Considering the amount to which we agree it is surprising we are
having a lengthy discussion :-)
Elliott Lash wrote:
> I agree with all of the above, although I'm not sure
> the typological argument is as strong as you'd like.
> Typologically speaking, 3rd singular present is the
> most likely to be verbless, even if all other persons
> (or tenses) have verb forms.
I agree, but I again I want to ask whether diachrony shouldn't be
taken into consideration. At least a few forms in OIr., which look
verbless, etymologically do have the copula incorporated. I think of
"ní": synchronically it looks like a perfect case of 0-verb, but it's
agreed upon that it continues *ne-est(i).
> Perhaps you're right, although, I wouldn't put it past
> Old Irish speakers at least to think about their
> language as a series of surface coincidences. Well,
> maybe not the learned classes, but certainly those
> everyday people.
Again I agree, and things like that did indeed take place to a quite
great degree when OIr. was transformed into MidIr. and ModIr.
> Again, I agree with the above. Perhaps, I was a bit
> hasty in my comparison. However, a plausible (but
> unprovable) scenario for 'nis', would perhaps be that
> Old Irish speakers, unaware of the historical
> background, analyzed 'did,id,d' as pronouns, hence
> extending pronominal forms elsewhere by analogy. Who
Well, again this is one of the cases where I would first of all try
to examine historical scenarios. "nís" is really a marginal, with the
appearance of being absolutely archaic. Beside one instance in Amra
Choluim Chille and the possible one in Milan, both of which are
unclear, all three or four other examples occur in law texts which
are notoriously archaic. It is exactly those archaic texts where I
wouldn't expect re-analyses of the type mentioned by you. If (with a
really big IF) "nís" is the oldest form of the 3pl. of the copula,
then the standard forms "nídat" and "nít" would be based on re-
analyses, and they would simply have introduced the ordinary 3 pl.
> Unfortunately, he shed no light on the copula. All I
> could piece together was that has something to do with
> *de. But this doesn't explain the double 'd' in -did,
> or the 'i' in either
> -did or -id.
Indeed it doesn't.
> Any ideas? Sims-Williams 1984 seems to think that the
> -id part has been added to -d (from *de (esti)) as a
> later extension. But the opposite view is found in
> Thurneysen 1897 and 1946, which say that the initial
> 'd' was added to an original -id/-d, by analogy with
> the other persons of the singular.
> Very confused indeed this subject is.
Indeed it is. And the problem is that it is quite impossible to try
to tackle those copula forms on their own, without looking at the
greater picture of Irish historical verbal morphology, and without
reference to other possible particles (or rather: to THE other
infamous particle *eti). But as soon as you do that, you are treading
explosive ground :-)