©Journal of Bioeconomics, 9:2, 2007
Synergy — otherwise unattainable combined effects that are produced by two or more elements, parts or individuals — has played a key causal role in the evolution of complexity, from the very origins of life to the evolution of humankind and complex societies. This theory — known as the “Synergism Hypothesis” — also applies to social behavior, including the use of collective violence for various purposes: predation, defense against predators, the acquisition of needed resources (food patches, nest-sites, water supplies, raw materials, territories, even mates), and the defense of these resources against other groups and species. Among other things, there have been (1) synergies of scale, (2) cost and risk sharing, (3) a division of labor (or, better said, a “combination of labor”), (4) functional complementarities, (5) information sharing and collective “intelligence” and (6) tool and technology “symbioses”. Many examples can be seen in the natural world — from predatory bacteria like Myxococcus xanthus to social insects like the predatory army ants (Eciton burchelli) and the colonial raiders Messor pergandei, mobbing birds like the common raven (Corvus corax), cooperative pack-hunting mammals like wolves, wild dogs, hyenas and lions, coalitions of mate-seeking and mate-guarding male dolphins, the well-armed troops of savanna baboons (with their formidable canines), and, closest to humans, the group-hunting, group-raiding and even “warring” communities of chimpanzees. Equally significant, there is reason to believe that various forms of collective violence were of vital importance to our own ancestors’ transition, over several million years, from an arboreal, frugivorous, mostly quadrupedal ape to a world-traveling, omnivorous, large-brained, tool-dependent, loquacious biped. The thesis that warfare is not a recent, “historical” invention will be briefly reviewed in this paper. This does not mean that humans are, after all, “killer apes” with a reflexive blood-lust or an aggressive “drive.” The biological, psychological and cultural underpinnings of collective violence are far more subtle and complex. Most important, the incidence of collective violence — in nature and human societies alike — is greatly influenced by synergies of various kinds, which shape the “bioeconomic” benefits, costs and risks. Synergy is a necessary (but not sufficient) causal agency. Though there are notable exceptions (and some significant qualifiers), collective violence is, by and large, an evolved, synergy-driven instrumentality in humankind, not a mindless instinct or a reproductive strategy run amok.
The thesis of this paper is that collective violence in human societies” ranging from gang fights in urban neighborhoods to the cataclysm of World War Two” is not a recent cultural invention, though many students of feuding and warfare over the years have claimed otherwise. Rather, our well-documented propensity for violent social conflict represents a cultural elaboration of a deeply-rooted evolutionary pattern; it is a variation (albeit with some unique characteristics) on a ubiquitous phenomenon in the natural world, and there is much to be gained from viewing it in this light.1
Furthermore, there is reason to believe that this important trait, which has been the cause of heart-breaking tragedies and immense human suffering in recent times, can be traced back in evolutionary history to the emergence of our earliest bipedal ancestors. A “plausibility argument,” based on the accumulating evidence in paleoanthropology and related disciplines, will be presented here for the hypothesis that collective violence for various purposes has also played a vital role in our evolution as a species; we have been full participants in this widespread behavioral pattern for several million years.
Many students of human violence view organized warfare “the apex of our undeniably violent propensities” as a categorically distinct and historically recent phenomenon. Political scientist Claudio Cioffi-Revilla (1996:1), for instance, argues that Alethal conflict among organized, armed, and opposed social groups” originated in the Neolithic. And historian John Keegan (1993:121), while acknowledging the evidence of Araiding, and Arouting, behaviors among hunter-gatherers, views these patterns as being “below the military horizon” meaning that they were/are not pre-planned and organized Acampaigns, and therefore not true warfare. He equates the emergence of warfare with the rise of a specialized Amilitary class.” (Similar views have been adopted by, among others, Marilyn Keyes Roper 1975, Robert, Connell 1989, and contributors to the edited volume by Jo Groebel and Robert A. Hinde 1989.) Many anthropologists over the years have also uncritically accepted and even promoted this interpretation, as Lawrence Keeley documents in War Before Civilization (1996).
However, I believe such a categorical distinction is not justified. It decouples warfare from the long pre-history of collective violence in humankind; it ignores our continuity with the rest of the natural world; and, most important, it contravenes the fact that modern warfare shares a unifying causal principle, a deep homology with all of the other, multifarious forms of collective violence in nature. Or so I will argue here.”
It is hardly a novel idea that human violence is somehow a part of the natural world, for better or worse. In fact, there is a vast literature on this theme, philosophical, political/historical and, increasingly, scientific, tracing back at least to the likes of Aristotle, Nicolo Machiavelli, Thomas Hobbes, Thomas Malthus and Herbert Spencer, as well as to Social Darwinists like Ludwig Gumplowicz (1883), Walter Bagehot (1884), William Graham Sumner (1911) and others. In more recent times, Sir Arthur Keith (1947), Raymond Dart (1959), Konrad Lorenz (1966), Robert Bigelow (1969), Keith Otterbein (1970) Robert Ardrey (1976), Richard Alexander (1979, 1990), Paul Shaw and Yuwa Wong (1989), Johan van der Dennen (1995), Lawrence Keeley (1996), Richard Wrangham and Dale Peterson (1996), and many others have advanced similar arguments. (My own earlier work on this subject can be found in Corning 1971, 1973, 1975; also Corning et al. 1977.) See especially the reviews by van der Dennen 1990, 1991, 1995, 1999. Van der Dennen reminds us of how often the early writings, and insights, have been overlooked by modern theorists.
Charles Darwin was one of the major figures in this “naturalistic” scholarly tradition. In The Descent of Man (1874), Darwin assigned a key role to collective violence in the evolution of humankind. He posited that competition between “tribes” was likely to have been endemic in human evolution and that such conflicts were often “settled” by violence. This dynamic, in turn, became the instrumentality for a process of differential selection between groups which, over time, favored the higher intellectual, social and moral faculties, not to mention advanced weapons and technologies. At the end of a frequently-quoted passage in his masterwork, Darwin concluded that: ASelfish and contentious people will not cohere, and without coherence nothing can be effected. “tribe rich in the above [higher] qualities would spread and be victorious over other tribes” (1874, I:200).
Darwin’s “scenario” has since come under deep suspicion, because (it is claimed) the mechanism of “group selection” is problematical as an agent of social evolution while the (supposedly) more robust mechanism of “kin selection” (or inclusive fitness) seems insufficient to explain altruistic behaviors in large groups of unrelated individuals (see, for instance, Cronin 1993; Wright 1995). As van der Dennen (1999: 170) puts it: “The problem is still unsolved.” Of course, the opponents of Darwinism reject his conflict-oriented scenario altogether. We will revisit this issue below, where I will describe some new thinking that may enable us to restore a full measure of credibility to Darwin’s explanation.
But more important, the theory that will be presented here speaks to the urgent need to understand more fully the underlying causes of collective violence. We must strive for a deeper understanding of what sustains this multi-faceted behavioral pattern, which is at once universal in humankind even routine, and yet remains elusive and unpredictable to a significant degree. The imperatives for doing so should be self-evident.
We do not currently lack for theories of collective violence, which identify many significant contributing factors. These will be discussed by other contributors to this volume. However, I believe that the current “inventory” of theories is insufficient, and that one crucially important (indeed necessary) causal agency remains underappreciated and sometimes overlooked or neglected. The theory, in a nutshell, is that various forms of functional synergy are a major causal agency, necessary but certainly not sufficient, in precipitating collective violence in nature and humankind alike. Synergy is, in fact, a universal “common denominator” and is very often a major inducement and “driver” for these behaviors. I call it a “bioeconomic” theory of collective violence. I will explain why in due course.
The Synergism Hypothesis
The theory of collective violence that will be proposed here is derived from a broader, “bioeconomic” theory of cooperation and complexity in evolution that I refer to as the Synergism Hypothesis. The theory was first proposed in a book-length monograph more than 20 years ago (Corning 1983), but it has only recently begun to draw the attention of evolutionary biologists and evolutionary anthropologists (see especially Maynard Smith and Szathmáry 1995, 1999). The history of this theory is recounted elsewhere, along with detailed explications (see Corning 1983, 1995, 1996a,b, 1997, 1998a,b, 2003). Accordingly, I will outline only the essential points here.
In the conventional Neo-Darwinian paradigm, natural selection is viewed as the primary “mechanism” of evolutionary change. Evolution is characterized as a process of “blind variation and selective retention,” in psychologist Donald T. Campbell’s (1974) popular slogan. Accordingly, paleontologist George Gaylord Simpson asserted that “The mechanism of adaptation is natural selection … [It] usually operates in favor of maintained or increased adaptation to a given way of life” (1967:219). Similarly, biologist Ernst Mayr informed us that “Natural selection does its best to favor the production of programs guaranteeing behavior that increases fitness”(1976:365). And Edward O. Wilson, in Sociobiology, assured us that “natural selection is the agent that molds virtually all of the characters of species”(1975:67).
The problem is that natural selection is not a “mechanism.” Natural selection does not do anything; nothing is ever actively “selected” (although sexual selection and predator-prey interactions involve special cases). Nor can the sources of causation be localized either within an organism or externally in the environment. In fact, the term natural selection is a metaphor for an important aspect, or property of the ongoing evolutionary process. (Darwin’s inspiration for his metaphor was the “artificial selection” practiced by animal breeders.) Natural selection is really an “umbrella concept” that refers to whatever functionally-significant factors (as opposed to historical contingencies, fortuitous effects or physical laws) are responsible in a given context for causing differential survival and reproduction. Properly conceptualized, these causal factors are always relational; they are defined both by organism(s) and their environment(s), and by the interactions between them.
This crucially important point can be illustrated with a textbook example of evolutionary change, “industrial melanism.” Until the Industrial Revolution, a “cryptic” (light-colored) strain of peppered moths called Biston betularia, predominated in the English countryside over a darker “melanic” form. The wing coloration of cryptic strain provided camouflage from avian predators (like thrushes) as the moths rested on the mottled trunks of lichen-encrusted trees. This gave them an advantage over the darker form (carbonaria), which stood out. As a result, the melanic form was relatively rare. But as soot progressively blackened the tree trunks in areas near England’s growing industrial cities, the relative frequency of the two forms was eventually reversed; the birds began to prey more heavily on the light, cryptic strain while the darker, melanic strain became less visible (Kettlewell 1955, 1973).
The question is, where in this example was natural selection “located”? What was the “mechanism”? The short answer is that natural selection included the entire configuration of factors that combined to influence differential survival and reproduction. In this case, an alteration in the relationship between the coloration of the trees and the wing pigmentation of the moths, as a result of industrial pollution, was an important proximate factor. But this factor was important only because of the inflexible resting behavior of the moths and the feeding habits and perceptual abilities of the birds. If the moths had been subject only to insect-eating bats, which use “sonar” rather than a visual detection system to catch insects on the wing, the change in background coloration would not have made any difference. Nor would it have mattered if there were not genetically-based differences in wing coloration that allowed for “selection” between the two alternative forms.
Hence, one cannot (technically) speak of “mechanisms” or fix on a particular “selection pressure” in explaining the workings of natural selection; these are really only shorthand expressions. One must focus on the interactions that occur within an organism and between the organism and its environment(s), inclusive of other organisms. Natural selection as a causal agency refers to the functional consequences produced by adaptively significant changes in a given organism-environment relationship.
What are the factors that are responsible for initiating such changes? In other words, what are the sources of the “variations” that precipitate natural selection? The answer, of course, is many things. It could be a functionally-significant mutation, a chromosomal transposition, a change in the physical environment, a change in one species that affects another species, or it could be a change in behavior that results in a new organism-environment relationship. In fact, a whole sequence of changes may ripple through any complex network of relationships. For instance, a climate change might alter the ecology, which might induce a behavioral shift to a new habitat, which might encourage an alteration in nutritional habits, which might precipitate changes in the interactions among different species, resulting ultimately in the differential survival and reproduction of alternative physical traits and the genes that support them. However, it is the functional consequences (the “payoffs”) of these changes that cause natural selection.
Another way of putting it is that natural selection does not “select” genes; it differentially rewards, or disfavors, genes and gene complexes based on the effects produced by the phenotypes from whatever source, including genetic, developmental, environmental, experiential and even cultural influences. (On this crucially important point, see especially Corning 1983, 2003; also Odling-Smee et al., 2003; Weber and Depew, 2003; Richerson and Boyd, 2004.)
The Synergism Hypothesis represents an extension of this line of reasoning. Simply stated, cooperative interactions of various kinds, however they may occur, can produce novel combined effects “synergies” that in turn become the causes of differential selection. The “parts” that are responsible for producing those synergies (and their genes) then become interdependent “units” of evolutionary change. In other words, it is the payoffs associated with various synergistic interactions in a given context that constitute the underlying cause of cooperative relationships “and complex organization” in nature. The synergy produced by the “whole” provides the functional benefits that may differentially favor the survival and reproduction of the “parts.” (It should be noted that synergy is defined here very broadly; it encompasses all forms of “cooperative” effects that are not otherwise attainable. For a comprehensive discussion of this definition, see Corning 2003.)
Although it may seem like backwards logic, the thesis is that functional synergy is the underlying cause of cooperation (and organization) in living systems, not the other way around. Synergy is not just a synonym for cooperation; it refers to the functional effects that are produced by cooperation. Often these effects may be positive, or beneficial, but many times the consequences may be negative, or may have no functional significance at all. So it is really, at heart, an “economic” (or, more precisely, a “bioeconomic”) theory of complexity in evolution.
Because this may be an alien idea, let me restate it in a slightly different way. The functional effects produced by cooperation (and by complex systems) are the very cause of cooperation and complexity in evolution. The “mechanism” (as it were) underlying the evolution of complex systems is none other than the combined functional effects that these systems produce. It is the synergies that are the proximate causes of natural selection, or “synergistic selection” as John Maynard Smith (1983) calls it. Synergistic effects represent an independent (yet also interdependent) source of the “variations” that may be favored by natural selection. (All this is developed in much greater detail in Corning 2003.)
In fact, this paradigm is very similar to the way economists tell us that markets work in human societies. When a new “widget” is developed, its ultimate fate , its survival and reproductive success, so to speak, is ultimately determined by how well it succeeds in the marketplace. If the widget sells well, the “supply” is likely to increase, or so economic theory tells us. If not, the widget will soon go extinct. Many factors “internal and external” may contribute to these synergies. Moreover, the synergies are always historically-contingent and situation-specific. They are not the predictable product of a prime mover, or the inexorable outcome of any self-organizing dynamic, much less the working out of some deterministic “law” of evolution. The historical context is always a part of the causal “matrix.” (More on this below.)
In case it needs to be stated explicitly, this theory is also fully consistent with Darwin’s theory, though it is definitely not gene-centered. It is also compatible with the work in recent years on a multi-level selection paradigm (see especially D.S. Wilson 1983;1997; Jablonka 1994; Jablonka and Lamb 1995; Wilson and Dugatkin 1997; Sober and Wilson 1998; Odling-Smee et al., 2003). By the same token, this theory is quite compatible with, although it also differs from, the various theories of “rational choice” that abound in economic and political theory. Synergistic functional effects provide the “benefits” that may, or may not, lead to “rational” (self-interested) choices in a given context.
One example will perhaps suffice to illustrate this theory for our purpose. It demonstrates how synergy undergirds social behaviors in evolution. Long before Walt Disney discovered meerkats (Suricata suricatta) the Afrikaner name for the mongoose and gave them a featured role in The Lion King, these highly gregarious small mammals were using the synergy principle in a number of different ways to cope with the challenge of survival and reproduction in the harsh environment of the Kalihari Desert and other marginal areas in southern Africa. Renowned for their ability to stand tall on their hind legs and scan the horizon while using their long tails to form a tripod for balance, meerkats live in elaborate underground burrows with multiple-family groups of up to 30 or more animals. (Among the many references on meerkats, see Macdonald 1986; van Staaden 1994; Doolan and MacDonald 1996a,b, 1997, 1999; Clutton-Brock et al., 1999, 2000, 2001; Koenig and Haydock 2001.)
Among other things, meerkats gain synergistic benefits from huddling closely together for warmth during the cold desert nights (joint environmental conditioning and heat-sharing); they also hunt collectively and defend their burrows with noisy displays and threatening charges (what I call a Asynergy of scale); they take turns standing sentry duty to watch out for predators, such as hyenas, jackals and eagles (cost sharing and joint risk reduction); and they use various signals to communicate with their companions, including sharp warning cries when danger appears (information sharing). There is even a rough division of labor (what I prefer to call a “combination of labor”). The males are primarily responsible for defending the burrow and its dozen or so entrances, while the females and immature males share in nurturing the infants. In addition, meerkats economize on the cost of building and maintaining their burrows by “house-sharing” with non-competitive solitary yellow mongooses and social ground squirrels (an example of symbiosis).
Before turning to the role of synergy in the evolution of collective violence, one further point is critically important. From a synergy perspective, it is immaterial whether or not a particular form of synergy is biologically determined, or socially learned, or invented de novo during the lifetime of an animal, or some combination of the three. What matters are the adaptive consequences for survival and reproduction. It is the bioeconomic payoffs that count.
Furthermore, for those who are concerned about the causes of a complex social behavior like collective violence, it is also important to note that there is mounting evidence that animals are active participants in the process of adaptation; they make choices, solve problems, have “minds.” Some of the current book titles on this topic are highly suggestive: Animal Minds, Wild Minds, How Monkeys See the World, The Evolving Mind, A History of Mind, The Evolution of Consciousness, Natural Theories of Mind, and The Imitation Factor. (A few of the many recent references on this issue include Whiten 1991; Povinelli 1993; Sperry 1993; M.S. Goertzel 1993; M.S. Dawkins 1988; Lewin 1994; Plotkin 1994; Macphail 1998; Humphrey 1999; Hauser 2000; Dugatkin 2000; Griffin 2001a,b. See also the special issue of the Journal of Theoretical Biology, Volume 171(1), 1994, devoted to this subject. The topic is also discussed in some detail in Corning 2003 and Weber and Depew 2003.)
By the same token, “culture” (broadly defined here as socially transmitted learning) is not a “late-model” human invention that is radically disconnected from the evolutionary process; it is an integral part of it. Many animal behavior studies in recent years (from birds to killer whales and bonobos) have convincingly shown that culture, and cultural evolution, is a significant aspect of the overall evolutionary process, though humans are obviously exceptional exemplars. Gene-culture co-evolution is the shorthand expression that has been most frequently employed to characterize this “interactional” dynamic, at least in humankind. But the same principle applies elsewhere in nature as well. (See especially, Bonner 1980; Mundinger 1980; Cavalli-Sforza and Feldman 1981; Corning 1983, 2001; Boyd and Richerson 1985; Durham 1991; Wrangham et al., 1994; Boesch 1996; Heyes and Galef 1996; Runciman et al., 1996; Deacon 1997; McGrew 1998; Ehrlich 2000; Laland et al., 2000, Corning 2003; Odling-Smee 2003; Weber and Depew 2003.) The uniquely cumulative nature of cultural evolution in humans is discussed by, among others, Corning 1983, 2003; Boyd and Richerson (1996), Richerson and Boyd (2004), and Boesch and Tomasello (1998), who refer to a “ratchet” effect.
But most important, and this point has only recently begun to come into better focus among evolutionary theorists, behavioral changes are often the “pacemakers” of evolutionary change, as Ernst Mayr argued many years ago in a classic paper on “The Emergence of Evolutionary Novelties” (1960). In other words, the innovations and “choices” that animals make in coping with the challenges of survival and reproduction very often also affect the course of the evolutionary process. (See especially Bonner 1980; Cavalli-Sforza and Feldman 1981; Corning 1983, 2003; Boyd and Richerson 1985; Bateson 1988; Plotkin 1988; Plotkin and Odling-Smee 1979, 1981; Bateson et al., 1993; Avital and Jablonka 1994; Laland et al., 2000; Odling-Smee et al., 2003; Weber and Depew 2003; Richerson and Boyd 2004.)
In fact, this idea is not of recent vintage. It goes back at least three centuries, to Jean Baptiste de Lamarck in his Zoological Philosophy (1963). Though his theory relating to the inheritance of acquired characters has long since been rejected, nevertheless Lamarck deserves credit for being perhaps the first to propose that changes in “habits” are important as initiators of evolutionary change. Indeed, Lamarck claimed that changes in habits “come first.” Darwin, in The Origin of Species, acknowledged Lamarck’s views but was more ecumenical: AIt is difficult to tell, and immaterial for us, whether habits generally change first and structure afterwards; or whether slight modifications of structure lead to changed habits; both probably often change almost simultaneously. (1968: 215).
One illustration of this “pacemaker” effect is a laboratory experiment with a species of crossbills (birds whose mandibles cross over at the tips) that was conducted by Craig Benkman and Anna Lindholm (described in Weiner 1994). Benkman and Lindholm, in their experiment, trimmed the beaks of a group of experimental crossbills (a procedure, like cutting your nails, that does not harm the birds). This “treatment” effectively incapacitated the birds; it confirmed that the cross-over alignment of the mandibles is absolutely essential to prying open the seed cones. In other words, the crossbills’ distinctive feeding strategy could not have arisen in the first place without a prior structural variation. However, the experiment also showed that, as the birds’ bills grew back, even a small degree of crossing-over was enough to allow them to open some of the cones, however inefficiently, and that their performance progressively improved as the beak tips regenerated.
What this experiment indicates is that, just as Darwin suggested, a “slight modification” may have opened the door to a new “habit,” which natural selection subsequently “rewarded” with differential reproductive success. The new habit in turn established a new organism-environment relationship in which the causal arrows then flowed the other way as subsequent beak modifications were “tested” in relation to the new feeding strategy. The new behavior favored a more pronounced (more efficient) crossing-over, and perhaps other genetic changes as well. As we shall see, this dynamic may have had an especially important bearing on human evolution.
Synergy and Collective Violence
As noted earlier, the theory of collective violence that I am proposing here is, in effect, a “sub-theory” of the Synergism Hypothesis. It represents the application of a broad hypothesis about the evolution of cooperation and complexity generally to a well-defined category of cooperative social behaviors. The thesis, in brief, is that synergistic functional effects of various kinds have been an important (necessary, though not sufficient) causal agency underlying all the manifold forms of collective violence in nature. To repeat, the claim is that synergy is a “common denominator.”
What does “necessary but not sufficient” mean in this context? Is synergy, then, the “cause” of collective violence? The answer is both yes and no. The reason for this ambiguity is that it is important to take account of the broader issue of causation in animal and human behavior. The evolutionary/adaptive paradigm is based on the “ground-zero premise” that survival and reproduction is the ongoing, inescapable and often challenging problem for all living organisms. Furthermore, this problem is multifaceted in nature, especially for complex mammals like humankind. Indeed, humans have an array of some 14 basic needs “domains,” which form the essential requisites for survival and reproduction. These needs are elucidated (and documented in depth) in an article entitled ABiological Adaptation in Human Societies: “Basic Needs Approach” (Corning 2000).
These ongoing survival needs; which can be subsumed under the broad term “problems of living”, are therefore primary “causes” of animal behavior. They define the “goals” of biological adaptation and can (in principle) be linked to specific psychobiological, social and even cultural “motivators.” However, the specific “resources” that are available for meeting these needs in a given context; say, foraging territories, various nutrients, water, shelter, nesting sites, physical safety, even mates, are ultimately limited in quantity and are commonly (though not always) subject to competition from other organisms of the same and different species. As biologists Felicity Huntingford and Angela Turner put it in their important study of Animal Conflict (1987): “Although animals may have mutually beneficial effects on each other, most encounters between animals are riddled with “conflicts of interest” (p.5). Huntingford and Turner divide these conflicts of interest into two broad categories, “resources” and “outcomes.” But if the acquisition of mates, the capture of prey, defense of a territory, or success in avoiding a predator, i.e., physical safety, can also be termed resources, then we can refer simply to the single category of “resources.”
Accordingly, animal conflicts are primarily focused on acquiring, or defending, the resources (and outcomes) that are required for survival and reproduction. The struggle for resources is the root cause of animal conflicts of interest, and of violent conflicts. Indeed, biologist Geerat Vermeij (1987, 1994), elaborating on a suggestion of Darwin, points out that competitors, predators and dangerous prey are major agents of natural selection. Vermeij calls it the “escalation hypothesis.” (As an aside, Richard Alexander 1985, 1987, and some other contemporary theorists, e.g., Robarchek 1989; Chagnon 1990b; Gat 2000, draw a distinction between “somatic” and “reproductive” interests. Apart from the fact that the term “somatic” is somewhat obscure and glosses over the more familiar term survival needs, it draws a categorical distinction that seems unnecessary. Competition for mates can be viewed as competition for another scarce “resource.”)
But if needs and “goals”, and the competition generated by these goals, form essential elements in the causal “matrix,” so do the physical, behavioral and “mental” capabilities of the particular organism. The organism’s motivational and behavioral propensities, whether biological, learned or situational, are also causally important, as shown by the research in humans on such phenomena as nepotism, xenophobia, “balance of power” calculations and the like. All of these causal influences may play an important role. Anthropologists Carol and Melvin Ember (1992), in a cross-cultural study of warfare across 186 pre-industrial human societies, concluded that the fear of resource instabilities may be a major contributing factor. But, as anthropologist Napoleon Chagnon (1990b:80) points out: AWhile conflicts of interest over resources may be inevitable, violent (competitive) resolutions of the conflicts are not. Sometimes other “solutions” can be found.
Moreover, it is the more or less universally acknowledged core premise of modern evolutionary biology that the individual organism is the basic “unit” of survival and reproduction. Individual interests are the “ground-state” in nature. So, the question that is begged by this presupposition is what accounts for “collective” violence? Why do individual organisms cooperate in various contexts, including some that may be life-threatening? All of the other causes identified by social scientists may be relevant to this issue, but they are insufficient to account for the prevalence of collective violence, which is often a high-risk form of cooperation in the natural world.
One additional ingredient that must be added to the causal “package” is functional synergy. The functional benefits, or payoffs, that can be generated by collective action are the necessary additional cause, the cause without which collective violence will most likely not occur. Synergy is the “difference that makes a difference,” to use the mantra of anthropologist Gregory Bateson. It is synergy of various kinds that determines the cost-effectiveness of collective violence, and the very fact that collective violence is so often contingent (as we shall see) is closely related to the ebb and flow of various kinds of functional synergy in a given situation.
Synergy is not a unitary phenomenon, however. As suggested in the meerkats example above, there are many different kinds of synergy. Indeed, as already noted, there can be negative synergy as well as positive synergy (see Corning 2003). There can also be high and low synergy, depending upon how effectively the opportunities for synergy are exploited. So the synergies associated with a given instance of collective violence must always be identified and assessed on a case-by-case basis. Below is an abbreviated discussion of some of the more common (and relevant) forms of synergy. This will be followed by a brief survey of the animal research literature related to three major categories of collective violence: (1) anti-predation, (2) predation (or “hunting” behaviors), and (3) resource competition (both within and across species).
Synergies of scale: This is one of the most common synergies associated with collective violence. The term refers to the well-established fact that the sheer number of participants may produce combined effects that could not be achieved by any individual, or even a smaller group. In nature, larger numbers are often decisive in violent conflicts, as we shall see. A larger coalition of male lions may have greater success in commandeering a group of females and gaining mating privileges. A larger group of chimpanzees typically have greater success in capturing prey. By forming raiding parties in the hundreds of thousands, army ants can overwhelm much larger prey. And in a recent study of meerkats, smaller groups (with fewer sentinels) were found to have higher rates of predation than do the larger groups (Clutton-Brock et al., 1999; Blumstein 1999). Likewise, in human societies, it is one of the axioms of warfare that “God is on the side of the big battalions,” as Voltaire observed. In the military science textbooks and war gaming situations, military planners typically call for the attacking force to have a multiple of the defenders’ troops, in order to ensure success. (These days technological advantages may offset a numerical advantage, but this is an exception based on another form of synergy in nature, animal-tool symbiosis. Some specific examples will be cited below.)
Cost and/or risk sharing: The meerkats also provide one of many examples where the costs or risks associated with individual defense or offense can be reduced by sharing them with other “like-minded” individuals. Many studies have shown that animals which live or forage in groups may be able to reduce the amount of time they spend looking out for predators while, at the same time, reducing their risks and increasing their feeding efficiency. (Among others, see Pulliam 1973; Caraco et al., 1980; Roberts 1996). Thus, according to Hoogland (1981), larger colonies of prairie dogs (Cynomys ludovicianus) enjoy both an increase in the ability to detect predators and less individual scanning effort. Similarly, animals that migrate in flocks may jointly reduce their risk of becoming the victim of a predator (see below).
A Division of labor: One of the most important sources of synergy in the natural world involves what the economists call a “division of labor.” Plato was perhaps the first social theorist to appreciate that synergy lies at the very foundation of human societies; the division of labor produces mutually beneficial results because different people may have different aptitudes, and specialization may increase a person’s skill and efficiency. Plato, in his great philosophical dialogue, the Republic, wrote: “things are produced more plentifully and easily and of better quality when one man does one thing which is natural to him and does it in the right way, and leaves other things” (1946, II: 68). The classical economist Adam Smith, in The Wealth of Nations (1964), provided us with one of the textbook examples. At a pin factory that Smith had personally observed, ten workers performing ten different tasks were able to manufacture about 48,000 pins per day. But if each of the laborers were to work alone, attempting to perform all of the tasks associated with making pins rather than working cooperatively, Smith doubted that on any given day they would be able to produce even a single pin per man.
The writers of modern-day economics textbooks are fond of using Adam Smith’s pin factory as an illustration of the division of labor, but this characterization downplays the synergy involved. Another way of looking at the pin factory is in terms of how various specialized skills, tools and production operations were combined into an organized “system.” It should really be called a “combination of labor.” The system included not only the roles played by each of the workers, which had to be precisely coordinated, but also the appropriate machinery, energy to run the machinery, sources of raw materials, a supporting transportation system and (not least) markets where the pins could be sold to recover production costs. Not only that but the pin factory required “management” — one or more persons responsible for hiring and training workers, for planning, for production decisions, for marketing and selling the pins, for payroll and bookkeeping, and so forth. In other words, the economic benefits (the synergies) realized by the pin factory were the result of the total system, including a complex network of production tasks and cooperative relationships.
The division/combination of labor is also widespread in nature. Here I will provide just two illustrations. A number of other examples will be described below. It happens that more than 50 species of aphids have morphologically specialized castes of “soldiers.” In a recent study of the gall-forming aphid (Pemphigus spyrothecae) by Foster and Rhoden (1998), it was found that galls containing an equal number of soldiers and non-soldiers were approximately ten times less likely to be attacked by a predator than galls that had no specialized soldiers.
The other example involves a symbiotic relationship between two species. The African honey guide (Indicator indicator) is an unusual bird with a peculiar taste for beeswax, a substance that is more difficult to digest even than cellulose. In order to obtain beeswax, however, the honey guide must first locate a hive and then attract the attention of a co-conspirator, such as the African badger (or rat). The reason is that the ratel has the ability (which the birds do not) to attack and dismember the hive, after which it will reward itself by eating the honey while leaving the wax behind for the birds. However, this unusual example of cooperative predation also depends upon a third co-conspirator. It happens that the honey guides cannot digest beeswax. They are aided by a symbiotic gut bacterium which produces an enzyme that can break down wax molecules (Bonner 1988). So this improbable but synergistic feeding relationship is really triangular. And, needless to say, the partnership would not work without the services provided by the bacteria. It is truly a combination of labor.
Functional complementarities: Closely related to a division/combination of labor is the concept of a functional complementarity. Many functional combinations in the natural world do involve labor “inputs,” but many more do not. For instance, some species of crabs form symbiotic relationships with sea anemones. They do not divide up a single task but provide complementary functions. The crabs provide legs and mobility for the partnership while the sea anemones, armed with an array of potent, poison-filled stingers (nematocysts), provide the crabs with a formidable defensive weapon against potential predators (Huntingford and Turner 1987:15, 42). Another example can be found in the honey guide-ratel partnerships described above. The process of searching for bee hives and destroying them involves an unusual combination of abilities, and the nutritional preferences for different parts of the hive represent a complementarity of goals which, in fact, is essential to the success of their partnership.
Information sharing and collective “intelligence”: Animals frequently communicate information to one another, whether “intentionally” or not. This fact was recognized by Darwin, who developed some of the underlying principles in one of his lesser-known but now more widely appreciated works, The Expression of the Emotions in Man and Animals (1965). Indeed, it is now recognized that communication processes represent an important facilitator for agonistic and violent behaviors generally, and that many forms of synergy may be involved. For instance, alarm calling, recruitment calling and other forms of signaling (like the rump displays in white tailed deer) are commonplace; many animals may be able to benefit from the information acquired by only one member of a group, and a collective effort (many eyes and ears) may greatly improve each individual animal’s odds of avoiding a predator. Sometimes there is an increased risk to the caller, but often these are minimal and the collective benefits may be substantial. Meerkats provide just one example.
Furthermore, as the research literature on animal communications has accumulated over the past half century or so, it has become evident that many species provide information to one another that is highly specific. A frequently cited example is the vervet monkeys (Cercopithecus aethiops) that emit different alarm calls to denote different kinds of predators (Seyfarth et al., 1980a,b). (For other examples, see Cheney and Seyfarth 1996). Many more species differentiate in their communications between different kinds of food sources, for instance, many birds (reviewed in Naguib et al., 1999; also see Bugnyar et al., 2001), primates (Hauser et al., 1993; Hauser and Marler 1993) and bats (Wilkinson and Boughman 1998). Ward and Zahavi’s (1973) well-known “information center hypothesis” for colonial roosting and breeding birds has also been supported in subsequent research (reviewed in Buckley, 1997). And the famous “waggle dance” of honeybees remains a classic example of “symbolic” communication in nature (von Frisch 1967; for updates that draw on more recent studies see Gould and Gould 1995; and Seeley 1995.)
Finally, there is a growing literature on collective “intelligence” and synergistic decision-making in social species, even in small-brained honeybees, ants and termites (see especially Deneubourg and Goss 1989; Camazine and Sneyd 1991; Franks et al., 1991; Seeley 1995; Saffre et al., 1999). There has been relatively little research on this phenomenon to date in social carnivores and primates (to my knowledge), though one classic example in baboons was described many years ago by Hans Kummer (1968). Of course, there is a large volume of research, especially in social psychology and the management sciences, on synergistic decision-making in human groups (Maier 1970; Hackman and Morris 1975; Hill 1982; Blumberg et al., 1983; Kernaghan and Cooke 1986, 1987; Cooke and Kernaghan 1987; Boehm 1996, 1999). Among other things, it is well-documented that groups generally can produce better (and more effectively implemented) decisions than do individuals. (A well-known consulting firm that specializes in organizational “team-building” is named Human Synergistics. See Lafferty and Associates 1975.) Anthropologist Christopher Boehm (1996, 1999) makes a similar case for the broader process of human evolution (see below).
Tool and technology “symbioses”: One of the traditional rubrics used to discriminate between warfare in humankind and other forms of collective violence in the natural world is our use of weapons and other “technologies.” In essence, tools and weapons represent a major form of synergy, a cooperative effect (or effects) that are not otherwise attainable. Among other things, weapons can increase the force or reach of a blow, or enable the user to strike a blow from a distance, as well as providing the ability to cut or penetrate the flesh of an adversary. Likewise, technologies that are designed to protect the user against the blows of an adversary, like a shield or body armor, represent important defensive weapons.
However, humans are hardly unique in being able to exploit these tool-related synergies. It is important from a theoretical standpoint to recognize that tool/weapon synergies are also commonplace in the natural world. In fact, a great many species, throughout the phylogenetic spectrum, have evolved specialized anatomical structures and armaments that serve in whole or in part for offensive or defensive purposes. The nematocysts deployed as defensive weapons by sea anemones were mentioned earlier, but even some bacteria produce a poison, called colicin, that is used to kill other bacteria. Many ants also have effective stingers. Crustaceans like crabs use their powerful claws (chelae) to threaten or engage an opponent. Stag beetles use their enlarged mandibles to defend themselves. The so-called gladiator frogs have sharp, scythe-like spines on their forelimb thumbs, which can inflict severe wounds on an opponent. Many birds use their beaks and claws as weapons, both in predation and in self-defense. Contrary to popular belief, the horns and antlers of many animals are not typically used for self-defense, but for display and for sexual competition with conspecifics. Male antelopes, for example, lash out with their sharp hooves when engaging a predator and use their horns primarily for pushing matches with rivals.
Naked mole-rats, which live in underground burrows, are endowed with outsized front teeth (thus their affectionate nickname, “sabre-tooth sausages”). The mole-rats’ teeth are used in part as digging tools for their extensive tunnel systems, which are carved out of hard-packed soils, and for procuring underground tubers and roots. However, the mole-rat “militia” (normally the largest males) also use their teeth as potent weapons against intruders, like snakes and other rodents. Likewise, among primates the long, sharp canines of savanna baboons are legendary. They provide formidable weapons for use in contests with one another and, often, against much larger predators. Finally, cataceans like whales and dolphins sometimes engage in open-mouthed attacks, though their tails are also commonly used as weapons. (See especially Huntingford and Turner 1987; Sherman et al., 1991; Kummer 1968; Strum 1987.)
Some theorists might object that there is still a fundamental difference between human weapons and the anatomical specializations of other animals. Humans fabricate tools/weapons from external objects and materials. True, but so do a number of other species. E.O. Wilson (1975: 172-175) provided a sampler, compiled from a number of sources. The list included solitary wasps, ant lions, archer fish, four species of finches on the Galápagos Islands, as well as the brown-headed nuthatch, the black-breasted buzzard, the Egyptian vulture, the black cocatoo, northern blue jays and sea otters.
Indeed, tool/weapon use is even more common (and sophisticated) among the primates, especially the great apes (chimpanzees, bonobos, gorillas and orangutans) and capuchin monkeys (Cebus apella, Cebus capucinus). Not only do capuchins make and use tools in diversified ways (Chevalier-Skolnikoff 1989; Visalberghi 1990; Westergaard and Suomi 1994) but they also use “clubs” in violent encounters with predators like snakes (Boinski 1988).
Chimpanzees are especially impressive tool-users. They frequently use saplings as whips and clubs (one famous incident will be described below); they throw sticks, stones and clumps of vegetation with a clearly hostile intent (but rather poor aim); they insert small sticks, twigs and grasses into ant and termite holes to “fish” for booty; they use sticks as pry bars, hammers, olfactory aids (to sniff out the contents of enclosed spaces), and even as toothpicks; they also use stones as anvils and hammers (for breaking open the proverbial tough nuts); and they use leaves for various purposes — as sponges (to obtain and hold drinking water), as umbrellas (large banana leaves are very effective), and for wiping themselves in various ways (including chimpanzee equivalents of toilet paper and “sanitary napkins”). Not only are chimpanzees proficient tool-users but they can also make tools. They break off small tree branches and strip them to fabricate ant “wands”; they use their bodies for leverage when they break down larger sticks to make hammers; they work leaves into sponges; and they carefully select stones of the right size and shape for the job at hand and will then carry them to their work-sites. (See especially Beck 1980; Goodall 1986; McGrew 1992,1993; Wrangham et al., 1994; Stanford 1996.)
Elephants are also proficient tool-makers and users, perhaps second only to chimpanzees in their versatility (Chevalier-Skolnikoff and Liska 1993). Among other things, they scratch or clean their ear cavities with grass or other vegetation; they scratch their bodies with sticks held in their trunks; they wipe cuts with clumps of grass held in their trunks; they reach toward inaccessible food with sticks and hit humans with sticks held in their trunks; they throw objects with great accuracy at other animals, at humans, and at human vehicles; they brandish or wave branches, apparently to chase away flies or to threaten other animals; they lay mats of grass over their backs to keep biting flies away; they pile branches or push over a large tree onto a fence, forcing it to sag so that they can walk over it; they even use sticks and stones for making “drawings” in the dirt.
In sum, tool use is widespread in nature, and one cannot legitimately cite tools and weapons as a clear-cut dividing line between humans and other animals. To be sure, we are much more accomplished tool/weapon makers and users than any other species. But the process of cultural invention by which we arrived at this pinnacle is an integral part of a very long story. Collective violence in hominids long predated the development of sophisticated modern weaponry, or so I will argue. (For a bibliography of earlier writings on the role of weapons in human warfare, see Divale 1973; also, see Bigelow 1969; Tiger and Fox 1971; Baer and McEachron 1982a,b; O’Connell 1989; van der Dennen 1995; Wrangham and Peterson 1996.)
The Sociobiology of Collective Violence
One key issue remains to be addressed before undertaking a brief survey of the literature on collective violence in non-human animals, following which we will reconsider the role of collective violence in human evolution. This relates to the underlying theoretical problem of explaining the evolution of collective violence in nature. How can it be accounted for within a Darwinian and sociobiological perspective?
First, some background is in order. The cardinal assumption of modern Neo-Darwinian theory is that the genes are free-agents that pursue their self-interests in relentless competition with other genes. This being the case, cooperation and social organization in nature are a theoretical conundrum, especially if one assumes that cooperation depends on altruism, or self-sacrifice. Early on, this is exactly what was assumed to be the case, and the proposed solution to this apparent problem was what has come to be known as “inclusive fitness theory” (or “kin selection,” in Maynard Smith’s term). Actually, the idea was first proposed by Darwin himself. In The Descent of Man (1874), Darwin coined the term “family selection” to suggest that sociality might first arise within families, where self-sacrifices would be offset by the benefits to close relatives.
When the foundations of modern population genetics were being laid in the 1920s and 1930s, the problem of sociality was raised once again, and again the solution that later came to be called inclusive fitness was proposed by the pioneer geneticists. “Insofar as it makes for the survival of one’s descendants and near relations,” wrote the formidable J.B.S. Haldane in his seminal book The Causes of Evolution (1932), “altruistic behavior is a kind of Darwinian fitness, and may be expected to spread as a result of natural selection.” However, it was William D. Hamilton’s pair of classic papers on “The Genetical Evolution of Social Behavior” (1964a,b) that formalized the idea and gave it wings. In the process, Hamilton reinforced the tendency to equate sociality with altruism. He identified only three kinds of social behavior C altruism (self-sacrifice), selfishness (gaining at another’s expense) and spite. Only later on did he add “reciprocity” to his list. Hamilton’s model asserted that sociality (read altruism) could be accounted for in terms of its benefits to close kin.
The key to inclusive fitness theory is the coefficient of relationship (r) between two individuals. According to the conventional wisdom of population genetics, two siblings have one-half their genes in common. So their coefficient of relationship is r’1/2. Likewise, two first cousins have one-eighth of their genes in common, or r’1/8. Assuming for the sake of argument that there is such a thing as an altruist gene, it could spread in a population via natural selection if the increases in the summed fitness of close relatives due to self-sacrificing behavior by an altruistic relative were sufficient to offset that relative’s loss of fitness. In formal terms, the gain-loss ratio (k) must exceed the reciprocal of the average coefficient of relationship, or k>1/r. Legend has it that Haldane anticipated Hamilton’s model with some barroom bravado: “I would gladly give up my life for two brothers or eight cousins.” Maynard Smith, a student of Haldane, says that his mentor later changed the number of cousins to ten; he wanted to show a profit (personal communication).
Inclusive fitness theory has proven to be an important predictor of social behavior in nature, in the limited sense that sociality is frequently correlated with close kinship. Many sociobiologists and behavioral ecologists assume that the relationship is also causal and all important. However, there are several problems with this presumption. One is that the sociality-kinship correlation is far from perfect. There are also many examples in nature, and more are being found all the time, where sociality is not tied to kinship or is only loosely associated, most notably in groups that include a mixture of close kin and unrelated cooperators (see below). In other words, kinship is not a necessary precondition for cooperation. This would seem to contradict the basic assumption of inclusive fitness theory. Another way of putting it is that if r=0, inclusive fitness theory would logically seem to predict that sociality will not occur. Yet sociality among unrelated individuals does occur, quite frequently.
To cite some examples: In birds, many species have been found where immature “helpers-at-the-nest” will aid not only their younger siblings but also unrelated nestlings. (Among others, see Ligon and Ligon 1982; Mumme et al., 1988; Clarke 1989; Mumme 1992; Haig et al., 1994; plus the general review in Brown 1987.) There are also studies in birds of food sharing that is not primarily directed to relatives (for examples, see Møller 1987; Parker et al., 1994; also the review in Brown 1987).
Communal nesting and breeding among unrelated animals is also commonplace, for instance, in kingfishers (Reyer 1984), paper wasps (Queller et al., 1990), manakins (McDonald and Potts, 1994), mongooses (Creel and Wasser 1994), and halictine bees (Kukuk and Sage 1994). Also, Bernasconi and Strassmann (1999) report that, across a dozen genera of ant species, unrelated queens co-found joint colonies, which enhances worker production and colony survivorship, even though there may be subsequent conflicts between the queens. (See also Hagen et al., 1988; Queller 1988; Sherman 1988; Rissing et al., 1989; Mesterton-Gibbons and Dugatkin 1992; Breed et al. 1994; Scott 1994; Strassmann et al., 1994; Sasaki et al., 1996; De Heer and Ross 1997; Keller 1997.)
The discovery that insect colonies may consist of multiple queens and/or multiple lineages, presents a major challenge to inclusive fitness theory. For instance, biologists Joan Strassman, David Queller and their colleagues, in a study of a multi-queen neo-tropical wasp (Parachartergus colobopterus), found that colony workers showed no preferential treatment toward their own kin (Strassmann et al., 1997). In a separate study, Queller (2000) found that 35% of the foundress nest mates in the social wasp (Polistes dominulus) are unrelated and (evidently) gain only in being able to survive until later opportunities for reproduction might arise.
Likewise, in mammals there are now many studies — in male and female lions, chimpanzees, capped langurs, meerkats, evening bats and more — that involve cooperation among unrelated individuals (see Packer and Pusey 1982; Moore 1984; Packer and Ruttan 1988; Scheel and Packer 1991; Stanford 1992; Wilkinson 1992; Grinnell 1995; Goldberg and Wrangham 1997; Clutton-Brock et al., 1999, 2000, 2001). Alarm calling is also common in many social animals and birds (see below) and may benefit both related and unrelated individuals, and even other species (reviewed in E.O.Wilson 1975; also Brown 1987). Avital et al.,(1998) report that the “adoption” of unrelated juveniles has been found in over 150 species of birds, a major adaptive puzzle. In addition, Bernard Chapais (2001) argues that many cases of cooperation that have been interpreted as being based on altruistic kin-selection could instead be viewed as examples of kin-based mutualism, or reciprocal altruism.
Then there is “allomothering,” “baby-sitting,” and “adoption” by unrelated conspecifics, which has been reported in close to 300 species (Stanford 1992; Avital et al., 1998; Clutton-Brock et al., 2000; Hrdy 2001). In one of two recent papers on “babysitting” by unrelated helpers in meerkats, Clutton-Brock et al., (2000) conclude that: “Firm evidence of close associations between levels of helping behaviour and kinship in other social vertebrates is also scarce.” Something more than kinship seems to be involved in social cooperation. That “something” is functional synergy.
Another serious problem with inclusive fitness theory is that it is completely irrelevant to the explanation of mutualistic associations (symbioses) between different species B for obvious reasons. (See the in-depth review in Herre, et al., 1999; also the reviews in Corning 1997 and 2003.) Symbiosis is now recognized to be a widespread and vitally important category of cooperative relationships in nature. Indeed, a great many species, like ruminant animals with their cellulose digesting gut bacteria, are totally dependent on various symbionts for their survival. And mutualistic symbiosis is based exclusively on functional, “bioeconomic” (cost-benefit) considerations, i.e., the synergies.
Other problems with inclusive fitness theory relate to some logical complications that are not often addressed. If sociality is strongly associated with kinship, what accounts for the fact that a great many species are not at all social? It seems that kinship is neither necessary nor sufficient for sociality to occur. Also, if kinship is presumed to be causally related to cooperation, how come the most common forms of cooperation in nature occur among altogether different species (symbiosis)? In symbiotic relationships, the partners have none of their genes in common (in theory), although this long-held population genetic dogma is rapidly coming unraveled; about 31% of the genes even in yeast cells have homologues in the human genome, and humans share some 99% of their genes with chimpanzees. We do not really give up 50% of our genes when we reproduce sexually but only 50% of the genes that have functionally different effects from those of our partners. Moreover, all genes are not created equal. Indeed, our genes may turn out to be of less importance in reproduction than what was disparagingly called “junk DNA” and was largely neglected until recently B the other 98 percent of the DNA in our chromosomes. But that is another story (see Gibbs 2003).
One answer to these objections has, no doubt, already occurred to some readers. It is also possible to have cooperation that is not altruistic. I call it “egoistic cooperation,” to differentiate it from “altruistic cooperation.” It may well be that close kinship is an important “inducement” for the evolution of altruistic cooperation in nature (group selection is another possible path), but there are many other forms of sociality that are mutually beneficial; they may impose short-term costs that are later offset by equivalent or greater benefits, or they may provide immediate net benefits to the cooperators. In short, cooperation can also involve relatively straightforward “economic” calculations of costs and benefits, and the main constraint may be how these costs and benefits are toted up and distributed among the cooperators. However, it took the sociobiologists a while to recognize this fact.
The decoupling of kinship, cooperation and altruism began with the publication of another landmark theoretical paper, Robert Trivers’s “The Evolution of Reciprocal Altruism” (1971). Although Trivers’ thesis has often been referred to as the Good Samaritan scenario, this is not correct. The biblical Good Samaritan acted without the expectation of repayment; it was an uncompensated act of charity to a stranger of a different “nationality.” But under Trivers’s scenario (he did not actually provide a formal model), the term “altruist” is a misnomer. The helper acts with the assurance that a low-cost, low-risk form of assistance performed now will be repaid with interest later on. It is really reciprocity with a delayed repayment schedule. Or, to be precise, it is an investment; the ultimate benefits will greatly outweigh the costs. Reciprocal altruism is really an oxymoron.
In fact, Trivers’s paradigm ultimately relied on synergy. Trivers illustrated his concept with a human example that, again, was presaged by J.B.S. Haldane. Suppose that two strangers (non-kin) find themselves alone together next to a body of water. One of the two falls in, with a 50% chance of drowning, absent a rescuer. If the stranger comes to the victim’s aid — with a 5% chance of drowning in the process — but is later repaid when the tables are turned, the aid-giver is actually a net beneficiary. It is really a form of risk-sharing synergy, as described above.
Of course, this was also a highly contrived scenario. It would take a lot of encounters among non-kin at the water’s edge, a lot of accidents, and a lot of drownings (because the rescuer gene would be a rare mutation, initially), for such a genetic change to spread through a large population. Indeed, Trivers’ model/scenario included the assumption that “the entire population is sooner or later exposed to the same risk of drowning.”3
More credible were the three classes of real-world examples that Trivers cited to support his thesis — cleaner fish, alarm calling in birds and various types of helping behaviors in humankind. Many more examples have also been identified in the years since Trivers’s paper was published. Perhaps most famous is the blood-sharing among unrelated vampire bats (Wilkinson 1984, 1988, 1990), although behavioral biologists have also found many examples in chimpanzees and other primates, including coalitions, food sharing and mutual grooming behaviors. These are described in some detail in Wilkinson (1988) and de Waal (1996).
However, all of these real-world examples depend on “personal relationships” (whether kin or not). As Trivers himself put it in his original paper: “It will be argued that under certain conditions natural selection favors these altruistic behaviors because in the long run they benefit the organism performing them.” Commenting on various human examples, Trivers noted that “all these forms of behavior often meet the criterion of small cost to the giver and great benefit to the taker.” In other words, it is not altruism at all but mutually-beneficial, win-win reciprocity, synergy.
A more decisive decoupling of kinship, altruism and cooperation occurred when John Maynard Smith (1974, 1982b, 1984) introduced game theory (specifically, the so-called “Prisoner’s Dilemma”) into evolutionary biology in the 1970s and early 1980s. The basic premise of game theory models (so called because they involve calculated moves by each of two independent, self-serving “players”), is that the cooperators are unrelated and that they will only participate if there is an expectation of net benefits. The very purpose is to design the game in such a way that cooperation will be a “stable” strategy and will allow a gene for cooperation to spread in a population, rather than being swamped by cheaters.
However, synergy was always the secret ingredient — the hidden key in these models. It was simply disguised and hidden in the numbers used to fill in the “cells” of various payoff matrices. For instance, in the famous tit-for-tat model presented by political scientist Robert Axelrod and biologist William Hamilton (1981), which should properly be credited to the legendary systems scientist Anatol Rapoport (who proposed the solution to them), the payoffs posited for different player “strategies” were defined in such a way that mutual cooperation would be more rewarding over time than “defection,” or cheating. In the original model, Axelrod and Hamilton awarded one point to each player for mutual defection; asymmetrical cooperation yielded five points for the defector and none for the cooperator; and mutual cooperation yielded a total of six points, evenly divided. Since defectors would always be penalized by reciprocal defection in the next round, after two rounds the mutual cooperators would begin to out-gain defectors and the differential advantage for mutual cooperation would increase steadily from there on. In other words, mutual cooperation can be a highly synergistic strategy. Though for the most part unappreciated, it is synergy that ultimately drives the tit-for-tat model.
Game theory has proven to be a fertile and productive theoretical tool, but it has also been limited by its narrow and constricting assumptions about the nature of cooperation. The reason why Axelrod and Hamilton’s tit-for-tat model was hailed as a great step forward was that it added the simple assumption that the cooperators would interact more than once. The game would be “iterative,” with each player’s subsequent “moves” being affected by what had come before; in effect, it added cumulative synergies to the game. (A corollary was that the players should start out by being cooperative and then respond in kind to whatever the other player does.)
A further step toward realism occurred when zoologist Martin Nowak and mathematician Karl Sigmund (1993) developed a new kind of game theory model called “Pavlov,” which allowed the players to punish cheaters. Called “win-stay, lose-shift,” this strategy permitted a player to exclude cheaters from subsequent rounds and, by implication, from future benefits. In other words, the players could manipulate the costs and benefits.
It turns out that Pavlov conforms well with reality. It is now recognized that “policing” of cooperation and the punishment of cheaters is common in nature, and that cooperation is not so constrained by the threat of cheating as the early game theory models implied (see Boyd and Richerson 1992; Frank 1995, 1996; Michod 1996). As Timothy Clutton-Brock and Gregory Parker (1995:209) point out in a review article on the subject: “Individuals often punish other group members that infringe their interests…Punishing strategies are also used to establish and maintain dominance relationships, to discourage parasites and cheats, to discipline offspring or prospective sexual partners and to maintain cooperative behaviour.”
There are other problems with the conventional game theory paradigm. For instance, there are many circumstances in nature where the alternative to a win-win cooperative effort is not zero (the lowest value possible in a game theory payoff matrix) but death. For an animal faced with the prospect of confronting a predator, cooperative defense might be the only logical choice. Cheating, or “defection” would be self-defeating in the most dire sense of the word. So the costs/benefits for not participating may also be a critically important consideration.
Another problem is that game theory models have not as a rule (at least to my knowledge) dealt with multiple interests, where cooperation in one area — say mutual grooming — may also affect cooperation in other areas, like hunting, meat sharing, coalition-building or mutual defense. Cooperative groups may represent ongoing, multi-interest coalitions. Nor does game theory as a rule address the sometimes complex interplay between the costs and benefits associated with various choices, or “strategies.” (A useful model related to this issue was developed by Lima, 1989; see also the discussion in Grinnell et al., 1995.)
On the other hand, Herbert Gintis (2000a,b, 2003; also Bowles and Gintis 2002;) has formulated an alternative game theory paradigm B especially applicable to human sociality Gintis believes, that does not rely on kinship or reciprocal altruism. He refers to it as the “strong reciprocity” (and aggressive punishment) model (see also Henrich and Boyd 2001; Boyd and Richerson 2002). Gintis also argues that it is not possible to model human behavior on the basis of narrow, “economic” self-interests alone and that our interests include prosocial elements. This theory has been supported by a growing body of research on the evolution of a norm of “fairness” (Fehr and Schmidt, 1999; Fehr and Gächter 2000, 2002; Nowak at al., 2000; Falk et al., 2001; Henrich et al., 2001; Fehr and Fishbacher 2002; Sigmund 2002).
Another serious objection is that inclusive fitness theory, reciprocal altruism, tit-for-tat and other game theory models exclude one of the most important forms of cooperation in nature, namely, interactions that produce combined effects (synergies) that are largely self-policing because they are interdependent. Maynard Smith and Szathmáry (1995) have suggested a useful metaphor to illustrate this paradigm. Suppose that two oarsmen decide to cooperate in rowing a small boat across a river. In one alternative configuration, a “sculling” arrangement, the oarsmen each have two oars and row in tandem. In this situation, it is possible for one oarsman to slack off (to cheat) and let the other one do most of the work. This represents the classical game theory relationship.
Now imagine instead a “rowing” arrangement. In this configuration, each oarsman has only one opposing oar. Now their relationship to the performance of the boat is interdependent. If one of the oarsmen slacks off, the boat will go in circles and will not reach its goal. Interdependence has the effect of making a cooperative relationship self-policing. Maynard Smith and Szathmáry (1995: 261) conclude that the rowing model is a better representation of how cooperation (and complexity) evolves in nature. “The intellectual fascination of the Prisoner’s Dilemma game may have led us to overestimate its evolutionary importance.” (See also Maynard Smith and Szathmáry 1999.)
In fact, many forms of cooperation in nature, especially in relation to collective violence, involve “teamwork” of various kinds (where “defection” would produce “zero” rewards in the payoff matrix for all concerned). I call it the “corporate goods” model of cooperation, to differentiate it from the traditional economic concept of “collective goods.” The latter refers to goods (benefits) that may be produced by only a few members of a group (say specialized military defense), yet all the members of the group share the benefits equally; the benefits cannot be divided up. Corporate goods, on the other hand, are jointly produced by the interdependent actions of a defined group via economies of scale, combinations of labor and the like. Sometimes these goods are “collective” in nature, but at other times the goods are divisible (say a large prey animal), and the precise distribution may involve ” political” (social) choices.
Of course, teamwork is more likely if those who cooperate get a proportionate share of the goods and those who “defect” are excluded. One classic example was cited by Darwin in The Descent of Man (1874). Darwin noted that “Hamadryas baboons turn over stones to find insects, etc., and when they come to a large one, as many as can stand around, turn it over together and share the booty.” Below are two other brief examples, one in baboons and the other a variant of the honey guide-ratel symbiosis described above.
The first example involves the development of cooperative hunting behavior in a troop of olive baboons (Papio annubis) known as the “Pumphouse Gang.” In the course of studying a group of 49 baboons on a huge ranch near Nairobi, Kenya, over a period of several years, Shirley Strum, Richard Harding and several other co-workers observed the emergence and spread of a new “cultural” pattern. At first it was confined to a few adult males that opportunistically pursued and captured newborn antelopes or hares. It was a solitary activity and there was no food sharing. But over the course of time the pattern changed. The amount of predation increased; females and juveniles began to participate; food sharing became more commonplace; hunting skills and efficiencies improved; most important, the troop began to evolve systematic searches and coordinated attacks (see Harding 1973; Strum 1975a,b; Harding and Strum 1976). Moreover, it was the synergies (the proximate rewards) that drove this behavioral changes, not genetic mutations or natural selection, or inclusive fitness or even reciprocal altruism. It involved interdependent, mutualistic “corporate goods.”
The second example provides a striking illustration of nature’s opportunism, as well as an interdependent form of “corporate” synergy. It happens that the African honey guide birds mentioned above also form symbiotic partnerships with humans, the nomadic Boran people of northern Kenya. Some years ago, biologists Hussein Isack and Hans-Ulrich Reyer (1989) conducted a systematic study of this behavior pattern and found that Boran honey hunting groups were approximately three times as efficient at finding bees’ nests when they were guided by the birds. They required an average of 3.2 hours to locate the nest compared with 8.9 hours when they were unassisted. The benefit to the honey guides was even greater. An estimated 96% of the bees’ nests that were discovered during the study would not have been accessible to the birds had the humans not used tools to pry them open. Moreover, the Borans’ use of smoky fires to suppress the bees’ defenses benefitted both the humans and their symbionts. (This unique partnership also provides an example of multiple synergies. The role of the endosymbionts that break down wax molecules was mentioned above. In addition, the joint Boran-honey guide effort is coordinated by a system of two-way communications, a repertoire of vocalizations that serve as signals.)
In both of these examples of cooperative predation, the “goods” (the benefits) were the result of “team-work;” they would have been produced much less efficiently, or not at all, if any of the participants “defected” (or cheated). Yet in a review article on “Cooperation Among Unrelated Individuals” in the Quarterly Review of Biology some years ago, Mesterton-Gibbons and Dugatkin (1992) concluded that there are only three behavioral categories that fit within their definition: (1) group-selected behaviors, (2) reciprocal altruism, and (3) by-product mutualism (the latter refers to cooperation that is a coincidental by-product of individual actions).
One problem with this typology is that by-product mutualism is a category that depends on making assumptions about what is going on in the “mind” of the animal. For instance, Dugatkin (1999: 118-120) uses an example of collective predation in a species of fish called wrasse (described in detail below) to illustrate by-product mutualism. However, it is not at all clear that the individual fish are acting alone and are oblivious to one another’s behavior. If a mobbing animal will only do so in concert with others and will reliably choose flight (or avoidance behaviors) when acting alone, mobbing cannot be called an incidental “by-product.” Indeed, Dugatkin is not even certain that this is an example of synergy. “Cooperation in group foraging certainly pays off for the wrasse, but whether they do anything as a group that exceeds the sum of their own individual actions is not clear.” I must disagree. Since success in raiding a damselfish’s nest can only be achieved by a large group of wrasse acting in concert, it is an unambiguous example of a synergy of scale, whatever their “motivations.”
However, a more serious problem with the Mesterton-Gibbons and Dugatkin typology is that it leaves out “mutualism”, corporate goods like the baboon and honey guide examples described above.4 As we shall see, the multifarious kinds of “team-work” that are found in nature are especially relevant to collective violence. From an evolutionary perspective, all cooperation is a “by-product” of the synergies it produces in relation to the ultimately “selfish” interests of the participants. That is precisely what the Synergism Hypothesis asserts. When cooperation involves altruism and the beneficiaries are closely-related kin, it is called inclusive fitness; when there are reciprocal acts of generosity, it is called reciprocal altruism; and when there are mutual benefits, it is variously called mutualism, or tit-for-tat, or win-win, or a cooperative effect, or whatever. All of these cooperative behaviors are “selfish” in the broad sense of being fitness-related, regardless of the immediate motivation. Even if the behavior is the “unintentional” aggregate effect of many individual actions, it may still be synergistic. And, according to this theory, it is the synergies that drive cooperative behavior, not the other way around.
This brings me to one final issue in sociobiology, the vexed controversy over “group selection.” I will not review this debate in detail here. (For a history and analysis, see Corning 1997, 2003; also D.S Wilson 1975, 1980, 1997; Wilson and Sober 1989, 1994; Wilson and Dugatkin 1997; Sober and Wilson 1998.) In brief, the albatross that has burdened this issue is the erroneous assumption that stable cooperation in nature, and thus the differential selection of competing groups, must be based on altruism. This notion can be traced back to Darwin himself. In its modern incarnation, it was adopted most conspicuously by George Williams (1966) in his legendary critique of evolutionary theory, as well as by William Hamilton (1964a,b) in his classic papers on inclusive fitness, by E.O. Wilson in Sociobiology (1975) and by other influential theorists of the past few decades. For instance, Wilson launched his massive 1975 textbook with the assertion that altruism is the “central theoretical problem of sociobiology” (p. 3).
As noted above, there are several other possible ways for stable groups to arise in nature that do not entail altruism, including even learned behaviors (positive feedbacks) that do not even enlist the services of natural selection. I submit that both the baboon and the honey guide-Boran symbioses provide examples. To reiterate the argument above, behavioral innovations, including even “culturally” transmitted behaviors, may become the “pacemakers” and instrumentalities of differential selection among groups. Below I will propose that the pacemaker scenario applies especially to human evolution, and, furthermore, that the synergies associated with collective violence played an important role in our transformation from an arboreal, frugivorous, mostly quadrupedal ape into a world-traveling, omnivorous, large-brained, tool-dependent, loquacious biped. But first, I will set the stage with a brief review of the extensive research literature on the role of collective violence in the rest of nature. As we shall see, every one of the three broad categories of collective violence that will be reviewed here are very likely to have played an important role in human evolution as well.
Collective Defense Against Predation
Edward O. Wilson (1975) called defense against predators one of the “prime movers” of social evolution. While this form of social behavior obviously has high fitness value, the real driver is functional synergy. Wilson tacitly acknowledged as much when he went on to quote an ancient Ethiopian proverb: “When spider webs unite, they can halt a lion.” (p.37). Many examples were cited by Wilson in Sociobiology, and many more have been reported since.
Even various “passive” forms of collective defense may provide synergistic advantages, especially synergies of scale, information sharing and joint risk reduction. For instance, flying foxes, or fruit bats (Pteropus), sleep in trees together in large aggregations. Any predator trying to climb the tree will alert the entire colony, which triggers a great commotion and a mass exodus (Neuweiler 1969). Likwise, in arctic ground squirrels (Spermophilus undulatus), researcher Ernest Carl (1971) found that he could sometimes stalk isolated individuals to within three meters without being detected, while organized groups were able to detect his presence and set up alarm calls as far as 300 meters away. Similar observations were made by King (1955) in prairie dogs (Cynomys ludovicianus), where colony-members use alarm calls to alert one another. When the alarm call is given, the prairie dogs quickly retreat into their underground burrows. Indeed, so effective is the prairie dog alarm system that researcher John Hoogland reports he and his crew observed only 22 successful predation events during 73,000 hours of observations over 16 years (cited in Holmes 1996).
Flocking is also a widespread form of passive defense in birds. Wading birds like redshanks (Tringa totanus), for example, respond to alarm calls from individual birds by congregating with much agitation and then flying off together at top speed (Goss-Custard 1970). Cresswell (1994) found that larger flocks of redshanks seem to benefit both from increased vigilance and a “confusion effect,” which reduced the overall success-rate of various predators. Similarly, eider ducklings (Somateria mollissima), when attacked by a predatory gull, will rapidly coalesce and then disperse en masse (Goss-Custard 1970).
What are the advantages (the synergies) for animals that form such aggregations against predators? One often-cited advantage is the “many eyes” effect, or the “collective detection effect.” (Among the many references, and some critical discussions, see especially Pulliam, 1973; Barnard and Thompson 1985; Brown 1987; Elgar 1989; Lima 1990, 1995; Roberts 1996.). Both a synergy of scale and information sharing (and cost reductions) may be possible when a group of animals collaborate in looking out for a predator and then alert one another. And a large group may well spot a predator much more quickly than any individual, or even a small group. Although some studies have produced ambiguous results, many more have been supportive. One example, an experiment involving groups of starlings (Sturnus vulgaris) and a model of a hawk, found that groups were much more effective than individual birds in spotting the model (Powell 1974). Another experiment, by Anders Møller (1987), utilized a stuffed owl as a surrogate predator to test the responses of different-sized colonies of swallows (Hirundo rustica). The pseudo-owl was detected much more quickly by the larger colonies.
Another common form of synergy in animal aggregations is a reduction in the time and energy that each individual must invest in maintaining vigilance. For example, Murton (1968) documented that wood pigeons (Columba palumbus) collect food at a slower rate when they forage alone than they do when searching for food in a flock; more of their time is spent scanning for predators. Powell (1974), in his starling study, also found that the individual birds spent less time in surveillance activities. And Pöysä (1994) determined that investments in vigilance by the members of a large population of teals (Annas crecca) in Finland decreased both with increasing group sizes and with proximity to neighboring groups. Likewise, in prairie dogs (Cynomys ludovicianus), the overall level of anti-predator protection increases with the size of the group, yet the amount of time each individual member needs to spend looking out for predators decreases (Hoogland 1979). Indeed, a reduction in individual vigilance as a function of increasing group size is one of the most frequently reported phenomena in the research literature on animal social behavior (see, for example, the discussions in Pulliam 1973; Barnard and Thompson 1985; Brown 1987; Elgar 1989; Lima 1995; Roberts 1996.)
Risk sharing is another one of the synergistic benefits of flocking and joint nesting. It is often referred to as the “selfish herd” effect. By merging into a large aggregation, each individual organism may reduce its personal odds of being a victim. Still another important synergy of scale in some flocking birds, observed by Tinbergen (1951) and supported by Mohr (1960), involves what could be called a deterrent effect, a result of an increase in the risk of injury to a predator. For instance, a tight formation of starlings (Sturnus vulgaris) presents a dangerous situation for a falcon, which customarily swoops down to attack its prey at great speed. The high probability of a collision with other birds in the flock provides an effective shield. Accordingly, falcons generally try to pick off strays. Still another synergy of scale, less common but nevertheless theoretically significant, involves the ability of some animals, for instance bombardier beetles, “stink bugs” and butterflies, to collectively enhance various noxious chemical defenses (aposematism) against predators (Eisner 1970). In a similar vein, Japanese honeybees (Apis cerana japonica) defend against raiding parties of giant hornets (Vespa mandarinia japonica) by surrounding each invader with a ball of some 500 or more bees and using their combined body heat to bake them to death (Ono et al., 1995).
One of the most unusual examples of a synergy of scale, however, is found in a primitive order of aquatic green algae called Volvocales. These organisms, which form integrated “colonies,” have been popular with students of biology since the 19th century because the great range of Volvocale species seem to mirror some of the earliest steps in the evolutionary trend toward complex multicellular organisms. The smallest of these species (Gonium) have only a handful of cells arranged in a disk. However, the Volvox that give the Volvocale line its name consist of up to 60,000 cells in the shape of a hollow sphere that is easily visible to the naked eye. Among the many functional advantages that have been documented in a detailed study of the Volvox by Graham Bell (1985), one of his most important findings concerned the fact that Volvox colonies enjoy a much higher survival rate than smaller colonies. It happens that these planktonic algae are subject to predation from filter feeders like the ubiquitous copepods, but there is an upper limit to the prey size that these predators can consume. Large, integrated, multi-cellular colonies like Volvox are virtually “immune” to filter feeders.
Needless to say, the many “active” forms of defense against predators in nature are also rife with synergy. A classic example is the defensive reaction of newly hatched owlfly larvae (Ascaloptynx furciger), which typically cluster together for self-protection. When threatened by an insect predator, the larvae react by snapping their impressive mandibles in a concerted way. In a comparative study, Henry (1972) showed that when an owlfly larva was isolated and acted alone it could easily be overwhelmed by a predator, but when the larvae were crowded together their united efforts served as an effective deterrent.
Defense against predators also seems to have been a primary incentive for the evolution of eusocial bees, and many studies have documented the role of “guard” bees against threats of various kinds, including human poachers (E.O. Wilson 1971). Indeed, many social insects, such as army ants (Eciton burchelli), deploy morphologically-specialized “soldier” castes, typically the largest members of the colony. Similarly, in the eusocial naked mole-rat colonies, the vital and dangerous role of defense is allocated to the largest workers in the colony, who respond to intruders like predatory snakes by trying to kill them, or by sealing off the colony’s underground tunnel system (Sherman et al., 1991, 1992).
Observations that herbivores also engage in collective defense go back to Darwin. In The Descent of Man (1874), Darwin noted that: “Social animals mutually defend each other. Bull bisons in North America, when there is danger, drive the cows into the middle of the herd, while they defend the outside” (p. 119). Similar defensive formations have been documented in musk oxen (Ovibus moschatus), the eland (Taurotragus oryx) and the water buffalo (Bubalus bubalus), among others (Tener 1954; Kruuk 1972; Eisenberg and Lockhart 1972). However, in African elephants (Loxodonta africana) it is the females with calves that do the job of defending the group (Douglas-Hamilton 1972). The synergies of scale include the advantage of having many eyes (and many ears and noses for that matter), as well as the deterrent effect of compelling a predator to face the massed defenses of several animals at once.
It is not surprising, therefore, that similar patterns of collective defense are found also among cetaceans like the group-living killer whales (Orcinus orca), the so-called “wolves of the sea” (Brown and Norris 1956; Martinez and Klinghammer 1970; Kirkevold and Lockard 1986; Ford et al., 2000; Mann et al., 2000; Rendell and Whitehead 2001). Similarly, bottlenose dolphins (Tursiops sp.) join forces to defend themselves against predators like sharks (Connor et al. 1992). Organized collective defense is also the rule for primates that occupy more dangerous terrestrial habitats (a paradigm that is also highly relevant to human evolution, as we shall see). This was first observed in howler monkeys (Alouatta villosa) in the 1930s by C.R. Carpenter and has since been studied by many researchers in various species. (In baboons, for example, see Chance 1955; DeVore 1963; Crook and Aldrich-Blake 1968; Kummer 1968; Strum 1975a,b, 1987).
By far the most impressive form of collective anti-predator behavior in nature, however, is commonly referred to as “mobbing.” Mobbing involves nature’s equivalent of the old military dictum that the best defense is a good offense. It is an especially common strategy in birds, which are often burdened with the disadvantage of being much smaller than their adversaries yet must stand and defend a fixed nest site or territory. Many species of birds engage in mobbing, certain hummingbirds, vireos, sparrows, jays, thrushes, warblers, blackbirds, ravens, finches, towhees and others (S.A. Altmann 1956; E.O. Wilson 1975). Some of the advantages were documented in a controlled study by Hoogland and Sherman (1976) involving 54 colonies of bank swallows (Riparia riparia) varying in size from 2 to 451 members. The study showed that larger colonies detected predators much more quickly, produced more vocal commotion, mobbed in greater numbers and were generally more effective in deterring a predator.
Predators are often at some risk against mobbers, and sometimes the consequences may be lethal. For instance, Gersdorf (1966) observed massive attacks by starlings against sparrow hawks which occasionally resulted in the intruder’s death. Similarly, Altmann (1956) observed how the Anna Hummingbirds (Calypte anna) often direct their counter attacks to the predator’s eyes. Wolf researcher David Mech has reported various cases of wolves being struck, and even killed, by their prey (Mech 1984; Mech and Nelson 1990). And Cowlishaw (1994) reports that, in four of eleven observations, male baboons killed attacking leopards while only one defender was killed.
The benefits associated with mobbing behaviors are for the most part overwhelmingly favorable, needless to say. For instance, Møller (1987) determined that successful nest predation in large colonies of swallows averaged only 1.2% and that potential predators were almost always deterred. And, in a study of twelve species of forest birds (all of which were subject to owl predation but only half of which were mobbers), Chris Pavey and Anita Smyth (1998) found that the six species that did not engage in mobbing were 8.75 times more likely to be attacked.
Of course, there is always some risk associated with being a participant in a mob, but some clever species have invented ways to minimize the risk. Sandra Pius and Paul Leberg (1998) documented that black skimmers (Rhynchops niger) gain an anti-predator advantage by nesting jointly with gull-billed terns (Sterna nilotica), which respond aggressively both to stuffed model predators and various real threats while the skimmers remain by-standers. In effect, the skimmers are free riders; the terns unwittingly serve as the skimmers’ “protectors.” Similarly, azure-winged magpies (Cyanopica cyana) breed near the nest-sites (and in synchrony) with Japanese lesser sparrowhawks (Accipiter gularis), which aggressively defend their nests against predators. The magpies benefit by having greatly reduced predation on their own nests (Ueta 2001). (Other examples are cited in Barnard and Thompson 1985 and Morris 1990.)
The synergies of scale associated with collective defense in birds are most often used against large predatory birds like hawks and owls, but they may also be deployed against other predators like snakes (Francis et al., 1989) and even against animals that are several times larger than any of the defenders. Acorn woodpeckers, for instance, are cooperatively breeding birds that also collaborate in stocking “communal granaries” with acorns (usually storage holes drilled into the trunk of a large tree), which they utilize as their joint winter food supply. (Because they also live in a temperate winter climate in Northern California, they do not migrate.) Given the vital importance of these food caches to the community, when the alarm call sounds the birds will rush to defend them with massed attacks against all comers, including strange birds, squirrels and other animals (Koenig and Mumme 1987; Koenig et al., 1995). No individual acorn woodpecker could build or stock one of these granaries alone, much less defend it. So the synergies associated with this behavior pattern are manifold.
Mobbing behaviors are also deployed by many terrestrial animals. Small primates, like marmosets (Callithrix aurita), will join forces to mob raptors and snakes (Corrêa and Coutinho 1997). Herds of axis deer (Axis axis) occasionally chase tigers and leopards, while barking loudly at them (Schaller 1967). Agoutis (Dasyprocta punctata) will mob snakes and other immobile predators (Smythe 1970). A troop of baboon males may mount a frenzied attack against potential predator, deploying their large canine teeth as the functional equivalent of weapons (DeVore 1972). And Mongolian gerbils (Meriones unguiculatus), like the acorn woodpeckers, amass collective food caches which they stoutly defend (Ågren et al., 1989).
Chimpanzees have also been observed engaging in mobbing behaviors on numerous occasions (Goodall 1986). Wrangham and Peterson (1996) note that chimpanzees often use their fists in violent confrontations and routinely use “clubs” and throw stones at their opponents. More to the point, Kortland and Kooij (1963) demonstrated (and captured on film), in a celebrated experiment with a troop located in Guinea, that chimps may also use “weapons” to thwart potential predators. When a stuffed leopard was dragged out from a hiding place near a troop of chimpanzees, the members began to hoot and bark, stamped the ground loudly with their hands and feet, charged the leopard waving branches broken off from saplings and threw branches in the leopard’s direction. Finally, the chimps warily approached the compliant leopard and used their “weapons” to pound on its back.
More recently, Christophe Boesch (1991) reported an attack on a trapped live leopard by a group of chimpanzees armed with “clubs” in the Taï National Park. By human standards, to be sure, these are only crude examples of animal-tool symbiosis in violent confrontations, but they are theoretically important. They challenge the conceit that warfare in humankind is a categorically different phenomenon from animal violence because we alone use weapons; it also supports the contention that human warfare has very deep roots in our evolutionary history (as we shall see below).
Surely one of the most awe-inspiring examples of collective anti-predation in nature is the full-fledged defensive charge of a group of African elephants (Loxodonta africana). Iain Douglas-Hamilton (1972), who conducted a multi-year study of elephant behavior in Tanzania, provided a vivid description of one memorable incident when, with no warning, a group of females with their calves in tow attacked his Landrover and systematically demolished it. Apparently the animals associated the vehicle with one of their most dangerous predators, humankind.
Though the attack of an angry elephant may be awesome in terms of sheer numbers, the defensive mobbing reactions of social insects, ants, bees, termites, etc., are among the most impressive social behaviors in all of nature. Tens of thousands of insect warriors can be recruited by the appropriate chemical and/or auditory signals to mount formidable and sometimes lethal “sting operations” against much larger predators (E.O. Wilson 1971; Hölldobler and Wilson 1990). Indeed, so effective are the defenses of eusocial insects like the army ants that they are classified as “top carnivores;” no other species regularly preys on these formidable predators. It is a powerful example of how the “balance of power” in nature is often the result of a synergy of scale. (We will return to this venerable concept in relation to human evolution.)
Predation in groups is another well-documented form of collective violence in nature, and it is equally dependent on synergies, especially various combinations of labor, synergies of scale, information sharing and cost/risk reductions. One of the more significant examples, theoretically, involves the predatory bacterium Myxococcus xanthus. Large colonies of these primitive, one-celled creatures are collectively able to engulf prey items that are much larger than themselves and are then able to secrete digestive enzymes in quantities (and concentrations) that would otherwise be dissipated ineffectively in the surrounding medium (Allee 1931).
Social insects are, of course, among the most impressive predators in nature, and a celebrated example is the foraging system of the South American army ants. In a single day, a raiding party of up to 200,000 workers, armed with potent stingers and fanning out from their nest in a dense, tree-like foraging pattern, might reap some 30,000 prey items, many of which are then split up and hauled back to the nest for all to share. Because they forage en masse, army ants can also collectively subdue much larger prey than would otherwise be possible, even lizards, snakes and nestling birds (E.O.Wilson 1975; Hölldobler and Wilson 1990).
In addition to this impressive synergy of scale, army ant foraging parties also utilize a division of labor. The task of hauling much of the booty back to the nest is assigned to a specialized caste of large workers called “sub-majors” (or porters) that team up to carry prey items which, if split into pieces, would be more than each individual ant could carry alone. Moreover, ecologist Nigel Franks (1989, 1991) has shown that the ability of army ant colonies to pool information and engage in a form of self-organized collective problem solving represents another synergistic property of the colony as a whole.
Some species of fish also prey together. The wrasse are a tropical reef fish that feeds on the abundant supply of eggs produced by the much larger Sergeant-major damselfish. A study by Susan Foster (reported in Dugatkin 1999, pp. 118-20 and mentioned earlier) showed that the female damselfish aggressively defend their nests. Thus no single wrasse, nor even a small group, could overwhelm the damselfish’s defenses. However, very large groups of wrasse can do so collectively. Since success in raiding a damselfish’s nest can only be achieved by a large group acting in concert, it is (to repeat) an unambiguous synergy of scale.
Birds sometimes engage in collective predation as well. Perhaps the best documented example is the common raven (Corvus corax), which effectively utilizes coordinated strategies for hunting small prey and for stealing prey items (or scavenging leftovers) from other carnivores, such as eagles, hawks, wolves and humans. Behavioral ecologist Bernd Heinrich’s recent book, Mind of the Raven: Investigations and Adventures with Wolf-Birds (1999), provides a richly-documented account of this highly intelligent, highly cooperative species. The book contains many anecdotes collected from a large number of credible observers, as well as personal observations and a number of experimental studies by Heinrich himself. His account conveys in a compelling way the cleverness and versatility of these remarkable birds, as well as their manifold uses of collective violence. Here is a brief sample:
Could pairing result in cooperative hunting partnerships? Economics provides a rationale for the raven’s gregariousness….Wildlife film maker Jeff Turner photographed ravens…at Cape Pierce in Alaska, in the spring of 1997, and described seeing a raven dive like a hawk and hit a kittiwake in the air…I saw ravens plucking fresh kittiwakes every day. Usually only one raven dove and hit a kittiwake again and again, eventually forcing it to the ground when it and others jumped on it and killed it. The ravens often worked in pairs. We also saw them go up to kittiwakes on the nest, grab a wing and yank them off, then they or a partner rushed in and took eggs. John R. Moran, an ornithologist, saw groups of ravens at the same time attacking and killing large gulls and geese.
Hunting behavior of ravens could be more efficient when two or more work together rather than alone, and numerous anecdotes indicate that ravens indeed hunt effectively in teams. Take squirrel hunting, for instance. A squirrel on a tree can easily escape almost any pursuer either by running to another side of the tree, or going up or down the tree…If a mobile hunting partner could position itself in the path of an escaping squirrel, the retreating animal could be cut off from its escape route. Gary Keene…saw something resembling this behavior: one raven chasing a gray squirrel across a road while another awaited on the other side.
The most commonly observed raven teamwork has been of ravens taking prey away from predators. Wildlife biologist George Schaller told me of watching raven pairs in Mongolia cooperate in snatching rats from feeding raptors. Similarly, in Yellowstone Park, Ray Paunovich reported seeing a red-tailed hawk with a ground squirrel. Two ravens approached. One distracted the hawk from the front while the other handily snatched the squirrel from behind. Carsten Hinnerichs saw the same maneuver repeated three times in a row in a field near Brück, Germany, where a fox was catching field mice. Terry McEneaney, Yellowstone Park ornithologist, observed two ravens circling an osprey nest where the female osprey was incubating. One raven landed on the nest rim and took a fish, then while the osprey was distracted by this thief, the other raven swooped down and stole an osprey egg….
Wisconsin ornithologist and Arctic explorer Ludwig Kumlien saw raven pairs succeed in killing young seals that lay basking near their ice holes. One raven would at first leisurely circle over the seal, then drop down beside the seal’s escape hole in the ice. The raven’s partner then drove the seal to the hole, where the first raven killed it by pecking it on the head (Heinrich 1999, pp. 132-33).
One of Heinrich’s conclusions is directly relevant to the thesis of this chapter. “According to some views, the birds are not cooperating if they seek only to satisfy their own chances of success without a conscious regard for the other. From an ecological-evolutionary perspective, however, the effect is critical and intent is irrelevant, the latter being another question altogether. Cooperation might occur whether it is intended or not. What matters for practical cooperation is payoff” (p. 135). In other words, what matters are the synergies.
In response to the possible objections of sociobiologists, or game theorists, who might question whether or not raven partnerships are stable (i.e., mutually beneficial) or merely exploitative, Heinrich responds: “To that question we do have an answer, and that answer is derived from observations in the aviary. First, I found that the raven partnerships can indeed last years, and as previously supposed, mate pairing appears to be almost irreversible. Furthermore, mutual tolerances among partners ensure that food secured or held by one is also available to the other; food is not defended against partners. Therefore, from the standpoint of evolutionary ecology [raven] pairs cooperate…” (p. 135).
Though hunting cooperatively is a limited phenomenon in nature, overall, it is more widespread, even legendary, in a number of mammalian species, including:
Hyenas (Crocuta crocuta) B Kruuk 1972; Frank 1986; Mills 1990; Hofer and East 1993; Roes 1994;
Lions (Panthera leo) B Schaller 1972; Packer et al. 1990a,b; Stander 1992; Grinnell et al., 1995);
Wild dogs (Lycaon pictus) B Bourlière 1963; Estes and Goddard 1967; Fanshawe and Fitzgibbon 1993;
Wolves (Canus lupus) B Murie 1944; Mech 1981, 1999, 2000; Mech and Adams 1999; Savage 1996; Hampton 1997;
Baboons (Papio) B Kummer 1968; 1971; Strum 1975a,b; 1981;1987;
Killer whales (Orcinus orca) B Brown and Norris 1956; Hancock 1965; Martinez and Klinghammer 1970; Baldridge 1972; Kirkevold and Lockard 1986; Bigg et al., 1987; Hoelzel 1991; Pyle et al., 1999; Guinet et al., 2000, Ford et al., 2000; Mann et al., 2000; Pitman et al., 2001; Rendell and Whitehead 2001;
Capuchin monkeys (Cebus capucinus) B Fedigan 1990; Robinson 1986; Miller 1999;
Chimpanzees (Pan troglodytes) B Kawabe 1966; Teleki 1973; Nishida et al., 1983; Ghiglieri 1984; Goodall 1986; Smuts et al., 1987; Boesch and Boesch 1989; Boesch 1994; Wrangham and Peterson 1996; Stanford 1996, 1999; Basabose and Yamagiwa 1997 (among others).
Chimpanzee collective hunting behaviors are especially well documented. Although the pattern and frequency of group predation varies significantly from one community to another, in general it can be said that chimpanzees are regular, proficient hunters. Over the years there have been various interpretations of this behavior pattern. In some cases (at Gombe, for instance), group hunting may be analogous to team sports in humans; the nutritional value may not be very great, and the pattern of meat sharing could be more nepotistic and Apolitical@ (Nishida et al., 1992; Stanford 1999; Mitani and Watts 2001). Nevertheless, given the extremely close phylogenetic relationship between chimpanzees and humans, the discovery that chimpanzees are systematic predators has great theoretical significance in relation to our understanding of human evolution (again, see below).
While social mammals provide the best-known examples of the synergies associated with collaborative predation, there are also many striking examples in other species. For instance, there are the groups of 15 to 20 orb-web spiders (Metabus gravidus) that combine their efforts to build a large web that can span a stream where their prey are especially abundant (Maynard Smith 1982a). There are the black headed gulls (Larus ridibundas) that will recruit collaborators whenever they spot fish in the water because they fare much better at catching them when they work in groups (Götmark et al., 1986). Also, there are the social spiders (Anelosimus studiosis) that cooperate in capturing (relatively) large prey, like crickets, and then will share the booty (Furey 1998). But perhaps the most remarkable example is the symbiotic partnership between “honeyguide” birds and various partners, including humans (as noted earlier).
What are the advantages of collective predation? A number of early studies suggested that larger groups of hunters were unequivocally better off. Schaller (1972), for example, concluded that lioness hunters were twice as successful when hunting together. He also observed that larger groups could capture larger prey, like giraffes and buffalo. Schaller’s measure of “success” referred to “efficiency” (captures per chase) and the number of multiple kills. Bourlière (1963) and Estes and Goddard (1967) drew similar conclusions about wild dogs. They noted that it would be impossible for wild dogs to capture larger, faster prey like adult wildebeest and zebras without a coordinated effort. Kruuk (1972:185) observed a total of 34 hyena hunting episodes and, again, concluded that larger groups were more successful. Various researchers have drawn similar conclusions about group hunting in wolves (see especially Murie 1944; Mech 1970, 1988, 1998, 1999, 2000; Hampton 1997). Palomares and Caro (1999), in a broad review of interspecies killing in mammalian carnivores encompassing 27 “killer” species and 54 “victim” species, concluded that, as a rule, group-hunting species kill larger prey than do individual hunters.
However, some other studies and analyses of hunting behaviors have suggested that the advantages associated with cooperative predation are always contingent. For instance, Craig Packer and Lore Ruttan (1988) systematically re-analyzed the data on hunting behaviors from 28 studies encompassing 60 different species and concluded that cooperative hunting was only advantageous when the quantity of meat acquired per capita exceeded what each individual animal could obtain on its own. Accordingly, factors such as the prey size, prey distribution, relative abundance and relative ease of capture all may enter into the calculus of whether or not the synergies, net of the costs, will be sufficiently “rewarding” for the participants. A study by Pasquet and Krafft (1992) in the social spider (Anelosimus eximus) was also consistent with this hypothesis. These two researchers found that the propensity for cooperation in capturing prey among the occupants of a communal web depended upon the prey type (small or large). Of course, the spiders also benefitted from having to exert less web production effort per spider, plus greater overall capture efficiency and the ability to capture and attack much larger prey (e.g., moths versus grasshoppers).
By the same token, Christophe Boesch (1994) points out that the manner in which the spoils are distributed and the level of policing of free-riders may also be crucially important. If participants are routinely disadvantaged, or if non-participants are routinely rewarded, the system may not be sustainable. Thus, the chimpanzees at the Taï National Park research site were found to hunt in groups 95% of the time, whereas chimpanzees at the Gombe site do so only 12% of the time. Cooperative hunting at Gombe is unstable, partly because there is a high success-rate for individual hunters, but also because non-hunters are not rigorously excluded from sharing the meat. Nevertheless, as Craig Stanford (1999:68) observes, as a rule “the larger the hunting party, the greater their odds of making a kill.”
Another aspect of group hunting in nature that has come into better focus in recent years is the fact that success often depends not on sheer numbers alone but on synergies that may derive from a tactical division/combination of labor. For instance, caribou and moose can mount formidable defenses, sometimes even deterring mass attacks by wolves (Crisler 1956). And Nelson and Mech (1994) published an account of a how single white-tailed deer (Odocoileus virginianus), which normally will flee a predator, achieved a stand-off against three attacking wolves. Thus, coordinated group hunting is normally an essential strategy for wolf packs. Both Murie (1944) and Crisler (1956) observed wolves systematically driving caribou toward other wolves that were waiting in hiding, just as the raven partners do (see above). And the vivid description by Mech and Adams (1999) of the killing of a Muskox (Ovibos moschatus) by a pair of wolves on Ellesmere Island, Canada, illustrates how crucial a team effort can be.
Likewise, a solitary hyena would be unlikely to be able to separate a wildebeest mother from its calf, but when hyenas hunt in pairs they can coordinate their efforts. One hyena may distract the mother while the other steals the calf (Packer and Ruttan 1988). Paul Stander (1992) identified a similar combination of labor in the 486 group-hunting episodes that he observed in a pride of lions in Namibia. Stander identified four stalking categories, which he called wings, centres, drivers and catchers, and noted that individual lions seemed to have preferred stalking roles. Christophe Boesch (1989; 1994) made similar observations among the chimpanzees at Taï National Park; some hunters serve as drivers, chasing a colobus monkey, say, through the trees toward other hunters who are lying in wait.
Finally, it has been found that factors other than hunting efficiencies/synergies may also affect the size of predator groups. These include such things as the distribution and quantities of other types of food resources that are available, the size of competitor groups, the challenge of having to defend an exclusive territory, and their own susceptibility (especially their infants) to attacks from other predators (Packer et al., 1990a; Wrangham et al., 1993). For instance, in the wild dog groups studied by Fanshawe and Fitzgibbon (1993), larger groups were not only more successful in capturing prey but were better able to defend their kills against preemption by spotted hyenas. Likewise, larger lion prides seem better able to protect their cubs against predators like leopards and nomadic lions and are more successful at defending exclusive territories (Schaller 1972; Packer et al., 1990b; Stander 1992). Just as there are numerous factors that may enter into defining the problem of survival and reproduction for any given species, so multiple forms of synergy may play a role in the adaptive responses to those problems.
Collective Violence Within and Between Groups
As noted earlier, animals not only prey upon (and defend against) one another, they also compete for various “resources”, food, water, nesting or sleeping sites, territories, allies and, of course, mating privileges. “Conflicts of interest” over needed resources are endemic in nature. But much of this competition goes on at the individual level, and there are various strategies short of violence for resolving these conflicts. One obvious example is out-migration when one competitor opts to re-locate rather than stand and fight.
For this reason, conflicts of interest may be “movers” but they are not sufficient to explain collective violence; to repeat, the “difference that makes a difference” is the collective potential for utilizing functional synergy in meeting these challenges; synergy provides the “bioeconomic” benefits that may make cooperative behaviors of all kinds, including violence, a desirable strategy. It is functional synergy that drives the evolution of cooperation, not the other way around. Applying this perspective to violent competition between groups, I will distinguish between (1) “internal” conflicts involving coalitions, alliances, or factions within a group or community and (2) “external” clashes that occur between different groups or communities, including those of other species.
Regarding the first category, Harcourt and de Waal, in the preface to their edited volume on Coalitions and Alliances in Humans and Other Animals (1992:v), note that: “The well-known human proclivity to form coalitions and alliances undoubtedly has a long evolutionary history.” Alliances of males to acquire or hold mates, for instance, is a common phenomenon in nature. It is particularly well-documented in our closest relatives, the chimpanzees (Harcourt and de Waal 1992; Wrangham and Peterson 1996).
Similarly, male lions often form alliances to “take possession” of, and defend, a group of female lions, though they may subsequently quarrel among themselves over individual mating privileges (Bygott et al., 1979 ; Packer and Pusey 1982; Packer et al., 1988; Grinnell et al., 1995). A pattern of mate-seeking and mate-guarding coalition behavior has also been observed in diverse baboon troops (See especially DeVore 1965; Packer 1977; Strum 1987; Noë 1986, 1992, 1994; Bercovitch 1988; Noë and Sluijter 1990; Harcourt and de Waal 1992), as well in male dolphins, where an unusual two-level pattern of alliances and “super-alliances” has been documented by Richard Connor and his colleagues (Connor et al., 1992, 1996, 1999).
Coalitions may also arise to resist or replace the dominant animal in a group (see especially de Waal 1982; Goodall, 1986; Harcourt and de Waal 1992). Ronald Noë (1992) found that coalitions against the dominant animal in baboon troops are most likely to occur, not surprisingly, among mid-ranking males that use their numbers (a synergy of scale) to offset the superior fighting ability of the alpha male. However, males do not by any means hold a monopoly over the synergies that can be derived from cooperation. In bonobos, for instance, coalitions of females form partnerships to impose a degree of control over the behavior of the much stronger males, including even their access to resources (de Waal 1996; Stanford 1999). (Indeed, bonobo groups may in fact be organized around female coalitions.) Likewise, lionesses may join forces to resist the tyranny of an abusive male (Schaller 1972). And Bernd Heinrich (1999) reports that coalitions of immature ravens frequently gang up against a possessive adult bird to “steal” a portion of a prey item for themselves (see also Marzluff and Heinrich 1991). As noted earlier, the synergies produced by coordinated individual actions can serve as an “equalizer”, in accordance with the age-old (even biblical) adage about “strength in numbers.”
Violent conflicts between groups, and even sometimes between species, provide the closest approximation in nature to human warfare, and, as suggested earlier, the resemblance is not simply a facile analogy. Conflicts of interest over the requisites for survival and reproduction often pit organized groups against one another, or precipitate collective action, including collective violence, as an instrumentality for securing and/or protecting needed resources of various kinds, meat on the hoof, foraging territories, nesting sites, water holes, mates, and more. But if conflicts of interest generate the motives for collective violence, it is the synergies that makes it a rewarding activity, an “adaptive” behavioral response.
Thus, packs of African wild dogs (Lycaon pictus) and spotted hyenas (Crocuta crocuta) are frequently in contention over the prey that are brought down by one or the other species (Estes and Goddard 1967). Male dolphins (Tursiops sp.) in Shark Bay, Australia, form coalitions to sequester and protect females and their calves from the depredations of other male alliances (Connor et al., 1992). Coalitions of male lions (Panthera leo) on the African Serengeti defend females and their cubs both against other groups of males and other predator species (Schaller 1972; Packer and Pusey 1982; Packer et al., 1990a). “Gangs” of elk (Cervus canadensis nelsoni) approaching a salt lick can drive away various competitors: porcupines, mule deer, even moose (M. Altmann 1956). The desert seed harvester ants (Messor pergandei) that form colonies with multiple queens are generally able to gain a significant advantage in raids against smaller colonies with single queens (Rissing and Pollack 1991). Rhesus monkeys (Macaca mulatta) in Northern India aggressively defend their home ranges against other rhesus groups (Lindburg 1971).
Groups of female southern sea lions (Otaria byronia) at Península Valdés, Argentina combine forces to protect their pups against harassment and infanticide from marauding groups of males (Campagna et al., 1992). Large groups of female vervet monkeys (Cercopithecus aethiops) living in the Amboseli National Park, Kenya, are able to acquire and defend larger home ranges than do smaller competing groups (Cheney and Seyfarth 1987). Established groups of rock pipits (Anthus petrosus) in Sweden form alliances with neighboring groups to jointly defend their territories against intruders (Elfström 1997). Groups of Australian magpies (Gymnorhina tibicen) in two different populations (one in New Zealand and the other in Queensland) were observed forming into groups of varying sizes, and it was found that their effectiveness in defending their territories increased significantly with group size (Farabaugh et al., 1992).
Spotted hyenas sometimes defend a permanent territory (Kruuk 1972) and sometimes defend a more fluid feeding range that reflects the movements of migrating prey species (Hofer and East 1993). In either case, though, the hyenas aggressively defend their resources from competitors of all kinds. Hostile interactions are also common among groups of primates, such as white-faced capuchins (Cebus capucinus) (Perry 1996). Though previously in doubt, it is now also clear that groups of male chimpanzees and bonobos consistently defend home ranges against other groups, as well as mounting raids against other groups (often for females) and even sometimes engaging in lethal group attacks, “warfare”, against other groups (Goodall 1986; Manson and Wrangham 1991; Harcourt and de Waal 1992; Stanford 1998, 1999). For an analysis of the factors that influence the “recruitment” of participants in chimpanzee group conflicts, see M.L. Wilson et al., 2001.
In sum, collective violence in nature is a ubiquitous phenomenon. It is found on every continent; it has been independently “invented” by widely diverse species on innumerable occasions; it is associated with an array of specific objectives that relate to the problems of living; and it is very often highly effective, that is, the benefits clearly outweigh the costs and the risks. Otherwise these behaviors would have been subject to adverse selection and would be rare or non-existent, or would not have evolved in the first place. To reiterate, the option of acting alone is the theoretical “ground-state,” so cooperative behaviors require an adaptive explanation. The unifying “explanandum” proposed here is synergy. There are many different motivations, “goals” and conditioning factors associated with collective violence, but the expectation/realization of positive synergy is a common denominator. With this evolutionary-adaptive framework providing a theoretical context, we turn now to the issue of human evolution.
Collective Violence and the Evolution of Humankind
No research domain, it seems, has been more conflicted with competing theories, and strong emotions, than the evolution of the human species. This is understandable. The scientific debate over our “origin story” inevitably affects our collective self-image. But more important, the argument hinges on a relatively small and painstakingly accumulated body of indirect evidence about events, processes and environments that existed many millions of years ago. Indeed, our assumptions are regularly being upended by new fossil finds.
Accordingly, any scenario for human evolution must be very tentative. However, as I hope will shortly become apparent, the scenario proposed here is not really new. Rather, it attempts to bring some existing scenarios and hypotheses into better focus, as well as making explicit some implicitly recognized aspects of the process and uniting them under a common causal explanation, namely, the Synergism Hypothesis. And, as we shall see, collective violence most likely played an important part. Although this scenario rests on certain “plausibility” arguments (and the judicious use of Occam’s razor), I believe it has merit. Since the history of this debate is recounted in more detail in Corning (2003), I will provide only an abbreviated discussion of this scenario here.
The starting point and touchstone for theorizing about human evolution is Darwin’s The Descent of Man (1874). People who find fault with Darwin’s speculations tend to forget that he was compelled to reason about our origins without the benefit of a single early hominid fossil. Given the paucity of evidence, Darwin wisely demurred from attempting to reconstruct a precise historical narrative. He concerned himself with the functional dynamics and the overall logic of the process. In Darwin’s view, the crucial first step in human evolution was the shift from the trees to a terrestrial life-style by our remote primate ancestors, though he could only speculate about why these creatures had abandoned the relative safety of an arboreal existence. Perhaps, he theorized, it was “owing to a change in [their] manner of procuring subsistence, or to some change in the surrounding conditions” (quoted in Wolpoff 1999a:187).
In any case, Darwin reasoned that this momentous change of venues by our ancient ancestors encouraged the development of bipedal locomotion, which in turn liberated the forelimbs and hands for other uses. In time, this important innovation allowed for the invention of tools and weapons, a sequence rather like the development of wings and flight skills in ancient, bipedal (theropod) dinosaurs. Tools/weapons in turn were able to offset our ancestors’ obvious anatomical disadvantages (notably their reduced canine teeth) and facilitated the adoption of hunting as a major subsistence strategy. Darwin also stressed the role of social cooperation, reciprocity and “mutual aid” in human evolution, especially in food-getting but also in conflicts with other groups and other species.
In modern terminology, Darwin proposed that natural selection operated at three “levels”, between individuals, between “families” of close kin, and between social groups. Indeed, Darwin believed that competition, including warfare, between various “tribes” played a major role in shaping the course of human evolution (as noted earlier). “Natural selection, arising from the competition of tribe with tribe…would, under favourable conditions, have sufficed to raise man to his high position.” The tribes that were the most highly endowed with intelligence, courage, discipline, sympathy and “fidelity” would have had a competitive advantage, he argued. In an oblique allusion to the role of synergy, Darwin observed that: “Selfish and contentious people will not cohere, and without coherence nothing can be effected. A tribe rich in the above qualities would spread and be victorious over other tribes; but in the course of time it would, judging from all past history, be in its turn overcome by some other tribe still more highly endowed. Thus the social and moral qualities would slowly tend to advance and be diffused throughout the world” (1874 I: 97,199-200). Finally, Darwin viewed increased intelligence and the development of language in humankind as later outgrowths of tool use and social organization. As we shall see, Darwin’s scenario still has much to recommend it; the factors that he identified are still in play among contemporary theorists.
In contrast with Darwin’s multi-step scenario, various “prime mover” theories of human evolution have been commonplace over the past century. Among others, there was the “killer ape” scenario (Raymond Dart), “man the hunter” (Sherwood Washburn and Chet Lancaster), “woman the gatherer” (Adrienne Zihlman and Nancy Tanner); the “food sharing hypothesis” (Glynn Isaac) and the “nuclear family” model (Owen Lovejoy). While these single factor explanations may draw attention to important aspects of the process, they have all ultimately proven to be insufficient. (Lately, the “climate-change hypothesis” seems to have become fashionable in some quarters, inspired in part by new evidence that past climate patterns have been much more unstable than we had previously thought was the case.)
However, many theorists nowadays shy away from a monolithic approach. As our understanding of the complexities of human evolution have increased, prime mover theories have come to seem much less satisfactory. Yet the ultimate theoretical question still beckons: How do we explain the evolution of humankind? What accounts for our remarkable evolutionary transformation? Is there some underlying principle involved? Or was it an accident, a meandering “drunkard’s walk” without any particular logic to it? I believe that there was an inner logic to the process of human evolution and that the Synergism Hypothesis has something to contribute. It provides a frame-shift that allows us to view the process from a bioeconomic perspective.
I first proposed in an earlier book (Corning 1983) that there was no prime mover in human evolution. Rather, various forms of synergy with significant “payoffs” for the immediate problems of survival and reproduction represented the common thread, and that the process was propelled by proximate behavioral innovations and choices. The evidence that has accumulated since then has reinforced this theory, I believe. In the truest sense, the evolution of humankind involved an entrepreneurial process, a pattern of behavioral invention, trial-and-error learning, “selective retention” and the subsequent natural selection of supportive anatomical changes. Moreover, much of our inventiveness has involved new forms of synergy. Synergy has played a key role in this transformation; it has generated potential bioeconomic benefits/payoffs of various kinds, synergies of scale, combinations of labor, functional complementarities, cost- and risk-sharing, tool and technology symbioses, and more.
In this scenario, behavioral changes are seen as having been the “pacemakers” of various progressive trends toward greater functional competence and complexity in hominid groups, as Darwin himself suggested. In other words, our hominid ancestors shaped the trajectory of their own biological evolution with their behavioral inventions. To a very considerable degree, our species invented itself though obviously there was no premeditated plan. As the zoologist Jonathan Kingdon put it in the title of his provocative 1993 book, we are the Self-Made Man. (In my own earlier book, I also alluded to archaeologist V. Gordon Childe’s classic 1936 study of the agricultural revolution called Man Makes Himself.) However, this paradigm can also be classified as a “gene-culture co-evolution theory;” behavioral and biological innovations are viewed as having been deeply intertwined (see Cavalli-Sforza and Feldman 1981; Corning 1983, 2003; Boyd and Richerson 1985; Durham 1991; Odling-Smee et al., 2003; Weber and Depew 2003; Richerson and Boyd 2004).
Two other preliminary points are in order here. One is that the assumptions that are made about the context of human evolution are critically important. The external physical and biotic environment, what Darwin referred to as the “surrounding conditions”, were obviously a key factor. But equally important was the “internal environment”, the array of specific biological survival needs that imposed day-by-day imperatives for our ancestors, as well as the biological endowment (the precursors or preadaptations) that conditioned their efforts to meet those needs. In many scenarios, the full context of human evolution is unstated, or downplayed, or used only selectively to develop a theory.
The other preliminary point is that, in speculating about the many unknowns in human evolution, Occam’s razor (or the principle of parsimony) provides an important analytical tool, a way of choosing among various explanations when “hard” evidence is limited or lacking. To be sure, Occam’s razor has a double-edged blade, so one must use it with care. But, all other things being equal, the simplest and most incremental (least radical) alternative is more likely to be the right one. By the same token, the scenario that implies the greatest rewards or payoffs for the least possible cost or risk (cost-effectiveness) is more likely to have been the one that our ancestors selected, by and large. Although it is unlikely that we will ever find conclusive evidence for many aspects of human evolution, we may be able to rank-order the different scenarios in terms of their relative plausibility.
The first and perhaps most fundamental point about the context of human evolution is that our ancestors had to cope with an environment that was highly variable over time (see Vrba et al., 1995; also Kingdon 1993; Foley 1995; Alley 2000a,b; Boyd and Richerson 2001). It is now known that there have been a total of 27 ice ages during the past 3.5 million years alone, and that many of these episodes involved major changes in land forms and local ecosystems in various parts of the world. For instance, during the deep freeze that began about 33,000 years ago and peaked about 20,000 years later, sea levels declined more than 400 feet from present levels as a mile-deep mountain of ice became trapped in the expanding northern glaciers. In fact, we now recognize that significant shifts in local micro-climates, and in the relative abundance of plant and animal species, can occur over a very short period of time (Allen et al., 1999; Alley 2000a,b; Keeling and Whorf 2000; Stanley 2000). Although tropical Africa has long been a fecund and bountiful area, there have also been many booms and busts.
The most important ecological influence on the course of human evolution, it now seems evident, was a major long-term change along the Eastern edge of the subtropical forest belt that spans the mid-section of Africa, in what is now (principally) Tanzania, Kenya and Ethiopia and extending as far west as Chad (Coppens 1994). Here a combination of factors gradually reshaped the landscape. One was a rifting process, due to the movement of the huge tectonic plates that run the length of the continent. The other was a shift in global climate patterns. Over time these influences greatly altered the terrain and the ecology of East Africa. Immense tracts of dense tropical forest areas, safe havens with plentiful resources for arboreal primates, were gradually converted into a more broken, “mosaic” pattern with many “patches” of woodland, an abundance of lakes and streams, large areas of marshland and more open savanna areas. Many of these areas came to be populated with large herds of herbivores (some 20 different species), along with a formidable number of carnivores, sabre toothed cats, lions, leopards, hyenas, cheetahs and more (Foley 1995).
Some theorists believe that these environmental changes were sufficient to account for the origin of the hominid line; they are said to have “induced” or “driven” or “forced” East African apes to shift from an arboreal lifestyle to terrestrial living and to adopt new survival strategies. As one well-known biologist put it, “the causes were ecological.” Biologist Patrick Bateson (1988) calls this the “billiard ball” theory of evolution. For one thing, hard evidence of a close correlation between climate changes and the major benchmarks in human evolution is sparse and contradictory (see especially White 1995). But more important, the deterministic model obscures the active role of the participants themselves. Some of the Eastern primates of that epoch chose to change their “manner of procuring subsistence,” in Darwin’s phrase. Others did not. Either way, these behavioral choices were fateful for the future course of hominid evolution. (In what follows, I will rely on the full-length textbook treatments by paleoanthropologists Bernard Campbell 1985, Richard Klein 1999, and Milford Wolpoff 1999a, with other sources cited below.)
The shift to a terrestrial life style most likely did not happen all at once. For one thing, it involved substantial costs and risks. As foraging ranges expanded, so did the time and energy required to exploit them, and the australopithecines of three to five million years ago were imperfect bipeds, competent but not as efficient as later Homo erectus/Homo ergaster. (I will side-step the debate surrounding various fossil remains and use the term “australopithecines” here generically to include such known predecessors as Ardepithecus ramidus.) More important, the exploitation of a mosaic environment introduced serious new risks from predators and competitor species, not to mention rival proto-hominid groups. Some theories of human evolution have downplayed these threats, but it was in fact a major challenge, with life and death consequences. (See especially Anderson 1986; Cheney et al., 1987; Dunbar 1988; van Schaik 1989; Cowlishaw 1994; Iwamoto et al., 1996; Wrangham 2001.) It is possible for a large primate to survive for many days without food, so long as fresh water is available, but a single encounter with a predator will very likely reduce its survival chances to zero.
There is ample contemporary documentation and some fossil evidence going back to Ardepithecus ramidus some 4.4 million years ago suggesting that our ancestors were indeed subject to “predation pressure.” Evolutionary biologist Robert Foley (1995) points out that there were no less than ten large carnivore species roaming East Africa in those days, compared with just two today. Lee-Thorp et al., (2000) singled out leopards (Felis), sabre-toothed cats (Merantereon) and hyenas (Crocuta) as likely to have been systematic predators on primates and australopithecines. Pack-hunting species like Palhyaena would have been particularly dangerous, but even giant eagles were a threat (see also Brain 1981, 1985; and Isbell 1995).
The proto-hominids that chose to venture into this changing and hazardous environment brought with them many important precursors/preadaptations (some prefer Gould and Vrba’s term “exaptations”), including stereoscopic color vision, a vertical climbing anatomy, dexterous forelimbs with manipulative hands, sociality, a relatively high degree of intelligence, a flexible dietary pattern (a major advantage) and, highly significant, a form of social organization that is based on a nucleus of closely-related males who are joined by unrelated females. Only 6% of the 167 primate species studied to date have male-based groups, and this may have been one of the keys to the emergence of the hominid adaptive pattern (Wrangham 1987; Lee 1994). One other important precondition had to do with the fact that our earliest proto-hominid ancestors were diminutive in stature, less than 3 feet tall. Even the much later Australopithecus afarensis of 3.5 million years ago are estimated to have weighed some 15 percent less on the average than modern chimpanzees. Wolpoff (1999a: 217-219) calls these proto-hominids “Miocene midgets.” In other words, our remote ancestors were extremely vulnerable to predators.
Accordingly, the question is: How did a diminutive ape with constrained mobility on the ground and no natural defensive weapons, but with a relatively large brain (slightly bigger than modern chimpanzees), manipulative hands and an omnivorous digestive system, solve the problem of shifting to a terrestrial habitat, broadening its resource base and, over time, greatly expanding its range? (By three million years ago, australopithecines had spread through much of Africa.) Social organization, group living, was very likely a key factor. In a patchy but abundant environment that was also replete with predators, competitor species and sometimes hostile groups of conspecifics, group foraging and collective defense/offense was the most cost-effective strategy. There were immediate payoffs (synergies) for collective action that did not have to await the plodding pace of natural selection. It is also likely that the earliest of these proto-hominid pioneers stayed close to the safety of the trees. But, as they began to venture further from their safe havens, the risks increased commensurately. (A number of other theorists over the years have endorsed the group-defense model, including George Schaller, Alexander Kortlandt, John Pfeiffer, Richard Alexander, Richard Wrangham and others.)
Note that the group-defense scenario does not assume hunting, male provisioning, monogamy or any of the other more radical innovations that have been proposed in the past. It assumes only that the synergies derived from acting collectively, foraging and reproducing as a group, were both immediate and mutually beneficial; the odds of survival were greatly enhanced. There may very well have been “group selection” between these groups, but it was not based on altruism. It involved jointly produced “collective goods.” Nor did it require “a cooperative gene.” It required only a degree of intelligence about means and ends, and costs and benefits that does not seem unreasonable in light of the findings about chimpanzees and bonobos. Moreover, because these groups were formed around a nucleus of closely-related males, individual selection, kin selection and group selection would have been aligned and mutually reinforcing, just as Darwin had supposed in The Descent of Man.
Why would the males defend the females and infants? For one thing, the males might not have known their paternity if the females followed a reproductive strategy of promiscuous matings and, perhaps, disguised ovulation (as bonobos evidently do). Another factor was that all of the infants would have been closely-related, “nephews,” “cousins,” or even younger siblings. A third point is that, in an extremely “K-selected” species with a very long reproductive cycle (several years, even in chimpanzees) and a relatively short life span, each offspring was relatively more valuable; the benefits of protecting the mothers and infants, and the costs of not doing so, were much greater. Finally, in a tightly organized, interdependent group it was not significantly more costly to defend the offspring of close kin than it was to defend one’s own progeny and oneself; it was not a matter of altruism, or reciprocal altruism but of teamwork in a win-win (or lose-lose) situation, a synergy of scale. An analogue here, as many other theorists have noted, is the organization of savanna baboon troops. (See especially Cowlishaw 1994, and other references cited therein. For a general review of primate social patterns, see Pusey 2001.)
Was there also a division/combination of labor? Contemporary hunter-gatherer societies, not to mention most modern societies, typically have a division of labor along sexual lines, and it is possible that a rudimentary version of this pattern existed also among the early australopithecines. Applying the principle of parsimony, it seems likely that the females would have been primarily responsible for carrying the infants and shepherding the juveniles, while the males served as the primary (though not necessarily exclusive) guardians for the group. Some evidence for this can be seen in the australopithecine fossils of 3.5-3.0 million years ago, where there appears to have been a sharp sexual “dimorphism;” the females remained very small while the males grew much larger, closer in size to modern pygmies. Many theorists attribute such size differences to sexual competition among the males. However, a division of labor between males and females can also create a “selection pressure” for dimorphism. If australopithecine males were close kin, this might have dampened status rivalries and mitigated male sexual competition. On the other hand, if the males came to play an important role in defending the group, larger size would have provided a significant functional advantage. An analogue can be seen in baboons, where the females average about 30 pounds and the males about 90 pounds, a far greater difference than is typically found in primates.
We may never know for certain about this and many other details relating to human evolution, but group living/group foraging and a cooperative division of labor allowing for increased access to a more dangerous but abundant terrestrial environment is likely to have been primordial in the hominid line. It would have involved the most limited, incremental behavioral changes with the most cost-effective payoffs for the participants; it was highly synergistic. Moreover, as time went on the group-living mode of adaptation led to other forms of social cooperation and more elaborate forms of synergy.
One other innovation may also have played a crucially important role in the transition of our ancestors from arboreal to terrestrial apes, namely a synergistic “soft technology” of wood and bone implements, and perhaps thrown objects as well. There have been many tool-use advocates over the years, from Darwin to Dart, Szalay, Washburn, Birdsell, Coursey and Mann (Wolpoff 1999a). However, many other theorists claim that the first “real” tools in hominid evolution were the manufactured stone cutters/choppers/scrapers that first appear in the fossil record some 2.5 million years ago. These theorists often downplay the importance of tools and envision early australopithecines as having been minimal tool-users at best. But this view, as Jonathan Kingdon observes, is obtuse. For one thing, the lack of fossil evidence is not evidence of a lack, as the old cliché goes. But more important, tool-use can have a revolutionary effect. It can be the functional equivalent of opening up a new ecological niche, or a whole new adaptive modality; otherwise unattainable sources of food can suddenly become a reliable, even abundant part of an animal’s diet. Also, it should be stressed that the payoffs are immediate (proximate); they need not await the workings of natural selection. (See also the discussions in Lewin 1993; Kingdon 1993; Odling-Smee et al., 2003; Weber and Depew, 2003.)
It seems unlikely that the australopithecines of three to five million years ago could not have adapted successfully to a terrestrial life-style and survived, even prospered, for perhaps three million years without the acquisition and skilled use of various natural objects, such as digging sticks, hammers, carriers and the like. As we noted, chimpanzees, elephants, capuchin monkeys and many other species are frequent users of tools for procuring food, and sometimes in conflict situations as well. It seems very improbable that the stone tools dating back to about 2.5 million years ago sprang from the mind of a “tool-challenged” predecessor.
By the same token, it seems likely that “weapons” also played an indispensable part in the successful transition to a terrestrial life-style. One can hardly exaggerate the value to a diminutive, relatively slow-moving biped, lacking the baboon’s outsized canine teeth, of being able to use a short stick (similar to the modern billy club) or a large femur, or even a well-aimed rock, as a defensive weapon (as Darwin argued in The Descent of Man). For the very same reason that policemen around the world still use billy clubs and rioters still throw rocks, the wielder of a weighted object can engage an opponent at much less personal risk (beyond arms’ length) and can strike a much more damaging blow. Even crude weapons, if skillfully used, would have tipped the balance of power in many confrontations. Indeed, it has been pointed out that the same hand-held wooden implements can double as digging sticks and defensive weapons.
Here, again, the principle of parsimony can be invoked. Consider the alternatives. A group of cooperating Miocene/Pliocene midgets without the benefit of defensive implements of any sort would have been at a great disadvantage against bigger, faster predators armed with large canines and sharp fore-claws. Conversely, a lone australopithecine, even armed with a club, would have been at a great disadvantage against a group of hungry hyenas. But a group of australopithecines traveling together in dangerous or unfamiliar country with digging tools/weapons carried at the ready would have been far more likely to hold their own in any life-and-death situation. These creatures may not always have been subject to predation, but even one incident in a lifetime would have been one too many.
Though even more speculative, it is also possible that the “arms race” which has been a hallmark of recent human history began (albeit very modestly) in the Miocene/Pliocene. If groups wielding simple “weapons” like digging sticks gained “leverage” in confronting predators, the same would have been true in confronting various competitors. Again, the principle of parsimony can be applied. The manufacture and skilled use of tools/weapons by these proto-hominids very likely entailed learned, socially-transmitted behavioral innovations, and the application of these innovations to new contexts would have been an incremental step with potentially large benefits.
Does this mean that the australopithecines were “killer apes” after all? I would argue that this image is greatly overdrawn. The challenge for our ancestors, as it is for us, was survival and reproduction. The “problems of living” were the driving motivation, and this involved an array of continuing, inescapable “basic needs,” as noted earlier. Violent confrontations with other groups can be very risky, especially when weapons are involved; the potential costs could outweigh any possible benefits. It is more likely that armed conflicts between groups were related to such vitally-important objectives as the acquisition or defense of food patches, water holes and sleeping sites. Rose and Marshall (1996) call it the “resource defense” model, but there must (logically) have been many resource “offensives” as well. So long as the overall population size of these proto-hominids remained small and mobile, and the vast East African environment offered many feasible alternatives, it is doubtful that most of these conflicts resulted in deadly violence. Of course, it is also very possible that the australopithecines followed the male chimpanzee practice of raiding other groups to acquire females, as Manson and Wrangham (1991), Foley (1995), Stanford (1999) and others have suggested. In any case, organized competitive violence, especially with crude weapons, represented another example of a collective, behavioral synergy that is likely to have been exploited at a very early date.
There is some “hard” fossil evidence to support the group-living/tool-using scenario. One compelling source comes from brain “endocasts”, impressions of the interior surface of australopithecine skulls made by paleoanthropologists Ralph Holloway (1975, 1983, 1996), Philip Tobias (1971, 1985) and others. What these surface impressions indicate is that, even as far back as 3.5 million years ago, australopithecine brains, while still quite small by comparison with our own, had already undergone a dramatic internal reorganization. The functional changes were of the kind that are associated with skilled use of the hands, complex social behaviors and more sophisticated communications abilities, though probably not language skills in our sense of the term. In other words, during its first two million years or so australopithecines had moved well beyond the cognitive abilities of modern chimpanzees (the brains of the later australopithecines were also some 20 percent larger) in ways that were compatible with the scenario described above; they were accomplished group-living, tool-using, loquacious bipedal apes, and the changes in their brain anatomy were a reflection of major changes in their behavior (Conroy et al., 1998; Falk 1998; Falk et al., 2000; also Rilling and Insel 1999).
A second form of hard evidence consists of the very changes in dentition, tooth size and shape, enamel thickness and other characteristics, that paleoanthropologists rely upon to differentiate among the various fossils discoveries. Australopithecine dentition was significantly different from that of the chimpanzees. The reduction of their canine teeth and the development of larger molars with thicker enamel implies that australopithecines had adopted a more diversified dietary pattern, with a larger percentage of tough, chewed plant foods (reviewed in Teaford and Ungar 2000). (There is also a suggestive parallel in a late Miocene Sardinian biped, Oreopithecus bambolii, described in Alba, et al., 2001). This, in turn, suggests that behavioral (and dietary) changes had preceded the natural selection of the relevant anatomical changes.
The third source of evidence involves a significant alteration in the hands of the austalopithecines, away from the primate and great ape pattern. These “handy apes” already had relatively dexterous hands; their thumbs were shortened and their fingers flattened, although their thumbs were still not fully opposable. The modern precision grip, or grips (there is more than one), was certainly not perfected in australopithecines, so it is unlikely that their tool-making was very advanced. However, anthropologists Nicholas Toth and Kathy Schick’s important research on early tool-making techniques has shown that even the crude, “flaked” stone tools of 2.5 million years ago required considerable manual dexterity and practice (Toth, 1987a,b; Schick and Toth 1993; Toth et al., 1993). Among other things, Toth and Schick showed that even the famous Kanzi, an intelligent and carefully trained bonobo, was unable to master these techniques. Thus, the australopithecines most likely could do things with their hands that chimpanzees could not, although they were obviously not as adroit as later successors (Steele 1999). Much has been made of the cognitive developments underlying the emergence of language, but the developments associated with skilled use of the hands were equally significant, just as Darwin supposed.
One other form of social cooperation (and synergy) may also have played an important role in the evolution of the australopithecines. It is well known that human females have a much more difficult birthing process than, say, chimpanzees and that the progressive enlargement of the hominid infant brain is partly to blame. Less widely appreciated is the fact that changes in the anatomy of the pelvic region accompanying the shift to bipedalism were also a complicating factor. Accordingly, the anatomical changes evident even in australopithecine females suggest that the human birthing process, with the infant facing the mother’s back, had already become a necessity; the infant’s head had to rotate in order to fit through a more constricted “pelvic aperture.” One consequence was that the childbirth process in australopithecines all but required assistance C midwifery. (A detailed review and analysis can be found in Wolpoff 1999a: 141-142, 271-273.) Thus, in addition to male-male and male-female cooperation, cooperation between females may have become one of the elements of the australopithecine adaptive pattern. In other words, there was a “social triad”, a three-way nexus of cooperation and synergy that created the basic scaffolding for a more elaborate and extensive pattern of cooperation over time.
In sum, humankind may have evolved from closely-cooperating, group-living predecessors that emerged more than five million years ago, and our remote ancestors’ social “coherence,” as Darwin put it, provided a foundation for the development of many other collective synergies as well. This is not to say that the influence of individual competition, status rivalries, internal social conflicts, etc., somehow magically disappeared. Then as now it is likely that there was a sometimes precarious interplay between competition and cooperation, between various individual self-interests and the interests of the group. No doubt differences in individual “personalities” also played a part, just as they do today even in chimpanzees and other primates. A dynamic tension between individual and group interests is a common phenomenon in social mammals (see especially de Waal 1996).
The key to australopithecine sociality lay in the relative costs and benefits to each individual for cooperation or non-cooperation. Why should the males, even if they are closely related, cooperate with one another? And why should the females help one another if they were unrelated and perhaps rivals for social status and the attentions of the males? Reciprocity and reciprocal altruism may help to explain it. But the benefits associated with being included in the group, and the high cost of ostracism, may also have been a major factor. The group was a vitally important survival unit (it produced collective goods that were measured in terms of life and death), which each individual had a stake in preserving and enhancing. In other words, the “public interest” was rooted in the group’s potential for generating collective synergies. For instance, a larger group was more likely, all other things being equal, to benefit from synergies of scale in confrontations with predators or competitors (not to mention potential prey). These collective synergies provided an overarching incentive for containing conflict and enhancing cooperation, and punishing cheaters and free-riders.
The same principle of collective synergy (and policing to maintain it) may well have contributed to the second major transition in human evolution. In the scenario described above, systematic group hunting was evidently not a part of the picture. The current consensus seems to be that the australopithecines may have opportunistically scavenged meat and hunted easily-captured small prey as components of a diversified food quest (see especially the review in Stanford 1999). No doubt seasonal fluctuations and the specific opportunities and constraints of different habitats played a role. However, there are also indications that major behavioral changes began to occur about 2.5 million years ago. A recently discovered 2.4-million-year-old species, Australopithecus garhi (or an as yet unidentified contemporary) at Gona, in Ethiopia, was already adept at transporting flaked stone tools over some distance and using them for chopping, cutting, smashing bones and perhaps skinning both antelopes and wild horses (Asfaw et al., 1999).
The importance of these “crude” Oldowan tools (so-named because they were first discovered at the Olduvai Gorge by Louis Leakey, father of Richard) can hardy be overstated. It really amounted to a technological revolution, because it enabled our ancestors to become systematic hunters (and scavengers) and to exploit the teeming herds of large animals that populated the open grassland areas in East Africa, and beyond. Once stone tools were deployed, moreover, the carcasses of these animals provided “raw materials”, horns, bones, skin and sinew, for many other uses. Just as digging sticks and hand-held weapons may have played a key role in the success of the early australopithecines, the invention of stone tools vaulted our ancestors into a new ecological niche. Equally significant, this adaptive revolution evidently predated the emergence of Homo erectus/Homo ergaster by several hundred thousand years. In other words, synergistic behavioral changes, a technological symbiosis, preceded and supported the major anatomical developments that are reflected in the fossil record much later on.
The specimens of Homo erectus/ergaster that begin to appear in the fossil record about two million years ago were strikingly “improved.” (Major sources for what follows include Rodman and McHenry 1980; McHenry 1992; Lewin 1993; Steudel 1994, 1996; Foley 1995; Klein 1999; Wolpoff 1999a.) First and foremost, these hominids were 50% taller than “Lucy” and her cousins (Australopithcus afarensis) and even greater than that by weight, or approximately equal in height to some modern Homo sapiens. Their brains were also more than double the size of the early australopithecines and showed evidence of further reorganization and a greater degree of lateralization, or asymmetry (associated with handedness and various cognitive specializations). Biologist Robin Dunbar (1996, 1998, 2001; Aiello and Dunbar 1993) makes a strong case for a close relationship between neocortex size, group size and the evolution of language in hominids. Homo erectus/ergaster also had longer legs (and shorter arms) and had perfected the striding gait of modern humans. Not only was this more energy-efficient (by some 30%) than the knuckle-walking technique of chimpanzees, but it was especially well suited for longer distance travel. Whereas chimpanzees in open areas may have ranges of 200 square miles, human hunter-gatherers typically exploit ranges of close to 700 square miles.
Another striking change in Homo erectus/ergaster was a sharp decline in the size differences between the two sexes to approximately modern human proportions (between 20% and 30%, compared to 10-15% in chimpanzees and bonobos). To some theorists this suggests that sexual competition among the males had declined; perhaps the nuclear family and more permanent male-female pair-bonding had emerged. But it is also likely that there were significant functional benefits in having the females be relatively larger. One advantage would have been the ability to travel at a faster pace and not hold back the rest of the group. Another would have been the ability of the females (and the group) to cover more ground and greatly expand their day ranges. Still another advantage might have been the ability to carry heavier infants over longer distances. Larger size also implies a larger birth canal and pelvic aperture, an important consideration in giving birth to larger infants. Certainly it would also have been an advantage for self-defense, when the females were foraging independently or were sequestered at “home bases.” In any case, the decline in sexual dimorphism was another possible example of how behavioral changes were the “pacemakers” for the anatomical modifications that occurred.
A final distinguishing feature in Homo erectus/ergaster was a significant reduction and refinement of the teeth and mandibles and changes in wear patterns, as well as a reduction in the size of the gut. These changes suggest yet another behavioral pacemaker, a dietary shift to one that required less chewing of course plant foods and greater consumption of meat. In turn, these anatomical changes facilitated the later development of the vocal tract and the progressive evolution of language skills. The case for systematic hunting and meat-eating in our ancestors was recently reinvigorated by anthropologist Craig Stanford (1999), and a number of theorists have pointed out that the anatomical changes in these hominids (especially their larger sizes, bigger brains, “expensive tissue”, and larger day-ranges) would have entailed a drastic increase in energy requirements. Leonard and Robertson (1994, 1997) estimate that overall energy needs would have increased by as much as 85%, judging by contemporary human foragers. Aiello and Key (2001), likewise, argue that the increased size and greater reproductive costs for the female (with shortened birth intervals and longer infant dependencies) would have imposed a drastic increase in their daily energy requirements. Estimated daily energy needs for Australopithecus afarensis was 1763 calories, whereas H. erectus might have required 2722-2896 calories. Aiello and Key conclude: “If Homo erectus females adopted this reproductive strategy, it would necessarily imply a revolution in the way in which females obtained and utilized energy…” (See also Martin 1981; Martin and MacLarnon 1985; Aiello and Wheeler 1995.)
A number of theorists have proposed that these nutritional needs could have been satisfied by so-called “underground storage objects” (USOs), a quaint euphemism for nutrient-rich tubors, rhizomes, corms, and other roots (see especially Klein 1999, 2000; O’Connell et al., 1999; also Wrangham et al., 1999). However, it is likely that these items were already a part of the hominid diet, tracing back perhaps to early australopithecines with digging sticks. Moreover, the quantities of energy (and protein) that could readily be obtained (and easily transported) from USOs would have been limited. On the other hand, the East African savannas of 2.5-2.0 million years ago teemed with herds of large game animals.
The most plausible explanation for the transition from australopithecines to Homo erectus/Homo ergaster, I (and some others) believe, is that a major behavioral shift occurred, a shift that was also the “pacemaker” for a cascade of anatomical changes (see Wood and Collard 1999; Wrangham 2001). In the half million years after stone tools became a standard part of their tool-kit, our hominid ancestors made the transition from a diminutive forager that opportunistically hunted and scavenged meat to a systematic group hunter and confrontational, or “power” scavenger (Blumenshine 1987) that relied on meat (as well as plant foods) to provide a more stable, abundant, high-quality, cost-effective food supply. These hominids joined the ranks of “top carnivores” and could hold their own in confrontations with other carnivore competitors, not to mention potential predators. This conclusion is not original, of course (see, for example, Washburn and Lancaster 1968; Shipman and Walker 1989, Wrangham and Peterson 1996; Stanford 1999; Wolpoff 1999a). But I would add that it is also the most parsimonious explanation for the anatomical changes that occurred.
The “trigger” for these changes may have been a significant shift in global climate to a more variable, “oscillating” pattern between 2.8 and 2.5 million years ago, though this was hardly the preeminant cause that some theorists claim.5 One consequence for various East African inhabitants was that they had to cope with a more arid, more seasonal environment with relatively less abundant (more widely scattered) plant foods and relatively more meat on the hoof. A shift to more systematic exploitation of large game animals was a cost-effective behavioral response to these changes. Meat provides twice as many calories as fruit and ten times as many calories as leaves, not to mention protein and other nutrients. And meat can also be obtained in large packages that are susceptible to collective acquisition, bulk transport and shared consumption. Moreover, a shift of emphasis to hunting-gathering, versus gathering-hunting, did not require a great adaptive leap; early hominids were most likely already opportunistic hunters.
Nor was big game hunting necessarily a dangerous activity. There are a variety of less hazardous alternatives. As Kingdon (1993) notes, coordinated tactics of various kinds can be used to ambush prey, or to panic and drive animals into mudholes, swamps, cul-de-sacs, “deadfalls” or even (in later times) into prepared trip lines, nets and traps. Later on, fire brands were probably also used for driving potential prey, as well as for deterring carnivore competitors. In other words, intelligent problem-solving is likely to have played a major role in the momentous change to a hunting/scavenging way of life.
Other scenarios are also possible, of course, but the hunting/scavenging/foraging plus food sharing/provisioning scenario seems most consistent with other evidence, tooth wear patterns, tool use patterns and the anatomical changes that are found in H. erectus/ H. ergaster. Over the course of time there were also progressive improvements in tool-making skills (as reflected in the Developed Oldowan and Aucheulean traditions), plus more selective use of raw materials, more complete “processing” of animal carcasses and evidence of more specialized tools for different uses, such as wood working, skinning, and plant food processing. It is also likely that there was an increased need for transporting food, stone “cores” and, very likely, water supplies over longer distances. Indeed, water became a much more critical resource with the adoption of the hunting/scavenging mode of adaptation in an open savanna environment. As anthropologist Rosemary Newman (1970) observed, we evolved into a “thirsty, sweaty animal” that needed as much as two quarts of water per hour while in hot pursuit of animal prey on a hot day (see also the analyses in Wheeler 1985, 1991; Foley 1995; and Wolpoff 1999a).
The package of behavioral synergies that undergirded the anatomical development of Homo erectus/ergaster in turn provided a foundation for the many improvements that followed. Among other things, this framework allowed for the elaboration of the group as a unit of collective adaptation, with greater social organization, more coordination of activities and especially a division (combination) of labor. One important example was the adoption of consistently-occupied “home bases” or encampments. This led to a significant improvement in economic efficiency for the group as a whole, because it allowed for a more elaborate combination of labor. Resources as needed, meat, plant foods, stone tool “cores,” animal skins, water, firewood, etc., could be carried to a safe haven and then shared and utilized through a network of reciprocities. (For a primate model, see Kortlandt 1992.) Indeed, Richard Wrangham and his colleagues argue that home bases arose through the introduction of cooking, which expanded the range of edible plant foods and encouraged the development of nuclear families (Wrangham et al.,1999, Wrangham 2001).6
In addition to fire and cooking, many other “soft technologies”, nets, weirs, traps, spears, shelters, containers, rafts and more, also, most likely, long predated the vaunted achievements of modern Homo sapiens. The long-term significance of these synergistic inventions was that they enabled Homo erectus/ergaster to generate food surpluses, which allowed for the growth of bigger, more mobile groups of physically larger, longer-lived individuals and, equally important, their more numerous offspring.
However, the invention of more efficient technologies was only half the story. “Culture”, the accumulated know-how and experience of the group, also became an increasingly important part of the hominid behavioral package. Larger cooperating groups were able to exploit many new opportunities for social synergy, including the sharing of costs and risks, pooling information, utilizing a combination of labor and, not least, many synergies of scale against competitors, predators, and prey. Likewise, mutual aid, or “succorant behaviors,” could increase the odds of surviving an injury or illness, and the joint policing of “free riders” and cheats could serve to reduce internal conflicts. Anthropologist Christopher Boehm (1996, 1997, 1999) has also stressed that cultural processes like group decision-making can play a significant role and even become a locus of group selection (see also Sober and Wilson 1998). All of this must also have encouraged the development of supportive psychological propensities. However, the changes were shaped by cultural changes; behavioral innovations were the “pacemakers,” and the social group as a multi-faceted survival enterprise became a “selective screen” for the development of various psycho-social traits, as Darwin long ago proposed.
Some theorists claim that culture did not evolve in humankind until much later, with the emergence of art, rituals, symbolic language and the like. But if culture is defined more broadly (following Bonner 1980) as a body of socially transmitted knowledge, skills and artifacts that are passed between generations via learning and teaching, then culture was already a vital part of the hominid life-style more than two million years ago. After all, the very existence of a tool-making “tradition” implies the ability to transfer the requisite tool-making skills between generations. And this was surely the tip of the iceberg. For instance, modern hunter-gatherers have “mental maps” covering hundreds of square miles, with the precise locations of literally thousands of water holes, plants, animals, natural hazards and other landmarks, most of which they have learned from their forefathers. (See also the discussion of “Unwritten Knowledge” in Diamond 2001.)
The fossil record also suggests that, in later Homo erectus/ergaster, culture became cumulative and an increasingly potent adaptive tool; new ideas and inventions were not only preserved and communicated to subsequent generations but were refined and improved upon over time. The group as a whole became a transgenerational repository of adaptive information and an engine for the invention of more synergies. Spears, for example, came to be made of better raw materials; they were more finely shaped and balanced; their tips were fire-hardened; barbed tips were added to increase their penetrating and holding power; wooden spear throwers were invented as a way to increase their range, striking force and accuracy; finally, bows and arrows were invented as a lightweight alternative that could increase the hunter’s range and precision, and (not least) multiply the hunter’s supply of “ammunition.” Each of these inventions represented a major economic advance. More food could be acquired more dependably with less time, effort and collective risk. (For recent finds relating to the antiquity of various technological improvements, see Bar-Yosef and Kuhn 1999, and Thieme 1999. Bar-Yosef and Kuhn date advanced blade production back to the Lower Paleolithic, and Thieme reports on the discovery of Neanderthal spears that are about 400,000 years old. Some bone tools have also been dated back to about 150,000 years, according to Bower 1997)
The last major transition in hominid evolution, the emergence of anatomically (and culturally) modern Homo sapiens, perhaps 100-150,000 years ago, is currently a focal point of controversy. The self-flattering image of humankind as the product of a saltatory leap of some kind seems irresistible (e.g., Wills 1993, 1998; Diamond 1997; Tattersall 1998; Klein 1999). However, the final “sprint” to humankind was preceded by a long period of progressive cultural and anatomical changes throughout the Middle Pleistocene (from about 750,000 to 250,000 years ago) and beyond (reviewed in Wolpoff 1999a). A number of theorists, Luca and Francesca Cavalli-Sforza (1995), Jared Diamond (1997), Ian Tattersall (1998), Christopher Wills (1998), Richard Klein (1999,2000), Paul Ehrlich (2000) and others, hold that the perfection of human language and the emergence of a more advanced technology were major factors in the modern human diaspora. It is significant that the timing of the African exodus, if true, coincided with the flowering and spread of the Aurignacian industry, which encompassed a range of technological improvements. These included more diversified and specialized (and efficient) tools made from various materials, more skilled manufacturing techniques, better cooking skills, more elaborate shelters, better food storage capabilities, greater use of marine resources, larger population-densities (approximating modern hunter-gatherers), longer occupation of different encampment sites and greater mobility.
Needless to say, a more advanced cultural “package” would have provided an important “economic” advantage, namely, the means to support a rapidly-growing population in diverse habitats. However, the Aurignacian technology may also have given our East African ancestors a major “military” advantage. It seems likely that the great human diaspora of 50,000 years ago was not a peaceful trek into virgin territory but a more hostile invasion of already occupied lands; the human wave was often (perhaps not always) accompanied by coercion and collective violence primitive “warfare.” This is not a new theory, but it deserves a new look.
I hasten to add that we are not talking here about formal wars of conquest or imperialism in the modern sense; the terminal Pleistocene humans were not necessarily more “warlike” in temperament, or seeking dominion for its own sake. More likely, the process was driven by a pressing need for resources to support a growing, mobile population in a changing environment. (The last major ice age began about 75,000 years ago, intensified about 33,000 years ago and peaked about 20,000 years ago.) Call it the “resource acquisition” model of primitive warfare, and of human evolution.
One reason for suspecting that a primitive form of warfare was involved in the final emergence of humankind is that the expanding human populations were not migrating into uninhabited territories, for the most part. Like the 15th century Europeans who discovered that their “New World” was already populated with “natives,” the pre-historic humans who migrated out of Africa found that other hominid groups had preceded them by hundreds of thousands of years. Peaceful trade relationships can arise when there is the possibility of mutually beneficial exchanges between groups. But an attempt to seize another group’s territory and resources is a zero-sum game. The “invaders” came as competitors who threatened the livelihood of the established residents. They would not have been welcomed.
The implication of this dynamic seems obvious. The human diaspora very likely involved many episodes of collective violence, the forcible displacement of the prior occupants by modern humans. However, this coercive pattern most likely would not have been successful without the synergistic advantages of a superior military technology, along with (very likely) superior numbers and (quite probably) improved language and organizational capabilities. There are two strong candidates for what might have provided a “competitive edge” in terms of military technology (others may have left no archeological traces): (1) spear throwers, which greatly increased the range and accuracy of their possessors, and (2) hafted stone axes with sophisticated obsidian blades and bitumin mountings. The advantages of these weapons over simple wooden clubs or primitive hand-axes in hand-to-hand combat should be obvious. It is noteworthy that the first evidence for the use of spear-throwers and advanced stone axes roughly coincides with the estimates for when modern humans first began to migrate out of Africa (Clark 1992; Farmer 1994; Klein 1999, 2000).
There are a great many examples of human conquests in more recent history that were based on some combination of larger numbers, better organization and decisive technological advantages. A chilling example, cited by Jared Diamond (1997), involved the total destruction of the Moriori hunter-gatherer society on the Chatham Islands (in the Pacific) in 1835 at the hands of 900 well-armed Maori agriculturalists from nearby New Zealand. The Maori first learned of the peaceful Moriori from a transcient Australian seal hunter. Excited by the report that the Moriori had no weapons, the Maori immediately organized a seaborne invasion. When the unsuspecting Moriori did not resist, the Maori raiding party slaughtered them with impunity. (Ironically, the two groups traced their ancestry back to a common Polynesian origin, but they had long since lost contact with one another.) (Other examples are cited in Keegan 1993; also Corning 2003.)
The warfare hypothesis also conforms to the principle of parsimony (Occam’s razor), I would argue. The use of organized collective violence involved only an incremental adaptive change, namely, the augmentation and application to other purposes of ancient behavioral patterns and weapons that were first developed by australopithecines and later perfected by Homo erectus/ergaster and archaic Homo sapiens groups in relation to defense (and offense) against predators and competitors in various species. Armed collective aggression for group defense/offense and for hunting/scavenging had been a major part of the hominid behavioral repertoire for at least two million years and possibly twice that long. (This idea also has many fathers, of course.) The role of “warfare” in the human diaspora of about 50,000 years ago does not, therefore, presuppose the evolution of a “warfare gene,” or a “killer ape.” Warfare was more likely to be rewarding to groups/tribes/nations that held a decisive military advantage, a preponderance of power. Some indirect evidence that a pre-historic shift in the “balance of power” played a key role in human evolution includes the massive wave of “overkills” (and extinctions) of many large game animal species, most likely caused by migrating human hunters (Martin 1967; Martin and Klein 1984; MacPhee 1999, especially the chapter by Martin and Steadman), as well as the rapid growth of human populations as they moved into new areas, and the ultimate occupation of many marginal, even extreme environments (in some cases, no doubt, under duress from superior competitors). Once again, Darwin=s group selection paradigm may warrant reconsideration.
In sum, potent new forms of cultural synergy with immediate functional benefits (both economic and “military”) were most likely responsible for the spread of modern humans out of Africa and around the world. Coercion is likely to have played a major part in this dynamic, but it would be wrong to treat warfare as a “prime-mover.” The ability to make war was itself the product of a synergistic package of capabilities.
More important, armed conflict is, fundamentally, an instrumentality for attaining various ends; it is not (usually) an end in itself. The casus belli between these hominid populations were very likely to have been influenced by a combination of factors, including (perversely) our very success in reproducing and expanding our numbers. Violent conflicts between groups are much more likely to occur in unstable or constricted environments, or when resource limits are reached. They are more likely when two populations are in direct competition for vital resources and there is little basis for cooperation. Even then, bloodshed is not inevitable. The odds of violence are also influenced by a more or less explicit calculus of costs and benefits, and risks. If war is politics by other means, as von Clausewitz famously suggested, the converse may also be true; politics may be war by other means.7
A shorthand slogan for this calculus is, again, a “balance of power” (or more to the point, an “imbalance of power”). But this venerable concept implies a many-faceted analytical process, not a narrow statistical exercise. It is well-documented that a disparity of power between groups can be a great inducement to violence, both in nature and in human societies (see especially Otterbein 1970, 1994; Pitt 1978; Alexander 1979, 1990; Robarchek 1989; Manson and Wrangham 1991; van der Dennen 1995; Wrangham and Peterson 1996; and Gat 1999). However, an imbalance of power is never a sufficient explanation for collective violence. Many societies with huge disparities in population, wealth and military power nevertheless live quite peacefully together (the example of the U.S. and Canada comes readily to mind), while some of the most bitter and bloody wars on record have occurred between societies that were evenly matched.
Equally important, the balance of power is an “umbrella concept” that encompasses a potentially very large nexus of contributing factors; it is quintessentially a synergistic effect. It includes not only the number of warriors and weapons that can be fielded by each side but the potency and quality of those weapons, as well as intangibles like tactical advantages, organizational and communications capabilities, training, leadership, the will and determination of the combatants, the “stakes” for the combatants (as von Clausewitz long ago suggested) and, not least, the available alternatives for each side.
The balance of power amounts to a very imperfect science. As many political and military leaders have learned to their regret (and at a terrible cost in lives), wars very often do not go as planned; the synergistic effects produced by these manifold contributing factors often cannot accurately be gauged until after the fact. And, sadly enough, the combatants may not appreciate this eternal verity before the fact. Indeed, many wars are caused in part by the delusions and miscalculations (and machinations) of leaders, and their followers. It is safe to say that any theorist who aspires to a deterministic, predictive theory of warfare is destined to fail; wars are inescapably enmeshed in the skein of history and in the minds and perceptions of fallible humans.
Accordingly, human evolution, a process that may have spanned six million years and is still ongoing, included three distinct transitions, three “great leaps forward,” to use the current rhetoric. And, in each of these major transitions, collective violence of various kinds may have played an important part. The first and in many ways the most important transition involved our ancestors’ shift from an arboreal to a terrestrial mode of adaptation. This momentous change, I have argued, was accomplished with a synergistic behavioral package that included a “triad” of social synergies and a critically important tool/weapon symbiosis. It is likely that this adaptive strategy also included collective violence, primarily for defense against predators but quite possibly for other purposes as well.
The second stage, which entailed a dramatic “hominization”, a suite of major anatomical developments, was the result of a synergistic new pattern of behaviors, including potent new tools, systematic group hunting and, quite likely, the exploitation of fire, the adoption of home bases and the invention of a more elaborate division/combination of labor. Finally, the world-wide diaspora that resulted in the replacement of archaic Homo sapiens and Neanderthals by modern humans about 50,000 years ago was also a synergistic, behavioral/cultural phenomenon as larger groups with more advanced technology and organization overwhelmed other, less advanced hominid populations, not to mention many other “megafauna,” in a world-wide spasm of extinctions.
There is good reason to believe that a propensity for collective violence is an ancient hominid trait; it has probably been a part of our behavioral repertoire for millions of years. Thus we are not radically different from the many other species that routinely use collective violence as an instrumentality for dealing with various “problems of living.” Our basic biological needs give rise to an array of instrumental resource needs, and the pursuit of these resources has often been fraught with competition, conflict and, not infrequently, collective violence.
Modern wars have added several layers of complexity to the patterns of violence that may still be found in more or less pristine form even today in some simple folk societies. If war can be likened to chess, modern wars are more like three dimensional chess or maybe they are many-dimensional, like string theory. But, metaphors aside, the challenge for students of modern warfare not to mention a global community that is daily assaulted and tragically scarred by collective violence is to understand its manifold causes. It is abundantly clear that a great many causal factors may influence the incidence of collective violence in complex human societies, as the various contributors to this volume convincingly argue.
However, there is one causal agency that is also reliably involved, namely, the potent functional effects produced by various synergies of scale, functional complementarities, combinations of labor, symbiotic war-making technologies, informational synergies, and so forth. Moreover, these synergies may play a decisive role both in instigating wars and in determining their outcomes. The recent war in Iraq provides a contemporary example. As the military analyst Kenneth Pollack observed in a book review about the Iraq war in The New York Times: “Most striking of all is the extent to which, for the first time in history, the American military waged a truly joint war in which all the services and all the branches of each service worked to create an integrated force far more powerful than the sum of its parts.” Pollack was talking about synergy, of course. (The aftermath of the war provides a very different lesson, needless to say.)
Synergy is a “common denominator” in collective violence. It is a major culprit. Yet, in the final analysis, synergy is only the servant of various human purposes, impulses and perceptions, and the inescapable conclusion is that, so long as the gains of collective violence are perceived to outweigh the costs, there will be no end to war.