This repository contains the supplementary data of the experimental studies described in Kiss, Pieper, Börner (2023), revised version after initial submission to Glossa. There is a folder for the two experiments, each containing an independent ReadMe file.
The experiments were conducted as part of our project Position and Interpretation of adverbial PPs in German clause structure, which is supported by a DFG grant (KI 759/8). The interaction between the syntactic position and the interpretation of adverbials in German clauses has been the subject of a recent debate. Typically, a hierarchy of adverbial classes is proposed. Order constraints between elements of different adverbial classes, as well as between adverbials and arguments are then derived from the membership in the adverbial classes. Alternative proposals attribute adverbial positioning solely to semantic constraints imposed by the adverbials on the phrases they modify. The focus of the first funding period (KI 759/8-1) was on analyzing adverbial PPs. The respective results of the first funding period indicate that order preferences originate in an interaction of adverbial-specific and global constraints.
There is a long-standing debate whether adverbials assume base positions in German clause structure. A major division can be drawn between analyses proposing base positions dependent on class membership, and analyses allowing free generation of adverbials unless adverbials from different classes modify the same event. Based on two experimental studies on event-internal modifiers we present an analysis which rejects base positions, as well as free generation of adverbials in isolation. The results indicate that event-internal modifiers occupy several positions in German clause structure. Adverbials are base generated but constrained by a structural condition together with linear precedence rules based on a ranking of thematic roles, agentivity, animacy, and the category of the phrase. Event-internal adverbials combine with a verbal projection (identifying the respective event variables) but require the identification of an additional individual variable with its reference phrase. Linear precedence rules apply to all event-internal adverbial types in both affirmative and privative interpretation, except for one linearization constraint based on thematic roles which only applies to affirmative comitatives. The interaction of the identification requirement with the linear precedence constraints accounts for the diverging pat-terns of serializations observed with event-internal modifiers.
We would also like to thank Sarah Broll for her support in the preparation of the experiments and fruitful discussions.