“The blood of martyrs is the seed of the church.” With that chilling dictum the third-century theologian Tertullian confessed the fundamental flaw of human altruism, an intimation that the purpose of sacrifice is to raise one human group over another. Generosity without hope of reciprocation is the rarest and most cherished of human behaviors, subtle and difficult to define, distributed in a highly selective pattern, surrounded by ritual and circumstance, and honored by medallions and emotional orations. We sanctify true altruism in order to reward it and by that means to promote its recurrence in others. Human altruism, in short, is riddled to its foundations with the expected mammalian ambivalence.

As mammals would be and ants would not, we are fascinated by the extreme forms of self-sacrifice. In the First and Second World Wars, Korea, and Vietnam, a large percentage of Congressional Medals of Honor were awarded to men who threw themselves on top of grenades to shield comrades, aided the rescue of others from battle sites at the cost of certain death to themselves, or made other extraordinary decisions that led to the same fatal end. Such altruistic suicide is the ultimate act of courage and emphatically deserves the country’s highest honor. But it is still a great puzzle.

What could possibly go on in the minds of these men in the moment of desperation? “Personal vanity and pride are always important factors in situations of this kind,” James Jones wrote in WWII, "and the sheer excitement of battle can often lead a man to death willingly, where without it he might have balked. But in the absolute, ultimate end, when your final extinction is right there only a few yards farther on staring back at you, there may be a sort of penultimate national, and social, and even racial, masochism— a sort of hotly joyous, almost-sexual enjoyment and acceptance— which keeps you going the last few steps. The ultimate luxury of just not giving a damn any more."

The annihilating mixture of reason and passion, which has been described often in firsthand accounts of the battlefield, is only the extreme phenomenon that lies beyond the innumerable smaller impulses of courage and generosity that bind societies together. One is tempted to leave the matter there, to accept the purest elements of altruism as simply the better side of human nature. Perhaps, to put the best possible construction on the matter, conscious altruism is a transcendental quality that distinguishes human beings from animals. But scientists are not accustomed to declaring any phenomenon off limits, and it is precisely through the deeper analysis of altruism that sociobiology seems best prepared at this time to make a novel contribution.

I doubt if any higher animal, such as an eagle or a lion, has ever deserved a Congressional Medal of Honor by the ennobling criteria used in our society. Yet minor altruism does occur frequently, in forms instantly understandable in human terms, and is bestowed not just on offspring but on other members of the species as well. Certain small birds, robins, thrushes and titmice, for example, warn others of the approach of a hawk. They crouch low and emit a distinctive thin, reedy whistle. Although the warning call has acoustic properties that make its source difficult to locate in space, to whistle at all seems at the very least unselfish; the caller would be wiser not to betray its presence but rather to remain silent.

Other than man, chimpanzees may be the most altruistic of all mammals. In addition to sharing meat after their cooperative hunts, they also practice adoption. Jane Goodall has observed three cases at the Gombe Stream National Park in Tanzania, all involving orphaned infants taken over by adult brothers and sisters. It is of considerable interest, for more theoretical reasons to be discussed shortly, that the altruistic behavior was displayed by the closest possible relatives rather than by experienced females with children of their own, females who might have supplied the orphans with milk and more adequate social protection.

In spite of a fair abundance of such examples among vertebrates, it is only in the lower animals, and in the social insects particularly, that we encounter altruistic suicide comparable to man’s. Many members of ant, bee, and wasp colonies are ready to defend their nests with insane charges against intruders. This is the reason that people move with circumspection around honeybee hives and yellow-jacket burrows, but can afford to relax near the nests of solitary species such as sweat bees and mud daubers…

Honeybee workers have stings lined with reversed barbs like those on fishhooks. When a bee attacks an intruder at the hive, the sting catches in the skin; as the bee moves away, the sting remains embedded, pulling out the entire venom gland and much of the viscera with it. The bee soon dies, but its attack has been more effective than if it withdrew the sting intact. The reason is that the venom gland continues to leak poison into the wound, while a banana-like odor emanating from the base of the sting incites other members of the hive to launch kamikaze attacks of their own at the same spot. From the point of view of the colony, as a whole, the suicide of an individual accomplishes more than it loses. The total worker force consists of twenty thousand to eighty thousand members, all sisters born from eggs laid by the mother queen. Each bee has a natural life span of only about fifty days, after which it dies of old age. So to give a life is only a little thing, with no genes being spilled…

Sharing the capacity for extreme sacrifice does not mean that the human mind and the “mind” of an insect (if such exists) work alike. But it does mean that the impulse need not be ruled divine or otherwise transcendental, and we are justified in seeking a more conventional biological explanation. A basic problem immediately arises in connection with such an explanation: fallen heroes do not have children. If self-sacrifice results in fewer descendants, the genes that allow heroes to be created can be expected to disappear gradually from the population. A narrow interpretation of Darwinian natural selection would predict this outcome: because people governed by selfish genes must prevail over those with altruistic genes, there should also be a tendency over many generations for selfish genes to increase in prevalence and for a population to become ever less capable of responding altruistically.

How then does altruism persist? In the case of social insects, there is no doubt at all. Natural selection has been broadened to include kin selection. The self-sacrificing termite soldier protects the rest of its colony, including the queen and king, its parents. As a result, the soldier’s more fertile brothers and sisters flourish, and through them the altruistic genes are multiplied by a greater production of nephews and nieces.

It is natural, then, to ask whether through kin selection the capacity for altruism has also evolved in human beings. In other words, do the emotions we feel, which in exceptional individuals may climax in total self-sacrifice, stem ultimately from hereditary units that were implanted by the favoring of relatives during a period of hundreds or thousands of generations? This explanation gains some strength from the circumstance that during most of mankind’s history the predominant social unit was the immediate family and a tight network of other close relatives. Such exceptional cohesion, combined with detailed kin classifications made possible by high intelligence, might explain why kin selection has been more forceful in human beings than in monkeys and other mammals.

To anticipate a common objection raised by many social scientists and others, let me grant at once that the form and intensity of altruistic acts are to a large extent culturally determined. Human social evolution is obviously more cultural than genetic. The point is that the underlying emotion, powerfully manifested in virtually all human societies, is what is considered to evolve through genes. The sociobiological hypothesis does not therefore account for differences among societies, but it can explain why human beings differ from other mammals and why, in one narrow aspect, they more closely resemble social insects…

To resolve the puzzle of human altruism we must distinguish two basic forms of cooperative behavior. The altruistic impulse can be irrational and unilaterally directed at others; the bestower expresses no desire for equal return and performs no unconscious actions leading to the same end. I have called this form of behavior “hard-core” altruism, a set of responses relatively unaffected by social reward or punishment beyond childhood. Where such behavior exists, it is likely to have evolved through kin selection or natural selection operating on entire, competing family or tribal units. We would expect hard-core altruism to serve the altruist’s closest relatives and to decline steeply in frequency and intensity as relationship becomes more distant.

“Soft-core” altruism, in contrast, is ultimately selfish. The “altruist” expects reciprocation from society for himself or his closest relatives. His good behavior is calculating, often in a wholly conscious way, and his maneuvers are orchestrated by the excruciatingly intricate sanctions and demands of society. The capacity for soft-core altruism can be expected to have evolved primarily by selection of individuals and to be deeply influenced by the vagaries of cultural evolution. Its psychological vehicles are lying, pretense, and deceit, including self-deceit, because the actor is most convincing who believes that his performance is real.

A key question of social theory, then, must be the relative amounts of hard-core as opposed to soft-core altruism. In honeybees and termites, the issue has already been settled: kin selection is paramount, and altruism is virtually all hard-core. There are no hypocrites among the social insects. This tendency also prevails among the higher animals. It is true that a small amount of reciprocation is practiced by monkeys and apes…

But in human beings soft-core altruism has been carried to elaborate extremes. Reciprocation among distantly related or unrelated individuals is the key to human society… Through the convention of reciprocation, combined with a flexible, endlessly productive language and a genius for verbal classification, human beings fashion long-remembered agreements upon which cultures and civilization can be built.

Yet the question remains: Is there a foundation of hard-core altruism beneath all of this contractual superstructure?… The question is important because pure, hard-core altruism based on kin selection is the enemy of civilization. If human beings are to a large extent guided by programmed learning rules and canalized emotional development to favor their own relatives and tribe, only a limited amount of global harmony is possible. International cooperation will approach an upper limit, from which it will be knocked down by the perturbations of war and economic struggle, canceling each upward surge based on pure reason…

My own estimate of the relative proportions of hard-core and soft-core altruism in human behavior is optimistic. Human beings appear to be sufficiently selfish and calculating to be capable of indefinitely greater harmony and social homeostasis. This statement is not self-contradictory. True selfishness, if obedient to the other constraints of mammalian biology, is the key to a more nearly perfect social contract.

My optimism is based on evidence concerning the nature of tribalism and ethnicity. If altruism were rigidly unilateral, kin and ethnic ties would be maintained with commensurate tenacity. The lines of allegiance, being difficult or impossible to break, would become progressively tangled until cultural change was halted in their snarl. Under such circumstances the preservation of social units of intermediate size, the extended family and the tribe, would be paramount. We should see it working at the conspicuous expense of individual welfare on the one side and of national interest on the other…

Individual behavior, including seemingly altruistic acts bestowed on tribe and nation, are directed, sometimes very circuitously, toward the Darwinian advantage of the solitary human being and his closest relatives. The most elaborate forms of social organization, despite their outward appearance, serve ultimately as the vehicle of individual welfare.

Human altruism appears to be substantially hard-core when directed at closest relatives,although still to a much lesser degree than in the case of the social insects. The remainder of our altruism is essentially soft. The predicted result is a melange of ambivalence, deceit, and guilt that continuously troubles the individual mind…Yet it is a remarkable fact that all human altruism is shaped by powerful emotional controls of the kind intuitively expected to occur in its hardest forms. Moral aggression is most intensely expressed in the enforcement of reciprocation. The cheat, the turncoat, the apostate, and the traitor are objects of universal hatred. Honor and loyalty are reinforced by the stiffest codes…

In summary, soft-core altruism is characterized by strong emotions and variable allegiances. Human beings are consistent in their codes of honor but endlessly fickle with reference to whom the codes apply. The genius of human sociality is in fact the ease with which alliances are formed, broken, and reconstituted, always with strong emotional appeals to rules believed to be absolute. The important distinction is today, as it appears to have been since the Ice Age, between the in-group and the out-group, but the precise location of the dividing line is shifted back and forth with ease.

Professional sports thrive on the durability of this basic phenomenon. For an hour or so the spectator can resolve his world into an elemental physical struggle between tribal surrogates. The athletes come from everywhere and are sold and traded on an almost yearly basis. The teams themselves are sold from city to city. But it does not matter; the fan identifies with an aggressive in-group, admires teamwork, bravery, and sacrifice, and shares the exultation of victory.

Nations play by the same rules. During the past thirty years geopolitical alignments have changed from a confrontation between the Axis and the Allies to one between the Communists and the Free World, then to oppositions between largely economic blocks. The United Nations is both a forum for the most idealistic rhetoric of humankind and a kaleidoscope of quickly shifting alliances based on selfish interests…

Can culture alter human behavior to approach altruistic perfection? Might it be possible to touch some magical talisman or design a Skinnerian technology that creates a race of saints? The answer is no. In sobering reflection, let us recall the words of Mark’s Jesus: “Go forth to every part of the world, and proclaim the Good News to the whole creation. Those who believe it and receive baptism will find salvation; those who do not believe will be condemned.” There lies the fountainhead of religious altruism. Virtually identical formulations, equally pure in tone and perfect with respect to in-group altruism, have been urged by the seers of every major religion, not omitting Marxism-Leninism. All have contended for supremacy over others…

“If only it were all so simple!,” Aleksandr Solzhenitsyn wrote in The Gulag Archipelago. “If only there were evil people somewhere insidiously committing evil deeds, and it were necessary only to separate them from the rest of us and destroy them. But the line dividing good and evil cuts through the heart of every human being. And who is willing to destroy a piece of his own heart?”

…The true humanization of altruism, in the sense of adding wisdom and insight to the social contract, can come only through a deeper scientific examination of morality… To the extent that principles are chosen by knowledge and reason remote from biology, they can at least in theory be non-Darwinian…

The philosophical question of interest is the following: Can the cultural evolution of higher ethical values gain a direction and momentum of its own and completely replace genetic evolution? I think not. The genes hold culture on a leash. The leash is very long, but inevitably values will be constrained in accordance with their effects on the human gene pool.