Swedish relative clause extractions: The Small Clause Hypothesis

Sammanfattning: On the basis of data from Swedish, this thesis investigates the Small Clause Hypothesis put forth by Kush et al. (2013). The hypothesis is suggested to account for the rare possibility of relative clause extraction, a phenomenon that poses a challenge for syntactic theories of locality. In brief, the hypothesis states that the possibility to extract from relative clauses is restricted to cases where the matrix contains a small clause-selecting verb. In that case the parser can reconstruct the complex noun phrase involving a relative clause as a small clause (from which extraction is not blocked). Language variation is claimed to be derivable from differences with regard to properties of the relative pronoun. A detailed investigation of the Small Clause Hypothesis and the analysis based on that, against data from Swedish, reveals that the predictions generated by the proposal are not borne out. First, a number of extraction examples retrieved from the literature constitute counterexamples to the claim that relative clause extraction is restricted to small clause-selecting matrix verbs. Second, Kush et al.’s (2013) assumptions about the role of the Swedish relative complementizer in the parsing process are implausible in light of data from other Scandinavian languages and extraction data from the relevant small clauses. Finally, the results of a controlled acceptability judgment experiment on Swedish relative clause extractions showed no statistically significant differences between matrix predicates. The conclusion of the thesis is that Swedish relative clause extractions do not provide any support for the Small Clause Hypothesis and therefore that another explanation for the phenomenon must be sought.