※ 國立清華大學語言所102學年度第二學期專題演講八 ※

演講者： William Snyder (University of Connecticut)
演講題目：Whether to Satiate
時間：2014 年 05 月 21 日 (星期三) 中午 12 時 30 分
地點：清大人社院 B305 室
摘要：
Certain sentence-types sound unacceptable at first, but if judged on multiple occasions, over a
period of time, start to sound better. This phenomenon has been termed “syntactic satiation,”
and the paradigm case is wh-extraction from an English whether-island (Who does John
wonder whether Mary likes?). Native speakers of English who haven’t previously judged
such sentences find them jarringly bad, but native speakers who are professional linguists
often report that they now sound fully acceptable.

In (Snyder 2000) I suggested that evidence about “satiability” is important for research in
theoretical syntax, because initially unacceptable sentences that differ in satiability are
probably violations of different constraints. I further suggested that it might be possible to
induce satiation in the laboratory, under controlled conditions, and presented results of a first
attempt. Here I provide new, supporting evidence from an improved version of that original
study. We demonstrate that the effect is robustly present, and persists for at least one month,
even when a methodological concern raised by Sprouse (2009) has been eliminated.
Nonetheless, different laboratories have had variable success at inducing satiation,
and in this talk I will explore a number of possible reasons