This thesis investigates the use of the pragmatic particles (PPs) 'enfin' and 'écoute' in French film dialogue, and their translations in British English subtitles. Using a corpus of nine films and eight episodes drawn from two television series – all released in the UK between 2005 and 2015, and equating to approximately twenty-two hours – the study identifies tokens across a much wider range of contexts than has previously been possible using traditional corpora. The main contribution is an analysis of PP functions. The results for 'enfin' show a different functional distribution of the particle to other corpora, with corrective 'enfin' occurring significantly less frequently. The relatively large number of tokens of performative and emotional (or affective) 'enfin' allows for an elaboration of these two categories, and a tendency is observed for 'enfin' to appear as an apparent disagreement mitigator in discussions between peers. With regard to 'écoute', it is argued that écoute1 functions as a face-threat mitigator in unequal relationships and écoute2 as an FTA, although the particle is multifunctional and some tokens exhibit characteristics of both categories. Attention is given to combinations of 'enfin' and 'écoute' with other particles: while there is a clear tendency for disagreement-mitigating 'enfin' to co-occur with 'mais', and for the precision and restrictive subcategories of the corrective to co-occur with 'je veux dire', other previously documented combinations ('enfin bon' and 'ben écoute') are not frequently occurring in the present corpus. The thesis also makes a significant contribution to the field of Audiovisual Translation (AVT). The English subtitles show high rates of omission for both particles consistent with previous research, with disagreement-mitigating 'enfin' particularly vulnerable to omission. However, the analysis reveals a surprising pattern regarding 'écoute': a clear division of labour between ‘look’ (used to translate more confrontational tokens) and ‘listen’ (more conciliatory and socially distant). The study includes an experimental analysis of the subtitles relative to their character limits, demonstrating a potential new approach for researchers wishing to investigate the impact of various subtitling constraints.

Las series de la televisión son una fuente importante para el entendimiento de las estructuras lingüísticas y socioculturales de una lengua. Según muchos investigadores, el habla televisiva no puede ser tratada como lengua hablada real; según algunos, sí es posible. Otros argumentan que existen muchas diferencias entre el habla auténtica y el habla televisiva, pero que también tienen mucho en común. El objetivo de este trabajo es contribuir a un mayor entendimiento de cómo se puede investigar fenómenos sociopragmáticos mediante el diálogo televisivo, lo que esperamos lograr a través analizar la serie televisiva La Casa de Papel por muestras de descortesía usada. Para llevar a cabo esta investigación, veinticuatro escenas de la primera temporada de La Casa de Papel fueron recogidas y analizadas. La descortesía expresada fue clasificada según cinco categorías de descortesía y cinco categorías de respuesta ante la descortesía. Los resultados muestran que existen en el corpus ejemplos de cada categoría de descortesía propuesta, algo que indica que el habla de la televisión podría tener las mismas estructuras sociopragmáticas que el habla real. Aunque se necesita más estudios en el área para comprobarlo, argumentamos que el diálogo televisivo podría ser una fuente adecuada por investigaciones sociopragmáticas de la lengua hablada. / Television series are an important source for the understanding of the linguistic and sociocultural structures of a language. While some claim that television speak can not be treated like real spoken language, others claim the opposite. This study positions itself in the latter category. The objective of the present study is to contribute to a greater understanding of how one could investigate sociopragmatic phenomenons by using television dialogue as a source. This we hope to achieve by analyzing the impoliteness used in the television series La Casa de Papel. To accomplish this investigation, twenty four scenes from the first season of La Casa de Papel were gathered and analyzed. The expressed impoliteness in these scenes were later classified according to five categories of impoliteness and five categories of impoliteness responses. The results show that the corpus contain examples of every category of impoliteness proposed, which strengthens the claim that television dialogue can possess the same sociopragmatic structures as real spoken language. Although more studies in the area is needed to ascertain this claim, we argue that television dialogue could be an adequate source for sociopragmatic studies of spoken language.