Abstract
The allative is one of the three distinct spatial cases that Hittite is known to have used in the course of its written history, alongside the locative and the ablative. The allative ending was -a, and with the i-stems, we find the ending (-i)-ya.
In some instances, it is known to have been used in dative-locative function. This feature is confirmed by the post-Old Hittite era, although the allative itself becomes moribund after the Old Hittite period.
The traditional hypothesis for this syntactic peculiarity is that it is motivated phonologically: the dative-locative of the i-stems was rather "unmarked", as the ending -i blended into the stem. This led speakers/writers to use the more overtly marked -iya even for the dative-locative, as -iya did not have any other grammatical function by that time.
Whether this hypothesis can be proved or disproved by the textual evidence is the purpose of this paper. An important argument in the discussion is the fact that there are indications that this feature actually appears already in Old Hittite.
The textual evidence shows that the case ending -a in its variant -iya in the i-and ai-stems is already used to mark location and beneficiary/recipient in OH/OS. It also refutes the standard view that the -iya ending spread in post-OH to disambiguate the multifunctional dative-locative singular ending -i, in those stem classes or elsewhere.