Dealing with altruism, Darwin1 recognizes that the selfish individuals are favored by evolution. At the same time, he has no doubts "that a tribe including many members who, from possessing in high degree the spirit of patriotism and fidelity, obedience, courage, and sympathy, were always ready to give aid to ea ch other and sacrifice themselves for the common good, would be victorious over most other tribes; and this would be natural selection".
This is puzzling and Darwin lets us down, laments Cronin2, suspecting group selectionism which she ridicules ("greater-goodism") and, joining Hamilton, concludes that Darwin dealt with human altruism, saw the problem, discussed it, but left it unsolved.
I assume that the discussion between individual selectionists, such as Dawkins 3, and the new-breed of group selectionists , D.S. Wilson4 or Sober, is well known. Whereas individual selectionists conclude that the number of altruists in a group must decrease, the group selections agree that this is so of isolated groups, but argue that when there are two or more groups with different ratios of altruists to egoists, the total
number of altruists in all groups increases.
I suggest that in some important respects the difference between the views of individual and group selectionists is irrelevant, because a third more powerful mechanism in humans ,meta-selection, is at play. What I mean by meta-selection is a higher-level-selection, such as when a breeder decides which quality ,e.g., altruism or egoism, to breed. If the breeder chooses one or the other, it is irrelevant whether individual or
group selection theory is right. I admit that there is no higher power in human society such as a breeder to do meta-selection. However, there is an equivalent power at work, and that is the group as- a-whole. Through group dynamics leadership emerges which becomes relatively independent and prescribes and forces certain behaviors for the good of the group (as
the leadership conceives it) , which differ from those the members would manifest were they governed by their own inclinations. The leadership manages to demand altruistic behavior andgives actual or symbolic social rewards, or promises of future rewards,
such as eternal life. But there is always a precarious balance between the leadership and the rest of the group or population. For example, the government demands that its citizens risk their lives in war, but if there were losses in Gulf War as great as those suffered by the Napoleonic army, the government would not survive.
I propose that the leadership has an important role throughout the evolution of humans in social exchange, which is one of the most important aspects of social dynamics. People have a concept and strong interest in balanced exchange, i.e., exchanging equivalent values, and in detecting those who cheat. Leda Cosmides5 assumes that this is a Darwinian algorithm. I believe it is, even though Cosmides' supporting experiments may not be as convincing as she thinks. Almost all people believe in fair or just exchange, but tend to exaggerate the value of their
own contributions.
The algorithm of social exchange is pre-programmed by evolution, but who programs the equivalencies of values which are, as is well known, so different at different times and circumstances? Who determines what is fair and just exchange? Here the leadership holds the key role, defining the equivalencies "for the good of the
group", but also in such a way as to advantage themselves. For example, in the first great empires rulers were allowed to have 1000 concubines. That was accepted as just and fair, though it was obviously at the expense of 1000 male altruists (as defined by sociobiology), presumably without them being altruistically inclined. What kind of individuals does the leadership favor? Throughout the ages, It favors those who accept the equivalencies as defined by the leadership, that is, individuals who are flexible to accept what the leadership defines as just and fair. This phenomenon has regressive and
progressive features, as the history of Germans in this century shows. Although mostly Christians, the majority was swayed by the Nazi ideology to accept the cruelties of anti-Semitism. To be so flexible, one needs to be endowed with a propensity to repress, with a "social filter"
(E. Fromm), which blinds one to the discrepancies in one's beliefs. However, after the War the German nation, as it seems, relatively quickly rejected the Nazi ideology. In general, those who rebel from the social consensus as defined and enforced by the leadership are minorities on two ends of the spectrum: on the altruistic end are those
fighting for the rights of the underprivileged, and on the egoistic end are those who ignore fair exchange or grossly misinterpret it, such as the criminals. The progress of human society rests in flattening the hierarchical pyramid towards a state in which everybody has some influence on defining the social exchange rules, without jeopardizing the welfare of the social system as a whole.
The main point of this essay is to introduce the possibilitythat group as a whole performs meta-selection. In the highest primates, the evolution has favored group organization and leadership, and in turn the leadership has been in a position to prescribe behaviors "for the good of the group" which nobody would engage in without
such leadership. This is reminiscent of Adam Smith's conclusion that an invisible hand transforms actions taken by individuals who are motivated only be desire for their o wn security and gain into forces that promote social goods that are not intended by the individuals.

REREFERENCES
1 Darwin (1871/1981). The descent of man and selection in relation to sex.Princeton: Princeton University Press.
2 Cronin, H. (1991).The ant and the peacock.Cambridge University Press.
3 Dawkins,R.(1089) Selfish gene,2nd ed. Oxford: Oxford University Press
4 Wilson, D. S.(1989). Levels of selection: An alternative to individualism in biology and the human sciences. In: E. Sober, Conceptual issues in evolutionary biology. Cambridge, Mass: The MIT
Press.
5 Cosmides, L. (1989).The logic of social exchange.Cognition, 31, 187-276.