On Sat, 15 Sep 2001, John Cowan wrote:

> Marcus Smith scripsit:
> > It makes as much sense to me to say that plural
> > -s (dogs), possessive -'s (John's), and 3rd person singular present -s
> > (walks) are the same thing.
> Well, actually, I do think they are.  For one thing, they have exactly
> the same morphophonemic alternation of /z/ - /s/ - /Iz/ after voiced,
> voiceless, and sibilant consonants.

Somehow I knew you were going to say that. :) I've also heard this kind of
story from quite well-known linguists, but I've never been able to buy
into it.

For one thing, the fact that they have the same alternation doesn't
mean much. You get exactly the same type of alternation with past tense
<ed>: walk[t] - beg[d] - load[Id]. This points to a common phonological
basis for the alternation, so it is not surprising for homophonous
morphemes to undergo the exact same process.

Another factor is one that has been discussed many times on this list: the
possessive -'s attaches to phrases rather than words. This is not true of
the plural or 3rd singular. They have different syntactic distributions.

I suppose you know all of this though.