This course is part of the L229A,B,C Laboratory Syntax sequence. We will focus on issues involved in making the study of WH-Constructions more empirical. More generally, we will talk about the elicitation of acceptability judgments and their relation to linguistic theory. On the empirical side, we ask what influences these judgements? On the theoretical side, we review accounts of what these judgments say about grammar. The course will be part seminar and part laboratory.
The goal is to provide you with the necessary background to conduct well-controlled studies of acceptability (and to analyze their results). Although acceptability judgments have been one of the primary linguistic research methodologies (probably still the dominant methodology), they have received alarmingly little methodological attention. It took over 30 years (Schütze, 1996) for the first systematic assessment of this methodology for linguistic research to be published. In addition to a variety of extra-linguistic and task-related factors (which we will discuss in class), acceptability judgments have been shown to correlate (among other things) with frequency of occurrence and with complexity of the material.
In this class we will focus on task-related effects, design-related effects, and processing/complexity-related effects. We will introduce magnitude estimation and (if there is interest) reading time studies as ways to investigate the relation between linguistic complexity (or other factor) and acceptability judgments.
As Schütze already concluded: if we want to use acceptability judgments as a research tool (whether elicited from so-called naive informants or from our friendly officemate), we should understand what we are measuring and how it is influenced by other factors.