Stephen Maitzen

According to global skepticism, we know nothing. According to local skepticism, we know nothing in some particular area or domain of discourse. Unlike their global counterparts, local skeptics think they can contain our invincible ignorance within limited bounds. I argue that they are mistaken. Local skepticism, particularly the kinds that most often get defended, cannot stay local: if there are domains whose truths we cannot know, then there must be claims outside those domains that we cannot know even if they are true. My argument focuses on one popular form of local skepticism, ethical skepticism, but I believe that the argument generalizes to cover other forms as well.

Skeptical theism combines theism with skepticism about our capacity to discern God’s morally sufficient reasons for permitting evil. Proponents have claimed that skeptical theism defeats the evidential argument from evil. Many opponents have objected that it implies untenable moral skepticism, induces appalling moral paralysis, and the like. Recently Daniel Howard-Snyder has tried to rebut this prevalent objection to skeptical theism by rebutting it as an objection to the skeptical part of skeptical theism, which part he labels “Agnosticism” (with an intentionally capital “A”). I argue that his rebuttal fails as a defense of Agnosticism against the objection and even more so as a defense of skeptical theism.

Newcomb's problem supposedly involves your choosing one or else two boxes in circumstances in which a predictor has made a prediction of how many boxes you will choose. We argue that the circumstances which allegedly define Newcomb's problem generate a previously unnoticed regress which shows that Newcomb's problem is insoluble because it is ill-formed. Those who favor, as we do, a ``no-box'' reply to Newcomb's problem typically claim either that the problem's solution is underdetermined or else that it is overdetermined. We are no-boxers of the first kind, but the underdetermination we identify is more radical than any previously identified: it blocks the very set-up of the problem and not just potential solutions to the problem once it has been set up. The defect is subtle, but it cripples every genuine version of the problem, regardless of variations in such things as the predictor's degree of reliability, the basis on which the prediction is made, or the amount of money in each box. The regress shows that, surprisingly enough, no one can understand Newcomb's problem, and so no one can possibly solve it

In a dense and fascinating article of some ten years ago, Toomas Karmo adds his voice to the chorus of philosophers who deny the possibility of soundly deriving ‘ought’ from ‘is.’ According to Karmo, no derivation containing an ethical conclusion and only non-ethical premises can possibly be sound, where ‘sound’ describes a deductively valid derivation all of whose premises are true. He also suggests that the only valid derivations of ‘ought’ from ‘is’ will be trivial ones. His argument has, to my knowledge, gone unrebutted; on the contrary, it has won recent endorsements, some philosophers evidently believing that he has finally put to rest the issue of the logical autonomy of ethics. Against that trend, I intend to rebut his argument, both by falsifying the taxonomy on which his argument relies and by offering a nontrivial and potentially sound ‘is’-‘ought’ derivation of my own.

Skeptical theism claims that the probability of a perfect God’s existence isn’t at all reduced by our failure to see how such a God could allow the horrific suffering that occurs in our world. Given our finite grasp of the realm of value, skeptical theists argue, it shouldn’t surprise us that we fail to see the reasons that justify God in allowing such suffering, and thus our failure to see those reasons is no evidence against God’s existence or perfection. Critics object that skeptical theism implies a degree of moral skepticism that even skeptical theists will find objectionable and that it undermines moral obligations that even skeptical theists will want to preserve. I discuss a version of the first objection and defend a version of the second.

Unfortunately, the book's weaknesses outweigh its strengths. Chief among the weaknesses is its spotty attention to relevant and important literature, both historical and contemporary. Even though Helm writes at length about assent, and even though he discusses Augustine, he completely ignores John Henry Newman, whose Grammar of Assent deserves at least a mention. Helm devotes more than a chapter to the relation between belief and the will and another chapter to fideism, yet he never mentions Louis Pojman's arguments in Religious Belief and the Will, an analytic-philosophical treatment of exactly those issues. He attacks evidentialism without confronting the most important defenses of that view, such as those offered in Feldman and Conee's "Evidentialism." He asserts the possibility of self-contradictory beliefs without considering the reasons offered by possible-worlds theorists and others for denying that a self-contradictory proposition can be a relatum in the belief relation. He simply asserts that "[n]o obligations follow logically from facts", without any indication that such a claim is both highly controversial and the focus of an enormous, decades-old literature.

The Knower Paradox has had a brief but eventful history, and principles of epistemic closure (which say that a subject automatically knows any proposition she knows to be materially implied, or logically entailed, by a proposition she already knows) have been the subject of tremendous debate in epistemic logic and epistemology more generally, especially because the fate of standard arguments for and against skepticism seems to turn on the fate of closure. As far as I can tell, however, no one working in either area has emphasized the result I emphasize in this paper: the Knower Paradox just falsifies even the most widely accepted general principles of epistemic closure. After establishing that result, I discuss five of its more important consequences.