Project C1

2nd Phase - Contextually licensed non-canonical word order in language comprehension

The sub-project C1 is a joint project of linguistics (grammar theory) and cognitive psychology to understand information-structurally marked forms, primarily word order variants with sentence-initial non-subject. Of central interest is the concept of contextual licensing, by which we understand that a licensing context reduces the processing costs when understanding a marked form, compared to the null context or the non-licensing context. In the case of spatial-relational statements, we identified the licensing types Given-New and Poset in the first application period. In the second application phase, a differentiated terminology of licensing types for transitive structures should be developed and experimentally validated on the basis of corpus data. We expect an important contribution to this from the consideration of the prosodic realization of word order variants with and without a licensing context. We use the innovative method of eye movement measurement when reading aloud in comparison to reading quietly. We hope to gain new insights into the interaction of the information-structural function of word order variation, prosody and context in understanding language.

Download full description

1st Phase - Contextually licensed non-canonical word order in language comprehension

Although word order in German is relatively unrestricted, isolated sentences are comprehended more easily (i.e., read more quickly) with initial subjects than with, e.g., initial objects. This is in accordance with the claim that sentences with initial subjects have canonical word order that is neutral with respect to information structure (IS). Non-canonical - topicalized or scrambled - word order signals an IS function to the reader/listener. Accordingly, an isolated non-canonical structure renders comprehension difficult since without context it does not serve the IS function signalized. The aim of the project is to identify IS functions of non-canonical word order and to validate these by means of licensing contexts, in which the sentence serves the IS function in question. A context licenses non-canonical word order if it facilitates the processing of a non-canonical structure to a larger extent than that of a canonical one. We assume that in suitable contexts non-canonical structures facilitate comprehension, for example by allowing faster integration of currently encoded information into the discourse model constructed so far. Further issues to be examined will be to compare word order to lexical means to encode IS, as well as to investigate the contribution of phonological IS correlates in speaking and listening.

Download full description

Principal Investigators

• Prof. Dr. Reinhold Kliegl
• Prof. Dr. Gisbert Fanselow

Former Staff Members

• Dr. Thomas Weskott
• Dr. Robin Hörnig
• Sabrina Gerth

Activities