--- In
cybalist@yahoogroups.com, "bobbyisosceles" <nervous@...> wrote:
> The "hu-/du-" opposition meaning "good/bad" appears in a number of
> compounds. Most of the time, these compounds are appended with -
> as in "huiti-", "good dwelling" or the Modern Persian
> word "domand", "enemy."
>
> However, I've seen in Sanskrit a potential sandhi combination H->r,
> as in "Duryodhana" < (duh: + yudh-) meaning "bad/dirty fighter."
>
> With that in mind, is the word "draoj-/draug-", meaning "lie,"
> really a combination of duH + aoj/aox to mean "bad word/bad
> utterance"? Looking only at the Modern Persian "durugh" would lead
> one to that conclusion, but the Avestan and Old Persian forms
> didn't have that epenthesis. Is there such a thing as a "zero
> grade" version of duH that became "dr"?
So far as I know there isn't any such zero-grade form of PIE *dus- in Indo-Iranian.
Iranian never has /r/ in formative prefixes deriving from Proto-Indo-Iranian *dus^- 'bad, wrong, difficult, un-, -less'. Avestan has both unvoiced dus^- and voiced duz^- (which are never confused with dur-); Old Persian has dus^-, Sogdian ds^-, Parthian dwj-, Middle Persian both duj^- and dus^-, New Persian duz-. No /r/'s in sight here.
The dur- form of the prefix dus.- in Vedic (< Proto-Indo-Iranian *dus^-) is due to the normal rules of external sandhi. Final /s./ is changed to /r/ before a vowel or a voiced consonant (namely, before 'soft' sounds). There are a few exceptions (seemingly older compound words) in which dus.- becomes du:- instead of the expected dur-; on these forms, check out the book snippet at
http://tinyurl.com/yavr7d7
Finally, the rare form duH- of this prefix you mention above is found in Sanskrit only before /k/, /kh/, /p/ and /ph/. However, this aspirated form of the prefix is neither present in Iranian nor, if reconstruction has an inherent logic, in Proto-Indo-Iranian.
Best,
Francesco Brighenti