*h₁i̯éh₁-/h₁ih₁-´

This lexeme is attested several times in inscriptions from Maraş, Tabal and Emirgazi, dating from the second half of the 13^th to the late 8^th century BCE.

[A.P.]

[A.P.]

As one of the basic level terms, this verb was semantically clear as soon as its cognate link with Hittite iya- ‘to do, make’ was established (cf. e.g. Kronasser 1956a:39, 169, 181 et passim), even though it was still conflated with the semantically very similar verb HLuw. /izzi(ya)-^(di)/ ‘to do, make’ at the time. Once /a(ya)-^(di)/ and /izzi(ya)-^(di)/ had been separated, the link with Hittite iya- and CLuw. a(ya)- ‘to do, make’ was re-established (see Hawkins 1975a:141, Hawkins & Morpurgo Davies 1975a:128, and Hawkins 1995a:98), and the meaning was explored again in detail by Morpurgo Davies & Hawkins 1987a:276-279. It can be observed e.g. in SULTANHAN, §13: hwa/i-pa-wa/i sa₅+ra/i-wa/i-ti-wa/i+ra/i-ia-´ wa/i-su-SARMA-ma-sa SERVUS-lá/í-i TONITRUS-hu-za-sa za-a-zi wa/i+ra/i-ia-zi-´ CUM-ni á-tà-a, approx. /kwippa=wa Sarrawattiwaraya Wasusarmas(sa) hudarli Tarhunzas zanzi warriyanzi anni ada/, translating as ‘Indeed, (the storm-god) Tarhunza made these assistances for Sarrawattiya, Wasusarma’s servant’ (cf. Hawkins 2000a:466, 469).

When used in a ritualistic context, the verb frequently denotes an honouring of a deity by way of a sacrifice, similarly to its CLuw. and Hittite cognates, see Rieken 2007a:267f.

The CLuw. cognate a(ya)-^(di) is attested both in the active voice, meaning ‘to do, make’, and the medio-passive, then meaning ‘to become’, see e.g. Sasseville 2021a:300f. It is reasonable to assume that the HLuw. verb shows same voice-dependent semantics, but since no medio-passive forms have been found attested so far, this cannot be verified.

[AH.B.]

The present lexeme represents a root verb with lenited endings, whose stem shows allomorphy between /a-/ mostly in the singular forms and /aya-/ throughout the remainder of the paradigm, see e.g. Sasseville 2021a:300f.

[AH.B.]

1. /an(n)iya-^(ti)/

This lexeme is well attested throughout the Cuneiform Luwian corpus.

[D.S.]

For the 2pl. pres. act. form, see Carruba 1968a:15 (followed by Melchert 2003b:192, Sasseville 2021a:300f.).

[D.S.]

Checked against photographs in the Konkordanz (Mainzer Fotoarchiv).

[D.S.]

This Luwian verb is etymologically equated with Hitt. iye/a-^(mi) ‘to do, make’ and translated accordingly also as a medio-passive ‘to become’ on contextual grounds, which has been generally accepted (Otten 1953a:82 The Hieroglyphic Luwian verb previously adduced is outdated, since the decipherment of the signs 〈za〉 and 〈zi〉 by Hawkins, Morpurgo Davies & Neumann 1973a., Laroche 1959a:23f., Melchert (CLL:3f.), Yakubovich 2010d:56-61, Sasseville 2021a:300f.).

[D.S.]

The stem inflects according to the lenited mi-conjugation. Yakubovich 2010d:56 wishes to separate the verb into two stems, a regular one and a previously reduplicated one functioning as an imperfective, which is not tenable and contradicted by the context of KBo 13.260, in which both allomorphic stems describe the exact same action. Moreover, the “reduplicated” formation based on a root verb would be expected to belong to the ḫi-conjugation. For the verbal stem class, see Sasseville 2021a:300f. and also see below under Reconstruction.

[D.S.]

1. ānni(ya)-^(di)

1. āyattar/n-
2. ^DAyant(i)- (DN) (Otten 1953a:82)

This lexeme is well attested across the entire Lycian corpus.

[D.S.]

The composition of this paradigm diverges from one lexical list to another (cf. DLL:2, Neumann 2007a:6f., Sasseville 2021a:106), because certain forms can be confused with the lexeme axa- c., a substantive, and others such as 3sg. pret. adẽ are so short that their division into a separate lexeme is not always obvious; cf. under Graphic Features. As pointed out by Melchert (DLL:9) the word dde (TL 44b.29) could be a mistake for a^!de (3sg. pret. ‘he made’), but the obscurity of the context calls for caution.

[D.S.]

One could also interpret me=munaiti (TL 90.3) as me=mu=n(e)=aiti and recover another instance of the 3pl. present form aiti, but the meaning of the sentence remains obscure.

The two forms axã at TL 44a.34, 35 have been treated as substantives (acc. sg. c.) by Melchert (DLL:7), on the pretext that a 1sg. form is contextually unexpected, which is not justifiable, since there are other cases of 1sg. verbal forms at various places in the text (TL 44a-c). Instead, we take them as verbal forms (thus with Cau 1999c:25), which is furthermore the only option for a predicate in each sentence (outdated Schürr 2009a:159; see Schürr & Dönmez 2015a:135-137).

Concerning TL 21.4, Melchert (DLL:2) tries to recover a preterit form ede from the segment meded (followed by Eichner 2007a:108). We suggest emending meded<ẽ> to accommodate for the syntax, i.e. m(e)=ede=(a)d<ẽ> (or m(e)=ed(e)=ed<ẽ>) prñnawã prñnawã (for =ede agreeing with acc. sg. c., cf. hrm̃ma-).

On the analysis of adru (a)de (TL 44b.47), see Neumann 2007a:4 and Lyc. A adru.

We take ãinẽ as an infinitive form of a(i)-^(di) with Schürr & Dönmez 2015a:143. The nasalization of the last vowel -ẽ is probably due to the fact that the infinitive occupies both the position and the role of the finite verb in the sentence.

[D.S.]

Torp 1898a:8 suggests for ade the meaning ‘he made’ on contextual grounds, followed by Meriggi 1936b:272 (and Meriggi 1937a:513, who includes aiti (TL 44c.17) as a form of the same verb ‘to do, make’). Mittelberger 1964a:75 n. 83 compares Lyc. a(i)-^(di) etymologically to Luwian ā(ya)-^(di) ‘to do, make’, which brings further confirmation. The semantic assignment is commonly accepted. For the brilliant identification of the 1sg. pret. ind. med. axagã with the meaning ‘I became’, see Melchert 1992c:189-191 (followed by Sasseville 2021a:106, contra Neumann 2007a:6f.).

[D.S.]

For the verbal stem class, see Melchert 1984a:159-161, Sasseville 2021a:106.

[D.S.]

1. as-^(ti)

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.]

There is general agreement on the grammatical interpretation (see Melchert (DLL:114), Schürr 2018b:66, Sasseville 2021a:106).

[D.S.]

Shevoroshkin 2002a:228 interprets eim̃ as an apocopated participial form of the verb a(i)-^(di) ‘to make’, which is entirely plausible (likewise Sasseville 2021a:106). Outdated Shevoroshkin 1971a:218 “personal pronoun”.

[D.S.]

1. as-^(ti)

Hapax in the so-called Longer Strategos Votive Inscription of unknown date (see the discussion s.v. śa) from Side.

[Zs.S.]

The attestation, the segmentation, and the grammatical analysis follow Pérez Orozco 2003a:106, Pérez Orozco 2007a:136 checked against the photographs in of Darga 1967a:Res. 8 (= Neumann 1968a:Taf. I.2 = Brixhe 1969a:Planche II), and Zinko & Zinko 2019a:426 Fig. 21.5. As Pérez Orozco 2007a:139 remarks, it may be present also in the completely unintelligible sequence ewśako³pebijaseñnebarśaka of the same inscription after ewś, but it cannot currently be proved. Its segmentation is relatively assured, since it is followed by the word ośad (see s.v.) and preceded by śa ‘and’ (see s.v.) and it is very probable that there is a word boundary between the two ‹a›-s (cf. already Neumann 1968a:89, 91, Neumann 1978b:874; Ševoroškin 1975a:162, 164; Eichner 1993a:127; Nollé 2001a:636). The grammatical analysis is based on the semantic analysis, cf. under Meaning.

[Zs.S.]

The reading of the second sign (N23) as ‹k› by Pérez Orozco 2003a:106 is not accepted by Schürr 2016b:150 n. 7 (“unklar”) and Rizza 2019a:546 (“speculative” and “uncertain”). See the discussion s.v. *kuar for the merits of the reading ‹k› and why it is followed here. Note that the reading ‹k› leads also here to a meaningful sequence (see under Meaning) unlike the traditional reading ‹b›.

[Zs.S.]

Setting aside the interpretation by Ševoroškin 1975a:164 (ab- ‘3rd person personal and demonstrative pronoun’, followed by Eichner 1988b:52 n. 30, Eichner 1993a:128) based on a reading which is not supported here, Pérez Orozco 2003a:106, Pérez Orozco 2007a:132, 136 suggested that ak is the Sidetic cognate of Luw. aḫa, Lyc. aχa ‘I made’ (followed by Lebrun 2012b:356). This proposal is formally fitting (on the reflex of the laryngeal see woskiji from the same root as Luwian wašḫaya-, Pérez Orozco 2007a:132). On the semantic side, S3 is a votive inscription and the votive act is expressed in the clause of ak (śa ak ośad anaϑemataś) shown by the word anaϑemataś (which probably does not contain any verb, contra some scholars, see s.v.): ‘and ak ośad the votive offerings (acc. pl.)’. Since ośad seems to be an adverb (see s.v.) and ak does not seem to be a recipient on morphological grounds (no traces of dative or anything similar), the remaining possibility is exactly the verb expressing the act of donation. Thus, Pérez Orozco’s proposal has good chances to be correct.

[Zs.S.]

Attested in a monolingual Carian official inscription of unknown date and in a Carian-Greek bilingual decree, dated to the end of the 4^th c. BC (Adiego 2007a:155 with refs.), both from Kaunos.

[Zs.S.]

The attestations follow Adiego 2007a:349 (for rejected cases see s.vv. adymδš and sñaidλo), checked against the photographs in Frei & Marek 1997a:Tafel I., IV. and Frei & Marek 2000a:86. The segmentation of ait seems assured as it is preceded by qrdsoλš, an obvious -oλ-derivative in nom.-acc. pl. c. of qrds (see s.vv.), and followed by 𐋆mali that re-appears some lines later (Melchert 1998c:35). The segmentation of aitusi is more problematic: it constitutes the last line of C.Ka 5, but everything else is conjectural. Since it is preceded by a sequence °mnos that may belong to mno ‘son and/or servant’ (see s.v.), its beginning seems assured. It is unclear if it should be further segmented, see the discussion under Meaning.

Both ait and aitusi are identified as plural forms (Melchert 1998c:35 and Adiego 1998b:25, Adiego 2007a:300f., 349), but they lack any trace of the nasal before the dental expected on Anatolian grounds (note that Simon 2020d even argues that the 3pl. pret. ending is -δ/-δane in Carian) and thus, they should be identified as singulars (cf. already Adiego 2000a:140f. on ait).

[Zs.S.]

Melchert 1998c:35 and Ševoroškin apud Melchert 1998c:35 n. 7 proposed to identify ait as a 3rd pl. pret. form ‘they have made’ based on the assonance with Lycian aitẽ ‘dto’ (cautiously followed by Adiego 2000a:140f. [but here as sg.] and Adiego 2007a:301, 349) and Adiego 1998b:25 (and cautiously in Adiego 2007a:300, 349) suggested to identify aitu in aitusi as a 3pl. imp. ‘they must make’ (it should rather be ‘they shall make’) based on the formal assonance to the quoted Lycian verb and the Anatolian 3pl. imp. endings (in Adiego 2007a:300 the possibility of yet another case of ait was also mentioned; the Greek part of this inscription has not been preserved). He admitted that the resulting element si remains unexplained (and entertained the possibility of a pronoun or an enclitic word in Adiego 1998b:25), but this is obviously not a lethal problem in a section that we do not understand at all. It is interesting to note that in Hieroglyphic Luwian some preterital medio-passive forms are created by adding a particle [si] (probably originating in a reflexive pronominal form *-soi) to the active forms (Rieken 2004b with refs.). This would fit both formally and semantically, but without further evidence it remains speculative if aitusi presents the same formation.

Both cases face the same problem: their explanation is both formally and (due to the general meaning of the verb) semantically fitting (and it can be even supported by the preceding words qrdsoλš and mnos as presumable direct and indirect objects), but since their immediate context is unclear, their identification cannot be proved at the moment.

[Zs.S.]

This lexeme is found in a variety of inscriptions, coming from Sardis, the Kaystros valley, and even Egypt.  Its earliest problematic attestation dates back to late 7^th – early 6^th century BCE (LW 49).  The stele LW/N 115 is datable to the mid-4^th century BCE, while the other texts can be dated only more loosely, within the period of 5^th-4^th century BCE.

[I.Y.]

Traditionally, the attestations listed above are assigned to two separate lexemes, i- and ii-, cf. Gusmani 1964a: 128, 131ff. There are, however, no compelling formal or semantic reasons to separate these two groups of forms. As shown in Melchert 2006b, iitλ represents a middle form with the ending -tλ, but there is no cogent evidence for the existence of dedicated preterit middle forms in Lydian (but cf. Sasseville 2021a:383). Therefore, the form il can belong to the same paradigm without exhibiting any voice differences from iitλ.

The hypothesis of Gusmani 1971a:7 that il has the meaning ‘here’ logically follows from his implausible analysis of the particle =im as a verbal form.

[I.Y.]

The form i in LW 49.1 may represent the abbreviated spelling of il. The grammatical function of i is reconstructed very tentatively based on ul in the following line. Note, however, that the alleged i may represent an accidental scratch (Gusmani 1964a:266).

[I.Y.]

The identification of i- and ii- as verbal forms is first found in Brandenstein 1931a:27f. Two hypotheses about the meaning of i- dominated the field of Lydian Studies in the twentieth century. According to one of them, first promulgated in Meriggi 1935a:102 assumes that the verb i- represents a variant of e- ‘to be’. But the assured instances of alternation between i and e in the same Lydian lexeme are not otherwise attested, and, in any case, the lexical identification between i- and e- would preclude the much more likely identification between i- and ii-. The second hypothesis, developed in Carruba 1960a:41,48, assigns to i- the meaning ‘to do, make’ and treats it as an unextended variant of in-^(t) ‘to do, make’. Being formally more satisfactory, it stumbles upon the absence of assured transitive attestations of i-. In particular, the recently found inscription LW/N 115 where il must form a part of the extended dating formula, speaks for the intransitive copula-like interpretation of this form (cf. kumẽd).

The middle ground between these two previously irreconcilable hypotheses became available with the identification of iitλ as a medial form in Melchert 2006b. It is now possible to retain the hypothesis of etymological connection between Lyd. i-, on the one hand, and Hitt. iye/a-^(mi) and CLuw. ā(ya)-^(di) ‘to make’, on the other hand, assuming at the same time that Lyd. i- reflects the etymological passive of the same Anatolian verb with the range of meanings ‘to become, emerge, represent’.  As a typological parallel from a closely related language, one can compare HLuw. /izzi(ya)-^(di)/, which is used with the meaning ‘to do, make’ in the active voice and with the meaning ‘to appear, become’ in the middle voice. From the contextual perspective, the meaning ‘to become’ appears to be the most appropriate in LW/N 115.3, while the meaning ‘to represent’ emerges from the analysis of LW 22 (Yakubovich 2017b). For the polysemy ‘to appear, represent’ cf. Russ. являться /javljat’sja/, which is used in both meanings. For a different semantic treatment of ii-, which is not conducive to merging it with i- into a single paradigm, see Melchert 2006b:1165.     

[I.Y.]

The synchronic and comparative considerations converge in assigning the root stem to Lyd. i-, but historical reduplication cannot be ruled out. There is no evidence that this verb was ever used in the active voice. Therefore, the formal representation i- can be proposed for this verb. 

[I.Y.]

1. aνi-
2. ani-

1. in-^(t)
2. inãn(i)-
3. fadi-

HW² I:6-18.