Mind, Technology, and Society Talk Series

Joe Pater
Professor, University of Massachusetts, Amherst
Time/Date: 3-4:30 p.m. Monday, December 7, 2015

Chancellor’s Conference Room, KL 232

Grammatical Agent-Based Modeling of Linguistic Typology


(joint work with Jennifer Culbertson, Coral Hughto, and Robert Staubs)

What effect does learning have on the typological predictions of a theory of grammar? One way to answer this question is to examine the output of agent-based models (ABMs), in which learning can shape the distribution over languages that result from agent interaction. Prior research on ABMs and language has tended to assume relatively simple agent-internal representations of language, with the goal of showing how linguistic structure can emerge without being postulated a priori (e.g. Kirby and Hurford 2002, Wedel 2007). In this paper we show that when agents operate with more articulated grammatical representations, typological skews emerge in the output of the models that are not directly encoded in the grammatical system itself. This of course has deep consequences for grammatical theory construction, which often makes fairly direct inferences from typology to properties of the grammatical system. We argue that abstracting from learning may lead to missed opportunities in typological explanation, as well as to faulty inferences about the nature of grammar. By adding learning to typological explanation, grammatical ABMs allow for accounts of typological tendencies, such as the tendency toward uniform syntactic headedness (Greenberg 1963, Dryer 1992). In addition, incorporating learning can lead to predicted near-zeros in typology. We show this with the case of unrealistically large stress windows, which can be generated by a weighted constraint system, but which have near-zero frequency in the output of our ABM incorporating the same constraints. The too-large-window prediction is one of the few in the extant literature arguing for Optimality Theory’s ranked constraints over weighted ones.

Suggested reading:

Pater, Joe and Elliott Moreton. 2012. Structurally biased phonology: Complexity in learning and typology. The EFL Journal 3:2. 1-44.


Joe Pater is Professor in the Department of Linguistics at the University of Massachusetts Amherst, and co-director of the Initiative in Cognitive Science. He received his PhD in 1997 from McGill University, and specializes in phonological theory, phonological acquisition and learning theory. His articles have appeared in journals such as Cognitive Science, Language Acquisition, Language, and Phonology, and he is the co-editor of the volumes Constraints in Phonological Acquisition (CUP) and Harmonic Grammar and Harmonic Serialism (Equinox). His website is https://blogs.umass.edu/pater/.