Elicitation and analysis of sexual concepts: overview of a cognitive sociolinguistic study

In this paper, we present a cognitive sociolinguistic approach to the variation of sexual concepts in discourse. Sexual concepts belong to the realm of linguistic taboos. These are concepts that are subject to certain restrictions of use, in a varying degree of interdiction according to cultures, social groups or individuals (Allan and Burridge, 1991, 2006). The direct consequence in actual discourse is that the utterance of sexual concepts is often a problem for speakers, and the linguistic solution they find for it reveals much about their conceptualization of sexuality. For instance, láser ‘laser’, nabo ‘turnip’, lo suyo ‘their thing’ are all expressions for the concept ‘penis’ that reveal different underlying conceptualizations of the same sexual organ. A cognitive sociolinguistic study of these semantic choices can help us in answering the following questions: how are these conceptualizations distributed in society? And how do they relate to different discourses about sexuality? This research questions forced us to deal with methodological difficulties in the phases of data collection and analysis of semantic frequencies. We decided to collect our own corpus of semi-structured interviews about sexuality in Madrid, Spain, which were designed as a semi-directed opinion survey, for the indirect elicitation of sexual concepts (MadSex, 1 million words). For the analysis, we rely on the Cognitive Sociolinguistics framework (Krinstiansen & Dirven) and we focus particularly on the analysis of variation at the conceptual onomasiological level. The sexual concepts were manually extracted and annotated according to the type of construal. We observed their overall frequencies and we analyzed the distribution of the construal types according to social variables of our speakers (gender, age, education, and micro-social variables such as stance towards sex talk). Our results up until now show significant differences among groups. For instance, when talking about body parts and physiology, men prefer to use metaphors while women prefer metonymies, and older generations tend to use more generic expressions. On the other hand, when talking about ideologically-marked concepts, such as ‘abortion’, the ideological position of the speakers explains their semantic preferences: pro-abortion speakers prefer the direct construal ‘to abort’, while anti-abortion speakers use expressions based on the concept ‘to kill’, among others. This kind of results highlights the social meanings present in the variation of sexual concepts in discourse. This is further proved by the fact that sexual variants can be quoted in order to (stereotypically) perform the speech of others (for instance, when quoting men’s voices, women will use metaphors)