PRINT May 1983


FOR SOME TIME NOW, in order to write this wretched article, I have been thinking about automatons. I read about large robots that weld sheet metal, install and mount car cylinders, put together eight thousand car doors in a day, and by their very presence make workers obsolete—reminding me of medieval images of the “triumph of death”; I console myself thinking of the nearly perfect artificial limbs for the mutilated, crammed with electronic sensors, that can be made today, be it in Italy or Japan. My house, too, is full of machines, which I sometimes turn on all at once, making a diabolical noise. Before long I may have to get myself an artificial heart (perhaps after my next, doubtless irremediable failure in love). But as for true automatons—I no longer see them around.

Granted, one still finds morbid fantasies in science fiction films. And one can imagine the Japanese inventing a new type of doll, inflatable, expensive, with real hair and silk skin, eyes that shine with a dull light, and nipples that project (discreetly) as far as a preset program allows. The personality is well defined, passive but capable of being excited by an intense glance or a rapid caress; even the doll’s loyalty can vary according to Its desires. Its mouth responds in aleatory fashion to such stimuli as sounds and the gestures of its partner, with seductive words, monosyllables, and sweet sighs put together from the soundtracks of the latest popular pornographic movies. I can imagine such a doll wearing out with use, becoming dried up, overweight, and covered with tiny wrinkles, and growing dependent on the affection of whoever happens along. Like a film exposed to too much light during development, this automaton fades and becomes useless.

But I have heard that in some laboratory they have isolated a substance that produces love. The traveling executive may soon carry with him not a doll, but an anti-emotion pill, and he will sleep absolutely tranquilly, pure as a monk. I do not look forward to the contrary solution (already possible today)—that as soon as a couple meet, at the first smile, they may each take an antitaboo pill and rush headlong to the natural conclusion for which Don Juan was so famous.

In other words, there is no longer a place for automatons in the world. With the exception of those at Disneyland, they have all been removed to what are rather like enormous death-filled battlefields—air-conditioned museums—where they wait, piled up in confusion, like the suits of armor built in the great Milanese factories of the 16th century. As for minor automatons, I can imagine myself on the Pacific coast, below William Randolph Hearst’s castle, amid huge piles of seaweed and jellyfish; in all this rottenness, dissolved in it and corroded, are the wind-up toys of our childhood—cars without wheels, broken-down trains, mice without tails, mutilated crabs, one-armed harlequins, and Mickey Mouses with broken noses.

Because of the great love I had for these things I’m afraid to visit the vast, morguelike museums that are the resting places of the most celebrated androids, such as Pierre Jaquet Droz’s boy who writes and draws, in Neuchâtel, or D. Roentgen’s and P. Kintzing’s dulcimer player in the Conservatoire des Arts et Métiers in Paris. Perhaps they are worn out from their long stints in the capitals of Europe, where they were discussed, admired, touched, and improved upon; in any case, there they are now, greasy and terribly sad. Who among the youth of today will feel the emotion I felt on seeing these for the first time? I remember the buzzing of the mechanism that set off the dulcimer player—with the slightest movement of metal the chest widened in a sigh, the eyes opened, the head tossed, then the hands and the elbow moved, emitting music. And the boy who draws, his pen first poised, then dipped in the inkwell, the excess ink shaken off, then the pen poised again on the paper and drawn across it, slowly producing figures and words, under one’s very eyes.

Could one possibly repeat the experience of the little monsters grouped together as though in a circus (almost as in Tod Browning’s 1932 horror film, Freaks), in an exhibition in the ’50s in Geneva, then the world center of watchmaking? Even if that show displayed monsters it was also filled with the songs and sounds of miniature birds, mere centimeters long, who batted their wings, opened their mouths, and hopped from one perch to another just like real creatures—but with voices like Luciano Pavarotti’s, or, better (given the time), like Lucia Del Monte’s. Would a contemporary gallery resist an art movement of artists turned mechanics, or would it succumb, and exhibit, for example, serpents, mice, toads, little dinosaurs with moving tails, and imaginary insects, such as those shown by a certain “Wolf, born in Carrara in 1934, lives in Rome,” at the Obelisco Gallery in Rome in 1965, in celebration of the idea of perpetual motion?

Probably not. With the coming of electronics and computers, automatons have been stripped of their hearts, their capacity for judgment, their impulse to be moved—to imitate and fulfill human emotions and desires. Once the woodman dreamed of a sort of giant who might with one hand grasp a tree and with the other cleave it in two with a magical axe. In such dreams automatons awoke. Something like the woodman’s dream actually came to pass in the 13th century, with the introduction of chutes down which tree trunks slid to converge on a kind of moving platform. A millwheel, in addition to pushing the platform, activated a vertical saw and a lathe, each of which in turn did its work on the wood. And if one wanted everything, truly everything, in the way of automation, it would have been sufficient to visit a theater stage from the 17th or 18th century. Here systems of winches, operated by scores of cables of varying lengths, could position hundreds of painted silhouettes in a matter of seconds, all with the tug of a rope. These sets––gardens, mountains, palaces, and cities, or fire and water in feigned catastrophes—moved up and down, alternating scenes to musical accompaniment. Angelic choruses rose in the air; Icarus fell; gods solemnly descended to princely courts.

By definition, then, the theater became the machine par excellence, the microcosm and symbolic center of the city, perhaps of civilization itself. And while the public observed exaggerated situations of joy or pain on stage, machines worked backstage to produce and reconcile these emotions in this purely speculative ambience. Some passions, of course, could only seem to find their resolutions on the battlefield; but here too, for millennia, engineers had been busy, devising and trying out mechanical means of war such as towers that moved on their own for short distances, trolleys that took soldiers up to the protective cover of a wall, and gigantic excavating machines that could divert rivers, parching enemies or sweeping them away by floodwaters, as was done at Mantua and at Lucca in the 14th and 15th centuries respectively.

All this is over now; at least it has ceased to be a major part, and is certainly not the apex, of our civilization. Except in a few science fiction films, the automaton today is no more than a symbol of passivity. Since what matters now is servility, the modern ideal would be to have living bodies rendered intellectually sterile, in a mild, less bloody version of the myth of Frankenstein. Like it or not, our current concept of automatons limits them to extensions of our physical possibilities, while intelligence, removed from their bodies, has been delegated to the microcircuit. The large, useless machines of Jean Tinguely are parables of the exhaustion of the myth, and his Homage to New York, which self-destructed at New York’s Museum of Modern Art in 1960, was an early celebration of the death of the machine in a sort of apotheosis and ruin. There is no longer a Pygmalion to believe in the possibility of giving life to his creation, the beautiful Galatea. We no longer talk of cybernetics; the terrors of the unconscious are placated by playing videogames rather than with dolls or toy trains. Best not to imagine what might happen if humanity’s latest servants were to make a mistake; better to spend the day pushing buttons, playing records or tapes, ever reducing our real participation in life to such forms of physical activity as automatons were originally and above all created for.

Throughout their history automatons have retained the capacity to analyze movement, transcribing it in linear fashion. At first there was an interesting relationship between machinery and presumed cosmological systems; later, with the progress of anatomical and neurological knowledge, machinery came to reflect research done into muscles and muscular functions. A second constant is the process of miniaturization, shrinking the universe to the dimension of a room. The building of automatons on a very small scale has been going on for quite some time, as is borne out by the story of mechanical fly built by Bishop Virgilio of Naples around the year 1000 to rid the marketplace of noxious insects. A third feature is simultaneity of movement—a spinning wheel’s single wheel may activate hundreds of spools, and so on. But what interests us here is really the mental concept behind automation.

Let’s take as an example the extremely sophisticated and much discussed astronomical clock built by Giovanni de’Dondi over the 16 years between 1348 and 1364. It became the property of the Duke of Milan and then of the King of Spain, was destroyed in a fire, and was recently reconstructed in two versions—one in the Smithsonian Institution in Washington, the other in the Museo Nazionale della Scienza e della Tecnica in Milan. The device comprised a sort of heptagonal turret with eight clock faces, one marking the time and seven with hands indicating the paths and situations of the planets. The clock provided an immediate horoscope without need of calculations or tables. It also functioned as a perpetual calendar. The motor consisted of the usual hanging weight, but since the speeds of the hands obviously had to vary according to the orbits and velocities of the planets, a highly complex system of gears was necessary. The elementary theory of gears had been articulated by Erone Alessandrino:

Wheels on the same axis will always move in the same direction, the direction of the axis. But wheels on two different axes, in contact with each other by means of gears, will move in opposite directions; one will turn toward the right and the other toward the left. If both wheels are equal in size, a single turn of one toward the right will correspond to a single turn of the other toward the left. But if they are different, one larger than the other, the smaller of the two will turn several times before the larger wheel will have completed a single rotation. The number of turns depends on the relationship between their dimensions.

Our knowledge of de’ Dondi’s invention is quite precise, since he left an illustrated manuscript, Tractatus astrarii, which is now codex D. 39 in the Biblioteca Capitolare in Padua (where he had been educated and was a professor). Before his achievement there had been tracts on astronomy that had included quadrants with card indicators that could be moved by hand, thereby attempting a view of the movements of the planets in relation to each other. The complexity of their courses, then, as far as it was known at the time (this was before Copernicus) had been acknowledged and depicted in graphic representations. But de’ Dondi’s clock was innovative.

It is worth noting how this device relates to the automaton’s involvement in motion. With the first efforts to measure time, there was an attempt to render its passage in linear or graphic terms. This is what happens with the gnomon or sundial. A subsequent system probably stemmed originally from biology, i.e., from the steady beat of the pulse, which suggests a means of dividing time into units ultimately quantified as seconds (a division which contains within itself the concept of minutes and hours). The hanging weight of the medieval clock translated these pulsing divisions of time into the motion of a pointer along the circumference of a circle—the motion of a hand around a clock face. The circular path of the hand was completed in a span of 24 hours (12, in modern clocks). But this was terrestrial time. To obtain astral times, based on extremely complex theories, it was necessary to translate the normal speed of the clock hand around the face into the speeds of the hands marking the orbits of the seven planets shown on de’ Dondi’s clock. This was done through precise calculation of the diameters of a series of 107 gears (not very many, really—little more than 15 per planet). Apart from the technical difficulties, the complexity of the mental process involved is most impressive.

Another method for transmitting movement is the use of systems of levers. Applicable here are not only Archimedes’ law—“Equal weights at equal distances [from the fulcrum] are in equilibrium, and equal weights at unequal distances are not in equilibrium but incline toward the weight which is at the greater distance”—but also the law by which the arc described by the end of the long arm of a scale is proportionally greater than that described by the end of the short arm. Following these very simple rules in the use of weights of varying sizes, and above all in pivoting levers at unequal distances from their ends (almost never at their centers), it was possible as early as the 14th century to construct a mechanical rooster at the top of a tower, which could flutter its wings. More levers were introduced and more movements became possible: from birds to horses to roaring lions to androids who play the dulcimer. The belly of an automaton, instead of a heart and intestines, contains levers and gears.

Series of spools or pulleys work like gears in that different-sized ones will have different effects. Variations in the sizes of pulleys, for example, alter the length of a cord or rope wound up per revolution, and thus the speed of movement of the object being hauled. Another factor is the distance between a pulley and the object to be hauled. In other words, paradoxically, simultaneous changes of different items of scenery in a theater are possible only if each object is moved (hoisted, pulled sideways) by capstans of different sizes.

With the coming of electricity, commands could be given via electromagnets and electric motors. But there remained the problem of how to subdivide the movement (now capable of more speed than before) and how to distribute it. Levers continued to be the only means for imitating the motion of human or animal—the body itself, in fact, can be seen as a system of levers activated by the muscles. This is a return to the mechanistic vision of the 18th century, whereby even the most complex organism, like man, could be reduced to certain elementary machinery. At least in the case of man, however, an analogous theory was advanced by a follower and friend of Michelangelo, the theoretician Vincenzo Danti, in Trattato delle Perfette Proporzioni (Treatise on perfect proportions, 1567): “All the limbs of the human body are made to serve the sense. . . every time the limbs perform their respective proportionate functions and perfectly, they will be optimally duty and service that they must perform. So it follows that proportion is none other than the perfection of bodily structure to the extent necessary to achieve the desired end.” In this theory of anatomical functionalism there seems to be a glimpse of more recent ideas about movement, such as those of the Futurists.

An important further area of motion is the control of it. Automatons generally employ intermittent rather than continuous movement—they carry out commands given sequentially. If protrusions or recesses are positioned around the surface of a wheel or cylinder that turns at a constant speed, these irregularities can issue directions to a system of levers or gears whose position is affected by direct contact with that surface. A music box, for example, may contain a revolving cylinder studded with carefully positioned pegs which in turning strike metal keys, causing them to ring in sequence. A needle on a record, a laser beam on a videodisc, react analogously to carefully contrived situations and can do so indefinitely.

Some of the applications of this technology have been extremely ingenious. The achievements of Islamic culture represent both a great resource and a connecting link between the Western world and the world of classical antiquity, especially in the realms of science and technology. An Arab treatise has given us a description of a miniature theater of flute-playing automatons. The movements of their tiny hands were achieved with levers activated by arrangements of small watermills; air for the flutes was compressed in cylinders, again by the action of water. But how to produce the rhythm and melodic line integral to music? A construction similar to a primitive phonograph was devised. In performing a piece, actual musicians recorded the movements of their hands by means of cords attached to their fingers and to needles which incised a rotating wax cylinder. The marked cylinder was then reproduced in terra-cotta, connected to levers, and set in motion, and the automatons, activated by the incisions in the terra-cotta, actually played. (A record player, by comparison, reproduces only sound, not movement.) Certainly this must have been an extraordinary invention; perhaps excavations will someday bring to light one of these cylinder recordings.

In the previously mentioned music box system the music is transcribed not by manual performance, but rather through mathematical measure. The duration of the music depends on the rotation of the cylinder; variations in the size of the studded pegs can affect how violently the metal key vibrates, and thus the volume of the notes. There is no limit to the quantity of sounds that can be produced simultaneously, and obviously, since the protrusions on the cylinder can activate levers, innumerable movements can be developed through appropriate mechanisms. At this point the limits of the results depend on the limits of the imagination.

Alchemy, too, an essential precursor of modern chemistry (even if its aims were misdirected), sometimes produced extremely refined toys. as the results of virtuoso experiments—marvels without practical value. Yet many sovereigns, especially at the end of the 16th century, grasped the economic potential of some of these inventions. In Pratolino, for example, the Grand Duke of Tuscany, aided by Bernardo Buontalenti, built grottos of automatons. Most were toys, but there was also a completely automated mill, a version of which was in fact sold in Yugoslavia under the equivalent of a modern patent. The interest of the Holy Roman Emperor Rudolph II in technology was evident from his science museum (also a tool of political propaganda), which has been described by such men of his court as Johannes Kepler and Tycho Brahe. The extraordinary illustrated manual Machinae Novae (New machines), edited in Venice in 1595, includes the early farm machinery of Faustus Verantius. Recognition of the practical application of some of the devices used in automatons of the past grows day by day through the discoveries of industrial archaeology.

The relationship between automatons and production is not, then, something new. (A refinement in computer technology originated with the automaton—the perforated computer card, a sort of negative of the studded cylinder we have discussed.) An exemplar, of course, is the weaving industry at the outset of the Industrial Revolution, in which ultimately the threads were so positioned in the mechanized looms that manual labor was reduced to sliding the bobbin back and forth. (In addition, of course, setting up the apparatus was extremely laborious and took not hours, but days.) This 18th-century invention did not constitute a breakthrough on its own; it had to be accompanied by other such contrivances as Richard Arkwright’s water frame (1769), which made possible the production of a stronger cotton thread, and James Hargreaves’ spinning jenny (1764), which gave a single worker control over 16 spindles. This statistic can be compared to the 14 machines per worker at today’s Fiat plant in Mirafiori, or the 10 machines per worker in the larger Japanese factories. It is a striking example of how automation has nurtured modern capitalism.

Automatons are above all an exercise in ingenuity on a highly specialized level, dealing with the use, the quantitative modification, and the qualitative differentiation of movement. They are a metaphor for, perhaps a proof of, the basic underlying law of the multiplicity of appearances. The builder of automatons does not create movement; he can imitate God only in his varied characterization and utilization of movement. An android, of course, requires more movable parts and more verisimilitude than a little bird covered in feathers, and it is in this field that the most refined research has evolved. The results are at times miraculous. In the Kunsthistorisches Museum in Vienna there is a small doll representing an extremely elegantly dressed woman playing the lute. Better than paintings and sculptures of the period (16th century), this figure analyzes contemporary feminine behavior:

A combination of stateliness and grace characterizes its movement. The feet move slowly, the figure seems to float. The skirt disguises the shape of the body completely. In contrast, the hands and head are in free and elegant motion. The head describes exquisite curves to the music produced by the sensitive delicate hands. The main features of the typical society woman are incorporated in a clockwork mechanism, the limitations of which naturally exaggerate the characteristics of the figures as a caricature does.

The writer is none less than Fritz Saxl, on the occasion of the Italian meeting of the British Academy in 1936. Analogous examples are rare but not unknown; I would mention the Munich praying figurine from around 1560 (now in the National Museum of American History in Washington, D.C.). Obviously these dolls who walk and gesticulate require compact mechanisms, the products of the miniaturization that reaches its apex in tiny singing birds, or even in wristwatches, which were and are authentic marvels.

The subject of automatons is more complex, of course, than the basic examples described here indicate, but I limit my discussion to those aspects that connect most directly to the problems of today. In summary, the study of automatons can also be seen as a vast field of exploration of movement, studied as much in its laws as in its applications. Movement is not a simple subject; its ramifications spread into many unexpected areas, and its study requires special aptitude. Etienne Jules Marey, for example, known as a photographer, was a skilled investigator of movement. He not only invented the chronophotograph, by which various positions of the human body in motion are superimposed in one image (as later occurs in Giacomo Balla’s painting and had earlier been explored by Leonardo), but also created scores of instruments for physiological examination. (Many are still in use today—the sphygmograph, the pneumograph, the cardiograph.) Among other things, Marey studied graphic methods for representing the reciprocal positions of trains in the French railway system.

A passion for movement is a way of analyzing and constructing the world. Perhaps I too am an investigator of movement. I dream, in fact, of someday holding an elaborate seminar at some university, rather like Buckminster Fuller. I would call forth various cultural situations of the day on a large computerized screen: the growth and diminution of the cities, the intensity of telephonic and communications, the rhythm of public and private means of transport, aerial and sea routes, the reciprocal analysis of books translated into various languages, the social effect of international travel, of records and films, etc. Thus one could ponder just what life is today and the destiny of the world (and not through the eyes of the large corporations or the rather inefficacious international organizations). In other words, I would like to be another Marey—adapted to the year 2000 and with access to the latest technology.

The continuation of research into movement seems certain, even if there is the same difference between an automaton from the past and a modern computer as there is between a wind-up watch and an electronic one. And perhaps at this point it is possible to specify the strange malady that has befallen the race of automatons. I believe it is connected to the process in Modern art which has destroyed traditional painting by preventing us from seeing the canvas as a window on the world, while also stressing physicality. So we have on the one hand the feeble suggestions of reality embodied in conceptual art, and on the other the hypothesis of figures interchangeable with reality, as in the sculpture of George Segal or Duane Hanson. This same polarization can be seen in the realm of automatons in more accentuated form. On the one hand we have circuits enclosed in plastic boxes; on the other, an urge toward biological creations, exact duplications of man, zombies. Traditional culture operates somewhere between the two poles, between conceptual abstraction and physicality without spirit, and thus has become obsolete.

At this point it is worthwhile to reread Karel Capek’s comedy—or better, tragedy—about automatons, R.U.R. (1921). The setting remains plausible: the island where the automatons are built could be Japan; the firm, Rossum’s Universal Robots, could certainly be located somewhere in Silicon Valley; and the daughter of a president (quite possibly the president of the United States) is also still credible in her desire to visit a factory creating a world market and achieving an absolute monopoly, placing in circulation more automatons than there are human beings. The office where the first act unfolds, with its big window looking down on rows of trucks, its large executive desk, and its map of major transport routes (which would now be aerial rather than naval), has also remained current. There would be modifications, of course: Domain, the general manager, would no longer need a robot in human form for typing, since a machine could now follow the sound of his voice on its own, producing typescript by integrating what was dictated with words stored in its memory. And Marius, the super intelligent robot who finally takes on an almost human dimension (a second Adam), would now have no need for arms and legs in order to memorize the contents of a great library; the information could be absorbed by a more direct process than the cumbersome one of reading.

Yet the basic program underlying the success of Rossum’s Robots is perfectly applicable today:

Nature has found only one method of organizing living matter. There is, however, another method, more simple, flexible, and rapid, which has not yet occurred to Nature at all. . . . Well, any one who’s looked into anatomy will have seen at once that man is too complicated, and that a good engineer could make him more simply. . . . A man is something that, for instance, feels happy, plays the fiddle, likes going for walks, and, in fact wants to do a whole lot of things that are really unnecessary. . . . But a working machine must not play the piano, must not feel happy, must not do a whole lot of other things. A petrol motor must not have tassels or ornaments. . . . And to manufacture artificial workers is the same thing as to manufacture motors. The process must be of the simplest, and the product of the best from a practical point of view.

Thus the automaton, which took form in the desire for a magic creature that could substitute for men and women, fulfilling their tasks, realizing their wishes, ceases to exist. The machine has been simplified. It has made itself less complex, less sensitive than human beings; the large robots on the automobile assembly lines are vulgar, without spirit or breath.

In postmortem, the automaton seems like a sort of benevolent angel upon whom we could project our desires and fears. These strange creatures, in whatever material they were made, were a living symbol of their antecedents—ancient gods, hibernating, who could be called to life by magical means to serve humankind. On a more modest level they anticipated future technological progress and so enabled us to take comfort in present asperity and failure. They were our kin, friendly or threatening, but in any case bearing the promise of change. Humanity has created many gods in its own image—angels, devils, legendary heroes—with which to fill the astral void, from which no radio observatory has ever received a meaningfully encoded message. And automatons in particular have been comforting doubles, in office work, in daily routine, in hard labor, in the dangers of war, in play. Their loss now leaves us truly alone. And we are left with the impression that we ourselves have become the true automatons—that we grow, react, decline, and die not according to nature but in step with a preestablished program, rigid and statistically verifiable.

Our process of decay is slower than that of many animals, faster than that of most trees, very rapid in terms of geologic time; our activities vary according to social and professional conditions but remain finite, however much we travel the world. In other words, the defects of automatons—their repetitiousness, their built-in limitations, their inability to change, their simple natures—are absolutely our own defects. And now automatons are not reproducing; they are whited tombs.

Eugenio Hattisti is an art historian who has written several monographs on artists of the Renaissance. He is president of the Italian Society for Industrial Archaeology.

Translated from the Italian by Meg Shore.