/adantiwal(i)-/, ara/i-ti-wa/i-li-

Both attestations are very hard to read due to abrasion and partly broken contexts, in addition to the signs being arranged in an unusual manner around a figure and on a cylindrical shape respectively. Moreover, the presence of the logogram markers in MARAŞ 5 suggests that the following word is to be seen as functioning as a noun rather than an adjective. The latter part of the word is clear, because the explicit spelling in both cases simplifies the interpretation considerably. Considering the beginning of the word, however, the forms are harder to reconcile. Clearly, rhotacism is in evidence with one form showing /l/, the other /r/. Rhotacism as such is expected here, as Rieken & Yakubovich 2010a:215-217 have shown that rhotacism is a wide-spread phenomenon in late HLuw. texts, i.e. the graphemes representing earlier /d, r, l/ alternate and prove that these sounds have merged into a flap [ɾ]. Rather more surprising is the presence of the logogram EDERE in MARAŞ 3, which in the verb is used to represent the root /ad-/ and would therefore suggest a reading /adlantiwalin/, i.e. a combination of the original /d/ with a merger /l/. Two possible explanations come to mind, one of which would be to assume a partly hypercorrect spelling. The other possible explanation sees the scribe aware of the obvious semantic connection between the verb /ad-/ ‘to eat’ and the adjective /adantiwal(i)-/ and using the logogram for that reason; the morphological connection would probably have been opaque to a contemporary speaker.

[AH.B.]