pa-(d)
‘to keep safe (?)’
- Language
- Lydian
- Grammar
- verb
- ID
- 857
1. Individual Anatolian Languages
1.1 Lydian
pa-(d)
‘to keep safe (?)’1.1.1Transmission
This lexeme is found in a metrical inscription (LW 14) from Sardis and on a fragmentary inscription from Manisa (LW 46). Both inscriptions are dated roughly to the 5th or 4th century BCE.
[D.S.]
1.1.2Forms
| 3sg. pres. act. | fapad | LW 14.18 (Sardis) | |
| šipad | LW 46.1 (Manisa) | ||
| 3sg. pret. act. | paal | LW 14.1 (Sardis) |
The attribution of these three forms to the same stem pa- goes back to Meriggi 1935a:109 n. 99 and Gusmani 1964a:195 and is adopted here for the sake of economy, although the semantic evidence is missing.
[D.S.]
1.1.3Meaning
According to Gusmani 1964a:74, the verb paal in LW 14.1 is intransitive and takes mrud ‘stele’ as its subject. Differently, Meriggi 1935a:109 n. 99 takes mrud ‘stele’ as the direct object of paal and suggests the meaning ‘to establish’. Sasseville 2021a:311f. follows the syntactic interpretation of Meriggi but suggests an etymological connection with Hitt. paḫš-(mi) ‘to protect’ and brings contextual support:
mru=m=it paal dum=iš qλ̣[…]ṣ factot (LW 14.1)
‘He, to whom they perform the payment, has kept the stele safe for me.’
In other words, the person who performs the action of pa-(d) towards the stele receives a payment (see under cato-(d)), which makes ‘to keep safe’ a plausible translation. The other attestations of šipad and fapad are found in fragmentary contexts, which hinders any verification of their semantics. In summary, the Lydian verb pa-(d) can be tentatively translated as ‘to keep safe (?)’.
[D.S.]
2. Reconstruction
2.1Reconstruction ‘to protect’
If the semantic and etymological analysis of Lyd. pa-(d) ‘to keep safe (?)’ is correct, a root verb of the lenited mi-conjugation can be reconstructed for Proto-Anatolian, i.e. PAnat. *pā́-/p-´(di) (< *péh2-/ph2-´) (Sasseville 2021a:314-318). Only the second and third laryngeal are possible from an Anatolian perspective. Regarding the diverging stem formations, the Hittite cognate paḫš- ‘to protect’ (also spelled paḫḫaš-) shows an s-extension (EDHIL:612), behind which the laryngeal was kept. The morphological origin of the Hittite formation in -š- is discussed in Oettinger 2013a:58 and HEG P:362-364 with further literature. On methodological grounds, both formations should be back-projected into Proto-Anatolian and their relationship discussed within the Indo-European frame.
[D.S.]
3. Proto-Indo-European
3.1Semantic Reconstruction
3.2Morphological Reconstruction
The PIE root *peh2(i̯)- ‘to protect, guard; to graze animals’ is well attested in the IE languages, with numerous verbal and nominal formations derived from it. Among the verbal formations, there is, e.g., an athematic root present PIE *péh2-/*ph2-´ > Lyd. pa-(d) ‘to keep safe’ vel sim. (cf. above the PAnat. reconstruction), Ved. pā́ti, YAv. pāiti ‘protects’. A present *p(e)h2-sk̑e/o- is the basis of Lat. pāscere ‘to feed, pasture’, TochB paskentär ‘they protect’. Among the many nominal formations, the joint evidence of Myc. po-me = /poi̯mēn/ nom. sg. m. ‘herdsman’, Gk. ποιμήν ‘id.’, and Lith. piemuõ ‘id.’ points to an already PIE word for ‘herdsman’, which can be reconstructed as PIE *poh2imén-.Cf., e.g., Schaffner 2005a:166–175 with a very thorough discussion and further lit.
[T.S.]
4. Transmission
TochB paskentär, TochA pāsantär ‘they protect’Cf., e.g., LIV²:460 with n. 3, DTB²:392f., Malzahn 2010a:698f.