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 5 Nagy's book (1992) which also examines the question of syllabic structure (ib. 133-158). My lists of Uralic and Finno-Ugric roots of UEW compiled for différent research objectives cover 90% of ail etymons and contain variants which cannot be revealed by Statistical data. The primary and secondary forms of Uralic etymons show variations relevant to our topic. In the first kind of variation there is a CVCV root which can be extended with a CV élément in différent ways: (C,)VC2

V ~ (COVC2V-C3V:

ips ~ ips-se 'taste'(UEW 83-84) dVC2

V-(C

3

V):

kine(-l3) 'tears' (UEW 159), saj3(-k3) 'ski; to ski' (UEW 429) The structures dVC2V-C

3

V and CiVC

2

C

3

V or those of dVC

2

and C,VC

2

C

3

V can alternate: piwtä (piwe-tä) 'to follow the spoor of a wild animal' (UEW 387) sah- (sälk3-) 'to get in or on smth' (UEW 434-5) par3 (parwa) 'heap, group' (UEW 356-7) I add an example in which a C\V root is extended to C]V-C2V: nu-m3 'upper part; heaven; god' (UEW 308-9) The alternation säte- (sälk3-) shows uncertainty in reconstruction of the etymon. These examples show that the formation of CVCCV structures from CVCV was still an active process in the period which can be reached by lexical reconstruction. Some variants, however, can reflect only the uncertainty of etymologies. Décsy (1990: 35) examined the list of etymons with consonant Clusters and found that: „A certain number of thèse words came intő being from an earlier trisyllabic séquence (CV+CCV)+CV so that the vowel of the second syllable was eliminated. The original third syllable may hâve been a derivative suffix. In any case, the syllable type CVC is a fascinating non-Uralic formula ... among the 129 words with CVC we find a disproportionally high number of uncertain etymologies and apparent affect words". Thus, the problem of medial consonant Clusters requires further study. For the proto-languages of Central Eurasia, the consonant Clusters were not typical. Nikolajeva (1988: 124-7) compared the parallels in Uralic and Yukagir proto-languages and established, as a rule, that in PYuk „the PU consonant Clusters hâve not been preserved" and PYuk had CVC where PU had etymons with consonant Clusters CVCCV, except nasals with homorganic obstruents and two cases with k + s and k + c (U nukse and pokcd). Baskakov (1979: 138-87) studied root formation in Turkic, comparing it with Mongol. According to the reconstruction based on the dictionary of Ancient Turkic, the Statistical data of Proto-Turkic roots show the dominance of CVC structures: 91.4% (with possible loss of first consonant) and only 5.6% of etymons had the structure CVCC (or (C)VCC), where the last consonants can be derived from suffixes of -VC structure. In Proto-Dravidian the typical root pattern was CVC with short or long vowel