--- In
cybalist@yahoogroups.com, "Richard Wordingham" wrote:
>
> > > Matasovic´:
> > > Another, less probable explanation would involve positing
> > > a special rule *h2rC > *arC before C[r]C > CriC
>
> > That's right. In fact, CRC > CaRC (where R stands for a resonant)
but
> > CLCstop > CLiCstop (where L stands for a liquid). Also notice
> > Matasovic´ made a typo (quoted verbatim by Brian).
>
> What's the typo? As I read the quote, the general rule is CrC > CriC,
bled by some other change in this case. However, the 2011 correction by
Matsovic' reads:
>
In the last line he wrote "CRC" instead of "CrC". Perhaps a Freudian
slip?
Although in the whole I regard Matasovic's as a good work, there's still
plenty of room for improvement. Leaving aside the cases of faulty "PIE"
reconstructions like this one (I don't blame him for adhering to the
standard model), there's a number of points I'm critical of:
- use of /f/ instead of /F/ "phi" in Proto-Celtic reconstructions.
- listing of pre-verb compounds as separate entries.
- lumping of Celtic words with a loose semantic connection but
phonetically similar under the same lemma (homonymy). There's also a
milder instance of this at the Proto-Celtic level.
- cases where Continental Celtic (Gaulish) differs so much from Insular
Celtic, making impossible the reconstruction of a single protoform.