Avestan fštāna Author(s): Eric P. Hamp Source: Zeitschrift für vergleichende Sprachforschung auf dem Gebiete der Indogermanischen Sprachen, 76. Bd., 3./4. H. (1960), pp. 273-275 Published by: Vandenhoeck & Ruprecht (GmbH & Co. KG) Stable URL: https://www.jstor.org/stable/40848057 Accessed: 13-11-2024 23:52 UTC JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms Vandenhoeck & Ruprecht (GmbH & Co. KG) is collaborating with JSTOR to digitize, preserve and extend access to Zeitschrift für vergleichende Sprachforschung auf dem Gebiete der Indogermanischen Sprachen This content downloaded from 152.118.24.31 on Wed, 13 Nov 2024 23:52:29 UTC All use subject to https://about.jstor.org/terms Avestan f Stana Von Eric P. Hamp, University of Chicago There is an interesting and well-known family of words for 'breast, teat9 for which the following may be mentioned: Skt. stánãu, Arm. stin, gen. stean, Gk. orrjviov arfj&oç Hesych.; Lith. spenys, OPruss. spenis, Oír. sine *spenio- (bó triphne cthree-teated cow'), ON. speni, OE. spann, OHG. spunne. On the basis of these forms, Porzig (Gliederung 163) assumes a Germanic-Baltic-Keltic *sp-, equating MHG spenen1) 'suckle', and considers this initial original. Specht (Dekl. 86) declares the st- to be „verhüllende Umgestaltung". I consider these judgments to represent, methodologically, steps backward so far as the relevant scholarship is concerned. Pedersen (KVG § 49. 1) also relates Arm. san 'Zögling9, but this may be left out of account as long as the other forms remain problematical and unagreed. We cannot however ignore the Avestan dual f Stana; Bartholomae (Altiran. Wb. 1030) also adduces modern Persian pistan 'breast, nipple9, but Hübschmann (Persische Studien 41 and 186, fn. 2) questions whether this is related to f Stana- or whether it is not rather *payastãna~. The related Pehlevi form is likewise ambiguous. Nevertheless, amongst the Pamir dialects we find Yidgha iScin and Munji yiSkvuna (Morgenstierne, Indo-Iranian Frontier Languages II, § 109, p. 73); these go back not to *st~, but to a mediate *$-, which must reflect the older fSt-, Again, we find (I-IFL II, § 56, p. 312) in Sanglechi St9nõk 'new-born kid9 < *fStanya-ka- (?), péStdn 'udder9 < *paya-fStana.Gershevitch (A Grammar of Manichean Sogdian, Oxford 1954, § 314. *) states: „In B. 9Stnh 'breast9 VJ 6a. 1028, Sogdian agrees with Skt. stana, against Av. f Stana-, NPers. pistan, cf. Meillet, BSL xxiii 106." However, it would seem from other paragraphs in Gershevitch's account, relating to the various Sogdian and Iranian *) Since MHG spen 'Brust' is probably < * 'spani, spenen may well be derivative, rather than coordinate, evidence. 18 Zeitschrift für vergi. Sprachf. LXXVI 3/4 This content downloaded from 152.118.24.31 on Wed, 13 Nov 2024 23:52:29 UTC All use subject to https://about.jstor.org/terms 274 Eric P. Hamp phonemes involved, that earlier *st observed St must reflect a mediat clusion is to be drawn from the fine (Essai de grammaire sogdienne I, § *f$). The prothetic vowel is of cour therefore have an apparent *$tana-. T forms, fits with the attested Avestan We might think of assuming for initial *s£- as in Sanskrit, an initia Reichelt (Av. Elementarbuch §17 trat*. Therefore, we are forced to ass an initial in part comparable to oth fSuyant-, Sogd. ßy'ws 'gentleman9, Bu Moreover, the internal treatment 'mit straffen Brüsten9 < *pstn (Re cluster for Iranian. That the form by the ablaut. There seems to be no methodologica from Bartholomae's (Grundriß § 83. 3) formula *(p)st(ë)no-. Trautmann (s. v.) assumes *psten- and *spen- side by side. See also most recently Pokorny IEW 990. In light of this, Porzig's alignments ignore the clear IndoIranian unity which obviously overrides the other dialect associations which he is discussing. If we accept, as I think we must, something like *pst(ë)no- for Indo-Iranian, as Bartholomae has posited, we now have opposed to this an Armenian and Greek apparent *stên-, and a European *sp-. Tentatively, I can understand this distribution only as two innovating patterns (of whatever details of development and origin) whereby Indo-Iranian is not directly involved as a major innovator. We might proceed, by internal reconstruction, to enquire how these initials might perhaps better be unified and understood. Indo- European permitted initial clusters of the form *pt- and *sC-, Therefore *spt- would be within the frame of the occurring pattern; all attested shapes could readily be deduced from this sequence with maximum economy. We know that Iranian was relatively retentive in the matter of initial consonant clusters : Av. f Sarama-, Sogd. Sß'rm- (for the metathesis, cf. the words for 'comb9 in modern Pamir dialects), Yidgha fSarm, Munji Sfordm, beside OCS srarm, Russ. sorom, etc. This content downloaded from 152.118.24.31 on Wed, 13 Nov 2024 23:52:29 UTC All use subject to https://about.jstor.org/terms Eric P. Hamp, Palatal before resonant in Albanian 275 This sequence remains, however, a pure construction until we can substantiate it. It is thinkable that we may see something of an earlier shape for the same base in Hittite iSpatar, gen. iSpannaS c Spieß (?)', but it should be noted that the forms iSpifa = i§pãi'sich satt essen9, i$pij,atar c Sättigung9 also occur. An old neuter *sp(e)t-r/*sp(e)t-n- would go a long way toward helping to throw light on our form. Perhaps we may see an ultimate relationship to forms listed in IEW, p. 981 or 983; but at present I see no clear choice. Palatal before resonant in Albanian Von Eric P. Hamp, University of Chicago 1. Albanian glun- 'knee9 The standard Albanian forms are Tosk gju, def. gjuri, pl. gjunjë. North Greg giù, def. giuni, pl. giuj. The dialects of Çamërija (south Tosk) and of the enclaves of Greece and Italy (exclusive of some of the Calabrian dialects, where the departure is normal) show an initial gì-. Borgo Erizzo (Dalmatian Geg) has (plural) gun. On the basis of the reflexes in North Greg and southern Tosk, the protoAlbanian form is clearly *glun~. This has long been known and accepted; it would be superfluous to adduce here the long list of separate dialect attestations that my files show. Wackernagel (Altindische Grammatik 3. i. 132) accepts the possibility that the Albanian form represents a dissimilation from *gnun-, and compares this to yovf at- : yóvv, parallel to Vedic drúnah, òoqFox- : dáru, ôóqv. The comparison is attractive in every way, especially when one considers that Albanian has also generalized *drun- > Geg dru, drûni, Tosk drü, druri. We may perhaps be troubled by the length of the vowel in Keltic : Oír. glún, Welsh Breton glin < *glünos (a neuter s-stem in Keltic; the Welsh plural glinyeu shows reshaping). It is hard to know just how this length fits; perhaps we have a cross with a laryngeal-extended stem *gnuX-, In any case we find a parallel long-vowel form in Albanian dri-zë ctree9 (the meaning varies in different dialects) < *drü-, plus the diminutive suffix -zë < *-diã. 18* This content downloaded from 152.118.24.31 on Wed, 13 Nov 2024 23:52:29 UTC All use subject to https://about.jstor.org/terms
