>Had "Old Greek" a voiceless aspirated alveolar trill?
Vox Graeca refers to two grammarians (well, all right, one grammarian and
Plato) who describe what seems to be a trilled, alveolar sound. (so not
guttural as in French or some German dialects; not fricative as in
English).
W S Allen says "Generally speaking [r] is a voiced sound, but in certain
environments in calssical Attic it seems to have been voiceless." He bases
this on:
(a) descriptions at the time
(b) the Byzantine tradition of breathings (with which I assume you are
familiar)
(c) dialect spelling variations
(d) an Armenian loan word
(e) modern Tsakonian
He points to modern Iceland where a voiceless alveolar trill is phonemic,
and contrasts with voiced ones.
He suggests the r was also voiceless in sequences such as thr, phr, chr.
The aspiration is usually etymological, but not always. (Initial r in Greek
never comes from PIE initial *r)
Modern Greek has only a voiced r
Peter