ALTRUISM AND WAR
BY CARMI TURCHICK
ALTRUISM AND WAR: DIVERGENCE AND CONVERGENCE IN GROUP IDENTITY DIFFERENTIATION
War is inseparable from, and impossible without, altruism. Group commitment combined with bias oriented in-group/out-group differentiation, and a psychological defense mechanism capable of convincing even the worst of us that we are basically good, allows us to employ altruistic impulses for war. Humans universally commit to, and identify with, groups which differentiate themselves culturally and psychologically. In-group/out-group differentiations that view the out-group as not having inherently different behavior or desire attributes allow in-group commitment to out-group inclusive higher level groups; for example, humanity. Differentiations that ascribe inherent positive attributes to in-group and, oppositionally, inherent negative attributes to out-group, begin the psychological justification process for racism, aggressive war, and genocide. Commitment connects individual self-image to the group. We are good, therefore our group is good. Conformity and commitment are strengthened by perceived threat, which causes differentiation to become oppositional. We altruistically kill and die.
In order to assert that altruism is a central component of the human psychology of war we must first attempt to establish that altruism exists in humans. To accomplish this task we must also show that it is indeed possible for altruism to have evolved and to have been selected for, it cannot exist if it could not have evolved.
The debate over altruism is quite heated and extensive, and spans many disciplines and sub-disciplines, including Economics, Game Theory, Evolutionary Psychology, Sociobiology, Ethology, Social Psychology, Cultural Anthropology, and Evolutionary Biology. Whole books are dedicated to the issue (Batson, 1991, Field, 2001) while others dwell on it extensively (Sober and Wilson, 1998, Wright, 1994, Boehm, 1999). Theoretical paradigms, such as gene-culture co-evolution (Tooby & Cosmides, 1992), are largely dedicated to solving the puzzle of altruism's evolution in humans. Game Theory has been used to attempt to show that gene-culture co-evolution would be required for altruism to win the selection struggle over selfish “defectors” (Gintis, 2003). Despite this extensive examination by the various interested disciplines there are still major unsolved questions, and even the definition of altruism itself is not agreed upon across disciplines. In the biologically based behavioral sciences, altruism is defined as behaviors that are largely outside of the common use definition of altruism: “selfless devotion to the welfare of others” (Funk & Wagnalls, 1993). This is often, with some slight variations, also the definition found in the social science based work on altruism. The biological definition, however, is "behavior by an individual organism that reduces its own reproductive fitness while improving the reproductive fitness of at least one member of the same species" (Field, 2001). The crucial difference is that the popular definition does not require that the actor's fitness be reduced; it allows the possibility for an act to be selfless but still result in increased fitness for the actor.
It may help our understanding of altruism to slightly alter this biological definition to read "behavior by an individual organism that reduces or risks the reduction of its own reproductive fitness while improving the reproductive fitness of at least one member of the same species." For example, a heroic effort to save someone from a burning building is not altruistic only if the person dies in the attempt, it remains altruistic if they live and do not pay a fitness cost. This can be a critical distinction because evolution is not a process of certainties, but rather one of odds. By recognizing that altruistic behavior need not always result in a loss of reproductive fitness, at least in humans, we see that the average fitness cost need not be as high as it otherwise would appear.
The critical problem for the evolution of altruism, with the biological definition, is that reduced reproductive fitness can only equal fewer (compared to selfish conspecifics) of the altruistic organism's offspring in the next generation, and this seems to only lead to the extinction of whatever individuals might arise with the new genetic predisposition for altruism in otherwise selfish populations. Many also assert that an established population of altruists would likewise be doomed when faced with selfish invaders.
Hamilton’s (1963) theory of kin-based altruism or "inclusive fitness," a willingness to sacrifice oneself for a sufficient number of genetically related individuals to maintain or enhance ones genetic representation for future generations, established one commonly accepted grounds for altruism to evolve. If a mother gives her life to save three of her children, then her genetic representation in the next generation increases. This kind of altruism therefore can be positively selected for by evolution. Even parental investment in the young, for example, which entails one individual increasing the fitness of another at a cost to themselves, is seen as kin-based altruism. Yet, few articles are written about the parents who altruistically feed their own children daily. Rather, we see articles about those who altruistically go out of their way to help unrelated strangers. Sometimes these altruistic acts reduce the relative fitness of not only the actor, but of their kin as well. Kin selection is an insufficient explanation for this behavior.
Reciprocal altruism (Trivers, 1971), and the similar Game Theory strategy of tit-for-tat (see Axelrod, 1984) show us that altruistic acts are in one's best interest in situations where one can expect the favor to be returned, and where a failure to return the favor can be punished. However, examples of altruistic acts where there is no possibility of the favor being returned, such as those resulting in the altruists’ death, are not hard to find. So these contributions also do not encompass the full range of observed altruistic behavior.
A third theory of the evolution of altruism asserts that groups with altruists are more fit than those without, and that this may result in increasing global representation of altruists, through differential reproduction, even though altruists within each group decline in number relative to non-altruists in their own group (Wright, 1945; Maynard-Smith, 1964; Wade, 1978; Sober and Wilson, 1998). This theory relies on group selection, evolutionary selection at the level of the group, an idea that has been controversial. Sober and Wilson, in Unto Others (1998), have made a detailed and extremely well supported case for the existence of group selection as part of their multi-level selection theory. The attempted rebuttal has largely consisted of statements that the author is “unconvinced” (Nesse 2001, Laland and Brown, 2002); a rebuttal that itself is extremely unconvincing. However, this line of reasoning fails to explain, as do the others, how altruism might initially arise and be selected for. It also cannot lead to the observed reality; human groups are universally comprised of individuals who primarily act altruistically, with the exception of a tiny minority. This is an assertion I will support in detail below.
None of the current dominant theories regarding altruism can begin to address heroic altruism; altruism that is non-reciprocal, non-kin related, and directed towards biological out-group. For example, Kathi worked in the Sanctuary Movement smuggling Central American refugees into America. The refugees were non-kin, they were strangers who in biological terms were out-group, and the chances for reciprocity were very slight (it is unlikely Kathi will ever need to be smuggled into Guatemala). Kathi risked five years in jail for each of the 2000 refugees she brought into the United States, if apprehended here, and far worse if she were caught on the Mexican side of the border. Further, she frequently made the choice between food, for herself and for her kin (her four children), and the gasoline that would enable her to drive yet again into Mexico to save refugees lives. How can we account for such acts of heroic altruism?
Current hypothesis tell us that Kathi was acting irrationally and that it is "obvious" that a group of altruists such as Kathi would be easily overrun by "selfish," by those who do not cooperate and who refuse to reciprocate altruistically, and thereby leave the altruist at a fitness disadvantage (Laland and Brown, 2002). If these hypothesis are correct, then we must either assert that Kathi does not exist, that Kathi's 'real' ultimate motivations are other than altruistic (for example, Kathi might be motivated by reputational considerations or the desire for the emotional reward), or accept that further explanations are still needed. Much of this current thinking is influenced by game theory, which frequently applies the Prisoners Dilemma game to the problem.
One common scenario of Prisoners Dilemma has two captured revolutionaries who must decide whether to “cooperate” - remain silent- or to “defect” by turning on their captured comrade. If both cooperate they each get one year in prison. If both defect then both get three years in prison. If one defects and one cooperates, the defector goes free and the cooperator faces the firing squad to be shot. The mathematics, where each result is ascribed a certain point value, appear to show that one should always defect. However, empirical studies show that Prisoners Dilemma players, who of course do not face scenarios including firing squads, usually choose cooperate, even when they know there will be only one round of play and therefore no chance for reciprocity or tit-for-tat (Field, 2001; Poundstone, 1992). Game theory concludes that players who cooperate are acting irrationally. However, some enormous assumptions have been made in the Prisoners Dilemma model and in its application to human behavior.
The players in Prisoners Dilemma experiments exist in social isolation, frequently they are even made anonymous to the other player. This hardly replicates the environment or conditions our ancestors evolved in, and evolution is about adapting to the environment. It is also assumed that the altruist will play blind, they will choose cooperate with no information about the other person involved, and that this also resembles reality. In fact one would be extremely unlikely to find oneself in a real Prisoners Dilemma type scenario with someone else that one did not already have a very well established idea and experience of their altruistic nature. One does not just walk into a bar and get accepted by a group of revolutionaries that night. There is a testing and initiation process that is universal in humans before one is fully accepted and trusted in a group (Brown, 1991). One must build a reputation, and groups also have reputations that affect the behavior of the individuals within them. This second point is often missed.
What would— in reality— be likely to happen in the above Prisoners Dilemma scenario? Let us suppose Jim defects and Brett does not. Brett is killed. Jim goes free; but this is not the end of the story. Rather, this is only the story at the selection level of the individual. At the group selection level we obtain a very different result. Humans universally seek to punish defectors (Brown 1991, Tooby & Cosmides 1992, Boehm 1999, Price, Cosmides, Tooby 2002), those who violate group norms, and especially those who betray the group (for example, whistle-blowers. See John M. Darley in Codes of Conduct 1996). Jim goes free, but now all of his in-group, his revolutionary friends, are his enemies. They are likely to try to kill him, and the government is likely to do little to stop that from happening. No one likes a traitor. In fact, choosing to defect likely results in one of three outcomes, all of which are negative, not fitness enhancing. Either you and your revolutionary comrade spend two more years in prison than you would have otherwise, or you go free but become a social pariah, or you both end up dead. This is true because, at the group selection level, a group has a strong interest in maintaining its fitness. Tolerating traitors is extremely damaging to a groups fitness, and the fitness of the individuals in the group, because it would encourage more defection, and more free-riders, in the future. It would establish your reputation for being a group of altruists easily and willingly exploited. Conversely, the group has an interest in rewarding commitment and loyalty. As Arnold Ludwig points out in King of the Mountain (2002), there is a positive correlation between going to prison as part of a movement that seeks to overthrow an existing government and later coming to power oneself. Over 20% of all rulers in the last century spent time in jail before coming to power. Within the group Ludwig terms "visionaries,” which includes revolutionaries, a full 68% of future rulers spent time in jail before coming to power. The choice to cooperate carries the potential for less jail time, a greater chance (albeit small) of becoming a head of state, enhanced status and reputation if one survives, and at worst one becomes a martyr for one’s cause and group. When the reality of selection at the group level is considered, the decision to cooperate becomes highly rational.
An additional source of confusion involving Prisoners Dilemma is the equating of cooperation with altruism. Cooperation in no way precludes the actor from a relative gain in fitness, fitting the common use definition of altruism, while the biological use of the term altruism does preclude such a gain.
Unfortunately, our altruists are not out of the scientific woods yet. A central problem has been the question of how altruism could initially arise. Once altruists dominate a group, the altruists can enforce conformity to their norm, as happens in band and tribal level societies even now (Boehm, 1999). This is also confirmed by those examining altruistic punishment of defectors and free-riders (Price, Cosmides, Tooby 2002, Fowler 2005, de Quervain, et. al 2004) The first altruists, however, would not be in the majority, so they would be unable to pressure or force others in their group to behave as altruists. Further, they would not be able to establish reciprocal altruistic exchanges, having only selfish to relate to. By biological definition, they would be less fit than the other members of their group, and therefore quickly lose out in the game of evolution if we only consider the individual level of selection. In order for the initial altruists to prosper, there must be counter-balancing selection pressure or pressures that allow them to thrive. Altruists have clearly thrived as, for example, Mansbridge (1990) observed in Beyond Self Interest, “A wide range of evidence suggests that humans routinely engage in altruistic behavior toward unrelated and very distantly related people even in the absence of rewards and punishment.” How do we solve this problem?
It is important to note that kin based altruism can evolve without facing this problem, and that it also can include enforcement and reciprocity mechanisms, which would be adaptive even for kin based altruism. So we need not view the newly evolved more general altruists as having been entirely defenseless against selfish exploitation.
Further, it is possible, even likely, that few expressed behavioral differences would exist until several altruists existed and a reciprocal cycle could be established. Much of our behavior, especially social behavior, is expressed only after the expected environmental stimulus triggers it. Only mentally unbalanced people act randomly. Even in vervet monkeys we see that grooming partners are more likely to take the altruistic risk of assisting one another against predators and other chimps ( Seyfarth and Cheney, 1984, cited in Trivers, 1985)We should expect that the first human general altruist would also not randomly express that altruism, rather they would be likely to express it only when some indication that reciprocity was likely existed. So the first altruist may, paradoxically, never have acted altruistically, leaving that honor to their descendants.
If the existing social system that our ancestors lived in was characterized by a single dominant male, and even two or three reciprocal altruists formed an alliance, they would be likely to achieve dominance. If dominance was achieved, then the resulting monopolization of mating opportunities might allow enough altruists to exist in the next generation to reach the point where altruistic punishment of defectors could be established. At that point dominance becomes a numbers and alliance game, the larger the alliance the more fit it is. Of course, the more members of the alliance, the more mating opportunities have to be shared. Humans seem to have taken this path to it's destination, the point where all males in a group are part of the dominance alliance and an egalitarian division of mating opportunities exists. This still presents a problem; why not stop at an alliance of just over half the males? There must be some selective pressure that makes it a more adaptive strategy to include everyone.
That selective pressure is supplied by group selection, and not just the simple group selection via differential group reproduction rates that is common in the current theories of altruism. Rather, the selective pressure is group selection, of the males, strengthened through the mechanism of war. War provides a second differential; differential mortality at the group level.
LeBlanc and Register (2003) make a strong case in Constant Battles that humanity has a long history of warfare. While debate still exists over archeological evidence that some believe shows a record of warfare going back over 20,000 years, there is abundant evidence that war is now universal, and has been so at least since the end of the last ice age 11,000 years ago. In addition to this evidence, we now know that chimpanzees also make war on each other and wipe out neighboring groups, or at least that they wipe out the neighboring groups males, and they do so with a method of raids that is very similar to human band level warfare and conflicts between gangs in more complex societies (see Wilson and Wrangham, 2003 for a good overview of the literature on this topic). This supports a hypothesis that war may be an evolved predisposition, and it also points to the possibility that when we are considering human group selection we are primarily considering male group selection (see also War and Gender, Goldstein, 2001). To be certain, women and children also die, but the group is wiped out when its males are wiped out. In both species territory is primarily defended, and war waged, by the males (Fishbein and Dess, 2003). Some assert that band and tribal level societies are typically composed of related males and unrelated females (Boehm 1999). (Patton (2000) offers a contradictory example.) We also see that the same pattern exists in chimpanzees, where it is the males that closely bond, the males that are philopatric, and the males who form groups for hunting and for war (Manson and Wrangham, 1991). However, in group selection terms it is not crucial that the group be kin, as "In all cases, a group is defined as a set of individuals that influence each others fitness with respect to a certain trait, but not the fitness of those outside the group." (Sober & Wilson, 1998). Regardless of whether the group is composed of kin or not, we may perhaps view warfare biologically as group level sexual selection, and also as male intrasexual competition at the group level.
In evolutionary theory there are two broadly recognized forces of selection, natural and sexual, as established by Darwin. Sexual selection readily allows the evolution of features that carry a fitness cost. The classic example is the male peacock's extravagant feathers, which evolved due to selective pressure from female peacocks, who selected the males with the “best” plumage as breeding partners. But sexual selection can also work through intrasexual competition, where individual males compete with each other for the chance to mate. These contests can injure and exhaust even the victors. If there is group level sexual selection, then we should expect that women would show a stronger preference for an individual male if he were part of what seemed a more fit male group than the same male in a less fit group. There seems to be no work done on this question. While there has been a great deal of conflict over the issue of group level natural selection, there has been little consideration of group level sexual selection or group level intrasexual competition. In fact these seem to be novel concepts. If the latter two acted as selective mechanisms for warfare and altruism, meaning that women prefer cooperative altruists and that warfare functions at the group level as two rams butting heads does on the individual level (perhaps the existence of ritualized warfare helps support this view), then there is no longer a requirement that altruism and warfare enhance fitness in natural selection terms. Sexual selection allows the evolution of features or behaviors that have a fitness cost. If this is the case, then we should also expect to find gender based differences in some related behaviors, such as variations in group commitment levels.
Fishbein and Dess (2003) point out that “The notion that human females display less in-group favoritism and more acceptance of different others is supported by studies with American high school and college samples...as well as in a large multinational study.” More evidence of an in-group favoritism gender gap can be found in the study of linguistics.
Richerson and Boyd (2001) point out that linguistic differentiation is part of group differentiation “More generally, dialect evolution seems to be set in motion whenever social cleavages form, resulting in the linguistic marking of many social boundaries.” (See also Tajfel, 1978.) Robin Dunbar in his Grooming, Gossip, and the Evolution of Language (1996) observes that, “...men and women differ rather strikingly in the way they learn accents. As they mature, boys tend to adopt their local regional working class accents, whereas girls tend to pick up a more neutral middle-class standard form of English.” These observations, taken together, further indicate that group selection in humans is primarily male group selection, an assertion that lends itself well to the hypothesis that war is a key component in that selection process.
Many who have argued against group selection in general, and the effects of war on selection in particular (see LeBlanc and Register, 2003 for a thorough discussion), have done so with the observation that there are frequent population movements between human groups, and the claim that wars infrequently result in the annihilation of an entire group. Rather, they assert, survivors simply go and join other related groups. As we see from the evidence above, it is the males who are being selected against in all cases, and the females who, if they are able to, go and join related groups. One hardly needs a citation to assert that females are frequently part of the victor’s spoils of war throughout our known history, and Wilson and Wrangham (2003) show that there is some evidence that this is the case at times in chimpanzee warfare as well. As LeBlanc succinctly states, "Rarely were prisoners taken in tribal warfare, except for women who were integrated into the victors’ society. The goal is annihilation of all men, women, and children, although men were the primary target." (LeBlanc and Register, 2003)
What can we derive from male group intrasexual selection, at least partially through the mechanism of warfare, in terms of the evolution of altruism? My research thus far has found only one other attempt to connect altruism and warfare (Patton, 2000), and that effort focuses on the reciprocal status gain of better than average warriors. Simple math shows that this status gain cannot provide incentive for the majority, who by definition will tend to be average and not above. Yet, when we consider the needs of an individual living in a world where there is imminent danger of an attack by another group of men, altruism on a daily basis may suddenly seem a rather obvious course of action.
Warfare tests the fitness of groups of men at the group level. While complex societies are rarely wiped out, the band and tribal level societies that represent the group size we evolved in may have been frequently eliminated by rivals. LeBlanc and Register (2003) offer various examples, clustered in the range of 25-30%, of groups wiped out every century. It is widely accepted that altruism generally increases the group fitness; the altruist usually pays a smaller fitness cost than the beneficiary gains in fitness. If we examine the common case of a group where there are different abilities for hunting, as Boehm (1999) has done, we see that sharing of the meat is universal, and that good hunters are frequently kept from using their talents to directly establish dominance by social norms. Yet there is no balance of reciprocity, there is no chance that the good hunter will suddenly lose his talents and the poor hunter will suddenly become expert. The good hunter will simply act altruistically and contribute meat to the poor hunters many more times than he will receive meat from him. How does the good hunter benefit if the egalitarian norms of the community, as in an example cited by Boehm (1999), require him to leave his kill at the edge of the village and return to the campfire empty handed with a story of failure? He benefits because of warfare. His life is partially dependant on the strength of the other males in his group. No man, no matter how fit in evolutionary terms, can stand alone against a group. Even if there have been men who could fight ten men and win, it can hardly be a stable strategy in evolutionary terms. The existence of one group of men who are willing to altruistically help each other, who are willing to form a coalition for war, requires that neighboring groups do the same or perish. “All for one and one for all” gives a decided advantage in warfare (and also against predation in general). If the penalty for out-competing altruists within the group is death at the hands of an out-group of altruists, then suddenly altruism becomes much more appealing in biological terms. Therefore, altruism conforms to the biological definition only at the individual level of selection; it decreases the relative fitness of the altruist compared to that of other in-group members. However, altruism at the individual level can result in mutualism at the group level; both parties become more fit at the group level (Sober and Wilson, 1998, state “In a population, altruistic behavior by many may result in benefits to many, including the altruist.”). With warfare operating as a selection mechanism, the fitness benefit at the group level of acting altruistically exceeds the fitness penalty at the individual level. By this line of reasoning, altruism can be seen as evidence of a history of warfare.
The hypothesis of altruism producing mutualistic results at the group level is supported well, including game theory modeling and biological examples (lions, who form kin-based male coalitions for the purpose of intrasexual competition), by Mesterton-Gibbons and Dugatkin (1992) and their concept of by-product mutualism. They observe that “There are three categories of cooperation among unrelated individuals: group selected behavior, reciprocal altruism, and by-product mutualism.... The mechanism for by-product mutualism is the common enemy of a sufficiently adverse environment. A prerequisite for this mechanism is the boomerang factor, that is, any uncertainty that increases the probability that a noncooperator will be the victim of its own cheating.” What better "common enemy" can a group have than a real common enemy, an out-group that creates a decidedly adverse environment when it wages war against the in-group? The cheater in such a group faces a very real danger of a boomerang factor. If he weakens other group members, and thereby the group, by cheating, the group may be wiped out and the cheater along with it. If he holds back in combat, the tide of battle might turn against his group, and he will share the groups fate. Ironically, if all men were selfish defectors we would quickly realize the answer to the utopian question, “what if they held a war, and nobody came?”
Of course, there is a spectrum of behavior that exists between selfish and altruist, as opposed to the common binary view. Some may only act altruistically so long as the risk appears slight, while others simply conform, and still others become heroic altruists, jumping on grenades and charging machine gun nests with pocketknives. Heroic altruists run into burning buildings, hold the position until everyone else has safely retreated, smuggle refugees, start orphanages in distant countries, become kamikazes, and become suicide bombers. They are willing to sacrifice altruistically for the group in ways that are neither kin related nor potentially reciprocal, particularly for those who die altruistically. (While many conflate altruism and morality, the behaviors that derive from our altruism are typically “moral” when directed towards an in-group, and typically “immoral” when directed towards an out-group on behalf of an in-group. However, as Bloom discusses in his Global Brain (2000), we also frequently see immoral acts directed against in-group members who do not conform.) Heroic altruism has been largely ignored in the literature on altruism, perhaps because the standard theories could not account for its existence. In our case, however, it fits in well with our hypothesis. It is highly advantageous in war, for example, for the group members to be willing to die for the group; to value their own lives lightly. There are examples of strategists creating this effect intentionally; by burning their own ships, placing the army where it cannot retreat, burning the bridges behind them, etc. As Sun-Tzu comments in his classic, Art of War, “Throw your troops into situations from which there is no way out, and they will choose death over desertion. Once they are ready to die, how could you get less than maximum exertion from your officers and men?” (in The Book of War, Caleb Carr, 2000). During most of our evolutionary past, there would have been little opportunity for retreat. Bands and tribes lived surrounded by enemies or potential enemies. In these cases there would have been “no way out.” The group either survived or perished together, and individual willingness to perish altruistically, to be heroic altruists for the group, increased group fitness.
It may seem that the argument above is the old Wynne-Edwards group selection, long since demolished, of adaptations evolving solely for the good of the group. However, instead the point is that individual fitness is highly dependant on group fitness, and that in the case of humans a group that became too dominated by selfish, in comparison to neighboring groups, would quickly be eliminated from the gene pool, both altruist and selfish alike.
We also have to return here to our slight alteration of the definition of altruism. Evolution is a process of averages and odds, not of certainties. If altruism evolved as the willingness to risk altruistic sacrifice, and not necessarily to lose that gamble, then this also helps us understand how altruism might have evolved. In many situations, having a group of individuals all willing to risk for one another greatly lowers the odds that any of them will pay the fitness cost of taking that risk. If one person is being attacked by a lion and no one helps them, they are doomed. If one person tries to help, their risk is great. If twenty come to help, the risk is quite minor for all of them. The more altruists, the lower the average fitness cost of being an altruist becomes. Comparing a group of altruists to a group of selfish, it is obvious that the selfish would feed many lions, and the altruists would not. The willingness to risk death increases ones odds of avoiding it.
The counter argument typically asserts that, regardless, it is obvious that selfish would out-compete altruists and replace them in each group. This view seems to neglect the central feature of selfish, that they do not cooperate, and a central fact about humans in general, that they often need to work together to survive at all. Given these two points it is possible to suggest that a better understanding of the dynamic between selfish and altruist might be found in models of parasite/host interactions; a group of altruists is effectively a host body for a selfish parasite. Three important insights result. First, altruists are a resource for selfish, and therefore we should expect competition between selfish for access. (In complex societies, for example, criminal groups lower their fitness by fighting each other, thereby lowering the cost of their existence to society.) Second, it might be possible that, just as lower virulence can evolve in diseases, lower virulence in the form of becoming less completely selfish may have evolved in selfish populations. High virulence that kills the host is maladaptive compared to low virulence that allows long term infection and reproduction. Third, the fitness of selfish should correlate with the ratio of selfish to altruist; the more selfish, the lower their fitness is. The result of these would be to lower the cost of altruism, and to decrease the fitness advantage of being selfish.
Perhaps then the solution to how altruism can be a stable evolutionary strategy is a combination of multiple factors which increase the fitness of altruists and decrease the fitness of selfish. These might include benefits to kin, reciprocity, by-product mutualism, altruistic punishment of selfish, interference by selfish of other selfish, female sexual selection preferences for altruistic males, and coalitional intrasexual competition in the form of war. While we may be unable to prove that the above would suffice, it would be equally difficult to show the opposite to be true. When we also recognize that we are neither pure altruists nor blind ones, that we all are somewhat selfish, that we all attempt not to be taken advantage of by selfish, and that our behavior changes depending on environmental stimuli, then it seems quite possible for evolution to select for altruism.
Supporting much of the above is Darwin’s own view from The Descent of Man (1871, p. 166), “It must not be forgotten that although a high standard of morality gives but a slight or no advantage to each individual man and his children over the other men of the same tribe, yet that an increase in the number of well-endowed men and advancement in the standard of morality will certainly give an immense advantage to one tribe over another. There can be no doubt that a tribe including many members who, possessing in a high degree the spirit of patriotism, fidelity, obedience, courage and sympathy, were always ready to aid one another, and to sacrifice themselves for the common good, would be victorious over most other tribes; and this would be natural selection. At all times throughout the world tribes have supplanted other tribes; and as morality is one important element in their success, the standard of morality and the number of well-endowed men will thus everywhere tend to rise and increase.”
Before we leave the topic of altruism we must briefly consider whether it is an ultimate motive psychologically, or whether deeper selfish drives underlie its expression. Sober and Wilson (1998) offer a very detailed discussion of this issue, which space does not allow us to explore fully. One common argument, which they mention, is that we act altruistically in order to gain the emotional reward; that our ultimate motivation psychologically is a selfish desire for feeling good. This idea fails to explain the motives of heroic altruists, many of whom gain no emotional rewards because their altruism results in their deaths. Further, this argument uses the strongest evidence that altruism is a psychologically ultimate human predisposition to argue against altruism. That acting altruistically universally results in an emotional reward is good evidence both that it is an evolved predisposition and that it has an important connection with fitness, as do all other universal behaviors with such a reward, such as mating, eating, etc. (One would hardly assert, for example, that the ultimate drive for eating is the emotional reward and not hunger, even despite the fact that we may indeed eat more when depressed.) This strongly implies that altruism is an ultimate psychological motive. Additionally, recent empirical studies show that empathy, which is a required ability for altruism to occur (see Batson et al. 1991 for empirical work connecting altruism and empathy), is also a genetically based predisposition, found even in new born babies, and is not ultimately a cultural or even a conscious decision (Keysers 2003). The Neuron article about Keysers' work makes it very clear, “Although Keysers said that empathy for others is often thought of as a matter of morals, ‘in our study, on the other hand, we show that empathy is a very basic, simple and automatic process,’ he said. Keysers explained that when we see the emotions on another's face, ‘we don't need to think about how that person feels.’ Instead, according to Keysers, we share the feeling of disgust because the insula is activated as if we were disgusted ourselves. ‘This sharing is automatic,’ he said. ‘Our subjects were not asked to share the emotion of the other person and did not report attempting to do so after the scan. It just happens.’ ” This shows that empathy is an ultimate motive— we do not feel empathy because of any deeper psychological need— and further supports the same conclusion for altruism as well.
WAR AND GROUP SIZE
Beyond altruism, what other adaptations might be advantageous for war? Even in chimpanzee warfare, a primary objective is to outnumber the opposing group (Wilson and Wrangham, 2003; Wilson, Hauser, and Wrangham, 2001). The aim of raiding style warfare, whether practiced by chimpanzees or humans, is to find and attack a single victim with a group. The human history of warfare and conquest, as well as the much smaller sample from chimpanzees, shows a clear pattern of the many conquering the few. (The current controversy over chimpanzee warfare does not weaken these assertions. Whatever affects human interference may have had on chimpanzee behavior, at most we can only have triggered a behavior pattern that was latent. No one asserts that warfare was in any way “learned” by chimpanzees from humans.) What adaptations, then, might allow for a larger group so that one might gain this crucial advantage? Robin Dunbar shows in his Grooming, Gossip, and the Evolution of Language (1996) that bonding by reciprocal grooming of fur requires a time and energy commitment that becomes impossible beyond a certain number of individuals in a group. There just is not enough time in a day for all of the members of a very large group to remain sufficiently bonded via grooming to maintain the group’s cohesion.
Language, Dunbar proposes, offered a solution to this problem, allowing humans to bond with less effort and time, and therefore, in greater numbers. Further, language allows, as Dunbar also points out, a greatly enhanced ability to enforce altruism. We gain the ability to become informed of an individual's non-cooperation without having direct knowledge of the infraction. This allows enforcement of cooperation by the group collectively, reducing the fitness cost to each individual to an insignificant level in most cases (Boehm, 1999). The function of language in promoting group cohesion can only further tilt the fitness equation towards altruism.
Language also allows for the formation of alliances between groups. Thus, two small groups may succeed together against one larger group, and many groups may band together to form a coalition of groups. By this process, a band may become part of a clan, the clan a part of a tribe, and the tribe in turn a member of a confederation of tribes. The process may eventually lead to the development of nations. But language alone is insufficient for these alliances to prosper. The altruistic commitment necessary for a band to go to war beside another band is similar to the commitment an individual must possess in order to go to war for his band. We certainly need to feel confident that our allies are committed and will altruistically go to war on our side. And, as we have established above, if we do not cooperate and act altruistically then we do not show up for battle. If an alliance has formed because of a mutual threat, then non-cooperation risks the same boomerang effect for the group as we see for the individual in the by-product mutualism concept described above. What we needed was a plastic psychological bonding mechanism, an ability to bond psychologically with a group larger than the number of individuals we can know and keep track of socially. That alone would prove insufficient, however. We also needed easy ways to keep track of who was in-group and who was out-group; there could be no reasonable expectation that we would be able to remember and recognize all of our hundreds or thousands of allies. Suddenly we needed the ability to distinguish between strangers who were in-group and strangers who were out-group, and we needed to make sure that we ourselves were not mistaken for out-group. In short, we needed to evolve our human group behavior predispositions.
HUMAN GROUP BEHAVIOR PREDISPOSITIONS
Humans are social territorial animals. This means we defend territories as a group, in our case a group of males. In order to do so we must be able to identify our in-group and differentiate it from out-groups. The advantages offered by belonging to a larger group require the acquisition of an adaptation, or set of adaptations, that allow us to commit to non-kin based groups, and even groups larger than we can personally bond with, and to quickly distinguish between in-group and out-group individuals. In addition to this, we need to be able to differentiate in-group from out-group both oppositionaly and exclusively or non-oppositionaly and non-exclusively, in which case we simply categorize them (Many in the social sciences go astray here, failing to distinguish between the two types of "other," see Staub, 2005, for example). We need to be able to differentiate enemies from allies and potential allies. In high school terms, football players are out-group to basketball players. However, they are not typically oppositionaly defined, one may belong to both groups and positive attributes of one do not imply oppositional negative attributes in the other. “Nerds” are out-group to both football and basketball players, and they are typically oppositionaly defined. One is either a “nerd” or a “jock,” and not both, though both groups may be composed of sub-groups that share no members with other constituent sub-groups. Beyond that, both the “nerds” and the “jocks” are likely to have higher group level commitments that they share, such as religious, political, or national loyalties. So the animosity between the groups may be highly situational and dependent on the group level of threat. This points to a highly plastic ability to commit to multiple in-groups of various sizes; we may even commit at the level of all of humanity.
At the individual level, the behavior predispositions we need to insure the fitness of our in-group are commitment, conformity, positive in-group bias, in-group based altruism, and identification with the in-group. Without commitment from its constituent members, an in-group would lose cohesion and quickly fail. Conformity insures that we are not mistaken for out-group, which could prove extremely negative for an individual’s fitness, and that we do not violate norms or commit other acts that might bring social sanction. Positive in-group bias, the favoring of fellow in-group members, also includes a tendency to rationalize and excuse transgressions of in-group that would be condemned when committed by out-group (Messick and Tenbrunsel, 1996;Weber, 1994). Finally, without in-group based altruism we would be unable to go to war as a group. Beyond that, repeated altruistic acts within the group also serve to strengthen social bonds and individual commitment to the group. For example, sharing food is a universal social bonding mechanism (Brown, 1991) and it is inherently altruistic. By identifying with our in-group we come to see the other members as being similar to ourselves (Bar-Tal, 1990), effectively allowing us to behave towards them as we would towards kin.
The last predisposition we need to address, before we examine the psychological process that brings us to war, is not group related; it is our own biased self-image.
A recent poll by the Barna Research Group showed that, of those who believe in Heaven and Hell, only .05% believe themselves to be going to Hell. Similar findings concerning other attributes are common. Thomas Gilovich cites a few examples, “A survey of one million high school seniors found that 70% thought they were above average in leadership ability, and only 2% thought they were below average. In terms of ability to get along with others, all students thought they were above average, 60% thought they were in the top 10%, and 25% thought they were in the top 1%!...A survey of university professors found that 94% thought they were better at their jobs than their average colleague (Gilovich, 1991).” We think very highly of ourselves.
Some suggest that it is psychologically healthy and adaptive for us to have a positively biased self-image (Mele 1997; Taylor and Brown, 1988), that a lack of positive self-deception is a sign of depression (see also Trivers, 2002). Michael J.C. Waller goes a step further and asserts that, “On the negative side, low self-esteem leads to physiological and behavioral changes that result in elimination from the gene pool (in van der Dennen et al., 1999).” This suggests that a positively biased view of oneself may be adaptive. Unfortunately, positive bias in self-evaluation also comes into play for individuals who act in ways that we easily define as evil in a moral sense. For example, the Nazi doctors who helped kill millions, and who tortured and killed tens of thousands directly in their experiments.
Lifton (1986), in his book Nazi Doctors, shows that even these doctors felt themselves basically good, “...the dominant tendency among these Nazi doctors was to present themselves to me as decent people who tried to make the best of a bad situation.” These Nazi doctors hoped Lifton would confirm their own views. Perhaps unsurprisingly, they were also unable to examine their actions in an unbiased way, “Yet none of them— not a single former Nazi doctor I spoke to— arrived at a clear ethical evaluation of what he had done, and what he had been part of.” The Holocaust stands out as one of humanity’s most unambiguously evil episodes, yet not one of these central perpetrators would admit that to themselves. They all maintained that they were basically good.
GOOD PEOPLE ALTRUISTICALLY GO TO WAR, COMMIT EVIL ACTS
There are psychopaths in the world, and some of them do lead others into war. However, all of the evidence indicates that they are a very small minority. There can be no doubt that the vast majority of men who go to war are not psychopaths. Indeed, some of the key indicators that someone is a psychopath are that they lack altruistic impulses and empathetic feelings. As we established above, if the vast majority were not altruistic they would not go to war. Yet, merely citing all the documented historical incidents of brutality, rape, torture, massacres of women and children, and ethnic cleansing during war would likely result in thousands of pages of citations. We must briefly mention here that a crucial misunderstanding has been the common conflation of behaviors towards in-group and towards out-group. That a person commits horrific immoral acts against out-group members has proven a poor predictor of like behavior towards in-group, (In many instances, the vast majority participate in the commission of massacres and the like together. It is difficult to have a massacre with just a few perpetrators. A positive correlation with immoral acts towards in-group would mean a society in complete internal chaos, and that is the opposite of what is typically observed.) Likewise, the reverse; those who commit criminal acts against in-group are not known for their willingness to go to war on the groups behalf. Immorality towards out-group and immorality towards in-group are largely unrelated psychological phenomenon; the former is driven by altruistic impulses while the later reflects the absence of those impulses. Having clarified this point, let us now move on to our central question: what is the psychological path to war and how does it take good people to a place that allows them to commit evil acts?
The path to war begins with our individual belief that we are good in moral terms. When we commit to an in-group it "...results in reducing the perceived differentiation between the self and members of the in-group..." (Brewer in Codes of Conduct, 1996). This finding is common. In Group Beliefs (1990), citing Wilder (1978), Daniel Bar-Tal states "The formation of a group reality "We are a group" was enough for individuals to assume that they were relatively similar to one another and dissimilar to the out-group members." This effect is apparently strong enough that it occurs even among individuals who were categorized as a group by researchers on the basis of consensual dislike (Turner et al. 1983 cited in Group Beliefs). Sober and Wilson (1998) state "Good deeds performed by the Yankees make one proud, foul deeds make one ashamed. ‘What they do reflects on me,’ or so the Yankees fans seem to feel. This same pattern of thinking is present in deeper and more pervasive types of identification— with family, clan, ethnicity, nationality, and religion...Human beings don't simply belong to groups; we identify with them." The first step then on the path to war is to apply our own inaccurate view of ourselves to the group. They are similar to us, and we are way above average and basically good, no matter what our past behavior might indicate to an outside observer.
This application of self-image to the group results in positive in-group bias, according to Cadinu and Rothbart (1996), who describe this as “self-anchoring.” “According to the self-anchoring principle, in-group perception in the minimal group paradigm will be largely based on self-perception. The second process implicated in in-group favoritism can be thought of as a differentiation process, according to which people infer out-group characteristics to be different from in-group characteristics. Because the in-group is assumed to be favorable, different comes to mean unfavorable.” In a series of four experiments, they generated significant empirical data to support their self-anchoring principle (Cadinu and Rothbart, 1996). Discussing the findings of Weber (1994), Brewer states, "This finding is one illustration of a general leniency bias in favor of in-groups. In-group and out-group members are equally likely to be given credit for unambiguously positive actions and to be blamed for unambiguously negative ones. When possible external justifications are present, however, in-group members are less likely to be held accountable or be blamed for negative behaviors or for failures to do something positive...The rule appears to be, when judgments are uncertain, give an in-group member the benefit of the doubt." De Vries (2003) states, “There is little empirical doubt that in-group identification affects in-group bias...meta analysis shows the relation to be positive and strong.” This bias also includes a consistent finding that in-group members favor fellow in-group members when given money to allocate, even in extreme minimal group scenarios where they have met neither fellow in-group nor out-group members, and are only in-group because they have been informed by the researchers that they are (Tajfel, 1978; Brewer in Messick and Tenbrunsel, 1996;many others). As Judd and Park (1988) put it "In-groups consistently show favoritism for members of their own group when compared with out-group members in evaluating performances, personalities, and behaviors. This ethnocentrism effect (as it was first labeled by Sumner, 1906) has been demonstrated even when the in-group versus out-group distinction is rather arbitrary, when the groups have no prior status outside of the experimental situation, and when equivalent information has been presented about all group members." Turner et al. (1983, cited in Group Beliefs) showed that individuals who were categorized as in-group exhibited similar positive bias behavior whether the group was selected based on mutual liking or on consensual dislike; in-group favoritism seems to be automatic even when all of the in-group members hate each other.
Favoring in-group makes perfect evolutionary sense; if we did not favor in-group there would be little point to having a group at all, and altruism would be out of the question. However, as Brewer points out, while positive in-group bias is a separate phenomenon from negative bias towards out-group, both can appear the same to out-group members in their milder forms. Bob may have been promoted over John because of positive in-group bias, but to John it looks just like negative bias against his out-group. John is likely to form a negative bias towards the in-group in response, and the cycle has begun. Before we can continue on the path, however, we must make a crucial distinction between the two forms group differentiation can take. There is a critical difference between the psychology of group categorization and the psychology of group oppositional differentiation.
One standard game boys play features the “good guys” against the “bad guys.” It has likely never yet been seen that the scenario has become “two groups of good guys who are competing over limited resources.” Yet many other games feature two teams without making reference to the morality of either one. Good guys vs. bad guys is an oppositional differentiation; if one group is the good guys, then the other must be the bad guys. In some ways the path to war is no more complex. We believe, as discussed above, that we are a good person and that our in-group is composed of individuals similar to us, whom we therefore are strongly predisposed to view as good (Cadinu and Rothbart 1996). If there is an out-group we are in conflict with, we become strongly predisposed to view them oppositionaly. We are good, and, oppositionaly, they are bad. However, because we are capable of commitments to many groups at many levels, including the group of humanity, the process becomes more complicated than the children’s game or the minimal group paradigm.
Sherif et al. (1961), in the famous Robbers Cave experiments, showed how readily negative bias and aggression can be generated when two groups are oppositionaly established and put into competition with one another. Just as importantly, the experiment demonstrated that cooperation and good relations can replace bias and aggression when the two groups are required to work together to achieve a new task. The introduction of a higher-level goal resulted in a higher-level in-group composed of both groups. This plasticity of commitment is a crucial factor for our path to war and, as we shall see, for our potential path to peace as well.
As discussed above, we are able to commit to many groups at many levels at the same time. So long as these groups do not become oppositionaly defined to each other there is no problem. Importantly, not all groups at the same level are oppositionaly defined towards each other. For example, in sports this can be seen as the difference between rival and non-rival teams. If you are a Cowboys fan you may also like the Bears, but you may not like the Redskins. Cowboys fans are partially defined as not Redskins fans, but there is not an oppositional differentiation towards all other teams. In-groups require one oppositionaly differentiated out-group, but other groups may be allies or enemies or neutral. Further, one may be committed to higher-level groups in common with even oppositionaly differentiated out-group members. The Cowboys fan may be committed to the same groups at the religious and national and political level as the Redskins fan. They may even be part of the same professional organizations, and sometimes even kin. However, at the level of which team they are fans of, and committed to, they are still capable of group violence against each other. While this is evidenced by the violence between rival soccer fans in Europe today, perhaps its most robust known expression was the Nika Revolt in Byzantium.
The contemporary historian Procupius gives us this description of the events leading up to the revolt of 532 A.D., which involved the Blue and Green fans of two opposing chariot teams.
“In every city the population has been divided for a long time past into the Blue and the Green factions; but within comparatively recent times it has come about that, for the sake of these names and the seats which the rival factions occupy in watching the games, they spend their money and abandon their bodies to the most cruel tortures, and even do not think it unworthy to die a most shameful death. And they fight against their opponents knowing not for what end they imperil themselves, but knowing well that, even if they overcome their enemy in the fight, the conclusion of the matter for them will be to be carried off straight away to the prison, and finally, after suffering extreme torture, to be destroyed. So there grows up in them against their fellow men a hostility which has no cause, and at no time does it cease or disappear, for it gives place neither to the ties of marriage, nor of relationship, nor of friendship, and the case is the same even though those who differ with respect to these colors be brothers or any other kin… .I, for my part, am unable to call this anything except a disease of the soul… .”
This offers a rather dramatic example of extreme group violence, even given conditions fairly similar to minimal group conditions commonly used by social psychology. As Procopius describes it, the situation was group over all. Group over kinship, friendship, family, and even over self. Yet there was no real basis for the conflict or even for in-group identification and commitment. And, as the Nika Revolt showed well, the two factions were able to quickly unite, despite their past enmity, against the higher level group represented by the Emperor Justinian. Perhaps the most singularly effective means of uniting two warring factions is to convince them both to make war on a third faction instead. How did these two essentially minimal groups go from self-anchoring and in-group bias to group warfare, with no possible concrete rational justification?
Because there are, in complex societies of today, nearly an infinite number of in-groups, and because one clearly does not typically have negative bias towards more than a small portion of them, we see that oppositional differentiation may be needed for such negative bias to arise. On the national level, we are now able to have conflicts with any of well over a hundred other nations. Despite that, we are at peace with the vast majority of them the vast majority of the time. In contrast to this, Reynolds, Falger and Vine (1987) point out that “...it must be recognized that in pre-industrial societies in particular, in-group attachments are usually intense and out-groups are not only regarded as inferior, but are often vilified as the very essence of evil.” Yet, it is extremely uncommon for a complete outsider, such as an anthropologist or missionary, to be met with aggression by the same tribes they later document as being highly aggressive towards out-group. The distinction lies in oppositional differentiation. The groups are oppositionaly differentiated from their neighbors, but not from strangers such as anthropologists from Europe or America. The anthropologist is simply neutral, neither in-group nor out-group. In complex societies, and on a world scale, this is how we regard most other groups, as neutral. We categorize them as other on some group levels but we do not oppositionaly define them, or psychologically exclude them, from our higher-level groups. So, for example, we may act altruistically to help starving children in many countries around the world. They are part of our in-group of humanity, and we have not been driven by an oppositional differentiation to exclude them from that group. Even in the case of chimpanzees it seems possible that a parallel process may take place. Wilson and Wrangham (2003) note that in captivity, “Newly introduced males are predictably aggressive to one another, but appropriate management can lead to eventual acceptance.” This process has not been seen in the wild, however. Perhaps the introduced male is not oppositionaly differentiated and is only viewed as a stranger, as we saw above for humans. Those same children and those same introduced males fare far worse when they are oppositionaly differentiated from the in-group.
In his book, Evil (1999), Baumeister recounts the massacre of a group of Native Americans, and the unfeeling use of a three-year-old boy for target practice by three American soldiers. There must be a psychological process that takes us from positive in-group bias and oppositional differentiation, seeing ourselves as the good guys and the out-group oppositionaly as the bad guys, to a willingness and ability to kill without emotion, to massacre and commit genocide, or to blow ourselves up in order to kill. There is no doubt that most Cowboy fans are unwilling to commit such acts against Redskins fans, but it is possible for them to become willing if soccer hooligans and the events leading up to the Nika Revolt are any indication. This path continues with conformity and its effects on group polarization.
CONFORMITY AND GROUP POLARIZATION
Individual conformity to in-group norms is strongly fitness enhancing. We can commit to groups far too large for there to be any possibility that all individual in-group members will know all other in-group members. Therefore, to avoid being mistaken for out-group and becoming subject to attack, and to continue to have altruistic interactions with fellow in-group, conformity is a necessity. Every society and group differentiates itself visually and culturally, and does so in an oppositional way. If one gang has blue bandannas (Crips) as an identification symbol, the other will choose red bandannas (Bloods). If one religion eats only spicy food (Sikhs), the other will eat only mild food (Hare Krishnas). Much of the time spent gossiping deals with the question of conformity, whether others are conforming and whether the gossipers are conforming (Dunbar, 1996). Best and worst dressed lists are effectively nothing more than a question of whether the famous people in question are conforming. But conformity is not only a matter of linguistic and visual signals. It also informs beliefs.
Solomon Asch’s famous experiment (Asch, 1951) in which 75% of participants either changed their minds or chose not to contradict the majority about the factual matter of how long a line was, due to the unanimity of those choosing the wrong line, is informative. We not only change our minds, but, as Asch also found, we seem to then have no conscious knowledge of having done so. There is empirical evidence that this effect extends to political policy views. Geoffrey L. Cohen (2003) showed that the primary salient factor governing whether an individual supports a policy is whether or not the political party they are committed to does so. “For both liberal and conservative participants, the effect of reference group information overrode that of policy content. If their party endorsed it, liberals supported even a harsh welfare program, and conservatives supported even a lavish one. The results are consistent with the contention that people base their attitudes on social meaning. Once the policy was socially defined as liberal or conservative, the persuasive impact of its objective content was reduced to nil. Participants also denied having been influenced by the stated position of Democrats and Republicans, and instead they claimed that their beliefs followed from an apprehension of policy content (guided by their personal philosophy of government).” The evidence, from both sources, that we are unaware of the groups’ influence in changing our beliefs is an important factor, both for the path to war and for in-group polarization.
Group polarization is an extremely well established phenomenon (Baron and Kerr, 2003; Sunstein, 2002). It has been shown empirically numerous times that “...deliberation tends to move groups, and the individuals who compose them, toward a more extreme point in the direction indicated by their own prediliberation judgments. For example, people who are opposed to the minimum wage are likely, after talking to one each other, to be still more opposed...(Sunstein, 2002).” Several theories have been proposed to explain why polarization of group views occurs, of which none include conformity, oppositional differentiation, or a predisposition for inaccurately assessing group views. Rothbart and Mauro (in Messick and Tenbrunsel, 1996) point out the consistent pattern that individuals view in-group beliefs as more extreme than they in fact are, and see out-group as even more extreme than in-group. If one perceives in-group beliefs to be more extreme than they are in fact, then one is likely to conform not to the actual group beliefs, but to one's erroneous view of them. As social group animals, we have a deep desire not to be ostracized by our in-group. In fact, we feel ostracism quite literally as pain, “Two key areas of the brain appear to respond to the pain of rejection in the same way as physical pain...(Eisenberg et al., 2003).” This underscores how important the group is to survival, but it also points to a likelihood that we will express views more extreme than our own, in order to conform to what we believe the group norm is and to avoid the possibility of ostracism. (We may even do this without conscious intent, although it is unclear whether the subconscious conformity described by Asch and Cohen can occur as an anticipatory phenomenon.) We can then see how polarization may occur. If all of the members of a group express views in their discussion of a topic that conform to their erroneous view of the group norm, then that reinforces for each that the more extreme view that they perceived is in fact the group norm, and they will be likely to change their beliefs to conform to it as in the cases above. As Hogg and Abrams (2001) assert, “In many intergroup contexts in-group norms are polarized away from the out-group so that they do not represent the central tendency of in-group positions, and the intragroup processes cause the in-group to become more extreme so that the gulf between in-group and out-group widens.” This is part of oppositional differentiation as well, if the out-group position is Z then ours must be A to establish clear boundaries, and insure that there is no mistaking us for out-group. Sunstein (2002) states, “It also matters whether people think of themselves, antecedently or otherwise, as part of a group, with a degree of solidarity. If they think of themselves in this way, group polarization is all the more likely, and it is likely too to be more extreme.”
While group polarization is important for politics and law, it is crucial for the process of preparing psychologically for war. We must take a very extreme view of the out-group to be able to feel good about doing horrible things to them, or even to do horrible things in a casual way, or as an undesirable duty. In fact, we define the out-group as not belonging to any of our higher-level in-groups. Group polarization is a central mechanism that brings us towards that extreme view.
DEFINING THE ENEMY AS OUT-GROUP TO ALL OF OUR IN-GROUPS
Most of us, with a few exceptions such as the Jainists and extreme animal rights group members, are willing to kill at least some other creatures and feel no particular guilt about doing so, or empathy towards the creature we kill. The ethnographic record has yet to document the band or tribal level culture that does not hunt, or to record hunters having moral misgivings about their actions. Even most vegetarians kill pest bugs like flies and mosquitos without any qualms. On the other end of the willingness-to-kill-spectrum, murder of in-group is universally condemned, except when the group approves of and demands it (Brown 1991). It is therefore essential that we exclude our enemies from every in-group commitment we have, including the in-group of humanity, in order to remain altruistic towards our in-groups psychologically. This exclusion takes the form of demonizing and dehumanizing our enemies, a phenomenon that is extremely well recognized and pervasive (Keen, 1986). By doing so, we place our enemies psychologically into the group of creatures that we kill, as pests or threats to ourselves, and out of the in-group that we do not kill. While comparisons of war to hunting are often made, desirable food animals are not commonly used for dehumanizing. Rather, snakes, rats, gorillas, diseases, and at best animals seen as unclean or having other negatives attributes, such as pigs, are used (Keen, 1986). We talk about the communist cancer, being busted by the pigs, and so on.
When we encounter a human universal, such as this one, we need to ask how it is adaptive; what are the fitness benefits? What is the utility? In order to have the ability to belong to multiple groups on various levels at the same time, and in order even to be able to bond in groups beyond the 150 or so individuals we can keep track of (Dunbar, 1996), we had to have a very broad definition of in-group. After bonding by kin and by people one knows, there seems to be no available concrete stopping point until we get to everyone, all humans, if we are to increase the size of in-group we can commit to. Of course, we maintain the ability to specifically designate in-groups and out-groups culturally, but the billions of people not specified in any way are effectively in-group, psychologically, when it comes to our altruism. This means that when a tsunami hits we send aid to that group of strangers we will never meet. It also means that in order to attack them as a group we have to clearly designate them out-group, as far out as possible. But to just do that it would seem that calling them a bunch of elk or trout would suffice; why do we instead go for such negative pest and threat animals? The only obvious greater utility is that doing so triggers our altruistic impulses to protect each other from snakes and rats and so on. Elk and trout are not threats. Dehumanization's utility is that it focuses our pro-social altruistic psychology on the task of killing the out-group, and frees us from the constraints of seeing them as part of the humanity level in-group as well. Recognizing this utility for dehumanization further supports the assertion that altruism is a central component of the psychology of war.
In the process of evolution it is extremely common for physical structures to be adapted for new uses. The same basic parts have become hands, flippers, wings and so on. There is no reason for us to expect the evolution of new behaviors to be any different. Why develop an entirely new psychological structure when an existing one can be only slightly adapted and work well? If war was selfish in motivation, or the act of “defectors,” there would be no need to exclude the enemy from one's in-groups, or to dehumanize them. It is only because war is an altruistic act, an act that is strongly pro-social towards the in-group, that the oppositional differentiation must occur. This can be further seen by examining the other major propaganda component of building a psychological willingness to go to war; defining the out-group as criminal defectors.
DEFINING THE ENEMY AS CRIMINAL DEFECTORS
We hardly need a lengthy series of citations to see that describing the out-group enemy in terms of criminal negative “defector” behaviors is common if not universal. The fiction of the Iraqis taking infants out of incubators in Kuwait in 1991 and leaving them to die is just one recent example. The long history of group violence is filled with such stories, from Jews killing babies and drinking their blood to African Americans raping and killing white women. The long list of typical accusations includes lying, thieving, raping, murdering and more. Closely related to this is the common assertion that the out-group is insane, or that the leader is a "madman."
Again we must ask why this is needed and what impulse it is likely to trigger, and again we come to altruism. Such fictional accusations attempt to trigger the listeners altruistic impulse to enforce conformity and defend the in-group against “defectors.” Boehm (1999) establishes very well that, in band and tribal level societies, the in-group collectively, altruistically, and unanimously acts against those whose behavior has gone beyond what is tolerated. The short list of punishments goes quickly from loss of status to ostracism, and then to banishment, and to death. What the above false accusations would not be likely to trigger is any action from those who are selfish “defectors”; if the psychology of war was not altruistic in nature it would be hard to see what purpose such accusations might serve. Bloom (2000) also establishes that non-conformity frequently leads to cruel acts against the unfortunate in-group member, both in humans (teenage girls in camp being one example he cites; the behavior was so cruel in this instance that one of the researchers quit and vowed to never study teenage girls again), and in other social animals, among whom even disfigurements caused by injury can lead to ostracism or outright attacks. These attacks, no matter how cruel and how immoral they may appear, are altruistic enforcement of conformity. The bully or the teenage girl that torments the fat or studious or poorly dressed child may be altruistically enforcing conformity to group norms. By defining the enemy as criminal defectors, or even as ugly or stupid or otherwise unfit, we may employ this same altruistically based conformity enforcement predisposition for going to war.
As discussed above, we are all too easily taken down the path to seeing the out-group as baby killers, rapists, or worse.
THE PATH TO WAR
Let us briefly summarize our journey thus far. We start with an unrealistic self-image that, among other things, asserts our morality and essential goodness irregardless of the evidence to the contrary. Committing to, and identifying with, an in-group connects our self-image to our view of the in-group. We are good and they are similar to us, therefore they are good. We are the good guys. As part of the essential process of group delineation and differentiation, without which there is only a bunch of people and no in-group, there must be oppositional differentiation with at least one out-group. We must say not only who we are, but also who we are not. Our positive bias towards in-group can easily be confused by out-group as negative bias towards out-group, and our oppositional differentiation is very likely to generate real negative bias. If we are the good guys, then by the process of oppositional differentiation the out-group must be the bad guys. This oppositional differentiation becomes salient when conflict or perceived threat occurs between the groups. Whatever the facts of the conflict, we are likely to be guided by our positive bias towards in-group, and our belief that we are the good guys, to excuse in-group actions and blame out-group. Conformity and group polarization process spur negative bias to extremes. Blame for conflict can readily be translated into negative bias as well; it cannot be the in-group’s fault. Increased perceived threat increases conformity and oppositional differentiation, which in turn strengthens the group polarization process.
As Sherif demonstrated in the Robbers Cave experiments, this process can easily and quickly lead to intergroup violence, with both sides feeling justified in their actions and biases. As we saw with the Blues and Greens of Byzantium, there is no requirement for even a shred of rational basis to exist for an all out war to occur.
We have also established, both from an evolutionary perspective, and from analysis of common psychological preparations for war— dehumanization, demonization, and criminalization of the enemy out-group— that altruism is a central part of the psychology of war.
What we have not established is the path away from war, the path to peace. A full understanding of the former will lead to an equal understanding of the latter. This then is where the paths of group identity differentiation diverge and converge, on the way either to war or to peace.
THE PATH TO PEACE
We have established that humanity is blessed and cursed with a highly plastic group commitment ability. We can readily commit to an in-group with even the most absurd of excuses; some social psychologists have gone as far as using a coin toss with the study participants watching. We can readily commit to in-groups of any size, including nations of over a billion people, all of humanity, and the high school chess club we play for. The vast majority of in-groups we commit to in complex societies never go to war or commit acts of intergroup violence. The national in-group we commit to is at peace with the vast majority of other national groups the vast majority of the time. We are far from finding ourselves in a constant struggle of all against all. The important question is, what makes the difference?
First, oppositional differentiation only requires one out-group; all other groups may be neutral or simply categorized. Second, commitments to in-groups may be more or less salient to an individual, and this can be manipulated and culturally directed. Third, we are individually committed to many different in-groups, usually including the in-group of humanity with at least some salience.
The strength of our commitment to the in-group of humanity is the crucial factor. It is the point at which the altruistic path to genocide and suicide bombings converges and diverges from the altruistic path to peace and non-aggressive heroic altruism.
Commitment to humanity generates positive bias towards humanity, and connects the individual self-image, by the process of self-anchoring, to all of humanity. The individual comes to see the in-group of humanity as basically good no matter what the evidence. This is highly suggestive of the state of the Social Sciences for the last four or five decades, the denial of war by anthropologists and archeologists being one good example. Much more importantly, strong commitment at this highest possible level can act as a counter-balance to the group polarization and bias formation inherent in group differentiation. This is what creates pacifists, who are no longer able to see others as wholly out-group to an extent necessary for violence against them.
The difference between an individual on the psychological path to war and an individual on the path to peace is generated by the salience of their commitment to the in-group of humanity. The individual on the path to war believes in inherent differences between human groups; they have not strongly self-anchored to the group of humanity and are able to view out-group as the bad guys oppositionaly. Hong et al. (2003) states, “Social identification effects are attenuated when people hold a malleable view of human character and thus do not view social identities as fixed, concrete entities.”
The individual on the path to peace and non-aggressive heroic altruism sees no inherent differences between human groups; they are strongly self-anchored and committed to the group of humanity, and have perhaps an unrealistic and biased view of the in-group of humanity as basically good.
One bit of evidence supporting this view of heroic altruists is the ubiquitous answer that one gets when asking them why they, for example, ran into the burning building to save its inhabitants. The standard response is, “I just did what anyone else would do.” Yet, if it were truly “what anyone else would do” then the question would not be asked, as it is, quite often when a similar action is taken. The answer shows us two things. First, the individual in question is committed to the in-group of humanity. They do not say, “I just did what any Christian would do,” or “I just did what anyone except the Baathists would do.” The belief that anyone would do it shows that at the level of humanity they include all groups as in-group. Second, it shows that they have an unrealistic positive bias towards humanity, a bias that is evidence of commitment and self-anchoring.
If any of the above is correct, then peace is indeed possible. If it is correct then relatively small cultural changes can intentionally direct us towards peace and positive heroic altruism. At the very least we can readily see that our plastic group commitment abilities can be both positive and negative. In terms of sheer numbers, more people are peacefully committed to the in-group of China today than were alive in the world only a short time ago. This suggests strongly that it should be theoretically possible to peacefully commit all of humanity to the in-group of humanity. The biggest theoretical hurdle might be the need for an oppositional out-group for the in-group of humanity. However, it is possible to unite groups in conflict in order to achieve mutual goals. Humanity seems to have enough mutual goals, such as taking action to slow Global Warming, that this course of action might be possible.
Equally important for peace may be keeping the implications of these assertions in mind when conflicts arise. The conscious mind can overrule the subconscious. As Mendelberg showed in her book The Race Card (2001), once an implicit racist message is revealed, and thereby becomes explicit, our receptivity to it reverses. If we make the processes of polarization, dehumanization, demonization, and criminalization that lead us to war explicit we may be able to inoculate ourselves against them. Given that they seem to be ubiquitous, it should not be hard to find them when there is a move toward war.
If we keep in mind the universality of positive in-group bias we may avoid reactively starting down the path of oppositional negative bias ourselves. If we can view others’ actions with an understanding of evolved human group behavior predispositions, then we can avoid ascribing inherent negatives to their group. Finally, if we can converge on an understanding that humanity has no inherent differences, then there may be peace.
Asch, S.E. (1951). “Effects of group pressure upon the modification and distortion of judgments.” In H. Guetzkow (ed.) Groups, leadership, and men. Pittsburgh, PA: Carnegie Press.
Axelrod, Robert. (1984). The evolution of cooperation. New York: Basic Books.
Barna Research Group (2003). “Americans describe their views about life after death.” http://www.barna.org/cgi-bin/PagePressRelease.asp?PressReleaseID=150&Reference=B
Baron, Robert S., Kerr, Norbert L. (2003). Group process, group decision, group action - 2nd ed. Buckingham; Philadelphia PA: Open University Press.
Bar-Tal, Daniel. (1990). Group beliefs: a conception for analyzing group structure, process, and behavior. New York: Springer-Verlag.
Batson, C. Daniel. (1991). The altruism question: toward a social psychological answer. Hillsdale NJ: Lawrence Earlbaum Associates.
Baumeister, Roy F. (1999). Evil: inside human violence and cruelty. New York: Henry Holt.
Bloom, Howard K. (2000). The Global Brain: the evolution of the mass mind from the big bang to the 21st century. New York: Wiley.
Boehm, Christopher. (1999). Hierarchy in the forest: the evolution of egalitarian behavior. Cambridge Mass. ; London : Harvard University Press.
Brewer, Marilynn B. (1996) “In-group favoritism: The subtle side of intergroup discrimination.” In David M. Messick and Ann E. Tenbrunsel (eds.), Codes of conduct: behavioral research into business ethics pp. 160-170. New York: Russell Sage Foundation.
Brown, Donald E. (1991). Human Universals. Philadelphia: Temple University Press.
Cadinu, Maria Rosaria and Rothbart, Myron. (1996). “Self-anchoring and differentiation processes in the minimal group setting.” Journal of Personality and Social Psychology. vol.70(4) April 1996 pp. 661-677.
Cohen, Geoffrey L. (2003). “Party over policy: the dominating impact of group influence over political beliefs.” Journal of Personality and Social Psychology. vol. 85(5) November 2003 pp. 808-822.
Darley, John M. (1996). “How organizations socialize individuals into evildoing” In David M. Messick and Ann E. Tenbrunsel (eds.), Codes of conduct: behavioral research into business ethics pp. 13-43. New York: Russell Sage Foundation.
Darwin, C. (1871). The descent of man and selection in relation to sex. London: Murray.
de Quervain, Dominique J.-F., Fischbacher, U., Treyer, V., Schellhammer, M., Schnyder, U., Buck, A., Fehr, E. (2004). "The Neural Basis of Altruistic Punishment." Science 305: 1254-1258
De Vries, Reinout E. (2003), “Self, in-group, and out-group evaluation: bond or breach?” European Journal of Social Psychology. 2003, 33, 609-621.
Dunbar, Robin I. M. (1996). Grooming, gossip, and the evolution of language. London: Faber and Faber.
Eisenberg, Naomi I., Lieberman, Matthew D., Williams, Kipling D. (2003). “Feeling the pain of social loss.” Science, October 10 2003, 302: 237-239.
Field, Alexander J. (2001). Altruistically inclined? : the behavioral sciences, evolutionary theory, and the origins of reciprocity. Ann Arbor MI: The University of Michigan Press.
Fishbein, Harold D. and Dess, Nancy K. (2003). “An evolutionary perspective on intercultural conflict: basic mechanisms and implications for immigration policy.” In Richard W. Bloom and Nancy Kimberly Dess (eds.) Evolutionary psychology and violence: a primer for policymakers and public policy advocates. Westport, CT ; London : Praeger.
Fowler, James H. (2005) "Altruistic punishment and the origin of cooperation." PNAS 102: 7047-7049
Funk and Wagnalls Standard Dictionary, 2nd ed. (1993). New York: HarperCollins.
Gilovich, Thomas. (1991). How we know what isn’t so: the fallibility of human reason in everyday life. New York: Free Press.
Gintis, Herbert. (2003). “The hitchhikers guide to altruism: gene-culture coevolution and the internalization of norms.” Journal of Theoretical Biology 220,4 (2003): 407-418.
Goldstein, Joshua S. (2001). War and gender: how gender shapes the war system and vice versa. Cambridge, United Kingdom: Cambridge University Press.
Hamilton, W. D. (1963). “The evolution of altruistic behavior.” American Naturalist 97: 354-356.
Hogg, M.A., and Abrams, D. (2001). Intergroup relations. Philadelphia PA: Psychology Press.
Hong, Ying-yi., Chan, Gloria., Chiu, Chi-yue., Wong, Rosanna Y. M., Hansen, Ian G., Lee, Sau-lai., Tong, Yuk-yue., Fu, Ho-ying. (2003) “How are social identities linked to self-conception and intergroup orientation?: the moderating effect of implicit theories.” Journal of Personality and Social Psychology. vol. 85 (6) December 2003, pp. 1147-1160.
Keen, Sam. (1986). Faces of the enemy: reflections of the hostile imagination. New York: Harper & Row.
Keysers, Christian., Wicker, Bruno., Plailly, Jane., Royet, Jean-Pierre., Gallese, Vittorrio., and Rizzolatti, Giacomo. (2003). “Both of us disgusted in my insula: the common neural basis of seeing and feeling disgusted.” Neuron, vol. 40, 655-664, 30 October 2003.
Laland, Kevin N. and Brown, Gillian R. (2002).Sense and nonsense: evolutionary perspectives on human behavior. Oxford: Oxford University Press.
LeBlanc, Steven A. and Register, Katherine E. (2003). Constant battles: the myth of the peaceful, noble savage. New York: St. Martins Press.
Lifton, Robert Jay. (1986). The nazi doctors: medical killing and the psychology of genocide. New York: Basic Books.
Ludwig, Arnold M. (2002). King of the mountain: the nature of political leadership. Lexington KY: The University Press of Kentucky.
Manson, J.H. and Wrangham, R. (1991). "Intergroup aggression in chimpanzees and humans." Current Anthropology, 32: 369-390.
Mansbridge, Jane J. (1990) Beyond self-interest. Chicago: University of Chicago Press.
Maynard-Smith, John. (1964). “Group selection and kin selection.” Nature 201:1145-47.
Mele, Alfred R. (1997). “Real self-deception.” Behavioral and Brain Sciences. (1997) 20, 91-136.
Mendelberg, Tali. (2001), The race card: campaign strategy, implicit messages, and the norm of equality. Princeton NJ; Oxford: Princeton University Press.
Messick, David M. and Tenbrunsel, Ann E. (1996). Codes of conduct: behavioral research into business ethics. New York: Russell Sage Foundation.
Mesterton-Gibbons, Michael and Dugatkin, Lee Alan. (1992). “Cooperation among unrelated individuals: evolutionary factors.” The Quarterly Review of Biology. Vol. 67, 3, (Sep., 1992) pp.267-281.
Nesse, Randolph M. (2001) “Natural selection and the capacity for subjective commitment.” In Nesse, Randolph M. (ed.) Evolution and the capacity for commitment, pp. 1-36. New York: Russell Sage Foundation.
Patton, John Q. (2000). “Reciprocal altruism and warfare: a case from the Ecuadorian Amazon.” In Lee Cronk, Napoleon Chagnon, and William Irons (eds.), Adaptation and human behavior: an anthropological perspective. pp.417-436. Hawthorne NY: Aldine De Gruyter.
Poundstone, William. (1992). Prisoners Dilemma. New York: Doubleday.
Price, Michael E., Cosmides, L., Tooby, J. "Punitive sentiment as an anti-free rider psychological device." Evolution and Human Behavior (2002) 23, 203-231
Procopius, History of the wars, I, xxiv, translated by H.B. Dewing (New York: Macmillan, 1914), slightly abridged and reprinted in Leon Barnard and Theodore B. Hodges, Readings in European history, (New York: Macmillan, 1958).
Reynolds, Vernon., Falger, V.S.E., and Vine, Ian. (1987) The sociobiology of ethnocentrism: evolutionary dimension of xenophobia, discrimination, racism and nationalism. London: Croom Helms.
Richerson, Peter J. and Boyd, Robert. (2001). “The evolution of subjective commitment to groups: a tribal instinct hypothesis.” In Randolph M. Nesse (ed.), Evolution and the capacity for commitment. pp.186-220. New York: Russell Sage Foundation.
Sherif, M., Harvey, O.J., White, B.J., Hood, W.R., and Sherif, C.W. (1961). Intergroup conflict and cooperation: the robbers-cave experiment. Norman, OK: University of Oklahoma Book Exchange.
Sober, Elliott, and Wilson, David Sloan. (1998). Unto others: the evolution and psychology of unselfish behavior. Cambridge Mass. ; London : Harvard University Press.
Staub, Ervin. (2005). "The origins and evolution of hate, with notes on prevention." In Sternberg, Robert J. (ed), The Psychology of Hate. Washington, DC: American Psychological Association.
Sunstein, Cass R. (2002). “The law of group polarization.” Journal of Political Philosophy. vol.10 (2) pp. 175-195 2002.
Sunzi, bing fa. (6th cent. B.C.) The art of war. In Caleb Carr (ed.) (2000). The book of war. New York: Modern Library.
Tajfel, Henri. (1978). Differentiation between social groups: studies in the social psychology of intergroup relations. London; New York: Academic Press.
Taylor, Shelley E. and Brown, Jonathan D. (1988). “Illusion and well-being: a social psychological perspective on mental health.” Psychological Bulletin. vol. 103 (2) March 1988, pp. 193-210.
Tooby, J., and L. Cosmides. (1992). “The psychological foundations of culture.” In J. H. Barkow, L. Cosmides, and J. Tooby (eds.), The adapted mind: evolutionary psychology and the generation of culture, pp. 19-136. New York: Oxford University Press.
Trivers, R. L. (1971). “The evolution of reciprocal altruism.” Quarterly Review of Biology 46: 35-57.
Trivers, R. L. (1985). Social Evolution. Menlo Park, CA: The Benjamin/Cummings Publishing Company, Inc.
Trivers, R.L. (2002). Natural Selection and Social Theory. New York: Oxford University Press.
Turner, J.C., Sachdev, I., and Hogg, M. A. (1983). “Social categorization, interpersonal attraction and group formation.” British Journal of Social Psychology, 22, pp. 227-239.
Wade, Michael J. (1978). “A critical review of models of group selection.” Quarterly Review of Biology 53: 101-14.
Waller, Michael J. C. (1999). "Group selection and the selfish gene: the units of selection problem revisited.” In Johan M.G van der Dennen, David Smillie, and Daniel R. Wilson (eds.), The Darwinian heritage and sociobiology pp. 3-16. Westport CO; London: Praeger.
Weber, J. G. (1994). “The nature of ethnocentric attribution bias: in group protection or enhancement?” Journal of Experimental Social Psychology 30: 482-504.
Wilder, D. A. (1978). “Perceiving persons as a group: effects on attribution of causality and beliefs.” Social Psychology, 41, 13-23.
Wilson, Michael L., Hauser, Marc D. and Wrangham, Richard W. (2001). “Does participation in intergroup conflict depend on numerical assessment, range location, or rank for wild chimpanzees?” Animal Behavior, 2001, 61, pp. 1203-16.
Wilson, Michael L. and Wrangham, Richard W. (2003). “Intergroup relations in chimpanzees.” Annual Review of Anthropology Oct. 2003, Vol. 32, pp. 363-392.
Wright, Robert. (1994). The moral animal: the new science of evolutionary psychology. New York: Random House.
Wright, Sewall. (1945). “Tempo and mode in evolution: a critical review.” Ecology 26:415-19.