Subject island effects are variable and unstable. Previous accounts, relying on well-defined structural properties of subjects, fail to capture this variability. We argue that subjects are islands only when they satisfy the Subject Criterion, which implies that at the interface they are interpreted outside the predicative nucleus of the clause, in a categorical structure (à la Ladusaw 1994); on the contrary, they are transparent for extraction when they undergo total reconstruction into the predicative nucleus of the clause, giving rise to a thetic structure. The thetic vs. categorical interpretation correlates with (non)presuppositionality and it is constrained by various factors, most notably the stage-level vs. individual-level nature of the predicate: this interaction of different factors can account for the observed variability, and is supported by our experimental evidence. The transparency of totally reconstructed subjects is not stipulated, but is shown to fall out from a top-down oriented syntactic computation, as proposed in Chesi (2004).
Subject islands and the Subject Criterion
CHESI C
2012-01-01
Abstract
Subject island effects are variable and unstable. Previous accounts, relying on well-defined structural properties of subjects, fail to capture this variability. We argue that subjects are islands only when they satisfy the Subject Criterion, which implies that at the interface they are interpreted outside the predicative nucleus of the clause, in a categorical structure (à la Ladusaw 1994); on the contrary, they are transparent for extraction when they undergo total reconstruction into the predicative nucleus of the clause, giving rise to a thetic structure. The thetic vs. categorical interpretation correlates with (non)presuppositionality and it is constrained by various factors, most notably the stage-level vs. individual-level nature of the predicate: this interaction of different factors can account for the observed variability, and is supported by our experimental evidence. The transparency of totally reconstructed subjects is not stipulated, but is shown to fall out from a top-down oriented syntactic computation, as proposed in Chesi (2004).I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.