Introducing the Intersex Research Corpus: Pathologized innate sex characteristics in biomedical research discourse 1950-2020

Abstract

<p>This talk begins to trace the development of academic biomedical discourse on intersex bodies across the seventy years between 1950 and 2020. The chosen methodology is corpus-assisted discourse analysis, which entails statistically supported analysis of a very large database of articles (i.e., a corpus of 1.3 million words). We have built the corpus of biomedical research articles because they are a key source of pathologization of certain innate sex characteristics in the past half century or so. As argued by Bruno Latour, scientific written language activates processes of social construction during the overall process of constituting scientific facts and findings (Latour 1987). These social construction processes are always difficult to perceive, criticize and scrutinize when one is inside a discourse community. Research articles, however, do provide important information about the communities in which they are generated and consumed (Latour 1987; Bazerman 1994). The first stage of analysis is a corpus-driven approach, meaning we subjected the corpus to analysis using specialist software without making hypotheses, to see what patterns present themselves in the data. Basic trends include a significant increase of mental health discourse in the corpus since 2006, with the semantic domains of ‘risk’ and ‘danger’ driving that focus in the texts as references to positive states of mind become scarce. People with pathologized innate sex characteristics are increasingly framed as ‘at risk’ of mental health issues, and cancer, discourses which of course were also present in the pre-2006 corpus, but as some other previous topics have receded these ones have grown in frequency. We hope discussion will generate fresh ideas for how to take a deeper dive into the corpus and explore the nuances of these findings.</p>