galaxyrocker wrote:
I wonder if it's one of those things where the relative particle aspect of it overcomes the copula part of it? Like would you say 'Cé ar thaitin sé leis' or "Cér thaitin sé leis'? I'd almost lean towards the former, since it's not part of the copula, tho I admit I'm not sure. I feel it could be the same case here?
I never thought about writing "cé ar" and I would tend to write always cér.
I'd think "cé ar" is old style orthography as "cé'r" and "cé'r'bh" is.
Phonologically, it is more natural to say
/k´e: r/ than
/k´e: ər/ because
/ə/ usually is silent next to long vowels. So, there is no difference in pronunciation between cér and cé ar.
Acc. to GGBC (13.80) cér just occurs as indirect relative in Cér leis and Cér díobh thú.
But GGBC's short explanations are not convincing at all.
What happens in other cases of cé + indirect relative? (e.g. Cé( a)rbh áil leis)
Why is
Cér leis é indirect relative at all? (Present tense
Cé leis é isn't.)
Why should
Cérbh é ... be direct (!) relative? (Present tense
Cé hé ... isn't. Direct relative of copula in past tense is ba/ab in other instances.
Other past tense copula forms in -r(b/bh) aren't relative: níorbh, gurbh, etc.
Acc. to FGB there is no such limitation to cér leis/díobh because there are other examples: Cér uaidh é? Cér chas é?
BTW:
Thanks to David who mentioned in an e-mail forms like
Cé hé gurb eól dó neart t’fheirgese? Who knoweth the power of thy anger? (P. Ua Laoghaire)
Standardized:
Cé hé arb eol dó ... So we get:
"Who likes" -
Cé hé ar maith leis ... "Who would like" -
Cé hé ar mhaith leis ..."Who should" -
Cé hé ar cheart dó ...etc.
Though such forms are still rare, they are at least used

E.g. ABN:
cé hé ar féidir dó a rá leis “Céard atá tú a dhéanamh?
(and who may say unto him, What doest thou? Eccl. 8)