This paper investigates the syntax of small clauses (SCs), with special focus on resultative constructions in Korean. I propose that resultative constructions in Korean are categorized into four sub-types, with two factors intertwined: (i) whether the SC is combined via complementation or adjunction and (ii) whether the SC-subject can be null or not. I provide evidence for this proposal, based on some symmetries and asymmetries between -lo resultatives and -key resultatives in Korean. I argue that -lo resultatives and -key resultatives behave differently because the former is merged as a complement while the latter is merged as an adjunct. I also show that the distribution of an SC-predicate is crucially affected by the presence or absence of a null subject, both in complement and adjunct resultatives. Empirical evidence for this claim is drawn from asymmetries in the distribution of SC-predicates, which include predicate fronting, predicate right-dislocation, and predicate omission. Overall, this paper provides further evidence for the claim that small clauses undergo cyclic Spell-out and linearization, and shows that movement within and out of an SC is restricted by general computational properties such as anti-locality and probe-goal Agree.
Abstract
1. Outlook
2. Two Types of Complement Small Clauses in Korean
3. Proposal
4. Two Types of Adjunct Small Clauses in Korean
5. Conclusion
References
(0)
(0)