The editors define adaptationism as “the claim that natural selection is the only important cause of the evolution of most nonmolecular traits and that these traits are locally optimal” (6). In this volume they have assembled a dozen essays by some of the most important theorists and researchers in the field in order to promote dialogue regarding one of the central issues in evolutionary biology. Although the issue is primarily a theoretical one, the book's audience is “the practical-minded biologist” with interests in “the intellectual, social, and organizational aspects of evolutionary biology” (15–16). The depth and range of the research represented in this excellent book will make it essential reading for anyone expecting to do work in this field.
The book divides naturally into four sections. A common theme in the first three chapters is the use of phylogenetic analysis as a means of testing optimality claims. Chapters 4 and 5 then proceed to discuss the theory of optimality and the notion of evolutionarily stable strategy models. Chapters 6, 7, and 8 have to do with the rationale and meaning of tests of optimality models, and the final four chapters constitute a survey of various positions in the debate over adaptationism.
In “A Likelihood Framework for the Phylogenetic Analysis of Adaptation,” David Baum and Michael Donoghue argue that two distinct approaches to the study of adaptation—the homology approach and the convergence approach—can be unified in a useful way. Proponents of both homology and convergence are skeptical about causal claims in adaptationist accounts, but Baum and Donoghue suggest that it is possible to give a framework for studying character evolution in terms of what they call “likelihood” that is both statistically rigorous and centered on giving an analysis of unique historical events in light of information from analogous characters.
In their “Adaptation, Phylogenetic Inertia, and the Method of Controlled Comparisons,” editors Hecht and Sober argue that natural selection and phylogenetic inertia are not mutually exclusive explanations of trait values or trait associations. Depending on the case, one, both, or neither explanation may be suitable.
Chapter 3, “Optimality and Phylogeny: A Critique of Current Thought” by Hudson Kern Reeve and Paul W. Sherman, suggests that historical methods for analyzing the effects of natural selection, genetic drift, and similar processes have been widely misused. In particular they claim that empirical evidence does not support the view that parsimony-based phylogenetic methods are sufficient to recover information about ancestor states, but rather shows such methods to be misleading under certain circumstances.
Joel Brown, in “Fit of Form and Function, Diversity of Life, and Procession of Life as an Evolutionary Game” (Chapter 4), applies evolutionary game theory in a striking and original way to explain how natural selection results in the phenomena named in the chapter title. According to Brown, evolutionary game theory, as a research agenda, has probably been eclipsed by genetic research programs, but is rapidly expanding and will eventually prove itself a useful explanatory approach.
In Chapter 5 (“Optimality and Evolutionary Stability under Short-Term and Long-Term Selection”), Ilan Eshel and Marcus Feldman distinguish between what they call short-term evolution (changes in relative frequencies of genotypes in the short run) and long-term evolution (mutations are tested by trial and error over the long run). The two processes are qualitatively different from each other; convergence to individual local optima, while unlikely in short-term evolutionary contexts, is common in long-term evolution, but since the two processes are so radically different it is a mistake to draw inferences about long-term evolution based upon the observable processes of short-term evolution.
Edward Allen Herre, Carlos Machado, and Stuart West present a study of sex ratio responses in populations of fig-pollinating wasps as a means of identifying considerations that are likely to be important in constructing and interpreting tests of optimality (Chapter 6, “Selective Regime and Fig Wasp Sex Ratios”). They are particularly interested in the question of the direction of fit of optimality models: do researchers use the models to test the precision of adaptation, or do they use adaptive responses to test the precision of the models?
In “Is Optimality over the Hill? The Fitness Landscapes of Idealized Organisms” (Chapter 7), George Gilchrist and Joel Kingsolver assess recent studies and adduce new examples to show the complementarity of fitness landscapes and genetic models as methods for describing and predicting adaptedness. They argue that the more “topographical detail” present in a fitness landscape, the better the understanding of the real significance of adaptedness in a population.
According to Kenneth Halama and David Reznick (Chapter 8, “Adaptation, Optimality, and the Meaning of Phenotypic Variation in Natural Populations”) more work must be done on adaptive phenotypic variation within populations. Populations can support multiple phenotypes (each adapted to some environmental subset) and this phenotypic plasticity plays an important role in creating variation among individuals.
In “Adaptationism, Optimality Models, and Tests of Adaptive Scenarios” (Chapter 9) Peter Abrams suggests that the evidence that has been traditionally adduced in support of optimality models in the literature has been weak and that quantitative models are in any case seldom possible. Like others in this volume, Abrams assesses the social aspect of the debate over adaptationism and concludes that both sides have, to a certain extent, neglected to say enough about the purpose of optimality models.
The prospects for a synthesis between adaptationism and developmental biology is Ron Amundson's topic in Chapter 10, “Adaptation and Development: On the Lack of Common Ground.” The main difficulty lies in fundamental disagreements over explanatory suppositions, and although there have been successful syntheses of radically divergent research programs in the past, Amundson is skeptical about the prospects for one between adaptationism and developmental biology.
Peter Godfrey-Smith describes three versions of adaptationism that can be found among working biologists, one understanding natural selection as a “fit” between organism and environment, another taking it to be the only necessary explanatory factor regarding natural populations, and the third adopting the normative attitude that natural selection should be the null hypothesis in any research program (Chapter 11, “Three Kinds of Adaptationism”). Godfrey-Smith holds that evidence for or against one version does not count as evidence for or against any of the other versions.
The last chapter in the collection (“Adaptation, Adaptationism, and Optimality”) finds Egbert Leigh relying on a version of the design argument to show that adaptation is self-evident. After reviewing the claims of D'Arcy Thompson, Fisher, and Levin regarding the nature of adaptedness, Leigh shows how to construct an optimality model, how to understand adaptive function, and how to determine appropriate levels of selection for a given model.
