Visit https://ebookfinal.

com to download the full version and

explore more ebooks

A First Course in Computational Algebraic

Geometry 1st Edition Wolfram Decker

_____ Click the link below to download _____

https://ebookfinal.com/download/a-first-course-in-
computational-algebraic-geometry-1st-edition-wolfram-
decker/

Explore and download more ebooks at ebookfinal.com

Here are some suggested products you might be interested in.
Click the link to download

Current Developments in Algebraic Geometry 1st Edition

Lucia Caporaso

https://ebookfinal.com/download/current-developments-in-algebraic-
geometry-1st-edition-lucia-caporaso/

Introduction to algebraic geometry 1st Edition Brendan

Hassett

https://ebookfinal.com/download/introduction-to-algebraic-
geometry-1st-edition-brendan-hassett/

Ring Theory and Algebraic Geometry 1st Edition Ánjel

Granja

https://ebookfinal.com/download/ring-theory-and-algebraic-
geometry-1st-edition-anjel-granja/

A first course in network theory First Edition Ernesto

Estrada

https://ebookfinal.com/download/a-first-course-in-network-theory-
first-edition-ernesto-estrada/
Computational geometry algorithms and applications 3e
edition Edition Berg

https://ebookfinal.com/download/computational-geometry-algorithms-and-
applications-3e-edition-edition-berg/

Algebraic Geometry and Arithmetic Curves 1, Corrected

Edition 2006 Edition Qing Liu

https://ebookfinal.com/download/algebraic-geometry-and-arithmetic-
curves-1-corrected-edition-2006-edition-qing-liu/

A First Course in Functional Analysis 1st Edition Orr

Moshe Shalit

https://ebookfinal.com/download/a-first-course-in-functional-
analysis-1st-edition-orr-moshe-shalit/

Abstract Algebra Applications to Galois Theory Algebraic

Geometry and Cryptography 1st Edition Celine Carstensen

https://ebookfinal.com/download/abstract-algebra-applications-to-
galois-theory-algebraic-geometry-and-cryptography-1st-edition-celine-
carstensen/

A first course in atmospheric radiation 2. ed. Edition

Petty

https://ebookfinal.com/download/a-first-course-in-atmospheric-
radiation-2-ed-edition-petty/
A First Course in Computational Algebraic Geometry 1st
Edition Wolfram Decker Digital Instant Download
Author(s): Wolfram Decker, Gerhard Pfister
ISBN(s): 9781107612532, 1107612535
Edition: 1
File Details: PDF, 1.32 MB
Year: 2013
Language: english
A FIRST COURSE IN COMPUTATIONAL
ALGEBRAIC GEOMETRY
African Institute of Mathematics Library Series

The African Institute of Mathematical Sciences (AIMS), founded in 2003 in

Muizenberg, South Africa, provides a one-year postgraduate course in
mathematical sciences for students throughout the continent of Africa. The
AIMS LIBRARY SERIES is a series of short innovative texts, suitable for
self-study, on the mathematical sciences and their applications in the
broadest sense.

Editorial Board

Professor J. Banasiak (University of KwaZulu-Natal)

Professor A.F. Beardon (University of Cambridge)
Professor P. Deift (Courant Institute of Mathematical Sciences)
Professor P.E. Dorey (Durham University)
Professor B.W. Green (Director, AIMS)
Professor F.W. Hahne (AIMS and STIAS)
Professor A. Iserles (University of Cambridge)
Professor P.E. Kopp (University of Hull)
Professor J.W. Sanders (Academic Director, AIMS)
Professor P. Sarnak (Institute for Advanced Study Princeton)
Dr. T. Tokieda (University of Cambridge)
Professor N.G. Turok (University of Cambridge and Perimeter Institute,
Ontario)
 A FIRST COURSE IN
COMPUTATIONAL
ALGEBRAIC GEOMETRY

WOLFRAM DECKER
and
GERHARD PFISTER
Universität Kaiserslautern

With Pictures by Oliver Labs

CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town,
Singapore, São Paulo, Delhi, Mexico City
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK

Published in the United States of America by Cambridge University Press,

New York

www.cambridge.org
Information on this title: www.cambridge.org/9781107612532

c Wolfram Decker and Gerhard Pfister 2013



This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.

First published 2013

Printed and bound in the United Kingdom by the MPG Books Group

A catalogue record for this publication is available from the British Library

ISBN 978-1-107-61253-2 Paperback

Cambridge University Press has no responsibility for the persistence or

accuracy of URLs for external or third-party internet websites referred to
in this publication, and does not guarantee that any content on such
websites is, or will remain, accurate or appropriate.
 Contents

Preface page vii

Prologue:
General Remarks on Computer Algebra Systems 1
1 The Geometry–Algebra Dictionary 11
1.1 Affine Algebraic Geometry 11
1.1.1 Ideals in Polynomial Rings 11
1.1.2 Affine Algebraic Sets 14
1.1.3 Hilbert’s Nullstellensatz 20
1.1.4 Irreducible Algebraic Sets 23
1.1.5 Removing Algebraic Sets 25
1.1.6 Polynomial Maps 29
1.1.7 The Geometry of Elimination 32
1.1.8 Noether Normalization and Dimension 37
1.1.9 Local Studies 45
1.2 Projective Algebraic Geometry 49
1.2.1 The Projective Space 49
1.2.2 Projective Algebraic Sets 52
1.2.3 Affine Charts and the Projective Closure 54
1.2.4 The Hilbert Polynomial 57
2 Computing 60
2.1 Standard Bases and S INGULAR 60
2.2 Applications 75
2.2.1 Ideal Membership 75
2.2.2 Elimination 75
2.2.3 Radical Membership 77
2.2.4 Ideal Intersections 78

v
vi Contents

2.2.5 Ideal Quotients 79

2.2.6 Kernel of a Ring Map 79
2.2.7 Integrality Criterion 80
2.2.8 Noether Normalization 82
2.2.9 Subalgebra Membership 83
2.2.10 Homogenization 83
2.3 Dimension and the Hilbert Function 84
2.4 Primary Decomposition and Radicals 90
2.5 Buchberger’s Algorithm and Field Extensions 94
3 Sudoku 95
4 A Problem in Group Theory Solved by Computer Algebra 101
4.1 Finite Groups and Thompson’s Theorem 101
4.2 Characterization of Finite Solvable Groups 104
Bibliography 112
Index 115
 Preface

Most of mathematics is concerned at some level with setting up and solving

various types of equations. Algebraic geometry is the mathematical discipline
which handles solution sets of systems of polynomial equations. These are
called algebraic sets.
By making use of a correspondence which relates algebraic sets to ideals
in polynomial rings, problems concerning the geometry of algebraic sets can
be translated into algebra. As a consequence, algebraic geometers have devel-
oped a multitude of often highly abstract techniques for the qualitative and
quantitative study of algebraic sets, without, in the first instance, considering
the equations. Modern computer algebra algorithms, on the other hand, allow
us to manipulate the equations and, thus, to study explicit examples. In this
way, algebraic geometry becomes accessible to experiments. The experimen-
tal method, which has proven to be highly successful in number theory, is now
also added to the toolbox of the algebraic geometer.
In these notes, we discuss some of the basic operations in geometry and
describe their counterparts in algebra. We explain how the operations can be
carried out using computation, and give a number of explicit examples, worked
out with the computer algebra system S INGULAR. In this way, our book may
serve as a first introduction to S INGULAR, guiding the reader to performing his
own experiments.
In detail, we proceed along the following lines:
The Prologue contains remarks on computer algebra systems in general with
just a few examples of what can be computed in different application areas.
In Chapter 1, we focus on the geometry–algebra dictionary, illustrating its
entries by including a number of S INGULAR examples.
Chapter 2 contains a discussion of the algorithms involved and gives a more
thorough introduction to S INGULAR.

vii
viii Preface

For the fun of it, in Chapter 3, we show how to find the solution of a well-
posed Sudoku game by solving a corresponding system of polynomial equa-
tions.
Finally, in Chapter 4, we discuss a particular classification problem in group
theory, and explain how a combination of theory and explicit computations
leads to a solution of the problem. Here, algorithmic methods from group the-
ory, number theory, and algebraic geometry are involved.
Due to the expository character of these notes, proofs are only included oc-
casionally. For all other proofs, references are given.
For a set of Exercises, see
http://www.mathematik.uni-kl.de/∼pfister/Exercises.pdf.

The notes grew out of a course we taught at the African Institute for the
Mathematical Sciences (AIMS) in Cape Town, South Africa. Teaching at AIMS
was a wonderful experience and we would like to thank all the students for
their enthusiasm and the fun we had together. We very much appreciated the
facilities at AIMS and we are grateful to its staff for constant support.
We thank Oliver Labs for contributing the illustrations along with hints on
improving the text, Christian Eder and Stefan Steidel for reading parts of the
manuscript and making helpful suggestions, and Petra Bäsell for typesetting
the notes.

Kaiserslautern, Wolfram Decker

October 2011 Gerhard Pfister
 Prologue:
General Remarks on Computer Algebra Systems

Computer algebra algorithms allow us to compute in, and with, a multitude

of mathematical structures. Accordingly, there is a large number of computer
algebra systems suiting different needs, ranging from the general to the special
purpose. Some well-known examples of the former are commercial, whereas
many of the special purpose systems are open-source and can be downloaded
from the internet for free. General purpose systems aim at providing basic
functionality for a variety of different application areas. In addition to tools for
symbolic computation, they usually offer tools for numerical computation and
for visualization.

Example P.1 M APLE is a commercial general purpose system. To show a

few of its commands at work, we start with examples from calculus, namely
definite and indefinite integration:
> int(sin(x), x = 0 .. Pi);
2
> int(x/(x^2-1), x);
1/2 ln(x - 1) + 1/2 ln(x + 1)

For linear algebra applications, we first load the corresponding package. Then
we demonstrate how to perform Gaussian elimination and to compute eigen-
values.
with(LinearAlgebra);
A := Matrix([[2, 1, 0], [1, 2, 1], [0, 1, 2]]);
 
2 1 0
 
 1 2 1 
 
0 1 2

GaussianElimination(A);

1
2 Prologue: General Remarks on Computer Algebra Systems
 
2 1 0
 
 0 3/2 1 
 
0 0 4/3

Eigenvalues(A);
 
2
 √ 
 2− 2 
√
 
2+ 2

Next, we give an example of solving numerically1 :

> fsolve(2*x^5-11*x^4-7*x^3+12*x^2-4*x = 0);
-1.334383488, 0., 5.929222024

Finally, we show one of the graphic functions at work:

> plot3d(x*exp(-x^2-y^2),x = -2 .. 2,y = -2 .. 2,grid = [49, 49]);

For applications in research, general purpose systems are often not powerful
enough: The implementation of the required basic algorithms may not be opti-
mal with respect to speed and storage handling, and more advanced algorithms
may not be implemented at all. Many special purpose systems were created
by people working in a field other than computer algebra: they had a desperate
need for computing power in the context of some of their research problems. A
pioneering and prominent example is Veltman’s S CHOONSHIP which helped
to win a Nobel price in physics in 1999 (awarded to Veltman and t’Hooft ‘for
having placed particle physics theory on a firmer mathematical foundation’).
1 Note that only the real roots are computed.
 Prologue: General Remarks on Computer Algebra Systems 3

Example P.2 GAP is a free open-source system for computational discrete

algebra, with particular emphasis on Computational Group Theory. In the fol-
lowing GAP session, we define a subgroup G of the symmetric group S11 (the
group of permutations of {1, . . . , 11}) by giving two generators in cycle2 no-
tation. We check that G is simple (that is, its only normal subgroups are the
trivial subgroup and the whole group itself). Then we compute the order |G| of
G, and factorize this number:
gap> G := Group([(1,2,3,4,5,6,7,8,9,10,11),(3,7,11,8)(4,10,5,6)]);
Group([(1,2,3,4,5,6,7,8,9,10,11), (3,7,11,8)(4,10,5,6)])
gap> IsSimple(G);
true
gap> size := Size(G);
7920
gap> Factors(size);
[ 2, 2, 2, 2, 3, 3, 5, 11 ]

From the factors, we see that G has a Sylow 2-subgroup3 of order 24 = 16. We
use GAP to find such a group P:
gap> P := SylowSubgroup(G, 2);

Group([(2,8)(3,4)(5,6)(10,11), (3,5)(4,6)(7,9)(10,11),
(2,4,8,3)(5,10,6,11)])

By making use of the Small Groups Library included in GAP, we can check
that, up to isomorphism, there are 14 groups of order 16, and that P is the 8th
group of order 16 listed in this library:
gap> SmallGroupsInformation(16);
There are 14 groups of order 16.
They are sorted by their ranks.
1 is cyclic.
2 - 9 have rank 2.
10 - 13 have rank 3.
14 is elementary abelian.
gap> IdGroup( P );
[ 16, 8 ]

Now, we determine what group P is. First, we check that P is neither Abelian
nor the dihedral group of order 16 (the dihedral group of order 2n is the sym-
metry group of the regular n-gon):
2 The cycle (4,10,5,6), for instance, maps 4 to 10, 10 to 5, 5 to 6, 6 to 4, and any other number
to itself.
3 If G is a finite group, and p is a prime divisor of its order |G|, then a subgroup U of G is called
a Sylow p-subgroup if its order |U| is the highest power of p dividing |G|.
4 Prologue: General Remarks on Computer Algebra Systems

gap> IsAbelian(P);
false
gap> IsDihedralGroup(P);
false

More information about P can be obtained by studying the subgroups of P of

order 8. In fact, we consider the third such subgroup returned by GAP and
name it H:

gap> H := SubgroupsOfIndexTwo(P)[3];
Group([(2,3,11,5,8,4,10,6)(7,9), (2,4,11,6,8,3,10,5)(7,9),
(2,5,10,3,8,6,11,4)(7,9), (2,6,10,4,8,5,11,3)(7,9)])
gap> IdGroup(H);
[ 8, 1 ]
gap> IsCyclic(H);
true

Thus, H is the cyclic group C8 of order 8 (cyclic groups are generated by just
one element). Further checks show, in fact, that P is a semidirect product of
C8 and the cyclic group C2 . See Wild (2005) for the classification of groups of
order 16.

Remark P.3 The group G studied in the previous example is known as the
Mathieu group M11 . We should point out that researchers in group theory and
representation theory have created quite a number of useful electronic libraries
such as the Small Groups Library considered above.

Example P.4 M AGMA is a commercial system focusing on algebra, num-

ber theory, geometry and combinatorics. We use it to factorize the 8th Fermat
number:
> Factorization(2^(2^8)+1);
[<1238926361552897,1>,
<93461639715357977769163558199606896584051237541638188580280321,1>]

Next, we meet our first example of an algebraic set: In Weierstraß normal form,
an elliptic curve over a field K is a nonsingular4 curve in the xy-plane defined
by one polynomial equation of type

y2 + a1 xy + a3 y − x3 − a2 x2 − a4 x − a6 = 0,

with coefficients ai ∈ K. In the following M AGMA session, we define an elliptic

4 Informally, a curve is nonsingular if it admits a unique tangent line at each of its points. See,
for instance, Silverman (2009) for a formal definition and for more information on elliptic
curves.
 Prologue: General Remarks on Computer Algebra Systems 5

curve E in Weierstraß normal form over the finite field F with 590 elements by
specifying the coefficients ai . Then we count the number of points on E with
coordinates in F.
F := FiniteField(5,90);
E := EllipticCurve([Zero(F),Zero(F),One(F),-One(F),Zero(F)]);
E;
Elliptic Curve defined by y^2 + y = x^3 + 4*x over GF(5^90)
#E;
807793566946316088741610050849537214477762546152780718396696352

The significance of elliptic curves stems from the fact that they carry an (addi-
tive) group law. Having specified a base point (the zero element of the group),
the addition of points is defined by a geometric construction involving secant
and tangent lines. For elliptic curves in Weierstraß normal form, it is conve-
nient to choose the unique point at infinity of the curve as the base point (see
Section 1.2.1 for points at infinity and Example P.6 below for a demonstration
of the group law).

Remark P.5 Elliptic curves, most notably elliptic curves defined over Q re-
spectively over a finite field, are of particular importance in number theory.
They take center stage in the conjecture of Birch and Swinnerton-Dyer (1965)
5 , they are key ingredients in the proof of Fermat’s last theorem Wiles (1995),

they are important for integer factorization Lenstra (1987), and they find appli-
cations in cryptography Koblitz (1987). As with many other awesome conjec-
tures in number theory, the Birch and Swinnerton-Dyer conjecture is based on
computer experiments.

Example P.6 S AGE is a free open-source mathematics software system which

combines the power of many existing open-source packages into a common
P YTHON-based interface. To show it at work, we start as in Example P.1 with
computations from calculus. Then, we compute all prime numbers between
two given numbers.
sage: limit(sin(x)/x, x=0)
1
sage: taylor(sqrt(x+1), x, 0, 5)
7/256*x^5 - 5/128*x^4 + 1/16*x^3 - 1/8*x^2 + 1/2*x + 1
sage: list(primes(10000000000, 10000000100))
[10000000019, 10000000033, 10000000061, 10000000069, 10000000097]

Finally, we define an elliptic curve E in Weierstraß normal form over Q and

5 The Birch and Swinnerton-Dyer conjecture asserts, in particular, that an elliptic curve E over
Q has an infinite number of points with rational coordinates iff its associated L-series satisfies
L(E, 1) = 0.
6 Prologue: General Remarks on Computer Algebra Systems

demonstrate the group law on this curve. The representation of the results takes
infinity into account in the sense that the points are given by their homogeneous
coordinates in the projective plane (see Section 1.2 for the projective setting).
In particular, (0 : 1 : 0) denotes the unique point at infinity of the curve which
is chosen to be the zero element of the group.
sage: E = EllipticCurve([0,0,1,-1,0])
sage: E
Elliptic Curve defined by y^2 + y = x^3 - x over Rational Field
sage: P = E([0,0])
sage: P
(0 : 0 : 1)
sage: O = P - P
sage: O
(0 : 1 : 0)
sage: Q = E([-1,0])
sage: Q
(-1 : 0 : 1)
sage: Q + O
(-1 : 0 : 1)
sage: P + Q - (P+Q)
(0 : 1 : 0)
Q + (P + R) - ((Q + P) + R)
(0 : 1 : 0)

Among the systems combined by S AGE are M AXIMA, a general purpose sys-
tem which is free and open-source, GAP, the system introduced in Example
P.2, PARI/GP, a system for number theory, and S INGULAR, the system fea-
tured in these notes.

S INGULAR is a free open-source system for polynomial computations, with

special emphasis on commutative and noncommutative algebra, algebraic ge-
ometry, and singularity theory. Like most other systems, S INGULAR consists
of a precompiled kernel, written in C/C++, and additional packages, called li-
braries and written in the C-like S INGULAR user language. This language is in-
terpreted on runtime. S INGULAR binaries are available for most common hard-
ware and software platforms. Its release versions can be downloaded through
ftp from

ftp://www.mathematik.uni-kl.de/pub/Math/Singular/
 Prologue: General Remarks on Computer Algebra Systems 7

or via your favorite web browser from S INGULAR’s webpage

http://www.singular.uni-kl.de/ .
S INGULAR also provides an extensive online manual and help function. See
its webpage or enter help; in a S INGULAR session.
Most algorithms implemented in S INGULAR rely on the basic task of com-
puting Gröbner bases. Gröbner bases are special sets of generators for ideals in
polynomial rings. Their definition and computation is subject to the choice of
a monomial ordering such as the lexicographical ordering >lp and the degree
reverse lexicographical ordering >dp . We will treat Gröbner bases and their
computation by Buchberger’s algorithm in Chapter 2. S INGULAR examples,
however, will already be presented beforehand.
S INGULAR Example P.7 We enter the polynomials of the system
x+y+z−1 = 0
2 2 2
x +y +z −1 = 0
x3 + y3 + z3 − 1 = 0
in a S INGULAR session. For this, we first have to define the corresponding
polynomial ring which is named R and endowed with the lexicographical or-
dering. Note that the 0 in the definition of R refers to the prime field of charac-
teristic zero, that is, to Q.

> ring R = 0, (x,y,z), lp;

> poly f1 = x+y+z-1;
> poly f2 = x2+y2+z2-1;
> poly f3 = x3+y3+z3-1;

Next, we define the ideal generated by the polynomials and compute a Gröbner
basis for this ideal (the system given by the Gröbner basis elements has the
same solutions as the original system).

> ideal I = f1, f2, f3;

> ideal GI = groebner(I); GI;
GI[1]=z3-z2
GI[2]=y2+yz-y+z2-z
GI[3]=x+y+z-1

In the first equation of the new system, the variables x and y are eliminated.
In the second, x is eliminated. As a consequence, the solutions can now be
directly read off:
(1, 0, 0), (0, 1, 0), (0, 0, 1).
8 Prologue: General Remarks on Computer Algebra Systems

The example indicates that >lp is what we will call an elimination ordering.
If such an ordering is chosen, Buchberger’s algorithm generalizes Gaussian
elimination. For most applications of the algorithm, however, the elimination
property is not needed. It is, then, usually more efficient to choose the ordering
>dp .
Multivariate polynomial factorization is another basic task on which some of
the more advanced algorithms in S INGULAR rely. Starting with the first com-
puter algebra systems in the 1960s, the design of algorithms for polynomial
factorization has always been an active area of research. To keep the size of
our notes within reasonable limits, we will not treat this here. We should point
out, however, that algorithms for polynomial factorization do not depend on
monomial orderings. Nevertheless, choosing such an ordering is always part
of a ring definition in S INGULAR.

S INGULAR Example P.8 We factorize a polynomial in Q[x, y, z] using the

S INGULAR command factorize. The resulting output is a list, with the fac-
tors as the first entry, and the corresponding multiplicities as a second.
> ring R = 0, (x,y,z), dp;
> poly f = -x7y4+x6y5-3x5y6+3x4y7-3x3y8+3x2y9-xy10+y11-x10z
. +x8y2z+9x6y4z+11x4y6z+4x2y8z-3x5y4z2+3x4y5z2-6x3y6z2+6x2y7z2
. -3xy8z2+3y9z2-3x8z3+6x6y2z3+21x4y4z3+12x2y6z3-3x3y4z4+3x2y5z4
. -3xy6z4+3y7z4-3x6z5+9x4y2z5+12x2y4z5-xy4z6+y5z6-x4z7+4x2y2z7;
> factorize(f);
[1]:
_[1]=-1
_[2]=xy4-y5+x4z-4x2y2z
_[3]=x2+y2+z2
[2]:
1,1,3

Remark P.9 In recent years, quite a number of the more abstract concepts
in algebraic geometry have been made constructive. They are, thus, not only
easier to understand, but can be handled by computer algebra. A prominent
example is the desingularization theorem of Hironaka (see Hironaka (1964))
for which Hironaka received the Fields Medal. In fact, Villamajor’s construc-
tive version of Hironaka’s proof has led to an algorithm whose S INGULAR
implementation allows us to resolve singularities in many cases of interest (see
Bierstone and Milman (1997), Frühbis-Krüger and Pfister (2006), Bravo et al.
(2005)).

When studying plane curves or surfaces in 3-space, it is often desirable to

visualize the geometric objects under consideration. Excellent tools for this are
 Prologue: General Remarks on Computer Algebra Systems 9

S URF and its descendants S URFEX6 and S URFER7 . Comparing these, we note
that S URFEX has more features, whereas S URFER is easier to handle.

Example P.10 The following S URFER picture shows a surface in 3-space

found by Oliver Labs using S INGULAR (see Labs (2006)):

S INGULAR Example P.11 We set up the equation of Labs’ surface in S IN -

GULAR. The equation is defined over a finite extension field of Q which we
implement by entering its minimal polynomial:

> ring R = (0,a), (x,y,w,z), dp;

> minpoly = a^3 + a + 1/7;
> poly a(1) = -12/7*a^2 - 384/49*a - 8/7;
> poly a(2) = -32/7*a^2 + 24/49*a - 4;
> poly a(3) = -4*a^2 + 24/49*a - 4;
> poly a(4) = -8/7*a^2 + 8/49*a - 8/7;
> poly a(5) = 49*a^2 - 7*a + 50;
> poly P = x*(x^6-3*7*x^4*y^2+5*7*x^2*y^4-7*y^6)
. +7*z*((x^2+y^2)^3-2^3*z^2*(x^2+y^2)^2
. +2^4*z^4*(x^2+y^2))-2^6*z^7;
> poly C = a(1)*z^3+a(2)*z^2*w+a(3)*z*w^2+a(4)*w^3+(z+w)*(x^2+y^2);
> poly S = P-(z+a(5)*w)*C^2;
> homog(S); // returns 1 if poly is homogeneous
1
> deg(S);
7

We see that S is a homogeneous polynomial of degree 7. It defines Labs’ sur-

6 http://surf.sourceforge.net
7 http://www.oliverlabs.net/welcome.php
10 Prologue: General Remarks on Computer Algebra Systems

face in projective 3-space. This surface is a ‘world record’ surface in that it

has the maximal number of nodes known for a degree-7 surface in projective
3-space (a node constitutes the most simple type of a singularity). We use S IN -
GULAR to confirm that there are precisely 99 nodes (and no other singularities).
First, we compute the dimension of the locus of singularities via the Jacobian
criterion (see Decker and Schreyer (2013) for the criterion and Sections 1.1.8
and 2.3 for more on dimension):

> dim(groebner(jacob(S)))-1;
0

The result means that there are only finitely many singularities. By checking
that the nonnodal locus is empty, we verify that all singularities are nodes.
Then, we compute the number of nodes:

> dim(groebner(minor(jacob(jacob(S)),2))) - 1;
-1
> mult(groebner(jacob(S)));
99

S INGULAR Example P.12 When properly installed, S URF, S URFEX, and

S URFER can be called from S INGULAR. To give an example, we use S URFER
to plot a surface which, as it turns out, resembles a citrus fruit. To begin, we
load the S INGULAR library connecting to S URF and S URFER.

> LIB "surf.lib";

> ring R = 0, (x,y,z), dp;
> ideal I = 6/5*y^2+6/5*z^2-5*(x+1/2)^3*(1/2-x)^3;
surfer(I);

The resulting picture will show in a pop-up window:

See http://www.imaginary-exhibition.com for more pictures.

 1
The Geometry–Algebra Dictionary

We will now explore the correspondence between algebraic sets in affine and
projective space and ideals in polynomial rings. More details, and all proofs not
given here, can be found in Decker and Schreyer (2013). We will work over a
field K, and write K[x1 , . . . , xn ] for the polynomial ring over K in n variables.
All rings considered are commutative with identity element 1.

1.1 Affine Algebraic Geometry

Our discussion of the geometry–algebra dictionary starts with Hilbert’s ba-
sis theorem which is the fundamental result about ideals in polynomial rings.
Then, focusing on the affine case, we present some of the basic ideas of alge-
braic geometry, with particular emphasis on computational aspects.

1.1.1 Ideals in Polynomial Rings

To begin with, let R be any ring.

Definition 1.1 A subset I ⊂ R is called an ideal of R if the following holds:

(i) 0 ∈ I.
(ii) If f , g ∈ I, then f + g ∈ I.
(iii) If f ∈ R and g ∈ I, then f · g ∈ I.

Example 1.2 (i) If 0/ = T ⊂ R is any subset, then all R-linear combinations

g1 f1 + · · · + gr fr , with g1 , . . . .gr ∈ R and f1 , . . . , fr ∈ T , form an ideal of
R, written T R or T , and called the ideal generated by T . We also say
that T is a set of generators for the ideal. If T = { f1 , . . . , fr } is finite, we

11
12 The Geometry–Algebra Dictionary

write T  =  f1 , . . . , fr . We say that an ideal is finitely generated if it

admits a finite set of generators. A principal ideal can be generated by
just one element.

(ii) If {Iλ } is a family of ideals of R, then the intersection λ Iλ is also an
ideal of R.
(iii) The sum of a family of ideals {Iλ } of R, written ∑λ Iλ , is the ideal gener-
ated by the union λ Iλ .

Now, we turn to R = K[x1 , . . . , xn ].

Theorem 1.3 (Hilbert’s Basis Theorem) Every ideal of the polynomial ring
K[x1 , . . . , xn ] is finitely generated.

There have been, since the original (Hilbert, 1890), quite a number of differ-
ent proofs for the basis theorem (see, for instance, Greuel and Pfister (2007)
for a brief proof found in the 1970s). One which nicely fits with the spirit
of these notes is due to Gordan (1899). Though the name Gröbner bases was
coined much later by Buchberger1 , it is Gordan’s paper in which these bases
make their first appearance. In fact, Gordan already exhibited the key idea be-
hind Gröbner bases, which is to reduce problems concerning arbitrary ideals
in polynomial rings to ones concerning monomial ideals. The latter are usually
much easier to solve.

Definition 1.4 A monomial in x1 , . . . , xn is a product xα = x1α1 · · · xnαn , where

α = (α1 , . . . , αn ) ∈ Nn . A monomial ideal of K[x1 , . . . , xn ] is an ideal generated
by monomials.

The first step in Gordan’s proof of the basis theorem is to show that monomial
ideals are finitely generated (somewhat mistakenly, this result is often assigned
to Dickson):

Lemma 1.5 (Dickson’s Lemma) Let 0/ = A ⊂ Nn be a subset of multi-indices,

and let I be the ideal I = xα | α ∈ A. Then there exist α (1) , . . . , α (r) ∈ A such
(1) (r)
that I = xα , . . . , xα .

Proof We proceed by induction on n, the number of variables. If n = 1, let

(1)
α (1) := min{α | α ∈ A}. Then I = xα . Now, let n > 1 and assume that the
lemma holds for n − 1. Given α = (α , αn ) ∈ Nn , with α = (α1 , . . . , αn−1 ) ∈
αn−1
Nn−1 , we write xα = x1α1 · · · xn−1 .
1 Gröbner was Buchberger’s thesis advisor. In his thesis, Buchberger developed his algorithm
for computing Gröbner bases. See Buchberger (1965).
 1.1 Affine Algebraic Geometry 13

Let A = {α ∈ Nn−1 | (α , i) ∈ A for some i}, and let

J = {xα }α ∈A  ⊂ K[x1 , . . . , xn−1 ].

By the induction hypothesis, there exist multi-indices

(1) (1) (s) (s)
β (1) = (β , βn ), . . . , β (s) = (β , βn ) ∈ A
(1) (s)
( j)
such that J = xβ , . . . , xβ . Let ℓ = max{βn }. For i = 0, . . . , ℓ, let Ai =
j
{α ∈ Nn−1 | (α , i) ∈ A} and Ji = {xα }α ∈Ai  ⊂ K[x1 , . . . , xn−1 ]. Using once
(1) (1) (si ) (si )
more the induction hypothesis, we get βi = (β i , i), . . . , βi = (β i , i) ∈ A
(1) (si )
such that Ji = xβ i , . . . , xβ i . Let
ℓ
(1) (si )
B= {βi , . . . , βi }.
i=0

Then, by construction, every monomial xα , α ∈ A, is divisible by a monomial

xβ , β ∈ B. Hence, I = {xβ }β ∈B .

In Corollary 2.28, we will follow Gordan and use Gröbner bases to deduce the
basis theorem from the special case treated above.

Theorem 1.6 Let R be a ring. The following are equivalent:

(i) Every ideal of R is finitely generated.

(ii) (Ascending Chain Condition) Every chain

I1 ⊂ I2 ⊂ I3 ⊂ · · ·

of ideals of R is eventually stationary. That is,

Ik = Ik+1 = Ik+2 = · · · for some k ≥ 1.

Definition 1.7 A ring satisfying the (equivalent) conditions above is called a

Noetherian ring.

Finally, we introduce the following terminology for later use:

Definition 1.8 We say that an ideal I of R is a proper ideal if I = R. A proper

ideal p of R is a prime ideal if f , g ∈ R and f g ∈ p implies f ∈ p or g ∈ p. A
proper ideal m of R is a maximal ideal if there is no ideal I of R such that
m  I  R.
14 The Geometry–Algebra Dictionary

1.1.2 Affine Algebraic Sets

Following the usual habit of algebraic geometers, we write An (K) instead of
K n : The affine n-space over K is the set

An (K) = (a1 , . . . , an ) | a1 , . . . , an ∈ K .

Each polynomial f ∈ K[x1 , . . . , xn ] defines a function

f : An (K) → K, (a1 , . . . , an ) → f (a1 , . . . , an ),

which is called a polynomial function on An (K). Viewing f as a function

allows us to talk about the zeros of f . More generally, we define:

Definition 1.9 If T ⊂ K[x1 , . . . , xn ] is any set of polynomials, its vanishing

locus (or locus of zeros) in An (K) is the set

V(T ) = {p ∈ An (K) | f (p) = 0 for all f ∈ T }.

Every such set is called an affine algebraic set.

It is clear that V(T ) coincides with the vanishing locus of the ideal T  gen-
erated by T . Consequently, every algebraic set A in An (K) is of type V(I) for
some ideal I of K[x1 , . . . , xn ]. By Hilbert’s basis theorem, A is the vanishing lo-

cus V( f1 , . . . , fr ) = ri=1 V( fi ) of a set of finitely many polynomials f1 , . . . , fr .
Referring to the vanishing locus of a single nonconstant polynomial as a hy-
persurface in An (K), this means that a subset of An (K) is algebraic iff it can
be written as the intersection of finitely many hypersurfaces. Hypersurfaces in
A2 (K) are called plane curves.

Example 1.10 We choose K = R so that we can draw pictures.

(i) Nondegenerate conics (ellipses, parabolas, hyperbolas) are well-known

examples of plane curves. They are defined by degree-2 equations such
as x2 + y2 − 1 = 0.
(ii) As discussed in Example P.4, elliptic plane curves are defined by degree-3
equations. Here is the real picture of the elliptic curve from Example P.6:
 1.1 Affine Algebraic Geometry 15

y2 + y − x3 + x = 0
(iii) The four-leaf clover below is given by a degree-6 equation:

3
x2 + y2 − 4x2 y2 = 0
(iv) The plane curve with degree-5 equation
49x3 y2 − 50x2 y3 − 168x3 y + 231x2 y2 − 60xy3
+144x3 − 240x2 y + 111xy2 − 18y3
+16x2 − 40xy + 25y2 = 0

admits the rational parametrization2

g1 (t) g2 (t)
x(t) = , y(t) =
h(t) h(t)
with
g1 (t) = −1200t 5 − 11480115t 4 − 19912942878t 3
+272084763096729t 2 + 131354774678451636t
+15620488516704577428,

g2 (t) = 1176t 5 − 11957127t 4 − 18673247712t 3

+329560549623774t 2 + 158296652767188936t
−1874585949429456255447,
2 See Definition 1.67 for rational parametrizations. The parametrization here was found using
the S INGULAR library paraplanecurves.lib.
16 The Geometry–Algebra Dictionary
h(t) = −45799075t 4 − 336843036810t 3
−693864026735607t 2 − 274005776716382844t
−30305468086665272172.

In addition to showing the curve in the affine plane, we also present a

‘spherical picture’ of the projective closure of the curve (see Section 1.2.3
for the projective closure):

(v) Labs’ septic from Example P.10 is a hypersurface in 3-space. Another

such hypersurface is the Kummer surface:

Depending on a parameter µ , the equation of the Kummer surface is of

type
2
x2 + y2 + z2 − µ 2 − λ y0 y1 y2 y3 = 0,
where the yi are the tetrahedral coordinates
√ √
y0 = 1 − z − 2x, y1 = 1 − z + 2x,
√ √
y2 = 1 + z + 2y, y3 = 1 + z − 2y,
 1.1 Affine Algebraic Geometry 17
3µ 2 −1
and where λ = 3−µ 2
. For the picture, µ was set to be 1.3.
(vi) The twisted cubic curve in A3 (R) is obtained by intersecting the hyper-
surfaces V(y − x2 ) and V(xy − z):
z

Taking vanishing loci defines a map V which sends sets of polynomials to

algebraic sets. We summarize the properties of V:

Proposition 1.11 (i) The map V reverses inclusions: If I ⊂ J are subsets of

K[x1 , . . . , xn ], then V(I) ⊃ V(J).
(ii) Affine space and the empty set are algebraic:
V(0) = An (K); V(1) = 0.
/

(iii) The union of finitely many algebraic sets is algebraic: If I1 , . . . , Is are ide-
als of K[x1 , . . . , xn ], then
s  s 

V(Ik ) = V Ik .
k=1 k=1

(iv) The intersection of any family of algebraic sets is algebraic: If {Iλ } is a

family of ideals of K[x1 , . . . , xn ], then
 

V(Iλ ) = V ∑ Iλ .
λ λ

(v) A single point is algebraic: If a1 , . . . , an ∈ K, then

V(x1 − a1 , . . . , xn − an ) = {(a1 , . . . , an )}.

Proof All properties except (iii) are immediate from the definitions. For (iii),
by induction, it suffices to treat the case of two ideals I, J ⊂ K[x1 , . . . , xn ]. Let
I · J be the ideal generated by all products f · g, with f ∈ I and g ∈ J. Then,
18 The Geometry–Algebra Dictionary

as is easy to see, V(I) ∪ V(J) = V(I · J) and V(I) ∪ V(J) ⊂ V(I ∩ J) ⊂ V(I · J)
(the second inclusion holds since I · J ⊂ I ∩ J). The result follows.
Remark 1.12 (i) Properties (ii)–(iv) above mean that the algebraic subsets
of An (K) are the closed sets of a topology on An (K), which is called the
Zariski topology on An (K).
(ii) If A ⊂ An (K) is any subset, the intersection of all algebraic sets containing
A is the smallest algebraic set containing A. We denote this set by A. In
terms of the Zariski topology, A is the closure of A.
(iii) If A ⊂ An (K) is any subset, the Zariski topology on An (K) induces a
topology on A, which is called the Zariski topology on A.
(iv) Topological notions such as open, closed, dense, or neighborhood will
always refer to the Zariski topology.
Along with treating the geometry–algebra dictionary, we will state some
computational problems for ideals in polynomial rings arising from its entries.
These problems are not meant to be attacked by the reader. They rather serve
as a motivation for the computational tools developed in Chapter 2, where we
will present algorithms to solve the problems. The first explicit S INGULAR
examples based on the algorithms, however, will be presented in the current
chapter.
Problem 1.13 Give an algorithm for computing ideal intersections.
S INGULAR Example 1.14
> ring R = 0, (x,y,z), dp;
> ideal I = z; ideal J = x,y;
> ideal K = intersect(I,J); K;
K[1]=yz
K[2]=xz
So V(z) ∪ V(x, y) = V(z ∩ x, y) = V(xz, yz).

Remark 1.15 The previous example is special in that we consider ideals

that are monomial. The intersection of monomial ideals is obtained using a
simple recipe: Given I = m1 , . . . , mr  and J = m′1 , . . . , m′s  in K[x1 , . . . , xn ],
with monomial generators mi and m′j , the intersection I ∩ J is generated by the
 1.1 Affine Algebraic Geometry 19

least common multiples lcm(mi , m′j ). In particular, I ∩ J is monomial again.

See Section 2.2.4 for the general algorithm.

Our next step in relating algebraic sets to ideals is to define some kind of
inverse to the map V:

Definition 1.16 If A ⊂ An (K) is any subset, the ideal

I(A) := { f ∈ K[x1 , . . . , xn ] | f (p) = 0 for all p ∈ A}

is called the vanishing ideal of A.

We summarize the properties of I and start relating I to V:

Proposition 1.17 Let R = K[x1 , . . . , xn ].

/ = R. If K is infinite, then I(An (K)) = 0.

(i) I(0)
(ii) If A ⊂ B are subsets of An (K), then I(A) ⊃ I(B).
(iii) If A, B are subsets of An (K), then

I(A ∪ B) = I(A) ∩ I(B).

(iv) For any subset A ⊂ An (K), we have

V(I(A)) = A.

(v) For any subset I ⊂ R, we have

I(V(I)) ⊃ I.

Proof Properties (ii), (iii), and (v) are easy consequences of the definitions.
The first statement in (i) is also clear. For the second statement in (i), let K be
infinite, and let f ∈ K[x1 , . . . , xn ] be any nonzero polynomial. We have to show
that there is a point p ∈ An (K) such that f (p) = 0. By our assumption on K,
this is clear for n = 1 since every nonzero polynomial in one variable has at
most finitely many zeros. If n > 1, write f in the form f = c0 (x1 , . . . , xn−1 ) +
c1 (x1 , . . . , xn−1 )xn + . . . + cs (x1 , . . . , xn−1 )xns . Then ci is nonzero for at least
one i. For such an i, we may assume by induction that there is a point p′ ∈
An−1 (K) such that ci (p′ ) = 0. Then f (p′ , xn ) ∈ K[xn ] is nonzero. Hence, there
is an element a ∈ K such that f (p′ , a) = 0. This proves (i). For (iv), note that
V(I(A)) ⊃ A. Let, now, V(T ) be any algebraic set containing A. Then f (p) = 0
for all f ∈ T and all p ∈ A. Hence, T ⊂ I(A) and, thus, V(T ) ⊃ V(I(A)), as
desired.
20 The Geometry–Algebra Dictionary

Property (iv) above expresses V(I(A)) in terms of A. Likewise, we wish to

express I(V(I)) in terms of I. The following example shows that the contain-
ment I(V(I)) ⊃ I may be strict.
Example 1.18 We have
I(V(xk )) = x for all k ≥ 1.
Definition 1.19 Let R be any ring, and let I ⊂ R be an ideal. Then the set
√
I := { f ∈ R | f k ∈ I for some k ≥ 1}
√
is an ideal of R containing I. It is called the radical of I. If I = I, then I is
called a radical ideal.
Example 1.20 Consider a principal ideal of K[x1 , . . . , xn ]: If
µ
f = f1 1 · · · fsµs ∈ K[x1 , . . . , xn ]
is the decomposition of a polynomial into irreducible factors, then

 f  =  f1 · · · fs .
The product f1 · · · fs , which is uniquely determined by f up to multiplication
by a constant, is called the square-free part of f . If f = f1 · · · fs up to scalar,
we say that f is square-free.
Problem 1.21 Design an algorithm for computing radicals.
The computation of radicals will be treated in Section 2.4.
S INGULAR Example 1.22
> LIB "primdec.lib"; // provides the command radical
> ring R = 0, (x,y,z), dp;
> poly p = z2+1; poly q = z3+2;
> ideal I = p*q^2,y-z2;
> ideal radI = radical(I);
> I;
I[1]=z8+z6+4z5+4z3+4z2+4
I[2]=-z2+y
> radI;
radI[1]=z2-y
radI[2]=y2z+z3+2z2+2

1.1.3 Hilbert’s Nullstellensatz

√
It is clear from the definitions that I(V(I)) ⊃ I. But even this containment
may be strict:
 1.1 Affine Algebraic Geometry 21

Example 1.23 The polynomial 1 + x2 ∈ R[x] has no real root. Hence, consid-
ering the ideal of the real vanishing locus, we get

I(V(1 + x2 )) = I(0)
/ = R[x].

Here, by the fundamental theorem of algebra, we may remedy the situation

by allowing complex roots as well. More generally, given any field K, we may
work over the algebraic closure K of K. Then, by the very definition of K, every
nonconstant polynomial in one variable has a root. This fact has a multivariate
analog:

Theorem 1.24 (Hilbert’s Nullstellensatz, Weak Version) Let I ⊂ K[x1 , . . . , xn ]

be an ideal, and let K be the algebraic closure of K. Formally, regard I as
a subset of the larger polynomial ring K[x1 , . . . , xn ]. Then the following are
equivalent:

(i) The vanishing locus V(I) of I in An (K) is empty.

(ii) 1 ∈ I; that is, I = K[x1 , . . . , xn ].

The proof will be given in Section 1.1.8.

Problem 1.25 Design a test for checking whether 1 is in I.

S INGULAR Example 1.26
> ring R = 0, (x,y,z), dp;
> ideal I;
> I[1]=972x2+948xy+974y2+529xz+15yz-933z2+892x-483y-928z-188;
> I[2]=-204x2-408xy-789y2-107xz+543yz-762z2-528x-307y+649z-224;
> I[3]=998x2+7xy-939y2-216xz+617yz+403z2-699x-831y-185z-330;
> I[4]=688x2+585xy-325y2+283xz-856yz+757z2+152x-393y+386z+367;
> I[5]=464x2+957xy+962y2+579xz-647yz-142z2+950x+649y+49z+209;
> I[6]=-966x2+624xy+875y2-141xz+216yz+601z2+386x-671y-75z+935;
> I[7]=936x2-817xy-973y2-648xz-976yz+908z2+499x+773y+234z+35;
> I[8]=-574x2+560xy-199y2+623yz+146z2-821x-99y+166z+711;
> I[9]=124x2-751xy-745y2+678xz-47yz+326z2-447x+462y+225z+579;
> I[10]=902x2+383xy-828y2+865xz-433yz-137z2-265x+913y-928z-400;
> groebner(I);
_[1]=1

Problem 1.25 is a special instance of the following problem:

Problem 1.27 (Ideal Membership Problem) Design a test for checking whe-
ther a given f ∈ K[x1 , . . . , xn ] is in I.

Remark 1.28 Let I ⊂ K[x1 , . . . , xn ] be a monomial ideal, given by monomial

generators m1 , . . . , mr . Then a monomial is contained in I iff it is divisible by at
least one of the mi . If f ∈ K[x1 , . . . , xn ] is any nonzero polynomial, we write it
22 The Geometry–Algebra Dictionary

as a K-linear combination of different monomials, with nonzero scalars. Then

f ∈ I iff the respective monomials are contained in I. See Section 2.2.1 for the
general algorithm.

Now, we discuss a second version of the Nullstellensatz which settles our

question of how to express I(V(I)) in terms of I:

Theorem 1.29 (Hilbert’s Nullstellensatz, Strong Version) Let K = K, and let

I ⊂ K[x1 , . . . , xn ] be an ideal. Then
√
I(V(I)) = I.
√
Proof As already said earlier, I ⊂ I(V(I)). For the reverse inclusion, let
f ∈ I(V(I)), and let f1 , . . . , fr be generators for I. Then f vanishes on V(I),
and we have to show that f k = g1 f1 + · · · + gr fr for some k ≥ 1 and some
g1 , . . . , gr ∈ K[x1 , . . . , xn ] =: R. We use the trick of Rabinowitch. Consider the
ideal
J :=  f1 , . . . , fr , 1 − t f  ⊂ R[t],

where t is an extra variable. We show that V(J) ⊂ An+1 (K) is empty. Sup-
pose on the contrary that p = (a1 , . . . , an , an+1 ) ∈ V(J) is a point, and set p′ =
(a1 , . . . , an ). Then f1 (p′ ) = · · · = fr (p′ ) = 0, so that p′ ∈ V(I), and an+1 f (p′ ) =
1. This contradicts the fact that f vanishes on V(I).
From the weak Nullstellensatz, we conclude that 1 ∈ J. Then we have 1 =
∑ri=1 hi fi + h(1 − t f ) for suitable h1 , . . . , hr , h ∈ R[t]. Substituting 1/ f for t in
this expression and multiplying by a sufficiently high power f k to clear de-
nominators, we get a representation f k = ∑ri=1 gi fi as desired.

Corollary 1.30 If K = K, then I and V define a one-to-one correspondence

{algebraic subsets of An (K)}

V↑↓I
{radical ideals of K[x1 , . . . , xn ]}.
As we will see more clearly in Section 2.2.3, the trick of Rabinowitch allows
us to solve the radical membership problem:

Corollary 1.31 (Radical Membership) Let K be any field, let I ⊂ K[x1 , . . . , xn ]

be an ideal, and let f ∈ K[x1 , . . . , xn ]. Then:
√
f ∈ I ⇐⇒ 1 ∈ J := I, 1 − t f  ⊂ K[x1 , . . . , xn ,t],

where t is an extra variable.

 1.1 Affine Algebraic Geometry 23

Based on the Nullstellensatz, we can express geometric properties in terms

of ideals. Here is a first example of how this works:

Proposition 1.32 Let K be any field, and let I ⊂ K[x1 , . . . , xn ] be an ideal.

The following are equivalent:

(i) The vanishing locus V(I) of I in An (K) is finite.

(ii) For each i, 1 ≤ i ≤ n, we have I ∩ K[xi ]  0.

Problem 1.33 Design a test for checking whether (ii) holds.

In Section 1.1.6, we will see that (ii) holds iff the quotient ring K[x1 , . . . , xn ]/I
is finite-dimensional as a K-vector space. How to compute the vector space di-
mension is a topic of Section 2.3.

Example 1.34 Taking the symmetry of the generators into account, the com-
putation in Example P.7 shows that the ideal

I = x + y + z − 1, x2 + y2 + z2 − 1, x3 + y3 + z3 − 1
⊂ Q[x, y, z]

contains the polynomials

z3 − z2 , y3 − y2 , x3 − x2 .

1.1.4 Irreducible Algebraic Sets

As we saw earlier, the vanishing locus V(xz, yz) ⊂ A3 (R) is the union of the
xy-plane and the z-axis:

Definition 1.35 A nonempty algebraic set A ⊂ An (K) is called irreducible,

or a subvariety of An (K), if it cannot be expressed as the union A = A1 ∪ A2 of
algebraic sets A1 , A2 properly contained in A. Otherwise, A is called reducible.

Proposition 1.36 Let A ⊂ An (K) be an algebraic set. Then the following are
equivalent:
24 The Geometry–Algebra Dictionary

(i) A is irreducible.
(ii) I(A) is a prime ideal.

Problem 1.37 Design a test for checking whether a given ideal of K[x1 , . . . , xn ]
is prime.

Corollary 1.38 If K = K, then I and V define a one-to-one correspondence

{subvarieties of An (K)}
V↑↓I
{prime ideals of K[x1 , . . . , xn ]}.

Proposition 1.39 If K = K, then I and V define a one-to-one correspondence

{points of An (K)}
V↑↓I
{maximal ideals of K[x1 , . . . , xn ]}.
Here is the main result in this section:

Theorem 1.40 Every nonempty algebraic set A ⊂ An (K) can be expressed

as a finite union
A = V1 ∪ · · · ∪Vs

of subvarieties Vi . This decomposition can be chosen to be minimal in the sense

that Vi ⊃ V j for i = j. The Vi are, then, uniquely determined and are called the
irreducible components of A.

Proof The main idea of the proof is to use Noetherian induction: Assuming
that there is an algebraic set A ⊂ An (K) which cannot be written as a finite
union of irreducible subsets, we get an infinite descending chain of subvarieties
Vi of A:
A ⊃ V1  V2  · · ·

This contradicts the ascending chain condition in the polynomial ring since
taking vanishing ideals is inclusion reversing.

Problem 1.41 Design an algorithm for finding the irreducible components of

a given algebraic set.

The algebraic concept of primary decomposition, together with algorithms

for computing such decompositions, gives an answer to both Problems 1.41
and 1.37. See Section 2.4.
 1.1 Affine Algebraic Geometry 25

If K is a subfield of C, and if all irreducible components in An (C) are points

(that is, we face a system of polynomial equations with just finitely many com-
plex solutions), we may find the solutions via triangular decomposition. This
method combines lexicographic Gröbner bases with univariate numerical solv-
ing. See Decker and Lossen (2006).
S INGULAR Example 1.42
> ring S = 0, (x,y,z), lp;
> ideal I = x2+y+z-1, x+y2+z-1, x+y+z2-1;
> LIB "solve.lib";
> def R = solve(I,6); // creates a new ring in which the solutions
. // are defined; 6 is the desired precision
> setring R; SOL;
//-> [1]: [2]: [3]: [4]: [5]:
//-> [1]: [1]: [1]: [1]: [1]:
//-> 0.414214 0 -2.414214 1 0
//-> [2]: [2]: [2]: [2]: [2]:
//-> 0.414214 0 -2.414214 0 1
//-> [3]: [3]: [3]: [3]: [3]:
//-> 0.414214 1 -2.414214 0 0
In this simple example, the solutions can also be read off from a lexicographic
Gröbner basis as in Example P.7:

> groebner(I);
//-> _[1]=z6-4z4+4z3-z2 _[2]=2yz2+z4-z2
//-> _[3]=y2-y-z2+z _[4]=x+y+z2-1

1.1.5 Removing Algebraic Sets

The set-theoretic difference of two algebraic sets need not be an algebraic set:

Example 1.43 Consider again the union of the xy-plane and the z-axis in
A3 (R):

Removing the plane, the residual set is the punctured z-axis, which is not de-
fined by polynomial equations. Indeed, if a polynomial f ∈ R[x, y, z] vanishes
on the z-axis except possibly at the origin o, then the univariate polynomial
26 The Geometry–Algebra Dictionary

g(t) := f (0, 0,t) has infinitely many roots since R is infinite. Hence, g = 0 (see
the proof of Proposition 1.17), so that f vanishes at o, too.
In what follows, we explain how to find polynomial equations for the Zariski
closure of the difference of two algebraic sets; that is, for the smallest algebraic
set containing the difference. We need:
Definition 1.44 Let I, J be two ideals of a ring R. Then the sets
I : J := { f ∈ R | f g ∈ I for all g ∈ J}
and
∞
I : J ∞ := { f ∈ R | f J k ⊂ I for some k ≥ 1} = (I : J k )
k=1

are ideals of R containing I. They are called the ideal quotient of I by J and
the saturation of I with respect to J, respectively.
Problem 1.45 Design algorithms for computing ideal quotients and satura-
tion.
Since the polynomial ring K[x1 , . . . , xn ] is Noetherian by Hilbert’s basis the-
orem, and since I : J k = (I : J k−1 ) : J for any two ideals I, J ⊂ K[x1 , . . . , xn ], the
computation of I : J ∞ just means iterating the computation of ideal quotients:
the ascending chain
I : J ⊂ I : J2 ⊂ · · · ⊂ I : Jk ⊂ · · ·
is eventually stationary. We will discuss how to compute ideal quotients in
Section 2.2.5.
Theorem 1.46 Let I, J be ideals of K[x1 , . . . , xn ]. Then, considering vanishing
loci and the Zariski closure in An (K), we have
V(I) \ V(J) = V(I : J ∞ ) ⊂ An (K).
If I is a radical ideal, then
V(I) \ V(J) = V(I : J) ⊂ An (K).
The theorem is another consequence of Hilbert’s Nullstellensatz. See Decker
and Schreyer (2013).
S INGULAR Example 1.47 We illustrate the geometry of ideal quotients by
starting from an ideal I which defines the intersection of the curve C = V(y −
(x − 1)3 (x − 2)) ⊂ A2 (R) with the x-axis:
 1.1 Affine Algebraic Geometry 27

> ring R = 0, (x,y), dp;

> ideal I = y-(x-1)^3*(x-2), y;

> ideal GI = groebner(I); GI;

GI[1]=y
GI[2]=x4-5x3+9x2-7x+2
> factorize(GI[2]);
[1]:
_[1]=1
_[2]=x-1
_[3]=x-2
[2]:
1,3,1

There are two intersection points, namely p = (0, 1) and q = (0, 2). The ideal
J = (x − 1)(x − 2) defines a pair of parallel lines which intersect the x-axis
precisely at p and q, respectively. We compute the ideal quotient I1 = I : J:

> ideal J = (x-1)*(x-2);

> ideal I1 = quotient(I,J); I1;
I1[1]=y
I1[2]=x2-2x+1
> factorize(I1[2]);
[1]:
_[1]=1
_[2]=x-1
[2]:
1,2

The resulting ideal I1 defines the intersection of the parabola C1 = V(y − (x −

1)2 ) with the x-axis which consists of the point p = (0, 1) only. In fact, the
x-axis is the tangent to C1 at p:
28 The Geometry–Algebra Dictionary

Computing the ideal quotient I2 = I1 : J, we may think of the result as defining

the intersection of a line with the x-axis at p:

> ideal I2 = quotient(I1,J); I2;

I2[1]=y
I2[2]=x-1

A final division also removes p:

> ideal I3 = quotient(I2,J); I3;

I3[1]=1

S INGULAR Example 1.48 To simplify the output in what follows, we work

over the field with 2 elements:
 1.1 Affine Algebraic Geometry 29

> ring R = 2, (x,y,z), dp;

> poly F = x5+y5+(x-y)^2*xyz;
> ideal J = jacob(F); J;
//-> J[1]=x4+x2yz+y3z J[2]=y4+x3z+xy2z J[3]=x3y+xy3
> maxideal(2);
//-> _[1]=z2 _[2]=yz _[3]=y2 _[4]=xz _[5]=xy _[6]=x2
> ideal H = quotient(J,maxideal(2)); H;
//-> H[1]=y4+x3z+xy2z H[2]=x3y+xy3 H[3]=x4+x2yz+y3z
//-> H[4]=x3z2+x2yz2+xy2z2+y3z2 H[5]=x2y2z+x2yz2+y3z2
//-> H[6]=x2y3
> H = quotient(H,maxideal(2)); H;
H[1]=x3+x2y+xy2+y3
H[2]=y4+x2yz+y3z
H[3]=x2y2+y4
> H = quotient(H,maxideal(2)); H;
H[1]=x3+x2y+xy2+y3
H[2]=y4+x2yz+y3z
H[3]=x2y2+y4

> LIB "elim.lib"; // provides the command sat

> int p = printlevel;
> printlevel = 2; // print more information while computing
> sat(J,maxideal(2));
// compute quotient 1
// compute quotient 2
// compute quotient 3
// saturation becomes stable after 2 iteration(s)
[1]:
_[1]=x3+x2y+xy2+y3
_[2]=y4+x2yz+y3z
_[3]=x2y2+y4
[2]:
2
> printlevel = p; // reset printlevel

1.1.6 Polynomial Maps

Since algebraic sets are defined by polynomials, it should not be a surprise that
the maps relating algebraic sets to each other are also defined by polynomials:

Definition 1.49 Let A ⊂ An (K) and B ⊂ Am (K) be (nonempty) algebraic

sets. A map ϕ : A → B is called a polynomial map, or a morphism, if there
are polynomials f1 , . . . , fm ∈ K[x1 , . . . , xn ] such that ϕ (p) = ( f1 (p), . . . , fm (p))
for all p ∈ A.

In other words, a map A → B is a polynomial map iff its components are

30 The Geometry–Algebra Dictionary

restrictions of polynomial functions on An (K) to A. Every such restriction is

called a polynomial function on A.
Given two polynomial functions p → f (p) and p → g(p) on A, we may
define their sum and product according to the addition and multiplication in K:
send p to f (p) + g(p) and to f (p) · g(p), respectively. In this way, the set of all
polynomial functions on A becomes a ring, which we denote by K[A]. Since
this ring is generated by the coordinate functions p → xi (p), we define:

Definition 1.50 Let A ⊂ An (K) be a (nonempty) algebraic set. The coordi-

nate ring of A is the ring of polynomial functions K[A] defined above.

Note that K may be considered as the subring of K[A] consisting of the

constant functions. Hence, K[A] is naturally a K-algebra. Next, observe that
each morphism ϕ : A → B of algebraic sets gives rise to a homomorphism

ϕ ∗ : K[B] → K[A], g → g ◦ ϕ ,

of K-algebras. Conversely, given any homomorphism φ : K[B] → K[A] of K-

algebras, one can show that there is a unique polynomial map ϕ : A → B such
that φ = ϕ ∗ . Furthermore, defining the notion of an isomorphism as usual by
requiring that there exists an inverse morphism, it turns out that ϕ : A → B is
an isomorphism of algebraic sets iff ϕ ∗ is an isomorphism of K-algebras.

Example 1.51 Let C = V(y − x2 , xy − z) ⊂ A3 (R) be the twisted cubic curve.

The map
A1 (R) → C, t → (t,t 2 ,t 3 ),

is an isomorphism with inverse map (x, y, z) → x.

By relating algebraic sets to rings, we start a new section in the geometry–

algebraic dictionary. To connect this section to the earlier ones, where we re-
lated algebraic sets to ideals, we recall the definition of a quotient ring:

Definition 1.52 Let R be a ring, and let I be an ideal of R. Two elements

f , g ∈ R are said to be congruent modulo I if f − g ∈ I. In this way, we get
an equivalence relation on R. We write f = f + I for the equivalence class of
f ∈ R, and call it the residue class of f modulo I. The set R/I of all residue
classes becomes a ring, with algebraic operations

f + g = f + g and f · g = f · g.

We call R/I the quotient ring of R modulo I.

 1.1 Affine Algebraic Geometry 31

Now, returning to the coordinate ring of an affine algebraic set A, we note

that two polynomials f , g ∈ K[x1 , . . . , xn ] define the same polynomial function
on A iff their difference is contained in the vanishing ideal I(A). We may, thus,
identify K[A] with the quotient ring K[x1 , . . . , xn ]/I(A), and translate geometric
properties expressed in terms of I(A) into properties expressed in terms of
K[A]. For example:

• A is irreducible ⇐⇒ I(A) is prime ⇐⇒ K[x1 , . . . , xn ]/I(A) is an integral

domain.

For another example, let I ⊂ K[x1 , . . . , xn ] be any ideal. Then, as one can show,
Proposition 1.32 can be rewritten as follows:

• The vanishing locus V(I) of I in An (K) is finite ⇐⇒ the K-vector space

K[x1 , . . . , xn ]/I is finite-dimensional.

Definition 1.53 A ring of type K[x1 , . . . , xn ]/I, with I an ideal of K[x1 , . . . , xn ],

is called an affine K-algebra, or simply an affine ring.

With regard to computational aspects, we should point out that it were cal-
culations in affine rings which led Buchberger to design his Gröbner basis
algorithm. In fact, to implement the arithmetic operations in K[x1 , . . . , xn ]/I,
we may fix a monomial ordering > on K[x1 , . . . , xn ], represent each residue
class by a normal form with respect to > and I, and add and multiply residue
classes by adding and multiplying normal forms. Computing normal forms, in
turn, amounts to computing remainders on multivariate polynomial division by
the elements of a Gröbner basis for I with respect to >. See Algorithm 1 and
Proposition 2.27 in Chapter 2.
S INGULAR Example 1.54
> ring R = 0, (z,y,x), lp;
> ideal I = y-x2, z-xy;
> qring S = groebner(I); // defining a quotient ring
> basering; // shows current ring
// characteristic : 0
// number of vars : 3
// block 1 : ordering lp
// : names z y x
// block 2 : ordering C
// quotient ring from ideal
_[1]=y-x2
_[2]=z-yx
> poly f = x3z2-4y4z+x4;
> reduce(f,groebner(0)); // division with remainder
-4x11+x9+x4
32 The Geometry–Algebra Dictionary

Closely related to normal forms is a result of Macaulay which was a ma-

jor motivation for Buchberger and his thesis advisor Gröbner. Together with
Buchberger’s algorithm, this result allows one to find explicit K-vector space
bases for affine rings K[x1 , . . . , xn ]/I (and, thus, to determine the vector space
dimension). In fact, as in Gordan’s proof of the basis theorem, one can use
Gröbner bases to reduce the case of an arbitrary ideal I to that of a monomial
ideal (see Theorem 2.55 for a precise statement).

S INGULAR Example 1.55 In Example P.7, we computed a lexicographic

Gröbner basis GI for the ideal

I = x + y + z − 1, x2 + y2 + z2 − 1, x3 + y3 + z3 − 1
⊂ Q[x, y, z].

By inspecting the elements of GI, we saw that the system defined by the three
generators of I has precisely the three solutions

(1, 0, 0), (0, 1, 0), (0, 0, 1)

in A3 (Q). In particular, there are only finitely many solutions. As said earlier in
this section, this means that the Q-vector space Q[x, y, z]/I has finite dimension.
We check this using S INGULAR:

> vdim(GI); // requires Groebner basis

6

In general, if finite, the dimension d = dimK (K[x1 , . . . , xn ]/I) is an upper

bound for the number of points in the vanishing locus of I in An (K). In fact,
given a point p ∈ V(I) ⊂ An (K), there is a natural way of assigning a multiplic-
ity to the pair (p, I). Counted with multiplicity, there are exactly d solutions.
See Section 1.1.9.

1.1.7 The Geometry of Elimination

The image of an affine algebraic set under a morphism need not be an algebraic
set3 4 :

Example 1.56 The projection map (x, y) → y sends the hyperbola V(xy − 1)
onto the punctured y-axis:
3 We may, however, represent the image as a constructible set. See Kemper (2007) for an
algorithmic approach.
4 In projective algebraic geometry, morphisms are better behaved.
Exploring the Variety of Random
Documents with Different Content
 A pretty good thing, and I suppose there’s nursemaids and
governesses for all the children, too, hey?

Maxwell
Why, as a matter of fact there are—but not exactly in the way
you mean. You see, they run a sort of a boarding school, too, the
people that run this model tenement do, and they take care of all the
children there—keep ’em night and day, feed ’em and dress ’em and
teach ’em trades and all that sort of thing. They let them come to see
you on Sunday, but you’re relieved of all responsibility and your wife
of all the trouble.

Ryan
I see. But what’s to become of my job when I’m going off to this
here model tenement to sleep nights? Don’t you know that a janitor
has to sleep in the building he’s taking care of?

Maxwell
Well, you’re not in love with your job, are you? You’d be willing to
do some other work where there was a chance of advancement and
better pay, wouldn’t you?

Ryan
Sure I would. Where am I going to get it?

Maxwell
Why, as to that, I guess Mr. Morris or I could get you a job
somewhere. You’re a big husky fellow and pretty steady, I know. I
could get you a job in the shipping department of a factory I am
interested in, if you wanted it.

Ryan
What would the hours be?

Maxwell (laughing)
 Why, you’re going pretty fast, Ryan. I don’t know just what the
hours would be yet, but I suppose they’d be from half past eight to
six or something like that. That’s easier than your hours now, isn’t it?

Ryan
Much easier. Now, what has all this got to do with my wife having
too many children?

Maxwell
Why, these people think that Mrs. Ryan has had too many
children. Some of these people are doctors, and they say it’s bad for
a woman’s health to have so many children.

Ryan
I see. Well, what would you like to have me do about it?

Maxwell
About the model tenement?

Ryan
No, about my wife having any more children.

Maxwell
Well, you know—you see I’m not—what they think is— Oh,
damn it, I don’t know about this part of the thing, Ryan. It’s out of my
line. I’m not a philanthropist. I’m just talking for these people
because they know I know you. About that question you’ll have to
talk to Mrs. Dannenberg or Mr. Morris.

Ryan
Mr. Morris—he’s the little guy that was standing on a chair when
I came in to-night, ain’t he?

Maxwell
Yes.
 Ryan
Is he a doctor?

Maxwell
No, I don’t think he is.

Ryan
He came into my place the other day selling some sort of a little
doctor book.

Maxwell
I know; he does that because he’s a philanthropist.

Ryan
Do people make a living by being philanthropists?

Maxwell
No—well, as a matter of fact, many people do make a pretty fat
living out of it, but Mr. Morris doesn’t. He’s a sculptor—a man that
makes statues, you know.

Ryan
Oh, I know what a sculptor is, all right. What statues did Mr.
Morris make?

Maxwell
Why, he made that statue over in the corner, for one thing.
(Ryan goes over and examines the purple statue.)

Ryan
Do you like this statue, Mr. Johnson?

Maxwell (laughing)
 I’m not a judge of such things, Ryan. But people who know about
art say that it’s very good indeed.

Ryan
Is it a good likeness?

Maxwell
It’s not supposed to be a portrait; it’s a sort of a fancy statue. It
represents the “Emancipation of Woman”—woman freed, you know.

Ryan
Freed from what?

Maxwell
Why, from overwork, and excessive child-bearing and all that
sort of thing, you know.

Ryan
Well, that lady’s freed from child-bearing, all right. She’s perfectly
safe on that score. (He goes back to his chair and pours out another
drink.) But now let’s get back to your friends’ proposition. I’m to get a
day job and come home nights—is that right?

Maxwell
That’s right.

Ryan
The children are taken off my wife’s hands, and she doesn’t have
to do any more cooking or washing or anything, hey?

Maxwell
That’s it, Ryan.

Ryan
 And the place we’d live would be pretty much like this apartment,
would it?

Maxwell
Just as comfortable, at any rate.

Ryan
My wife would have no work to do; I’d work in the daytime and
come home nights—there’d be no kids to take care of—it would be
pretty much like the life that you and your wife have, wouldn’t it, Mr.
Johnson?

Maxwell
Ah—why, yes, Ryan, pretty much the same. What do you think of
the proposition?

Ryan
Well, I’ll tell you. Of course I wouldn’t like to have the children go
away—though they are a lot of trouble—but I suppose they’d be
better taken care of than we could do, so, if that was all there was to
it, I’d say go ahead. And it’d be all right for me, too, if I was a
bachelor. But it’s on account of my wife that I feel I’ve got to say,
“Excuse me!”

Maxwell
Why, you don’t begrudge your wife a little rest, do you, Ryan?

Ryan
Would I begrudge the old woman a little rest? Yes, by God, I
would, if a little rest meant having nothing to do all day except sit
around and talk to her friends and run around town. It’s just on that
account that I say nix to your whole proposition. Now you think I’m a
slave driver, I suppose. Well, I ain’t, Mr. Johnson, but I’ve lived with a
woman thirteen years, and what I’m telling you I didn’t get out of no
books—it’s facts!
 Listen, Mr. Johnson. What you want us to do is to live just the
sort of life that you and your wife live—no children to take care of, no
washing nor cooking nor nothing. Well, what I say is, excuse me!
That may be all right for you and your wife—she don’t need no
housework nor children nor nothing to keep her busy. Her time is full
of all sorts of useful things—I know that. (Maxwell stirs uneasily and
looks at the floor.) But with Annie, my wife, it’s a different proposition
altogether. She’s one of them women—and there’s a lot more of
them than you think—that can’t stand living in a nice regular
apartment with nothing to do. I know because we tried it when we
was first married. It’s like what the old fellow said: “Satan finds some
mischief still for idle hands to do”!

Maxwell
True enough! And—

Ryan
Listen, Mr. Johnson. When we was first married, Annie was one
of these here idle hands they tell of. She was an idle hand for three
years, and Satan, as the old fellow said, certainly found some
mischief for her to do. Not anything real bad—there’s no real harm in
Annie—but it was mischief all right. For the first three years we was
living in a nice apartment in Brooklyn. I wasn’t a janitor, then; I was
driving a truck. I was out all day and I got home every night at six or
seven o’clock. All Annie has to do is to get my breakfast and supper
and keep the place clean. Does she do it? Sure she does, for the
first four or five weeks! Then she gets to making friends with other
women in the building and going out to matinees and vaudeville and
all that sort of thing. That’s all right—I can afford it—I don’t care if
she has a good time; but then what does she do? I give her two
dollars in the morning to go out and buy a good supper for me when
I come home. I come home and she ain’t in yet, and the lady in the
next flat gives me the groceries that she’s sent home. And what is
they? A little chipped beef and a box of Saratoga chips and some
baker’s bread. About fifty-cent’s worth. When she gets home I ask
her where she’s been. Why, Mrs. Eindorfer has took her to a
spiritualist meeting, and she’s spent the rest of that money to look
into a glass ball or have her fortune told or some such foolishness.
Now this goes on for nearly three years. It ain’t all spiritualists’
meetings; it’s all sorts of things. She makes all sorts of friends,
women and men, too; I had to beat a couple of ’em up. The flat
wasn’t kept up; I run into debt; my meals wasn’t cooked right or on
time, and Annie was half sick all the time just from running around
entertaining herself. I ain’t blaming her. She wasn’t to blame. And
what was to blame? The apartment house was to blame. When
Peter was born, after we’d been married three years, and I gave up
trucking and moved out of that apartment house and got a job as
janitor, everything was all right. And everything’s been all right ever
since.

Maxwell (thoughtfully)
And the apartment house was to blame?

Ryan
Believe me, Mr. Johnson, the only part of an apartment house to
live in is the basement, where you can have a regular home. I been
a janitor for ten years, and I seen these apartment houses do queer
things to families. They don’t seem to have no children when they
live in apartment houses, that’s one thing. And there ain’t no coal to
bring up and the washing goes out, and there ain’t nothing for them
to do but just make fools of themselves. And sometimes there’s a
good many divorces been caused by these here apartment houses.
And there’d be a good many more divorces if a lot of husbands knew
what went on when they was downtown at business.
Understand me, Mr. Johnson. I don’t mean you and your wife at
all. You ain’t that sort of people, but what I do say is for my wife, and
for a lot of women with more education and more money than she’s
got, the only sort of life is doing housework and taking care of
children all day long. So Annie and me will stay down in the
basement, much obliged to you, unless we go out of New York to live
in a little house in the country sometime. And Annie’ll have just as
much work to do there. She’s one of them women that wasn’t meant
to be idle. And now I guess I’ll go downstairs.
(He rises and goes toward the door. Maxwell sits silent for a
moment and then rises a little unsteadily. He holds out his hand to
Ryan for a second and then drops it and starts.)

Maxwell
Ryan, I—er—why, I guess you’re right, after all. I’ll tell my friends
what you said.

Ryan
All right. No hard feelings, I hope.

Maxwell
Not at all; that’s all right. Good night, Ryan.

Ryan
Good night, Mr. Johnson.
(He goes out.)

Maxwell
There’s a man that’s master in his own home, at any rate. (He
lights a cigar and walks around the room with his hands in his
trousers pockets, coming to a halt in front of the purple statue. He
looks at it reflectively.) Satan finds some mischief still—(A pause)—
for idle hands to do. For idle hands to do. For idle hands to do. For
idle hands—
(The doorbell is rung violently. Maxwell starts and runs out into
the hall. He returns with Helen, who is very much out of breath. As
she runs into the room the combs drop from her hair, which falls over
her face and shoulders. She throws herself on the chaise-longue.
Maxwell sits beside her and tries to push her hair back from her
face.)
Nellie! What’s the matter?
 (Helen sobs without answering.)

Helen
It’s those nasty Martins and that nasty policeman and that nasty
Lionel Morris.
(The bell rings again. Maxwell goes to the door and admits
Lionel.)

Lionel
Oh, I’m awfully glad you got back all right, Miss White. I jumped
into a taxi as soon as that brute of a policeman came, and then I met
all the rest of the crowd at the studio and everybody said, “Where’s
Miss White?” So I came right up here to find out if you’d got home.

Maxwell
For God’s sake, will somebody tell me what’s happened?

Lionel (sitting on the pianola bench)

Why, you see—

Helen
Be still. You see, Max, we were all at the Mortons’ studio, and
Adrian Wolfe made a speech about those nasty striking miners in
California or wherever it is that everybody is wearing mourning for
and parading and all that and this—and Mr. Morris said: “Let’s walk
up to Union Square and hold an open air meeting to protest.” So we
went up there and I made a speech and there was a crowd and I
saw a policeman there, but I thought it would be all right, and then
Mr. Morris made a speech and he said something about trampling on
a bloodstained flag, and the policeman told him to stop, and he
called the policeman a myrmidon, and some more policemen came
and broke up the meeting, and he ran away and wouldn’t help me;
and I ran down into the subway, and I don’t see how he dares show
his face in here!
 Maxwell
I’ll talk to him presently, but quiet down a little. You’d better go in
your room and fix up your hair.
(Helen rises to go out. She stops in the doorway and turns to
Maxwell.)

Helen
I won’t go to that nasty Amaranth this summer, Maxwell.

Maxwell
No, you won’t go to Amaranth.

Helen
Then will you get an automobile?

Maxwell
No, I won’t get an automobile.

Helen
Then what—

Maxwell
I am going to take that two thousand dollars and buy, with the
assistance of the building and loan association, a small house in a
city called Joplin, in the State of Missouri. It will not be a large house,
but I think that you will not find the time hanging heavy on your
hands. My brother has a wholesale grocery there, and I dare say he
will take me into the business, especially as I have a little money to
invest. And I’ll come home to luncheon every day. Missouri is a fertile
State. My brother has six children.

Lionel
But, Miss White—Mr. Johnson!
 (Helen goes down the hall to her room. Maxwell walks up to
Lionel.)

Maxwell
My wife’s name is not Miss White but Mrs. Johnson—Mrs.
Maxwell Johnson, of Joplin, Missouri. Get that? Do you know what
keeps me from dropping you down the elevator shaft?

Lionel
What—what do you mean?

Maxwell
It’s the janitor. Yes, Ryan, the fellow down in the basement with
nine children that you and Mrs. What’s-her-name wanted to
segregate. He told me all about you to-night. You’re nothing but a by-
product! The apartment house is the real devil in this pretty little play
—the apartment house is responsible for Feminism and Socialism
and Anarchism and Eugenics and pups like you. You’re just a sort of
bad substitute for the movies—that’s all you are. The apartment
house breeds the whole bunch of you—the apartment house and its
artificial, lazy, good-for-nothing life.
(Lionel starts toward the door hurriedly, but stops as if shot
when the telephone bell rings close to his ear. He comes back into
the room and Maxwell goes to the telephone.)

Maxwell
Hello!... What’s that?... Yes, this is Mr. Johnson.... No, I don’t
think so. Hold the wire and I’ll see. (With his hand over the
transmitter he looks into the room.) Nellie!

Helen (coming into the living room with her hair

in a long braid, wearing a blue tea gown)
Yes, Max?

Maxwell
 What is your name?

Helen
Why, Helen, of course, stupid.

Maxwell
Helen what?

Helen
Helen Johnson.

Maxwell
Not Miss Helen White?

Helen
No! No! No!

Maxwell (smiling)
Well, that’s all right, then. There’s a cop down-stairs with a
warrant for the arrest of a Miss Helen White and a Mr. Lionel Morris,
charged with making incendiary speeches in Union Square. They
think that Morris’s taxicab stopped at this building, and the policeman
is going through all the apartments. He’ll be here in a minute. (In the
receiver.) All right, Sam, it’s all right. Thanks for tipping me off.
(Maxwell reënters the room and sits on the pianola bench. Helen
reclines, with some dignity, on the chaise-longue. Lionel crouches
behind the purple statue.)

Maxwell (meditatively)
Satan finds some mischief still—

Helen
What are you saying, Max?

Maxwell
 Oh, I was just thinking of the janitor. I had quite a talk with him
after you left.
(The doorbell rings, and Maxwell admits a large policeman.)

Policeman
Excuse me, sir; it’s just a matter of form. I’m looking for a couple
of them Anarchist-Suffrage-I. W. W. bugs. It’s just a matter of form.
The man’s name is Lionel Morris and the woman’s name is Helen
White. Are you Lionel Morris?

Maxwell
No; my name is Maxwell Johnson. The janitor knows me, and so
do a lot of people in the building.

Policeman
Thank you, sir. It’s just a matter of form. Now, madam—it’s just a
matter of form—are you Helen White?

Helen
No, I am not Helen White. I am Mrs. Maxwell Johnson.

Policeman
Thank you, madam, thank you; it’s just a matter of form. You see
these parties is incendiaries; they called me a mermaiden. Now, just
two more questions—it’s just a matter of form: Is Miss Helen White
here?

Helen
No, Helen White is not here.

Maxwell
And I’m glad she isn’t here, officer.

Policeman
 You may well be that, sir; you may well be that. Now, is Lionel
Morris here?
(There is a pause, during which the purple statue shakes
slightly.)

Maxwell
Well, what do you think about it, officer? Take a look around the
place. Want to look in the dumb waiter or down the kitchen sink?

Policeman
Oh, I know he’s not here, Mr. Johnson, and I’m sorry to have
troubled you. Much obliged to you. Good night, sir.

Maxwell
Have a drink before you go?

Policeman
Well, I hadn’t ought to, but I guess I will, thanks.
(Helen pours the whiskey and Maxwell and the policeman lift
their glasses.)

Policeman
Well, here’s how, sir.

Maxwell
Here’s Joplin!

Helen
Oh, I’ll drink that.
(She takes a sip from Maxwell’s glass.)

Policeman
What’s that, something new?
 Maxwell
No, it’s old as Adam and Eve.

Policeman
Well, it’s a new one on me. Thank you, sir. Good night.

Maxwell
Good night. (The policeman goes out. After the door slams shut,
Lionel stands up, but remains behind the purple statue.) Hadn’t you
better go to some other apartment house? The cop’s gone down the
elevator. He’ll be gone by the time you get downstairs. (Lionel goes
out and as he turns he brushes against the purple statue, which
topples on its pedestal. He bangs the door shut after him violently,
and the statue falls to the floor and breaks into several pieces.
Maxwell and Helen look at it for a moment and then turn to each
other and laugh.)

Helen
Oh, look what’s happened to the “Emancipation of Woman”!

CURTAIN
 Transcriber’s Notes
Punctuation, hyphenation, and spelling were made
consistent when a predominant preference was found in the
original book; otherwise they were not changed.
Simple typographical errors were corrected; unbalanced
quotation marks were remedied when the change was
obvious, and otherwise left unbalanced.
Illustrations in this eBook have been positioned between
paragraphs and outside quotations. In versions of this eBook
that support hyperlinks, the page references in the List of
Illustrations lead to the corresponding illustrations.
*** END OF THE PROJECT GUTENBERG EBOOK JOYCE KILMER
***

Updated editions will replace the previous one—the old editions

will be renamed.

Creating the works from print editions not protected by U.S.

copyright law means that no one owns a United States copyright
in these works, so the Foundation (and you!) can copy and
distribute it in the United States without permission and without
paying copyright royalties. Special rules, set forth in the General
Terms of Use part of this license, apply to copying and
distributing Project Gutenberg™ electronic works to protect the
PROJECT GUTENBERG™ concept and trademark. Project
Gutenberg is a registered trademark, and may not be used if
you charge for an eBook, except by following the terms of the
trademark license, including paying royalties for use of the
Project Gutenberg trademark. If you do not charge anything for
copies of this eBook, complying with the trademark license is
very easy. You may use this eBook for nearly any purpose such
as creation of derivative works, reports, performances and
research. Project Gutenberg eBooks may be modified and
printed and given away—you may do practically ANYTHING in
the United States with eBooks not protected by U.S. copyright
law. Redistribution is subject to the trademark license, especially
commercial redistribution.

START: FULL LICENSE

THE FULL PROJECT GUTENBERG LICENSE
 PLEASE READ THIS BEFORE YOU DISTRIBUTE OR USE THIS WORK

To protect the Project Gutenberg™ mission of promoting the

free distribution of electronic works, by using or distributing this
work (or any other work associated in any way with the phrase
“Project Gutenberg”), you agree to comply with all the terms of
the Full Project Gutenberg™ License available with this file or
online at www.gutenberg.org/license.

Section 1. General Terms of Use and

Redistributing Project Gutenberg™
electronic works
1.A. By reading or using any part of this Project Gutenberg™
electronic work, you indicate that you have read, understand,
agree to and accept all the terms of this license and intellectual
property (trademark/copyright) agreement. If you do not agree to
abide by all the terms of this agreement, you must cease using
and return or destroy all copies of Project Gutenberg™
electronic works in your possession. If you paid a fee for
obtaining a copy of or access to a Project Gutenberg™
electronic work and you do not agree to be bound by the terms
of this agreement, you may obtain a refund from the person or
entity to whom you paid the fee as set forth in paragraph 1.E.8.

1.B. “Project Gutenberg” is a registered trademark. It may only

be used on or associated in any way with an electronic work by
people who agree to be bound by the terms of this agreement.
There are a few things that you can do with most Project
Gutenberg™ electronic works even without complying with the
full terms of this agreement. See paragraph 1.C below. There
are a lot of things you can do with Project Gutenberg™
electronic works if you follow the terms of this agreement and
help preserve free future access to Project Gutenberg™
electronic works. See paragraph 1.E below.
1.C. The Project Gutenberg Literary Archive Foundation (“the
Foundation” or PGLAF), owns a compilation copyright in the
collection of Project Gutenberg™ electronic works. Nearly all the
individual works in the collection are in the public domain in the
United States. If an individual work is unprotected by copyright
law in the United States and you are located in the United
States, we do not claim a right to prevent you from copying,
distributing, performing, displaying or creating derivative works
based on the work as long as all references to Project
Gutenberg are removed. Of course, we hope that you will
support the Project Gutenberg™ mission of promoting free
access to electronic works by freely sharing Project
Gutenberg™ works in compliance with the terms of this
agreement for keeping the Project Gutenberg™ name
associated with the work. You can easily comply with the terms
of this agreement by keeping this work in the same format with
its attached full Project Gutenberg™ License when you share it
without charge with others.

1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside
the United States, check the laws of your country in addition to
the terms of this agreement before downloading, copying,
displaying, performing, distributing or creating derivative works
based on this work or any other Project Gutenberg™ work. The
Foundation makes no representations concerning the copyright
status of any work in any country other than the United States.

1.E. Unless you have removed all references to Project

Gutenberg:

1.E.1. The following sentence, with active links to, or other

immediate access to, the full Project Gutenberg™ License must
appear prominently whenever any copy of a Project
Gutenberg™ work (any work on which the phrase “Project
Gutenberg” appears, or with which the phrase “Project
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookfinal.com