Speakers
Description
This ongoing research to obtain a master’s degree in Linguistics at the Federal University of Ceará explores the extraction of specialized collocations and their analyses for the creation of a corpus-based bilingual glossary of legal discourse in American English and Brazilian Portuguese. The specialized collocations were extracted from a legal English corpus constituted of the subtitles from 134 episodes of the North American TV Series “Suits” (CS), which was submitted to analysis using the software Sketch Engine (Kilgariff et al., 2014). The comparable corpus English Web 2021 (enTenTen21) was chosen to find further evidence of usage and co-occurrence. Fromm (2011) categorizes television series depending on their use of terminology from completely fictional to a portrayal of real-life communication. CS can be considered a type of specialized text (Pavel, 1993), exploring the variety of legal discourse (legalese), encompassing here its written and printed aspects and its oral elements (Hoffman, 1998 apud Finatto, 2015). Monteiro-Plantin (2014) indicates lexical unit combinations are multiword units of relative stability, with a certain degree of idiomaticity, and conventionally used in specific situations. Specialized collocations would be the ones found in specialized discourse, such as legal terminology (Bevilacqua, 2005). Following Orenha-Ottaiano (2016, 2021) and making use of Corpus Linguistics with the help of the Sketch Engine tool, we were able to compile CS, counting 1,203,293 tokens and 960,603 words. Using different Sketch Engine tools (Keywords, Word Sketch and Concordance), we collected data to compose a list of keywords in descending order according to their frequency in CS. It was manually analyzed to exclude grammatical words and select the relevant results. Respecting the descending order of their keyness score (likelihood of the unit pertaining specific terminology), we selected fourteen words to serve as nodes for researching the collocations. The selection of candidates for specialized collocations was done considering a typicality score not lower than 3 (Glabasova, Brezina, & McEnery, 2017, Orenha-Ottaiano, 2016), as it would otherwise indicate that the co-occurrence of the words involved does not suggest sufficient stability or fixity, resulting in 41 candidates. The Concordance tool presented the nodes in context, organized by lines, centered and highlighted to facilitate the analysis of the elements that accompany them to define whether they are valid candidates for a lexical combination with compositional meaning. Those combinations that present relevant frequency throughout the research corpora were then analyzed in their morphosyntactic formation. According to Sinclair (1991), a collocation is composed by a node and a collocate, the node being the search word of research interest, while collocate would be what accompanies it. Hausmann (1985) understands the collocation as being composed by a base and a collocate, the base as an independent element, semantically autonomous and understandable/ translatable regardless of the collocation, while the collocate would be a modifying concept, interpretable within the collocation and depending on it for translation. Once all the collocations are analyzed, the construction of the glossary will be based on the methodological approach of Faulstich (2011) and the analysis of a corpus of ten lexicographic products focused on legal terminology selected from the web. The goal of this research is to assist language users in navigating the specificities of legal terminology and to explore the lexicographic approach towards languages for specific purposes. Furthermore, the corpus being constituted of a cultural product is an aspect which narrows the gap between academy and community, whom these results should serve in the first place.
