*h₂éu̯h₂-eh₂-/*h₂uh₂-éh₂-

This lexeme is attested 15 times in the HLuw. corpus with the relevant inscriptions originating from Maraş and Tabal in the centre, Amuq in the south, and Karkamiš in the south-east of the HLuw. area. The oldest inscription, dating to the 13^th century BCE, is found further to west in the Konya Plain. The bulk of the texts dates to between the 10^th and the late 8^th centuries BCE.

[AH.B.]

The meaning ‘grandfather’ was assigned to the present lexeme at an early stage in HLuw. studies (already Meriggi 1937b:97, followed by Bossert 1946a:144) on both a contextual and an etymological basis. For one, the word frequently appears in various Anatolian languages in contexts with other kinship terms like ‘father’ and others from the male line as well as in pairs with ‘grandmother’. For another, the etymological connection between the individual Anatolian languages is very transparent regarding this short lexeme, and so is the connection to the wider IE family.

The assigned meaning is borne out by the HLuw. contexts, e.g. in MARAŞ 4, §8: mi-pa-wa/i-tà tá-ti-i AVUS-ha-ha LEPUS-pa+ra/i-hi sa-ta, approx. /ammi=ba=wa=ada tadi huha=ha tabarahi asta/, meaning ‘But to my father and grandfather there was authority’.

A particular combination is attested in YALBURT frag. 4, §3: a-mi-zi/a (VIR₂)TÁ AVUS-zi/a, approx. /amminzi tadinzi huhanzi/ or perhaps even compounded as /amminzi tadi=huhanzi/, ‘my fathers (and) grandfathers’, mimicking closely the Hittite equivalent atta- ḫuḫḫa- ‘father (and) grandfather’ (see e.g. Martínez Rodríguez 2019b:189).

[AH.B.]

The attested forms show that the word is a common gender a-stem and not subject to i-mutation, cf. e.g. Hawkins, Morpurgo Davies & Neumann 1973a:26-28.

[AH.B.]

1. /huhahanna-/

1. /huhad(i)-/

The only occurrence of this word comes from a MS fragment of the Kuwattalla tradition.

[E.R.]

Checked against the photographs in the Mainzer Fotoarchiv. – Laroche’s (DLL:46) reading ḫu-u-ḫa-x was completed in Starke 1985b:126 with -t[i], which is fully compatible with the traces.

[E.R.]

Laroche’s (DLL:46) proposal to identify the word as the Cuneiform Luwian equivalent of Hitt. ḫuḫ(ḫ)a- (sic) gained full credibility when Starke 1985b:126 completed the reading to ḫu-u-ḫa-t[i] (abl./instr.), which made obvious the syntactic parallelism with the abl./instr. na-a-ni-ia-t[i] ‘brother’, another kinship term. Support comes also from the cognates recognized in Hieroglyphic Luwian, Lycian and Carian (see below). For the translation as ‘grandfather’ established in the handbooks, see CLL:71, HW² Ḫ:638, HEG A-K:260, HED H:357, EDHIL:353.

[E.R.]

The assignment of the common gender is based on the meaning referring to a human individual, but the form of the abl./instr. does not allow to pin down the stem class. Thus, both a mutating -a/i- and a non-mutating -a- are possible; of these only the latter is compatible with the a-stem suggested by the spelling of the attestations of the word in Hieroglyphic Luwian and by Lycian.

[E.R.]

1. ^mḪuḫa-zalma-
2. ^mḪuḫa-ziti-
3. ^mḪuḫa-nani-
4. ^mḪuḫa-šarpa- (Laroche 1966b:70, HEG A-K:261)

Rejected compounds:

1. *ūḫḫa-...

1. ḫūḫadalla/i-

This lexeme is found in decrees/dedicatory inscriptions from Limyra and Xanthos as well as on the inscribed pillar of Xanthos, dated contextually to shortly after the ascension of Artaxerxes II in 405-404 BCE and stylistically to around 400 BCE (Borchhardt & Eichner 1997-1999a:19f.). It is also found in personal names, see below.

[D.S.]

Neumann 2007a:137f. adduces the word ]uga in the grave inscription TL 96.1, which one could restore as [x]uga. This is indeed attractive, but unverifiable due to the badly preserved state of the inscription; cf. Kalinka 1901a:73. For the attestation xugahi in the new inscription N 337, see Christiansen 2012a.

[D.S.]

The spelling xux- (see also in personal names) instead of xug- is not surprising, considering the vacillation in aga vs. axa ‘I made’, which indicates an ongoing phonological merger of g and x; cf. Eichner 1988a:146f.

[D.S.]

The contexts in which xuga- is attested have always made clear that it is a kinship term. Pedersen 1945a:25f. equated for the first time Lyc. xuga- with Hitt. ḫuḫḫa- (and Lat. avus) and thus translated it correctly with ‘grandfather’ (followed by Neumann 1961a:51, 70, Carruba 1978a:864, Oettinger 2002a:97, Melchert (DLL:86), Neumann 2007a:137f., Christiansen 2012a:145, Shevoroshkin 2012a:471, Martínez Rodríguez 2019b:190); cf. Lyc. xn͂na- ‘grandmother’.

[D.S.]

1. Epn͂xuxa- (PN) (Carruba 1969c:270, 274f.)
2. Apn͂xuxa- (PN) (Seyer & Tekoğlu 2009a:222)

This lexeme is found in the inscription of the inscribed pillar of Xanthos, dated contextually to shortly after the ascension of Artaxerxes II in 405-404 BCE and stylistically to around 400 BCE (Borchhardt & Eichner 1997-1999a:19f.).

[D.S.]

Kalinka 1901a:43 reads xug̣[ … ] (followed by Friedrich 1932a:68). Shevoroshkin 2012a:471 reads and restores xu[sttite] instead of xug[a-. Schürr 2018b:59, 87 reads xu[ and leaves the restoration open as either xu[ga- or xu[pa-. However, the letter g̣ can still be partly seen in the autography of Kalinka 1901a:41, whereas Schmidt 1868a:Taf. VII.2 draws only a vertical stroke. Therefore, the reading xug̣^?[ is accepted here.

[D.S.]

For the semantic elucidation, see under Lyc. A xuga-.

[D.S.]

Reconstructed from Carian personal names attested in Egypt (Memphis and Luxor / Thebes, also C.xx 1 is probably from Egypt, see s.v. qtblem), but also known from Greek transmission as Γυγος and Ιδαγυγος (Adiego 1994a:33, 35, 37f., Adiego 2007a:334, 361, 408; cf. Zgusta 1964a:191 §451-4; Blümel 1992a:12, 14; LGPN V.B:93, 208). There is also a widespread view that the name of the Lydian king, Gyges, is identical to this Carian name, regularly adapted in Neo-Assyrian as Gūgu > Guggu (Aro-Valjus 1999, on this internal and regular formal variation in Neo-Assyrian see e.g. Simon 2020f:244 n. 15 with refs.) and in Lydian as ‹kuka-› (Adiego 2007a:334f.; Yakubovich 2010d:95; Melchert 2013c:35; Garnier & Sagot 2020a:179; due to semantic reasons Schürr 2019d:6f. remained sceptical [see under Meaning] and it was rejected by Sasseville s.v. kuka-, since he sees an inherited word in ‹kuka-›, which, however, does not explain the Neo-Assyrian and Greek renderings; only for the sake of completeness shall the rejection by Hawkins 2013a:9 without a single argument [“speculative”] be mentioned). The Carian forms date to c. 650-470 (on the Memphite ones see Kammerzell 1993a:163‒172, the graffito from the tomb of Montuemhat in Thebes dates around the mid-7^th c. BCE [during the rule of Psammetichus I]; and the date of C.xx 1 to the 6th c. is not assured, see s.v. qtblem; the Luxor one cannot be dated), the Greek ones to the 5th c. (and perhaps to the 4th c., LGPN V.B:93, 208).

[Zs.S.]

The attestations follow Adiego 2007a:361, 408, 419 (with refs. to forms that were misread earlier, see Adiego 1994a:38), checked against the photographs of the inscriptions, if available (O. Masson 1978a:Pl.VII.2 [E.Me 17], Pl.XX.2 [E.Me 43a], Pl.XX,I.1 [E.Me 44a]; E.Lu 2 was published only as drawing [The Epigraphic Survey 1998a:Plate 206 (“GR 21”)]; E.Th 44 is unpublished). Adiego 2007a:106f. proposed to read xr-quq as š^?r-quq, but according to the drawing (no photograph is available, as per above), the first sign, which is clear and not fragmentary, is not an ‹š› (the lower diagonal cross-bar is missing). Unfortunately, no other similar sign could have been found and the only attested fitting Carian sequence is indeed šrquq and thus, one can only speculate about a local sign form. All in all, Adiego’s proposal is plausible, but cannot be proved yet.

[Zs.S.]

Adiego 1994a:33 (here with a question mark), Adiego 2007a:334 identified it with the Anatolian word for ‘grandfather’ (Cuneiform Luwian ḫūḫa-, etc., see s.v.). Although this is based only on the formal assonance, it is generally accepted (Yakubovich 2010d:95, Simon 2011b:543 [but cf. below], Melchert 2013c:35; Brosch 2016a:7f.; Serangeli 2020a:207). However, based on the forms in Greek transmission there exists also an alternative explanation since the early research, which proposed to identify *quq with a water-bird based on the Greek word γύγης ‘a waterbird, perhaps ‘Eurasian bittern’’ (see s.v., Fick 1909b:12 and Neumann 1961a:70; cf. also Schürr 2019d:6f.). Nevertheless, as Nocentini 2012a:318 (cautiously followed by Simon 2019a:297) pointed out, these explanations do not necessarily exclude each other, since there are bird names based on similar terms (barbagianni ‘barn owl’ or compère-loriot ‘oriole’). A secondary bird name is, however, not very probable, since these bird names are not based on the word ‘grandfather’ and they are compounds. Having said that, this does not exclude that the meaning of *quq refers to a bird. Since both animal (bird) names and kinship terms are used in Anatolian onomastics, both are equally possible. A solution is provided by the compound names: although the meaning of d- in dquq is highly problematic (see s.v.), that of šr- is clear, a local adverb ‘up’ or ‘above’ (see s.v.). Such an adverb can hardly be combined with a bird name, but place adverbs do appear in Anatolian onomastics in combination with kinship terms (Melchert 2013c:40, he interprets šr-quq as “super/hyper-grandfather”), and thus, the meaning ‘grandfather’ is preferred here, too.

[Zs.S.]

Rejected cognates:

1. γύγης

The lexeme is only found as a personal name attested on coins (LW 69): kukal=iṃ (Weidauer Nr. 114), [k]ụkal=ịṃ (Weidauer Nr. 115), kukal=iṃ (Berk, Ltd. Nr. 2), [k]ukall=i[m] (Numismatic Fines Arts Inter., Inc., Auction 30 (8 Dec. 1992) Nr. 95) (see Browne 2000a, outdated reading as rkalim in Gusmani 1980a:152, Melchert 1994a:332 et passim, Browne 1996a). It has been correctly equated with the name of the Lydian ruler Gyges, i.e. kukal=im ‘I am (a coin) of Gyges’ (Browne 2000a:178, Yakubovich 2017b:288; cf. Dale 2015a). Grošelj 1951a:256 equates the Lydian name of Gyges with Hitt. ḫuḫḫa- ‘grandfather’ based on the gloss of Hesychius: γυγαί· πάπποι (cf. Heubeck 1953a:271). Heubeck 1961:47 accepts that Gyges is etymologically related to Lyc. xuga- ‘grandfather’ (followed by Milani 2001a:284, Sh. Hawkins 2013a:133, 175). Neumann 1961a:69f. does not accept it on phonological grounds. Although it was believed that the Lydian name had to be borrowed from Carian because of the retention of the laryngeals, this has now proven unnecessary, because it has been recently shown that the second (and third?) laryngeal is preserved as Lyd. k (Oettinger 2017f and Oettinger 2021a; cf. with more precise conditioning Sasseville 2021a:165 n. 49, Yakubovich 2019b:402). The rendering of Lydian word internal k with the Greek gamma is paralleled by Lyd. šrkaštus → Συργάστης, whereas a Lydian word initial k such as in katowa- is rendered by Greek kappa, Καδοας/Καδως. One Lydian phonological phenomenon that has not been addressed yet is whether the original lenited laryngeal fell together with the fortis one, just like it tends to happen between Lycian x and g; cf. under Reconstruction.

[D.S.]

HW² Ḫ:638.

The word γυγαί is only known from a gloss of Hesychius (“πάμποι”, which is practically always corrected to πάπποι ‘grandfathers’) and usually connected to the Anatolian words for ‘grandfather’, Hittite ḫuḫḫa-, Cuneiform Luwian ḫūḫa-, Carian quq, etc. (Hammerich 1948a: 70; Whatmough 1949a: 288; Grošelj 1951a: 256/264; Brandenstein 1954a: 65; Heubeck 1959a: 62f.; DELG:239; GEW I:331; EDG:290; Simon 2019a:296f.; cautiously Neumann 1961a:71 [“mag”]; and not Anatolian according to Serangeli 2020a:207). Only Simon 2019a:296f. (cf. also Simon 2018a:388) investigated the precise origins of this word and argued on phonological grounds that it can reflect only the Carian word (“possibly”, according to Garnier & Sagot 2020a:179). Since, however, the gloss was not labelled by Hesychius, it is not possible to decide if we are dealing with the transcription of a Carian word or (perhaps rather, exactly because of the missing label) with a Greek word borrowed from Carian (that might have been restricted to Greek speakers in and around Caria).

[Zs.S.]