I was going for a sort of reductio ad absurdum with my comment on the first declension genitive and dative singular being the same case. If we look at it entirely disconnected from any inflectional morphemes then we would have, who knows, some 25 or 30 cases including at least two for subjects of sentences.
I see. Not sure I quite understand. Do you mean if you ignore the overt markings you can analyse sentences in such a way that in Latin or English there are something like 20 or 30 cases? There's probably way more than that. I once listened to a nervous little East german professor read stiffly from his poorly translated paper why there were hundreds of different kinds of and, logically speaking, not grammatically. I think the Roman, Greek, and Indian grammarians did a fine job of abstracted the overt morphological markers from the grammatical functions when it came to case. If you look at Latin's five declensions, you get a good feel for there being five cases, and maybe an extra one thrown in for the 2nd declension vocative. That some cases in certain declensions have identical endings did not seem to confuse them very much, but who knows, maybe you're on to something. You should write it up and send it to Language or Linguistic Inquiry.
This is all in aid of my internal rantings against the Huddleston/Pullum categorization of bush as a preposition in the sentence 'On hatching, the chicks scramble to the surface and head bush on their own.'
I don't know. I've always felt that the fiddly bits of parts of speech that are left over in the linguistic rag bag after all the heavy lifting and categorization are done, make for interesting perusal. There is something about prepositions, verbal particles, and adverbs, that is all messy and overlappish. Pullum's discovery is at least interesting, but then again if it rubs you the wrong way, rant on, dude! You'll be in good company. every linguistic conference I've ever gone to is full of ranting, peevish linguists disagreeing with each other and sometimes with earlier versions of themselves.
