The research presented in this study investigates the phenomenon of category mismatches in coordination, specifically focusing on the syntactic structures that allow for the conjoining of elements from different syntactic categories, such as noun phrases (NPs) and complementizer phrases (CPs). Conducted by a team of linguists, this work addresses a significant gap in the literature regarding the limitations and allowances of the Law of the Coordination of Likes (LCL), which traditionally posits that only elements of identical syntactic categories can be coordinated. Recent discussions have suggested that this law may be too restrictive, and this study aims to explore the nuances of this debate through empirical evidence from German.

The methodology employed in this research includes a series of acceptability judgment experiments designed to assess how speakers of German perceive mismatching DP-CP coordinations. The study utilized a Latin square design to control for various factors, including the selectional properties of the verbs used in the coordination. Participants were presented with sentences that either conformed to or violated the expected syntactic structure based on the selectional requirements of the verbs. This design is notable for its rigorous approach to isolating the effects of syntactic category mismatches, allowing for a comprehensive analysis of how these structures are processed by native speakers.

Key findings from the experiments reveal that speakers of German exhibit a strong preference for coordinations where the selected DP appears as the leftmost conjunct, while the unselected CP is positioned to the right. This preference contradicts predictions made by both linear adjacency and processing-based accounts, which suggest that the order of conjuncts should favor the rightmost element due to its proximity to the selecting verb. The results indicate that participants rated the acceptability of mismatching coordinations significantly higher when the selected element was positioned first, thereby supporting a bottom-up derivation model that posits an asymmetric structure in these coordinations.

The broader significance of this research lies in its implications for various fields such as syntax, language processing, and computational linguistics. By challenging existing theories on coordination and selection, this study contributes to a more nuanced understanding of syntactic structures and their acceptability across languages. Additionally, the findings may inform the development of more sophisticated natural language processing systems that account for syntactic variability and the complexities of human language use. This research thus not only enriches theoretical discussions but also has the potential to enhance practical applications in language technology and machine translation systems.

Source: glossa-journal.org