20th WCP: Evolutionary Ethics and Biologically Supportable Morality

Introduction

The term ‘evolutionary ethics’ denotes an approach to naturalistic moral philosophy which seeks to explain how moral traits and behavior evolved. Sophisticated versions of evolutionary ethics do not argue that the moral judgments of each and every individual can be predicted given only the tenets of evolutionary theory. Rather the aim is usually to show that human beings possess moral traits because such traits confer a selective advantage. The motivation for this kind of view lies in a broader naturalism: if moral philosophy is to be founded on a naturalistic understanding of human beings and their place in the world, and if evolutionary theory gives us the best (kind of) account of the natural history of human beings, then moral philosophy will need to be brought in line with (some version of) evolutionary theory. Shaping moral theory so that it is possible to explain the selective advantage of moral traits and behavior is thus the vocation of evolutionary ethics.

One of the intriguing problems confronting evolutionary ethics is to solve the apparent paradox of altruism. According to evolutionary theory, natural selection entails that in general only the fittest individuals in any given biological population will survive and reproduce. An organism’s evolutionary telos, or goal, seems to be to promote its own fitness in order to survive long enough to reproduce. In situations where an organism confronts a choice between enhancing its own fitness and enhancing the fitness of others, it would seem to follow that the organism will (or "ought to," or should be expected to) choose to enhance its own fitness. (1)

The paradox arises because empirical facts seem to contradict this prediction of evolutionary theory. In a wide range of cases, and among organisms of differing levels of sentience and sapience, individual organisms frequently behave in ways that promote the fitness of some group (especially but not always a kin group) at the expense of their own individual fitness. This phenomenon, which Elliott Sober labels "evolutionary altruism," calls for reconciliation with evolutionary theory. (2) Although it might be a mistake to think that moral traits among humans is coextensive with their actions that would correctly be described as altruistic in an evolutionary sense, it is probably true that some morally required actions promote the fitness of other people at the expense of the fitness of the individual. (3) Hence the paradox of altruism appears to have implications for evolutionary ethics as well, even if the relationship between morality and altruism needs further elucidation. If some moral traits are altruistic in the evolutionary sense, then the evolutionary explanation of altruism will constitute a part of the explanation of morality.

Darwin’s own response to the paradox of altruism was to tweak evolutionary theory. He appears to have thought that it was a mistake to think that natural selection operates in every case on individuals; sometimes selection can operate on groups. (4) This conception of group selection" answers the paradox of altruism by explaining why altruistic behavior can promote an organism’s evolutionary telos. The selective advantage of the group over competing groups (which do not contain altruistic members) entails improved prospects of survival and reproduction for group members. (5) In this way, it can be shown that the evolution of altruistic behaviors is consistent with natural selection, provided that we reinterpret natural selection to allow group selection.

I wish to argue that something is amiss with the picture that is developing here: group selection cannot explain the evolution of morality, at least not universalistic morality. Morality is typically understood to involve elements that take into account just one group, namely the group of all moral beings, whereas group selection essentially involves many competing groups. At bottom, morality is a one-group phenomenon and group selection is a many-group phenomenon. After spelling out just what the problem is, I will offer several suggestions about how to overcome it.

Evolutionary Altruism

Sober distinguishes three concepts, which he calls ‘evolutionary altruism’, ‘psychological altruism’, and ‘morality’. (I’ll return to ‘morality’ shortly.)

Evolutionists speak of altruism and selfishness. The terms also occur in ordinary language and in the theories of social psychologists. One difference between the evolutionary and the pyschological concepts is straightforward. You don’t have to have a mind to be altruistic or selfish in the evolutionary sense. Evolutionary altruists confer fitness benefits on others at their own expense. (6)

Sober goes on to discuss the example of a plant that leaches insecticide into the surrounding soil. If a neighboring plant benefits from the presence of the insecticide as much as the plant that produces it, then the leaching plant is altruistic and the free-riding plant is selfish. This is true even though neither has a mind.

Sober observes that evolutionists have devised several possible explanations of "helping behavior," whether this kind of behavior counts as evolutionary altruism in a particular instance or not. Kin selection, group selection, and "reciprocal altruism" are all "possible evolutionary scenarios" that explain such helping behavior. Not all helping behavior will count as altruism in the evolutionary sense, however. Since a parent’s reproductive success depends on the survival of its offspring, helping one’s own offspring counts as evolutionary selfishness. So does reciprocal altruism, which confers a (net) fitness benefit. (7)

In general, evolutionary altruism has two necessary conditions: a trait is altruistic in the sense in question if (1) group members who have the trait are less fit than those who lack it; and (2) groups of altruistic individuals (or altruistic groups) have greater average fitness than groups of selfish individuals (selfish groups). (8)

It will be crucial to my argument later to observe that the evolution of this kind of altruism requires group selection. In particular, the existence of evolutionary altruism presupposes competition among groups, as the second condition entails the existence of both altruistic and selfish groups, with a selective advantage conferred on the former. Sober observes this interesting consequence of evolutionary altruism: if it can be shown that human beings evolved by strictly individual selection, then we cannot possess any traits of evolutionary altruism.

Psychological altruism, in contrast to the evolutionary variety, requires possession of a mind, and in particular it requires the capacity to be motivated. Sober represents motivational structures in terms of preference sets, whose members may be "other-directed" or "self-directed" preferences, depending on whose welfare is predominantly at stake (though a preference may be both other- and self-directed). (9) Extreme altruists always seek to satisfy other-directed preferences, whether doing so satisfies self-directed preferences or not. Extreme egoists, contrarily, always seek to satisfy self-directed preferences. Moderate altruists try to act so as to satisfy both other- and self-directed preferences, but when push comes to shove and a choice must be made, other-directed preferences generally prevail. Similarly, moderate egoists also try to act in ways that satisfy both other- and self-directed preferences. But if they must choose, self-directed preferences win.

It is not especially important to our discussion whether a given action is easily recognizable as altruistic or egoistic. Third-person and even first-person classifications may be difficult. And, as Sober notes, "These four motivational structures do not describe enduring personality types. Rather, they describe the ways an individual might be motivated in a given choice situation." (10) Sober points out two features of this model of motivation: (a) agents choose so as to maximize preference satisfaction. This feature is also an axiom of decision theory, which supplies tools for modeling evolutionary scenarios (as I discuss below). Second, (b) all forms of psychological altruism and egoism entail that, given a choice between benefitting both oneself and another on one hand, and denying a benefit to both oneself and another on the other hand, individuals will choose the former.

Notice that the two forms of altruism, evolutionary and psychological, are conceptually independent. The proximal mechanisms of motivated behavior are distinct from their distal explanations in evolutionary terms. Sobel discusses all of the permutations, but consider just two of them. Suppose that we humans are evolutionary altruists. We might also be psychological altruists, and if so then we would give more weight to our other-regarding preferences in situations where those conflicted practically with our self-regarding preferences. But with the same evolutionary disposition, we might instead be psychological egoists. As Sober describes it, "Suppose we have a psychology in which we act only to maximize our own pleasure and minimize our own pain, but we happen to take pleasure when others are well-off and feel pain when they are not. Under this scenario, we would be pychological egoists through and through, while all the time being evolutionary altruists." (11) So even if we were to discover that human beings are all psychological egoists, that alone would not settle the question of whether we were evolutionary altruists.

Conclusions About Morality

The most common objection to evolutionary ethics is to present some version of the "naturalistic fallacy" or to argue that it is impossible to derive a normative "ought" from a descriptive "is." (12) The worry is that the descriptive hypotheses that constitute evolutionary theory can never issue in the kind of normative prescriptions that constitute—or at least follow from—morality. It may be difficult to see, for example, how the fitness-enhancing character of a behavior entails that I ought morally to engage in that behavior. Evolutionary ethicists have struggled with this objection, but I am not concerned either to press or answer it here.

Sober offers a different objection when he notes that evolutionists have usually been concerned with morality in a very special way. They are often especially concerned to explain the evolution of the self-sacrificial behavior that morality sometimes requires or praises. Some evolutionists have resorted to group selection hypotheses to explain this behavior, whereas others have instead sought to demonstrate a selfish payoff. As Sober remarks, however, neither approach succeeds in explaining morality. "Belief in morality may promote self-sacrifice, but so may many other proximal mechanisms. The evolutionary story just described does not explain why morality evolved rather than some other proximal mechanism that could have achieved the same end." (13) Sober sets a high standard for explaining morality: a successful explanation shows not only that morality could have evolved, it also explains why morality and not, say, "schmorality" evolved (where "schmorality" names a collection of traits and behaviors that would also explain self-sacrificial behavior). In short, what has been explained is at most the possibility of morality or the compossibility of morality with evolution, not the existence of morality.

My concern, however, takes a somewhat different form. I’m willing to grant the evolutionary ethicist that a normative "ought" may be derived from a descriptive "is," and I’ll settle for an explanation of the possibility that morality evolved. I think that evolutionary ethics still faces a serious problem.

Suppose that evolutionary altruism evolved, and hence that group selection is responsible for it. Suppose further that the relationship between morality and evolutionary altruism can be specified. Now, evolutionary altruism confers fitness benefits on the members of the group who possess the altruistic trait, and these benefits presumably provide a selective advantage over other groups. The benefits of altruism thus devolve on one’s fellow group members at the expense of conspecifics of other groups.

The person who acquires a reflective awareness of this fact about evolutionary altruism is in a bind. For distinctively moral concern is often taken to extend not only to members of one’s group, but to all moral beings as such (whatever beings that may include). (14) Moral traits and behaviors may have evolved because they conferred a selective advantage on groups who possessed them. But when moral agents try to act from consideration for the well-being of all moral beings, they may become aware that they are in fact acting so as to promote the fitness of their own group (whatever that may be) at the expense of other groups of moral beings. In short, morally motivated agents seem unable to exhibit truly moral concern for the well-being of moral beings outside their group. In a way, reflective moral agents cannot be moral if morality evolved by group selection.

I see three ways out of this position. First, we could develop an alternative hypothesis to group selection in order to explain morality. This project seems especially difficult and unappealing—what else could selection act on other than individuals and groups?—and it does not appear to be a live option in current evolutionary theory.

Second, we could conclude that morality did not evolve. This solution, which a number of evolutionists and sociobiologists embrace, ought to be taken very seriously. Some theorists simply contend that morality is a product of culture rather than evolution, and it serves to keep our baser instincts in check. A more sophisticated line of thought might be that morality expresses features of human nature that evolved for other purposes, but which humans with their large brains adapted to moral ends. Finally, I think evolutionists like Richard Dawkins, who discuss evolution in terms of genes rather than individual organisms and species, might fall under this position. For if one is an eliminativist about behavior, then the problem of explaining the evolution of distinctively moral behavior will not arise. (15)

Third, we might try to argue that our beliefs about the character of morality, the scope of the moral community, and the nature of moral concern are all radically mistaken. If one adopted an error theory, then the claim would be that our beliefs about the scope of moral community were necessary for moral traits to evolve, even though they were (and are) false. On the other hand, one might instead seek to show that universalistic conceptions of morality are not biologically supportable. The idea here would be to suggest that philosophers have created the problem with their view that moral concern must extend to all moral beings rather than predominantly to one’s group. A revisionist approach to the explanatory ends of moral theory could help bring morality in line with evolutionary theory. But whatever way we go, the work of evolutionary ethics is just beginning.

Notes

(1) My use of the term "choice" at this point should be interpreted loosely: an organism with a choice is in a position to exhibit two distinct forms of behavior and subsequently exhibits one of them. No intentionality is presupposed. For all I mean by it here, phototropism exhibits choice.

(2) See the discussion in Elliott Sober, From a Biological Point of View (Cambridge: Cambridge University Press, 1994) and especially Elliott Sober, "Evolutionary Altruism, Psychological Altruism, and Morality: Disentangling the Phenotypes," in Evolutionary Ethics, ed. M. H. Nitecki and D. V. Nitecki (Albany: SUNY Press, 1993).

(3) But see below for a discussion of the perils of drawing conclusions for morality from the theory of evolutionary altruism.

(4) See Charles Darwin, On the Origin of Species (London: Murray, 1859), chap. 7. The literature concerning the correct interpretation of Darwin is vast and convoluted, and so it is a vexed question whether Darwin really offers a hypothesis of group selection (here or elsewhere). Nothing for my argument hangs on the resolution of this exegetical question; others besides Darwin have suggested that altruism evolved by group selection, and on Sober’s account of it, evolutionary altruism evolved only if group selection occurred.

(5) That is, if a group truly enjoys a selective advantage over competing groups, it follows that group members are more likely to survive and reproduce than members of competing groups. This entailment is part of the logic of the term ‘selective advantage’.

(6) Elliott Sober, "Evolutionary Altruism, Psychological Altruism, and Morality: Disentangling the Phenotypes," pp. 204–205.

(7) For the introduction of the concept of reciprocal altruism, see R. Trivers, "The Evolution of Reciprocal Altruism," Quarterly Review of Biology 46:35–57.

(8) Sober, op cit, p. 205.

(9) Ibid, p. 207.

(10) Ibid, p. 210.

(11) Ibid, p. 211.

(12) R. J. Richards claims to derives an "ought" from an "is." Unfortunately, the "ought" he derives is a predictive/explanatory "ought" rather than a normative "ought." The sense in which water ought to boil at 100 C is not the same as the sense in which I ought to take care of my children. See R. J. Richards, "Birth, Death, and Resurrection of Evolutionary Ethics," in Evolutionary Ethics, ed. M. H. Nitecki and D. V. Nitecki (Albany: SUNY Press, 1993).

(13) Sober, op cit., p. 212.

(14) I don’t mean to suggest that one’s duties are the same to everyone or that everyone’s well-being figures equally into the moral agent’s deliberations. The point is just that, on one ordinary theoretical conception of morality, moral beings enter the deliberative picture in ways that nonmoral beings do not.

(15) Dawkins, of course, has difficulties of his own, including the challenge of explaining the sense in which genes "behave" (e.g., they "strive"). See Richard Dawkins, The Selfish Gene (Oxford: Oxford University Press, 1976) for the basic view.