LIETUVIŲ KALBOTYROS KLAUSIMAI, VI

LIETUVOS TSR MOKSLŲ AKADEMIJA

ON A SLAVIC-BALTO-FINNIC ISOGLOSS

B. A. LARIN

I

In the preface to the *Dictionary of Regional Great Russian*, edited by A. Kh. Vostokov, we read (pp. II—III): “In the old Russian language we find the nominative replacing the accusative with an infinitive. This feature has been preserved in the present-day regional Northern dialect. People say: glove take off, sister accompany, water pour, river cross, and so on.” In the same year, 1852, P. A. Lavrovsky identified such constructions in Old Russian writings and contemporary Russian folklore¹. Subsequent research proceeded along these two lines.

In F. I. Buslaev’s *Historical Grammar of the Russian Language*², the function of infinitival constructions “denoting the necessity or possibility of an action” was defined, and vivid examples were collected, beginning with 1229 (the treaty charter of Prince Mstislav Davydovich of Smolensk with Riga and Gotland) and ending with “still in use in regional dialects, for example, in Arkhangelsk: ‘buy out if the right hand [is]’.” Examples from a seventeenth-century collection of proverbs are given:

To kill a lamb is not to destroy a soul. Without money, water to drink. To speak truth is to lose friendship. To give a loan is to be chilled forever. To love a game is to buy a domra. The first little song—sing it blushing.

From *Ancient Russian Poems* (Kirsha Danilov’s collection): “Though the leg be broken, the doors must be taken off. Where Yermak might winter.”

¹ П. А. Лавровский, О языке северных русских летописей, СПБ, 1852, стр. 122—124.

² F. I. Buslaev, *Historical Grammar of the Russian Language* (1st ed. 1858, 2nd ed. 1862, § 196).

87

I have quoted so much from F. I. Buslaev because later Russian syntacticians take their point of departure from him, repeat his examples, and gradually enlarge the body of this material.

In an 1872 article, Vs. Miller³ cited a Vedic parallel to the construction in which an object in the nominative accompanies an infinitive: sa asya mahimā na samnaçē (literally, “his great might not to attain”), thereby considerably extending the chronological scope of the study of this turn of phrase.

In three pages of his book *From Notes on Russian Grammar*⁴, A. Potebnya further broadened the field of research in several directions. He cited five examples from medieval Belarusian writings (the oldest dates to 1387) and one from contemporary Belarusian vernacular speech, thereby refuting the view, established by his time and still repeated today, that this phenomenon was confined exclusively to the Russian North.

He connected the origin of the construction with a pattern of grammatical evolution that he substantiates in this study: “from impersonal constructions to personal ones.” He reconstructs the ancient form of the formula Such is the truth: for a Rusyn to take from Riga and on the Gotland shore as follows: Such is the truth [there is] for a Rusyn to take…, where he considers the word truth the subject, and taking (a noun in the dative case) the object or an adverbial of purpose belonging to the predicate. Later, in his view, the construction became impersonal, and the word truth, while remaining formally in the nominative, was reinterpreted as a direct object of taking (now an infinitive). This explanation of Potebnya’s—which seems debatable to me—opened the way to a profound historical analysis of this puzzling syntactic construction. After A. A. Potebnya’s work, it is no longer possible either to project the modern Russian understanding of this construction into the past or to regard it as immutable and unambiguous throughout all of the history accessible to us.

³ This article, written in response to a questionnaire for the Fourth Archaeological Congress in Kazan, was published in the *Moscow University News* for 1872, part IV, and was subsequently revised into the book *A View of the Lay of Igor’s Campaign*, Moscow, 1877. It should be noted that the author did not include the Vedic parallel in this book, since the construction cited is not typical of Vedic and differs from the Russian examples in that the infinitive had not yet crystallized in the Vedas: numerous and varied nominal formations were in use, in which the case form of the noun was still readily understood.

⁴ A. A. Potebnya, *From Notes on Russian Grammar*, II, 416–18, Kharkov, 2nd ed., 1888. Since many people continue to regard this construction as specifically North Russian even to the present day, I reproduce here the Belarusian evidence found by A. A. Potebnya: 1) From Grand Duke Jogaila’s charter of 1387: “and in every matter and case he (Skrigala) is to administer justice and his people”; 2) from *Acts of Western Russia*, I, 352: “For a Vitebsk man, to return all belongings in kind” (1503); 3) ibid., I, 92: “to give them a measure of mead”; 4) ibid.: “to give him half a measure of mead, and half a Lithuanian groat” (1480); 5) ibid., II, 27: “if anyone should obtain a sword against someone, then a hand to be cut off; and if anyone wounds someone, then a neck to be cut off” (1507). Modern example: “To give half of Marko the bogatyr” (Romanov, *Belarusian Collection*, III, 338). Academician E. F. Karsky (*The Belarusians*, issue 2, reprint 1956, p. 163, and issue 3, p. 395) found a further nine examples in monuments from the fourteenth to sixteenth centuries, as well as several examples from folk speech.

88

After interpreting the construction “nominative with the infinitive,” A. Potebnya had no difficulty explaining Old Russian and modern dialectal constructions with надобо (надо), (н)требь with the nominative case (не надобо моя дань, 1490; не требь ти молчание “Lexicon palaeoslovenico = graeco = latinum F. Miklosich’a”; modern али сила тебе надоть несчетная). Potebnya was the first to link these constructions with infinitival ones, which aids the historical understanding of both. In his view, the nominative case of the noun in these constructions was and remains the expression of the subject, while (есть) взяти, (есть) надобо is the predicate.

A. V. Popov⁵, who wrote that “the origin of this type of construction is entirely clear,” and D. N. Ovsyaniko-Kulikovsky⁶, who called the form of the noun in these constructions a “pseudo-nominative,” followed Potebnya.

A. N. Sobolevsky⁷ and E. F. Karsky also accepted Potebnya’s theory. Ignoring Buslaev’s and Potebnya’s observation that masculine nouns could occur in the nominative in this construction (знать сова по перью, — знать сокол по полету; ленивой и по платью знать), Sobolevsky asserts that this construction admitted only singular feminine nouns. Like Potebnya, he regards it as a later result of the decomposition of the primary construction—a combination of a direct object in the nominative with an imperative or a participle (Дай бог ему здоровье и мьзда спасения от бога прияти Prologue, 1383; Взявъ у воеводы отписка, 1634; соямя рубашка—in the Domostroy).

In an early study⁸ of the language of the Novgorod charters, A. A. Shakhmatov ventured a very bold conjecture for his time, which, let us note at once, has only now begun to receive further support: “It is highly probable that in the fourteenth century, for words denoting animate objects, the same pattern was observed as in the modern language; but perhaps, alongside the genitive-case form = the accusative-case form, there may also have existed an accusative singular and plural form identical to the nominative singular and plural, as in the cases: держати тивунъ (No. 2); тивунъ свой держати (Nos. 1 and 9); слати осетрьникъ и медовара (Nos. 1 and 3); при осетрьника (Nos. 6, 7, and 9); всякий гость пустити (No. 12), and certain others do not prove anything against this, since these may be nominative cases used in place of the expected accusatives under the influence of the infinitive, just as we find here: вся таль пустити (No. 12), грамота изрѣзати (No. 12), та грамота дати ти назад (No. 3), продати ти дань своя (No. 10), дати исправа (No. 16).” Shakhmatov was the first to recognize that the coincidence of nominative and accusative forms in masculine and neuter nouns cannot be grounds,

⁵ А. В. Попов, Сравнительный синтаксис именительного, винительного и звательного падежа в санскрите, зенде, греческом, латинском, немецком, литовском, латышском и славянских наречиях, „Филологические записки“, 1879, вып. VI, стр. 47.

⁶ Д. Н. Овсянико-Куликовский, Синтаксис русского языка, 2-е изд., СПБ, 1912, стр. 205, 207 и 202.

⁷ А. И. Соболевский, Русский исторический синтаксис, литографированное издание, СПБ, 1892—93, стр. 57 сл.—80 сл.

⁸ А. А. Шахматов, Исследование о языке Новгородских грамот XIII и XIV вв., СПБ, 1884, стр. 202.

89

to rule out the possibility of a construction “nominative direct object with an infinitive” containing masculine and neuter nouns, not only feminine nouns (whose nominative and accusative forms differ). Moreover, he proposed a special category of the less common “nominative direct object of the animate category” in place of the accusative, which had long since merged with the genitive, and which was preserved chiefly with the infinitive.

In a later study⁹ on the language of the Dvina charters, he continued his reflections on these constructions. Having recognized as the norm in these texts for “animate nouns” the identity of the singular and plural genitive forms with the accusative form, he adds (p. 129 ff.): “let us note here that конь воронь (а дал… полишестадцать сороковъ бѣлки, а пополонка конь воронь № 90) is not an accusative, but a nominative. Likewise, the form бобры (Се купи… и у стороньнихъ рѣкахъ рыба ловити и бобры бити (No. 108) is not a plural accusative, but a nominative.” As we see, he also assigns these examples to the construction “nominative instead of accusative in the direct object.” Shakhmatov draws a clear distinction between: (a) a predicate expressed by a finite verb (се купил… а далъ… and the like), in which cases the plural nominative appears in place of the accusative because “the transitive verb governing it is at a considerable distance from the governed word”; and (b) a predicate expressed by an infinitive (p. 131): “the replacement of the accusative by the nominative seems to be explained by the fact that the infinitive, being by its very origin a noun, did not possess the full verbal meaning that other verbal forms possessed.” Here, following Miklosich, he is prepared to recognize the noun in the nominative as the subject. Finally, in his last work on syntax¹⁰ he is inclined to derive the construction “infinitive with the nominative case” from “надо(бѣ) + infinitive.”

This, in fact, brings the theoretical discussion of the question to a close. The minor corrections and additions made later were not as significant as the widening of the geographical or linguistic boundaries of the construction under study as materials from diverse sources accumulated. For example, in the “Oath Charter of Prince Dmitry Olgerdovich of Bryansk to King Jagiełło, 1388,” published by V. A. Rozov¹¹, we find: слюбу ѣсть держати и цѣлую правду и чиста вѣра… A few years later, Prof. E. Timchenko¹² also found the nominative direct object with an infinitive in a number of Old Ukrainian sources. Nevertheless, as late as 1927, S. P. Obnorsky¹³ calls this an “old North Great Russian feature” and goes on to write: “the existence of this feature in Belarusian, while it is absent in South Great Russian and Little Russian, is characteristic,” citing Kolosov’s “Survey of the phonetic and formal features of the vernacular Russian language” (Warsaw,

⁹ А. А. Шахматов, Исследование о Двинских грамотах XV в., ч. I—II, СПБ, 1903.

¹⁰ А. А. Шахматов, Синтаксис русского языка, вып. 1, Лд., 1925, стр. 12.

¹¹ Южнорусские грамоты, собранные Владимиром Розовым, т. 1, изд. Отд. рус. яз. и словесности АН, Киев, 1917, стр. 42—44.

¹² Е. Тимченко, Номінатив і датів в українській мові, у Київі, 1925 р., стр. 15 сл.

¹³ С. П. Обнорский, Именное склонение в современном русском языке, вып. 1, Л., 1927, стр. 266.

90

1878), while ignoring the works of Ukrainian linguists. In this connection, P. Bitsilli¹⁴ cited from I. I. Sreznevsky’s collection „Monuments of the People’s Language“ another excerpt from the text of a Ukrainian folk duma: „Чи ще́ж тобѣ не далася тяжкая неволя знати“. In his article he noted that the construction was used predominantly with an independent infinitive and set the chronological limit for its use in literary Russian at the first quarter of the eighteenth century, explaining its disappearance as the conscious removal of an „anomaly in formal and semantic terms“. This supposedly happened „all at once… which prompts us to link this phenomenon with the Petrine reform that stirred thought.“ The language of eighteenth-century writing and literature has been studied too little to accept P. Bitsilli’s conjecture without reservation. I can cite an example from Novikov’s „The Painter“ (1772, fol. 15): Once you have sung the prayer service, you can bring him a glass of wine. The period of use of the construction may perhaps be extended closer to our own time. And how can one disregard the construction’s very wide prevalence in colloquial Russian! Living speech, even in its vestigial features, breaks through into literary language. In the second edition of his play „Yegor Bulychov and Others“¹⁵, M. Gorky, in defiance of proofreaders and editors, restored Bulychov’s line: „Much was beaten and plundered in youth; in old age one must save soul“, rather than save one’s soul, as the „guardians“ of a planed-down literary language wanted to print.

In a 1935 work, B. Unbegaun¹⁶ gathered documentary evidence from the first half of the sixteenth century for the use of the nominative as a direct object with feminine nouns (and nouns such as *rukhlyad’*, *rat’*, and *doch’*), noting that animate masculine nouns always occur in the genitive-accusative case form.

In his dissertation, Ya. A. Sprinchak presented a systematization of the modal functions of infinitival constructions¹⁷. F. P. Filin¹⁸ published excerpts from materials on this construction gathered by Leningrad dialectologists on expeditions for the Dialectological Atlas of the Russian Language, selecting valuable new data that уточняют and correct the traditional account, for example, the nominative as direct object with a dependent infinitive (*I want to drink cold water* — Vologda); with personal forms of the verb (*and fish, we really do love [it]* — White Sea region; *today the stove they heated* —

¹⁴ P. Bitsilli, The Use in Russian of the Nominative Form of Feminine Nouns in -а with an Infinitive, „Collection in Honor of Prof. L. Miletić on the Seventieth Anniversary of His Birth (1863—1933)“, Sofia, 1933, pp. 199—208.

¹⁵ In the almanac „The Year Sixteen“, Almanac 1, Moscow, 1933. Also reprinted in the 30-volume Collected Works, vol. 18, p. 106.

¹⁶ B. Unbegaun, La langue russe au XVI siècle (1500—1550). La flexion des noms, Paris, 1935, p. 128—131.

¹⁷ Ya. A. Sprinchak, Syntactic Constructions in the Law Code of Ivan the Terrible: a) The Construction Infinitive plus Feminine Singular Nominative, „Scientific Notes of the Herzen Leningrad Pedagogical Institute“, vol. XX, Leningrad, 1939, pp. 133—143; b) The Construction „infinitive with a feminine noun in the nominative“ in the History of the Russian Language, „Scientific Notes of Dnipropetrovsk University“, XIX, 1941, pp. 3—47.

¹⁸ F. P. Filin, On the Use of the Nominative Case Form of Feminine Nouns in -а with Accusative Meaning, „Bulletin of the Dialectology Section of the Institute of the Russian Language“, issue 1, Moscow—Leningrad, 1947, pp. 17—22.

91

Vologda region); with an imperative (give me a berry — White Sea region; cut off a crust of bread — Pudozh district); with the adverb видно ‘visible’ (from here, the hut is visible — Leningrad region); with a converb (a week having passed — Leningrad region). Rephrasing Shakhmatov’s early idea, F. P. Filin asserts the development of a “general accusative category,” a “general accusative case,” but only from feminine nouns ending in -а.

Following F. P. Filin and citing material collected later by the Leningrad scholars, V. Georgieva¹⁹ points to instances of the nominative case being used instead of the partitive genitive: Cut some grass for the calf — Arkhangelsk; How many fish do you need to catch? — Krasnye Gory, Estonian SSR; We have little strength now — Novgorod. V. Georgieva proposes considering separately and independently the combinations of the nominative case with nado ‘need to’ (p. 47), although she herself has cited quite a few examples of the type: the water needs wiping up — White Sea area; the insole needs to be put in — Leningrad region (p. 43), which point to a close connection between these constructions.

V. I. Borkovsky²⁰ gave a detailed account of his predecessors’ work and presented rich, well-organized material on the nominative used for the direct object in the oldest Russian charters. In explaining the origin of the construction “nominative case with an infinitive,” he considers A. A. Potebnya’s hypothesis the most convincing; as for its geographical distribution, he fully agrees with M. A. Kolosov’s claim that Novgorod was the center from which it spread (see pp. 345 ff.), disagreeing on this point with Buslaev, Vs. Miller, Potebnya, and Timchenko.

The latest work on this question known to me was written by M. I. Pi-gin²¹. It reiterates the main positions of earlier scholars, beginning with Potebnya, but without any references to the relevant literature. The illustrative material (both dialectal and from older written sources) was collected independently. If this were taken as the outcome of more than a century of Russian linguists’ efforts to explain the construction under study historically and genetically, one would have to consider the question settled, given the way scholarly thought moves in a closed circle. This, however, is not the end, but only a first step toward solving the problem.

II

As early as the beginning of this century, constructions with a nominative-case direct object used with an infinitive, a converb, a gerund, etc. were distinguished in syntactic studies of Lithuanian (J. Jablonskis) and Latvian (J. Endzelīns). Russian linguists, however, did not notice this and until quite recently considered such constructions, as we have seen, specifically Russian, even Northern Great Russian. Needless to say,

¹⁹ V. Georgieva, Syntactic Constructions Formed by Combining the Nominative Case with an Infinitive or the Adverb “nado” in Modern Russian Dialects, “Materials and Studies in Russian Dialectology,” vol. III, Moscow–Leningrad, 1949, pp. 40–58.

²⁰ V. I. Borkovsky, Syntax of Old Russian Charters, Lviv University Press, Lviv, 1949, pp. 338–351.

²¹ M. P. Pigin, Constructions Containing a Combination of an Infinitive with the Nominative Form of a Noun in the History of the Russian Language, “Scholarly Notes of the Karelian-Finnish University,” vol. IV, issue 1, Petrozavodsk, 1954, pp. 84–111.

92

that no one had even tried to look for analogies beyond the Indo-European languages. Such narrowness of outlook led to violence against the material in its interpretation and historical analysis.

The nominative structure of the sentence dominant in Indo-European languages was regarded as the only possible one for languages of this family, and everything that did not conform to it was forced into the nominative model at all costs, or interpreted as an “anomaly,” a decomposition or deformation of the original nominative sentence. Thus, in order to substantiate his hypothesis concerning the origin of the “nominative with infinitive,” A. A. Potebnya had to insert the copula “is” into the phrase Такова правда узяти Русину… so that the nominative case правда would become the subject, and есть узяти the predicate, and the disrupted canon could thereby be reconstructed.” This procedure was what most appealed to all of Potebnya’s followers. But in the Treaty Charter of Mstislav Davydovich with Riga and the Gotland coast of 1229, from which this example is taken, our phrase occurs 14 times and never with the verb есть. Another construction with есть occurs once: Тая правда есть Руси изъ Смолынеска до Гочкого берега, that is, “this law is valid (applies) for the Rus’ from Smolensk to the Gotland coast.” But this construction contains no infinitive at all and is in no way a prototype of the construction under investigation. And if, among the relatively few instances of the nominative case used for a direct object with a dependent infinitive (dependent on another verb in a finite form), the verb есть does occur as a rare exception, it stands outside this phrase, governing it, with the meaning “should, ought to”; that is, it is a pleonastic expression of modality already inherent in the infinitival phrase.

To understand our “anomaly” within the nominative system, we must set aside the Indo-European canon. The indisputable existence of an “ergative system” in Caucasian and other languages, or of a “passive system” in certain Asian languages, gives grounds for taking a different approach to interpreting “anomalies.” It is entirely legitimate to view such clear anomalies as the one we are studying either as substrate intrusion or as a survival (relic) of a pre-nominative system in Indo-European languages. K. K. Uhlenbeck’s well-known article²² “Agens und Patiens im Casussystem der indogermanischen Sprachen” (“Agent and patient in the case system of Indo-European languages”) drew general attention to the coincidence of the accusative marker in the masculine and feminine singular with the nominative-accusative marker of the neuter, which points to an ancient, single case for the subject in a passive construction and for the object in an active one (after transitive verbs). The nominative case of the “acting entity” (agens), marked by “S,” developed later, unevenly, not in all Indo-European languages, and only with masculine and feminine nouns. Uhlenbeck identified parallels in Greenlandic, Basque, and other languages. This opens up a far-reaching

²² C. C. Uhlenbeck, Agens und Patiens im Casussystem der indogermanischen Sprachen, „Indogermanische Forschungen“, Bd. XII, Strassburg, 1901, S. 570—572.

93

historical perspective on the pre-Indo-European system of nominal inflection. As early as 1876, A. Leskien²³ convincingly denied the existence of an “S”-marked nominative in Slavic languages, which may point both to the relatively late final differentiation of the nominative and accusative in individual Indo-European languages and to the persistence, in some of them, of the “passive nominative.”

Prof. S. E. Malov (speaking in response to my paper on this subject in the 1930s) saw an analogy to the constructions under consideration in a feature of the structure of Turkic languages: the nominative form, which consists of the bare stem without any markers (as did the most ancient Indo-European form of the nominative of neuter nouns: Sanskrit madhu, Greek μεϑυ, and others), served from the earliest written records to express an indefinite object, both after finite verb forms and after formations corresponding to the infinitives and participles of Indo-European languages.

In Finnish,²⁴ a nominative similar to the Turkic one is used as the case of the direct object (with accusative meaning) in the singular in the following instances: 1) after the imperative: Syö kala! “eat the fish!”; Anna kirves! “give me the axe!”; Löytäkää kassa ja ostakaa lippu “find the ticket office and buy a ticket”; 2) after the first infinitive or a participial phrase with a verb expressing obligation, or with adverbs and predicatives such as нужно, стоит, хорошо, полезно: Minun täytyy maksaa velka “I must pay the debt”; Teidän täytyy lukea tämä kirja “you should read this book.” The same form expresses the subject with a verb in the passive form (sana sanotaan “the word is spoken”). In other words, here we have a typical “passive nominative,” which can express the subject with a passive predicate and the object in other cases. Similar constructions are also found in Estonian. The nominative, used to express the object, occurs after the imperative (Anna vikat ja luisk peremehe kätta “Give the scythe and whetstone to the master!”), after the -da infinitive expressing obligation or necessity (Õpetükk tuleb selgeks õppida “The lesson must be learned”), and with a dependent infinitive (Tal oli kavatsus kirjutada kiri “He intended to write a letter”), as well as with an impersonal predicate (Kolhoosis lõpetati kevadkülv “The spring sowing was completed on the collective farm”).²⁵

²³ А. Leskien, Die Deklination im Slavisch-Litauischen und Germanischen, Leipzig, 1876. Некоторые ценные соображения можно найти в статье Э. Э. Андерсена, К вопросу о генезисе падежей именительного и винительного, „Русский Филологический Вестник“, кн. 4, 1912, стр. 43—54.

²⁴ This unmarked nominative is sometimes called the accusative. See, for example, H. Jensen, Finnische Grammatik, I, Hamburg, 1934, S. 32 ff.; L. Hakulinen, Развитие и структура финского языка, ч. 1, Фонетика и морфология, § 47, М., 1953; ч. 2, Лексикология и синтаксис, § 30—31, М., 1955. Here (p. 224) Setälä’s view is cited: the indefinite object was expressed by the nominative (unmarked) form, and the more definite object by the accusative. This fully corresponds to the distinction in the expression of the direct object in Turkic languages; see the reference above to S. E. Malov. See also Arne Runeberg, Some observations on linguistic patterns in a bilingual society, Part I, Helsingfors, 1951, p. 37.

²⁵ See 1) N. Remmel, J. Valgma, E. Riikoja, Eesti keele grammatika VIII—X klassile, Tallinn, 1959, lk. 169—170. 2) E. Päll, E. Totsel, G. Tukumtsev, Eesti ja vene

94

The use of the nominative as a direct object in Finnish and Estonian (as in Turkic languages) fully corresponds to Old Russian and modern dialect constructions with a nominative object: 1) with the infinitive, used mainly to express volition, as well as impersonal and temporally indefinite action; 2) with the imperative; and 3) with adverbial participles. There is also a parallel in the use of this nominative-object form with finite verb forms. Let us examine the Russian material in greater detail.

III

The geographical range of constructions with the nominative as a direct object, as has already been noted (by F. P. Filin and V. Georgieva), is steadily shrinking. In the last century, the main cause has been the spread of literacy and the increased influence of the literary language. Earlier, the leveling force of the dominant nominative sentence pattern must have played a role, as did the accumulation of new synonymous means for expressing those syntactic functions that had previously been expressed by archaic constructions, remnants of a different linguistic structure.

The materials cited by Ukrainian scholars make it certain that constructions with a nominative object existed in the medieval Ukrainian-Belarusian literary language, into which such constructions could have entered only from the living speech of the population of the Lithuanian-Rus’ state. It may therefore be acknowledged that five or six centuries ago these constructions were known in all three East Slavic languages as relics, but they were most widespread and most persistent in Great Russian dialects and in the business language of Muscovite Rus’.

The claim that these constructions were and are unknown in South Great Russian dialects also proved inaccurate. In a number of works by V. I. Sobinnikova²⁶, it has been shown that both the Voronezh charters and modern Voronezh dialects have the construction “infinitive with a nominative direct object”: чинить рас-

keele korvutav grammatika. E. Päld, E. Totsel, G. Tukumtsev, Comparative Grammar of Estonian and Russian, Tallinn, 1962, p. 246. I will add several more examples: 1) a direct object in the nominative with an infinitive, when the action is said to be necessary; 2) a direct object in the nominative with participles corresponding to our adverbial-participial constructions; and, finally, 3) with an impersonal predicate, which is analogous to certain Slavic and Baltic constructions. These examples were kindly provided to me by O. N. Semenova from the syntactic card index of contemporary literary Estonian at the Institute of Language and Literature of the Estonian SSR: 1. Homme tuleb arst kutsuda “tomorrow one will have to call a doctor.” Õpetaja audis tüdrukula käsi kala ära puhastada “The teacher ordered the girl to clean the fish.” Talud said käsu sulane või tüdruk mõisa saata “The farms received an order to send a farmhand or a girl to the manor.” 2. Söömine lõpetatud mindi õm “having finished the meal, they went into the yard.” Värsked ajalehed läbi loetud “having read the latest newspapers...” Uksed uuesti hoolega lukku käänatud “Having carefully locked the door again...” 3. Eeva aidati veoautole “Eeva was helped onto the truck.” Nüüd lastud kohe poiss kuninga ette “now the boy was immediately admitted before the king.”

²⁶ V. I. Sobinnikova, Common and Dialect Features in the Language of Regional Written Documents of the 17th–Early 18th Centuries (Based on Voronezh Charters), “The Initial Stage in the Formation of the Russian National Language,” Leningrad, 1961, p. 214; The Simple Sentence in Voronezh Dialects, Voronezh, 1962, pp. 23, 192.

95

права; отдать грамотка на Воронеже; та вотчина продать; дать купчая; велено на Воронеже тюрьма делати (modern izba бялить; трава касить; K. Filatov had already cited in 1898: Он ковыль-травка рвал, as V. N. Sobinnikova pointed out). In the 1930s I went on an expedition with students from the Rostov Pedagogical Institute to the Don region, where we recorded: “глава причасать некуда”, “адезына вздеть”.

Even more important is the considerable variety of constructions with a nominative direct object and the variety of their syntactic functions.

The reasons given above show that it is unjustified to exclude masculine and neuter nouns on the grounds that they do not form constructions with a nominative direct object (at least in the older sources). Examples in which masculine nouns occur as nominative direct objects were already cited by F. Buslaev and A. Potebnya.

With the imperative form, a nominative direct object was noted by N. M. Karinsky in the Pskov Hexaemeron of 1374: “Cover me with a divine robe, send your precious mercy (host)”²⁷. The Paris “Dictionary of the Muscovites” of 1586 contains two entries: Appelez la chambrière! Possavy chelouza = chaya “Call the serving woman!” and Combien vault ung renard povr? Seto datty lychisa sorna? “How much is a poor fox worth? What should one give for a black fox?” (see B. A. Larin, “The Paris Dictionary of the Muscovites of 1586,” Riga, 1942, pp. 64 and 143). In Rybnikov’s recorded epic songs, I, 202, 401:

“Take as much of the gold treasury as is needed, and take for yourself a mighty army!”

For finite forms of the past and present tenses, the oldest example of a nominative direct object was cited by V. I. Borkovsky²⁸ and V. Georgieva: And now I have learned of your right, with whom I am—with Vitebsk (a charter from James, Bishop of Polotsk, to the people of Riga, c. 1300). From the Pskov Chronicle, V. Georgieva cited: They laid the foundation of the Church of the Nativity, built of stone, beyond the Stepa, and built a new bridge in Pskov²⁹. A. I. Sobolevsky’s Lectures gives³⁰: The princes and boyars gave them alms. From “Great Russian Folk Songs,” collected by A. I. Sobolevsky, VII, 206: “And they call him a friend, they pour him a bowl of drink… From records of the Seliger Expedition of 1936: стоплю баня (village of Igoshevo); съезла телега сена (village of Bereznik); она несёт чашка (village of Zalesye), дали нам нова телега (village of Moshonka).

The oldest examples with an adverbial participle occur in the Russian Justice: Закладаюче городня; куна взяти (according to the Synodal copy of 1282). А первое пагуба исплативше, а въ прочь князю поточити (according to the Ferapontov copy of the 16th century) (see Russian Justice, ed. by Acad. B. D. Grekov, I, Moscow–Leningrad, 1940, pp. 132, 257).

B. Unbegaun cited from a charter of 1491: Let them report, my lord, having plowed their fields, they are to be entirely at the disposal of the tsar, Menli Giray³¹.

²⁷ N. M. Karinsky, A Study of the Language of the Pskov Hexaemeron of 1374, p. 244.

²⁸ V. I. Borkovsky, Syntax of Old Russian Charters, p. 348.

²⁹ V. Georgieva, Materials and Studies in Russian Dialectology, vol. 3, Moscow–Leningrad, 1949, pp. 48–50.

³⁰ A. I. Sobolevsky, Lectures on the History of the Russian Language, 4th ed., St. Petersburg, 1907, p. 237.

³¹ B. Unbegaun, La langue russe au XVI siècle, Paris, 1935, p. 130.

96

In the Domostroi, Buslaev had already pointed out: soyami shirt.

From the Moscow charter of 1634 (according to the Old Russian Dictionary card file): Having taken a note from the voivode. From a seventeenth-century medical book (Shchukin collection): brew bran porridge, and apply it hot (to the patient’s leg). From the seventeenth-century “Instructions on How to Treat”: And, while planting cabbage, water it with the same water. From the records of the Seliger Expedition of 1936: A week having passed (village of Goristaya).

All these constructions have parallels in Finnish and Estonian (see above), as well as in Lithuanian and Latvian (which will be presented below).

We shall examine the construction “nominative of the complement with an infinitive” in Old Russian texts according to its functions.

A) We regard the most ancient function of the construction as expressing spontaneous necessity, inevitability, which in translation into modern language may be conveyed by the words it will come to pass, it will be necessary:

Brother, Prince Vladimir, here we must drink the honeyed cup (Zadonshchina); It is no longer for us, brothers, to be in our own land… and we must kiss the green grass (ibid.)³²; And if they redeem the land of our ancestral estates from us, then that land must be given to the village elder and all the Vlachs, and the deed of gift issued, and the monastery courtyards vacated (Acts of the Kholmogory and Ustyug Eparchies, II, 32; 1536–37); …in the morning, a dish of molasses and a small loaf each, and in the evening, porridge to eat (The Journey of Trifon Korobeinikov, 1594); Without money—water to drink (Collection of seventeenth-century proverbs, ed. P. K. Simoni); You have stripped me bare, tavern-keeper, guide to the wretches; only from shame can the southern side be recognized (“The Feast of the Tavern Signs,” manuscript of 1666, ed. V. P. Adrianova-Peretz). A little path must be laid out for me, a little road trodden for me (from a song recorded in the village of Berkovo, Belozersky District).

B) To the second group we assign the expression of recommendation, request, advice, or instruction, which can be conveyed in translating this construction by adding the word should, ought to: From the deed of gift to the Suzdal Spaso-Evfimiev Monastery by Pozharsky, 1587: And Archimandrite Levkey, together with the brethren, is to have our parents entered in the Synodikon, and memorial services sung for them and a mass served by the brotherhood; From the seventeenth-century Lubchanin herbal: Therefore this herb should be taken by those whose constitution is cold and dry; From the “History of the Seven Sages” (ed. F. Bulgakov, Society of Lovers of Ancient Writing, 1878): If a man’s wife dies, he is to take another; From Avvakum’s “How God Created Man”: (Adam) should have understood the devil’s wiles and strengthened his wife in keeping the commandment; From P. K. Simoni’s “The Bookbinder’s Manual”: A pit should be dug beneath the incense burner. From the seventeenth-century “Instructions on How to Treat”: To stop a nosebleed, dilute salt thoroughly in cold water and smear it on the head; From a fifteenth-century charter: And they divided the gardens between themselves. Semen is to build the middle partition… Grigory is to build the rear partition; From the bylina “Solovey Budimirovich” (ed. Lyatsky):

³² Воинские повести древней Руси, изд. В. П. Адриановой-Перетц, М.—Л., 1949, стр. 39—40, серия „Литературные памятники“.

7. Kalbotyros klausimai VI.

97

And for you, girl, it is not fitting to be thus; rather, stay at home and carry water.

From Paus’s Collection of Proverbs, early 18th century, fol. 12b. As the cup you pour for a friend, so shall you drink yourself.

C) The following group: an order to a subordinate, a state law. The additional meaning of the construction may be rendered by the word “must.” I shall cite only two or three examples (a great many are found in the Russian Justice, the Law Codes, and the Code of Tsar Alexei Mikhailovich): From “Instructions to Afanasy Pashkov on his appointment as voevoda in the land of Dauria (1655):”

And in all administrative matters concerning service people, merchants, and tradesmen, they are to be judged and their cases settled without delay, justly, according to the sovereign’s decree. From the interrogation of Vasily Bolotnikov in 1622: And he was sent alone to Kazan, and he was ordered to give up his wife and live in Kazan.

D) A special group comprises purposive constructions in which the modality of the infinitival predicate is specified, determined, or emphasized either by a relative pronoun or by the conjunction in order that (and, in later language, by the particle would). This group appears to be of later formation:

…and a whole bucket, in which to boil the wheat, 1643 (Materials for the History of Russian Icon Painting, ed. I. E. Zabelin, Moscow, 1850, pp. 6–7); “A fine piece of cloth was bought, with which to cover the icon,” 1703 (Archive of the Onega Monastery); “Magmed Amin, Tsar of Kazan, sent his man to you to arrange a marriage, so that he might take your daughter for himself,” 1491 (Collection of State Charters and Treaties, part V, p. 10); “Our sovereign sent word throughout many states that he, a great sinner, might find a bride for himself,” 1581 (Monuments of Diplomatic Relations between the Muscovite State and England, St. Petersburg, 1883, II, p. 6); “He volunteered to go to Vyatka with a treasonous letter, so that he might likewise stir up disorder in Vyatka,” 17th century (Acts of the Archaeographic Commission, II, p. 198); “We have stood surety for the peasant Efrem Ondrenov’s son, that he may plow the land and build a yard and new chambers” (Acts of the Muscovite State, vol. I, St. Petersburg, 1890).

E) In a number of texts, our construction occurs together with the words possible, possible; but even without these words, it can express the possibility of the action denoted, as can be seen in the fourth example: And to gather all the service and willing men together and discuss with them whether it might be possible to establish a field, and whether it might be close to the fort (Instructions to Afanasy Pashkov, 1655); “…and those surplus lands, with all their people… for 15 versts, and on those surplus lands and in those tracts it is possible to establish a sizable new settlement,” 1680 (Acts concerning the construction of towns in Siberia); How can such a thing be done: to take a wife from her living husband (Epic song, recorded by Academician Shakhmatov in the village of Gorka, Olonets). And for them, sire, the Greeks, to abandon the Orthodox Christian faith up to this time—it is no wonder, 17th century (Petition of the Solovetsky Monastery to Tsar Alexei Mikhailovich).

98

F) Adjacent to group E is also a group expressing the conditionality or contingency of the action expressed by our construction:

And this is set down as an everlasting disgrace: if a mistress give herself to a slave, 17th century (Kotoshikhin, On the Russian Realm); To brew beer is to lose the flask (Collection of Proverbs, 17th century, ed. Simoni); Lower the axe this far, and the head is chopped off (Rybnikov, Songs, I, 382).

All constructions without an infinitive, with a nominative before the words nadobě, nadobno, nado, do not belong to the subject of our study, although they are close in modal meaning to infinitival constructions. Constructions with nado(b) alongside an infinitive with a nominative direct object were mentioned above. In these, necessity is expressed pleonastically both by the infinitival construction and by the additional word nado(b). I doubt that verbless constructions with nado(b) can be derived from verbal ones as secondary forms “with the verb omitted.” The formula attested in the Synodal manuscript of the Russian Justice (1282): “A mother’s share is not due to the children” is perfectly clear without the verb “to take” or any similar verb, although it probably corresponded to fuller formulas containing an infinitive.

Summarizing the observations made, let us emphasize the features characteristic of all functional varieties of the construction “infinitive with a nominative direct object”:

1) expression of an action that is possible, desirable, necessary, or contingent, but is not actually taking place now or did not actually take place earlier;

2) the action’s more or less evident impersonal character: either it is not specified at all with respect to the agent, the actor, or only a possible category or class of actor is indicated, i.e. the acting person is indefinite;

3) although the predicative is overwhelmingly expressed by an infinitive, these constructions may also take imperative forms, participles, and even finite verb forms.

As we saw above, these are precisely the features that determine the choice of the nominative “passive” or “with accusative meaning” to express the direct object in Finnish and Estonian (and also in Turkic languages).

In Finnish, these ancient (and likewise residual) constructions found a new and fairly stable functional use. In the Russian literary language they fell out of use by the beginning of the 19th century, while in folk dialects they still survive sporadically, especially in the Russian North. Their displacement from the literary language resulted from the accumulation of more clearly differentiated syntactic (or lexical) means for expressing the modal and other meanings conveyed by these archaic constructions (a form of purposive clauses with the conjunction in order that and finite predicates was developed in place of infinitival constructions, and conditional clauses with the conjunction if; necessity was expressed by the words must, necessary, should, ought, obligatory). The indefiniteness of the object acquired clear lexical expression (some, any, some kind of, any kind of, etc.).

99

IV

Let us turn to an examination of the material from the Baltic languages.

In Lithuanian, especially in vernacular dialects, the following constructions are widespread:

1) the nominative of the direct object with an infinitive;

2) the nominative of the direct object with a participle.

The first syntactic construction can also occur independently (example: kaip riešutas perkrimsti “how to crack a nut” — Žemaitė³); it is usually introduced either by verbs: reikia “one must,” tūri “troubles, worries,” teko “had to,” tiko “happened” and the like, or by adverbs: sunku “difficult,” lengva “easy,” sveika “healthy,” gera “good” and the like. The construction is also found after (ne)galima “(im)possible,” (ne)laikas “(not) the time,” prieina “it is necessary” and the like.

Most linguists tend to interpret residual linguistic phenomena on the basis of the prevailing norm; this reflects the general pattern of assimilating relics, both among the common people and in scholarly circles. In this case³, the infinitival and participial constructions with the nominative case had to be understood according to the model of a simple sentence of nominative structure. The remarkable Lithuanian linguist J. Jablonskis showed ingenuity and found an elegant solution, but his conception lacks sufficient historical perspective³⁴. He interpreted these constructions as a variety of nominative structure supposedly characteristic of Lithuanian, in which the nominative case together with an infinitive or participle forms a “compound subject” (and in some cases a predicate). This may be accepted if the sentence is divided into only two “parts” (as Shakhmatov did): the subject and the predicate, with all related words included in each part and syntactic relations within the first and second parts disregarded. In the complex sentence: Ne laikas šiėnas džiovinti, kadà lietùs ima lyti “It is no time to dry hay when it starts to rain” (Leipalingis), the phrase šiėnas džiovinti “hay to dry” may be considered the subject, and ne laikas “not the time” the predicate. But this does not eliminate the internal syntactic relations within the construction šiėnas džiovinti, where šiėnas is still the object of the infinitive džiovinti “to dry.” These internal relations within the subject part are somewhat muted, but they are not eliminated or lost, just as the understanding of the nominative form in šiėnas has not disappeared, even though here it functions as an object, contrary to the norm of Indo-European languages. Therefore, if we speak not of the “subject part” as one half of the sentence, but of the subject in the narrow sense, we must recognize džiovinti as the subject, with the direct object šiėnas.

³³ Žemaitė, Raštai, V, Vilnius, p. 550.

³⁴ See in “Selected Works,” the reprint edition of Jablonskis’s grammatical works prepared by J. Palionis: J. Jablonskis, Rinktiniai raštai, vol. I, p. 161, § 197–198; p. 454–59, §§ 20 c, 21 e; p. 560–64, §§ 6–7.

100

On the other hand, I also find questionable the splitting of the construction Nominativus cum infinitivo, which J. Jablonskis allows in his analysis of the second group of examples containing it (§ 21, e). In the sentence „Jai namai matyti“ (“the house is already visible”), he separates matyti—as a predicate now—and namai becomes the subject, chiefly because it is in the nominative case. But the Lithuanian infinitive does not have the passive meaning “to be visible,” which alone could justify treating namai as the subject rather than the object. This impersonal sentence has no subject, and namai matyti can be regarded as a single (in Shakhmatov’s terms, “indivisible”) predicate, although the function of an object in the word namai is fully discernible in it. This is even more evident in the following example by Jablonskis from the same group (a popular saying recorded in Žagarė): Berigų kiaulė: nei burną prausti, nei kojos aūti, nei poteriai kalbėti “It’s good to be a pig: neither to wash your mouth, nor to put on shoes, nor to say prayers”; more literally: “neither to wash your mouth, nor to put shoes on your feet, nor to say prayers.” The Russian translation, using one word for each Nom. c. Inf. phrase, conveys the unity of these phrases well. I would therefore consider the analysis of this sentence more accurate if Jablonskis had identified each of these phrases as a predicate in its entirety, rather than only the infinitives (praūsti, aūti, kalbėti). What, then, are the nominatives burną, kojos, poteriai? Subjects? But the verbs praūsti, aūti, kalbėti are clearly transitive in these combinations; consequently, these nominatives can only be defined as “direct objects.”

Thus, the nominative case of the direct object is used with infinitives and converbs in Lithuanian dialects (especially in Dzūkian and eastern dialects, and in part of Samogitian), and is therefore also reflected in the language of writers (in Krėvė, Žemaitė, and others). I shall give examples with various introductory components:

1. Nominative of the direct object with an infinitive: Dabar man rūpe, iš kur rągūtės gaut (pažyčyti) (from B. A.’s card file for the Dictionary of the Lithuanian Language) “Now I’m wondering where to get a sled?”; Nāšlei pačiai reik(ia) rūpinti visi namū reikalai (from the same source) “The widow herself has to take care of all the household affairs”; Nutiko prieš lietų šienas pasigauti (from the same source) “It so happened that the hay was brought in (carted away) before the rain”; Sunkiausi darbeliai reiks man dirbt (from the same source) “I’ll have to do the hardest jobs”; Vėlus metas, reikia vaikai guldyti (from the same source) “It’s late; the children must be put to bed”; Mokydavo kukorių, kaip geriau valgymą virt (from A. Baranauskas’s notes, published by F. Specht) “He taught the cook how best to prepare a meal”; Tai labai gera miegól miegėlys (from the same source) “It is very good to sleep a sweet sleep”; Rugėliai pjauti labai nusibosta (from the same source) “It gets very tiresome to reap rye”; Man teko ir mergáuti, ir tinginys vyras gauti (from J. Jablonskis’s notes) “I had to spend time as a maiden and end up with a lazy husband”; Baisi yra ir paminėt tie žodže (from A. Baranauskas’s notes) “It is frightening even to recall those words”; Nāms pastatīt — ne kepūre pakēlt (recorded by J. Šukys) “Building a house is no mere taking off of a cap”; Man gyvuliai sutvarkīt (from the same source) “I still have to get the livestock taken care of.”

101

9

All these are impersonal sentences with clearly expressed generality and modality. The last examples show that in popular speech the Nom. c. Inf. construction is used without introductory words, as a completely independent construction.

2. Nominative of the direct object with converbs. (All examples are from the card file of the Dialectological Atlas of the Academy of Sciences of the Lithuanian SSR.) Kur gāvus tas kālis “If only one could get that fur somewhere”; Linai rā.ūnunt reik’a šī.ltai apsviīkte “When harvesting flax, one must dress warmly”; Kad geri rōtai gōvus “If only one could get a good cart”; Gerai bu.tū stipras darbinikas pasišūndžus “It would be good to hire a strong farmhand”; Kur ča rō.dus gēras mēistras “Where could one find a good master here?,” that is, “Where can one find a good master?”; Bu.t g’erai tāks arkli.s gō.vus “It would be good to get a horse like that”; Vyt kad gāvus nemāžas šmōtas duōnos “If only one could get a decent-sized piece of bread”; Kat gāvus nemāžas p’emtū “If only one could get a herdsman—not a little boy”; Kožnas dārbas dirbunt mōksla rāika “Every job requires skill when doing it”; Kat_kaip nē būt miežai pasējus “If only I had somehow sown the barley”; Tai kās_če man padār’us “And what would I have done?”

This construction, too, as the recorded examples show, has an impersonal, generalizing, and in a certain sense irrealis meaning in popular speech—that is, it is opposed to the assertion of an event that has occurred or is occurring. The marked variety of the nominative (the S-nominative) became established in Lithuanian as the only possible form, but the functions of the unmarked nominative have not disappeared without a trace: they survive residually to this day in constructions where the nominative of the direct object occurs with infinitives and converbs, just as the functions of neuter pronouns and predicative forms of adjectives persist despite the disappearance of the neuter form. It seems that no one now disputes the proposition that a grammatical category sometimes arises before a clearly differentiated form for expressing it has developed in a language. As we see in this case, a grammatical category sometimes also dies out neither immediately nor soon after the form expressing it has been lost from the language.

V

In Latvian, remnants of the nominative of the direct object with an infinitive occur with a governing verb in the dative, or after adverbs with modal meaning, as well as with the indeclinable (i.e. impersonal) form of the past passive participle (examples below).

Academician J. M. Endzelin, in his extensive comparative grammar of Latvian³⁵, describes this “anomaly” as follows: “When

³⁵ J. Endzelin, Lettische Grammatik, Heidelberg, 1923, § 392, S. 409—410 и то же без всяких изменений в латышском переводе: J. Endzelīns, Latviešu valodas gramatika, Rīgā, 1951, § 392, lpp. 553—554. Именительным подлежащего, а не дополнения считают номинатив в этих конструкциях и авторы „Musdienu latviešu literārās valodas gramatika“, I, Rīga, 1959, § 452, 763; II, Rīga, 1962, § 428, 10.

102

When the infinitive was still a living nominal case form, the object of the action expressed by the infinitive could stand in the nominative—as the grammatical subject. But even after the infinitive became a purely verbal form, its complement (object) nevertheless remains in the nominative, as the grammatical subject, in cases where the infinitive depends on a modal adverb: piōsiniuojīs akmins grūti kustināt “a moss-covered stone is difficult to move” (Treiland, Materials on the Ethnography of the Latvian Tribe, Moscow, 1881). Cf. Lithuanian akmuō lēngva (ratuōs) indēti ir sunkiāsias “Viltis,” 1908, no. 77, and examples in Jablonskis’s Lithuanian Syntax, pp. 19 and 23; similarly (at least in dialects), also in cases where the infinitive depends on a debitive or an imperative: jāiet rija kult BW 946 “one must thresh the barn”;...vaiga issukāt galva (Kaletos) “one must comb the head”; zirgs vajadzēs mazgāt (Bezzenberger) “the horse will have to be bathed”; vajdzēja celtīs gails kaut “one had to get up and slaughter the rooster”; gribis koads dorbs padarīt (Nereta) “one wants to get every task done”; kungam ēst tej maizite (in a folk song) “the master must eat this bread.” Cf. Lithuanian: reikia kultuvē padarýti “one must make a flail,” and Russian dialectal: надо душа спасти (Sobolevsky, Lectures on the History of the Russian Language, 4th ed., pp. 197 ff.), or “you will lose your wild head” (V. Miller, Essays on Russian Folk Literature, p. 86).

We also find a direct object in the nominative in constructions of the type: viņam duots zirni ēst “he was given peas to eat” (Mühlenbach, Latvian Grammar, § 352).

As we can see, Potebnya’s conception recurs here: at first the nominative expressed the subject; now, even if it has not been replaced by the accusative, it expresses the direct object.

Dialect material was collected for the Atlas of the Latvian Language using a questionnaire that included almost all the examples cited above by Academician J. Endzelīns³⁶. The alternation in the dialects (as, indeed, in the literary language) between nominative and accusative forms in infinitival constructions following a debitive proves beyond any doubt that, in the contemporary linguistic consciousness of Latvian speakers, this is an object, not a subject. Otherwise, consistency would require introducing into grammar the category of “subject accusative,” which no one has done or does. Yet it is precisely this inconsistency (contrary to Academician Endzelīns’s views) that the authors of the Grammar of the Contemporary Latvian Literary Language have adopted (see above). The distribution of responses to question no. 270 in the Atlas questionnaire: jāsāk pļaut siens — or sienu? “one must start mowing hay” (nominative or accusative?) is as follows: in 109 localities, the nominative; in 6 localities, alternation between nominative and accusative; and in 31 localities, only the accusative (the latter in most Livonian dialects and occasionally in Latgale).

³⁶ Latviešu valodas dialektoloģijas Atlanta materiālu vākšanas Programa, Rīgā, Latvijas PSR Zinātņu Akadēmijas izdevniecība, 1954, NN 270, 272, 273, 274, 276, 277, 280, 281. Selections from the Atlas materials were kindly sent to me by S. Rage and D. Zemzare, staff members of the Institute of Language and Literature of the Academy of Sciences of the Latvian SSR, to whom I express my deep gratitude.

103

With a dependent infinitive (question no. 272), the nominative is much less frequent: jāiet plaut siens is attested in only 73 localities; in 29 localities there is sienu (accusative), in 35 localities siena (genitive), and in 11 localities alternation between nominative and genitive. The geographical distribution of the responses is essentially no different from that described above.

The nominative direct object occurs somewhat less frequently in phrases (question no. 276): (ne) vajag malka skaldīt “(one) must chop firewood”; nav brīv ne vārdiņš teikt “one may not say even a word”; gribas kāds darbs padarīt “one wants to get every task done”; and also in constructions such as: vajadzēja celties gailis kaut “one had to get up and slaughter the rooster”; tagad vajag spēks “now strength is needed”; vajag nākt piens dzert “one must come to drink milk.”

The nominative is used much more frequently in folk dialects with an infinitive dependent on an adverb: pūsbūnojs akmens grūti kustināt “a moss-covered stone is difficult to move”; aka nav viegli rakt “a well is not easy to dig” (question no. 280).

But the accusative predominates in all responses to questions 272, 276, 277, and 280. A more precise numerical estimate cannot yet be given here because of the lack of material³⁷.

As we can see, Latvian, even in folk dialects, has preserved fewer traces of ancient constructions with a nominative direct object, although it is rightly considered the most Finnicized language. Unification (the elimination of these surviving constructions under the influence of the dominant expression of the object in the accusative) has been most radical in the Livonian dialects, that is, clearly in the newer, mixed dialects; most remnants of this archaism are found in the central and eastern dialects, which are relatively more “pure” and faithful to the Indo-European tradition. The most persistent construction proved to be Nominativus cum Infinitivo after the debitive (this construction is optional in the literary language and common in most folk dialects). The other constructions have noticeably fallen out of use (even since the time of Mühlenbach’s and Endzelīns’s observations).

All the Latvian constructions listed and illustrated, like the analogous Russian and Lithuanian ones, are characterized by impersonal (subjectless) usage, a generic object, and the more or less clearly expressed irreality of an action that has not yet taken place.

Conclusions.

1. It is perfectly clear that a certain degree of grammatical distinctiveness in each language in no way detracts from the analogous nature of Finnish, Estonian, Russian, Latvian, and Lithuanian constructions with a nominative direct object.

³⁷ Here are some more examples recorded for the Atlas with a nominative direct object: Nav viegli maisi nest “It is not easy to carry sacks” (Aurī, near Dobele); Šitais grūti izrēķināt “This is difficult to work out” (Vilzeme, near Limbaži); Pelašķi ir grūti iezert “Yarrow infusion is difficult to drink” (Braslava, near Limbaži); Maiss ir smagi celt “The sack is difficult to lift” (near Rūjiena) Abuliņš ir grūti plaut “Clover is difficult to mow” (there too); Apesta maize grūti pelnēt “It is difficult to earn back the bread that has been eaten” (near Lejasciems).

104

2. After comparing them all, there is no doubt that these constructions included nouns of any gender as their object (the unprovable masculine nominative forms in Russian are quite evident in Lithuanian and Latvian), and that imperative forms and converbs were originally just as normal in constructions with a nominative direct object as the infinitive. These verbal formations are known in all five languages under comparison. The combination of a nominative direct object with finite verb forms is found only in Russian, Finnish, and Estonian. From this it may be concluded that the loss of the distinction between the “active and passive” nominatives and the displacement of the passive form by the active (S-form) began specifically with finite verb forms.

3. Both the geographical distribution and the history of these constructions argue against the supposition that the “anomalous” constructions were borrowed. The survival of remnants of a pre-nominative sentence structure, corresponding to a different formation of the verb, a different relationship among nominal inflection forms, and different case functions, can surprise us only by its relative abundance and stability—explained by the adaptation of these constructions to a new formation in the development of the sentence, to a new system of nominal and verbal inflection and word formation.

4. The distribution of a series of similar constructions across the areas of two Baltic, two Finnic, and (formerly) three East Slavic languages attests to the great antiquity of this phenomenon not only a) because they are rudimentary in character and are clearly dying out in most of these languages; b) because the related Baltic and Slavic languages could not have constituted a unity for at least two thousand years; but also because c) the Slavs and Balts came into contact with the Finnic languages no later than those same two thousand years ago. d) It would be implausible to suppose that constructions with a nominative direct object developed independently and in parallel in each of these languages: all the constructions considered are so closely alike, and yet in all the languages mentioned they fall outside the prevailing norm as “anomalies” and cannot be satisfactorily interpreted within a nominative system as the result of an internal, independent evolution of the simple sentence in each language, or as some variety thereof.

The most plausible explanation is that these rudimentary constructions go back to the language of a common substrate, the tribal dialects of the Baltic region and northern Europe, with which both the Finnic tribes and the first waves of Baltic settlers from the southeast became intermixed as they moved into their present areas of settlement, followed finally by the vanguard of the Slavs advancing from the south³⁸.

³⁸ The correspondences for this construction in Turkic languages noted by Prof. S. E. Malov should be traced and studied by Turkologists; these correspondences may apparently be included among the features characterizing the “Samoyedic” linguistic system. If this is confirmed, the antiquity and survival of the construction we are studying in Slavic, Baltic, and modern Finnic languages will become even more evident.

105

Only with the help of archaeologists can we proceed further in substantiating this hypothesis, or reject it if the history of the material culture of the Balto-Slavic-Finnic areas does not accord with it.

In conclusion, it should be recalled that there are a number of observations concerning other linguistic anomalies within the same group of languages which, when studied further, may yield isoglosses converging into a bundle and thus may define more precisely the contours of the supposed substrate. The fixation of stress on the first syllable is known, in addition to the Baltic-Finnic languages, in Latvian and in Lithuanian dialects bordering the Latvian SSR³⁹.

The Latvian debitive, the Lithuanian permissive, and the optative represent departures from Indo-European verb formation, with Finnish and Estonian parallels. The numerous Old Lithuanian forms of variants of the local case (inessive, adessive, illative, allative) also clearly approach the Finnic-Estonian declensional system in this respect and diverge from Indo-European models. A number of lexical and phraseological correspondences—for example, those noted by Academician J. Endzelin⁴⁰—may be considerably increased; this is one of the successive tasks facing lexicologists among Slavic, Baltic, and Finnic scholars. At the present stage in the development of linguistics, research into grammatical and lexical-phraseological relations among the Finnish, Baltic, and Slavic languages, successfully begun almost a century ago by the Danish linguist V. Thomsen, should be resumed.⁴¹ The rapprochement and mutual assistance between Slavic linguists and Baltic and Finno-Ugric scholars, which has noticeably intensified in recent decades, is a guarantee and prerequisite for our further success in uncovering the most ancient contacts between the tribes and languages of the Baltic region and the Russian North.

ON A COMMON SLAVONIC, BALTIC AND FINNISH ISOGLOSS

B. LARIN

Summary

Constructions of Infinitive and Verbal adverbs with the Nominative of Direct Object known in the areas of two Finnish, two Baltic and three East Slavonic languages go back to the languages of the Baltic ant North Russian substratum.

The antiquity of such constructions is indicated by their rudimentarity, the discrepancy in Nominative system of the sentence and correlation with archaic passive and ergative system.

³⁹ This may also be compared with the unsystematic shift of old stress toward the beginning of the word in northern Russian dialects (zasukha, dobycha, za lesom, zábolʹ, golomya, etc.).

⁴⁰ I. M. Endzelin, On Latvian-Finnish linguistic relations, collection of articles “In Memory of Academician L. V. Shcherba,” Leningrad, 1951, pp. 292–304.

⁴¹ V. Thomsen, Begøninger mellem de finske og de baltiske (litauisk-lettiske) Sprog, København, 1890.

106

The tradition of such constructions may be attributed to the beginning of our era, because at that time and not later the contacts between Baltic tribes, Slavs and Western Finns were established as well as their geographical distribution over the territories they have occupied since the first centuries A.D.

ON A SLAVIC-BALTIC-FINNISH ISOGLOSS

B. LARINAS

Summary

1. The infinitive construction with a nominative direct object, which has survived to this day in the northern Russian dialects, has long attracted the attention of researchers. However, as long as all attempts to study it historically and genetically remained within the bounds of a single language and did not escape more or less accidental selection of facts, it continued to be the subject of confused and fruitless disputes. The shortcomings of a synchronic approach to linguistic phenomena that do not conform to prevailing language norms are especially clearly revealed here.

2. This construction can be explained more convincingly only by studying its distribution and use both chronologically and territorially (from the perspective of linguistic geography).

3. In the Russian literary language, this construction has been observed from the 11th–12th centuries to the end of the 18th century. It occurs most often in northern dialects, but is occasionally found in others as well, including southern dialects.

4. The nature and origin of this construction can be determined even more precisely by comparing it with analogous syntactic patterns in the Baltic and Baltic Finnic languages.

5. Data from the Finnic languages show that this construction is connected with an ancient function of the nominative, characteristic of the period when the nominative had not yet been differentiated from the accusative but was already opposed to the instrumental, purposive, and locative cases. This system of case relations, better preserved in the Baltic Finnic languages, helps explain the use of an object in the nominative not only with the infinitive but also with personal verb forms. Such constructions have been observed in old Russian texts and in present-day dialects.

6. In this way, we identify an archaic isogloss running through the territory of the Slavic, Baltic, and Baltic Finnic dialects. It encourages us to seek other common features of the “Baltic linguistic area,” and further traces of contacts between Slavic, Finnic, and Baltic tribes that lasted two (or more) millennia.