Overview
Data: Sprouse et al. (2016) on variation in the strength of island effects on acceptabiliy judgments. We will use the data collected for that paper, which can be found here, in this module.
Theory: Sprouse (2018) on the relationship between acceptability and grammaticality. We will specifically be concerned with his discussion in Section 3.3 of what apparent gradience in acceptability implies about discreteness v. continuity in grammatical representations.
In this first module of the course, we are going to focus on minimally extending standard statistical models used in analyzing acceptability judgments–generalized linear mixed effects models–in order to probe the nature of the grammatical representations that drive acceptability judgments. We will consider two possibilities discussed by Sprouse (2018): (a) that the grammatical representations underlying acceptability judgments are discrete (or categorical); and (b) the grammatical representations are continuous (or gradient).
The basic recipe, which we will repeat through the course, is (i) to define two or more (families of) models–in this case, one that assumes that the grammatical representation is categorical and another that assumes the representation is gradient; (ii) to fit both models to the data from some acceptability judgment data–in this case, to the data collected by Sprouse et al. (2016); and (iii) to compare how well the two models fit the data, weighed against some measure of how parsimonious (or conversely, complex) each model is.