In Morals by Agreement, David Gauthier convinced many of us—including Peter Danielson, author of Artificial Morality, the latest successor to MA—that morality can best be understood as a set of intramental, strategic responses to patterns of otherwise dilemmatic, game-theory-reducible interactivity. More particularly, Gauthier and Danielson are of a mind that: (a) characteristic of our interactive circumstances are the Prisoner's Dilemma and its cognates; (b) these are circumstances in which our pre-moral, straightforward maximizing (SM) disposition fares considerably worse than (virtually any species of) “constrained maximization” (CM)—the latter being the dispositional essence of morality—and (c) it should not surprise us, therefore, that (or, at least, if) selective pressure has rendered most of us just so morally (i.e., CM) disposed. In short, what Danielson calls the “fundamental” justification for morality—that is, one “that does not appeal to any of the concepts of [morality itself]” (AM, p. 19)—is that: conventional wisdom notwithstanding, nice guys—though, adds Danielson, not necessarily the nicest guys—finish first!
The threat and promise of artificial morality
