Semantic Roles and Decomposition of Agency

Social network analysis relies on a mapping of relationships or interactions among persons or groups. Often a particular relationship is assumed (e.g., marriage or trade ties), but all sorts of relationships can be taken into account. One way to deal with social network analysis of texts is to extract all event structures consisting of an actor, undergoer, and event, grammatically realized as subject, (in)direct object, and verbal predicate. Leviticus 17–26 contains 181 different verbs resulting in a complex social network. This prompts the question of how to differentiate between various verbal events in a quantifiable way.
The question is addressed by invoking linguistic theories of agency since all events presuppose a certain amount of effort on behalf of the agents. By implication, verbs can be measured according to agency required, that is, amount of force and volition. This chapter develops a framework for capturing agency from verbal events on the basis of the Role and Reference Grammar theory for deriving semantic roles from the lexical aspect of verbs, also known as Aktionsart. It is argued that dynamicity and causation are the two most important verbal features for decoding agency.