Human automata

 “I think everybody should be a machine”,

Andy Wharhol, Interview with Gene Swenson, Art News (1963).

The term human automaton refers to machines capable of imitating human characteristics such as appearance, movements, actions or even mental processes. This idea is related to devices already present in the ancient world, in civilizations like the ancient China or the Greek culture. In this latter case, references can be found not only in Greek mythology but also in the Hellenistic world, where automatons were intended as tools, toys, religious idols, or prototypes for demonstrating basic scientific principles, although never intended to equal nature: according to the Greek culture artefacts could never become as real as natural objects.

Yet, the manufacturing tradition of automata continued in the Greek world well into the Middle Ages. At that time, also in the Islamic culture, complex humanoid automata were described and built. The interest for this devices revived in the Renaissance (Leonardo’s articulated mechanical robot for the Ludovico Sforza festivities is a good example of it). Later, whith Renaissance technology and mecanicist philosophy, the idea of the human being considered as a machine gained importance. The most important contributions to this conception were made first through Descartes’ division of man into a mechanistic body and an inmaterial soul, and later by J. Offray de la Mettrie, one of the leading supporters of the metaphor of man as a machine, as well as by Vaucanson’s famous automata. La Mettrie considered that animals and humans were like mere automatons or machines and denied the existence of the soul as a substance separated from matter:

Comme une corde de Violon ou une touche de clavecin frémit et rend un son, les cordes du cerveau, frappées par les rayons sonores, ont été excitées à rendre ou à redire les mots qui les touchaient. Mais comme telle est la construction de  ce viscère, que dès qu’une fois les yeux bien formés pour I ‘optique ont reçu la peinture des objets, le  cerveau ne peut pas ne pas voir leurs images et leurs différences : de même, lorsque les signes de ces  différences ont été marqués, ou gravés dans le cerveau, l’âme en a nécessairement examiné les rapports; examen qui lui était impossible sans la découverte des signes, ou I ‘invention des langues. Dans ces temps, où I ‘Univers était presque muet,  l‘âme était à l‘égard de tous les objets, comme un  homme qui, sans avoir aucune idée des proportions, regarderait un tableau, ou une pièce de sculpture: il n’y pourrait rien distinguer; ou comme un  petit enfant (car alors I ‘âme était dans son enfance) qui, tenant dans sa main un certain nombre  de petits brins de paille ou de bois, les voit en général d’une vue vague et superficielle, sans pouvoir  les compter ni les distinguer. Mais qu’on mette  une espèce de pavillon, ou d’étendard, à cette pièce  de bois, par exemple, qu’on appelle mât, qu’on en  mette un autre à un autre pareil corps; que le premier venu se nombre par le signe 1 et le second  par le signe ou chiffre 2 ; alors cet enfant pourra les  compter, et ainsi de suite il apprendra toute I ‘arithmétique. Des qu’une figure lui paraitra égale à  une autre par son signe numératif, il conclura sans  peine que ce sont deux corps différents; que 1 et 1  font deux, que 2 et 2 font 4,* etc.

Julien Offray de La Mettrie, L’homme machine, 1747

In modern times, some artists have reinterpreted this vision: Fritz Kahn, a physician and writer of popular science books, was the responsible for the idea of several illustrations (made by others on his instructions) representing human physiology as operated by machines and depicting the body as a factory, relating it to the industry and his spaces, so common at that period of time. This link shows an animated and interactive application made in 2009 by Henning M. Lederer and based on Kahn’s poster “Man as Industrial Palace” from 1926.

The first apparatus that can be considered as intended to imitate mental processes (a mental automata), probably more symbolic than real, are the calculating machines. In the 18th century the concept of “mathematical automaton”, that corresponds to the model designed by Pascal, was that of a system composed of gears and other pieces intended to produce the movement of printed or engraved figures and to performe the rules of arithmetic. A century later, Babbage’s machines, especially the analytical engine, marked an important step forward in the process of automatization, not only of mathematical operations but also of mental tasks. The purpose of his machines, that can be considered as an extension of the human brain, was to avoid human errors in the calculus, and they reflected Babbage’s idea of translating the division and organisation of physical tasks into mental processes (as Gaspard De Prony did in the Enlightenment to produce logarithmic and trigonometric tables for the French Cadastre). Some authors attribute him the intention to automate intelligence but Babbage himself dismissed the idea that his machines were able to think; as Ada Byron stated:

The Analytical Engine has no pretensions whatever to originate anything. It can do whatever we know how to order it to perform. It can follow analysis; but it has no power of anticipating any analytical relations or truths. Its province is to assist us in making available what we are already acquainted with”.

Ada Augusta Lovelace, “Sketch of the Analytical Engine invented by Charles Babbage Esq. By L. F. Menabrea, of Turin, with notes upon the Memoir by the Translator”, 1843

Although his machines, and especially the analytical engine, were provided with memory and were capable of taking decisions, it was not human being but industry that Babbage had used as a model. As a matter of fact, at that time also factories were considered as automata, an idea that is well reflected in this quote from Andrew Ure:

Some authors, indeed, have comprehended under the title factory, all extensive establishments wherein a number of people co-operate towards a common purpose of art; and would therefore rank breweries, distilleries, as well as the workshops of carpenters, turners, coopers, &c., under the factory system. But I conceive that this title, in its strictest sense, involves the idea of a vast automaton, composed of various mechanical and intellectual organs, acting in uninterrupted concert for the production of a common object, all of them being subordinated to a self-regulated moving force. If the marshalling of human beings in systematic order for the execution of any technical enterprise were allowed to constitute a factory, this term might embrace every department of civil and military engineering; a latitude of application quite inadmisible.

Andrew Ure, Thre Philosphy of Manufactures, Londres, 1835, pp. 13-14.

For a long time, it was not accepted culturally and socially that machines could be capable of reproducing mental faculties. This mistrust continued well into the XXth century, when the first computers were designed:

It was towards the end of 1946 that the British public first began to hear about computers, which the press would insist on calling electronic brains. In November of that year, Lord Louis Mountbatten used this term to describe the ENIAC in an address he gave to the British Institute of Radio Engineers. Hartree wrote to the Times protesting, but adding that it was true that an electronic calculating machine “can be set up in such a way as to exercise a certain amount of judgement”. He was challenged to explain exactly what this meant by a correspondent who observed that the left hand would seem to be giving back what the right hand had taken away. In a second letter, Hartree defended the use of the word judgement, but in terms so obscure that his readers can have been left little the wiser.

In spite of all protests, the press continued to use the term electronic brain, along with certain variants, for example, mechanical brain and giant brain. This, together with the general use by computers designers of terms –such as memory– based on physiological analogy, made some people jump to the conclusion that the claim was being made that computers were something more than mere machines. The matter surfaced in the press in June 1949, when Sir Geoffrey Jefferson, Proffesor of Neurosurgery at Manchester, delivered the Lister Oration to the Royal College of Surgeons, taking for his title “The mind of Mechanical Man.” Jefferson evidently thought that dubious statements were being made and that he should refute them. His flights of rhetoric about machines not being able to write sonnets and know that they had written them, or feel emotion, were eminently quotable and were a gift to the press. “When we hear it said that wireless valves think”, he said, “we may despair of language.”

Turing was asked for his comments. What he actually said will never be known, but what came through in the Times was that the main preoccupation of the University of Manchester was to find “the degree of intellectual activity of which a machine was capable and to what extent it could think for itself”. This was too much for Dom Illtyd Trethowan of Downside Abbey who wrote to the Times expressing the conviction that all responsible scientist would be quickly to dissociate themselves from this work; he mentioned Butler’s Erewhonians who felt the necessity to guard themselves against the possible hostility of the machines and stated his own belief that men are free persons. In the true traditional style of Times letter writing, he ended by demanding to know how far Turing’s opinions were shared, or might come to be shared, by the rulers of the country.

Maurice Wilkes, Memoirs of a Computer Pioneer, The MIT Press, 1984, pp. 195-196.

Today, the interest in automata is related to the intense debate over artificial intelligence, the limits of machines capable of thinking and making decisions, and the shift in certain human initiatives such as creativy or responsibilities. Some of the machines that are being developed are aimed at solving a specific problems or exectuting specific task, and others seek to imitate human beings, including their mental processes.

For further reading see Bernadette Bensaude-Vincent and William R. Newman (eds.) (2007), The artificial and the natural: an evolving polarity, The MIT Press; Víctor Guijarro & Leonor González (2010), La quimera del autómata matemático. Del calculador medieval a la máquina analítica de Babbage, Madrid, Cátedra (especially chapters III and VII); and Simon Schaffer, “Enlightened Automata”, in William Clark, Jan Golinski, and Simon Schaffer (eds.), The Sciences in Enlightened Europe, Chicago University Press, 1990.

Advertisements