En réponse à bnathyuw <[log in to unmask]>: > > > > still find it quite unlikely, but would like to see > the evidence as it could win me over. Not seen it unfortunately... i'ld have > thought the arguments allen uses ( both for the > voiceless rho ( altho he's not so convinced by |rr| > being /rr_0/ and for the alveolar rho ) are pretty > strong. he looks at ancient descriptions of the > phonology, as well as how the sounds were transcribed > in foreign languages ( and some of the middle eastern > languages would presumably have transcribed /R\/ as a > gutteral consonant rather than the /r/ they use ). Well, that may not mean much, since I've seen the French r transcribed by Arabic people with an alveolar trill, while they do have at least a voiceless uvular trill (and indeed Arabic people speaking French as a second language pronounce generally the 'r' as an alveolar trill instead of an uvular trill or fricative, something they have in their L1 and thus should be able to pronounce without a problem). It seems that when a language has only one rhotic, it is mapped by foreigners with what they consider to be their most common rhotic rather than with the sound that should sound nearest in their language. > also, the transfer of aspiration across the |r| in > phrouros > pro-horos suggests at least that /r/ ( or ) > /R\/ had a voiceless allophone. > That I didn't say was wrong. I even saw a description which explained that the only Greek rhotic was a voiceless uvular trill. > bn > > ( having just looked again at the site you cite ( > hmmph ! no pun intended ) i get the impression that > this is a recommendation of how to pronounce ancient > greek names if you're a modern english person. as > englishers may be more familiar with and able to > produce a french |r| than an italian or greek one, > this may be the origin. > > just a thought ) > I somehow doubt that nowadays English-speakers are more familiar with the French 'r' rather than the Spanish 'r'. Actually, seen that alveolar flaps appear quite frequently in English dialects (whether as an allophone of /t/ or an affected or regional way to pronounce /r\/) both in Britain and in America, I'd say exactly the contrary, and that the description of it as a French 'r' was specifically to prevent its pronunciation as a Spanish or Italian 'r'. Christophe. http://rainbow.conlang.free.fr Take your life as a movie: do not let anybody else play the leading role.