The syntax-semantics interface concerns the correspondence between the syntactic structure of sentences and their logical form. Since the two broadly correspond, mismatches provide insights into the mental mechanisms structuring and interpreting sentences. While most work at the syntax-semantics interface has concerned clauses, this project investigates nominal constituents (DPs). In particular, it targets potential mismatches at the DP-border, where material seems to belong to DP syntactically, but not semantically. Relevant phenomena, where there already has been some work, are floated quantifiers, nonlocal readings of adjective, event-related quantification, and focus-affected quantification. Furthermore, we investigate recently discovered data from relative measurements such as "percent" and "third". The data vary across languages and even German dialects. To compare theoretical options requires us to study this variation closely in the broader context of variation in measurement structures and other DP-border phenomena.
The starting point of the present project is a novel link between morphosyntax and semantics. It is known that measures in German allow either genitive "100 Gramm dunkler Schokolade" or appositive case on the second noun: "100 Gramm dunkle Schokolade". But relative measures show different interpretations for the case patterns: the genitive "30 Prozent der Studierenden arbeiten hier" concerns the ratio of chocolate-eating students to all students, while the appositive "30 Prozent Studierendearbeiten hier" has a ’nonconservative’ construal: it concerns the ratio of student workers to all workers. Preliminary results show that these two construals of relative measures can be expressed by measurement structures in several other languages (e.g. Modern Greek, Korean, Tamil). In all these languages, the nonconservative construal requires focus on the second noun ("Studierende" in the example).
The main goal of the project is a theory of measurement structures and the syntax-semantics of DP. Nonconservative construals stand in conflict with the Conservativity Universal (Keenan and Stavi 1986), a well established generalization. One way of reconciling the two is to assume nonconservative construals involve a syntax-semantics mismatch at the DP border. To determine whether this analysis is correct requires us to understand the DP border generally. The project puts together a detailed understanding of three areas: the morphosyntactic properties of measurement structures, the semantic properties of measurement structures, and the properties of other phenomena at the DP-border. The main languages investigated are German, Greek and other European languages, Tamil, Korean, and Hebrew, while for other languages a questionnaire is developed. The project uses primarily informant based investigation, but also makes use of corpora.