The paper presents IMAGACT, a freely accessible online ontology of action verbs. Based on corpora of spoken Italian and English, it has been developed at the Universities of Florence and Siena and the CNR Pisa under the direction of Massimo Moneglia. More than 1000 particularly frequently verbalized actions are represented by short animations or film scenes and assigned to the corresponding verbs in currently about 20 languages. The ontology can be queried as a bilingual dictionary, for explicit language comparison and via the visualized action types. The article focuses on the integration of German verbs into the database, which was carried out by the author. The comparison of German and Italian is particularly interesting as the languages follow typologically different patterns: German as an endocentric language uses a multitude of specific verbs to refer to the given actions, while Italian, with its preference for general verbs, uses significantly fewer verbs to refer to the same number of actions. This leads to divergence problems for Italian-speaking learners of German as L2. The typological differences are confirmed by IMAGACT. At the same time, the platform offers a tool to facilitate language acquisition and translation.
