ON THE QUESTION OF THE CAUSES AND MECHANISM OF THE TRANSFORMATION OF ky, gy, xy TO k’i, g’i, x’i IN OLD RUSSIAN

The article discusses the interaction of phonological and phonetic factors within the framework of the change traditionally described in historical phonetics as the transition of ky, gy, xy to k’i, g’i, x’i, which occurred in Old Russian dialects in the 12th–13th centuries. The explanation of its causes and mechanism is so far an unsolved problem of East Slavic historical dialectology. The diffi culty lies in the fact that within the limits of one phonetic change, two diametrically opposed processes were combined: palatalization of the velar and fronting of the next non-front vowel [y] to [i]. Having subjected the existing phonetic and phonological explanations of the change to critical consideration, the author puts forward a hypothesis according to which the transformation ky, gy, xy > k’i, g’i, x’i consisted of three diff erent phonological changes.
The key was the fi rst one — replacing /y/ → /i/ after the velars: /ky/, /gy/, /xy/ > /ki/, /gi/, /xi/. It was refl ected in written records of the 12th–13th centuries. This phonological change within the framework of the drag chain shift — Кy → Кi → Ći — motivated by the results of the second palatalization of velars and its mechanism was a kind of fi lling gaps in syntagmatic system (k#, g#, x#) that arose after *ki, *gi, *xi > ći, źi, śi. The South dialects of Old Russian that implemented the second palatalization but did not develop a hardness/softness contrast before /i/, /e/ were the nuclear zone of the change ky, gy, xy > ki, gi, xi. In the Old Novgorod dialect, where the second palatalization did not occur and the combinations ki, gi, xi were preserved, the change ky, gy, xy > ki, gi, xi was delayed.
The second change — the accommodation of the velar to /i/ in the new combinations ki, gi, xi (< ky, gy, xy) — was allophonic (palatalized allophones [k’, g’, x’] appeared before /i/) and therefore could not be refl ected in written records of the time.
The third and fi nal change was the latent phonologization of [k’, g’, x’], as a result of which independent phonemes /k’, g’, x’/ appeared, which were included in the correlation of hardness/softness. This paradigmatic change was also not refl ected in written language.