Audiovisual texts can be reconfi gured across languages and cultures in multiple ways, as outlined in Chapter 1. Traditional forms of audiovisual translation, such as subtitling and dubbing, remain primarily anchored to discourses on correspondence or equivalence between source and target texts. Emerging assistive varieties such as audio description, on the other hand, call for an alternative understanding of translation as mediation – thus acknowledging the active role that audio describers play in selecting what needs to be transferred to their sensory impaired viewers and how best to do so on a case by case basis. Consequently, audiovisual translation accommodates multiple ontologies , i.e. different answers to the question ‘what is audiovisual translation’. According to Chesterman (2006: 10), this co-existence of ontological views under a common label calls for a concept of translation that has a ‘fl exible cluster shape rather than a prototypical form’. Understanding translation as a cluster concept thus allows for a productive interplay between multiple theorizations that foreground different aspects of the translation phenomenon.