Cognitive foundations of emotion verbs complementation in Croatian/Kognitivni temelji komplementacije glagola za izrazavanje emocija u hrvatskome jeziku.
1.1 Complementation of emotion verbs and semantic motivation of grammatical structure
When speaking about the relationship between grammar and meaning, different contemporary models of grammatical description are generally divided into formal and functional approaches. Formal approaches separate syntactic relations from semantics and pragmatics i.e. they study and describe syntactic structure in isolation using a formal methodological apparatus which is not based on semantic or pragmatic terms. The prototype of such an approach was set up with the publication of Chomsky's transformational grammar, i.e. Syntactic structures (1957), a work based on his doctoral dissertation The Logical Structure of Linguistic Theory (1955). The main implications of such an approach are reflected in the thesis of the autonomy of syntax, which is also one of the foundations of transformational grammar from its very inception. (1) On the other hand, functional approaches study syntactic relations in the light of their determination i.e. motivation by semantic and pragmatic factors. Different traditional functional approaches (e.g. Dik 1978, 1989, Van Valin, LaPolla 1997, Van Valin 2001) to grammatical description separate the meaning of linguistic units from their use. (2) In relation to that, cognitive (constructional) approaches (Langacker 1987, 1991, 1999, 2008, Goldberg 1995, Croft 2001) form a separate model of grammatical description within the functional framework, in which semantics and pragmatics are not separated (3). Regardless of that difference, all functional approaches describe structural relations in the light of their semantic determination i.e. motivation by semantic and pragmatic relations.
In the context of these differences, emotion verbs provide a very interesting field for interdisciplinary study. Emotions, as an extralinguistic phenomenon, are part of everyday human experience and are thus a very important component of human cognitive structure. In order to demonstrate that human affective experiences are organized in a systematic fashion, psychologists have studied empirically the principles and the nature of emotion categorization i.e. their clustering into larger groups according to different aspects and degrees of their similarity (see Shaver, Schwartz, Kirson & O'Connor 1987). Given the results of this kind of psychological research, and the insistence of the cognitive linguistic community on the intimate connection between language and other human cognitive abilities, our motivation for exploring how the principles of extralingusitic categorization of emotions are manifested in grammatical relations should come as no surprise.
The aim of this paper, is, therefore, to demonstrate that formal facts related to the complementation of emotion verbs in Croatian are actually clear manifestations of encyclopedic knowledge i.e. of emotional experience, and its systematic extralinguistic categorization. We will show, in other words, that this systematic pattern is reflected, in a clear and unambiguous way, in grammatical relations, giving them a solid semantic motivation.
1.2 Psychological studies and the definition of emotions
Emotions can be defined as experiences or states triggered off by a certain event, situation, action, other people, our thoughts, expectations and plans. According to the basic hedonistic tone, emotions are divided into pleasant and unpleasant. They also differ in relation to their intensity and duration, which are directly connected with their activation capacity i.e. the fact that some emotions encourage reaction in the form of physical activity, while others exclude that possibility. In one recently published psychological encyclopedic manual (Strickland 2000: 218) emotions are defined as "a reaction, both psychological and physical, subjectively experienced as strong feelings, many of which prepare the body for immediate action. In contrast to moods, which are generally longerlasting, emotions are transitory, with relatively well-defined beginnings and endings. They also have valence, meaning that they are either positive or negative. Subjectively, emotions are experienced as passive phenomena. Even though it is possible to exert a measure of control over one's emotions, they are not initiated--they happen to people."
The first influential theory of emotions in modern times--the James-Lange theory--was formulated independently in the 1880s by the American psychologist and philosopher William James and the Danish physiologist C. G. Lange (1834-1900). Both scientists arrived at the view that the physiological manifestations of emotion precede the subjective ones--rather than trembling because we are afraid, we are afraid because we tremble. Even though the brain responds to a threatening situation by activating peripheral responses, we do not consciously experience the emotion until these responses are activated. Thus, the central nervous system itself does not actually produce the emotion. Over the following decades, this theory drew a widespread response and criticism. An alternative model of emotional experience was formulated in 1927 by Walter Cannon (1871-1945), who proposed that emotions do originate in the central nervous system. Cannon argued that nerve impulses first pass through the thalamus, from which subjective responses are routed through the cerebral cortex, directly creating the experience of fear at the same time that physiological responses are passing through the hypothalamus. The Cannon-Bard theory, whose name reflects later modifications by Phillip Bard, thus delineated the psychological and physiological components of emotion as simultaneous and argued that the experience of emotion comes directly from the central nervous system. Some more recent theorists have once again moved closer to the James-Lange model. The 1962 Schachter-Singer theory restores James's emphasis on the interpretation of physiological responses but adds another element--a cognitive evaluation of what caused the responses. This theory thus contradicts James's assertion that emotion is communicated solely on the basis of physical feedback, asserting that this feedback by itself is not clear enough to specify a particular emotion. Rather, the brain chooses one of many possible interpretations and "labels" the feedback pattern, and it is this labeling that results in the experiencing of a particular emotion (see Strickland 2000: 218).
People in general have a great knowledge about emotion. When given posed or natural photographs of common emotional expressions, people around the world can reliably name the emotion being expressed (Ekman, Friesen, & Ellsworth, 1982a, 1982b). People from a variety of cultures agree on which emotion generally follows a particular set of abstract antecedents, such as insult, loss, and danger (Boucher & Brandt, 1981; Brandt & Boucher, 1984; Ekman, 1984; Roseman, 1984; Sullivan & Boucher, 1984). Both children and adults can report and agree on typical antecedents of several common emotions (Harris, 1985; Masters & Carlson, 1984; C. Smith & Ellsworth, 1985). They can also talk about methods for controlling the expression of negative emotions (Hochschild, 1983; Johnson, 1983; Masters & Carlson, 1984; Saarni, 1979, 1984). Children as well as adults agree about the similarity or distinctiveness of a diverse array of emotions, and their similarity ratings, when multidimensionally scaled or factor analyzed, reliably form a two--or three-dimensional structure, with positive-negative valence, activity or arousal, and potency or dominance being the most frequently obtained dimensions (e.g., Averill, 1975; Osgood, Suci, & Tannenbaum, 1957; Russell, 1978, 1979, 1980: Russell & Ridgeway, 1983; Schlosberg, 1952, 1954). In that sense, an extensive body of research and theory in cognitive and cognitive-social psychology suggests that the various components of emotion knowledge are likely to be parts of an organized whole, and that systematic quality has been schematically presented in Figure 1 as three-dimensional frame of emotional categorization.
[FIGURE 1 OMITTED]
In this model every concrete emotion can be located in relation to the way in which it combines three key parameters: hedonistic tone or evaluation which divides emotions into pleasant and unpleasant (symbol E), degree of potency (symbol P) and ics activation potential (symbol A). Study of human intuitive emotional categorization by Shaver et al. (1987) showed that, regardless of the fact that the number of categories respondents could establish was not fixed in advance, emotional domain can be structured within a three-dimensional model. In that model different concrete emotions form clusters in relation to their proximity in the given "coordinate system" i.e. more or less expressed similarity according to three given criteria.
The three-dimensional solution corresponds well with three-dimensional solutions obtained earlier by, for example, Schlosberg (1954), Osgood et al. (1957), and Russell and Mehrabian (1977). The first, horizontal dimension corresponds with what Osgood et al. called "evaluation." The second and third, extending back and up, respectively, are formed by splitting intensity dimension into two separate vectors, which Osgood et al. called "potency" and "activity". (Shaver et al. 1987: 1071)
In respect to what has been said above, in the central part of this paper we will try to see if the described nature of categorization of emotions has any impact on grammatical relations i.e if it is formally reflected in grammatical structure, in particular in the case marking of emotion verb complements. (5) If we manage to demonstrate this, it will mean that a grammatical category can be determined by the very nature of human categorization and conceptualization of extralinguistic phenomena. This, of course, will be in perfect tune with the functional thesis about the motivation of grammatical structure by semantic relations, which we have described above.
2. Semantic analysis of emotion verbs and case coding of their complements
In this section, the character of affective states expressed by specific emotion verbs will be clearly and unambiguously related to the basic spatial concepts of origin (point of departure), directedness, goal and parallelism, and through that, to the main criteria of emotion categorization in psychological research. Specific verb groups will be described using the methodology of Cognitive Grammar. Thus, it will be shown that case marking of their complements is highly motivated by the extralinguistic categorization of emotions which is largely based on spatial concepts.
Theory which is today known as Cognitive Grammar first appeared on the world linguistic scene under the name space grammar (Langacker 1982). Thus, it could be said that from its very inception cognitive grammar emphasized spatial foundation of human conceptualization and, as a result, largely of the grammar itself. R. Langacker (1987: 489) defines grammar as a structured inventory of conventional linguistic units. That definition implies that grammar is not merely a list of mutually independent categories but that grammatical categories are determined by interdependent relations within the system. Feature of conventionality stems from the symbolic nature of linguistic units and the fact that the relation between their phonological and semantic pole (see Langacker 1987: 76-81) is arbitrary. In other words, grammar consists of those aspects of cognitive organization on which the speaker's knowledge of linguistic conventions rests. In line with that, R. Langacker defines cognitive grammar as a usage-based model and emphasizes the importance of the study of symbolic and semantic aspects of linguistic (grammatical) units. As a matter of fact, he places the semantic component in the centre of the grammatical description, claiming that language is by its nature symbolic. Here, it is important to note that grammar in this theory, in principle, is not separated from lexicon but these two levels of linguistic description form a continuum. Semantic analysis in the following part of this section will be based on these methodological assumptions. That analysis will be carried out on the verb groups divided according to the formal criterium of case coding of their complements. However, these formal facts and perceived patterns will, in that sense, be considered only as reflections of conceptual-semantic regularities which will be recognized, established and explained by detailed analysis.
2.1 Genitive complements
Genitive complements, either without prepositions or in a prepositional phrase od+genitive (6), with verbs such as bojati se (to be afraid of), plasiti se (to be scared of), strahovati (to fear), gnusati se (to loathe), stidjeti se (to be ashamed of), sramiti se (to be ashamed of), groziti se (tu shudder), libiti se (to hesitate), zazirati (to have an aversion to) and so on are primarily motivated by the nature of emotion they express i.e. emotion that is characteristic of the subject/trajector. Since these are, almost without exception, extremely negatively marked emotions, most often different types of fear with smaller or larger activation potential, the motivation for the genitive coding of their complements is completely clear and logical if we have in mind the concept of ablativity (departure as one of the core meanings of the genitive).
In such scenarios, genitive complements as landmarks with the semantic role of a broadly understood effector (7) function as some kind of impulse to the subject/trajector. That impulse triggers off interaction between these two elements in the scenario (thinner dashed two-sided arrow in Figure 2), and results in a specific psychological state of the subject that enables us to connect it with the semantic role of experiencer. Since we are talking about negative experiences, and negative emotions result in our instinctive departure from their causes (bold dashed arrow in Figure 2), the semantic motivation for the genitive coding of complements/landmarks as conceptual origins of ablativity in this group is obvious, unambiguous and psychologically rooted. Apart from this basic aspect of ablativity, in such scenarios another aspect of ablativity is implicitly or explicitly present through the category of cause (8) which is related to the genitive complement/landmark. The cause, as the initiatior of the event structure in the action chain determines the departure of the subject/trajector from the genitive landmark (compare Bojim se njihova posjeta (zbog moguce reakcije mojih roditelja) / I am afraid of their visit (because of the possible reaction of my parents), Grozim se takvih situacija (zbog problema koji poslije nastaju) / I shudder at these kinds of situations (because of the problems that may arise) etc.).
[FIGURE 2 OMITTED]
2.2 Dative complements
Dative complements, just like genitive complements, are also determined by the nature of emotion as the primary criterion of case marking. But, contrary to the genitive complements, which are the result of negative emotions expressed by verbs and are therefore compatible with ablative meaning, dative complements fit the schematic meaning of the dative, viz. directedness or adlativity (9). The reason is that verbs with dative complements in most cases express positive emotions, which result in the subject's/trajector's wish to get closer to the referent of the dative complement/landmark. These verbs are: radovati se (to look forward to), veseliti se (to look forward to), obradovati se (to be happy about), razveseliti se (to be happy about), zavidjeti (to envy), diviti se (to admire) etc.
The mentioned verbs most often have animate referents (10) as complements. These referents are affected by the action of some more concrete or more abstract entity belonging to the personal sphere of the dative complement. As such, the dative-marked complements may be claimed to have the semantic role of the target person (TP) (11), a sort of generalized dative role which usually subsumes the dative roles of a recipient, benefactive, malefactive and experiencer (12) (e.g. Veselim se njegovu dolasku / I look forward to his arrival, Radujem se njegovoj odluci / I look forward to his decision, Obradovao sam im se kada su stigli / I was happy to see them when they arrived etc.). Scenarios with dative complements with given verb types are semantically somewhat more complex than the previously described scenarios with genitive complements because the dative complement/landmark functions both as the effector and the target person (13). As genitive complements, dative landmarks are some kind of effectors of ablative-causal nature, which trigger off some kind of psychological experience in the subject/trajector and initiate its action. Here, this action is of adlative nature because the resulting emotions are positive. This enables the dative complement to function as the target person i.e. the object of the trajector's desire to get closer to the entity that is seen as positive.
This ablative-adlative chain of causes and consequences is even more obvious if we look at the following examples expressing explicit causality: Veselim se njihovu dolasku jer ih dugo nisam vidio / I look forward to their arrival because I haven't seen them in a while; Radujem se njegovoj odluci jer bi se moglo puno toga promijeniti / I look forward to his decision because it could change a lot of things etc. One of the rare verbs which govern dative complements, and which do not express positive emotions is the verb zavidjeti (to envy). But even that verb fits into the adlative concept because of the subject's/ trajector's desire to get closer to and to finally possess a certain entity, characteristic or something else that is located in the personal sphere of the dative referent/landmark. The scenario with dative complements of positive emotion verbs is presented in Figure 3. The emphasis is on the central concept of adlativity which is marked by a bold dashed arrow.
[FIGURE 3 OMITTED]
2.3 Accusative complements
As explained above, the primary psychological-semantic criterion for genitive and dative complements is the nature of the emotion expressed by the verb. On the other hand, in the case of verbs governing the accusative, either without a preposition or with the preposition na in the indirect object, this criterion is not very important. In this case, the key criterion is the activation potential of the emotion, which was described in the introduction. As opposed to emotion verbs with dative complements, all verbs with accusative complements refer to extremely strong positive and negative emotions such as mrziti (to hate), voljeti (to love), obozavati (to adore), prezirati (to despise), ljutiti se (to be angry), naljutiti se (to get angry), razljutiti se (to get angry), srditi se (to be angry), rasrditi se (to get angry), bjesnjeti (to be mad), razbjesnjeti se (to get mad). The intensity of emotions enables the inclusion of these accusative meanings into the schematic concept of target. These verbs open up possibilities for a more direct contact of the subject/trajector with the accusative referent/landmark. In Figure 4, this is shown by an even bolder dashed arrow directed from the trajector to the landmark. As we can see in this Figure, the only difference between dative and accusative scenarios that matters, at least for our discussion, is precisely that possibility of a concrete or abstract direct affectedness of the landmark. The other relations remain the same and will not be repeated here. Our claims can be nicely captured by comparing, for example, verbs diviti se (to admire) and zavidjeti (to envy), which govern dative complements, with verbs voljeti (to love), obozavati (to admire) or mrziti (to hate), which govern accusative complements. As we can clearly see, the first set of verbs expresses emotions of lower intensity and activation potential. Such verbs signal the directedness of the subject's emotion towards the dative landmark, their distance and the subject/trajector-to-complement/landmark flow of energy in the action chain. On the other hand, when dealing with verbs that govern accusative complements we talk about emotions of much stronger intensity which result in a greater likelihood of action towards the object. At the same time, this decreases the distance between them and allows for action with an energy flow running in both directions. The intensity of these emotions, especially if they result in some kind of direct contact, very often provokes some kind of reaction on the part of the referent realized as the object. (14)
[FIGURE 4 OMITTED]
2.4 Instrumental complements
Instrumental complements, either without prepositions with emotion verbs such as ponositi se (to be proud of), diciti se (to be proud of), oduseviti se (to be thrilled) or in prepositional phrases such as s(a) / za / nad + instrumental with the verbs suosjecati (to sympathize with), tugovati (to mourn), zaliti (to regret), zalovati (to mourn), zudjeti (to long), ceznuti (to yearn), patiti (to suffer), ludovati (to be crazy about), sazaliti se (to feel pity for), are motivated by the schematic concept of parallelism. With these verbs, the concept of parallelism is realized in its more abstract i.e. non-spatial sense, in relation to the nature of the emotions expressed by these verbs or the activation potential inherent in these emotions.
The prototype concept of spatial parallelism i.e. overlap of the paths of the nominative trajector and the instrumental landmark, so obvious in the instrumental sociative and instrument meanings (e.g. Ivan sece s prijateljem / Ivan is taking a walk with a friend, Ana putuje automobilom / Ana travels by car etc.), is also discernible in the scenario coded by the verb suosjecati / to sympathize with. The emotion which is marked by the verb includes affective identification of the experiencer (trajector) with the effector (landmark) (e.g. Suosjecam s prijateljem koji je izgubio obitelj / I sympathize with a friend who has lost his family). The identification of these two participants in the process designated by the verb, which is fostered by the metaphoric mapping of concrete spatial relations to abstract relations in the emotional domain, fits easily into the concept of parallelism on the conceptual level. In examples of this type, the trajector identifies with the emotional state of the landmark and, in that sense, maintains a metaphorical equidistance analog to the physical equidistance between the trajector and the landmark in scenarios of parallel physical movement. This metaphoric equidistance provides the conceptual-semantic motivation for the instrumental coding of the landmark. This is represented in Figure 5 by a bold dashed two-sided arrow connecting the trajector and the landmark.
[FIGURE 5 OMITTED]
In the case of instrumentals without a preposition, in combination with verbs ponositi se and diciti se (to be proud of), we are dealing with extremely positive emotions. But unlike verbs with dative complements, which also express positive affective states, in this case landmarks coded by the instrumental are, so to say, part of the trajectors themselves. To put it simply, the feeling of pride can ocur only in relation to some entity which represents part of us i.e. which we consider ours. We identify ourselves with it on the affective level and thus it is kept in our proximity (e.g. Ponosio se svojim djetetom / He was proud of his child, Dicio se ratnim odlikovanjima / He was proud of his war medals etc.). Here, it is very interesting to analyse the relationship between these verbs and their antonyms stidjeti se and sramiti se (to be ashamed of) which require genitive complements (e.g. Stidio se svoga djeteta / He was ashamed of his child, Sramio se ratnih odlikovanja / He was ashamed of his war medals). This contrast certainly has to do with the negative nature of the emotions expressed by these verbs and with the human instinct to establish ever greater distance from the entities which cause shame. With the verbs ponositi se and diciti se, in turn, which express positive emotions, such distancing will never become an issue. So, the initial minimal distance between the trajector and the landmark remains the same during the whole process. Having in mind this equidistance, the described scenario fits, even in this case, the schematic concept of parallelism, motivating the instrumental coding of the landmark itself. These relations are shown in Figure 6.
[FIGURE 6 OMITTED]
In Figure 6 the large circle represents the trajector, and the square located within this circle is the landmark, i.e. the effector which causes the emotion expressed by the verbs ponositi se and diciti se. Since the landmark, in the case of these two verbs, represents a part of the trajector, so to speak, and the emotions they express are highly positively marked, the experiencer will not wish to distance himself from the effector. Their relationship will remain the same during the process denoted by the verb. This equidistance represents the conceptual-semantic foundation of the instrumental coding of the landmark and is also marked by a bold dashed two-sided arrow. The same idea of equidistance, coupled with the expression of the positive emotion, also explains the morphosyntactic coding of the complement of the verb oduseviti se (to be thrilled). Put simply, there is an entity which causes a positive emotion. As a landmark it is initially close to the trajector and maintains this equidistant position relative to the trajector for some time.
The concept of spatial parallelism is also very close to the scenarios marked by the verbs tugovati (to feel sad about), zaliti (to feel sad about), zalovati (to mourn), zudjeti (to yearn for), ceznuti (to yearn for), patiti (to suffer) and ludovati (to be crazy about), which appear in combination with prepositional instrumental expressions of the type za + instrumental (e.g. Obitelj je mjesecima zalovala za pokojnikom / The family mourned the deceased for months, On jos uvijek tuguje/zali za bivsom suprugom / He is still sad about his ex wife, Oni zude/ceznu za boljim ivotom / They yearn for a better life, Moj prijatelj dugo je ludovao za njom / My friend was crazy about her for a long time). In these examples, the emotional relationship of the trajector and the landmark is metaphorically conceptualized as a constant attempt to follow and eventually reach the desired, but inaccessible entity. The landmarks, who are the objects of unfulfilled desire, constantly "evade" the trajector. The trajector is nevertheless, intent on reaching the entity, and so metaphorically follows it for some time. In this metaphorical pursuit of a desired entity, which, for some reason, is bound to fail, the distance between the trajector and the landmark remains the same during the entire process.
This relationship is schematically presented in Figure 7, where the circle and the square represent the trajector and the landmark, respectively. Thin dashed arrows symbolize their metaphoric movement. During that movement the landmark moves away from the trajector (ABL marks the relation of metaphorical ablativity). At the same time, the trajector, represented on the other side of the Figure, persists in trying to get closer to the landmark. It tries to decrease the metaphorical distance between them (ADL marks the relation of metaphoric adlativity). As the result, the initial distance, represented by a bold dashed two-sided arrow in Figure 7, remains the same. Here again, the scenario fits into the instrumental scheme of parallelism.
[FIGURE 7 OMITTED]
The verb sazaliti se, which is used in combination with prepositional instrumental constructions such as nad + instrumental expresses an emotion which is short-termed, and thus, does not possess a big activation potential. It is not very probable that such an emotion would prompt the experiencer to act towards the effector. Due to its limited duration, the distance which divides these two elements will also remain the same (e.g. Na putu kuci sazalio se nad prosjakom koji je stajao na kisi / On the way home he took pity on the beggar who stood in the rain). This fact becomes even more obvious if we compare this verb with the verb zaliti (to feel sorry for) which opens up a place for an accusative complement. That verb expresses a long-lasting and a more intensive affective state which results in a greater likelihood of action towards the landmark and in the neutralization of its initial distance from the trajector. As far as the preposition nad (literally above) is concerned, its usage in this case is motivated by the conceptual orientational metaphors GOOD IS UP and BAD IS DOWN (see Lakoff and Johnson 1980), because pity is, as a rule, felt by the person in a better position, and that position is metaphorically conceptualized in terms of supralocativity. The relationship is schematically represented in Figure 8. Here, the circle also symbolizes the trajector/experiencer and the square represents the landmark/effector. In view of what has been said so far about the emotion expressed by the verb sazaliti se (to take pity on), the equidistance between the trajector and the landmark is represented by a bold dashed two-sided arrow. Such equidistance represents again the conceptual-semantic motivation for the instrumental coding. It also provides the link to the prototypical instrumental sociative and instrument meanings, where the equidistance between the trajector and landmark is realized through physical movement.
[FIGURE 8 OMITTED]
The semantic analysis presented above is by no means the whole story or an end in itself. It was meant to provide a platform for future discussions of the semantic determination of syntactic structure. Of course, since emotions are universal psychological category it could be added that, if the observed patterns of emotion verb complementation do indeed have the proposed conceptual motivation, such a motivation should not be language-specific. But if we take for an example complementation of the same verb group in Polish, another Slavic language, we will see at least partially different patterns of complementation. However, that fact is not an argument based on which the main thesis of this paper, and the analysis described in the second chapter, could be challenged. Although emotions are universal category of human experience, grammatical relations themselves, just as their conceptual-semantic motivation, are always language specific to a certain degree. The aim of this paper was, above all, to shed light on conceptual-semantic motivation of case coding of emotion verb complements in Croatian. Presence of the perceived and described motivation patterns as the foundation of case coding of the emotion verb complements can surely be, to a certain point, found in other languages as well, even in non-Slavic ones. On the other hand, in languages where formal facts do not fit into the analysis carried out in this paper it is surely possible to determine existence of different conceptual-semantic patterns which are reflected in different grammatical relations as well. In that way, it could be shown that in those cases as well, grammatical relations are semantically motivated. This very fact actually supports the main thesis of this paper whose aim was to show that it is possible to establish, present and explain semantic motivation of grammatical relations in specific form which is characteristic for a concrete language. Its aim was not to prove that described conceptual-semantic patterns are, by themselves, universal. Since formal linguistic facts in the framework of cognitive grammar are seen as reflection of semantic relations, and these relations in turn in the view of conceptual semantics represent the result of individual and the broadest collective experience of concrete linguistic community, it is clear that in different languages identical conceptual-semantic motivation patterns cannot exist. Different formal relations emerge in languages of more or less divided cultural communities on the basis of their different collective experiences. However, it is important to note that in any of these languages grammatical facts can be looked at in the light of semantic motivation i.e. conceptual-semantic motivation pattern is possible to be established in their grammatical descriptions as well. That pattern can, to a lesser or higher degree, overlap with patterns in other languages but it is primarily characteristic for that concrete language and is reflected in specific formal relations as a result of specific experience of that linguistic community. To conclude, authors believe that similar conceptual-semantic motivation patterns could potentially be established and described in other languages as well. Their differences in fact represent specific realizations of universal principles of human conceptualization which are mainly based on basic spatial concepts. This conclusion can therefore be regarded as an additional argument on behalf of the thesis of "weak" autonomy of syntax i.e. a critique of the thesis of its "strong" autonomy, in line with definitions given by R. Langacker (2005: 103-104).
Averill, J. R. (1975). "A semantic atlas of emotional concepts". JSAS: Catalog of Selected Documents in Psychology, 5, 330.
Belaj, B. (2009). "Prostorna znacenja na razini slozene recenice". Zbornik radova 37. seminara Zagrebacke slavisticke skole Prostor u jeziku/Knjizevnost i kultura sezdesetih, Zagreb, 43-69.
Belaj, B. (2011). "Gramatika i znacenje". Zbornik Milorada Nikcevica: Izmedu dviju domovina, Osijek, 479-517.
Belaj, B., G. Tanackovic Faletar (2008). "Space, conceptualization and case meaning: a cognitive account of the dative in Croatian", usmeno izlaganje odrzano na medunarodnom znanstvenom skupu Cognitive Linguistics between Universality and Variation, Dubrovnik, 30. rujna -1. listopada 2008. (in press).
Boucher, J. D., M. E. Brandt (1981). "Judgment of emotion: American and Malay antecedents". Journal of Cross-Cultural Psychology, 12, 272-283.
Brandt, M. E., J. D. Boucher (1984). "Judgments of emotions from antecedent situations in three cultures". Paper presented at the meeting of the International Congress of Cross-Cultural Psychology, Acapulco, Mexico.
Chomsky, N. (1957). Syntactic Structures. The Hague/Paris: Mouton.
Chomsky, N. (1977). Essays on form and interpretation. Amsterdam: Elsevier North Holland.
Chomsky, N. (1955/1975). The Logical Structure of Linguistic Theory. Chicago: The University of Chicago Press.
Croft, W. (1995). "Autonomy and functionalist linguistics". Language, 71, 490-532.
Croft, W. (2001). Radical Construction Grammar: Syntactic Theory in Typological Perspective. Oxford: Oxford University Press.
Dajbrowska, E. (1997). Cognitive Semantics and the Polish Dative. Berlin--New York: Mouton de Gruyter.
Dik, S. C. (1978). Functional Grammar. Amsterdam: North Holland.
Dik, S. C. (1989). The theory of Functional Grammar, part 1. Dordrecht: Foris.
Ekman, P. (1984). "Expression and the nature of emotion". In K. S. Scherer & P. Ekman (Eds.), Approaches to emotion, Hillsdale, NJ: Erlbaum, 319-343.
Ekman, P., W. V. Friesen, P. C. Ellsworth (1982a). "What are the similarities and differences in facial behavior across cultures?" In P. Ekman (Ed.), Emotion in the human face (2nd ed.), Cambridge: Cambridge University Press, 128-143.
Ekman, P., W. V. Friesen, P. C. Ellsworth (1982b). "What emotion categories or dimensions can observers judge from facial behavior?" In P. Ekman (Ed.), Emotion in the human face (2nd ed.), Cambridge: Cambridge University Press, 39-55.
Goldberg, A. E. (1995). Constructions: A Construction Grammar Approach to Argument Structure. Chicago: University of Chicago Press.
Harris, P. (1985). "What children know about the situations that provoke emotion". In M. Lewis & C. Sam-hi (Eds.), The socialization of affect. New York: Plenum Press, 161-185.
Hochschild, A. (1983). The managed heart: The commercialization of human feeling. Berkeley: University of California Press.
Hopper, P. J., Thompson, S. A. (1980). "Transitivity in Grammar and Discourse". Language 56 (2), 251-299.
Jackendoff, R. (1990). Semantic Structures. Cambridge: MIT Press.
Johnson, D. L. (1983). The development of children's theories of emotional control. Unpublished master's thesis, University of Denver, CO.
Lakoff, G., Johnson, M. (1980). Metaphors We Live By. The University of Chicago Press: Chicago and London.
Langacker, R. W. (1982). "Space Grammar, Analysability, and the English Passive". Language, 58, 1, str. 22-80.
Langacker, R. W. (1987). Foundations of Cognitive Grammar, Vol. 1. Stanford, California: Stanford University Press.
Langacker, R. W. (1991). Foundations of Cognitive Grammar, Vol. 2. Stanford, California: Stanford University Press.
Langacker, R. W. (1999). Grammar and Conceptualization. Berlin: Mouton.
Langacker, R. W. (2005). "Construction Grammars: cognitive, radical and less so". Ruiz de Mendoza, F. J., S. M. Pena Cervel (ur.), Cognitive Linguistics, Internal Dynamics and Interdisciplinary Interaction; Cognitive Linguistics Research 32, Berlin--New York: Mouton de Gruyter, 101-157.
Langacker, R. W. (2008). Cognitive Grammar: A Basic Introduction. New York: Oxford University Press.
Lyons, J. (1977). Semantics, Vol. 2, Cambridge: Cambridge University Press.
Masters, J., & Carlson, C. R. (1984). "Children's and adults' understanding of the causes and consequences of emotional states". In C. E. Izard, J. Kagan, & R. B. Zajonc (Eds.), Emotions, cognition, and behavior. Cambridge: Cambridge University Press, 438-463.
Masters, J., & Carlson, C. R. (1984). "Children's and adults' understanding of the causes and consequences of emotional states". In C. E. Izard, J. Kagan, & R. B. Zajonc (Eds.), Emotions, cognition, and behavior. Cambridge: Cambridge University Press, 438-463.
Newmeyer, F., J. (1998). Language Form and Language Function. Cambridge: The MIT Press.
Nuyts, J. (2008). "Pattern versus process concepts of grammar and mind: A cognitive-functional perspective". Jezikoslovlje, 9,1-2, 87-107.
Osgood, C. E., G. J. Suci, P. H. Tannenbaum (1957). The measurement of meaning. Urbana: University of Illinois Press.
Palic, I. (2010). Dativ u bosanskome jeziku. Sarajevo: Bookline.
Roseman, I. J. (1984). "Cognitive determinants of emotion: A structural theory". In E Shaver (Ed.), Review of personality and social psychology (Vol. 5). Beverly Hills, CA: Sage, 11-36.
Russell, J. A. (1978). "Evidence of convergent validity on the dimensions of affect". Journal of Personality and Social Psychology, 36, 1152-1168.
Russell, J. A. (1979). "Affective space is bipolar". Journal of Personality and Social Psychology, 37, 345-356.
Russell, J. A. (1980). "A circumplex model of affect". Journal of Personality and Social Psychology, 39, 1161-1178.
Russell, J. A., A. Mehrabian (1977). "Evidence for a three-factor theory of emotions". Journal of Research in Personality, 11, 273-294.
Russell, J. A., D. Ridgeway (1983). "Dimensions underlying children's emotion concepts". Developmental Psychology, 19, 795-804.
Saarni, C. (1979). "Children's understanding of display rules for expressive behavior". Developmental Psychology, 15, 424-429.
Saarni, C. (1984). "An observational study of children's attempts to monitor their expressive behavior". Child Development, 55, 1504-1513.
Schlosberg, H. (1952). "The description of facial expressions in terms of two dimensions". Journal of Experimental Psychology, 44, 229-237.
Schlosberg, H. (1954). "Three dimensions of emotion". Psychological Review, 61, 81-88.
Shaver, P., J. Schwartz, D. Kirson, C. O'Connor (1987). "Emotion Knowledge: Further Exploration of a Prototype Approach". Journal of Personality and Social Psychology, 52, 1061-1086.
Smith, C., E. Ellsworth (1985). "Patterns of cognitive appraisal in emotion". Journal of Personality and Social Psychology, 48, 813-838.
Strickland, B. R. (ur.) (2000). Gale Encyclopedia of psychology (2nd edition). Detroit--New York--San Francisco--London--Boston--Woodbridge, CT: Gale Group.
Sullivan, B. N., J. D. Boucher (1984). "Commonalities of meaning in antecedents to emotions across firm cultures". Paper presented at the meeting of the International Congress of Cross-Cultural Psychology, Acapulco, Mexico.
Saric, Lj. (2008). Spatial Concepts in Slavic, A Cognitive Linguistic Study of Prepositions and Cases. Weisbaden: Harrasowitz Verlag.
Tanackovic Faletar, G. (2010). Semanticki opis nesamostalnih (kosih) padeza u hrvatskom jeziku i njegove sintakticke implikacije (unpublished dissertation). Filozofski fakultet: Osijek
Van Valin, R. (2001). An Introduction to Syntax. Cambridge: Cambridge University Press.
Van Valin, R. D. Jr., R. J. LaPolla (1997). Syntax, structure, meaning, and function. Cambridge: Cambridge University Press.
Faculty of Philosophy, Osijek
Goran Tanackovic Faletar
Faculty of Philosophy, Osijek
(1) For detailed discussion on the relation of grammar and meaning in contemporary formal and functional linguistic theories see Newmeyer (1998) and Belaj (2011) and for discussion on autonomy of grammar and functional approaches see Croft (1995).
(2) In traditional functionalism the pragmatic level if often considered a special descriptive level which is connected with the semantic and the syntactic level by linking rules (see Van Valin and LaPolla 1997).
(3) On the differences and similarities between traditional functional linguistic approaches and constructional approaches in Cognitive Linguistics see more in Nuyts (2008).
(4) See graphical presentation in Shaver et al. (1987: 1069).
(5) Of course, since these general theses can be proved only by systematic and non-selective analysis of corpus data, authors will analyse all emotion verbs which open up space for genitive, dative, accusative and instrumental complements. A list of analysed verbs has been compiled not only through introspection but also by systematic dictionary analysis.
(6) It should be noted that in the genitive with the preposition od and in other prepositional phrases, the nominal word is not governed by the verb, but by the preposition heading the prepositional phrase. But in this paper we shall ignore this formal criterion because it is not relevant for our discussion.
(7) The semantic role of the effector is here understood as an effector of a broader type, which encompasses prototypical inanimate referents and animate referents which lack the semantic features of volition and intention, in relations to the action expressed by the verb, i.e. unvoluntary agents. Contexts with unvoluntary agents cover examples such as Sramim se njegovih postupaka prema nekim ljudima / I am ashamed of his actions toward some people, Zazirem od ljudi kao sto je on / I have an aversion to people like him, Vec danima strahujem od njegove reakcije / I've feared his reaction for days etc. In such cases, the action of the referents coded by the genitive does not have to be, and most often is not, directly or intentionally directed towards the subject. In other words, the reaction of the subject in the form of some kind of departure from the genitive referent is not direct, but is an indirect consequence of its action.
(8) In localist theories, the category of cause as an abstract concept of a secondary order is explained by the primary spatial meaning of ablativity. For more on that, and other abstract meanings elaborating the category of space see in Lyons (1977: 718-724) who gives a good and concise overview of the localist assumptions, and in Belaj (2009).
(9) On the meaning of directedness as the most schematic meaning in the taxonomy of dative meanings see Belaj and Tanackovic Faletar (2008) and Tanackovic Faletar (2010). For more on schematic meanings of other dependent cases see Tanackovic Faletar (2010).
(10) Animate referents prevail, but there are also many examples with inanimate referents (e.g. Veselim se zimskim praznicima / I am looking forward to the winter vacations, Radujem se godisnjem odmoru / I am looking forward to the holidays etc.).
(11) On personal sphere as the unifying schematic meaning of the dative and, consequently, on the semantic role of target person ascribed to dative referents see Dabrowska (1997), Palic (2010) and Saric (2008).
(12) Target person, as the general semantic role of the dative, subsumes these microroles of the dative in general, but not in the case of these verbs because here we only talk about the target person which is not realized in any of these subroles.
(13) This is also one of many pieces of evidence that makes untenable the [theta]-criterion, as one of the main principles of Principles and Parameters Theory (PPT). Namely, this criterion states that every argument can be ascribed only one thematic role, and every thematic role is ascribed to one and only one argument. For a detailed discussion on the untenability of the [theta]-criterion see Jackendoff (1990: 59-60).
(14) In accordance with R. Langacker and his view of transitivity, i. e. the 'billiard ball model' (see Langacker 1991), as well as with Hopper and Thompson (1980), it is also very important to emphasize the fact that, in this kind of view, the transitive construction actually bears 'a meaning of its own'. Keeping that in mind, we may say that it is actually not the accusative complement in itself that is linked to particular aspects of emotions, but that the whole construction is involved in the form-meaning pattern.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Znanstveni radovi|
|Author:||Belaj, Branimir; Faletar, Goran Tanackovic|
|Date:||Dec 1, 2011|
|Previous Article:||Modeling epistemic distance: the Croatian l-participle/Modeliranje epistemicke udaljenosti: glagolski pridjev radni u hrvatskome.|
|Next Article:||Syntax of Croatian and German verbs of incomplete meaning exemplified on modal verbs/Sintaksa glagola nepotpunog znacenja u hrvatskom i njemackom...|