Natural Selection as a Population-Level Causal Process

Abstract

Recent discussions in the philosophy of biology have brought into question some fundamental assumptions regarding evolutionary processes, natural selection in particular. Some authors argue that natural selection is nothing but a population-level, statistical consequence of lowerlevel events (Matthen and Ariew [2002]; Walsh, Lewens, and Ariew [2002]). On this view, natural selection itself does not involve forces. Other authors reject this purely statistical, population-level account for an individual-level, causal account of natural selection (Bouchard and Rosenberg [2004]). I argue that each of these positions is right in one way, but wrong in another; natural selection indeed takes place at the level of populations, but it is a causal process nonetheless. argue for a third logical possibility: natural selection is indeed a causal process, but it operates at the population level.The issues at stake in this debate will be fleshed out in more detail in the discussion below. However, at the outset it is important to distinguish this debate from the more well known debates over the units (or levels) of selection; unfortunately, similar terminology makes this distinction difficult. One obvious difference between the debates is that the units of selection debates assume a causal basis for natural selection, whereas the existence of selection's causal basis is one of the points of contention in the debate at hand. With regard to the issue of levels, however, the difference between the two debates can best be seen by the following example.Suppose that a person believes, in terms of the levels of selection debate, that selection (either sometimes or always) acts on organisms. Such a person still might ask whether that selection Millstein, p. of 3 of 41process was acting on individual organisms or populations of organisms. 1 It is the latter question that concerns us here, and indeed, throughout this paper I have, for the sake of simplicity, assumed organismic selection, although I believe that similar arguments can be made for other units of selection. Now, it may turn out that the present debate has interesting consequences for the debates over the units of selection, but I will not explore such consequences here.Further terminological confusion arises because of the distinction being made between an individual and a population. Indeed, one might wish to speak of an 'individual' population rather than a class or type of population. However, the reader should understand that, in what follows, 'individual' refers to an individual organism, whereas a population refers to a particular spatiotemporal collection of organisms.Some philosophers might object to the very idea of population-level causality. In response to this concern, I will argue in section 2.1 that anyone who accepts the reality of frequency-dependent selection is already committed to population-level causality. Moreover, in section 2.2 I will argue that population-level causality is consistent with three commonly accepted accounts of causality; thus,...

A critical review of the statisticalist debate

“…Metaphysical, in the sense such an analysis tries to identify the nature of evolutionary phenomena prior to or independent from a particular scientific theory that deals with these phenomena. It is in this context that Millstein distinguishes discriminate and indiscriminate sampling processes, as two types of causal processes that are affected or not by phenotypic differences between organisms (Millstein, 2002(Millstein, , 2005(Millstein, , 2006. This is an ontological distinction made without resorting to any conceptual apparatus of evolutionary theory.…”

“…Millstein (2006Millstein ( , 2008 analyzes some case studies where field biologists try to establish causal relationships between fitness on the one hand and phenotypic or genetic characters on the other, in order to confirm their adaptation hypotheses that the traits in questions were formed by selection and not by drift. Another classical example is Andersson (1982)'s field study of sexual selection in which he confirmed tail length of widowbirds affect their mating chance (a surrogate measure of fitness) by experimentally manipulating the phenotype.…”

Scientific Realism, the Semantic View and Evolutionary Biology

“…Even if we accept, for the sake of the argument, that there could be a way to make this abstractness compatible with the claim that Price's Equation gives us the deep structure of biological evolution, there remains a problem. The problem is that several authors support the idea that it is possible to give a causal interpretation of the FTNS [50][51]. Indeed, the causal interpretation of the evolutionary principles "shows adaptive evolution as a genuine causal process, where fitness and selection are both causes of evolution" [47, is simply a mathematical tautology whose truth follows from the definition of the terms.…”

Fitness: Philosophical Problems

“…There are debates about how, precisely, drift is to be distinguished from selection. Some, for example, argue that selection and drift represent distinct causes (Hodge 1987;Millstein 2006), while others hold that the distinction is best understood in terms of outcomes (Matthen and Ariew 2002;Walsh 2007). But regardless of how one attempts to make this distinction, it is clear that if fitness is understood in terms of realized outcomes only, then the theoretical resources for distinguishing selection and drift do not exist.…”