Markedness and the Language User1

ABSTRACT. The traditional approach to markedness over the past half century has been to posit certain language universals, build a theory based on these universals, and then attempt to fit speaker/listener variability into the common theoretical mold. In this paper I propose an alternative approach that begins at the level of the individual language user. The basis of markedness is argued to be the cognitive process, EXPECTATION, which is quantified in terms of probability. This approach correctly predicts patterns in individual languages to diverge from so-called universal patterns since EXPECTATION is tied to the individual speaker/hearer and defined according to the context, linguistic and/or extralinguistic, in which a pattern occurs. Since seemingly similar language patterns can differ in many ways from one language to the next due to e.g., the context in which it occurs, the person using it, the frequency with which it appears, etc., we correctly predict cross-linguistic variability based on differing levels of EXPECTATION. Yet, we also predict similarities across languages given that language is spoken by humans with common physiological and cognitive abilities