*arou̯eh2-
‘to free’
- Language
- Proto-Anatolian
- Grammar
- verb
- ID
- 3338
1. Individual Anatolian Languages
1.1 Hieroglyphic Luwian
/arawa-(ti)/, a+ra/i-wa/i-
‘to set free’1.1.1Transmission
A hapax legomenon, this lexeme is attested once in an inscription from Karkamiš, which is dated to the late 8th century BCE.
[AH.B.]
1.1.2Forms
| 3sg.? pret. act. | /arawatta/ | a+ra/i-wa/i-ta | ADANA 1, §2 (Karkamiš) |
In the editio princeps (Hawkins, Tosun & Akdoğan 2013a:3), the form 〈a+ra/i-wa/i-ta〉 in ADANA 1, §2 is analysed as a 3pl. pret. act., i.e. /arawanta/ ‘they set free’, but see discussion of the meaning below.
[AH.B.]
1.1.3Graphic Features
1.1.3.1Phonological Interpretation
The phonological interpretation of this lexeme largely depends on its belonging to the same word family as HLuw. /arawann(i)-/ ‘freeman’. If this very likely connection holds true, then the interpretation of the form can be seen as secure, save for the ending, which remains ambiguous between a 3sg. and a 3pl. pret. act., cf. under meaning below. The word is at least partially cognate with Hittite arawaḫḫ-(i) ‘to free’ and probably also Lydian rawa- (see under lemma).
[AH.B.]
1.1.4Meaning
The link to the slightly more frequently attested noun /arawann(i)-/ ‘freeman’ is quite evident and was first suggested in the editio princeps by Hawkins, Tosun & Akdoğan 2013a:3. It receives further support from the context, because the verb occurs in ADANA 1, §2, a dedicatory inscription on a stele for the storm-god providing details about who set it up. One clause reads wa/i-mu za-ti (“MÁ”)ma-sa-hu-na-li (DEUS)TONITRUS-hu-ti PRAE? a+ra/i-wa/i-ta, i.e. /wa=mu zatti mashunalli Tarhunti parran arawatta/ ‘He (scil. the country-lord from the previous clause) set me free? before this Tarhunza of Prosperity’. Orthographically, the attested form is ambiguous regarding number and can stand for 3sg. /arawatta/ or 3pl. /arawanta/, but since the preceding clauses only mention the lord of the person who set up the stele without introducing any further actants, it seems more appropriate to assume a singular verb form.
[AH.B.]
1.1.5Stem
The lexeme is likely a factitive verb in /-a-/, derived from a putative adjective */arawa(/i)-/ ‘free’, which is not attested in HLuw., but whose direct cognate can be found in Hittite arāwa- ‘free’, see Sasseville 2021a:65 and Rieken & Sasseville 2014a:304f. Luwian factitives in /-a-/ usually inflect according to the ti-conjugational pattern (Sasseville 2021a:65).
[AH.B.]
1.2 Lydian
rawa-
‘to release’1.2.1Transmission
This lexeme is attested in two votive inscriptions found in the areas to the east and south of Sardis respectively. The first inscription can only be approximately dated to 5th or 4th century BCE, the second one mentions the 12th regnal year of Alexander and thus can be dated to the late 4th century BCE.
[I.Y.]
1.2.2Forms
| 3sg. pret. act. | [r]ạwal | LW 45.4 (Kogamos valley) | |
| rawal | LW 50.5 (Tire) |
1.2.3Graphic Features
Despite the fragmentary preservation of the first form, its restoration seems secure, because the preceding word faτaν is identical in both inscriptions.
[I.Y.]
1.2.3.1Phonological Interpretation
A word beginning with r- is unusual not only for Lydian but also for the Anatolian languages in general and can be most easily explained by assuming a historical aphaeresis (Gérard 2005a:51).
[I.Y.]
1.2.4Meaning
The second context makes it clear that rawal denotes a favorable action by Zeus in response to which the commissioner of the inscription erects a votive object. Accordingly, the meaning ‘to grant’ was proposed in Brandenstein 1931b:35 and accepted in Gusmani 1964a:189. The question is how to reconcile this meaning with the formally attractive comparison with Hitt. arawaḫḫ-(i) ‘to free’, offered in Gusmani 1980a:86. One possibility is to start with the basic meaning ‘release’ and assume that it was used in these contexts for the revelation of the divine will (for example, via oracular inquiry), which was favorable for the commissioner of the inscription.
[I.Y.]
1.2.5Stem
The comparison with the Hittite stem arawaḫḫ-(i) speaks on face value for the factitive stem rawa-(t), but as long as the absence of lenition is not confirmed through the synchronic attestation of diagnostic forms, this stem assignment must remain tentative (Sasseville 2021a:65).
[I.Y.]
1.3 Hittite
arawaḫḫ-(i)
‘to set free’1.3.1Literature / Comments
HW 2nd ed. A:257
2. Reconstruction
2.1Reconstruction ‘to set free’
Although the verbal stem class of Lyd. rawa- is not assured, it is here equated to the one of HLuw. /arawa-(ti)/ and Hitt. arawaḫḫ-(ḫi) ‘to set free’ and PAnat. *arou̯eh2- ‘to set free’ can be reconstructed (for the gathering of the three cognates, see Sasseville 2021a:65). The derivational base posited by Sasseville 2015a:285 n. 9 is not the adjective in *-u̯o-, but the abstract noun *arou̯eh2- (Lyc. arawa-) according to the principle of conversion. This implies that both the adjective in *-u̯o- and the abstract noun in *-u̯eh2- were present in Proto-Anatolian, for which there is no counterevidence.
[D.S.]
For the lemma head see PAnat. *ár-eh2-.

