Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives


SRB Archives

This review appeared in Volume 8 (1) of the Semiotic Review of Books.

Animal Morality

Anne C. Zeller

Good Natured. The origins of right and wrong in Humans and other Animals. By Frans de Waal. Cambridge, MA. Harvard University Press 1996. ISBN 0-674-35660-8. 296 pp.

De Waal's book has an explicitly stated theme: "an investigation into how social environment shapes and constrains individual behaviour" (10). He begins it by references to T. Henry Huxley's 1893 Lecture on Evolution and Ethics which was one of the first academic discussions of morality to take our animal past into account. Huxley saw human morality as a sword to slay our animal past, and nature as morally indifferent. Recent thinkers such as George Williams extended this idea with the conception that because animals do not share human moral standards, morality is a biological stupidity, since he feels it contradicts the `me first' principle. There have been many alternative view points discussed in the century since Huxley but for de Waal they can be summarized in Dewey's metaphors of what the Gardener is doing in the garden - cultivating the plants of morality or rooting out the weeds of our bestial animal nature. However, to de Waal this is really not the central question. He takes it as given that morality is not a veneer superimposed on human nature, but an essential core of our existence as social animals. I feel that the evidence he presents in the concluding chapter of a neurobiological function in the brain which controls conscience, lying, reliable behaviour etc. could be more usefully presented here as a foundation for his position rather than after the argument has been completed. DeWaal spends the next 218 pages discussing the rationale for this argument and providing data from a wide range of observations on various species of primates, primarily macaques and chimpanzees, as well as from other researchers' work on a variety of species. This material is illustrated by a large number of photographs which are a definite addition to the book.

He does discuss Biology's focus on function at this point and approaches it from the perspective that since altruism does occur among animals it is clearly a product of natural selection even if it appears to be as contradictory to the principle of selection as the ability of bumble bees to fly. In order to explore the premise of the animal and adaptive bases of moral behaviour he gives a brief reference to the past investigations of animal nature in ethology, but moves quickly into a discussion of what he calls the "New Improved Sociobiology". This he contrasts with "Calvinist Sociobiology" which has as its foundation the dual natures of humanity: its bestial ancestry and the achieved morality of the angels. We must battle our genes (our original sin) to live a moral existence. This viewpoint is presented as a manifestation of gene-centered sociobiology in which all behaviour is directed towards genetic evolutionary success, whether it is theft or Mother Theresa's efforts in the slums of Calcutta. In other words, it does not matter how altruistic the behaviour appears, its true function is the promotion of an individual's own genetic survival. The "New Improved Sociobiology" which de Waal espouses is based to a large extent on Trivers' formulation of Reciprocal Altruism. In this approach behaviour arises not directly from genetics, but on a foundation of co-operation for continued individual survival, which itself may have genetic underpinnings. The underpinning permit cooperative social behaviour to arise in much the same way that genetics permits language to arise in humans. These foundations do not mandate the behaviour nor constrain its variability. Therefore, it is possible to have a tremendous range of social behaviours, from extreme conflict to mutualism and symbiotism. For social animals, life in a social group is a necessity, and in order for society to survive concepts of right and wrong must exist. Otherwise the grouping of animals is not a social unit but merely an aggregation of individually competing animals. Social groups are often based on mutually supporting kin who promote their own and each other's survival. But groups are rarely made up only of kin, and interactions with others in the group are necessary.

Here is my own personal difficulty with an argument that requires both kin selection and reciprocal altruism in order to cover all the observations. If you risk yourself to support kin, in order that your genes will make up a greater proportion of the next generation's gene pool than they did of yours which is a definition of evolutionary success, it is clearly counter productive to aid a non-related individual to raise his/her gene fraction in the next generation as well. Trivers seems to get around this by arguing for a much more complex interpretation of living systems than direct competition because living systems have integrity and value at different levels. Success is necessary at the ultimate level of genetic survival, but this level will never be reached if proximal day to day behaviour is not successful also. Individuals take their circumstances into account so as to choose the best course of action. This approach leads to the more recent idea of cognitive ethology which looks at animals at a level far more complex than the classical "instinctive behaviour", and "fixed action patterns". Animals are seen as calculating, knowing individuals and the language that is used to describe them allows both positive and negative labels. As well as using label of "suckers", "grudgers" and "cheaters" as classical sociobiolgical research frequently does, labels such as "friendship", "reconciliation" and "greeting" are also valid. De Wall argues "If animals can have enemies they can have friends; if they can cheat they can be honest.."(19). He deals with the charge of anthropomorphism in two ways. The first is the argument just used above, that if "spite" and "greed" are allowable terms then altruism and sharing must be also. The second argument is that it is the behaviour that he is categorizing rather than the nature of the animal and these behaviours occur repeatedly under conditions of reliable observation and experimental manipulation. The book is full of observations of a range of primates and other animals and accounts of experiments which support de Waal's arguments about morality. Many of these observations are anecdotal, but most of them are made by experienced researchers who have compiled data over long periods of time. Thus, this less quantitative data is still valid support for arguments, especially when it can be supplemented by well designed experiments. Why, for example, would a capuchin bring prized food from a container in the corner, unreachable by the capuchin in the next cage, over to their adjoining cage bars and drop some there, which the other reached for and ate, if it were not a manifestation of sharing, possibly done with the potential for reciprocal altruism at a future time. Chapter one ends with a brief discussion of some historical background for the idea of group selection, especially from the work of the Russian Kropotkin. He contradicted Darwin's ideas of individual struggle by observing that in very harsh climates cooperation was often much more key to survival than competition. More current ideas of group selection regard it as one level of a nested series of opportunities for selection, where competition can occur at the level of the individual, the kin line or the group. After all, if a group is necessary for survival then the maintenance of the group deserves some effort on the part of its members. Even Adam Smith came around to this view point in his later writings with the argument that each working for himself still fosters the common good. Richard Alexander has developed a term "indirect reciprocity" for how an individual could be rewarded by the group for altruistic behaviour. It builds on de Waal's argument concerning the importance of social bonds as a foundation for moral behaviour. Individuals who act altruistically find that others in the group interact with them more positively, thus encouraging such behaviour in a proximal sense as well as giving it ultimate reproductive value. When a system of mutually reinforcing positive relations is set up it tends to develop a group norm for behaviour. In order for this type of reinforcement to operate among animals they need a good memory for events, a stable relationship and the opportunity to help others. These factors underlie the lives of most group oriented primates and de Waal uses the term "community cooperation" to refer to this idea.

"I do not necessarily mean that animals make sacrifices for their community, but rather that each and every individual has a stake in the quality of the social environment on which its survival depends. In trying to improve this quality for their own purposes,they may help many of their group mates at the same time." (31)

This social input can proceed from the top down, as in the concept of "control animal" or from the lower levels up as in the development of alliances among female chimpanzees to curb the disruptive behaviour of aggressive males. Episodes of both types are described with the ensuing modified responses of those who have been subjected to social sanction.

One very interesting aspect of the development and learning of ethical principles is de Waal's suggestion that they are learned very early in life on the model of a critical learning period. He rests this argument on the observation that morally correct behaviour in any society is far too complex to be learned by trial and error and far too variable to be genetically programmed. If young animals are not raised socially they have a very difficult time learning the norms of a group (appropriate behaviour and responses to others). Also, some moral rules reinforce species typical predispositions, such as the more reconcilliative nature of stumptails macaques than of rhesus. However, when young rhesus were housed with somewhat older, larger stumptails, the rhesus gradually learned not the stumptail motor patterns of reconciliation, but the appropriate time and frequency for it according to stumptail standards. Even when the stumptails were removed after five months, the rhesus continued their modified and unrhesus-like behaviour.

After these introductory chapters examining the historical, philosophical and biological substructure for his position, de Waal sets out a list of the features which he argues underlie human moral behaviour and devotes the next five chapters to examining them. These features are sympathy, empathy, mutual aid, a sense of fairness, and mechanisms of conflict resolution. His investigation utilizes a number of examples from various species of primates which he himself has studied, such as chimpanzees, rhesus, stumptail and longtail macaques, and capuchins. In addition, he discusses lemurs, baboon and other types of animals such as dogs, elephants and dolphins. His contrasting example of a non-society-living animal is the domestic cat. Recognizing that there might be resistance to the term "sympathy" used for animals, he substitutes the term "succorance" or aid to non-offspring in an attempt to help or alleviate suffering or distress of others. This type of behaviour, he argues, probably derived from nurturant, parental care in an evolutionary sense. His definition is basically that succorance is a recognition that another is in difficulty and the impulse to do something (or not do something) about it. His clearest examples are situations of handicapped young who are given much more latitude for inappropriate behaviour than normal ones. A little rhesus female with autosomal trisomy (rather like Downes syndrome) was very slow in development and in learning to deal with her captive environment. Her mother was not particularly helpful towards her although tolerating her clinging for a long time; but her juvenile sister protected and rescued her from awkward situations. The sister, who was not full size, held and comforted Azalea, and other adults groomed her twice as much as they did her peers. Azalea did not seem to learn the potential to be hurt in aggressive encounters even after several bad experiences. Thus, after she was 2-1/2 years old she died as the result of a fight, but not because she was attacked as an individual. De Waal recounts three or four cases of similar situations in which blind, palsied, or injured infants of monkeys, apes and lemurs received extra care and tolerance from non-parental group mates. I have seen this in longtail macaques, where a palsied young individual maintained right of way on a branch, and even the adult males went around her, although her healthy age mates were threatened out of the way. Chimpanzees may go even farther by consoling individuals they have injured themselves. This can take the form of returning to the victim and inspecting and grooming the wound. The care-giver is thus seen to be sensitive to the situation of the other and act in ways which are appropriately helpful. As with reconciliation, what is at issue here is not merely an emotional claim of sympathy, but actual, appropriate behaviour in a timely fashion with a particular outcome.

The goal of studying such behaviours is to get at testable ideas and repeatable observations in order to validate the occurence of, and motivations behind, the behaviour. The principle of parsimony in explaining such behaviours is two-fold. First, it is not useful to invoke an explanation of a higher capacity if a lower one will do. In other words, is it possible to explain these types of behaviour by the principle of "learned adjustment" to changed conditions rather than by cognitive empathy; which suggests a choice to act appropriately was made? The second aspect of parsimony is that since primates have a shared evolutionary background, do similar behaviours in humans and higher primates arise from homologous or analogous processes? If they are analogous it implies the evolution of divergent processes for similar behaviour which, although not impossible, is not very parsimonious. De Waal argues that in order to investigate the problem we need to be able to at least suggest the possibility of higher cognitive processes and test predictions made on a model of homology. If, for example, the only rationale for aid was ultimate genetic value, why did other animals not take advantage of handicapped ones who were surely in no way able to retaliate.

The argument supporting cognitive processes is based on the parallel that the more complex an organism's relation is with its environment, the better it must know itself and its social surroundings. Humans can look at the world through the eyes of others (i.e. role play) which allows "emotional contagion" which implies that they can feel for others also. Several experiments with chimpanzees by Menzel and Povinelli suggest that they too can project what another individual must know, based on what type of opportunity he has had to learn. If the chimpanzee has seen one human watch where food is placed it will believe that individual rather than a blind observer when each one suggests a different possible location in which to search for food.

In children, the capacity for "cognitive empathy" appears at about the same time as mirror, self-recognition. This awareness of understanding how another might view a particular situation is a major aspect of deception which is well documented behaviour in chimpanzees. This argument does not propose that human and chimpanzee capabilities for sympathy are equivalent, but that in each case we have attachment to the victim, modification of behaviour, and some level of caring behaviour, whether it is consolation, grooming, aiding, sitting with, or not harassing. Humans clearly recognize the other's experiences as belonging to the other and this perspective-taking converts helpful behaviour into cognitive altruism. In a wide variety of human situations, sympathy and altruism are more likely to occur with those we are personally attached to, and in conditions which can afford it. The "have nots" are less likely to be altruistic than the "haves", but among humans there seems to be a positive emotional reward for aiding others, such that morality seems to function to protect and nurture sympathy. Empathy arises from a German word Einfuhlung or "feeling into" which is what underlies the capacity to recognize the other as "other". This is not necessarily a "rationalist" ability. Most primates notice how their conspecifics react to the environment and react in similar ways. This can be at the "lower" level of a "flock response" of starlings to a predator, or at the more complex level of learning about the dominance hierarchy through observation. Most primates seem to learn this without having to take or give a lesson to every single other member of the group.

This leads to a discussion of norms of behaviour. There are appropriate and inappropriate ways to interact with others based on social as well as physical factors. These qualify as "prescriptive rules" since they are upheld by reward and punishment. Sometimes behaviour that is acceptable in one context is not in another. For example, de Waal observed a situation in which the alpha male chimpanzee found a subordinate male mating and chased him off. This is acceptable behaviour. However, the alpha was very persistent in his chasing and eventually the females began to close ranks and bark and piloerect until the alpha ceased harassing the subadult. This type of social control, which has been observed a number of times, indicates that rank is a negotiated position among chimpanzees. If enough of the group harasses a high ranking individual, he will usually modify his behaviour or discover that he cannot find allies when he needs them. This respect for rules only operates if the opinions of the others matter. Among rhesus for example, alliances are usually from the top down and the sense of social regularity is different. Control behaviour by high ranking baboons and macaques is an effective indicator and regulator of social norms. I have seen an adult male barbary macaque jump on and bite an adult female who had been very rough with her young juvenile daughter over the previous few days. Immediately the mother's behaviour changed and the daughter was not harassed into screaming for the next while. Control behaviour not only breaks up potentially dangerous fights, it can force combatants to reconcile with grooming and affiliative gestures. I saw a group of barbary adult females surround two quarrelling subadult females and vocalize at them until one presented to the other who groomed her.

In many ways de Waal argues that conflicts, and their resolution is a necessary aspect of the development and maintenance of social life. Conflicts allow animals to aid each other, to reconcile with each other, and in some cases, to modify their social position with respect to each other. You do not get a complete acceptance of the "status quo". There is always a balance of two conflicting strategies, to probe at the social order looking for ways to advance your position, balanced with the conservation of the current state. This latter is usually more valued by high and low ranking individuals, while it is the mid-ranking ones who tend to push. In order to resist pressure from below, alliances are usually formed between closely ranked animals who frequently demonstrate reciprocal mutual aid. Another arena of mutual aid is food sharing. Some individuals tend to share more than others, and in more than 5000 opportunities which de Waal arranged for food sharing to occur, it was definitely reciprocal.

If you shared a lot, you got shared with a lot. This is less true of wild chimpanzees sharing meat, but females rarely had their own kills, and were shared with on a regular basis nevertheless. It is possible that this generosity translated into future favours, but this has not been clearly ascertained.

Certainly among bonobos (pygmy chimpanzees), female sexual activity is very frequent in the context of food sharing. This element of patterning, and the presence of social rules is made most evident in situations in which the rules are broken. Even adult chimpanzees who expect certain levels of sharing from others with whom they have shared in the past will go into positive tantrums of screaming and throwing themselves on the ground like youngsters do in weaning tantrums if their expectations are not fulfilled by others. In situations where a high ranking animal is extremely aggressive, lower ranking ones may form an alliance and harass the aggressor until it ceases such behaviour. Mothers spend quite a bit of time disciplining their own young and establishing the levels and types of behaviour that will be tolerated. Young animals may tease and torment older ones to learn what their reaction will be and where the boundaries are. The expected social norms may be pushed by some, but a majority of the group usually acts to maintain them.

All of these behaviours support the position that social norms exist and that considerable social unrest, including sanctions, will ensue if they are not adhered to. Rules are also seen in children's play where conflict resolution must be learned. Janet Lever's research suggests that little girls seem to avoid conflict by fostering turn-taking behaviour, and stopping the game if it appears to lead to high levels of conflict. On the other hand she says boys learn to compete and operate on a rule-based morality. De Waal wonders if this research is culturally limited, but suggests that female primates do seem to work hard at their social connections. If serious conflicts erupt between female primates they are not likely to repair the relationship, while males (especially chimpanzees) seem to use cycles of conflict and reconciliation to establish and maintain social bonds. In fact, de Waal spends some time discussing the social utility of conflict since it provides an arena for the development of social relations. Not only do the conflicting individuals have an opportunity to reconcile after a fight, but among chimpanzees the whole group responds with hoots and loud calls suggesting that the state of social harmony is important to all. This conflict resolution would probably not occur if it was not based on strong attachments and a need for cooperation. In fact, de Waal goes to the extent of stating that human aggression should be studied as a social rather than an antisocial behaviour. He argues with the Seville Statement on Violence which promotes the idea that humans should overcome their "genetically based violent nature". De Waal challenges that paradigm by saying that aggression exists at both a biological and social level and the levels of cooperation required for survival must be balanced by the competition for resources that each individual needs. He discusses the concept of a "social cage" in which even free ranging animals are constrained in their behaviour by the necessity of surviving. In captivity, social closeness is intensified because there is no escape, yet primates manage to adapt. We study this adaptability of primates in order to learn how to manage the stresses of our life. When you cannot leave a situation you must become politically astute in order to survive staying in it. This argument provides a basis for understanding why rules (read dominance hierarchy) are more clearly manifested in captive conditions than in many free range situations. In the wild there is an opportunity to be more flexible, to escape an oppressive situation, or to try to change your position. In captivity, however, because the rules and the social control maintaining those rules are much more manifest, it is more difficult to modify the system.

Conflict between attached individuals causes discomfort almost on a parallel with physical separation, according to de Waal. The strength of reunion euphoria is evident in greeting ceremonies seen in a wide range of species. Greeting behaviours are an embodiment of positive social affiliation. Reconciliation behaviours operate in the same way as can be seen in experiments in which monkeys must cooperate to achieve a common goal. In this experiment after the animals had interacted aggressively they were put in a position where they had to sit close together and cooperate for a reward. This close interaction occurred much more often if the animals had engaged in reconciliation behaviours before they had to cooperate. As the experiment continued the rate of reconciliations became much higher which allowed many more successful cooperative ventures. Reconciliation is a learned social skill, sensitive to social configuration and reflecting the strength of the bond between animals. The rate of nervous self scratching rises after conflicts, but fall after reconciliations. Peace making may also increase the amount of time that animals sit together.

Thus the focus of de Waal's argument is that conflict has its socially positive functions under circumstances where it can be resolved. This approach differs markedly from the Seville Statement on Violence which sees a biological argument concerning the function of violence as "justifying" it. However, de Waal's position is not that violence is a biologically given (as in "On Aggression") but that it is an aspect of social dynamics which if used well can restrain raw contests over resources. Since reconciliation works best where positive social bonds are strongest, and individuals are responsive to the sanctions of the whole group, clearly it can have positive group-wide consequences.

Not only does de Waal disagree with the Serville Commission's statement, he disagrees with statements about the "jungle like" aspects of inner city life promulgated by various government agencies in the United States. He argues that it is not just the crowding of cities that is the problem but the lack of all types of resources including space. De Waal comes from the Netherlands, which is a very densely populated country and he cites evidence that under conditions of crowding in primates the frequency of alliance behaviour, grooming rules, reconciliations and greetings rise even higher than rates of aggression. In female, rates and intensity of aggression rise more than in males, but rather than de Waal's explanation that male rates are already high because they are already competing at a maximum rate, it seems to me that the argument that females have more interest invested in physical resources than males in order to maintain reproductive viability may be an explanation for this difference. Also, since females are less willing to re-establish bonds after a major fight, they may continue to harass a loser to death in a confined space. Thus captivity for primates reveals their potential for adaptability rather than the limits of their adaptability. Therefore, rather than bureaucrats blaming city dwellers for their intrinsically violent nature which are promoted by the "behavioural sink" of crowding stress, de Waal suggests that:

"...human biology has everything to do with environment. Not only has it been shaped by past environments, it also determines how we respond to current environments. Insofar as we share these responses with other primates, study of their group life under a variety of conditions may help us see where the remarkable human plasticity comes from and what it takes to stay out of the behavioural sink". (203)

As in discussions of primate language, culture, and politics, morality in animals (primates) is not the same as it is in humans but it rests on the same foundations. These foundations are not enough to accept the position of those proposing the "Great Ape Project" that suggests the inclusion of higher primates under the umbrella of human rights. However, even though humans must place human wellbeing at the core of their morality this does not mean that we can ignore that moral behaviour has valid roots in our evolutionary past, which can also be manifested in our biological kin. De Waal's metaphor is of a tower. He says that just because we can see farthest from the top of the tower, does that mean that the lower floors are not part of the tower.

Anne Zeller is an Associate Professor in the Department of Anthropology and Classical Studies at the University of Waterloo (Canada). She has recently been conducting fieldwork in Indonesia (Kalimantan) on the adult-infant social relations in Macaca fascicularis (crab-eating macaque). She is the author of a study of visual and vocal communication in Macaca sylvanus in Priamates: Recherches Actuelles. J.J. Roeder and J.R. Anderson (eds) Paris: Masson (1990). Among her other publications are: "The interplay of kinship organization and facial communication in the Macaques." In: Evolution and Ecology of Macaque Societies. J.E. Fa and D.G. Lindberg (Eds) New York: Cambridge University Press (1996), and "Evidence of Structure in Macaque Communication." In: The Ethological Roots of Culture. R.A. Gardner, B. Gardner, and B. Chiavelli (Eds.) Netherlands: Kluwer Academic Publisher, 1994 (15-39). Her most recent work involves comparisons of paintings and drawings by chimanzees, gorillas, orangutans and human children. This has been presented at two meetings in 1996.

Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives