Correlation between consonantal duration and the quantity and presence of the following vowel in the Finnic varieties of Ingria (Lower Luga area) (Зависимость длительности согласного от долготы и наличия последующего гласного в прибалтийско-финских идиомах Ингерманландии района Нижней Луги) (original) (raw)

2022, Ануфриев А. А. и др. (ред.). Проблемы языка: Сборник научных статей по материалам Девятой конференции-школы «Проблемы языка: взгляд молодых ученых» (28–29 апреля 2022 г.). М.: Институт языкознания РАН, 5-26. doi: 10.37892/978-5-6049527-1-9-1

https://doi.org/10.37892/978-5-6049527-1-9-1

The paper studies a correlation between the duration of a consonant and the length or presence of the following vowel in the Finnic varieties of Ingria of the Lower Luga region: dialects of the Finnish, Ingrian and Votic languages, as well as in a mixed Siberian Ingrian/Finnish variety which also originates from the same region. Our first hypothesis (a) about a compensatory (inversely proportional) lengthening of preceding consonants which accompanies the final vowel loss was confirmed on the basis of the three idiolects with a high percentage of final vowel loss (two Southern Lower Luga Ingrian and one Siberian speaker). These idiolects can be considered to be at an intermediate stage of the final vowel reduction between the Finnish language, which has no reduction, and the Estonian language, where the short final vowel has been completely lost from the disyllabic words of the structures discussed in the paper (CVVCV and CVCxCyV). In our data, C2 directly preceding the reduced V2 (i.e. the short intervocalic consonant in the CVVCV structure or the second consonant of an intervocalic cluster in a CVCxCyV structure) is lengthened if the final vowel is lost. In Estonian, either the long vowel of the first syllable (in the *CVVCV > CVV:C structure) or the first consonant in the cluster (in the *CVCxCyV > *CVCx:Cy structure) had been lengthened. On the other hand, C2 lengthening in our data is often phonetically manifested as a slight post-aspiration of this consonant rather than through a prolongation of its main articulation. Durations of C2 and C2x were the longest in the Siberian data, which represent the most advanced stage of the reduction and loss of V2 among the studied varieties. Our second hypothesis (b) about an anticompensatory (directly proportional) lengthening of the word-initial consonant (C1) before a long vowel (V1), as compared to the position before the short V1, was also generally confirmed. However, a statistically significant increase in the duration of C1 before a long V1 was found only for the Votic speaker, while for the rest of speakers, it was observed only as a weak insignificant trend. The compared structures CVVCV and CVCxCy, however, differed not only in the length of V1, but also in the length of C2 (single consonant vs. cluster), so their comparison was not perfect to study this effect. In general, the type of interaction between the duration of some segments and the length or presence of other segments in the studied Finnic varieties seems to depend, at least partially, on the stage of the final vowel reduction and loss. This type of interaction may also be different for the first and second syllable. The correlation between C1 duration and V1 length does not depend on the degree of V2 reduction, while the correlation between C2 duration and the presence of V2 apparently does.