*ti, *tu

This lexeme is attested 20 times across in inscriptions from Assur, Tabal and Karkamiš, dating maximally from the 9^th to the late 8^th century BCE.

[A.P.]

At ASSUR letter f+g, §52, the pronoun hosts a clitic chain: 〈-ha-wa/i-za〉, approx. /=ha=wa=anz/, ‘Also, to us’ with the quotative particle intervening.

[A.P.] [AH.B.]

For a long time, the 2sg. forms were thought to be unattested in the HLuw. corpus, and only the realisation that the forms of the 2sg. and 3sg. enclitic pronouns are identical closed the assumed data gap. Regarding the orthotone allomorphs, the data proved to be particularly challenging, and the oblique stem long went without connection to it for the partly fragmented and ill-understood ASSUR letter f+g, §16 clause it occurs in. The only fully preserved context is ASSUR letter f+g, §52: ti-ha-wa/i-za tu-wa/i-na INFANS-ni-na CUM-ni *77-ti-sa, approx. /ti=ha=wa=anz tuwin niwarannin anni …-ttis/, translating as ‘Also, you shall give us your child in marriage!’ (cf. Hawkins 2000a:537). Even for the more frequently attested enclitic allomorphs, the general acceptance of their analysis as 2sg. took some time, not least because the functional analyses of /=du/ and /=di/ differed from scholar to scholar, see already Kammenhuber 1955b:362 and the correction of HHG:129, 131f. by Mittelberger 1963a:93 (later incorporated by Meriggi 1980a:317) and similar later notes e.g. by Marazzi 1990a:66f. and Melchert 1994a:218. Naturally, this realisation also led to an enhanced understanding of some passages in the ASSUR letters and other texts, e.g. the KIRŞEHİR letter, §3: (BONUS)sa-na-wa/i-ha-wa/i-ra/i PUGNUS.PUGNUS-i, approx. /sannawi=ha=wa=di …-li/, meaning ‘May you, too, stay well!’ (cf. Akdoğan & Hawkins 2010a:2, 4f.).

As pointed out by Hawkins 2000a:538f., differentiating the object and reflexive use of the pronouns proved difficult for some time and impacted the analysis and understanding of a number of constructions, but /=di/ has since been identified as the reflexive, whereas /=du/ serves the acc. and dat. object function (see e.g. Melchert 2003b:189f., Rieken 2004b:184, and Yakubovich 2015b:6.4).

Outside the genre of the letters, 2sg. pronouns, both orthotone and enclitic, are very rare, much like their plural counterparts. The very short inscription KARKAMIŠ A18a is one of the few memorial texts that address the intended reader (usually either a passer-by or a deity) directly: EGO-mi ^Ipi^?-sa-mi-tá-sa tara/i-wa/i-ni *a-wa/i-ti i-zi-ia-sa, approx. /ammu=mi Pisamittas tarawanni a=wa=di izziyassa/, meaning ‘I am Pisamitta, the loyal (one). You be reverent!’ (cf. Hawkins 2000a:194).Note that Hawkins 2000a discusses the verbal form, but refrains from a definitive analysis, noting that the enclitic pronoun /=di/ can be either 3sg. or possibly 2sg. reflexive. His very tentative suggestion “he honours himself” is, however, nonsensical in the context. Similarly, HİSARCIK 1, §4 pleads directly with the gods: wa/i-tu-u 9-ta i+ra/i-wa/i-ti x-wa/i-ha, approx. /wa=du nuwatta irwadi …-ha/, translating as ‘I x-ed you nine times with (a) gazelle(s), (come on, keep alive your worshipper Kurti, your servant!)’ (cf. Hawkins 2000a:483).

[AH.B.]

The orthotone stem shows stem allomorphy with /ti/ for the nom., whereas the oblique cases are represented /tu/, see e.g. Yakubovich 2015b:6.4. Although the data are too scarce to prove it, it is likely that the oblique cases are no longer distinguished here either. Similarly, the 2sg. clitic pronouns rely on two forms as well, viz. /=du/ for the acc. and dat. objects, and /=di/ for the reflexive, mimicking the distribution of forms of the 1sg. (see e.g. Melchert 2003b:189f., Rieken 2004b:184, and Yakubovich 2015b:6.4). From an etymological viewpoint, it is likely that the dative used to be represented by /=di/ perhaps still in the 2^nd millennium, but since the 2sg. pronouns are relatively infrequent in the corpus for genre reasons, the relevant data, if any, have yet to be uncovered; see Yakubovich 2010d:168-171 for a plausible path of development. Despite their rarity in the attested data, the prevalence of the 2sg. pronouns in the spoken language was likely the cause for their forms being extended to the 3sg. for the acc./dat. and reflexive.

A previous abl./instr. form has been lexicalised with a specialised meaning, see /tuwadi/ ‘at your place’, while /tuwi(ya)-/ represents a derived adjectival possessive.

[AH.B.]

1. /tuwadi/

2. /tuwi(ya)-/

The 2^nd sg. Pronoun is attested mainly the Old Hittite festival text for the Storm-god of the sky and the Sun-goddess of Arinna, though it is also found in a ritual fragment (CTH 770) and a birth ritual (CTH 520).

[D.S.]

Only secured attestations of the 2^nd person singular are included in the above table, although the 3sg. =du ‘him’ and refl. =ti ‘him-, themselves’ look exactly like their 2^nd sg. counterparts, which can make an assignment to either of the two difficult.

The only certain case of nom. tī ‘you’ is KUB 35.133 iv 12’, where it is followed by a 2^nd sg. impv. verb (Starke 1990a:195 n. 637, CLL:225). The attestation of ti-ip-pa-wa[ (KBo 8.17 obv. 3’) interpreted as ti=ppa=wa[ ‘but you…’ is less than assured, especially since the rest of the text does not use the quotative particle =wa, but cf. CLL:225 and Kloekhorst 2014b:594. The segments ti-iḫ-ḫa-ti[(-) (KUB 35.100 rev. 2) have also been interpreted as ti=ḫḫa=ti[ (CLL:225, Kloekhorst 2014b:594), although the preceding word -d]u-ut-ta surely represents the particle chain =d]u=tta, which renders the two scholars’ interpretation very unlikely. Likewise, ti-i-iḫ-ḫa-ku-w[a(-) (KUB 35.106, 7’) is dividable as tī=ḫḫa=kuw[a (CLL:225, Kloekhorst 2014b:594), although the perspective of Rieken 2017a:23 to take tīḫḫa as a verbal form in the 1^st sg. pret., i.e. tīḫḫa=kuw[a is more attractive still. In sum, all these forms are not probative as attestations of the 2^nd singular personal pronoun ‘you’.

The cases of 2^nd sg. refl. in CLuw. are missing in the table by Melchert 2003b:189, but the first one adduced here (VBoT 60 obv. 7’) is made highly probable by its occurrence together with a verb in the 2^nd sg. impv., i.e. a=wa=ti kūri ‘Cut for yourself!’, but cf. kūri-. The second one is also clear, zunnimiš=ti ‘You are made alright!’; see Melchert 2003b:203 and under the Glossenkeilwort of zunni-^(ti).

Cases of =ti acting as the actual dative personal pronoun of the 2^nd singular are not attested yet. For a tentative proposal of reading one example of =ti as acc./dat. ‘(to) you’ in Ištanuwa Luwian; see CLuw. duwayā-⁽ⁱ⁾.

[D.S.]

Checked against the photographs in the Konkordanz (Mainzer Fotoarchiv).

[D.S.]

A discussion of the 2^nd sg. pronouns in CLuw. is found in Melchert 2003b:189f. and Yakubovich 2010d:166-171.

[D.S.]

1. tūwi(ya)-

The reflexive pronoun =ti is well attested across the Lycian corpus.

[D.S.]

Only a representative number of attestations is listed in the table above. Contra Carruba 1977a:311, there is no reflexive particle =di to be segmented anywhere; cf. przzidi(je)-, mahanahidi, tubehid-.

[D.S.]

Laroche 1958b:171f. recognizes on comparative grounds the existence of a Lycian reflexive particle =ti to be distinguished from the relative pronoun =ti-. He further compares the use of the particle in Lyc. =ti prñnawate ‘he built for himself’ with Hitt. =za parnawaišket ‘he took as residence’ (KUB 1.1+ iv 62-63). His analysis is followed by Carruba 1969b:91f., DLL:65, Neumann 2007a:356f., Gehrisch 2018a, Christiansen 2020a and Melchert 2021c. Borchhardt & Eichner 1997-1999a:62f. (and Kloekhorst 2011a) try to negate the existence of the Lycian reflexive pronoun, seeing instead the relative pronoun everywhere, which is easily refutable with the phrase tibe=ti=p(e)=ewẽ pijeti=tdi : e[ … ] (N 324.27), where a relative pronoun =tdi (dat. sg.) is clearly used together with a particle =ti, which cannot be interpreted as a second relative pronoun within the same clause. Moreover, relative pronouns in Lycian canonically occupy the position after the verb, except in conditional clauses where they are fronted (see under lemma Lyc. A =ti).

[D.S.]

This lexeme is attested 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.]

For the analysis of utetu as ute=tu ‘said to him (?)’, see Sasseville 2021d:183.

[D.S.]

The Lycian B particle sequences containing =tu are in Neumann 2007a and Gehrisch 2018a:90f. barely analyzed.

[D.S.]

Melchert (DLL:132) takes =tu as an enclitic pronoun ‘to him/her’, but ‘to you’ seems less likely to him. Sasseville 2018a:310f. brings evidence from Greek inscriptions composed for the same Lycian ruling dynasty that =tu might be translated with ‘to you’ in beneficial formulas. Schürr 2018b:82 assumes likewise that =tu might mean ‘to you’, since ‘to him’ is expressed by =i in Lycian. However, it should be pointed out that =i ‘to him’ is so far only found in the Lycian A dialect.

[D.S.]

1. tuwi(je)-

This particle is found in the inscriptions from Sardis and the Lower Hermos valley. It is already attested in the archaic inscription LW 54 (late 6^th or early 5^th century BCE), but the majority of inscriptions where it occurs can be broadly dated back to the 5^th or 4^th century BCE. Noteworthy is the unusually high concentration of this particle in LW 13.

[I.Y.]

Note that =τ in LW 13.12 may represent a morphophonemic variant of =s.

[I.Y.]

The reflexive function of =(i)τ is affirmed in Carruba 1969b:46, but essentially with reliance on its Anatolian cognates. The reason for that is that the majority of the relevant contexts still remain obscure. Perhaps the clearest one is the passage in LW 54.3, where the negligent owner of the tomb is threatened with “stealing it from himself”, i.e. forfeiting it. The reflexive particle =(i)τ is not restricted to a particular person or number of the subject (Yakubovich 2010d:180f.).

The interpretation of Gusmani 1964a:216f., according to which =(i)τ represents a free variant of =(i)t is easily falsifiable on syntactic grounds. The particle =(i)τ appears in the same syntactic position as indirect pronominal clitics, e.g. =mλ ‘to him’, while =(i)t occupies the slot after both indirect and direct pronominal clitics. The same consideration falsifies the approach of Melchert 1991a:136, 139-140, where instances of =τ are analyzed as combinations of =t and the emphatic particle =(i)s.

[I.Y.]

These pronominal forms are found across the Palaic corpus.

[D.S.]

For the possibility of seeing another case of the 2sg. reflexive pronoun, see Pal. šūšte. Pace Yakubovich 2010d:175, we are reluctant to accept the interpretation of =ti as a 2sg. dat. pers. pron.; the two passages he adduces are here interpreted otherwise, i.e. with =ti as 3^rd sg. refl. (see also Sasseville 2019a:27).

[D.S.]

Checked against the photograph in the Konkordanz (Mainzer Fotoarchiv).

[D.S.]

The existence of a reflexive enclitic particle =ti in Palaic goes back to Meriggi 1963a:5, who convincingly compares =ti azzikī ‘eat!’ (KUB 35.165 obv. 15 and 20) with the reflexive use of Hittite verb for ‘to eat’, i.e. ed-/ad-^(mi) (+ =za) ‘to eat’ (followed by Carruba 1970a:74, Yakubovich 2010d:174). The reflexive particle -ti can be used with a verb in the 2^nd person or 3^rd person singular.

The tonic form tū was already identified as a pronoun indicating the 2^nd person singular dative by Kammenhuber 1959a:39 (followed by Carruba 1970a:74). Carruba 1970a:75 analyzes the clitic form =du as a personal pronoun indicating the 3^rd person singular, but Melchert 1984b:35 points out rightly that some examples must be 2^nd person singular (followed by Yakubovich 2010d:175), which is supported by the tonic form tū belonging only to the 2^nd person. The extension of =du to the 3^rd person singular must be considered secondary (see Yakubovich 2010d:178).

[D.S.]

HEG W-Z:702-708