Religion in Human Evolution: From the Paleolithic to the Axial Age

by Robert N. Bellah

  Play

  If we are correct, following de Waal, in finding empathy, or following EiblEibesfeldt, love, as the basis of all social bonding, then play will need to be understood as involving some kind of bond. But play, as we will see, is a kind of event, an activity that begins and ends, and it takes place in the context of daily life, from which it is to some degree differentiated. If mammals from a long time ago are often “social,” as seems to be the case, though in varying degrees, what kind of society characterizes the daily life from which play is distinct?

  The most obvious form of social bonding arises from the very practice that makes it possible: parental care. Although it varies by species, some relationship between mother and child and, less often, between father and child, continues after the offspring have become independent. Siblings, cared for by the same mother, can be rivalrous even in species with low social complexity: two pups may seek to nurse from the same breast. Yet siblings often appear to share a degree of trust that may be weaker in other relationships. In short, something like protokinship probably goes way back in the mammalian line, and can be found even among reptiles and fish.76 Without language there can, of course, be no kinship terms, yet the recognition of kinship is often present. For example, though any other animal may be a play partner, siblings are especially likely to be; and the play between mother and child seems to be part of “cherishing” from very early on.

  Yet kinship does not supply the only basis of social order of most mammalian societies, to the extent that they have societies, or bird societies either. In fact, the more social the species, the more likely it will be organized also in terms of a dominance hierarchy. Although the well-known pecking order in a group of chickens, where every single individual is ranked in terms of dominance to every other individual, is rare, some kind of ranking from the dominant male (and it is almost always a male) to middle-ranked members of the group, to the very lowest, who may be on the boundary of exclusion, is common. It was Abraham Maslow in early work with rhesus monkeys and chim panzees in the 1930s who first coined the term “drive for dominance,” and though he did not think there was a drive for submission, because all shared the drive for dominance, he did argue that submissive behavior in order to placate the dominant by admitting inferiority was certainly important.77

  Dominance and the attempt to attain it when one doesn’t have it would appear to mobilize aggression, not empathy. And because in almost all mammal societies that have dominance hierarchies it is males who are at the top or fight to be at the top, we might imagine that there are two moralities, differentiated by gender: females in terms of empathy, males in terms of dominance. This would be especially the case among primates, where male dominance hierarchies are highly developed and parental care is almost exclusively the domain of the mother. Yet things are seldom so simple. De Waal notes that dominant males enforce obedience to the rules of the group on younger members by inflicting punishment, occasionally severe, for transgressions, but then comments:

  There is no single individual from whom infants and juveniles receive more aggression, however, than their own mother. Usually, of course, it is of the nondamaging kind, but bites and even injuries do occur. Irwin Bernstein, a well-known American primatologist, interprets it as socialization, in which mothers teach their offspring to inhibit particular behaviors that may get them into trouble. Even though maternal aggression may not be to the youngster’s immediate advantage, it promotes the caution and behavioral control required for survival in a hierarchically structured social environment?s

  Also, counterintuitively or perhaps not, although males fight more often among themselves, they are better at resolving conflicts amicably than are females.79 Dominant males are certainly looking out for themselves: they eat first and, if there is little, most; they attempt, almost always without total success, to monopolize mating with the females of the group. Yet by curtailing fights between their subordinates, sometimes taking the lead in hunting, and distributing resources, including social acceptance, they serve the group as well as themselves. (As usual it is both/and, not either/or.) Thus de Waal summarizes:

  Not surprisingly, given this integrative function, formalized hierarchies are best developed in the most cooperative species. The harmony demonstrated to the outside world by a howling pack of wolves or a hooting and drumming community of chimpanzees is predicated on rank differentiation within. Wolves rely on each other during the hunt, and chimpanzees (at least the males, who are by far the most hierarchical sex) count on the other members for defense against hostile neighbors. The hierarchy regulates internal competition to the point of making a united front possible.80

  De Waal finds that, especially among primates, dominance hierarchies are variable in their degree of despotism. Rhesus monkeys, for example, are despotic, and any challenge from below is severely punished. Chimpanzees, however, are quite different: “Even though we cannot go so far as to call chimpanzees egalitarian, the species has certainly moved away from despotism toward a social arrangement with room for sharing, tolerance, and alliances from below. Although high-ranking individuals have disproportionate privileges and influence, dominance also depends to some degree on acceptance from below.“81 It is even possible for a coalition of females to oppose an alpha male who is acting too harshly, and, because other males have their own reasons not to come to his rescue and females are large and strong enough that several of them can subdue a single male, the alpha has little choice but to back down.82

  Having looked briefly at the social structure of ordinary life among highly social mammals, especially primates, we need now to look at play, an activity that by its very definition is not ordinary life. I want to focus on play because I think it is the first example in evolutionary history of one of Schutz’s multiple realities other than the world of daily life. According to Johan Huizinga in his Homo Ludens, play is the ultimate source of virtually all human cultural systems: myth and ritual, law, poetry, wisdom, and science.83 Cultural systems, as Geertz uses the term, are multiple realities at the human cultural level.

  I will turn to Gordon Burghardt, whose splendid book The Genesis ofAnimal Play is the best recent treatment of the subject, for a fairly complex definition of play, complex because of the many dimensions that students of animal play have noted. Burghardt sums up by indicating five things that must in some way always be present before we can call something animal play:

  1. Limited immediate function

  2. Endogenous component

  3. Structural or temporal difference

  4. Repeated performance

  5. Relaxed field84

  The first criterion indicates that play is “not fully functional in the context in which it is expressed,” that it “does not contribute to current survival.“85 If, according to Darwin, evolution can be characterized as “the struggle for existence,” and according to Spencer as “the survival of the fittest,” then play is something different from the “paramount reality” of the world of daily life in evolutionary history, and the something different is the first alternative reality.86

  The second criterion is that play is something “done for its own sake,” pleasurable in itself, spontaneous and voluntary; it is not a means to an end.87 This is what Burghardt means by speaking of its “endogenous component.” The third criterion, “structural or temporal difference,” indicates that play may use behaviors from ordinary life, like fighting, chasing, wrestling, but without the aim that such behavior would ordinarily have. It uses features of ordinary life “playfully,” for their own sake, and not to achieve the aim that they have in ordinary life. This is one of the bases for seeing play as not “serious,” though that is an issue that will need further consideration. Burghardt points out that this criterion does not mean that play is “completely unstructured, free from rules,” and, as a result purely “creative.” Indeed, he says that “if these claims were true we would never recognize any behavior as play.“88 The fourth criterion is that play behavior
is “performed repeatedly in a similar, but not rigidly stereotyped, form.” It is, then, “something that is repeatedly performed, often in bouts, during a predictable period in the animal’s life (which in some cases can be virtually lifelong).””

  The fifth and final criterion is related to the first one: play behavior “is initiated when an animal is adequately fed, healthy, and free from stress (e.g. predator threat, harsh microclimate, social instability), or intense competing systems (e.g. feeding, mating, predator avoidance). In other words, the animal is in a’relaxed This criterion is important for helping us understand the origin of play and the reason why it is limited, largely but not absolutely exclusively, to mammals and birds, and also why it is often limited to the young, though in some species it continues throughout life. One can think of a variety of conditions that would produce a relaxed field-we could almost say, analogously, relaxed selection-but an obvious one is parental care. Young animals whose primary needs are taken care of by others, who are fed and safe, are the ones most likely to play. Also a hierarchical social structure that provides some relief from aggression within the group and a more adequate defense against external dangers could provide conditions that encourage play, not only for the young, but for adults as well. This might be especially true for a hierarchical structure like that of the chimpanzees, one that has moved away from despotism toward something that begins to look like a “constitutional monarchy.” The idea of a “relaxed field” doesn’t explain why animals play, but it is a beginning.

  Students of animal play have discerned three major forms of play: locomotor, object, and social play. Burghardt speaks of locomotor-rotational play as it can involve not only movement from place to place but movement in one spot, involving various kinds of turn. This is usually the earliest form of play in the life of the animal and is often solitary. Burghardt gives the example of “the gambols of foals released from barn stalls into a field.“9’ Object play is also often solitary and involves an animal interacting with an object with no purpose other than to play. Anyone who has ever had a cat knows what object play is, but human infants interacting with toys is another obvious example. Social play involves at least two animals, but sometimes more. As Burghardt says, “social play can take many forms, but the most common are quasi-aggressive behavior patterns such as chasing, wrestling, pawing, and nipping.” As to the salience of social play, he notes that it “is interesting to watch, involves many often complex and often balletlike movements, and appears to presage the use of these behavior patterns in more serious adult behavior.“92 Social play has the most possibilities for further development and will be discussed further below, but first we will need to consider one more of Burghardt’s classifications of play, this one developmental.

  Burghardt, organizing a great deal of previous work on animal play, speaks of primary, secondary, and tertiary play processes, all of which must meet the five criteria for the definition of play. Primary process play is not the result of direct natural selection and may have no selective consequences: it is the earliest form of play. Play of this type may have no role in subsequent behavior, or it “may serve as a `preadaptation’ or `exaptation’ providing variation that can be selected.” If I may interpret Burghardt, it would seem that play is the result of something almost like “relaxed selection,” such as parental care would provide, and it is just in situations of relaxed selection that genetic possibilities already existing, which is what “preadaptations” and “exaptations” indicate, previously suppressed because nonadaptive, are then released into behavior. Once play behavior has come into existence, it may be selected for various functions, which is what secondary and tertiary play pro cesses describe, but some degree of primary play process survives in species that have also developed significant elaborations.

  Secondary process play covers many of the functions that have been offered to explain the evolution of play. Secondary process play is “behavior that, once it occurred, evolved some role, although not necessarily an exclusive or even a major one, in the maintenance or refinement of normal development of physiological and behavioral capacities. Play may serve to maintain the precision of predatory, defensive, and social skills, neural processing, and physiological capacities.”

  Tertiary process play is “play behavior that has gained a major, if not critical, role in modifying and enhancing behavioral abilities and fitness, including the development of innovation and creativity.” Although the transition between secondary and tertiary processes is not a sharp one but more of a continuity, tertiary process play points to the rich possibilities of the development of play among human beings where culture adds an array of possibilities for further elaboration.93

  As we have seen, parental care seems to be an important precondition for the development of play. Burghardt makes some generalizations about the relation of kinds of parental care to the degree of play development. We have noted above the difference between precocial species, where the young are born virtually viable, and altricial species, where the young are born helpless. In some precocial species, such as cattle and horses, the mother usually gives birth to only one offspring and the offspring may stay with the parents longer than the offspring of some altricial species, where a large number of helpless infants mature quickly to independence. Thus “the parental care system as a whole needs to be considered. Nonetheless, altriciality may be a useful marker in identifying animals in which play is prominent because species with altricial young often play more, or more complexly, than even close relatives that are more precocial.“94

  Burghardt emphasizes the complexity of play and the very uneven development of research on key aspects of it, which make it difficult to account for the origin of animal play, yet he offers a tentative set of hypotheses that he calls “the surplus resource theory of play.“95 One aspect of this theory is that the longer the period of parental care, the more likely the offspring will have the energy and, often, the intelligence, to need some form of expression to avoid what we would call It is in response to this more or less prolonged period of the relaxation of selection pressures that primary process play arises. Primary process play is a response to the absence of specific pressures, not to such pressures themselves. Nevertheless, the absence of selection pressures means that highly specific instinctual capacities to deal with the environment-predation, flight, mating-normally present from birth among animals lacking parental care, in their earliest appearance gradually atrophy through disuse among animals with extended parental care, that is, they tend to be genetically deselected. What takes the place of the deselected instincts, usually quite precise in their behavioral implications, are more generalized play behaviors, but ones that are now available to selection as secondary play processes-that is, play behaviors such as wrestling, running, chasing, and so forth-that could, in a general way, help to hone skills that will be useful in the “real world” once the young are on their own. Burghardt is clear that play did not originate to provide these functions, but that functions can develop out of play behavior as an activity whose good was originally not for any function at all. Once secondary play process has arisen, there is the possibility that play will give rise to novel activities not previously part of the species repertory. In other words, play is a new kind of capacity with a very large potentiality of developing more capacities, what Burghardt calls tertiary play processes, some of them quite extraordinary.

  We need now to look at some of the features of social play among nonhuman animals to see just how some of these new capacities might have arisen. Our examples will come mainly from canids and primates, dogs and chimpanzees in particular, because those are the best studied. What is most striking in animals whose social structure is more or less strongly hierarchical is the equality that characterizes play. Marc Bekoff and Jessica Pierce put it strongly:

  We want to stress that social play is firmly based on a foundation of fairness. Play only occurs if, for the time they are playing, individuals have no other agenda bu
t to play. They put aside or neutralize any inequalities in physical size and social rank. As we will see, large and small animals can play together, and high-ranking and low-ranking individuals can play together, but not if one of them takes advantage of its superior strength or status.

  After all is said and done, it may turn out that play is a unique category of behavior in that asymmetries are tolerated more so than in other social contexts. Animals really work at reducing inequalities in size, strength, social status, and how wired each is to play … Play is perhaps uniquely egalitarian. And if we define justice as a set of social rules and expectations that neutralize differences among individuals in an effort to maintain group harmony, then that’s exactly what we find in animals when they play.97

  Burghardt notes some of the particular ways in which animal play is egalitarian. One common way is role reversal: “One animal chases the other; when the gap closes, the chased individual may suddenly swing around and begin chasing the chaser up trees, around bushes and rocks, and so on. One animal may be on top in a play wrestling match and then appear at the bottom.” 98 In this example, role reversal occurs within a play bout, but the reversal may also occur across bouts: “That is, one animal may chase another one day and be chased the next.“99

  Burghardt calls the behavior of the older, stronger, or higher-status animal involved in play with a younger, weaker, or lower-status animal “selfhandicapping,” and notes that “self-handicapping implies some kind of mutual intentionality in aspects of animal social play.“100 De Waal comments on rhesus monkeys, who do not seem to notice temporary impairments in their play partners right away, but if the impairment lasts longer than a couple of weeks, they do take it into account and adjust to the impaired animal as they would to a younger one. “I always admire the complete control of adult males at play; with formidable canines, they gnaw and wrestle with juveniles without hurting them in the De Waal argues that play inhibitions are probably produced by conditioning, are “learned adjustments”: “From an early age, monkeys learn that the fun will not last if they are too rough with a younger On this account play would be an expression of the plasticity and openness to learning that arises when parental care limits the need for early instinctive self-preserving behavior.