Identifying stance in legislative discourse: a corpus-driven study of data protection laws

Mirroring public ideologies and value systems in legislative discourse, stance not only functions as a powerful instrument for legislators to balance legal values and stakeholders’ interests but also acts as a valuable reference for individuals to understand legislative texts. This study conducts a corpus-driven analysis of stance expressions in legislative discourse. Using three self-compiled corpora that incorporate data protection laws from the United States, the European Union, and China, we apply Hyland’s stance model to contrastively analyse evidence of hedging, boosting, self-mention, and attitude markers across these jurisdictions and eventually propose a specialised research model of stance in law. This study unveils the nature of modesty and sufficient discursive space of data protection laws, as well as legislative values and public ideologies conveyed by different jurisdictions within the broader socio-legal cultural context. Besides uncovering the legal constructiveness of data protection laws, the results also suggest that the overall representation of stance in data protection legislation aligns with its performance in legislative discourse, which showcases a legislative tendency to achieve an overtly neutral appearance through covert stance expressions.