Speaker: Timothy J. O’Donnell (MIT)
Title: Productivity and Reuse in Language
Date/Time: Thursday, Nov 17, 12:30-1:45p
Location: 32-D461
A much-celebrated aspect of language is the way in which it allows us to make “infinite use of finite means.” This property is made possible because language is fundamentally a computational system: Novel expressions can be composed out of a large inventory of stored, reusable parts.
For any given language, however, there are many more potential ways of forming novel expressions than are actually used in practice. For example, English contains suffixes which are highly generalizable (e.g., -ness; Lady-Gagaesqueness, pine-scentedness) and suffixes which can only be reused in specific words, and cannot be generalized (e.g., -th; truth, width, warmth).How are such differences in generalizability and reusability represented? How can the child acquire this system of knowledge? This set of related questions can be called the “problem of productivity.”
I will discuss a mathematical model of productivity and reuse which addresses the problem by treating it as a structure-by-structure inference in a Bayesian framework. I will compare this model to four other probabilistic models which formalize historical proposals from linguistics and psycholinguistics.
I will discuss the evaluation of these proposals on two very different sub-systems of English morphology: the English past tense, which is characterized by a sharp dichotomy in productivity between regular (i.e., -ed) and irregular (e.g., sing/sang) forms, and English derivational morphology, which is characterized by greater variability in productivity: from wholly productive (e.g., -ness), to productive in certain contexts (e.g., -ity; agreeability), to completely unproductive (e.g., -th). I will discuss several different aspects of the development and the adult state of these two systems, possibly including developmental overregularization, suffix ordering phenomena, and the generalizability of suffix combinations.