In this lecture, I propose to take the formula of “cybernetics for the 21st century” seriously. This formula holds a paradoxical dimension, insofar as the signifier “cybernetics” - and even worse, the prefix "cyber" – are largely demonetized and appear outdated. However, the hypothesis that can be made is that we are living in a "third phase" of cybernetics, which unlike the first and second, does not say its name.

This phase is characterized by the fact that the artifacts projected by cyberneticians are now entering society on a large scale. This is the case for the “Machine Learning Tsunami,” which is based on the neuroconnectionist models that were those of the cyberneticians; but it is also linked to the fact that information technologies have left the Internet to irrigate new industrial sectors, in robotics, transport, logistics, etc. From this point of view, the analogy is striking between Wiener's 1948 description of the automated factory as "the complete animal with sense organs, effectors and proprioceptors" and technical objects such as the autonomous vehicle or Amazon's robotic warehouse. It is only now, with a 70-year time lag, that cybernetic artifacts are taking hold, with the form of autonomy and agentivity that the first cyberneticists attributed to them.

But what is interesting in this "unstated" return of cybernetics is the discordance of times. The mode of production of science and technology has changed radically since the immediate post-war period, with the passage from sciences nationalized for the war, a regime of science against which cybernetics presented itself in part as a counter-model, to a regime of techno-scientific production largely privatized and piloted by large corporations. It is particularly interesting in this context to go back to the first cybernetics and revisit their arguments, not to transfer them as they are to the present situation as a set of forgotten discursive resources, but to measure the gap that separates us from them.

This "return to cybernetics" has a double interest. First, in the field of philosophy. We know Heidegger's famous formula about cybernetics as the metaphysics of the atomic age. But in fact, one of the characteristics of cybernetics is to have produced a philosophical discourse. Cybernetics is characterized by a particular style of science, marked not only by a radical interdisciplinarity, from mathematics to engineering sciences, to psychology, life social sciences, but also by the production of a philosophical reflection, directly linked to technical development. One of the most important aspects of this "spontaneous philosophy" lies in the internal debates within cybernetics concerning the nature of information. Cyberneticians relativize the symbolic dimension of information in favor of an alternative conceptualization, which I call "the signal" rather than "the code," that returns to information as a "physical term."

Now, this informational materialism also has effects from a political point of view. This is the second interest of this "return to cybernetics." There is, in Wiener's work, a transfer between these ontological debates about the nature of information and a political stance about the impact of the new technologies that cybernetics promotes. It is therefore this nexus - ontology and politics of information - that I wish to present, confronting it with the contemporary situation.

For this talk, I will begin with a historical review of early cybernetics to situate the movement and assess its successes and failures. I will then return specifically to the question of the interpretation of information as a theoretical term, drawing in particular on the discussion of the analog/digital distinction during the 7th Macy conference. Finally, I will present and discuss a catalogue of precepts for information politics, as they can be extracted from Wiener's writings.