> 1. Did I see you?
> [...] the same as, and probably due to, the structure which is
> present in
> Welsh, where the same sentence could be rendered:
> 2. Wnes i dy weld di?
>> 2. Wnes i dy weld di?
>> with the preterite of 'gwneud' (vn: 'doing') as an auxilliary.
> I have been wondering whether this is really a good case of Celtic
> influence. While such constructions do indeed occur in Welsh,
> earlier than
> in English, they are only optional (as I understand it), whereas in
> English they have become compulsory.
We were discussing this same structure in Old Irish in January:
> do-génat do chlanna iar tain do-grés a cin d'imresain (SnR 1407-1408)
> 'Your children afterwards will always contest their guilt.'
> This is even more striking as it involves "do·gní"-periphrasis + a
> do- infinitive. As for classification, ordinary VNs and do-
> infinitives (= "do" + VN) should be kept apart to see if a pattern
> arises there.
Saltair na Rann (AD 988) predates the Norman invasion of Ireland, so
there is no possibility of English influence here. Question: is do-
periphrasis (aka "irrelevant 'do'" -- ;-) ) something that Old Irish
and Brittonic both had as part of their common Celtic patrimony?
McWhorter argues that such a structure is exceeding rare among the
world's languages, and only that fact that it has taken root in an
international language like English makes it seem common.