Semantic analysis of spatial prepositions in standard Italian

U ovom se radu istražuje prostorno značenje onih članova prijedložnog sustava u standardnom talijanskom koji svoja primarna značenja ostvaruju upravo u domeni prostora. Istraživanje je provedeno unutar kognitivnolingvističkog teorijskog okvira, pri čemu je Vandeloiseov (1986, 1991) funkcionalni pristup prostornim prijedlozima u francuskom jeziku korišten kao osnova za teorijsko-metodološki model kojim je provedena semantička analiza talijanskih prostornih prijedloga. U radu su kritički razmotreni pristupi prostornim prijedlozima općenito, a njihova je ukupnost podijeljena u dvije stupnjevito ustrojene kategorije: (i) 'dominantno prostorne' i (ii) 'funkcionalno-prostorne'. Kritički su prikazane i najvažnije semantičke studije posvećene talijanskim (prostornim) prijedlozima kako bi se stekao uvid u dosadašnja postignuća na ovom području (Devoto 1940, Parisi i Castelfranchi 1970, Crisari 1971, Weinrich 1978a, Pennacchietti 2006). Osim spomenutih semantičkih studija, proučen je doprinos talijanske gramatikografije (1951.-2016.) ovom području te su podjelom svih analiziranih izdanja po kriteriju obrade kategorije prijedloga (tradicionalistički, sintaktički i Schwarzeov smjer talijanske gramatikografije) izvedeni zaključci o načinima definiranja članova ove kategorije, njihovim klasifikacijama i načinima značenjske raščlambe. Spoznaje do kojih se došlo kritičkim razmatranjem međunarodne i talijanske literature relevantne za predmet ovoga rada iskorištene su u razradi teorijsko-metodološkog modela. Pokusnom primjenom radikalno polisemijskog pristupa na jedan dio prostornoznačenjske strukture prijedloga su potvrđena je potreba za uspostavom jasnih kriterija kojima se mogu međusobno razgraničiti različita prostorna značenja istog prijedloga i definirati ono primarno. Kao kriterij za utvrđivanje odjelitog prostornog značenja prihvaćen je kriterij različitog prostornog odnosa između trajektora i orijentira prema 'modelu dosljedne polisemije' (Tyler i Evans 2003). Međutim, bilo je nužno jasno definirati što se može odrediti kao razlika u prostornom odnosu, pa su s tim ciljem teorijski razmotreni i međusobno razgraničeni pojmovi 'prostornoga' i 'funkcionalnog' u značenju prijedloga. Sagledani su čimbenici dinamike sila i naposljetku je izdvojen raspored trajektora i orijentira u odnosu na osi trodimenzijskog koordinatnog sustava kao pokazatelj različitog prostornog odnosa. Kao kriterij za definiranje primarnog značenja određen je stupanj kompleksnosti pojma koji ga određuje. Nadalje, izrađen je prijedlog simboličkog prikaza (kompleksnih) pojmova koji određuju prostorna značenja prijedloga i on je primijenjen na konkretne elemente obuhvaćene analizom. Teorijski su sagledani paradigmatski odnosi među prijedlozima, a posebna je pozornost posvećena problemu prijedložne antonimije. Zaključak da se primjenom Pennacchiettijeve (2006) varijable '(retro)aplikativnosti' može utvrditi prijedložna antonimija omogućio je utvrđivanje homonimije u slučaju oblika da. Provedeno istraživanje je uporabno utemeljeno, a korišten je korpus pisanog talijanskog jezika CORIS/CODIS Sveučilišta u Bologni. Korpusna je analiza u ovom radu kvalitativnog karaktera, pri čemu je važnu ulogu imala sustavna introspekcija istraživača poduprta sudovima izvornih govornika. Istraživanjem je obuhvaćeno ukupno dvadeset i dva prijedloga i prijedložna izraza: su, sopra, sotto, al di sopra, al di sotto, davanti (a), dietro (a/di), di fronte a, alle spalle di, dirimpetto a, a destra di, a sinistra di, di/a fianco a/di, vicino a, lontano da, accanto a, in, dentro (a/di), fuori (di/da), a, da1 i da2 . Za sve su navedene prijedloge i prijedložne izraze izrađena uporabna pravila koja su u nekim slučajevima popraćena ograničenjima izbora, a svakom je uporabnom pravilu pridružen simbolički prikaz složene strukture pojma koji određuje prostorno značenje. Svaka značenjska varijanta se prikazuje analognim simboličkim prikazom, čime se ukazuje na nepostojanje stvarnog dispariteta između pojmovne i leksičke polisemije. Kompleksna struktura pojmova koji određuju prostorna značenja prijedloga obuhvaća funkcionalne i prostorne značenjske odrednice. Njihova je struktura posljedica čovjekovog utjelovljenog iskustva i enciklopedijskog znanja (Žic Fuchs 1991, 2009), a holistički karakter takvih snopova različitih osobina ustrojenih prema načelu obiteljskih sličnosti omogućuje govornicima korištenje istog oblika za kodiranje izazito širokog raspona srodnih, ali ne potpuno jednakih prostornih odnosa.The main objective of this dissertation is to provide a cognitively plausible and usage-based semantic analysis of spatial prepositions in standard Italian. This study of prepositions was carried out within the cognitive linguistic framework. In order to find the most convenient theoretical and methodological model to be employed in this research it was necessary to take into consideration various approaches to spatial prepositions in general, beyond the realm of Italian semantic literature, which is comparatively scarce in terms of systematic research on prepositional meaning. Therefore, various approaches to the semantics of spatial prepositions at the international level were critically appraised and as a consequence a thorough critical account of their main features, advantages and disadvantages was provided. Namely, chapter 2 is dedicated to a brief historical overview of the development of semantic research on prepositions, including some recent contribution to this field. The totality of all approaches was divided into two main categories: (i) 'dominantly spatial' (geometrical and logical) approaches and (ii) 'functionalspatial' (functional) approaches. Within the first category, apart from radically spatial approaches, special attention was given to Annette Herskovits and her semantic analysis of English prepositions (1986) as well as to the further elaboration of her model by Di Tomaso (1996). On the other hand, Claude Vandeloise was recognised as a true founder and the main representative of 'functional-spatial' approaches in general. His approach to the semantics of French spatial prepositions (1986) was estimated as the most suitable model for the analysis of spatial prepositions. At the heart of Vandeloise’s approach is his refusal to acknowledge the validity of geometry and logic as the disciplines to rely on when describing the spatial relations as they are conceptualised by speakers and consequently encoded. Vandeloise opted for an anthropocentric view that hinges on the encyclopaedic knowledge and embodied experience of the speaker. The usage rules Vandeloise (1986, 1991) proposed for the French prepositions are based on the predominantly (but not exclusively) functional concepts: localization, concepts of naive physics such as the vertical axis, the bearer / burden relation, container / contained relation, physical accessibility and accessibility to perception, potential encounter, and general and lateral orientation. Chapter 3 is entirely devoted to five semantic studies of Italian prepositions that were evaluated as having the greatest impact on the field’s advancement (Devoto 1940, Parisi and Castelfranchi 1970, Crisari 1971, Weinrich 1978a and Pennacchietti 2006). Through a critical appraisal of their methodologies, some language-specific notions related to the Italian prepositional system were gathered. The most valuable contribution to the theoretical and methodological model of this study is Pennacchietti's distinction of applicative and retroapplicative prepositions. This variable was subsequently used as a tool to establish prepositional antonymy in this research, which made it possible to distinguish polysemy from homonymy. Chapter 4 deals with the category of prepositions in Italian grammar books. In point of fact, the influence of semantic analyses proposed by grammarians is extremely strong, and due to the lack of extensive semantic literature on prepositions these accounts of the category in question represent very valuable sources for the semantic research of prepositional meanings. Sixteen Italian grammar books are encompassed by this analysis: Battaglia i Pernicone (1977 [1951]), Regula i Jernej (1975 [1965]), Lepschy i Lepschy (1981), Dardano i Trifone (1985 [1983]), Rizzi (1988), Castelvecchi (Serianni) (1988), Sensini (1997), Andorno (2003), Salvi i Vanelli (2004), Patota (2006), Trifone i Palermo (2007), Schwarze (2009) and Ferrari i Zampese (2016). According to the criterion of the approach to prepositions three main directions within Italian grammaticography were identified: (i) ‘traditionalist’, (ii) ‘syntactic’ and (iii) ‘Schwarze’s approach’. The main patterns of classification and semantic analysis of prepositions were viewed and assessed. Christoph Schwarze's dual (both formal and functional) classification was evaluated as the most original and valuable of all due to the German author's independence from the Italian traditionalist patterns of classifying prepositions. It is the first attempt to provide a classification of Italian prepositions based on the semantic (not only formal) criteria and to incorporate relevant notions generated within the cognitive linguistic framework in the semantic description of Italian prepositions. In terms of theory and methodology the crucial part of the dissertation is chapter 5. In this part Vandeloise's model was re-examined and a number of theoretical questions were answered. With the aim of verifying what kind of approach to prepositional polysemy should be adopted – ‘lumping’ or ‘splitting’ in Taylor’s terminology (2012), an experimental radical polysemy analysis was conducted on one segment of the semantic structure of the preposition su. Consequently, a need for establishing principled criteria for discerning distinct spatial meanings for each preposition under investigation emerged. For this reason, the criteria proposed by Tyler and Evans (2003) within their ‘Principled Polysemy Model’ were adopted. However, the notion of ‘a spatial configuration holding between the trajector and the landmark which is distinct from the other senses conventionally associated with that preposition’ required further clarification. As a result of the ensuing discussion on the ‘spatial’ and the ‘functional’ in the meaning of spatial prepositions, in which the factors of force dynamics were regarded, a different position of the trajector and the landmark with respect to the axes of the three-dimensional system was defined as the criterion for establishing a distinct sense within the domain of space. Furthermore, the question of defining the primary meaning was considered. The criterion of the converging linguistic evidence as proposed by Tyler and Evans (2003) was tested on two Italian prepositions (su and sopra) and was judged as not fully viable. Therefore, the criterion for determining the primary sense of spatial prepositions was found in the level of complexity of the concepts that are used as primitives in the formulation of usage rules for each preposition. In this chapter, a method of symbolic representation of the (complex) concepts was developed and proposed. The figure of an octopus was chosen to represent the complexity of prepositional meaning, and its tentacles as symbols of active or non-active features pertaining to the concept. In this way the attribution of cognitive value to the graphic representation, which often occurs with network models, is supposed to be avoided. In the last part of chapter 5, paradigmatic relations between prepositions were considered, with special attention to the problem of prepositional antonymy. The very concept of antonymy in general was discussed, as well as specifically prepositional antonymy (from both the structuralist and post-structuralist perspective). The implementation of Pennacchietti’s variable of (retro)applicativity was advocated and proposed instead of a corpus-based methodology. In chapters 6-12 the theoretical and methodological model that resulted from the theoretical discussion and empirical verification in chapter 5 was put into practise and used for the purpose of analysing the spatial meanings of the following prepositions: su, sopra, sotto, al di sopra, al di sotto (Chapter 6), davanti (a), dietro (a/di), di fronte a, alle spalle di, dirimpetto a (Chapter 7), a destra di, a sinistra di, di/a fianco a/di (Chapter 8), vicino a, lontano da, accanto a (Chapter 9), in, dentro (a/di), fuori (di/da) (Chapter 10), a (Chapter 11) and da1 and da2 (Chapter 12). As it can be deduced, the chapters are organised in terms of the semantic criteria, namely the (complex) concepts that determine the spatial meanings of the prepositions. For all prepositions usage rules were proposed (with or without selection restrictions). The usage rules and selection restrictions were formulated with a view of covering all the prepositional uses verified in the examples taken from the University of Bologna’s corpus of written Italian CORIS/CODIS. The analysis showed that su and sopra are double near-synonyms as they are bisemic in the same way in the domain of space. The prepositional hyperonymy-hyponymy relationship was recognized for some pairs of nearsynonyms, e.g. davanti a – di fronte a; dietro (a/di) – alle spalle di, in – dentro (a/di). The only case where homonymy instead of polysemy was proved was in the case of da (da1 and da2 ). The main conclusion of the empirical analysis and the overall theoretical research boils down to the idea that the only way to account for prepositional meanings that often diversify even within the domain of space is by postulating complex concepts that consist of functional and spatial components that are organised according to Wittgenstein’s principle of family resemblances. The pre-linguistic and holistic nature of these concepts allows speakers to use prepositions for encoding a wide array of spatial relations. The structure of the concept that determines the meaning is available as a whole even in situations where not all components are activated. It is presumed that the functional nature (which implies inclusiveness based on associative links between individual features) of most of the components allows them to be arranged in clusters and exist in the speaker’s mental lexicon as wholes and not as individual distinct meanings. It is pragmatics, i.e. the speakers’ knowledge of the world and space (Žic Fuchs 1991, 1991/1992, 1997, 2009), that makes the use of such a limited group of lexemes an effective means for encoding the immense complexity of spatial relations