corpora.
This introduction starts with a foreword to the corpus-driven approach adopted (§1.1), which is further discussed in Chapter 2. Section 1.2 presents the research questions connected to the descriptive aim of the research. Lastly, Section 1.3 summarises the content to be found in each of the subsequent chapters.
←3 | 4→
1.1. The corpus-driven approach
In his seminal book, Toury (1995: 5, 11, 15) repeatedly alludes to a corpus or problem as the object of study of Descriptive Translation Studies (DTS). By focusing on a practical challenge for subtitlers, that is, the translation of fictive orality, the present monograph reports on a corpus study.
TS have long worked with digitised corpora, especially since the mid-1990s (Baker 1995). In TS, Corpus Linguistics is borrowed as a methodology ‘to study many of the processes involved in transferring information, ideas and concepts from one language to another’ (Picchi and Peters 1997: 253).
For the purposes of this research, the Corpus of Police Procedurals (CoPP) has been compiled. It is an English–Spanish parallel corpus including the source dialogue transcript and the DVD subtitles of 15 episodes of contemporary police procedurals; five episodes from three different TV series: Castle (ABC, 2009), Dexter (Showtime, 2006) and The Mentalist (Warner Bros, 2008).
The approach to the corpus has been data-driven. This means that preconceptions about the language material to be found in TV dialogue and subtitling have been kept to a minimum, and have been used exclusively to serve as a guideline to the research questions presented in the following section. Accordingly, manual annotation of the corpus has been undertaken from scratch and the results presented in this volume are those that have been qualitatively deemed as most relevant. Two sets of data have been annotated: syntactic and lexical. While ‘[a];n analysis of specific linguistic features necessarily shows a partial view of the data’ (Saldanha 2009: online), the selection has been intended to be as impartial as possible, in accordance with the recommendations made for corpus-driven translation research.
The corpus-driven methodology adopted has been complemented with corpus-based approaches that have allowed for a closer look at aspects discussed in specialised literature on spoken and written language. The analyses have combined quantitative and qualitative approaches on the basis that ‘the two approaches have complementary strengths and weaknesses’ (Biber 1988: 52). As specified by Keith (2008), quantitative studies entail ←4 | 5→data reduction, inference, discovery of relationships and exploration of processes that may have a basis in probability; thus, their scope is useful for the purposes of this study seeking to find norms. Qualitative approaches to corpora, on the other hand, ‘enable very fine distinctions to be drawn’ (McEnery and Wilson 1996: 70), which is an added value for descriptive research.
1.2. Aim and research questions
In his exploration of the genre of the novel, Bakhtin (2004/1981: 416) proposed the idea that ‘it makes no sense to describe ‘the language of the novel’ because the very object of such a description, the novel’s unitary language, does not exist’. Along the same line, this book does not intend to define the language of police procedurals and their subtitling, which, similarly, is not unitary. Rather, a corpus of three different series has been compiled in order to gather data to answer the research questions below. The series have been selected to be comparable among themselves (Chapter 4), so as to foster the detection of norms in dialogue construction and translation of the genre, which motivates the decision to refer to ‘police procedurals’ or ‘crime fiction’ in general throughout the book. Careful extrapolation of the results found in each production, in relation to the identification of genre norms, has been supported with the help of statistics, which have been used to assess the significance of the differences observed in the quantitative analyses of the CoPP.
This study pursues a linguistic characterisation of the dialogue and subtitling of the CoPP. Without further specification, this would be a rather unfeasible goal as the study could cover an unlimited range of linguistic aspects of TV dialogue and subtitling. The corpus-driven approach, however, prevents very specific questions from being drawn, since fine-grained research questions are typically the result of preconceptions or prior knowledge about the material in hand – and, as stated, this goes against the nature of corpus-driven analyses, which try to keep preconceptions to a ←5 | 6→minimum. Thus, from a general point of view the study strives to answer the three following research questions (RQ) which, as will be seen in the subsequent chapters, only become specific as corpus research advances:
RQ1.What are the main syntactic and lexical features of the source text (ST) and the target text (TT) in the CoPP, in connection to their hypothetical nature as intermediate genres in the continuum from spoken to written language?
RQ2.More specifically, what are the most salient features of fictive orality in the STs and the TTs?
RQ3.Are TV dialogue and subtitling genre-oriented?
The study seeks to find out how – if at all – TV dialogue and subtitling differ with regard to the occurrence in them of signs of fictive orality. Broadly speaking, the literature suggests that TV dialogue holds elaborate features of fictive orality affecting both the levels of syntax and lexicon, whereas subtitles have been reported to make use of rather simple, unmarked or standard language. Given the fact that interlingual subtitling involves transfer from one language into another, some of the results yielded by the study may reveal differences in the language system (English or Spanish) instead of differences in text type (TV dialogue or subtitling). The combination of corpus-based and corpus-driven techniques is expected to shed light on this contrastive analysis.
This introduction details the aim and scope of the study, the research questions underpinning it and the methodological approach adopted. The following chapters introduce the framework for the research conducted and present the results of the analyses. More specifically, the book is structured as follows.
Chapter 2 employs the cross-disciplinary notion of norm as a unifying thread to discuss the importance of identifying patterns of behaviour in ←6 | 7→the three disciplines directly connected to the study here reported; namely, Television Studies, Linguistics and Translation Studies. Consequently, the chapter shows how norms invariably refer to identifiable patterns of behaviour either in the form of genre invariants of TV fiction, prototypical features of spoken or written language, or recurrent types of translation solutions, all of which are traceable by means of corpus-driven research.
Chapter 3 sets out the framework of the two objects of study: scripted dialogue and subtitles. It provides theoretical information deemed relevant for interpreting
