The number of people still reached almost a thousand, most of them living in the village of Zasečiai, with fewer in Pagiriai, Narčiai, Strela, Teteikiai, and elsewhere. It is not known how much more material worthy of publication Polish linguists will find in J. Rozwadowski’s archival papers, but the publication “The Lithuanian Dialect of the Zietela Area in the Navahrudak Region” will remain one of J. Rozwadowski’s most valuable works on Lithuanian, enriching Lithuanian dialectology. For this, we offer our deepest gratitude to all the Polish linguists who contributed to preparing this work for publication.

Aloyz as Vidugiris

SIGNIFICANT WORKS FOR THE CENTENARY OF AN EMINENT LINGUIST

Twentieth-century European linguistics is inextricably associated with the name of the Polish linguist Jerzy Kuryłowicz. Drawing on the methods of structural analysis, this exceptionally broad-minded scholar opened up new perspectives in comparative and general linguistics, and put his innovative ideas into practice in numerous monographs and other works, whose subject matter encompasses not only Indo-European languages but also those of other families (especially Semitic). The enduring vitality of his ideas and traditions is clearly demonstrated by two major collections of articles compiled by Wojciech Smoczyński and published on the occasion of the hundredth anniversary of Kuryłowicz’s birth. One of these collections, devoted to Baltic and Slavic studies, appeared as the fourth volume of the international journal “Linguistica Baltica,” well known to our linguists (Kuryłowicz Memorial Volume Part Two, Kraków, 1995, 333 p.). The other, even larger collection is devoted to general and methodological issues in linguistics and the development of various languages (mainly Indo-European) (Kuryłowicz Memorial Volume Part One. Analecta Indoeuropaea Cracoviensia Vol. II, Kraków,1995, 593 p.). We would like to briefly discuss that collection here. The works published in it show how Kuryłowicz’s ideas continue to be developed and are reflected in current research on the development of various languages.

Kuryłowicz’s works are particularly important to Lithuanian historical linguistics for their methodological and theoretical principles. Of the many articles published in the collection, I would like to draw attention here to those that examine issues which might prompt our linguists to reflect on and investigate more closely the development of corresponding phenomena in Lithuanian and other Baltic languages.

The eminent linguist’s biography, scholarly works, and their methodological postulates are discussed in the collection’s introductory article by its compiler and editor, Smoczyński (“Jerzy Kuryłowicz 1895–1978. On the hundredth anniversary of his birth”, XI–XXIV). Applying a functional approach to language in comparative Indo-European grammar, Kuryłowicz studied chiefly not individual linguistic elements, but their relations to one another and the genealogy of their systems; he defined the relations among these linguistic units in terms of their oppositions, cases of neutralization, and hierarchy.

248

By investigating changes in oppositions and the direction of their development, one can establish the relative chronology of linguistic changes. It is worth noting that the methodology of internal reconstruction defined and tested by Kuryłowicz was also employed in the works of Lithuanian linguists Jonas Kazlauskas, Vytautas Mažiulis, and others; it remains in successful use today.

Axel Holvoet’s article (“Jerzy Kuryłowicz on verbal aspect”, 19–24) should be of interest to researchers of tense and aspect categories in Lithuanian and other Baltic languages. It further develops ideas about an undifferentiated aspect/tense category reconstructed for Proto-Indo-European, within which the imperfective present was opposed to the perfective past (preterite). In Slavic languages, aspect is directly independent of tense, while the imperfect became a past tense; Kuryłowicz therefore regarded this state of affairs as a late stage in the development of the tense/aspect category. Comparing data from Slavic, Romance, and Greek, Holvoet concluded that the different aspect stems of Slavic verbs could have arisen independently of tense forms marked for aspect. Thus, the Slavic imperfective (imperfective-aspect) forms may have been closer to the old imperfect than previously thought. Holvoet also emphasizes that further study of the aspect category must carefully distinguish two types of aspect opposition: one directly associated with the expression of tense and one independent of it.

Leszek Bednarczuk’s article (“Typologie et isomorphisme du système indo-européen”, 39–48) attempts to survey the principal features of the grammatical system of Proto-Indo-European from the perspective of modern linguistic typology. Among these, the author identifies the major role of postpositional derivation, syncretism of inflectional endings, more highly developed nominal categories and less developed verbal ones. Relations between words were expressed by cases and postpositional adpositions (i.e., postpositions), as well as by postpositional conjunctions and numerous particles. Word order varied, but the SOP (or SOV) pattern predominated. The author argues that the ergative or active clause structure hypothesized by some linguists left no clear traces in the oldest Indo-European-language texts. The use of the accusative as the case of the direct object is considered an argument against reconstructing an ergative clause structure for the period when the Indo-Europeans had an inflectional grammatical system. The article suggests that agglutinative and isolating-type elements found in some Indo-European languages may have arisen through the influence of languages belonging to other families and types. External influence from other language types is also invoked to explain the narrowing of the verbal tense system, the auditive (or indirect mood) category, ergative clause construction, and verb-initial position in some languages. Nevertheless, it should be borne in mind that in ancient Indo-European languages we observe more agglutinative features, and the functions of relational morphemes are more differentiated than in later developmental periods, when morpheme fusion and syncretism grew stronger. This is also clearly evident in the history of Lithuanian inflection, especially declension. Thus, at least some agglutinative phenomena (for example, the greater semantic independence of relational morphemes, especially suffixes) may have been-

249

… inherited; there is no need to explain them through contact with other language types.

The development of Baltic infinitive and supine (purpose) constructions helps clarify the results of the study of Vedic data presented in José Luis Garcia-Ramón’s article (“Zur Konkurrenz von Dativ und Akkusativ von Nomina actionis und Abstrakta im Indogermanischen”, 101–114). They show that, from the earliest times, the accusative of verbal abstracts had the meaning of direction (Ziel), while the dative expressed purpose (Zweck). In contextually conditioned cases with verbs of motion (Vedic ā-yā ‘go’, vah ‘ride’, etc.), however, the distinction between these meanings was neutralized, and forms of both cases could express purpose. The accusative was more often used with unprefixed and intransitive verbs, the dative with prefixed and transitive verbs. Baltic infinitives, too, developed from the dative expressing purpose, while the supine developed from the accusative; however, the supine is generally found only with verbs of motion. Nevertheless, instances of supine use with verbs meaning ‘want’ and ‘be able’ in Lithuanian and Latvian dialects (cf. the author, LKK XXXIII, 100) suggest that formerly, in Baltic as well, datives and accusatives of verbal abstracts—the precursors of infinitives and the supine—were used in a manner similar to that in Old Indic during the Vedic period.

The use of Sanskrit infinitives in -tum, which derive from accusatives of verbal abstracts, is also discussed in Jost Gippert’s article (“Zur Syntax des Infinitivs auf -tum im Altindischen”, 255–277). The abundant data, examined in detail, show that in impersonal constructions with the neuter adjective form śākyam ‘possible’, these infinitives were associated with both the accusative and the nominative, much as in the Lithuanian construction galima pjauti rugius / rugiai ‘it is possible to reap rye / rye can be reaped’, and so on. The author’s conclusion is therefore worth noting: such impersonal Sanskrit constructions are of recent origin, having developed from personal ones. This is useful to bear in mind when investigating the origin of the so-called object nominative in Lithuanian and Latvian.

Balticists investigating the development of the verbal tense system should take note of Paolo di Giovine’s article on the preterite with long root vocalism in the western Indo-European languages (“Sul preterito a vocalismo radicale lungo nelle lingue indoeuropee ‘occidentali’”, 115–129). As is known, Baltic past-tense forms with a long vowel have so far been interpreted in different ways: some scholars, along with Nicolas van Wijk, Christian Stang, and Jonas Kazlauskas, regard them as ancient and inherited, while others agree with Kuryłowicz, who considered them more recent, having arisen within the Baltic languages. The article’s author, who published a substantial book on the Indo-European perfect in 1990 (Studio sul perfetto indoeuropeo. Parte I, Roma: „La Sapienza“), presents new arguments for the view that preterite forms with long root vocalism in Latin, Celtic, and Germanic are not reflexes of the ancient perfect, but innovations that arose independently in each language group as the role of reduplication diminished and the perfect itself disappeared. The origin of the Indo-European perfect in a periphrastic construction with a verbal noun and a copula is also discussed by Witold

250

Stefański’s article (“L’origine du parfait en indo-européen”, 131–135). In 1976, an intriguing hypothesis by Eriko P. Hampo concerning possible traces of the augment characteristic of the Indo-Iranian, Greek, and Armenian perfect in the past-tense forms of prefixed Lithuanian verbs was published in the journal “Baltistica” XII (1), 25–29: those traces are said to be shown by the retraction of stress onto the prefix from an erstwhile stressed augment that had stood before the root and later disappeared (cf. paduoda – padavė, etc.). In the article published in this collection, the author similarly reconstructs an augment (PIE e-) in forms of the Albanian verb dha ‘give’. Comparing them with the corresponding Baltic verb forms, he considers this possibly a common feature of Albanian, Baltic, and Slavic (“Albanian dha ‘give’”, 395–398).

Important questions concerning the formation of the verbal system in the Indo-European languages are also considered in Reinhard Stempel’s article on the relationship between the stative, perfect, and middle (“Stativ, Perfekt und Medium: Eine vergleichende Analyse für das Indogermanische und Semitische”, 518–528). The opposition between state and action is considered primary and the oldest. It is argued that stative verbs were once undifferentiated with respect to tense, mood, and voice, and formed the basis of the Indo-European perfect. The category of voice and the active–middle opposition are attributed only to forms of verbs with an action meaning. The author discusses cases in which middle forms have taken on stative and resultative meanings from original stative verbs, and points to parallels in Semitic languages.

Lithuanian cardinal numerals septyni ir devyni ‘seven’ and ‘nine’ have long been derived from reconstructed PIE forms *septṇ/septṃ and *neun. The corresponding Germanic and Anatolian forms are examined in Kenneth Shields Jr.’s article (“On the Indo-European origin of the Germanic cardinals ‘7’ and ‘9’”, 505–512). It is argued that the Indo-European numeral for ‘7’ had the root *sep-, to which the “non-singular” elements *-t and *-n were added. This produced the doublets *sep-t and *sepṇ > *sepṃ, and their contamination gave rise to *septṃ, whose reflex in Latin is septem. However, even before *sepṇ changed into *sepṃ, a second “non-singular” element -t had been added to it in the Germanic and Anatolian dialects, thereby producing the proto-form reconstructed for Germanic *sepṇ t. Similarly, the “non-singular” element *-n had been added to the root *new-, and in Germanic and Anatolian languages also *-t. The author considers the original forms *sepṇt and *newṇt to have once existed in at least two dialects—Germanic and Anatolian. According to this interpretation, the Germanic numerals for ‘7’ and ‘9’ should be derived from proto-forms that differ somewhat from those of the corresponding Baltic numerals.

The articles mentioned here, which are more closely related to problems in Lithuanian historical linguistics, make up only a small part of the works published in the book. It contains 61 articles, of which 3 are devoted to Kuryłowicz’s life and work, 8 to general and methodological questions, 9 to problems in Indo-European studies, 4 to Hittite, 2 to Tocharian, 7 to Indo-Iranian, 2 to Armenian, 7 to Greek, 1 to Albanian, 3 to Latin, 5 to Celtic, 6 to Germanic, 3 to Semitic, and 1 to Turkic languages. The collection includes Woicie-

251

…cho Smoczyński’s bibliography of Kuryłowicz’s works (XXXV–XLVII), and, at the end of the book, indexes of words and names. This excellently prepared and very rich book is in no way inferior to the best Western European publications of its kind. Together with the second volume dedicated to Kuryłowicz, mentioned above (Linguistica Baltica 4), and the memorial collection for Jan Safarewicz prepared by the same Wojciech Smoczyński (Analecta Indoeuropea Cracoviensia Vol. 1, Cracoviae, 1995), this may serve as an example of how we, too, should honor the memory of Lithuania’s most distinguished linguists.

Vytautas Ambrazas

A GUIDE TO LITHUANIAN LANGUAGE STUDIES

T E R J E M A T H I A S S E N. A SHORT GRAMMAR OF LITHUANIAN, Columbus, Ohio, 1996: Slavica Publishers, INC, 256 p.

Over the past decade, we have seen a resurgence of interest in Lithuanian language studies in many European countries. Lithuanian language courses are offered at universities not only in Latvia, Germany, Poland, and Russia, but also in Italy, Sweden, Norway, Finland, the Czech Republic, and other countries; important Lithuanian studies research is conducted at many of them, and new periodicals in Baltic linguistics are being launched¹. Yet until now there has been a serious shortage of English-language Lithuanian grammar guides suitable for university studies. This gap is being addressed by the recently published *A Short Grammar of Lithuanian* by Terje Mathiassen, a professor at the University of Oslo.

The author of the grammar is a Balticist and Slavist well known to Lithuanian linguists, continuing in Norway the tradition of Lithuanian studies established by his teacher, Christian Stang. He speaks Lithuanian fluently and has published many important works on Lithuanian linguistics. Having taught Lithuanian at the University of Oslo for a long time, he has accumulated extensive experience in the field, which he has drawn on in his grammar.

The grammar consists of a short introduction and 14 chapters. The introduction provides some information about the place of Lithuanian among other languages, its main dialects, and its present state. The first chapter (21–36) describes the phonological system of Standard Lithuanian, its realization in spoken and written language, as well as vowel and consonant alternations, stress, and pitch accent. Chapters 2–7 (37–177) discuss morphological topics according to word classes (the so-called parts of speech), their morphological categories, and inflection. In view of the book’s users’ practical needs, the description of forms and their functions is not rigidly separated, one from the other; the boo-

¹ One of the most recent is Res Balticae. Miscellanea italiana di studi baltistici. 1, 1995; 2, 1996; 3, 1997, Pisa.

252