
Rucker, Rudy, 308, 337
Rudiments of Ramsey Theory (R. Graham),
450
Russell, Bertrand, 175, 182, 184, 317, 327,
438,542,588,592,613
Russell's paradox, 613
Rust Engineering Company, 391-92
Ryan, Ralph J., III, 490-91
Saal, Harry J., 24
Saccheri, Giralamo, 177
Sachs, David, 308
717
Sackson, Sid, 513
"Sacred Beetle, The" (Fabre), 647
Saint Matthew Passion U. S. Bach], 351
Sallows, Lee, 30
Salmon, Wesley C., 345, 607
Samuel, Arthur L., 471
Samuelson. Paul A., 298
Samuelson, Robin, 208
Sandiford, Peter J., 478-79
Sandoz, Maurice Yves, 196
Santa Maria del Fiore, 65
Sartre, Jean-Paul, 603
satisfaction:
average, 550
expectation of, 549-50
Sauve, Leopold,441
Savant, Marilyn vos, 283-84
S a w y e ~ W. W., 10
scaling:
fluctuations, 355
fractals, 355
noises, 352-55
Scantion-Princeton, 71
scarab, 646-47
Scarecrow of Oz, The (Baum), 649
Scarne, John, 493
Schaffer, Karl, 208, 211
Schattschneider, Doris, 224
Scherer, Karl, 244
Schillinger, Joseph, 633, 638
Schillinger System of Musical Composition,
The (Dowling and A. Shaw), 633
Schlafli, Ludwig, 178, 219
Schlegel, Richard, 332-34, 337
Schlesinger, G., 535
Schmidt, Hermann, 263
Schmitt, Peter, 90
Schoenberg, I. J., 43
Schofield, A. T., 154
Scholasticism, 532, 536
Schopenhauer, Arthur, 602, 606
Schor, J. Paul, 630
Schorpp, Charles B., 16
Schroeder, L., 686
Schultz, Philip, 652
Schwartz, Abe, 111
Schweppes Building, 71
Science and Music (Jeans), 628
science fiction:
relativity of time in, 532-22
time travel in, 525, 535
Science Fiction Puzzle Tales (Gardner), 263
Scientific American Book of Mathematical
Puzzles & Diversions (Gardner), 16,
70,377,398,477
"Scientific Dreamers' Worldwide Cult," lOB
scientific method, 504
Scott, George D., 131
Scott Joplin Melody Dicer, The, 636
screws, helices in, 118, 122
718
Scriven, Michael, 567, 569,574
Second Miscellany of Puzzles (Barr), 255
Second Scientific American Book of Mathe-
matical Puzzles and Diversions, The
(Gardner), 70, 395
"Secret Miracle, The" (Borges), 533
Sedlak, Francis, 171
Segal, Harold A., 538
Segre, Beniamino, 683-84
self-similar fractals, 56
"Self-Similar Sets 5" (Bandt), 56
SergeI's Square, 60
sets:
cardinal number of, 612-13
finite, 327-28
infinite, 327-38, 612
set of all possible, 612-13
set theory, 327, 334, 340, 342,347,613-12
Cantorian, 336-37, 346, 347
non-Cantorian, 335, 342
Zermelo-Fraenkel,613
severed head trick, 193-94
"Sex Bias in Graduate Admissions" (Bickel,
Hammel, and O'Connell), 548
Shakespeare, William, 121, 538
Shannon, Claude E., 378
Sharp, R., 279
Shaw, A., 633
Shaw, George Russell, 158
Sheep and Goats problem, 429, 429,
434-36,435
shells, of mollusks, 120, 121
Shepperd, J.A.H., 158
Shoemaker, Sydney S., 534-35, 538
short bow ties, 78, 84, 85
Shortz, Will, 609
Shulman, Polly, 380
Siberut, 653
sides, as topological property, 227-30, 223,
234,235
"Silver Blaze" (Conan Doyle), 601
Silver Dollar Game without the Dollar.
376-77,376
Silver Dollar Game with the Dollar, 376,
377
Silverman, David L., 308, 309
Sim, 441, 442, 444
Sime, Sidney H., 651-52
Simmons, Gustavus J., 25, 441
simplicity, 553-64
complexity vs., 553-56, 562
data vs., 556
in geometry, 558-59, 558
measurement of, 556-57, 562-63
in nature, 556
in physics, 553
and pi, 557
pragmatic factors in, 555
SIMP (Simplified Integrated Modular
Prose), 635, 636-37
Index
Simpson, E. H, 547, 547,548
Simpson's paradox, 546-48, 547
Singmaster, David, 8, 144
singularity, 614
Sistine Chapel, 601
Sisupalavadha, 30
Sites, Richard L, 479
Six Characters in Search of an Author (Pi-
randelIo), 317
6-space,143-44
Sixth Book of Mathematical Games from
Scientific American (Gardner), 26
"64-Square Madhouse, The" (Leiber), 480
Skinner, B. F., 580
Slade, Arthur (fictional character), 150-60
Slade, Garth, 654
Slade, Henry, 155, 156, 157,245,246
Slaughterhouse-Five (Vonnegut), 518
Slaughter on Tenth Avenue, 351
Sleator, Danny, 491
sliding-block puzzles, 71, 596
Sloane, N.J.A., 147
Slocum, Jerry, 407
Smalley, Richard, 657
small steIlated dodecahedron, 216-20
Smith, W.H.B., 109
Smith, W. Whately (Whately Carington),
154
Smullyan, Raymond M., 455-58, 460, 609,
613
Smullyan's ball game, 455-58, 459, 460
snails, 120
snakes:
hoop, 649
sidewinding. 646
Snakes of Hawaii (V. Knight), 600
Snort, 375-76, 376
Snow, C. P., 65
snowflake curves, 321-23, 322, 324,
355-57,356
problem, 321-22. 322
soccerball, tessellation of, 655
Social Choice and Individual Values
(Arrow), 297
Socrates, 365
Soddy, Frederick, 139-41
Sokolnikoff, Ivan S., 279
Solidac, 632
solids:
of constant width, 40-41. 41
convex, 90
Platonic, 216,218,640,657,658
semi-regular, 641
"Solution of the Four-Color-Map Problem,
The" (Haken and Appel). 682
Soma, 60, 70. 71, 398-408, 399, 400, 401,
402,404,405
impossibility problem. 402, 402, 408
impossibility proof for, 402-3, 402, 403
spatial imagination and. 402
Index
Somap, 407
Son of a Giant Sea Tortoise (Madden), 23
Sorensen. Roy, 576
"Sorrow of Search, The" (Duns any) , 319-20
"Sound of Thunder, A" (Bradbury), 523
"South Wind, The" (Duns any), 317
Space Child's Mother Goose, The (Winsor).
227
spaceships,415,420,421
space-time, 152,162,171,247,323,517-18
black holes and pathologies of, 528
graphs,614-17,615
non-Euclidean, 617
in quantum mechanics, 518
space-time continuum, in a planiverse, 97
"Spacetime Donuts" (Rucker), 308
Spacetime Physics (E. Taylor and Wheeler),
684
spectral densities, 351, 354
llf,358
spectrum analysis, 155
Spencer, Jeffrey, 208
Spencer, J. H., 444
spheres, 64, 221,223, 235,613,640,642,
643,650
Cartesian formula of, 137
hyperspheres and, 137-49
packing, 128-36
packing of, 141-44, 142
tessellation of, 655-56
as 3-sphere, 137
topological properties of, 228
spherical creatures, imaginary, 649
spheroids, 64
75, 76
spiders, 652
Spinfinity, 122
spiritualism. 154, 155. 157, 245-46
spokes, 85
sporadic groups, 380, 409
Sprouts, 485-92
five-spot, 488
four-spot, 488
FTOZOM (fundamental theorem of zero-
order moribundity) in, 489
misere, 485, 487, 488, 489, 491
one-spot, 487
order of moribundity in, 489
six-spot, 488
three-spot, 486-87,486,488
topology of, 485, 487
two-spot, 487-88, 487
squared circles, 63
square knot, 240-41, 243, 248, 250,251
with six alternating crossings problem,
241,250,251
square lattice, 135, 135
square numbers, 128-29, 128, 129, 134
palindromes in, 25, 26
square pyramids, 129-30, 131
719
squares,46,59, 235
points in, 342-43, 343
unfolding of, 168. 169
Stadler. Maximilian, 629-31
Stampolis, Peter. 111
starfish. 85. 86
stargazer, 645
star pattern, 82, 83, 85
star polygons, 38, 39
statistical theory, 390
Statter, Arthur C., 654-55
steady-state theory, 332-34, 337.614
steel sphere packing problem, 136
Stein, Gordon. 157
Steiner, Rudolf, 649
Steinhaus. Hugo. 461
stellated dodecahedron, small. 216-20
stellated polygons, 50, 51, 52. 54. 56, 57
Stephenson, Charles W., 406
steriverse.95
Stem, Erik, 208, 211
Stem, Marvin, 569
Sterne, Laurence, 600
Stevens, Wallace, 366
Stewart, Ian, 183, 211
sthenometer, 687
Stillwell, John, 262, 263
Stockholm. 60-61, 62, 67
Stone, Arthur H" 385-86. 389. 395
Story of Psychic Science, The (Carrington),
686-87
Stover, Mel, 159. 160, 294
straight rule, 551
Strange Loop Gazette (Kim), 670
Strauss, Richard, 351
Structure and Evolution of the Universe,
The (Whitrow), 99
Studies in Words (C. Lewis), 621
subellipses.61,63
subgraphs,438,451
Sub-Mechanics of the Universe, The
(Reynolds), 597
subsets, 340
cardinal number of. 612-13
of real numbers. 344
succotash, 491
Summa Theologica (Thomas Aquinas),
536
sum of two squares problem. 666, 672
sun pattern, 82, 83, 85
supercircles, 63, 64, 67
supereggs, 63,65,67-70, 69, 71
superellipses, 59-72. 62, 70, 72
in furniture design, 61-63
in urban planning and architecture,
60-61,66-67.71
superellipsoids,64
superpoints. 617, 618
superpredictors. 588
superspace, 597,614,617-18
720
superspheres, 64, 67
157-58, 248, 323
supertasks, 34Q-49
infinity machine, 345, 346
marble transfer machine, 345
moving-marble, 347
Thompson lamp, 344-45
superultimate question, 602-3
surreal numbers, 369-82
Surreal Numbers (Knuth). 369-70, 594-96
Swann. Donald. 122
Swedish Broadcasting Company, 575
Swift, Jonathan, 47
switching theory, 18
symbolic logic, 11
Symbolism and Belief (Bevan), 536
symmetry,187-224
Escher and. 215
types of, 189
in words, 198-99. 198, 200-203, 200,
201,202,206-7.206
Symmetry Aspects of M. C. Escher's Peri-
odic Drawings (MacGillavry). 215-16
symmetry operations, 189, 560
"Symmetry Types of Periodic Sequences"
(Gilbert and Riordan), 18
symphonies, 628
Symposium of Two-Dimensional Science
and Technology, A (Dewdney). 108
System Development Corporation. 504-5
System of Logic. A (Mill). 541
"Tachyonic Antitelephone. The" (Benford.
Book, and W. Newcomb), 519-20
tachyons, 519-20
Tac-Tix, 70, 398
Tait, Peter Guthrie, 248
tangents, 39-40
tangrams, 398
Tanner, L. R. 477-48
Tao Te Ching (Lao-tzu), 597
Tarski, Alfred. 316-17,665
ta-ta.651-52
uTa-Ta, The" (Sime), 652
Taylor, Brook, 11
Taylor, Edwin E, 684
Taylor. Herbert, 269
Taylor, R, 599
Taylor theorem, 11
Teeko, 493
temperature, Conway's theory of, 379
temporal nothingness, 601, 604-5
Temptation of St. Anthony (Flaubert), 647
Tenaza, Richard R, 652-53
tennis,60B
tessellations, 179,213,215,222, 332, 333,
667
honeycombs and. 646
periodic, 74,81
of the plane. 213
Index
of soccerball, 655
of spheres, 655
tesseracts, 162-68,170-71,173
cube problem, 168, 172, 173
generation of, 163-65, 164
3-space projection of, 167, 167
unfolding of, 168-69
see also 4-cube
tetraflexagons, 396
problem, 427-29, 427, 428, 434
tetrahedrons, tetrahedral, 41, 42, 208, 440,
442,640,641,642,658,671
numbers, 129-30, 136
pyramids, 130, 131, 135
tetrahexaflexagons, 389,392,393
tetrominoes, 412, 413, 497
Thales of Miletus, 176
THEM (Two-way Hexapawn Educable Ma-
chines), 478
theology, hyperspace and, 152-54
theorems:
on curves of constant width, 40
nontheorems vs., 672-73
simplicity and, 558-59, 558
"Theory of Knots, The" (Neuwirth), 239
Theory of Probability Ueffreys), 289
Theory of the Mechanism of Survival, A (W.
W. Smith), 154
theosophy, 649
thermodynamics, 526
of molecules, 617
Thinking Machines (Conklin), 471
Thomas, Gnostic Gospel of, 153
Thompson, D'Arcy Wentworth, 119,643,
646,654
Thompson, Perronet, 181
Thomson, James F . 344-45
Thomson lamp supertask, 344-45
Thoreau, Henry David, 190, 553-54
Thorp, Edward, 481
Thorvaldsen, BerteL 69
3-cube, 143, 172
three-dimensional space, 94, 451, 597
see also 3-space
three prisoner paradox problem, 277-78,
280-84
"Three Sailors' Gambit, The" (Dunsany),
480
3-space, 134, 137, 138, 151-54, 155, 157,
162,164,167,169,171,178,213,
216,255,257,261,318,518,612,
613,617
Klein bottles in, 230-31
knots in, 249
tiling of, 205
universes, 617
3-sphere, 137-38
Three Spheres (Escher), 221, 223, 223
Through the Looking Glass (Carroll),
318-19,537,596,608
Index
Thurber, James, 429
ticktacktoe, 471-72,479, 480, 485, 493,
494,494,495,501,502
generalized,493
Harary (animal), 493-503, 495, 496, 498,
499
moving-counter, 493
Tierney, John, 284
Tietze, Heinrich, 249
tiling, 73-93
Euclidean, 181
of floors, 46
Harary ticktacktoe and, 499
of hyperbolic plane, 181
nonperiodic, 73-77, 74, 76,77-91,77,87,
87,88
periodic, 73
polygonal, 76
of 3-space, 205
TImaeus (Plato), 56
time, 169,619
absolute, 531
causality and, 586
quantized, 539
relativity of, 531-33
stopping of, 531, 532-35
Time and the Gods (Dunsany), 537-38
TIme Machine, The (H. Wells), 517, 518,
529
time travel, 517-30,535,664
branching time paths and, 522-24
paradoxes in, 520-24
in science fiction, 518-21, 522-24
thermodynamic problem with, 526
and traveler's position in space-time,
526
TIme Travel and Other Mathematical Bewil-
derments (Gardner), 73
"Time Traveller's Story, The" (H. Wells),
517
"Time Without Change" (Shoemaker),
534-35
TIny Alice (Albee), 315
Tipler, Frank, 528
Titzling, Otto, 686
Toeplitz, Otto, 40, 643
"To His Coy Mistress" (Marvell), 654
top, nontransitive, 301, 302
390,493,
671
of Aulonia hexagona, 642-43, 657-58
embedded properties in, 262
intrinsic properties in, 262
of Klein bottles, 230-37
of knots, 239-53, 645
properties in, 227-30
of Sprouts, 485, 487
of to ruses, 229-30, 229,254-55
Topology of 3-Manifolds and Related Topics
(Fort),259
721
Topology Seminar, Wisconsin, 1965 (Bing
and Bean), 257
"Topology" (Tucker and Bailey), 255
Topsys & Turvys (Newell), 190-91
Topsys & Turvys Number 2 (Newell),
190-91
toruses,235,250,254-70,254,650
-cutting problem, 230, 236-37, 237
knotted,241-42,241,242,246,255,
256-57, 250259-61,260,262,
266-67,267
one-hole, 261
problems, 262,263-64,263, 264, 266-69,
267,268,269,270
-reversed problem, 257, 265-66, 265
reversible cloth, 255-56, 256
Sprouts on, 488, 491
topology of, 229-30, 229, 254-55
two-hole, see two-hole torus
T6th, 1. Fejes, 646
Touche, 493
tournament paradox. 298, 300-301
"Town in the Library in the Town in the Li-
brary, The" (Nesbit), 315
Transcendental Physics (Zollner), 155, 245
transfinite numbers, 320, 336-37, 341-44,
371.596
cardinal, 342
ordinal. 457
transition probabilities, 634
transitivity, 286-97
trapezoids, 46-47,47
rep-4,50
Treatise on Probability; A (Keynes), 288
tree graph, 457,458, 459, 460,464-65
"Trees and Ball Games" (Smullyan), 456-57
Trei, Peter G., 652
"Trial and Error" (D. Michie), 472
triangles, 56, 74, 178,181,655
equilateral, 36,43,46,52,56,181,273,
275,321-22,640
hyperbolic, 180, 180
isosceles, 47, 47, 73, 85, 219
problem, 17,19, 19
rep-3, 48,56,57
rep-4, 48-50, 49
rep-5, 48, 56, 57
Reuleaux, 36-37, 3 ~ 4 0 - 4 1 , 4 3
scalene, 56
triangular:
lattice, 135, 135
numbers, 128-29, 128, 129, 134
Trigg, Charles W., 24, 25
trihexaflexagons,386, 386, 388,391-92
trip-lets, 660-62
triplets, nontransitive, 303-5, 303, 304, 308,
309,310
Tristram Shandy (Sterne), 600
Tritter, Allan, 378
Tromelin, Count de, 686
722
Trotter, Hale F., 241-42
tubes, topological properties of, 228
Tucker, Albert W., 255
Tuckerman, Bryant, 386, 388-89, 390, 395
Tuckerman, Louis B., 389, 390,392
Tuckerman traverse, 388-89, 389, 392
Tukey, John W., 386, 389-90, 395, 607
Turing, Alan, 662, 665, 668, 669
Turing Game, 668
Turing machines, 417, 664
Turing Omnibus, The (Dewdney), 108
Turn About, Think About, Look About
Book, The (B. Gardner), 195
Turner, Dean E., 184
Turner, J. D., 67
Turner, S. D., 291-93
Turney, Peter, 173
Twain, Mark, 520, 523, 524-25, 636
1\veel, L. E. van der, 687
twiddled bolts problem, 124-25, 125, 127
twin paradox, 684
1\vitchit, 71
two boy problem, 277, 282
2
c
,344
"Two-dimensional Matter" (Dash), 98
Two-Dimensional Science and Technology
(Dewdney), 94, 108
two-dimensional space, 94-113
twofold symmetry, 189
of numbers, 194-95
of words, 194, 194
two-hole torus, 264, 266, 266, 269, 269
knots in, 257-58, 258
problem, 261-62, 261, 266, 266
Two Mysteries, The (Magritte), 668-69, 669
two parrot problem, 282
2-space, 138
2-sphere, 137
Ulam, Stanislaw M., 135, 300, 324,450, 553
Ultimate Aleph, 334
Ulysses Ooyce), 662
unabexaflexagons, 392
Unamuno, Miguel de, 317-18
unexpected egg paradox, 569-70, 570, 573
Unexpected Hanging and Other Mathemati-
cal Diversions, The (Gardner), 94
unexpected hanging paradox, 567-79,568
unexpected spade paradox, 570-72, 571
universal set, 611-12
universe, universes:
of discourse, 611-12
infinity of, 614-19
nature of, 613-22
parallel, 539, 618-21, 622
shrinkage of, 614, 622
3-space, 617
Unsolved Problems in Geometry (R. Guy et
al.), 172
UNu, 26
Index
upside-down:
faces, 192-94
optical illusions, 192, 193
pictures, 190-92, 191,192,195,196-97
words, 207
Upside-Downs of Little Lady Lovekins and
Old Man Muffaroo, The (Verbeek),
191, 192
Uptegrove, Neil, 394
urchin, Kepler's, 216-18
Uses of Infinity (ZippinJ. 338
van Fraassen, Bas, 607
Vardi, !lan, 208
Varieties of Religious Experience, The (W.
James),317
Vasari, Giorgio, 65
Vegetable Staticks (S. Hales), 132
Venn, John, 612
Venn diagrams, 613-15, 613, 614
Verbeek, Gustave, 191, 192, 196
vertical symmetry, 189,190
Villa-Lobos, Heitor, 358, 633, 633
viruses, 643
helical, 643
icosahedral, 643
Visions of Symmetry (Schattschneider), 224
Visitors from Oz (Gardner), 236
Visual Encyclopedia of Science Fiction,
The,539
Voderberg, Heinz, 75, 75
Void,598
volvox, 650-51
"Von Aubel's Quadrilateral Theorem"
(Kelly),560
Von Aubel's theorem, 558-59, 558, 559-60,
561
VOlUlegut, Kurt, 518
von Neumann, John, 594
Voss, Richard F., 351, 353, 354, 359-65
voting paradox, 297-300
Vout, Colin, 375
Waerden, B. L. van der, 641
Wagner, Richard, 351, 636
Wainwright, Robert T., 417, 419
Walden, William, 481
Walden (Thoreau), 190, 553-54
Walker, David N., 281
Walker, Kenneth, 654
Wallis, John, 176
Walter, W. Gray, 322
waltzes, 631, 632
Wang, Hao, 76
Wang dominoes, 76
Ward, D. R., 183
Warhol, Andy, 600
Warmus, M., 461-62
wasps, 646
Waterfall (Escher), 664
Index
Watson, G. N., 133
Watson, James D., 117
Watts, Harry James, 36-37. 43
Watts Brothers Tool Works, 37, 41, 43
Watts square-hole drill, 36-37, 37, 43
Way Things Are, The (Bridgman), 347
weather, in a planiverse, 97-98
We Build Our Own Computers (Bolt et al.),
481
Weiland, Andrea, 478
Weinbaum, Stanley G., 648
Weisbach, Michael, 501
Welcker, 629
Wells, H. G., 150,291,517,518,520,529,
532-33
Wells, Mark B., 430
Weyl, Hermann, 327, 344
whales, 645
"What Is the Most Amazing Approximate
Integer in the Universe?" (Good), 684
What Mad Universe (F. Brown), 524-25
What Shall We Do Tonight?, 193
"What the Tortoise Said to Achilles" (Car-
roll),316
Wheeler, John Archibald, 171, 524, 527-28,
597,614,617,618-19,684
Wheelers, 649, 650
wheels, in animal propUlsion, 649, 652-54
Wheels, Life, and Other Mathematical
Amusements (Gardner), 184, 420
"When the Earth Screamed" (Conan Doyle),
649
When the Sleeper Wakes (H. Wells), 520,
532
Whistler, Rex, 191, 191
White, Morton, 192
"White Christmas," 637
Whitehead, Alfred North, 182, 184, 532,
554,562
Whitehead, J.H.C., 5
Whitehead's Philosophy of Science (Palter),
182-83
white holes, 614
white music, 352, 362-64, 363, 367
randomizers and, 352-53, 352
white noise, 352, 354, 358-60, 360, 362
Whitrow, G. J., 99
Whitworth, William A., 278
"Why Animals Run on Legs, Not on
Wheels" (Diamond), 653-54
Why Knots, 247
Whyte Gallery, 213
Wiener, Norbert, 60, 471
"wild" structure, 369
William Lowell Putnam Mathematical Com-
petition of 1953,437
Williams, Ben Ames, 3-4, 6-8
Willink, Arthur, 154
Will to Believe, The (W. James), 291
Wilson, J. Walter, 8
723
Wilson, Marguerite, 408
Wilson, Rosina, 30
Winchell, Paul, 194
Winkel, M., 632-33
Winning Ways (Berlekamp, R. Guy, and
Conway), 111, 380, 407
Winograd, Terry, 664
Winsor, Frederick, 227
Wisner, R. J., 392
Wiszowaty, Kenneth W., 478
Witch of Agnesi, 66
Witten, Edward, 158, 248
Wolk, Barry, 305, 308
Wood, Robert W., 193-94
wood lice, 653
648
words:
inverted into numbers, 195
in map-fold puzzles, 424-25
mirror symmetry of, 198, 201, 202, 206-7
palindromic, 206
reflected into words, 195
symmetry in, 198-99,198,200-203,200,
201,202,206-7,206
twofold symmetry of, 194, 194
upside-down, 206, 207
724
World and the Individual, The (Royce), 344
world-line loops, 521
world lines, 518, 614-15
World o/the Unseen, The (Willink), 154
World War II, 60, 425
worms, 84, 85
Wright, Frank Lloyd, 119
Yaglom, 1. M., 43, 560
years, twofold symmetry of, 194-95
Y gnetovich, Vladimir, 491
Zapf, Hermann, 71
Zeltmacher, J. J., Jr., 236
Zeno's paradox, 344, 345--46, 347,663
staccato runner and, 345--46, 347
Zeno's Paradoxes (Salmon), 345
Zermelo-Fraenkel set theory, 613
zero, 592-93, 595
zero case, 17
zero curvature, 182, 598
Zippin, Leo, 338
Zollner, Johann Karl Friedrich, 155,245
zoo, mathematical, 640-59
Zoological Society of San Diego, 194, 640
ZOONOOZ, 194, 198, 640
Index