MadSex: Collecting and Analysing a Sociolinguistic corpus of sexuality in the city of Madrid

Pizarro Pedraza, Andrea [UCL]

The social motivation of linguistic taboo is widely acknowledged in the literature (Allan & Burridge, 1991, 2006). Nonetheless, empirical studies on the sociolinguistic variation of taboo concepts are rare, particularly in spoken discourse. The analysis of taboo language inevitably faces two major problems: the choice of an appropriate corpus and the methodological difficulties in the analysis of taboo expressions. In Spanish, the available spoken corpora lack both sufficient taboo expressions and the social information needed for their quantitative analysis. Furthermore, the extreme variation in the expression of taboo concepts (metaphors, metonymies, vagueness, etc.) limits the use of automatic data extraction tools. With these problems in mind, we decided to create a spoken corpus on the topic of sexuality in Madrid. The methodology was planned to collect abundant and comparable data, both linguistic and social. We wrote a questionnaire designed to indirectly elicit sexual concepts through face-to-face, recorded interviews. During fieldwork, we collected 54 interviews in two districts of Madrid. The corpus is pre-stratified by gender, age and level of education, although broader social information is available. After transcription, the Madrilenian spoken corpus of Sexuality (MadSex) reaches ca. 1 million words and includes varied sexual concepts. For the analysis of the semantic variation, we adopt Cognitive Sociolinguistics (Kristiansen & Dirven, 2008) as a framework. The results of our case studies until now shed new light on relevant topics for linguistic taboo, such as the role of metaphor in the expression of taboo concepts or the tabooisation of feminine sexuality.