The representation of migrants in Spanish judicial decisions: using corpus data to refute hate speech

The phenomenon of immigration and its depiction in media texts have been examined profusely within the field of corpus-based discourse analysis (Gabrielatos and Baker, 2008; Baker et al., 2013; and Blinder and Allen, 2016). This research seeks to present it as reflected in a corpus of 600 judicial decisions issued by Spanish courts in the years 2016 and 2017. This analysis was motivated by the rise of extreme right-wing parties in Europe in recent years. Such parties dehumanise immigrants and portray them as a threat to the welfare state. On first examination, the results appear to dissociate immigration and crime since a considerable percentage of the keywords obtained (about 20 percent) revolves around three major topoi (namely, `family', `territory/access' and `legal punishment') and there is no evidence of any major offences or crimes amongst the top-ranking lexicon. The study of the collocate networks of the keywords within the category `legal punishment' confirms our initial perception; in fact, out of twentyone collocates, only the word delito (`crime') itself collocates with terms referring to typified crimes such as violencia (`violence'). In parallel, the data were triangulated using the text-classification software UMUTextStats (Garcia-Diaz et al., 2018). The results of this second analysis also confirm our initial observations.