Moore's 'p and I do not believe that p': more Metadox than Paradox

Moore's Paradox gets its claws into your brain and doesn't let go, I find. Apparently, Wittgenstein 'once remarked that the only work of Moore’s that greatly impressed him was his discovery of the peculiar kind of nonsense involved in such a sentence as “It’s raining but I don’t believe it”.' He discussed it at some length in the tenth section of Investigations. Wikipedia (of course) lays it all out neatly:

Moore's paradox concerns the putative absurdity involved in asserting a first-person present-tense sentence such as 'It's raining but I don't believe that it is raining' or 'It's raining but I believe that it is not raining'. The first author to note this apparent absurdity was G.E. Moore. These 'Moorean' sentences, as they have become known:

• can be true,
• are (logically) consistent, and moreover
• are not (obviously) contradictions.

The 'paradox' consists in explaining why asserting a Moorean sentence is (or less strongly, strikes us as being) weird, absurd or nonsensical in some way. Subsequent commentators have further noted that there is an apparent residual absurdity in asserting a first-person future-tense sentence such as 'It will be raining and I will believe that it is not raining'. There is currently no generally accepted explanation of Moore's Paradox in the philosophical literature.

Since Jaakko Hintikka's seminal treatment of the problem, it has become standard to present Moore's Paradox as explaining why it is absurd to assert sentences that have the logical form: (OM) P and NOT(I believe that P), or (COM) P and I believe that NOT-P. Commentators nowadays refer to these, respectively, as the omissive and commissive versions of Moore's Paradox, a distinction according to the scope of the negation in the apparent assertion of a lack of belief ('I don't believe that p') or belief that NOT-P. The terms pertain to the kind of doxastic error (i.e. error of belief) that one is subject to, or guilty of, if one is as the Moorean sentence says one is. ...Moore presents the problem in a second, distinct, way:

• It is not absurd to assert the past-tense counterpart, e.g. 'It was raining but I did not believe that it was raining'.
• It is not absurd to assert the second- or third-person counterparts to Moore's sentences, e.g. 'It is raining but you do not believe that it is raining', or 'Michael is dead but they do not believe that he is'.
• It is absurd to assert the present-tense 'It is raining and I don't believe that it is raining'.
• I can assert that I was a certain way (e.g. believing it was raining when it wasn't), that you, he, or they, are that way, but not that I am that way. Why not?

Many commentators—though by no means all—also hold that Moore's Paradox arises not only at the level of assertion but also at the level of belief. Interestingly imagining someone who believes an instance of a Moorean sentence is tantamount to considering an agent who is subject to, or engaging in, self-deception (at least on one standard way of describing it).

What Moore identifies is a mouthfeel sort of question: there is nothing wrong with the statement; it just feels as though there is.

I wonder if this is a sort of meta-version of a more familiar issue with our mind's capacity to crunch data. We take in lots of discourse; mostly it makes sense, but sometimes it is gibberish. Our minds are quick at sorting the one from the other (distinguishing whether my 3-year old is actually communicating to me, or just babbling doubledutch). But there are statements that seem the latter and then resolve into the former: the example that comes to mind is from Stevens's 'Emperor of Ice-Cream':

Let the wenches dawdle in such dress
As they are used to wear, and let the boys
Bring flowers in last month's newspapers.
Let be be finale of seem.
The only emperor is the emperor of ice-cream.'

That 'Let be be finale of seem' line looks like spam-gibberish, with its unidiomatic reduplication of 'be' near the beginning, and its strange way with nouns. But it makes clear, perfect sense. I'm assuming that this particular knight's move, from 'nonsense' to 'ah, I see!' is precisely the effect that Stevens was going for.

Is there something similar happening, on a conceptual rather than a merely semantic level, with 'it's raining but I don't believe that it is raining'? It inhabits something that looks, at first blush, like a flat contradiction (as it might be: 'I believe it is raining and I do not believe that it is raining'), but in fact it isn't in that form. It's in as perfectly logical a form as 'It was raining but I didn't believe that it was raining.'

Except that it is different? (This is the 'bites, doesn't let go' part of this). 'It was raining but I didn't believe that it was raining' is simple enough: it means 'I was mistaken' (for instance: 'I thought it was the sound of applause; but it was just the sound of rain on the conservatory roof'). 'I believe it is raining and I do not believe that it is raining' would be a way of talking about my split personality, or the ability of human beings to live with contradictory things. But 'it's raining but I don't believe that it is raining' is a deeper statement; and maybe that's why it has such a striking mouthfeel. Perhaps its depth doesn't emerge until we replace its trivial belief with a nontrivial one. 'The universe is godless, but I don't believe the universe is godless' is a better iteration of the paradox. You see what I mean.