On Mo, 9.02.2009, 15:54, Neil McLeod wrote:
> (The attestaions are all quite late. As the word is an o-stem, I wonder
> if it is a cpd with -crann, 'wood', perhaps a late form of *'fo-chrann',
This explanation, although not impossible as such, involves a number of
additional assumption: vowel assimilation *fochrann > *fachrann,
generalisation of the lenited stem variant without the "f" from phrases
like *isind fhochrunn "in the underwood"(which goes against the tendency
to rather introduce initial "f" even where it does not belong). This
explanation would also leave the variant with initial "e", i.e. "eachrann"
unexplained - this could be a hypercorrect spelling.
I assume that a compound *fochrann would have remained etymologically
transparent at all stages of Irish (or Gaelic). It's hard to see why it
should have developed away from its etymologically transparent form.