> There's no situation where "she accused (Mel and) I" works.
That is true.
> It happens because of people "hyper correcting" to match the inverted sentence. "(Mel and) I accused her" which they get corrected to use by the teacher
That is also true. But your own example is telling you that there is also no situation where "Mel and I accused her" works. Sentences in that form do not naturally occur in English; they must be taught in school. The correct form would be "Me and Mel accused her".
Intrusive R isn't like that. It is a feature of the language, not something dictated to speakers by a formal education system.
> The correct form would be "Me and Mel accused her".
No it isn't. You'd never say "Me accused her". If you don't like the inversion its perfectly fine to say "I and Mel accused her", but using "me" for the subject is never correct.
If you believe otherwise, you are tragically misinformed and miseducated.
Using "me" in subject context is, besides being wrong, a social low-class marker. It does not come naturally to English speakers; it only ever occurs through exposure to the misuse.
I'm a bit skeptical of this overall, but it's obviously complicated to get data that filter out the effects of sociolectal solidarity. (Can you get people to speak "naturally" but without priming them to speak or to avoid speaking in a particular register or sociolect!?)
which can become conventionalized and eventually part of native speakers' intuitions even though they can be contrary to a more logically consistent or simpler understanding of agreement.
> Using "me" in subject context is, besides being wrong, a social low-class marker. It does not come naturally to English speakers
This is not even a theoretical possibility. There are no low-class markers that don't come naturally, because no one is going to make an effort to present themselves as low-class.
It takes obvious effort to overcome received habits from one's local idiolect, as for example underclass people did in England, last century, deliberately adopting an (imperfectly reproduced version of) the overclass rhotic affectation in an attempt at economic advancement.
You are getting really disingenuous, now. Who are you performing for?
You like saying that, but so far you've shown no evidence that you have any idea what you're talking about.
Compare the article on intrusive R I linked before:
> This is now common enough in parts of England that, by 1997, the linguist John C. Wells considered it objectively part of Received Pronunciation, though he noted that it was still stigmatized as an incorrect pronunciation, as it is or was in some other standardized non-rhotic accents. Wells writes that at least in RP, "linking /r/ and intrusive /r/ are distinct only historically and orthographically".[17]
John C. Wells has patiently explained to you that intrusive-R pronunciation, while not expected of anyone, is (mostly) not considered to require correction, or to reliably unmask a bounder.
> There's no situation where "she accused (Mel and) I" works.
That is true.
> It happens because of people "hyper correcting" to match the inverted sentence. "(Mel and) I accused her" which they get corrected to use by the teacher
That is also true. But your own example is telling you that there is also no situation where "Mel and I accused her" works. Sentences in that form do not naturally occur in English; they must be taught in school. The correct form would be "Me and Mel accused her".
Intrusive R isn't like that. It is a feature of the language, not something dictated to speakers by a formal education system.