Abstract

In this paper, we provide a game-theoretic examination of indirect utilitarianism by comparing the expected payoffs of attempts to apply a deontological principle and a utilitarian principle within the context of the Prisoner’s Dilemma (PD). Although many of the best-known utilitarians and consequentialists have accepted some indirect form of their respective views, the results in this paper suggest that they have been overly quick to dismiss altogether the benefits of directly enacting utilitarian principles. We show that for infallible moral agents, what we call ‘non-autonomous agents’, direct utilitarianism dominates indirect utilitarianism via deontology in terms of achieving the maximized utilitarian outcome, but only in underlying games where the maximized utilitarian outcome involves unequal payoffs. In other situations, indirect utilitarianism implemented through Kantian deontology either ties or dominates direct utilitarianism in terms of achieving the maximized utilitarian outcome. We also examine the two different moralities on the assumption that fallibility, which is a form of autonomy, is an aspect of moral agency by introducing Endogenized Morality Models (EMM’s). We believe that just as indirect utilitarians worry about the cost of applying moral principles, so too they should worry about the fact that humans have both pro-social and materialistically selfish motivations and hence are fallible moral agents. We show that there are conditions under which fallible autonomous utilitarians achieve higher expected material and psychic payoffs than fallible autonomous deontologists and conditions under which they do not.