Automated detection of style rule violations in legislative texts


Good legislation must not only meet criteria of substance but also comply with formal and linguistic standards. Manual assessment of the compliance of legislative texts with these standards is extremely costly. The Centre for Legislative Studies has therefore collaborated with the Institute of Computational Linguistics to investigate whether the linguistic assessment of legislative texts can be supported by computers. The goal was to develop methods by which violations of certain style rules specific to legislative drafting can be detected automatically. The project focused on German-language legislation in Switzerland.




Dr. Stefan Höfler (Project leader)
Dr. Kyoko Sugisaki (Institute of Computational Linguistics)
Prof. em. Dr. Michael Hess (Institute of Computational Linguistics)
Prof. Dr. Felix Uhlmann (Centre for Legislative Studies)


SNF (Personen- und Projektförderung), Forschungskredit der Universität Zürich


March 2011 – February 2015