Ernst Fehr and Simon Gächter have introduced the concept of altruistic punishment to explain cooperation [in group projects for the "public good"]. Following Fehr and Gächter, I define altruistic punishment as punishment inflicted on a defector from cooperation, which is costly to the punisher and without material gain. In a series of public goods experiments that pitted private return against public welfare requiring cooperation in a group project, cooperation was found to flourish when altruistic punishment was possible and to break down when it was ruled out.SOURCE: The Killing Trap: Genocide in the Twentieth Century, by Manus I. Midlarsky (Cambridge U. Press, 2005), pp. 108-109
Subjects were given the opportunity to invest in a group project with monies handed to them, or to keep the funds. Individually, if they chose to invest, they would receive less than if they kept the money, but collectively the group as a whole would receive more, if all invested.... Subjects could punish others after information was provided as to how much each had invested. But each punishment of another subject was costly. Specifically, subjects who chose to punish were required to forfeit an amount equal to one-third of the monetary punishment imposed on a defector. Thus, the punishment is altruistic. Defectors – those who refused to cooperate – were punished even when material self-interest was sacrificed by cooperators. Participants who chose to punish defectors by withholding monies themselves had to sacrifice monetary rewards....
"Negative emotions towards the defectors are the proximate mechanics behind altruistic punishment." Concerning altruistic punishment, Fehr remarks, "It's a very potent force for establishing large-scale cooperation, every citizen is a little policeman in a sense. There are so many social norms that we follow almost unconsciously, and they are enforced by the moral outrage we expect if we were to violate them." The greater the extent of deviation from cooperation by defectors, the more heavily they were punished by cooperators. It was "punishment per se [that] provided the motivation, not some consequence anticipated by the player." Instrumentality was not especially relevant....
Altruistic punishment has been robustly established. These findings stem from public goods experiments, but are readily generalizable to social groups seeking a basis for cooperation in the absence of a functioning external authority. Given an extreme, even life-threatening environment, such as massive economic failure followed by war, the particular form of altruistic punishment chosen can be severe.
Even as they were clearly losing their respective wars, Hitler proceeded mercilessly with his extermination campaign, Enver Paşa lent his approval to the Armenian genocide, and the Hutu extremists were rapidly eliminating Rwandan Tutsi. Hitler would die by his own hand, Enver in battle to unite Turkic peoples against the Soviets, and many of the Hutu génocidaires in the refugee camps of northern Congo.
In a chapter somewhat disturbingly redolent of "the freakonomics of genocide" Midlarsky highlights altruistic punishment as a factor that not only motivates followers to cooperate in a genocidal project they otherwise find extremely distasteful (or worse), but also motivates leaders to persist in genocide at the expense of losing larger battles against their enemies.
On a more familiar level, "altruistic punishment" might well describe the motivations of political activists who would rather purge their party of defectors than win the next election.