IE-Semitic connections

[Larry Trask]

On Fri, 5 Feb 1999, Rick Mc Callister wrote:

> Basque <z> is /s/, so the idea that it may have come from Romance
> *septe > *sepce, *sepci; *sepse, *sepsi and then metathesized to
> /saspi/ sounds sounds interesting. The problem is that [afaik] none
> of those forms are documented in Ibero-Romance --I don't know about
> S. Gallo-Romance. There is also the question of whether open /E/
> would go to /a/ in Basque. Larry Trask would know that.

Both Romance high-mid /e/ and low-mid /E/ are borrowed into Basque as
/e/ at all periods.  (Basque has only a e i o u/.)  After Romance /E/
and /O/ were diphthongized to /ie/ and /ue/, these too were generally
taken into Basque as /e/, as in <leku> `place', from some Romance reflex
of Latin LOCU of the approximate form *<lueco>.

Latin /pt/ was assimilated to */tt/ very early in Iberian Romance, and
there was no */ps/ stage.  The same thing happened in the ancestors of
French and of Italian.  I have no data for southern Gallo-Romance, but I
would be surprised if anything different happened here, since this kind
of cluster assimilation seems to have been a very widespread feature of
spoken Latin (Vulgar Latin).

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk