Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita

ARTŪRAS JUDŽENTIS Vilnius Academy of Arts THE SYNTACTIC RELATION BETWEEN SUBJECT AND PREDICATE: CHANGING CONCEPTIONS KEYWORDS: syntactic relations, the syntactic relation between subject and predicate, mutual connection. ABSTRACT The article surveys changes in conceptions of the syntactic relation between subject and predicate in Lithuanian linguistics from around the middle of the twentieth century to the present day. It presents an overview of the more significant grammatical works on this question. It examines how this relation can be explained using the theoretical tools of modern grammar. INTRODUCTION This article aims to draw attention to an important theoretical and practical problem in Lithuanian syntax, provoke debate, and contribute to finding solutions. It therefore briefly reviews and evaluates the more significant works on syntax that have had a greater impact on the development of Lithuanian linguistics and appeared since the 1960s, in which the syntactic relation between subject and predicate is examined. This topic is examined in greatest detail in Ramunė Vaskelaitė’s doctoral dissertation Models of Syntactic Relations in Lithuanian Syntax (2009). This article pays less attention to various terminological issues concerning syntactic relations, inconsistencies in their classification and their causes; instead, it highlights general differences in the theoretical foundations of description, offers a possible periodization of conceptions, draws a clear distinction between traditional and newer explanations of the nature of this relation, and extends the review to the most recent work in this field. It may be said that the article completes and sharpens BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 2 doi.org/10.35321/bkalba.2020.93.11 some notes and conclusions from the dissertation. The article’s third section presents a brief account of this relation based on the new conception and sets out the arguments supporting it. The article employs a methodology characteristic of review studies: primary sources—the more important works on Lithuanian syntax—are examined, compared, and assessed from historical and theoretical perspectives. Their study leads to general conclusions about changes in the conception of the syntactic relation between subject and predicate in Lithuanian linguistics. 1. THE THREE-WAY CLASSIFICATION OF SYNTACTIC RELATIONS In works on Lithuanian syntax (e.g., DLKG), the syntactic relation between subject and predicate is distinguished at the highest level of the classification of syntactic relations, alongside subordination and coordination. The basis for distinguishing this type of relation was probably the fact that it connects the so-called principal parts of the sentence—the subject and the predicate. Because of the concept of the predicative centre of the sentence that they express, a special syntactic relation was discerned between them at the syntactic level: mutual dependence expressed by agreement, or mutual association. However, this conception of the relation did not emerge all at once; it was gradually refined. It is important to emphasize that the first attempts to distinguish a third type of syntactic relation alongside coordination and subordination arose not through the natural development of the study of Lithuanian, but only after independence and the freedom to think had been lost. The dissemination of Lithuanian linguistic traditions in the Soviet Union was impeded, ties with émigré and Western scholars were severed, while everything originating in Russian scholarship was encouraged and supported. Under such conditions, the position in Lithuanian syntax that two types of syntactic relations should be distinguished—subordinating and coordinating relations (Dambrauskas 1939: 84)—was displaced by the teaching of three types of syntactic relations. Gradually, in Lithuanian linguistics, the conception of BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 3 doi.org/10.35321/bkalba.2020.93.11 the third type—the predicative relation (later given a more explicitly syntactic name)—became established. 1.1. The subject–predicate relation as a semantic (and syntactic) phenomenon Some time after the Second World War, in Lithuanian works on syntax from the 1960s and the early 1970s, the predicative relation was understood less as a syntactic and more as a semantic phenomenon—a semantic relation between words. The name given to this relation also reflects this understanding. This is evident from the works of Bronius Kalinauskas and Jonas Balkevičius. 1.1.1. In B. Kalinauskas’s works, the predicative relation is understood as a semantic relation—it is grouped together with object, attributive, and adverbial relations. As Ramunė Vaskelaitė notes in her dissertation Models of Syntactic Relations in Lithuanian Syntax, at that time “the conception of syntactic relations was influenced by linguistic ideas put forward by so-called Soviet specialists in syntax” (Vaskelaitė 2009: 13), so this term and its conception were most likely also borrowed from Russian linguistics. That it was still new to Lithuanian linguistics can be seen from B. Kalinauskas’s article Words and Word Combinations in the Sentence (Kalinauskas 1963). Semantic relations between words are conflated there with syntactic ones: “In a sentence, there are various syntactic relations between words, namely: predicative (between the subject and predicate), which form the logical basis of the sentence; object (between the predicate and objects); attributive (between modifiers and the words they modify); and adverbial (between the predicate and various adverbials)” (Kalinauskas 1963: 3). In the teaching resource The Syntax of Lithuanian Word Combinations (Kalinauskas 1972), published almost a decade later, predicative syntactic relations are not mentioned; however, when word combinations are classified according to their dependent component, predicative, attributive, object, and relational (or adverbial) combinations are listed. B. Kalinauskas explains the importance of identifying semantic relations in a sentence as follows: “Dependent components of word combinations, attached to their main components, enable us to express certain relations between real-world phenomena or objects, certain judgments and generalizations, since we can make judgments about reality only by comparing phenomena with one another. This is why it is very important to identify the semantic relations between the main and dependent components of a word combination” (Kalinauskas 1972: 15–16). BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351-7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 4 doi.org/10.35321/bkalba.2020.93.11 Thus, Kalinauskas regards predicative relations as a semantic connection between the words that make up combinations. The words linked by these relations in a sentence are very important: “These word combinations are the most important in the sentence. They form the centre of the most important communicative unit of language—the sentence” (Kalinauskas 1972: 16). Kalinauskas also identifies what he considers the principal syntactic (termed grammatical) feature of predicative word combinations: “Their main grammatical feature is that both members of such a combination can equally be either the main or the dependent member, depending on which component we ask the question from.” This feature of theirs, picked up by the school, is not mentioned in later, more general works on syntax. Kaip rašo R. Vaskelaitė (2009: 58), vienas pirmųjų veiksnio ir tarinio ryšį atskirai nuo prijungiamųjų ryšių aptarė Jonas Balkevičius. 1.1.2. Although in J. Balkevičius’s The Syntax of Contemporary Lithuanian (1963), predication is understood as a semantic (logical) and syntactic (grammatical) property of the sentence, these two levels do not seem to be sufficiently distinguished from one another: “The expression of affirmation, negation, or a question is an essential, the most important feature of a sentence, called p r e d i c a t i v i t y. The predicativity of a sentence is conditioned by the logical and grammatical relations between two parts of the sentence, one of which, namely the s u b j e c t, denotes the real-world phenomenon to which the affirmation, negation, or question raised in the sentence pertains, while the other—the p r e d i c a t e—reveals the content of the affirmation, negation, or question <...>.” “In syntax, the logical and grammatical relations between the subject and predicate are called p r e d i c a t i v e r e l a t i o n s, while the subject and predicate are called the p r e d i c a t i v e c e n t r e of the sentence” (Balkevičius 1963: 17). BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351-7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 5 doi.org/10.35321/bkalba.2020.93.11 The predicative relation is expressed by agreement: “predicative relations <...>, whose grammatical expression is the agreement of the main parts of the sentence <...>” (Balkevičius 1963: 26). However, as can be seen from the table on page 41, Grammatical Relations between Words in a Sentence, this agreement is distinguished from agreement as a means of expressing a subordinating relation. Later, in works on Lithuanian syntax, the subject–predicate relation came to be understood and defined primarily as a syntactic phenomenon, sometimes with the qualification that it conveys semantic predicative relations. 1.2. The relationship between subject and predicate as a syntactic phenomenon In the 1970s, when the concept of a reciprocal syntactic relationship linking the predicative center of the sentence had fallen out of favor in Russian linguistics itself (for example, it was no longer adhered to in the academic Russian grammar edited by Natalia Shvedova and published in 1980), the predicative relationship, now understood primarily as a syntactic phenomenon, became familiar and established in Lithuanian linguistics. Efforts were made to find a Lithuanian term conveying this new understanding. In Vytautas Sirtautas’s dissertation (1969), the academic Grammar of the Lithuanian Language (1976), and the syntax textbook by V. Sirtautas and Česys Grenda (1988), it is still called predicative (or predicational); however, in Vitas Labutis’s Problems of Word Combinations (1976), on the basis of the syntactic nature of the relationship, it is proposed that it be called mutually conditioning. Labutis later replaced this designation with the term two-way interface. Vytautas Ambrazas, in the Grammar of Contemporary Lithuanian (1994), chose the term reciprocal interface. 1.2.1. In Vytautas Sirtautas’s dissertation The Verbal Predicate in Contemporary Lithuanian (1969; published only in 2019), relatively little space is devoted to discussing the syntactic relationship between subject and predicate. Moreover, it is primarily the expression of this syntactic relationship that is examined—various cases of agreement and non-agreement, or lack of concord (Chapter V, On the Agreement of the Verbal Predicate). To name the relationship between the main parts of the sentence itself, V. Sirtautas, following L. Barkhudarov, chose BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 6 doi.org/10.35321/bkalba.2020.93.11 predication, clearly distinguishing it from sentence predicativity (or predicative force): “Predication does not coincide at all with predicativity. If this term were to become established, it would be convenient (because of its brevity, among other things) for describing subject–predicate relations” (Sirtautas 2019: 68). He contrasts the predicative syntactic relationship with subordination and coordination. It is supposedly a reciprocal relationship of “mutual dependence.” V. Sirtautas understands agreement as a means of expressing the syntactic relationship between subject and predicate: “agreement is an excellent manifestation of the relationship between subject and predicate, but it is also clear that agreement is not necessarily required for such relations to be realized” (Sirtautas 2019: 325); “Although agreement is not a necessary means of realizing the expression of their [subject and predicate] relationship (cf. I run), nevertheless (in cases of agreement) it is a clear manifestation of the correlativity of the main parts of the sentence and their mutual dependence” (Sirtautas 2019: 327). 1.2.2. Grammar of the Lithuanian Language In the third volume, devoted to syntax and published in 1976, the syntactic relationship linking the components of a word combination is not discussed separately; instead, three types of combinations are distinguished solely by the nature of the relationship: subordinate, predicative, and coordinate (LKG 1976: 9). The components of combinations formed by the predicative relationship “cannot exist one without the other”; these combinations “always form sentences,” apparently because they are therefore simply equated with sentences and are not discussed separately in the chapter on word combinations. We learn more about the predicative relationship in the chapter on simple two-member sentences, where the concept of the main sentence part—the subject—is discussed (LKG 1976: 279–280). The definition itself states that “[t]he subject interacts directly with the predicate”; these sentence parts “each require the other” (LKG 1976: 279). The grammar draws attention to the fact that the meaning (or, more precisely, the role) of a word functioning as the subject is understood differently depending on how the word functioning as the predicate is expressed: “The subject is characterized by the predicate. Usually the subject denotes the agent of an action, but it can also denote the object of a property or action. The meaning of the subject depends in part on the forms of the predicate.” BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 7 doi.org/10.35321/bkalba.2020.93.11 Thus, the active-voice verb forms that make up the predicate show that the entity denoted by the subject is the agent of the action <...>, whereas passive-voice forms show that the entity denoted by the subject receives the action from elsewhere and is its patient <...>. “ The grammar states that ‘[T]he forms making up the subject and predicate are often congruent with one another.’ When congruence is understood as a type of syntactic relation, two cases are identified. In sentences with a simple predicate, ‘[T]he nominative subject and the first- and second-person verb forms making up the predicate agree in number <...>. ’ On the other hand, ‘[T]he nominative subject is related to the verb forms making up the predicate with respect to the category of person as well. When the subject is the nominative form of a first-person pronoun, the predicate is a first-person verb form (Aš einu; Mes einame); when the subject is the nominative form of a second-person pronoun, the predicate is a second-person verb form (Tu eini; Jūs einate); and when the subject is the nominative form of a noun, a third-person pronoun, or any pronominal noun, the predicate is a third-person verb form (Tėvas, jis, tėvai, jie, kažkas eina).’ In sentences with a compound predicate, ‘the nominative subject and the adjectival or adjective-like form making up the predicate agree in gender, number, and case <...>’ (LKG 1976: 280). This concerns agreement of the predicative (or predicate attribute) form with the subject. The use of the copula in a compound predicate with the subject in terms of person and number is not discussed. As can be seen, this grammar devotes little attention to syntactic relations. Its third type is still called by a term from the domain of semantics: the predicative relation. Its distinctive feature is grounded in the mutual dependence expressed by congruence. 1.2.3. In V. Labutis’s Žodžių junginių problemų (1976), the chapter “Word Relations by Character and by Direction” identifies the basis for classifying word relations: “By the character of a word relation we mean how and to what extent the elements linked to one another syntactically affect one another. The character of a relation depends on whether the units being linked are syntactically unequal or BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 8 doi.org/10.35321/bkalba.2020.93.11 lygiaverčiai. Taip pat čia svarbu ryšio kryptis“ (Labutis 1976: 38). Pagal šiuos du požymius jo skiriamos trys žodžių ryšių atmainos: prijungimas, sujungimas ir tarpusavio sąlygojimas. Pastarojo ryšio pobūdis nusakomas taip: „Savitas to ryšio ir sintaksinis charakteris: čia jungiami du sakinio dalių lygmenyje lygiaverčiai vienetai (panašiai kaip sujungimo atveju), bet tie vienetai, kaip veiksnys ir tarinys, vienas kitą sąlygoja (artima prijungimui), ryšys čia yra abipusis <...>. Tas ryšys gali būti vadinamas t a r p u s a v i o s ą l y g o j i m u (plg. Burlakova 1975, 21–22)“ (Labutis 1976: 40–41). Tačiau kaip galima V. Labutis (bene vienintelis tradicinės krypties lietuvių kalbos sintaksės darbuose) nurodo ir kitokį šio sintaksinio žodžių ryšio aiškinimo būdą: „Semantinės krypties sintaksės tyrinėtojai, griežtai žiūrėdami žodžių ryšių lygmens bei santykio su sakinio dalimis, ryšį tarp subjekto vardininko ir veiksmažodžio laiko prijungimu, o jo reiškimo būdą – valdymu <...>“ (Labutis 1976: 41). In addition, Labutis clearly distinguishes the syntactic nature of a word relation from its expression (or marking), which he describes in the chapter “Ways of Expressing Word Relations.” He distinguishes four types of expression of word relations: congruence, coordination of forms, government, and adjunction; the second is characteristic of the mutual-relation type: “In terms of expression, we called it [i.e. the relation of mutual conditioning] coordination of forms” (Labutis 1976: 40). “Coordination of word forms is the possible (but not obligatory) mutual adjustment of forms of words that have independent categories” (Labutis 1976: 31). 1.2.4. In the textbook by Vytautas Sirtautas and Česys Grenda, Lietuvių kalbos sintaksė (1988) three types of grammatical word relation are distinguished: predicative, coordinate, and subordinate. “A predicative relation is a relation between the subject (in the nominative) and the predicate, most often its finite form <...>” (Sirtautas, Grenda 1988: 37). The syntactic relations between subject and predicate are discussed in the section on sentence constituents devoted to them. “The subject is connected with the predicate by predicative relations. This connection is obligatory. If there are several homogeneous subjects and several homogeneous predicates, each subject is necessarily connected with each predicate <...>. This is a very important feature of the subject and BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351-7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 9 doi.org/10.35321/bkalba.2020.93.11 a feature of the predicate relations that distinguishes them from relations between other sentence parts. <...>. Another feature emerges from the relation between coordinated subjects and the predicate, e.g.: Berries have been bought. Mushrooms and berries have been bought. Mushrooms, berries, and everything else have been bought. This clearly shows that the components of the subject determine the forms of the predicate (and which gender form is dominant). Thus the predicate is directly dependent on the subject. Here we should also recall the earlier correlation between subject and predicate relations (Many factories had been profitable / Many plants had been profitable), which makes the mutual relation between the main parts of the sentence unmistakably clear” (Sirtautas, Grenda 1988: 85). Among other things, the following is noted almost in passing: “It is inaccurate to regard the subject merely as a dependent (or ‘independent’) acceptor of the predicate; moreover, the notion of a ‘left-hand’ accent is even vaguer than that of the subject” (Sirtautas, Grenda 1988: 85–86). This is a reference to valency theory and its various forms. The first remark refers to the view that a noun functioning as the subject in a sentence is dependent on the verb functioning as the predicate. As can be seen, the textbook authors consider this view inaccurate. The authors clearly distinguish syntactic relations from their morphosyntactic expression. This is shown by the following remark: “Sometimes the subject–predicate relations are unnecessarily equated with agreement, which, of course, is also one of the indicators of those relations” (Sirtautas, Grenda 1988: 86). 1.2.5. In the grammar edited by Vytautas Ambrazas, Grammar of Contemporary Lithuanian, whose first edition appeared in 1994, types of syntactic relations are described, beginning with mutual connection. “A syntactic relation between units of sentence structure that condition one another is called mutual connection. The word form through which one unit is incorporated into a sentence is, in turn, grammatically dependent on the latter. Thus, the relation of mutual connection is reciprocal <...>. In typical cases it is expressed through the correlation of word forms (mutual agreement)” (DLKG 1994: 478). The verb’s relation with the nominative case of a noun or pronoun is identified as the most salient instance of this syntactic relation, and it is argued that the nominative depends on the verb (in a sentence formed with an impersonal verb, the subject nominative BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 10 doi.org/10.35321/bkalba.2020.93.11 cannot be present), while the person and number of the verb are selected in turn with regard to the form of the noun (equated with the third person) or pronoun (DLKG 1994: 479). Syntactic relations involving the verb būti functioning as a copula (forming compound predicates) are examined accordingly. It can be seen that the concept of mutual correlation conflates two distinct levels of syntactic description—the relation itself and its expression—even though the preface to the grammar emphasizes the aim of distinguishing two other levels: matters of form and of the content (meaning) it denotes. Syntactic relations should be distinguished not between word forms, but between words as lexical units, since it is meaning, not word form, that determines a word’s combinability. One word may be dependent on another not grammatically but semantically, because it complements and elaborates the meaning of that word. A grammatical (or morphosyntactic) word form can only indicate or express this dependency. V. Ambrazas uses the term subordination for “the syntactic relation between units of sentence structure, one of which (the dependent unit) depends on the other (the head unit) and is incorporated into the sentence only through the head” (DLKG 1994: 479). In this case, although the syntactic relation is distinguished on the basis of not one but two features, it is clearly separated from its expression: “In typical cases the dependent unit has a particular marker of subordination—a case ending, an inflectional suffix, or a function word (a preposition or conjunction); only occasionally is word order in the sentence such a marker” (DLKG 1994: 479–480). Unfortunately, the dependent word’s dependence on the head is understood as a property of its form rather than of the head word’s meaning, as shown by the subsequent division of subordination into strong (obligatory) and weak (optional), distinguished by whether the dependent form is required for the grammatical structure of the sentence. Coordination is defined accordingly as “the syntactic relation between units of sentence structure that are independent of one another and are incorporated into the sentence independently or through an identical relation with some other unit” (DLKG 1994: 482). Here again we can see two distinguishing features of the syntactic relation BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351-7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 11 doi.org/10.35321/bkalba.2020.93.11 features: a) whether a word form is dependent or independent of another form, and b) the way it is incorporated into the sentence. This threefold classification of syntactic relations has remained in place to this day; it was repeated in the fourth edition of this grammar (2006) as well. 1.2.6. Vitas Labutis in his textbook for higher education institutions, Lietuvių kalbos sintaksė (1994) essentially recognizes a binary classification of syntactic relations, but ultimately gives way to the already established tradition and distinguishes as many as four types of syntactic relations: Nevertheless, from a structural point of view—that is, in a highly generalized account of syntactic structure—two types of relations are clearly contrasted: subordination and coordination. “Word subordination is a relation between syntagmatically unequal constituents, in which one constituent dominates the other semantically and, most often, formally. It is a one-way relation between constituents” (Labutis 1994: 26). Between the word forms constituting a compound predicate (nemokėjo atsakyti, atrodė baugūs ir paslaptingi), V. Labutis discerns a peculiar type of syntactic relation: “Here we regard the second constituent of the predicate as subordinated in a particular way to the first. Such a relation could also be regarded as a separate type of relation—for example, word attraction” (Labutis 1994: 27). “Word coordination is a relation between constituents at the level of word forms that are equal and do not dominate one another” (Labutis 1994: 27). Although V. Labutis acknowledges that “the word form realizing a feature <...> structurally determines all word forms directly dependent on it, that is, also the form of the word functioning as the subject, which, moreover, is often semantically also the subject actant of the word functioning as the predicate,” he nevertheless argues that “the feature itself must be attributed to something; essentially, it cannot exist without the object or entity whose feature it is.” He therefore sees grounds for distinguishing yet another type of “structural relation”—a two-way connection: “A two-way connection is the generalized structural relation between words functioning as subject and predicate, in which each syntagmatic element acts upon and conditions the other” (Labutis 1994: 28). BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351-7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 12 doi.org/10.35321/bkalba.2020.93.11 V. Labutis also discusses a fourth type of syntactic relation—the appositional relation: “In words that qualify one another (in a sentence, they most often function as adverbials), the qualifying word seems to dominate semantically, but the qualification can easily be dispensed with, and the seemingly principal constituent can also do without its supposed syntagmatic partner. <...>. This relation cannot be considered coordination either, since the constituents are not semantically equivalent. The relation between words that qualify one another can be called a p p o s i t i o n, or superimposition” (Labutis 1994: 28–29). V. Labutis clearly distinguishes between the levels of syntactic relations and their expression. He calls the expression of syntactic relations “formal relations between words” or “ways of expressing relations between words.” 1.2.7. In Jonas Balkevičius’s Syntax of Predicative Constructions in the Lithuanian Language (1998) syntactic relations receive little attention. First of all, the matter is summarized: “The interaction of two words forming a predicative syntagma is termed in various ways: mutual dependence (Ambrazas 1986: 9), bilateral connection (Labutis 1994 I: 17), or simply predicative relations (Balkevičius 1963: 17) or a predicative relation (Sirtautas, Grenda 1988: 37). These are synonyms, any one of which could suffice” (Balkevičius 1998: 22). The term co-domination is then introduced, along with the reason for adopting it: “the interaction of a dominant and a subdominant will be called co-domination, in view of the semantic correspondence of the terms.” The new term is used as a synonym for the earlier ones, with one qualification: “the traditional terms mutual dependence, predicative relation <...> are used to summarize only the relations between the main parts of a personal sentence <...>, whereas ‘coordination’ will denote the interaction of both members of the syntagma—the dominant and the subdominant—not only in a personal sentence but also in an impersonal one.” Elsewhere in the study, three traditional “ways of attachment” are distinguished: agreement, government, and adjunction; partial adjunction is considered a subtype of the latter—“an additional semantic relation between an inflected dependent constituent and a constituent from another word group.” The examples given are: Brolis grįžo pėsčias, Lapę gyvą pagavo, etc. (Balkevičius 1998: 40). As can be seen, here too semantic relations are confused with syntactic relations, and syntactic relations with their expression. BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 13 doi.org/10.35321/bkalba.2020.93.11 1.2.8. Although only a few articles in the Encyclopedia of the Lithuanian Language (1999, 2008) are devoted to syntactic relations, they nevertheless convey well the development of V. Ambrazas’s theoretical ideas, and are therefore worth briefly discussing here. The article “Syntactic Relations” distinguishes three types: subordination, mutual connection, and coordination. Subordinating relations are those between syntactic units “one of which depends on the other (the main one) and is incorporated into the sentence only through the main one.” The distinguishing feature of mutual-connection relations is that they are reciprocal: “the word form through which another is incorporated into the sentence is also grammatically dependent on the latter.” The second edition of the encyclopedia adds the note that “Mutual connection can also be considered one of the types of subordination, that is, reciprocal dependence.” Coordinating relations are those between syntactic units “which do not depend on one another and are incorporated into the sentence independently or through an identical relation with some other unit in it” (LKE 2008: 491–492). Each type of relation is discussed further in a separate article. The article on subordination adds that “[A] dependent unit usually has a particular marker of subordination” (LKE 2008: 435). In Lithuanian, this may be word endings, inflectional suffixes, or function words. A distinction is made between strong subordination (when “the dependent unit is necessary for the sentence structure to remain intact”) and weak subordination (when “the dependent unit can be omitted without damaging the grammatical structure of the sentence”). Atskirai apibrėžiant tarpusavio sąsają (LKE 2008: 548), pasakyta, kad tai yra „abipusis priklausomybės ryšys tarp sakinio sandaros vienetų, kurie formaliai sąlygoja vienas kitą“. Matyti, kad priklausomybė čia suvokiama tik kaip žodžio formos, o ne paties žodžio priklausomybė. Deramai neskiriamas sintaksės reiškinys (žodžio priklausomybė) nuo morfosintaksinio reiškinio (priklausomybės raiškos). BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 14 doi.org/10.35321/bkalba.2020.93.11 At the end of the article, it is noted that mutual dependency relations “are sometimes called predicative,” but this designation “does not correspond to the logical content of predication,” because “predication can be expressed not only by [mutual dependency] but also by syntactic subordination relations.” First of all, this remark is contradictory if we take into account that, as the author stated earlier, “mutual dependency can also be considered one of the varieties of subordination.” Moreover, by calling a syntactic relation “predicative,” we conflate the semantic and syntactic levels of linguistic analysis. Atskirame straipsnyje apie sujungimą (LKE 2008: 523) papildomai pasakyta, kad tokiu ryšiu siejami „sudėtinio sakinio dėmenys arba sakinio dalimis einančios savarankiškų žodžių formos“. Vadinasi, sujungiamieji ryšiai būdingi ne tik sakinio, bet ir frazės (arba žodžių junginio) lygmeniui. Dar pabrėžiama, kad sujungimas reiškiamas ne žodžių galūnėmis ar priesagomis, o „sujungiamaisiais jungtukais, žodžių tvarka ir intonacija“. Skiriamas uždaras (kai jungiamos tik dvi žodžių formos ar tik du sakinio dėmenys) ir atviras sujungimas. 2. A BINARY CLASSIFICATION OF SYNTACTIC RELATIONS After Lithuania regained its independence, for a time the study of Lithuanian syntax continued, through inertia, along the path that had been compulsorily chosen during the occupation and had become customary. Although innovations such as generative grammar, Louis Hjelmslev’s glossematics, and valency theory reached us through Russian translations and the press of the so-called democratic countries, the gap from the linguistic thought developing in the West continued to widen. The theoretical accretions of the Soviet period are also present in Lithuanian grammar works from the last decade of the previous century. They are reissued and continue to be relied on in schools and scholarly works. Lithuanian grammatical scholarship began to turn more toward the West only at the beginning of the new century. The rapid renewal of grammatical theory began when, at the Lithuanian BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 15 doi.org/10.35321/bkalba.2020.93.11 Language Institute, the series Studies in Lithuanian Grammar began to be published; this renewal is associated above all with the work of Axel Holvoet. The series also proposed, for the first time, a new classification of syntactic relations. 2.1. The classification of syntactic relations was discussed in the very first volume of this series, Studies in Syntactic Relations (Holvoet, Judžentis, eds., 2003). Although the concept of a syntactic relation was not discussed in the volume, this syntactic phenomenon was clearly distinguished from the morphosyntactic level—that is, from phenomena of a syntactic nature expressed by morphological means. Syntactic relations are also distinguished from the semantic relations on which they are based. The classification of syntactic relations presented in the volume’s first article is based on syntactic equivalence: “Syntactic relations are customarily classified according to whether the linguistic units they connect are syntactically equivalent or non-equivalent. The concepts of coordination and subordination are based on this distinction” (Holvoet, Judžentis 2003: 11). The article categorically rejects the previously customary practice of distinguishing three types of syntactic relations. A critique of this practice is given in the first footnote: “In more recent works on Lithuanian linguistics, several criteria are conflated when classifying syntactic relations—syntactic equivalence (the linguistic units either depend on one another or do not) and the direction of the relation (unidirectional or mutual; on this basis, three types of syntactic relations are distinguished at the highest level of classification <...>” (Holvoet, Judžentis 2003: 11). Cases of mutual dependency (or mutual dependence, predicative, and the like) previously distinguished are now unreservedly assigned to the subordinative type of relation. The article notes that the concept of coordination is essentially applied only to constructions of syntactically homogeneous units, made up of words that occupy the same syntactic position in sentences—that is, words that have the same kind of relation with other words in the sentence. A syntactic relation linking units of different syntactic status, by contrast, is considered subordinative. BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 16 doi.org/10.35321/bkalba.2020.93.11 The necessity of distinguishing syntactic relations from their morphosyntactic markers is emphasized: “To examine types of syntactic relations as a purely syntactic phenomenon, one must, of course, take morphosyntactic phenomena into account, since they reflect syntactic relations; however, morphosyntactic phenomena and syntactic relations cannot be equated” (Holvoet, Judžentis 2003: 12). 2.2. In the third volume of the Lietuvių kalbos gramatikos darbų, Gramatinių funkcijų tyrimai (Holvoet, Mikulskas, eds., 2005), the concept of syntactic dependency (the Lithuanian equivalent of the English term dependancy) was introduced, and subordination is identified with it. Dependency is defined as “a type of syntactic relation between words (or other linguistic units) in which one occupies a higher position than the other in the syntactic hierarchy” (Holvoet, Judžentis 2005: 11). In other words, in a subordinating relation, one word’s syntactic valency, which depends on the predicativity it denotes, is stronger than the other’s. The article then utočifies two distinguished types of subordination—government and modification. 2.3. In the chapter “Syntactic Dependency” of Rolandas Mikulskas and Axel Holvoet’s online-published outline of a new Lithuanian grammar (Holvoet, Mikulskas 2009b), the concept of subordination (which also encompasses cases previously called mutual linkage) is further refined. It states that the direction of subordination (what is subordinated to what) is generally perceived intuitively, but may also be defined more strictly (Holvoet, Mikulskas 2009b: 3). The directions of syntactic and morphosyntactic subordination may even be opposite (Holvoet, Mikulskas 2009b: 4). The chapter also emphasizes the importance of distinguishing syntactic subordination (e.g., noriu ledų) from morphosyntactic subordination (e.g., noriu pasivaikščioti). It is noted that morphosyntactic government usually marks a syntactic relation, whereas agreement may not indicate subordination: “agreement cannot be an indicator of syntactic dependency” (Holvoet, Mikulskas 2009b: 6). This is a new conception of agreement as a morphosyntactic phenomenon. BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 17 doi.org/10.35321/bkalba.2020.93.11 In the chapter authors’ view, the type of subordination known as government always expresses semantic dependency between words. In this case, “the syntactic relation is grounded in the semantic relation and reflects it.” This is not always so, however: agreement often, but not necessarily, marks a semantic relation between words. For example, in the sentence Sutikau Joną susirūpinusį, the agreement of the participle susirūpinusį with the noun Joną indicates a semantic relation between them, but does not indicate a syntactic relation (Holveet, Mikulskas 2009b: 7). 2.4. Axel Holvoet offers perhaps the most thorough critique of the distinct subject–predicate syntactic relation distinguished in traditional Lithuanian grammar in his article “Argument Hierarchies and Grammatical Functions” (Holvoet 2009a: 26–30). He emphasizes that the mutual-linkage relation in DLKG is grounded in morphosyntax, insofar as the verb’s agreement with the subject is understood as evidence of its dependence on the subject. On the other hand, the same verb governs the subject’s nominative case. Thus this relation, distinguished at the highest level of the hierarchy of syntactic relations, “is defined on the basis of criteria introduced only at a lower level of the classification (government and agreement).” A. Holvoet shows, drawing on Western linguistic theory, that agreement can mark not only the dependent but also the head of a syntactic dependency—that is, the predicate rather than the subject. In other words, the predicate’s agreement with the subject indicates that the subject depends on the predicate. Once agreement is understood this way, one must abandon the third distinguished type of syntactic relation—mutual linkage—since there is no longer a feature distinguishing it from subordination. The article goes on to ask whether agreement can indeed be regarded as an indicator of a syntactic relation. Drawing on Igor Mel’čuk and William Croft, it shows that agreement marks not a syntactic but only a semantic relation; it “points directly to the word (or words) with which the agreeing word is related” (Holvoet 2009a: 29). BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 18 doi.org/10.35321/bkalba.2020.93.11 3. THE NATURE OF THE SYNTACTIC RELATION BETWEEN SUBJECT AND PREDICATE As the overview shows, the new classification of syntactic relations differs substantially from the earlier one in that it does not recognize a third type of syntactic relation—the reciprocal relation (or whatever else it may be called)—and treats it as subordination. How, in this case, are instances of the subject’s relation to the predicate explained? DLKG gives two, one might say, typical examples of this relation, reflecting its two variants: the relation between the sentence subject and predicate (Mes gailėjomės draugo) and the relation between the subject and the copula (in a compound predicate) (Brolis buvo piktas). Let us discuss them in a little more detail. 3.1. Regarding the first example, DLKG states that “The nominative pronoun mes ‘we’ in this sentence is dependent on the valency of the verb gailėjomės ‘regretted’; if we used the verb pagailo [i.e. an impersonal verb], we would have to change the nominative to the dative, cf.: Mums pagailo draugo. ‘We felt sorry for our friend.’ However, the person and number form of the verb gailėjomės is in turn selected with regard to the form of the pronoun mes ‘we’ (cf.: aš gailėjausi, tu gailėjaisi, jie gailėjosi ‘I regretted, you regretted, they regretted’), so the relation here is reciprocal” (DLKG 1994: 479). As for the first dependency (government), it should be noted that what depends on the verb is not the pronoun’s nominative case, but the pronoun itself. Since it functions as the subject in the sentence, it is marked by the nominative case customary for the subject of a personal sentence. The verbs gailėtis ‘regret’ and pagailti ‘feel sorry for’ are connected only by a common root and its meaning; otherwise, they are verbs with different meanings, morphology, and syntactic properties. Therefore, replacing one with the other proves very little. The verb pagailti also requires an argument with a subject-like meaning, whose syntactic role in an impersonal sentence may be explained in various ways (as an object or even as a non-prototypical subject), but which is marked not by the nominative, but by the dative. Thus, the word functioning as the subject is dependent on the verb functioning as the predicate. This is a type of subordination known as government. BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 19 doi.org/10.35321/bkalba.2020.93.11 It is marked by the nominative case of the pronoun (or, in other cases, of a noun or a word used nominally), or such a nominative serves as its marker. As for dependency in the other direction—the selection of the verb’s person and number form according to the form of the pronoun—various views have been expressed in recent Lithuanian linguistics. According to one of them, the verb’s person and number form is selected independently of the pronoun functioning as the subject, in accordance with the circumstances of communication. Since these circumstances are the same when selecting a personal pronoun, the grammatical person and number forms of the pronoun and the verb coincide. This explanation is also supported by the entirely possible sentence expressing the same proposition, which is even more frequently used in spoken language: Gailėjomės draugo ‘We regretted our friend.’ For Lithuanian, this explanation of verb-form selection would be entirely sufficient. Only typological data (cf. Holvoet, Mikulskas 2009a: 4, footnote 1) compel us to acknowledge that in this case the verb is nevertheless in agreement with the pronoun. It might seem that the authors of the new Lithuanian grammar, by recognizing agreement between the verb and the pronoun, accept the explanation given in DLKG. Moreover, in an effort to defend the idea of a reciprocal relation, V. Ambrazas proposed in his later works that it be understood as double subordination, with the two dependencies running in different directions (Ambrazas 2005: 167–168; LKE 2008: 548). However, this is only a misleading impression. For agreement is not regarded as an indicator of a syntactic relation in the framework of the new grammar (Holvoet, Mikulskas 2009a: 6). It indicates only the semantic relation of the verb to its required argument, but does not mark a syntactic relation (Holvoet, Mikulskas 2009a: 7). Thus, the syntactic relation between subject and predicate is unidirectional—the predicate requires the subject, which is a characteristic feature of subordination. In the Foundations of General Syntax, A. Holvoet further refined the previously presented concept of agreement. Agreement is a means of marking the head and dependent member of a construction. Deictically, agreement indicates the verb’s arguments or discourse participants; thus, agreement is not a means of marking a syntactic relation (cf. Holvoet 2009b: 130). A. Holvoet BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351-7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 20 doi.org/10.35321/bkalba.2020.93.11 relies on Christian Lehmann, who argues that “agreement is inherently a means of determining reference: by copying the grammatical features of their arguments—person, number, and gender—verbs help determine their reference” (Holvoet 2009b: 129). 3.2. The second DLKG example is explained in this grammar very much like the first. The copula būti also determines the noun’s nominative form, since “[I]f a finite form is replaced by a passive participle or an infinitive, the nominative must likewise be replaced by another case, cf.: Brolio būta pikto, (Kaip) broliui būti piktam. Meanwhile, the person and number form of the verb functioning as a copula is selected with reference to the nominative subject, cf. Mes buvome pikti; Aš buvau piktas” (DLKG: 479). The copula būti requires complements (governs them) just like other verbs. Thus, it takes a noun (not its nominative form). Depending on the circumstances of the utterance and the intended meaning, a grammatical form of the copula is selected, while its arguments (dependents) are marked with different cases. The impersonal past participle form būta presents the circumstances obtaining at the time of speaking differently, and therefore the subject it requires is marked atypically—with the genitive case. The same applies to the infinitive būti, which takes a noun functioning as the subject (or complement) in the sentence. It is marked with the dative case. It has already been discussed that, in number and person, the copula agrees with the word functioning as the subject. It should be emphasized once again that this agreement does not indicate another syntactic relation (or a different direction of that relation), but merely signals again the semantic dependence of the noun on the verb, which is also marked by the choice of the noun’s case. In other words, this dependence is marked twice in the sentence: syntactically—by the choice of the noun’s case (nominative), and semantically—by the verb’s agreement in number and person with that noun. BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351-7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 21 doi.org/10.35321/bkalba.2020.93.11 4. CONCLUSION The development of the understanding of the syntactic relation between subject and predicate in Lithuanian linguistics can be summarized briefly as follows. The established binary division of syntactic relations, which took hold in the 1950s and 1960s, was replaced by a ternary one, assigning the subject–predicate relation the role of conveying sentence predicativity. Initially, it was understood as an indeterminate semantic and syntactic relation; later, in the second half of the 1970s, its syntactic nature was clarified and its name correspondingly changed. Because the principles underlying the division of syntactic relations were not clearly defined, the nature of the subject–predicate relation remained incompletely elucidated. Lithuanian linguistics continues, to this day, to grapple in many respects with a contradictory understanding of the subject–predicate relation. At the beginning of this century, a new way of dividing syntactic relations was proposed. It essentially extends the tradition of the binary division of syntactic relations in the language scholarship of independent Lithuania, but draws on a new level of theoretical thought in Western linguistics. According to this approach, the syntactic relation between subject and predicate is considered a dependency relation: the verb takes (governs) a nominal expression functioning as its subject argument, and itself agrees with it in person and (not always) number, thereby expressing the semantic relation between the predicate and its argument. Regardless of how agreement in the form of the predicate of a finite clause is understood, the new division of syntactic relations has at least one advantage over the earlier one: it distinguishes one fewer type of syntactic relation than earlier works on syntax. At least with respect to the theoretical foundations of grammatical description, this is clearly more economical. Moreover, the new theory is more economical in its principles for dividing syntactic relations: it relies on only one feature of the phenomena being classified. As has been shown, the earlier division relies on at least two features (not to mention that they are not properly distinguished). As becomes clear from the survey presented in this article, the reciprocal relation (then called predicative) began to be distinguished in Lithuanian works on syntax only after Lithuania was occupied, and not without BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 22 doi.org/10.35321/bkalba.2020.93.11 rusiškų gramatikų poveikio (dar plg. Vaskelaitė 2009: 13). Atsižvelgiant į tai, jau būtų laikas atsisakyti šio lietuvių sintaksės teorijos sovietizmo, nelaisvos sąmonės vaisiaus ir grįžti prie Vakarų mokslinės minties, kurios savaiminę sklaidą Lietuvoje grubiai nutraukė sovietų okupacija. LITERATŪRA Ambrazas Vytautas 2005: Axel Holvoet, Artūras Judžentis, red., Sintaksinių ryšių tyrimai [rec.]. – Acta linguistica Lituanica 52, 167–170. Balkevičius Jonas 1963: Dabartinės lietuvių kalbos sintaksė, Vilnius: Valstybinė politinės ir mokslinės literatūros leidykla. Balkevičius Jonas 1998: Lietuvių kalbos predikatinių konstrukcijų sintaksė, Vilnius: Mokslo ir enciklopedijų leidybos institutas. Dambrauskas Leonardas 1939: Žodžių santykis sakinyje. – Gimtoji kalba 6, 84–86. DLKG 1994 – Dabartinės lietuvių kalbos gramatika, red. Vytautas Ambrazas, Vilnius: Mokslo ir enciklopedijų leidykla. DLKG 2006 – Dabartinės lietuvių kalbos gramatika, red. Vytautas Ambrazas, 4-oji pataisyta laida, Vilnius: Mokslo ir enciklopedijų leidykla. Holvoet Axel 2009a: Argumentų hierarchijos ir gramatinės funkcijos. – Gramatinių funkcijų prigimtis ir raiška, red. Holvoet A., Mikulskas R., Vilnius: Vilniaus universitetas, asociacija „Academia Salensis“, 1–36. Holvoet Axel 2009b: Bendrosios sintaksės pagrindai [Bibliotheca Salensis IV]. Vilnius: Vilniaus universitetas, asociacija „Academia Salensis“. Holvoet Axel, Judžentis Artūras, 2003: Sintaksinių ryšių tipai. – Sintaksinių ryšių tyrimai, red., Hovoot A., Judžentis A., Vilnius: Lietuvių kalbos institutas, 11–35. BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 23 doi.org/10.35321/bkalba.2020.93.11 Holvoet Axel, Mikulskas Rolandas 2005: Sintaksinės priklausomybės tipai: papildymai ir patikslinimai. – Gramatinių funkcijų tyrimai, red. Holvoet A., Mikulskas R., Vilnius: Lietuvių kalbos institutas, 11–38. Holvoet Axel, Mikulskas Rolandas 2009a: Gramatinės funkcijos (rankraštis). Prieiga internete: http://old.lki.lt/LKI_LT/images/Padaliniai/Gramatikos_skyrius/Gramatines%20funkcijos.pdf Holvoet Axel, Mikulskas Rolandas 2009b: Sintaksinė priklausomybė. Prieiga internete: http://old.lki.lt/LKI_LT/images/Padaliniai/Gramatikos_skyrius/Priklausomybe.pdf. Holvoet Axel, Judžentis Artūras, red., 2003: Lietuvių kalbos gramatikos darbai 1: Sintaksinių ryšių tyrimai. Vilnius: Lietuvių kalbos institutas. Holvoet Axel, Mikulskas Rolandas, red., 2005: Lietuvių kalbos gramatikos darbai 3: Gramatinių funkcijų tyrimai, Vilnius: Lietuvių kalbos institutas. Holvoet Axel, Mikulskas Rolandas, red., 2009: Gramatinių funkcijų prigimtis ir raiška [Acta Salensia I], Vilnius: Vilniaus universitetas, asociacija „Academia Salensis“. Kalinauskas Bronius 1963: Žodžiai ir žodžių junginiai sakinyje. – Kai kurie gramatikos ir stilistikos klausimai, Kaunas: Valstybinė pedagoginės literatūros leidykla, 3–23. Kalinauskas Bronius 1972: Lietuvių kalbos žodžių junginių sintaksė: mokymo priemonė, Vilnius. Labutis Vitas 1976: Žodžių junginių problemos, Vilnius: Vilniaus valstybinis universitetas. Labutis Vitas 1994: Lietuvių kalbos sintaksė, Vilnius: Vilniaus universiteto leidykla. Labutis Vitas 1998: Lietuvių kalbos sintaksė, 2-as patais. leidimas, Vilnius: Vilniaus universiteto leidykla. LKE 1999: Lietuvių kalbos enciklopedija, par. K. Morkūnas, red. V. Ambrazas, Vilnius: Mokslo ir enciklopedijų leidybos institutas. BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351–7204 ARTŪRAS JUDŽENTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 24 doi.org/10.35321/bkalba.2020.93.11 LKE 2008: Lietuvių kalbos enciklopedija, par. K. Morkūnas, red. V. Ambrazas, 2-as patiksl. ir papild. leidimas, Vilnius: Mokslo ir enciklopedijų leidybos institutas. LKG 1976: Lietuvių kalbos gramatika 3: Sintaksė, vyr. red. K. Ulvydas, V. Ambrazas, A. Valeckienė, Vilnius: Mokslas. Sirtautas Vytautas 2019: Veiksmažodinis tarinys dabartinėje lietuvių kalboje [Bibliotheca actorum humanitaricorum universitatis Saulensis 18]. Šiauliai: Litera. Sirtautas Vytautas, Grenda Česys 1988: Lietuvių kalbos sintaksė, Vilnius: Mokslas. Vaskelaitė Ramunė 2009: Sintaksinių ryšių modeliai lietuvių kalbos sintaksėje:humanitarinių mokslų daktaro disertacija, Vilnius: Lietuvių kalbos institutas, Vytauto Didžiojo universitetas. Gauta 2020 11 29 Priimta 2020 12 30 THE RELATION BETWEEN SYNTACTIC SUBJECT AND PREDICATE: CONCEPTUAL CHANGES Summary The article reviews the development of the concept and classification of the relation between syntactic subject and predicate during the last seven decades in Lithuanian linguistics. This type of syntactic relations was singled out only in the Soviet period, its separation was influenced by the works of Russian syntax. Initially, its concept did not clearly distinguish between the semantic and syntactic levels of sentence analysis, it was called a BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351-7204 ARTŪRAS JUDŽĖNTIS. Sintaksinis veiksnio ir tarinio ryšys: sampratos kaita | 25 doi.org/10.35321/bkalba.2020.93.11 predicative relationship. Later, terms more syntactic by nature (mutual dependency, interdependency) were proposed. Attempts were made to explain this type of relations, but it failed to do this on the basis of one criterion. Both the direction of relation and it’s morphosyntactic notation have been relied upon, often without proper distinction between these criteria. At the beginning of this century, with the rapid renewal of Lithuanian grammar theory, there was a come back to the dual division of syntactic relations based on one feature – syntactic equivalence. The term interdependency was abandoned; its cases were interpreted as subordination, marked at morphological level by the selection of case of the argument word. The semantic relation between the verb and its argument is also marked on the verb that is agreed in person (and number) with his argument. At the end of the article, it is shown how several examples of the relation between syntactic subject and predicate are interpreted in the Lithuanian grammar and how they could be interpreted on the basis of the theoretical grounds common to Western linguistics. KEYWORDS: syntactic relations, relation between syntactic subject and predicate, interdependency. ARTŪRAS JUDŽĖNTIS Vilniaus dailės akademija Maironio g. 6, LT-01124 Vilnius [email protected] BENDRINĖ KALBA 93 (2020) http://journals.lki.lt/bendrinekalba ISSN 2351-7204