zum Inhalt springen

Philosophische FakultätLingCologne

Predicting the Development of Case Marking Paradigms in Heritage German?

Judith Schlenter1, Marit Westergaard1,2

1UiT The Arctic University of Norway, 2NTNU Norwegian University of Science and Technology

In an eye-tracking experiment, we showed German native speakers (n=35) two visual scenes depicting an event involving three animate entities, and at the same time they heard a sentence that included a ditransitive verb. The participants’ task was to select the target scene. One half of the 24 ditransitive sentences followed the order indirect object > direct object (IO-DO), the other the order direct object > indirect object (DO-IO). For a fully counterbalanced design, we additionally manipulated the order of masculine and feminine nouns after the verb (Table 1). A previous acceptability judgment task with a separate participant group (n=48) revealed higher ratings for IO-DO than for DO-IO, irrespective of noun order.

Preliminary analyses suggest that it is not just the order of objects, but also the perceptual salience of the case cue that influences the processing of ditransitives: If dative and accusative are marked by der and die respectively, participants prefer the target over the competitor scene prior to the onset of the second object (Figure 1, right), indicating that they are able to predict its thematic role. However, if the difference between dative and accusative corresponds to the difference between dem and den, we observe no stable target preference prior to the second object. Moreover, den corresponds to the condition with the lowest accuracy (73%) and longest response latency.

Our findings extend previous research showing listeners’ predictive use of case marking on sentence-initial arguments to postverbal arguments while controlling for factors such as animacy and perceptual salience.

Table 1. Example item in all four experimental conditions. Cross-splicing is applied to the critical segment for a prediction effect (in grey) after article offset. The rightmost columns show the respective target and competitor scene for a sentence.
Figure 1. Fixation proportions for the target (red, solid line) and competitor scene (blue, dot-dashed line) from article offset aligned to zero for masculine nouns (left) and feminine nouns as first object (right). The order indirect object > direct object is shown at the top and the order direct object > indirect object at the bottom. The dashed vertical line indicates the average onset of the second object. Note that only correctly answered trials are included into the eye-tracking analyses.