*méd^hu-

The lexeme mattu- ‘wine’ is attested at least three times in the Hieroglyphic Luwian corpus. A fourth instance may be found in (“VITIS[”])mi-t[u- in TELL TAYINAT 2, but the lexical assignment is not assured. All attestations are confined to inscriptions of the 8th century BCE.

[D.S.]

nom./acc. sg.	/mattu=sa/	ma-tu-sà	SULTANHAN, §26 (Tabal), BOR, §7 (Tabal)
		(VITIS)ma-tù-sà	AKSARAY, §4c (Tabal)
		(“VITIS[”])mi-t[u-	TELL TAYINAT 2, 2b-a (Amuq)

Hawkins 2000a:373 suggests tentatively that the instance of (“VITIS[”])mi-t[u- in TELL TAYNAT 2, 2b-a is a dialectal variant of ma-tu-. He suggests to restore (“VITIS[”])mi-t[u-sa-ha].

[D.S.]

Based on the contexts and the determinative (“VITIS”), Hawkins & Morpurgo Davies 1987a:280-283 conclude that mattu- is a commodity connected with wine. This lexeme is either translated as ‘wine’ (Payne 2012a:101) or as ‘sweet wine’ (cf. Starke 1990a:191 n. 624, Giusfredi 2010b:243-251), but its quality of sweetness is not assured. According Hawkins 2000a:477, /mattu-/ is the general word for ‘wine’ as opposed to /wiyani(ya)-/, which means ‘vine’ (likewise Maier 2013a:229 n. 1250).

[D.S.]

This lexeme is most likely a neuter u-stem (Hawkins & Morpurgo Davies 1987a:282, Melchert 2003b:198, pace Starke 1990a:191), although a late extension in -s- or in -d- is still possible based on the current attestations.

[D.S.]

Attested in Luwian rituals in MS and NS.

[Zs.S.]

nom.-acc. sg.	ma-ad-du	KBo 7.68 ii 10’	(CTH 764, NS)
abl.-instr.	ma-ad-du-ú-[wa-ti]	KUB 35.103 ii 15’	(CTH 766, MS)

Attestations following CLL:144 (cf. also HED M:100) checked against the photographs of the Konkordanz. The generally accepted restoration ma-ad-du-ú-[wa-ti] is based on the structurally parallel sentence of KUB 35.103 ii 8’-9’ (Starke 1980b:79 n. 23, Starke 1985b:222 n. 53). Alternatively, Carruba 1969b:72 suggested ma-ad-du-ú-[(n)a^?-ti], which is, however, not supported by the photograph, has not gained supporters and has been abandoned later by Carruba himself (Carruba 1982a:49).

[Zs.S.]

While early scholars did not translate this word (Meriggi 1957a:198 n. 17 [who, however, cautiously compared it with Gr. μέθυ]; Laroche 1959a:70; Carruba 1982a:49; Weitenberg 1984a:292), nowadays it is generally translated as ‘wine’ (Hawkins; Morpurgo Davies 1987a:282f.; CLL:144; Haas 2003a:252; HED M:100; Kloekhorst 2008c:484; Lebrun 2009e:11; Zehnder 2010a:154; Francia 2014a:11). Nevertheless, Starke 1980b:79 n. 23 (based on the quoted idea of Meriggi) translated it as ‘süßes Getränk’, as the substantivization of an adjective ‘süß, lieb, angenehm’ (similarly Tischler, first as an adjective ‘sweet’ [HEG L-M:165, as well as van den Hout 2003a:304], later [Tischler 2008a:115, Tischler 2016a:217] ‘ein süßes Getränk, Traubensaft, Honigwein, o.ä.’, originally only as the meaning of the HLuw. form). However, Starke 1990a:191 n. 624 preferred more cautiously only ‘(süßes) Getränk’ and translated the word as ‘wine’ on the basis of the identification with HLuw. matu- ‘wine’ (see s.v.). Ünal 2016a:963 combined these approaches and translated it as “sweet wine”. While acknowledging the meaning ‘wine’, Weeden 2013a:81 n. 33 claims that the supporters of the Luwian interpretation of the personal names Maddunāni and Madduwatta “would need to assume that the meaning ‘wine’ is a substantivized use of the adjective ‘sweet’” and that “the older meaning may have been preserved in composition” (accordingly translating these names as ‘sweet brother’ and ‘sweet father’). No such need exists, however, since, on the one hand, maddu- can also be simply a replicated onomastic element of those families and, on the other hand, the meaning of such names are not necessarily to be taken literally (Hawkins 2013a:8). Finally, Durnford 1975:49 translated it as ‘honey, mead’, with question mark. Durnford was led to this translation by his methodologically dubious assumption that “such a word does occur in a text [sic – Zs. S.] as yet untranslatable, and it is more profitable to consider names which contain this element”, while assuming a divine name *Maddu ‘Honey-god’ based on the parallelism of the names Maddu-nāni- and *Arma-nāni- translated as ‘Having the Moon-god as brother’ by him and citing Tiwata-wiya- as well (translated as ‘Sun-woman’ by him), which has nothing to do with the question. Nothing, however, requires that the first member is a divine name and not a simple appellative, but even if this is the case, there is no hint for a meaning ‘honey’ (except the etymological considerations, which are, however, not probative) and not provided by Durnford, either.

In order to define the meaning KBo 7.68 ii 10’ is not helpful, since maddu- appears here in a list of sacralized foods (different types of breads, crumb) and objects (table, a vessel, a wooden object) connected to eating (KBo 7.68 ii 6’-11’), thus maddu- can belong to either of these groups (since the text is fragmentary, it cannot be assured that maddu- denotes the liquid missing from this list). The text KUB 35.103 ii 15’ in itself does not help either, since the only clear thing is that maddu- represents a purification or cleaning substance. Nevertheless, the meaning of its adjectival derivate, *madduwi(ya)- as ‘of wine’ is assured (see s.v.), thus it can be extrapolated to maddu- as well. Nothing points, however, to a sweet drink, a sweet(ened) wine, honey wine, mead, not to mention grape juice: these assumptions are based purely on etymological considerations (cf. under Reconstruction) and thus they are not probative.

Finally, as mentioned above, in KUB 35.103 ii 15’ maddu- is used for cleaning. If this was strictly ritual, the meaning ‘wine’ is appropriate. Nevertheless, as Haas 2003a:252 points out, “wine” was used for normal, non-ritual cleaning as well. Since, however, wine is not very appropriate for cleaning, one can entertain the possibility of a wine-based substance, for instance ‘wine vinegar’ (if this is correct, it represents one more argument against the assumptions of Durnford, Starke and Tischler).

[Zs.S.]

I. [235]*madduwi(ya)-

[Zs.S.]

^NINDAma-du[ is a hapax from a Hittite ritual (KBo 8.91 rev. 3, CTH 479 [MS]). Due to its determinative, it is generally agreed that it means ‘a kind of bread, pastry’ (HEG L-M:165, Tischler 2008a:115, Tischler 2016a:217; Ünal 2007a:439, Ünal 2016a:346; CHD:s.v.; note, however, that Hagenbuchner 2002a did not treat it in her catalogue of bread types), where the only point of disagreement is that ma-du[ could not only be a noun but also an adjective modifying NINDA (as pointed out by HED M:100).

This word was repeatedly discussed: CHD:s.v. pointed to the cross-culturally well-known possibility that the names of some bread types derive from geographical names and cautiously compared ^URUMaddunašša “along” with the element maddu- in personal names (Laroche 1966b:Nr. 352a, 793, 794), which is probably identical to CLuw. maddu- ‘wine; wine vinegar^?’ (emphasized by HEG M:165), see the discussion under the Lydian ghostword †maddun-. HEG M:165 and HED M:101 connected madu[ exactly with this CLuw. word and thus proposed ‘sweet bread’ and ‘sweet bread, honey bread, wine bread’, resp. (despite the fact that nothing points to a sweet component in maddu-, see the discussion under CLuw. maddu-). HED M:100f. even assumed a Luwian origin based on this assonance.

Set aside that both comparisons must be excluded on phonological grounds since the compared words are written with geminated consonants, unlike madu[, the join KBo 42.103 rev. 3. identified by Trémouille 2002b:848 (not cited in the quoted discussions) revealed that the full form of this word is ^NINDAma-du-˹up˺-pa-aš-ši-iš, making all of the above discussions obsolete.

[Zs.S.]

This Anatolian word is assumed to have been borrowed into non-Anatolian languages as well.

Greppin 1982a:70 proposed that the Armenian word mat‘owz ‘mulberry’ is a loan from HLuw. matu- ‘wine’. This view was rejected by Simon 2013b:109 on phonological and semantic grounds.

Watkins 2001a:51 suggests that the Etruscan word for ‘wine’, matu, represents a borrowing from a “Luvian-like language” on the basis of CLuw. maddu- and HLuw. matu-. Set aside the highly problematic and unsolved issue of the existence of Anatolian loanwords in Etruscan and that there is no evidence for the Anatolian origin of the Etruscans (see most recently the discussion in Simon 2015a:4–6, 9 n. 42 with refs. [cf. also Simon 2016b:330–332] contra Beekes 2002a; Kloekhorst 2012c:49f. and esp. Bachvarova 2007a:170, Bachvarova 2012a:149–151, Bachvarova 2016a:361f.), the central problem is that there is no evidence (and not provided by Watkins either) for the meaning ‘wine’ of this Etruscan word. It appears only in the sarcophagus inscription AT 1.96 (Pallottino 1978a:237; Rix 1991b:140; Meiser 2014a:252), whose meaning, however, is only partially understood: the sentence of matu (line 8) belongs to the unintelligible part, where, however, a word with the meaning ‘wine’ does not really fit, thus Watkins’s suggestion has no basis. Cf. AT 1.96 (following Meiser 2014a:483):

arnθ: aleθn2as: ar: clan:	‘Arnθ Aleθnas, son of Ar(ntθ),
ril 3XXXVIII:	age 38,
eitva: ta4mera: śarvenas	having quadrupled the eitva tomb chamber,
5clenar: zal: arce: 6acnanasa:	having issued two sons,
zilc mar7unuχva: tenθas: ẹθl	having held (the position of) governor and maruships, thus?
8matu: manimeri	matu in the funerary monuments?’

[Zs.S.]