> >> acht óen ba mraithem,
> > Shouldn't it be acht óenar ba mraithem?
> I gave this some thought, since "óenar" would be my natural
> choice here, biased as I am toward Modern Irish. But DIL
> indicates that "óenar" is quite rare outside the dative usage
> with the possessive ("messe m'óenur" = I by myself, alone, etc.).
> Meanwhile, Kavanagh's Lexicon of the Wb. Glosses gives examples of
> "óen" used as an indefinite pronoun ("óin di airchinchib assiae" = one
> of the leaders of Asia), so I went with it.
I have nothing at hand to check myself, but maybe óen could be used as an indefinite personal pronoun if it was clear from the context (e.g. in a partitive construction) that
it referred to a person. Therefore maybe: ba óen diib ba mraithem ?
I agree with you that we moderns, being educated in ModIr., tend to overestimate the semantic range of the personal numerals in OIr. These are far more rarely used in OIr.
than in MIr. or ModIr. I noticed this when I wrote the relevant chapter for my "Introduction". Here is what I wrote there in paragraph 47.1.1. "Personal Numerals":
1. The numbers óenar and from triar to deichenbor are compounds of the cardinal numbers and a neuter o-stem abstract probably derived from the word fer 'man.' Attested
forms apart from the nominative singular are: gen. sg. oínair, prep. sg. oínur; gen sg. triir /t'r'i/r'/, prep. pl. trírib /t'r'ir'0b'/ 'in groups of three' (Fél. Prol. 210); cethrairib 'in
groups of four' (Thes. I 497, 16); prep. sg. cóiciur /kog'ur/.
2. dias 'two men, a pair' is a feminine a-stem: gen. sg. deisse /d'es'e/, prep. and acc. sg. diis, díis /d'i/s'/, prep. pl. deissib /d'es'0b'/ 'in pairs' (Fél. Prol. 210).
3. Numbers above 10 are expressed in the same way that we already learnt in chapter 12.4., i.e. only the digit is expressed by the personal numeral: dias ar fichit '22
persons,' deichenbor ar dib fichtib ar trib cétaib '350 persons (lit.: ten persons on twice twenty on three hundred).'
4. It would seem that in Old Irish the personal numerals were used to signify 'a group of persons,' specifically 'men.' If any other type of person was meant, apparently
ordinary numbers were used, e.g. trí maicc 'three sons,' trí (!) seithir 'three sisters,' trí soír 'three free persons' etc. in the Old Irish Triads. Another, perhaps younger,
construction was to use the personal numeral followed either by the counted word in the genitive plural, e.g. cethrur airech 'four noble-men' (LL 480), tánaicc Calcus + a
thriar ban 'Calcus and his three wives came' (St. Ercuil 1952), or in a partitive construction with the preposition deL 'of, from,' e.g. dechinbor ar dib fichtib ar tríb cétaib de
epscopaib '350 bishops.' This construction is not restricted to persons only, but can be used for animals and things also.
5. A very frequent use of the personal numerals is in the plain singular prepositional case immediately after a possessive pronoun, for example tusu t'óenur 'thou alone,'
at·taam ar ndiis i cuimriug 'we two are in bonds' (Wb. 32a28), a cethrur /a g'eqrur/ 'the four of them.'
6. The plain prepositional case of a personal numeral without a preposition or possessive pronoun before it signifies accompaniment, for example at·tá mí nád·n-imthet rí
acht cethrur 'there is a month when a king journeys only with four (companions)' (CG 535-536)
SnaG 206-207 and 262
David GREENE, 'Celtic,' in: Indo-European Numerals, ed. J. Gvozdanovic, Berlin and New York 1992, 516-520
Maybe I will have to reformulate this paragraph in view of what you noted above.
Merry Christmas everyone