The argument of my larger discussion is that the purportedly evolutionary explanations being advanced in psychology and the social sciences, for example, in “evolutionary psychology” and “cognitive anthropology,” are not the genuine article. The roots of the failure to generate evolutionary explanations lie in the methods social scientists employ. First, those who use “reverse engineering” rather than historical methods to derive “evolutionary” explanations of the features of the human psychology they posit, consistently fail to rule out alternative explanations. Thus, they fail to make a plausible case that these features are adaptations. Second, a common line of reasoning is that the discovery of an alleged “universal” in human behavior constitutes evidence of some innate psychological mechanism, and innateness constitutes a prima facie case for adaptation. This methodological assumption suffers from some well-known problems, as I later note. But a minimum requirement for making the case for adaptation on its basis is demonstrating that the psychological feature posited -- the cognitive “mechanism” or “predisposition” -- guides rather than merely fits an aspect of behavior. Proponents, I contend, consistently fail to establish this stronger claim. In the end, cognitive psychology contributes little other than the notions of “cognitive modules” and “cognitive mechanisms” to the research here considered, evolutionary biology little other than the notions of natural selection and adaptation.
Several qualifications are in order. First, few of the arguments I advance are original. Many figure in critiques of evolutionary psychology, the most visible and influential of the relevant research programs, and were prominent features of the criticisms leveled at human sociobiology more than two decades ago. Regrettably, recurrent themes in the current literature demonstrate that these arguments need to be rehearsed again.
Second, although I arrive at skeptical conclusions concerning the present and future prospects of these research programs, I am a Darwinian, to paraphrase W. V. Quine in another context, to the extent that anyone in their right mind would be. But one can grant that aspects of human behavior and psychology stem from evolved capacities, and maintain that social scientists have not succeeded in their efforts to construct evolutionary explanations of them -- and are unlikely to do so given the methods they employ (cf. Dupre 1998, Richardson 2001). Alternatively put, the substance of my critique concerns the evidential warrant of the methods employed and hypotheses advanced in these research programs. It is not motivated, as advocates of these programs often suggest, by naive views about humans, or fear that “hard-headed” investigations into human nature will uncover some nasty truths about it (at the risk of being overly facetious, there is little to fear in this regard), or an opposition to any and all forms of reductionism.2
Commonly, these programs assume a specific model of conceptual integration, according to which biology explains psychology and psychology explains culture. More specifically, we are told that evolutionary psychology supplies “the middle causal theory” linking biology and the social sciences: “the psychological mechanisms that come between theories of selection pressures on the one hand and fully realized sociocultural behavior on the other” (Cosmides et al. 1992:6). But, of course, conceptual integration so construed is not the only way to conceptualize the relationship between the social/behavioral sciences and biological/natural sciences. We can grant that the objects studied by the various sciences differ in their compoundedness and think of them as being arranged on something akin to a ladder. And we can grant that, assuming physics is at the “bottom” of this ladder, there are constraints that flow upward: that is, that economic cycles and social institutions are subject to the laws of physics, that human biology dictates that we cannot fly even if a desire to do so should turn out to be an innate feature of human psychology, and that evolutionary processes contributed to aspects of human psychology and behavior, as well as to our biology. We can also grant that the assumption that there is a “lower level” description available for objects that have dubious identity criteria (e.g., the assumption that there is a physical-state description of “ideas” or “beliefs”) finances or underwrites our present commitment to such objects. We can even maintain that such assumptions are not just presuppositions of science as we know it, but results of science to date. All of this is compatible with there being unique causal relationships at each level that a full account of what goes on would include (Nelson and Nelson 2000b, cf. Dawkins 1989, Dennett 1995, Dupre 1998). Moreover, on the arguments canvassed here, no genuine integration of evolutionary biology is being achieved in the research under consideration.
The criticisms I offer are also not motivated by “political” considerations of the simple-minded or vulgar sort evolutionary psychologists often attribute to their critics.3 But they do challenge the “value-freedom” and apolitical status evolutionary psychologists claim for their hypotheses, and for those of evolutionary biology and human sociobiology they seek to appropriate. I also develop an argument Philip Kitcher advanced some years ago4 to maintain that we are warranted in bringing the most stringent standards of evidence to bear on hypotheses positing “innate” features of human psychology precisely because these claims typically carry significant social implications.5
Stringent standards, but defensible ones, and the acceptance of a “new” research program or theory may be justified on grounds other than empirical results.6 Such grounds obtain when rival research programs or theories have been shown to be bankrupt or there is a perceived need, again based on developments in the sciences, for a new discipline. But even in such cases (and it is not clear that either situation presently obtains7), we need to ask whether the new program is promising. At a minimum, this calls for an evaluation of its core methodological assumptions, research questions, and hypotheses in light of closely related successful theories. The research programs under consideration claim to use evolutionary biology and, by so doing, to provide evolutionary explanations of human psychology and behavior. Thus, it is appropriate to assess their methods and explanations using core tenets and accepted methods of evolutionary biology. It is the argument of the next several sections that when we do so, these programs are less than promising.
Finally, I will contend that, in the case of evolutionary psychology, those of us who argue against the viability of this research program are not aptly described as “cutting off a new research program at the knees.”8 As will become clear, many hypotheses being advanced in this research program are well-worn hypotheses of human sociobiology and, in some cases, Social Darwinism before that, to which evolutionary psychologists have simply grafted technical notions from cognitive science.9
The next two sections outline some significant difficulties attendant to making a plausible case, let alone demonstrating, that a trait or capacity is an adaptation, and some formidable obstacles to using historical methods to make the case that features of human psychology are adaptations. In the fourth section these arguments are among those brought to bear on the methods employed in evolutionary psychology to demonstrate the failure to make the case for adaptation. The fifth section is devoted to an analysis of parental investment theory, a mainstay of evolutionary psychologists’ explanations of alleged evolved predispositions toward mating and parenting. An analysis of an argument Richard Dawkins has offered for asymmetrical sex roles makes it clear that even quite sophisticated theorizing in evolutionary biology, indeed theorizing not thought to apply to human behavior, can be deeply informed by sociocultural and value-laden assumptions. In the sixth section I use a hypothesis from cognitive anthropology to explore what is needed to make the case that a specific cognitive mechanism or predisposition guides human behavior, and to show that, in terms of this hypothesis and others, there is a failure to make this case. In my conclusion, I maintain that there is a special demand to be made of scientists advancing hypotheses that carry social implications that in no way violates the boundary traditionally (if problematically) assumed between science and politics. In such cases, we should expect scientists to uphold and meet the highest epistemic standards.
Adaptation so understood is, as G.C. Williams makes the point, “a special and onerous concept” (Williams 1966: vii). Williams was in part responding to misapplications of the notion to traits only shown to be conducive to fitness (Williams 1966:vii). Arguments insisting on this distinction are common and originate with Darwin. Elliot Sober’s claim that a trait is an adaptation if and only if is “selected for” is representative (Sober 1993), as is Richard Burian’s criticism of the assumption among some population geneticists that a correlation with increased fitness is sufficient to demonstrate that a trait is an adaptation (Burian 1992). 13
This points to a second reason why adaptation is a difficult notion: making the case that a given trait or capacity is an adaptation requires ruling out alternative explanations that are equally viable. As earlier noted, a trait or capacity may be the result of biological processes other than selection and/or neutral in adaptive significance. In addition, and, again, as Darwin recognized, current function may reflect conversion and cannot itself establish original function or causal history (Darwin 1859). Finally, as Daniel Dennett makes the point, when the “trait” in question is an adaptive behavior or capacity exhibited by a species displaying some degree of behavioral flexibility, equally viable explanations include the discovery of “a forced move” or “good trick”-- and at least in the human case, cultural descent, codiscovery, and information transmission (Dennett 1995). 14
Reflecting these issues, it is a common argument that to make a plausible case, let alone demonstrate, that a given trait or capacity is an adaptation requires knowledge and use of history. Sober emphasizes, for example, that “adaptation is a historical concept” (1993:84). Burian and Ernst Mayr argue that it is because correlation with increased fitness does not indicate the historical origin of a trait that it is insufficient to demonstrate adaptation (Burian 1992:11 and Mayr 1983, cf. Brandon 1978). This yields a third difficulty. For reasons explored in some detail below, the kinds of historical information required to rule out alternative explanations or to identify the causal process(es) that resulted in a given trait or capacity are often difficult to come by -- sometimes simply because selection tends “to cover its own tracks ... to destroy the variation on which it acts” (Sober 1993:69).
We will have reason to return to most of the arguments so far summarized. Here I note their general implication: generating an adaptation explanation is hard work.
Evolutionary psychologists, as much if not more than other social scientists, are clear about basic evolutionary notions. They claim to demonstrate that the features of human psychology they posit are adaptations and distinguish this from claims that they are presently adaptive. They recognize that making a plausible case for, let alone demonstrating, adaptation requires history. Finally, they claim to have and use the requisite historical information. The Introduction to The Adapted Mind (Cosmides et al. 1992), a collection whose editors and contributors include prominent members of the discipline, cites each claim as a central premise of the research program.16
In this section, I postpone the question of whether evolutionary psychologists make even a minimally plausible case for the specific “evolved psychological mechanisms” they posit. I focus here on the nature of the reasoning that yields “the history” to which they appeal, the reasoning they use to identify specific “adaptive problems” facing ancestral populations, and the reasoning they use to predict specific “psychological mechanisms” selected for in light of “known adaptive problems.” I contend that, in each case, reverse engineering is the primary method employed. As the use of this method to derive adaptation explanations has its champions as well as its critics, I begin by summarizing arguments on each side. I then explore cases that demonstrate that evolutionary psychologists fail to meet the standards outlined by supporters of this method and graphically illustrate the pitfalls its critics cite.
By its nature, reverse engineering is an ahistorical method of explanation in which, beginning from the assumption that a phenomenon is the result of some process of building or “design” (that is, is an artifact in the most general sense), one infers “why” it is the way it is or “why” it came to be. Its use in evolutionary theorizing is uncontroversial when history is irrelevant -- for example, when it is used to derive an explanation of a trait’s present adaptiveness. But when used to generate adaptation explanations -- that is, when present traits or capacities are used to infer historical conditions and causal processes -- it is at least controversial.17
Some, including Dennett, maintain that reverse engineering is not only appropriate but indispensable to adaptation explanations, on the grounds that biology just is engineering and the work of natural selection just is R and D. Thus, Dennett argues, “the engineering perspective on biology is not merely occasionally useful, not merely a valuable option, but the obligatory organizer of all Darwinian thinking, and the primary source of its power” (Dennett 1985, 185; cf. Pinker 1994, Pinker and Bloom 1992). Critics of the method, according to Dennett, were motivated by “a misplaced fear about what it might entail” (185). Dennett’s reference here to “a misplaced fear” is misleading, for critics of the use of this method express more than one concern. But he does defend the method against one of them: the charge that beginning from the assumption that a given trait is an adaptation encourages what Steven Jay Gould and Richard Lewontin call “Just So Stories” (Gould and Lewontin 1979). Critics of the method, Dennett argues, “were reacting against a certain sort of laziness: the adaptationist who hits upon a truly nifty explanation for why a particular circumstance should prevail, and then never bothers to test it -- because it is too good a story, presumably, not to be true” (1995:242). However, these problems can be overcome by adhering to
good rules of thumb to be followed by the prospective reverse engineer ... (1) Don’t invoke adaptation when other, lower-level explanations are available (such as physics) ... . (2) Don’t invoke adaptation when a feature is the outcome of some general developmental requirement ... . (3) Don’t invoke adaptation when a feature is a by-product of another adaptation.(Dennett 1995:247)
Recall that Dennett insists that there are additional requirements for making the case for adaptation when those things to be explained are behaviors or capacities exhibited by “nonstupid species,” in particular the need to rule out equally viable explanations such as the discovery of “a good trick.” Moreover, in the human case, Dennett concurs with Gould and Richard Dawkins (who don’t themselves agree on much) that further requirements need to be met. As Dennett makes this argument, “the very considerations that in other parts of the biosphere count for an explanation in terms of natural selection -- manifest utility, obvious value, undeniable reasonableness of design -- count against the need for any such explanation in the case of human behavior” (1995:487-8, emphasis in original, cf. Gould 1980, Dawkins 1989). To make a plausible case that a trait or capacity enjoying these virtues is an adaptation requires ruling out co-discovery of a good trick, information transmission, and the like.
Critics of reverse engineering point to other problems Dennett does not address (or address adequately). These include: that assuming the ubiquitousness of adaptations threatens to render the hypothesis of natural selection circular; that explanations that assume a trait is an adaptation cannot rule out alternative explanations; and that specifying a priori or a posteriori the conditions that render a trait “optimal” or adaptive is largely “a test of the ingenuity of theorists ...” (Lewontin 1984: 242; cf. Maynard Smith 1984).18 It is the second and third criticisms that are of most relevance here. That an adaptation explanation is tested, as Dennett calls for, is not sufficient to overcome them. As his own arguments suggest, the evidential warrant of a hypothesis is in part a function of the status of rival hypotheses, not just its testability or confirmation.
Let me clarify the position on reverse engineering that informs the balance of my discussion. If we understand natural selection as an algorithmic process that builds as well as sorts and winnows, and I think Dawkins and Dennett are right in characterizing it this way, then although the explanations that result are not as solid as those arrived at by historical methods, reverse engineering is an appropriate method for generating adaptation explanations. But all such explanations are subject to the norm of explanatory power, with the objects being explained in part determining the kinds of alternative explanation that need to be ruled out. In the case of physical traits and capacities, these include conversion of function, byproducts of adaptations, mutation, migration, and so forth. In the case of features of human psychology, the list of viable alternatives is much larger.
It is time to bring these several lines of argument to bear on evolutionary psychology. I begin with what evolutionary psychologists claim to know about “the Pleistocene way of life” in which, they contend, a host of cognitive mechanisms and predispositions were selected for. A close look reveals that the general outline of this “history” is a reconstruction of the ancestral conditions and selection pressures that led to human intelligence and language first proposed by “Man, the hunter” theorists of the 1950s and 1960s, and subsequently expanded upon by human sociobiologists.19 According to this theory, the relevant ancestral populations were “Pleistocene hunter-gatherers” and “the hunting adaptation” provided the selection pressures propelling the emergence of a large and complex brain, language, and social organization.
It is not the emphasis on the role of hunting in propelling the directions of human evolution that is of interest here (for despite valiant efforts to resuscitate it, it remains controversial if not largely discredited) but the accounts of “Pleistocene life” that anthropologists proposed and evolutionary psychologists continue to assume and develop. In interpreting fossils and artifacts, and in drawing inferences about ancestral behavior and forms of social organization, architects of the theory relied in no small measure on then-current ethnographic studies of contemporary hunter-gatherers and then-current accounts of social behavior exhibited by some nonhuman primates. They relied, in short, on reverse engineering -- using accounts of behavior in these present groups to draw inferences about the relevant ancestral population.
As previously noted, we are not in a position to determine “the” ancestral population, “its” ecological setting, and so forth, in which uniquely human traits and capacities emerged. Indeed, although evolutionary psychologists’ explanations assume a relatively sudden emergence of human cognitive capacities, there is not consensus within evolutionary theorizing as to whether human cognitive capacities emerged gradually or suddenly, or emerged earlier than artifacts would suggest and remained latent until provoked by a change in some aspect of the environment. Even setting these considerations aside (and although one would not know it reading most accounts evolutionary psychologists provide of “the Pleistocene way of life”20) the history that resulted has been and remains controversial. Its central features were subjected to substantial critique in the 1970s and 1980s, and on precisely the grounds critics of reverse engineering cite. Anthropologists and biologists demonstrated that there was insufficient evidence to warrant the emphasis on hunting by identifying alternative selection pressures as likely to have been implicated in the emergence of distinctively human characteristics. They also demonstrated that there was insufficient evidence to support the reconstructions of social organization among Pleistocene hunter-gatherers, including assumptions about divisions of labor by gender, by identifying alternative models equally commensurate with available evidence by way of artifacts and fossils. Together with primatologists critical of the theory, they pointed out that neither contemporary hunter-gatherers nor nonhuman primates constitute our ancestral population.21
Other criticisms of this reconstruction included charges that the ethnographic accounts of contemporary hunter-gatherer groups were ethnocentric and androcentric, and that anthropomorphism and androcentrism characterized the models of primate behavior appealed to. Likewise, it is worth noting the emergence in the 1990s of critiques in archaeology and paleobiology that challenged traditional interpretations of fossils and artifacts, and traditional hypotheses about prehistoric social life, as unwarranted by available evidence and often androcentric.22 For one need not take a side in any of these debates to acknowledge that the history to which evolutionary psychologists appeal is controversial, and far more so than they typically acknowledge.
I next turn to specific hypotheses advanced in evolutionary psychology that are representative in their reliance on reverse engineering. The first hypothesis, advanced by David M. Buss and others, proposes that women are endowed with an evolved psychological mechanism to prefer men with good economic prospects (Buss 1999; cf. Symons 1979). As I later explore, evolutionary psychologists often appeal to “parental investment theory” to predict sex-differentiated predispositions towards mating and parenting. They also predict sex-differentiated cognitive mechanisms based on what they claim to know was an adaptive problem our ancestors faced, knowledge in turn partly based on studies of contemporary hunter/gatherer groups. This method is one of two the editors of The Adapted Mind advocate. 23
By combining data from paleontology and hunter-gatherer studies with principles drawn from evolutionary biology, one can develop a task analysis that defines the nature of the adaptive information-processing problem to be solved ... . Once one understands the nature of the problem, one can then generate very specific, empirically testable hypotheses about the structure of the information-processing mechanisms that evolved to solve it.(Cosmides et al. 1992:11)
Cosmides et al. maintain that because this method generates testable hypotheses, “it is immune to the usual (but often vacuous) accusation of post hoc storytelling” (1992:11). That is, the abilities to predict a cognitive mechanism and to test that prediction are supposed to insure that the hypothesis in question is not a “Just So Story.” But neither insures that the hypothesis in question meets the norm of “explanatory power,” that is, that it is more viable than rival explanations, or that it is warranted by available evidence. Finally, the researcher who uses this method to identify an adaptive problem is again relying on a “history,” including paleontological and archaeological hypotheses concerning fossils and artifacts, that is itself largely the product of reverse engineering.
One argument Buss advances for the hypothesis outlined above concerning women’s mating preferences is representative in these several respects, and instructive in others.24 It is representative in that Buss maintains that the prediction of this preference is derived from our knowledge of an adaptive problem Pleistocene females faced. They needed to identify a male able to invest in them and their offspring, given that males would have varied in terms of their access to resources and that females would have been burdened with frequent pregnancies, breast feeding, and child rearing, and thus could not sustain themselves and offspring by gathering (77). This “adaptive problem” is identified using ethnographic studies of contemporary hunter/gatherer groups (79-80; 110-111). Finally, Buss reports that numerous cross-cultural studies confirm that contemporary women are “the descendants” of female ancestors endowed with an evolved psychological mechanism to prefer males with resources, a mechanism they in turn have inherited.
Buss’s argument is instructive because, of course, the “adaptive problems” surrounding mating and parenting constitute only a subset of those impacting on inclusive fitness.25 But evolutionary psychologists, like human sociobiologists before them, devote a great deal of attention to identifying and/or explaining alleged sex-differentiated predispositions to mating and parenting strategies. 26 In addition, many hypotheses they advance in this arena, including that on which we are focusing, were mainstays of human sociobiology. What’s “new” are the notions of “evolved psychological mechanisms” (and predispositions) that evolutionary psychologists borrow from cognitive science and graft onto the hypotheses in question. Finally, as I next explore, in no other domain is evolutionary psychologists’ reliance on reverse engineering or the problems that result more obvious.
For the sake of argument, let us grant the account of present hunter/gatherers Buss offers. Let us also set aside one obvious way in which his reasoning is circular: his use of contemporary hunter/gatherers to identify ancestral problems on the basis of which he allegedly predicts that contemporary women will have inherited specific psychological mechanisms. Setting these things aside will allow us to focus on a far more serious and general problem with adaptation explanations arrived at using reverse engineering. What is the nature of the evidence for the hypothesis that identifying a mate with resources was an adaptive problem facing our female ancestors? It cannot be the cross-cultural surveys of preferences among contemporary women Buss undertakes and cites. These are described as testing a prediction derived from the prior identification of the historical adaptive problem. Can the adaptive problem be derived from behaviors attributed to contemporary hunter/gatherers? Again, although it requires that we assume that observing these groups is akin to time travel, let us grant that these groups afford insights into our evolutionary history and, although controversial, let us also grant that they exhibit the behaviors Buss claims. Let us further grant that the case has been made for the historical adaptive problem earlier outlined. Then we need to ask what grounds there are for believing that an adaptation in the form of a psychological mechanism was its solution. After all, there is an obvious rational explanation for this preference: given that men control the vast majority of resources in most if not all cultures, the preference is highly rational if not what Dennett calls “a forced move.” Indeed, freshman in an introductory course on Darwin identified this alternative in about a minute. All of this reveals that if the reasoning here isn’t outright question begging, it is something close to it. To accept this explanation, we must begin by assuming that an adaptation was the solution, an assumption that is not warranted and, indeed, was that which was to be tested and confirmed.
A more defensible hypothesis from the perspective of evolutionary theory is Lena Cosmides and John Tooby’s hypothesis positing that humans are endowed with an innate cognitive mechanism for detecting cheaters (Cosmides and Tooby 1992).27 Cosmides and Tooby propose the mechanism to explain an apparently common ability by test subjects to solve a logic problem couched in terms that involve detecting a cheater, and an equally common inability to solve the same puzzle presented in abstract mathematical terms. Such a mechanism, they maintain, would have been selected for because of its obvious benefits in maintaining social contracts. Their reliance on reverse engineering is obvious; the psychological mechanism is inferred from inconsistencies in logical reasoning, and the “adaptive problem” it is taken to have solved is inferred from reconstructions of Pleistocene life earlier discussed. This methodological approach is the second the editors of The Adapted Mind describe as central to evolutionary psychology.
... researchers can start with a known psychological phenomenon, and begin to investigate its adaptive function, if any, by placing it in the context of hunter-gatherer life and known selection pressures ... to try to understand what its adaptive function was -- why that design was selected for rather than alternative ones.( Cosmides et al. 1992:10)
For reasons to be explored in the next section, it is not the commonness of the inconsistency in logical reasoning that warrants the consideration of an evolutionary explanation. It is because this phenomenon might be characterized as “irrational” that it can plausibly be considered to fall into the category Francis Crick called “frozen accidents” and Dennett, alluding to the arrangement of keys on a keyboard, calls “QWERTY” phenomena -- phenomena whose present maladaptive nature suggests historical, and in this case evolutionary, origin. There are all too many viable explanations for “rational” aspects of psychology and behavior.
Nevertheless, Cosmides and Tooby’s hypothesis suffers from the general problem to which adaptation explanations arrived at by reverse engineering are prone. There are equally plausible alternative explanations for the phenomenon in question. These include: that humans are generally less adept at solving logic puzzles presented in mathematical terms than they are at those presented in applied contexts, that they are less adept at solving logic puzzles couched in unfamiliar terms than familiar, that they often suffer from math phobia, and so forth. Indeed there is a deeper problem. Unless we assume that human cognitive capacities are the product of a process capable of producing “perfect” artifacts (and natural selection is no such thing), the fallibility involved and typically overcome in a basic logic course, in no way calls for a specific adaptation explanation. Finally, as any student in a basic course in evolution knows, two requirements for natural selection are variation and heritability. No case has been made for either.
The status of sexual selection as an evolutionary mechanism remains a matter of ongoing investigation and debate in evolutionary biology. Some view sexual selection, as Darwin viewed it, as an evolutionary mechanism distinct from natural selection (e.g., Dawkins 1989); others as a useful way to think about processes that are, in the end, subcategories of natural selection (e.g., Bateman 1948); and some maintain that if it is a distinct mechanism, its role is “very meager” (Mayr 1972).
Nor do the disagreements end here. One of Alfred Russell Wallace’s objections to sexual selection was the notion of “female choice,” which together with intrasexual selection constituted sexual selection as Darwin defined it. As noted in a recent survey of the relevant literature, disagreements remain about intersexual selection (Spencer and Master 1992). Some concern the methodological difficulties in demonstrating mate choice. Critics maintain that many experiments and field studies claiming to demonstrate female choice do not succeed in eliminating alternative explanations (e.g., Lambert et al. 1982). To cite just one example, it was long assumed that females of lek species tend to mate with males at the center of the lek because males so located are “superior.” Partridge and Halliday (1984) among others suggest an alternative explanation: that the central area of the lek provides more protection from predators during mating. The general criticism of explanations appealing to intersexual selection is familiar; too often, critics argue, alternative explanations are not considered or ruled out.
Against this background, it might seem unnecessary to undertake an analysis of parental investment theory, a cluster of hypotheses its proponents claim to derive from sexual selection. But parental investment theory figures prominently in evolutionary psychology, and both it and the way it is used in the discipline illustrate all of the problems so far identified. Arrived at by reverse engineering, circular reasoning marks the arguments for the theory and the uses made of it. Although it has been subjected to devastating critique in the last three decades, evolutionary psychologists present it as a “mainstream” theory of evolutionary biology that enjoys extensive confirmation. Finally, even its most sophisticated formulation, offered by Dawkins and considered below, is deeply informed by sociocultural assumptions. I begin by briefly summarizing its history.
Parental investment theory traces its origins to some 64 experiments on fruit flies undertaken by geneticist Angus John Bateman in 1948. Bateman took what he found to be a much greater variance in male reproductive success to be significant and to entail “a nearly universal dichotomy in the sexual nature of the male and female … a combination of an undiscriminating eagerness in the males and a discriminating passivity in the females,” including human males and females (1948:365).
In an article that became the second most-cited article in human sociobiology, R. L. Trivers (1972) introduced what is now called “parental investment theory.” Describing Bateman’s paper as his “key reference,” Trivers proposed that the notion of “parental investment” explains Bateman’s results. Gametic dimorphism alone, he argued, dictates that sperm are “cheaper” than eggs, and because female investment in offspring typically extends to gestation or egg-sitting, nursing in mammals, and so forth, it is significantly higher than male investment. This differential investment explains the asymmetry in breeding potential Bateman found and predicts the differences in reproductive strategies (the “undiscriminating” males and “coy” females) traditionally “observed.”
As evolutionary psychologists like to point out, parental investment theory is predictive and thus, they claim, avoids the charge of a priori reasoning or ad hoc theorizing. I turn shortly to the question of its predictive success. But it should be clear that the arguments just summarized rely on reverse engineering -- Bateman’s from experimental results on fruit flies, Trivers’ from those results supplemented by the notion of “parental investment.”
Evolutionary psychologists rely heavily on parental investment theory to derive hypotheses predicting sex-differentiated adaptive problems and predispositions. The hypothesis earlier considered, that women have an evolved predisposition to prefer mates likely to be good providers, is but one of a number of hypotheses claimed to follow from parental investment theory. 28 Men are predicted to be predisposed to engage in behaviors (including coercive behaviors) to insure paternity, to be predisposed to value fertility in prospective mates more than women do, and to be endowed with cognitive mechanisms that allow them to recognize fertility, to name just a few.29 These predictions do follow from Trivers’ hypotheses when these are supplemented with the notions of “predispositions” and “evolved psychological mechanisms.” But like the reconstruction of Pleistocene life on which evolutionary psychologists rely, Trivers’ hypotheses have been subjected to substantive critiques, most leveled by other biologists. Even a brief summary of these arguments suffices to show that the theory lacks the status evolutionary psychologists claim.
Trivers’ core assumption about the relative cheapness of sperm has been challenged on several grounds. These include that males produce millions of sperm for each egg a female produces and, in many species, high sperm production is required for successful insemination; that in some species, high sperm production is correlated with shorter life-span; and that males of a number of species produce costly accessory secretions.30 Hence, critics argue, a more accurate comparison of parental investment would invoke the minimum number of ejaculations required for fertilization and the cost of a single egg (e.g., Lanier et al. 1975, Dewsbury 1982, Tang-Martinez 2000).
Nor does field research bear out the theory’s predictions. Again the literature documenting counterexamples is extensive. An extensive survey article published in 1986 by Sandra Hrdy detailed numerous counter-examples to the “coy female” hypothesis that demonstrated that females of a number of species seek to mate more than once or twice, including when they are not ovulating and/or are pregnant (Hrdy 1986).31 Indeed, Hrdy reported that “no fewer than six different models to explain how females might benefit from mating with different males” had been proposed to explain reports dating back to 1975 of polyandrous mating (1986:127). In short, for more than 25 years there have been abundant counterexamples to the “coy female” hypothesis predicted by parental investment theory. Finally, this hypothesis seems straightforwardly incompatible with the arguments evolutionary psychologists and others advance for “sperm competition,” a third type of sexual selection purported to select traits to insure paternity in light of polyandrous behavior.
Trivers and other advocates of the theory also take differential investment in offspring to entail that the interests of the sexes are inherently conflicting (gleefully described as “the battle of the sexes” in publications intended for a lay public): that males are inherently interested in acquiring as many mates as possible, females inherently interested in acquiring a mate who will provide consistent support for them and their offspring. But, critics argue, given that sexual reproduction is the primary function of courtship in sexually reproducing species, an inherent conflict between the sexes is highly unlikely (e.g., Tang-Martinez 2000). As Spencer and Masters (1992) make the argument, “the male--female communication system is subject to strong stabilizing selection.” Thus, “unusual or fussy individuals” (an allusion to the “coy females” predicted) would be at a distinct disadvantage, either because they would be more likely to be rejected or more likely to reject suitable mates (1992:301).
Finally, critics point to obvious cases of circular reasoning in applications of the theory. In one well-known example, Trivers notes that in dung flies, “the male who first leaps on top of a newly arrived female copulates with her.” But the lack of female choice in this case does not necessarily challenge parental investment theory, Trivers maintains, for it may “result from the prima facie case the first male establishes for his sound reproductive abilities” (Trivers 1972: 170). As Spencer and Masters note, in this passage “the theory is quite obviously untestable; even when female choice is entirely absent, it is possible to explain the phenomenon in terms of female choice” (1992: 296). Reflecting these and the other lines of argument just summarized, they conclude that
Uncritical acceptance of [sexual selection’s] ubiquitous occurrence (for example, in Trivers 1972) does no service to evolutionary theory ... heritable variance in reproductive success, based on characters that are not favored by natural selection, must be demonstrable before sexual selection may be invoked unequivocally.(Spencer and Masters 1972:301, emphasis added)
Here Spencer and Masters allude to Darwin’s argument for sexual selection and it is instructive to return to his reasoning.
Recall that Darwin proposed the mechanism of sexual selection to explain phenomena that he thought natural selection could not explain: traits that seemed to provide no advantage in terms of survival and might work against it. Darwin did not assume that every trait was selected for; but he did view the preponderance of such traits as calling for some sort of explanation (if for no other reason than that they might constitute counterexamples to the hypothesis of evolution by natural selection). In other words, the traits to be explained by sexual selection were those that, if one only assumed natural selection, seemed irrational.
Add the components of sexual selection, however, and they no longer seem so. But this does not mean that in order to successfully use sexual selection in explaining some behavior, all that must be established is its utility. It has precisely the opposite effect. Sexual selection sets severe limits on the traits to be explained on its basis: to exclude behaviors or capacities that might be “good tricks” discovered by a nonstupid species and traits natural selection can explain, and to include only those for which there is evidence of heritable variation in reproductive success.
There are, then, a number of serious challenges to Trivers’ theory. Indeed, no less a champion of it than Dawkins in the first edition of The Selfish Gene acknowledges in the second edition that the assumption that sperm are cheap relative to eggs is deeply problematic. But Dawkins remains convinced that there is “a fundamental asymmetry in sex roles” and uses an extensive endnote annotating the original text to propose an alternative explanation (Dawkins 1989:300-1).
It is worth considering. For one thing, Dawkins maintains that the separation of sperm and egg is itself the product of a more fundamental asymmetry in the sexes that closely parallels that advocated by evolutionary psychologists. (So although their appeals to parental investment theory are dated, Dawkins’ explanation might provide the alternative evolutionary psychologists need.) In addition, Dawkins’ explanation is arrived at by reverse engineering and suffers from the problems to which such explanations are prone. Finally, although more sophisticated than some hypotheses we have encountered, Dawkins’ explanation is deeply informed by gender stereotypes.
I quote his explanation (1989:300-1) in its entirety, interspersing comments that point to what I take to be significant problems. Dawkins begins by asking us to “Suppose that we start with two sexes that have none of the particular attributes of males and females. Call them by the neutral names A and B.” Assuming that the As and Bs have none of the attributes of males and females -- and we do need to assume this given that Dawkins is offering an explanation of them -- they do not have sperm or eggs. In what sense, then, are these organisms sexes? In the body of this chapter, Dawkins maintains the standard definition of the sexes, noting that “one fundamental feature ... used to label males as males, and females as females, throughout animals and plants” is that male gametes are smaller and more numerous than female gametes (1989:141). So, at this point the As and Bs are at most proto sexes.
Now, any animal, whether an A or a B, faces a trade-off. Time and effort devoted to fighting with rivals cannot be spent on rearing existing offspring, and vice versa. Any animal can be expected to balance its effort between these rival claims.
Perhaps. But other rival claims are equally plausible. “Sleep now or eat?” “Look for a mate or groom?” “Look for prey or fight with conspecifics over food or territory?” Or, if the protosexed organisms (which have somehow graduated to animals) were, as is likely, too simple to have such options, then “Drift here or there?” Of course if they were this simple, than fighting rivals and parenting were not likely among their behavioral options.
Dawkins next introduces the possibility that the groups “may” have come to “differ.”
The point I am about to come to is that the As may settle at a different balance from the Bs and that, once they do, there is likely to be an escalating disparity between them.
To see this, suppose that the two sexes, the As and the Bs, differ from one another, right from the start, in whether they can most influence their success by investing in children or by investing in fighting. (I’ll use ‘fighting’ to stand for all kinds of direct competition within one sex.)
Again, why suppose that the differences that “may” obtain between groups of organisms fall on these axes as opposed to others? And why at this point is “fighting” restricted to intersexual competition -- or, if we grant the reasonableness of so restricting it, why assume it wouldn’t have been in the best interest of all of these organisms?
Initially the difference between the sexes can be very slight, since my point will be that there is an inherent tendency for it to grow.
Say the As start out with fighting making a greater contribution to their reproductive success than parental behavior does; the Bs on the other hand start out with parental behavior contributing slightly more than fighting to variation in their reproductive success. This means, for example, that although an A of course benefits from parental care, the difference between a successful carer and an unsuccessful carer among the As is smaller than the difference between a successful fighter and an unsuccessful fighter among the As. Among the Bs, just the reverse is true. So, for a given amount of effort, an A can do itself good by fighting, whereas a B is more likely to do itself good by shifting its effort away from fighting and towards parental care.
I don’t know whether to attach any significance to it, but there are subtle differences in Dawkins’ wording here. For As, fighting “makes a greater contribution” to their reproductive success than parental behavior; for Bs, parental behavior “contributes slightly more” than fighting to such success. An A “can do itself good” by fighting; a B is “more likely” to do itself good. Perhaps Dawkins shares my suspicion that in terms of simple and as yet unsexed organisms, it is less plausible to attribute “parenting” than it is “competing.” The latter, after all, might be accomplished by drifting faster or sideways.
In any event, Dawkins maintains that if we grant his “starting” assumptions, we are well on our way to an explanation of both a fundamental asymmetry in sex roles and gametic dimorphism.
In subsequent generations, therefore, the As will fight a bit more than their parents, the Bs will fight a bit less and care a bit more than their parents. Now the difference between the best A and the worst A with respect to fighting will be even greater, the difference between the best A and the worst A with respect to caring will be even less. Therefore an A has even more to gain by putting its effort into fighting, even less to gain by putting its effort into caring. Exactly the opposite will be true of the Bs as the generations go by.
The key idea here is that a small initial difference between the sexes [sic] can be self-enhancing: selection can start with an initial, slight difference and make it grow larger and larger, until the As become what we now call males, the Bs what we now call females.
Dawkins concludes by noting that his explanation of asymmetrical sex roles reverses the common trajectory of such explanations. The separation of sperm and egg does not explain all the “characteristics” of males and females. Instead, an initial division between organisms that gained from parenting and those that gained from competing explains both it and such characteristics.
There is nothing wrong with the logic of Dawkins’ argument. But I suggest that his explanation is either circular or a “Just So Story,” the central assumptions of which we have no reason to accept. Questions and comments interjected along the way reflect both worries. Although Dawkins does not assume gametic dimorphism, he does assume that for some protosexed organisms, the available and good reproductive strategies were fighting rivals and parenting. In the initial stages of the argument, he is careful to couch what he says about “initial” and “slight” differences in terms of possibilities. But differences in reproductive strategies, let alone of these kinds, too closely mirror what is to be explained.
Dawkins also maintains that, in relation to one another, each of these reproductive strategies would have undergone runaway selection. This may hold in the cases in which all the Bs and all the As settled, respectively, on parenting and fighting. But as Dawkins notes in the same work, field research reveals four combinations of reproductive strategies, and computer simulations indicate that eachcan attain stability (i.e., constitute an evolutionary stable strategy): males desert (fight rather than parent), females care (parent); males care, females desert; males care, females care; and males desert, females desert. Stability can also be reached in populations that display either the pair “male deserts/female parents” and “female deserts/male parents,” or the pair “both desert” and “both parent” (Dawkins 1989:302). But given these results, the situation that we are asked to suppose started the ball rolling toward sex and “asymmetrical sex roles” is again indistinguishable from what Dawkins claims to explain. It is just as plausible that all the organisms with which the story begins would have settled on “fight” or settled on “parent.”
Consider the alternative: that Dawkins’ explanation does not presuppose what it was meant to explain. Then we have been given no reason to assume important features, however abstract, of the hypothetical situation he describes -- including an initial difference between two groups of unsexed organisms that involved tendencies toward “competing” and “parenting.” The concluding sentences of Dawkins’ argument are these: “The initial difference can be small enough to arise at random. After all, the starting conditions of the two sexes are unlikely to be exactly identical” (1989: 301). They reflect, respectively, a “Just So Story” whose details aren’t argued for, and circularity.
I have argued that Dawkins’ explanation of asymmetrical sex roles presupposes rather than explains them, an all too common problem in the domain of sex differences.32 But there is an irony worth noting in concluding this section. Although Dawkins notes that the stereotypes of coy females and promiscuous males may “strike a chord” when we think of human behavior, he also maintains that in this and other cases there are significant difficulties in applying the principles involved to humans (Dawkins 1989:164). It is in this work that he maintains that “we must begin by throwing out the gene as the sole basis of our ideas on evolution” for understanding human evolution and modern humans, and that he introduces the notion of “memes” as units of “cultural transmission.,” The notion of memes, he contends, is a more appropriate explanatory notion than genes in the study of human behavior (Dawkins 1989: 189-201). Publications in evolutionary psychology are replete with appeals to sociobiology, including to Dawkins’ work, but I have yet to find this argument positively cited.
I have argued in this and the previous section that although reverse engineering is an appropriate method for deriving adaptation explanations, the explanations making use of this method that are being advanced in evolutionary psychology and human sociobiology fail to meet the standards advocates of the method call for and illustrate the problems its critics cite. I have not yet considered whether social scientists and psychologists make even a minimally plausible case for the cognitive mechanisms they propose and/or the “respectability” of these purported mechanisms as “scientific objects.” I now turn to these questions.
Cognitive anthropologist Scott Atran proposes that universal features of folk taxonomies for living things, features found across cultures and throughout history, suggest that humans are endowed with a cognitive module aiding in the classification of, and thus inductive reasoning about, plants and animals (Atran 1995). Atran’s first line of argument is from universality. He appeals to anthropological research that suggests several relevant universals: first, that “virtually all humans in all cultures” divide living things using the categories “plants” and “non-human animals”; second, that whatever the particular constitution of a culture’s life form groups or taxa, “the life form level or rank” divides living things in hierarchical and mutually exclusive partitions; and, finally, that plants and animals are perceived to be “natural kinds,” whose “essence” determines their morphology and behavioral characteristics (1995:206-8). Atran also cites a psychological study reporting that infants pay more attention to animate than inanimate objects and can in their first year “distinguish plastic representations of animals from plastic representations of all other things” (1995:206). In a third argument, Atran appeals to a study comparing the folk taxonomy of the Itza of Guatemala with that of university students raised in rural Michigan. This found cases in which the two folk taxonomies would group animals (e.g., felines and canines), or classify individual species, in the same way and not in keeping with scientific classification. Atran takes the study to suggest that “there are at least some universal cognitive factors at work in folk-biological classification that are mitigated or ignored by science” (1995:212).
On the basis of these arguments, Atran proposes “an innate living-kind module”: an innately driven “research program” that compels people “to deepen and extend the domain of information relevant to living kinds into an all-embracing taxonomy” (Atran 1995:220). As the tendency to classify objects as “natural kinds” does not extend to inanimate or inert objects, Atran contends that it is plausibly explained as an adaptation, noting the evolutionary advantage to “acquiring ‘automatic’ competency” in understanding living species and being able to reason inductively about them (1995:222).
Presumably, Atran views the research devoted to infants’ attention spans as ruling out information transmission and the like as plausible explanations of the purported universals in folk taxonomies (although it seems quite a leap to infer an innate research program on its basis). Likewise, Atran may take the assumption of “natural kinds” characteristic of folk taxonomies to be just the kind of irrational phenomenon lending itself to an adaptation explanation. Finally, the innate research program he proposes would certainly be advantageous in the ways he proposes. But although Atran may have succeeded in making a case that the innate research program he posits fits aspects of infant and adult behavior, he has not made the case that any such program guides it.34
Insofar as it is possible to identify an alternative research program that equally well fits the behavior in question, no case has been made that the program proposed guides it.35 We can use examples from Quine’s argument for the indeterminacy (or inscrutability) of reference to illustrate the problem. An innate research program that takes “kinds of event” rather than “kinds of living thing” as its basic ontological category is extensionally equivalent to that which Atran proposes, as is a research program whose basic ontological categories are Platonic forms and instantiations of them. Thus, there is no evidence to warrant the conclusion that one of these “unconscious programs” is guiding the behavior in question. Moreover, if Quine is correct about the indeterminacy of reference, arguments for cross-cultural universals in taxonomies are nonstarters. There is no “fact of the matter” in terms of the ontologies we or anyone else is committed to (Quine 1969).
To return to the general argument of this section, the need to demonstrate that a proposed cognitive algorithm or predisposition does more than fit some aspect of human behavior carries serious implications for evolutionary psychology. A core methodological assumption of this research program is that the universal human nature it posits exists at the level of “evolved psychological mechanisms,” not of (expressed) cultural behaviors (Cosmides et al. 1992:5). Of course it would be remiss not to note that, as we have seen, evolutionary psychologists do appeal to “expressed behaviors” in their attempts to devise evolutionary explanations.36 But given their claims to be identifying a human nature existing at the level of psychology, the rationale of this discipline depends on the ability to demonstrate guidance, not mere compatibility. The warrant for any posited cognitive mechanism consists, in part, of there being no alternative algorithm or predisposition that equally fits the behavior in question.37 This is a high bar. Reverse engineering cannot catapult evolutionary psychology over the explanatory hurdles with allusions to the technical notions of “cognitive modules” and “adaptation.”
Citing human sociobiology’s empirical failings and significant social implications, Kitcher quotes the narrator at the conclusion of The Great Gatsby as noting that “carelessness that results in the destruction or diminution of human life is unforgivable” (1985:10). There is no excuse for failing to meet standard norms such as explanatory power, for ignoring relevant and obvious counterevidence, and/or for misrepresenting the status of the hypotheses to which one appeals. Science so characterized just is bad science. When the hypotheses generated carry significant implications for social policy and/or human self-understanding and aspirations, as do many of those advanced in evolutionary psychology, such research is appallingly irresponsible.
1 Jack Nelson and Paul Roth provided extensive feedback on earlier versions of this paper, and Robert Richardson also provided helpful feedback. Earlier versions were presented to the STA Colloquium at Washington University and the Philosophy Colloquium at the University of Washington. I am grateful to members of these audiences for helpful comments and criticisms, particularly Garland Allen, Ken Clatterbaugh, Arthur Fine, Andrea Woody, and Alison Wylie. Parts of sections four and six were written jointly with Jack Nelson and presented at the PNP Colloquium at Washington University (Nelson and Nelson 2000a). Again, thanks to members of the audience for their comments and criticisms.
2 Contributors to Barkow et al. (1992), Buss (1999), and Schubert (1989), offer arguments charging critics with one or more of these motivations.
3 For example, the editors of Sex, Power, Conflict: Evolutionary and Feminist Perspectives, who are evolutionary psychologists, insist there is a clear distinction between the goals, rationale, and methods of evolutionary psychology, on the one hand, and those of feminism (more correctly, what they characterize as feminism) on the other. Predictably, they claim that evolutionary psychology is concerned with “how things are,” and “feminism” with “how things should be” (Buss and Malamuth 1996:3-8). They fail to acknowledge that many feminist critiques of evolutionary psychology are offered by scientists, indeed biologists, and emphasize lack of evidential warrant for hypotheses being advanced in the discipline.
4 Kitcher’s argument, outlined in my concluding section, occurs in his critique of human sociobiology in Vaulting Ambition (Kitcher 1985).
5 I assume a holistic account of evidence, according to which the evidence for any specific hypothesis or claim includes both the observation consequences of that hypothesis or claim (together with the larger theory or theories within which it is embedded) and the relationship of the hypothesis or claim in question to other accepted theories (Nelson 1996).
6 I use scare quotes for “new” because I will later contend that in the case of evolutionary psychology’s approach to mating and parenting strategies, the hypotheses advanced are well-worn hypotheses of human sociobiology to which evolutionary psychologists simply add technical notions they borrow from cognitive science.
7 I leave it to those better versed in recent social science literature to respond to claims that older social science paradigms are dead and/or that integrating evolutionary biology into their explanations is the only way for the social sciences to achieve respectability (as argued for in Brown 1991, Cosmides et al. 1992, Rosenberg 1980, and Schubert 1989, among other places). Sometimes the accounts given by advocates of these programs of the “alternatives” available to social scientists are unduly limited, as is Cosmides et al.’s description of “The Standard Social Science Model” in chapter 1 of The Adapted Mind (1992). And sometimes, the accounts of available alternatives are ridiculous. For example, Buss (1999) cites “seeding theory” and “creationism” as the only alternatives to an evolutionary approach to human psychology and behavior.
8 I am grateful to Ken Clatterbaugh, Arthur Fine, and Andrea Woody whose comments on an earlier version of this paper made clear the need to make this argument explicit.
9 A number of research programs in the social sciences seek or claim to incorporate evolutionary theory in their explanations of human psychology and culture, and there are differences in the nature of the relationship they envision with evolutionary biology. (Compare, e.g., Brown 1991 with Schubert 1989.) I emphasize evolutionary psychology for several reasons. It is the most visible of these programs, outside as well as within academia and I confess to having no patience with the practice of writing off publications aimed at the lay public as “howlers” that are not worthy of serious attention. Such was the response two decades ago to the critiques feminist scientists leveled at human sociobiology; feminists, critics charged, were focusing on an obvious case of bad science that no one took seriously. But its closest descendant, evolutionary psychology, propounds many of the same hypotheses -- albeit dressed up in the garb of technical notions from cognitive science. As discussed in my concluding section, many of the hypotheses advanced to a lay audience carry significant social implications and thus do warrant serious attention by philosophers of science. As much to the point, the goals, research questions, and methodological/metaphysical assumptions of evolutionary psychology are influential in other social sciences. This is reflected in the use of “evolutionary psychology” by other social scientists to describe their research, and in their appeals to notions such as “cognitive modules” and “evolved psychological mechanisms.” Used here, the phrase denotes the narrower research program but many of the arguments I advance apply more broadly.
10 As explored below, the methods used to arrive at evolutionary explanations also differ because some are more feasible in explaining cases involving microevolution than they are in cases of macroevolution, and because they make use of different kinds of historical information.
11 There are of course disagreements about the “units of selection,” and a number of arguments that natural selection only works at the level of organisms, not that of genes (see, e.g., Dawkins 1979, Mayr 1988: ch. 8; Ruse 1988: ch. 4 provides a good overview of these disagreements). I emphasize its role in selecting for traits because this is the emphasis of the research I consider.
12 Arguments representative of the views outlined in this paragraph include Brandon (1984), Burian (1983, 1992), Gould and Lewontin (1979), Maynard Smith (1984), Mayr (1983, 1988), Richardson (2001), and West-Ebelard (1992).
13 Cf. Brandon (1984), Mayr (1988), and West-Ebelard (1992). Ernst Mayr notes that the majority of evolutionary biologists now recognize that many phenotypic traits are by-products of evolutionary processes and are “quite cautious” in what they attribute to natural selection (Mayr 1988:136).
14 Dennett uses “a good trick” to denote “a behavioral talent that protects [an organism] or enhances its chances dramatically” (1995:77-78); and, borrowing from chess, “a forced move” to denote an adjustment in behavior to achieve “the one and only one solution to staving off disaster” (1995:128).
15 Here I diverge from Richardson’s approach, which takes evolutionary psychology as using historical methods but not successfully.
16 Cosmides et al. note that:
the central premise of The Adapted Mind is that there is a universal human nature ... that exists primarily at the level of evolved psychological mechanisms, not of expressed cultural behaviors... . A second premise is that these evolved psychological mechanisms are adaptations, constructed by natural selection over evolutionary time ... . A third assumption ... is that the evolved structure of the human mind is adapted to the way of life of Pleistocene hunter-gatherers ... .(Cosmides et al. 1992:5)
17 Good discussions of the issues involved include Dennett (1995), Lewontin (1977, 1978), Maynard Smith (1984), and Richardson (2001).
18 Cf. Dupre (1998), Gould and Lewontin (1979), and Richardson (2001).
19 For example, Washburn (1961), Lee and DeVore (1968), Wilson (1978), Tooby and DeVore (1987).
20 In this sense, Buss (1999) is an exception for he does note some of the debates I next mention. But he also contends that the hunting adaptation and “Man, the Hunter” theory provide the most viable explanations of the evolution of language and social organization, and what evolutionary psychologists and human sociobiologists describe as the sexual division of labor.
21 The literature mentioned in this paragraph is extensive. Representative critiques can be found in Bleier (1984), Haraway (1978a, 1978b), and Tanner and Zihlman (1976).
22 See the works cited in n.21 above, Conkey and Spector (1984), and Wylie (1997), for overviews of the critiques cited in this paragraph.
23 In other words, technical terms from cognitive science are taken to yield a “new” causal theory that links evolutionary biology and the social sciences.
24 Buss also offers an argument based on parental investment theory for this hypothesis. I discuss parental investment theory in the next section.
25 As Sarah Hrdy puts this point,
[In the case of debates about sexual selection theory], as always in such debates, reality exists in a plane distinct from that predefined by the debate. In this case, reality is hours and hours, sometimes months and months, of existence where sexual behavior is not even an issue, hours where animals are walking, feeding, resting, grooming.(Hrdy 1986:124)
26 For example, David Barash noted that sociobiology “relies heavily on male-female differences” (Barash 1977:283). Similarly Buss maintains that “if selection has not sculpted psychological mechanisms designed to solve adaptive problems posed by sex and mating, then evolutionary psychology would be ‘out of business’ before it even got off the ground” (Buss 1999:97).
27 I explore below why this hypothesis is more defensible from the perspective of evolutionary theorizing, but there are other compelling criticisms of this hypothesis. See, for example, Fodor (2000).
28 For example, in Buss (1990), Daly and Wilson (1981, 1983, 1990), Symons (1979, 1992). In The Adapted Mind alone (Barkow et al. 1992), 11 of 18 articles draw on one or more aspects of Trivers’ theory of parental investment to derive hypotheses positing mating and/or parenting strategies.
29 See the works by Buss, Daly and Wilson, Wilson and Daly, Symons, and Thornhill and Thornhill in the reference list.
30 Current hypotheses as to why there are so many more sperm relative to eggs do nothing to further the case that male investment is minimal. They include that sperm “compete” with each other; that the relatively large number of sperm reflects the “hostility” of the female reproductive tract; and that females in some species are able to defer fertilization and “select” sperm after numerous copulations. Tang-Martinez (2000) provides an extensive survey of the biological literature challenging parental investment theory.
31 See, for example, Eaton (1978), Hrdy (1986), Gladstone (1979), Hunter et al. (1993), Lanier et al. (1975), Smith (1988), Tang-Martinez (2000), and Van Voorhies (1992).
32 For example, Buss 1996 advocates a claim he attributes to Symons 1992: “To an evolutionary psychologist, the likelihood that the sexes are psychological identical in domains in which they have recurrently confronted different adaptive problems over the long expanse of human history is essentially zero” (301). But of course that the sexes have consistently confronted different adaptive problems is not yielded by historical research. It is either taken to be entailed by parental investment theory or simply assumed and thus circular. Finally, although Buss attributes this claim (which he cites without quotes) to Symons (1992), it is not so stated in this article although it can be taken as an implication of Symons’ arguments.
33 This section builds on an argument advanced in Nelson and Nelson (2000a).
34 Since writing this section, I have read recent publications by Atran that include more nuanced and technical arguments, and back off some of his earlier conclusions about the exact nature of the cognitive mechanism underlying folk taxonomies. But the point of the present discussion -- that demonstrating that a proposed cognitive mechanism fits an aspect of human behavior is not sufficient to demonstrating that it guides it -- applies to this recent work as well.
35 A classic argument for the difficulties in establishing the stronger claim was made by Quine in his critique of Noam Chomsky’s claim that an innate language structure facilitates language acquisition and underlies verbal behavior (Quine 1972). It is one thing, Quine pointed out, to take language at face value and speak of every known language being translatable into a language structured by referential position, bound variables, logical operators, and the like. It is another to claim that this is the only way to translate or parse the language (it is not), and yet a further step to suggest there is evidence that infants use or prefer one set of unconscious rules to another, extensionally equivalent set in understanding and generating linguistic output (Nelson and Nelson 2000a). “What is more than trivial in the new doctrine,” Quine argued, is that “it imputes to the natives an unconscious preference for one system of rules over another, equally unconscious, which is extensionally equivalent to it” (1972: 443).
36 As we saw in the last section, evolutionary psychologists use studies of contemporary hunter-gatherer groups to identify ancestral adaptive problems; they appeal to cross-cultural surveys to support hypotheses positing evolved psychological mechanisms; and they build from findings in psychological research to derive hypotheses positing cognitive mechanisms (e.g., the algorithm for detecting cheaters earlier discussed). Perhaps Cosmides et al. (1992) emphasize that the human nature they posit is at the level of psychological mechanisms to forestall the kinds of counter-examples in “expressed cultural behaviors” that bedeviled human sociobiologists. Evolutionary psychologists also emphasize their focus on psychological mechanisms in arguments that theirs is a “genuinely new discipline,” not (as it might appear) a version of human sociobiology (e.g., Buss 1999:ch. 1).
37 John Dupre offers a similar argument:
With arguments for the existence of [evolved mental modules] almost exclusively a prioristic, and evidence for the behavior they are supposed to generate equivocal at the very best, we should treat this school of genetic determinism [evolutionary psychology] with all the respect that thinking people have come to accord sociobiology.(Dupre 1998:162)
Atran, Scott 1995: Causal constraints on categories and categorical constraints on biological reasoning across cultures. In Dan Sperber, David Premack, and Ann James Premack (eds.), Causal Cognition: A Multidisciplinary Debate. Oxford: Clarendon Press, 205-33.
Barash, David P. 1977: Sociobiology and Behavior. New York: Elsevier North‑Holland.
Barkow, Jerome H., Leda Cosmides, and John Tooby (eds.) 1992: The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York and Oxford: Oxford University Press.
Bateman, Angus John 1948: Intra-sexual selection in Drosophila. Heredity 2, 349-68.
Bleier, Ruth 1984: Science and Gender. Elmsford, NY: Pergamon.
Brandon, Robert N. 1984: Adaptation and evolutionary theory. In Elliott Sober (ed.), The Nature of Selection: Evolutionary Theory in Philosophical Focus. Cambridge, MA: MIT Press, 58-82.
Brandon, Robert N. 1990: Adaptation and Environment. Princeton, NJ: Princeton University Press.
Brown, Donald E. 1991: Human Universals. Philadelphia: Temple University Press.
Burian, Richard M. 1978: A methodological critique of sociobiology. In A.L. Caplan (ed.), The Sociobiology Debate. New York: Harper & Row, 376-95.
Burian, Richard M. 1983: Adaptation. In Marjorie Grene (ed.), Dimensions of Darwinism: Themes and Counterthemes in Twentieth Century Evolutionary Theory. Cambridge, UK and New York: Cambridge University Press, 287-314.
Burian, Richard M. 1992: Adaptation: Historical perspectives. In Evelyn Fox Keller and Elisabeth A. Lloyd (eds.), Keywords in Evolutionary Biology. Cambridge, MA: Harvard University Press, 7-12.
Buss, David 1996: Sexual conflict: Evolutionary insights into feminism and “the battle of the sexes.” In David M. Buss and Neil M. Malamuth (eds.), Sex, Power, Conflict: Evolutionary and Feminist Perspectives. New York and Oxford: Oxford University Press, 296-318.
Buss, David M. 1999: Evolutionary Psychology: The New Science of the Mind. Boston: Allyn and Bacon.
Buss, David M. and Neil M. Malamuth (eds.) 1996: Sex, Power, Conflict: Evolutionary and Feminist Perspectives. New York and Oxford: Oxford University Press.
Conkey, M. W. and J. D. Spector 1984: Archaeology and the study of gender. Advances in Archaeological Method and Theory 7, 1-38.
Cosmides, Leda and John Tooby 1992: Cognitive adaptations for social exchange. In Barkow et al. (eds.), The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York and Oxford: Oxford University Press, 163-228.
Cosmides, Leda, John Tooby, and Jerome H. Barkow 1992: Introduction: Evolutionary psychology and conceptual integration. In Barkow et al. (eds.), The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York and Oxford: Oxford University Press, 3-18.
Daly, Martin and Margo Wilson 1981: Abuse and neglect of children in evolutionary perspective. In Richard D. Alexander and Donald W. Tinkle (eds.), Natural Selection and Social Behavior: Recent Research and New Theory. New York: Chiron, 405-16.
Daly, Martin and Margo Wilson 1983: Sex, Evolution and Behavior: Adaptations for Reproduction. Boston: Willard Grant Press.
Daly, Martin and Margo Wilson 1988: Homicide. Hawthorne, NY: Aldine de Gruyter.
Daly, Martin and Margo Wilson 1990: Killing the competition. Human Nature 1, 83-109.
Darwin, Charles  1975: On The Origin of Species by Means of Natural Selection, or, the Preservation of Favored Races in the Struggle for Life. A Facsimile of the First Edition, ed. Ernst Mayr. Cambridge, MA: Harvard University Press.
Dawkins, Richard 1989: The Selfish Gene, 2nd edn. Oxford: Oxford University Press.
Dennett, D. C. 1995: Darwin’s Dangerous Idea: Evolution and the Meanings of Life. New York: Simon and Schuster.
Dewsbury, D.A. 1982: Ejaculate cost and male choice. American Naturalist 119, 601-10.
Dupre, John 1998: Against reductionist explanations of human behavior, Part 1. The Aristotelian Society Supplementary vol. LXXII: 153-71.
Eaton, R. L. 1978: Why some felids copulate so much: A model for the evolution of copulation frequency. Carnivore 1, 42-51.
Fodor, Jerry A. 2000: The Mind Doesn’t Work that Way: The Scope and Limits of Computational Psychology. Cambridge, MA: MIT Press.
Gladstone, D. 1979: Promiscuity in monogamous colonial birds. The American Naturalist 114 (4), 545-57.
Gould, Stephen J. 1977: Biological potentiality vs. biological determinism. In Philip Appleman (ed.), Darwin: A Norton Critical Edition. New York and London: W.W. Norton & Co., 460-65.
Gould, Stephen J. 1980: Sociobiology and the theory of natural selection. American Association for the Advancement of Science Symposia 35, 257-69.
Gould, Stephen J. and Richard C. Lewontin 1979: The spandrels of San Marco and the Panglossian paradigm: A critique of the adaptationist programme. Proceedings of the Royal Society B205, 581-98.
Haraway, Donna 1978a: Animal sociology and a natural economy of the body politic, Part I. Signs 4 (1), 21-36.
Haraway, Donna 1978b: Animal sociology and a natural economy of the body politic, Part II: A political physiology of dominance. Signs 4 (1), 37-60.
Hrdy, Sarah 1986: Empathy, polyandry, and the myth of the coy female. In Ruth Bleier (ed.), Feminist Approaches to Science. New York: Pergamon Press, 119-146.
Hunter, F. M., M. Petrie, M. Otronen, T. Birkhead, and A. P. Moller 1993: Why do females copulate repeatedly with one male? Trends in Ecology and Evolution, 8, 21-26.
Jones, S. 1999/2000. Darwin's Ghost. The Origin of Species Updated. New York: Random House.
Keller, Evelyn Fox and Elisabeth A. Lloyd (eds.) 1992: Keywords in Evolutionary Biology. Cambridge, MA: Harvard University Press.
Kitcher, Philip 1985: Vaulting Ambition. Cambridge, MA: MIT Press.
Lambert, D. M., P. D. Kingett, and E. Slotten 1982: Intersexual selection: The problem and a discussion of the evidence. Evolutionary Theory 6: 67-78.
Lanier, D. L., D. Q. Estep, and D. A. Dewsbury 1975: Copulatory behaviors of golden hamsters: Effects on pregnancy. Physiological Behavior 15, 209-12.
Lee, Richard B. and Irven DeVore (eds.) 1968: Man the Hunter. Chicago: Aldine, 159-213.
Lewontin, Richard C. 1974: The Genetic Basis of Evolutionary Change. New York: Columbia University Press.
Lewontin, Richard C. 1977: Adaptation. In Richard Levins and Richard C. Lewontin (eds.), The Dialectical Biologist. Cambridge, MA: Harvard University Press, 65-84.
Lewontin, Richard C. 1978: Adaptation. Scientific American 239 (3), 156-69.
Lewontin, Richard C. 1984. Adaptation. In Elliott Sober (ed.), Conceptual Issues in Evolutionary Biology: An Anthology. Cambridge, MA and London: MIT Press, 234-51.
Lloyd, Elisabeth A. 1996: Pre-theoretical assumptions in evolutionary explanations of female sexuality. Philosophical Studies 69, 139-53.
Maynard Smith, J. 1984: Optimization theory in evolution. In Elliott Sober (ed.), Conceptual Issues in Evolutionary Biology: An Anthology. Cambridge, MA and London: MIT Press, 289-315.
Mayr, Ernst 1972: Sexual selection and natural selection. In Bernard Campbell (ed.), Sexual Selection and the Descent of Man. London: Heinemann, 87-104.
Mayr, Ernst 1983: How to carry out the adaptationist program. American Naturalist 121, 324-34.
Mayr, Ernst 1988: Toward a New Philosophy of Biology. Cambridge, MA and London: The Belknap Press of Harvard University.
Nelson, Lynn H. 1996: Empiricism without dogmas. In Lynn H. Nelson and Jack. Nelson (eds.), Feminism, Science, and the Philosophy of Science. Dordrecht: Kluwer, 95-119.
Nelson, Lynn H. and Jack Nelson 2000a: On Quine. Belmont, CA: Wadsworth/Thompson Learning.
Nelson, Lynn H. and Jack Nelson 2001b: Quine on reconstituting empiricism. Paper delivered at the PNP Colloquium, Washington University, September 2000.
Partridge, L. and T. Halliday 1984: Mating patterns and mate choice. In J. R. Krebs and N. B. Davis (eds.), Behavioral Ecology: An Evolutionary Approach, 2nd edn. Oxford: Blackwell, 222-50.
Pinker, Steven 1994: The Language Instinct: How the Mind Creates Language. New York: William Morrow.
Pinker, Steven and P. Bloom. 1992: Natural language and natural selection. In Barkow et al. (eds.), The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York and Oxford: Oxford University Press, 451-93.
Quine, W. V. O. 1969: Ontological relativity. In Ontological Relativity and Other Essays. New York and London: Columbia University Press, 26‑68.
Quine, W. V. O. 1972: Methodological reflections on current linguistic theory. In Donald Davidson and Gilbert Harmon (eds.), Semantics of Natural Language. Dordrecht: D. Reidel, 442-54.
Richardson, Robert 1996: The prospects for an evolutionary psychology: Human language and human reason. Minds and Machines 6: 541-57.
Richardson, Robert 2001: Evolution Without History: Critical Reflections on Evolutionary Psychology. In Harmon R. Halcomb (ed.), Conceptual Challenges in Evolutionary Psychology: Innovative Research Strategies. Dordrecht: Kluwer, 327-373
Rosenberg, Alexander 1980: Sociobiology and the Preemption of Social Science. Baltimore and London: The Johns Hopkins University Press.
Ruse, Michael 1988: Philosophy of Biology Today. Albany, NY: SUNY Press.
Schubert, Glendon 1989: Evolutionary Politics. Carbondale and Edwardsville, IL: Southern Illinois University Press.
Smith, S. M. 1988. Extra-pair copulations in black-capped chickadees: The role of the female. Behaviour 197: 15-23.
Sober, Elliott 1984a: The Nature of Selection: Evolutionary Theory in Philosophical Focus. Cambridge, MA: The MIT Press.
Sober, Elliott (ed.) 1984b: Conceptual Issues in Evolutionary Biology: An Anthology. Cambridge, MA and London: The MIT Press.
Sober, Elliott 1993: Philosophy of Biology. Boulder, CO and San Francisco, CA: Westview Press.
Spencer, H.G. and J. C. Masters 1992: Sexual selection: Contemporary debates. In Evelyn Fox Keller and Elisabeth A. Lloyd (eds.), Keywords in Evolutionary Biology. Cambridge, MA: Harvard University Press, 294-301.
Symons, Donald 1979: The Evolution of Human Sexuality. Oxford: Oxford University Press.
Symons, Donald 1992: On the use and misuse of Darwinism in the study of human behavior. In Barkow et al. (eds.), The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York and Oxford: Oxford University Press, 137-159.
Tang-Martinez, Z. 2000: Paradigms and primates: Bateman’s principle, passive females, and perspectives from other taxa. In S. C. Strum and L. M. Fedigan (eds.), Primate Encounters: Models of Science, Gender, and Society, Chicago: University of Chicago Press.
Tanner, Nancy M. and Adienne Zihlman 1976: Women in evolution, Part I: Innovation and selection in human origins. Signs 1 (3), 585-608.
Thornhill, R. and N.W. Thornhill 1987: Human rape: The strengths of the evolutionary perspective. In C. Crawford, M. Smith, and C. Krebs (eds.), Psychology and Sociobiology. Lubbock, TX: Texas Tech University Press, 269-291.
Thornhill, R. and N.W. Thornhill 1992: The evolutionary psychology of men's coercive sexuality. Behavioral and Brain Sciences 15, 363-421.
Tooby, John and Leda Cosmides 1992: The psychological foundations of culture. In Barkow et al. (eds.), The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York and Oxford: Oxford University Press, 19-136.
Tooby, John and Irven DeVore 1987: The reconstruction of hominid behavioral evolution through strategic modeling. In W. G. Kinzey (ed.), The Evolution of Human Behavior: Primate Models. Albany: SUNY Press, 183-237.
Trivers, R. L 1972: Parental investment and sexual selection. In Bernard Campbell (ed.), Sexual Selection and the Descent of Man. Chicago: Aldine, 136-179.
Van Voorhies, W. A. 1992: Production of sperm reduces nematode lifespan. Science 360, 456-58.
Washburn, Sherwood L. (ed.) 1961: Social Life of Early Man. Chicago: Aldine.
West-Ebelard, M. J. 1992: Adaptation: Current usages. In Evelyn Fox Keller and Elisabeth A. Lloyd (eds.), Keywords in Evolutionary Biology. Cambridge, MA: Harvard University Press, 13-18.
Williams, George C. 1966: Adaptation and Natural Selection. Princeton, NJ: Princeton University Press.
Wilson, E.O. 1978. On Human Nature. Cambridge, MA.: Harvard University Press.
Wilson, M. and M. Daly 1992: The man who mistook his wife for a chattel. In Barkow et al. (eds.), The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York and Oxford: Oxford University Press, 289-322.
Wylie, Alison 1997: The engendering of archaeology: Refiguring feminist science studies. Osiris 12: 80-99.