The transformation of science fiction: A quantitative analysis of the emotional pattern at the hugo awards

Abstract

Science fiction as a literary genre has been in constant evolution since its origins, at the beginning of the last century. However, in recent decades this evolution has accelerated from being a conservative genre to adopt approaches closer to political progressivism (Oleszczuk, 2017). The aim of this research is to determine whether this transformation is also seen in the stylistic construction of the novels, and more specifically, in their emotional pattern, conducting a dual synchronic study of the 1960s and 2010s. Specifically, it has been analysed 24 novels that received the Hugo Award, probably the most relevant in the field of science fiction and, to a lesser extent, fantasy, 12 from the 1960s and another 12 from the 2010s. Three computational linguistic techniques were applied to this corpus of documents: sentiment analysis, clustering and principal components, in order to identify homogeneous groups of novels. The results show the grouping into three clearly differentiated clusters, with dissimilar emotional patterns, and which correlate strongly with the decade in which they were written. This confirms that, in addition to the thematic evolution, there has been a substantial change from a stylistic point of view.