Biology and the unity of science

by Claus Emmeche

This comment was printet in Sats - Nordic Journal of Philosophy, vol. 2(No.1): p. 153-162 (2001).
Jan Faye's recent work, Athenes Kammer: En filosofisk indføring i videnskabernes enhed, is a clear and engaging book, written in Danish and intended to be a philosophical introduction to the unity of the sciences -- as its subtitle indicates. In addition to the arguments for unity of science, the book contains an interesting exposition of Faye's views on classical themes in philosophy of science, such as the nature of theory, models, laws, explanation, realism and antirealism. Only a few of the themes will be addressed in this note, which will focus on the unity of science question mainly from the perspective of philosophy of biology.

Faye draws upon a gallery of established and, within this corner of academia, well-known professional philosophers; he calls the book "a situational report and a contribution to the debate" (p. 7). But who is this report written for? I guess some philosophers, and those few scientists who may pay attention, will think that a debate about the unity of science, declared by Faye to be a central topic, is somewhat outdated or only relevant in the context of a history of analytical philosophy. This is sad, I think, because the question is a good point of departure to reflect upon basic values of all sciences that academics should be ready to defend, especially in periods like the present, when science in the sense of free, critical inquiry is under pressure from many sides. On the surface the book is written for a well-informed philosophically interested lay audience, but it is too essayist to be suitable as an introduction to the subject. The arguments are often condensed and presuppose some previous knowledge of the philosophical positions mentioned. One might get the idea that the book is written under the cover of addressing a more general public, while the real target is local colleagues who have a professional interest in the questions raised. But this cannot explain why the work is so careless about relating to discussions about unity of science which have been conducted in a Scandinavian or Danish context a decade ago, and where one can find interesting viewpoints, also from other fields than philosophy, e.g., Fink & Hastrup 1990 (in which Hans Fink developed a strong ontological non-emergentist and non-reductionist argument for unity, not addressed by Faye); Bak & Dal 1990; Collin 1990; Køppe 1990.

The book is worth reading for anyone interested in philosophy of science. Yet, I must confess I was a little disappointed after having read the book, though Faye may not be the one to blame. I did not find anything with particular relevance to my interest in biology and the interfield between theoretical biology and philosophy of biology - no crucial points with implications for the question about the role of biology in the unity of sciences, or about in what specific sense also biology as a certain cluster of research areas are united to (or in union with) the other sciences, except for the general thesis of Faye's that unity comes through methodology (the scientific method), and this must pass for biology as well. This complaint may sound idiosyncratic, but I think it could have been asserted from the point of view of the philosophy of the other special sciences as well. An important insight, however, can be gained by reflecting upon this fact, namely, about various ways of doing philosophy of science. As in some of Faye's previously published works (e.g., Faye 1993, 1998) which the book summarises, the focus is 'science' at a very high level of generality and the material is approached with some aprioristic ways of thinking, searching for general conditions that must hold for something, such as scientific knowledge in general, to be possible at all. I consider Faye's approach a legitimate one -- important studies can be made about general constituents of science (as explored in the book's discussion of cognitive purposes of science, facts, explanations, ways of reasoning, interpretations, and so forth) - but it is still such a distant sort of enterprise that insight into the particular ways of doing scientific work within individual sciences cannot be revealed from that distance (for instance, Faye's own previous work on Niels Bohr's philosophy demanded more concrete and detailed studies of the peculiarities of quantum physics). We can roughly characterise Faye's approach in Athenes Kammer as standing in an old tradition of general philosophy of science by comparing that approach with three other ones according to this somewhat simplistic scheme:


  • General Philosophy of Science (GPoS)

  • General Philosophy of (a particular) Science-n (GPoS-n)

  • History and Philosophy of (a particular) Science-n (HP-n)

  • Studies of Philosophy in Science (SPiS)

I will briefly characterize these approaches and explain to what extent Faye's approach is of the first kind.

One cannot accuse Faye of being a positivist, but to the extent that his approach can be characterized as GPoS, he stands in the tradition of logical positivism in his way of approaching the unity question -- even though the positivists would have unity based on ontological arguments within a physicalist ontology, or upon epistemic arguments via theory reduction, while Faye's claim is an argument for methodological unity via the generality of a scientific method. Pace all the highly different forms of this method (that Faye denotes 'practical-instrumental methods', p. 86), basically, in Faye's perspective, scientific method reduces to forms of inference ("induction, abduction, and inference to the best explanation", p. 86). This rather logocentric view of method may appear a little strange when seen in light of the current interest within philosophy of science and science studies in all the material and pragmatic aspects of method, such as diverse forms of scientific experiment and the whole technical culture necessary to do experimental work in at least some corners of natural science, such as high energy physics (Galison 1997) or molecular biology (two fields with highly different epistemic cultures, cf. Knorr-Cetina 1999). And in its basic metaphysical orientation, this approach to method as abstract forms of reasoning is in strong contrast to what has been called 'the naturalistic turn' in philosophy of science (e.g., Callebaut 1993).

Now, closer to the reality of science, but as a continuation of GPoS, we can have "General Philosophy of Science applied to science-number-n" (GPoS-n) in which a philosopher applies the battery of concepts from GPoS to a particular field, for instance, biology or sociology. Here, one should not be seduced into believing that a certain body of methods and theories from a GPoS (whatever that would be) can be applied equally smoothly to all kinds of science (and one may indeed find good arguments in Faye's text against that view). There is presently among philosophers and historians of science some consensus that GPoS (the dominant approach until at least the mid-1960s) took physics, often understood as an axiomatic system of theories, to be the prototype of a science, what was expected to hold generally for all science was very often characteristic peculiar to physics. Thus the very idea of an "application" of general principles to the study of particular sciences often proved to be problematic due the fact that what GPoS took to be general principles were not so general indeed, and as soon as philosophical investigations of other fields than physics entered the scene, it became apparent that there were problems with the received view. An example of GPoS-n is the very early phase of philosophy of biology. Philosophy of biology is still a relatively new field (with important forerunners, such as Woodger 1929) but was at its inception an odd mixture of a protest against the dominance of physics as a prototype of science within philosophical circles and a continuation of the basic themes of what Putnam (1962) dubbed "the received view" within GPoS of, for instance, theories as syntactic structures, a view that can be found in some early works (Ruse 1973, compare Hull 1974) of the field (for a criticism, see Thompson 1989). The received view had its roots in logical positivism, and again, this view (of, for instance, theory structure) is not Faye's, but the implication of his own approach is, so far I can tell, a restatement of the tenet that it is possible (a) to achieve a universal philosophical theory of science -- namely science as being a general method (in the abstract sense as a logic of discovery and justification) -- apart from particular realizations of that method (within literary theory, geology or sociology), and then (b) to apply that theory in particular studies of these realizations (particular instances of GPoS-n).

This tenet, the unified science idea and the GPoS-n approach to philosophical investigations of particular language games of science came under attack in the period following the mid-1960s, not so much as a result of the general decline of analytic philosophy, or Wittgenstein's late philosophy, but rather as a result of the historicist turn in philosophy of science, inaugurated by Thomas Kuhn's Structure of 1962. This started an intellectual movement within science studies in which scholars from history of science cooperated with philosophers, and philosophers began to take the actual, complicated detailed historical records from the annals of individual sciences much more into consideration, somewhat de-emphasizing the pursuit of generality and focusing on case studies with a more special interest for the specificity of particular disciplines. The lessons of this History and Philosophy of Science-n (HP-n) movement are visible, of course, also in Faye's exposition, especially in chapter 8 surveying theories and models, but Faye is in strong opposition to a dominant view within the HP-n tradition, namely the view that it no longer makes sense to talk about a scientific method as a specific phenomenon. In a recent discussion about this topic on an electronic philosophy list (HOPOS-L: A Forum for Discussion of the History of the Philosophy of Science), Greg Ranson expressed the view that "a large and clear majority of philosophers of science today would reject the idea that there is any single fixed thing called "scientific method"", and Richard Burian added that he "share[s] with many colleagues the view that useful methods are, by and large, context specific and exhibit limitations that show up when they are stretched too far. Attempts to [...] expand methods to apply generally make them so abstract as to become rules of thumb -- or, if not, into poor tools in many particular contexts. The problems of "one size fits all" methods remind me of the awkwardness of "one size fits all" garments." (HOPOS-L Digest, #2001-12, 17 Jan. 2001). Another writer raised the sentiment that "any philosopher who TODAY insists on there being A scientific method has been asleep for the last twenty years!" (Joseph C. Pitt, ibid., 18. Jan.).

So, has Faye been asleep? No, because he certainly acknowledges the diversity of methods in the natural and social sciences and the humanities. He operates with a distinction, which might have been developed more in detail, between the aforementioned practical-instrumental methods (which are descriptively diverse) and a normative idea of a set of methodological prescripts common to all sciences (p. 30), a set he later on (as I read him) reduces to rules of inference, but which could have been characterized more inclusively and in continuity with the 'practical' or pragmatic level of science as a normative activity of the community of researchers, very much in the tradition of the philosophy of C. S. Peirce (who is not mentioned in the book). Still, the problem with pointing to logic as distinctive for the unity of science is that logic, and logical inferences, are also present in many other forms of human intellectual activity. Thus, behind the unity of science thesis lures the problem of demarcating science from other intellectual activities, and the philosophical problems with a clearcut demarcation. This problem, including considerations of practical or theoretical motivations for a demarcation in particular situations, is not sufficiently addressed by Faye. The motivation for a demarcation is mainly found implicitly in the Platonic picture of science that Faye's book depicts, and which is revealed in the description of science like the goddess Athena, threatened by the evil social constructivists who cannot see the flashes of lightning She hides in her Olympic chambers (p. 187).

It is nevertheless refreshing that Faye admits the normative dimension in his description of methodological prescripts common to all sciences against common wisdom in much of the now flourishing HP-n tradition. This reluctance concerning prescriptions was very adequately described by Ian J. Slater in the above mentioned discussion: "The denial of 'scientific method' is, for many philosophers, a way of denying this a priori normative approach to philosophy of science. It is not a claim that there are no common methods used in science. [...] It is rather an objection to the source of authority, and for many philosophers, historians, and sociologists, the failure of Popper (and others) to propose a consistent and comprehensive methodology (or a methodology that a significant number of scientists actually use) has led to the outright rejection of any attempt to prescribe such a thing. The best it is possible to do on this view is to describe what scientists do, and discuss method in the descriptive context. And that's why philosophers deny that there is a scientific method, yet claim that there are scientific methods. They are denying the a priori normative generalized rules, but affirming the fact that you can describe various methods that scientists use in practice." (HOPOS-L Digest, #2001-12, 18 Jan. 2001). Against this dominant trend, I can follow Faye some way in his affirmation of common scientific ways of thinking, though I would insist more on its character of being of a normative and social phenomenon in addition to its logical character. These aspects are inseparable; logic is a normative science, as Peirce remarked. This brings us to a distinctive feature of science, one that is related to its special social character where certain norms are crucial in governing the actions of the individual scientists, and thus, to another and more pragmaticist way of stating Faye's methodological unity thesis: The unity of the various sciences derives from the ethical dimension of scientific inquiry as a process of work (cf. Schmidt 1990): Within the particular scientific institution, a scientist does not have to be recognized as much as he or she must avoid disrespect and disagreeableness, that is, acceptability is not recognition but avoidance of surpassing the limits of acceptability. What defines these limits is a complex set of normative practices (not to be analysed in this note), and among them, the methodological norms that Faye's work is about. Social studies of science -- even though some may commit philosophical fallacies (such as that of inferring from empirical sociology to epistemological relativism) -- can help explore the normative basis of scientific unity across the plurality of disciplines.

The last approach listed, "Studies of Philosophy in Science" (SPiS) is probably the one which is farthest removed from Faye's very general and abstract one. It is not a separate tradition within philosophy of science as a discipline; the designation is used here to point to an important aspect of the relation between science and philosophy, namely, that although scientists seldom care about philosophy of science, they cannot escape being confronted with philosophical problems in their work as scientists, especially when they face basic problems that (in Kuhn's terms) cannot be solved within the strictures and presuppositions of a given paradigm. They are then forced to consider the adequacy of the basic conceptual schemes within the field, its metaphysical obligations, and the possibility and implications of changing the interpretation of the theoretical language or eventually modifying that language. Dealing with theory and theory change in science on this basic level involves genuine philosophical considerations. An example from biology is the shift from an essentialist to a variational notion of species as the units of evolution, triggered by the Darwinian revolution. Here, philosophic and scientific inquiry go together. Accordingly, SPiS denotes an approach to philosophy of science where scientists or philosophers investigate foundational problems within a given field, but not in isolation from scientific work in that field (like in contemporary quantum physics where few physicists care about the different philosophical interpretations). The separation of philosophy and science is illusory in such situations where 'philosophy' is part and parcel of 'science'. In evolutionary biology there has been quite a lot of work done on such themes as the species problem, teleology, and the units of selection problem, with participation from both biologists and philosophers. A recent introduction to the philosophy of biology (Sterelny and Griffieth 1999) testifies much of this work.

I mentioned that one cannot find in Athenes Kammer remarks about in what specific sense also biology, as a certain cluster of research areas, is united to (or in union with) the other sciences, except for methodology; Faye does not subscribe to ontological arguments for unity. But then a question for which Faye offers no answer arises, namely in what sense can we have 'local unity' within a major field, that is, how and why can we talk about big fields like physics, chemistry, and biology as falling in certain 'internally united' clusters? Contemporary biology for instance consists of a wealth of disciplines, each dealing with certain aspects of living systems, their form, function, genesis and relations to other systems -- molecular biology, cell biology, physiology, immunology, ecology, ethology, taxonomy, and developmental and evolutionary biology -- to mention the most important fields. Even though the specialisation into research areas is high, most practitioners of these fields have an intuition of being part of a coherent, integrated science of life within which one can find a set of shared general principles, e.g., expressed in the geneticist Theodosius Dobzhansky's famous saying that "nothing makes sense in biology, except in the light of evolution". If unity of science only comes through method, there would be no explanation of the 'natural' clustering of specialised disciplines other than tradition, convenience of terminology, or institutional organization of education based upon certain conventional clusters. One could not allow that any real intrinsic structure pertaining to the phenomenal domains under study can be the explanation of why both immunology and genetics are natural parts of biology, while cosmology and particle physics both are parts of physics. My point here is not that all disciplines fall neatly into distinct main clusters; on the contrary, a lot of exciting developments in science take place within interdisciplinary areas such as geophysics, biophysics, cognitive science, bioinformatics, brain science, etc., but my point is that there is still a certain structure to be found among all specialised research areas, so that any one discipline is more closely related to a special subset of disciplines than to any arbitrarily chosen discipline. This structure is not explainable by the general, high-level and abstract 'unified method of science' thesis. I guess Faye would here allow for a more detailed level of special methods that might contribute to explain a certain family-resemblance of a cluster of research areas within the humanities as compared to a cluster of disciplines within, say, biology. But a more direct, and in my view correct way to understand the clustered structure of disciplines would refer to the subject matters of study, and in that sense allow for ontological as well as epistemological arguments for the unity of biology. In biology, an ontological argument would be that life or living systems all share a common evolutionary history and have certain emergent features in common which distinguish living systems from other complex physical non-living systems (cf. Emmeche 1997, 1998). Furthermore, one can observe that, as a result of the development of the biological disciplines in the past century, areas that formerly were seen as mainly separated (see Mayr's thesis of 1961 of two kinds of biology, functional and evolutionary) become more integrated in approach as well as in outlook and understanding, so that one no longer finds the same kind of sharp separation between taxonomy, evolutionary biology, genetics and developmental biology as one did 40 years ago. This trend towards integration of theories (and concepts) in biology is a kind of epistemological unity that is not reducible to Faye's general methodological unity thesis, though not incompatible with it. Integration as a central feature of scientific knowledge in general, including integration in biology, was dealt with in a similar book by Collin in 1990. The emergent ontology of living systems and the integration of evolutionary, cellular, genetic and physiological approaches to the scientific understanding of such systems can be seen as supplementary explanations of the coherent and special character of the life sciences.

Thus one could argue for a 'local' unification of the life sciences on ontological and epistemic grounds. Furthermore, no logical necessity forces one to dismiss an argument for 'global' unification that is based not just on methodology but also on epistemology and ontology, just because a bad metaphysics of reductionist physicalism or phenomenalism within the positivist philosophy of science failed to provide it. Indeed, as a framework for this unified view of science, a non-reductive pluralist materialism is a promising candidate, as the work of Køppe (1990) showed. Such a framework preserves the vision of science as giving a grand and ultimately coherent narrative of the world, although it is informed by the postmodern condition that for such big narratives the story has to be told by irreducibly many non-identical voices. The danger of a purist view of unification based solely on methodology is that it degenerates to a general statement about a necessary relation between the concept of science and the concept of method.

Note added.
An English translation of Faye's book is forthcoming with Ashgate Press. The issue of Sats where this comment was published also contains comments by Finn Collin and Paisley Livingston, and a response by Jan Faye. I thank Paisley Livingston for linguistic corrections.


Bak, Thor A. and Erik Dal, eds. Videnskabens Enhed-? København: Munksgaard, 1990.

Callebaut, Werner. Taking the naturalistic turn - or how real philosophy of science is done. Chicago: The University of Chicago Press, 1993.

Collin, Finn. Videnskabsfilosofi. Enhed og mangfoldighed i videnskaberne. København: Museum Tusculanums Forlag, 1990.

Emmeche, Claus. "Autopoietic systems, replicators, and the search for a meaningful biologic definition of life". Ultimate Reality and Meaning 20 (4) (1997): 244-264. Online:

Emmeche, Claus. "Defining life as a semiotic phenomenon". Cybernetics & Human Knowing 5 (1) (1998): 3-17.

Faye, Jan. 'Virkelighedens sprog eller sprogets virkelighed?' Philosophia årg. 22(1-2) (1993): 25-46.

Faye, Jan. 'Enhedsvidenskab - fugl Fønix?' Filosofiske Studier, bd. 18 (1998): 6-31.

Faye, Jan. Athenes Kammer. En filosofisk indføring i videnskabernes enhed. København: Høst, 2000.

Fink, Hans and Kirsten Hastrup, eds. Tanken om enhed i videnskaberne. Århus: Aarhus Universitetsforlag, 1990.

Gallison, Peter. Image and Logic. A material culture of microphysics. Chicago: The University of Chicago Press, 1997.

Hull, David L. Philosophy of Biological Science. Englewood Cliffs, New Jersey: Prentice-Hall, 1974.

Knorr-Cetina, Karin. Epistemic Cultures: How the Sciences Make Knowledge. Harvard University Press, 1999.

Køppe, Simo. Virkelighedens niveauer. De nye videnskaber og deres historie. København: Gyldendal, 1990.

Mayr, Ernst. "Cause and effect in biology". Science 134 (1961): 1501-1506. (Reprinted in: E. Mayr: Towards a New Philosophy of Biology. Cambrifge, Mass.: Harvard University Press, 1988).

Putnam, H. 'What theories are not'. In E. Nagel, P. Suppes, and A. Tarski (eds.): Logic, Methodology and Philosophy of Science. Stanford, Calif.: Stanford University Press, 1962.

Ruse, Michael. The Philosophy of Biology. London: Hutchingson & Co, 1973.

Schmidt, Lars-Henrik. 'Det videnskabelige arbejdes etik'. In Fink, Hans and Kirsten Hastrup (eds): Tanken om enhed i videnskaberne. Århus: Aarhus Universitetsforlag, 1990.

Sterelny, Kim and Griffith, Paul E. Sex and Death. An introduction to the philosophy of biology. Chicago: The University of Chicago Press, 1999.

Thompson, Paul. The Structure of Biological Theories. New York: SUNY Press, 1989.

Woodger, J. H. Biological Principles. A Critical Study. London: Kegan Paul, 1929.