Dupré, J. (1997) Review of Brandon's "Concepts and Methods in Evolutionary Biology" British Journal for the Philosophy of Science 48 292-296. 

Review of Brandon's "Concepts and Methods in Evolutionary Biology"

John Dupré
Department of Philosophy
Birbeck College, University of London
Malet Street, London, WC1E 7HX
(0171) 631 6383/6549  Fax:  0171 631 6564
School of English and American Studies
University of Exeter, Exeter, EX4 4QH
(01392) 264267/264257  Fax:  01392 264361

ROBERT BRANDON (1996)Concepts and Methods in Evolutionary Biology Cambridge, Cambridge University Press, paper 00000 

This book is a collection of essays by a leading philosopher of biology and spans his career over almost the last twenty years. Most of the topics that have been of concern to philosophers of biology in this period are touched on to some extent, and the collection of these essays in a convenient volume will certainly be welcomed by everyone working in this field. The essays are arranged chronologically, and divided into three sections. Although the chapters in the first section have substantial interconnections, being involved with fundamental conceptual issues in evolutionary theory, on the whole there is not much attempt to tie the book into anything more than a sequence of independent essays. The later sections, moreover, cover a quite diverse range of topics. The whole is neither more nor less than the sum of the parts. For these reasons I shall, in this review, more or less follow the chronological sequence of the collection.

The first section of the book, containing papers published in the late seventies and early eighties, addresses some of the most widely discussed issues in the philosophy of biology: the nature of adaptation; teleological, or functional, explanation; the structure of evolutionary theory; and the units of selection. These are valuable papers that have contributed significantly to the development of discussion on these topics, and that helped to establish Brandon's reputation as a major contributor to the field. I would, however, be skeptical of the claim on the back cover that the book could serve well as an introduction to the field. The papers occupy recognisable positions in the evolution of various debates, and it is hardly surprising that in some respects the field has developed in directions that could not have been predicted 15 years ago. As one example, Brandon deliberately avoids discussion of the term "function" in his paper on teleological explanation. But subsequent discussion of these issues has been focused very heavily on this concept. A contemporary survey of the topic would also need, for example, to address the contrast between the selective benefits that led to the evolution of a trait and the selective benefits that have currently or recently maintained the trait in the population.

The first section concludes with a brief piece on the units of selection problem or, as Brandon reformulates it, the levels of selection problem, surely the topic to which more trees have been sacrificed than to any other in the philosophy of biology. The ideas introduced here are treated much more fully and successfully in a later chapter (chapter 8), so that the earlier chapter is of mainly historical interest. Brandon's basic point, introduced in the earlier paper but much refined in the later, is that progress on this topic requires a sharp distinction between the level of the biological hierarchy at which selection takes place and the units which are selected. In the most familiar scenario, for instance, selection takes place at the organismic level, but what are selected are genetic units of various sizes up to and including the entire genome. Deploying to good advantage David Hull's now well-known distinction between interactors and replicators, Brandon describes a hierarchy of interactors any of which may turn out, empirically, to be the level at which selection is taking place. At the end of the earlier piece Brandon offers a motivation for the problem of the levels of selection, but leaves it to others to motivate the units of selection problem. In the later piece he argues that much previous work on the topic has provided a hierarchy incoherently confusing interactors and replicators, and he offers no very compelling function for a hierarchy of replicators. Brandon thus offers both clarification of the standard units of selection problem, and a defence of a particular stance on the problem, specifically a pluralistic one on which the important question is where the causal action of selection is found to lie. The solution strikes me as right-headed, and the later paper, originally published in an anthology not widely read by philosophers, is perhaps as good as any philosophical treatment of the topic.

The middle section of the book begins with interesting papers on human sociobiology and on the origins of language from the perspective of the ecological conditions under which phenotypic plasticity would be selectively advantageous. The first paper concludes that the kinds of unstable environment that favour phenotypic plasticity will also favour cultural transmission. Our own species, of course, is a plausible candidate for such a set of adaptations, and this will raise serious doubts about the simple application of adaptationist arguments to human behaviour. This is especially true because, as others have also noted, the patterns of phenotypic transmission made possible by culture will not necessarily promote biological optimality. Some of the lessons of this discussion are applied in the second paper (coauthored with Norbert Hornstein) to the formulation of a possible set of scenarios for the evolution of language. I'm not entirely sure how to evaluate an exercise of this kind, confessedly speculative rather than empirical, but certainly the perspective seems a plausible one. (On a carping note, the reader is irritatingly reminded of the drawbacks to collections of reprinted papers by almost seven pages of nearly verbatim overlap between these two pieces.)

This section of the book also contains, in addition to the paper already mentioned on levels of selection, a paper on species co-authored with Brent D. Mishler. The paper includes a useful critical but generally positive discussion of the species as individuals thesis, and advocates an account of species (first introduced by Mishler and M.J.Donoghue) grounded on the important distinction between criteria of grouping and of ranking. Their "phylogenetic species concept" is monistic with regard to grouping: organisms classified together should be monophyletic, that is, descended from a single ancestral organism or (cohesive) group of organisms. But it is pluralistic with regard to ranking. That is to say, there is no unique criterion on the basis of which an appropriately monophyletic group is denominated a species. This is one of the more controversial areas in the philosophy of biology, with regard to both the monism and the character of the grouping criterion, though the position Brandon and Mishler advocate has become a significant contender. It is not, however, a position that could be sensibly evaluated without a fairly serious immersion in the wider debate.

The last and most recent set of papers addresses rather broader topics. Particularly welcome is a pioneering attempt to address the relation between theory and experiment in evolutionary biology. The paper includes a helpful classification of empirical work on evolution along the dimensions from hypothesis testing to more disinterested "Baconian" description or measurement and from manipulation to pure observation. Most interesting, perhaps, is Brandon's conclusion roughly that there is lots of good theory and lots of good data in evolutionary biology, but not much connection between them. Brandon views this optimistically as a motivation for exciting further work rather than an embarrassment for the field. Others might be less complacent about the value of theoretical work, most notably in population genetics, that appears so distant from any readily imaginable connection with empirical reality.

One of Brandon's greatest contributions to the philosophy of biology has been his insistence on the necessity of a sophisticated analysis of the concept of the environment for providing an adequate account of evolution. This is a central point of his monograph, Adaptation and Environment. In this light I found the contribution to the present volume (co-authored with Janis Antonovics) a bit disappointing. I suppose this was in large part because the centrepiece of the paper is a detailed analysis of a computer simulation of a highly simplified evolutionary process, and I am generally skeptical of the value of such exercises. The paper does, however, introduce Brandon's useful distinctions between external, ecological, and selective environments.

The book concludes with the broadest topic, the defence of what Brandon calls "mechanism" against reductionist and holist rivals. The general position defended semed to me sensible and convincing. My major reservations were terminological. Most strongly, I found Brandon's cooptation of the term "mechanism" surprising. (That, I suppose, is unsurprising since I have devoted a good deal of time to defending what I have described as anti-mechanism, but I agree with most of Brandon's actual metaphysical views.) Brandon's characterization of the kind of mechanisms that science looks for is that "a mechanism is just a causal pattern." This is intentionally "vague and open-ended" to avoid any illegitimately a priori restriction on the kinds of causal patterns science may actually discover. It specifically allows, contra reductionism, for top-down causality as well as bottom-up or, as Brandon prefers, "inside-out" causality, and also for indeterministic regularities. The only restriction on nature it presupposes is that "there are discoverable causal regularities in the world", which Brandon takes to be true though not necessarily so. (He does not say whether he thinks that every event occurs in accordance with some causal regularity, which I take it would be the only controversial reading of this claim.) Though, as I have said, I find this position entirely sympathetic, I do not think it sits happily under the rubric of mechanism. Pedantically, I suppose, mechanism should be understood as an obsolete seventeenth century view of the world as composed solely of, and to be understood wholly in terms of, impenetrable atoms of matter in motion. Understood as relevant to contemporary views, coherence with this traditional sense suggests that mechanism is at least broadly reductionistic and, if not deterministic, at least explicitly committed to the causal closure of the physical. Perhaps part of the problem is that Brandon is not very sympathetic in his treatment of the positions he opposes to mechanism. Reductionist thinking, in particular, has advanced beyond the formulations by Nagel in 1961. The upshot of this is that I do not find in this essay a persuasive argument that there is any basic defect in the traditional debate between reductionists and anti-reductionists. Mechanism, if it is to be conceived as an ongoing metaphysical project rather than a historical curiosity should be conceived as embracing the most sophisticated contemporary versions of reductionism, and Brandon's own account strikes me as a plausible version of anti-reductionism. And, of course, as his formulation illustrates, anti-reductionism need not be equated with any mystical doctrines about vital forces or spirits.

Brandon is clearly seen in this collection as a philosopher more concerned to influence the course of biology than of philosophy (though the last piece is a notable exception). Most of his work is concerned with issues that will be familiar only to professional philosophers of biology and biologists, and he shows no interest in issues concerning the social or political context of science or even, in any general sense, with the epistemic or ontological status of science. Given these aims and limitations, however, it provides a first-rate selection of essays spanning a large range of topics arising in contemporary evolutionary biology. The book will be of considerable value to a variety of professional students of evolutionary theory, and will be a useful resource for postgraduate level seminars in the philosophy of biology.