I never thought my question about the meaning and origin of the Indo-European perfect/stative would stimulate this much discussion. It's fascinating to read, however, and informative.
I have another question. This is one that I submitted to the website "Ask-a-Linguist" (from which I received instructions to join LinguistList and thence Cybalist), but the answers that I got simply said that the answer to my question is unknown. Perhaps someone on Cybalist will be able to shed more light on this question.
My question is, what was the probable pronunciation of Proto-Indo-European "r"? Specifically, was it pronounced trilled, with vibration, as in the majority of today's languages, or was it more like the commonest English pronunciation of "r", an alveolar or retroflex approximant, with no vibration? I know that English is rather special in preserving the original pronunciation of PIE "w" (although I hear that Flemish Dutch and a variety of Danish also preserve this pronunciation), and I wondered whether English was also alone in preserving the original pronunciation of "r", or whether English changed the original pronunciation of "r".
I have found evidence to support both possibilities. On the one hand, it is well-known that a former /z/ sound (from earlier /s/) became something written "r" in Latin and Germanic. Now I can see how /z/, a fricative continuant, could evolve into a continuant approximant, i.e. English-style /r/, but to me it does not seem likely that the /z/ sound would (as a first step) become a non-continuant vibration, the /r/ of most of the world's languages. If this /r/ from /z/ was the same /r/ as that from PIE /r/, it suggests to me that PIE /r/ was a continuant, more like common English /r/.
Moreover, PIE /r/ could also function as a vowel, in the same way that /j/ (or /y/) and /w/ corresponded to the vowels /i/ and /u/. It seems to me more likely that a continuant would be used as a vowel, rather than a vibration. North American English has such a vowel in words like burn, bird, earth, herd, etc.
Finally, according to several scholars and even one Sanskrit Grammarian (Panini), Sanskrit "r" was pronounced "made with the tip of the tongue turned up into the dome of the palate. It thus resembles the English smooth r, and, like this, seems to have been untrilled" (so according to William Dwight Whitney, in Sanskrit Grammar, and corroborated by German sources; his conclusion is partly based on the influence Sanskrit /r/ had in the euphonic processes of the language). This would suggest that the (common) English pronunciation of /r/ is ancient. (However, Whitney points out that other Sanskrit grammarians describe Sanskrit /r/ as being made "at the roots of the teeth", i.e. at the position of vibrated r, but no mention is made of any vibration by any of these grammarians.)
On the other hand, however, there is some evidence (as it seems to me at least) that PIE /r/ was a trilled sound. One is that the combination *sr became /str/ in many languages (e.g. English stream). This suggests that the *r in this combination was trilled (I think that if you try pronouncing /sr/ with English untrilled r, you will see that it seems unlikely that a /t/ sound would be inserted, whereas /sr/ with a trilled /r/ has an acoustic resemblance to /str/ with trilled /r/ -- though I may be wrong on this one). Another is that the combination *wr- existed in initial position. If one pronounces /wr/ with English untrilled /r/, it is rather difficult to detect a /w/ before the untrilled /r/. But if the /r/ is pronounced trilled, one can readily discern the /w/ before it. This suggests that PIE /r/ had to be trilled.
There is also some suggestion that English, in dialects besides Scottish, once had a trilled /r/ that subsequently became untrilled. This is the development of the medial combination /dr/, which became /dhr/ (dh = English th in there) in most words in English, e.g. mother, father, gather, weather, from Old English modor, faeder, gaderian, weder, respectively. If one pronounces /dr/ with untrilled English /r/, one can see that this combination seems stable and relatively easy to pronounce. However, if one pronounces /dr/ with a trilled (non-uvular) /r/, one can see that the /d/ is liable to change to /dh/ (th in there). This suggests that English /r/ was formerly trilled, and thus is not inherited from PIE.
Does anyone have any decisive comments to make about this? I.e., is English untrilled alveolar or retroflex /r/ another relic of PIE like /w/, or is this untrilled /r/ a comparatively recent innovation, and PIE had trilled /r/? I would like to know if English is more conservative than other languages as regards this consonant /r/, as it is with the consonant /w/.
Andrew Jarrette