Pastabos dėl sintaksinio dviprasmiškumo sąvokos ir kiti susiję dalykai

Acta Linguistica Lithuanica LV (2006), 1 -53 Notes dél syntactic ambiguity and other related matters ROLANDAS MIKULSKAS Lietuviy kalbos institutas The article contains a polemic with Axel Holvoet’s analysis of Lithuanian sentences of the type Onutés Zalios akys. The author disagrees with Holvoet’s claim that sentences of this type are syntactically ambiguous between predicative possessive (‘belonging’) constructions and copular constructions whose subjects are discontinuous as result of fronting. In the author’s opinion syntactic ambiguity would entail semantic ambiguity, whereas the constructions under discussion have a well-defined meaning (they give a description of a person based on a characteristic attribute), entitling us to describe them as an autonomous construction. The sentence type under discussion originated, in the author’s view, by reinterpretation of a copular construction with a subject constituent made discontinuous by syntactic movement motivated by topicalisation. The author also polemicises with Holvoet’s theory of thematic-rhematic structure, a variety of that advanced by Bogustawski, pointing out that the notion of marked pattern of thematic-rhematic structure is not clearly defined. In the third volume of the Lithuanian language grammar work Gramatiniy funkcijy tyrimai (Holvoet & Mikulskas 2005), in chapter 6 Intranzitive sentence types: existential, locative possessive sentences, in subsection ir 6.6. ( 153-159) Axel Holvoet discusses one of the possessive constructions of an indefinite syntactic structure, which is represented by such sentences as: (1) Onutés Green eyes. (= 31)! (2) Peter's difficult character. (= 32) The original numbering of Article i8 quoted in the illustrative example is shown in parentheses. RoLanpas MIKULSKAS Based on the previously formulated existential and locative sentences, on the one hand, and possessive possession and belonging sentences, on the other hand, the author of the article analyzes the syntactic structure of the sentences in question and at the same time attempts to determine their place between syntactic constructions that are close in functional terms. Having tested the whole arsenal of the possibilities of syntactic division, the author arrives at the conclusion that due to the possibility of two syntactic divisions, such sentences as (1) and (2) are to be considered syntactically ambiguous, and the very construction they represent can be described as an intermediate one between possession and belonging constructions. The author of the article believes that such a sentence type could even be called hybrid (Holvoet 2005: 158). It is possible to find not one or two reasons why we should consider this conclusion of the author unsatisfactory. I will try to present some of them below. 1.1. Already in the first light the conclusion itself is formulated. In fact, the article shows quite convincingly that no syntactic breakdown model at hand can be applied without flaws to the analysis of the sentence in question. How long does it take for the syntactic structure of such sentences to become bi-pramatic? After all, the author’s observation that,,the construction in question has a special meaning“ (ibid., 156) seems to direct him to the search for its most transparent syntactic interpretation. And syntactic ambiguity also presupposes semantic ambiguity! Even if we agree with the author’s assertion that »the actual division is an aspect of the semantic structure of the sentence« (Holvoet 2003: 96), the semantic ambiguity of the sentences in question cannot be discussed in this post, because the article analyzes only,,non-- significant* variants of these sentences. (t. y. without a contrast kiré, segmental indicator or i8keltinés theme). 1.1.1. The ambiguous semantic structure, it seems, is more appropriate to talk about in those cases, when there is a basis to assign two different meanings to a sentence that is identical in all respects. The classical semantic basis for ambiguous syntactic interpretation is provided by the different quantification of action in the sentence. Such ambiguous sentences are widely known in the linguistic literature as: (3) Parents of multiple children do not pay the childcare fee. (a) ‘Several parents don’t pay the tax yet’ (e.g., malignant or unable), Dél sintaksinio dviprasmiSkumo (b) ‘All parents who have more than one child in the kindergarten are exempt from the fee’ (i.e., are exempt from the fee). (4) Parents of many children have little free time. (a) ‘Many parents have little free time’ (i.e. they work a lot, are at home), (b) ‘Many parents have little free time’ (i.e. care for many children takes away from them). The possibility of such a double interpretation was first pointed out by Lakoff (1970: 405), later discussed by Jackendoff (1977: 60), Hudson (1984: 79-81), and the cognitive causes of semantic (and syntactic) ambiguity were revealed by Langacker (2000: 82-83). The essence of the syntactic ambiguity of such sentences as (3) and (4) consists in the fact that the quantifiers (some, many) in the possessive phrase are (a) in the case of interpretation included in the logical operators (negative no-, quantifiers maZi) present in the other sentences, and (b) in the case of interpretation not included, i.e. the possessive phrase is integral in relation to the predicate negation or the second quantifiers. About the (a) interpretation cases it is usually said that the apprehension of the action of the second quantifier of their predicate negation is ar ,,pay* (wide scope). According to Langacker, the semantic specificity of the possessive phrases used in these sentences, and more specifically the ambivalentity of the terms of kinship that make them up, provides the basis for the dual interpretation of these sentences. As we know, the terms of kinship in semantic terms are of a relational nature: one term of kinship is defined through its relation to another. So, for example, the TEVO idealized cognitive model (ICM; Lakoff 1987: 68-76) always implies a relationship primarily with the child(ren) (son(s), daughteir r(s)) somewhat weaker - su zmona (baby mother). In cognitive grammar, such phrases as the father of the child (-y), as well as other possessive constructions, are usually analyzed based on the TasKo (reference point) arskarra (Langacker 2000: 80-81, plates 171-193; Langacker 1991: 37-39, plates 167— 180). In general, in cognitive analysis the possessor (child(ren)) is regarded as a point of reference through which the possession (the possession(s)) becomes available (identifiable) to the perceiver, which functions as a target in relation to the point of reference. In terms of the syntactic relationship between the components of the so-called possessive construction, the target usually functions as the main dému (head), o reference point as a — modifier jo (change). Taéiau Rotanpas MikuLsWhat when possessive phrase is made up of du kinship terms, there is reason to believe, its kad main démens, for example, tévas in the semantic structure there is always an ‘anticipateir d’ place for the essential jo ,,argument” — of the child -y), which can biti ,,filled* or remain implicit. Thus, to say galima that the components of the phrase in question are linked by two complementary syntactic dependencies: firstly, the relationship of modification to a lesser - extent ir (probability that it will be activated is mazZesné) supplementation — (in other terms — control) relations. The cognitive basis for such ambivalentity is that both terms of kinship, one to a greater extent, the other to a lesser extent, elaborate each other's — substructures. — Now, turning to (3) ir (4) sentence, we can ai8kiau say, kodél (a) in interpretation cases ju posesyviniy fraziy quantifiers are more »accessible* predicate negation to the second ar quantifier of the sentence: Siy fraziy posesorius (pregnant) ¢it as a more traktuojamas or less freely selectable ar add-on to the main domain (teva) That the latter one can replace the whole possessive phrase (so bit ir quantifiable), obviously shows the spoken sentence ir (a) paraphrasing. 1.1.2. In this long digression, I wanted to say that the use of a certain,,syntaxically ambiguous thing“ in the definition of a sentence has a semantic struktira cognitive basis. rimtas ir This, of course, does not mean that in every case where a sentence can be understood in two ways, it is necessary to talk about ambiguojo us syntax. For example, even such a seemingly simple sentence as John's temperature went up, and that can have two interpretations. Under typical circumstances iSheard such a saying, Of course, first we will think that rose Ion Chinese temperatiira, t. y. He's probably sick. But it's not hard to imagine ir such ,,technical“ circumstances of this statement (e.g., work in the boiler room), in which it is necessary that the device, whose indicators uz Jonas responds, temperatiira. This is ir not a pana’iy sakiniy primaji aktanta (John) su the second (temperatiira) is also linked by the functional model of the reference entity, although the possessive relationship between such actants is only implied. In the first, prototypical interpretation, the patient’s dative profiles the reference ta8ka itself (John's person), o the second — su Area associated with this reference point (reference point’s dominion), In this case, the responsibility — of the individual is within the scope of However, in such an analyzed sentence we can hardly consider the syntax as ar ambiguous, because the reasons for the twofold interpretation are internal, jo depending on the context, and are determined by the structural peculiarities. o ne Del sintaksinio dviprasmiskumo 1.2. I§ straipsnio ai8kéja, kad tokiy kaip (1) ir (2) sakiniy semantinei specifikai apibiidinti menka atrama yra ir autoriaus ankstiau aptartos dvi predikatiniy posesyviniy sakiniy turéjimo priklausymo konstrukcijos. True, the actual part of the sentences representi- - ng these ir constructi- — ons can in some respects be compared with the actual part of the sentences under consideration, and thus become an argument for the su latter’s place among the former (i.e. between two types of possessive sentences). However, the functionally comparable types of sentences differ greatly: neither the possessive nor the belonging meanings correspond to the functional purpose of (1) and (2) — “to identify a person by a certain distinctive attribute”. (Holfoet 2005: 155). If we 8i consider the argument of semantic inadequacy to be extremely important, then we can doubt whether the construction in question is considered possessive. The possibility of such a hypothesis will be discussed later, but if it proves to be true, then the su test of ir correlation of the genitive of the subject of the predicate-negative, which helps to distinguish existential sentences from locative and, accordingly, possessive sentences from belonging sentences, such as (1) and (2), cannot be applied. All this in mind the definition of the design in question when hybridinés can be misleading. 2.0 Before proceeding to the more detailed analysis of the syntactic structure of the sentences (1) and (2), it is useful to discuss some more general issues of the relationship between the structure of the sentence constituents and the grammatical function. When semantic arguments lead to the conclusion that the same sentence may have alternative constituent structures, it is natural to assume that an alternative syntactic division corresponds to an alternative distribution of grammatical functions. Let us examine whether this is actually the classical case of alternative two logical operators in a sentence, when there is a serious basis to talk about the syntax of such a sentence. ir dviprasmiskuma. So let us return briefly to the just ka discussed (3) and (4) sentences of the dual interpretation of the principle. 2.1. As we have seen, two interpretations of such sentences are based on the i3 possibility of lemia distinguishing two syntactic jy relationships within a noun phrase, which is a possessive construction of a specific verb: in one case (a), the main deem — possession is only an addition to the other deem — possessor, in the other case (b), it is modified by the latter. However, it ju should be noted that in both cases the main — dému representing the noun phrase, and therefore its grammatical function, remains the same (tévai). Thus, it cannot be said that the sentence level doubles j ROLANDAS MIKULSKAS for the interpretation of such sentences is relevant the dual distribution of its grammatical function, —a certain indirect response exists only within the subjective noun phrase. 2.2. That the ambiguity of this type of sentence is not related to the distribution of grammatical functions, is shown even more clearly by examples of simpler noun phrase tiniy structures, e.g.: (5) Everyone here will not risk. (a),Not everyone will su risk it’, (b),Not everyone will risk it’. (6) Mazai Zmoniy reads a lot of books. (a), There are few people who read many books', (b), There are many books that are read by only a few'. Siy sakiniy (a) in the case of interpretation, the first logical operator involves the second in its j operation apreptj, (b) in the case of interpretation o — atvirkS¢iai. For example, (5) sakinio (a) the interpretation is determined by the fact that the quantifier each aprepia the sentence negation, t. y. a negative predicate does not risk attributing to all elements of the assumed reference mass. Interpretation (b) determines the disposition of the logical operators: the negation aprepia the quantifier, the negative predicate is t. y. attributed only to some elements of the assumed reference. masés 2.2.1. The fact that the dual interpretation of the sentences in question cannot be linked to an alternative distribution of su grammatical jy functions, makes the reasons for ambiguity to be ju sought in an alternative syntactic division. TaGiau at least the i§ first Glimpse seems to both In the case of interpretation of this sentence we have to establish the constituent ta structure itself. Of course, unless we postulate a generative example of some deep structure. We recall that in the generative tradition the factor of syntactic breakdown is extremely important for the semantic analysis of the sentence. Generative semantics representation vu (plg. The ambiguity 1969) of Lakoff's su theorem with two logical operators is attempted to explain by postulating two different deep structures. For this logical operators were equated with actionaZodziy predicates, in deep structure definibréZianciám separate propozicijas. Such a position was necessary to adopt in order to show that in a deep structure one logical operator is involved through syntactic subordination of another operation. j Thus, by means of syntactic derivation (together with the application Dél sintaksinio dviprasmiskumo of a certain logical operatory,,degradation“ rule, assigned to the corresponding nouns of the structure of the deepness Salutiniy sentence) from the structure of the univerbal,,transformation“ in terms of syntactic division are derived two different structures of constituents of the »deepness«, responding to an alternative logical dual interpretation. operatoriy saveika. Jos ir yra laikomos atsakingomis uZ kalbamy sakiniy 2.2.2. It was soon noticed that such a structural allocation of alternative constituents of the sentence in question, formulated in terms of syntactic subordination, was not sufficient to cover all the possible needs. For example, such an analysis cannot distinguish the principle i8 in cases where neither of the logical operators used in these sentences is involved in the aprept of the other. For example, (6) can also be understood in the sense that just a small number of people read a large number of books. It's the third time Interpretation of this sentence, which is determined by its quantification disposition, when neither of them is dominant over the other. 2.3. Cognitive grammar devotes considerable attention to the issues of quantification and negation in mutual connection in a sentence (see Langacker 1991: 118-141). Representatives of this linguistic direction, unlike generativists, do not emphasize the importance of the syntactical division of a sentence for its semantic analysis, and in the case under consideration the different quantificatory and negative mutual relations are not connected with alternative constituent structures, especially since such structures can be distinguished only at the level of deep speech sentence syntactic analysis, which is prohibited by the methodological framework of cognitive grammar. For only conceptualizations of structures with phonological expression are recognized- *. Bypassing complex cognitive schemes and avoiding specific terms, I will only briefly describe how they represent the so-called ambiguous sentences of different interpretation, highlighting only those aspects of the analysis that can help to solve the questions raised in the discussion. 2.3.1. First of all, it should be noted that the representatives of cognitive grammar refuse to reduce the semantic analysis of spoken sentences to their logical propositions? The basic postulate of cognitive grammar requires that the phenomenon of language represents only phonological structures, semantic structures and symbolic relationships between them, i.e. symbolic structures. Taylor (2002: 292) formulated the $j limitation as a symbolic thesis. Rotanpas MIKULSKAS considering such a view too restrictive. Accordingly, the so-called logical elements in jy analysis are not considered logical operators in the strict sense of this word, but are defined as units of complex semantic structure. The interaction between two quantifiers or between a quantifier and a negation is examined through the conceptual links (correspondences) connecting these substructures. For the interpretation of a sentence, it is not very important which path of composition is followed in forming the Sudeten language structure — the sentence. Thus, one or another division of the structure of a sentence with the same phonological expression does not determine the possibility of different interpretation of that sentence. Such possibilities arise only when the sentence is conceptualized in different ways as a complete compositional structure. Different conceptualizations of the same sentence do not depend on its typological division, or on the different priority of the compositional levels, but on the possibility of the type of process profiled by the sentence to form different combinations of its instances at certain stages (levels) of the compositional path, which specifications have an inherently specific hierarchical structure and which can be further modified and quantified. 2.3.2. To make it clearer, what is being said, as an example, let's consider (6) Sentence. Such a sentence with quantifiers is rational to consider as the basis of conceptualization not as a single instance of a process type (number), but as its replicated variant, the multiplicity of which shows the plurality of the process actancy. In this case, such a multiple instance represents a sudétiné structure. Zmonés reads books. We can immediately say that 8i structure can be conceptualized in three ways, or on its basis we can distinguish three specifications of the subjective process type: (i),‘several people read several books’, (ii),‘the same person reads several books’, (iii),‘the same book is read by several people’. The simplest is to explain (i) the possibility of further conceptualization of the specification: the quantifiers maZai and much respectively simply clarify the books readan¢iy Humans ir are reading books in total. The alternative quantifiery apprehension ratio depends on which of the two —(ii) or (iii) the type — specification will be chosen as the basis for the sentence galutinei conceptualization. In other words, the higher level of conceptual organization replicates the two types of subjective processes by vienas iS ji Syntactic ambiguity with instances of the first or second actor. Thus, selecting the replicated process type (count) (ii) specification results in the interpretation of the (6) sentence (a), where j quantizer little operation apreptj included quantizer much. In the option (iii) type specification we have the interpretation of (6) sentence (b), where the quantifiery apreption disposition is inverse. 2.3.3. It is important to understand correctly what is meant when it is said that different interpretations of such sentences are caused by the already selected process type specifications, which are gliding in a certain hierarchical structure, at a higher level of conceptual organization and causing the effect of different quantification of operation. The whole hierarchical order of Si conceptual organization can be understood dynamically, i.e. as a kind of procedural asymmetry of conceptualization, the direction of which is related to the choice of specific and unspecific quantifiable reference by the participants of the replicated process. For example, if (6) is chosen as the basis of the sentence interpretation (ii) process type specification, the referents of the quantifiable subject noun (Zmonés) are perceived as specific individuals in the further stage of conceptualization, the referents of the object noun (book), even when it is quantified (many books), remain non-specific. In cognitive terms, the referents of the first actor are perceived as instanced in the physical space, while the referents of the second actor remain in the mental space defined by the type specification. In the final stage of the sentence conceptualization, the first quantizer (maZzai) allocates the 8j replicated process (ii) type specification element ‘many books’ to all specifically treated individuals (Zmonéms). In the case of interpretation (b) of the sentence in question, when the process type specification (iii) is chosen as the basis for further conceptualization, the referents of the object noun (book), the adverbs, quantified as specific individuals, 0 the subject noun (small Zmoniy) referents remain in the mental space of the type specification. This alternative conceptualization strategy is associated with specific and non-specific reference in the language of logic and is called the property of different quantifiers. 2.3.4. In a slightly different way, though also through the specific and non-specific references, alternative conceptualizations of sentences with subject quantifiers and predicate negations can be explained. Such a sentence of negative predicate in kato turinj galime apibrézti kaip (gramatiniu laiku) kalbovo resp. klausovo the space of discourse localizes a situation in which a certain process is eliminated. For example, (5) the negative predicate of the sentence divides, it is not very suitable and the RoLANDAS MIKULSKAS qualification,,actual’*, because only i8 part and only in »special cases« it is dependent on the actual premise of the discourse space®. Thema and rhema can only be considered as discourse categories in cases of contrasting division or thematic intervention. 3.2.2. It appears that the subject(s) allocated in the case of layered topical division of the sentence has little in common with the subject(s) allocated in the cases of contrasting division or theme separation. Only the latter relates to the logic of discourse and can be defined in the traditional terms of Zinoma vs. new information or presupposition vs. focus. Meanwhile, the components of the layered actual division are defined in more cognitive terms of reference vs. predication. In the latter case, the choice of the subject determines the characteristics of the constant categorical noun: vitality, personality, definition, etc. According to Holvoet, these properties, together with the semantic roles played in the sentence, constitute the characteristic of “inherent topicworthiness”. But these characteristics are also determined by the choice of grammatical function (first of all — the subject) (sic!) (Holvoet 2003: 87). Thus, the actual division, understood as a coherent distribution of reference and predicate acts, and especially distinguishing its unsigned variant, in fact, becomes inseparable from the distribution of grammatical functions in the sentence. Therefore, the layered actual division can save little for the syntactic analysis of sentences: based on it, we risk generating tautological statements. For a certain tautology is already present in the very definition of the subject of such an actual division: the subject of the sentence reflects the dramatization of the unsymmetrical subject, the latter being defined as a subject coinciding with the position of its subject.... Ne veltui gelbéjantis if keblios padéties griebiamasi formaliy gramatiniy funky skyrimo kriterijy. For example, it is stated that,,the character, although functionally reflects the dramatization of the unsigned theme, can be defined by (most notably the subject-coding) formal positions” (Holvoet 2005: 146). not the function's disposition in it. In other words, the semantic contrast of these two sentences consists primarily of the appropriate distribution of grammatical functions of the actants of the same verb in the lexical and referential values corresponding to the specificity of the given situation. (Jonas, garsus Journalist) atžvilgiu. O The subject in both sentences is the same -Jonas, only in the second sentence it should be considered Zyméta, because it does not coincide with the subject, But for such a "zymétity" it can still be defended (See below). * About Siq savoka Zr, Langacker (1991: 97), pg. Mikulskas (2006: 37). 16 Del sintaksinio dviprasmiskumo 3.2.3. Although it is acknowledged by introducing the latter variant of the layered actual sentence division that the latter division sometimes does not coincide with the distribution of grammatical functions and thus avoids the complete identification of the latter two sentences, however, on the one hand, the latter distribution of the layered functions, but different, on the other hand, the impression is created that the latter deviations from the aktualiosios skaidos variantas taip pat yra priklausomas nuo gramatiniy canonical model are only considered as exceptions confirming the latter rule (cf. Holvoet 2005: 145). And this, as we shall see later, contradicts the facts of language. 3.2.4. Since pabréZiama,,,that,,zyméta (layered —R. M.) actual division’s character is based on a deviation from a certain typical, grammatically encoded, actual division model* (Holfoet 2003: 88-89), it becomes clear that this “symmetry” is of a different nature than in cases of contrasting divisions or themes. For in the latter cases, the deviation from the neutral accentuated rem jprastos ar The word order is motivated by the logic of discourse (kontextiniy presupozicijy), and not by one or another distribution of grammatical functions corresponding to real situations. Of course, the prototype of the actual sentence division is undoubtedly those cases when the primary subject of the sentence coincides with the subject, the secondary subject su jo o — su object. Transitive sentence canon corresponding This distinct grammatical status of the action containing the figure of verbal relation and its linear development at the same time testifies to the relative ir cognitive importance of the information encoded by the subject in relation to the information conveyed by the object. There is reason to believe that when Sia asymmetry (most often already encoded in the actual verb structure) is reflected in the semantinéje actual sentence division, conceptualization (processual) is most smooth. jo But it ar makes more sense Name of Zymetas from This deviates from other models of actual division, especially if they represent the distribution of sentences grammatical functions verb actant semantic role or lexical resp. jy referentiniy veréiy”? in relation is functionally motivated, ZodzZiy o order seems normal? ° Plg. Langacker (2001: 44) said: ,,In particular, discourse tends flow most smoothly to when the participant chosen subject also the pivot corresponding salient topic. as is to a Ne always 1Seina talks about the distribution of grammatical functions in a sentence in relation to the jo semantic role of the verb actants. Some different sentence predicates 17 Rotanpas MIKULSKAS 3.2.4.1. In what sense do I call the distribution of grammatical functions functionally motivated!? First of all, when speaking about adequate conceptualization of the fragment of reality, it is profiled by appropriate sentence models. It is reasonable to assume that the adequacy of the chosen sentence model to a given situation ultimately depends on the disposition of the archetypal semantic role played by its actor in that model and on the degree of its assimilation. For example, the canonical transitive sentence is based on the so-called ENERGY TRANSFER GRANDINES ICM”, cf. : (7) Boy spyré ball ir (he) isdauzé langq. In this sentence, the position prototipiskai of the subject 0 is the agent, the position of the object (pozicijas) -exposure to the active agent (and dél padétj to űrvében biivj ar változgató) inanimate objects, respectively muveris (it is also an instrument) ir patient. ir That the distribution of such a sentence grammatiy functiony cognitive is motivated, atzvilgiu too much irodinéti is not necessary. Structured asymmetry of such sentences glides already in the past jy tense. Sodziy reSmeje ir confirming prototypiniy jo actants. On the structure of the canonical transitive sentence, it can be said that ikoniSkai ji repeats ENERGY TRANSFER GRANDINES pattern asymmetry. 3.2.4.2. In other cases, the asymmetry of the grammatical function disposition is not so marked (we could assume that ¢it is a lesser semantiniy roles dramatization laipsnis). Then the asymmetry of perception, which is necessary for struktithe conceptualization of sentences, can be implied by the corresponding models of i§ cognitive reality, e.g.: When we re-examine the verbs, we usually have to talk about the different distribution of grammatical functions of the verb’s actants in relation to the lexical and referential values. “The nature of functional motivation can be understood in different ways. The same term can also be used in a diachronic sense, referring to the so-called systemic pressure constraints, for example, when the motivation for the structural rearrangement of the older, inherited language creates the emergence of a newer, more grammatical, structural one. In this sense, the term functional motivation is also used by Holwoet, who explains the emergence of a newer possessive construction of the habeo type in place of the mihi est type as a result of the greater grammatization of the former. The newly chosen possessive predicate allows the subject of the possessive sentence to be given the status of a subject and thus,,equalizes* the discrepancy in this respect that exists in analogous mihi est-type sentences (cf. Holvoet 2005: 149). 2 Plagiau Zr. Langacker (2000: 23-34; 2002: 209-225). 18 Dél sintaksinio dviprasmiskumo (8) The child has temperatiiros. (9) *Temperatirai is a child. The distribution of grammatical functions in such and similar sentences in terms of semantic roles is determined by the inherent asymmetry of possessive relations, which is perhaps most pronounced when the whole relates to its part. 3.2.4.3. The syntactic asymmetry of some, for example, possessive constructions profiling the relationship of kinship depends on the perspective of conceptualization chosen by the speaker, e.g. : (10) These two children have a mother. (11) $i mother has two children. It can be reasonably assumed that the semantic basis of the conceptualization of these two sentences constitutes approximately the same situation of reality, and thus they encode different profiles of that situation. du Be without doubt, there ir are different circumstances in which one or another sentence may be implied. Thus, we could argue that the semantic contrast in such sentences consists of two different perspectives of conceptualization of the actual situation. The choice of one or the other perspective fixes the alternative distribution of grammatical functions, only in this ir case not in terms of semantic roles played by the verb actants, but in terms of lexical and referential values. Gramatiniy functiony disposition asymmetry This is related both to the specificity of ne the fragment of su reality being conceptualized and to the conceptualization method itself, which in this case constitutes the content of functional motivation. 3.2.4.4. In other cases, the alternative choice of the figure of kinship relationship by the subject of the sentence is determined by the coding of two different real situations and, accordingly, generates two sentences of different semantic content, e.g. : (12) AtvaZiavo nuotakos tévas. (13) Dad's bride has arrived. In these sentences we use posesyviniy konstrukty main démens modifier choice ir jo (then ir jy disposition functional motivation) leads to two completely different real situations. 3.2.4.5. The motivation of syntactic distribution in possessive constructions that profile kinship relationships resembles the padétis In this regard, the 19 RoLANnpDAS MIKULSKAS in equatorial constructions. For example, in the sentences (14) and (15) the grammatical functions are divided into actants and lexical functions, respectively. choice of the perspective lemia (kalbéjimo aplinkybéms adekvatus) i§ esmés tos patios situacijos of conceptualization of referential values, e.g.: (14) My wife is the editor of the publication. (15) The editor of the publication is my wife. However, such a choice may be limited by the specifics of actual categorization. For example, one-sided asymmetry of the grammatical function disposition is only possible in inclusion relationships in equivalence sentences, cf. : (16) Lithites are of the kaciy family. (17) *Katees is a litity genus. 3.2.4.6. In some cases, two different meanings of the conceptualization perspective of the i§ actual situation in the language can be used as predicates of the same direction, forming so-called conversion pairs: buy vs. sell; duodenum vs. take; give vs. to receive gifts; lending vs. to ir borrow etc., plg.: (18) John gave Anna gely. (19) She received from John the gift of géliy. In such sentences the distribution of grammatical functions in relation to semantic roles depends on the choice of the corresponding predicate for the purpose of communication. Perhaps worth ir noting is the generalization that the speaker prototypically chooses the so-called agent perspective (about the nature of such a strategy, see. Langacker as 2001: 43) in the (18) sentence, but why is the recipient’s choice of subjects in the sentence turéty biti (19) considered less functionally motivated if it meets the communication tasks? 3.2.4.7. From the perspective of ir ja conceptualization, the distribution of grammatical functions corresponding to semantic roles in language can be expressed only by lexical, but by grammatical means, of which ne passivation is to be mentioned first, e.g. : ir i8 (20) Laidq is married to a well-known journalist. (21) The show is hosted Zinomo Journalist. DaZnai prefer passive constructions because of the logic of discourse. ir 20 Dél sintaksinio dviprasmiskumo For example, the identification of the noun reference in the anaphora, and therefore ir its correct ju interpretation, is facilitated by such a grammatical funkcijy disposition, when the noun is also the subject of the sentence, cf. : (22) Do not hesitate to buy this computer program. Recently 1 was iS esmés updated. Although in terms of grammatical structure, passive constructions are often reasonably regarded as synonyms of the corresponding active constructions, (plg. Langacke2002: 229-230), r, however, shows (22) that from the perspective of discourse, the distribution of grammatical functions can also be considered motivated and natural. 3.2.4.8. Summarizing what has been said, we can say that various sentence models are functionally motivated if they are grammatical functions. As we have seen, the direction of this asymmetry dispozicijos reiSkiama asimetrija yra ikoni8ka profiliuojamai situacijai. may be determined by an objective cause, most often derived from the process of word designation itself, but may also be determined by a subjective cause, such as the choice of the perspective of conceptualization of the ja actual fragment, ir which in turn is often supported by the logic of the discourse. 3.2.5. The concept of the functional motivation of the distribution of grammatical functions is relevant to the discussion as far as it can be captured in the relevant sentence models. (resp. zymétuma). Taking into account only the motif of the grammatical structure of the sentence, we should first call Zyméts all those sentences, whose grammatical function asymmetry of disposition is in a direction compatible with that of the mentioned ENERGY TRANSFER GRANDINES ICM vector. The structural asymmetry of the Past-Proximal direction is taji, kaip atsimename, ikoniSkai atliepia kanoninis tranzityvinis sakinys. perhaps most evident in passive constructions (Of course, the distribution of grammatical functions is not a mirror reflection of the corresponding active construction: agents of passive constructions never become direct objects). jy Dad ,,grammatical” passive constructions The asymmetrical is the most obvious. 3.2.5.1. The following sentence models are also considered to be grammatically correct: (23) and (24), e.g.: (23) The inhabitants regained their expropriated property. (24) This hat is Antano's. 21 RoLANDAS MIKULSKAS There is reason to believe that the semantic basis of the sakiniy, conceptualization of the predicates that form the conversion pairs is the same idealized cognitive model. However, the direction of the process underlying such an idealized model coincides with the direction of su the process profiled by only one member of the conversion pair. Thus, although the subject of (23) the sentence (the inhabitants) performed veiksmas (against) true su tam conditions both physically and mentally can be treated as active (subject's agentiness accordingly differs in the syncretic recipient-beneficiary role), still due to such activity »direction« is the direction of the movement su of a good (property) that overlaps the main one, in the energy transmission chain - GAVIMO ICM (no schema Zr. Langacker 2002: 227). This direction corresponds only to the direction of the profiled process given by the other member of the conversy pair. The (24) “active” member of the possessive affiliation relationship in the sentence is Antanas only todél ne that is a living creature, but dél that he ir It can be assigned a certain role ir of beneficiary, whose psychic energy vector is directed towards the possession (hat). However noun Antanas néra The subject of this sentence. Subject (and possessive) position in sentences of this type (plg. still hat Si belongs to Anton) usually takes dead thing A substantive noun. It is understood that it is very be difficult to speak about such a sentence action as the direction of the profiled relationship of dependence. proceso Jy structural asymmetry is based on the asymmetry of the cognitive possessive relationship model. The latter is based on the fact that possession is always identified by the possessor's noun. atzvilgiu. tampa The reference point of this cognitive operation (reference point). Obviously, the direction of conceptualization of such a (24) possessive relationship as a sentence is toward the Singa phonological sequence. su ju in the direction of,,reading“ (processual in the narrow sense) in relation to the sutampan¢iai grammatical asymmetry of the disposition. Todél all the sentences profiling the relationship of belonging, despite the ordinaryness, jy also turéty biti considered Zymétais. Zymetais turéty biti laikomi ir daugelis kity fraziy modeliy (between jy ir us ripping, which is represented by sentences), (1) ir (2) kuriy grammatiy functiony distribution only disposition’ ar jy 8 pg. : (® The child is dying. (ii) The snake kills the child. In both sentences the grammatical functions of the verb actants deviate in one way or another from the canon of the transitive sentence, although pasiskirstymas jy their adequacy to the situations 22 Dél sintaksinio dviprasmiskumo being conveyed ar (functional motivation) is not in doubt. We could reasonably assume that the commonality of many of these jy sentences stems from the need for ir communication and the repeated conceptualization of a real situation. i$ ty Dél and is to kalbéti about pagrindo sentence types. atskirus Calling such sentences By considering only the deviation of jy the grammatical structure from the canonical model, we in many respects underestimate the empirically proven facts of language. It seems that it makes more sense to ,,Zymétas vs. neZymétas“ sentence (or typically) ju to apply the definition based on the concept of functional motivation discussed above. 3.2.5.2. Of course, the division of sentences j The ir classification of non-symmetrical syllables based solely on the criterion of canonical grammatical structure is much simpler. It may be implied that the rémimasis This classification of functional motivation criteria may be problematic in many cases, so the possibilities of application are still to be investigated. However, it is jo already foreseen that functional motivation is useful in cases where there is a basis to talk galime about outdated sentence models, especially about sentence types. The very existence of the conventional sentence model indicates that there is some kind of ar functional jy motivation for its emergence. Todél jreguliniai sakiniai modeliai ir corresponding jy Zodziy order de definitione turéty biti laikomi neZymétais. Adhering In this position, the symbols are called only individual implementations of various sentence models, which are primarily related to the unusual actual division, the latter being su understood as a category of discourse. Consequently, the division, which understands the functionSkai motivated aktualioji grammatiy functiony distribution, is biity nezyméta. It is understood that such a non-Zyméta actual division of the sentence must not necessarily coincide with su the hierarchy of grammatical functions. It is not difficult to imagine that consistently following funkatliekamy semantic role is the same, but the disposition in phonological sequence is reversed jy (as the ir actual partition). Although in both cases the disposition of grammatiy funkyy is motivated in its own way, there is reason to believe that (i) the sentence ji is more common than (ii) in the sentence, despite the fact that in the latter the direction of the asymmetry it expresses coincides su ENERGY TRANSFERS GRANDINES ICM vector. ™ Gia does not accept the fact that the hierarchy of grammatical ir functions itself is in some respects ambiguously defined. The Omen has indirect objects that encode various archetypal semantics. For example, the indirect object of the dative in the above mentioned hierarchy often vaidmenis takes a higher padét) than in the case of the ciné motivation principle. In an effort to keep the §j ski tum, we would avoid the following problems: we had to talk about the sentence model: which grammatical 23 Rotanpas MIKULSKAS functions are not divided, and which actual functions are divided. Only the cases of contrasting division and separation of the sentences, especially those that are recognisable in the sentence of the most common word order, are to be called symmetrical. The actual part of the predicate focus: or rema, iScales only that word, whose information answers the doubts or questions expressed in the sentence presupposition. We could say that the actual part of the sentence is closely connected with its specific (in Holvoet's terms, additional) presuppositions. 3.2.6. It is no secret that the order of words in a sentence is closely related not only to the distribution of grammatical functions, but also to the actual division (cf. Lyons 1999: 230). However, my intuition says that in such languages as Lithuanian, where the order of words in a sentence is relatively free, there is such a specificity. In view of these things, Holvoet (2003: 96) states that,,the actual division is an aspect of the semantic structure of the sentence, which may or may not be reflected in the word order, BUT IS ONTOLOGICALLY INDEPENDENT FROM THE WORD ORDER” (except my —R. M.). It is true that some cases of the Zymétos division are recognizable: mi only by the use of a contrasting crossword, corresponding intonation or /and segmental indicators. At the same time, it should be noted that all cases of layered actual division, which, as it is easy to understand, represent what we call functionally motivated sentence models, are essentially related to the order of the words in them. Holvoet also implicitly acknowledges this, stating that,,the important difference between Zyméta's non-contrast and Zyméta's contrastive actualistic division is that the latter does not always reflect the word order (ibid., 90). Thus, zyméta non-contrastive (i.e. layered) actual division reflects all the order of the words in the sentence’®. Of course, there are no objects of the instrumental or accusative. This is explained by the fact that the role of the experiencer in the hierarchy of initiative comes immediately after the role of the agent, and is therefore quoted higher than the roles of the instrument or the patient (see Langacker 2002: 238-239; cf. Holvoet 2003: 88). 8 See also ibid., 91: "If the actual division deviates from the model encoded by the grammatical functions of the unsigned, the order of the word becomes the only or main indicator of the actual division" (iScited by me — R. M.). 24 Dél sintaksinio dviprasmiskumo In the case of a layered actual division, it is all the more appropriate to talk about the corresponding unsyméta model of Word order (ibid., 86). 3.2.6.1. The relative independence of the actual division from the ZodZiy order is illustrated in Holvoeto- ’s (2003: 96-97) thesis by the following sentence pair, cf. : (25) Jonukas is very gifted. (= 48) (26) Very talented this boy. (= 49) It is stated, esa Siy the actual part of the sentence is not different. In both cases, the subject is the noun Jonukas, whose referent is said to be gifted in both sentences. One important detail is forgotten: These two sentences can be compared in terms of the division of meaning only if, after reversing the order of the words (26) in the sentence, the name Jonukas is accompanied by the demonstrative noun that gives the description of definition to the postverbal noun phrase. In this case, the definition of the proper noun is not sufficient to protect oneself from an undesirable (26) interpretation of the sentence, according to which Jonukas would indicate new, reminiscent, information of the sentence and would almost automatically be iStariamas with a contrasting accent, cf.: (27) Very gifted (is) Jonuxas, (0 not Petriukas). Such undesirable effect of the word order is related to the specificity of Lithuanian language in ensuring the coreferentiality of the discourse. As we know, Lithuanian language does not belong to the languages in which the definiteness or indefiniteness of noun phrases is regularly indicated by the corresponding articles. In Lithuanian language, in the actual space of discourse, minimally essential correferentiality is ensured by other means, such as demonstrative words, defining adverbs, relative quantifiers insignificant or common words, etc. (cf. Mikulskas 2006: 33). Some facts indicate that one of the most important means of coordinating the essential reference of discourse in Lithuanian language is the order of words in the sentence. It is quite understandable that the noun phrase said at the beginning of the sentence, in the prototypical position of the subject, is mostly defined, since its referent is already introduced into the actual space of the discourse: usually only the noun phrase whose referent is identified can become a point of support for the further discourse. Both of these figures, or at least one i3 jy —trajectory (in the case of the intransitive actionaWodzio), are inseparable elements of the conceptual structure of the actionaWodzio. They can be understood as nominal references in the most general sense of the reference point, without which it is difficult to imagine any process. Cognitive terms define the trajectory and boundary of verb-verb relationship and form the conceptual basis of the main grammatical functions — subject and object. Additional semantic and syntactic characteristics of this figure (the so-called archetypal roles of agent, patient, muver, instrument, experiencer) are induced from the use of language, which reflects the prototypical participants of the corresponding cognitive models (the main one - the GRANDINES OF ENERGY TRANSFER ICM was mentioned earlier). 3.3.2. 25 RoLaNDAS MIKULSKAS Now let us return briefly to the layered actual division of the sentence. It, as we remember, factSka is concerned with the problems of the process of the sentence, not with the logic of the discourse. The assertion that two acts of reference are not possible at the same time, which is a kind of justification of the layered actual division, is related to the specificity of attention concentration and thus has a psychological basis. At the same time, it implicitly refers to the time parameter that is important for the conceptualization of its syntactic structure. It is the use of this procedural time that determines the distinction between primary and secondary themes in the actual section. It is easy to notice that they are very close in their essence to the relative-length figures of the cognitive grammar — the trajectory and the limit-center. It is their disposition that grammatically encodes the subject and object of the sentence. Formally, this coding is expressed in nominative or accusative syllables or in the corresponding ZodZiy order. It is also easy to notice that the same parameters important for the “inherent subjectiveness” of noun phrases also constitute the grammatical characteristics of the subject of the sentence. I3 of what has been said, it becomes clear that by defining the subject and object of the sentence as figures of the syntactic relationship of the first and second cognitive levels, the essence of this subject assumes the functions of the primary and secondary subjects during the layered actual division. The subject itself is thus released for the analytical needs of the discourse. In other words, the cognitively introduced trajectory and boundary characteristics resolve the layered actual divisions that are important to Salininkas, i8 32 Dél sintaksinio dviprasmiskumo reference vs. predicate dichotomy, sentence processing problems. 3.3.3. Thus understood the subject and subject sqvokas unites their definiziamuy noun phrase common function —point of reference. Only 8ia function of the subject nouns perform each in their domains. temos ir The domain of a single topic can comprise from one to the whole of a sentence. It is precisely their all-relevant sentence content that helps to correctly interpret the same thematic ir noun. The domain of the subject as a point of reference does not override the sentence boundary. In other words, the subject noun helps to properly understand the content of the action-word relation. Second by importance In this respect there is the object of the sentence. Thus, in interpreting the content of the sentence, the nouns of the subject and object form a kind of reference point, which is comparatively formed sudétiniy posesyviniy lokatyviniy construction grandine, elements su tomis, kurias ar (plg. the lawyer of my mother's sister's husband; The third volume of the Lithuanian language grammar work can be found on the website, on the top shelf of the Soninés bookcase, next to the number of the Journal Acta Linguistica Lithuanica). 3.3.4. To limit the ir subject of the subject helps not only different points of reference of the noun domains, which coincide only in the jy marginal case, but also different points of reference of the noun domains in relation to you. Temos the noun usually remains in the iSoré of the interpretative field it creates. [ sentences are drawn to its domination as a point of reference solely according to the internal logic of the development of the discourse. In contrast, the subject noun, which serves as a reference point, is an immanent element of the sentence structure. This element is already present in the abstract form in the conceptual structure of the verb itself, and is only semantically clarified when forming finite jo sentences by o selecting the actants corresponding to ir the situation it is encoded in. 3.3.5. If we assume that the common theme helps to understand properly the contents of the whole sentence, then in each i8 In this sentence we find a noun phrase (dazZnai jvardinj her variant), co-authors8kq su ta tema’®. We could say that such a theme functions as visy 8 This kind of padétis is somewhat idealized. Interpretation j of common themes in language lauka (especially if we take the last paragraph) may be included such sentences, in which the subject is not mentioned in ir the noun phrase can be approximately reproduced only impheitically. ir biti 33 ROLANDAS MIKULSKAS Siy noun phrase common predecessor. The spoken verb phrases become a kind of supporting ta8kais (pivots), allowing the predicted properties of the sentences to be assigned to the appropriate speaker, which ir indicates the subject. It is understood Si that the correferential noun phrases funkcija only appear in the course of the discourse. Based on his preliminary observations, Langacker believes that the first (2001: 12-13) candidates for the co-referential support point function are in order of precedence the subject, the object, or the two possessors. Jo 1000 mg/ day. This trend would only be confirmed. The data discussed above show that it is quite common to say a direct indirect object at the beginning of a ar sentence in Lithuanian. Sis leidžia faktas sėti when lietuviy kalboje Langacker's hierarchy of sentence parts candidate for the function of a — supporting point is probably as sharp, cf. : — ne (38) You should try the miisy pie: (a) here is baked 1s according to a special recipe, (b) here 4 is baked according to a special recipe, (c) Cia resLa Jo ugmisoma by prescriptiypatingg on, (d) cia so TeSL4 is mixed according to a special recipe. 15 first It is not easy to say which commentai§ ry variant is preferred in the language. Plg. but: (39) Yesterday, they took me to the i§ hospital by mistake. A week ago, Sam had an appendicitis. (40) Today Lietuvos - Suspects arrested at the Russian border. Is sy confiscated a large quantity of contrabandiniy cigareteciy. Tviréiau The priorities of the choice of the sentence part for the role of the co-referential support ta’ko could be discussed only after more detailed research. 3.4. In summary: In the context of the current discussion on the topic of sentence division, we can say that the development of cognitive grammar only allows to define the subject specifically for ne the purposes of discourse analysis, but naujai allowed to distinguish more clearly the non-significant ir ir Significant cases of meaning breakdown in sentences. O this helps to better model the conditions of ir alternative,,reading“ of that particular syntactic structure. 4. | meaningful division of the sentence, not being more flexible than the layered actual division concept allows, let us analyze more closely the existential 34 Because of syntactic ambiguity, the locative ir predicate of a sentence does su not have the semantics of the syntax. ir Previously, we have observed that those typical syntactic structures (of the same grammatical distribution) different funkcijy conceptualization of the sentence depends to some extent on the actual division. ir jo Now, let's try to answer the i question, ar Zyméta actual part of the sentence can change the type of the sentence. 4.1. At the beginning, let us recall what is said about the type sentences in the article under discussion (Holvoet 2005: 139-147). Here analyzed sentences perraLet's say as (41) ir (42): (41) There is a sqbook in the desk. (= 1) (42) The notebook is on the table. (= 2) These du sentences seem to be composed iS ty a certain semantic element. According to the author of the article, ju the common feature is that one, ir the other ir requiring ,,the presence of a certain thing in a certain place“. As the author writes, the »(1) ir (2) sentence prefix does not have different grammatical functions in pasiskirstymu: both cases the same noun (notebook) goes factor. PrieSprieSa related to su the actual part of the sentence, arba more precisely — su jo nezyme modelium <...> In an existential sentence, the subject forms the place where the existence of a certain thing is asserted; rema forms a statement about The existence of this thing. In the locative sentence the existence of the thing is assumed, the reminder o statement forms only in a certain place" jo buvimas (ibid., he 140). further correctly observes that the difference in actual division is not the only That's the difference. IS tikryjy ty difference, according to the author of ir the article, there are more — ,first of all they relate to the verb’s purpose. In the su existential biti sentence The meaning of this action is bitent existential it affirms the object — (bitybés) egzistencija’ (ibid., 140-141). 4.1.1. Thus, it is acknowledged at the same time that the difference in actual division is not the only cause of semantic contrast in sentences. TaGau remains unclear circumstances that prompt the sentence actionWodziui (41) biti give exactly ,,existential“ meaning, when in a locative (42), sentence The meaning of this verb, according to the author's zodiac, »is generally difficult to specify* (ibid., It is not 141). difficult to believe that the definition of the subject noun in the locative sentence and the indefiniteness in ir jo the existential sentence remain the only constants, independent of the actual division (in the sense that they are primary in relation to them), the elements of the semantic structure of the sentence, allowing the action Zodiacs to be conceptualized accordingly, ju giving them a,,localization“ need in the first t. y. 35 ROLANDAS MIKULSKAS case and an,,existential“ meaning in the second. In order to explain the connection of the interpretation of the existential verb with the undefinedness of the postverbal noun phrase, it is sufficient to delve into the ontology of the predicate of its actual existence. It seems that to assert the existence of an object is accepted only when the reality of its status is doubted. We could reasonably say that formally Si is ontologically doubtful because of the indefiniteness of the noun phrase in question (not for nothing This effect of (in)definiteness related to the predication of existence is widely discussed in the literature; Zr. Lyons 1999: 236-246). Of course, in saying this, we should not forget that the main function of the description of definition is to ensure the correferentiality of the noun phrase denoted in the actual discourse space, or, more simply, to indicate the referenty identity of the noun phrase and its antecedented. It has been previously shown that the undefinedness of postverbal noun phrases in Lithuanian sentences is prototypically implied by the support of its position. A closer examination of the expression of the predicate of existence allows us to clarify that such an indication of the indefiniteness of the noun phrase related to the actual division of the sentence and the corresponding ZodZiy order only iconically reflects the ontological conditions of the spoken proposition, but is not directly related to the referential nature of that phrase in reality. After all, the defining description itself opens the possibility of unique identification of the referent of the already introduced or newly introduced noun phrase. It is obvious that the status of an object in reality and its status as the referent of a noun phrase in the discourse space are not the same thing: the referent of a defined noun phrase can be instantiated and identified in the mental (e.g., assumption, expectation) space. However, the strong correlation between the indefiniteness of the you proposition and the indefiniteness of the postverbal noun phrases suggests that these statuses are often identified in the self of the speakers. Usually there is no reason to doubt the actual existence of the dél defined noun phrase of the referent, we would say, it is,,naively,, assumed. And vice versa, in a new discourse (undefined noun phrases) the referent does not have such an “existential” presupposition, so it can be positive if necessary. 4.1.2. A prototypical way of correlation of the indefiniteness of existential proposition and postverbal noun phrase can be found in the attempt to imagine that (41) affirms the existence not of any, but of a specific (recently mentioned in the conversatio36 Dél sintaksinio dviprasmiskumo n) brooch. (t. y. presence at the counter). This interpretation of the sentence is likely, but less likely, cf. : (43) Is there a sqskaity sqbook on the desk? (a) "Yes, there's that notebook on the desk. (b) Yeah, that notebook is on the desk. This example, among other be things, shows that a definite noun phrase, confirmed by the existence of its referent, is easier to say at the beginning of the sentence. This is ir understandable, because (b) the sentence rhyme only forms an ta affirmation of the verb word is. 4.1.3. If the (41) sentence is still speaking about a specific, previously mentioned, stitching, then the place of reading in it will most likely be marked with contrasting accents, cf. : (44) In the wheels there is a notebook, ne (but on the table). It's easy to see that This sentence in relation to the actual division is The second variant of the (42), locative, sentence: shows only the ta contrasting ne consonant of the rhyme (Srazciwe), but its unusual presence at the beginning ir of the word. Therefore, the sentence has nothing to do with an (44) existential statement. su 4.1.4. If we correctly relate the ontological conditions of the predication of existence to the undefinedness of postverbal su noun phrases, then at least at first glance it is difficult to explain the quite usual use of i8 inherently defined real nouns in existential sentences, cf. : (45) The room has Jonas. But the inherent definition of the name so used implies that the holder is known to the jo speakers. Such a definition is of semantic nature (its basis consists of REALY NAME ICM) it does not ir in any way indicate a correferential noun phrase in the actual j space of discourse, or, in other words, it does not imply an antecedents. Since the implication is a condition of the subject of the noun phrase, the proper noun, without it, can freely Si occupy a reminiscent position in the existential sentence. 4.1.5. The specificity of the use of proper nouns in existential sentences This is the iconic aspect of the sentence. In a naive worldview The question ar of the existence of man as a living creature is identified with the su jo question of life: a dead man does not exist! Type in question 37 RoLanpas MIKULSKAS When asserting in sentences the existence of a person, the holder of a name, it is rarely meant his life: existence is understood as the presence of that person in a certain place. Well, the vitality of that person, i.e. the existence in general, is usually not questioned in such sentences, it is simply assumed, arba priimamas kaip duotybé. Taip, kalbédami apie asmenis, galétume skirti ju existence (being) in some place existence (living) in ir general. When asserting the existence of a thing in spoken-type sentences, it is usually also localized. Only in this case, it seems, the presence of things in some place is identified with their existence in the world in general. 4.1.6. From this observation we can draw one important conclusion: existential sentences with verbs rarely talk about the existence of a thing or person in general, the statement of their existence is important to the speaker or listener only insofar as these objects are within their reach or accessible to their control. In other words, the existential meaning of such a sentence predicate (be) prototypically implies the beneficiary. If this is true, then the appearance and use of the posesyvinés mihi est type construction alongside the existential verb biti sentences is quite understandable. There is also an easy way to paiSkinama and lokatyvy use in the latter: the location of the object is indicated along with Zymi and its accessibility degrees. 4.1.7. We could reasonably assume that the prototypical localization of the existence of some thing in the sentences of the type in question is reflected by the indefiniteness of the noun it is named, rather than by its indefiniteness in such ways that it signals that that noun has no antecedents in the actual space of the discourse. In other words, the thing whose existence is affirmed in such a sentence is usually introduced to the discourse for the first time. In such a real space of localized object existence and actual discourse, the iconicity of the unlocalized* noun (its referent is not uniquely identified in this space) is reversed. It is easy to understand that its basis is the identification of the accessibility of the object to the participants of the conversation in reality with the “accessibility” (i.e., recognition) of the relevant noun referent in the actual space of the discourse. In view of all this, the actual division of existential sentences of the type under consideration and the corresponding word order in them can be considered functionally motivated. Todél nezymétu can deservedly be called not only (42), locative, sentence, but also inverse ZodZiy order (41), existential, sentence. 38 Dél sintaksinio dviprasmiskumo 4.2. The reverse aktualioji division is the first distinctive j feature of the existential locative sentence of the type in question. ir If This characteristic is essential for typological jy classification, it varies according to the These sentence division variants don't change the jy sentence type so much, ir t. y. egzistencinis the sentence becomes ar the locative inverse. Examples such as the one mentioned above are (44), not misleading. Nors The order of this sentence ZodZiy is the same as ir (41), existential, sentence, incorrectly jj by the latter, su because in fact it is the marked variant of the locative, sentence. (42), Apparently, we are talking about the typological dependence of sentences. jy ir ja the corresponding word order determines as far as it helps to recognize ji the indefiniteness ar of the subject jy noun in one case and the definiteness in another. ir For example, in a locative sentence, the subject-subject noun suffix at the beginning of the sentence simultaneously (42) indicates the biting of the locative verb and the noun ir definition: only the identified referent can to be localized. biti In order to interpret such a sentence as existential, we first have to "delete" the definition of the jo subject noun. This is usually done only by contrasting it with the reminisce kiréiu status, e.g. : (46) Notebook is in drawer, (0 not pencil). Bet This sentence is the eighth variant of (41), the existential sentence. 4.3. When the defining description of the subject noun is given explicitly, typologically we talk about the independence of the sentence from The actual divisions of jy Zyméta are even more obvious. At least it is, taip when the thing- © Plg. : (1) Antanas came home late in the evening. Varkin was completely off his feet. Gi) Sigitas arrived only at night. Bricavininea until namy pavézéjo co-workers. It is easy to notice that the shortest possible distances separate the supporting noun phrase at the beginning of the sentence from the antecedential theme in the phonological sequence. We could reasonably assume that this ensures its corrective defined interpretation in the most economical way. ferentiskumut Although there apibreztumas is an essential attraction to noun phrases, traditionally called simply subject, trait, nederéty The figures of this actual division identify a definite noun su phrase - These qualities are defined on different grounds. It is understood that the actual part of the su sentence correlating with the preverbal definition of ir the darktavardinés phrases with the postverbal indefiniteness is not Zyméta. yra In fact, such an tik actual division can be considered as a characteristic of the semantic structure of sentences that are classified in a typological sense. 39 RoLanpas Mrkutskas naminés frazés neapibrézZtuma existential sentences Zymi genitive of the undefined quantity, e.g.: (47) Staléiuje is sqsiuviniy. This sentence could be the answer to the question What is on the table? When the subject noun of such a sentence is removed from the subject, its existential dimension remains easily perceived, cf.: (48) Is there (where) sqsiuviniy? Sqsiuviniy (it) is (on the table). The positive predicate of this sentence's rhyme continues to exist is. A sentence of the same Zodziy order may have another riies Zyméta actual division as rema contrasting kiréi distinguishing only place (sra:cruse- ). Of course, the presupposition of such a sentence is somewhat different, cf.: (49) Where (still) is sqsiuviniy? (Jy (= sqsiuviniy) is) starciuse. Interestingly, the existential predicate of such a division of the sentence is on the subject side and thus has the status of a presupposition. | this is pointed out by démesj and Holvoet (2005: 142) in their analysis of the analogous sentence Saldainiy is dézutéje (= 8). Although the actual part of such sentences as (49) and (= 8) resembles the actual part of the corresponding locative sentences, we could, in Holvoet's opinion, consider them existential sentences. And 8this is kodél: In these sentences, the presupposed existence of the subject-verb-referent of the phrase is of a different kind than in the locative sentences. In the latter, the reality of the dél-localized referent (i.e., the dél of his existence in reality) is not in doubt. In the first sentences, however, only the assumption of the actual existence of the noun referent of the subject dél jy is made. We could even say that the existence of this referent is presupposed only in the hypothetical, or imaginary, mental space. | the presupposition of such a hypothetical existence most clearly indicates the genitive of the indefinite quantity. Undoubtedly, such sentences as (49) and (= 8) have the function of confirming the existence of the referrer of this supposed noun phrase. However, it must be noted that this confirmation of existence is interpreted in the spoken sentences in a distinct way. It is likely that the answer to the question in the (49) example will most likely be —in the desk. However, the localive alone cannot confirm the possible existence of the thing, only together with the verb word be (remember that if the thing where Although yRa, then it yRa circumscribed). Thus, although the predicate in (49) belongs to the subject, it forms one predicate 40 Del sintaksinio dviprasmiskumo with the locative in the rem in the stald. More generally, the $j predicate can be expressed as a bit in the space. Although formally such a predicate resembles a typical predicate of a locative sentence (see Holvoet, Cizik-ProkaSeva 2005: 74-78), on the level of its discourse, due to the additional function of confirming the existence of the referrer, it is considered existential. In so saying, we could actually agree with Holvoet’s (2005: 142, 2 ignaSa) possible alternative (= 8) interpretation of the sentence, according to which its predicate syncretically means both existence and location. In other words, the verb biiti in this sentence is both existential and locative at the same time. Then we could describe such types of sentences as (49) and (=8) as intermediate between existential and locative. But then we should admit that the last actual part can at least strongly,,- wash* the foundations of the typological dependence of the sentence. Anyway, if such sentences, guided by more functional than formal considerations, are still assigned to the existential type, then hardly without compromising the facts of language we can call the mandatory locatives in them peripheral (or optional) parts of the sentence (see Holvoet 2005: 146). 4.4 Thus, no matter how varied the meaning division is, existential sentences remain existential, and locative sentences remain locative. The actual partitioning itself does not change their typological dependencies. We will see if this statement is applicable more broadly. 5. Finally, we can proceed to the analysis of sentences such as (1) and (2). For the sake of simplicity we will discuss only the first one of them, presenting it as (50): (50) Onutés Zalio ojos. When discussing structures such as (50) in Difference, Holvoet notes that the binomial modifier cannot be omitted, “because probably under no circumstances could we consider ZodZiy seky Onutés akys, Petros charakteris to be sentences”. According to the author of the article,,,<.- ..> the adverb Gia is important because it is the main predicate of the sentence“ (Holvoet 2005: 153). I8 tiesy This statement is only true if the diphthong is perceived as a predicate in the sentence in question. This interpretation implies that the subject noun akys with the possessive genitive Onutés forms a non-verbal compound, cf. : 41 RoLANDAS MIKULSKAS Interpreted as a possessive possession construction sentence then (50) reminiscent of a dative construction, which in Lithuanian language is relative su (Holfoet 2005: 150), plg. : (69) For the child (is) a style. (= 23) (70) ?Onutei (years) Green eyes. The Mihi est type construction for the meaning of possession is still used in Lithuanian in cases where the possessor can be perceived as an experienced experiencer of a certain thing. ir — I8 principle ir (70) statement is impossible, but it sounds archaic, although similar construction — Bijiinéliui Zali leaves (Sucks 1998: 162) - is evidenced in dialects. §j sentence in his article quoting Holwoat (ibid., the 159) possibility that the dative In this construction there has to be an older genitive. uz But if it is correct that the possessive construction of the type mihi est is started by an existential sentence, then it is at least incorrect in terms of origin to compare it with a construction in which the possessor is the genitive of the noun’’. ja su For the origin of the latter construction is linked to the su o Holvoeto (ibid., 159), ir posesor's i&kélimu of i§ the relevant noun sentence (j argumenty lygmenj). Dél of this origin is not suitable for the spoken construction ir su existence predication associated with the negative genitive test. 5.5.3. Sentence (50) meaning is predikacija interpreted as a relationship of possession between a ir person and a good attribute helped by certain specific suchy # 18 principo ,,possessively* reinterpretate the genitive of the spoken type of a sentence noun may be replaced by a dative, especially in those cases where the genitive Zymétas of the subject of the possessive relationship (posesorius) is perceived as ir The coefficient of this ratio. We could even say that In this case, posesyvine the sentences reflect the arguments of the so-called possessor-- separation j clause between the minor stages. It is easy to notice that if the genitive of the possessor of such sentences were replaced by the dative of the beneficiary, they would in su fact coincide with the possessive construction of the mihi est type. It is possible that now such a kad change is more difficult to imagine, rather blocked) only because the possessive (0 possession construction of — dative in the current Lithuanian language is almost displaced by newer habeo type constructions. su However, it is more likely that in the past part ,Possessively* reinterpreted sentences of the type in question, after the semantic dative was replaced by the genitive, coincided with possessive sentences of the mihi est type, which appeared in existential su sentences. i8 Additional diachronic studies were required to confirm this hypothesis. 48 Dél sintaksinio dviprasmiskumo contexts, in which those attributes of a person become,,in demand”, the disposal of them can be in one way or another beneficial to that person, cf.: (71) Onutés (is) beautiful legs and a lean waist —galéty participate in a beauty contest. possessive construction with PanaSiomis aplinkybémis vartojama (71) sakinj drasiai galétume keisti the possessive predicate, cf.: (72) Onuté has pretty legs and a nice waist —galéty to participate in a beauty contest. 5.6. Summarizing all this, we can say that for various possible (50) two types of sentence models provide the basis for sentence interpretations: the connection construction and (typologically close to the locative sentences) the possessive belonging construction. In both cases, the sentence under consideration is Zymét with regard to the relevant division. (50) in the sentence as the inverse of the possessive possession clause of the Zodziy order Zalios akys is Onutés Zyméta variant can be understood only in special circumstances, which allowed the object of possession (eye) to be considered relatively independent. When for the interpretation of the (50) sentence we take the construction of the conjunction Onutés akys yra Žalios, then in the latter’s respect it may have a double significance. Since the author of the article discussed (50) the neutrality of the sentence in terms of intonation and contrasting cross-section, it is most acceptable to consider it as a connection construction with the iSkeltine theme. However, the fact that the word order of such sentences as (50) also appears to be neutral makes it necessary not to overexpose the structure of its origin and to seek other than the solution of the subject motivation to justify this order. 5.6.1. If (50) the word order of the sentence is considered to be nezyméta, then it can be based only on its constituent structure, in which the genitive of the beginning noun (Onutes) is no longer considered to be part of the possessive noun compound. The analysis of possible interpretations showed that the likelihood of such a sentence division is related to the possibility of conceptualizing jj, on the one hand, as an equative or as ascriptive construction, and on the other hand, as a possessive possession construction. Alternative directions of conceptualization of the sentence under consideration also presuppose the distribution of alternative grammatical functions in it. 49 Rotanpas MikuLsKkas 5.6.2. If it is correctly stated that the main purpose of sentences such as (50) is to define a person by attributes that are proper to him, then an equative or an ascriptive interpretation should be considered more acceptable. The characteristics of such an interpretation of the functions of language allow us to speak about a distinct type of semantic sentences. However, in a favourable context, the possibility of possessive interpretation of such sentences cannot be ruled out. It is also difficult not to notice the formal inadequacy of the sentence for both interpretations. In the first case, the genitive of the noun (Onutés) is the subject of the sentence, in the second case it is the possessive predicate argument. Both the subject and possession constructions posesoriy encoded in the genitive for the active sentence of Lithuanian language néra bidinga. 5.6.3. Almost all the aforementioned (50) possibilities of interpretation of a sentence are discussed in Holvoet's article, but in search of the only correct syntax, the structure of such a sentence is ambiguous. It jo interpretacijos jis padaro metodologiniu podiiiriu neteisinga iSvada, seems to be more correct to assume that by choosing one or another direction of conceptualization of these sentences, their syntactic structure becomes unambiguous. Furthermore, if there is any basis to talk about a separate type of the sentences discussed, it can be called hybrid not because the whole construction can be defined as an intermediate between the possession and belonging constructions* (ibid., 158) but because the inherited form of these sentences does not correspond to the new trends in their conceptualization. LITERATURA Bocustawskt, A. 1977: Problems of the Thematic-Rhematic Structure of Sentences. Warsaw: University of Warsaw Press. Downine, P., Noonan, M., eds., 1995: Word Order in Discourse. Amsterdam: John Benjamins. Givon, T. 1984: Syntax: A Functional-Typological Introduction. Vol. 1. Amsterdam: John Benjamins. Gotppere, A. E., ed., 1996: Conceptual Structure, Discourse and Language. Stanford, CA: Center for the Study of Language and Information. 50 Syntactic ambiguity Forrest, L.B. 1996: Discourse Goals and Attentional Process in Sentence Production: The Dynamic Construal of Events. In: Gotppere, ed., 1996, 149-161. Hotvoet, A. 2003: Aktualiosios splitting problems. In: Ho.vorr & Jupzentis, ed., 2003, 79-98. Hotvorr, A. 2005: Intranzitive sentence types: existential, locative ir possessive sentences. In: Hotvorr & Mikuskas, red., 2005, 139-160. Hotvoer, A., Crztk-ProKaseva, V. 2005: Additional ir circumstances. In: Ho.vorr & Mikutskas, red., 2005, 65-92. Hotvoet, A., Jupzentis, A., red., Sintaksiniy 2003: ry3iy tyrimai. Vilnius: Lithuanian Language Institute. Hotvoet, A., Mixutskas, R., red., Gramatiniy 1.) funkcijy tyrimai. 2005: Vilnius: Lithuanian Language Institute. Hotvoer, A., Mikutsxas, R., red., The study of noun 3.) compounds. 2006: Vilnius: Lithuanian Language Institute. (Lietuviy kalbos gramatikos darbai, 4.) Houvoer, A., TAMULIONIENE, A. 2006: Description category. In: Ho.vorr & Mikutsxas, ed., 2006, 11-32. Hupson, R. A. 1984: Word Grammar. Oxford: Basil Blackwell. Jackenporr, R. 1977: X-Bar Syntax: A Study of Phrase Structure. Cambridge, MA / London: MIT Press. Laxorr, G. 1969: On Derivational Constraints. Papers from the Fifth Regional Meeting of the Chicago Linguistic Society. Chicago: Chicago Linguistic Society, 117139. Laxorr, G. 1970: Repartee, or a reply to "Negation, conjunction and quantifiers". Foundations of Language 6, 389-422. Laxorr, G. 1987: Women, Fire, and Dangerous Things: What Categories Reveal about the Everyone. Chicago / London: University of Chicago Press. Lancacker, R. W. 1991: Foundations of Cognitive Grammar. Vol. 2, Descriptive Application. Stanford, CA: and Stanford University Press. 51 Rotanpas MikuLskas Lancacker, R. W. 2000: Grammar Conceptualization. Berlin / New York: Mouton de Gruyter. LANGACKER, R. W. 2001: Topic, Subject, and Possessor. In: Simonsen & ENDRESEN, eds., 2001, 11-48. Lanoacker, R. W. 2002: Concept, Image, and Symbol: The Cognitive Base of Grammar. Berlin /New York: Mouton de Gruyter. Li, Cu. N., ed., 1976: Subject and Topic. New York: Academic Press. Li, Cu. N., Tuomson, S. A. 1976: Subject and Topic: A New Typology of Language. In: Li, ed., 1976, 457-489. Lyons, Cur. 1999: Definiteness. Cambridge: Cambridge University Press. Keenan, E. L., Comrie, B. 1977: Accessibility and Universal Grammar. Linguistic Inquiry 8 (1), 63-99. Mixuiskas, R. 2006: A descriptive bicolor perspective. In: Hot feed & Mixuisxas, red., 2006, 33-65. Nuyts, J., Peperson, E., eds., Language and 1997: Conceptualization. Cambridge; Cambridge University Press. Simonsen, H. G., Business, R. Tu., eds., 2001: A Cognitive Approach to the Verb: Morphological and Constructional Perspectives. Berlin / New York: Mouton de Gruyter. Suxys, J. 1998: "Lietuvių kalbos linksniuotieir ji prieveiksmiir ai: vartojimo norma". Kaunas: Light. Taytor, J. R. 2002: Cognitive Grammar. Oxford: Oxford University Press. Tomuin, R. S. 1995: Focal Attention, Voice, and Word Order. In: Downinc & Noonan, eds., 1995, 517-554. Tomuin, R. S. 1997: Mapping Conceptual Representations into Linguistic Representations: The Role of Attention in Grammar. In: Nuyts & PEDERSON, eds., 1997, 162-189, 52 Syntactic ambiguity van Hock, K. 1995: Conceptual Reference Points: A Cognitive Grammar Account of Pronominal Anaphora Constraints. Language 7 (2), 310-340. vaAN Corner, K, 1997: Anaphora and Conceptual Structure. Chicago: University of Chicago Press. Rolandas Mikulskas Lithuanian Language Institute P. Vileisio 5, 10308 Vilnius, Lithuania mikulskai@takas. It 53