The Politics of Peace (and War) in Preliterate Societies

The Adaptive Rationale behind Corroboree and Calumet

by JOHAN M.G. van der DENNEN, Center for Peace and Conflict Studies, Department of Legal Theory, University of Groningen, The Netherlands


Peaceable preindustrial (preliterate, primitive, etc.) societies constitute a nuisance to most theories of warfare and they are, with few exceptions, either denied or `explained away'. Contending theories have also tended to severely underestimate the costs of war to the individuals as well as to the communities involved. Materialist theory, as formulated by Ferguson, is one such exception: "[I]n contrast to the Hobbesian view, we should find nonwar, the absence of active fighting, in the absence of challenges to material well-being" (Ferguson, 1984). Where the costs of initiating violence outweigh the benefits, war is expected to be absent (Durham, 1976; Ferguson, 1984, 1990, 1994). There is no theoretical reason to deny the possibility of peaceful societies. Indeed, "there may be alternative peaceable and militaristic trajectories of evolution" (Ferguson, 1994).
The capability to make peace (peaceability) and the readiness to make war (warlikeness) are, it will be argued, not Platonic essences but the outcomes of a rational (Realpolitical) cost/benefit calculus (though the benefits of war or peace to the warrior-participants may not always be prima facie obvious) and an adaptive response (in the Darwinian sense) to particular sociopolitical ecologies.
Most peoples seem to prefer peace when they can afford it, i.e., when they can solve the internal problem of the `young male fierce warrior syndrome' (especially prevalent when the warrior role is rewarded with social status and/or sexual privileges), and the external problem of being left `in peace' by other peoples. One reason why it is the young males (especially young bachelors) who are eager to initiate and conduct war, Keeley (1996) explains, is that they have the least to lose and the most to gain from successful combat: spoils and loot ("booty and beauty"); honor, status and renown; sexual access to both ingroup and outgroup women, etc.
The ecological roots of peace may be as complex as, or even more so than, the roots of violence and war. There may be as many reasons for peaceability as there are for belligerence: Intercommunity nonviolence may be a response to overwhelming odds; it may be the taming effect of defeat; it may be enforced by colonial or imperial powers; it may be the result of isolation and/or xenophobia; it may be due to a negative cost/benefit balance of war, making peace more opportune under the given circumstances; it may be due to a voluntary decision to abstain from or abandon violence, or to a nonviolent ethic or pacifistic ideology; or some combination of all these factors. As Dentan (1992) reminds us: "[P]eaceability is not disability, not a cultural essence unrelated to a people's actual circumstances". Thus, warlike people are quite capable of peacefulness, while peaceable peoples are perfectly capable of intergroup violence under altered circumstances.
If war is so universal and ubiquitous as has been claimed by advocates of the Universal Human Belligerence theorem (see Van der Dennen, 1990), the mere fact of peace constitutes a problem, and we would have to develop a theory of peace as an abnormal, anomalous condition. Gregor (1990) has actually proposed such a perspective: "Political systems are so volatile and war is so contagious that its existence should occasion little surprise. It is peace that needs special explanation".
In this contribution I shall argue that the claim of universal human belligerence is grossly exaggerated; and that those students who have been developing theories of war, proceeding from the premise that peace is the `normal' situation, have not been starry-eyed utopians; and that peace - the continuation of potentially conflictuous interactions between discernible groups of human beings with other means (to paraphrase the famous Clausewitzian dictum) - in primitive peoples is just as much a deliberate and rational political strategy, based on cost/benefit considerations and ethical judgments, as is war.

The Security Dilemma

General Robert E. Lee is reported to have said that "it is a good thing that war is so horrible or else we would grow too fond of it". The statement by Davie (1929) that "Men like war" is as apodictic as it is general (referring to all men), and obstinately reiterated to the present day. Lately, Van Creveld (1991) stated (with a similar universal pretense): "However unpalatable the fact, the real reason why we have wars is that men like fighting, and women like those men who are prepared to fight on their behalf".
Jane Goodall (1986) observed a great eagerness in young prime male chimpanzees for the behaviors involved in `lethal male raiding' parties, but she also pointed out quite emphatically that there are distinct individual differences in this eagerness.
Fox (1991; Cf. Klineberg, 1964) seems to advance what may be called a Bad Seed or Rotten Apple theory of war: One rotten apple soon spoils the whole basket. Similarly, one or a few percent of hyperaggressive or belligerent males distributed more or less at random throughout the megapopulation would be sufficient to create a rampant war complex among all the demes involved (e.g., Papua New Guinea, Amazonia.)
There is a much more `tragic' variant of this theory in which no one has to harbor ill will. The expectation or suspicion thereof is sufficient for an internecine war complex to develop. Virulent war complexes do not have to be explained by some evil streak in human nature, but can be understood - at least in part - as the result of a war trap (Tefft, 1988, 1990), from which nobody can disengage on penalty of annihilation.
Richerson (1995) advances what he calls the `evolutionary tragedy' hypothesis (not unlike the `tragedy of the commons'): Warfare is liable to evolve even if it makes everybody worse off. It results from the perversion of the situation (the perfidious logic of the war `game') rather than that of the actors involved. The only practical way to avoid victimization by aggressors is to deter attack by being conspicuously prepared to fight, and display a credible ability and will to inflict unacceptable damage on would-be attackers.
Primitive societies, like modern nation-states in an anarchic state system, are trapped in a security dilemma (e.g., Elias, 1978). Simple game-theoretical analysis reveals why such a situation results, most of the time, in an equilibrial stalemate of mutual deterrence (assuming short-term rational choices of actors) even if none of the actors harbors evil intentions or sinister motives (or is equipped with aggressive/violent/belligerent drives, urges or instincts).
The security dilemma in which (primitive) peoples find themselves has the formal structure of a Prisoner's Dilemma (PD) in which individual short- term rational behavior leads to a collectively irrational outcome: All parties involved defect and lose (in terms of casualties, destruction of property, costs of war preparations, opportunity costs, etc.).
In a relatively stable socio-ecological environment (in which each society knows its and others' place, numerical strength, retaliatory capacity, etc.) to be on the alert and be prepared to defend itself may be a beneficial strategy resulting in a kind of peace through insulation with only sporadic and incidental flares of overt violence.
In this case, which has the formal structure of an iterated PD, diplomacy and peace become viable options. In such an iterated PD situation, when both parties know each other more or less intimately, and expect future (reciprocal or mutually beneficial) interactions, mutual suspicion and xenophobic fear can give rise to mutual caution and diplomatic maneuvering, but probably only if there also is a higher (e.g., tribal) authority to stop the private-enterprise (revenge) raiding of the young male warriors, or relax the obligations of the blood feud (and the concomitant male ideal of the macho warrior, and material rewards and social privileges attached to the warrior role). "The Mohave Indians of the Colorado River valley are by reputation a warlike tribe" Stewart (1947) relates, "although my informants insisted that the people as a whole were pacifically inclined. It was asserted that, while war was disliked by a majority of the Mohave, battle was the dominant concern of the kwanamis (`brave men'), who were responsible for the recurrent hostilities and over whom there was no effective control" (Stewart, 1947).
As Goldschmidt (1994) points out, the problem of internal dissatisfaction with existing peace treaties among preindustrial societies is a recurrent one. The problem is caused partly by (a) distrust and fear; and (b) inability to restrain the (entrepreneurial raiding of the) warriors. "Even when the population is war weary" Goldschmidt (1994) concludes, "even when there is a genuine need for peace, the peace is fragile precisely because there remain those who feel that their masculinity, by which we mean their social identity, is lost if they do not press their cause", that is, the hatchet will not be ceremonially buried, when there is no acceptable face-saving device (peace with honor) for the `fierce' warriors.
Nevertheless, even in a situation of chronic insecurity, the acceptance of mitigating rules of combat, of a common law of war and peace, is in accordance with enlightened self-interest (Mühlmann, 1940). Rules for war mitigation and a common law of war and peace can, Mühlmann holds, gradually develop (only?) in a situation of hereditary enmity.
Paradoxically, the Plains Indians' war complex with its emphasis on individual feats of bravery and bravado (as exemplified by counting coup; touching the enemy, whether alive or dead, was considered to be the ultimate act of bravery) actually limited violence, so that warfare, though incessant, boiled down to a series of small-scale raids of a few `braves' striking coup and stealing horses, which were far more important objectives than killing the enemy.

The Inventory of Allegedly Peaceful Societies

`Simple' human societies, according to Knauft (1991, 1994) place great emphasis on generalized reciprocity and far less on balanced competition or negative reciprocity. Concomitantly, collective military action or warfare tends to be rudimentary or absent. This contrasts in aggregate terms with more complex, sedentary, horti- and agricultural societies, among which subsistence and demographic intensification are associated with increasing property ownership and status inequality, and increasingly competitive politicoeconomic and military rivalry (e.g., Fried, 1967).
Accordingly, we should be able to find a number of such `simple' societies without war, or with only rudimentary war, in the literature. Swanton (1943) surveyed the anthropological literature and found that there were about as many societies that were peaceable as warlike. Leavitt (1977) found war absent or rare in 73% of hunting and gathering societies (n=22), 41% of simple horticultural (n=22), and 17% of advanced horticultural societies (n=29). Holsti (1913), Hobhouse, Wheeler & Ginsberg (1915), van der Bij (1929), Numelin (1950), Textor (1967), Bonta (1993, 1997), and Van der Dennen (1995), among others, present inventories of a great number of peaceful peoples.
On the other hand, Otterbein (1970), in a sample of 50 societies, found only four or five to have engaged "infrequently or never" in any type of offensive or defensive war. Ross (1983) found twelve societies engaged in warfare "rarely or never" out of a sample of ninety societies. Also Jorgensen (1980) identified seven peaceful societies in his study of northwestern North America. Among these peaceful societies are the Monache, Panamint, Battle Mountain and Hukundika Shoshone, Gosiute, Kaibab Paiute, Wenatchi, Columbia Salish, Copper Eskimo, Cayapa, Lapps, Gonds, Tikopia, Semang, Dorobo, and Mbuti. Most of these peaceful societies, as Keeley (1996) notes, were recently defeated refugees living in isolation, or were forcefully pacified, or both (cf. also Ember & Ember, 1992).
Nevertheless, the evidence of a substantial number of peoples without warfare, or with mainly defensive and/or low-level warfare (i.e., seldom exceeding the level of petty feuding or desultory skirmishes) does not support the view of universal human belligerence. It does not support the equally erroneous view of universal peaceability either. Rather, it supports Mühlmann's (1940) and Dentan's (1992) view that peace as well as war are the results of illuminated and opportunistic self-interest in the political arena.
Van der Bij (1929) concluded that primitive peoples were peaceful because they were primitive. Steinmetz (1929), on the other hand, concluded that primitive peoples were primitive because they were peaceful. Steinmetz thereby reiterated the statement by Gumplowicz (1892) that peaceful peoples were evolutionarily stultified and remained on the level of monkeys. Gumplowicz, by the way, admitted that ethnology offers numerous examples of such peaceful peoples, without giving any explanation of why and how such monkey-like peaceful peoples have been able to survive in so warlike a world as he envisaged.

Peace as the Normal Condition

"The question has been raised whether the traditional view of early society as one of constant warfare is really justified by the facts. There is, in fact, no doubt that to speak of a state of war as normal is in general a gross exaggeration" Hobhouse, Wheeler & Ginsberg (1915) concluded in their extensive survey of some 650 primitive peoples. Similarly, Quincy Wright (1942) stated: "No general golden age of peace existed at any stage of human history nor did any general iron age of war. Neither the Rousseauian nor the Hobbesian concept of natural man is adequate".
In even the most warlike societies, the vast preponderance of time is spent in the pursuit of ordinary, peaceful activities (Gregor & Sponsel, 1994).
The unsentimental military analyst Turney-High (1949) proved, in several parts of his work on primitive war, to be a perceptive and keen psychologist. He observed that "Cold-blooded slaughter has really never been approved by the bulk of mankind. All have understood the amenities of peace to a greater or less degree... Peace, then, seems to be the normal situation in the minds of even warlike peoples".
Similarly, Keeley (1996) notes, warfare, whether primitive or civilized, involves losses, suffering and terror, even for the victors. "Consequently, it was nowhere viewed as an unalloyed good". At some level, even the most militant warriors recognized the evils of war and the desirability of peace (e.g., Jalé: Koch, 1979; Kapauku: Pospisil, 1963; Jivaro: Karsten, 1967; Apache: Opler, 1983). Even the fierce Jivaro head-hunters regarded their incessant warfare as a curse. Additional evidence of the universal preference for peace, Keeley contends, is the ease with which some of the most warlike of tribal peoples accepted colonial pacification or pacified themselves (e.g., the Waorani or Auca; vide infra).
In discussing the Inevitability Belief (i.e., the belief that war is `natural' and, therefore, inevitable), Ferguson (1989) notes:

[T]he claim for universality [of primitive war] can only be advanced by relying on several dubious procedures: letting one cultural subdivision with war represent a broader cultural grouping which includes some groups without war; letting war at any point in time count, and disregarding what may be much more typical periods of peace; and when these fail, falling back on the untestable assertion that a peaceful people might have had war before the Westerners arrived. Even if we focus on societies where warfare is an undisputed occurrence, periods of active warfare involving a given group usually are relatively brief. The vast majority of humans, living or dead, have spent most of their lives at peace. So one can agree with Hobbes that politically autonomous groups have the potential for war, but this tells us nothing about why real war occurs. Contrary to the Hobbesian image, peace is the normal human condition.

Purification Rituals: Ambivalence toward the Enemy

We have been led to think that disregard for enemy life and his feelings are characteristic of warfare, Turney-High (1949) states, but this is not necessarily so, as evidenced by ambivalent feelings toward the enemy and guilt-expiating ritual, both of which seem to be universal and betraying `bad conscience'.

War and killing push men into some kind of marginality which is at least uncomfortable, for there seems to be a basic fear of blood contamination, an essential dread of human murder. If man did not consider human killing something out of the ordinary, why has there been such common fear of the enemy dead, the idea of contamination of even a prestigeful warrior of the we-group? We have seen that the channeling of frustration into hatred toward the enemy is good for the internal harmony of the we- group, but the enemy is human, too. Humanity is capable of ambivalent attitudes toward its enemies (Turney-High, 1949).

In a chapter of his The Golden Bough, aptly entitled "Taboo and the Perils of the Soul", Frazer (1890) was the first to acknowledge the existence, and summarize the available evidence of disculpation ritual, taboos and purification ceremonies (or lustration), indicative of some sense of guilt, in the post-war behavior of primitive peoples. The purpose of the seclusion and the expiatory rites which the warriors who have taken the life of a foe have to perform is, he points out, "no other than to shake off, frighten, or appease the angry spirit of the slain man" (cf. Kennedy, 1971; Goldschmidt, 1988; Keeley, 1996).
In his Totem und Tabu, Freud (1913) was so impressed by these examples of disculpation ritual among primitive peoples that he discussed the subject at length, connecting the expiatory ceremonies following the killing of an enemy with the general ambivalence of taboo.
Much of the post-war ritual activity in primitive societies seems clearly to indicate the expiation of guilt. Various kinds of ritual penance after killing were widespread in primitive (and ancient) societies. Fasting, sexual abstinence, and separation were common, as were ritual responsibilities such as sacrifices for vows given. Often the returning warrior was considered spiritually polluted or contaminated and had to undergo additional purification (cleansing) rituals. The Pima, for example, regarded the killing of an enemy to be such a dangerous act that a Pima warrior withdrew from battle the moment he killed his opponent to begin his rites of purification, or lustration (Kroeber & Fontana, 1987).
"There has existed" Turney-High (1949) concludes his perceptive review, "a dread of taking enemy life, a feeling that if the life of a member of the we-group was precious, so was that of a member of the other-group. Fear of death-contamination has demanded expiation or purification among many folk".

Prudent Feuders

There are a number of instances of tribal communities that do not support individual members in their personal vendettas against outsiders for fear that such revenge actions may escalate intercommunity violence which would prove detrimental to the collective interests of the whole community. There are three ways for kin units such as patriclans to avoid unnecessary feuds: (1) They may send the culprit into exile; (2) they may renounce the clan's responsibility to avenge him, giving other clans a free license to hunt him down, or the community may even turn a murderer over to the victim's kin; or (3) his own clan may put him to death (A.Moore, 1978; S.Moore, 1972; Boehm, 1985, 1986). Boehm (1986) comments: "A clan system of collectivized self- defense and liability `works' only if clan members are reasonably prudent in committing homicides or in otherwise stimulating members of other clans to kill them. Too much heroic aggressiveness can embroil a clan in so many feuds that it faces serious decimation or cannot earn its subsistence. Warriors living in feuding societies [such as the Pathans (Pashtun) and Montenegrins] are aware of these costs, and mostly they behave accordingly - that is, prudently. They try to be as aggressive as honor demands, but also try not to initiate feuds recklessly or pointlessly".

Peacefulness Does Not Equal Pusillanimity or `Gentleness'

When Gregor (1990) tried to find comparative data to complement his study of the relatively peaceful Xingu communities, he was frustrated by the minimal number of peaceful peoples he could find. He writes:

Other researchers, who have combed the literature more systematically than myself, have reached the same conclusion. Thus Richard Sipes notes in his study of war and combative sports: `Relatively peaceful societies are not easy to find. I had to investigate 130 societies to find eleven, of which five were rejected because of insufficient information' (1973: 68). Similarly, Otterbein (1970) found only four peaceful cultures among the fifty in his study of the evolution of war. Turning to advanced, state-level societies the searcher for peace becomes even more disheartened...
The societies that come closest to fitting the model of the truly peaceful culture are small in scale and primarily hunters and foragers. This conclusion is in keeping with research on war by Wright... and others who have positively associated war with community size and cultural development. Peaceful peoples also tend to be geographically isolated. Otterbein (1970), for example, finds that societies lacking in military organizations, such as the Copper Eskimo, the Dorobo and the Tikopians, live on islands, mountain tops, arctic wastelands and plateaus surrounded by malaria infested jungles. In some cases this isolation is a strategic adaptation to dealing with more aggressive societies that surround them. In most instances, however, peaceful societies appear to achieve their status by evading rather than solving the problems of intertribal relations (Gregor, 1990).

Isolation, splendid or not, seems prima facie to be the most prominent condition for peacefulness. So much so, in fact, that Mühlmann (1936, 1940) virtually identified peaceful peoples with Rückzugsvölker (litt. evading/retreating peoples).
Why could Gregor find so few peaceful peoples? One of the reasons might be simply because his criteria were wrong. In order to classify a people as `peaceable', some scholars demand not only absolute proof of the absence of intercommunity warring and feuding, but also the absence of every trace of intragroup violence, manifestations of aggression, and even conflict. They quite unrealistically require these societies to be `gentle' and pusillanimous in all walks of life.
As Turney-High (1949) already observed: "Such warless people have by no means been friendly and pacific. They have not been ignorant of how to shed human blood, nor have they abhorred it. Neither have they been without social institutions which formalized man-killing... Field ethnology no more demonstrates that a warless people are per se a kindly one than it shows that a monogamous tribe is sexually chaste" (Turney-High, 1949).

Bellicosity Does Not Equal Aggression

Whatever function aggression or violence may serve in the life of the individual or the small group, Malinowski (1941) already observed, it does not serve the same function between political units. Wars between bands, tribes, states or similar political entities are not just magnified quarrels between individuals. Warfare is not just simply aggregated individual aggression.
The profound misunderstanding about aggression and warlikeness, and the fundamental confusion concerning `nonaggressive' and `peaceful' is perhaps best exemplified by Heelas (1989; Cf. also Dentan, 1992), who devotes his whole contribution discussing definitions of aggression in his Search for Peaceful Peoples.
It may be important to note that `peace' as used by Heelas and Dentan, and by many other Anglo-American authors, refers to the absence of physical violence generally (including intra- and intergroup violence), while in most other languages `peace' (except in such metaphors as `peace of mind', etc.) refers preferentially or exclusively to the absence of `war' (as collective, organized, armed and violent intergroup or interstate conflict).
Montagu (1978) makes the same distinction between intragroup or intergroup `aggression' and implies that they may vary independently: "When reference is made to aggressive societies we have to be quite clear whether the reference is to intragroup or intergroup aggression. There are societies in which intergroup aggression is high but in which intragroup aggression is low, as among a number of New Guinea peoples. There are some societies in which aggression is high both within the group and between groups, as among the Yanomamö. There are societies in which both inter- and intragroup aggression is low, as among the Toda of Southern India, and there are some societies in which both inter- and intragroup aggression are nonexistent, as among the Tasaday of Mindanao, in the Philippines" (Montagu, 1978) (The Tasaday have in the meantime been exposed as victims or perpetrators of a hoax).
The only reasonable criterion for peacefulness is the presence or absence of offensive war or warlike behaviors (which implies that it is an intergroup phenomenon), and not the presence or absence of any and all forms of intragroup violence, or aggression, or conflict. The confusion rests on the, mostly implicit, assumption that war in some unspecified way is the result of the collective outpouring of accumulated `raw aggression'. In a previous publication (van der Dennen, 1986) I have tried to outline the fallacies involved in this kind of reasoning, especially the cumulative fallacy (the confounding of levels-of-analysis).
According to Kennedy (1971) and numerous other authors (see van der Dennen, 1986), aggression is obviously correlated with, and an integral aspect of war, but the relationships between war and aggression are reciprocal, complex, and mediated by intervening variables. There is no simple cause and effect relationship, and as White (1949) and others have long contended, there is probably more evidence to support the proposition that war produces aggression than the reverse. Ember & Ember (1992, 1994) found empirical evidence that among primitive peoples socialization for aggression is more likely to be a consequence than a cause of war. Grudges of `unemployed' warriors after coercive pacification have sometimes been (mis)construed as evidence of some kind of innate bellicosity.
Robarchek & Robarchek (1992), discussing the Waorani in Amazonia (who are probably unique in deliberately and consciously abandoning feuding and warfare), draw attention to the often limited options available in a hostile environment: "In such a situation, where warfare is endemic [and rampant], a people's options are rather limited: they can either flee, fight back, or be overwhelmed. Given the sociocultural environment of the region (and with no safe refuge available), engaging in at least defensive warfare becomes a functional necessity for group survival. Warfare, under these conditions, is contagious; once one group adopts it as a tactic for advancing its ends, others must either take it up or be destroyed".
The result is a more or less stable balance of terror with constant raiding among the various social groups (Robarchek & Robarchek, 1992). In such a situation, fear, as Whiffen (1915) long ago, and Mühlmann (1940) and Meyer (1977) more recently, pointed out, seems to be the predominant war motive.
There is, furthermore, a strong androcentric bias in the accounts relating aggression and warfare in primitive societies; "with a sleight-of-hand extension of man into Man... Woman is either ignored or presented as innately less aggressive than man. The arguments for a biological difference in the sexes in this regard are far from conclusive, but in cases where such a difference is put forward, the general conclusions of humanity's aggressive nature are not revised" (Howell & Willis, 1989).
These authors also draw attention to the fact that aggression/violence/warlikeness, though considered `natural' (particularly or exclusively in males) is also condemned as `bad', while its perceived opposite, peacefulness, carries with it the negatively valued connotations of being passive and inert, qualities which are associated with females. One might go so far as to state that for many males in primitive communities, as well as in our Western culture, `peaceful' equals `weak' equals `unmasculine/feminine' equals `impotent' equals `emasculated/castrated'.

Primitive War as a Post-Contact Phenomenon

The effects of contact with `civilized' states and colonialism in the warfare patterns of primitive peoples have, until recently, not sufficiently been acknowledged. Virtually all over the globe such contact has exacerbated warfare within and among nonstate societies to a degree we are only beginning to sense (e.g. Blick, 1988; Ferguson, 1992a,b; Ferguson & Whitehead, 1992; Sponsel, 1994).
"Accepted wisdom even now holds that `primitive' cultures are typically at war and that the primary military effect of contact with the West is the suppression of ongoing combat. In fact, the initial effect of European colonialism has generally been quite the opposite. Contact has invariably transformed war patterns, very frequently intensified war and not uncommonly generated war among groups who previously had lived in peace. Many, perhaps most, recorded wars involving tribal peoples can be directly attributed to the circumstances of Western contact" (Ferguson, 1992b). A consequence of this is, as he explains elsewhere (Ferguson, 1990), a systematic exaggeration of images of warlike behavior in supposedly `first contact' accounts.

The Characteristics of Peaceful Peoples

Fabbro (1978) analyzed five peaceable primitive societies, including the Semai, the Siriono, the Mbuti, the !Kung, and the Copper Eskimo. To these `traditional' groups, Fabbro added two literate peaceful communities for reasons of comparison, the Hutterites and the Islanders of Tristan da Cunha. Contemporary peace groups, such as Hutterites, Mennonites and Amish, living in permanent communities based on a common religion, are also called `cenobites'.
A peaceful society, according to him, is one that is not involved in internal (i.e., intracultural) collective violence; one that exhibits relatively little interpersonal violence; one that provides no special role for warriors; and one that has values and sanctions precluding violence as a means for resolving conflict. Peaceability should not be confused with pacifism, which is only one genre of peaceability (Dentan, 1992).
McCauley (1990) presented the results of a study of the Semai and two other peaceful societies, the Buid of the Philippines, and the South American Xingu River conglomeration of tribes. Various combinations of the peaceable communities mentioned above were also present in the analyses of Gregor (1990) and Dentan (1992, 1994). From the combined analyses of this rather small sample a number of patterns emerge:

* All peaceful societies are essentially small, local, face-to-face, communities with very low degree of social stratification, and open and egalitarian decision-making.
* The `traditional' societies do not maintain an exclusive monopoly over an area of land. Other groups may come and go, and in times of shortage an incumbent band may share the food and water resources with another less fortunate group. But conflicts within these groups are also partly responsible for personnel changes, fission being used as a dissociative conflict resolution form.
* The traditional societies produce little or no economic surplus. Material inequality between individuals on a long-term basis is, therefore, impossible. As a corollary, leadership remains on the level of personal authority rather than coercive power because there is no surplus to appropriate.
* The differences in child-rearing practices between the traditional and the cenobite societies are open to a number of possible explanations. Cenobites generally are more authoritarian with children than are peaceable `refugees' like Semai, and they approve the spanking and whipping of children as corporal punishment of last resort (Dentan, 1992, 1994). The socialization of nonaggression (e.g., Montagu, 1978; Irwin, 1990) may be a relatively minor factor in the creation of peaceability (e.g., Riches, 1987; Dentan, 1992; Eibl-Eibesfeldt, 1993), though some cross-cultural studies find a positive correlation between child abuse and neglect or harsh socialization practices and bellicosity (e.g., Levinson & Malone, 1980; Ross, 1992).
* Many of the peaceful societies develop what Gregor (1990) calls an `antiviolent' value system; cultural norms and ideologies which discourage both intra- and intergroup violence (an important component of which seems to be Gelassenheit at least among cenobites). Nonviolence is supported by stigmatizing quarreling, boasting, stinginess, anger, and violence, and by according prestige for generosity, gentleness, and conflict avoidance. This value system is supported by supernatural beliefs (McCauley, 1990).
* Peaceability and nonviolence among primitive peoples and cenobites seems to stem from (a psychology of) defeat: "Defeat tamed them... those that survived did so by learning virtues of political accommodation or withdrawal from temporal affairs" (Barkun, 1986; see Dentan, 1992, 1994). Or, as Bigelow (1969) put it "their `peacefulness' was imposed on them by force". `Islets of peaceability' can arise as an adaptive response to defeat by neighboring peoples when there are relatively unpopulated areas (called `refuges' or `enclaves') to flee to.
* Peaceable `refugees' tend to be insulationist and xenophobic. Lacking the oppositional frontier processes that create peaceable `refugees', cenobites need specific mechanisms to maintain the boundaries between their people and the `others' by means of physical isolation. Peaceable peoples like Semai contrast themselves with the peoples they fear, creating a counterculture. The antiviolent value system is embodied in a contrast between the peacefulness of the ingroup and the violence of outsiders. Outsiders are bloody, violent, dangerous, ugly, evil, animal-like and, in a real sense, less than human. Children are warned against outsiders and, especially, about behaving like outsiders. Apparently, "hating violence requires violent people to hate" (McCauley, 1990).
* The gender-equality characteristic of many egalitarian band-level societies is not a necessary correlate of peacefulness among enclaved peoples, although the two phenomena can co-occur.
* None of the peaceful societies would seem to operate on the premise that its members would automatically refrain from violence (even though aggressive models are absent). Even the most peaceful of these societies employ various forms of social conditioning and indoctrination to constrain and deflect the tendencies to resort to violence, as well as community inducements to discourage violence, and instructions in the virtues and arts of nonviolent conflict resolution. Tribal cosmology, rituals, legends, religious and ethical concepts and precepts reinforce the nonviolent norms of the society. And social ostracism is typically inflicted on individuals who violate these norms (S.Brown, 1994).

The 52 Peaceful Societies investigated by Melko (1973) are not really societies (in the ethnological sense) but particular historical periods of particular civilizations (such as the Han and T'ang dynasties in China) without major internal physical conflicts. Yet, some of his findings may be summarized for reasons of comparison.

* No one form of government, no one economic system, no one structure of society, no one system of education seems to be essential to peace.
* Moderate powers seem to have had the advantage over great powers in maintaining peace. They are strong enough to resist attack, but not strong enough to become overextended. Small powers that have been successful in maintaining peace have refrained from interfering in the affairs of their neighbors. Great powers seem to succeed in attaining peace only if they conquer all other great powers within range.
* Peace is the normal internal condition for a society. Conflict involving physical fighting is exceptional. When it occurs, most people involved in it are not fighting most of the time. Most people in most places in most periods of history have not been killed or injured in war.

A Typology of Peace

Tefft (1988, 1990), among many other, pointed to the fact that tribes involved in so-called `restrictive' (or rather `restricted') wars often have institutionalized checks limiting the level of intertribal violence (e.g., Polopa: D.Brown; Bete: Balandier, 1986; Meru: Fadiman, 1980). Many societies from all over the world seem to make a clear distinction between `real' wars and more game-like, ritualized wars (also called `agonistic' wars) (e.g., Rappaport, 1968).
In the next section, I shall present a typology of peace and more fully discuss some of the strategies and mechanisms of peace-making and peace-keeping in primitive societies.
According to Service (1975), it is the evolution of the various causes of peace (instead of the `causes of war') that can be studied in the human record, and "a large and essential part of the evolution of political organization is simply an extension and intensification of peace-making means".