Might could have explained that better

The local "grammar guy" in my father's paper is, in my admittedly very limited exposure to him, a peever who can't explain his positions even when he's right. Today he tackles double modals, not that he calls them that. He doesn't even seem to know what the problem is with them:

Have you ever said "might could?" Many have.

This expression and a few others like it are nonstandard. They are more common in speech than in writing, and they are heard in regional American speech, especially in the South, according to Webster's Dictionary of English Usage.

And then he trots out this contradiction, without even seeming to notice:

The helping verb "might" seems to intensify the notion of possibility or speculation, Webster's says.

The problem with these expressions is that only one of the verbs is necessary.

Apparently, he thinks "might do", "could do", and "might could do" all mean the same thing. (At least that's what he's saying.) Clearly they don't. (I say 'clearly'  because, as a double modal user from a family of them, it's obvious to me that "might" is modifying "could".)

What is true is not that only one is necessary, but that in MSE only one is permitted. To say the same thing as "might could" in MSE, you don't simply drop one modal. The one you don't use must be replaced with an adverb (he possibly could, maybe he could) or another, non-modal-auxiliary verb (he might be able to).

ps - my father said he thought "might" wasn't even a verb here, but an adverb. This may in fact be the same feeling that most double-modal speakers have: might = probably/perhaps.