NOTES

Raymond C. Kelly, 2000, Warless Societies and the Origin of War, Ann Arbor, MI: University of Michigan Press.

In the Preface Kelly explains that for many years he has been teaching an undergraduate course on the anthropology of war at the University of Michigan. His book addresses some of the most frequent major questions that students bring to class--- Is war a primordial, universal, and pervasive feature of human existence, or something that developed at a certain time in prehistory? Are there peaceful societies which lack war? If so, what are they like?

Reading the book is like working a jigsaw puzzle, the author gradually puts pieces in place, and the pattern gradually begins to emerge, although at times somewhat tediously and repetitiously. Nevertheless, Kelly is quite explicit about the purpose and basic method of his book (p. ix): "My principal objective is to present a general model for the initial evolution of war that is grounded in the comparative analysis of ethnographic data and then to apply this to the interpretation of pertinent data in the archaeological record." Subsequently in the "Introduction" Kelly (p. 1) states: "The central objective of this study is to elucidate the conditions under which warfare is initiated in sociocultural contexts where it did not previously exist, and to decipher the origin of war in that sense."

In the "Introduction" Kelly (pp. 1-2) observes that, from the archaeological evidence available so far, it appears that warfare was rare to absent until quite late in human prehistory, around 7,500 to 7,000 B.C., and is related to the development of agriculture and sedentary villages. Later warfare is associated with competition over trade routes, and then the development of hierarchical and centralized political organization. The implication is that earlier hunter-gatherer societies were warless. This is particularly significant because hunter-gatherers societies represent about 99% of human existence, evolving some three million years ago. There is a paradox, however, because the ethnographic record of historically known hunter-gatherer societies indicates that many have sporadic interpersonal and intergroup violence if not warfare.

Conclusions about the temporal and spatial distribution of warfare pivot on whether it is defined broadly or narrowly. Kelly (pp. 4-5) defines warfare as armed conflict with lethal weapons, multiple killings (not considered murders or criminal acts), collective effort, organized activity, between groups [inter-group, not intra-group, thus the we/they topology emerges], division of labor, advanced planning (not spontaneous eruption of violence), legitimate, morally appropriate and justified killing, in which participation is laudable, esteemed, prestigious, and there is social substitutionability (any individual will do, so no distinction between military and civilian personnel). Because of
social substitionability, warfare differs from capital punishment in which a specific individual is targeted for revenge or retribution (pp. 5-6).

Kelly goes on to note that: "Although war entails lethal violence between individuals who reside in separate social groups, not all acts of intergroup lethal violence exhibit the full ensemble of distinctive features that characterize war. Distinguishing war as a specific form of intergroup lethal violence is essential to elucidating the initial evolution of war ...." (p. 6). In warfare the concepts of injury to one's group as a whole, responsibility of one's group for counteraction, and other group liability for retribution are crucial. In short, the key to crossing the boundary between war and other forms of intergroup violence is whether social substitutionability is absent or present (p. 6). Furthermore, social substitutionability marks the beginning of the coevolution of war and society (p. 7). Kelly discusses the ethnographic case of the Gebusi from the fieldwork of Bruce Knauft to illustrate these principles (pp. 7-10). Kelly concludes that the Gebusi do not have war, because they do not have social substitution that is its hallmark (p. 10).

In Chapter 1, "The Category of Peaceful Societies," Kelly develops the most perceptive summary and analysis of Fabbro's classic study yet available. Also Kelly goes on to develop Fabbro further with updated ethnographic material. In pursuing the origin of war Kelly (p. 11) thinks that it necessary to first identify the distinctive features of warless or peaceful societies, and then the critical differences between them and warlike societies. Also Kelly (p. 12) notes that there are different levels of intensities of peace. He admits that "...there are no concrete example of societies that entirely fulfill our highest aspirations for peace, nonviolence, equality, and social justice" (p. 13).

Kelly says that: "The critical issue that arises at this juncture is whether one form of violence begets another?" (p. 17).
Then he identifies a series of dependent null hypotheses about higher degrees of violence and concludes that one form does not necessarily lead to a higher form (p. 19). However, he suggests that war does evolve from other forms of violence when a new conceptual threshold with a distinctive logic is reached, namely, social substitutionability (p. 21).

Kelly touches on gender differences as well. He notes that females are less violent than males (pp. 31-33). Furthermore, the pattern of same sex violence is quite different in females than in males, the former often over sexual rivalry (pp. 33-34). For males, food is more important as a trigger of violence. Male-male violence is not predominantly over women (pp. 33-35). Also there is a wider range of forms of violence among males than females, and male violence involves more weaponry, lethality, and causes (p. 35).

In the first chapter Kelly concludes that violence is not a unitary phenomenon, that violence does not necessarily generate more violence, and that warless societies do exist and are not scarce (p. 37).

In Chapter 2, "Warless and Warlike Hunter-Gatherers: A Comparison," Kelly first summarizes his main points or conclusions so far: 1. Fabbro's peaceful societies lack war but are not devoid of homicide and other forms of violence. However, they lack the concept of social substitution. 2. A homicide may or may not be avenged, but if avenged this does not necessary lead in turn to further violence in some endless cycle of blood feuding. 3. Warless societies aren't explained by childhood enculturation practices that prevent or reduce anger or that lead to more effective mechanisms of conflict management and resolution. (Physical violence is itself a principal vehicle of conflict resolution). 4. Warless societies have an intrinsic limit on the extent to which one violent act leads to another. That is, violence is specific, not generalized, it is targeted at an individual rather than a group (p. 42). Kelly concludes that this implies that war originates from the transformation of one type of collective violence into another higher form of collective violence, instead of from a condition of peaceful nonviolence to one of armed conflict. This is the transformation from individual to group responsibility including social substitutionability, as in moving from capital punishment to blood feud to full war (p. 43).

Next Kelly asserts that the key to the origin of war is the difference between unsegmented and segmented societies (p. 44). Unsegmented societies possess only the minimal complement of universal social groups--- the nuclear family (parents and any children) and local community. Bilateral descent is common, that is, kinship is traced along both the maternal and paternal sides. Unsegmented societies have no level of organization beyond the local community, even though other communities may be in the neighborhood. The local group is not a subunit of any larger organizational entity (pp. 44-45). They may have bride service, but not bridewealth (p. 48). Also they lack the concepts of group liability and social substitionability, and many do not even respond to homicide (p. 54). When they do avenge a murder by killing the murderer, it doesn't go any further into a cycle of blood revenge feuding (p. 56). Thus, murder is a relational loss to individuals, not to the group. This may be because the unsegmented hunter-gatherer band has no fixed membership, but composition is fluid as nuclear families come and go, mainly moving to where game or other food resources are available (p. 57). They have no food storage (p. 71). These unsegmented hunter-gatherer societies are warless. (Also see pp. 118 and 128).

Kelly (p. 45) writes that: "Segments are units that are equivalent in structure and function. Segmental organization is the combination of these like units into progressively more inclusive groups within a segmentary hierarchy." Two of the most basic features of segmented societies are descent groups and intergroup bride exchange (pp. 46-47). They may also have bridewealth (p. 48). Segmented societies have the concepts of group injury, group liability, and group responsibility, and together these provide the basis for collective violence in the form of feud or war (p. 48).

Next Kelly identifies six logical stages in the evolution of war: 1. no counteraction to murder; 2. capital punishment legitimate, but no one (individual or group) designated to execute murderer; 3. some individual kin of murdered individual responsible to avenge murder by killing murderer; 4. kin group responsible; 5. prefer to kill murderer, but if not available then any member of the murderer's group liable; 6. kill murderer and/or any member of murderer's group. [Note that this sequence proceeds from individual to group, or from specific to general, regarding responsibility and liability].

Kelly observes that exogamy, marrying out into another group, covaries with less war between the groups (p. 62), but can also become a means for military alliance (p. 64). Also food storage limits mobility, stimulates population increase, and contributes to the development of wealth or inequalities (p. 68). Eventually warfare becomes a residential or territorial concern of the group, rather than merely kin group vengeance.

In Chapter 3, "The Origin of War: A Transitional Case," Kelly discusses the exceptional case of the Andaman Islanders--- even though they are an unsegmented hunter-gatherer society, they have war, both internal (intra-cultural) and external (inter-cultural). Thus, from Kelly's perspective the Andaman Islands are a natural laboratory to study war (p. 81). (See pp. 77, 81, 96, 105, and 118).

First, Kelly summarizes his two main conclusions: (1) "Warfare is not an endemic condition of human existence but an episodic feature of human history (and prehistory) observed at certain times and places and not others" (p. 75). (2) "War thus originates as a transition from one form of retributive collective violence to another, that is, as a transition from capital punishment to blood feud, with these representing different patterns of vengeance...." (p. 75).

In the case of the Andamans, they do not fight over women, but over resources and territory, especially pigs and honey (p. 98). They have intergroup intracultural conflict when a group encounters another at a hunting resource (p. 100). Also when a lone male is discovered from another group he may be killed (p. 101). The Andamans launch spontaneous attacks, opportunistic ambushes, and preplanned surprise raids (p. 91), and they also have intergroup confrontations (p. 101). Casualties are mostly males, rather than children and females as opportunistic victims, because the Andamans have no concept of group responsibility and liability. [There are numerous quite remarkable similarities or parallels, but not necessarily identities, between the Andaman aggression and that of the Gombe chimpanzees and the Amazon Yanomami, both also circumscribed (e.g., see pp. 106-108)].

In this chapter Kelly is more explicit than before about the basis of war, developing an explanatory analysis clearly grounded in the theoretical frameworks of cultural evolutionism, ecology, and materialism. The causes of war in the Andamans are natural resource competition and competitive exclusion in overlapping ecological niches and territories, although Kelly doesn't employ the term competitive exclusion. Also he repeatedly mentions the circumscription of island ecology. That is, islands are isolated and bounded, so there is no additional space readily available for expansion into new areas to exploit more resources. Also, the Andaman Islanders had an unusually high population density, thus potential for resource depletion and ensuing competition (pp. 91-96). In particular, violence was likely to erupt when coastal and inland bands encountered each other along their territorial boundaries (p. 95). In short, among the Andamans, war appears to be a spacing mechanism, keeping the population spread out over the landscape in relation to resources, although Kelly doesn't use the term spacing mechanism (pp. 96, 99).

Kelly notes that internal war may be triggered by factors other than resource competition, and that peace may be achieved by nonviolent conflict resolution. In contrast, external war is endemic and resource competition is the main cause (p. 102). However, again, there is only individual liability and responsibility, not group (p. 103). (Also see p. 105).

Kelly (p. 119) concludes the Andaman case: "While external war is unremitting and constitutes a condition of existence that defines the boundaries of the niches exploited by two populations, internal war originates as an alternation of war and peace, that is, as a war/peace system. In light of this, the origin of (internal) war is also at the same time the origin of peace (as a socially constructed condition)."

In Chapter 4, "The Early Coevolution of War and Society," Kelly (121) begins with this statement: "The origin of war is a question of enduring interest because the conclusions reached are of central relevance to our conceptions of human nature, and such conceptions inform the political philosophies that shape and legitimize our social institutions. The origin of war is thus much more than a matter of antiquarian curiosity." Then he counters the view of human nature and war advanced by Thomas Hobbes and others by asserting that the ethnographic and archaeological evidence reveals that war was not the primeval, universal, and pervasive condition of early human societies (pp. 121-122, 124, 160).
However, Kelly asserts that capital punishment is a cultural universal and in some situations may lead to blood feud (p. 123).
Also he says that societies without a concept of social substitutionability have the lowest frequency of warfare (p. 128).

Kelly stresses again the importance of the definition of war (122-123, 139, 140-142). He points out that war is not a unitary phenomena (p. 122). Also he notes that war is episodic and alternates with peace (p. 124).

Kelly distinguishes between the social type (unsegmented or segmented) and the economic type of a culture. He notes that whereas the economic type varies with ecology, the social type doesn't necessarily do so. Thus, the economic type and social type are not necessarily closely correlated (pp. 127, 129).

A new consideration is introduced in this chapter, the political factor, and this in terms of uncentralized or centralized political organization which covary with war in different ways (p. 130). The different motives of war are political control, economic gain, social status, and defense (p. 131). The social type is correlated with the type of conflict and in turn with their frequency and distribution (p. 135).

Kelly clearly identifies two main points regarding the correlations between ecology, society, and war: l. "Warfare is typically rare to nonexistent within and between unsegmented foraging societies inhabiting environments characterized by low resource density, diversity, and predictability at densities below 0.2 persons per square mile" (p. 133). 2. "Spontaneous conflicts over access to resources occur both within and between unsegmented foraging societies in environments that are rich in naturally occurring subsistence resources, that are characterized by high resource density, diversity, and reliability, and that support population densities in excess of 0.2 persons per square mile. The incidence and severity of conflict is amplified by higher population densities and/or environmental circumscription...." (p. 136). These statements point to ecology as the key to whether or not warfare occurs in unsegmented hunter-gatherer societies, or more specifically, resource availability and also population density and/or circumscription or compression (p. 144). [Compression refers to when the foraging area of a group is reduced because of encroachment from another group or society]. In short, resource reliability and abundance are associated with war rather than paucity, according to Kelly (pp. 135, 138).

Here Kelly first mentions yet another type of intergroup aggression which he calls sponatenous violence, this added to homicide, capital punishment, feud, war (p. 136). The transition to war is the movement from capital punishment to spontaneous conflict (p. 139). Thus, war derives from other lesser types of violence, not from peace (p. 140). Furthermore, war is different in segmented societies (p. 146), and segmented societies may trigger war with unsegmented societies (p. 147). [See p. 143 for more detail on Kelly's model of the origin of war].

Next, Kelly tests his ethnology (cross-cultural analysis based on comparison of ethnographies) against the archaeology record and discovers remarkably close agreement (p. 158). He asserts that "nearly all early Upper Paleolithic societies were warless" (p. 147). The earliest conclusive evidence of war is from the Nubians around 12,000-14,000 B.P. (p. 148). Group responsibility and group liability can be detected in the archaeological record from multiple deaths including of children and from pincushioning (pp. 150-151). Pincushioning is when multiple arrows or other weapons are used on an individual even after death. He cites the cave art depiction of an "execution group" in Spain dated at 5000-3000 B.P. (p. 155). Also Kelly indicates that the first evidence of a fighting force is in the Mesolithic (p. 155). However, Kelly cautions that multiple deaths can also be caused by diseases and/or starvation as well as war (pp. 156-157).

Then Kelly finally summarizes the main points of his study:

1. War and society coevolve.

2. War is not primordial, but has a definite origin in the relatively recent prehistoric past, after 10,000 B.P.

3. Various forms of violence--- including homicide, capital punishment, and spontaneous conflict over resources--- extend far back into prehistory, thus there was no utopian condition of nonviolence [killing absent] as well as peace [warfare absent].

4. Within one's own local group, however, lethal violence was quite rare--- only about once/100 years.

5. In most unsegmented societies, homicide usually does not lead to further violence, such as a blood revenge cycle, unlike in many segmented societies.

6. Also aspects of positive peace are present, including peace maintaining mechanisms in unsegmented societies.

7. Peacemaking coevolves with war, and war and peace alternate, thus, even warlike societies are not always engaged in war and may experience extended periods of peace.

8. The future can be peaceful, at least at times, since alternation of war and peace are part of the nexus of human nature, war and society (pp. 160-161). To phrase this last point in another way for emphasis--- peace is just as much a part of human nature and its evolution and prehistory as is war.

Kelly has provided a most impressive meticulous, penetrating, and important critical analysis of the ultimate origin of war based on comparing ethnographic data on unsegmented and segmented hunter-gatherer societies and then testing this against the archaeological record where it is verified. His analysis rests on theory drawn from cultural evolution, ecology, and materialism. Pivotal is resource competition leading to violent conflict and possibly competitive exclusion. However, he distinguishes between resource abundance and scarcity, and also brings in other ecological factors like population density, circumscription, and compression. Furthermore, Kelly goes beyond materialism to include mentalistic factors, especially the social concept of group injury, responsibility, liability together with social substitutionability as at the root of warfare. However, he does not go into the influence of world view, values and attitudes which mentalists would emphasize.

While Kelly's study is ingenious and convincing, there are a few weaknesses, although they are not devastating to his overall argument.

1. More attention needs to be given to the impact of foreign contact on the Andamans. Kelly asserts that in the Andaman Islands there was war before European contact (p. 92). However, it is not clear the extent to which their situation may have been influenced by the British, and likely even earlier external forces from Arab, Chinese, Indian, and Burmese contacts. Kelly admits that the Andamanese hostility to outsiders may be a defensive response to slave raiders (p. 81). Also Kelly notes that during conflicts between Andamans and British, the former would try to take Western tools (pp. 84, 86-87, 89). This is reminiscent of the cases where trade goods and other Western contact influences have triggered or at least intensified and often transformed indigenous warfare as documented in War in the Tribal Zone (R. Brian Ferguson and Neil L. Whitehead, eds.).

2. More consideration needs to be given to diversity in cultures and in aggression. More problematic is the sample size of ethnographic cases used from cross-cultural comparisons to derive generalizations. Sample size ranges from seven to only a few dozen. However, there are nearly 7,000 distinct cultures extant today, and there were probably tens of thousands to hundreds of thousands during the course of four million years of human cultural evolution. Extrapolating from such a small sample to generalize about the whole of hunter-gatherer existence may be suggestive, but can it really be conclusive, especially given what is known about the enormous diversity of hunter-gatherer societies historically? Similarly, because war and other forms of human aggression encompass tremendous variation and variability in every respect, just how realistic is it to offer a general model for the origin of war in light of this?

4. The situation may be far more complicated than portrayed. Despite the intricacies of Kelly's study and argument, he repeatedly exhibits simple dualistic thinking--- hunter-gatherers fall into only two social types (unsegmented and segmented); political organization is either uncentralized or centralized; historic ethnographic societies are either foragers or farmers; and resources are either scarce or abundant. It is not always so simple and clear.

5. Kelly's study would have been stronger had he considered some other especially relevant studies, such as Bonta on peaceable societies, or Dyson-Hudson and Eric A. Smith on territoriality. Also he should have included more commentary and refutation of potentially competing approaches and information, such as Napoleon Chagnon's sociobiological explanation of Yanomami aggression and Richard Wrangham's work on Gombe chimpanzee aggression. While there is some passing mention of positive peace and peacemaking,
a final chapter focused on this would have added a great deal to the book. After all, Kelly asserts that most of hunter-gatherer existence is peaceful (warless) if not nonviolent. Isn't that just as important to document and explain?

Finally, is there anything of relevance in Kelly's study for current and future warfare?

1. If nonviolence and peace prevail in most hunter-gatherer societies, especially unsegmented ones, and they represent 99% of human existence, then this means that human nature evolved under these conditions and warfare is an episodic and temporary aberration in human experience. Those interested in peace need to think about war, but they also need to think far more about peace. Nonviolence is not achieved through violence, and peace is not achieved through war. Those who think so are restricted by the negative concept of peace as nothing more than the absence of war. Peace has far more potential than is generally appreciated.

2. Kelly distinguishes war from capital punishment by the presence of social concepts of group injury, group liability, group responsibility, and social substitutionability. Modern societies have transcended many of these obstacles to peace. Police and judicial systems target specific individuals for their criminal acts. In other words, social substitutionabilty no longer applies, except in the case of terrorism. An increasing number of countries including most of those in Europe have abandoned capital punishment all together. Military doctrine and tactics in recent warfare by the U.S.A., among other countries, attempts to minimize civilian casualties, something made feasible by advances in military technology (for example, so-called smart bombs). Military intervention has increasingly aimed at peace restoration and maintenance, even if some force is initially employed.

3. Finally, as Margaret Mead, Ashley Montagu, and many other anthropologists since have argued, and as Kelly demonstrates once again, war is a social invention not biological imperative. It evolved under certain conditions very late in human prehistory. Humanity continues to evolve and can evolve away from war and further toward peace. Increasingly it is being recognized that while conflicts are inevitable, there are many nonviolent and peaceful alternatives for their resolution. Indeed, this is one of the major reasons for the United Nations and many other international and national agencies it has had some genuine successes. The long-term perspective offered by anthropology, and in particular by studies such as that of Kelly, reveals that war is only a relatively recent aberration, and looking ahead in the long-term it may eventually be abolished or simply become obsolete like blood feuding, slavery, apartheid, capital punishment, and so on. While human rights have ancient roots their systematic formal institutionalization internationally is largely post-WWII, and yet much progress has been made a lot more is likely to be made in the future. Respect for human rights is peace, as the Zapotec of Mexico recognize.

Further Reading

Boehm, Christopher, 1986, Blood Revenge: The Enactment and Management of Conflict in Montenegro and Other Tribal Societies, Philadelphia, PA: University of Pennsylvania Press (especially pp. 202-236).

Bonta, Bruce D., 1993, Peaceful Peoples: An Annotated Bibliography, Methuchen, NJ: Scarecrow Press.

Bonta, Bruce D., 1996, "Conflict Resolution Among Peaceful Societies: The Culture of Peacefulness," Journal of Peace Research 33(4):403-420.

Chagnon, Napoleon A., 1997, Yanomamo, Fort Worth, TX: Harcourt Brace.

DiCanio, Margaret, 1993, "Revenge," The Encyclopedia of Violence, New York, NY: Facts on File, pp. 226-229.

Ericksen, Karen Paige, and Heather Horton, 1992, "Blood Feuds: Cross-Cultural Variations in Kin Group Vengeance," Behavior Science Research 26(1-4):57-85.

Ferguson, R. Brian, 1992, "Tribal Warfare," Scientific American 266(1):108-113.

Ferguson, R. Brian, and Neil L. Whitehead, eds., 1992, War in the Tribal Zone: Expanding States and Indigenous Warfare, Santa Fe, NM: School for American Research.

Levinson, David, 1994, "Feuding," Aggression and Conflict: A Cross-Cultural Encyclopedia, Santa Barbara, CA: ABC-CLIO, pp. 63-66.

Otterbein, Keith F., 1997, "The Origins of War," Critical Review 11(2):251-277.

Otterbein, Keith F., and Charlotte S. Otterbein, 1965, "An Eye for an Eye, a Tooth for a Tooth: A Cross-Cultural Study of Feuding," American Anthropologist 67:1470-1482.

Rice, O.K., 1982, Hatfields and the McCoys, Lexington, KY: University of Kentucky Press.

Robarchek, Clayton, and Carole Robarchek, 1998, Waorani: The Contexts of Violence and War, New York, NY: Harcourt Brace College Publishers.

Sponsel, Leslie E., 1996, "The Natural History of Peace: A Positive View of Human Nature and its Potential," A Natural History of Peace, Thomas Gregor, ed., Nashville, TN: Vanderbilt University, pp. 95-125.

Waller, Altina L., 1988, Feud: Hatfields, McCoys, and Social Change, Chapel Hill, NC: University of North Carolina Press.

Wrangham, Richard, and Dale Peterson, 1996, Demonic Males: Apes and the Origins of Human Violence, New York, NY: Houghton Mifflin.