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

10 LÁSZLÓ DEZSŐ with a consonant cluster after the first vowel. However it is not present, or is exceptional in a number of proto-languages; if present, its frequency is much lower than that of CVC(V). Further, it is subject to structural constraints and it may possibly derive from CVC(V) structures. The roots with initial consonant cluster only exist in Indo-European. Benveniste's scheme can present the structural complexity of roots if arranged in structural hierarchy: (1) CéC, (2) CéCC, (3) CCéC which can also be interpreted from the diachronic point of view: (1) is primary, where also (2) and (3) are present, as in Indo-European; their diachronic sequence is to be clarified. 2. Consonants in Etymons 2.1. Some Statistical Data of Etymons Décsy gives the statistical data of Indo-European etymons based on Pokorny and other sources (Décsy 1991: 35) and those of Uralic etymons of UEW (Décsy 1991: 25) in detail. I shall sum up his data, adding some comments. The Indo-European labial stops: p, ph, b, bh will be indicated as p/b; the dentals: t, th, d, dh as t/d; the velars kj, kjh, gj, gjh; k, kh, g, gh and kw, kwh, gw, gwk as k/g (Décsy uses/ instead of diacritics). The glottalic theory proposes a different system of stops (cf. Lehmann 1993: 98) which has no impact on the figures. The data of total number of Indo-European stops and sibilants: p/b (623), t/d (776), p/b and t/d (1399); k/g (1379 of which 310 are palatalized); s (676). The data of the total number of Proto-Uralic etymons: stops/? (109), / (77), к (179); palatal and dental stops or affricates (tj/c) (51), interdental fricatives dh, dhj or 8, 8' (20); affricate с (24), sibilants s, s (88). (I have also used Décsy's characters: tj, dh, dhj). The summarized figures of dentals /, tj, dh, dhj (148) and/? (109) will be 257 and will exceed the number of velars (179). In both proto-languages the number of velars (k/g and k) is very high being close to that of p/b and t/d in Indo-European (1399: 1379). In Uralic the number of/ is very low, with other dentals it will be higher (148); however, if p is also added (257), their number exceeds that of к (257: 179). Presenting dentals, I have accepted Décsy's interpretation of с (tj) as the palatal variant of t. Counting Indo-European s velars, I did not consider palatal к in comparing labials and dentals with velars. If in Uralic we count only p + t + 5 (206) as opposed to к (179) then the difference will be less. If in Indo-European we do not count k/g then p/b and t/d will remain 1399, but k/g will decrease to 1069. The relation-