On pg. 169 fn. 11 I claim that Go. aleina 'ell' which would normally be phonologized as /ali:na/ is a mistake for *alina with a short medial syllable. This may indeed be correct but there is some evidence that might support the reality of long i, viz. MW elin 'ellbow' which points to a proto-form *oli:na:. I see there is an article that I will have to read by Dirk Boutkan in Amsterdamer Beiträge zur älteren Germanistik Vol. 41, 1995 that seek to justify *ali:na for all Germanic.
Update: Boutkan makes a good case that the Germanic forms can be derived from a Proto-Germanic *ali:no:. He further argues that the word is a loan from Proto-Celtic *oli:na: which come from *ole:na: a derivative of the hysterokinetic n-stem continued in Gk. ὠλήν (usually ὠλένη). I'm not sure if that is the best way to handle these complicated data.
Update: Boutkan makes a good case that the Germanic forms can be derived from a Proto-Germanic *ali:no:. He further argues that the word is a loan from Proto-Celtic *oli:na: which come from *ole:na: a derivative of the hysterokinetic n-stem continued in Gk. ὠλήν (usually ὠλένη). I'm not sure if that is the best way to handle these complicated data.
Maybe you were right with your suggestion that got. aleina is a writal error for +alina < *o(:)lena:; the Celtic form could intead stem from *o(:)le:na:, as Boutkan thought. Difficult word.
ReplyDeleteBest, Sergio