Epistemicism is notorious for making a pair of counterintuitive claims.1
According to epistemicism there was a nanosecond during which I stopped being a child: I was a child at the beginning of it, but not at the end.
Moreover, according to the epistemicist, it is not possible to know when that critical nanosecond occurred. It is these consequences of epistemicism that regularly invite the incredulous stare and which have led people to try and find alternative theories that respect classical logic; the most famous alternative being supervaluationism.Unfortunately for this project, it is possible to prove that there's a nanosecond at which I stopped being child, and that it is unknown which nanosecond it is, from a couple of eminently plausible premises. As we noted in chapter 1, apart from classical logic, all we need to derive the existence of a nanosecond at which I stopped being a child are the premises that I was a child after one nanosecond of my life had passed, but not after several billion had passed. But once we have accepted the existence of such a nanosecond it would be madness to suppose that we know which one it is. If you are not immediately convinced by this latter claim ask yourself which number it is: if you are unable to produce a satisfactory answer, I would suggest that this is because you do not know which number it is.
Thus the project of finding a more palatable alternative to epistemicism that accepts classical logic is pretty much a no-go if one takes these above consequences of epistemicism to be the source of the unpalatability. However, some theorists have suggested that the radical component of epistemicism is not the thesis that there is an unknown nanosecond at which I stopped being a child, but the claim that vagueness amounts to nothing more than this special kind of ignorance.[115] [116] What explains these astonishing theses asserting the existence of unknown cutoff points, according to these theorists, is the thought that when there is vagueness about where a boundary lies, there is also no fact of the matter about where it lies.
Unfortunately, the locution ‘there’s no fact of the matter’ is somewhat mysterious without further explanation.
Since we are assuming classical logic, it follows that Harry is either bald or he isn’t, even when there is no fact of the matter about whether Harry is bald. Thus one of the following must hold: either (i) Harry is bald but there’s no fact of the matter about whether Harry is bald, or (ii) he isn’t bald but there’s no fact of the matter about whether he’s bald. Both disjuncts are equivalent to something of the form ‘P but there’s no fact of the matter about whether P’.But this result does not seem to be consistent with our pretheoretic understanding of the locution ‘there’s no fact of the matter’. This suggests ‘there’s no fact of the matter’ is at best a semi-technical notion in need of further explanation. At this juncture the epistemicist could triumphantly point out that ‘P but it’s impossible to know whether P, is perfectly consistent, and seems to play the right role in explanations as well— perhaps the semi-technical talk of there being no fact of the matter is just code for talk about things being unknowable for certain kinds of reasons. This would be an outright defeat for the classical theorist I am describing: the challenge, then, is to find some more technical notion that is consistent with the consequence that something can be true even when there’s no fact of the matter, but excludes an epistemicist reading of it.[117]As I have warned earlier, the task of attempting to find an uncontroversial classification of theories of vagueness as ‘epistemicist’ or ‘not epistemicist’ is not a productive one. That said, there is a set of questions that I think helps clarify the view that vagueness is just a matter of ignorance or uncertainty. If one can identify distinctive features of vagueness-related uncertainty that are dramatically different from ordinary uncertainty, it becomes harder to maintain that there’s little more to vagueness than ignorance. For if uncertainty about whether Harry is bald is just the same as uncertainty about where I left my keys, for example, we shouldn’t expect to see very dramatic differences. In recent years this strategy has been pursued by a number of authors.[118] The operative thought in all these cases is that when one is uncertain about something because it is believed it to be borderline one is in a very different kind of state than when one is uncertain about some ordinary matter such as where a set of keys has been left. However, these theorists typically end up rejecting the probability calculus; a thesis quite central to the present project (see chapter 7). In this chapter we shall explore, drawing on some analogies with expressivism about other subject matters, some distinctive features of beliefs about the vague that are predicted by the theory of vague propositions developed in chapter 6.
8.1