Nyelvtudományi Közlemények 96. kötet (1998-1999)

Tanulmányok - Dezső László: Typological Comparison of Root Structuring in Uralic and Early Indo-European. [Az uráli és korai indoeurópai tőstruktúrák tipológiai összevetése] 3

Typological Comparison of Root Structuring in Uralic and Early Indo-European 7 *kunce ~ кисе 'star' (UEW 210-1); PTung *kucV~ *kocV(< *kutV~ *kotV) (Sinor 1990: XXXIV) *kince (künce) (?kice (küce)) 'nail' (FU ? U) (UEW 157); PTung *kVcV (< *kVtV) ~ PMong *kVtV~ *kVdV(Sinor 1990: XXXIV). The Altaic proto-languages only have the form without nasal, which makes it probable that the variant with nasal is secondary in Uralic. In the last two Altaic forms с derives from t; such a phonetic change is plausible also in Uralic and could be typical for Pre-Uralic. Instead of UEW I have quoted Sinor's excellent article, because it shows the possibilities of parallels in the study Uralic-Altaic not only for phonology, but also for semantics, giving us insights „into the distant past of Central Eurasia." (Sinor 1990: XXXIV). We can add further data without Altaic parallels: wenÓ3 ~ wec3 'whole, all' (UEW 568-9) wac3 ~ wanca 'root' (UEW 548-9) The variation -rjs-: s is also possible: pirjse-me :pise-me 'lip' (UEW 383—4) This group of clusters differs from the earlier ones. The data of Altaic protolanguages suggest that, in PU and PFU, an earlier root with final, affricate or stop was later extended with an inserted nasal to a homorganic nasal and affricate (stop) cluster. However, there is another explanation: the root had a final nasal followed by an affricate or stop which was lost in Altaic. The examples in the preceding groups (a) worjke/a and (b) sar/кз seem to support the second alternative. At any rate, this type of clustering requires particular treatment (for Proto-Dravidian cf. Krishnamurti 1972: 237 ff). These proto-languages attest roots of CVCV, CVC and CVCCV structures with short vowels, however long vowels can also occur in CV:C roots which will require explanation. If we consider Décsy's generalization of CV chain as a possible prototype, then the prototypical structuring in roots, bases and inflected words will manifest a sequence of CV syllables which was dominant in PU according to Décsy (1990: 34-5). However, in the course of change in Proto-Uralic and in the other proto-languages of the area, the chain of CV syllables was modified. In the languages of the area examined there was an important restriction to the root structure: the root could not have consonant clusters in initial position. Such clusters may arise from the shift of accent to a further vowel and from the loss of the unaccented syllable, but this did not take place because the accent was fixed on the first syllable in PU. The CVC structure with accent is also hypothesized for early Indo-European, but later the accent changed position. The vowels could have normal and reduced grade or could be dropped. In Altaic and Dravidian the vowels could also be long. Long vowels may originate with the combination of a laryngeal with a short vowel in Dravidian (Krishnamurti 1997). A similar process could also take place in the change from Uralic