Re: [tied] aks.ara, Sanskrit; is there an IE cognate?

There are several certain cases where the correspondence Skt. ks. : Gk. kt/pt/khth/phth results from metathesis in *TK clusters (the "earth" and "bear" words are among the best evidenced). There are also some less certain one, and my personal opinion is that *g^Hs- or *gWHs- should be admitted as the source of ks.- : khth-/phth-. One such case is the "yesterday" word. I am not sure into which category the "perish" word falls; my *gWHs- is a tentative guess.

The "carpenter's root" was very likely just *tek^s-. This is what nearly all branches suggest. The odd man out is Greek, where <tekt-> may have been back-formed from <tekto:n>, which in turn is perhaps a by-form of older *tek^so:n -- possibly a case of analogical rule inversion motivated by historically justified cases of -ks- alternating with -kt-.

Piotr

----- Original Message -----
From: petegray

To: cybalist@yahoogroups.com

Sent: Wednesday, June 20, 2001 8:36 PM

Subject: Re: [tied] aks.ara, Sanskrit; is there an IE cognate?

The word in question (Skt a-ksara ~ Gk phthisis) is one of a handful discussed in the literature, where Skt ks. corresponds to Greek kt, pt, khth, or phth. At one stage it was suggested that PIE had the phoneme /T/, called in English "thorn". Hence the title of an article of Schindler in Die Sprache: "A Thorny Problem". His analysis shows the following patterns: Pokorny Vedic Avestan Greek k'T ks. s^ kt kT ks. xs^ kt kwT ks. - pt g'hT ks. - khth gwhT ks. Gz^ phth The best known of these are the "earth", "fish", "yesterday", "bear" roots. The consonant following the root vowel (ie at the end of the root) is never a stop. This is also true for all roots in Greek where pt alternates with p, or where Greek pt corresponds to p outside Greek. Hittite apears to have tk or tek Toch A has tk, Toch B k or ks Armenian collapses these with *ks Albanian has dh, dj Baltic and Slavic collapse the k' forms with the k's, and the other words are not attested. Germanic dehsa (axe) suggests -ks- < (Pokorny) *tek'T while guman, gestern point only to a guttural. Celtic: the regular reflex is a /t/ Italic sitis, texo, ursus suggest *ks humus, homo, heri only the guttural appears. Greek : the dialects occasionally show ks or sk in alternation with kt Schindler then goes on to analyse the various options for what was in PIE, but his conclusion is out of sync with Venneman's analysis in the "New Sound of Indo-European" (1989). Venneman concludes that we simply have a metathesis of an original *tk, *dgh, or without metathesis a straightforward development from *gwdh, or *gti. In particular the phthisis ~ ksiti- ~ a-ksara word is taken back to *gwdhi-. Does that answer the question? Peter ----- Original Message ----- From: "Piotr Gasiorowski" <gpiotr@...> To: <cybalist@yahoogroups.com> Sent: 19 June 2001 23:43 Subject: Re: [tied] aks.ara, Sanskrit; is there an IE cognate? I think aks.ara = a-ks.ara- 'not melting away, imperishable' (opposite of ks.ara-), hence 'something permanent, record, writing', a derivative of {ks.ar-}, cf. ks.arati 'flows, melts away, vanishes'. The latter often regarded as related to Gk. phtheiro: 'destroy, spoil' (< *phther-jo:), phthora: 'destruction, decay, deterioration', though the semantic evolution behind these meanings is not quite clear. The protoform would apparently be something like *gWHs-er-, maybe ultimately cognate to *gWHs-ei- (with a different "root extension") as in Skt. ks.in.a:ti/ks.in.oti 'destroys, exhausts' Gk. phthin(w)o: 'decay, perish; destroy', and in the celebrated adjective (Gk. aphthito-, Skt. aks.ita-) of the poetic formula *n-gWHsitom k^lewos 'imperishable fame'. Further cognates, anybody? Piotr ----- Original Message ----- From: S.Kalyanaraman To: cybalist@yahoogroups.com Sent: Tuesday, June 19, 2001 11:23 PM Subject: [tied] aks.ara, Sanskrit; is there an IE cognate? aks.ara = unalterable; sacrifice; sword; a letter, a vowel, a sound, a word (Skt.)

Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.