Edited by Tibor Laczkó and Catherine O. Ringen
[Approaches to Hungarian 12] 2011
► pp. 63–84
This paper discusses the experimental results and theoretical implications of our investigations of syntax–prosody mapping in Hungarian sentential embedding constructions. In experiments controlled for potential effects of factivity, contextual givenness and contrastive focus, we tested the predictions of competing theories seeking to identify the property yielding diverging syntactic and semantic effects in object clauses. Our findings suggest that factivity does not have a direct syntactic and hence prosodic correlate, contra e.g. Kiparsky & Kiparsky (1970), and these findings are compatible with an alternative analysis proposed by de Cuba & Ürögdi (2009) claiming that factivity is irrelevant for syntax, and clauses are differentiated by ‘referentiality’. We found givenness effects with all embedded clauses (factive and non-factive, ‘referential’ and ‘non-referential’), evidence that givenness is independent of these factors.