Sentence first, arguments after: Mechanisms of morphosyntax acquisition
Friday, February 2nd, 2022, 9:00 am, via Zoom.
In natural languages, there are complex patterns linking the structure of words (morphology) and the arrangement of words within sentences (syntax). For example, in German, verbs in second position have different morphology than verbs in final position, as in ‘dein Bruder möchte in den Zoo gehen’ (“Your brother wants to go to the zoo”). In language acquisition, knowledge of this type of morphosyntactic contingency develops relatively early, but we have very few mechanistic ideas about how learning takes place.
In this talk, I explore the statistical learning mechanisms that underlie learning of morphosyntactic contingencies. I contrast two general approaches that a learner might take. One approach might be to learn the position of prosodically prominent open-class words (“verbs go 2nd or last”) and then fill in the morphological details. Alternatively, one could work in the opposite direction, learning the position of closed-class morphemes (“-te goes 2nd and -en goes last”) and fitting open-class items into the resulting structure. This second approach is counter-intuitive, but I will argue that it is the one learners take. Evidence comes from a series of miniature language experiments with children and adults. In these experiments, subtle mathematical distinctions in learners’ input had significant effects on learning, illuminating the biased statistical computations underlying distributional analysis. Taken together, results suggest that learners organize knowledge of language patterns relative to a small set of closed-class items—just as patterns are represented in modern syntactic theory.