Online research in philosophy

In ethics, it is commonly supposed that we have both positive duties and negative duties, things we ought to do and things we ought not to do. Given the many parallels between ethics and epistemology, we might suppose that the same is true in epistemology, and that we have both positive epistemic duties and negative epistemic duties. I argue that this is false; that is, that we have negative epistemic duties, but no positive ones. There are things that we ought (...) not to believe, but there is nothing that we ought to believe, on purely epistemic grounds. I also consider why the parallels between ethics and epistemology break down at this particular point, suggesting that it is due to what I call the infinite justificational ‘fecundity’ of perceptual and propositional evidence. (shrink)

In this article I show that the argument in John Harris's famous "Survival Lottery" paper cannot be right. Even if we grant Harris's assumptions—of the justifiability of such a lottery, the correctness of maximizing consequentialism, the indistinguishability between killing and letting die, the practical and political feasibility of such a scheme—the argument still will not yield the conclusion that Harris wants. On his own terms, the medically needy should be less favored (and more vulnerable to being killed), than Harris suggests.

What if human joy (more technically, utility) went on endlessly? Suppose, for example, that each human generation were followed by another, or that the Western religions are right when they teach that each human being lives eternally after death. If any such possibility is true in the actual world, then an agent might sometimes be so situated that more than one course of action would produce an infinite amount of utility (or of disutility, or of both). Deciding whether to have (...) a child born this year rather than next is a situation wherein an agent may face several alternatives whose effects could well ramify endlessly on such suppositions, for the child born this year would be a different person—one who preferred different things, performed different actions, and had different descendants—from a child born next year. It has recently been suggested that traditional utilitarianism stumbles on such cases of infinite utility. Specifically, utilitarianism seems to require, for its application, that all experience of pleasure and pain cease at some time in the future or asymptotically approach zero. If neither of these conditions holds, then the utility (and disutility) produced by each of two alternative actions may turn out to be infinite, and utilitarianism thus loses its ability to discriminate morally between them. (shrink)