The Experience of Matter (Still) Calls for the Symbol

When Claude Shannon demonstrated, in his 1937 master’s thesis, A Symbolic Analysis of Relay and Switching Circuits¹, that George Boole’s logic could be used to describe symbolically the functioning of electronic circuits, that logic had been lying dormant in libraries since its creation in 1854, even though several electronic circuits were already operational at the time², built and industrialized empirically and on the basis of test boards³. The ENIAC, a computer built in the United States in 1945 by a team led by John William Mauchly and J. Presper Eckert, was not itself designed on the basis of Boolean algebra – nor of the von Neumann architecture, for that matter – because Mauchly and Eckert “did not think in terms of logical states, but in terms of electrical pulses routed through different parts of the machine⁴”

Besides Boolean algebra, other logical systems existed at the time, but proved less useful because they were ill-suited to a technical reality that was taking concrete shape experimentally. This is the case, for example, of logics based not on sets of values but on mathematical functions, such as the lambda calculus invented by Alonzo Church in the 1930s, and the combinatory logic developed in 1929 by Haskell Brooks Curry⁵. The three-valued logic invented by Jan Łukasiewicz did not find any direct practical application either, for lack of coherence with the concrete electronic components favored in practice. These alternative logics found interest in mathematics and theoretical computer science, but not in the description of concrete technical configurations, as was the case for Boolean algebra, which was better suited to the existence of on/off dispositifs, able either to let the current through or to stop it.

Thus, it would be a mistake to see in the birth of computers an invention that first takes place in thought and is subsequently implemented materially. On the contrary, it is a technical evolution already at work, which finds, among the multiple symbolic expressions available at the time, the one that suits the material functioning of the dispositif. The scientific principles, sometimes abstract, that begin to envelop the technical object while affecting its evolution (its individuation) must not be considered as the essential core defining the nature of the technical object, for the latter already existed, the outcome of an experimental process in which it evolves in contact with humans and with its environment. In the words of Gilbert Simondon:

The study of the schemas of functioning of concrete technical objects has a scientific value, for these objects are not deduced from a single principle; they are the testimony of a certain mode of functioning and of compatibility that exists in fact and was constructed before having been foreseen: this compatibility was not contained in each of the separate scientific principles that served to construct the object; it was discovered empirically⁶.

It is above all through the progressive implementation of empirical improvements that technics advances, sometimes thanks to happy accidents. In this light, technics appears first of all as a thought in act, emerging from the manipulations performed on the materiality of the world. Abstraction and symbolic representation come afterwards: to model, to organize, to guide standardization and ensure transmission, while making it possible to optimize the development of technics so that it can reach higher levels of complexity.

1. Claude E. Shannon, A symbolic analysis of relay and switching circuits [en ligne]. Master’s thesis. [s. l.] : Massachusetts Institute of Technology, 1937. disponible à l'adresse : https://dspace.mit.edu/handle/1721.1/11173.↩︎

2. Such as needle and ratchet telegraphs and electromagnetic communication relays.↩︎

3. In fact, in a movement that had already begun in 1920, electrical engineering, which never ceased to be an art of practice, let itself be guided more and more by mathematics. Boole’s logic, for its part, was applied only gradually, and it never ruled out the need for engineers’ tinkering. See on this subject Maarten Bullynck, The engineer’s “peculiar sort of frustration” with Shannon’s Boolean algebra [en ligne]. Université Paris 8, 2016. disponible à l'adresse : https://shs.hal.science/halshs-01541609v1.↩︎

4. Paul E. Ceruzzi, A history of modern computing. 2nd ed. Cambridge (Mass.) : MIT press, 2003.↩︎

5. Jonathan P. Seldin, «The logic of church and curry ». dans : GABBAY, Dov (dir.), The handbook of the history of logic. [s. l.] : Elsevier, 2009, p. 5–819.↩︎

6. Gilbert Simondon, Du mode d’existence des objets techniques, op. cit., p. 53.↩︎