World Building in Spanish and English Spoken Narratives (Advances in Stylistics)

index.cfm -Visit the website and download the text-world diagrams: www.bloomsbury.com/lugea-world-building -Use the VUE User Guide wiki to help browse the diagrams: https://wikis.uit.tuft s. edu/confl uence/display/VUEUserGuide/ENGLISH+USER+GUIDE -To see key features of each story's text-world diagram, open 'Windows' and then 'Layers' and 'Map Info' Contents List of Figures x List of Tables xi Preface xiii 7.9 Boulomaic modal-worlds 7.10 Boulomaic modal expressions 7.11 Deontic modal-worlds 7.12 Deontic modal expressions 7.13 Epistemic modal-worlds 7.14 Epistemic modal markers 7.15 Evidential markers 7.16 Hypothetical-worlds and conditional tense In the worlds-based theories used in linguistics, psychology, philosophy and literary semiotics, we frequently use the term 'world(s)' to refer to clustered representations of aff airs: fi ctional, possible or mental. Th is study aims to compare the linguistic apparatus available to and used by Spanish and English speakers when they are asked to create a narrative world. According to Werth (1999) , reference is the key linguistic tool in creating a text-world; that is, a mental representation of discourse. One of the main ways of referring in language is through deictic reference, whereby events are related spatially, temporally and personally to the speaker. A signifi cant development of Text-World Th eory has been Gavins' ( 2005 , 2007 ) identifi cation of modality as a phenomenon that excludes information from our perception of the way things currently are (as opposed to how they should or might be). Deictic reference and modality, then, are both what we call 'world-builders'; they relate referents and propositions in language to a originating viewpoint from which a text-world can be built in discourse. Because of their centrality in world building, these two phenomena will be the key features under investigation in this study. Deixis and modality ## Initial defi nitions Th e two linguistic phenomena that this study compares cross-linguistically are deixis and modality. As they are described in detail in Chapter 2 , a brief defi nition of each will suffi ce here. Th e word 'deixis' comes from Ancient Greek δειξις (meaning display or demonstration ) and describes those terms which, in order to be interpreted, require knowledge about the context in which they are used. For example, the demonstrative this or Spanish este , the pronouns he or él , and temporal adverbs such as now or ahora are all deictic terms as we cannot know their referent if we do not know the space, time or person at their source. In fact, these examples respectively represent the three 'core' deictic categories: space, person and time. Deictic terms have fi xed grammatical meanings, but the referent changes with each instance of use. As such, deixis is a notoriously tricky object of linguistic study ( Levinson 2006 : 97) and interests linguists from a broad range of fi elds including pragmatics, semantics, psycholinguistics, cognitive linguistics, computational linguistics, as 1 Introduction 'who sees' and 'who speaks' into one viewpoint. Instead, he identifi es three kinds of focalization, with varying degrees of narratorial access to the characters' psyche. Although Genette fails to give concrete linguistic criterion for these categories, other authors (e.g. Fowler 1996 ; Simpson 1993 ) have also identifi ed a range of ways for the narrator to have access to the characters' perspective, for which they cite deixis and modality as key factors in doing so. Although it is not possible to describe all the taxonomies here, suffi ce to note that deictic terms do not necessarily contribute to a unifi ed deictic centre, but can be used to blend the viewpoint of narrator and character, as in the following example of FID: 'Mary turned and stared. Were these the tulips she had seen here yesterday?' ( Jones 1997 : 71). Although the narrator uses the third person 'she' , which indicates a personal distance from Mary's viewpoint, the proximal deictic spatial adverbs these and here indicate spatial proximity. Th is blurring of viewpoints is mirrored on the temporal plane in this example, as the events are related in the past perfect, indicating the narrator shares a diff erent temporal deictic centre from Mary, yet the use of deictic temporal adverb yesterday shows a shared deictic centre. Despite the fact that the narrator is external to the character, (s)he has access to her perception of events shown here through the deictic choices. It is clear that deixis is fundamental in encoding the perceiving subject in narrative, be that a monologic or plural subject. As mentioned in the previous section, subjectivity can be expressed through modality as well as deixis, and it is the use of modalized propositions that also express 'ideological point of view' ( Fowler 1996 ;Uspensky 1973 ) or 'conceptual point of view' ( Chatman 1978 ). As Simpson (1993) reveals in his Modal Grammar of viewpoint, through the narrator's expression of belief, obligation or desire, the existence of a perceiving subject in the narrator is asserted, or alternatively, the omniscient narrator shows access to the character's inner thoughts. Th is is certainly the case with epistemic modality, which involves the most internalized perception of events, leading Simpson to declare that '[t]he epistemic system is possibly the most important regarding the analysis of point of view in fi ction ' ( 1993 : 48). It is clear, then, that modality as well as deixis contributes to establishing viewpoint in narrative. McIntyre ( 2006 : 36) points out that many of the existing frameworks for the analysis of point of view do not adequately account for how changes in point of view are wrought in a text. Th e cognitive approaches to point of view in narrative that were alluded to at the start of this section are better equipped to deal with shift s in perspective over the course of a narrative text and are explained in detail later (Section 1.2.1). working in various disciplines within the social sciences at the State University of New York, Deictic Shift Th eory attempts to explain how readers are guided through a narrative text and how they empathize with various points of view. It identifi es a 'deictic fi eld' , whereby a set of deictic expressions all relate to the same deictic centre which, in the fi ctional world, could be a narrator or character. Readers are typically able to suspend their own egocentric deictic fi eld and switch to that of whichever person is indicated through the language of the text. Th is, then, is the fi rst deictic shift in the reading process. Subsequently, the various shift s in deixis throughout the narrative, brought about by changes in tense or pronouns, indicate to the reader the shift s in viewpoint. Th ere is no doubt that Deictic Shift Th eory off ers a very dynamic understanding of how the deictic centre moves in a text, which was lacking in the traditional stylistic taxonomies of point of view (as mentioned in Section 1.1.2). However, its potential effi cacy in this particular study is doubtful. Th e model is primarily a reader-orientated one, in that it follows the deictic clues laid out for the reader to interpret. As such, Deictic Shift Th eory is primarily designed to account for narrative comprehension. In contrast, this project studies narrative production to compare how Spanish and English speakers produce deixis, in order to fi nd out if they use deictic dimensions (e.g. space, time, person) in world building diff erently. Th e fi nal reservation I have with the model is more general, and this relates to the sole emphasis it puts on the interpretation part of the communication process. Th e position taken here is that deixis is conducive to a more relational and sociocentric approach (as elaborated in Chapter 2 ), so the phenomenon demands a model that considers all participants in the production and reception of deictic terms. Another drawback of Deictic Shift Th eory for this study is that it does not deal with how modality is used in the construction of narrative viewpoint. One theory that does deal with both deixis and modality is Text-World Th eory ( Gavins 2007 ;Hidalgo Downing 2000 ;Semino 1997 ;Werth 1999 ), a framework that analyses the mental representation that users of language co-create in their minds when they participate in discourse. In this way, it also fulfi ls another of the criteria for current purposes, in that it considers the production and reception of language. As explained at the outset of this chapter, Text-World Th eory posits that discourse participants create a mental representation of the discourse (the text-world) and use deictic terms as 'worldbuilders' to establish spatio-temporal coordinates therein ( Werth 1999 : 180-209). Like Deictic Shift Th eory, it then tracks the shift s in deixis throughout the discourse, as generated by changes to the spatio-temporal coordinates to which both producers and addressees align themselves in the joint text-world. Th e use of modality generates 'modal-worlds' ( Gavins 2005 ; 2007 ), where the propositions contained are held at a remove from the text-world as they are not verifi able by participants. Th us, the textworld model allows for the analysis of how deixis and modality work together in 'world building'; that is, constructing the viewpoint from which all discourse participants access the narrative text-world. Th e text-world is said to be infl uenced by various factors of 'discourse-world' , which includes all aspects of the setting surrounding the language event, as well as all the discourse participants' individual knowledge and experience. In this way, the model is, the universal, the cultural and the individual -making it extremely useful in the comparison of a language pair for the benefi t of translation. In making a case that translating meaning across languages is possible, Jakobson argued that this is because 'cognitive experience' is universal ( Jakobson 2000 : 115). Regardless of the particular language we use, our cognitive experience is shaped by two vital things that everyone has in common: a mind and a body through which we experience the world. Th erefore, to some extent our phenomenological experience and the cognitive mechanisms with which we process that experience are comparable. Th e existence of about sixty basic meaning elements in all the world's languages means that 'all human cognition rests on the same conceptual bedrock' ( Wierzbicka and Goddard 2004 : 144). Cognitive approaches to language, borrowing heavily from cognitive psychology, emphasize how we use boiled-down, abstract representations (such as image-schemata [ Langacker 1987 ], mental models ( Johnson-Laird 1983 ) or scripts [ Schank and Abelson 1977 ]) to be able to conceptualize, store and retrieve the information which is used to process language. In concentrating on the fundamental image schemata and cognitive processes behind the linguistic forms, Cognitive Linguistics overrides linguistic and cultural diff erences, addressing universal ways of encoding the world through language. In doing so, it provides an ideal vehicle for accessing the common ground upon which we can base a cross-linguistic comparison. However, despite this fundamental universality, there is clearly a wide range of individual and cultural variation in how we cognitively process language. As Tabakowska remarks, 'While the physical, sensual dimension of perception is universal for all human beings, the psychological dimension is conditioned by sociological and psychological factors, and based upon [...] experience, conditioned by individual cultures and by people's individual make-ups ' ( 1993 : 59). Cognitive Linguistics addresses this seemingly infi nite variation by taking seriously the context surrounding the discourse. In fact, it takes a particularly broad view of what constitutes context, encompassing social and personal experience and knowledge as contextual factors which shape meaning construction ( Langacker 1999 ; cf. Porto Requejo 2007 for a concise account). Th at is, the particular experience and knowledge of each individual or each culture, as an element of the context, will infl uence the production and reception of a text. In relation to the reception of literary texts, Stockwell notes that a cognitive account can off er 'a unifi ed explanation of both individual interpretations as well as interpretations that are shared by group, community and culture ' ( 2002 : 5). It is this capacity of Cognitive Stylistics to account for the individual, the universal and the cultural that makes it the ideal approach for a contrastive analysis of Spanish and English world building in narrative. Narrative style ## Style in language or style of a language? Since its inception, Text-World Th eory has been used and developed mostly by those who work in the fi eld of Stylistics to analyse the ways in which texts use phenomena such as modality, reference or metaphor to achieve stylistic eff ects. For stylisticians, 'thinking for speaking ' ( 1987 , 1996a ) a 'weaker' form of Whorfi anism whereby our language may not condition exactly how we think, but it does condition how we conceptualize events for verbalizing them. His research has focused on the ways in which speakers of diff erent languages encode motion events in narrative, building on Talmy's ( 1985Talmy's ( , 1991 ) ) distinction between satellite-and verb-framed languages, and he has found that languages tend to concentrate on either the manner of an action or the path that it takes. For example, in describing the same scene an English speaker is likely to say, he swam across the river , whereas a Spanish speaker is more It is hoped that the results of the fi rst research question will tell us not only how Spanish and English speakers use modality and deixis in narrative, but also what that means for their perspective-taking in general, how that might refl ect cognition and how it might help in translation. ## Th e frog stories Th e diffi culty with comparing how Spanish and English speakers use deixis and modality to build a narrative text-world is in fi nding comparable data. Using natural data, how can we compare narratives when the narrators will inevitably be taking perspective on diff erent textual content? Th e aforementioned children's picture book that Slobin used in his research was fi rst used in a PhD thesis some decades ago ( Bamberg 1985 ), and this has since become a popular narrative elicitation method in linguistic research. Th e methodology uses a wordless picture book ( Mayer 2003 ) that tells the story of a little boy and his dog on a hunt for their frog, and the resulting spoken narratives have come to be known as 'the frog stories' ( Berman and Slobin 1994 ). Berman and Slobin (1994) applied the frog story methodology to speakers of diff erent ages in fi ve languages -English, Spanish, German, Turkish and Hebrew -to explore how they developed the skills to relate events in narrative. Since then, frog story research has fl ourished; so much so, that twenty years aft er their original publication, a volume detailing all the subsequent frog story research has been published ( Strömqvist and Verhoeven 2004 ). While the early work by Berman and Slobin studied the development of narrative styles in child speakers of various languages, the picture book has been used for a variety of linguistic research projects, gathering narrations from native speakers or second-language learners, using spoken means, sign language or gesture. Most of the studies have had contrastive or typological interests; there now exists data for more than thirty global languages and the work continues. Th e frog story's success is testament to the simple effi cacy of the methodology, which provides an extra-linguistic visual stimulus to elicit crosslinguistic comparable data. Th e frog story methodology off ers a promising solution for the current research aims, as it allows for the elicitation of narratives that have roughly the same textual content, but leaves the deictic and modal choices down to the narrators' preference. According to Slobin, 'the strength of the frog story lies in the wordless presentation of a readily understood plot, with sufficient complexity to allow for detailed analysis of temporal, causal and spatial dimensions of events ' ( 2004b : 115). Thus the visual input stimulates the narrators to conceptualize, formulate and articulate the events using spatio-temporal deixis, allowing for the analysis of Spanish and English use of deixis in relation to common conceptual content. Using the frog story method, they found that the languages examined 'presented a distinctive style, especially in representations of time and space' ( Slobin 2005 : 115). The pictures used to elicit the narration form a sequence of events which evokes: "Text World Theory is a powerful framework for discourse analysis that, thus far, has only been used in monolingual Anglophone stylistic analyses. This work adapts Text World Theory for the analysis of Spanish discourse, and in doing so suggests some improvements to the way in which it deals with discourse - in particular, with direct speech and conditional expressions. Furthermore, it applies Text World Theory in a novel way, searching not for style in language, but for the style of a language. Focusing principally on deixis and modality, the author examines whether Spanish speakers and English speakers construct the narrative text-world in any patterned ways. To do so, the 'frog story' methodology is employed, eliciting spoken narratives from native adult speakers of both languages by means of a children's picture book. These narratives are transcribed and subjected to a qualitative text-world analysis, which is supported with a quantitative corpus analysis. The results reveal contrasts in Spanish and English speakers' use of modality and deixis in building the same narrative text-world, and are relevant to scholars working in language typology, cross-cultural pragmatics and translation studies. These novel applications of the Text World Theory push the boundaries of stylistics in new directions, broadening the focus from monolingual texts to languages at large."--Bloomsbury Publishing. Cover Half-title Title Copyright Dedication Contents List of Figures List of Tables Preface 1. Introduction 2. Deixis and Modality 3. The Frog Story Corpus 4. Text-World Theory 5. Departures from the Text-World 6. Analysis of Temporal World Building 7. Analysis of Spatial, Personal and Modal World Building 8. Conclusions References Appendix I Appendix II Index