"Such and suchlike: a cross-linguistic reconnaissance"

This talk is a preliminary exploration of the semantic and formal properties of the English word suchand some of its counterparts inother languages. The proposal is that suchwords are 'demonstrative similatives' or, equivalently, 'similative demonstratives', i.e., their meanings lie at the intersection of the semantic dimensions of similarity and demonstration. We show that this kind of classification is straightforward for languages like Latin and Sanskrit, in which the counterparts to suchoccur in both similative and demonstrative paradigms. In these so-called 'correlative' paradigms the demonstrative similatives occur in systems with both non-demonstrative similatives and non-similative demonstratives. English suchand French teldo not occur in correlative systems, and the treatments of suchand telin their respective traditions show a great amount of confusion, in part as a result of the absence of these paradigms, and also because both elements differ strongly from ordinary demonstratives. We also show how the Sanskrit similative paradigm (and that of Kannada) extends the one of Latin in allowing similatives to encode similarity with the speaker and the hearer(s). The talk ends with the methodological point: if more than one language-specific category can do justice to the specifics of one language, the better category is the one that does double duty as a cross-linguistic category.