Re: écagne ,and ConLand names in translation(was: RE: R V: Old English)

At 04:29 PM 4/3/2000 -0500, Thomas Weir wrote:
>Raymond Brown wrote:
>
> > In the Hellenistic period the pronunciation of zeta was clearly [zz] (as
> > word initial one assumes just [z]).  In the earlier period it appears to
> > have been varied between [zz] and [zd] according to dialect.
>
>That's something that's always struck me as counterintuitive:  why would
>a protoform like *[dyeus] become [zdeus], rather than [dzeus]?  The latter
>is a straightforward example of palatalization (cf. Japanese t --> ts / _u),
>while the former, unless you invoke haphazard metathesis from an originally
>palatalized [dzeus], requires you to explain the appearance of that [z] from
>somewhere else in the system, which AFAIK would be very problematic.
That's bothered me too since I first read about the [zd] pronunciation, but
I think I have a plausible explanation. From what I've read, some Greek
dialects had [dz] instead. This would be a not surprising outcome of
palatalized *[d]. I believe that the original affricate was *[dz], and in
most Greek for some reason it metathesized to [zd]. Why would it do that?
Well, I don't know, but it's not unheard of; Hebrew and Aramaic at least,
and IIRC Arabic too (so perhaps all Semitic langs) had a regular metathesis
when a coronal stop immediately preceded a sibilant, for example *[tS] >
[St]. Note also [sk] from [ks] in Middle Eastern variants of the name
Alexander, such as the city Iskerderun (the al- probably dropped off
because it was seen as the Arabic article).