Inspired by this answer by Arnaud Fournet I have this question: Do Old Indic (Vedic, Sankrit) words beginning with a voiceless aspirated stop (like ph, th, or kh) have cognates in other branches of Indogermanic? What are typical regular sound correspondences?
asked May 8, 2019 at 14:26
-
2Wikipedia lists Sanskrit ph, th, kh as the regular correspondences for PIE *p, *t, *k before an original laryngeal– b aMay 8, 2019 at 14:41
-
1Well, yes, but that's circular. Laryngeals are posited in the first place because of their effect on surrounding sounds.– jlawlerMay 8, 2019 at 21:17
-
@jlawler It's not circular if there's independent evidence for the laryngeals in question, as in some cases there is.– TKRMay 9, 2019 at 1:14
Add a comment
|
1 Answer
As a rule, the voiceless aspirates are a dialectal feature. In most languages, they cannot be distinguished from plain voiceless. As regards *kh, the typical pattern is Old Indian kh, Armenian x, Greek k(h). For example, in a potentially onomatopeic verb: *kakh- "to laugh", where probably no laryngeal is involved. As regards *th, typically Old Indian th, Armenian t, Greek t(h), very often preceded by initial s- as in *st(h)eH- "to stand", cf. *pont(H)- "path, way through", *ost(H)- "bone". As regards *ph, typically Old Indian ph, Armenian ph, Greek p(h), a good example is *(s-)phol- "to fall".
-
1Wikipedia (for what it's worth) agrees with you on /kh/, and lists Skt खादति khā́dati :: OArm խածանեմ xacane, from PIE kʷh₂ed-. May 8, 2019 at 23:42
-
1What do you mean by "dialectal feature"? Does this indicate an irregular and sporadic correspondence, or can it be consistently traced to certain dialects? May 9, 2019 at 8:18
-
2It seems that the creation of a fourth series of consonants: the voiceless aspirated series, occurred only in a subset of probably contiguous languages: Greek, Armenian and Indo-Iranian. In the other languages, the phonotaxis of C+H did not create an additional series: voiceless C+H=C. The verb *kakh- "to laugh" has a variant in Germanic *hlah- < *klakh-, so Germanic is somehow close to this subset of "Central Indo-European" languages. So this isogloss is dialectal in the sense that only a few languages are involved.– user23769May 9, 2019 at 10:27