Nyelvtudományi Közlemények 91. kötet (1990)

Tanulmányok - Larsson, Lars-Gunnar: The Origin of Mordvin M šaba, E žaba ’child’ and Cheremis šußo 147

THE ORIGIN OF MORDVIN M SÁBA, E ZABA 'CHILD' AND CHEREMIS SUßO 149 Proto-Slavic *suka, attested in Russian suka 'bitch'. Although Skr. säva'the young of any animal' apparently is not too well supported by other Indo-European words, it is worth being examined as the possible origin of mdE zaba, M saba and Cher. sußo 'child'. As to the semantic correspondence there are hardly any objections to raise against the alleged Indo-Iranian origin of these words in the Volgaic languages. A fluctuation between the meanings 'child' and 'the young of an animal' has parallels in other languages, e.g. in Swedish where unge has both these meanings. When it comes to the phonetic development of the Indo-Iranian word in the Volgaic languages, there are problems. The non-initial stop in the Mordvin forms seems dubious, evén if it might be possible to assume a variation of b and v. Certainly, one would hâve expected a v, cf. mdE rau, ravo 'the river Volga' < Indo-Iranian, e.g. Avesta ravan- 'river' (Joki 1973. 307). Furthermore, the représentation of an Indo-Iranian *$- by Mordvin s- is not usual, cf. instead Finnish suka 'Borste; Bürste, Striegel, Kamm'/Mordvin suva 'Granne, Achel, Spreu' < Indo-Iranian, cf. Skr. süka 'Granne des Getreides, Stachel eines Insekts' (Joki 1973. 315, cf. also Collinder 1960. 55). The initial voiced sonant of mdE zaba, however, is explained as a spontaneous variation between voiced and voiceless consonants, illustrated by Paasonen (1893. 11) by examples as mdE sul'nams, zol'nams 'rieseln'. In addition to this, it seems suspicious that Walde-Pokorny (1959) do not give any words from other Indo-Iranian languages that would correspond to Skr. säva-. Neither does Mayrhofer (1976) in his etymological Sanskrit dictionary, although he gives examples from Middle Indic and Modern Indic languages. One's doubts are increased by the theory of Kuiper (1948. 67, 136) that Skr. sava- is a loanword from Munda languages; Kuiper compares it with Santali debe 'dwarfish, small'. If this theory is correct, then Skr. sävawould be an Indián word in the sense that it has originated on the Indián subcontinent and would not have any old Iranian cognâtes. Skr. säva- 'the young of any animal', which, with regard to its meaning, seems to fit very well as the origin of mdE zaba, M säba and Cher, sußo 'child', consequently causes two problems. The first is the consonantism of the Mordvin words, which does not seem to be the regulär représentation in an Indo-Iranian loanword. The second problem is that it is not at all clear how this word would have passed over from languages in India to the Volgaic languages. To regard mdM saba, E zaba 'child' and Cher, sußo as an Indo-Iranian loanword causes problems. These problems can, however, be avoided if the word is regarded as a loan from the language of the Gypsies. In the différent sources on Romany, there are words that seem to fit as the origin of the word Nyelvtudományi Közlemények 91. 1990.