LIETUVIŲ KALBOTYROS KLAUSIMAI XL (1998)

Reviews. Review articles. Notes. Opinions. Reviews. Observations Überblicke. Rezensionen. Anmerkungen

Poljakov Oleg, The Problem of the Balto-Slavic Sprachgemeinschaft [Heidelberger Publikationen zur Slavistik, A. Linguistische Reihe, Band 8. Hrsg.: Baldur Panzer], Frankfurt am Main: Peter Lang, Europäischer Verlag der Wissenschaften, 1995, XXX 336 p.

The book by Oleg Poljakov, associate professor in the Department of Slavic Philology at Vilnius University, which is under review, is perhaps the most substantial study since Janis Endzelīns’s “Slavic-Baltic Studies”¹ devoted to one of the, one might say, perennial problems of historical linguistics—the problem of the earliest relations between the Baltic and Slavic languages. The book consists of a “Preface” (pp. XIII–XIX), three parts (pp. 1–302), “General Conclusions” (pp. 303–305), and a bibliography (pp. 307–334). In the “Preface,” drawing on theoretical principles, he characterizes the principal methods used to address the problem of the earliest relations between the Baltic and Slavic languages, describes his research method as "Systemanalyse, systematische Behandlung", emphasizing that this problem can supposedly be solved not by enumerating individual phenomena common to these languages, but by analyzing isoglosses at all linguistic levels and thus avoiding the arbitrariness of opposing and mutually contradictory hypotheses (p. XVII); he also sets out a rather modest aim, which is “aber nicht, endgültige Lösungen vieler Probleme vorzuschlagen, sondern eine neue Diskussion über das Problem der balto–slav. Sprachgemeinschaft und der Sprachverwandtschaft überhaupt anzuregen, die der modernen Entwicklung der Sprachwissenschaft entsprach” (p. XIX). As becomes clear later in reading the book, O. Poljakov pursued much more ambitious aims in this study. In the first part, “Introduction. The History of the Problem of Balto-Slavic Linguistic Community and Questions of Methodology” (pp. 1–58), he describes in detail the methods for investigating the earliest relations between the Baltic and Slavic languages, distinguishing Brugmann’s method from Meillet’s method and addressing the question of the origins of the Baltic and Slavic languages in general. He also reviews the state of research into the earliest relations between the Baltic and Slavic languages at individual linguistic levels, drawing on the most recent linguistic literature: lexical (following W. Mańczak, he regards vocabulary as an important factor in determining the degree of relationship between languages; he mentions shared derivational suffixes in these languages), phonetic-phonological (he surveys only matters of prosody and accentuation), morphological (he deals only with the verbal system), syntactic, and even “extralinguistic” (after reviewing research, mostly from a methodological perspective, into the Indo-European homeland, he—

¹ Эндзелин И. М. Славяно–балтийские этюды, Харьков, 1911.

243

—concludes that evidence from other disciplines, such as archaeology, can neither confirm nor refute the linguistic community of the Baltic and Slavic languages). It is striking that, in reviewing the state of the problem of the earliest relations between the Baltic and Slavic languages at different linguistic levels, he mentions, comments on, and criticizes only arguments directed against a Baltic-Slavic community, while leaving untouched the arguments of its proponents, contenting himself with the general remark that their arguments, too, are debatable (p. 58).

From a methodological point of view, O. Poljakov’s observation is noteworthy: the problem of the earliest relations between the Baltic and Slavic languages can be solved by listing and enumerating not individual linguistic phenomena or groups of phenomena, but by analyzing isoglosses at all linguistic levels—and, most importantly, these shared phenomena and elements cannot be projected onto the same chronological plane (p. 8).

The second part, “Phonetic-Phonological Problems of the Balto-Slavic Linguistic Community” (pp. 59–192), examines quite varied phenomena in the phonetic-phonological systems of the Baltic and Slavic languages that arose at different times, such as: the fate of the Indo-European syllabic sonants l, m, n, r; the merger of Indo-European velar and labiovelar dorsal consonants; lengthening of root vowels in the Baltic and Slavic languages (the so-called Winter’s law); the merger of Indo-European aspirated and unaspirated plosive consonants; palatalization of Indo-European palatal dorsal consonants; changes in consonant clusters; the change of Indo-European *s after i, u, r, k; reflexes of Indo-European nasal consonants at the end of a word; and word-final laws in the Slavic languages. Considerable space is also devoted here to matters of Baltic and Slavic accentology. After surveying in detail the intonational systems of Latvian and Old Prussian and of the Lithuanian Samogitian dialects, he arrives at the interesting conclusion that the original character of Balto-Slavic intonations (acute—the falling accent; circumflex—the rising accent) has been preserved in Standard Lithuanian and the Aukštaitian dialects.

The most controversial topics in this chapter are, of course, Winter’s law and the operation of de Saussure’s–Fortunatov’s law in the Slavic languages.

The third part, “Declension in the Baltic and Slavic Languages and Their Relations” (pp. 193–302), is devoted to the declensional systems of the Baltic and Slavic languages. In O. Poljakov’s view, these are more important than the conjugational systems for resolving the question of the earliest relations between the Baltic and Slavic languages, because declension is “static,” changing and having changed less. He adopts an eight-case system as the basis for his analysis (including the vocative), which was characteristic of a later stage in the development of the Indo-European proto-language.

The conclusion O. Poljakov draws from his analysis of the relations between Baltic and Slavic declension systems regarding the earliest relations between the Baltic and Slavic languages is unexpected: after the disintegration of the Indo-European proto-language, there was already a common source for Baltic and Slavic declension, which could supposedly be called a “proto-language” (“Ursprache”) or a “linguistic community” (“Sprachgemeinschaft”) (p. 302). But when O. Poljakov adds that this common source was of the same kind as that of Indo-Iranian, everything becomes clear: after the disintegration of the Indo-European proto-language, there existed a common Balto-Slavic language, from which later developed

244

the Baltic and Slavic languages. After all, the existence of the Indo-Iranian proto-language is an accepted fact in Indo-European studies.

There is no doubt that the publication of O. Poliakov’s study marks a new and important stage in research on the ancient relations between the Baltic and Slavic languages, quite apart from the fact that a book published abroad and written in German is a significant biographical milestone for O. Poliakov as a scholar. In this connection, it must be stressed that, by writing and publishing this book in German—the classical language of Indo-European studies—O. Poliakov has also served Lithuanian linguistics. Through O. Poliakov’s book, the work, interpretations, hypotheses, and ideas of Lithuanian linguists have come to the attention of Western scholars. It is very useful, indeed essential, for Lithuanian and Latvian linguists to study O. Poliakov’s book carefully. One immediately sees that it was written not only in Lithuania but also in the centers of Western scholarship and culture. The latest Indo-Europeanist literature is reviewed and cited, and the findings and hypotheses of the most recent research are drawn upon. For every question examined, the literature is reviewed in detail, explanations and hypotheses are presented, and nothing is left unsaid. But O. Poliakov also has his own point of view: he substantiates and argues for it, engaging in polemics with the views of other researchers. Thus, O. Poliakov’s book is a scholarly study. A positive feature of O. Poliakov’s scholarly method that should be highlighted is his examination of facts and phenomena of the Baltic and Slavic languages within the broad context of data from Indo-European languages: these facts and phenomena appear in a new light, either as archaisms or as innovations. Access to the latest research on Indo-European languages and its findings, together with the book’s author’s systematic approach, made it possible to reconsider in fundamental ways many claims of Baltic historical comparative linguistics, for example, the history of neuter nouns in the Baltic languages (pp. 197–209).

As already mentioned, O. Poliakov’s book also contains controversial points that are indeed worth debating, but, of course, not in the tone and style of W. Hock.²

O. Poliakov distinguishes two principal methods for investigating the earliest relations between the Baltic and Slavic languages and their character—the Brugmann method and the Meillet method (p. XIV). In O. Poliakov’s view, their essence is as follows: by enumerating common exclusive features and facts of the Baltic and Slavic languages (the so-called “eight points of Brugmann”), one asserts the existence of “eine balto-slavische Sprachgemeinschaft” (after the breakup of the Indo-European proto-language); by pointing to identical or similar correspondences and parallels in the histories of other Indo-European languages, one recognizes that the Baltic and Slavic languages developed in parallel and independently (the Meillet method). These two methods, these two points of view, were the reason for the best-known linguistic debate, in which many scholars took part. Some researchers, O. Poliakov goes on to say, recognize the existence of a “balto-slavische Sprachgemeinschaft,” while others deny it, “ausgehend paradoxerweise von den methodologischen Vorstellungen K. Brugmanns (pro) oder A. Meillets (con-”

² See Linguistica Baltica 5–6, 1996–97, 227–233.

245

tra)” (p. XIV). This gives rise to two categories of researchers: “Anhänger dieser Theorie und Gegner dieser Theorie (d. h. der balto-slavischen Sprachgemeinschaft).” Such a division of fundamental positions on the problem of the earliest relations between the Baltic and Slavic languages is schematic, too broad, and therefore does not reflect reality. Let us recall that O. Poliakov considers the Meillet method “im großen und ganzen ein großer methodologischer Fehler” (p. 14); we know whose side O. Poliakov’s sympathies are on—he is a fervent advocate of the unity of the Baltic and Slavic languages. His preconceived, a priori commitment to the existence of this unity diminished the persuasiveness and evidentiary force of an otherwise good and interesting book.

The reality is that there are neither opponents nor proponents of a Balto-Slavic linguistic community; rather, there is a problem of i n t e r p r e t a t i o n, which is at the heart of research into the earliest relations between the Baltic and Slavic languages. The Baltic and Slavic languages share a set of isoglosses that link them to one another and distinguish them from other Indo-European languages. How did these isoglosses arise? Is it necessary to assume the existence of a separate Balto-Slavic language after the breakup of the Indo-European proto-language, and thus to rely on the concept of a proto-language, which in fact goes back to A. Schleicher’s “Stammbaumtheorie” in the mid-nineteenth century? Or might this set have arisen in some other way—through the parallel development of the Baltic and Slavic languages, as A. Meillet conjectured, or through their convergent development and secondary convergence, as J. Endzelin and many other researchers believed? The great French comparativist A. Meillet attempted to find another solution to the problem: he discerned an element of parallelism in the development of the Baltic and Slavic languages. Methodologically, this was the most important point in addressing the problem of their earliest relations. In A. Meillet’s view, the similarity between the structures of the Baltic and Slavic languages was determined by the following factors: 1) the Baltic and Slavic languages arose from identical dialects (de parlers indo-européens sensiblement identiques) of the Indo-European linguistic community (l’unité indo-européenne), which were not separated by any major isogloss; 2) in their development, the Baltic and Slavic languages did not undergo radical breaks in their systems ((aucune fracture brusque du système); in other words, they are conservative and archaic and therefore retained many ancient features; 3) the Balts and Slavs lived at all times in neighboring geographical regions under similar civilizational conditions, and this was reflected in their languages. In addition, borrowings—especially those into the Baltic languages from Slavic—enhanced the similarity between the Baltic and Slavic languages.³

N. Trubetzkoy, too, did not rule out the possibility of convergent development between the Baltic and Slavic languages, noting that consistent divergent development would probably have led to the emergence of transitional, intermediate languages or dialects.⁴

Even before the Second World War, atlases of the main national languages of Europe—Italian, French, German, etc.—had been compiled. Atlases covering several languages proved particularly important for the study of phenomena in language history and language contacts, such as, for example, the eight-volume work by K. Jabergo and J. Judo, “Italy and

³ M e i l l e t A. Les dialectes indo-européens, Nouveau tirage, Paris, 1922, 40–41.

⁴ N. Trubetzkoy Selected Works on Philology, Moscow, 1987, 48.

246

Pietų Šveicarijos kalbų atlasas“ ((I tomas, 1928). Kartografiniu kalbinių duomenų pavaizdavimu – pirmiausia klasikiniu izoglosų bei jų pluostų pavidalu – buvo padėti arealinio geografinio indoeuropiečių kalbų tyrimo pagrindai. Šie atlasai iškėlė įdomių faktų, rodančių, kad kalbiniai reiškiniai peržengia dialektų ir kalbų arealų ribas, įsiterpia į kitos kalbos plotą. Indoeuropiečių kalbotyroje atrasta komparatyvinį metodą ėmus taikyti ir genetiškai negiminingoms kalboms, paaiškėjo, kad bendrų, tipologinių fonetinės ir gramatinės sistemų ypatybių gali turėti ir skirtingų struktūrinių tipų kalbos⁵. Visa tai tyrinėtojams kėlė mintį, kad dėl kitų kalbų poveikio ir didelės svetimų elementų akumuliacijos kalbos gali iš esmės pasikeisti – iš vieno struktūrinio genetinio tipo pereiti į kitą, pavyzdžiui, iš fleksyvinio į agliutinatyvinį⁶. Todėl Prahos funkcinės lingvistikos atstovai skyrė genealoginę kalbų giminystę (parentė, Verwandschaft) ir antrinę tipologinę kalbų giminystę (affinité, Affinität). Genealoginės giminystės sąvoka, išreiškianti ryšį tarp kalbų, išaugusių iš vieno kelmo, vadinamo čia prokalbe (Ursprache), čia kalba-baze, buvo būdinga visos XIX a. istorinės lyginamosios kalbotyros mąstymo aparatui. Antrinė tipologinė kalbų giminystė atsiranda tarp genetiškai nesusijusių kalbų ir apima bendrus jų bruožus, atsiradusius dėl konvergentinės raidos ir interferencijos (Prahos lingvistų terminas). Balkanų kalbų struktūrų panašumas leido tyrinėtojams imti kalbėti apie kalbines sąjungas (alliance linguistique, Sprachbund): kaimyninės, kad ir negiminingos kalbos, ilgą laiką palaikiusios pastovius intensyvaus bendravimo ryšius, viena kitą „užkrečia“ ir įgyja panašių tarimo įgūdžių, bendrų garsų bei fonotaktikos elementų, taip pat panašių gramatinės struktūros ypatybių⁷.

Apparently, the similarity between the Baltic and Finnic languages of the Baltic region also arose from these areal convergent factors. The Finnic languages of the Baltic region not only have a large body of words borrowed from the Baltic languages, but have also converged with them in their phonetic systems⁸; they show striking similarities in grammatical structures as well. Compare, for example, the present-tense paradigms of Lithuanian miegoti and Estonian magama ‘to sleep’:

Lithuanian Estonian I sleep mina magan ‘I sleep’ you sleep sina magad ‘you sleep’ he, she sleeps tema magab ‘he/she sleeps’ we sleep meie magame ‘we sleep’ you sleep teie magate ‘you sleep’ they sleep nemad magavad ‘they sleep’

⁵ Я кобсон Р. Характеристике евразийского языкового союза, Париж, 1931.

⁶ Трубецкой Н. С. Op. cit., 29–30, 53–58.

⁷ Jakobson R. Selected Writings. – I. Phonological Studies, The Hague – Paris, 1962, 137–201.

⁸ Anttila R. An Introduction to Historical and Comparative Linguistics, New York–London, 1972, 174.

247

As can be seen, we are dealing not with isolated, separate facts that are therefore quite possibly coincidentally similar, but with a fragment of a morphological system whose structural and functional similarity even extends to the material identity of its elements (plural endings -me, -te and the verb roots themselves mieg-, mag-). In the Estonian paradigm, the third person is expressed by the bare stem—as in the Baltic languages—since -b-/-va- are present-tense markers and -d- is the plural formative. The structural similarity of the paradigms is further enhanced by the fact that the historical ending of the first-person singular was *-m, and the consonant m at the end of a word regularly became n—as in the Baltic languages. In the Baltic languages, we find a materially identical element in the inflection of athematic verbs, cf. ei-m-i ‘I go’. The only difference between the paradigms is that the element -t-/-d- functions in Estonian in the second-person singular, whereas Lithuanian has it in the third person, cf. es-t-i. Other Indo-European languages, e.g. Hittite, have it in both the third and second persons, cf. Hitt. e-eš-ta ‘you were; he was’.

If convergent development was capable of bringing about such a striking coincidence in languages belonging to fundamentally different genetic structural types, what then is there to say about the Baltic and Slavic languages, which developed from a single language—even if the Indo-European proto-language was divided into dialects—and which existed as neighbors for millennia under similar geographical and civilizational conditions, their speakers maintaining close and intensive contacts.

O. Poliakov attaches great importance to the common Baltic and Slavic lexicon and to prosodic and ablaut phenomena—all of which, he argues, demonstrate the genetic, rather than parallel and independent, origin of the shared Baltic and Slavic facts. It is quite possible that some morphological innovations, such as the genitive singular ending of o-stems *-ā (Lithuanian vilko, Latvian vilka), -a (Old Church Slavonic vlъka) and the instrumental singular ending of ā-stems -ān (Lithuanian ranka, Old Church Slavonic rǫkǫ), arose at the time when the common Baltic and Slavic lexicon was taking shape (e.g., names for body parts: galva, Old Church Slavonic glava; ranka, Old Church Slavonic rǫka; delnas, Serbian-Croatian dlan; ragas, Old Church Slavonic rogъ; pentis, Old Church Slavonic pěta). But does this mean that there was a single language? The o-stem genitive singular ending is indeed a striking Baltic-Slavic isogloss, distinguishing these languages from the other Indo-European languages, since it is a paradigm innovation formed by replacing the old genitive form with the ablative form (p. 216). But Prussian preserved the old inflection (deiwas ‘of the god’, with -as from Indo-European -oso). The logical conclusion would be that, when the genitive was replaced by the ablative, quite possibly at the same time as the common Baltic and Slavic lexicon arose, the Baltic and Slavic languages were two independent dialect continua. Otherwise, the genitive would have been replaced by the ablative in Prussian as well. The same is apparently indicated by certain phonetic phenomena, such as the fate of Indo-European *s after i, u, r, k: if a Balto-Slavic language had existed after the breakup of the Indo-European proto-language, words such as teisūs, viešulas, blusà, mūsas, mūsė, ausis, daūsos, klausýti, saūsas, tausýtis, retaining an intact s (compare Old Church Slavonic tichъ, Old Russian vichъrъ, blъcha, tъchъ, mъšica, Old Church Slavonic ucho, duchъ, slušati, suchъ, po-tuchnoti), would hardly have survived.

According to the latest research, which O. Poliakov reviews in detail (pp. 202–205), Prussian neuter nouns with the nom.-acc. singular ending

248

-an < PIE *-om (e.g., assaran, creslan: Lithuanian ežeras, krėslas) is a superarchaism. Thus, the emergence in Lithuanian of neuter forms in predicative function (e.g., Kaip man gera! ‘How good I feel!’) and the appearance in Slavic languages of neuter nouns in -o (ozero, krėslo) in place of archaic neuter forms in *-om are innovations shared by the East Baltic and Slavic languages. O. Poliakov’s explanation that the old neuter form *geran (cf. Old Prussian labban) was replaced by gera by analogy with u-stem adjectives, whose neuter never had -n < *-m (cf. gražu and Hittite idalu ‘bad’), is entirely plausible. The result of analogical leveling may also be the Slavic neuter in -o. When nom. acc. sing. *-om regularly changed phonetically to *-b, the neuter category was in serious danger of disappearing. To prevent this, the ending -o may have been introduced into neuter nominal forms from the pronominal declension (cf. to ‘this’) or perhaps also from the s-stem (cf. slovo < *slovos). O. Poliakov agrees with this classical Indo-Europeanist treatment of the phenomenon (A. Leskien, K. Brugmann, F. Fortunatov). Thus, the Proto-Indo-European neuter category in *-om was eliminated independently in Baltic and Slavic, but the phenomenon is not late: Baltic-origin loans in Baltic Finnic languages already lack -an (cf. Finnish silta ‘bridge’, heinä ‘hay’).

Further cases of parallel and independent development can be found in the nominal declension system. The i-stem nom. plur. ending *-ies > Lithuanian -ys (cf. ugn-ys ‘fires’) and Slavic -bje < *-i̯ies (cf. Old Church Slavonic gostbje) replaced the PIE inflection *-eies in this case (cf. Sanskrit agnayas ‘fires’) independently in each language group. This is evident from the fact that the old PIE inflection survives in the Lithuanian dialectal (Aukštaitian) form ugnies, with -ies < *-eis < *-eies. O. Poljakov also accepts this development (p. 274).

It is not sufficiently clear whether the i-stem acc. plur. ending -is (Lithuanian akis, Latvian acis ‘eyes’) and Slavic -i (Old Church Slavonic ogni ‘fires’) constitute “eine gemeinsame ostbaltish-slavische Neuerung” (p. 276), since *-ins > -i is the only instance of such a phonetic change, and is therefore highly problematic. Rather, one would expect here the same result as in the jo-stem acc. plur. form, where *-i̯ins yielded a nasal-vowel reflex; cf. the pronouns Old Church Slavonic je ‘them’, ježe ‘which’, and also the i-stem gen. plur. Old Church Slavonic gostbji, with -bji from *-i̯ōm. The Baltic and Slavic inflection in this case is likewise probably the outcome of parallel, independent development. A clear case of parallel development in the vowel system is the change of the diphthong *eu to Lithuanian iau and Slavic ju, since during the contact between Baltic and Baltic Finnic, Baltic eu was still intact; cf. Finnish leuka ‘Unterkiefer, Kinn’ < Baltic, cf. Lithuanian liaukà ‘gland’ (for the semantic connection, cf. Lithuanian geležūnys ‘gland’: Latvian dzēdzs ‘Kinnlade, kiefer’; Latvian žuokli ‘Kinladen; Mandeln’). It is also not entirely clear why the i-stem gen. sing. form with the ending *-eis (ugnies, Old Church Slavonic ogni) constitutes a Baltic and Slavic innovation (p. 270). After all, this ending occurs in Germanic (gen. sing. Old High German ensti, Old Saxon ensti, Old English ēste ‘favor, grace’, with Germanic *-eiz < PIE *-eis), Oscan (aeteis), and probably Old Indian (gátes ‘going’)⁹.

⁹ B a m m e s b e r g e r A. Die Morphologie des urgermanischen Nomens, Heidelberger, 1990, 124–125.

249

Alongside dat. plur. forms with Germanic *-miz, which are probably of instr. plur. origin (cf. Saitchamimis in the oldest Germanic sources), it is now generally accepted that there were also dat. plur. forms with the ending *-maz <PIE *-mos, for example, the Gothic dat. plur. dagam ‘to the days’ is derived from *dagamaz with syncope of the third syllable and subsequent assimilation *dagamz >dagam(m)¹⁰. The Slavic dat. plur. ending -ть may continue either *-mus or *-mos, while Baltic has instances in which the vowel a changes into the vowel u, especially next to labial consonants (cf. Lithuanian mazgas: muzgas, paskui: puskui, kaponė: kuponė, sambarė: samburė, Latvian jūrmala: jūrmule). Thus, the dat. plur. *-mos and instr. plur. *-mis or *-mis inflection is a dialectal feature inherited from Proto-Indo-European, forming an opposition to the inflection *-bhi(i)os. found in other Indo-European languages (Aryan, Greek, Italic, Celtic).

The two examples given below illustrate O. Poliakov’s method of attempting to prove the existence of a Balto-Slavic language after the dissolution of the Indo-European linguistic community. The o-stem nominal nom. plur. o *-oi (vilkai, Old Church Slavonic vьlci) is said to be “eine gemeinsame genetische Isoglosse im Baltischen und Slavischen” (p. 236), whereas the same ending in Greek (λύκοι) is taken to be the result of parallel, independent Greek development. In explaining the origin of the o-stem dat. sing. inflection (o vilkui, Old Church Slavonic vьlku), O. Poliakov relies on V. Mažiulis’s theory that Baltic and Slavic shared the ending *-ō here, which was later replaced in Slavic by u-stem *-u<-*-ou¹¹, but rejects Mažiulis’s claim that this ending was common Indo-European, since with the addition of *-i it was found in Hittite, Aryan, Greek, and other languages. This, he argues, diminishes the impression of Balto-Slavic unity (pp. 223–224).

The statement that the reflex of the long diphthong *ōi ui “im Inlaut nie vorkommt” (p. 217) is inaccurate; therefore, he rejects the development *ōi > *uoi > ui, claiming that it cannot be supported by the facts (p. 219). However, there are words in which ui may derive from the diphthong *ōi¹². The possibility of a root *-ōi is also indicated by Lithuanian kuomija ‘dwelling place, place (of a wolf),’ which is from the same root as kaimas. Only here the reflex of *ōi was different—in the position before m, i disappeared¹³. With this in mind, and knowing that the dative singular of o-stems, *-ōi (from *-o +-ei), is well attested in Indo-European languages, the traditional explanation that *-ōi changed 1) to *-ō (with -i lost through sandhi) >Lithuanian -uo and 2) via an intermediate stage *-uoi >Lithuanian -ui >-u (with the subsequent loss of -i) is not entirely impossible. The claim that Baltic and Slavic did not preserve root words with diphthongal roots, and that this is another feature they share, is also inaccurate ((p. 290)¹⁴. That the dative singular ending of u-stems -ui (sūnui) < *-uoi < *-ōi derived from *-ou-ei (p. 282) is phonetically impossible. But O. Polyakov’s idea that the dative singular ending of u-stems

¹⁰ B a m m e s b e r g e r A. Op. cit., 45–46.

¹¹ M a ž i u l i s V. Baltų ir kitų indoeuropiečių kalbų santykiai, Vilnius, 117 ir tolesni.

¹² K a z l a u s k a s J. Lietuvių kalbos istorinė gramatika, Vilnius, 1968, 120 (su literatūra).

¹³ K a r a l i ū n a s S. Baltų kalbų struktūrų bendrybės ir jų kilmė, Vilnius, 1987, 183.

¹⁴ K a r a l i ū n a s S. Op. cit., 188.

250

-uo (in Samogitian and Gervėčiai šūnuo) may have developed from the loc. sing. *-ōu (cf. Samogitian turguo <’-uo-ēn and <’-ōu, and Old Indian śūnaú <’-ōu).

The history of consonant stems in Baltic and Slavic languages and the development of their inflection are certainly not special evidence of Baltic-Slavic linguistic activity (p. 299), since the consonant-stem declension type in these languages is a clearly preserved Indo-European archaism, showing that Baltic and Slavic languages derive from the Indo-European proto-language. Some shared changes in the declensional paradigms of Baltic and Slavic languages (e.g., stem endings with the normal ablaut grades -en-, -er-) are conditioned by the general tendency in language development toward simplification and leveling of paradigms, and are therefore late and arose independently in each language group. In general, too, the history of the declensional systems of Baltic and Slavic languages does not support the conclusion that, after the breakup of the Indo-European linguistic community, there was a common source of Baltic and Slavic declension of the kind found in Indo-Iranian (p. 302). What is common to Baltic and Slavic noun declension is, first, a state of Indo-European archaism preserved in many cases; second, reflected dialectal features of the northwestern area of the Indo-European proto-language, from which the Baltic, Slavic, and Germanic (as well as Tocharian) languages emerged; and third, innovations in Baltic and Slavic languages arising through convergence, at the levels of morphology, phonology, prosody, and lexical semantics, as a result of the neighboring position and long-lasting intensive contacts of the Baltic and Slavic peoples, as well as similar civilizational and geographic conditions.

There are also more questionable reconstructions or comparisons, as well as one or another mistake. Perhaps OCS azъ can indeed be reconstructed as *eg’hom (p. 180), cf. Old Indian ahám ‘I’; but Lithuanian aš (cf. dialectal and Old Lithuanian ėš) and Latvian es <Baltic ēz can only derive from Indo-European *Hēg’h₃ with zero-grade vocalism, which existed alongside *Heg’ēh₃ with normal ablaut grade in the Latin and Greek forms ego and ἐγώ. Although the origin of the diphthong ui remains unexplained, as O. Poliakov correctly observes (pp. 217–218), it is certain that Lithuanian lūisas ‘lame’, kuīnas ‘without horns, hornless’ and tuikti ‘become dizzy, confused’ have no etymological connection, respectively, with lūsis ‘Lynx lynx’, kūnas ‘body’ and tūkti ‘become fat’ (p. 218). Moreover, Chr. Stang does not connect them either¹⁵. Lūisas seems to be related to lūsas ‘lame’, lūšinti ‘make lame’, lušónas ‘cripple’, ap-lūšti ‘cripple’ and, further, lūžas ‘lame’; kuīnas may be etymologically related to kuīnas ‘club, cudgel’, kuīnis ‘stick, rod’ and, further, through the intermediate form *kain-, to keinis, keinys ‘stick for striking “ripkai”’ (for the semantic connection, cf. Lithuanian baužas ‘without horns, hornless’: Latvian bauzis ‘hornless bull’, bauze ‘stick, rod; loom beam…’), while tuikti undoubtedly belongs to the root found in tvaikti ‘…become dizzy, get giddy’. Prussian naryti s ‘be defiant’ (p. 155) is in fact Lithuanian narytis (-ijasi) ‘show stubbornness’ LKŽ VII 564, which is used in Prussian Lithuanian writings instead of the Slavic loanword found in other dialects, narãvytis ‘same’, and therefore has nothing to do with nóras. Because of the different root structure, Lithuanian

¹⁵ S t a n g Chr. S. Comparative Grammar of the Baltic Languages, Oslo–Bergen–Tromsö, 1966, 72.

251

mokėti, laukas, toli, sotus, on the one hand, and OCS mogą, logą, Russian dalb, OCS sytb, on the other (pp. 108–110), represent different roots. Because of the different root vocalism, Lithuanian glodùs and Latvian gluds (p. 109), as well as Lithuanian brėžti and braũkti (p. 114), cannot be etymologically related. The relationship in origin between Lithuanian nãmas and dimstis ‘courtyard, estate’ (p. 109) is indeed unclear¹⁶, but the possibility cannot be ruled out that namas and Latvian nams have an independent root 'nem-. OCS trepetati and Russian pryskatb are also better not connected with Lithuanian drebėti and Russian bryzgatb (pp. 104–105), but treated as separate roots¹⁷.

The difference in the initial consonants of Lithuanian strãzdas, Latvian str azds, and Slavic 'drozdъ (p. 104) should be explained by Siebs’s law. Lithuanian geĩsti does not contain '-ds- (p. 100), but consists of the root geid- and the infinitive formative *-t(e)i, and therefore has *-dt- at the morpheme boundary (> *-tt- >-st-). Page 94 states that pirštas, OCS prъstъ have Lithuanian š and Slavic s from ‘k’, whereas on p. 115 they are assigned to words that have in Lithuanian š derived from Indo-European *s after r. Page 293 indicates that the o-stem gen. sing. ending *-ā (vilko, OCS vlъka) derives from Indo-European *-ās; on p. 210, however, it is correctly argued that this is the old ablative ending (*-ōd). On pp. 178–179 there is Finnish paimen ‘herdsman’; on p. 208 piemen is correctly given. On p. 179 we find Prussian menschon ‘of the fish’; on p. 263, correctly, ‘of the flesh’. Prussian guntwei is not 1st plural (p. 69). Instead of Lithuanian taišyti (p. 96), Prussian sachis (p. 101), Old Indian ugnis (p. 110), Lithuanian taisytis (p. 116), Latvian bārti (p. 141), Lithuanian sù vienà šviesà (p. 226), instr. sing. bãdu (p. 227, note 269), gen. dual. Dviijų karvių (p. 260, note 330) should read, respectively, tašyti, sackis, ágnis, taušytis, bārt, sù viena šviesà, badù, dvijų karvių.

Simas Karaliūnas

THE LINGUIST’S RETURN

Tamara Buch. Opuscula Lithuanica. Wydal Wojciech Smoczyński, Warszawa, 1998: Katedra Językoznawsta Ogólnego i Bałtystyki Uniwersytetu Warszawskiego, 381 p.

Our well-known Germanist Ina Meikšinaitė recently quoted an excerpt in her published memoirs about her sister Tamara Buchienė, taken from a letter she wrote to Monsignor Vincentas Mincevičius, who lived in Rome: “And what is life doing to me! I always only wanted to live in Kaunas, walk along the banks of the Nemunas; I needed pine forests, lakes, silence. Why did I have to live in various countries, among all kinds of people, speak different languages and—at this time—lose

¹⁶ F r e a n k e l E. Litauisches etymologisches Wörterbuch, I, Heidelberg–Göttingen, 1962, 483.

¹⁷ Ф а с м е р М. ЭСРЯ III (1987) 391, IV (1987) 99.

252