Coding and Decoding

A review of Code: From Information Theory to French Theory, Bernard Dionysius Geoghegan, Duke University Press, 2023.

Is there a pathway that goes “from information theory to French Theory”? Straying away from the familiar itineraries of intellectual history, Bernard Dionysius Geoghegan invites us to take a path less trodden: a detour that allows the reader to revisit famous milestones in the development of cybernetics and digital media, and to connect them to scholarly debates stemming from fields of study as distant as structural anthropology, family therapy, and literary semiology. Detours and shortcuts are deviations from linear progression, reminding the traveler that there is no one best way to reach a point of destination. Similarly, there are several ways to read this book. One is to start from the beginning, and proceed until the end, from the birth of communication science during the Progressive Era in the United States to the heydays of French seminars in sciences humaines in the Quartier latin before mai 68. Another way is to start from the conclusion, “Coding Today”, and to read the whole book in reverse order as a genealogy of the cultural analytics used today by big data specialists and modern codifiers of culture. A third approach would be to start from the fifth and last chapter on “Cybernetics and French Theory” and to see how casting cultural objects in terms of codes, structures, and signifiers relates to previous methodologies of treating communication as information, signals, and patterns. The common point of these three approaches to reading Code is to emphasize the crossing of boundaries: disciplinary boundaries between technical sciences and the humanities; political demarcations between social engineering and cultural critique; and transatlantic borders between North America and France. The gallery of scientists and intellectuals that the book summons is reflective of this broad sweep: Norbert Wiener, Warren Weaver, Margaret Mead, Gregory Bateson, Claude Lévi-Strauss, Roman Jakobson, Jacques Lacan, Roland Barthes, and Luce Irigaray are seldom assembled in a single essay; yet this is the challenge that Code raises, inviting us to hold together disciplines and methodologies that are usually kept separate.

The empire of code 

Let’s start from the present and move it from there. “Coding” now mostly means writing lines of code or computer software using a programming language such as JavaScript, Python, or C++. Codes can also designate social norms or cultural imperatives governing acceptable behavior in a certain context or within a subgroup. To “know the codes” means to be able to navigate a certain social world without committing blunders or impropriety. Of course, social scientists have taught us that social rules are best obeyed when one is not conscious of their imperium. Social norms must become embodied knowledge to be played spontaneously, and the best performance has the charm and immediacy of the natural, the innate, the unrehearsed. Culture cannot be recitated as a learned lesson or a set of rules. When social life is reduced to a system of codes, decontextualized from its rich background and reformatted for transnational circulation, then it becomes a simulacrum. This is why we should worry about the extension of the domain of the norm that is fueled by the twin forces of globalization and digital technologies. We are witnessing the weakening of the notion of culture, once thought of as a set of evidences shared and anchored in a territory, and today reduced to a corpus of explicit norms and cultural markers, which circulate on a global scale. The crisis in culture that Hannah Arendt diagnosed in 1961 has now given way to culture’s opposite: the reign of the explicit, the quantified, the normative. The disappearance of high culture as a shared implicit within territorial and social boundaries gives way to the sequencing of small bits of cultural content that are recombined to form a marketized commodity, as in UNESCO’s heritage list of intangible assets. These packets of texts and images circulate through networks that separate them from their point of origin and delivers them to the right place. If the network changes, due to congestion or broken links, routers can use an alternative interface to reach destination. 

There is a growing disconnect between the territory in which we live and the cultural references that we manipulate. National or religious identity is redefined as a set of cultural markers and signs of belonging that are decomposed and recomposed into new individual selves that are both unique and interchangeable. Coding implies normativity. We need new norms and regulation because things that seemed obvious, at least within a given cultural space, are no longer so. If everything is open to discussion and contestation, then we must make the rules explicit and as detailed as possible. This codification of social practice considerably reduces inner spaces of freedom and nonnormativity: the intimate, the private, the unconscious. Normativeness is the consequence of coding, the passage to the explicit, the quantification of affects. A grammar, for example, is a code and when we make a mistake, we are corrected. Contrary to language, code is acquired by apprenticeship or formal training: one must know the rules to practice coding, whereas it is not necessary to know grammar to practice a language. Coding follows a model of communication that makes each term explicit, where the receiver understands exactly what the emitter wants to say. This applies to social interactions, where what was previously left unsaid now needs to be specified, and even to the use of language, with the spread of global English and the standardization of public expression. In a multicultural context, it is recommended to speak as clearly as possible without using allusions, cultural references, and humor. The spread of artificial intelligence and chatbots will only reinforce this trend: in order to make ourselves understood by machines, or to allow machines to communicate between themselves, we must separate language from culture and minimize the noise generated through the process of encoding and decoding. 

The age of the seminar

This becoming-code of all cultural contents and social interactions has a long history. A surprising milestone in the advent of code is to be found in the works of philosophers, literary critics, and semioticians that are sometimes bundled together in the United States under the label of “French Theory.” Coding and decoding were definitely code words in French intellectual discussions during the 1960s and 1970s. “Assez décodé !” (Stop decoding/stop fooling around) was the title of a popular essay in 1978 that took aim at Roland Barthes’ new literary criticism and the abuse of technical jargon. Geoghegan identifies the 1960s as the period when “culture as communication” gave way to a preoccupation with “culture as code.” Cybernetics and information theory acted as both model and test bed for this transformation. They were part of a broader trend of social transformation based on the import of American technologies and institutions to fit postwar France’s condition. Techniques of management and human engineering were adopted en masse by an increasingly technocratic France. Funding from American foundations, tracing back to fortunes accumulated by robber barons and with links to the Cold War intelligence apparatus, supported the creation of research institutions that set new modes of organizing critical inquiry in the humanities and social sciences. A new research center and central forum for teaching social sciences was created within the Ecole pratique des hautes études as the “sixième section,” better known as the Ecole des hautes études en sciences sociales or EHESS. It modeled aspects of its study program on the social sciences in the United States, distancing itself from previous modes of scholarly organization in French universities. Its scope was resolutely transdisciplinary and experimental.  It pioneered the use of statistical methods and mathematical models in the humanities. Indeed, there is a book to be written on the fascination, some would say the math envy, exerted by mathematics and formal science on French social scientists as diverse as Claude Lévi-Strauss, Pierre Bourdieu, and Jacques Lacan. One locus for such collaboration was Lévi-Strauss’s research seminar on the utilization of mathematics in the social sciences,  which let to long-lasting interdisciplinary collaboration between scientists and social critics.

The research seminar thus became a key site for the clinical analysis of the human condition, remote from the elegant discussions in cafés and salons that previously exemplified intellectual authority in France. The seminar was the domain of the expert, the specialist, the fieldworker. It displayed science in the making, and opened its ranks to any social scientist who had new research results to share, regardless of academic position or social authority. Later on, Michel Foucault would label this new kind of postwar thinker a “specific intellectual” whose political responsibility was akin to that of the “nuclear scientist, computer expert, and pharmacologist.” Structuralism imposed itself as the dominant paradigm, with its emphasis on codes, systems, communication, economy, and even informatics patterning of signs. The promise of scientific precision and far-reaching advances attracted younger scholars eager to chart bold yet rigorous programs in emerging research areas. Human sciences as envisioned by Claude Lévi-Strauss had one great aim: “the consolidation of social anthropology, economics, and linguistics into one great field, that of communication.” In particular, “social anthropology,” he wrote, “can hope to benefit from the immense prospects opened up to linguistics itself, through the application of mathematical reasoning to the study of phenomena of communication.” Lévi-Strauss was an enthusiastic reader of Shannon and Weaver’s Mathematical Theory of Communication (1949). One of his early papers on the relevance of cybernetics on linguistics argued that engineering models of communication could be transposed onto all other fields of human activity, including linguistics, economic transactions, and the circulation of women within primitive systems of kinship. Through the 1950s, Lévi-Strauss sought to establish a physical infrastructure equal to the tasks of his emerging structural anthropology. His ascension to a chair at the Collège de France in 1960, and his concomitant establishment of the Laboratory of Social Anthropology, presented him with the long-sought opportunity to establish a research laboratory. One of his first initiatives was to acquire a copy of the Human Relations Area Files, a searchable database of two million index cards compiling ethnographic findings. Vast regimes of human data were disassembled into informational units for cross-cultural analysis. They were part of a global apparatus of knowledge that, paradoxically, unmoored cultures from local and embodied reality. Headquartered in Paris, UNESCO offered an early vehicle for bringing these new political techniques to the world.

Back to the future

Code insists on the transatlantic origins of the dominant paradigm in the sciences humaines, both institutionally and in terms of substance. The history of structuralism and poststructuralism has often been told, with an emphasis on the John Hopkins conference of 1966 that spearheaded the reception of French contemporary thought in North America. Here Geoghegan goes further back in time to highlight the way European nascent human sciences were incorporated into emerging logics of US communication science during World War II. As war swept Europe, the Rockefeller Foundation mobilized to bring threatened European intellectuals under the umbrella of US wartime science. An early recruit was Russian-born linguist Roman Jakobson, who founded the Linguistic Circle of New York in 1943 as a successor to the celebrated Prague Linguistic Circle, mixing structural linguistics initiated by the Swiss linguist Ferdinand de Saussure with diverse insights from fields including Russian formalism, avant-garde art such as futurism and cubism, and relativity theory developed in atomic physics. For Saussure, language was like a game of chess: one did not simply speak but selected from among a field of possibilities prefigured by formal constraints and anticipated threats. With Jakobson, language became probabilistic and combinatoric, ordered on principles that followed the direction of cybernetics and communication science. Much as Warren Weaver and Claude Shannon used probabilistic sequences to predict series of words, phrases, and sentences, Jakobson described phonemes as probabilistically encoded and decoded series. Another Rockefeller foundation initiative was the establishment of the Ecole libre des hautes études in New York, which recruited Claude Lévi-Strauss but declined to support Jacques Lacan. Under Jakobson’s influence, Lévi-Strauss ceased to study the empirical facts of indigenous kinship and focused instead on the relations among terms that constituted a kinship system proper. With the aid of a French mathematician, he even found algebraic expressions for his kinship studies. The linguistics seminar Jakobson and Lévi-Strauss held at the Ecole libre made a field trip to AT&T headquarters to witness the performance of the Voder, a synthetic speaking device, in 1944. According to Geoghegan, the Ecole Libre was a methodological crucible, nudging French scholars away from a concern for social equality and redirecting them in technocratic directions. As he remarks, “this was indeed a strategy of political transformation of the sort that would become a pillar of American ‘nation building’ in decades to come.”

The last thesis proposed by Geoghegan—or the first if you follow the book order, from chapter one to chapter five—, is that cybernetics wasn’t an invention of World War II and the Cold War, as science historians sometimes assume. Code shows that “links among the Rockefeller, Macy, and Carnegie philanthropies forged in the 1930s and 1940s, well before the United States’ entry into World War II, guided subsequent initiatives in cybernetics, information theory, and game theory.” The roots of the project lie in Progressive Era technocracy and its agenda to transform social strife into communication engineering problems available for technical problem-solving. Welfare policies, not warfare, were the test bed for the rise of the communication sciences, and its first deployments were to be found in the colony, the clinic, the asylum, and the urban ghetto. As Geoghegan observes, “dreams of cybernetic post-humanism depended on disappearing the bodies of native persons and other subjects regarded as less than human.” Anthropologists Ruth Benedict and Margaret Mead thought that all existing human cultures were distributed along a great “arc” which covered the whole range of possible cultural traits. Each culture then selects along this arc a “pattern” of human possibilities that fits its environment and forms a coherent whole. After his pathbreaking master degree thesis that laid the groundwork of information theory, Claude Shannon’s PhD dissertation, completed in 1940, applied Boolean algebra to the orderly processing of eugenic data. The celebrated Macy Conferences on Cybernetics, initially convened in 1942, brought together mathematicians, anthropologists, engineers, and scientists from other disciplines, and popularized notions such as reflexivity, feedback loops, and error correction mechanisms. Scientific networks cultivated in the 1930s and consolidated in wartime military projects laid the foundation for interdisciplinary communication projects well into the 1950s.

Return to sender

There is a tendency to downplay the links between the natural sciences and the dominant paradigms in the humanities. This book show that the history of the human sciences in the twentieth century cannot be separated from the rise of the communication sciences. Fields such as anthropology, psychology, and semiotics served as experimental laboratories for the engineering of a society of digital media and codified culture. Far from trailing behind engineers and natural scientists, human scientists spearheaded the reconceptualization of cultural forms as forms of code that could be decomposed and recombined using mathematical tools. Efforts to transform the humanities and social sciences into a single field, the human sciences, oriented toward communication, cannot be separated from the rise of scientific philanthropy. The Rockefeller Foundation and a host of like-minded philanthropies funded by robber barons (e.g., the Ford Foundation; the Josiah Macy Jr. Foundation; the Wenner-Gren Foundation) lavished generous funding on interdisciplinary research linked to research programs inspired by cybernetics and information theory. Their midcentury interest in these fields reflected progressive hopes to submit divisive political issues for neutral technical analysis. The long-standing aim of American philanthropies to reorient the humanities toward exact, quantifying, empirical, and rule-governed theoretical analysis found fertile ground in postwar France. Even if we should use the expression “French Theory” with caution, there was a theoretical impetus toward formalization, even a “math envy,” that shaped the dominant paradigms of structuralism and poststructuralism. A cybernetic turn of mind influenced French structuralists’ talk of codes, systems, and communication. While Barthes’s contrarian attitude or Lacan’s extravagant vocabulary carried a critique of technocratic rule, their seminars fit within the period’s emphasis on experts, codification, and structures. Their effort to remake French thought also ended up remaking American thought along the way. If we summarize the standard model of communication as a message sent by an addresser to an addressee through a channel involving operations of coding and decoding, the development of French Theory on American campuses was a case of return to sender.