The authors suggest that PA fighters do risk death, but in conflicts between groups, it is the groups with more PAs that have a higher chance of winning. Thus, while parochial altruists are selected against (because they risk their lives for their group), they are also selected for (because they kill off more members of less-PA groups in violent conflicts). Moreover, the losing side's numbers are replenished by conquerors' genes (thus becoming more PA).
The authors contend that archeologically-derived estimates of group warfare are consistent with their scenario for the evolution parochial altruism. One would think that other, more recent, historical examples could also be used, e.g., between city-state warfare in classical Greece.
The paper's innovation is that a seemingly "irrational" behavior from selfish genes' point of view could nonetheless evolve. The genes that cause their bearers to die in patriotic battles may die, but their competing alleles on the losing side may suffer more.
Science 26 October 2007:
Vol. 318. no. 5850, pp. 636 - 640
The Coevolution of Parochial Altruism and War
Jung-Kyoo Choi1 and Samuel Bowles2*
Altruism—benefiting fellow group members at a cost to oneself—and parochialism—hostility toward individuals not of one's own ethnic, racial, or other group—are common human behaviors. The intersection of the two—which we term "parochial altruism"—is puzzling from an evolutionary perspective because altruistic or parochial behavior reduces one's payoffs by comparison to what one would gain by eschewing these behaviors. But parochial altruism could have evolved if parochialism promoted intergroup hostilities and the combination of altruism and parochialism contributed to success in these conflicts. Our game-theoretic analysis and agent-based simulations show that under conditions likely to have been experienced by late Pleistocene and early Holocene humans, neither parochialism nor altruism would have been viable singly, but by promoting group conflict, they could have evolved jointly.