Between Belief and Disbelief

Abstract

Sometimes instead of believing or disbelieving a proposition one does not take a stand on it. This essay explores that middle ground. We begin by distinguishing a variety of different attitudes or cognitive relations one might have to a proposition that one does not believe or disbelieve. We argue that identifying all of them as suspending judgment or withholding judgment neglects important differences. We then discuss epistemic evaluations of these attitudes and relations. Finally, we examine the implications of our findings for the attitudes that a philosopher might take toward contentious philosophical theses. Evidentialism, our preferred view of epistemic justification, is helpful in clarifying the issues and addressing the problems.

Appendix

John Turri (2012) develops an epistemic puzzle. He offers an example in which the subject has no evidence for or against a proposition and yet, Turri argues, withholding judgment is not epistemically justified for the subject. We have been defending the evidentialist view that withholding judgment on a proposition is always epistemically justified for anyone whenever a person grasps a proposition and it is a determinate fact that the person’s evidence supports neither the proposition nor its negation. So we should address Turri’s argument.

Here is an example very much like Turri’s, with a couple of small changes that are intended to strengthen its challenge to our evidentialist view.

P is a mathematical proposition that S understands. S has no evidence for or against P. S is polling 100 people about P, people who S knows to be both mathematicians and epistemologists who have studied the literature on epistemic justification. (We have added to Turri’s example that the polled mathematicians have this epistemological expertise.) S polls the 100 about P as they leave a meeting room in which they have engaged in a discussion of P’s truth value. In answer to S’s inquiry about P each of the 100 tells S that S’s withholding judgment on P is not epistemically justified (We have changed Turri’s example so that the 100 assert that S’s withholding is unjustified, where Turri has them saying that the withholding “is not the thing to do.”) The mathematicians say nothing more to S about either P’s truth value or the justification of any attitude toward P.

It is clear that after S’s polling S still has no evidence for or against P. So in our evidentialist view neither believing nor disbelieving is justified for S. In our view under such circumstances S’s withholding judgment on the proposition is epistemically justified.

Turri argues against the justification of S’s withholding judgment:

I find it difficult to accept that in a case where all the evidence directly indicates that withholding is not the thing to do, withholding is nevertheless the thing to do. We seem to be owed some explanation of why this should be so, especially since it would constitute a dramatic deviation from the effects of expert advice in other cases. Other things being equal, we think we should follow the experts’ advice when they recommend: believing, disbelieving, withholding, not believing, and not disbelieving. Why would not withholding be any different?” (2012, 363)

As a preliminary point we can now explain why we added to the story that the 100 have known epistemological expertise. We did so in order to have S know that their assertions about justification are well informed about both the math and the justification. So S has excellent reason to think that 100 people with ample relevant expertise assert that S’s withholding judgment on P is unjustified. As Turri observes, when experts make judgments within their expertise, our justified response defers to those judgments unless we have some special reason not to do so. Turri asks what the special reason is in this case.

Suppose that S does not have epistemic expertise. In particular, S does not have good reason to think that the experts are mistaken in this case. With this additional supposition of S’s having no independent epistemological evidence, S’s deferring to the 100 experts is quite justified. The justified deference concerns S’s attitude toward the contents of their testimony. That content is what their expertise supports. Their known expert testimony renders S justified in accepting what they tell S. S is justified in believing that S’s withholding judgment on P is not justified.

It does not follow that what the 100 tell S is true. Not only does this not follow, but also the story gives no reason to think that the 100 are right. Turri says that “the evidence directly indicates that withholding is not [justified]” (363). It is important to recognize who receives this indication. The expert testimony indicates to S that S’s withholding is not justified. But the story gives those who consider it no indication that this testimony is correct. The story simply stipulates that the relevantly knowledgeable 100 assert to S that S’s withholding is not justified, without giving any reason to think that these assertions are true. In the story the 100 do not identify anything about S’s withholding judgment that gives those who consider the story reason to think that S’s withholding is unjustified. Nor does the story give us any evidence that the 100 assertions that S’s withholding is unjustified are backed by some good evidence for that evaluation. The story does not provide any new information about what makes withholding justified or unjustified.

Evidentialism gives reason to think that S’s withholding on P is justified. Throughout the story S lacks evidence for or against P. Withholding is the sensible attitude for S, in light of all S has to go on about both the truth value of P and the justification of S’s attitude toward it. S’s evidence does not make any other attitude even a little bit reasonable. Either belief or disbelief would be quite unreasonable for S in light of the equal probability that the proposition has the other truth value and S’s lack of any higher order evidence in favor of either belief or disbelief in particular being justified. Withholding judgment on P is the noncommittal frame of mind that matches this lack of indication. It is fitting; it is justified. So in our view withholding judgment on P is the epistemically justified attitude for S.

Again, we think that in the story S is justified by the expert testimony in believing that the withholding is not justified. Thus our view has it that in this odd sort of case what someone is justified in believing to be the person’s justified attitude toward a proposition conflicts with what the person’s justified attitude actually is. This is unusual but not incoherent. S’s justification for the proposition that S’s withholding is unjustified is misleading expert testimony concerning which the epistemically uninformed S has no competing evidence. What makes it true that withholding judgment on P is justified for S is S’s lack of evidence for or against P. Misleading testimony can justify untrue epistemic beliefs in the presence of conflicting epistemic facts that are not apparent, just as misleading testimonial justification can justify untrue non-epistemic beliefs in the presence of conflicting non-epistemic facts that are not apparent. S does not see the epistemic facts. S’s having justification for believing that S’s withholding is not justified would be some excuse for S if S were to take some other doxastic attitude toward P in order to avoid what S has justification to think would be unjustified withholding. Nevertheless, S’s withholding would be justified.