🚀 go-pugleaf

In article <9969@fs3.cam.nist.gov>, koontz@cam.nist.gov (John E. Koontz X5180) writes:
> In article <1993Jan4.032130.12753@trl.oz.au>, jbm@hal.trl.OZ.AU (Jacques Guy) writes:
> |> ... if memory serves, Latin r --> s.
> 
> Sorry, it's the other way around, i.e., s > r (via z?) / V__V, cf.
> genus (nom. sg.) vs. generis (gen. sg.) (example by memory, Latin not my 
> strength)

Good enough, though.  Essentially *all* intervocalic -s- in Latin became
-r-.  (-ss- was, of course, maintained, since neither -s- was intervocalic).
The few examples of intervocalic -s- in classical Latin mainly reflect
earlier -ss-, simplified after a long vowel.  -ss-, in turn, mainly
reflects /tt dt dht/, all of which may already have been assibilated
in Indo-European.

The opposite change, [r] to a sibilant, is rarer but does occur.
French _chaise_ 'chair' (beside expected _chaire_ 'speaker's platform')
< Gk. Lat. _cathedra_ is the classic example. Assibilation also occurs
in some Slavic languages, where Polish <rz> represents [z^] deriving
from earlier /r/ in certain environments and Czech <r^> is a "rolled
post-alveolar fricative" deriving from earlier /r/.

--Leo Connolly