> Since "lá/laa/lae" is neuter, that instance of acc. "lae" is
> ipso facto evidence for nom. sg. "lae", too, right?
Yes. But this reasoning would equally apply if the word were masculine.
> BTW, both "laa" and "lae" are disyllabic, right, like modern
> Scottish Gaelic "latha"?
> What do you think of Thurneysen's
> suggestion (GOI §284.3) that OI "láa/laa/lae" is a short form
> of "la(i)the"?
Thurneysen's suggestion that "lae" is due to dissimilation in the frequent combination "lathe brátho" "Doomsday"
seems a bit forced to me. In my eyes, "lae" is much too well established in Irish to be explainable from a single phrase -
especially in the case of a central word like "day". This would imply that the early Irish went around talking about the
end of the world all time. Even if we allow greater religiosity than today, this is rather unlikely. I would rather think of two
different derivations from a single root *la-, one with the suffix *yo-, the other one with a more complex suffix -tiyo-.