*diréi̯e/o-(di)
‘to roast’
- Language
- Proto-Luwic
- Grammar
- verb
- ID
- 2218
1. Individual Anatolian Languages
1.1 Cuneiform Luwian
tīr(a)i-(di)
‘to roast, blaze (?)’1.1.1Transmission
This lexeme is found in a festival text of the cult of Ištanuwa, within the mythological text of the uninvited deity and within a ritual with similar mythological context involving a sick child, Kamrušepa, and the Sun-god. The latter context is not written entirely in Luwian, even though the Hittite passage prior to the invocation specifies that the following speech is in Luwian, but it nevertheless contains a mix of Hittite and Luwian lexemes.
[D.S.]
1.1.2Forms
| 3pl. pret. ind. act. | ti-ra-in-ta | KUB 35.135 iv 11’ | (CTH 772, NS) |
| part. nom./acc. sg. n. | te-ra-a-im-ma-an-za | KUB 35.107+ ii 9’ | (CTH 764, MS) |
| [te]-ra-a-im-ma-an-za | KUB 35.107+ ii 3’ | (CTH 764, MS) | |
| ti-ra-˹i˺-[-im-ma-an-za] | KBo 12.100 obv. 12 | (CTH 764, LNS) |
Outdated is Otten 1953a:102 n. 193, who takes teraimmanza in KUB 35.107+ ii 9’ as an accusative plural attribute to the following dIM-za, which is no longer possible on grammatical grounds.
[D.S.]
1.1.3Graphic Features
Checked against the photographs in the Konkordanz (Mainzer Fotoarchiv). Pace Otten 1953c:97 and Starke 1985b:237, we see no traces of the sign 〈te〉 at the edge of the line of KUB 35.107 ii 3’ and, therefore, it is here marked as completely restored (thus also Steitler 2017a:390). The finite form ti-ra-in-ta recovered in KUB 35.135 iv 11’ is based on the approach that the word begins with the sign 〈ti〉, differently Starke 1985b:322 ([ … ]x-ti-ra-in-ta). The restoration at KBo 12.100 obv. 12 follows the suggestion of Starke 1990a:244 n. 152.
[D.S.]
1.1.4Meaning
Melchert (CLL:227) and Torri 2010a:386 regard the participle tiraimmanza is an epithet of the sky, i.e. tappaš-, which would be broken off in each case, although the restoration [ tap-pa-aš-š]a te-ra-a-im-ma-an-za (KUB 35.107+ ii 9’) is contextually supported (followed by Steitler 2017a:390f., 396). Although no meaning has been assigned yet (not even by Starke 1990a:230), it is possible to suggest one using the combinatory method. The phrase [tappaš]a tirāimmanza found in the Luwian passage surely equates to Hittite wand[a]niya nepiši ‘in the blazing sky’ found in the Hittite passage of the same text (KBo 9.127 + KUB 36.41 i 11’); see CLuw. wandani(ya)-. Therefore, tiraimma/i- should likewise mean ‘blazing’. Further evidence for these semantics can be drawn from the finite form tirainta. In the Hittite passage preceding the Luwian incantation with tirainta, it is said that the cult officials roast the heart and liver (recoverable despite the fragmentary context). Therefore, Luw. tirainta should equate to Hitt. ḫappinit zanu-(mi) ‘to cook, roast with fire’. This is in agreement with the status of the participle as an epithet of the sky, describing its fire-red color, thus ‘blazing sky’.
[D.S.]
1.1.5Stem
Based on the currently available attestations, it is impossible to tell whether the verb belongs to a stem in -(a)i-(di) or -a(i)-(di). However, the etymological comparison with Lyc. B zireime/i- contributes evidence for a stem in -(a)i-(di). For a possible derivative from the same root, see CLuw. tīran-.
[D.S.]
1.2 Lycian B
zir(e)i-(di)
‘to warm (?)’1.2.1Transmission
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.]
1.2.2Forms
| part. nom./acc. pl. n. | zireima | TL 44c.32 (Xanthos) | |
| part. abl./instr. | zireime{me}di | TL 44c.44-45 (Xanthos) | |
| [z]ireimedi | TL 44d.2 (Xanthos) |
The word zireime/i- was rightly recognized as a participle in -me/i- corresponding to Luw. -mma/i-; see Gusmani 1989-1990a:73 and Starke 1990a:68.
[D.S.]
1.2.3Graphic Features
Gusmani 1968c:3 n. 9 restores [z]ireimedi at TL 44c.44-45 followed by Neumann 2007a:436 and Schürr 2018b:86. Borchhardt & Eichner 1997-1999a:25 restores it as well, but with a question mark. Melchert (DLL:138) takes zireimemedi in TL 44c.44-45 as an adjective derivative in -me/i- derived from the substantivized zireime/i-, which is formally possible, but not likely from the contextual perspective; see below. Therefore, we take zireime{me}di as a mistake resulting from diplography (thus also Neumann 2007a:436 and Schürr 2016d:176).
[D.S.]
1.2.4Meaning
Meriggi 1938a:146 translates zireima at TL 44c.32 with ‘decorated, ornamented’ (followed by Gusmani 1989-1990a:73), taking it as an attribute to sbirte, which he interprets as ‘trophies’. However, an agreement between the two nouns in not possible on grammatical grounds. Gusmani 1968c:3 n. 9 posits tentatively the meaning ‘famous’ (followed by Shevoroshkin 1971a:224), which is a mere guess. Differently, Melchert (DLL:138) discredits the semantic assignment ‘famous’ and specifies that it modifies some kind of offering. Shevoroshkin 2014a:252 regards it as a compound made of zi- and (e)reime/i- (followed by Schürr 2018b:86) and translates ‘provision supply’, which is not remotely persuasive and completely unsupported.
Sasseville 2021d shows that zireime/i- in TL 44c.32, 44-45 refers to waxssa, which he translates as ‘bread’ on both contextual and etymological grounds, and therefore suggests the meaning ‘warm’ for zireime/i-, adducing CLuw. tiraimma/i- ‘roasted, blazing (?)’ in etymological comparison.
[D.S.]
1.2.5Stem
The participial stem in -eime/i- points out to a verb in -(e)i-(di).
[D.S.]
2. Reconstruction
2.1Reconstruction ‘to roast’
If we equate CLuw. tiraimme/i- with Lyc. B zireime/i-, we obtain a Proto-Luwic causative-iterative (or denominative) in *-éi̯e/o-, i.e. *diréi̯e/o- ‘to roast’. The initial voiced alveolar is necessary to explain the development to Lycian B z- parallel to PAnat.*d- > Lyd. c, although a following glide as in the word for ‘god’ (cf. Lyd. ciw- and Lyc. B ziw- < *di̯eu̯-) may have played a part (Sasseville 2021d:182). Therefore, it is possible (though not assured) that the root of CLuw. tiraimme/i- with Lyc. B zireime/i- should be reconstructed as *di̯er-. A different approach would be to see the -r- as suffixal and assume a denominative verb.
[D.S.]

