How Crime Narratives Shape Subjective Assessments of Guilt

Crime reporting is a prevalent form of journalism with the power to shape public perceptions and social policies. How does the language of these reports act on readers? We seek to address this question with the SuspectGuilt Corpus of annotated newspaper crime stories. For SuspectGuilt, annotators read short crime articles and provided text-level ratings concerning the guilt of the main suspect as well as span-level annotations indicating which parts of the story they felt most influenced their ratings. SuspectGuilt thus provides a rich picture of how linguistic choices affect subjective guilt judgments.

In addition, we build BERT-based predictive models based on SuspectGuilt and show that these models benefit from in-domain pretraining with unlabeled data. Besides, we experiment with joint supervision from the text-level ratings and span-level annotations. Such models might be used as tools for understanding the effects of crime-reporting on a large scale.