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 15 reliable basis. In Ci position the three stops are found in 45.1% of ail occurrences (k - 22.5%, p - 15.6%, / - 7.0%); the initial position is dominated by k, no other consonant has even a similar frequency. The use of ail résonants is relatively limited 34%. In C2 position the frequency of stops is low: 11% (k -5.7%, p - 2.2%), / - 3.8%o), the use of sibilants and affricates changes but little: 19.4%>. The frequency of résonants is much higher in C2 than in initial position: 63.8%). Thus in Ci position the ratio of obstruents and résonants is 2:1 in favor of the former, in C2 position the ratio is the same but in favor of the résonants. The change is due to the fréquent use of stops in Ci and to their low frequency in C2, because the share of sibilants and affricates remains nearly the same in both positions. The phonotactic rules of Uralic and Finno-Ugric proto-languages are treated in detail in Bakró-Nagy (1992: 133-58). Gamkrelidze and Ivanov (1984: 19-21, 139-143) establish rules concerning the use of stops of the same root, in Ci and C2 positions: (1) in a root there can only be stops with différent articulation, i.e. a stop with the same features cannot figure in both Ci and C2 position, (2) in roots there are only combinations of stops homogeneous concerning the feature voiceless-voiced. The latter rule is inhérent in the System of Uralic stops which is lacking voiced stops. The former rule is also valid (cf. Bakró-Nagy 1992/3: 4-6). Gamkrelidze and Ivanov state further rules (1984: 145-147) concerning the Ci and C2 positions grouping stops according to the place of articulation: „anterior", „medial": G [-anterior, -medial], D [-anterior, +medial], B [+anterior, -medial], i.e. into posterior, medial, anterior groups. The relationship between Ci and C2 can logically be decessive: from front to back, or accessive: from back to front, i.e. backwards or forewards. Both directions are possible with one modification: if there is a medial stop (D), the combinations B-D and D-G, i.e. the backward movements are usual, but the opposite: G-D and D-B are very rare. I have checked the Uralic and Finno-Ugric etymons of UEW and have found the following possibilities of stop séquences. As in Indo-European both decessive (backward) and accessive (foreward) movements are possible: examples of decessive séquence: Ci - p, C2 = /: pata (358); Ci =/?, C2 = k: päk3 (361) with equal frequency (4 examples); accessive séquence: Ci =k,Cj = t: kat3 (131), Ci = k, C2 —p: kepä (146-7), the former is more fréquent, but the numbers are small (5:3). The medial position has no role in the séquences: B-D (p - t), G-D (k - t) are equally possible. When / occupies the Ci position, it can be followed by k as C2: taka (506-7) three times, but not by p; i.e. backward movement is possible, forewards is not. In other terms D-G is admitted, D-B is not; thus, of the two séquence rules rarely occurring in Indo-European, one is prohibited, actually, the most infrequent one. The more complex structures with three consonants can be related to CVC structures in both proto-languages. As we have seen, in CVC structures any kind of consonant can occur in both positions in Indo-European, if C = obstruent, H =