The Limits of Evolutionary Explanations of Morality: The Inclusivist Anomaly and its Implications for Evoconservatism

Transcription

1 The Limits of Evolutionary Explanations of Morality: The Inclusivist Anomaly and its Implications for Evoconservatism Abstract. Traditional conservative arguments against the possibility of major moral progress or fundamental institutional reform relied on under-evidenced assumptions about the limitations of human nature. A number of contemporary thinkers from a variety of disciplines have attempted to fill this empirical gap in the conservative argument by appealing to evolutionary theory. These evoconservatives infer from evolutionary accounts of morality that it is unrealistic to think that cosmopolitan and other inclusivist moral ideals can meaningfully be realized. In this paper, we argue that contemporary evoconservative arguments are no more successful than their pre-darwinian predecessors, because they have overstated the explanatory reach of evolutionary theory and underestimated the scope of the explanandum. We argue that no adequate evolutionary explanation has been given for several important features of contemporary human morality, namely the very inclusivist moral commitments that prominent evolutionary explanations appear to rule out. We examine putative selectionist and byproduct explanations of morality and show that they fail to account for significant aspects of the morality of substantial numbers of existing human beings. We argue that to the extent that evolutionary explanations of morality succeed, they do not imply that morality is severely constrained, such that inclusivist morality is a utopian fantasy that is incapable of shaping important moral beliefs, practices and institutions. We go on to attribute the evolutionarily anomalous features of human morality that we have identified to a flexible capacity for open-ended normativity, which we argue presents a serious obstacle to theorists who wish to draw substantive moral and political lessons from human evolutionary history. Finally, we consider the implications of open-ended normativity for empirically informed versions of the ought implies can thesis. Introduction Increasing philosophical attention is being paid to evolutionary explanations of morality, reflecting a more general trend toward empirically informed ethics. Whether evolutionary explanations of morality succeed is an important philosophical matter in its own right, compelling us to investigate the nature of the explanandum as well as the adequacy of proposed evolutionary explanations. Yet in addition, some authors maintain that evolutionary accounts of morality have important implications for ethics and political philosophy. Conservative thought in the secular tradition has long held that human nature places severe limitations on the plausibility of systematic social and moral reform (Kirk 2001), but it has offered little in the way of scientific evidence to support this skeptical claim. A number of contemporary thinkers from a variety of disciplines, whom we will refer to as evoconservatives, have attempted to fill this empirical 1

2 gap in the conservative argument by appealing to the prevailing evolutionary explanation of morality to show that it is unrealistic to think that cosmopolitan and other inclusivist moral ideals can meaningfully be realized. In this paper, we argue that contemporary evoconservative arguments are no more successful than their pre-darwinian predecessors, because they have overstated the explanatory reach of evolutionary theory and underestimated the scope of the explanandum. First, we summarize the predominant evolutionary account of morality and show that even if it is vindicated, this leaves unexplained a number important features of contemporary human morality namely, the very cosmopolitan and other inclusivist moral commitments that prominent evolutionary explanations appear to rule out. We argue that these features of the morality of substantial numbers of human beings, which we refer to collectively as the inclusivist anomaly, are not amenable to any standard type of evolutionary explanation, whether of the selectionist or byproduct variety. Second, we show that to the extent that evolutionary explanations of morality succeed, this does not show that morality is severely constrained such that progressively inclusivist moralities and institutional orders that increasingly rely on inclusivist moral commitments are a utopian fantasies or destined to play a trivial role in human, ecologyically irrelevant fantasy. Third, we attribute the evolutionarily anomalous features of human morality that we identify to a flexible capacity for open-ended normativity the capacity to change our behavior through reflection and modification of existing norms. We sketch an alternative evolutionarily informed explanation of inclusivist morality that adverts to this capacity, but which has no conservative implications. We argue that the capacity for open-ended normativity presents a 2

3 serious obstacle to theorists who wish to draw substantive moral and political lessons from human evolutionary history. Finally, in light of these results, we conclude that Principle of Minimal Psychological Realism according to which moralities and moral theories must take into account evolved psychological capacities in framing their conceptions of moral virtues and duties is far less informative than its proponents have assumed, and cannot salvage the evoconservative argument. Given the capacity for open-ended normativity, the boundaries of what humans can do and be are so indeterminate that the naturalistic philosopher s version of the ought implies can thesis provides little guidance for the evaluation of moral theories or proposals for institutional reform. 1. Selectionist Explanations of Morality: The Received View In broad strokes, morality in the relevant evolutionary literature refers to a social commitment to authority-independent and/or preference-independent norms, modulated by other-directed and inward-directed moral emotions and judgments, and typically enforced through institutionalized sanctions. Painting a finer-grained picture would involve filling in specific moral content, such as a sense of fairness, prohibitions against particular behaviors (e.g., murder, stealing, gluttony, etc.), conceptions of virtues, specific punitive reactions to norm violations, and so on. Why think that morality at any level of description might be amenable to evolutionary explanation? One reason is that moral systems are spatiotemporally ubiquitous in human societies. Moral rules structure the behavior of all known hunter-gatherer bands, nomadic tribes, and sedentary agricultural populations, on all habitable continents and across all ecological niches and modes of subsistence. Moral systems are presumed to have been in place at least since the origins of behaviorally modern humans in the upper Paleolithic, and possibly much 3

4 earlier as evidenced by high levels of cooperative foraging in the paleoanthropological record. Although moral systems vary considerably, they exhibit significant commonalities in form and content (Brown 1991; Knauft 1991; Joyce 2006; but see Prinz 2008). 1 In any system with variation and heredity, including biological and cultural systems, such patterns cry out for a selectionist explanation, or else indicate the existence of pervasive constraints. Furthermore, moral systems present as adaptively configured so as to foster cooperative social structures, resulting in a functional match to coordination problems that is incredibly unlikely to arise by chance alone (i.e., in the absence of selection). 2 The received view among evolutionary theorists who believe that human morality can be given a selectionist explanation goes roughly like this. Morality developed and spread among small, scattered hunter-gatherer groups in the middle-to-late Pleistocene, where it was selected for the effect of managing patterns of interaction that resulted in costly intra-group conflicts. In particular, morality helped solve collective action problems by reducing free-riding, enabling individuals to resist temptations to act selfishly, and preventing dominant individuals from monopolizing the fruits of cooperation generating an evolutionary return that was greater for each individual than would have been possible if each had acted alone or as part of a group that did not cooperate 1 Cross-cultural universality should not be taken, in itself, to imply innateness, but as theoretical work on adaptation, developmental systems, and cultural evolution has shown, evolutionary explanation is not wed to any particular mode of trait acquisition. An evolutionary response to selection only requires like-begets-like heritability, and this condition may be met by genetic, epigenetic and cultural modes of transmission. Thus, it does not matter for present purposes whether moral traits are genetically prepared or socially acquired or both, so long as they were proper objects of selection or discernable byproducts of other adaptations. The underlying mode of acquisition may be relevant here insofar as it bears on the flexibility of moral morphology, but both genes and cultural variants act to bias human behavior and neither is determinative of the same (see 2). 2 To say that something is an adaptation is a strictly backward-looking statement it is a claim about the selective etiology of a trait, not about its present utility or current contribution to survival and reproduction. Thus, to the extent that morality is associated with reproductive costs in the modern environment, this does little to undermine the selective-etiological claim that significant aspects of morality are adaptations. 4

5 effectively (Haidt 2012; Boehm 2012/2001; Sterelny 2012; Kitcher 2011; Joyce 2006; Wright 2000). The morality of hunter-gatherer societies, which is widely regarded as the ancestral state of human morality, is ubiquitously anti-hierarchical, in the sense that individuals with authoritarian or despotic tendencies are actively and normatively opposed by alliances among the rank and file, group decisions are not dominated by any single individual, and sharing is a central organizing principle that governs the distribution of certain highly prized resources, such as meat (Boehm 2001). Anthropologists and philosophers of science characterize this type of morality as egalitarian. 3 Violations of egalitarian norms in hunter-gatherer bands especially attempts to monopolize resources or to exercise authority over fellow group members are met with forceful sanction ranging from social ridicule to ostracism to execution (ibid). Egalitarian moralities are inferred to have existed at similar frequencies in prehistoric hunter-gatherer societies, dating back at least to the origins of behaviorally modern humans and possibly as early as the emergence of Homo sapiens itself. 4 Various accounts have been offered to explain stable cooperation in moderate-sized nonkin groups, such as reciprocal altruism (Trivers 1971), indirect (reputation-based) reciprocity (Alexander 1987), and punishment-reinforced cooperation (Boyd et al. 2003). There is, however, broad agreement on the basic Darwinian logic: In a population of competing cultural groups, 3 For the sake of clarity and convenience we will follow this convention, though we will consider problems with this characterization below. 4 A number of contemporary evolutionary theorists have converged on the hypothesis that cooperative foraging was the key ecological design problem that prompted the evolution of the egalitarian ethos (e.g., Sterelny 2012; Boehm 2001). Though somewhat speculative, the hypothesis is as follows: Early in human evolution (~400kya), there was a shift to hunting large dangerous quarry, particularly during frequent periods of glaciation when edible plants and small game animals were scarce. For much of human history this cooperative feat was accomplished with rudimentary stone-tipped wooden spears and other non-projectile weapons. This required not only meta-cognitive capacities (such as shared intentionality ) that were lacking in the last common ancestor of humans and chimpanzees, but also sophisticated normative mechanisms for underwriting the equitable distribution of the spoils once the fruits of cooperation were realized. 5

6 those that developed effective moralities, i.e. moralities that were capable of avoiding the lost opportunity costs associated with cooperation failures (Kitcher 2011), were more likely to pump hominins into the next generation, to persist as groups, to sustain and transmit their social structures, and/or to give rise to offspring groups. 5 These ecological conditions, so the argument goes, conferred a reasonably high probability on the evolution of morality in broad strokes, and go some way toward explaining its more specific contours, such as our evaluative attitudes toward kin, kith, strangers, non-reciprocators, gluttons, cheaters, murderers, and the like. Focusing on the prosocial effects of prehistoric morality can obscure its darker side. Ethnographic and modeling work indicates that the evolution of egalitarian and other altruistic moral norms in moderately sized populations hinges on institutionalized punishment (Boyd et al. 2003; Boehm 2001); and the evolution of punishment appears to pose a higher-order altruism problem that only group-level selection can solve. Group-level selection, in turn, is only strong in the context of frequent and frequently lethal intergroup conflict (Bowles 2009; Boyd et al. 2003; Boyd and Richerson 2002). Groups that contained more altruists and moralizing punishers and hence more cooperative social structures tended to outperform groups with less effective moralities in economic and military contests between groups. A striking feature of the received selectionist explanation, therefore, is that it implies morality is essentially an intra-group affair. The same ecological conditions and selection 5 There is continued controversy over the level at which selection must operate in order to stabilize cooperative interactions among non-kin. A growing chorus of biologists, anthropologists, and philosophers of science now argue that robust cooperation in moderate-sized groups of non-kin is only likely to evolve through a process of selection at the group level, given the costs of altruism and norm enforcement to individual fitness within groups (Sober and Wilson 1999; Boehm 2001; Boyd, Gintis and Richerson 2003; Bowles 2009/2008; Wilson and Wilson 2007; Haidt 2012; for a partially dissenting view, see Sterelny 2012). For purposes of this paper, it does not matter whether selection for moral traits can be cashed out at the level of individuals in a group-structured population, or at the level of cultural groups proper, since in either case a selectionist explanation would be vindicated. We will not consider evolutionary explanations of morality at the level of cultural variants themselves, since the received selectionist explanation conceives of moral traits as parts of the individual or group phenotype, rather than as units of selection in their own right. 6

7 pressures that made moral traits adaptive would have imposed a fitness cost on extending evolutionarily excessive moral consideration to out-group members. Just as free-riding on ingroup members will tend to undermine group performance in a competitive intergroup arena, so too will excessive moral consideration toward members of the out-group, particularly given that there were (until perhaps very recently in human history) no institutional capacities for punishment at the meta-group level. The selectively optimal combination appears to have been reasonably expansive moral consideration toward members of one s in-group (with a caveat for gender), and highly strategic including predatory, antagonistic and apathetic behavior toward strangers, who were often distrusted, dehumanized, and delegitimized (Navarette and Fessler 2006; Pinker 2011). 6 Careful examinations of the ethnographic and archeological records attest to this pattern (Pinker 2011; Bowles 2009; Boehm 2001; Keeley 1996). Pervasive intergroup violence is not an artifact of civilization, but rather a deeply rooted aspect of the human condition: Morality was born and coevolved in a Darwinian crucible of intergroup conflict. 7 To say that this is the received selectionist explanation of morality is not to say that it is the received view. Some prominent moral psychologists and philosophers of science argue that allegiance to specifically moral norms is an evolutionary byproduct of adaptive tendencies toward norm compliance in general (Machery and Mallon 2010; Sripada and Stich 2006), or that certain moral norms are byproducts of moral emotions and non-moral capacities (Lieberman 6 Our claim is not that intraspecific aggression is always adaptive, as the costs of aggression will often outweigh its benefits. Nor do we intend to claim that cooperation between groups was never fitness enhancing. Under certain conditions, intergroup hostility can lead to lost opportunity costs, such as the benefits of material trade and mate exchange that would have flowed from non-antagonistic interactions. Nevertheless, patterns of intergroup homicide in pre-state humans, as well as in common chimpanzees, indicate that intergroup predation often reaps evolutionary rewards, and this would have been particularly true for weapons-wielding hominins with the cognitive prowess to make case-by-case risk-benefit calculations. 7 Sterelny (2012, 190) argues, contra Bowles and his collaborators, that cooperation and altruism are the fuel of war, but not warfare s child. Even if Sterelny is right that primitive cooperative capacities predate intense intergroup conflict, they were likely coopted and honed in coevolution with the latter. The skills required for hunting dangerous game are readily transferrable to hunting dangerous hominins. 7

8 2008; Prinz 2008). Edouard Machery (personal communication) holds that the generic capacity for norm following is an adaptation, but maintains that morality per se cannot be an object of selection because moral norms cannot be adequately delineated from non-moral ones. Others remain unconvinced that genuinely altruistic motivations exist or are cross-culturally robust (Stich 2013). To further complicate matters, when some theorists maintain that morality didn t evolve, what they mean is that it didn t evolve via gene-based selection, whereas they accept that specific moralities could have been culturally selected for. Our aim here is neither to evaluate the standard selectionist account nor its detractor theories. Instead, we will argue that even if a selectionist explanation of certain aspects of morality could be given along the lines sketched above, this would still leave much of contemporary moral morphology beyond the scope of evolutionary explanation altogether. This, in turn, will be taken to show that morality is not constrained by human evolutionary history to the degree that evoconservatives and others might presuppose. 2. Evoconservatism and Two Types of Morality It is important to recognize that many proponents of the standard selectionist explanation of morality do not subscribe to a conservative brand of politics, nor have they suggested that the evolutionary explanations they give, if vindicated, would have any conservative moral or political implications. Philip Kitcher (personal communication; see also note 15), for instance, maintains that although morality has the evolutionary function of solving cooperation failures within groups, its emergence prompted an ongoing ethical discussion, which due to our deliberative faculties can go in any number of directions, including inclusivist ones. 8

9 Nevertheless, authors from a variety of disciplines have inferred from the received selectionist explanation of morality that the content of human morality is seriously constrained and evolutionists have done little to disabuse them of this notion. These evoconservatives, as we will call them, contend that the ecological challenges our distant ancestors faced generated selection pressures for evaluative tendencies that circumscribe effective moral commitments to members of one s own kin, group, tribe or nation and that these putative facts about human evolutionary history significantly constrain the shape of plausible moralities and the scope of other-regarding concern. This, in turn, is thought to suggest that cosmopolitan and other inclusivist moral principles (see 3) are not appropriate or realistic for beings like us. Stephen Asma (2013), for instance, argues that moral emotions cannot stretch indefinitely to cover the massive domain of strangers and nonhuman animals, given that our other-regarding dispositions were limited by evolutionary design to our affective communities of kith and kin (see also Asma 2012, 45-46). U.S. appellate judge and legal theorist Richard Posner defends species-based moral discriminations by appealing to similar evolutionary considerations (Posner and Singer 2001), which are also read as supporting Jack Goldsmith and Eric Posner s contention that it is a mistake to try to create an international legal order grounded in cosmopolitan moral principles because we should not expect individual altruism to extend to people who are physically and culturally more distant (2005, 212). Francis Fukuyama (2002, ), a prominent conservative bioethicist, holds that political orders must be grounded in a substantive conception of human nature that pays heed to our evolved biases toward kin and ingroup as well as the evolutionarily evidenced limitations of our capacity to sympathize with all human beings. Leading psychologist Jonathan Haidt, who has stressed the moral psychological significance of in-group loyalty, expresses a similar view about the limits of inclusivist morality: 9

10 It would be nice to believe that we humans were designed to love everyone unconditionally. Nice, but rather unlikely from an evolutionary perspective. Parochial love love within groups amplified by similarity, a sense of shared fate, and the suppression of free riders, may be the most we can accomplish (2012, 245). Larry Arnhart (2005), a self-proclaimed proponent of the Darwinian Right, goes further in arguing that evolved constraints on human morality make moral impartiality not only implausible but also morally repugnant insofar as it entails a rejection of our evolved human nature. Thus, evoconservatives believe that there are significant psychological constraints on the shape of human morality, that these constraints are essentially fixed, and that they result in an other-regard that is effectively restricted to groups. The chief improvement of evoconservatism over traditional conservative philosophies is that it appeals to contemporary evolutionary theory to ground its empirical claims about the moral limitations of human nature. In short, evoconservatives hold that the content of morality in particular the scope of moral duties and the class of beings who are recognized as having moral standing is severely constrained due to evolutionary history. This in turn limits the set of social practices and institutions that are feasible and, according to some natural law theories, morally desirable. 8 This view has much more radical implications than those who endorse it acknowledge. If human morality is explainable according to the selectionist logic that evoconservatives endorse, then it is an understatement to say that inclusivist morality is a nonstarter. It implies that the scope of moral consideration tout court is very limited, not just the scope of equal basic moral 8 There is a much weaker evoconservative claim that might be distinguished here. This weaker view holds that selectionist explanations of morality imply limited sympathy or feelings of positive regard for distant strangers, but that this psychological claim in itself has no conservative political implications. That is to say, humans may develop institutions and cultural practices that allow them to treat distant strangers as being worthy of moral consideration, even equal consideration, even if they are incapable of loving them (to use Haidt s words). In other words, social practices and institutions may produce inclusivist morality, or a broadened range of what Kitcher calls behavioral altruism, without unlimited sympathy or love. Some of the writers discussed above are unclear as to whether they are only making the psychological claim or also making the mistake of assuming that if the psychological claim is true, then conservative moral or political conclusions follow. Some, including Posner and Goldsmith, clearly make the mistaken inference from the former to the latter. As we will see, to do so is to fail to appreciate how cultural developments, in particular institutions, can expand our capacities for behavioral altruism. 10

11 consideration. In other words, it implies that not only is it implausible to expect people to regard all human beings as worthy of equal basic moral consideration, but also that it is implausible to expect people to regard most human beings as worthy of any moral consideration at all. It is important to disentangle claims about adaptive constraints on moral morphology from claims about the underlying mode of acquisition. The innateness of a character is often associated with its rigidity, un-alterability or resistance to modification. Yet it is widely acknowledged that both genes and environment (including culture) can bias human development toward particular outcomes. We should not assume that behaviors produced by gene-based dispositions are more difficult to alter either at the individual or populational level than behaviors produced by culturally acquired dispositions. This is a matter of developmental contingency: In some cases, outcomes that result from cultural biases can be more entrenched and hence more difficult to modify than outcomes that result from biological difference-makers. So although the evoconservative inference hinges on claims about the rigidity of human moral behavior, the plausibility of these claims need not turn on a resolution of the moral innateness debate (even if many evoconservatives assume that it does). 9 Although tthe received selectionist explanation does not make any explicit claims about constraints on the shape of human morality, but it could be read to suggest, in line with the evoconservative inference, that the only sort of morality that humans are capable of engaging in is what we will refer to as morality as cooperative group reciprocity (Gauthier 1989). Indeed, evolutionary theorists have explicitly linked the evolutionary narrative to a strategic, prudence- 9 Some evolutionary psychologists (e.g. Haidt 2012) have proposed that human morality clusters along innate, content-specific foundational attractors, such as justice, harm, in-group loyalty, sanctity, and so on; even if this were true, if there are great differences between the relative weights placed on these foundations across cultures (as there seems to be), then this will not have any obvious evoconservative implications. 11

12 based theory of morality, such as that of Gauthier (see e.g. Frasner and Sterelny, unpublished). 10 According to morality as cooperative group reciprocity theories, moral standing is something that members of a cooperative group confer on one another and only on one another. Individuals excluded from this reciprocal arrangement have no moral standing at all, and hence there are no moral duties constraining how out-group members should be treated. Moral status is conferred on individuals who can either disrupt or contribute to cooperation i.e., on the basis of strategic capacities relative to a cooperative scheme. This is not to say that each moral action necessarily involves explicit strategic calculations but rather, that our motivation for acting morally emanates stems from the fact that doing so was and is in our cooperative self-interest. As we noted earlier, in the circumstances in which human morality emerged in the environment of evolutionary adaptation ( EEA ) a highly instrumental attitude toward strangers, not an acknowledgement that all human beings count morally, was most likely conducive to fitness. This strategic conception of morality is not only consistent with evolutionary explanation it is Formatted: Space After: 0 pt arguably central to evolutionary theories of morality. It is helpful here to distinguish morality as cooperative group reciprocity from what we refer to as subject-centered between two quite different types of moralitiesy (Buchanan 1990). Formatted: Font: Italic The first type we have already noted: Morality as cooperative group reciprocity, which ties moral standing to strategic capacities. Subject-centered morality, in contrast, recognizes individuals as Formatted: Font: Not Italic having moral standing on the basis of their possessing certain non-strategic capacities, such as practical rationality, responsiveness to reasons, sentience, and so on. We will now argue that insofar as it ties moral status to non-strategic and non-group-based properties, subject-centered 10 Alexander (1987: 3), for instance, writes that understanding human morality requires that we view societies as collections of individuals seeking their own self-interest. Even evolutionary theorists who recognize that contemporary human morality is not confined solely to morality as cooperative group reciprocity (e.g., Joyce 2006, ch. 4) still view reciprocity relations as dominating contemporary moral behavior. 12

13 morality (and its associated moral culture and institutions) constitutes a glaring anomaly from the standpoint of the received selectionist account. 3. The Inclusivist Anomaly One major flaw in the evoconservative appeal to evolutionary theory is that contemporary morality, as experienced and exhibited by significant numbers of people, is strikingly more inclusive than one would expect if selectionist explanations were the whole story, or even most of it. We will highlight in particular four anomalous aspects of this inclusivity. First, significant numbers of people now regard at least some nonhuman animals as proper subjects of moral consideration; that is, they believe that there are moral constraints on how we are to treat animals that do not derive from contingent human interests or sensitivities. There remains, of course, much disagreement over precisely what treatment is due certain nonhuman animals and from what moral principles such obligations are derived. However, there is an increasingly broad-based consensus that animal cruelty is a wrong to animals qua moral subjects (DeGrazia 2009) a moral judgment that is enshrined in the laws of developed nations. Animal blood sports are widely illegal and seriously punishable, increasing legislation is aimed at enhancing the welfare of agricultural animals, and there are significant, institutionalized constraints on the use of certain non-human animals in medical experimentation with some usages, such as research on great apes, having been prohibited categorically because of the high subject-centered moral status that is attributed to these animals. Second, many people regard valid moral norms as universalizable; that is, they regard it as incorrect to say, for example, that X is permissible for me but not for you, or for Blacks but not for Whites, or for men but not for women without adducing a morally relevant differencemaking feature. Importantly, belonging to or identifying with a particular group, such as a race, 13

14 gender, religion or ethnicity, is widely and increasingly held not to be an acceptable differencemaker when it comes to ascriptions of basic moral status, including political and civil rights (for a discussion, see Pinker 2011, ch. 7) and this belief is widely institutionalized. Third, there is the culture of human rights: Many people now recognize that all human beings ought morally to be treated in certain ways by their own governments, irrespective of whether there are local laws are in place that protect those rights, and irrespective of the contingent strategic properties that people possess. This is the foundation of cosmopolitan moral principles, by which we mean principles that accord an equal basic moral status to all human beings, irrespective of group membership and strategic capacities. These principles have been codified in international human rights law, which has been incorporated into the domestic law of, and legally binds, over two hundred nations. Although enforcement of human rights by international institutions is weak, there is substantial enforcement through domestic courts in a growing number of countries. The concept of basic inalienable rights, which served as the bedrock of modern constitutional democracy as well as the motivation for the anti-torture, abolitionist and decolonization movements (Buchanan 2012), is an affirmation of the equal status of all people regardless of their group membership and the threats they pose to cooperation and thus constitutes an explicit rejection of reciprocity-based theories of morality. Fourth, is the emergence of a subject-centered morality that compels us to recognize the moral standing of individuals who pose no threat to us or who do not contribute to cooperative goods. Even if a vulnerable minority group can safely be exploited or oppressed without incurring any long-term risks to the majority group, it is widely held that such treatment is inconsistent with the moral status of those individuals. Likewise, it is widely held that persons who lack strategic capacities, such as severely disabled individuals or young children, may not 14

15 justly be denied access to social resources or excluded from the class of beings that are proper subjects of moral concern. The best explanation for this constellation of considered judgments is that the moral worth of such persons is thought to derive from properties other than their strategic capacities. The third feature, the emergence of human rights culture, can be seen as a shift from a cooperative group-based morality to a subject-centered morality, insofar as human rights are thought to be grounded in the dignity of the human individual and where the possession of dignity does not depend on one s strategic capacities or group membership. None of these inclusivist moral features are plausibly explainable in standard selectionist terms, that is, as adaptations that arose in the EEA. The survival of human groups in the late Pleistocene and early Holocene depended crucially on the exploitation of animals, which lack the relevant strategic capacities. Early human groups that treated nonhuman animals as subjects of moral worth would have paid a high fitness price, for this would have placed severe restrictions on the exploitation of animals for protein and other valuable materials like skin and bone, and as beasts of burden. This would have been particularly true for nomadic hunter-pastoralist tribes and other post-neolithic populations who relied more heavily on domesticated animals for their subsistence. Competition among these groups would have placed a fitness premium on maximizing control over animal domesticates and their life cycles. Similarly, the tendency to universalize moral judgments may have been adaptive if it were restricted to members of one s own group, along the lines discussed above; but it is hard to see how the tendency to universalize would have contributed to a group s survival if it were extended to out-group individuals regardless of their strategic capacities. Doing so would have had two negative consequences: First, it would have made the group vulnerable to predation by groups that did not acknowledge that moral judgments or norms should be universalizable, and 15

16 second it would have limited the group s ability to exploit other vulnerable groups in fitness enhancing ways. Nor is the core commitment of human rights culture the belief that every human being has certain basic moral entitlements something that is explainable in terms of morality as cooperative group reciprocity. Especially in cases of armed conflict, but in many other kinds of interactions as well, members of a group that act on a commitment to human rights (which prohibits a no-holds-barred approach) may be disadvantaged, rather than advantaged, in fitness terms. The fact that nations could enhance their overall productivity by oppressing certain groups, or by withdrawing basic measures of support for, say, certain disabled individuals or children, is not considered morally legitimate grounds for doing so. Subject-centered morality thus comes with attendant fitness costs that morality as cooperative group reciprocity avoids. We have shown that inclusivist moral commitments cannot plausibly be explained as adaptations derived in the EEA. We think it is equally implausible that these inclusivist features could be explained as objects of selection in a more recent environment. Consider, for instance, the view that the social environment has changed so profoundly, due to the increasing interconnectedness of human communities, that a more inclusive morality is actually a groupbeneficial trait; and further, that the spread of inclusivist moralities in recent human populations is due to the advantages or fitness benefits they conferred. If that were true, then the inclusivist anomaly would vanish. Note, however, that even if inclusivist morality could be explained in this way, the evoconservative argument would have little force, since it would imply that moral inclusivity is limited not by a rigid, evolved moral psychology but rather by ecological circumstances that make it beneficial which leaves open the possibility of further significant expansions of the moral circle. 16

17 Philip Kitcher (2011) appears to favor something along these lines. He holds that the function of ethics is to replace altruism failures with behavioral altruism, and that this replacement is constitutive of moral progress. Though Kitcher does not explicitly define altruism failure, we infer from his analysis that this refers to fitness costs that result from noncooperation in circumstances where cooperation would be fitness enhancing (for a similar reading, see FitzPatrick 2012). Kitcher seems to believe that moral progress is achieved only when morality-as-reciprocity enables us to expand our circle of cooperators to include previously excluded strategic partners (236, 307). 11 It is worth noting that on this interpretation of Kitcher, many of the putative achievements of subject-centered morality, such as basic rights for persons with disabilities or children, will not count as instances of moral progress. This is because such moral inclusions do not avoid fitness costs, as they do not implicate a group of persons who would, if treated well, contribute to the net cooperative good, or, if treated poorly, undermine it. Nevertheless, Kitcher argues that in the current environment, the costs that arise from social practices and institutions that disregard the interests of some of the world s population are so severe that a more cosmopolitan morality is actually prudential. Emphasizing the strategic capacities of oppressed and marginalized groups, Kitcher contends that inegalitarian distributions cannot be long maintained given the technological possibilities for violent retaliation now increasingly available to the poor and oppressed (311). Kitcher seems to believe that recent expansions of our moral circle has begun to expand, and that it will continue to do so, are due to the presence of ecological conditions that make such expansions fitness enhancing or advantageous. 11 On the other hand, if Kitcher s notions of altruism failure and moral progress are not descriptive but rather normative, as he has suggested in personal communication, then Kitcher must advert to a moral theory whose content and force is independent of the evolutionary function of morality. In this case, it is unclear what role the evolutionary function is playing in Kitcher s substantive ethical project, particularly if it does little to constrain the shape of human moralities (see 2). 17

18 As we see it, there are two problems with this view. First, it clearly cannot account for one dramatic departure from morality as cooperative-group reciprocity: The growing recognition that there are moral constraints on our treatment of nonhuman animals, which lack strategic capacities and whose unrestrained exploitation continues to have advantages (e.g., economic). The fact that Kitcher (2011) recognizes the difficulty that the animal welfare movement poses for his view suggests that Kitcher s account of ethics is more closely bound to cooperative groups than he has acknowledged. Second, under present and foreseeable conditions, the costs of social practices and institutions that discount or disregard the interests of the world s worst off people fall disproportionately on the world s worst off. It hardly seems likely that the richest societies are suffering any major disadvantages or a loss of reproductive fitness because of their support for the deeply inegalitarian global order. One might argue that the exploitation of vulnerable populations could lead to terrorism and other forms of blowback against powerful nations and that this gives powerful nations a wholly self-interested reason not to exploit vulnerable peoples but rather to bring them into the cooperative fold. But the empirical linkages here are too dubious and contingent to ground a global expansion of the moral circle. Thus, inclusivist morality is not merely a scaled up contemporary version of the strategic, group-restricted morality that arose in the EEA. In sum, standard selectionist explanations of morality not only fail to cite ecological conditions and selection pressures that can explain these inclusivist features of modern morality they render these features inexplicable (cf. Lazari-Radek and Singer 2012) Could inclusivist features of morality, which entail doling out evolutionarily excessive doses of altruism, be explained instead as a product of sexual selection: That is, as hard-to-fake signals of vigor that are appealing to the opposite sex in the mating arena? To make this case, one would need to show that acquiring inclusivist moral traits supplies an advantage in sexual competition that outweighs its straightforward costs to fitness outside of the mating context, and that this advantage has resulted in the prevalence of these traits in human populations. As neither of these claims seem plausible to us, we will not examine them here. 18

19 4. Byproduct Explanations of Inclusivist Morality Even if selectionist explanations of inclusivist morality fail, the latter could still be afforded an evolutionary explanation if it can be shown to be a byproduct of other adaptive features. For instance, some theorists argue that cultural moral norms, such as the incest taboo, are not objects of selection in their own right but incidental byproducts of disgust reactions and other moral sentiments (Lieberman 2008; see also Prinz 2008). In this section we will consider whether inclusivist morality can be given an evolutionary byproduct explanation and, if so, whether this might have any evoconservative implications. We will consider three types of byproduct explanation that might be put forward to account for inclusivist morality. The paradigmatic byproduct explanation is what we will call a causal byproduct explanation. This describes the scenario in which one trait is causally linked to another trait that is selected for, thereby hitchhiking its way to populational prominence. We might say that functionless trait X is an evolutionary causal byproduct of adaptive feature Y only if X is causally related to Y such that when Y occurs X reliably accompanies it. The term byproduct is often used in the evolutionary psychological literature as a catchall for any trait that cannot be given a plausible selectionist explanation (Buss et al. 1998), such as art, music, and science. However, a causal byproduct explanation must do more than simply show that some other type of explanation is implausible; it must provide a positive account that meets the standards of adequacy for scientific explanation. In their classic architectural spandrel analogy, Gould and Lewontin (1979) compared (initially) functionless byproducts to the unavoidable, roughly triangular, geometric space created by resting a dome on top of contiguous arches. A spandrel in the evolutionary sense is 19

20 any necessary, predictable side consequence of selection for another trait, be it genetic, structural, physiological, genetic cognitive or behavioral (Gould 1997). An instructive example relates to the large and fully erectile clitoris of the female spotted hyena, which is comparable in size to the male counterpart s penis and is explained as a byproduct of selection for increased aggression (Gould 1983). The causal pathway from adaptation to byproduct is postulated to run as follows: Female hyenas that are more aggressive tend to be socially dominant and thus able to commandeer more resources for their offspring; consequently, hyena populations experienced selection for increased female aggression; increased aggression in mammals is typically produced by increasing levels of testosterone; and a direct side effect of increased levels of testosterone in females is an enlarged clitoris. Here, the same reasonably well understood proximate mechanisms that produce the underlying adaptation (aggression dominance) are also shown to reliably produce the byproduct (a hypertrophied clitoris). So far as we are aware, no one has so much as sketched such a causal pathway in the case of inclusivist morality. Just as altruism is not an unavoidable byproduct of nepotism, inclusivist morality is not an unavoidable byproduct of group-restricted morality. Indeed, humans were perfectly capable, for hundreds of thousands of years, of restricting the universalizability of their moral judgments to members of their own group. aand without itin the absence of a description of the relevant causal connections, there is no basis to reach any evoconservative conclusions about the plausibility or ultimate potential scope of moral inclusivity. Whether the sympathy or other-regard one feels for kith and kin will translate into sympathy or other-regard felt toward all human beings and even nonhuman animals will depend on the conceptual, cognitive and affective motivations that underlie group restricted morality and how they interact with the social environment. In the absence of these details, 20

21 there is simply no basis to say that the extension of sympathy or other-regard to strangers or out-groups is inevitable or likely. Indeed, humans were perfectly capable, for hundreds of thousands of years, of restricting the universalizability of their moral judgments to members of their own group. Just as altruism is not an unavoidable byproduct of nepotism, inclusivist morality is not an unavoidable byproduct of group-restricted morality. More importantly, even if we set aside the matter of causal linkages, inclusivist moral features are not plausible candidates for byproduct explanation because they have not reliably accompanied any of the putatively relevant adaptations thought to have arisen in the EEA. For tens or hundreds of thousands of years, human beings possessed the whole suite of cognitive and emotional adaptations that plausibly underpin morality such as capacities for norm following, perspective taking, preference for consistency in belief, and the parochial altruism characteristic of group-restricted morality. And yet very few if any human beings exhibited inclusivist moral features until very recently in human history. Further, there are still many people, and even entire cultural groups, whose morality lacks one or more of the above inclusivist features. This time lag problem is fatal to causal byproduct explanation here, 13 for it shows that inclusivist morality is not a necessary, inevitable, predictable, enjoined or even highly likely result of selection for group-restricted morality or any other adaptation listed above, and thus it is not amenable to causal byproduct explanation. Therefore, even a complete description of ecological conditions of the EEA, in conjunction with evolutionary laws (such as natural selection), would not confer a high 13 The power of this point can be illustrated by imagining a similar pattern in the context of a paradigmatic byproduct explanation, such as the hyena s clitoris: If there was a one hundred-thousand-year temporal gap between increased hyena aggression and clitoral enlargement, this would completely vitiate the byproduct explanation, as selection for increased aggression would no longer be sufficient for, or confer a high probability on, or be a difference-making cause of, the hypertrophied clitoris. 21

22 probability on inclusivist morality neither as an adaptation nor as a causal byproduct. Accounts of evolutionary explanation that advert to difference-making causes fare no better in explaining the proliferation of inclusivist morality as a causal evolutionary byproduct. Why do humans increasingly exhibit inclusivist morality rather than more truncated forms of morality? The prehistoric evolution of basic moral capacities may be a precondition for the more recent emergence of inclusivist moral features in the hominin lineage, just as a nervous system is a precondition for the development of complex cognition in general but this is a far cry from an explanation. The existence of basic moral adaptations may help to explain why humans exhibit inclusivist moral capacities while, say, chimps (which presumably lack these basic moral adaptations) do not; but it does not help to explain why some behaviorally modern humans exhibit inclusivist moral features while other behaviorally modern humans do not, since both possess basic moral adaptations. Whatever the crucial difference-makers here might be, they will not be evolved psychological capacities. Formatted: Font: 10 pt A second type of byproduct explanation involves selection for some generic or overarching capacity, which in turn enables the development of some lower-level or nested capacity. For instance, one might describe astrophysics as a byproduct of selection for symbolic thought (which is often thought to be associated with the evolution of language). In the case of inclusivist morality, the claim would be that a range of generic adaptive capacities, such as reasoning, theory of mind, norm-following and so on, in conjunction with as-yet-unspecified sociocultural circumstances, conspire to produce inclusivist morality as a byproduct. Those who endeavor to provide this sort of byproduct explanation of inclusivist morality are in good company. In The Descent of Man, Darwin not only offered a proto-selectionist 22