Re: [tied] tt/st/ss

Sean Whalen napisał(a):

> Latin does show tt>ss>s after r.
> I can't come to
> any other conclusion than an intermediate tt>st, which
> had similar (though probably not identical) metathesis
> in Italic, Celtic, and Germanic.

But PIE *-st- consistently remains -st- in those groups!

> There's no reason to assume a PIE stage of tt>tst.
> This would mean that d(h)t>tt already, but Baltic and
> Slavic show that devoicing before voiceless stops
> hadn't occurred yet (among others); Sanskrit shows
> dht>ddh.

Why?

(1) You can have affrication independently of devoicing, e.g. -d(H)-t-
realised as *-dz(H)-t- (or, as an alternative, restoration of voicing
through analogy in some branches).

(2) What BSl. evidence points to the preservation of voice before *t?

>> The 'wrist' example
>> is unconvincing, as
>> the underlying root is more likely *wreik^-.
>
> Why? This would assume *wrikY+s+ti+s, what's the
> first *s doing there? A meaning of 'bending forward,
> rolling' seems to allow a match with Lith. risti 'to
> roll'; but no other examples of *wrikYstis.

Semantically, Germanic forms like OE wraxlian 'wrestle', wrigian 'turn,
move', wrixl 'change' fit very well. The suffix -st(h2)-, whatever it
is, occurs also in the Germanic and Balto-Slavic words for 'fist' and
the Balto-Slavic 'finger' word (*pirs^ta- < *pr.sth2o-).

Piotr