Extraction from English RCs and Cross-Linguistic Similarities in the Environments That Facilitate Extraction

Abstract

In the first two decades following Ross’s Constraints on Variables in Syntax, a picture emerged in which the Mainland Scandinavian (MS) languages appeared to systematically evade some of the locality constraints proposed by Ross, including the relative clause (RC) part of the complex NP constraint. The MS extraction patterns remain a topic of debate, but there is no consensus as to why extraction from RCs should be so degraded in English (compared to MS)—or why it should be so acceptable in MS (compared to English). We present experiment results which indicate that English should be counted among the languages that allow extraction from RCs in at least some environments. Our results suggest a negligible island effect for RCs in predicate nominal environments and a substantially reduced island effect for those in canonical existential environments. In addition, we show that the size of the island effect resulting from extraction from an RC under a transitive verb is substantially reduced when the transitive verb is used to make an indirect existential claim. We present arguments that patterns of RC sub-extraction discovered in Mainland Scandinavian languages are mirrored in English, and we highlight methodological innovations that we believe may be useful for further investigation into this and other topics.

Keywords

• islands
• relative clauses
• island effects
• experimental syntax
• <i>wh</i>-movement
• canonical and noncanonical existentials