Differential Agent Marking and animacy

This paper investigates the cross-linguistic effects of animacy on overt Agent marking on the basis of a 200-language sample. It is shown that animacy-driven Differential Agent Marking (DAM), with different case markers for animate and inanimate Agents, is typologically rather uncommon. In order to account for this type of DAM and its scarcity, it is argued that a related phenomenon needs to be taken into consideration: in a considerable number of languages, inanimates cannot be construed as the Agent of a transitive clause. In this paper, both animacy-driven DAM and this restriction against inanimate Agents are explained using the notion of unexpectedness. Inanimates are unexpected to occur as Agents because they are not readily associated with Agent semantics. Because of this unexpectedness, inanimate Agents can either be avoided, or they can receive special case marking (i.e. DAM). Since the first strategy is preferred, this explains why animacy-driven DAM is not very common. Finally, it is argued that the results of this study are also relevant from a broader theoretical perspective. They reveal both symmetries and asymmetries between the Agent role and the Object role, and they point out a number of problems with the Silverstein hierarchy.