Primates and Philosophers_How Morality Evolved

by Frans de Waal

  Appreciating this point ought to be the first stage in an inquiry about the evolutionary history that links the psychological dispositions of our ancestors to the capacities that underlie our contemporary moral behavior. De Waal demolishes his favored version of Veneer Theory by being very clear about the starting point—that, after all, is a project to which he has devoted much of his life—but he is considerably less clear as to the nature of the terminus. The vague talk about “building blocks” and “direct outgrowth” comes in because de Waal hasn’t thought as hard about the human phenomenon he takes to be anticipated or foreshadowed in chimpanzee social life.

  There’s a polar opposite of Veneer Theory, one we might call “Solid-to-the-Core Theory” (STCT, for short). STCT claims that morality is essentially present in our evolutionary ancestors. Perhaps in the glory days of human sociobiology some people were tempted to flirt with STCT, supposing, for example, that human morality reduces to dispositions to avoid incest (and similar simple tendencies) and that these have evolutionary explanations that apply to a wide range of organisms.1 STCT effectively takes the terminus of the evolutionary process that yields human morality to be the same as some prehuman starting point. It is no more, but no less, plausible than Veneer Theory as de Waal characterizes it. All the interesting positions lie somewhere in between.

  De Waal prefaces his lectures with a quotation from the late Stephen Jay Gould, indeed from a passage in which Gould was responding to sociobiological accounts of human nature. I think it’s worth reflecting on another observation of Gould’s, the comment that when we utter the sentence “Human beings are descended from apes” we can change the emphasis to bring out either the continuities or the differences. Or, to vary the point, Darwin’s phrase “descent with modification” captures two aspects of the evolutionary process: descent and modification. What is least satisfactory about de Waal’s lectures is his substitution of vague language (“building blocks,” “direct outgrowth”) for any specific suggestions about what has descended and what has been modified. Lambasting a view like his “Veneer Theory” (or like STCT) is not enough.

  III

  In fact, de Waal provides a little more than I have so far granted. He has been attuned to developments in evolutionary ethics (or in the evolution of ethics) during the past fifteen years, a period in which the naive reductions favored in sociobiological accounts have given way to proposals of an alliance between Darwin and Hume. The sentimentalist tradition in ethical theory, in which, as de Waal rightly sees, Adam Smith deserves (at least) equal billing with Hume, has won increased favor with philosophers. As it has done so, would-be evolutionary ethicists have felt the appeal of what I shall call the “Hume-Smith lure.”

  The lure consists in focusing on the central role of sympathy in the ethical accounts offered by Hume and Smith. So you first claim that moral conduct consists in the expression of the appropriate passions, and that sympathy is central to these passions. Then you argue that chimpanzees have capacities for sympathy, and conclude that they have the core of the psychology required for morality. If there are worries about what it means to talk about the “central” role of sympathy or the “core” of moral psychology, the primatologist or evolutionary theorist can shift the burden. Hume, Smith, and their contemporary champions sort out the ways in which sympathy figures in moral psychology and moral behavior; the primatologists demonstrate the sympathetic tendencies at work in primate social life; the evolutionary theorists show how tendencies of this type might have evolved.2

  My characterization of this strategy as “the Hume-Smith lure” is supposed to signal that it is far more problematic than many writers (including some philosophers, but especially nonphilosophers) take it to be. To understand the difficulties we need to probe the notion of psychological altruism, recognize just what types of psychological altruism have been revealed by studies of primates, and relate these dispositions to the moral sentiments invoked by Hume, Smith, and their successors.

  De Waal wants to recognize nonhuman primates as having dispositions that are not simply egoistic, and it’s useful to think of “psychological altruism” as a catchall term for covering these. As I understand it, psychological altruism is a complex notion that involves the adjustment of desires, intentions, and emotions in light of perceptions of the needs and wishes of others. De Waal rightly distinguishes the psychological notion from the biological conception of altruism, defined in terms of the promotion of others’ reproductive success at reproductive cost to oneself; as he points out, the interesting notion is one that only applies in the context of intentional behavior, and it can be disconnected from any thought of assisting the reproductive success of other animals.

  More precisely, psychological altruism should be thought of in terms of the relation among psychological states in situations that vary according to the perception of another’s need or desire. Although an altruistic response can consist in modification of emotions or intentions, it may be easiest to introduce the concept in reference to desire. Imagine an organism A, in a context in which the actions available have no perceptible bearing on another organism B, and suppose that A prefers a particular option. It may nonetheless be true of A that, in a context very similar to the original one, in which there is a perceptible effect on B, A would prefer a different course of action, one that A takes to be more conducive to the wishes or needs of B. If these conditions are met, then A meets a minimal requirement for having an altruistic disposition towards B as a beneficiary. The conditions are not sufficient, however, unless it is also the case that A’s change of preference in the situation where B’s interests are an issue would be caused by A’s perception that the alternative action accorded more closely with B’s desires or needs, and, furthermore, that the switch was not generated by a calculation that pursuing the alternative would be likely to satisfy others of A’s standing preferences. All this is a way of spelling out the thought that what makes a desire altruistic is a disposition to modify what is chosen in a situation where there is a perceived impact on another, that the modification aligns the choice more closely with the perceived wishes or needs of the other, that the modification is caused by the perception of those wishes or needs, and that it doesn’t involve calculation of expected future advantages in satisfaction of standing preferences.

  An illustration may help. Suppose that A comes across an item of food, and wishes to devour it all—that is, in the absence of B, A would devour it all. If B is present, however, A may choose to share the food with B (modifying the wish that would have been operative in the context in which B was absent), may do so because A perceives that B desires some of the food (or maybe that B needs some of the food), and may do so not from calculating that sharing will bring some further selfish benefit (for example, that B will then be likely to reciprocate on future occasions). Under these circumstances, A’s desire to share is altruistic with respect to B.

  We can think of the same structure as applying in the case of emotions or of intentions—a modification of the state that would have been present that is caused by the perception of the wants or needs of the other and that does not come about through the calculation of future benefit. Yet even if we restrict attention to the case of altruistic desire, it should be plain that there are many kinds of psychological altruism. As my disjunctive formulation, “wishes or needs,” already suggests, an altruist may respond either to the perceived wants or to the perceived needs of the beneficiary. Typically, these are likely to be in harmony, but, when they diverge, altruists have to choose which to follow. Paternalistic altruism responds to the needs, rather than the wishes; nonpaternalistic altruism does the reverse.

  Besides the distinction between paternalistic and nonpaternalistic altruism, it’s also important to recognize four dimensions of altruism: intensity, range, extent, and skill. Intensity is marked by the degree to which the altruist accommodates the perceived desire (or need) of the beneficiary; in the food-sharing illustration it’s easy to present thi
s concretely, as the fraction of the item the altruist is willing to assign the beneficiary.3 The range of altruism is marked by the set of contexts in which the altruist makes an altruistic response: to take an example from de Waal, two adult male chimpanzees may be willing to share across a range of situations, but, if the stakes become really high (with the possibility of monopolizing reproductive access, say), an erstwhile friend may act with utter disregard for the other’s wishes or needs.4 The extent of altruism is expressed in the set of individuals towards whom an altruist is prepared to make an altruistic response. Finally the skill of the altruist is measured by the ability to discern, across a range of situations, the real wishes of the intended beneficiary (or, for paternalistic altruists, the real needs of the intended beneficiary).

  Even if we ignore the complications of elaborating a similar approach to emotion and intention, and even if we disregard the distinction between paternalistic and nonpaternalistic altruism, it’s evident that psychological altruists come in a vast array of types. If we think of a four-dimensional space, we can map “altruism profiles” that capture the distinct intensities and different skills with which individuals respond across a range of contexts and potential beneficiaries. Some possible profiles show low-intensity responses to a lot of others in a lot of situations; other possible profiles show high-intensity responses to a few select individuals across almost all situations; yet others are responses to the neediest individual in any given situation, with the intensity of the response proportioned to the level of need. Which, if any, of these profiles are found in human beings and in nonhuman animals? Which would be found in morally exemplary individuals? Is there a single ideal type to which we’d want everyone to conform, or is a morally ideal world one in which there’s diversity?

  I pose these questions not as a prelude to answering them, but as a way of exposing how complex the notion of psychological altruism is and how untenable is the idea that, once we know that nonhuman animals have capacities for psychological altruism, we can infer that they have the “building blocks” of morality, too. The demise of Veneer Theory, as de Waal understands it, tells us that our evolutionary relatives belong somewhere in altruism space away from the point of complete selfish indifference. Until we have a clearer view of the specific kinds of psychological altruism chimpanzees (and other nonhuman primates) display, and until we know what kinds are relevant to morality, it’s premature to claim that human morality is a “direct outgrowth” of tendencies these animals share.

  IV

  De Waal has made a powerful case for the existence of some forms of psychological altruism in the nonhuman world. His best example, to my mind, is one he offered in Good Natured, and which he retells here, the tale of Jakie, Krom, and the tires. His description shows convincingly that the juvenile, Jakie, modified his wishes and intentions from those he’d otherwise have had, that he did so in response to his perception of Krom’s wishes, and that the modified wishes were directed at satisfying her perceived desire; although hard-line champions of psychological egoism may insist that the change was produced by some cunning Machiavellian calculation, it’s extremely hard to arrive at a plausible hypothesis—Krom is a mildly retarded, low-ranking adult female who is not in any great position to help Jakie, and the idea that this might raise his standing with onlookers is scotched by the absence of other members of the troop.5 What this reveals is that Jakie was capable of a psychologically altruistic response, of at most moderate intensity (there was little cost in interrupting his activities to help with the tires), towards an individual with whom he had a standing relationship, in a context where not much else was going on.

  Other examples are a lot less convincing. Consider the capuchins, the cucumber, and the grape. When de Waal’s report of his experiments appeared, some enthusiasts were prepared to hail them as demonstrating a sense of fairness in nonhuman animals.6 I take a sense of fairness to involve psychological altruism, as I have understood it, for it depends on not being content with a situation one would have seen as satisfactory precisely because one recognizes that the needs of others haven’t been met. In fact, de Waal’s experimental study reveals no kind of psychological altruism, but simply an animal’s recognition of the possibility of a preferred reward that it has not received, and a protest that results from the selfish wish for that reward.

  In my judgment, the most convincing examples of psychological altruism are those of the Jakie-Krom type, cases in which one animal accommodates its behavior to the perception of a wish, or a need, of another animal with whom it has often interacted, or of instances in which an older animal attends to the perceived needs of the very young. These are quite enough to show that nonhuman animals aren’t invariably psychological egoists—and, indeed, to suppose that we are likely to share the same capacities and the same status. But how relevant is psychological altruism of these types to human moral practice?

  Some ability to adjust our desires and intentions to the perceived wishes or needs of others appears to be a necessary condition for moral behavior.7 But, as my remarks about the varieties of psychological altruism should have suggested, it’s not sufficient. Hume and Smith both believed that the capacity for psychological altruism, for benevolence (Hume) or sympathy (Smith), was quite limited; Smith begins the Theory of Moral Sentiments with a discussion of the ways in which our responses to the emotions of others are pallid copies. Both would probably recognize the full range of de Waal’s studies, from Chimpanzee Politics through Peacemaking among Primates to Good Natured, as vindicating their central points, showing (in my terms) that psychological altruism exists, but that it is limited in intensity, range, extent, and skill.

  Far more importantly, they would distinguish this first-order psychological altruism from the responses of the genuinely moral sentiments. Hume’s Enquiry Concerning the Principles of Morals closes with the identification of the moral sentiments with “the party of humanity.” I interpret him as supposing that we have a capacity for refining the original, limited, dispositions to respond to the wishes and needs of our friends and children. Through proper immersion in society, we can be brought to expand our sympathies, so that we eventually become moved by what is “useful and agreeable” to people, not only when that conflicts with our selfish desires but even when it is at odds with our more primitive, locally partisan, altruistic responses.

  Smith is far more explicit than Hume about how this enlargement of sympathy should proceed. He takes it to involve reflecting upon—mirroring—the judgments of those with many different perspectives around us, until we are able to combine each point of view, with its peculiar biases, into an assessment that expresses a genuinely moral sentiment.8 Without the impartial spectator, Smith’s “man in the breast,” we only have our limited and idiosyncratic sympathies, types of psychological altruism that may be necessary if moral responses are to develop in us but that fall a long way short of morality.

  So I think the Hume-Smith lure is just that. It’s an invitation to students of animal behavior to demonstrate psychological altruism in their subjects, on the assumption that any kind will do, because “Hume and Smith have shown that altruism is what morality is all about.” I think a lot more work needs to be done. Fortunately, de Waal’s studies are valuable in showing us how it might proceed.

  V

  The role of Smith’s impartial spectator (or of Kant’s inner reasoner, or of a number of other philosophical devices for directing moral behavior) is especially evident in cases of conflict. The most obvious conflicts are those that pit a selfish impulse against an altruistic one. In these cases, you might think, the verdict of morality is that the altruistic one should win, so that a key step in the evolution of ethics is the acquisition of some capacity for psychological altruism. But that is far too swift. We need impartial spectators (or some equivalent) because our altruistic dispositions are too weak, often of the wrong kinds, and because conflicting altruistic impulses need adjudication.9 What happens when there’s no internal adjudi
cator can be seen if we consider de Waal’s earlier studies in light of his later defenses of psychological altruism.

  Chimpanzee Politics and Peacemaking among Primates reveal social worlds in which there are limited forms of psychological altruism. The societies are divided into coalitions and alliances, within which, some of the time, the animals cooperate. Some of the cooperation may be based on the identification of future advantages, but there are occasions on which the hypothesis that one animal is responding to the needs of another without calculating future benefit appears quite plausible. If you try to plot the psychological altruism manifested on the dimensions I delineated above, you’ll find that de Waal’s chimpanzees (the species for which there are most data) are quite limited in the intensity, range, and extent of their altruistic tendencies.

  The limitation on range is especially important because, as Peacemaking among Primates makes especially vivid, cooperation among these animals, and the psychological altruism that often underlies it, is always breaking down. When an ally fails to do his part, the social fabric is torn, and has to be repaired. De Waal documents the time-consuming ways in which primates reassure one another, the long bouts of grooming, for example, that follow ruptures within alliances.

  If you look at this behavior through the eyes of Adam Smith—both moral philosopher and social theorist—there’s an obvious thought. These animals could use their time and energy much more efficiently and profitably than they do, were they to have some device for extending and reinforcing their dispositions to psychological altruism. A “little chimp in the breast” would provide them with a smoother, more functional society, with greater opportunities for cooperative projects; maybe they could even interact with animals whom they didn’t see on a daily basis, and their group size could grow. Because they have some forms of psychological altruism they are able to have a richer social organization than most other primate species. Because those forms of psychological altruism are so limited they are socially stuck, unable to achieve larger societies or more extensive cooperation.