puwa-
‘goat’
- Language
- Lycian A
- Grammar
- subst
- ID
- 982
1. Individual Anatolian Languages
1.1 Lycian A
puwa-
‘goat’1.1.1Transmission
This lexeme is attested once in an undated grave inscription, and it is also found as a personal name in TL 147.1 (Neumann 2007a:294, Schürr 2019a:562).
[D.S.]
1.1.2Forms
| nom./acc. coll. | puwa | TL 102.3 (Limyra) |
1.1.3Meaning
The noun puwa represents something that is to be paid to a precinct by the potential offender. Based on parallel formulas, it is expected to represent some kind of sacrificial animal. Melchert (DLL:53) suggests from an etymological perspective the meaning ‘goat’ (followed by Schürr 2019a:560-564, Christiansen 2020a:249). The different interpretation suggested by Neumann 1974a:109 to take me=ttleiti puwa as me=ttleiti=pe uwa, though syntactically possible (contra DLL:53), is not convincing on combinatorial grounds, because the emphatic particle =pe is not used in the parallel formulas. Gehrisch 2018a:142f. suggests instead separating the sequence into me=ttleiti (e)p(i) uwa and translates ‘bulls’, which is not supported by comparable formulas and, thus, not persuasive. Outdated Carruba 1974a:580 n. 7.
[D.S.]
1.1.4Stem
While Melchert (DLL:53) suggests that puwa is a collective formation derived from a thematic noun in -e(/i)-, i.e. puwe-, Schürr 2019a:562 suggests that it is a collective to a common gender puwa-. Both assumptions are possible, but cannot be proven until further attestations are found.
[D.S.]
2. Reconstruction
2.1Reconstruction
Melchert (DLL:53) suggests an etymological connection with Av. būza- ‘goat’ and reconstructs the pre-form *bhūg̑o-, which is phonologically fine, because *g̑ is lost in Luwic. Schürr 2019a:560f. accepts this etymology and adduces more Indo-European cognates. Furthermore, he suggests a new inner-Anatolian cognate, i.e. Luw. *puntarr- ‘stubbornness’ continued in puntar(r)iye/a-(mi) ‘to be stubborn’ (found only in Hittite Transmission). He argues that this etymology for the Luwian cognate is semantically superior to the one Starke 1990a:378f. suggested, i.e. Hitt. pūwā(i)-(mi) ‘to crush’ (welcomed by HED PE/I/U:120, not mentioned in EDHIL:681).A different but far-fetched etymology is suggested in HEG P:649.His etymology is indeed attractive from a semantic point of view and cannot be refuted on phonological grounds. However, the question arises whether Luw. *puntarr- is derived from the noun denoting an animal, i.e. *puwa- + -ntarr-, or whether both Lyc. puwa- and Luw. *puntarr- are derived independently from the same (verbal) root; cf. HLuw. (VIA.PUGNUS)pu- ‘to take, hold’ (Melchert 2016a:211).
On the other hand, Simon 2020a opens the possibility of seeing different semantics of Luw. *puntarr-, which would speak against Schürr’s etymology; cf. under lemma.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
Based on the meanings of the cognates listed sub Transmission above, the original meaning of PIE *b(h)ug̑o- can be reconstructed as ‘he‑goat’ (cf. IEW:174).It is not possible to decide on the basis of external facts whether PIE *b(h)ug̑o- is the wild or the domesticated goat. Based on the tentative etymology presented here, it may be deduced that this kind of goat is already domesticated, because calls to summon animals are not used before their domestication. — In some languages the meaning was broadened to males of other animals, cf. ModE buck ‘male deer’, ModHG (Reh-)Bock ‘roebuck’, etc.
3.1.1Root etymology
It is hardly possible to reconstruct the meaning of a root underlying Lyc. A puwa- ‘goat’, because the only known cognates (cf. Transmission above) go back to a PIE noun meaning ‘he‑goat’. Due to the lack of a corresponding root, it may be tentatively presumed that the word for ‘he‑goat’ under investigation is of onomatopoetic origin and based on a call used to summon goats, cf., e.g., EWAhd II:218 with reference to Rohlfs 1925a:671f.Cf. also the clear example Germ. (Bavarian) dig ‘(interjection) chick! (call)’ → ‘(sbst.) chicken’, diminutive pl. digai ‘(young) chicken’ (BWB III:1711.) and the interjection Germ. (Bavarian) Bock, Bogi-Bogi (“Als Lockruf für Ziegen”, BWB II:1491).
3.2Morphological Reconstruction
Phonologically, Lyc. A puwa- ‘goat’, YAv. būza- ‘he‑goat’, and Arm. bowc ‘lamb’ form an impeccable equation, pointing to a proto-form PIE *bhug̑o- ‘he‑goat’ (cf. EIEC:229).Whether PGerm. *bukka(n)- ‘he‑goat’ is a loanword from PCelt. *bukko- ‘id.’ or vice versa and whether the geminate has to be explained by sound laws or by “expressive gemination”, remains a matter of debate, cf. the literature above sub Transmission with further references; the same is true for OInd. bukka- ‘goat’.The latter can be interpreted as a thematic stem derived from a root PIE *bheu̯g̑- (or *bhu̯eg̑‑), which is not a root listed in LIV² or IEW. Alternatively, it may be understood as a thematic derivative from an interjection PIE *bug̑! (cf. Root etymology). On the basis of the facts presented here, it seems best to reconstruct a noun stem PIE *b(h)ug̑o- ‘he‑goat’.
3.2.1Root etymology
Given the fact that there is no independent evidence for an underlying root PIE *bheu̯g̑- (or *bhu̯eg̑‑), it seems legitimate to look for another way to etymologize PIE *b(h)ug̑o- ‘he‑goat’. EWAhd II:218 refers to Rohlfs 1925a:671f., who concludes that (Ital., Arab., etc.) calls for summoning goats often consist of an anlauting voiced labial, a vowel, and a (pen‑)ultimate velar, hence bVK(V). This coincides with the fact that all cognates of Lyc. A puwa- synchronically attest anlauting b‑; cf. also the otherwise doubtful OInd. bukka- ‘goat’ above.Cf. also Ital. becco, Fr. bique, Span. boque ‘he‑goat’, which are not mentioned above sub Transmission but by Rohlfs 1925a:671. Furthermore, Blažek 2013a:45 lists some allegedly genetically related words from non-PIE languages like (North) Beja bok m. ‘he-goat’.Thus, it may be proposed that the word-initial consonant was rare PIE *b- and did not undergo sound change in Germanic and Armenian, which is a valid assumption in cases of onomatopoeia; cf. also PAnat. *u̯āi- ‘woe’. On the other hand, the second consonant, which is reflected as Lyc. A ‑w‑, YAv. ‑z‑, Arm. ‑c, and PGerm. *‑k(k)‑, clearly underwent sound change and is thus reconstructable as PIE *g̑. Therefore, the presumably underlying interjection might be reconstructed as PIE *bug̑! ‘(call to summon goats)’.
[A.O.]
4. Transmission
(?)OInd. bukka- ‘goat’ (Lex.), Hindī bokā ‘id.’, etc.According to EWAia III:358, the Indic lexeme onomatopoetically imitates the bleating of sheep but may also ultimately go back to OInd. *bhuja- (= YAv. būza- ‘he‑goat’), cf., e.g., KEWA II:436.
EWAia III:358, KEWA II:436f.
YAv. būza- ‘he‑goat’ (hapax, Yt.14.25),The long vowel of YAv. būza- can go back to PIE *‑u- (or *‑ū- < *‑uH‑), cf. de Vaan 2003a:284–291, §§10.2, 10.2.1.buziia- ‘of the goat’ (hapax, V.5.52), MPers. buz ‘goat’
Bartholomae 1904:969f., de Vaan 2003a:288, MacKenzie 1971a:20
Arm. bowc ‘lamb’According to Klingenschmitt 1982a:183, Arm. bowcane- ‘to feed’ is connected with PIE 2.*bheu̯g- ‘to be of use for sb.’ (LIV²:84f.).
EDAIL:193
PCelt. *bukko-, cf. OIr. bocc, MW bwch ‘he‑goat’The Celtic word is regarded as a loanword from Germanic by EDPG:82, Kroonen 2011b:154, and Schumacher 2007a:174–176; undecided EWAhd II:217; vice versa Orel 2003a:61.
EDPC:83, Henry 1900a:40, Schumacher 2007a:174–176
PGerm. *bukka(n)- ‘he‑goat’, cf. ON bukkr ‘id.’, OE bucca ‘id.’, ModE buck ‘male deer’, OHG bock ‘he‑goat’, ON bokki ‘he-goat, person of influence’Maybe < *bhug̑‑no- via Kluge’s Law (cf. Kroonen 2011b:154) or onomatopoetic gemination (cf. EWAhd II:217 and above on OInd. bukka‑).
EDPG:82, Orel 2003a:61f., de Vries 1977a:48, 64, EWAhd II:216–218, AWB s.v. boc, Holthausen 1934a:38, Bosworth & Toller 1898a:132, Bosworth & Toller 1921a:110