Strong vs. Weak Islands
MetadataVis full innførsel
The focus of the chapter is weak islands (WI), i.e. selective islands. Strong (absolute) islands are considered only to set the stage for the discussion of WIs. Up until the late 1980s nothing much beyond wh-islands had been thought to be a WI. Beginning with Relativized Minimality, an ever-growing range of WIs has been recognized. Thus, theories of weak islands have mushroomed, each coming with a significant set of new data and important new connections to other domains. The chapter emphasizes the correspondence between data sets and theories. Less attention is given to proposals that primarily recast the theoretical account of some narrow range of data. Section 2 explains the distinction between absolute versus selective islands. Section 3 introduces a range of classical strong islands and various types of explanations, among them subjacency, repair by ellipsis, and processing. Sections 4, 5, and 6 enumerate the kind of extractions that are sensitive to WIs and the factors that induce a WI. Under the rubric of weak-island sensitivity, it discusses arguments vs. adjuncts, referential or existentially presupposing vs. non-referential expression, D-linking, individual vs. non-individual expressions and how many-phrases, functional readings and event-related readings, split constructions, negative polarity licensing, and cross-sentential anaphora. Under the rubric of weak-island inducers, it considers intervention effects due to wh-islands, negatives and other affective operators, response stance and non-stance in contrast to volunteered stance predicates, extraposition islands, VP-adverbs, and finally quantifier scope islands. Section 7 is devoted to various theories of islands, starting with ECP and subjacency, moving on to Relativized Minimality. Theoretical problems and the diversity of the data motivate the transition to the scope theory that comes in two versions, the algebraic semantic and the dynamic semantic ones. Section 8 concludes.