Thesis abstract:

«p style="text-align:justify"»«span style="color:black"»This research aims to use computationally-informed discourse analysis to explore news discourse about the participation of women in criminal offenses, in relation to their motherhood and general assigned role of caregivers, as means of identifying circulating ideologies about motherhood and crime. The study is, then, located in the area of computational social science, as it is interested in the social science inquiry about ideologies regarding crime and motherhood through the combination of socially grounded theories and computational methods. News discourse, as part of media discourse, is selected as its pervasive influence in socially shared ideologies is widely recognized (Fairclough, 1995; Talbot, 2007; Thompson, 1990; Van Dijk, 2006; Wodak and Busch, 2004). Theoretically, this research is informed by Systemic Functional Linguistics (SFL) and Corpus-based discourse analysis. SFL provides a model of language as a stratified semiotic system by which interpersonal, ideational, and textual meanings are negotiated in social life (Halliday and Matthiessen, 2014). Interpersonal meanings organized in the APPRAISAL system will be considered (Martin and White, 2015). The corpus approach to discourse analysis will rely on techniques that show quantitative patterns across texts in the selected corpus (Tognini, 2010; Bednarek and Caple, 2017). The methodology includes the building and analysis of a specialized corpus of news articles posted on Chilean online news sites, following a three-pronged approach (Bednarek, 2008, 2010) that considers large and small corpus analysis, alongside qualitative discourse analysis of selected texts.«/span»«/p»