Re: Latin _ursus_ (was: HORSA vs. EXWA)

--- In cybalist@yahoogroups.com, "Tavi" wrote:

> --- In cybalist@yahoogroups.com, "Brian M. Scott" wrote:

>>> AFAIK, only a "satem" language would develop a sibilant
>>> from a former affricate,

>> PIE *TsT > PGmc. *ss

It's hard to find a sibilant affricative that demonstrably doesn't arise from palatalisation. German /ts/ is one of the few examples I can think of.

> I meant an *uncondicioned* development.

But the affricate /ts/ we are considering *is* a conditioned development! Would you accept Old French /ts/ > /s/? There are contexts where it did not arise by palatalisation.

> It's a pity nobody here has
> considered my proposal that Hittite hart(ag)ga- mightn't be the
> reflex of a "thorny cluster".

But < HAR-TÁG-GA-AS > /hartkas/ *has* a thorny cluster, namely -tk-.

(Having a space within the angle brackets < > prevents the brackets and contents disappearing when answering through the *groups* web interface.)

Richard.