Transdisciplinarity, a term originally chosen in the context of the philosophy of science and the organization of the sciences in order to get away from a superficial notion of interdisciplinarity, has become a fashionable term in contemporary reflections on forms of scientific research and organization. There are conferences on the concept of transdisciplinarity - as, for instance, at the beginning of the year 2000 in Zurich (International Transdisciplinarity Conference: Joint Problem Solving among Science, Technology and Society) - and the literature on this concept is becoming overwhelming. As a result, there is a risk that reflections on transdisciplinarity in relation to the scientific practice they are intended to promote become a matter of course. Thus it is worth reminding oneself of the origins of this concept, thereby reconnecting the methodological discussion with scientific and academic practice.
I introduced the concept of transdisciplinarity in 1986, at a symposium on the ideology and practice of interdisciplinarity at the Bielefeld Center for Interdisciplinary Research (ZiF), and again in 1987 at a symposium on the disciplinary system of the sciences and the task of science politics at the University of Constance.1 I then characterized it as a new principle of research, which was suggested by the development of the sciences itself. But this characterization had no effect on the usual rhetoric of interdisciplinarity. It seemed only a new word for an old thing, namely that which at the time (and subsequently as well, when "Ringvorlesungen" and the Studium Generale were gaining in importance in the day-to-day of the university) was adequately addressed by the term "interdisciplinarity", that is, by acknowledging the need to complement disciplinarity with a dialogue between the disciplines.
As I said, this has all changed today. Suddenly transdisciplinarity is on everyones lips, and not just there where the philosophy of science and university pedagogy reflect upon the disciplines (a reflection which is often quite far removed from academic reality), but also in actual research practice. It sometimes seems as though a new research paradigm were announcing itself, namely, a methodological orientation that might lead beyond the established fields and disciplines. What does this development mean? What is actually new, and what is merely clothed in new terminology? I want to address these questions in the following, in which I draw on some older reflections of mine, as well as on a lecture on the relationship between philosophy and the natural sciences which I held in 1999 at the spring meeting on solid-state physics of the German Physical Society in Munster.2 I will begin with a few words on the notions of disciplinarity and interdisciplinarity.
In the course of a long institutional route, our academic system has become disturbingly unfathomable.3 This is the case not only with regard to the ever accelerating growth of knowledge in all scientific fields, but also with regard to the organizational and institutional forms of academic research. There is an increasing particularization of disciplines and fields; whereas the capacity to think disciplinarily - that is, in terms of larger theoretical units - is decreasing.
Thus it is no surprise that there has been much talk about the desirability of interdisciplinarity, and this for some time now. Sitting alone on ones disciplinary island, one is likely to be drawn to ones mates on neighboring islands, whereby it is perhaps not so important who these disciplinary neighbors are. The borders between fields and disciplines, to the extent that they are still observed at all, threaten to become less institutional borders than cognitive ones. And thus the concept of interdisciplinarity comes to include the notion of an improvement, which should lead in time to a new scientific and academic order. Interdisciplinarity is in consequence neither something normal, nor indeed something really new, nor simply the scientific order itself. When it succeeds, it corrects defective academic and theoretical developments, thereby making clear that we have lost the capacity to think in larger disciplinary units. A coherent whole should be regenerated out of particularities, and we should thereby regain something that was the academic norm in the history of European academic institutions before the "discovery" of interdisciplinarity. Nevertheless, it is not this "institutional" perspective, that is to say the re-establishing of real disciplinarities, that should be in the foreground here, but instead the role of structures and strategies in research extending beyond fields and disciplines (and thus indirectly in teaching as well).
Here one should first make clear that these fields and disciplines came into being in the course of the history of the sciences, and that their borders are founded primarily neither in objects nor in theory, but that they are historical as well. At the same time, their historical identities are shaped by definite research objects, theories, methods, and goals, which often do not comprise a coherent disciplinary definition, but in fact interfere interdisciplinarily. This is expressed not only in the fact that disciplines are governed in their work by methodological and theoretical concepts, which cannot themselves be generated within each discipline, but also in the fact that the problems addressed by academic disciplines often cannot be enclosed within a single disciplinary frame. Thus in the history of the theoretical description of Heat, for instance, disciplinary responsibility often changed. At first, Heat was considered as an internal motion of matter and thus as an object of physics. It became an object of chemistry, however, in the light of the caloric theory formulated by Boerhaave at the beginning of the eighteenth century and later developed by Lavoisier, since it was then considered to be a kind of matter. Finally, Heat changed its disciplinary allegiance yet again with the kinetic theory, and once more became an object of physics. This shows that it is not the objects (alone) which define a discipline, but the manner in which one deals with them theoretically. This is often clear enough in the context of research, but not necessarily in teaching.
This example from the history of science can be generalized so as to show that there are certain problems that escape the confines of a single discipline. Far from being marginal ones, these are often central problems, as for example the environment, energy, and health. There is an asymmetry between the development of problems and that of disciplines, and this asymmetry is accentuated by the fact that the development of fields and disciplines is determined by increasing specialization. Ecological problems are complex, and they may be solved only through the cooperation of many disciplinary competencies. The same is true of energy and health. But this means that the term interdisciplinarity is concerned not merely with a fashionable ritual, but with forces that ensue from the development of the problems themselves. And if these problems refuse us the favor of posing themselves in terms of fields or disciplines, they will demand of us efforts going as a rule well beyond the latter. In other words, whether one understands interdisciplinarity in the sense of re-establishing a larger disciplinary orientation, or as a factual increase of cognitive interest within or beyond given fields or disciplines, one thing stands out: interdisciplinarity properly understood does not commute between fields and disciplines, and it does not hover above them like an absolute spirit. Instead, it removes disciplinary impasses where these block the development of problems and the corresponding responses of research. Interdisciplinarity is in fact transdisciplinarity.
While scientific cooperation means in general a readiness to cooperation in research, and thus interdisciplinarity in this sense means a concrete cooperation for some definite period, transdisciplinarity means that such cooperation results in a lasting and systematic order that alters the disciplinary order itself.4 Thus transdisciplinarity represents both a form of scientific research and one of scientific work. Here it is a question of solving problems external to science, for example the problems just mentioned concerning the environment, energy, or health, as well a principle that is internal to the sciences, which concerns the order of scientific knowledge and scientific research itself. In both cases, transdisciplinarity is a research and scientific principle, which is most effective where a merely disciplinary, or field-specific definition of problematic situations and solutions is impossible.
This characterization of transdisciplinarity points neither to a new (scientific and/or philosophical) holism, nor to a transcendence of the scientific system. Conceiving of transdisciplinarity as a new form of holism would mean that one was concerned here with a scientific principle, that is to say a scientific orientation, in which problems could be solved in their entirety. In fact, transdisciplinarity should allow us to solve problems that could not be solved by isolated efforts; however, this does not entail the hope or intent of solving such problems once and for all. The instrument itself - and as a principle of research, transdisciplinarity is certainly to be understood instrumentally - cannot say how much it is capable of, and those who construct and employ it also cannot say so in advance. On the other hand, the claim that transdisciplinarity implies a transcendence of the scientific system, and is therefore actually a trans-scientific principle, would mean that transdisciplinarity was itself unbounded, or that it was bounded by arbitrary terms which were themselves beyond scientific determination. Put otherwise: transdisciplinarity is - and remains deliberately - a science-theoretical concept which describes particular forms of scientific cooperation and problem-solving, as opposed to forms lying outside of scientific boundaries. For what could be the point of looking to trans-scientific considerations, i.e. at relations lying outside the scope and responsibility of the sciences, to find an organizing principle for the latter?
Furthermore, pure forms of transdisciplinarity are as rare as pure forms of disciplinarity. For the latter are most often realized and understood in the context of neighboring scientific forms, for instance in the sociological components of a historians work, or the chemical components of a biologists. To this extent, disciplinarity and transdisciplinarity are also principles governing research, or ideal forms of scientific work, hybrids of their normal forms. What is important is only that science and academic research are conscious of this, and that productive research not be bounded by out-dated restrictions (which are mostly a product of routine) confining it to given fields or disciplines. Such a confinement serves neither scientific progress, nor the world which, in reflecting on its own problems, seeks less to admire science than to use it.
In other words, transdisciplinarity is first of all an integrating, although not a holistic concept. It resolves isolation on a higher methodological plane, but it does not attempt to construct a "unified" interpretative or explanatory matrix. Second, transdisciplinarity removes impasses within the historical constitution of fields and disciplines, when and where the latter have either forgotten their historical memory, or lost their problem-solving power because of excessive speculation. For just these reasons, transdisciplinarity cannot replace the fields and disciplines. Third, transdisciplinarity is a principle of scientific work and organization that reaches out beyond individual fields and disciplines for solutions, but it is no trans-scientific principle. The view of transdisciplinarity is a scientific view, and it is directed towards a world that, in being ever more a product of the scientific and technical imagination, has a scientific and technical essence. Last of all, transdisciplinarity is above all a research principle, when considered properly against the background I have outlined concerning the forms of research and representation in the sciences, and only secondarily, if at all, a theoretical principle, in the case that theories also follow transdisciplinary research forms.
What may seem quite abstract here has long found concrete forms in academic and scientific practice. Indeed it is being increasingly encouraged institutionally, for instance in the case of new research centers which are being founded in the USA, in Berkeley, Chicago, Harvard, Princeton and Stanford,5 where a lot of money is in play. The "Center for Imaging and Mesoscale Structures" under construction in Harvard calls for a budget of thirty million dollars for a building of four thousand five hundred square meters. Here scientists will be investigating questions it would be senseless to ascribe to a single field or discipline. The focus is on structures of a particular order of magnitude, and not on objects of a given discipline. And there are other institutional forms possible, which are not necessarily housed in a single building, for instance the "Center for Nano-Science (CeNS)" at the University of Munich.
Such centers are no longer organized along the traditional lines of physical, chemical, biological, and other such institutes and faculties, but from a transdisciplinary point of view, which in this case is following actual scientific development. This is even the case where individual problems, as opposed to wide-scope programs, are the focus, as for example in the case of the "Bio-X"-Center in Stanford6 or the "Center for Genomics and Proteomics" in Harvard7 . Here, biologists are using mature physical and chemical methods to determine the structure of biologically important macro-molecules. Physicists like the Nobel prizewinner Michael Chu, one of the initiators of the "Bio-X" program, are working with biological objects which can be manipulated with the most modern physical techniques.8 Disciplinary competence therefore remains the essential precondition for transdisciplinarily defined tasks, but it alone does not suffice to deal successfully with research tasks which grow beyond the classical fields and disciplines. This will lead to new organizational forms beyond those of the centers just mentioned, in which the boundaries between fields and disciplines will grow faint.
Naturally this holds not just for university research, but for all forms of institutionalized science. In Germany at the moment, these present a very diverse picture, which ranges from university research, defined by the unity of research and teaching, to the Max Planck Societys research, defined by path-breaking research profiles in new scientific developments, to large-scale research, defined by large research tools and temporally constrained research and development tasks (that were earlier quite openly announced as lying in the national interest), to the Fraunhofer Institutes research, defined by economically relevant application, to industrial research, defined through its tight connection between research and development.
But the logic of this system, which bears witness not only to scientific reason, but also to extraordinary efficiency, is becoming problematic. For it leads to the autarchy of the component systems, whereas in fact - as in the case of the new centers I just mentioned - the emphasis should be on networking at the lowest institutional level, and not on the expansion of independent systemic units on the high institutional plane. For Germany, this means that temporary institutionalized research cooperatives should take the place of component systems which are increasingly opposed and isolated. And this can be easily justified from the point of view of the sciences: The scientific system must change, when research changes. At the moment, the situation in Germany is rather the reverse: It is not research that is searching for its order, but rather an increasingly rigid order which is already laid out in component systems that is searching for its research. And in such a case, the scientific order becomes counterproductive. This cannot be the future of research, or of a scientific system like that of Germany. As we have seen, the increasing transdisciplinarity of scientific research has wide-ranging institutional consequences - at least it ought to have them.
One dominant scientific development that leads to transdisciplinarity seems to revive the concept of the unity of nature, first as the concept of a unified physical theory - if there is only one nature, then all natural laws must be part of a unified theory of nature9 - and then in the form of increasingly transdisciplinarily ordered scientific research. If nature does not distinguish between physics, chemistry, and biology, why should those sciences that investigate nature do so in an unshakeably disciplinary manner? Set against the background of the transdisciplinary orientation of modern research programs, the original concept of the unity of nature does indeed seem to regain its reality. But our subject is not this concept, but rather the claim that transdisciplinarity is not merely a philosophical fantasy, but a part, indeed an essential part, of newer scientific research. Let me give you three examples of this.
The idea of investigating and creating functional structures in the laboratory on an order of magnitude between 10- 9 and 10- 6 meters, that is, individual atoms, molecules, and small groups of atoms, goes back to a visionary lecture of Richard P. Feynman, in which the latter discussed the possibility of storing and reading information in very small spaces, thereby anticipating a number of lithographic methods which are now in use.10 On his own account, Feynman was inspired by biology, in which such small and highly functional units are in fact at work. Why shouldnt it be possible to make them artificially?
Today, physicists and chemists work hand in hand manufacturing such nanostructures. Whereas physicists as a rule begin with given structures, for instance a surface structure, and then modify them with physical methods, chemists begin at the level of atoms and molecules, and then combine these systematically. All domains of nanotechnological research are closely networked, such that progress in one area generally leads to progress in other ones. Among the most important developments in nanotechnology are the synthesis of carbon rings (fullerenes), the production of microscopic tubes from carbon atoms11 and the artificial linking of small numbers of carbon atoms12 .
There are questions and research directions whose results can hardly be uniquely assigned to physics or the philosophy of science. Among these is the quantum mechanical measurement process. How can it be that measurements of a quantum mechanical system still lead to a definite and unique result, even when the measurable state was prepared to be a superposition of eigenstates of the measurable observables? Does the wave-function collapse in the act of measurement instantaneously into one of the eigenstates contained in the superposition (as supporters of the Copenhagen interpretation suppose)? Or do we only perceive a subset of the actual wave-function after the measurement (as, for instance, the many-worlds and the many-minds interpretations suggest)? Or is the measurement process a "real" process that takes place on an extremely small time-scale, whose genuine non-linear stochastic dynamics goes well beyond the fundamental assumptions of quantum mechanics, and strictly speaking contradicts them?13 Further questions might be directed at the unification of quantum mechanics, the theory of special relativity,14 and the role of non-local interactions in physics. Philosophers, practiced in the art of making fine distinctions, and in dealing with concepts requiring clarification (for instance, non-locality), can prove to be useful partners of physics. It goes without saying, that philosophy does not seek to answer questions that science, in this case physics, is better able to answer.
Furthermore, computer science is also of use here. If one follows the Copenhagen interpretation, a quantum system loses information when it is measured. The reason for this is that, before the measurement, the system is in a superposition-state, whereas after the measurement, it takes, qua projection, an eigenstate of the operator which is correlated with the observable. The remaining information of the state is thereby lost. Once the concept of information is introduced into quantum mechanics, the theory of information can be drawn upon for a further analysis of the measurement process, whereby a link is established to possible technical applications (for instance "quantum cryptography" and "quantum computers"15 ). The research principle of transdisciplinarity does not concern only the cooperation of distinct disciplinary competencies, but indeed reaches to technology.
The expression derives from the neurophysiologist John Eccles and the philosopher of science Karl Popper. It is the title of a collaborative book,16 and refers to a view which derives from the "dualistic" independence of psychical and physical states or processes, and which furthermore maintains the independence and identity of the self (or of the consciousness of self) in contrast to that of its physical representation. According to this view, which has its origin in Descartes, the brain belongs to the self, not the self to the brain. The self is the programmer of the computer "brain" - the pilot, not the piloted. Eccles translates this idea into neurobiological language. Here the self controls and interprets the neuronal processes; it actively seeks brain events which lie in its domain of interest and integrates them into a unified and conscious experience. It constantly scans collective interactions of large numbers of neurons which are open to an interaction with the world of mental states and events ("liaison brain"). The unity of conscious experience, according to Eccles central thesis, "is provided by the self-conscious mind and not by the neural machinery of the liaison areas of the cerebral hemisphere".17
As is well known, there are monistic conceptions which are opposed to such a view, which indeed appears a little speculative, for instance physical reductionism such as the identity-theory. In this context, all attempts to view mental processes as directed from "beyond", that is to view them as detached from the substance of the brain, are seen as a recurrence of philosophical naïvety. Such monistic alternatives can be extended by information-theoretical considerations, in which psychic and mental processes can be regarded as complex data-transformations, just as Mind and Body, Matter and Consciousness can be regarded as different structures of data-states. These approaches are also monistic and physico-materialistic in their structure. The brain is viewed as a "calculating device".
Both here and in the above-mentioned monistic conception, it is clear that problems that were originally purely philosophical have to a great extent become interconnected with research in neurophysiology and neuropsychology by means of empirical connections and dependencies between physical and psychical states and processes. Even in the case of the mind-body problem, the future of philosophy would appear to lie in the sciences, that is, in the scientific refinement of its intuitions. Far less controversial, although, as the example of Popper and Eccles shows, also far from uncontradicted, would be a so-called pragmatic dualism, on which view there is no reason to suggest a factual or absolute difference between mental and physical states and processes. Instead, it suggests that we have no grounds for dismissing such a distinction from the outset, and that it may well be better to understand cognitive concepts as explanatory concepts.18 Put otherwise: If everything that we know about the world is at least in part a construction of the brain, it follows that what biology knows about the brain cannot itself be independent of such constructions. Philosophical reflections are in this case also not external to scientific research, but a part of a research program which is transdisciplinarily oriented even in its more strictly natural-scientific parts.
The natural sciences are not only taking more and more transdisciplinary routes, routes which can also sometimes be philosophical ones, but they are also occasionally quite philosophical, since many of their theories have the character of interpretations. By this I mean that there are often several ways of expressing a theory, so that it is often not scientific considerations in the narrow sense that lead to the selection of one of these possibilities, and furthermore that the interpretations that attach to a theory are often not unambiguous. Let me take another example from physics.
There are many ways of interpreting quantum mechanics. Thus the Copenhagen interpretation characterizes the quantum world as a very peculiar one, in which particles no longer travel along paths, as they did in classical physics. In addition, the principle of causality is violated, such that a deterministic description of the world seems no longer possible in principle. But in a "Bohm-world", as one might call the world described by Bohms theory,19 things look differently than they do in such a "Heisenberg-world". Aside from an additional non-local force that does not appear in classical physics, and to which all observations that appear strange from the standpoint of classical physics might be reduced, this world is barely distinguished from the familiar world of Newton. Both interpretations - the Heisenberg-world and the Bohm-world - are accordingly completely distinct conceptually, but are nevertheless empirically equivalent formulations of the same theory. Thus one cannot decide on the basis of observations alone for the one or the other variant of quantum mechanics.20 In other words, scientific theories are often equivocal in the sense that distinct theoretical approaches can often be adequate to the same empirical data and that the same theoretical approach can be interpreted in different ways. The interpretation of quantum mechanics is in this regard not essentially different from an interpretation of the Kantian theory of space and time, which counts as a very philosophical theory indeed.
And this means that a certain inherent perspectivity of knowledge, that is to say the dependence of knowledge on its theoretical form, is always with us, even in the case of science. For knowledge can appear clothed in diverse forms of representation (diverse theoretical approaches which explain the same things), and a theoretical approach can itself be variously interpreted. The world to which we refer with our theories is in this regard a "Leibniz-world": Knowledge does not refer to an "absolute" world, whose essence remains cognizable even in the absence of our theories, and the world is not exhaustively circumscribed by some "absolute" knowledge.
Does that not mean that we are confronted by a fundamental paradigm change, in which it is not so much theoretical conceptions - as for instance the transition from Aristotelian to Newtonian physics or that from Ptolemaic to Keplerian astronomy - which are changing, but rather the order of our scientific knowledge, and thus also our scientific research and training, which is in the balance? I do not think that it will come to that, and this for the reasons I mentioned in my analysis of the concept of transdisciplinarity. For it is not the standards of rationality, nor with them the methods and forms of theoretical construction which are changing, but the organizational forms of science and scientific research. Once again: Transdisciplinarity is a scientific research principle that is active wherever a definition of problems and their solutions is not possible within a given field or discipline. It is not a theoretical principle that might change our textbooks. Just like an orientation towards fields or disciplines, transdisciplinarity is a principle that guides research and a form of scientific organization, albeit in such a manner that it removes the disciplinary impasses which are more the product of institutional habit than of scientific necessity.
That such impasses can also determine the logic of a scientific system, in so far as such a system is made up of component systems which follow their own institutional logic, has also been emphasized. If research takes on increasingly transdisciplinary forms, then temporary research cooperatives are the appropriate organizational form, and not isolated component systems. So once again: The scientific system must change, when research changes. It looks as though the research system still has to learn this. Transdisciplinarity would in this sense be the gadfly of the scientific order.
1 "Die Stunde der Interdisziplinarität?", in: J. Kocka (Ed.), Interdisziplinarität. Praxis - Herausforderung - Ideologie, Frankfurt/Main 1987, pp. 152-158; "Wohin geht die Wissenschaft? Über Disziplinarität, Transdisziplinarität und das Wissen in einer Leibniz-Welt", Konstanzer Blätter für Hochschulfragen 26 (1989), No. 1-2 (98-99), pp. 97-115, also in: J. Mittelstraß, Der Flug der Eule. Von der Vernunft der Wissenschaft und der Aufgabe der Philosophie, Frankfurt/Main 1989, pp. 60-88.
2 J. Mittelstraß, Zwischen Naturwissenschaft und Philosophie. Versuch einer Neuvermessung des wissenschaftlichen Geistes, Konstanz 2000 (Konstanzer Universitätsreden 205).
3 For the following, cf. J. Mittelstraß, "Interdisziplinarität oder Transdisziplinarität?", in: L. Hieber (Ed.), Utopie Wissenschaft. Ein Symposium an der Universität Hannover über die Chancen des Wissenschaftsbetriebs der Zukunft (21./22. November 1991), Munich and Vienna 1993, pp. 17-31, also in: J. Mittelstraß, Die Häuser des Wissens. Wissenschaftstheoretische Studien, Frankfurt/Main 1998, pp. 29-48.
4 The following discussion of the concept of transdisciplinarity is based on an earlier short characterization in: J. Mittelstraß, "Ein Prinzip faßt Fuß", in: GAIA. Ecological Perspectives in Science, Humanities, and Economics 7 (1998), No. 1, pp. 1-2 (Editorial).
5 Cf. L. Garwin, "US Universities Create Bridges between Physics and Biology", Nature 397, January 7, 1999, p. 3.
6 Cf. L. Garwin, ibid.
7 Cf. D. Malakoff, "Genomic, Nanotech Centers Open: $ 200 Million Push by Harvard", Science 283, January 29, 1999, pp. 610-611.
8 Cf. L. Garwin, ibid.
9 Cf. C.F. v. Weizsäcker, Die Einheit der Natur. Studien, Munich 1971.
10 R. Feynman, "Theres Plenty of Room at the Bottom", Engineering and Science 23 (1960), pp. 22-36.
11 Cf. P.M. Ajayan and T.W. Ebbesen, "Nanometre-Size Tubes of Carbon", Reports on Progress in Physics 60 (1997), pp. 1025-1062.
12 Cf. R.A. Broglia, "Wires of Seven Atoms - Feynmans Very, Very Small World", Contemporary Physics 39 (1998), pp. 371-376.
13 Cf. G.C. Ghirardi, A. Rimini, and T. Weber, "Unified Dynamics for Microscopic and Macroscopic Systems", Physical Review D 34 (1986), pp. 470-491.
14 Cf. T. Maudlin, Quantum Non-Locality and Relativity. Metaphysical Intimations of Modern Physics, Oxford 1994 (Aristotelian Society Series 13).
15 Cf. H. Weinfurter and A. Zeilinger, "Informationsübertragung und Informationsverarbeitung in der Quantenwelt", Physikalische Blätter 52 (1996), No. 3, pp. 219-224.
16 K.R. Popper and J.C. Eccles, The Self and Its Brain, Berlin, etc. 1977.
17 Op. cit., p. 362.
18 M. Carrier and J. Mittelstraß, Geist, Gehirn, Verhalten. Das Leib-Seele-Problem und die Philosophie der Psychologie, Berlin and New York 1989 (English, revised and expanded: Mind, Brain, Behavior. The Mind-Body Problem and the Philosophy of Psychology, Berlin and New York 1991).
19 D. Bohm, "A Suggested Interpretation of the Quantum Theory in Terms of Hidden Variables", parts I-II, Physical Review 85 (1952), pp. 166-179, pp. 180-193.
20 Cf. J. T. Cushing, Quantum Mechanics. Historical Contingency and the Copenhagen Hegemony, Chicago and London 1994.