Introductory Note

This article was written for the Scientific Council of the CNRS (Centre National de la Recherche Scientifique) where it was presented on 14 January 2002. It was used as the introduction to a full-day working session titled “The evolution of knowledge domains, interdisciplinarity and core knowledge”. It was intended as both a general reflection and a proposal for action.

1.

It will certainly be understood that my remarks that follow can by no means cover all the knowledge domains within the CNRS. This might be a task worth pursuing, though it could only be done by a group of researchers with the qualifications and time that are not at my disposal. I have thus chosen the more limited goal of reflecting upon the question of the evolution of knowledge domains in the medium term and of proposing, as far as possible, bold and nonconformist readings that can help in thinking about the current situation. More precisely, I’ll begin by putting the question in an historical perspective, hoping in this way to draw out the blind spots or the perfunctory, and perhaps simplifying, frameworks of analysis.

I’ll begin with a few general remarks resulting from a reading of some of the reports written a year ago by various divisions of the National Committee of the CNRS[1], reports by the heads and directors of interdisciplinary departments and programs. The first remark concerns the very strong tension I feel between three persistent points. One has to do with the defense of what constitutes the primary objective of the CNRS, basic research: a ubiquitous notion found in nearly all texts, but difficult to define today. The second point has to do with the imperious duty of trans- or interdisciplinarity, affirmed as the primary value by all the authors and directors of the institution; yet, the notion is ambiguous, especially as interdisciplinarity is not a good in itself but an approach which, only in certain contexts, can turn out to be of central importance. The last notion concerns the need to take into account the economic valorization of research carried out by the CNRS. These three points often give rise to feelings of incompatibility, to orientations that are experienced as being irreconcilable, as leaving little room for mutual accommodation. The argument I’d like to make is that these three points constitute intellectual and practical goals that must be “held together”, even if they are in part antinomic, and that one ought to perhaps reformulate the problem in order to better situate the stakes and the contradictions. These terms are not, in fact, neutral; they often carry with them strong values (or a certain tendency to euphemism, as with the expression of valorization)—and they perhaps mask certain realities that would be interesting to uncover.

My second remark concerns the topic I was asked to address: the evolution of knowledge domains. My problem is that the idea of a knowledge domain is perhaps too restrictive for thinking about the construction of the sciences today. The expression tends to refer to systems of utterances detached from the know-how they convey and without which they have little impact; to knowledge separated from instruments, material and technical devices that alone make its existence possible; to systems of (pure) knowledge with no connection to the concrete and differentiated spaces of production. The notion also leads one to believe that bodies of knowledge are “naturally” developed for themselves (knowledge for the sake of knowledge), that they are not linked to interests and goals, that they are not formed by the social, economic and political contexts in which they’re found. To the contrary, it seems to me that over the past 150 years, and even more so in the last decades, we have been dealing with forms of techno-knowledge that vary according to their place of production (the academic world, the firm, start-ups, experts), and that adequately grasping them requires that one not forget the complex structure linking utterances of knowledge, material devices, the space in which they’re developed and economic and political aims.

The dynamic of the growth of the sciences deserves attention as well, since it consists not just of one dynamic, but of several: there are multiple dynamics that do not necessarily pass from the basic to the applied, or from the scientific to the industrial. These dynamics certainly follow theoretical or experimental inquiry, but they also follow other logics, instrumental, institutional, political—or in feed-back. I therefore suggest that it is better to think of the question in terms of institutionally and economically situated practice domains, of mixed regimes of production that define the forms of knowledge whose logic varies according to the institutions in which they’re inscribed.

2.

Let me clarify these ideas and their importance in developing a strategy, by way of a rapid historical review of the knowledge production regimes beginning in the second half of the nineteenth century. What characterized the period from 1860-1900 was, first of all, the expansion of research and teaching laboratories in universities in the areas of physics, physiology, and psychology—that is to say, a multiplication of places where learning was by doing, for both researchers and students. A second novel element: new industrial sectors (underwater telegraphy, electrotechnology, organic chemistry, the radio) established laboratories of “basic” research. A third element: institutes for standards and testing were created by national governments, industries, the military and universities. The best example of this is the PTR of Berlin, the first of its kind, where technical norms were developed on the basis of the most recent knowledge (the Cavendish laboratory occupied this role at Cambridge, under the leadership of the great Maxwell himself). The last element: the efficiency of these institutional creations strictly depended upon the politico-economic context, especially the patent policies in force in different countries (in France, for example, patents are more often issued for products, while in Germany they are issued for processes, which has led to enormous differences in the dynamics of innovation).

From 1900 to 1940 the industrial laboratory became the norm in many industrial sectors (but to a much lesser extent in France). It was in this way that the new physics of electrons, begun by J.J. Thomson, was introduced to the labs of GE and ATT (making up the Bell labs), initially to optimize electric lighting and intercontinental communications. Interdisciplinarity came upon the scene at that time as a management tool: as a concerted demand and deliberate action, this interdisciplinarity was not in fact born within the university, but in industrial research laboratories. Because there were concrete problems to solve, because it was necessary to reduce the bottlenecks in technical development, and such reduction could only result from a state-of-the-art science that was still under way, the heads of these laboratories called on and recruited individuals from complimentary professions and with complimentary skills. Bell’s research program on materials during the 1930s is the perfect example of this: since it had wagered on new types of objects (the transistor would be the best known product), chemists, physicists specialized in metals, magnetism or electronics, experts in quantum mechanics (including many Europeans), crystallographers—and engineers from many other specialties—were brought together within a common space. Each one had a wide margin of latitude within their own program, but was required to participate in formally organized exchanges—the company’s patent specialists, for example, circulated among the various groups in order to identify opportunities for future meetings.

I’d like to note here four things: (1) it was these programs that demonstrated at that time the interest of interdisciplinarity (there was no reason for it to be so intensely expressed within the university); (2) it concerned more than a simple mix of academic disciplines, rather, different professions were put into action; (3) what was basic was not considered as a separate category giving rise to the applied, but as a working logic that would bear no results unless enriched and guided by what was happening elsewhere; (4) this in turn led to a recomposition of the disciplines: if it was in the industrial space that solid-state physics was concretely established, it was only after its assimilation by the university that it became a discipline.

In the years of the Second World War and the Cold War there emerged other types of professions and combinations of disciplines (around nuclear energy, electronics and quantum optics, around the molecular approach in biology, etc.). More importantly, mathematics, along with the new computer, entered into the heart of scientific and engineering practices. It was the creation of the H-bomb that brought forth the idea of simulation, as developed in Monte Carlo methods; it was at the RAND Corporation— an organization of several hundred researchers created by the American Air Force, bringing together mathematicians, logicians, physicists, engineers, economists as well as experts in organization and the social sciences—that game theory and systems analysis gained greater currency. The places in which science was constructed were diversifying. Areas such as Silicon Valley or Route 128 emerged and military and industrial think tanks placed scientists at the center of their activities. (One mustn’t forget that at the time it was a matter of winning the Cold War—a war that was carried out by the deployment of techno-scientific tools). Scientists were thus called on to think about submarine warfare or the organization of R&D—and this put them in a position to change their practices. Academics and basic researchers were both consultants to and the beneficiaries of various structures, which led them to meet regularly with their colleagues, compare their approaches and tools, to become inter-disciplinary and inter-professional.

This explains a second characteristic element of techno-scientific practices during those years: the preoccupation, if not the obsession, with gadgets (to use the term of American physicists of that time), with deriving from all their research the instruments, technical objects, the “black boxes” later reappropriated by industry or other disciplines. The war-era work on radar gave rise to resonance techniques (this work was carried out by Félix Bloch and many others at Stanford and elsewhere); the work on molecular jets and optical pumping led to masers and lasers, especially as a result of Isidor Rabi and Townes at Columbia; and war electronics made possible various types of detectors used in the physics of cosmic rays and later in industry. This work, stimulated from the outside, became an integral part of scientific work—often resulting in the creation of new companies. In short, instrumentation and instrumental and technical developments became a constituent part of the work of physicists. It’s worth noting, once again, that it was a new form of life that emerged, another way of being in the world that took shape, another definition of self and legitimate practices that came to light—and that this emerged because the social context changed and otherwise stimulated scholars. Lastly, these new practices came to define the norms that were established internationally since they were the bearers of a new industrial, economic and political efficiency.

The landscape has changed once again over the past two or three decades. The greater performative capacity of techno-science—especially by way of molecular biology, biotechnologies, the physical technologies of communication and information—and the change in the main political and economic regulations (what’s grouped under the heading of globalization, the withdrawal of States, the rise of financial capital, etc.), along with the concomitant change in the modes of knowledge production, have led to the academic world’s loss of its central role, reduced investment in basic research on the part of large companies, drastic changes in rules concerning intellectual property, evidence of having to create start-ups, changing values in scientific communities, etc. One could say that there is a tendency to pass from a mode of production capable of balancing together two systems, one being an open and public science and the other a private science, to a mode in which the role of the former is reduced—at least in certain regions of the world such as the United States and the United Kingdom. Under the influence of the liberal economic revolution and the change in social values, the knowledge production regime inherited from the Cold War and centered on the university has found itself placed in doubt, much to the benefit of a more private production regime of techno-scientific goods. The change in patent policies and the tendency to broadly extend the patentable areas of application have been decisive here: they made up the privileged means by which the commercial university has been able to alter the previous balances. These changes bring forward important questions, questions that are decisive for all research and for the CNRS in particular.

3.

Having completed this rapid overview, I want to return to the questions that currently concern us. I’ll address them by considering the notions of the basic, of valorization and interdisciplinarity.

Regarding the basic, I’ll say two things that are partially contradictory—but this tension is a fact that must be addressed. What the brief historical overview showed above is, first, that “the basic” must not be conceived of as something given or evident, something that might serve as the origin of technologies. In a knowledge-based liberal economy, which is one way of defining our societies, “the basic” doesn’t have (anymore?) its own being and place. The world of innovation and development, in order to succeed, requires the conjunction of different logics, their spatial integration, their temporal phasing. The basic is intimately mixed with the technical, the instrumental and industrial, with management, capital risk, appropriation policies and patents, with engineering sciences—and the mentalities of the various actors.

This first remark does not imply, however, that one needn’t defend “basic” work’s right to autonomy, but just the opposite, whether it concerns physicists, biologists or philosophers. Even if it doesn’t have an intrinsic definition, it’s a way of being in the world, a way of asking questions that are worth defending. The CNRS, because of its size, can fulfill here an important function. If one considers that the liberal logic (and especially that of companies) can lead to the neglect of the long-term in favor of the short-term—like the neglect of certain global and collective interests expressed by society—the CNRS and it researchers must identify the research domains that will be abandoned by the market economy, and can decide to make them “exist”. A classic example is the study of transgene flow, regularly ignored by agricultural biotechnology companies whose goal, understandably, is to put to market as quickly as possible the GMOs they developed, but a study which many “basic” researchers have taken up again. The CNRS already does this type of thing, especially in its multidisciplinary programs, but it must make this matter one of its raisons d’être. I would say that, in the new socio-economic context, one of its basic missions must be, quite precisely, to promote a type of research of the long-term that’s critical, but that risks being neglected by economic actors who are pressed for time.

The question of valorization, which is symmetric to the preceding one, is subject to the same remarks. Placing valorization at the center of the problem is of capital importance for the institution today. The CNRS does considerable work in this area, but it could perhaps still do more. The question of the frame of mind of researchers (their desire for involvement in development, if you will) is clearly a delicate one: it is political and goes well beyond the CNRS’s sphere of influence. Yet, considering the case of the Bell labs cited above, the CNRS could put into place a policy of systematically surveying the work carried out in its research units, a policy led by professionals whose task would be to actively identify possible fertilizations, and initiate exchanges. The economic potential of their work often remains invisible to researchers, immersed in their own logic, and on-going investigations aimed at valorization, investigations entrusted to specialized personnel or to important figures responsible for bringing together the work taking place in different areas, could provide significant benefits.

However, we must also calmly consider our relationships to industry and the business world since the logic of the latter can be short-sighted, as I just noted, but especially since it’s essential for society that other points of view are regularly considered. Being aware of the importance of valorization does not imply giving without expecting anything in return, it does not imply not being respected as a full partner, it does not imply not asserting one’s rights (on the timescale of research, for example) and one’s values. In short, it is up to an active CNRS, managing its own discoveries and cross-fertilizations to decide the social and economic uses of these, and to put them to public debate if it judges it useful for the general community.

Interdisciplinarity, as such, can be approached through two questions: What are its ends, and What are its tools? Interdisciplinarity, at a first level, is inherent to all scientific work. By definition, students are trained through more or less interdisciplinary groups, and researchers are socialized in the paradigms and working methods of diverse communities. With the intervention of creativity, these frameworks regularly find themselves to be in misalignment—a new interdisciplinarity appears, seeds for new ideas pass between neighboring fields, and novel frameworks spread and take root. And so it goes repeatedly, as seen in the regular overhauls of the National Committee’s divisions. At this level of interdisciplinarity, only the professionals of the domains concerned can judge what will be the best possible arrangements. The idea of creating transversal programs (and not necessarily reorganizing the committees) is another solution, a solution that takes into account both the fact that there is not just one possible arrangement and the need to maintain several parallel structures. The same “object” can be approached from many different angles, and care must be taken to maintain this multiplicity (I could develop this in much greater detail in the case of history as a discipline, with its different temporal registers, its various scales of analysis, its different connections with other approaches, the anthropological or economic, for example; but the point is rather banal and not worth developing among this group of readers).

However, the question of interdisciplinarity as it’s addressed by different segments of the CNRS today goes beyond the knowledge industries’ “customary” reconstructions. It covers—as is the case with most interdisciplinary programs—other issues and aims. Although it’s certainly a matter of helping us do our work well, it’s also a matter of making us more sensitive to what many call the social demand. If we admit that the institution’s inclusion among the concerns of the society that maintains and finances it is both important and legitimate, then two aspects need to be considered. The first was just mentioned, that is, the active and original contribution that the CNRS can bring to economic innovation and development. Such action implies a policy of valorization and vigilance, a policy that must be actively updated regarding what’s happening within and outside its walls. The issue here is much more than a kind of interdisciplinarity as understood in the earlier sense; indeed, the notions of profession, on the one hand, and of interests on the other, lead us to frame the question in a completely different way.

The other aspect is that of public debate (on GMOs or the greenhouse effect, for example), and the various positions it can assume—which leads to a third level of social interactions involving the sciences, a level that goes well beyond the notions of interdisciplines, inter-professions and the defense of interests, whether for single participants or research companies. I’d like to note three things. First, it’s worth repeating that all studies have shown that the growing fears in society with respect to the practices of industrial technoscience are due less to a mistrust of “science” than to a worry about, or refusal of, the means of social and political regulation, on the one hand, and the exclusively reductionist, technological and technocratic manner of addressing problems on the other—with the corresponding neglect of a variety of necessary approaches. The response to these concerns is not, therefore, to find new pedagogical methods or means of popularization (it would be necessary, above all, to instruct an uninformed public); it is not a matter of bad communication.

The second point: research domains having to do with studies on the environment and its management—the introduction of genetically modified organisms, laboratory- assisted human reproduction, impact measures of technology, nuclear energy and waste, planning for health crises, planetary equilibria (ozone holes, climatic changes, etc)—imply radically new forms of exchange that must be invented. It is not, in fact, a matter of making those who produce knowledge, techniques or industrial devices “collaborate” around the same question (in order to reconfigure a domain or to innovate, for example); rather, it’s a matter of intervening in debates that potentially involve the society as a whole, that are distinguished by an infinite variety of sensibilities and values, and whose outcome is, for each society, the choice of its future and the implementation of preventive actions.

In this task of defining the social issues for itself, all the facets of scientific work can be put to action. The sciences are among the primary actors (they set things in motion by contributing to the deployment of technical systems), and also among those actors located at the end of the cycle (science is asked to help regulate those things to whose change it had previously contributed). Yet, unlike what one sees in questions of interdisciplinarity, the problems here are completely framed from outside (for example, due to an unexpected health crisis), they unfold in timeframes defined by social and political needs (and not by those of research or innovation)—and they take on greater importance according to radically diverse criteria and must be accepted as such (conflict is inherent to politics). The intrinsic complexity of questions, like the inability to control work schedules, presents the sciences here with problems of which they have only imperfect knowledge. Transdisciplinarity is thus no more than one set of questions, and not necessarily the most critical, that are broader than the ones concerning the choices that society wants to make for its future (does it want a productivist agriculture or not?), such as the forms that the political or social debate takes or must take on. Experts and scientists certainly play an important role in these debates (think of the question of climate controls, unimaginable without modeling), but the stakes are such that they cannot be the main contributors nor the only judges. It is even likely that their research programs, including the basic questions that they define as those of their disciplines, will be redefined due to their involvement in these dialogical exchanges. And I think it is good to be aware of this, to expect it, and to want it.

The tools of interdisciplinarity will now, finally, be noted. They’ll obviously vary according to the meaning one gives to the expression “interdisciplinarity”, and each person can easily modify them. I will only draw attention to two general aspects, since the first is already suggested by the post-war situation, but also because it can be seen today and read in division and department reports —namely, that the development and transfer of instruments, techniques, and materials is the decisive tool of interdisciplinarity. By extension, one must also include collection techniques, the management and processing of data, software, calculation and modeling techniques, information sharing—all of them objects and practices that ought to be valorized. Perhaps certain departments and divisions that still live in relative isolation, due to their objects of study, could be decisive here (Nuclear and Particle Physics, for example, might come to mind). As the researchers from this department note, their strength is in the tools they develop.

The second remark concerns the place of the human sciences in the definition of the CNRS’s programs. If one agrees that the public debate is a space to take seriously, as much in order to analyze it as to assure that it unfolds in the best possible manner, one will note, first of all, that the human sciences are not without importance (this concerns life, ethics, law, the body, the economy, agriculture, the food industry, sustainable development, the precautionary approach, “governmentality”, etc.) I note that they are present in all of the interdisciplinary programs of the CNRS, yet I wonder if everyone really believes in their importance. One often has the impression that it’s still an “expensive hobby”, so to speak, something that cannot be central since there is nothing to be learned from this exchange. I therefore stress that if one has understood that these questions are by now, and for a long time to come, our horizon of thought and work, and that social issues can no longer be dealt with only in a “top down” fashion, then it’s better that the ear they lend to one another be profoundly attentive. It must be so between the scientists and “laymen” who reconstruct the social bond on a daily basis; it must be so as well—and wouldn’t it be a good start?—between the hard sciences, the human and the social sciences.

[1] The National Committee of the CNRS, an authority that is in part elected and in part nominated by the government, is divided into more or less disciplinary divisions. The National Committee plays an essential role in recruiting researchers and it periodically publishes situation reports.