1In 2018, the French Ministry of National Education launched an initiative to promote the cognitive sciences and neurosciences, with the aim of reforming teaching methods and tackling academic failure. In 1988, Hermès devoted its third issue, “Psychologie ordinaire et sciences cognitives” (“Ordinary Psychology and the Cognitive Sciences”), to an examination of the role that the cognitive sciences might play in restoring “ordinary psychology” following the downgrading of the behavioral sciences in the face of such notions as “mental representation,” “belief,” “knowledge,” “plans,” and “goals.” The aim of that issue was in fact to test the validity of these notions in a context where we would refrain from assigning to behaviorism the task of explaining learning or strategies for the inculcation of knowledge. Indeed, for behaviorism, ordinary psychology (that of common sense) is false: we act not because we have certain mental representations, but because we are conditioned and subject to a certain degree of reinforcement of our responses to repeated stimuli. The cognitive sciences, on the other hand, first of all accept a certain proximity with ordinary psychology. We have internal states that are like mental representations, which allow us to store information on our surroundings and to act in one way or another. These representations require scientific elucidation, which will derive its resources from computing (cognitivism) or brain science (connectionism). In any case, they are the starting point for the cognitive sciences, and they allow for continuity between what is implicit within common sense and an elaboration of the scientific models to be tested. Twenty years before the decision was taken to return to the cognitive sciences in the French education system, Hermès argued the case for bringing a more scientific approach to the concepts used spontaneously by teachers in their work.

An examination of cognition

2The mastermind of Hermès’ third issue (1988) was Pascal Engel, who led a seminar on the philosophy of mind as part of the communication science program devised and directed by Dominique Wolton. It was Engel who showed the first theoretical interest in justifying the journal’s inclusion of cognition in its subtitle, alongside communication and politics. A re-reading of his presentation highlights the relevance of the assessment that he undertook. Who, out of Jerry Fodor, Patricia Churchland, or Donald Davidson should set the tempo for the research to be supported on the issues that stood before the journal? What should it be? A modularity of the mind to justify mental attitudes, an “eliminativism,” where the description of the mind falls back on neurophysiological explanations alone, or an “anomalous monism” sophisticated enough to avoid the reductionism of the neurosciences—as well as the idea that the mind must obey the laws encapsulated in its own architecture? Engel chose Davidson. He was not sure whether Stanislas Dehaene, the recently installed president of the Education Ministry’s scientific council, might opt for eliminativism, this being favorable towards neuroscience. At least, thirty years ago, we set the scene that would help identify the key issues around the trends that are now bringing science to the education process and removing it from the empiricism of pedagogues. We should also recognize the fact that we have set out more than one argument to resist the sirens of neuroscience.

3In the early days of Hermès, the presence of researchers from the Center for Research in Applied Epistemology or Crea (a CNRS unit at the École Polytechnique) enhanced the focus on the cognitive sciences in the course of the internal debate. Pascal Engel played the leading role, as I have mentioned; Alain Boyer, Pierre Livet, Jean-Pierre Dupuy, and I addressed the central question of communication with reference to the work we were conducting on the Rue de la Montagne-Sainte-Geneviève in the Sorbonne. We felt it important to examine what the notion of information had to bring to the “ordinary” concept of communication, such as might be associated with common sense in the vein of Emmanuel Kant, the Frankfurt School, Paul Lazarsfeld, or Dominique Wolton. Crea made it its specialty to investigate the genealogy of cybernetics in order to better describe the cognitive and social sciences. This could not fail to have an impact on the editorial board’s work on developing forthcoming issues. From the journal’s very first edition in 1988, Jean-Pierre Dupuy made the direction clear: “The Habermasian and neo-cybernetic trends are not unrelated.” Put another way, what was required was a joining of approaches between communication, as arising from Jürgen Habermas’s work on the public space (promoted at the time by Jean-Marc Ferry), and the output from Nobert Wiener and the Macy Conferences. It was an exciting path to pursue. Thirty years on, we can say that it was richly exploited. The simple analyses advanced by Dupuy on the works of Adam Smith provided a platform of sorts for a joint consideration of politics and communication in a theoretical space opened up by cognition. If only we had at least an idea of how the journal’s first readers reacted to the exposé on Smith’s concept of “sympathy,” founded on the principle of “contagion” or “imitation.” Empathy is a popular theme at the present time, no doubt since neuroscience established it from an extrapolation of the discovery of mirror neurons. In 1988, sympathy was above all a representation of a common sense to which philosophy and the social sciences could bring a political dimension. Jean-Pierre Dupuy provided an explanation of how Adam Smith’s “invisible hand,” being the centerpiece of the mechanistic concept of the political economy, could be used to stage individuals’ games of self-love with themselves—games that could be interpreted as a self-referential loop linking individuals with themselves via the intermediary of society. Aside from the rather rough, demonstrative requirements, the issue was clear for the researchers involved in the debates among Hermès’ editorial board members: the links between cognition, communication, and politics truly existed; moreover, they were required for the cause of the public space and democratic values. It only remained to get out and explore, without fear of trespassing in the garden of various disciplines as grouped together by the cognitive sciences.

4The linguistic sciences could not avoid being requisitioned, of course. The linguist Georges Vignaux was quick to back the process of joining up with the cognitive sciences. The epistemologist Alain Boyer played a role in the work that resulted in Hermès issues 15 and 16 in 1995, under the title “Argumentation et rhétorique.” Both Vignaux and Boyer helped ensure that the communicational dimension of politics gained historical and epistemological clarity. Thanks to them, rhetoric and argumentation, which form part of any political space, were able to deal with cognition, which was understood as being the domain where information and the recording of information were decoded and used. Over the course of the journal’s publication, the notion of information—which has become central to the paradigmatic representation that society from the 2000s has given itself—does not accept the many ways in which it is accepted: it belongs to the register of neuroscience, and it merges in with the idea of signal; information is defined as messages that are determinable in terms of probability, as is conceded by artificial intelligence theorists; it assumes an epistemic subject endowed with so-called propositional attitudes and capable of learning, as demanded by cognitive psychology; information is the translation of an event that has disturbed the equilibrium of a given subject with its surroundings, in accordance with how ordinary psychology would conceive it; and it calls for a naturalistic (even environmentalist) interpretation by becoming an objective commodity, drawing its authority not from any object, but from an active process of cognitive extraction. Having become a notion with almost unlimited semantic extension, information attracts interpretation and fresh perspectives in the pages of Hermès—which human and social scientists are still making good use of today. The success of issue 39, edited by Virginie Paul and Jacques Perriault (2004), judiciously entitled “Critique de la raison numérique” (“Critique of Digital Reason”), illustrates the impact that the analysis of the various approaches to information has had in universities—without sufficient clarity always being achieved on the cognitivist theories that underlie digitalization strategies. Another issue, edited by Vincent Liquète and Susan Kovacs, headed “Classer, penser, contrôler” (“Categorize, Think, Control”) (Hermès 66, 2013), offers an account of the steps that have been taken to classify information and thus facilitate its conversion into knowledge—the construction of meaning from the informality of data. Indexation, classification, and referencing, which require the management and exploitation of digitalized knowledge, would indeed appear to be of scientific interest to a journal that considers cognition, communication, and politics as inseparable.

An assumed anti-reductionism

5The above remarks on the polyphonic descriptions of information apply all the more to the branches of science that seek to understand the functioning, and in particular the mechanisms, of information capture. In providing a wide definition of the cognitive sciences, Pascal Engel (1988) instigated a program that the journal has fulfilled constantly, even if implicitly: “All disciplines that, in one way or another, examine the ways in which systems, whether natural or artificial, record, store, and process pieces of information, and have an impact on the world through this information, are ‘cognitive.’” However, if we were to go through past editions and identify the different theoretical models of cognition among their pages, would that mean giving in to rationalization after the event? Even if the journal has never set itself the task of challenging or assessing these models, it has nevertheless used them as interpretative tools for the realities inherent in communication and politics. In taking this approach, the journal has avoided unilateralist dogmatism and is fulfilling the scientific function that ought to be expected from researchers.

6In issue 67, devoted to the subject of interdisciplinarity, Dominique Wolton (2013) used a manifesto to make an appeal for eschewing discipline (or embracing indiscipline). Thanks to this, a community of researchers proved able to accommodate different contradictory models at the same time and, in doing so, promoted negotiation-communication, which is as necessary for science as it is for the political space. In fact, it is not difficult to take past issues of Hermès and identify the three main models that explain intelligence and cognitive skills: 1) cognitivism, which contends that cognition involves processing recurring information through the controlled manipulation of symbols; 2) connectionism, which understands cognition as the result of emerging configurations of discrete elements in a problem context; and 3) enaction, which associates cognition with a history of coupling self-organized systems. Credit for this third and final concept, which goes beyond the references to computers or the brain made by the first two, can largely go to the Chilean biologist Francisco Varela, a former Crea researcher whose work has been drawn upon in several Hermès articles (see Le Blanc 2014, for example). Pierre Livet cross-referenced the works of Varela and his autopoietic approach to the living when examining the origin of the conventions that bind communities together without them even knowing the level of organization they have achieved. This is indeed a fine demonstration of the synergy of cognition, communication, and politics (see, for example, Livet 1988). There is certainly no orthodoxy in our journal; rather, there is a salutary documentary pragmatism, which although it does not contend that “anything goes,” as Paul Feyerabend does, it asserts that nothing should be rejected a priori, as long as it can successfully be exposed to argument. The cognitive sciences have passed this test under the three models I have just mentioned. They have now been metabolized, if I may put it that way, within the often turbulent culture of Hermès. “Metabolized,” perhaps, but that is not to say that they are considered routine: every time they are held up, Hermès’ antireductionist tendencies—which lead us to reject technological determinism, for instance—will come to the fore.

7Antireductionism is part of the matrix at Hermès. This was shown in an applied manner in the issue recently edited by Franck Renucci, Benoît Le Blanc, and Samuel Lepastier, under the title “L’Autre n’est pas une donnée. Altérités, corps et artefacts” (Hermès 68, 2014). To reduce is always to undermine the other—to eliminate it or dissolve it somehow. In some respects, reduction is essential to science, which seeks to eliminate chance, and hence break free from otherness. The resulting antireductionism can bring some perspective, however: whereas methodological reductionism is necessary to understand types of reality, ontological reductionism makes an improper attempt to take possession of the ultimate element from which it should be possible to deduce everything—the atom, the neuron, the gene, the cogneme, and so on. In issue 68, Franck Renucci interviews Francis Wolff, who makes a call to “resist reduction” (Wolff 2014). Indeed, that is where the journal’s essential epistemological program lies. In this sense, the “metabolization” of the cognitive sciences represents a prerequisite: it means we can be immediately suspicious of any attempt to remove the otherness that helps to make up knowledge relationships; it maintains the discontinuity that can delimit problems or situations that are subject to the cognitive act, even though the latter may be of an enactive nature. To paraphrase Wolff, we are discontinuists. We do not believe, for example, that humans are animals like any other, and that they can therefore be the subject of scientific investigation deriving solely from a determinism of natural phenomena. This is not an entirely easy thing to assert, in particular having heard Claude Lévi-Strauss announce—following Friedrich Nietzsche—that the role of the human sciences is to dissolve human beings in nature. And yet, that is the challenge: to allow for otherness as a resisting force, which demands experience from communication; and to accept otherness as an argument for objecting to the cognitive sciences, which are showing signs of having had enough of the Other, and of willingly reducing communication to the homeostasis of the cyberneticists, rather than the intersubjectivity of philosophers. Cognition is a truncated expression of the challenge of always compelling the cognitive sciences to play with and against themselves. Hence, within a single issue of Hermès, the reader will find a child psychiatrist’s vigorous critique of the neurosciences, which are found guilty of disregarding the body incarnate, desire, and language—and thereby ignoring what makes us more than simple animals (see Falissard 2014)—and a knowledge engineer’s indictment of the distorting tendency affecting cybernetics among the NBIC fields (see Claverie 2014). Strictly speaking, it would be wrong to say that antireductionism is part of Hermès’ DNA or religion: it has become a spontaneous commitment for the journal, as indispensable as morals were for Descartes.

The body revived

8To conclude, a few words on the increasingly frequent presence of the body in Hermès. It does no good for the editorial board to exercise undue caution for fear of acknowledging the growing power of virtual technology and the machinization of human beings in technologized communities. It should rather be considered that questioning “what the body can do” (Spinoza) is not a matter for either communication or politics. We may note that the cognitive sciences are once again the order of the day, bringing knowledge to the mechanisms of sensoriality. Brigitte Munier and Éric Letonturier were well aware of this when they devised issue 74, headed “La voie des sens” (2016). One possibly surprising effect of this issue was that it decoupled the cognitive sciences from the transhumanist pretensions of getting rid of the biological body. But no—cognitivist theorists are not in fact all committed to the program of technological convergence. Among them are researchers who have demonstrated that the senses are not mere information sensors, but also operators of communication and indicators of human beings’ resistance to the intimate. When Benoît Le Blanc and Bernard Claverie (2016) discussed the auditory attention mechanisms that allow us to sift out relevant details from among the barrage of information we receive, they did not fail to mention the inability of robots to perform the same task. When André Holley (2016) presented his assessment entitled “À propos du renouveau des recherches sur l’olfaction” (“On the Revival of Research on Olfaction),” he demonstrated the fertility of a branch of the cognitive sciences that is not forever doomed to feed the various fantasies generated by the NBIC fields. With this issue devoted to the senses, the body made a triumphant revival in all its carnal density—far removed from the techno-scientific reduction that only sees a passive surface destined to disappear, thanks to machines.

9At the beginning of this article, we recalled that the cognitive sciences are now in demand in the classroom. Let us hope that Hermès’ engagement in the field of cognitive science has demonstrated its resistance to the threat of scientism at a time when we are supposedly basing approaches to education on rules derived from scientifically established facts. While it is not clear whether the instigators of future reforms will heed this advice, the journal can at least offer its readers contributions that avoid the pitfalls of dogmatism. The ordinary psychology of teachers can be examined in many other areas—i.e. cognition, communication, and politics—regarding such urgent topics as attention, the emotions, motivation, and intention. Pedagogical strategies aimed at increasing the mental retention of information, at organizing distributed learning, and at shedding light on the pathways of metacognition come under the anti-reductionist approaches espoused by Hermès. Put briefly, the projects started up under the ambitions of the Ministry of National Education in 2018 have already been long defined by a journal that has never proposed cognition as the sole determinant of the causal factors of the competences required for learning. Although this is not an offer of service, we wish only that the overview we offer here of how the concept of cognition has been developed epistemologically in Hermès may resonate among school reformers.