*k̑o-/k̑i-/k̑e-

A highly frequent function word, the demonstrative is attested approx. 365 times in the HLuw. corpus. Its attestations span the entire geographical HLuw. area except the far west. The temporal spread is equally extensive, and attestations can be found on inscriptions from every century between the late 13^th and the early 7^th century BCE.

[AH.B.]

The list of forms given represents but a selection of the over 360 forms found in the corpus. Care has been taken to provide a good impression of the range of spelling variation for the various forms within the paradigm, as well as the temporal and geographical distribution of the attestations.

In KARATEPE 1 Ho., §51, the reading is usually given as 〈za-i^!-si-i〉, but the order of signs is not entirely clear, cf. the remarks by Hawkins 2000a:65.

Not accepted here is the new reading offered in Weeden 2010b:49f. for TOPADA, §4, who reads 〈zi/a-tara/i〉 instead of 〈tara/i-zi/a〉, because the combination of the numeral ‘three’ with a plural noun is anything but unexpected (cf. Bauer 2014a:88f.), thereby voiding his main argument.

[AH.B.]

The word was first identified as the proximal demonstrative by Bossert 1950-1951a:283, 290 in the third instalment of his edition of the Karatepe bilingual, although he relied on the old reading of the sign *377 as 〈ī〉 instead of 〈za〉. His semantic interpretation was followed by Meriggi 1957a:199, 221f. and received further support from Laroche 1958a:85 with a collection of forms and text passages.

The meaning and possible cognate links of the present lemma were the reason that the sign *377 was eventually fixed with the reading 〈za〉, cf. the detailed argumention in Bossert 1960c, later followed with broadened argumentation and a full consideration of the implications by Hawkins, Morpurgo Davies & Neumann 1973a.

[AH.B.]

The endings attach directly to the root of the demonstrative in nearly all cases. Only the dat./loc. pl. appears to be further derived (Melchert 2003b:190f.), perhaps on the basis of its singular counterpart. The abl./instr. form /zin/ was the last of the paradigm to be identified by Goedegebuure 2007a:320-327, 334, who views them as old adverbial forms that became incorporated into the demonstrative paradigm, a development mirrored by the corresponding non-proximal deictic demonstrative /aba-/ ‘that’ and /abin/, see under lemma.

/za-/ ‘this’ can be used both adnominally and usually preceding its noun (Bauer 2014a:61), or pronominally, then referring back to a referent that has already been established in the discourse, either by explicit mention or by virtue of its physical proximity.

[AH.B.]

1. /zadi/ ‘here’
2. /zin/ ‘on this side’

There are more than 40 attestations of the pronoun, of which a representative sample is documented here. The forms are widely distributed across festivals and incantations, beginning in the OH period (CTH 771), but the earliest manuscripts show MS.

[E.R.]

Reading and morphological interpretation of za-˹at-ti˺ in KBo 13.260 iii 16’ according to Sasseville 2021a:516f. n. 5 against zạ-tụ-pạ^?-tị^? as misread in Starke 1985b:262. – Restoration of  za-˹a˺-[at-ti] in KUB 35.54 iii 23 follows Yakubovich, Mouton and Puértolas Rubio (pers. comm.). – Reading of za-˹a˺ in KUB 35.54 iii 26 follows Starke 1985b:68 n. 31, but with different morphological analysis. – Attribution of ˹za˺-ši-i-in in KUB 35.13, 20’ not assured.

[E.R.]

[z]a-am= in KUB 35.103 iii 4 and KUB 25.37 ii 36’ shows assimilation of nasal /n/ to /m/ before the following labial consonant in clitic =pa ‘but’; similarly za-a-(am-)mu (← zan + =mu) in KUB 35.136++ iv 3’, 6’, etc.

[E.R.]

The function of zā- as demonstrative pronoun was first established by Meriggi 1957a221, based on first attempts by Otten 1953a:98. This was accepted by Laroche 1958a:85-87, who recognized the semantic equivalence with Hitt. kā- ‘this’. Melchert 2009a:152 then claimed correctly that not only proximal deixis is shared by the Luwian pronoun, but also speaker-orientation as established for Hitt. kā- in Goedegebuure 2003a. For general acceptance of the semantic value, cf., among others, DLL:112, CLL:274, HW 2nd ed.:428, HEG:A-K 456-459, HED:K 11, Goedegebuure 2010b:84-87, Kloekhorst 2012b.

Most attestations are used with spatial, speaker-oriented meaning (‘this (one) here with me’) referring to the ritual patron or his belongings, less frequently in anaphoric function (KBo 29.2 + KUB 35.52 ii 3’f. with dupl. KUB 35.54 ii 26’, KBo 13.260 iii 5’). Syntactically, the pronoun occurs in both adnominal and pronominal use.

[E.R.]

The stem with its switch between final -a- and -i- is typical of pronouns, while the lengthening of the stem vowels is restricted to the demonstratives zā-/zī- and apā-/apī-.

[E.R.]

1. zāni
2. zātī
3. zanna/i-
4. zašta/i-
5. zattan
6. zāwi(n)

This lexeme is found in various grave inscriptions 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 around 400 BCE. (Borchhardt & Eichner 1997-1999a:19f.).

[D.S.]

For the grammatical assignment see under Meaning.

[D.S.]

The division of the sequence siketesi has long been debated in the literature. Friedrich 1932a:66 separates it into sike and tesi. Hajnal 1995a:236 follows Friedrich and provides a convincing lexical explanation for the sequence si=ke, which is discussed under Meaning. Shevoroshkin 2002b:122 separates it into s(e)=iketesi, interpreting it as the proclitic conjunction se (followed by DLL:116) and a verb in the 2^nd person singular, but this makes no sense syntactically; see under Meaning. Neumann 2007a:326 leaves the matter open, and Schürr 2016d:170 takes siketesi as a full word, positing a genitival adjective.

[D.S.]

Hajnal 1995a:236 provides the best account of the word si, attested twice in the sequence si=ke tesi. He interprets Lycian B si as a demonstrative pronoun related etymologically to Luw. za- going back to PIE *k̑o-, and takes si as a nominative plural agreeing with erẽpliz ‘deities’. However,  erẽpliz must be accusative plural, i.e. the direct object of the 3^rd sg. verb xixbati. Therefore, it is preferable to analyze si as a dative singular agreeing with tesi ‘oath’ and view the particle =ke as coordinating si tesi with the dative/locative singular zppli.

trqqiz=kke=pe me=de zppli xixbati qetbeleimis si=ke tesi uwedris erẽpliz sebe leli pinati masasi tulijel^![i] (TL 44c.34-36)

‘The Storm-god, on the other hand, summons to the feast and to this oath all invigorating deities and gives a speech to the assembly of gods.’

The concept of summoning the invigorating deities for an oath finds a parallel in the Cuneiform Luwian ritual of Zarpiya, the physician of Kizzuwatna (CTH 757), which brings contextual evidence.

nu-za ke-e-da-ni le-en-ga-i ku-ut-ru-e-ni-eš e-eš-tén nu EGIR-an-da GEŠTIN 9-ŠU ši-pa-an-ti PA-NI ^GIŠBANŠUR ^dIn-na-ra-u-wa-an-da-aš (HT 1 i 57’-59’)

‘(Gods) be witnesses to this oath! Afterwards, he libates wine 9 times before the table of the invigorating deities.’

Therefore, the segmentation si=ke tesi ‘and for this oath’ is the best interpretation, because it works on syntactic, grammatical and contextual levels.

[D.S.]

The word is attested in funerary inscriptions (a Greek – Carian bilingual from Athens from the end of the 6th c. [Courbin 1954a:106; and not “525/520 BC” [Adiego 2007a:164] or “late 6th century BC (525-520 BC)” [Adiego 2019a:33]: the sculptor is probably identical to Aristocles, whose Aristion-stela is dated to c. 510 on stylistic grounds [Neudecker 1996a]] as well as on a false-door stela from Memphis, c. 500-470 [Kammerzell 1993a:162, 170f.]) and on votive objects partly of unclear provenance from the 6th c. (C.Ha 1: c. 500 [Jucker & Meier-Brügger 1978a:105–108, not “6th century BC”, contra Adiego 2019a:31], on its probable Egyptian origin see Meier-Brügger 1994a:113, Schürr 2010e:134, Simon 2019e:294, Simon forthcoming contra Adiego 2007a:144, Adiego 2019a:32f.; the 6th c. date of C.xx 1 is not assured [Gusmani 1978c:67, not “late 6th – 5th century BC”, contra Adiego 2019a:32], on its probable Egyptian origin see Schürr 2010e:134 and Simon 2019e:289, Simon forthcoming contra Meier-Brügger 1994a:113; Adiego 2007a:160, Adiego 2019a:32f.), partly from Kaunos from 500-480 BC (Çörtük & Gander & Holler 2018a:112f.).

[Zs.S.]

The attestations follow Adiego 2007a:319f., 410, 413 (cf. also Hajnal 1995b:23) supplemented by “C.Ka 10” checked against the published photographs (Courbin 1954a:109 Fig. 10; O. Masson 1978a:XII.1; Jucker & Meier-Brügger 1978a:Pl. 2; Gusmani 1978c:Pl.I & II; Çörtük & Gander & Holler 2018a:113 Fig. 1, resp.). Since the last letter of snṇ is fragmentary, other readings, notably snụ and snṃ are in theory also possible (see the detailed discussion in Çörtük & Gander & Holler 2018a:113–115), but the choice of the publisher (M. Gander) is probable on both contextual (votive inscription) and epigraphic grounds (the remaining part of the third sign fits the letter n by far better).

It is possible, but currently cannot be proved that sñ in sñaidλo of C.Si 1 also belongs here (Simon 2019e:292 [who argues that ñ appears originally in word-final position], cf. also Adiego 2000a:152, Adiego 2007a:413). Snś (E.AS 8 and C.Kr 1) and sñis (C.Ka 1) are not included here (contra Hajnal 1995b:23, O. Henry 2007a:96, and Adiego 2007a:291, 413, resp.), since they are personal names (Duhoux 2008a:64f.; Schürr 2013c:27; Simon 2019e:292, Simon 2019b:3f. with discussion on their relation). Schürr 2013c:27 cautiously proposed that also san belongs to this personal name, but this does not fit the context and eliminates the parallelism of the Carian text and the Greek text. Sδisa (a title, a profession, or an ethnic adjective, see Simon 2019b:3–6) has also been excluded (contra Hajnal 1995b:23). Hajnal 1995b:23 with n. 30f. proposed to analyse the obscure form biδślemsa (in the graffito of E.Ab 30) as a substantive biδślem (originally a compound participle) with the demonstrative pronoun sa, but there is no evidence for that and it does not clarify the inscription further.

As Adiego 2007a:320 points out, there are two possible explanations for the two nominative singular forms sa and san: either there is a gender distinction (he sees sa as common gender and san as neuter gender; san is identified as nom.-acc. sg. n. also by Melchert 1993d:79, Melchert 2010f:180) or san represents a sort of “extended form” (as in Melchert 2010f:180). According to Adiego, the latter explanation is supported by the acc. sg. form snn (he also quotes the “striking parallel” of the paradigm of Lycian ebe- [cf. already Neumann apud Melchert 1993d:80 n. 5], but this analogy depends upon the controversial interpretation of ebẽñnẽ, see s.v.). Alternatively, he proposes that both n-s in snn are accusative markers, which would support the former explanation.

Both suggestions are, however, based on unmotivated assumptions, such as the existence of an “extended” form or the double accusative marker. Since the acc. sg. snn is not identical to any of the attested nominatives (sa and san), it can represent only the acc. sg. c. (identified as such also in Melchert 1993d:80, Melchert 2010f:178, 180): if it were neuter, it should be identical to one of them. This conclusion can be avoided only if one assumes a third nominative singular form, *snn, which is clearly against Occam’s razor. Moreover, although the gender of the words in agreement with san (śjas ‘funerary monument’ vel sim.) and sa (upe ‘stele’) is independently not known, the acc. sg. snn is in agreement with orkn, a type of vessel (in C.Ha 1 and C.xx 1; “C.Ka 10” is fragmentary). The most straightforward analysis is that this is an acc. sg. c. with the usual Carian acc. sg. c. ending -n (see most recently Melchert 2010f:178). Although strictly speaking one cannot exclude that also orkn is a nom.-acc. sg. n., but this is again clearly against Occam’s razor (note that the plausible etymological connection with Greek ὕρχη ‘earthen vessel used for salting fish’ [see most recently Simon 2019a:302–304 with refs. and here s.v.] also supports the analysis of -n as ending, but it does not define its gender). If, however, snn is the acc. sg. c., then san must be the nom. sg. c. and sa the nom.-acc. sg. n. (against Adiego’s description, but in agreement with Hajnal 1995b:23 [who mentioned only one, problematic argument for this view and who would interpret sa phonologically as /saⁿ/, although there is no evidence for that] and partially with Melchert 2010f:180 [only for sa, as he takes also san as a nom.-acc. sg. n.). Accordingly, the pronoun is an n-stem (see already Hajnal 1995b:23). This analysis has the advantage of providing the most economical solution and avoiding the uncomfortable assumptions of a double accusative marker and of the double nominative forms either in commune or in neuter gender.

[Zs.S.]

There is a general and justified agreement regarding the meaning, which is based on the Greek – Carian bilingual G 1 and the context of the funerary and votive inscriptions (Adiego 1992a:33, Adiego 2007a:288, 319f., 410, 413, Adiego 2010b:154, 172, Adiego 2019a:30, 32; Melchert 1993d:79, Melchert 2002b:309, Melchert 2010f:180; Hajnal 1995b:22f.; Yakubovich 2005a:248; Duhoux 2008a:64f.; O. Henry 2007a:96; Simon 2008b:461; Çörtük & Gander & Holler 2018a:115–117 [Gander mentions [115] a translation ‘bronze’ in case of C.xx 1, but the literature quoted there does not contain such a proposal and the contexts of snn argue against such a translation, “C.Ka 10” [if it indeed contains snn] even excludes it being made of marble]).

[Zs.S.]

Last updated: 26 March 2020

This lexeme is found across the Palaic corpus.

[D.S.]

Melchert 1984b:28-30 argues convincingly for the existence of a Palaic demonstrative pronoun functioning as an enclitic (contra Kammenhuber 1959a:29-31, Carruba 1972a:25f.). For the enclitic status, he compares Pal. =apan found at the end of the sentence initial clitic chain. From the contextual perspective, Melchert points out that the enclitic pronoun ka- occurs after nouns designating items, which lie in front of the speaker and which are pointed out to the hearer. This speaks for the function of ka- as a pronoun with visual proximal deixis, i.e. referring to an object being physically present within the near sight range of the speaker.

In 2007, the new fragment KBo 48.178 was published carrying a duplicate text of KUB 35.165 rev. 9’-15’, where the enclitic pronoun appears several times. Furthermore, this new fragment carries more Hittite text preceding the Palaic speech than the passage in KUB 35.165. The important phrase, which is partly preserved is the following one: [nu ^UZUšuppa ^UZ]^UZAG.DU^ḪI.A GABA^ḪI.A SAG.DU^MEŠ [ … PA]NI DINGIR^MEŠ tianzi ‘They set in front of the gods [the sacralized meat], the shoulders, the thighs, the heads […].’ (KBo 48.178 l.col. 5’-6’). This brings confirmation that the various nouns mentioned refer to body parts of the sacrificed animals.

Regarding this enclitic pronoun, Palaic distinguishes between singular =kat and plural =ga (Melchert 1984b:30), but does not seem to distinguish between genders. Therefore, there is probably no gender distinction to be interpreted out of the sign form =kat (for a parallel phenomenon see the New Hittite situation of the enclitic pronoun =šet/=šmet). The Palaic participle wakkakanta agreeing with the body parts cannot distinguish graphically between the neuter singular and plural, i.e. pl. n. /wakkaganta/ and sg. n. /wakkagant/ with retention of the final alveolar after the nasal.

In 2017, a new join published by Soysal 2017a gives us the sentence ginu=kat karšandu ‘they must cut off this leg here’ (KUB 35.159 + Bo 4790 ii 8’/4’), which offers further evidence that words ending with the element =kat under discussion are to be interpreted as nominatives or accusatives and not as ablatives.

Eichner 2010a:51-53 missed the publication of the fragment of 2007 and pursued the idea that the forms in -(k)at are instrumentals (thus Starke 1990a:71-74). Eichner argues that the body parts, being in the instrumental case, refer directly to the sacrificial animals, e.g. they are fat at the head’. This must be rejected because of the Hittite passage preceding the Palaic one in KBo 48.178 l.col. 5’-6’, which makes it clear that all the body parts are being laid down in front of the gods. The Hittite text explains what the Palaic speaker does and says, and not something different. Another passage requiring a noun ending in =kat to be the direct object is KUB 35.163 (and the fragment KBo 13.265), in which a-ar-ra-kat and lu-u-ki-it form a small sentence right at the beginning of the Palaic speech. The verb lūkī- ‘to break (into pieces)’ is transitive and requires a direct object. Therefore, an interpretation in the ablative here would fail to account for the syntax.

[D.S.]

1. kiat
2. gāni

HW 2nd ed. K:1-4, Goedegebuure 2014a:32-36, 277-335