Re: Celtic, semitic, etc.

Thomas Leigh wrote:

>>A far as I know, there are very few scholars left who support evan a "Semitic
>>influence" theory, much less a "pre-Celtic substrate" theory.
And Raymond Brown replied:

>I had a feeling that 'twas so.
My two cents:  Speaking as an Austronesianist and a dabbler in
historical linguistics (I was conned into teaching a historical linguistics
course last semester and learned quite a bit in the process), I have to
say that I find the evidence for a historical connection between (insular)
Celtic and Semitic (or Afro-Asiatic) to be less than convincing.  IIRC, the
kinds of shared traits which are usually cited (VSO order, inflected
prepositions, articles, N-Adj order, etc.) are inevitably structural rather
than lexical.  While it's true that structural traits such as word order are
highly susceptible to areal influence, it's also true that certain structural
properties tend to cluster together in language after language, even when
those languages are demonstrably unrelated (or so distantly related that
the historical connection is irrelevant).  Observations along these lines
are what led Greenberg to posit all those word order universals.

Just about all of the properties which Celtic and Semitic are said to
share can also be found in other verb-initial languages all over the
world.  Malagasy, for example, has inflected prepositions, articles,
and N-Adj word order--and nobody would seriously claim that Malagasy
has a Celtic or Semitic substrate (or vice versa).

Matt.