Neil McLeod wrote:
> David Stifter wrote:
> > "bú" is only the acc. pl. The nom. pl. is not attested before the
> > Middle Irish period. It's "baí" then, we would it expect to be "boí"
> > in Old Irish.
> Could you elaborate on what you mean by 'not attested before the
> Middle Irish period'?
> The nom. pl. of bó is attested several times in Crith Gablach (c. AD
> 700). The forms there include:
> bae (CIH 778.24), ba (777.26, 777.34), bai (CIH 779.12, 779,29).
> Perhaps all of these have been modernised by later scribes, but the
> rest language was generally left in Old Irish form.
I wasn't aware of the CG forms, but they fit in the overall picture.
The reconstructable Common Celtic nom. pl. of "cows" is *bowes. This
should have given *boí in Classical Old Irish. As far as can be seen
this form is not attested, but what we find are forms like the ones
you cited from CG in texts that surely date to the OIr. period. Forms
like "baí" are quite obviously modernisations by later scribes: the
confusion of the diphthongs "oí" and "aí" is a typical feature of
Middle Irish (or perhaps Late Old Irish, but I am not sure we can
date it precisely).
> Thurneysen (p 217) gives the Old Irish nom. pl. as 'baí'. He does
> state that it developed from *boí,
What Thurneysen presumably means is what I explained above.