*h1ép-/h1p-´
‘to seize’
- Language
- Proto-Anatolian
- Grammar
- verb
- ID
- 3612
1. Individual Anatolian Languages
1.1 Cuneiform Luwian
āp(p)-(ti)
‘to seize, hold’1.1.1Transmission
The verb is attested in an incantation in a ritual of the Kuwattalla tradition.
[E.R.]
1.1.2Forms
| 3sg. pres. ind. act. | a-ap-ti | KBo 29.22, 5’ | (CTH 760, NS) |
The assignment of a-ap-ti at KBo 29.22, 5’ depends on the reading of the preceding signs (see under Graphic Features). – For another possible instance of this verb, see CLuw. appa.
[E.R.]
1.1.3Graphic Features
Checked against the photographs in the Mainzer Fotoarchiv. – Sasseville 2021a:261f., 558, (contra Starke 1990a:199) reads n[i-i]š a-ap-ti, which, if correct, would leave no doubt about the verbal interpretation and morphological analysis. Unfortunately, the reading of the putative negation is not entirely clear, but it is more likely than not.
[E.R.]
1.1.4Meaning
Due to the fragmentary context, the suggested meaning ‘to seize’ (Sasseville 2021a:261f., 558) relies entirely on the assumed relationship with Hitt. ēp(p)-/app-(mi) ‘to seize, hold’ and the Luwian reflexes of this root in Hittite transmission.
[E.R.]
1.1.5Stem
Sasseville 2021a:261f., 558 identifies the verb as a root stem of unlenited mi-conjugation.
[E.R.]
1.1.6Derivatives
1.2 Lycian A
app-(ti)
‘to take’1.2.1Transmission
This lexeme is attested in a grave inscription (TL 29) including a long biography of the owner dating to the reign of Alexander the Great, as well as in a dedicatory inscription (TL 40) dating to the tenth year after the Persian occupation of Xanthos (first half of the 4th century BCE). Contra Shevoroshkin 2002c:405f., there is no evidence of this verb in Lycian B; instead we see the Lycian adposition epe.
[D.S.]
1.2.2Forms
| 3sg. pret. act. | apptte | TL 29.9 (Tlos) |
| epp!tte (?) | TL 40d.2 (Xanthos) |
The form ap[.]di (TL 94.3) cannot belong to the same lexeme due to its lenited conjugation (pace Gusmani 1962a:169, DLL:3). Instead, it may belong to ep(e)i-(di), although the obscurity of the word [..]edehe in the sentence hinders a reliable interpretation. Pace Melchert (DLL:14), there is no need to separate epatte from apptte by making the former a stem in -a-(ti). Moreover, whereas Hajnal 1995a:184f. sees a univerbation of a noun *epa- with a verb *tate ‘he set up’ (instead of expected tade with lenited ending) to explain the strange geminate -tt- in the ending, Eichner 2017a:292f. suggests a medio-passive form in -tte comparable to Hitt. -ttati, with syncope and assimilation of *-td- > Lyc. -tt-, which is far-fetched and not supported by the general preservation of the Lycian cluster -td- (see tdi in the relative pronoun ti-, and tddãta). The emendation suggested under Graphic Features settles the problem and renders these awkward solutions unnecessary.
[D.S.]
1.2.3Graphic Features
Since the letters a and p look very much alike, it is tempting to emend the form epatte to epp!tte, which would better account for the gemination of the ending following a consonant and unify this form with the other apptte, with epp!tte showing a secondary vowel harmony (Sasseville 2021a:281; considered unsatisfactory by Réveilhac 2021a:560 n. 13 because of the four consecutive consonants). Nonetheless, this emendation needs further evidence and thus should be considered tentative.
Differently, Schürr 2007b:115 emends the form apptte to apa!tte in analogy to epatte, which does not solve the problem as it does not explain the unexpected gemination of the ending. Eichner 2017a:292f. makes the same emendation in order to retrieve another “medio-passive” form in -atte; but see under Forms.
[D.S.]
1.2.4Meaning
The meaning of apptte (and epatte) has long been clear not only on contextual grounds, in which it refers to the appropriation of Lycia into Alexander the Great’s dominion (TL 29.9), but also on etymological grounds with Hitt. ep-/app-(mi) ‘to seize, take’, and it was translated accordingly (Gusmani 1962a:169 with n. 34, 169, Laroche 1974b:139, Carruba 1977a:280, Carruba 1978b:176, Starke 1990a:320, Eichner 2017a:292f., Gehrisch 2018a:39, Sasseville 2021a:281f.). In the context of TL 40d.2, however, the direct object xruwata ‘thank offerings’ expressed in the first clause is assumed to be implied in the second clause (see Sasseville ibid. and cf. van den Hout 1995c:158 and Eichner ibid., contra Gehrisch 2018a:59).
[D.S.]
1.2.5Stem
Lycian app-(ti) reflects a root verb of the mi-conjugation (Sasseville 2021a:281f.). Outdated is the analysis as a preverb and a verb in Borchhardt & Eichner 1997-1999a:67.
[D.S.]
1.2.6Derivatives
1.3 Palaic
āp-(ti)
‘to seize’1.3.1Transmission
This lexeme is found in a festival text.
[D.S.]
1.3.2Forms
| 3sg. pret. ind. act. | a-ap-ta | DBH 46/2.101, 7’ | (CTH 753, NS) |
Its interpretation as a verbal form is assured due to the following niwa=aš=ti, which contains a particle chain. What appears to be at first sight a second attestation within the same context is most something else, because weḫenta must be the finite verb of the sentence: (-)]˹a-ap?˺-ti=pa ú-e-ḫe-en-t[a] (DBH 46/2.101 + KUB 35.159, 10’/14’). Moreover, the word (-)]˹a-ap?˺-ti could be acephal.
[D.S.]
1.3.3Meaning
The word āpta is found in a fragmentary context pertaining to the dissection of a sacrificial animal. The meaning ‘to seize’ for Pal. āp- is drawn from the etymological comparison with Hittite ēp-/ap-(mi) ‘to seize’, which is solid, but lacks contextual confirmation.
[D.S.]
1.3.4Stem
As expected, the weak grade has been generalized just like in other Palaic verbs; cf. āš-/aš-(ti) ‘to be’, hwā-(di) ‘to walk’.
[D.S.]
1.4 Hittite
ēpp-/app-(mi)
‘to seize’1.4.1Literature / Comments
HW 2nd ed. E:44-88
2. Reconstruction
2.1Reconstruction ‘to seize’
Comparing CLuw. āp-(ti), Pal. āp-(ti), Lyc. A app-(ti) with Hitt. ēpp-/app-(mi) ‘to seize’, a PAnat. root verb of the mi-conjugation can be reconstructed, i.e. *h1ép-/h1p-´ ‘to seize’ (EDHIL:243, pace Oettinger 1979a:88 “acrostatic”, unclear HEG A-H:107f.). As per Melchert 1994a:313 and Kloekhorst 2006a:104, the vowel -a- in Lyc. app-(ti) must reflect a generalization of the weak stem, although not in the way described by Melchert, who uses the back vowel harmony from the outcome of *n̥ in the ending *-n̥ti from an acrostatic paradigm.Outdated is also the anaptyptic interpretation with subsequent lenition of the ending, which is now contradicted by the emergence of CLuw. āp-(ti).. The same generalization of the weak stem must be adduced for Pal. āp-(ti). See also the causative-iterative *h1op-éi̯e/o- in Lyc. ep(e)i-(di).
There is also evidence for a lengthened grade *h1ēp- in CLuw. īppatarr-išša-(i) ‘to distrain’, which would reflect an instrument noun *h1ēp-e-tro- ‘instrument of seizing’. For the lengthened grade in another Anatolian instrument noun formation, see HLuw. /ist(a)rtt(a)-/ ‘throne’ < *h1ēs-tro- + -to-.
There is also potential evidence for an acrostatic formation of the same root, i.e. *h1ḗp-/h1ép-, which however depends on the etymology, see CLuw. *ip(a)- (ipama/i-). Note also that an instrument noun is also found as the base of CLuw. ipatarma, though with lenition of the labial consonant of the root.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
3.2Morphological Reconstruction
Based on the root semantics, we would expect a root aorist *h1ép-t, *h1p-ént ‘seized’ as the primary tense stem (cf. Eichner 1975a:82, LIV²:237 s.v. *h1ep-). The PAnat. mi-conjugation verb *h1ép-/h1p-´ ‘to seize’ seems to continue this root aorist in line with the “tēzzi principle” (cf. Malzahn 2010a:267f.) according to which new present stems could be formed to already existing preterites (e.g., root aorists or imperfects) by replacing the endings with the respective primary endings. Compare Hittite tēzzi ‘says’, Lycian tadi ‘puts’ from root aor. *dhéh1-t ‘put, placed’ (cf. also Jasanoff 2003a: 3, 151; Eichner 1975a:81, 88, 93) note that the one attestation of a 1pl pret. act. appuen ‘we seized’, according to EDHIL:243 “the only plural preterite form of a mi-verb to show zero grade instead of regular full grade”, might reflect the old ablaut behavior of the root aorist, on which see Malzahn 2004a. As for present stem formations within PIE, a root aorist could typically be paired with a Narten root present. Such a present *h1ḗp-ti, *h1ép-n̥ti ‘seize(s)’ is not attested directly but might explain the almost consistent long vowel allomorph in the Vedic and Avestan averbo (see above in the table for an alternative explanation). It might also have left traces in some Luwian derivatives (especially ippattarišša-(i) ‘to distrain, take possession’), see Proto-Anatolian *gei̯bh- or *h1ḗp-?, and in Hittite epurā(i)-(mi) ‘to move earth, build a siege ramp’ (Eichner 1973a:79f.). A causative/iterative *h1op-éi̯e/o- is attested in several branches, in some of which it probably constitutes a recent creation (e.g., Vedic abhy-āpayati ‘to bring to an end’ [ŚBr.]). Interestingly, Albanian (j)ep (3sg), (j)ap (1sg) ‘to give’ (from causative *h1op-éi̯e/o- ‘to make someone seize’) matches the most likely semantics of Lycian epeite (see Lyc. ep(e)i-(di) ‘to give (vel sim.)’).
[S.H.]
4. Transmission
Hittite ēpp-/app-(mi) ‘to seize’, Cuneiform Luwian āp-(ti), Palaic āp-(ti), Lycian A app-(ti) ‘to take’, Lyc. ep(e)i-(di) ‘to give (vel sim.)’.
See above and EDHIL:242f. s.v. epp-zi / app-.
Vedic ā́pa (perf.) ‘has reached, obtained’, ā́pat (aor.) ‘obtained’, apsanta (desid.) ‘want to obtain’, ī́psati (desid.) ‘wishes to obtain’, āpnóti ‘obtains’, āptá- ‘reached, obtained’the long vowel of the perfect *h1e-h1(o)p- spread throughout the averbo in Indo-Iranian (except for the desiderative), see LIV²:237 s.v. *h1ep- note 4..Cf., e.g., EWAia I:167 s.v. ĀP, Werba 1997a:157f. s.v. *ap, LIV²:237 s.v. *h1ep-.
Old Latin apiō, -ere ‘to bind, snag’ (Paul. ex Fest.), apīscor, -scī, aptus sum ‘to, seize, get, obtain’ (Naev.+), perf. co-ēpī ‘began/have begun’, Umbrian *opē- (3rd sg. fut. ipv. upetu, ppp. upet, etc.) ‘accept, take’ (*h1op-éi̯e/o-), Old Latin opet (CIL² 4) ‘accepts’, optō, -āre ‘wishes’.Cf., e.g., WH I:57f. s.v. apīscor, EDLIL:47 s.v. apīscor, Weiss 2010a:95f., LIV²:237 s.v. *h1ep-, LIV2+ s.v. *h1ep-.
Albanian (j)ep (3sg), (j)ap (1sg) ‘to give’ (< *h1op-éi̯e/o-).Cf., e.g., LIV²:237 s.v. *h1ep-, Schumacher & Matzinger 2013a:978 s.v. jep.