Darwinian Reductionism: Or, How to Stop Worrying and Love Molecular Biologyby Alexander Rosenberg
After the discovery of the structure of DNA in 1953, scientists working in molecular biology embraced reductionism—the theory that all complex systems can be understood in terms of their components. Reductionism, however, has been widely resisted by both nonmolecular biologists and scientists working outside the field of biology. Many of these
After the discovery of the structure of DNA in 1953, scientists working in molecular biology embraced reductionism—the theory that all complex systems can be understood in terms of their components. Reductionism, however, has been widely resisted by both nonmolecular biologists and scientists working outside the field of biology. Many of these antireductionists, nevertheless, embrace the notion of physicalism—the idea that all biological processes are physical in nature. How, Alexander Rosenberg asks, can these self-proclaimed physicalists also be antireductionists?
With clarity and wit, Darwinian Reductionism navigates this difficult and seemingly intractable dualism with convincing analysis and timely evidence. In the spirit of the few distinguished biologists who accept reductionism—E. O. Wilson, Francis Crick, Jacques Monod, James Watson, and Richard Dawkins—Rosenberg provides a philosophically sophisticated defense of reductionism and applies it to molecular developmental biology and the theory of natural selection, ultimately proving that the physicalist must also be a reductionist.
"Rosenberg provides an accessible review of current ideas on the 'wiring' of . . . gene complexes and why they help account for morphological evolution. He is one of the first philosophers to conside the implications of 'evo-devo' . . . and seizes the opportunity to promote a reductionist interpretation that was simply not possible with population genetics."
Bruce H. Weber
"Rosenberg's book tackles very difficult issues in the philosophy of biology in subtle and innovative ways."
Edward M. Engelmann
- University of Chicago Press
- Publication date:
- Edition description:
- New Edition
- Product dimensions:
- 6.00(w) x 9.00(h) x 0.90(d)
Read an ExcerptDarwinian Reductionism
Or, How to Stop Worrying and Love Molecular Biology
By Alex Rosenberg The University of Chicago Press
Copyright © 2006 The University of Chicago
All right reserved.
What Was Reductionism?
LET US DISTINGUISH functional biology from molecular biology. Functional biology is the study of phenomena under their functional kind-descriptions-for example, organism, organ, tissue, cell, organelle, gene. Molecular biology is the study of certain classes of organic macromolecules. This distinction is not entirely satisfactory, for many of the kinds identified in molecular biology are also individuated functionally, as the example of DNA replication at the end of the introduction illustrated. What makes a kind functional is that its instances are the products of an evolutionary etiology-a history of random variation and natural selection. Since natural selection operates at the macromolecular level, some of its kinds will be functional too. But the functional/molecular distinction is a convenient one that reflects widespread beliefs about a real division in the life sciences. Let's employ it as a handy label for the two parts of biology whose relationship is disputed between reductionists and antireductionists.
Employing it, let's briefly review the nature of the dispute about how functional and molecular biology are related. Reductionism is a metaphysicalthesis, a claim about explanations, and a research program. The metaphysical thesis that reductionists advance (and antireductionists accept) is physicalism, the thesis that all facts, including all functional biological facts, are fixed by the physical and chemical facts; there are no nonphysical events, states, or processes, and so biological events, states, and processes are "nothing but" physical ones. The reductionist argues that the metaphysical thesis has consequences for biological explanations: they need to be completed, corrected, made more precise, or otherwise deepened by more fundamental explanations in molecular biology. The research program that reductionists claim follows from the conclusion about explanations can be framed as the methodological moral that biologists should seek such macromolecular explanations. (Note that reductionism is not the evidently indefensible thesis that all biology is molecular biology, that molecular biology not only provides the explanans [what does the explaining], but also uncovers all the facts to be explained [the explanandum, plural explanantia]. This is not reductionism, for it affords no role to functional biology. It is some kind of eliminativism no reductionist has ever advocated.)
Antireductionism does not dispute reductionism's metaphysical claim, but denies it has implications either for explanatory strategies or methodological morals. The antireductionist holds that explanations in functional biology need not be corrected, completed, or otherwise made more adequate by explanations in terms of molecular biology.
The disagreement over the adequacy of explanations in functional biology drives a significant methodological disagreement with consequences for the research program of biology. The reason is simple: if the aim of science is explanation and at least some explanations in functional biology are adequate, complete, and correct, then the methodological prescription that we must search for molecular completions, corrections, or foundations of these functional explanations in molecular processes will be unwarranted. Consequently, molecular biology need not be the inevitable foundation for every compartment of functional biology. If the aim of science is explanation, and functional explanations are either false or incomplete and molecular explanations are either (more) correct or (more) complete, and thus more adequate, then biology must act on the methodological prescription that we should seek macromolecular explanations. If at its explanatory base all biology is molecular biology, then all biologists, or at least all those who seek complete and correct explanations, will have eventually to be molecular biologists as well as functional biologists.
This reductionist conclusion is one that most philosophers of biology, and those biologists sympathetic to them, believe to have been safely disposed of. Once we recognize the proximate/ultimate distinction in explanation and Dobzhansky's dictum about biology, the demonstration that reductionism is false can be left as an elementary exercise in Introductory Philosophy of Science class. Reductionism requires laws to be reduced. But, as will soon be clear, there are no laws in biology to be reduced; or if there are any laws in biology, they govern evolution by natural selection and are not open to reduction to physical laws. No laws, no reductionism. QED.
If this argument is too telegraphic, we may, trespassing on the impatience of au courant biologists and philosophers, rehearse some of its details-suppressed premises, enthymemic inferences, conversational implicatures, and all. We may divide our down and dirty history of reduction into two parts: the vicissitudes of reduction in general, and its narrowly biological problems in particular.
THE ECLIPSE OF POSTPOSITIVIST REDUCTION
To begin with the general problems, we need to recall how the original exponents of reduction, certain logical empiricists and their successors, supposed reduction was to proceed. And we need to remember some of the qualifications added to the original model in order to bring it into contact with the history of science. Reduction is supposed to be a relation between theories. In the anglophone locus classicus, Ernest Nagel's Structure of Science (1961), reduction is characterized by the deductive derivation of the laws of the reduced theory from the laws of the reducing theory. The deductive derivation requires that the concepts, categories, and explanatory properties or natural kinds of the reduced theory be captured in the reducing theory. To do so, terms that figure in both theories must share common meanings. Though often stated explicitly, this second requirement is actually redundant, as valid deductive derivation presupposes univocality of the language in which the theories are expressed. However, as exponents of reduction noted, the most difficult and creative part of a reduction is establishing these connections of meaning, that is, formulating "bridge principles," "bilateral reduction sentences," "coordinating definitions" that link the concepts of the two theories. Thus, it was worth stating the second requirement explicitly. Indeed, early and vigorous opponents of reduction as the pattern of scientific change and theoretical progress argued that the key concepts of successive theories are in fact incommensurable in meaning, as we shall see immediately below.
From the reception of Watson and Crick's discoveries, reductionists began to apply their analysis to the putative reduction of Mendelian or population genetics to molecular genetics. The difficulties they encountered in pressing Watson and Crick's discovery into the mold of theoretical reduction became a sort of poster child for antireductionists. In an early and insightful contribution to the discussion of reduction in genetics, Kenneth Schaffner (1967) observed that reduced theories are usually less accurate and less complete in various ways than reducing theories, and therefore incompatible with them in predictions and explanations. Accordingly, following Schaffner, the requirement was explicitly added that the reduced theory needs to be "corrected" before its derivation from the reducing theory can be effected. This raised a problem that became nontrivial in the fallout from Thomas Kuhn's Structure of Scientific Revolutions (1961) and Paul Feyerabend's "Reduction, Empiricism and Laws" (1964). It became evident in these works that "correction" sometimes resulted in an entirely new theory, whose derivation from the reducing theory showed nothing about the relation between the original pair. Feyerabend's examples were Aristotelian mechanics, Newtonian mechanics, and relativistic mechanics, whose respective crucial terms, impetus and inertia, absolute mass and relativistic mass, could not be connected in the way reduction required.
No one has ever succeeded in providing the distinction that reductionism required between "corrections" and "replacements." Thus, it was difficult to distinguish reduction from replacement in the crucial cases that really interested students of reduction. This was a matter of importance because of reductionism's implicit account of scientific change as increasing approximation to more fundamental truths. It was also Schaffner who coined the term "layer-cake reduction" to reflect the notion that synchronically less fundamental theories are to be explained by reduction to more fundamental theories-at the basement level, some unification of quantum mechanics and the general theory of relativity; above these, physical and organic chemistry; then molecular biology and functional biology; at the higher levels, psychology, economics, and sociology. Synchronic reduction is supposed to be explanatory because on the account of explanation associated with reduction, the deductive-nomological (D-N) model, explanation was logical deduction, and the explanation of laws required the deduction of laws from other laws. Synchronic reduction is mereological explanation, in which the behavior of more composite items described in reduced theories is explained by derivation from the behavior of their components by the reducing theory. Thus, reduction is a form of explanation. Diachronic reduction usually involves the succession of more general theories that reduce less general ones by showing them to be special cases which neglect some variables, fail to measure coefficients, or set parameters at restricted values. As the history of science proceeds from the less general theory to the more general, the mechanism of progress is the reduction of theories. But if there is no way to distinguish reduction from replacement, then the incommensurability of replacing theories makes both the progressive diachronic and synchronic accounts of intertheoretical relations impossible ideals.
More fundamentally, reductionism as a thesis about formal logical relations among theories was undermined by the increasing dissatisfaction among philosophers of science with the powers of mathematical logic to illuminate interesting and important methodological matters such as explanation, or theory testing. Once philosophers of science began to doubt whether deduction from laws was always sufficient or necessary for explanation, the conclusion that intertheoretical explanation need take the form of reduction was weakened. Similarly, reductionism is closely tied to the axiomatic, or so-called syntactic, approach to theories, an approach that explicates logical relations among theories by treating them as axiomatic systems expressed in natural or artificial languages. Indeed, "closely tied" may be an understatement, since deduction is a syntactic affair, and is a necessary component of reduction. But, for a variety of reasons, the syntactic approach to theories has given way among many philosophers of biology to the so-called semantic approach to theories. The semantic approach treats theories not as axiomatic systems in artificial languages but as sets of closely related mathematical models. The attractions to philosophers of biology of the semantic approach must be manifest. For a science like biology, without laws of the sort we meet with in physical science, can hardly display axiomatized theories; and one in which mathematical models figure so centrally in explanations is immediately amenable to analysis from the semantic perspective. (For a useful introduction to this conception and its attractions for biology, see Thompson 1988; Lloyd 1993.) But, on the semantic approach, the very possibility of reduction by deductive derivation of the axioms of one theory as theorems of the other became moot. The semantic approach treats theories as families of models, and models as implicit definitions, about which the only empirical question is whether they are applicable to phenomena. For reduction to obtain among theories semantically characterized requires an entirely different conception of reduction. On the semantic view, the reduction of one theory to another is a matter of employing one (or more) model(s) among those that constitute the more fundamental theory to explain why each of the models in the less fundamental theory is a good approximation to some empirical process, showing where and why it fails to be a good approximation in other cases. The models of the more fundamental theory can do this to the degree that they are realized by processes underlying the phenomena realized by the models of the less fundamental or reduced theory. There is little scope in this sort of reduction for satisfying the criteria for postpositivist reduction. (We will return to the role of mathematical models in the explanation of biological processes in chapter 4.)
To the general philosophical difficulties that the postpositivist account of reduction faced, biology provided further distinct obstacles. To begin with, as Hull first noted (1974), it is difficult actually to define the term gene as it figures in functional biology by employing only concepts from molecular biology. In other words, the required "bridge principles" between the concept of the gene as it figures in population biology, evolutionary biology, and elsewhere in functional biology and as it figures in molecular biology could not be constructed. And all the ways philosophers contrived to preserve the truth of the claim that the gene is nothing but a (set of) string(s) of nucleic acid bases could not provide the systematic link between the functional "gene" and the macromolecular "genes" required by a reduction (see chapter 4 below for more details). There is, of course, no trouble identifying "tokens"-particular bits of matter we can point to-of the population biologist's genes with "tokens" of the molecular biologist's genes. But token-identities won't suffice for reduction, even if they are enough for physicalism to be true.
The second problem facing reductionism in biology is the absence of laws, either at the level of the reducing theory or the reduced theory, or between them. If there aren't any laws in either theory, there is no scope for reduction at all. That there can be no laws in biology, with the one exception of the laws that govern evolution by natural selection and their consequences, is the major conclusion of the next chapter. For the moment, let's assume there are none; antireductionists will by and large grant the assumption, for it strengthens their case for autonomy and against reduction (see Kitcher 1984, for example).
Of course, the first problem, that of "defining" functional genes in terms of macromolecules, is really not very different from the problem of identifying laws linking functional genes to macromolecules, since the "bridge principles" reduction requires will have to be laws of nature. Thus, the argument (to come in detail in chapter 4) that there are no biological laws makes impossible fulfilling either reductionism's criterion of connection or its criterion of derivation by deduction.
Whereas the antireductionists were at most able to show that the criterion of connectability with respect to the Mendelian and the molecular gene was not fulfilled as the two theories were in fact stated, we can go much further in vindicating their conclusion. We can demonstrate that the criterion required by "layer-cake" reductionism cannot be satisfied as a fundamental matter of biological process. As we saw, individuation of types in biology is almost always via function: to call something a wing, a fin, or a gene is to identify it in terms of its function. But biological functions are naturally selected effects. And natural selection for adaptations-that is, environmentally appropriate effects-is blind to differences in physical structure that have the same or roughly similar effects. Natural selection "chooses" variants by some of their effects, those which fortuitously enhance survival and reproduction. When natural selection encourages variants to become packaged together into larger units, the adaptations become functions. Selection for adaptation and function kicks in at a relatively low level in the organization of matter. Accordingly, the structural diversity of the tokens of a given Mendelian or classical or population biological or generally "functional" gene will mean that there is no single molecular structure or manageably finite number of sets of structures that can be identified with a single functional gene.
Excerpted from Darwinian Reductionism by Alex Rosenberg Copyright © 2006 by The University of Chicago. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Meet the Author
Alex Rosenberg is the R. Taylor Cole Professor of Philosophy and Biology at Duke University and the author of many books, including Economics—Mathematical Politics or Science of Diminishing Returns? and Instrumental Biology, or The Disunity of Science, both published by the University of Chicago Press.
and post it to your social network
Most Helpful Customer Reviews
See all customer reviews >