A New Resource for German Causal Language

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 5968–5977 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 5968 A New Resource for German Causal Language Ines Rehbein and Josef Ruppenhofer Data and Web Science Group Archive for Spoken German University of Mannheim, Leibniz Institute for the German Language Mannheim, Germany Mannheim, Germany ines@informatik.uni-mannheim.de ruppenhofer@ids-mannheim.de Abstract We present a new resource for German causal language, with annotations in context for verbs, nouns and adpositions. Our dataset includes 4,390 annotated instances for more than 150 different triggers. The annotation scheme distinguishes three different types of causal events (CONSEQUENCE, MOTIVATION, PURPOSE). We also provide annotations for semantic roles, i.e. of the cause and effect for the causal event as well as the actor and affected party, if present. In the paper, we present inter-annotator agreement scores for our dataset and discuss problems for annotating causal language. Finally, we present experiments where we frame causal annotation as a se- quence labelling problem and report baseline results for the prediciton of causal arguments and for predicting different types of causation. Keywords: Annotation of causal language, causal tagger, German 1. Introduction Understanding causality is crucial for making sense of the world around us. Thus, understanding causal expressions in text is an important prerequisite for Natural Language Understanding (NLU). Causality, however, is also known to be a concept that defies an unambiguous definition, thus posing a challenge not only for automatic NLU systems but also for human annotators. Several proposals have been made that describe causality from a philosophical point of view, such as the Counterfac- tual Theory of causation (Lewis, 1973), theories of proba- bilistic causation (Suppes, 1970; Pearl, 1988) and produc- tion theories like the Dynamic Force Model (Talmy, 1988). Counterfactual Theory tries to explain causality between two events C and E in terms of conditionals such as “If C had not occurred, E would not have occurred”. However, psychological studies have shown that this not always coin- cides with how humans understand and draw causal infer- ences (Byrne, 2005). Probabilistic Dependency theories, on the other hand, try to explain causality based on the un- derlying probability of an event to take place in the world. The theory that has had the highest impact on linguistic an- notation of causality is probably Talmy’s Dynamic Force Model which provides a framework that tries to distinguish weak and strong causal forces, and captures different types of causality such as “letting”, “hindering”, “helping” or “in- tending”. While each of these theories manages to explain some as- pects of causality, none of them seems to provide a com- pletely satisfying account of the phenomenon under con- sideration. This problem of capturing and specifying the concept of causality is also reflected in linguistic annota- tion efforts. Human annotators often show only a moderate or even poor agreement when annotating causal phenomena (Grivaz, 2010; Gastel et al., 2011), or abstain from report- ing inter-annotator agreement at all. A notable exception is …

Content truncated to 500 words