*pói̯h₁-/-ih₁-´

One of the most well-attested verbs, this lexeme can be found 88 times in the corpus. The evidence covers most of the HLuw. geographical area from Kocaoğuz in the west to the Commagene in the east, and Boğazköy in the north to Tell Ahmar in the south. Likewise, the texts date to all periods of the HLuw. era from Empire around 1300 BCE to late Iron Age in the early 7^th century BCE.

[AH.B.]

The above table presents merely a selection of the 88 forms attested. Care has been taken to create an accurate picture of the temporal and spatial distribution of the present lexeme, although broken attestations have been omitted unless they contribute information not found elsewhere.

[AH.B.]

Representing a basic level term from the verbal domain, /piya-/ is frequently attested, and its meaning was assigned at an early stage in the study of HLuw. by Hrozný 1933a:232, not least due to the very pictorial sign *66 being used, albeit with readings still askew (e.g. /pa/ instead of /pi/) and just a small number of the available forms correctly identified.Cf. e.g. Hrozný 1933a:197 on KARKAMIŠ A1a, §8, where the erroneous interpretation of the signs as 〈pá-ā-u〉 prevented the author from attributing the form 〈pi-ia-ha〉 to the same paradigm. This quickly found acceptance, see e.g. Meriggi 1935b:48 and Gelb 1935a:12, who pointed out the cognate link to Hittite, while its CLuw. cognate received its own corrected interpretation by Meriggi 1957a:219 n. 67 at a much later date.

With multiple attestations in various contexts, the present verb provided the analytical platform on which the verbal forms in /-mina/ received their new grammatical interpretation as gerundives, see Melchert 2004c:356f.

Because it is one of the basic verbs, /piya-/ is used in a wide range of contexts. It can denote the actual action of giving, but similarly is used in broader, partly metaphorical contexts as in giving or contributing to a feast, e.g. in CEKKE, §11 (Hutter-Braunsar 2019a:261f.). In addition to that, the verb appears not only by itself, but also in combination with a number of preverbs, which create more specialized meanings. Hawkins & Morpurgo Davies 1982a:94f. can show that /anni ahha piya-/, usually written as CUM-ni ARHA pi-ia-, denotes ‘to sell’, see e.g. KARKAMIŠ A4a, §1. Combined with the preverb PRAE into /pari piya-/, the verb denotes ‘to hand over’ as e.g. in BABYLON 1, §1 (Hawkins 2000a:294, 393f.).

[AH.B.]

One of the most well-attested verbs in the HLuw. corpus, /piya-⁽ⁱ⁾/ serves as a prime example for its verbal stem class, viz. the verbs in /-iya-⁽ⁱ⁾/. Like its Hittite cognate pāi-/-i(ya)-^(ḫi) ‘to give’, the sequence /-iya-/ is part of its root instead of a suffix, and the verb is classified as a root verb. Again like its Hittite cognate, it inflects according to the ḫi-conjugation, but in contrast to it, shows no stem allomorphy, instead relying on the weak stem throughout (cf. Frotscher 2012a:157f., Sasseville 2021a:408-410, 425-427).

[AH.B.]

1. /pibassa-/
2. /piya-/
3. /piyatta-/
4. /piyaRiya-/
5. /piyattara(/i)-/

The verb is attested in both incantations and OH festival texts, among them the Ištanuwa festivals. The earliest manuscripts show MS.

[E.R.]

The imperative form pāiu was interpreted as a form of pāi-/pā-^(mi) ‘to go’ in Meriggi 1957a:219 n. 67, but the ditransitive use in KBo 4.11 rev. 54 du-uš-ša-ni-ia-al-la-aš-mi a-ia-tar pa-a-i-ú ‘the d. shall give me a.’) rather speaks for the assignment to an Ištanuwian ablauting verb pāi-/-i(ya)-⁽ⁱ⁾ ‘to give’ (first suggested in Rosenkranz 1952a:80; Bossert 1954a:101, see also CLL:163, Sasseville 2021a:408-410; still undecided DLL:77).

[E.R.]

Checked against the photographs in the Konkordanz except for Privat 40. – Reading of pí-u-un-ni in KBo 4.11 obv. 27 is difficult to confirm on the photograph, but probably correct.

[E.R.]

Already Forrer 1929a:207 assumed that Luw. piya-⁽ⁱ⁾ means ‘to give’ based on the equation of the names ^mPiyamaradu and ^mSUM-maradu (with the spelling SUM-ma- for the Luw. participle piyamma/i- ‘given’). After initial skepticism (see the polemic criticism by Sommer 1932a:75f.), at least the translation of the stem piya-⁽ⁱ⁾ as ‘to give’ was accepted, cf. DLL:82, CLL:178, HEG A-K:378, HED PA:55f., Oettinger 1979a:510, Jasanoff 2003a:138, Sasseville 2021a:408-410.

The semantic assignment of piya- is further supported syntactically by its ditransitive use with direct and indirect objects, as well as contextually because its imperative occurs in a wish for blessings that the Sun-god is asked to bestow on the Ḫatti lands (KBo 4.11 rev. 53-55).

[E.R.]

While the Ištanuwian dialect still preserves both the strong and the weak stem (pāi- vs. pi-), the weak stem pi- is generalized everywhere else (CLL:163, Jasanoff 2003a:94, Sasseville 2021a:426). The classification as -iya-⁽ⁱ⁾ verb is based on the ablaut seen in the Ištanuwian forms and the comparison with Hieroglyphic Luwian /piya-⁽ⁱ⁾/, where the significant 3sg. pres. pi-ia-i is attested.

[E.R.]

This lexeme is found in various inscriptions from Lycia.

[D.S.]

The discovery of the trilingual inscription brought decisive evidence for the semantic assignment of pije-^(ti) as ‘to give’ (equivalent to Greek διδόναι ‘to give’) (Laroche 1979c:66, though meaning already suspected by Laroche 1973b:94f.). The previous popular translation ‘to determine’ hence became obsolete (see Neumann 2007a:271 with further literature, DLL:49f.). Bryce 1975a argues that the verb pije-^(ti) in grave inscriptions does not simply mean ‘to give’, but rather ‘to allocate’ or ‘to make available’. In other words, it would be used when granting the right to additional people to be buried in the grave or in other parts of it (see Christiansen 2020a:226).

[D.S.]

1. mahanepi[jẽme/i-] (PN) (Neumann 2007a:191f.)
2. arm̃mazapi[me/i-] (PN) (Onur & Tekoğlu 2020a:23f.)

1. *pije-
2. pibije-^(ti)
3. pijata-
4. *pijatra-

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.]

Contra Gehrisch 2018a:95, who takes pijelu as a nominal form and even as the first member of a compound pijelumlez. As shown in Sasseville 2014-2015a:110-112, there is no suffix **-ela-, which would run counter to the rules of Lycian vowel harmony, but only -ala- and -ele/i-. The assignment of pijelu as a verbal form of the 1 singular imperative (cf. Hitt. -llu, CLuw. -(l)lu) is supported by context, which shows other verbal forms in the same person. For the phrase muxssa pijelu ‘I must give an invocation ritual (vel sim.)’ and a parallel in Hittite, see under muxssa-.

[D.S.]

The semantic assignment of Lyc. B pije-^(ti) is due to its equation with Lyc. A pije-^(ti) (DLL:124). Both forms attested are ditransitive, i.e. with a direct and indirect object, which renders the meaning ‘to give’ perfectly adequate (thus also Schürr 2018b:97).

[D.S.]

1. pina-^(ti)

Adiego 2017a:14 suggests recovering in the sequence οαιαρπι- // ποσιτι (S 2.3-4) the word αρπι-, which he equates to a personal name Αρπιας found in Greek Transmission. On etymological grounds, he sees the name as a compound made of ar- and piya-, i.e. a form of the Luwian verb piya-⁽ⁱ⁾ ‘to give’. Though possible, it is not provable. More evidence is needed.

[D.S.]

Reconstructed from Carian personal names in Greek transmission from Caria between the 5^th/4^th and 2^nd centuries (see under Meaning).

[Zs.S.]

The attestations follow Adiego 2007a:339 (cf. also Blümel 1992a:19). Hajnal 1995b:24 wanted to identify the participle also in the personal name Kbjom, but this is improbable, see the discussion s.v.

[Zs.S.]

Adiego 2007a:339 compared Ερμαπις and Μασσαραβις as well as Νετερβιμος with the well attested Anatolian (Luwic) compound names of the structure X-piya- and X-piyamm(i)- based on the Luwian verb piya- ‘to give’ calling attention to the practically identical forms of Luwian Armapiya- (Laroche 1966b:39 No. 135; Adiego 2007a:331 [see there for an explanation of the Anlaut, contra Sayce 1893a:122, who claimed an Egyptian origin], see already O. Masson 1959a:170, Houwink ten Cate 1961a:133, 176, and Blümel 1992a:13) and Lycian Natrbbije͂mi / Ἀπολλόδοτος (also Adiego 2007a:332 and Melchert 2013c:41). Formally and semantically speaking this is certainly an attractive analysis and thus, it is accepted here, too. Nevertheless, some details deserve more attention:

1. Μασσαραβις is explained by Adiego 2007a:332 as masara-bi- that “seem[s] to contain an r-form of the stem [known in Luwian as māššan(i)- ‘god’ ‒ Zs. S.], as in Sidetic”. There are two problems with this analysis: first, the n > r rhotacism is not attested yet in Carian and, second, the Carian word for ‘god’ (mso / *msn-, msñ-) is only attested with an [n] (see s.v.). A possible solution is provided by the fact that a Carian god identified by the Greeks with Dionysus is called Μάσαρις according to Stephen of Byzantium (Adiego 2007a:7f., 332 with ref.). Assuming a formally fitting i-mutating a-stem for Μάσαρις, Μασσαραβις may represent a theophoric name based on the Carian name of Dionysus (although note the slight difference in the sibilants) ‒ interestingly enough, the son of Μασσαραβις is called Διονύσιος (LGPN V.B:274; although Zgusta 1964a:302, Blümel 1992a:19 and Adiego 2007a:332 call attention to the theonym in connection with the personal name, they do not explain their relationship. Whether Μάσαρις is connected to the Carian word for ‘god’, as Adiego 2007a:332 assumes [with a misleading misprint -σσ-], is a different question and must remain open, but due to the lack of rhotacism in Carian it is less probable).
2. Thus, we have three, morphologically and semantically regular theophoric names. The problem is that only Ερμαπις shows a phonologically regular form with [p], both other names have [b], which is unexpected and irregular. The difference cannot be phonologically motivated since both Ερμαπις and Μασσαραβις have [b] in the same, post-vocalic position. One could at best assume the spread of a secondary, allophonic variant [bi-], but as yet there is no evidence for such a phonological change in Carian. Another option would be to look for chronological or geographical differences. There is definitely no chronological distribution since these forms are attested contemporarily (Ερμαπις: 5^th/4^th c. ‒ 2^nd c. vs. Μασσαραβις and Νετερβιμος: 459 ‒ c. 354, see LGPN V.B. s.vv.). It is remarkable, however, that Ερμαπις is attested in the main Carian area (Halicarnassus, Mylasa, Pladasa), but the forms with [b] are attested in the Carian periphery, Μασσαραβις in Miletus and Νετερβιμος in Kaunos, at the Lydian-Carian language border and close to Lycia, respectively. Although the first members of these compounds do not fit to a Lydian or Lycian origin (the Lydian word for ‘god’ is ciw- [see s.v.] and no cognate of māššan(i)- has been detected yet; the Lycian equivalent of “Neter-” is Natr(i) and, accordingly, Νετερβιμος cannot be explained as the Greek transcription of Natrbbije͂mi, contra Blümel 1992a:20 n. 65 following an idea of O. Carruba and D. Schürr), both languages show similar or identical phonological changes that could explain the [b] in these names (stops are underlyingly voiceless in Lydian with voicing in certain environments and Lycian shows voicing next to /r/, cf. esp. Natrbbije͂mi [contra Kloekhorst 2008a:125, whose argument for a voiceless fricative is not compelling]). Considering the existence of Lydian-Carian and Carian-Lycian language contacts as well as the special position of the dialect of Kaunos (on these issues see most recently Schürr 2019d:6‒14), it might be suggested that the unique phonology of these forms resulted from the contact with the neighbouring languages.

[Zs.S.]

Rejected derivatives:

1. bebi
2. bebint
3. bebnd
4. †biδ-
5. binq
6. †pint-
7. pisñ(o)
8. psnλ(o)

All the forms of this verb occur in the inscriptions of various genres that are all found in Sardis, except for TL 50, which is found in the Middle Kaystros valley. The earliest text is TL 99 (6^th century BCE), the latest is probably TL 50 (late 4^th century BC), the other texts can be dated only more loosely to the 5^th or 4^th century BCE.

[I.Y.]

The appurtenance of the form pil in LW 99 to this paradigm is questionable. The inscription is fragmentarily preserved, and the interpretation of pil as a pronominal form cannot be fully ruled out.

The late pill probably reflects the analogical extension of the stem variant pid- followed by anticipatory assimilation *pidl > pill (Gérard 2005a:76).

[I.Y.]

The identification of pi- as a verbal form is found in Brandenstein 1931b:14f., the correct grammatical analysis of pidν is due to Vetter 1959a:44. While -d- is still treated as a root extension in Gusmani 1964a:79, the correct analysis of the stem-formation of pi- is found in Melchert 1992a:38.  For the semantic view point, its most transparent form is fa-pil, but it was nevertheless translated as ‘to produce’, even when its connection with pi- had been acknowledged, see Gusmani 1964a:113 and cf. Gusmani 1980a:53.  For the correct semantic analysis of pi(d)- in LW 50, see Gusmani 1960a:282) more comprehensive discussion with Anatolian parallels is found in Gusmani 1964a:79f.

[I.Y.]

1. pida-
2. pin-
3. pita-^(d)
4. pita-^(t)
5. pitocν

1. pīš-

CHD P:18-55, EDHIL:614-617, HED PA:39-57, HEG P:376-384