On indirect reciprocity: the distinction between reciprocity and altruism, and a comment on suicide terrorism

American Journal of Economics and Sociology, The, Nov, 2004 by Jan Tullberg

IN THIS PAPER, I discuss the relationship between altruism and reciprocity and its relevance to normative and descriptive ethics. On the descriptive level, the paper argues against the strong tendency in biology to dissolve altruism. Normatively, the article argues against the dominant position in philosophy, which states that altruistic norms and acts are desired. An analysis of the events of September 11, 2001 indicates that institutionalized reciprocity and reciprocal reputation are insufficient for understanding those terrorists' behavior. However, altruism, particularly in the form of metaphysical reward, seems to provide a most important explanation. This judgment supports the two general conclusions of the paper. There should be a strict separation between reciprocity and altruism, instead of using the term "indirect reciprocity" as a wide gray zone. Real indirect reciprocity, i.e., reciprocal reputation and institutionalized reciprocity, is socially valuable. Altruism, sometimes presented as indirect reciprocity, is more of an obstacle than an asset to a democratic society.

Introduction

ALTRUISM is a central concept in moral philosophy as well as in sociobiology. However, this concept is certainly problematic, since in moral philosophy it tends to constitute the very bulk of what is called morality, whereas in sociobiology altruism as an explanation of human behavior tends to disappear. Apparent altruistic behavior can instead be explained by kin selection, reciprocity, and extensions of reciprocity (in other words, by natural selection), and these biological explanations leave little room for a behavior to be classified as altruism in a strict sense (see, e.g., Trivers 1971; Wilson 1979; Alexander 1979, 1987). The analysis in this paper acknowledges altruism, understood as actions that involve a net cost (see, e.g., Sesardic 1995), as a real factor in human societies. Arguments for this judgment will be presented later. However, this acknowledgment does not imply an affirmation of the common judgment in much of moral philosophy, which states that altruism is both important and beneficial. The value of altruism as a social factor needs to be reconsidered in light of an improved understanding of human behavior.

In his important book on moral systems, Alexander uses the term "indirect reciprocity," defined as acts in which "'the return is expected from someone other than the recipient of the beneficence" (1987, p. 85). Unfortunately, there are few theoretical investigations of cooperation including more than two individuals (Legge 1996). Boyd and Richerson (1989), investigating indirect reciprocity, however, concluded that it is more important to identify noncooperators and exclude them from cooperation than to focus on one's own treatment. They also concluded that a weakness of indirect reciprocity is that it is much more vulnerable to errors than is direct reciprocity. Recent theoretical work by Nowak and Sigmund (1998) claims that cooperation entirely based on indirect reciprocity can be evolutionarily stable, and that one determining factor for this is the availability of information about the cooperative quality of other individuals in the population. Work by Leimar and Hammerstein (2001) introduces a quality dimension in the "good standing" that improves the theoretical possibilities of indirect reciprocity.

In Alexander's (1987) 21 categories of human social behavior, altruism is nowhere to be found, and the reason given is that "we are not yet shown any beneficent behavior that cannot easily and logically be linked to nepotism and reciprocity" (1987, pp. 87, 160). According to this view, what humans name altruism is basically indirect reciprocity, with the return given in less obvious ways and without it having to be understood or consciously expected by the actor. For instance, the return could be elevated status, which in a second step could bring advantages.

Similarly, Wilson (1979) distinguishes between two basic forms of social behavior, namely, "hard-core" and "soft-core" altruism, corresponding to kin selection and reciprocity, respectively. He tends to place strict altruism in neither of these categories. Thus, in sociobiology altruism is exposed to the special problem of extinction, as succinctly pointed out by Trivers (1971, p. 35).

Neither "kin altruism" nor "reciprocal altruism" is altruism in a strict sense, as these phenomena can be understood from a selfish gene perspective. Indirect reciprocity might simply be a form of reciprocity, as indicated both by the term and by how most sociobiologists understand behavior. On the other hand, however, it might contain patterns that involve altruism. It is therefore of value to analyze the broad concept of indirect reciprocity as defined above, since it can include several types of social interactions that may vary in their probability of giving the actor a net return, or, in other words, in their probability of excluding or including altruism. Elsewhere, Tullberg and Tullberg (1994, 1996, 1997) have argued at greater length about the rationale of doubting the value of altruism as an ethical and social ambition, and this doubt makes the distinction between reciprocity and altruism important.