Jungties konstrukcijos ir jų gramatinis kontekstas

Acta Linguistics Lithuanica LXI (2009), 113 - 156 Grammatical context of the ir ju conjunction construction ROLANDAS MIKULSKAS Lietuviy kalbos institutas The aim of the article of is to contribute to a satisfactory classification of copular constructions. The author starts out pointing out verbocentric by theory grammar yields that a little for adequate analysis of copular and related constructions. an of A survey the of functions performed the verb by ‘be’ the in constructions under consideration shows that there is a semantic interdependence between verb meaning and the syntactic frame (argument structure). In view this of interdependence construction a grammar cognitive or grammar approach seems most promising. The author argues that little is to be gained from a proliferation of lexical meanings for the verb ‘be’ as a means of accounting for the properties of the individual constructions; rather, the focus should be on the aspects syntactic and of semantic structure most crucial the that are predication, and their to correlation with the function the on whole construction. of More specifically, with reference to copular constructions, particular attention should paid the be semantic to and/or referential the properties of (adjectival or substantival) nominal predicates. 1. IZANGA: CONSTRUCTIONS SU VEIKSMAZODZIU BUTI SYNTAKSINS ANALYSIS SUNKUMAI’ 1.0. Conjunction constructions are usually called such sentences as John is a litigation, John is a teacher ir John is the teacher who instilled in us the love for our mother tongue. Further dividing the types, the j first sentences are assigned by du some authors to the ascriptive type, the last to the equative type of connection constructions o — (plg. Bolvoet & Cizik-ProkaSeva 2005: 79). In order to achieve grammatical integrity of the aprag, the construction of the connection is often discussed- ‘NuofirdZiai thank you Kristina Lenartaite, carefully read the first version of this article for your valuable comments. Of course, I am ir responsible for the fragility of the article uz myself. 113 ROLANDAS MIKULSKAS mos alongside other constructions of similar structure, in which the predicate a8j also forms the finite verb bit form. Usually it is a locative construction is on the desk by the book; (Sqsiuvinis comitatyvinés jelent’més / annak anyaga John is su Birute), existential construction (Stalciuje / next to the book is sqsiuvinis), possessive possessive constructions ir (respectively priklausymo The child has a runny nose temperatiiros / kepuré is ir Si Antano), active perfect construction (I haven't tasted varliy in my life), passive (perfect imperfect) constructions ir (The city was completely destroyed during the night attack; Teacher John visy childhood was loved) ir some other Variations of this construction. In the literature there are various typological parallels between the latter type of constructor and the ir These structures are compared with the su above-mentioned connection structures. For example, the common ontology of the existential-loir cative construction is not in doubt. (plg. Bolvoet 2005: 139-147). Su These two types do not represent actual divisions, verbs are necessary, and finally, in ir terms of syntax, the interpretacijos other mentioned types of possessive du possessive — constructions ir are — comparable. (ibid., Some 147-153). cases with active and passive ir constructs and su predicative participants can deservedly be regarded as a binomial variant of the ascriptive type of conjunction construction) su predikatyvu (plg. John is always neatly dressed ir John is always twarking; The apartment is completely abandoned ir The apartment is poor). 1.1. When sentences are understood as a certain content of a proposition, the distinction between their types is made primarily by the designation of the proposition, or its function. For example, j ju individual intransitive sentence types are intended to affirm the subject-noun-- referent experienced bisen (Bulvés boils), existence (Riisyje still remained bulviy), statine padéti (location) (The book is on the table), to assign to that referent some characteristic (Room is early) etc. etc. ir In terms of syntactic construction types, we first rely on the syntactic interpretation of the sentences that instance these constructions. The sentence type understood in this way can be realized by several syntactic constructions, for example, a possessive possession relationship between two entities affirming proposition can be realized by two constructions — [NP,,,, has NP... One! it ENP, 4p ya NProws prices» Plg. Child has slogq temperatiiros / ir The child has a runny nose temperatures. / Dependency Grammar Model The linguists, who exist in the verb-centric model of the sentence, in interpreting the syntax of the 114 Jungties konstrukcijos ir jy gramatinis kontekstas constructions rely on the dependency relationships of the sentence action semantics, on the basis of which they define the grammatical functions of syntactic units. Such a syntactic analysis is more difficult to implement in cases where the verb of the sentence does not have rich semantics and various elements complementing the semantics of the verb are used to express the predication of the sentence. The difficulty of syntactic interpretation is inevitable when analyzing sentences with verbs, the meaning of which, even in its most general features, is often difficult to understand. For example, in the works of Lithuanian grammar, when speaking about conjunction constructions, the meaning of identify between two (referenti8ky) entities (which in essence is the same as the function of the conjunction in the logical structure of the proposition) is recognized only for the verb biti of the equivalence construction, while the verb biiti of the ascriptive constructions (as they are understood) is considered to be »semantically identical« (see Holvoet & JudZentis 2003: 32; Holvoet & CizikProkaSeva 2005: 79-80). It is understood that going along the path of this analysis of the equative construction John is the person you talked about, the complement of the verb of “identity” that person you talked about has to be assigned the status of argument as well as the subject noun John. It is much more difficult, following this logic, to recognize the status of argument of ascriptive constructions, "semantically clear" connections of binominal or noun complements, especially since they, as is recognized, are more suitable to be called predicates. In order to ensure the consistency of the construction of the connection, the authors of the Lithuanian language grammar have recognized certain,,equative-type meanings“ for ascriptive constructions of verbs (Holvoet & Cizik-Prokageva 2005: 80). Intuitively, the decision to assign a certain need for “identity” to the verbs in the whole construction of the connection seems justified (in the third part of the article I will go to this problem and try to justify this intuition more thoroughly). However, the aforementioned authors are silent about the consequences of such a decision: it remains unclear whether complements in newly invented,,identity“ predicates in ascriptive constructions are,,lifted“ to the status of arguments, or whether they remain predicates. 1,2. Similar syntax interpretation is followed in Lithuanian grammar works when comparing the sentences of existential and locative constructions Stalciuje yra sqsiuvinis and Sqsiuvinis yra staléiuje (Holvoet 2005: 139-147). In the first sentence, the verb biti acknowledges the 115 ROLANDAS MIKULSKAS meaning of,,existence”, and the locative becomes,,a peripheral circumstance”. In the second sentence, in the absence of any basis to assign any meaning to the verb word to be, the same locative inevitably becomes a “biological circumstance”, because without it the main predicate of the sentence would not be played (more precisely, that predicate requires the locative morpheme of the verb word complementing the noun). Since the prediction of the existence of any real object is difficult to conceive without its location, intuitively such a syntactic analysis seems to be inconsistent with the nature of the real thing. And the same distribution of grammatical functions in both types of constructions seems to emanate from the semantics of the same action, the affirmation of the existence of the localized thing and the localization of the existing thing. In other words, such an action of both constructions — existential and locative — is attributable to the need for a ‘localized/localizable existence’. 1.3. According to the same logic, in the syntax of Lithuanian grammar works, when interpreting the (old-fashioned, but still used in Lithuanian) possessive construction of the mihi est type (He was only seven years old) and the belonging construction (This little boy is mine), these two possessive constructions have to be evaluated differently from the point of view of typology. The action of the first construction being to acknowledge,,existence* means that the noun encoded by the dative has the status of argument to its possessor and $i possessive construction is considered a variant of the existential construction. The action of the second, possessive, construction, being encoded by its genitive possessoriy, I have to consider as a,,bity modifier”, thus equating the noun with the,,bity circumstance* of the locative construction in terms of its grammatical function and syntactic status, —due to such syntactic interpretation the construction in question becomes closer to the locative, rather than the possessive type of construction (see Holvoet 2005: 151-153). It is reasonable to ask whether the parallels drawn during this syntactic interpretation between apparently different purpose constructions help to properly assess their typological dependence. After all, the analysis of the interpretative possibilities of such sentences as Onutes's Green Eyes shows that, under favorable contextual circumstances, the possessive genitive can also have the status of argument (Mikulskas 2006b: 41-50). About the fact that genitive of,,iskeltinio“ be possessor of loss galéty replace with,,iskeltinio“ possessor of dative (i§ esmés sutampantj senovinés mihi su est type construction dative), speaks and Holvoets (ibid., 159), discussing the dialect-fixed example for Bijiinéliui Zali lapai, cf. 116 Jungties konstrukcijos ir jy gramatinis kontekstas Green leaves of Bijiinélio. Indeed, although the genitive and dative in Lithuanian form a fairly clear opposition in functional terms, in some syntactic contexts these syllables encode very similar meanings. For example, the dative sign is transcoded j the corresponding genitive posesoriy, e.g. This room is for you/you. (situation when allocating rooms to guests). Even the adnominal genitive can in some cases predikatinio beneficiento vaidmuo be didelio reik’més nuostolio gali biti be interpreted as iSorin (sentence part), 0 not as internal (noun part) posesorius, cf. The professor's portfolio lay on the table and the Assistant supported the professor's /professorio- 's portfolio, —only in the first sentence the attributive genitive status is not in question. Such contextual approximation of the genitive and dative functions is not surprising: after all, the dative roles — recipient, addressee, beneficial and experiencer — which imply indirect participation in the event profiled by the sentences, can easily be extended to the role of possessor?. Although the functional approximation of genitive and dative, which only occurs in specific syntactical contexts, is rather marginal in language, it may prompt the genitive of the mentioned possessive possession construction (Sis rasiklis is my) to be interpreted as iSorinj posesoriy, thus acknowledging its argument status. Such an analysis of this construction would be desirable, because then we could consider both the constructs of possession and belonging as two different instances of the same possessive relationship. Unfortunately, as we will see later (section 2.2.2.1), after a closer look at the structure and genesis of the construction of belonging, this desired direction of its analysis seems unlikely: the predicative function of the genitive and its attributive nature in the construction in question do not allow it to be easily considered as an argument for the action of the word, and it is worth looking for other alternatives for the interpretation of this genitive. 2 An important argument for the functional proximity of genitive and dative in Indo-European languages is provided by Geoffrey Haig 18in his hypothesis about the origin of past tense ergative constructions in Iranian languages. According to the linguist Shi, the ergative alignment construct in the past time evolved from the old Persian mand kartam type construction (long discussed and interpreted in literature), which in the 15th century was an External Possessor Construction. Batent The genitive with the iSorin argument status in the construction (mand) natiralia: perémé senovés Persian dative verb functions, primarily benefactive (Haig 2008: 41-88). 117 ROLANDAS MIKULSKAS 1.4.These short problems, which arise when applying verbocentric analysis method to constructions with verbs, show that in order to adequately define such sentences by typological dependence, verb semantics alone is not sufficient. The structural similarities generated during such analysis between the different types of spoken constructions can sometimes even obscure the functions performed by these constructions in the language. In order that the typological characteristics attributed to these constructions correspond more closely to their intuitively perceived purpose, more attention should be paid to the semantic nature of their verb complements and the referential properties of the nouns used in them. To define one or another type of construction helps the construction of instanced sentences in the Zodziy order and the corresponding non-Zyméta actual section. All these things in one way or another reflect the distinct types of constructions assigned in ci8kai motyvuoti. Kalbant apie aktualiosios sakinio skaidos svarba sakiniy the language, in other words, are functiontypology, it should be noted that the actual division correlating only with the preverbal noun-word definition and the postverbal noun-word undefinition is to be considered as undefined. Only such an actual division can be considered as a typologically classifiable property of the semantic structure of sentences (see Mikulskas 2006b: 34-41 for the justification of this idea). The third part of this article presents the typology of the conjunction constructions and will be based not so much on the syntactic interpretation derived from the semantics of their verbs, as on the revelation of the predicative nature of semantics and the functional analysis of the referential properties of nouns used in this position. But before that, in the second part, we will try to look deeper into the semantics of the verb beiti used in all the above mentioned constructions. 2. SEMANTIC PROPERTIES OF THE ACTION 2.0. Only by a more thorough examination of the semantic properties of the verb biti used in the above-mentioned constructions, we will see more clearly to what extent this action word galéty biti is considered to be a determining factor in the construction syntax (i.e. argument structure), or in terms of formal semantics — a semantic function (this term is widely used by the authors of the work on the grammar of the Lithuanian language, see Hovoet & JudZentis 2003: 15t.; Holvoet & Seméniené 2004: 15; Hovoet & JudZentis 2005: 16t.). Undoubtedly, the term semantic functionator is applied to the verb of the sentence. 118 The construction of the conjunction and the yy nuoja verbocentrinj sakinio modelj. Sio modelio taikymo Salininkai remiasi prielaida, kad sakinio veiksmaZodj subkategorizuojanti argumenty struktira grammatical context (subcategorization frame) is derived (i$introduced) i8 its so-called lexical meaning. It is understood that the application of the verb-centric model in syntactic analysis is problematic when the lexical meaning of the verb is difficult to define (as we have seen in the example discussed earlier). But in all cases, when the action word of a sentence appears to be full-fledged from a semantic point of view, can those features of the syntactic structure, which ultimately determine the typological dependence of that sentence, be derived from its meaning? In other words, does the concept of the sentence action as a semantic function, based on syntactic analysis, adequately reveal the structural and semantic characteristics of the construction that the sentence instances? 2.1. In language it is not difficult to find constructions in which to call verbs semantic functions is inaccurate, because the,,subcategorizating argumenty structure“ of those verbs is not 100 percent defined by semantics. A good example in this regard is the so-called locative alternances: we could reasonably say that argumenty configurations in these alternating constructions largely depend on which element of the situation being profiled is encoded as a direct object, 0 already the action word meaning is adjusted according to the corresponding construction (by saying this, together with Adele Goldberg (1995) we recognized the specificity of the construction as a syntactic structure and the superiority of that need to the meaning of the sentence verb), e.g.: (1) Tévas su sunumi (per dienq) malkas sukrové j kiigi (i.e. a certain cut-off (wrong) quantity of wood was cut). (2) Tévas su tunumi (per dienq) sugar malky kigi. The verbs in this construction can be defined as (su)loadi,,(- pabirus daiktus) kiigio tvarka (su)délioti‘ and (su)loadi,,GiS pabiry daikty) (su)formuoti kiigj‘ (the possibility of inserting in both sentences a time circumstance modifying the duration of the loading action shows that this action is not instantaneous, its result is not achieved at once, but gradually, gradually forming and increasing the malky kigi). Thus, when considering these alternating constructions, it would be more accurate to talk about their verbs and corresponding argument structures linked by a mutual but not unilateral relationship of dependence, i.e. not a relationship of head: dependent (see Croft 2001: 248). Of course, we should not forget that in both 119 ROLANDAS MIKULSKAS constructions there are two different realizations of one, more general, need. The basis for postulating such a common meaning arises when we realize that the semantic contents conveyed in the alternating vartojamy veiksmazodziy krauti, ir krauti, reik’més pagrista gali biti laiconstructions are different aspects (or profiles) of the same (smaller load) scenario. It is the general meaning (ja Seizi Iwata, distinguishing it from the phrases of general meaning (angl. "Silver-plated sculpture of the "Silver-plated Buddha" (Silver-plated Buddha)". Lexical Head Level Meaning); Zr. Iwata 2005: 361-363) includes the bit-level elements of that loading scenario: loading, loading object, loadable form, and the associated more general semantic roles — agent, muver, (destination) locator. 2.1.1. Even more difficult is the application of semantic functionality to English so-called transitive (double object) constructions and to verbs with alterning causative movement constructions, e.g.: (3) She gave her boyfriend a new Mercedes. (4) She gave a new Mercedes to her boyfriend. It is difficult to argue that the arguments for structural changes in these two constructions and the related differences in constructive meaning are introduced in the same way as in the action to give. We could reasonably argue that the differences in meaning in the constructions in question are determined by different structural frameworks, in which the same predicate is used, respectively [give NP NP] and [give NP [to NP]]: In these structures the place of the direct object is taken by nouns representing different semantic roles, respectively recipient or receiver. Of course, in order to be precise, we should admit that the meaning of the verb to give in the constructions under consideration depends to some extent on a specific, subcategorizating rém for each i§ jy badingo“. This dependency is particularly significant when the constructs in question use verbs that are more closely related to the trajectory of the transmitted object, e.g. (5S) They sent me another brochure. [send, NP NP] (6) We sent the new letter to all the applicants. [send, NP [to NP]] We could argue that in the meaning of the predicate send, the semantic component of the effect on the recipient (his personal sphere) is more pronounced, because the sentence prorecipient, 0 in the meaning of the predicate send, is profiled in the sentences of the same situation by selecting a move, pabréZiama its trajectory of movement (toward the recipient). In this case, however, we could assume that in the mind of the user of language there really exists a more general send meaning that encompasses both meanings, meaning the agent-caused movement of the mover into the sphere of the recipient. But it is hardly purposeful to assert that such an abstract need sanctiones the argumentative structure of one or another construction. On the other hand, since the semantic variants of the verb to send (send, and send,) in the constructs under discussion are only relevant to the extent that they are supported by different structural frameworks of the construct, these predicates can only be called semantic functions with large iConditions (see Langacker 2008: 239-244). It seems more accurate to say that the lexical meaning of the verb to send (defining a certain semantic-interpretative frame) is compatible with the semantic 120 Jungties konstrukcijos ir jy gramatinis kontekstas filiuojamos situacijos riboZenkliu (tiesioginiu objektu) pasirenkamas pats specifications of both constructions. 2.1.2. Furthermore, trying to define the verb bit in various constructions may have a meaning, we will discuss the practice of syntactic analysis of alternating constructions. We do this for two reasons. First, it is obvious that the meanings that can be attributed to the verb in question depend greatly on the type of constructions in which it is used. Secondly, some constructions with the verb be can be deservedly considered alternating: for example, we could think that the raison d’être of the coexistence of two constructions — existential and localising — is to bring out different aspects of the essence of that actual situation of “localized existence”. By grouping other spoken constructions on a scientific basis, we can assign more general meanings to the action words of such groups, thus avoiding the postulation of,,more specific“ lexical elements in cases where a more specific action word is more difficult to define. 2.2. Let us proceed to a more detailed discussion of the verb bit meaning, which it has in the above mentioned constructions. It is difficult to imagine how one or another principle is chosen in accordance with the basic postulate of cognitive grammar —the Contentual requirement (or also the Symbolic theses) —to operate in the grammatical apraxis of the brotherhood with unnecessary units of language (see further Zr. Langacker 1987: the approaching is not surprising, after all, some constructions of verbs have facts when they affirm the possession of something, but about the presence of something in someone's personal sphere, cf. : 97–98; Taylor 2002: 38–60; plg. Mikulskas 2006a: 40). 121 ROLANDAS MIKULSKAS zicijy ir su ne ko ko (16) At home we have peliy. The final part sakinyje i8 of the verb is implied ,,owners“ we (along with su their living environment at home) — first helps lokalizuopeliy location, so, as in the case ti of many other posesyvinése constructions, it ir factuSkai teatveic references ta8ko (reference point) function (more about The role of this cognitive property in the processes of perception ir and its importance in possessive construction ontology Zr. Langacker 2000: 171-202). It is important to note that the locative in such types of sentences (16), as is a byte element, or Sios (actually existential) nature of the construction: it specifies the scope of validity of the na reference point named by the processor's nouns, or otherwise defined field of cognitive search. — User In this construction, the biting locative is more ir existential than manyti, possessive, ji especially since we cannot consider such a construction as locative either: after all, the At home we have peliy to say almost ta the same, ka ir existential sentence There is peliy at home, but in the first sentence the predicate turéti additionally indicates the presence of peliy ,,Possessor“ (mes) in the living environment, in other words, the presence of peliy in the house for ju residents in a sense is ,,actual’, we would say, it is directed to j ju the personal sphere. So We have a friend. The affirmation of,,intentional” existence is as in the previously ir discussed mihi type construction: est only in one case the original possessive construction implies about i8 the affirmation of existence, in another case the inverse (16) original existential construction implies a possessive o relationship. Please note that biitent — i8 (12) These implications, o ne the need for the verbs used in the constructions, the "lexics" determine the jy linguistic function by ir typological dependence. This fact persuades the mind that when determining the significance of ar one other construction tipa, be in it used very little, less my important becomes the ne direction of conceptualization of the construction (i.e., its common use, ir or function) determining other semantic elements, such as the duration modifiers of the place of action, which are similarly ar ir Salininky is daznai the periphery of the Finnish preposition. South Tokyo ,,periferiniy* elementy of importance for the alternative interpretation of sudétiniy predikaty together ir 128 Jungties konstrukcijos ir jy gramatinis kontekstas tions conceptualization, we will have to look further at the analysis of constructions with predicative participants (see 2.5.1.1., 2.5.2.). 2.2.2.1. It seems that there is no more serious basis for interpreting the genitive of the possessive construction (13) as iSorinj (Antano) posesoriy, thus giving it the status of the verb's biiti argument: it is obvious that the genitive su This action word forms the predicate of the subordinate construction. Nesinoréty Sios construction actionto assign a word to a metonymy ir ,belonging’ meaning, because in ja another type of belonging construction explicitly The word belongs to the same action. As already mentioned, the literature refers to the predicate of the belonging construction biti gen. su + structural panaSumas j previously aptarta locative construction su of the predicate biti + lok. It is noted that, first, both constructions are linked by a similar model of actual division: both possessive and locative nouns are said ir Siu construction remose; secondly, the main predicate constructiontenka svoris abiejose actionZodzZio se biti nounZodni complement indirect verb (or corresponding prefix) morphemes: in the first case possessive genitive, in the second — case locative (or a locative prefix). Structural parallelism of both constructions will be clearly seen if we draw attention to the fact that the main predicate in démesj j them is The corresponding meanings of (or nearby) morphemes are inherent: as the location function is vo inseparable from the indication of the place, so the genitive function is inseparable ir from the function of kinship, property (belonging), totality, or more generally dalies / from the possessive relationship. — The question arises, could ar we find a proper semantic explanation for the j structural similarity of the construction of belonging to the localative construction? In other words, is there a more ar serious functional and cognitive basis for interpreting the possessive construction,,locatively*? After all, if the construction of belonging could be regarded as a variant of the locative construction, then its verb form could ir easily be assigned the meaning of location, as, by be the way, the It is a homonymous form of apraea. I8tirkime This opportunity. Remembering that possessive relations with the genitive are most commonly expressed in the attributive noun compound, we could even say that in the construct of belonging that we have, the genitive ir in the predicative position is taken over by the corresponding attributive noun compound, vartoi§ t. y. Antano kepuré > Si The analysis of Antano. Cognitive 129 ROLANDAS MIKULSKAS the possessive noun in grammar is based on the reference point (Langacker 1991: 167-180). The essence of this analysis consists of the assumption that in possessive noun combinations the genitive (-’s morpheme or of prefix) encodes not only and not so much the possessor, but a cognitively more elevated entity, which facilitates the identification of another, communicants’ object, entity. Thus, the so-called possessive construction is also used in language for the reference needs of nouns: the denotation of the genitive in it often refers not so much to the possessive relationship as to facilitate the recognition of the other noun-word referent (,,possession“) (in fact, in the sentence Someone stole the boy’s bicycle, it is invariably spoken of the boy’s bicycle — it was a bicycle he had borrowed from a friend; plg. also in this sentence: My train arrives exactly ten minutes later. It is important to note that, although the cognitive attribute of a reference is significant for more than one grammatical construction, it can be defined independently of the linguistic expression, because it reflects the general specificity of the perception process: some (less familiar) essentials are usually recognized through others (more familiar to us essentials), which are one way or another related to the first ones and can point to them. In order to interpret the function of the reference point »locatively- *, it is not at all difficult, especially when the possessive predicate content consists of the relationship of the whole and the part, the place meaning of the genitively encoded noun and the function of the reference point in the possessive construction naturally coincide, e.g. the window of the living room, the edge of the table, etc. Considering the cognitive function of the possessive construct of the noun, we could argue that,,possession” in it is naturally located in the cognitively more accessible,- ,possessor*’s* environment. It must be assumed that in the predicative position neither the cognitive reference point functions. Still, the purpose of the construction in question — not the clarification of the priklausymo santykj reikSdamas genityvas visiSkai nenustoja Sios bendrespossessive reference — is that it is often interpreted as a description of the subject by means of words, so that its referent is identified, at least in the actual space of the discourse. The reading that kepuré (13) in the sentence is localized in Anton's view bitty is considerably strained. And this is understandable: although the genitive of the belonging construction may be etymologically related to the genitive of the corresponding attributive compound, the purpose of the construction itself in language is different. Usually, $i construction is used when it is necessary to specify the owner of the object already known to the communicators. Dél to posesumas in the ownership construction prototipi8kai esti 130 The construct of the conjunction and the grammatical context yy are inanimate object, 0 possessor —person or (more rarely) animal (see This leash is miisy Pifo). The function of the spoken construction ikoniSkai is reflected and by it instancijuojanciy sentenceiy nezyméta actual part: in the rem the communicants are told the owner of the object. The semantic limitations of this construction also show that the possessive relationship expressed in the construction of belonging has a different purpose in language than in attributive noun combinations. The construction of belonging usually does not refer to the so-called inalienable possession, or the relationship between part and whole, cf. Your eyes are beautiful, but? “The beautiful eyes are yours.” And it is precisely the genitive of the attributive noun compound, which profiles the relationship of part and whole, as we have seen, that can naturally be interpreted,,locatively* — as a reference to the object that is located in the space. All this shows that neither the functional nor the cognitive basis of the possessor’s genitive of the belonging construction can be interpreted as the reference point of the location of the subject’s noun-denoted possession. It is precisely dél Siy semantic motif that we cannot consider as a variant of the locative construction. Then the action of this construction is to assign to the word biti a metonymy independent, belonging ‘meaning’, although, as we have seen, in language we have In this case, we need independently, without construction support, able to convey the verb belong. But such a synonymy of the verbs used in different constructions is unfavorable only to those who persist in maintaining the thesis of the verb as a semantic function of the sentence. 2.3.The semantic characteristics of matome, kad Sio veiksmazodzio predikatinis potencialas teikia menka pathe verb beiti in the constructions (7)-(13) are discussed, which form the basis for typological classification of the constructions. Tik constructions (11),? South is limited by such a saying in the language possibilité Zr. Mikulskas (2006b. 42-43). ®Of course, if we looked closer at the ontology of the construction of belonging, the meaning of,belonging‘ assigned to its verb,,ad hoc‘, we could also consider the meaning of one of the basic needs of this verb, the meaning of,identity‘, in well-proven constructions of connection, a certain variant or instance, cf. Si hapure is a dependent of Antan, —this extension of the construction (13) is not entirely arbitrary, because it fits well with the attributive origin of its genitive. On the other hand, too much emphasis is placed on the ontological parallelism of the possessive spoken construction, which obscures the attribution of the corresponding syntactic and semantic structure aspects of the construction, which define its identity. 131 ROLANDAS MIKULSKAS (12) ,existential* predicates could be recognized as having a relational independentness, allowing complements to be assigned ju arguments of status®. Other constructs of verbs discussed are too dependent on the type of the construction in which they are used. No The word bit is present in the se following constructions ‘identity’, ‘slocation’, or the meaning of,belonging‘ is not so much that it can be assigned some savarankigka, argument structure (,,subcategorizing i8 réma*) alone — This verb can hardly be regarded as a semantic function of the finite sentence of the syntactic structure, as it — is often done, when analyzing — sentences, prada forms more contentful lexical use verbs. kuriy Apparently, in the constructions ir of identity, locality and belonging, it makes sense to talk only about Syntactic distribution of CONJECTIVE PREDICTS. In fact, a certain position of the subject noun in an ascribed (ascriptive) construction, for example, The student is right, is the predicative binominal that represents the position, the governing conjunction ta determines what semantic characteristics the noun can have in the o ne 3 position of the subject. This means that In these similar ir constructions, the complementation relationship of the subject noun (pupil) directly connects the action to the word ne biti, but with all su the sudétic predicates su (bit right). Relevant constituencies (démetvarka of sentence i§ formation) ny such analysis of spoken constructions syntaksés represents formally, p.g. [Student,,,., [is [right]],,.,], or INP..,, [V [Adj]] 21. Obviously, by analyzing the identity constructions of the sentence prada (English) clausal head) formed verbs, we will consider the verb bit of the conjunction, ne but the whole substantive structure of this verb is formed by a binominal noun complement. su ar In fact,,,connection* actionazodj biti In the context of this syntax analysis, it is not very appropriair te to call (8in tradition we use termina borrowed identity i§ propositions, as logical structures, analyses), because For the jo primary function of ne connecting du the essentials of the identity proposition, 0 (predicate) subjunctive ar nouns to give the predicate of the sentence processuality, or temporality, the dimension of the verb — 8i esminé property is necessary for the finite sentence (t. y. such a sentence, which by indicators of grammatical time, person, modality is linked to su the ‘Vis délto ir These predicates cannot be regarded as full-fledged semantic functions in the context of such arguments, because the existential interpretation jy depends largely on the pragmatic characteristics of the constructs in which they are used, such as The actual divisions are not listed in the ir ja atitinkanéios Zodziy order sentence. 132 Jungties konstrukcijos ir jy gramatinis kontekstas situation (,,ground“, angl. ground, make). In order to identify the word bit assigned in the constructs under discussion, we could call jj a grammatical specificator. In the present-day New Lithuanian grammar, the specificator is called the verb biti of the Sudetian predicate in the active perfect sentence, cf. I've never eaten Varliy in my life. Specifier term Gia is also applied to the analysis of other compounds, whose elements in terms of internal and external syntactic relations take different forms. Based on the criteria of Arnold Zwicky (Zwicky 1993: 292-315) for defining the function, base and origin, the New Grammar distinguishes between the inner origin of the compound and the external origin of the compound (or base) when describing the aforementioned compound syntax. The specificator plays a primary role within the compound (Holvoet & Mikulskas 2009a: 10-13). In the Lithuanian language, in the construction of the perfect form, the verb biti is usually assigned the function of an auxiliary verb, which in essence coincides with the definition of the grammatical specificator. Similarly, we could also call the auxiliary verb bits of passive (perfect and imperfect) constructions as specificators, e.g. The city was completely destroyed during the night attack; The house was still under construction when the accident happened. When we talk about identity constructions, we always stick to the traditional term of connection. Although the grammatical function of the verb is undeniable and can be compared with the function of the auxiliary verb in the aforementioned analytical forms, its structural function in identity constructions is also obvious — it allows us to talk about the relationship of such a verb’s autonomy with respect to the second deémen of the subjective predicate and to attribute to it the aforementioned need for identity. An additional argument for distinguishing the action verb from the auxiliary verb (specificator), as we will see below (sections 2.5.1.-2.5.2.1.), is provided by the possible alternative interpretation of constructions with predicative participants. Subjective predicates used in these »ambiguous« constructions allow cognitive analysis to distinguish two semantic structures of the verb being: one when it performs the function of an auxiliary verb Gia, and the other when it goes through the conjunction Ga. © I continue to use the term analytical form* of the verb to refer to verb-bit combinations with participants Gia because of its usualness, thereby not requiring my opinion about the paradigmatic status of this form in Lithuanian language (see otherwise Holvoet & Seméniené 2004: 36-41). 133 ROLANDAS MIKULSKAS 2.4. What has been said about the semantics of verb bits should be sufficient to conclude that we cannot judge a verb by the constructions in which it is used, i tokio so, by distinguishing the types of verb bits in a construction, we can barely judge by their semantic characteristics. ir su jo pasiremti. Veiksmazoar Thus, the need for construction absolutely or almost absolutely dominates in relation to the lexem. Using constructional grammar terms, spoken constructions are defined as syntactic galétume structures oriented to j the needs of discourse, Gias functionally motivated actuality ir division has a specific significance. In constructive grammar, the provision is observed that only those j verbs whose semantics, or more precisely, whose leksemos, — ty leksemy represented semantic-interpretative rhemes, allowing to do ta (Semantinés dermés principas (The Semantic Coherence Principle); Goldberg 1995: 50). Depending on the need (from the semantic-interpretative remo pobtidzio) esmés that can be integrated several iS constructions pats veiksmaZodis j (plg. 2.1.-2.1.1. sk. discussed alternating constructions). Since only predicates of a certain semantics can be used in a particular construction, From the i$ perspective of the verb, we could say that the variation is at ir konstrukcijos least partially dependent on i8 the necessity of the verb. In defining the verb beiti needimes we could hardly rely on any ar easily recognizable Structuring knowledge about certain nonverbal real situations by immediate semantic-interpretative frameworks. South pernelyg abstract reiki This is the action the possibilities of,semantic compatibility’ are extremely wide, specifically ir they are revealed only by the use of one ji ar In the construction discussed above. For example, the possibility of using the verb biti in an ascriptive equative construction ar speaks of the compatibility of meaning with a certain jo semantics of the identity su proposition, the possibility of using the verb biti in a locative possessive mihi 0 est type $i construction speaks of the compatibility of ar meaning with a certain semantics of the existence proposition. jo su tikra To be sure, it is necessary to note that there are certain possible changes in the internal semantic structure of verbs used in spoken constructions, only susije su ju the choice of grammatical form. These changes can even lead to a change in the morphosyntactic form of the complement, thus correcting the syntactic form of the construction itself, e.g. : ir 134 Jungties konstrukeijos ir jy gramatinis kontekstas (17) Stinau, be strong. (18) Bik Zmogus /Zmogumi, don't behave like that. The imperative verb used in these sentences determines not only their aspectual characteristics, but also their lexical requirements. This is the essence of the imperfective action. The word being form refers to an action of perfective nature (i.e. an action which in principle can be completed or whose duration can be limited). Dél to predicate bik Gia reiSkia active, subject control presupposing action. The meaning of this predicate in the examples discussed can be defined bitty as,behave as a certain bidu‘ and,behave as a kam‘ respectively. However, it is too bold to say that the change in the meaning of the action word, in connection with the choice of its grammatical form, fundamentally changes the structure of the argument of the construction and thus its typological dependence. 2.5 Far more serious consequences for the typological dependence of identity constructions may have an alternative interpretation of these subjective predicates. The literature has extensively discussed cases, when the participants forming the semantic basis of such predicates (mainly the perfective meaning) can be perceived in spoken constructions as both,,substantives*, and,,.verbs*. For example, in the works of Lithuanian grammar (Holvoet & Pajédiené 2004: 134) discussing two possible interpretations of the sentence I am sure that there is both good and bad Zmoniy, it is said that one way we understand $j sentence when we perceive the sentence as a binary predicate controlled by the conjunction I am, and another way — when we understand the sentence as an analytical action form of the word (which is often called a verb-action compound). So in one case we have an ascriptive type construction, in the other case — perfect construction. Let's look at such construction alternatives more closely. 2.5.1. First, let us discuss the use of the active participle in the constructions in question. Usually, the participative forms made in intransitive verbs can be interpreted here in two ways, because they can record the ultimate fear of the subjective noun referent who has experienced the effect of a certain action. This fixation of the subjective referent’s acquired belief forms the semantic basis for the,,binomial* interpretation of the spoken participle form. It should be noted that such a verb binominal can also occupy an attributive helper in a noun compound, which is common in this part of the language, cf. We do not need people who are certain of their righteousness.,,The internal semantic structure of the subjective predicate interpreted by the verbs* is certain and in cognitive terms we could describe it as follows. The perfect participle form profiles the event Gia only in the final (resultative) stage of the action of the event that has occurred: This profile can be compared to the profile of the other static relationship Zymin¢iy biidvardziy. The processual dimension to give this binomial predication of the content verb esu inner semantic structure essentially constitutes the same profile of the static relationship that is scanned over time (in the mind) and is perceived only abstractly (so this action word is an imperfective action). In the formation of 135 ROLANDAS MIKULSKAS a subject predicate, the specific content of the participle form semantically specifies, or elaborates, the corresponding abstract verb substructure of the word. The elaborated substructure (place of elaboration) is freely chosen, since all elements of the imperfect action word’s internal structure (all profiled aspects of the abstractly perceived static relationship) do not change over time. 2.5.1.1.,,The interpretation of the verbal participle form of a spoken word may lead to i8 the discovery of non-significant elements of a sentence, usually assigned to the periphery of its syntactic structure, cf. More than once I have been convinced that there are both good and bad Zmoniy. In this sentence, the adverbial modifier ne kartq, by indicating the repetition of the profiled event in the past as the subject predicate esu jsitikines, also indicates that each such recurring event had its beginning and end, i.e. a certain duration. In cognitive terms, the participant In this substantive form, all the stages of the action of the occurrence are profiled, which, by forming a substantive predicate, elaborate all the relevant elements of the inner structure of the action Word I am, they can be imagined as a sequence of scanned stages of a certain (only abstractly perceived) verb-verb relationship that changes over time. Such semantic structure of the action word biti, as befits an auxiliary verb, is indifferent to the imperfection of perfection. ir / Ar The subjective predicate of the perfective or imperfective action of the participle depends only on the aspectual nature of the verb that forms the basis of the participant's operative basis and semantic interpretation: for example, the predicate (not kartq) is the certain perfective action of the participle, and the predicate (not kartq) is the heard (tq melodijq) —imperfective action." Perfection / imperfection ¢here and elsewhere I understand langakeriSkai (Langacker 2002: 85-91), as an action characterizing the inner structure of the process of word forming, which is not confused with the spoken analytical forms of meaning. 136 Jungties konstrukcijos ir jy gramatinis kontekstas 2.5.2.,,Substantives* can be interpreted as participants of passive action in a passive construction. Then the $i construction i§ is essentially no different from the construction of the ascriptive conjunction. The position of the binomial predicate in the construction of the connection Participants of the passive connection, like other binomials, can also be used attributively, cf.: (19) We have seen that the city is completely destroyed. (20) The completely destroyed city looked terrible. (21) These mushrooms are edible. (22) We know edible mushrooms. »ActiveZodisks* interprets the active participants of the process and transforms the joint structure into a passive one. Such an interpretation, in addition to a favourable context, may be prompted by adverbial modifiers modifying the duration of the participant- 's assumed action", cf.: (23) The city was completely destroyed during the night attack. (24) Vélyvq autumn collected and stored, These mushrooms are eaten with delight in winter months. The compound predicates of these sentences were destroyed and are eaten are considered analytical forms of action-verbs (destroy and eat) formed on the basis of participatory action. The cognitive explanation of the formation of such a form may be similar to the previously presented explanation of the formation of analytical verb forms with the participants of the active line. First of all, let's note that perfect has its own character. The latter is usually characterized by the ability of the action-oriented phrase to record, by more or less dramatic means, the importance of the action that has taken place to the communicants. Since the category of perfectiveness/imperfectiveness defines the possibilities of expression of the aspect given to a specific action by the word, it cannot be identified with the category of the event/process factor that fixes the dichotomy of Sin possibility and realization (see Mikulskas 2005: 26-27). ‘Not with all ri8iy predicates such modifiers: may be used. For example, in the two sentences The village was surrounded on all sides by mixky and Béglys was finally surrounded by police only in the second sudétine action. The Zodine form was surrounded The duration of the action can be modified. Dél iSilginiy objecty predikation specifics (see Mikulskas 2005: 32-38) in the first sentence the same form can only be used in static space padétj (misky i3sidéstyma circle around the village). The interpretation of the semantic limitation of the "active participle" form in the spoken construction coincides in essence with its "adverbial" interpretation (about § rei8kinj plaiau Zr. Langacker 2002: 145-147). 137 ROLANDAS MIKULSKAS Perhaps therefore the first daZnas are intuitively assigned to the ascriptive type of the connection construction (cf. Holvoet & Cizik-ProkaSeva 2005: 79). [8ties, when the subject name word referent in such sentences is assigned to a broader referenty class, the metonimic because it can be associated with an essential member of that class. When the predicative noun in the inclusive type construction and the predicative adverb in the ascriptive type construction are connected by the relation of action, such a metonimic connection between the class member and its constituent feature becomes almost unavoidable, and the needs of both constructions differ only slightly, cf. : (36) Peter is a fool. (37) Peter is a fool. Thus, we can not in all cases find such pairs based on the action relationship, on the basis of which we would assign the reiSkian@- ias connection constructions in the inclusion relationship to the ascriptive type. Furthermore, it is difficult not to note the fact that the functional curve of the conjunction differs in the ascriptive type and the inclusive type of constructions: in the first type of construction, the function of this finite verb form is primarily grammatical, while in the second type of construction it is also relevant to the entire proposition in an ontologically important relation of identity”. After all, we will agree that the logical inclusion relationship is a variant of the more general identity relationship: the name-word referent is drawn into a broader class on the basis of some generalized characteristic (type description), identifying jj with other members of that class. Thus we came to the idea that the two previously distinguished types of conjunction sentences —inclusive and equative —are united not only by noun predicates, but also by a common ontology. It seems that conjunction sentences with noun predicates should be treated as one type. Therefore, the question remains unanswered, how nouns of non-relative nature are able to express the main predicate of such a sentence? In other words, where is their predicate potential? 1” Of course, therefore, analyzing the constructions of connections with noun predicates, we could assign to them the needs of,,inclusion” or,,identity”, but then we would distinguish them greatly from the analogy constructs with binomial predicates, in which the function of the connection is first recognized as grammatical. Thus, in order to treat all connection constructions as essentially the same structural archetype, we should in all cases consider the structural rather than the semantic function of the connection in them. 144 Jungties konstrukcijos ir jy gramatinis kontekstas 3.2.1. If the noun in the predicate position is perceived as an instance of the implied type* (and the instanced example always refers to its type), then the relative aspect is easily demonstrated: such nouns represent the relationship between the representative of the individual (instanced) type and the type itself (or between the respective descriptions). The effect of inclusion (and the predicative potential of the noun predicate) is created by the subject's nameword referent through the identity relationship with the type instance by involving it in the type description defined by the broader referenty circle. Dél understands that the reference of a predicative noun to the description of a type (and the relative subtype of that noun) is easier to understand in cases where it is interpreted in an undefined and non-specific way, i.e. as a noun denoting a freely chosen representative of a type’’. But also in the so-called equative construction the predicative nouns defined are defined, i.e. the noun referents are uniquely identifiable instances of their own type). Since in these constructions the type description provided by the predicative noun is a much narrower reference savaip nurodo j atitinkamy tipu deskripcijas (jau vien tik todél, kad pagal than in the case of indefiniteness (for example, in the sentence (34) it is provided by the entire restrictive compound teacher, who instilled in us love for the native language, which practically can only have one referent), the meaning of the inclusion relationship is more difficult to understand — it is reduced to a minimum. However, the difference between type description and instance of that type is significant for the relative notion of predicate noun, and can be easily understood, especially if the instance, as in the example (34), is explicitly indicated by the defined indicator that. Thus, in both cases we can speak about the relationship between the noun and its predicate, which arises from the position it takes in the syntax — the predicate. +8 Plagiau apie tipo ir jo instancijos skirt) 2r. Langacker (1991: 51-73); plg. Mikulskas (2006a: 34-35). In such a sentence as John is the teacher, we could also treat the identity relationship as a direct inclusion of the individual (John) in the more general class (the teacher), in which case the simple noun (not a noun!) teacher is named. This analysis is supported by some examples of article languages, cf. fr. Jean est professeur (but cf. John is a teacher), but it is more difficult to apply it to equative constructions, where the defined predicative noun j type description refers only indirectly, through the instance of a uniquely identifiable representative of that type (cf. Langacker 1991: 69). 145 ROLANDAS MIKULSKAS 3.2.2. A few words should be said about such sentences as (31), (32) and (33), which, depending on whether the predicative noun is interpreted definitely or undefinitely, can respectively be interpreted as requiring an equation or an inclusion relationship. IS eilés These examples will be discussed. The sentence (31) seems to have equal chances of being interpreted ambiguously. For the interpretation of this sentence, the idealized cognitive model of SCHOOL (ICM; Lakoff 1987: 68-76) is supported by the principle of a situation in which the possessor of the reference (school class) may have more than one teacher, so long as the possessive description of the teacher does not necessarily mean a uniquely identifiable individual. And the description miisy lietuviy kalbos mokytojas in the sentence (32) the same SCHOOL ICM is rather encouraged to interpret definitely, and the sentence then expresses an equative identity relationship, because the middle and upper school classes usually have one teacher of some discipline. Therefore, the referrer of the description mentioned in this sentence may not be uniquely identified, for example, if the possessor of the name is all the students of the school, and there may be several Lithuanian language teachers in the school. I§ The analysis of the first sentence (33) of Zwingli should not differ from the analysis of the sentence (32) discussed earlier. Therefore, it seems that the type of teacher mentioned in the example of the description mdsy lietuviy kalbos mokytojas (33) is rather perceived as a certain role that is performed or instanced by a single individual in a given period. This case is known in the literature as “role definition”, whose specificity —the possibility of a single referent —makes the sentence (33) closer to the equative construction. 3.2.3. Summarizing what has been discussed, we can distinguish two main types of connection constructions: ascriptive and identity type constructions. Their difference in meaning depends on what - adverbs or nouns - are their predicates. Such a typology of the construction of the connection based on different parts of the language is very clear and is motivated by the fact that the verbs, due to their relative nature and their essence, are predisposed to pre- *° Fauconnier’s (1985: 35-81) differentiation of role and its value as a specific variant of the distinction,,type vs. instance- “. The definition of the role derives from its uniqueness in a certain historically determined world reality type. The definition of the role naturally takes over its values, which are acquired through the type of world instanciacijas konkreéiu laikotarpiu (plagiau Zr. Langacker 1991: 71-73; plg. Mikulsit represents (2006a: 54-55). 146 Jungties konstrukeijos ir jy gramatinis kontekstas in the position of the predicate to mean the main predication of the sentence, 0 nouns acquire such power only in the position of the predicate. Furthermore, depending on the fact that the predicative name ar word is interpreted in them definitely undefined, the identity ar constructions may be further constructions of the equivalence inclusive subtype. 3.3. Finally, it is skiriamos j necessary to ir discuss some peripheral cases of conjunction constructions, where the predicates are proper nouns, e.g.: (38) Let me introduce myself, I am Jonas Petraitis. (39) I am John. (40) My name is John. (41) My name is John. (42) Miisy's teacher is Jonas Petraitis. Real names usually provide a descriptive definition, i.e. their referent is a uniquely identifiable person. But nevertheless. The listener may not recognize the person referring to the real name, then the speaker uses the name with an indefinite indicator, cf. Miisy's teacher is such as Jonas Petraitis, — In this example, the real-name referent is a specific but undefined person. Therefore, it is not difficult to imagine a conjunction sentence in which the proper name is interpreted non-specifically, e.g. These two clauses are Sutkai, —in this example the predicative noun j denotes the whole Seima or family, which usually includes more than two members, so the inclusive subjunctive Gia of the proposition is easily understood. It seems that a much more problematic analysis is that of the construction of the conjunction, in which the proper name of the predicate means the only, even if undefined, referent. How should we understand such a nameZodiniy predicative in relation to pobiedj? | what type of description does the real name specify, or what type can the real name denotate instance? No matter how semantically poor is the description of the type represented by the proper name, it is significant from a theoretical point of view. In this regard, let us consider the sentence (38) — when speaking, the subject probably does not want to communicate to the listener by his name, but to present himself as a person uniquely associated with that name. However, in addition to unique identification, the name Jonas Petraitis also indicates that its holder is a person of Lithuanian nationality of male sex. In addition to nationality and gender, and sometimes also origin or position in 147 ROLANDAS MIKULSKAS society, the minor form of a name can also indicate the young age of its owner (e.g. Antanukas, Petraiciokas) and other things that we usually ignore when using names. These,,unnecessary* things form an empirical basis, even minimal one, for establishing the relation of instance between the type of person they represent and its specific representative. The type description provided by a real name can be particularly easily recognized when that name is associated in society with a historically significant role of its holder. The predicative potential of such a name, used in the position of a predicate, is obvious, cf.: (43) He is Vytautas Landsbergis, and that says everything. Finally, the implications associated with real names can lead to the general use of real names. When such nouns are used as predicates in a conjunction construction, there is no doubt about the predicative potential of dél jy, cf.: (44) That my boy is a real einstein. 3.3.1. The sentence (39) can be considered the answer to the question What is his name? or answering the question Who are you? In the first case, the analysis of the sentence turéty beiti is related to the analysis of such sentences as (40) and (41), and in the second case — to the discussed analysis of the sentence (38). Indeed, in a different interpretation of the sentence (39), its predicate John refers specifically to an individual who, moreover, can be implicitly uniquely identified as the person mentioned above, and the implication of recognition can be considered a description of the type instanced by the real name — it can be read explicitly as necessary, cf. : (45) I am John who called you yesterday. 3.3.2. Since the sentences discussed represent an identity relationship between two uniquely identifiable referents—persons, they can deservedly be regarded as instances of the construction of the equative subtype of the conjunction. The essence of IS is no different from that of the sentence (42), unless we consider it to be an answer to the question What is the name of this teacher? However, at first glance it seems that in this sentence the relation of identity between the two substantive referents is different from the case discussed earlier. Indeed, if we take the answer to the question What is the teacher? in sentence 3i, the subject noun is undefined and apparently has a wider reference than the proper name in the 148 Jungties konstrukcijos ir jy gramatinis kontekstas predicate position. The relation of noun to predicate in such a sentence is reversed than in the cases of inclusion relations discussed earlier (cf. John Peters is (miisy) a teacher). However, if we treat the subject noun and the teacher in the sentence as a certain role (see section 3.2.2), which in a particular period can only be performed (instanced) by a single individual (even if not recognized by the questioner), then the problem of the possibility of ontological relation of identity between two supposedly different referents is solved: the relation of identity is established between two specific referents, only the referent of the predicative noun (real name) can still be a uniquely identifiable person. 3.3.3. When sentence (39) has the same meaning as sentences (40) and (41), the identity relationship in it is not between the uniquely identifiable first-person (a8) referent and the holder of the real name (John) — the specific person instantiating the male gender individual (in Lithuanian), but between (the holder of the uniquely identifiable name) and the name itself. In other words, the target of this identity proposition is not the person itself, but only a certain, appellative aspect of it. This aspect is expressed explicitly in the sentences (40) and (41). Let us remember that in one aspect or another, the subject name of the person referred to in the sentence (a specific person) can be categorized in other inclusion clauses, for example, by indicating his profession, social status or membership in the Seimas and the like (see I am a pensioner /grandparent). It is understood that in such sentences as (39), (40) and (41), the referent of the predicate noun —the actual noun itself—is not a uniquely identifiable entity, cf.: (46) AS I am John, and my neighbor is also John. (47) AS my name is John, and my neighbor- 's name is also John. We could distinguish the different types of relationships (and predicative potentials) between proper names in the sentences in question by the fact that they refer to the relationship between the name given to a particular person (John) and the place of that name in the nomenclature (religious tradition, prevalence, etc.) or its implicational potential (historical, cultural associations, etc.) — the latter constituting the description of the proper name per se. All such sentences can deservedly be assigned to the inclusive subtype of identity-- type conjunction construction, but while preserving their specificity, we could distinguish them as a variation of the appellative conjunction sentence. Thus, the peripheral connection construction cases can be assigned to the types and subtypes discussed earlier in one way or another. 149 ROLANDAS MIKULSKAS 3.4.The typological classification of the constructions of the connection based on the semantic (part of the language) and/or referential properties of the predicative Words used in them, intuitively led by the assumption that the distinct construction types and subtypes are if essentially different instances of the same structural archetype. This is the assumption on which the terms prototype and periphery were based in the discussion. However, when discussing the peculiarities of the different constructions of the conjunction and the different expressions of the predicate, we have drawn attention to the different designations of these constructions in the language that motivate these peculiarities. In fact, the names of the different types and subtypes of conjunctional constructions —ascriptive, inclusive, equative constructions — are formed on the basis of different purposes. So we could leidZianti visas jas priskirti tai patiai, pavyzdZiui, tapatumo reikSmés konslegitimately ask, do the same form — X is Y, but different purpose constructions have something in common? In other words, are the discussed types of connection constructions united by some common meaning, structure, category? This is a fundamental question. Therefore, depending on how we answer it, we will be talking about homonyms of completely different constructions or about separate typological instances of the same construction. 3.4.1. In discussing the various types of conjunction sentences with substantive predicates, we have found an ontological commonality that allows us to talk about their common structural expression — the identity relationship. But the same relation of identity in its own way underlies the ontology of the connection of sentences with binomial predicates. As we have already mentioned, the predicate of a verb is of a relative nature. Simply put, this predication cannot be perceived (and discussed) autonomously, as a self-existent property, but only as a characteristic of a certain thing (qualitative or quantitative). Thus, in the semantic structure of the adverb there is always an anticipated place for the abstractly perceived object, and we will not overexpose the truth of the sentence John is an intelligent structure and the image of John is an intelligent object (Man). Thus, we could argue that the attribution of properties to the subject nameword referent in sentences of the adjective type is also realized through the affirmative of the identity relationship. For Gia only one participant of the identity relationship is the referent of the predicative adverb of the assumed noun, instanced not in physical reality, but only in the mental space of implications. In this respect, however, the case of the inclusive subtype of connection in sentences is similar: we could reasonably assume that one such sentence profiles the identity relationship only for structural reasons. The inclusion of a word referent into a broader class of referents satisfying the description of a certain type, or, in other words, its assignment to a certain, only in the mental space existing, univariate type of being, is structurally possible in the construction of the connection only through 150 Jungties konstrukcijos ir jy gramatinis kontekstas lyviy — vardazZodinio predikatyvo referentas — tikrovéje instancijuojamas a freely chosen representative of that type, instantiated for that purpose in a specific context. More generally, such a “structured” instance constitutes the functional motivation of the unspecified reference of the undefined noun (see Langacker 1991: 68). It is understood that the Zodji referent of such a noun does not exist outside the bounds of the sentence, and is also a mental phenomenon, albeit in a slightly different sense than the implied substantive element of the semantic structure of the predicative adverb. 3.4.2. In summary, we can affirm that the profiled identity relationship in the cases of the discussed connection constructions is a common element of meaning for the whole category, allowing us to talk about identity meaning constructions alongside the above-mentioned locative, existential, and belonging meaning constructions, in which the same structural identity expression function (or,identity‘ meaning) is essential for the connection verbs in all cases. This common element of meaning, which underlies the construction of the structural archetype of the connection, does not prevent its functional differentiation. As we have seen, depending on what the subject noun referent is identified with in the connection constructions, they can perform the functions of sub-sign assignment to the referent, class inclusion of the referent or equalization, or equation, of two specific referents. 4. CONCLUDING REMARKS 4.1 The article ends with a more general consideration of the extent to which the predicative potential of the verb bit contributes to the functional differentiation of the constructions in which it is realized. After all, in order to adequately classify these constructions, on the one hand, we have to take into account their purpose, on the other hand, it is difficult not to pay attention and during their analysis to a certain ontological commonality of constructions of different purposes, which allows them to be united in certain structural archetypes. In the second part of the article, the analysis of the possibility of a word to have meaning in various constructions allowed to define 151 ROLANDAS MIKULSKAS its predicate potential, or the range of meaning whose supporting points would constitute the meaning of ‘existence- ’, explicitly manifested in ta egzistencinéje konstrukcijoje, ir ,tapatumo* reikSmé, lengvai atpaZistama connection constructions. The other discussed constructions of verb bits, the “lexical” meanings of which are defined “tautologically” by the semantics of the complements of the verbs or by the corresponding function of the construction, may be related to one or another of the above-mentioned basic meanings in terms of their origin. In fact, as we have seen, there are the verbs of the basic locative and mihi est type possession constructions, although each in its own way, connected with the predication of existence (see sections 2.2.1., 2.2.2.), and the verb of the belonging construction connected with the predication of identity: after all, we can assert that the relationship of belonging in the latter construction is realized panaphically as and only in the ascriptive construction of connection discussed in section 3.4.1. — through the assertion of identity between two beings, in which one is only implied (see section 8 iSna8a). These things suggest that all the constructs discussed can be derived from two structural archetypes — existential and identity. Since the predicative potential of the verb being is rather limited, constructions adapted to the various communication needs of the structural archetypes mentioned above may be made by means of,,internal” rearrangement — reinterpretation — of the basic structure, or by means of,,internal” rearrangement of the basic structure, for example, by changing individual elements of this structure according to the need (for example, in the archetypal identity structure, the possessive genitive of the predicate,,- is3ius” is replaced by the belonging construction) and/or by changing its unsymmetrical actual division (the most prominent example — the case of existential versus localative constructions). 4.2. The fact that the same basic structure can be adapted to the needs of several different propositions, accordingly rearranged, is clearly demonstrated by the various interpretation possibilities of such sentences as Onutés (yra) Zalios akys, which have been extensively discussed in recent Lithuanian literature (see Holvoet 2005: 153159; Mikulskas 2006b: 41-50). These types of sentences are often interpreted as a construction of attributing a certain attribute to a person — then, it must be assumed, it is made in the way of splitting the theme of the corresponding connection construction (e.g. Onutés akys yra Žalios). However, in a favourable context, the sentence may have a possessive interpretation, close to the possession construction of the mihi est type. The construction of truth for Onutei is Zalio akys (Green eyes) which is not yet completely foreign to 152 Jungties konstrukcijos ir jy gramatinis kontekstas Lithuanian language, although it already sounds very well. It is not excluded that in the Lithuanian language, with the disappearance of the old mihi est type possession construction (which was replaced by the newer habere type construction), the dative could be changed to the genitive at least occasionally, when the context permitted it (about the possibility of this functional approximation, see section 1.3). Thus, the dual interpretation of the aforementioned basic structure (Onutés (is) Green Eyes) can be linked to two Sources of its origin. Based on the above interpretations, we could say that in language there are two homonymous constructions — attribution of an attribute to a person and possessive possession (Of course, by saying this we should ignore the fact that only the first interpretation is sufficiently common in usage that it can undoubtedly be called a construction). Then we have to associate the action of homonymic construction of the word biti with two different needs of the basic structure — ‘identity’ and ‘existence’. These meanings can be »considered« only in the presence of one or another (formally) interpretation of the same sentence. Thus, although they are “recognized” as beings of such constructions, they hardly have any prognostic value in determining the identity of such constructions. 4.3 Similarly, in the constructions discussed in the article, the verbs,,recognizing* the structure of origin and the inherited, reminiscent* needs, can hardly help much in determining the identity of those constructions. Even the recognition of those meanings says little about the purpose of the new constructions in which the verbs are used. Moreover, as we have seen on many occasions, the semantic properties of the verb often change with the changing function of the construct: for example, the verb of a locative construction that is ontologically (and perhaps etymologically) connected with the predication of existence, due to its subjective noun definition, no longer has the potential to quantify the referent of that noun to the verb of the existential construction (see section 2.2.1). Therefore, the semantics of the action word only adapts to the meanings of constructs adapted to the various needs of communication. In defining the identity of a construction, much more important are those elements of its syntactic and semantic structure that determine the adaptability of the construction itself to a specific function of language, in other words — its functional adequacy. | These elements and worth paying closer attention to tipologiSkai klasifikuojant konstrukcijas. 153