Chapter 4. Socially-Sanctioned Group Violence: Features, Examples, and Sources

4.1 Introduction

In discussing socially-sanctioned violence, it is useful to begin with a brief recapitulation of certain points from Chapter 3. There I observe that there are two key differences between the forms of individual and group violence which are illegal and morally wrong in a given society, and the forms which, though considered wrong in the abstract, are legal and morally condoned in specified contexts because they are viewed as the lesser among evils or ‘necessary’ (that is, useful) evils. These differences concern the proportion of a society implicated in a given practice of violence, and the role of learned norms in motivating the individual’s participation in the action.

In virtually all societies, violent acts which go against social norms—especially potentially lethal types of violence—are conducted by a very small minority of the population; in contrast, socially-sanctioned forms of group violence are performed, actively supported, or at least tolerated by the great majority of people.

The difference in participation is mirrored in differences in motivation: Individual and small group forms of violence which go against norms are conducted in spite of or in defiance of social norms. Such counter-norm violence is likely to occur as a product of: (1) emotionally- or physically-induced lowering of learned inhibitions against violence in a situation where the perpetrator is provoked (crimes of passion); (2) a psychopathic personality developed during childhood as a result of physical or social problems or deficiencies; or (3) the longer- or shorter-term adoption of the norms of a subculture of violence, such as those fostered by youth gangs, the drug subculture, or organized crime.

In contrast, individual participation in socially-sanctioned forms of group violence is conducted in compliance with social norms. The motivation for violent acts of this kind lies in feelings and ideas that are very nearly the opposite of those associated with acts of socially-banned violence. In acts of socially-sanctioned group violence, the individual maintains control of his or her actions, including the deliberate conduct of violence, in order to meet society’s expectations. Each person takes part either as an active participant or as a non-interfering observer out of a combination of mutually-reinforcing motives: obedience to authority, desire for social approval and acceptance, fear of social disapproval and punishment, and the internalization of the socially-taught moral views that support the action. In this respect, the motivation for participating in socially-sanctioned violence is identical to the motivation for engaging in the entire range of socially-accepted, non-destructive behavior which predominates in any viable society.

In addition, two other factors typically facilitate participation in socially-sanctioned forms of group violence and may also play a role in the conduct of some forms of socially-condemned violence: these are the routinization of the activity, which obliterates the individual’s sense of initiative and responsibility; and the cultural dehumanization of the intended victim in a variety of ways before, during, and after the violence. The impact of both factors is to reduce sympathetic or empathic feelings for the victim, and the feelings of guilt and remorse normally associated with the action.

Examples of socially-sanctioned group violence: The following sections of this chapter review some widely-practiced forms of socially-sanctioned group violence, with a view to illustrating the special motivations and social organization that characterize this form of behavior. This brief survey has several purposes:

First, it underscores that war is by no means unique as a sanctioned form of violence practiced around the world for centuries or millennia. The existence of other widespread forms of institutionalized violence may not seem supportive of the thesis that war is amenable to abolition; but the fact that virtually all of the other practices have died out does lend support to the thesis.

The second purpose is to derive from the now-abandoned forms of socially-sanctioned group violence a means of artificially distancing ourselves from the practice of war. Televised news makes war appear to be an unending and omnipresent aspect of the human condition. But when we look back at earlier, now-abandoned practices of socially-sanctioned violence, we are reminded that at the time when they were legitimate, they, too, seemed to be a lamentably unavoidable part of the human condition. This suggests a useful analogy: Contrasting modern views with contemporaneous views of now-abandoned practices suggests a way to think about war as if it had been—and, thus, could be—abolished. The general idea is to focus not on the rationales currently associated with war, but on the features that war shares with now-abolished forms of violence: the inhumanity, the barbarity, the indiscriminate arrogation by society of the right to destruction of individual lives. When we think back to earlier forms of violence, the rationales that legitimized them at the time fade into the background; what remains in the foreground is the monstrosity of the behavior. If we think about war from a parallel future vantage point, as though looking back at an activity that had been abolished, we would dissociate the barbarism of this instrument of power from the ends to which it is put. In other words, our moral assessment of the ends—whether achievable by other means, or abandoned along with war—would no longer have a bearing on our moral assessment of the means.

Third, surveying the justifying contexts and reasons for earlier practices of socially-sanctioned violence helps to clarify the relationship between utility and morality in this realm of human behavior. In the case of older forms of socially-sanctioned violence, an argument was made that on balance, despite the harm to the victim (if harm was acknowledged), society was better off with the practice than without; but in each case, this moral assessment was a function of a more general set of values embedded the society’s culture and world view. This suggests that a form of violence which is condoned in one society may not be condoned in another not because the link between cause (the violent practice) and effect (the social good) is no longer believed to be valid, but because the effect which is valued in the first society has become irrelevant, or a matter of indifference, or a social bad in the second.

Finally, the review of now-abandoned practices suggests a useful metaphorical handle on the role of moral beliefs in the conduct of war. Generally speaking (across time and culture), learned moral beliefs reinforce the developmentally acquired, internalized ability not to express hostility in the form of violent action: socially-taught, articulated moral beliefs strengthen the inarticulate acquired barrier to (or control over) violent behavior. When a given form of violence is socially sanctioned, however, social norms lower the conscious part of the internal barrier to violence. As long as the practice is legitimate, the role of moral beliefs is not only (and perhaps not mainly) to motivate the action, but to permit it, that is, to help overcome inarticulate, internalized inhibitions against violence. Once a practice is banned, moral beliefs spring back to their more usual position of reinforcing the individual’s self-control and inhibitions against violence.

Defining socially-sanctioned group violence: In surveying socially-sanctioned forms of violence, I have used a broad definition of ‘violence.’ The definition is not so broad as to include Johan Galtung’s ‘structural violence’: that is, harms that are inflicted on individuals by virtue of an inequitable form of social organization, such as starvation, poverty, illiteracy, disease, high infant mortality, indignity, and lack of any influence on public policy.104 The exclusion of these forms of harm does not indicate that they are less harmful than active violence; in many cases, the opposite is certainly true. But the purpose here is to identify common features and sources of active forms of violence, and there is reason to believe that the features and sources of structural violence may diverge substantially.

With respect to physical violence, I have used a very broad definition, including three main variants:

Intentionally-caused (or permitted) physical hurt, physical injury, or death: Activities of this kind are assault, rape, and homicide, and include massive and particularly brutal forms of such acts, such as torture, genocide, or war.

The systematic, coercive deprivation of individual freedom of action with the threat of physical harm or death: Practices in this area are slavery and lesser but still severe forms of involuntary servitude, including most of those where the victim is called a serf, villein, bondman, or indentured servant.

Physical violation of the bodies of the dead: The main actions in this area involve various forms of cannibalism, and the use of human body parts as materials for worked goods, such as skulls for bowls, bones for needles or necklaces, skin for fabric, etc.

The inclusion of the second and third areas—slavery and the violation of dead bodies—stretches the definition of ‘violence’ slightly; but in all three areas, the practice involves active, physically aggressive acts against the body of another human being. Moral objections to activities in the three areas have shared features, and the rationalizing moral justifications for activities in the three areas overlap substantially in form and content. Thus, the moral positioning of socially-sanctioned ‘violence’ taken in the narrowest sense (the first area) is clarified by including in the set of surveyed socially-sanctioned practices those in which the violence is threatened (the second area) or symbolic (the third).

Given this broad approach to ‘violence,’ I define institutionalized socially-sanctioned forms of group violence, including war, as practices which have the following features in common:

Each involves the deliberate infliction, or the coercive threat of infliction, of bodily harm or physical violation on some humans by others.

Each is condoned by society as a whole and generally involves either action by individuals who represent the society as a whole, or publicly visible, legal action conducted by a large fraction of the population, with the support (or, at a minimum, non-interfering tolerance) of the rest.

In each case, members of the society generally view the activity itself as morally wrong (bad, unethical, abhorrent) outside the specific context in which it is socially approved or required.

The society explains and justifies the activity in specified context as a means to a greater good, or the lesser among evils.

There are five forms of socially-sanctioned group violence which share these characteristics and which have been practiced in virtually all parts of the world (that is, practiced extensively on every continent) over centuries or millennia. They are:

• Ritual cannibalism
• Ritual human sacrifice
• Slavery
• Physical punishment for violating law or custom (including execution, flogging, burning, drawing and quartering, branding, and the cutting off of body parts; and the violent treatment of suspected law-breakers, including interrogation with torture and trial by ordeal)
• War

In addition to these widely-practiced forms of institutionalized violence, there are many other practices of socially-sanctioned violence which have been less widely practiced. Most tend to be found only in certain parts of the world. Some were socially-sanctioned but involved individual rather than group practices: that is, they were conducted not by or on behalf of society as a whole, but by individuals on an ad hoc basis, often out of sight of the community, or by a cultural subgroup. In many cases, practices of this kind may have been tolerated by the society, but not strongly approved; and the moral justification may have rested more heavily on individual philosophy than public morality.

Organization of the chapter: In the next section of the chapter, I give a cursory overview of some practices of violence that were socially sanctioned but not widely practiced. These activities give a sense of the range and nature of the violent acts comprised by the abstract category of ‘socially-sanctioned, institutionalized violence’ and the importance of variable, culturally-derived world views and moral beliefs in accounting for these practices.

In the following section, I survey in somewhat more detail three of the practices which, like war, were independently developed and practiced in many different cultures in all parts of the world: corporal punishment for violations of law or custom, human sacrifice, and slavery. The fourth such practice, ritual cannibalism, is reviewed in more detail in Chapter 5.

The subsequent sections of the chapter uses these diverse examples as the basis for a speculative consideration of the processes which led to the start of various processes, and the process which led to their demise. The chapter concludes with a discussion of the relationship between efficacy and morality in the practice and abolition of various forms of socially-sanctioned violence.

4.2 Some Socially-Sanctioned Forms of Physical Violence and Violation

The varieties of human hurtfulness are mind-boggling. If we were able to identify all of them, the list of violent practices which were socially approved in one part of the world or another for some period of time would probably run to hundreds of activities, or perhaps thousands, depending on how fine-tuned the typology, and how small the group credited with ‘social approval.’ In identifying some of the myriad contexts and forms of violence, this brief survey makes no claim to offer a representative sample. The purpose here is merely to insure that the reader does not restrict the scope of socially-sanctioned violence to the five most widely practiced forms, which are the focus throughout most of this essay, four of which often or always involve lethal violence.

What leads to the extraordinary variety in the forms of socially-sanctioned violence is not so much the specific physical impact of an action on a (generally live) human body (though that, too, can be greatly varied), but the context and meaning of the action. Thus, this list groups the forms of socially-sanctioned violence by the context rather than the physical form of the action.

Political violence: Torture of prisoners conducted by individuals who represent a society (or a prevailing religion) has already been mentioned as a form of legal punishment. Torture has also been common as a means of interrogation or a means of persuading a victim to ‘confess’ to something or to do something he or she would not do otherwise. Terrorist attacks and political assassination are other forms of political violence which are sanctioned in some quarters.

Family and other community punishment: The deliberate physical punishment of some people by others, outside of the framework of court-imposed punishment and at the discretion of the perpetrators, has been legal and socially-condoned more often than not over the past 10,000 years. Violence by some members of families toward others—mainly by men toward their wives—has been sanctioned in many cultures for millennia. Dobash and Dobash (1979) document ancient practices in which wives were considered chattel, which husbands could treat as they liked, including killing. Whipping, ear-boxing, and spanking of children at home and at school has a long tradition in the Western world and in some other cultures. Some professions involve organizations which have the right to inflict physical punishment on their members. This is particularly true of military service and warrior societies. In religious organizations, bodily humiliation, mutilation, or flagellation may be approved or required.

Birth control: Infanticide has been widely practiced as a form of birth control. In very early times and among cultures operating at the subsistence level, infanticide of new-born children of both sexes was used to keep the population constant. Subsequently and even today, population statistics show a disproportionately large number of boys as a result of the killing of new-born girls in countries where male offspring bring prestige or are valued as warriors or as sources of support for their parents in old age.

Rituals relating to puberty, adulthood, and marriage: Scarification, genital mutilation, and other ritual forms of cutting, mutilating, or distorting the natural shape of the body are common. Well-known examples include the binding of women’s feet in China, elongated ears or necks in parts of Africa, and genital mutilation in various countries. (Related bodily but non-violent symbols of men’s control over women include the requirement for full body covering among Muslim women, and the orthodox Jewish practice of having women shave their heads and wear wigs, so that they will be unattractive to men other than their husbands.)

Mortuary rituals (or lack thereof): In some simple societies, the bodies of the dead are treated in ways that people in most other cultures would consider a violation and degradation of the deceased. Shocking treatment includes not only the forms of ritual cannibalism discussed in Chapter 5, but also cases of double burials, in which bones are dug up after the flesh has decayed, and put through various secondary processes; and cases in which bodies are left in the open in or near a village to decay or be consumed.

Trophies and soul appropriation: Many of the cultures which practice ritual cannibalism or human sacrifice have other, related practices which involve some form of violation or degradation of the body (and, more to the point from their perspective, the soul) of others. Common practices are scalping, head-hunting, and head-shrinking. Less common are practices which involve the use of human bones as the raw material for tools, weapons, and utensils.

Practices in medicine and related fields: The histories of medicine, alchemy, magic, shamanism, sorcery, and voodoo are replete with examples of recipes and procedures that involve parts of human bodies, particularly but by no means exclusively blood, hair, nails, and ground bones. In addition, the practice of medicine has included more direct forms of physical ‘violence or violation,’ the most common of which are surgery, blood-letting, and enemas. In some cultures, seriously ill patients were killed or expected to commit suicide. In others, those considered physically undesirable were killed, including malformed children and women who had miscarriages.

Sports and entertainment: Despite the padded gloves, the objective of boxing is to physically hurt the loser and, if possible, beat him into losing consciousness. This is a milder version of blood sports conducted for entertainment, which include the fights to the death among Roman gladiators, playing a game like soccer with a human head in one South American tribe, and in medieval times the potentially lethal sport of jousting.

Preserving honor and conflict resolution: Many cultures have well-defined customs involving the use of violence, including lethal violence, as a means of preserving, restoring, or signaling honor, or responding to dishonor. These include blood feuds, a common form of ‘low-intensity’ war among tribes and chiefdoms in all parts of the world. A variant of this custom appeared in renaissance Europe in the form of the duel to the death as a response to an insult, which lasted for several centuries. The readiness of the Japanese samurai to take their own or others’ lives upon their masters’ orders is legendary.

4.3 Human Sacrifice, Slavery, and Corporal Punishment

It is, perhaps, surprising that there are as many as five distinct practices of socially-sanctioned group violence and violation which, as far as can be determined today, have arisen independently on every continent, and which in many cases have had substantially similar morally justifying frameworks. This convergence would be less surprising if the practices were universal, but that is not the case. None of the practices has existed at all stages of political and economic development; each is found predominately, though not exclusively, in cultures with a particular degree of scale and complexity. But only some cultures with the relevant degree of complexity and scale had a given practice: In other cultures with a similar political and economic structure, the practice was simply absent, or it was known to exist elsewhere but considered abhorrent and unthinkable.

The similarity yet diversity in actual practices of these common forms of socially-sanctioned group violence is consistent with the ‘multilinear’ view of cultural evolution developed in the 1950s by Julian Steward (1955), one of the pre-eminent American anthropologists of the century. The underlying concept is that human beings have certain tendencies or capacities, which are expressed in certain forms of social organization and social behavior, and which emerge in some cases but not others as a result of factors which may be too complex or subtle to identify. This perspective accounts for customs and patterns of behavior which develop independently around the world, and yet should not be viewed as necessarily embedded or pre-programmed in human nature. Human sacrifice, slavery, and corporal punishment for violating law or custom (along with ritual cannibalism, discussed in the next chapter) are such practices. Like them, I contend, war is common, yet not a feature of the human condition inextricably embedded in human nature; instead, like them, war is susceptible to being abolished. To support the case that these are, in fact, practices comparable to war, the following discussion underscores the features which they share with war: their prevalence as confirmed social institutions; their inhuman cruelty; the morally-based rationales and affect, which recognize but excuse the cruelty involved; and the importance of the stakes perceived to be at issue in the societies concerned.

Human Sacrifice

Broadly speaking, human sacrifice takes two main forms. The main meaning of the term refers to a religious ritual in which individuals are killed in order to communicate with or influence the behavior of a god or gods. In addition, the term is sometimes used to include death rituals in which family members, servants, and others are killed and buried with the head of a household or other highly-placed individual, generally in order to accompany and continue to serve that individual after death. Even though this second practice differs substantially in goal and tenor, I have included it under the general rubric of human sacrifice for convenience, distinguishing between the two by calling the first ‘religious sacrifice’ and the second ‘mortuary sacrifice.’

The sacrifice of human victims occurs mainly in agricultural societies (that is, not in hunter-gatherer or industrial societies). The practice is situated at a particular juncture in political-social development, with respect to both the form of social organization and the length of the period during which the practice is prevalent. There is a far larger group of customs of ‘religious sacrifice’ which do not involve human victims. Such customs predominate in the religious practices of both more and less developed societies. Among some simple bands and tribes which have a custom that can be considered a religious sacrifice (or analogous to religious sacrifice), the victim is the totem animal of the group. Among agricultural societies with a long history of religious sacrifice, early practices involving human victims typically gave way to animal sacrifices, then vegetable products (representing the harvest), and, finally, strictly symbolic offerings, for example, the wine and bread use in the Christian sacrament of the Eucharist.

The practice of human sacrifice by early agricultural societies existed in all parts of the world where large agricultural societies were found: in China and Japan;105 in India and across the Indian subcontinent;106 in the Middle East and throughout the circum-Mediterranean area from Carthage (Tunisia) to Greece and, in its earliest days, Rome;107 in the Mayan, Aztec, and Inca empires of Central and South America;108 and in parts of Africa and North America.109

In many cases, the purpose of human sacrifice was to persuade the gods responsible for fertility (of land, animals, or people) to provide bountiful crops, healthy babies, and productive livestock. Sometimes the sacrifice ritual had aspects associated with the fertility of the land. For example, the blood of the victim might be poured into the earth (Egypt), or the victim might be cut up and the body parts distributed around the ground (India).110

The most vivid examples of the killing of sacrificial human victims in a ritual that was a public occasion are those which survived into the modern period, such by the Aztecs in Mexico and the Khonds in India, and those for which detailed accounts or well preserved archaeological remains exist, such as for the Mayans and the Incas. The Aztecs and Mayans stabbed human victims who were laid across an altar, in rituals designed to maintain the order of time, the sun, and the seasons. The Khonds had a three-day ritual once or twice a year, at the end of which the victim was in some cases stupefied or dead; then the flesh was cut from the bone in small pieces, one for each household, which buried their piece in their plot. The victims, called Meriahs, were bought or raised for the purpose, kept in enclaves and fed and clothed until one or another was chosen to be sacrificed. Among the Khonds and the Aztecs, as well as other societies with this direct type of human sacrifice, the victim, once selected, was believed to take on the qualities of the god to which the sacrifice would be made. This made the victim a suitable means of communicating between man and god; and it generally made the victim holy, sacred, and powerful. Where victims were chosen from among the people, rather than taken from among captured prisoners of war or kidnapped from another tribe, they were sometimes honored and feted before the event, and given their choice of all of life’s pleasures.

Even when the victim was a stranger and when the rite involved both celebration and great cruelty, a sense of evil-doing might still pervade the proceedings. For example, the following words, uttered by the sacrificing priest, were recorded as opening part the Khond ritual:

We obeyed the Goddess and assembled the people. Then the victim child wept, and reviled, and uttered curses. All the people rejoiced except those with whom the child had dwelt and the Jani. They were overwhelmed with grief; their sorrows prevailed entirely over their expectations of benefit, and they did not give either their minds or their faith to the Gods.... Oh Deity! why have you instituted this miserable heart-rending rite? The Earth Goddess told the Jani to reply to the victim: “Blame not us, blame your parents who sold you, what fault is ours? The Earth Goddess demanded a sacrifice; it is necessary for the world. If the tiger begins to rage, the snake to poison, fevers and every pain afflict the people, a sacrifice becomes necessary.”111

In many other cases, the relationship between the religious sacrifice and the well-being of the community was indirect. Special sacrifices might be offered during natural catastrophes, such as drought, flood, or pestilence, to quell the presumed anger of the gods. The Carthaginians and extended settlements of Phoenicians around the Mediterranean had a long-lasting practice of sacrificing children by burning them on an altar during times of crisis, including losses in war as well as natural catastrophes. For example, in 310 BC, they conducted a massive sacrifice of 500 sons of nobleman, who were burned in a large pit, in an attempt to avert a defeat at Syracuse.112 Most references to human sacrifice in the Old Testament are diatribes against the practice of sacrificing children to Molek or Baal; but some references suggest that at certain periods the Israelites considered Yahweh and Molek identical, and also took part in the ritual.113

More generally, sacrifice was a means of maintaining a good relationship with the gods, expressing various feelings typical of a subordinate, such as gratitude, humility, or repentance for wrong-doing. One source, surveying the region from Egypt to the Indus valley during the first two millennia BC, identifies “propitiatory, dedicatory and expiatory rites involving children and adults on special occasions” as having been practiced through the region.114

Another version of a generalized effort to maintain good relations with gods was the practice of ‘foundation sacrifice,’ that is, burying a child or adult under the foundation of a house or public building. This practice was common throughout Europe, the Middle East, and Asia from ancient times and persisted in many places, against social norms, into the middle ages and even the early modem period. In the British Isles, the Druids supported the practice, and sacrificial victims have been found at Stonehenge, suggesting an origin fully independent of the practice in the Middle East and Asia.

Finally, a common form of sacrifice, conducted as part of a public ritual, was the killing and burial of family members and servants who accompanied an important person to the grave. The monumental examples of this practice are well known: the group burials with Kings in the pyramids in Egypt, and the massive underground burials of ancient Chinese rulers, in which clay or wooden figures were later substituted for live or dead human bodies. Less well known practices were widespread, going back not just to the early city states, but to some of the oldest agricultural villages in southeast Europe (present day Yugoslavia, Turkey, Romania, and Bulgaria), where some graves include not just finery and utensils but wives or children. In some cases, the scale of sacrificial burials was massive. The ancient city-state of Ur has 16 tombs for royal burials dating from around 2300 BC. The number of retainers in these tombs varied from 12 to 80.115 In Japan in 2 BC, the entire personal retinue of Prince Yamatohiko, brother of the Emperor, was buried alive, and “for several days they died not, but wept and wailed at night.”116 Just as animals were substituted for humans in religious sacrifices, clay and straw human figures and symbols for humans were substituted for humans in later burials in Egypt, China, and Japan.

The practice by which the widow of a dead man was expected to go to her death, called suttee after the Indian word ‘sati,’ is noted in historical records from many parts of the world. It seems to have been occurred persistently, however, over a period of one to two millennia, mainly in China and India, where it was extolled by the predominant religions (Shintoism and Hinduism, respectively). In the case of China, the values associated with the practice were honoring the dead and showing self-discipline. India, where the wife was expected to die on the funeral pyre or on a later memorial pyre, the dead husband was believed to go to a god-like afterlife which the wife could share. In India, suttee was legally banned by the British in 1829, and it is believed to have ended on any significant scale by around 1860.117

Slavery and Related Conditions

Slavery: Slavery has been practiced more recently than any other common form of socially-sanctioned violence except war and capital punishment. While slavery is not, in itself, a form of physical violence, it creates a situation in which the victim is perpetually threatened with violent assault, up to and including death, and in which many individuals regularly suffered physical abuses. Many if not most slaves in 19th century America were beaten and raped (or forced into sexual relations); many were permanently separated from their families when they were sold; and they remained permanently under a penalty of death for attempting to escape to be reunited with their families or for any other reason.

The most well known practices of slavery are the enslavement of blacks from Africa in the United States in the 18th and 19th centuries, and the practice of slavery in ancient Greece and Rome. Each of these cases is considered special and atypical, in part because the practices in modern America and in the classical world differed substantially. But slavery has been practiced in all parts of the world at one time or another, continuously or intermittently throughout recorded history, and among some simple societies without written history; and in every case the quality of ownership is the same. The life of the slave belongs to the owner, who can take it without having committed a crime or having to account for his action.118

Among indigenous groups in the Western hemisphere, throughout North, Central, and South America, there was a form of slavery substantially different from that in most other parts of the world: Men of enemy tribes captured in war were considered to be slaves (possessions), but they were adopted, married into the tribe, and treated as a son or brother for several years. Then, at a certain time (whose identifying parameters are not known), they would be informed that the moment for the final treatment of captured male warriors had arrived. This treatment was to be put to a painful, torturous death and, often, subjected to ritual cannibalism. Women captured in war were generally used for slave labor, but might also marry into the tribe. They were not automatically slaughtered in a ritual sacrifice; but in the case of some North American tribes which practiced ritual human sacrifice to the gods of fertility (such as the Pawnee), female slaves from another tribe were the preferred victim. Because of the small size and simple, egalitarian organization of hunter-gatherer bands and tribes, those which made slaves of captured members of enemy tribes had only a few such individuals among them; and the slaves generally lived ordinary lives indistinguishable from those of others in the group up to the time of their deaths.119

Slaves were also the victim of choice for human sacrifice in agrarian societies and early states in parts of the Middle East and other areas.

Serfdom: In many places for centuries after slavery became illegal, conditions that resemble slavery were widely practiced.120 These are the forms of compulsory servitude identified by the following terms: coloni (in late Roman society), serf, villein, bondman, thrall, and, in more recent times, indentured servant. Most (but not all) people in these groups were farm hands; some were household servants or worked in factories or at trades. What distinguishes these conditions of servitude from slavery is that what is owned is not the body but some or all of the daily labor of the individual. The exact legal and social positions of the individuals called serf, villein, bondman, or thrall differed from one country to the next and over time. Serfs, bondmen, and thralls were never classified as ‘free men’: villeins seem to have moved over time between the free and unfree status in England and France. For convenience, the following discussion uses the term ‘serf’ to identify all of the ‘unfree’ persons who were not outright slaves.

Serfs were ‘owned’ in the sense that their own and their children’s labor was owned in perpetuity; they were identified in records as being owned and sold; and they represented an enumerated part of the inheritable holdings of an estate at death. In some cases the serf could purchase his own freedom (in rare cases where he could cobble together the means to do so). In most cases, however, buying freedom was not an option not only because the serf was too poor, but also because the serf was bound to the land and would not be released by the master even for payment unless a replacement were provided—and no willing replacements were to be found.

Serfs were probably subjected to the same kinds of punishment suffered by slaves for failure to meet their masters’ work requirements. A detailed multi-year daily diary of serf affairs for a mid-19th century Russian estate shows that three-quarters of the serfs were beaten, many several times, during the course of a typical year; and the average beating, conducted with a birch rod, involved 30 lashes.121 In addition, serfs were frequently struck or beaten by their superiors; rape and forced sexual relations were common; and in Russia serfs could be transported to Siberia for disobedience.122 In principle, the masters of serfs were not legally permitted to inflict random or arbitrary violence on them, nor to kill them, whereas masters had both rights with slaves. In practice, there was no court in which masters were held accountable for violence toward serfs, short of homicide, and they could even get away with murder. Thus, the physical abuse to which masters could subject serfs with impunity was a key difference between serfs and poor free sharecroppers.

There were feudal societies built on serf labor in the pre-Colombian American city states, in the ancient Middle East, and throughout the circum-Mediterranean area. In addition, the condition of serf or villein was common throughout the British Isles, Scandinavia, Europe, and Russia continuously or intermittently from around 1000 AD (when slavery left from late antiquity was dying out) to the mid-19th century.123

One remarkable feature of serfdom in Europe, which is instructive with respect to socially-sanctioned violence more generally, is that it did not develop in a straightforward evolutionary sequence, arising initially as a weakened form of slavery and then gradually disappearing. In every case which I have had an opportunity to review (England, France, Spain, Hungary, and Russia), serfdom represented not a rise from slavery, but a loss of freedom by farmers who had previously been free men, but who, suffering under harsh conditions of one kind or another, found themselves increasingly deprived of various rights and freedoms. In England this occurred between 600 and 900 AD when assaults and migrations from Scandinavia were leading to the deaths of farmers and the loss of their land, and the farmers chose to place themselves under the protection of feudal lords.

In France, Catalonia, and Hungary, the transition occurred after the onset of the Plague in the 14th century, when labor was scarce and large landholders became more demanding. In Russia, serfdom was introduced between 1600, when peasants had the legal right to move from estate to estate between harvests, and 1700, when that right was abolished and the peasants were legally bound to the land. On the part of the Czar, this was part of an on-going effort with several goals: to enforce tax law and increase tax income; to identify potential recruits for military service; and to improve agricultural productivity by establishing a system of mutual obligation between the landowners, who were bound to set aside registered lands for the private use of the serfs, and the serfs, who were bound to give a certain number of days’ labor per year to farming the landowners’ land. Very quickly, however, the main effect of the law was to bind the serfs to the landowners, who viewed themselves as owning the serfs and conducted business accordingly.

In nations where serfdom was practiced, serfs generally made up the great majority of the peasant population. The first Norman census of England, in 1086, for example, reckoning male heads of households, lists 283,000 individuals, of which two-thirds (195,000) were ‘villeins’ or ‘bordari’ (a related group). The other major groups were: slaves (25,000), major tenants who ran large estates (23,000), free men (generally small farmers who rented property) (12,000), the landowning principals and the nobility (9,000), and burghers and townspeople (8,000). The remaining 15,000 comprised craftsmen, small cottagers, religious and military professionals, and so on.124 In Russia at the time of the last census before the abolition of serfdom, the 10th national census, held in 1859, serfs represented 37.5 percent of the population of European Russian, 22.5 million out of a total of about 60 million.125 What these examples suggest is that even though in most nation-states the existence of a large class of slaves gradually disappeared after the fall of Rome, farm workers making up a large proportion of the population were subjected to treatment comparable to that of slaves for centuries.

Corporal Punishment

Over the range from simple to complex societies, the punishments meted out by representatives of a society for violations of law or custom show a clear trend, first toward increasingly severe forms of corporal punishment and then away from corporal punishment and toward incarceration in prison. Simple societies of hunter-gatherers, which were organized in mobile extended family groups of 25–50 persons, had no means of imprisoning individuals who violated the accepted rules of behavior, nor could they support an individual who was maimed in a way that interfered with productive work. Many such societies had ingenious means of conflict resolution, penance or expiation, and reconciliation. Where an offense was considered so grave as to warrant punishment, the options were generally limited to two forms: death or exile (which may have led to death). Since there were no courts or police, the aggrieved party and kin, or the group as a whole, were responsible for apprehending and killing the defendant. In some cases, where the violation was particularly severe and ritual cannibalism was known, the consumption of the dead body by the group might follow the execution with an intent such as preventing any possibility of afterlife or rebirth.

In all parts of the world, more complex societies introduced a variety of forms of physical punishment. Most often, there was an effort to make the punishment fit the crime. Rape would be punished by castration; theft by cutting off a hand or a foot; slander or the betrayal of an oath by cutting out the tongue. The use of horrible forms of torture to extract confessions is well known from the Spanish Inquisition and from the practices of 20th century dictators who tortured political opponents.

In agricultural societies with rudimentary systems of law and prior to the development of prisons, blood feuds of the same kind that occurred among the simplest bands were common. Where frequent, blood feuds were generally not illegal. In theory, retribution by an aggrieved person’s kin against the perpetrator of an offense was supposed to be proportionate to the crime and to put the matter to rest. In practice, however, there were self-interested agreements about both the nature of the crime and the nature of the response. This often led to perpetual and, in some cases, escalating acts of (generally lethal) violence by each group of kin against the other, each justified not by the original crime but by the lack of justice in the last preceding act of revenge or retribution.

In Anglo-Saxon England, the kings attempted to dampen the tendency to wanton violence in society and the blood feuds that followed by establishing an elaborate system of fines (‘wergeld’ or man-money) for every conceivable form of physical aggression, and for theft, adultery, and rape. Around the year 600, King Æthelbert issued 90 ‘laws,’ the vast majority of which simply list the fine for a meticulously described violent offense.126 Comparable laws issued by King Alfred three centuries later maintained an elaborate system of fines, but added two entirely new features: first, the laws cover many activities other than violent crime and theft, most of which involve some betrayal of trust or promise; and second, for many crimes, the fines indicated in the older law have been replaced with forms of corporal punishment.127 In one case, where the rule seems to come directly from the Old Testament (which is the case for several other passages), it is difficult to know whether a practiced punishment is being described, or rather a general principle of equity: “If any one thrust out another’s eye, let him give his own for it; tooth for tooth, hand for hand, burning for burning, stripe for stripe.” In another case, the specificity of the crime suggests an actual punishment: “If a male serf commits rape on a female serf, the fine is his testicles.” Several other situations of aggressive and presumably unwanted sexual advances are described, involving free men and single free women or slaves or nuns: In these cases, the penalty is a fine, or marriage or both (in the case of a free woman). Thus the main difference between the two sets of laws concerns crimes punishable by death. In the early set, there were none. In King Alfred’s laws, death is listed as the only penalty (fine is not an option) in the following cases: murder (except by a priest, in which case the penalty is losing all he has and been expelled from the ministry); stealing and selling a free man; cursing one’s parents; causing the death of a pregnant woman; killing a thief for breaking and entering in daylight, unless unavoidably compelled to do so; being a witch; making sacrifices to pagan gods (this may be another extract from the Old Testament); or plotting against the life of the king or the lord one is sworn to serve.

During the late Middle Ages and Renaissance, more elaborate and horrible corporal punishments were devised, such as death by burning (for women; men were hanged or had their head cut off), death by being drawn and quartered, and various forms of mutilation and branding.

The public spectacles associated with gruesome forms of capital punishment and mutilation in 16th-19th century France (and other parts of Europe) are described in detail by Foucault (1979). This is the period when deliberate cruelty and the excruciating, long-lasting pain of corporal punishment reached an apogee in the Western world. Foucault then analyses the process by which torturous forms of physical punishment and capital punishment were gradually abolished in this part of the world, and replaced with incarceration in prison.

Some forms of corporal punishment now banned as cruel and extreme in the West are still practiced elsewhere. Two examples reported in the news in recent years are the instance of flogging an American youth in Hong Kong (for vandalism, involving wantonly destroying the finish on a series of parked cars, along with some friends); and a case in Saudi Arabia of two members of the royal family (a man and a woman) being publicly beheaded for having committed adultery.

4.4 The Declining Tolerance for Violence

Over the last two centuries, the worldwide abolition of slavery and serfdom and the decline in physical violence as a form of legal punishment have been paralleled by two other developments. First, there has been a steady decline in social approval of various forms of violence committed by one individual against another. This is illustrated in the abolition of dueling, the outlawing of the beating of children by parents and teachers, the growing public opposition to wife-beating, and, in many countries, the banning of boxing. In Western nations, we are rapidly approaching the point at which no form of interpersonal violence is socially sanctioned.

Second, the purposes for which war is socially sanctioned have shrunk in a parallel manner, reflecting the same declining tolerance for violence. Consider, for example, the role of warfare in creating, extending, and maintaining empires. For centuries the use of armed force for this purpose was considered more or less legitimate, and service in armed forces dedicated to this end was an honorable profession. Since 1945, however, with Germany’s defeat and the end of colonization, the use of force by great powers to acquire and rule far-flung empires has, little by little, been abandoned; and the morally-justifying view that imperial administrations are good for the subject populations has been replaced by the view that all people have the right to self-determination.

In sum, despite the unprecedented rate of killing in war and genocide over the course of the 20th century, and the recently rising rates of violent crime in the wealthier nations (after at least a century of steady decline), there is a great deal of evidence that in Western culture and world culture over the past several hundred years, there has been a marked shrinking in the set of socially-sanctioned forms of individual and group violence; and over the past century, a parallel shrinking in the set of socially-accepted reasons for war.

4.5 The Rise of Institutionalized Forms of Violence

If mature, reasonably socialized adults in all cultures are predisposed to avoid violent behavior, then what is the source of culturally-shaped beliefs about just or acceptable forms of violence in which individuals may or must participate?

A substantial body of literature in political science and sociology argues that the organization of society itself is the source of socially-sanctioned violence. People band together, the argument goes, in order to form ‘attack’ or ‘defense’ units, that is, groupings which permit them to defend the territory and resources on which they rely for survival, or, alternatively, to attack others and seize the resources available to them.

The ethnographies of simple societies suggest, however, that this conception of the origin of society, and the origin of socially-sanctioned violence, is much too narrow. Judging by the ways of life of the small bands with simple cultures which survived into the past five centuries, the primary function of the most basic social groupings was not to permit group violence nor to defend against attack, but to increase the chance of survival in other ways: that is, to provide jointly enough food, shelter, and, when needed, surrogate parents to ensure that each generation would survive long enough to raise the next generation to child-bearing age. This survival function would have predisposed the earliest forms of social organization among hunter-gatherer societies, which occurred at a time when the average life spans was generally very short (about 30 years)128 to discourage internal violence and to protect against external violence—but not to originate violence. Even later, when larger, more settled societies arose around the practices of fishing, herding, and horticulture, the chief function of the increased scale of society was not attack or defense, but the division of labor and construction of settlements which produced more stored foodstuff, giving more people access to a more reliable food supply, and thereby supporting larger populations with longer life spans.

Violent attacks by marauders and by empire-builders, seeking to appropriate the wealth accumulated by others, arose alongside the more settled life style. But such predatory violence followed rather than caused the rise of settlements and accumulated wealth, and settled defenders always far outnumbered mobile predators. This means that the rise of a warrior class and organized warfare (as distinct from small-scale, haphazard theft and marauding), which was associated with the growth of agriculture and cities in all parts of the world, was less a cause of the development of early city states than a product of it.

Against this backdrop, I suggest, the earliest socially-sanctioned forms of group violence are likely to have been rooted in specific experiences in which particular forms of violence undertaken in an ad hoc manner became strongly associated with survival. The predominant case would be that of a band or tribe facing starvation which found that taking food from the territory of another group, or seizing food supplies stored by another group, represented a route to survival. Over time, repeated violation of earlier inhibitions against attacking others—lapses from ordinary adult discipline made under duress—could lead to the development of a more positive ethos of conquest and destruction. An evolution of this kind might account for development of the aggression-justifying belief systems which arose repeatedly among nomadic pastoralists in the steppes of Asia—first as long ago as 2000 BC, among the Indo-European groups that spread from the Pontic-Caspian region into Europe, the Middle East, and South Asia, displacing more complex, settled, peaceful agrarian societies in those regions;129 and then much later in the West, among the Goths (including but not limited to the Scythians, Vandals, and Huns) in the 5th century AD;130 and in the east, among the Mongols, led by Genghis Khan, in the 12th century AD.131 Given the millennia-long repetition of outward expansion and aggression, it seems likely, however, that the glorification of conquest and empire was not merely handed down from one generation to the next, but reappeared from time to time in response to the exigencies of a nomadic lifestyle in which a group of men on horseback could rapidly attack and plunder a farming community or, alternatively, be attacked and lose everything to another group on horseback.132

To take a very different example, early agricultural societies seem likely to have introduced human sacrifice in an ad hoc manner, when faced with exceptional adversity. In all parts of the world, the first sedentary societies probably used the quantities of storable produce generated by agriculture in an effort to smooth out their food supply over the course of the year. From time to time, however, this aim must have been frustrated by plagues, floods, and droughts, leading to widespread famine. Drawing a parallel with their own social and personal experience of stability and security as a function of food, the leaders of these societies conceived of a natural universe controlled by gods who needed food to stay in a good humor and keep the world stable and predictable. Scaling up from daily life, and attempting to trade one evil for another (the root concept of sacrifice), the political-religious leaders of the early societies came up with human flesh as the preferred food of the gods, or, in another manner of speaking, the sacrifice of a human being as a concession of ultimate power and authority to the gods—in either case, propitiating forces beyond their own control in the hope of being granted, in return, conditions for a good harvest during the coming year. Unlike warfare conducted for the purpose of seizing food, acts of human sacrifice generally did nothing to alleviate the crises that may have prompted the creation of this ritual. But like acts of war aimed at seizing food, acts of human sacrifice were justified on the grounds that the evil being done was the lesser of two evils in the situation.

Over the course of human history, there must have been innumerable cases in which people in a given culture experienced some link between own vulnerability to pain, loss, suffering, and death and protection from such ills with the deliberate infliction of pain or loss on others; and having made this link, inferred a more general moral rule from it, such as, ‘human sacrifice will lead to good crops’ or, ‘a life of military conquest will earn glory as well as wealth and security.’ Moreover, through cultural dispersal, normatively loaded beliefs about just or desirable forms of group violence are likely to have spread from one culture to others which imitated or adopted them because they offered new solutions to shared problems.

4.6 The Demise of Institutionalized Forms of Violence

If institutionalized, socially-sanctioned forms of group violence tend to appear initially as a by-product of crisis response, why and how do socially-sanctioned forms of group violence end?

What we know about the cessation of once-widespread practices that have been abolished—ritual cannibalism, ritual human sacrifice, and slavery—suggests two main mechanisms:

• A practice can be abolished through some form of coercion imposed from outside the culture. This happened in many parts of the world over the past several centuries, when conquerors and missionaries from Europe encountered what they considered abhorrent and unthinkable customs.
• A practice can fall into disuse as a result of changes in society which lead to changes in the moral beliefs that once justified it. This seems to have happened with human sacrifice in the circum-Mediterranean area in ancient times, and with ritual cannibalism throughout the world at various times.

In the latter case, within the context of a larger world view which itself is changing, the practice may be increasingly perceived as either unnecessary as a means of achieving a given end, or unnecessarily harmful in relation to the objective. In what follows, I discuss each of these processes briefly, and then look at their interactive functioning.

The US civil war is probably the most widely cited instance of the abolition of a form of socially-sanctioned group violence as a result coercion imposed from outside the society: people in the North who opposed slavery imposed their view on those in the South who thought it should be legal, and the Northerners made war to impose their view. The exploration and colonization of the world by Europeans over the past five centuries offers many other examples of new standards of behavior being abruptly imposed by outsiders, with the result that many practices were changed, abolished, or died out with the populations that had practiced them. Among other practices, this applies to ritual forms of cannibalism and human sacrifice practiced in many parts of the world. In the Americas and the Pacific, explorers, conquerors, and missionaries insisted that primitive tribes instantly cease such practices. In Africa, India and other parts of Asia, British and other European colonial rulers took a hands-off approach at first, but eventually banned various customs of human sacrifice, including suttee (the burning of a widow on her dead husband’s pyre) and fertility rites involving the sacrifice of human victims.

Direct and indirect evidence suggests that in the more distant past—from one millennium to several millennia ago—the practices of ritual human sacrifice and ritual cannibalism were once common, but that they ceased as a result of changes in economic and political organization. Among simple groups encountered in the modern period—bands or tribes with a basic hunter-gatherer economy, or tribes with the same supplemented by some systematic horticulture or herding—ritual cannibalism appears to have been surprisingly widespread, with an incidence of 50 percent or more. Ancient writing and some archaeological evidence suggest that in the parts of the world that developed first, when large, settled, complex agricultural societies arose in various regions, ritual cannibalism died out and ritual human sacrifice became common. For both ritual cannibalism and ritual human sacrifice, the practices, where they occurred, were associated with cultural beliefs that were common among all societies with similar economies, including societies that did not practice ritual cannibalism or ritual human sacrifice. And ritual human sacrifice was, for the most part, abandoned by cultures that had more developed economies, incorporating much more extensive reliance on money (legal tender) and on trade with other cultures. Another example of association between a given form of socially-sanctioned violence and a particular stage of development is offered by slavery which was banned when modern political and industrial growth made it no longer tolerable or profitable.

The extraordinarily close correlation between successive stages of political and economic development and the rise and demise of given forms of socially-sanctioned violence implies that the latter is a function of the former, unmediated by any significant contributing role of moral belief or world view. If this were the case, then we should infer that war is likely to end if and only if there is an appropriate change in worldwide political and economic development—and that moral beliefs about just war have, indeed, little bearing on the matter.

There is, however, a more plausible interpretation of the association between political and economic development and sanctioned forms of group violence. This is that increasing complexity and scale of political organization and increasing control over natural threats to physical survival are associated with world views, and changes in world view, which lend themselves to the moral justification of specific forms of sanctioned group violence. If this is the case, then we would expect to find what history shows: a given form of sanctioned violence is sometimes but not always associated with a given degree of political and economic complexity, but virtually never occurs in societies considerably more or less complex.

For both cannibalism and human sacrifice, there is evidence of the demise of the practice in given cultures not after a change in development, but while the culture was still operating at the level of development normally associated with each practice. This suggests that a change in world view and in moral beliefs about group violence may precede and help cause the change in political and economic development, rather than follow from it.

The same may be said, in a much weaker form, about the imposition of abolition from outside the practicing culture. Explorers, missionaries, and imperial conquerors typically imposed broader horizons on local cultures, requiring change not just in the practice of certain forms of socially-sanctioned violence, but in a web of interlocking behaviors, beliefs, and values. In many cases, forms of ritual cannibalism by simple tribes were abolished almost instantly at the demand of imperial administrators, who (not surprisingly) treated the practices with contempt and horror, as inhuman and intolerable. The rapid collapse of practices which (anthropologists found in retrospect) were integrally incorporated in the equivalent of religious-philosophical-political organization and belief systems suggests that when confronted with an utterly alien world view, the entire set of beliefs supporting the ritual practice simply caved in. These beliefs were not replaced by alternative beliefs or practices, designed to accomplish the same ends as ritual cannibalism. Instead, people’s lives were deprived of those ends and other social ends incompatible with those of the colonizing culture.

Regardless of the proximate cause of change—views imposed from outside, changing political and economic structure, or simply weakening adherence to old beliefs—the moral views that supported previously sanctioned forms of violence tended to seem solid as long as they went substantially unchallenged, yet ultimately vulnerable to rapid and comprehensive change. A metaphor for this vulnerability might be that of a thick-skinned balloon, which can withstand knocks and bounces and being seriously bent out of shape, but which will collapse instantly if pricked by a sharp object. The culturally-shaped moral beliefs that support socially-sanctioned forms of violence can be solid for centuries, yet collapse and vanish in a few years. I attribute this unusual combination of resilience and vulnerability to the fact that the justifications for sanctioned violence are always working against an underlying propensity to non-violence. In terms of resilience, this means that the barrier to change must be kept high; that is, the social investment in maintaining the moral reasoning must be substantial, and the social means of fending off challenges and defections extensive. But in terms of vulnerability, it means that a moral cognitive dissonance is always waiting, like a tectonic fault, for an opportunity to settle into a more stable, consistent configuration of beliefs with regard to violence.

4.7 Goals, Efficacy, and Morality in Institutionalized Violence and Violation

In looking back at older forms of socially-sanctioned violence, we tend to classify them according to contemporary world views and standards of socially-sanctioned violence. We tend to distinguish between rituals of violence and violation which, by our lights, were based on false ideas about the world—rituals involving superstition, shamanism, magic, or implausible religious beliefs—and violent practices of a kind that continue today, which may have had the desired physical impact on the prospects for survival, or the prospects for thriving, such as socially-sanctioned war to obtain food, or socially-sanctioned use of foreign populations as slave labor to increase the wealth of one’s own society. In other words, the modern mind gravitates toward a functional assessment of violent practices: did they serve a practical end? And if so, were they, perhaps, justified by objective interests rather than by variable, culturally-determined moral beliefs?

In this discussion, the efficacy of a sanctioned form of violence as a means to expected or desired ends is not at issue. What is at issue is the motivation for the action among those who do it: Is the practice driven or permitted by unique, culturally-shaped moral beliefs according to which the form of violence is just and acceptable? Or—and here we have the relevant alternative—is the violence motivated by human needs which are common across cultures, such as survival needs, which do not involve culturally-variable moral beliefs?

The way to identify a given act of violence as rooted in culturally-variable moral beliefs, on the one hand, or innate human needs, on the other hand, is by looking carefully at the claimed benefit of the action.

The issue is readily confused because most forms of socially-sanctioned violence are called, or can be called, ‘necessary evils.’ Taken out of context, this term suggests that the violent action is necessary for survival. For the most part, however, the term is meant to show a link between two seemingly contrary expectations: the expectation that a given action will harm or cause pain to one person or many people (thus, it is ‘evil’), and the expectation that some good will come out of the (thus, it is ‘necessary’ or, more precisely, useful). For example, a person who says ‘Spare the rod, spoil the child’ is saying that spanking is a necessary evil. Or, to take another case, people believe that punishment which hurts criminals is a ‘necessary evil’ because if inadequately punished, crimes will be more common.

The use of socially-sanctioned lethal violence is sometimes but not often a survival-related ‘necessary evil’ in the sense that if one person does not die, one or more others will die, with certainty and speed. Most sanctioned forms of violence, though called ‘necessary evils,’ are not actions which, if suspended, would result in the prompt and certain death of one or many individuals. (A form of lethal violence which is a ‘necessary evil’ in this sense is a sniper’s killing a crazed individual who cannot otherwise be stopped from killing people randomly with an automatic weapon.)

Most if not all socially-sanctioned forms of group violence are institutionalized: that is, the context and reasons are identified in advance, the procedures are specified, and the participants (the violent actors) are known and, in many cases, go through extensive training. But violence which is carefully planned and prescribed in advance of the event is rarely undertaken as a matter of survival, strictly and narrowly speaking. Moreover, in the rare cases where pre-planned violence might actually be used to ensure the survival of most members of a group, it is likely that the same energy, directed at developing a non-violent means of survival, could prevent the rise of the situation in which lethal violence was needed.

For the most part, the claimed benefit of institutionalized, socially-sanctioned group violence is a social good less urgently or comprehensively needed than survival—and, by the same token, more closely bound to a specific culture and its world view. In most cases, the physical survival of members of the society is taken for granted: what is at issue are the political, economic, or cultural conditions under which the group will live. The goals for which people in a given culture may be prepared to inflict or suffer lethal violence are many. They include political conditions like creating or preserving an empire, or achieving self-determination; acquiring, keeping, or losing wealth; the toleration or non-toleration of religious, cultural, or ethnic self-expression, and so on.

If we look back at roughly comparable societies, one of which sanctioned a given form of violence and the other did not, what we would find most often is not an association between violence and survival, but rather an association between violence and another socially-defined ‘good’ which a modern observer might or might not consider worth killing or dying for. Political scientists routinely attempt to assess war as an instrument of politics by assessing the outcome of wars in terms of costs, risks, and losses to the aggressor compared with gains to the aggressor, where the costs and gains are measured in terms of wealth, territory, population, natural resources, political influence, or other socially-defined ‘goods.’ The conclusion typically drawn is that war was a rational or wise choice if the gains outweighed the costs, or an irrational or poor choice in the opposite case. The implication is that the purpose and goals of war—and in those senses, its cause—was gain of the relevant kind.

What such analyses overlook is that war can be perceived and assessed not only as a more or less successful means to given ends, but also as a means to ends which a given society may consider unacceptable. This distinction is all-important in any case where war (or another form of group violence) is not being employed as a means to survival, in the most basic, literal, physical sense of the term. Only in cases where a society is facing mass starvation (or a comparable environmentally-linked catastrophe, like mass death by freezing or by drowning), or facing deliberately-inflicted genocide, and when violent action might literally save more lives than it would cost in the threatened society can it be argued that a socially-sanctioned form of violence may be amorally driven by objective ‘interests.’ In other words, only when many lives are immediately at stake not during the prosecution of violence but as a certain fact in the absence of violence can we infer that the violence may be entirely dissociated from a special, culturally-determined system of moral beliefs. Even in such situations—for example, in a society facing mass starvation—the view that one’s own survival is an adequate reason for killing another human may be morally intolerable; and in those cases, individuals may starve rather than kill, as a result of learned moral beliefs. Thus, in the final analysis, the defensive killing of attackers may be the only form of socially-sanctioned violence motivated by a practical end in which the reason for killing other human beings need not be supported by culturally-shaped moral beliefs, but could be the result of a culturally-invariant will to kill, if necessary, in order to defend oneself (or others) from being killed.