Monday, October 19 at 2 PM in AP&M 4301.
Base identity as (phonologized) perseveration
Paradigm uniformity refers to the force militating against stem changes, or, in other words, against having multiple forms of the same stem. Especially abhorrent to paradigm uniformity are cases in which the allomorphs of the stem are phonologically dissimilar. Why do languages have paradigm uniformity? I propose that a major motivation for paradigm uniformity is perseveration within the production system. Specifically, when a speaker intends to produce a novel or relatively inaccessible form of a word, they activate other forms of the same word. Articulatory units comprising those forms also become activated, clamoring to be produced, i.e. copied into the output under construction. The copying is accomplished by the same mechanism as motor perseveration within the production system. I argue that copying is a normal part of the production process for a morphologically complex word. When there is no stem change, there is no cost to copying. However, if stem change is required, it is possible to copy too much. When too much is copied, the seeds for paradigm uniformity effects are sown. I present novel evidence for this production-internal motivation for paradigm uniformity from experimental data documenting a bias against labial palatalization (p-->tʃ) and from computational simulations of the production process in a connectionist framework.