The Formation of the East Baltic stative verbs *stāw-ē- and *dēw-ē-

THE FORMATION OF THE EAST BALTIC STATIVE VERBS *stâw-ëAND *dëw-ëAlfred Bammesberger University ofFreiburg im Breisgau The East Baltic Stative verbs *stâw-ë- and *dêw-ë- do not derive from alleged Indo-European perfect forms in -w-. Their starting point must be seen in the otherwise well-known roots IE *stä- and *dhë-. The -w- is ultimately due to the active perfect participle, whose strong form IE *-wbs was abandoned in Baltic (and Slavic), but left its trace in a phonologic rule according to which vowel-initial suffixes were linked to roots ending in a long vowel by means of -w-.1 The etymological connection of Lithuanian stovëti2 'stand', Latvian stâvêt { sto-, would be reduced to *stä-. If intervocalic -j- was lost at some stage in the history of Baltic (actually occurring sequences VjV may be secondary), the lsg. would also be aberrant, since from *stajö we would expect a form ending in a diphthong, *stau. Furthermore, the non-present stem *sta-ëwould probably contract (to *stä > *stö- or *stë- ?) Thus we may set up the following Lithuanian paradigm for *sta-ë- and contrast it with the forms of *min-ë- : lsg. *staumin-iù 2sg. *staimin-i 3sg. *staimin-i lpl. *staimemin-ime 2pl. *staitemin-ite inf. *stoti or *stëti minëti The paradigm of *stoti or *stëti was obviously too abnormal to be maintained; in particular, forms like *stau, stai etc. could not be perceived as belonging to a stative verb. On the face of it, OCS stojati8 definitely gives the impression of being the expected stative from *stä-, and can thus be considered more archaic than East Baltic *stäw-e-,7 whose origin will be examined below. In the Indo-Europeanist literature, *stäw-e- and *dëw-ë- are interpreted as continuing an archaic root shape, and play an important role in the discussion of certain perfect forms which now will be briefly taken up. Ever since Mahlow (1879: 144), a varying number ofBaltic and Slavic present formations have been considered as deriving from the IE perfect. Li. stovëti is given among Mahlow's examples. Van Wijk (134) re-examined the question and added dèvëti to the list. Both *stâw-ë- and *dëw-ë- are said to be based on perfect stems, namely *stä-w- and *dhë-w-, to the roots *stä- and *dhë- ; and the antiquity of the root shapes *stä-w- and *dhë-win these perfect formations is allegedly proved by the Sanskrit perfects tasthau and dadhau. Characteristic of this line of thinking is the statement by Fraenkel (1965:915-16): 'Lit. stovëti, lett. stävet sind auf einem im Ai. durch tasthau "ich habe gestanden" vertretenen Perfekt aufgebaut'; Fraenkel takes a similar position elsewhere (1962:92) with regard to dèvëti. With minor variations, *stâw-ë- and *dëw-ë- are projected back to IE root shapes *stä-w- and *dhë-w- in all the relevant literature that I have seen. Thus Watkins (1969:224) writes that *stâw-ë- is based on an old perfect which was characterized by a w-extension; similarly (p. 92), he derives *dëw-ë- from IE *dhë-w- and invokes Skt. dadhau. (Comparable statements 6 Vaillant (1966 : 77) mentions stojatiin the series ofë-verbs whichcontinuethe original o-grade of the perfect, but adds that the root vocalism is equivocal. Vaillant's assumption (1950:106-7) that a was lost in Slavic and Baltic even in first syllables is unlikely in view of Li. statyti 'build', which indicates a /o-participle IE *st3-tó- > Li. *stätas. Therefore OCS stojati may well reflect the zero-grade IE *sta-. But full-grade *sta- is also thinkable (see Cowgill 1973, fn. 27). 7 Specht (41) postulated Slav. *stavëti corresponding to Bait. *stâw-ë-, and saw the proof for *stavêti in the causative staviti, whose 'ti allein durch *stavêti verständlich wird'. It will be shown below that staviti can be satisfactorily explained on the basis of the root *stä-, and does not require the existence ofan ë-verb 'stavëti. Furthermore, it would be very hard to understand what purpose *stavêti could have served if stojati was available; on...