*=pe

This lexeme is attested more than 400 times across all regions from the 13^th to the early 7^th century BCE.

[A.P.]

The above table offers a representative selection, omitting broken attestations.

All attestations occur in clitic chains, cf. the discussions below; cf. also HLuw. /=ppa/, whose existence as a separate particle is not assured.

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

The phonological interpretation of this word is generally regarded as secure. As a clitic, the word carries no stress and is therefore only found in Wackernagel position, i.e. attached to the first stressed word of a HLuw. clause. While the HLuw. writing system is unable to convey that the initial plosive of the particle is lenited, the quality becomes evident in the evidence for the cognate CLuw. =pa, which is consistently spelt with singleton P (see Melchert 2003b:209 and below under lemma).

[AH.B.]

The word is a very frequently found particle in the HLuw. corpus and connects clauses. The commonly used translation is ‘but’, yet it should be noted that its meaning is in many cases less adversative than the choice of English but may imply and therefore often goes untranslated. Instead, /=ba/ serves to structure the text, because it is used for signalling changes in the informational flow, e.g. a turn in the narrative, a side remark, or the beginning of a new event etc., cf. e.g. Laroche 1958b:167 and Melchert (DCL:s.v.) for CLuw., and Melchert 2003b:208f. for both varieties of Luwian.As Melchert notes, delineating the functions of /=ba/ and /=ha/ may be less straightforward esp. towards the end of the HLuw. era. The two HLuw. versions of the KARATEPE 1 bilingual do not always employ the same particle in some places, see e.g. §8 and §14. This can be observed e.g. in KARKAMIŠ A2+3, §9 and its context: *a-mu-pa-wa/i-tu za-ia (DEUS)TONITRUS-sa DEUS.DOMUS-tà BONUS-sa₅〈+ra/i〉-ti-i za-la *261.PUGNUS-ru-ha, approx. /ammu=ba=wa=du zaya Tarhuntas …-hada wassaradi zala …-ruHa/, translating in the context as ‘(… and in my holdings, one sheep stood for ten homers of barley.) I/=ba/ myself supported these temples of (the Storm-god) Tarhunt for him (scil. the Storm-god) with goodness. (He/=ba/ who smashes my name …)’ (cf. Hawkins 2000a:109).

[AH.B.]

Indeclinable and cognate with CLuw. =pa, the HLuw. particle also assumes the same slot in the clitic chain attaching to the first stressed word in the clause, viz. the very first slot otherwise used for the conjunctive particle /=ha/ ‘and, also’ (cf. e.g. Laroche 1958b:161, Carruba 1985b:95, and Yakubovich 2015b:7.1); the two conjunctions are mutually exclusive, like the corresponding Hittite particles =(m)a ‘but’ and =(y)a ‘and, also’, cf. e.g. Carruba 1985b:82, EDHIL:537f., and Melchert (DCL:s.v.).

The connective is also found combined with the two negative particles /na/ and /ni/, forming a new connective with the disjunctive meaning ‘or’, cf. e.g. Yakubovich 2015b:6.6.

[AH.B.]

1. /niba/
2. /naba/

The particle occurs more than 50 times in all text types except for the festival descriptions (from MS onwards). A full list of attestations is offered in DCL:s.v.

[E.R.]

CLuw. =pa does not appear to have any allomorphs.

[E.R.]

Checked against the photographs in the Mainzer Fotoarchiv.*

[E.R.]

In Yakubovich 2022b:214f., enclitic =pa is described as a “weak sequencer”, but with its semantics equivalent to that of Hitt. =a/=ma (Melchert 2003b:208f.), it also includes a contrastive and counter-expectational meaning. As such, it can indicate a change of the frame of the action/situation, i.e. a change of topic or the temporal, locational, and conditional frame (see also CLL:161, DCL:s.v.).

[E.R.]

*The forms in the table were checked together with Marian Wehrstein.

This particle is well attested across the Palaic corpus.

[D.S.]

Checked against the photograph in the Konkordanz.

[D.S.]

Palaic pa is confined to the clitic chain following the first accented word of the sentence. Contra Carruba 1970a:66, the single spelled -pa as opposed to (=)p-pa is not confined to the position after consonant; cf. KUB 32.18 i 3’, 4’, 5’, DBH 46/2.101, 10’. The function of =pa seems to be contrastive, e.g. aḫuwanti ni=pp(a)=aš ḫašanti ‘They drink, but they do not quench their thirst.’ In some cases, it may indicate a sudden change of subject (or perspective), e.g. (the king and queen serve the bread preparation) ^dzaparwāš=an=pa=ti takkuwāti ‘Zaparwa on the other side looks at it.’

[D.S.]