Automata and the Origins of Mechanism and Mechanistic Philosophy

Author(s): Derek J. de Solla Price

Source: Technology and Culture, Vol. 5, No. 1 (Winter, 1964), pp. 9-23
Published by: The Johns Hopkins University Press and the Society for the History of
Technology
Stable URL: https://www.jstor.org/stable/3101119
Accessed: 17-01-2019 20:54 UTC

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/3101119?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

The Johns Hopkins University Press, Society for the History of Technology are
collaborating with JSTOR to digitize, preserve and extend access to Technology and Culture

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
Automata in History *
AUTOMATA AND THE ORIGINS OF MECHANISM AND
MECHANISTIC PHILOSOPHY

DEREK J. DE SOLLA PRICE.,.

HISTORIANS OF THE Mechanistic Philosophy customarily proceed

from the reasonable assumption that certain theories in astronomy and
biology derived from man's familiarity with various machines and
mechanical devices. Using everyday technological artifacts one could
attempt with some measure of success to explain the motions of the
planets and the behavior of living aninrnls as having much of the
certainty and regularity reproduced in these physical models. Indeed,
the steady advancement of technology and the increase in familiarity
with machines and their fundamental theory is usually cited as the
decisive factor in the growth of mechanistic philosophy, especially
toward the beginning of the instrument-dominated Scientific Revolu-
tion in the sixteenth and seventeenth centuries.
It seems clear that any interpretation of the interaction between
the histories of technology and philosophy must assign a special and
nodal role to those peculiar mechanisms designed by ingenious artificers
to simulate the natural universe. In this light we shall now exainine
the history of such simulacra (i.e., devices that simulate) and automata
(i.e., devices that move by themselves), whose very existence offered
tangible proof, more impressive than any theory, that the natural uni-
verse of physics and biology was susceptible to mechanistic explication.
It is the burden of this paper to suggest further that in the history
of automata is found plain indication that the customary interpretation
puts the cart before the horse. Contrary to the popular belief that

"The following two papers were presented on January 19, 1963, at the William
Andrews Clark Memorial Library in Los Angeles, as the first and second parts
of a symposium on "Automata and Simulated Life as a Central Theme in the
History of Science " organized by the Library and the Division of Medical
History, University of California (Los Angeles). The authors acknowledge with
gratitude the efforts of Dr. C. Donald O'Malley, which made the meeting possible.
.,. Dr. Price is Avalon Professor of the History of Science at Yale University
and an Advisory Editor of Technology and Culture. He is the author of Science
Since Babylon (1961) and Little Science, Big Science (1963).
9

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
10 Derek /. de Solla Price
science proceeds from the simple to the complex, it seems as if mechan-
istic philosophy-or mechanicism, to use the appropriate term coined
by Dijksterhuis 1 -led to mechanism rather than the other way about.
We suggest that some strong innate urge toward mechanistic explana-
tion led to the making of automata, and that from automata has evolved
much of our technology, particularly the part embracing fine mechanism
and scientific instrumentation. When the old interpretation has been
thus reversed, the history of automata assumes an importance even
greater than before. In these special mechanisms are seen the pro-
genitors of the Industrial Revolution. In the augmenting success of
automata through the age of Descartes, and perhaps up to and including
the age of electronic computers, we see the prime tangible manifesta-
tion of the triumph of rational, mechanistic explanation over those
of the vitalists and theologians.
Our story, then, begins with the deep-rooted urge of man to simulate
the world about him through the graphic and plastic arts. The almost
magical, naturalistic rock paintings of prehistoric caves, the ancient
grotesque figurines and other "idols" found in burials, testify to the
ancient origin of this urge in primitive religion. It is clear that long
before the flowering of Greek civilization man had taken his first
faltering steps toward elaborating pictorial representations with auto-
mation. Chapuis 2 points to the development of dolls with jointed arms
and other articulated figurines such as those from ancient Egyptian
tombs (from the XII Dynasty onward) and takes these as proto-
automata. Interestingly enough, it is from just such figurines as these,
representing scenes of battle, in ships, in bakeries, and so on, that the
modern historian of technology often obtains his most valuable infor-
mation about the crafts and everyday life of deep antiquity. 3
Perhaps the next level of sophistication is also found in ancient
Egypt: talking statues worked by means of a speaking trumpet con-
cealed in hollows leading down from the mouth. Two such statues
are extant: a painted wooden head of the jackal God of the Dead is
preserved in the Louvre, and a large white limestone bust of the god
Re-Harmakhis of Lower Egypt is in the Cairo Museum and was
described in technical detail by Loukianoff. 4 Jointed and talking figures
1 E. ]. Dijksterhuis, The Mechanization of the World Picture (Oxford, 1961),

3n.
•Alfred Chapuis and Edmond Droz, Automata (New York, 1958), pp. 13-29.
•See Charles Singer et al., eds., A History of Technology, Vol. I (Oxford,
1954), 427, 437, plate 13 A.
• Gregoire Loukianoff, " Une statue parlante, ou Oracle du <lieu Re-Harmakhis,"
Annales du service des Antiquites de l'Egypte (Cairo, 1936), pp. 187-93.

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
 Automata and the Origins of Mechanism II

are not confined to Egypt but probably occurred early in civilization

and are widespread. The articulated masks to be worn over the face,
found in Africa, and the famous Wayang figures of flat, jointed leather
for traditional Indonesian shadow plays are pointers in this direction.
Primitive animism may lie at the very root of animation.
It seems that by the beginning of Greek culture the process of
natural exaggeration in mythology and legend had produced at least
the concept of simulacra able to do more than merely talk and move
their arms. Daedalus, as well as imitating the flight of birds, is said
(ps. Aristotle, De Anima, i, 3) to have contrived statues that moved
and walked in front of the Labyrinth, guarding it, and Archytas of
Tarentum (fourth century B. C.) is said to have made a flying dove of
wood worked by counterweights and air pressure. That such a tradi-
tion, supported by devices probably no more complex than those of
ancient Egypt, was taken seriously, is indicated by the use of moving
statues to deliver oracles and by their later Roman counterpart, the
neuropastes.
An impressive use of an automaton resulted from the murder of
Julius Caesar. On the day of Caesar's funeral, the city of Rome was
the scene of great confusion and tumult over the death of its idol.
Mark Antony was to deliver the funeral oration, and he was determined
to arouse the populace to take action against the conspirators. The
scene is vividly described by Walter:
An unendurable anguish weighed upon the quivering crowd. Their
nerves were strained to the breaking point. They seemed ready
for anything. And now a vision of horror struck them in all its
brutality. From the bier Caesar arose and began to tum around
slowly, exposing to their terrified gaze his dreadfully livid face
and his twenty-three wounds still bleeding. It was a wax model
which Antony had ordered in the greatest secrecy and which auto-
matically moved by means of a special mechanism hidden behind
the bed. 5
This realistic automaton did as much as Mark Anthony's words to
create a riot of the populace at the funeral, which contributed to one
of the greatest revolutions in history.
As a literary and imaginative theme, the simulacrum or statue that
comes magically to life without mechanical intervention has been
with us from the early legends of Vulcan and Pygmalion to the
medieval Golem of Jewish folklore, the Faustus legend, the affair of
Don Juan's father-in-law, and several miraculous animations of holy
•Gerard Walter, Caesar: A Biography (New York, 1952), p. 544.

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
12 Derek ]. de Solla Price
images. A variant tradition in which animation is secured by scientific
but non-mechanical artifice is seen in the homunculus of Paracelsus
that was to be hatched alchemically from a basis of semen nourished
by blood, and in the monster of Frankenstein in which lightning
supplied the electric vital fluid.
Although there seems to have been a continuous and strong tradition
leading man to simulate living animals and even man himself, in early
Greek times there did not exist the technological skill to materialize
this dream more extensively than in speaking tubes and simple jointed
arms. Perhaps the most crucial point in the early history of automata
is that this skill seems to have been acquired in the search for a different
variety of automaton, the astronomical model.
The prehistory of cosmological simulacra is apparently less extensive
than that of biological models and is later in appearing. Coming to
grips with the basic astronomical phenomena probably required a
level of sophistication considerably higher than was necessary for a
reasonably basic appreciation of the movement of living things. At all
events, the beginnings of astronomical representation may be seen in
the famous star-map ceilings of Egyptian tombs, the flower-pot shaped
clepsydras with their celestial ornamentation, and in the goddess Nut
arching over the celestial vault and providing primitive mechanism
for the disappearance of the Sun by swallowing it at the end of the
day's journey. Perhaps it is not altogether fanciful to see the astro-
nomical zodiac as the first primitive coming together of a cosmic model
and a set of animal models.
In the Babylonian area are found also representations of the celestial
bodies and the beginnings of a primitive pictorial notion of the structure
of the universe. From that same area comes moreover the highly
sophisticated but non-pictorial mathematical astronomy that achieved
the first spectacular success of scientific prediction, a prediction based
upon an acute sensitivity for the pattern of natural number rather
than for any perception of mechanism or even of geometrical form,
but nonetheless deterministic in its findings of precise and regular
order in the most common astronomical phenomena.
Babylonian theory was exquisitely complicated and probably never
understood in its entirety by any Greek, but the basic principle of
mathematical regularity and the fundamental parameters of the motions
could easily be comprehended and transmitted, so that they formed a
secure foundation for much of the science that flourished in that age
still called "the Greek Miracle." The almost mystic dominance of a
regularity of number, which led to Pythagoreanism, and the rationality
of celestial motions, translated from Babylonian form into the geo-

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
 Automata and the Origins of Mechanism 13

metrical imagery of the Greeks, formed the basis of all Hipparchan

and Ptolemaic astronomy and much of Greek mathematics. By the
time of Plato it seems likely there existed artifacts, perhaps even with
simple animation, simulating the geometrically-understood cosmos.
Brumbaugh 6 has pointed out that much of Plato's imagery seems to
derive from models that were more than mental figments. Certainly
by the time of Eudoxos (ca. 370 B. C.) we find a geometrical model
of planetary motion having every appearance of relation to an actual
mechanism of bronze rings.
Perhaps the most telling evidence is found in the writings of Ctesibius
(300-270 B. C.), who lived, as did Straton the physicist, in the period
between Aristotle and Archimedes. Thanks to the monumental labors
of A. G. Drachmann, 7 we now know that the basic mechanisms of
water-clocks and other devices familiar from the writings of Vitruvius
(ca. 25 B. C.) and Heron of Alexandria (ca. 62 A. D.) go back to this
time. Recent excavations at the Agora of Athens and at Oropos confirm
the existence there of monumental water-clock edifices from at least
the third century B. C. onward; that little gem of architecture, the
Tower of Winds in the Roman Agora of Athens, built by Andronicus
Cyrrhestes ca. 75 B. C., agrees so well with this theory and is so per-
fectly preserved (except for the centerpiece mechanism) that from
it one can essay a reconstruction of this entire class.
It would be a mistake to suppose that water-clocks, or the sundials
to which they are closely related, had the primary utilitarian purpose
of telling the time. Doubtless they were on occasion made to serve
this practical end, but on the whole their design and intention seems
to have been the aesthetic or religious satisfaction derived from making
a device to simulate the heavens. Greek and Roman sundials, for
example, seldom have their hour-lines numbered, but almost invariably
the equator and tropical lines are modelled on their surfaces and suit-
ably inscribed. The design is a mathematical tour-de-force in ele-
gantly mapping the heavenly vault on a sphere, a cone, a cylinder, or
on specially placed planes. The water-clocks, powered by the fall of a
float in a container filled or emptied by dripping water (as in the
Egyptian clepsydras), did not only indicate the time by means of
scale and pointer. At first they seem to have been used to turn a simple
model of the sun around with a celestial sphere; certainly this was
•Rohen S. Brumbaugh, "Plato and the History of Science," Studium Generale,
Vol. XIV (1961), pp. 520-27.
• A. G. Drachmann, "Ktesibios, Philon and Heron; A Study in Ancient
Pneumatics," Acta Historica Scientiarum Naturalium et Medicinalium, Vol. IV
(Copenhagen, 1948), and The Mechanical Technology of Greek and Roman
Antiquity (Copenhagen; Madison, Wisconsin; and London, 1963).

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
14 Derek ]. de Solla Price
the earliest type of model lmown in the analogous development in the
Chinese cultural area. 8 Later, presumably by the time of Hipparchus,
the principle of stereographic projection provided a fiat map of the
heavens bearing the Sun and Moon that could be turned to display most
impressively an artifically rotating sky; this device was later adapted
into the astrolabe, the most important of all medieval scientific instru-
ments of computation. 9
From the evidence of the Tower of Winds, these monumental public
structures contained much more than the astronomical model and its
powering clepsydra. Ingenious sundials were added all around the
octagonal tower, and on top was a bronze Triton weather-vane which
pointed to eight relief figures personifying the winds, mounted on a
frieze surrounding the top of the building. Within the structure,
around the walls, were probably mounted parapegma calendars on
which were tabulated daily astronomical and meteorological events,
events that could be confirmed visually from the central astronomical
showpiece and from the weathervane.
Judging from the texts of Heron, Philon, and Ctesibius, as collected
by Drachmann; from the tradition of automatic globes and planetaria
made by Archimedes; and from the few extant objects (on which I
have previously commented elsewhere); 10 we may say that the tech-
nology of astronomical automata underwent a period of intense de-
velopment. The first major advances seem to have been made by
Ctesibius and Archimedes, and the subsequent improvement must
have been prodigious indeed, seeing that it made possible, by the first
century B. C., the Antikythera mechanism with its extraordinarily
complex astronomical gearing.11 From this we must suppose that the
writings of Heron and Vitruvius preserve for us only a small and
incidental portion of the corpus of mechanical skill that existed in
Hellenistic and Roman times.
Even though we lmow so little about this sophisticated technology,
only indeed that preferred part of it that was committed to writing
and copied into preservation, its characteristics are obvious. So obvious,
•Joseph Needham, Wang Ling, and Derek J. de Solla Price, Heavenly Clock-
work (Cambridge, 1959).
•Derek J. de Solla Price," Precision Instruments to 1500," Ch. 22; "The Manu-
facture of Scientific Instruments from c 1500 to c 1700," Ch. 23, in Singer et al.,
A History of Technology, Vol. Ill (Oxford, 1954).
10 Derek J. de Solla Price, " On the Origin of Clockwork, Perpetual Motion

Devices and the Compass," United States National Museum Bulletin 218: Con-
tributions from the Museum of History and Technology, Paper 6 (Washington
D. C., 1959), pp. 81-112.
11 Derek J. de Solla Price," An Ancient Greek Computer," Scientific American

(June 1959), pp. 60-67.

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
 Automata and the Origins of Mechanism 15

that I am surprised previous scholars have not drawn the inevitable

conclusions. Amongst historians of technology there seems always
to have been private, somewhat peevish discontent because the most
ingenious mechanical devices of antiquity were not useful machines
but trivial toys. Only slowly do the machines of everyday life take
up the scientific advances and basic principles used long before in the
despicable playthings and overly-ingenious, impracticable scientific
models and instruments.
We now suggest that from Ctesibius and Archimedes onward we can
see the development of a fine mechanical technology, originating in the
improvement of astronomical simulacra from the simple spinning globe
to the geared planetarium and anaphoric clock. Partly associated with
and partly stemming from these advances we see the application of
similar mechanical principles to biological simulacra. We suggest that
these two great varieties of automata go hand-in-hand and are in-
dissolubly wedded in all their subsequent developments. In many
ways they appear mechanically and historically dependent one upon
the other; they represent complementary facets of man's urge to
exhibit the depth of his understanding and his sophisticated skills by
playing the role of a do-it-yourself creator of the universe, embodying
its two most noble aspects, the cosmic and the animate.
In support of the thesis that astronomical clockwork and biological
automata are complementary to each other, the following evidence is
submitted: (a) both types of simulacra see their first extensive develop-
ment at the same time; (b) the techniques used are found at first only
in them, seeping slowly, and much later on, into other instruments and
machines; and ( c) throughout the entire medieval, Renaissance, and
even modern evolution of fine mechanism a central role is played by
great astronomical clocks whose principal characteristic is the combina-
tion of astronomical showpiece with the automatic jackwork of imi-
tation animals and human beings.
In Graeco-Roman times the deepest complementarity exists between
the clepsydra principles used in astronomical models and clocks and
the almost identical inner workings of the Heronic singing-birds and
other parerga. Less closely related but still significant are the statuettes
holding indicating pointers on the scales of the waterclocks and the
Triton figurine with wind gods surmounting the tower of Cyrrhestes.
It may be significant that Rhodes, which was a center of astronomy in
the first century B. C., and Delos, which manufactured sundials like a
Greek Switzerland, were both famed for their automatic statues; even
the Colossus of Rhodes is said by Pindar to have been animated in
some way.
But since by now we strongly suspect that we know only a fragment

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
16 Derek ]. de Solla Price
of the original fine mechanical tradition of classical times, let us turn
next to the Middle Ages. One may reasonably suppose that later
examples often preserve, with little refinement, an ancient source. The
ample evidence of many well-edited texts and a couple of extant instru-
ments testifies to the existence of a more or less continuous, and re-
markably homogeneous, tradition of mechanical water-clocks, mainly
from Islam but extending without change to contemporary Byzan-
tium, 12 and with some modifications even as far as China and perhaps
India. This tradition seems to have been transmitted to Europe without
much change or dilution during the medieval renaissance of the twelfth
and thirteenth centuries; during the next century it became conflated
with other lines of development and was thus transformed into the
essentially modem principle of the mechanical clock, which preserves
so much of the feeling and motivation of the old ideal. In particular
there was preserved the special complementary relation between the
clockwork and jackwork.
In the typical Islamic clock, which was in its heydey from about
800 A. D. to 1350 A. D. and which may be very close to the lost
Hellenistic originals, power is provided by a float in a vessel filled or
emptied by dripping water. This power is harnessed, either directly
by having a chain or string pull a block along a straight channel, or
rotationally by having the string wind around a pulley, or by using
a geared pinion and rack. The straight motion may trip a series of
levers one by one, opening a set of doors, moving a set of figurines,
or letting a series of balls fall into gongs and sound at set intervals.
The circular motion may be used to animate automata, moving their
heads or bodies or rotating their eyeballs, or to turn a globe or stereo-
graphic map of the heavens and perhaps also, by appropriate gearing,
models of the sun and moon placed upon the heavenly representation.
In a refinement, the dripping water may be caught in another vessel
which is suddenly and periodically emptied by an automatic syphon
or a balancing-jar; the apparatus then works rather like a faulty modern
lavatory cistern that flushes itself as soon as the tank is full. The
ensuing rush of water may then spin a water wheel to move other
automata or it may enter a vessel, displacing air so as to blow the
whistle or sound the organ pipes that provide the singing of the
mechanical birds or other manikins.
These mechanisms, though undoubtedly impressive, are mechanically
simple and Heronic. They are described in detail by Ridwan and
10 Note also the traditional Heronic jackwork described by Gerard Brett, " The

Automata in the Byzantine ' Throne of Solomon,' " Speculum, Vol. 29 (July
1954), pp. 477-87.

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
 Automata and the Origins of Mechanism 17

al-Jazari 13 (both early thirteenth century), and there are texts de-
scribing their appearance in Damascus and Gaza.14 There is also evi-
dence of two fairly simple clocks (fourteenth century) of this type
extant in Fez, Morocco,1 5 and of a quite complex geared astrolabe
designed by al-Biruni ca. 1000 A. D. 16 and attested by an example made
in Isfahan in 1221 A. D. 17 We know also that such devices were re-
putedly owned by Harun al-Rashid and Charlemagne 18 in the ninth
century A. D. and by Saladin in the twelfth century.
Just before the transmission to Europe in the thirteenth century of
the corpus of knowledge about clockwork and automata, that learning
somehow became intertwined with concepts of perpetual motion (an
idea apparently unknown in Classical antiquity) and of magnetism and
the mystery of magnetic force. This intertwining may have originated
with the accounts of travellers returned from China telling about the
clocks of Su Sung and the related work on the magnet being done
there. 19 Also toward the end of the thirteenth century came the
purely astronomical elaboration of complicated equatoria; these were de-
signed to compute the positions of planets and afforded a more complete
geometrical simulation of Ptolemaic theory than the older, somewhat
Aristotelian models embodying simple uniform rotation. Many such
devices are seen in the Alfonsine corpus, which contains also designs
for a rotating drum with leaky compartments filled with mercury
which acts as the regulatory agency of an astrolabe clock. Elsewhere
in the Islamic sources are the elements of the weight drive, used not
for the mechanical clocks to which it was later adapted but for pumping
water. 20
With the transmission to medieval Europe of all these ideas and
techniques seems to have come a burst of exuberant interest which was
further stimulated by the flowering of the craft guilds. The drawings
1 • Eilhard Wiedemann and Fritz Hauser, " Ober die Uhren im Bereich der

islamischen Kultur," Nova Acta Abh. der Kaiser!. Leop.-Carol. Deutschen

Akademie der Naturforscher, Vol. 100, No. 5 (Halle, 1915).
"H. Diels, "Ober die von Prokop beschriebene Kunstuhr von Gaza," Abh.
der Koniglich Preussischen Akademie der Wissenschaften, Philos-Hist. Klasse,
No. 7 (Berlin, 1917).
1 • Derek J. de Solla Price, "Mechanical Water Clocks of the 14th Century in

Fez, Morocco," to be published in the Proceedings of the Xth International

Congress of the History of Science, Ithaca N. Y. and Philadelphia, 1962.
1 • E. Wiedemann, "Ein Instrument <las die Bewegung von Sonne und Mond

darstellt, nach al Biruni,'' Der Islam, Vol. 4 (1913), pp. 5-13.

17 Price, "On the Origin of Clockwork .... ,'' loc. cit., pp. 98-100.

1 • Note also the interesting astronomical model described in F. N. Estey,

"Charlemagne's Silver Celestial Table," Speculum, Vol. 18 (1943), pp. 112-17.
1 • Price, " On the Origin of Clockwork .... ,'' loc. cit., pp. 108-10.

••Hans Schmeller, "Beitrage zur Geschichte der Technik in der Antike und

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
18 Derek ]. de Solla Price
in the famous notebook of Villard de Honnecourt (ca. 1254)-more
likely the album of a guild rather than the work of an individual-
show a clocktower for a mechanism that was probably a clepsydra-
driven bell chime. Other drawings show a simple rope-and-pulley
apparatus for turning an automaton angel (which is interpreted quite
unauthoritatively as an escapement by Fremont 21 ) and another rope-
driven automaton bird. Also from the thirteenth century are ample
records and even illuminations showing church water-clocks; there is
the preoccupation with perpetual motion of Robertos Anglicus in his
search for an astronomical simulator, and a similar preoccupation
"solved" through magnetic power by Peter Peregrinus. 22
By 1320 the clock, presumably a water-clock, had been adapted by
Richard of Wallingford to the working of complicated automata based
on the principle of the equatorium and demonstrating with great
ingenuity the exact motions of Ptolemaic astronomy. Not long after-
ward, in 1364, Giovanni de Dondi had built his great clock in Padua;
we know from a full manuscript description that this was a true
mechanical clock with weight-drive, verge-and-foliot escapement,
seven magnificent dials with a panoply of elliptical and normal gear-
wheels and linkwork to show all the astronomical motions, a fully
automated calendar showing Easter and other holydays, and-a little
dial for telling the time. The clock of de Dondi, though matching in
complexity and ingenuity any seventeenth century product of the
clockmaker's art, is somewhat anomalous in our history, for it has no
biological jackwork. However, we know that this was a firm tradition
by then, for it appears in the first monumental astronomical clock
of the cathedral of Strassbourg. From this most famous and influ-
ential series of three successive clocks (1354, 1574, 1842) has been
most fortunately preserved (in the local museum) the large bronze
automated cock which surmounted the structure. Crowing and mov-
ing most naturalistically on the hours, the cock accompanied with its
actions the carillon, the other manikins, and the astrolabe dial and
calendar work. By this time mechanical ingenuity was able to produce
automation of the bird figure; the complicated arrangement of strings
and levers became a reasonable simulacrum for the musculature and
skeleton of a real bird. 28
bei den Arabern," Abh. zur Geschichte der Naturwissenschaften und der Medizin,
No. 6 (Erlangen, 1922).
21 C. Fremont, Origine de l'horloge a poids (Paris, 1915).

••See Lvnn White, Medieval Technology and Social Change (Oxford 1962),
pp. 120-29, 173.
••Alfred Ungerer, Les Horloges Astronomiques et Monumentales Jes plus
remarquables de l'antiquite jusqu'a nos jours (Strassbourg, 1931), pp. 163-65.

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
 Automata and the Origi.ns of Mechanism 19

From this time forward, the great astronomical cathedral clocks,

complete with jackwork, sweep Europe, growing in number but per-
haps lessening in mechanical complexity during the sixteenth, seven-
teenth, and eighteenth centuries. The only interruption occurs during
that remarkably dead period of intellectual and economic depression
in the second quarter of the fifteenth century. Apart from this, one
can trace the steady evolvement of the clockmakers' fine metal-working
craft to its finest manifestation in the craft of the instrument maker
which was to dominate the development of learning during the Scien-
tific Revolution.
Accompanying the European popularization of water clocks and
mechanical clocks during the Middle Ages came a flood of literary
allusion based partly upon the clocks, partly upon travellers' tales of
parallel traditions of technology in Constantinople and the Orient, and
partly upon a revival of the classical mystique of magically animated
figurines. 24 The clock itself, in its debasement from astronomical mas-
terpiece to mere time-teller, becomes so familiar that it assumes allegori-
cal significance in such disquisitions as Froissart's L'Horloge Amoureuse
and in the tract L'Horloge de Sapience, whose manuscript illuminations
have offered recent scholarship so much detailed insight into early
mechanics and instrument-making. From Heronic sources, perhaps
Byzantine, perhaps transmitted through Arabic to medieval Europe,
come many allusions to brass trees full of singing birds, set in motion
by water power, by the wind, or by bellows.
More magically still, Albertus Magnus (as many other philosophers)
is said to have made a brazen head, and he especially is credited with
the feat of havin$ constructed a mechanical man-a robot, to use the
term coined by Capek-from metal, wax, glass, and leather. We know
no specific details of any such automaton made by Albertus, but we
may suppose that at about this period the art of automaton-making in
Europe had recovered a level of sophistication and verisimilitude prob-
ably not much inferior to that demonstrated in the Strassbourg clock.
Albertus's most famous pupil, St. Thomas Aquinas, stated emphati-
cally in his Summa Theologica (Qu. 13, Art. 2, Reply obj. 3, Pt. II)
that animals show regular and orderly behavior and must therefore
be regarded as machines, distinct from man who has been endowed
with a rational soul and therefore acts by reason. Surely, such a near-
Cartesian concept could only become possible and convincing when
the art of automaton-making had reached the point where it was felt
that all orderly movement could be reproduced, in principle at least, by
" Merriam Sherwood, " Magic and Mechanics in Medieval Fiction," Studies
in Philology, Vol. 44 (October 1947), pp. 567-92.

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
20 Derek ]. de Solla Price
a sufficiently complex machine. It is remarkable that at this very time
figures of apes become popular as automata-they had been used inter
alia by the Islamic clockmakers-being endowed with an appearance
similar to that of man but behaving as a "beast-machine." This is
probably the line that led to such literary and philosophical devices
as the Yahoos of Jonathan Swift, beasts shaped like men but without
rationality; it is also the line that made philosophically important the
emergent possibility of exhibiting mechanically many manifestations
of apparent rationality.
Of such kind were the mathematical calculating machines that began
with all the early astronomical automata, proliferated during the six-
teenth and seventeenth centuries, and culminated in the first true digital
computer of Pascal, the Pascaline of 1645. Of such kind were the
remarkably constructed musical automata during the same period,
particularly such impressive devices as that built by Achilles Lagen-
bucher of Augsburg in 1610; this seems to have had a large array of
instruments that were programmed by a sort of barrel-organ device
and is said to have performed with taste and to good effect. In these
mathematical and musical automata we see the first insidious intrusion
of mechanicism into areas that formerly had seemed typical of the
rationality distinguishing man from the beast-machine. Consequently,
at this moment in time, just before Descartes, began the reaction against
automata and the turning back to that mechanistic philosophy which
had been their original inspiration.
Related to water-clocks, and producing an almost independent line
of evolution for automata in the Renaissance, was the art of water-
works, a technique in which there was almost legendary proficiency
in Roman times. From a pair of beautiful Norman drawings of the
waterworks of Canterbury Cathedral 25 and its vicinity, we can surmise
that the ancient skill was in the hands of able craftsmen by about
1165 A. D., and thenceforth are found the clepsydras of churches
and monasteries, depending on an adequate supply of dripping water.
During the Middle Ages there seems to have been some production
of hydraulically operated automata; authority is lacking, but it is
probably safe to assume that they were close to the Heronic tradition
in their basic design.
At the close of the thirteenth century a particularly famous set of
such water-toys was built for Due Philippe, Count of Artois, at his
castle of Hesdin.26 It is described in detail by the Duke of Burgundy
in 1432, and one gathers that along with the spouts for wetting fine
••Robert Willis, The Architectural History of the Conventual Buildings of
the Monastery of Christ Church in Canterbury (London, 1869), pp. 174-81.
•• Sherwood, loc. cit.

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
 Automata and the Origins of Mechanism 21

ladies from below and covering the company with soot and flour were
quite a large number of animated apes covered with real hair and
sufficiently complicated to need frequent repair. This "pleasure
garden," in all its extravagant bad taste, became the talk of the civilized
world and was probably the ancestor of those famous and somewhat
more decorous French and English fountains and waterworks of the
late sixteenth and seventeenth century whose elegant automata im-
pressed the public and revived in sensitive philosophers the old urge
of mechanicism.
Yet another line of development deserves consideration, though it
does not directly relate to automata; that is the use of optical tricks
to produce apparently magical effects. There is some inkling of this
in the writings on optics of Classical antiquity, but plainer mention
is made by the medieval Polish scholar Witello. During the Renaissance
these optical illusions became quite a popular hobby among the ex-
ponents of natural magic and the perpetrators of mechanical trickery.
As a link between the Middle Ages and the Renaissance there is, as
in every aspect of the history of technology, the figure of Leonardo
da Vinci. His mechanical prowess in automata extended to the illus-
tration of Heronic hodometers and a planetary clock mechanism very
like that of de Dondi, both making use of gears. 27 His work on flying
machines is well known, but in the present context it may be refreshing
to regard it, not as a means for man to fly, but as the perfection of a
simulacrum for the mechanism of a bird. He is also reputed to have
made at least one conventional automaton, a mechanical lion which
paid homage to Louis XII on his entry into Milan by baring its brazen
chest to reveal a painted armorial shield of the sovereign.
In the Scientific Revolution of the sixteenth and seventeenth cen-
turies the dominant influences were the craft tradition and the printed
book. Both played crucial roles in raising automata, astronomical and
biological, to a new height of excellence. In the craft centers, par-
ticularly those of the central city-states of Nuremberg and Augsburg,
there grew up the first fine workshops of skilled clock and instrument
makers. According to our interpretation of the history of automata,
it is no accident that these cities and the whole Black Forest area have
been regarded until today as the chief centers for the manufacture of
both clocks and dolls. It is equally telling that the product particular
to them is the cuckoo-dock, a debased descendent in the great tradi-
tion of the Tow er of Winds and the Clock of Strassbourg, but one in
27 Derek J. de Solla Price, " Leonardo da Vinci and the Clock of Giovanni

de Dondi," Antiquarian Horology, Vol. 2, No. 7 (June 1958), p. 127. See also
letter from H. Alan Lloyd following, Antiquarian Horology, Vol. 2, No. 10
(March 1959), p. 199.

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
22 Derek ]. de Solla Price

which is still seen that highly significant liaison between the cosmic
clock and the biological artifact.
At Augsburg and Nuremberg during the sixteenth century the
masterpieces of the clockmakers were usually extremely elaborate
automaton clocks, in which tradition are the brothers Habrecht, makers
of the second Strassbourg clock in 1540-74. From about 1550 there
are preserved the first of the new series of automated manikins in
which the mechanism is considerably advanced beyond the old Heronic
devices. 28 For the first time, wheelwork is used instead of levers, gears
instead of strings, organ-barrel programming instead of sequential
delay devised hydraulically. The skill was so well known that Melanc-
thon wrote to Schaner in 1551, on the publication of the Tabulae
Resolutae, " Let others admire the wooden doves and other automata,
these tables are much more worthy of (true scientific) admiration."
At this stage, half a century before the birth of Descartes, other
technologies begin to influence the automaton-maker, and his reaction
to these in turn affects strongly quite different branches of the sciences
as well as technology and philosophy. One felicitous example is the
use of the armorer's craft by Ambroise Pare (ca. 15 60) for his design
of artificial limbs-partial automata to complete a man who had become
deficient. Then again, the draining of the Low Countries and English
Fens aroused new interest in the hydraulics of pumping engines, and
out of urban development came new ideas in massive waterworks and
portable engines for fire pumps. All these increased the technical
skill of those who would devise the fountains and automata that were
to be the wonder of St. Germaine-en-Laye, Versailles, and other places.
Of the influence of the printed book we may note that although
Vitruvius's De Architectura appeared in an incunable edition (Rome,
1486), the works of Heron had to wait until 1573 (Latin) and 1589
(Italian). Thus, although the simple waterclocks and sundials described
by Vitruvius were available throughout the Scientific Revolution, the
Heronic corpus did not begin to exercise its greater influence until
the last two decades of the sixteenth century. By that time the craft
tradition was already in full swing and the Habrecht Clock at Strass-
burg had been completed. So, by the time of Shakespeare, man's
ancient dream of simulating the cosmos, celestial and mundane, had
been vividly recaptured and realized through the fruition of many
technological crafts, including that of the clockmaker, called into being
in the first place by this lust for automata. The new automata were to
capture the imagination of the next generation, including Boyle and
Digby 29 and Descartes himself. Their very perfection would lead to
••Ernst von Bassermann-Jordan, Alte Uhren und Ihre Meister (Leipzig, 1926).
•• Digby's work is specially interesting as the first complex mechanization of

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms
 Automata and the Origins of Mechanism 23

the next phases: automation of rational thought-a stream that leads

from Pascal and Leibnitz 30 through Babbage to the electronic com-
puter, of memory by means of the punched tape first used in sixteenth
century Augsburg hodometers, and of the cybernetic stuff of respon-
sive action perceived dimly in the Chinese south-pointing chariot,
decisively in the thermostatic furnace of Cornelius Drebbel, and more
usefully than either in the steam-engine governor of James Watt.
Descartes, at the time when the crucial change of direction was
about to be made, was probably one of the first philosophers to sense
what its characteristics would be. Long before he published his
Discourse, and perhaps before he had become interested in theology,
he toyed with the notion of constructing a human automaton activated
by magnets. One of his correspondents, Poisson, says that in 1619 he
planned to build a dancing man, a flying pigeon, and a spaniel that
chased a pleasant. Legend has it that he did build a beautiful blonde
automaton name Francine, but she was discovered in her packing case
on board ship and dumped over the side by the captain in his horror
of apparent witchcraft. There is probably no more truth in these
rumors than in similar stories about Albertus Magnus and many others,
but it does at least suggest an early fascination with automata. And
the mention of magnets further suggests the desire to enlarge their
potentialities by the use of forces more potent than the mechanical
means of the time, an ambition surely presaging the idea that mechan-
ism, now richer in technique than ever before, could simulate the uni-
verse to that deeper level of understanding which was indeed soon to
be attained.
Descartes's place in all this, then, is that of one who stands on a
height scaled and begins the ascent to the next plateau, which is
suddenly revealed with greater clarity, though distant still. In many
ways it is like the balance between materialism and vitalism that
would come with Wohler's synthesis of urea; and, just as in that case,
there is deceptively slow adjustment to it among philosophers and a
feeling that no ground has been gained or given by either side. From
the Lascaux Caves to the Strassbourg Clock, to electronic and cyber-
netic brains, the road of evolution has run straight and steady, oddly
bordered by the twin causes and effects of mechanistic philosophy
and of high technology.
plant physiology and as a clear and stated example of influence by the machines.
Sir Kenelme Digby, Two Treatises: ... The Nature of Bodies; ... The Nature
of Man's Soul; ... In Way of Discovery of the Immortality of Reasonable Souls
(London, 1658), pp. 255-59.
••Note that this line of argument makes it sigcificant that both men were
philosophers, mathematicians, and pioneers of calculating machines.

This content downloaded from 201.174.75.10 on Thu, 17 Jan 2019 20:54:51 UTC
All use subject to https://about.jstor.org/terms