Refugees’ dehumanization in the Spanish media: A corpus-assisted study within the semantic preference framework

Abstract

Abstract The dehumanization of migrants and refugees in the media has been the object of numerous critical discourse analyses and metaphor-based studies which have primarily dealt with English written news articles. This paper, however, addresses the dehumanizing language which is used to refer to refugees in a 1.8-million-word corpus of Spanish news articles collected from the digital libraries of El Mundo and El País, the two most widely read Spanish newspapers. Our research particularly aims to explore how the dehumanization of the lemma refugiado is constructed through the identification of semantic preferences. It is concerned with synchronic and diachronic aspects, offering results on the evolution of refugees’ dehumanization from 2010 to 2016. The dehumanizing collocates are determined via a corpus-based analysis, followed by a detailed manual analysis conducted in order to label the different collocates of refugiado semantically and classify them into more specific semantic subsets. The results show that the lemma refugiado usually collocates with dehumanizing words that express, by frequency order, quantification, out-of-control phenomenon, objectification, and economic burden. The analysis also demonstrates that the collocates corresponding to these four semantic subsets are unusually frequent in the 2015–16 period, giving rise to seasonal collocates strongly related to the Syrian civil war and other Middle-East armed conflicts.