One of the books is Steven Pinker's The Language Instinct and one of the topics he examines is irregular plurals and why, in certain situations, they become regular; why, for example, do we say the Toronto Maples Leafs instead of leaves, why low-lifes instead of low-lives, and why Walkmans instead of Walkmen (this was in 1994, long before the iPod, when Walkman's still ruled the portable personal music market).
One of the reasons that he rejects on p. 142 is that "the books are closed on irregular words;" no more will be countenanced. He counters this with the example of oil-mice (a group Chinese peasants who scavenge oil from unprotected wells and a term that doesn't seem to have caught on.)
Having dispensed with this idea, Pinker posits another reason: headlessness.
"A headless word is an exceptional item that, for one reason or another, differs in some property from its rightmost element, the one that it would be based on if it were like ordinary words. A simple example of a headless word is a low-life--not a kind of life at all but a kind of person."Now, as I said, this was about 14 years ago, so this has probably been pointed out already, but Pinker supplies the example that contravenes his own analysis and doesn't even seem to realise it: oil-mice, a kind of person, not a kind of mouse. In Pinker's theory, oil-mouse is a headless noun and should be pluralised as oil-mouses.
This isn't the first time one of Pinker's examples has come back to bite him. His cute assertion that "no mere mortal has even flown out to center field" was examined on Language Log where Mark Liberman concluded that "when an actual sportswriter-type human being is writing prose about baseball, at least in the texts indexed by the Google News Archive, more human beings 'flew out to center field' than 'flied out to center field'."
Oh well, back to the drawing board.