Moore famously responded to the skeptics’ modus ponens by shifting the argument to a modus tollens and justifying the move by appealing to a Moorean fact. The classic example is as follows. The skeptic says:
(1) If S does not know that S is not a brain in a vat, S does not know there is
an external world.
(2) S does not know that S is not a brain in a vat.
(3) S does not know there is an external world.
And Moore says in response that
(2*) S knows there is an external world.
is more reasonable to believe than (2), and so concludes
(3*) S knows that S is not a brain in a vat.
More precisely, the Moorean fact in question is:
(4) It is more reasonable to believe (2*) than (2).
Now, suppose a persistent skeptic denies (4) and says:
(5) It is not more reasonable to believe (2*) than (2).
I say the skeptic who denies (4) is a persistent skeptic because (4) is a very modest proposal; definitely more modest than (2*). And typically, more modest proposals are more reasonable to believe than less modest ones. But what happens when the skeptic asserts (5)? Can Moore make another appeal to common sense to once again outmaneuver the skeptic, or does the skeptic’s denial of (4) drag Moore back into the skeptic’s playground?
If the dialectic so far is valid (i.e., no questions are being begged, fallacies being committed, etc), what would prevent Moore from positing second-order Moorean facts? For example, in response to the persistent skeptic, why can’t Moore say
(6) It is more reasonable to believe (4) than (5).
After all, if (4) is more modest than (2*), wouldn’t (6) be more modest than (5), thus even harder to deny than (4)? I guess another way of putting the question is this: does the modesty increase with higher-order Moorean facts?
Objection: What makes (4) more modest than (2*) is the fact that (2*) is about what one knows, whereas (4) is about what is reasonable to believe. But both (4) and (5) are about what is reasonable to believe, so (4) is not more modest than (5).
This is a good reply, but there may be a decent rejoinder. Consider first the definition of a Moorean fact. On the construal above, a Moorean fact is about what is reasonable to believe. But now two things bear saying. First, exactly how is what is known being distinguished from what is reasonable to believe here, such that there are things that are reasonable to believe that are not known while things that are known are reasonable to believe? Second, the canonical definition of a Moorean fact is about what one knows (Cf. David Lewis in “Elusive Knowledge”: “A Moorean Fact [is] one of those things that we know better than we know the premises of any philosophical argument to the contrary.”). Granted, Lewis’ definition is a bit odd in that it requires degrees of knowledge; but what if we understand a Moorean fact as such? Well, it seems clear that the objection under consideration would be sufficiently dealt with. But the question then is whether the dialectic above changes in light of this reworded definition.
Consider second the nature of the persistent skeptic’s response. Actually, the skeptic can deny (4) in two ways. He could say either of the following:
(5) It is not more reasonable to believe (2*) than (2)
(5*) It is more reasonable to believe (2) than (2*)
If the skeptic denies (4) by asserting (5), he apparently is not making a claim about what is reasonable to believe at all; he is denying a claim about what is reasonable to believe, presumably on some grounds more reasonable to believe than (4). What are those grounds? Does the skeptic thus reach a stalemate with Moore over what is more reasonable to believe by asserting the grounds for (5)?
If the skeptic denies (4) by asserting (5*), then again he must have some grounds independent of skeptical scenarios to support it. His grounds for asserting (5*) must be independent of skeptical scenarios because skeptical scenarios are his grounds for asserting (2). If he appealed to the same grounds, he’d be engaged in circular reasoning. But if he has additional grounds for (5), what are they? And again, are they more reasonable to believe than (4)? Do we reach a stalemate again?