No CrossRef data available.
Published online by Cambridge University Press: 22 April 2026
[OP hya hyā tya, functioning as relative and as article, rarely as personal pronoun, is a contamination of pIE demonstrative *so sā tod and relative *ḽos ḽā ḽod; it is a late formation in Iranian dialectal times, which is why the anteconsonantal t is not spirantized; it is not genetically the same as the Vedic demonstrative sya-s syā tyad, with which it is phonetically identical. But the category of the definite article was not well developed in OP, as examples show. OP hyāparam ‘afterward’ is abl. hyād + aparam ‘later’; there is no evidence that -d survived final after a long vowel, which would have prevented crasis. OP patiy is not only prefix, preposition, and postposition, but also adverb, as is shown by the use of hyāparam with and without a preceding patiy (as well as by other examples).]
1 The pIE relative *ḽos ḽā ḽod survives in OP as a component of hya hyā tya, but otherwise only in nom.-acc. yaciy ‘whatever’ (DNb 35, 36, 57) and in yātā conj. ‘until’, prep. ‘unto’; yaθā ‘as, as well as, when, after, that, so that, because’; yadā- ‘where’; yadiy ‘if, when, whenever’; yaniy ‘where’; yāvā ‘as long as’. I normalize yaniy XV 22 rather than yanaiy because I take the word to be the loc. sg. (pAr.) *yasmi, Skt. yasmi-n, Av. yahmy-a ‘where’, which gives OP *yahmi remade to yaniy with the -n- of instr. *yanā; cf. inst. tyanā DB 1.23, DSf 42, avanā XPh 49, 51–2. Similarly the pronominal abl. in (pAr.) *-asmād gave OP ahmād and then -anā in (abl.) avanā DSe 38, DSf 31, to ava- ‘that’.