Maria Polinsky

University of Maryland


Maria Polinsky is Professor of Linguistics, Associate Director of the Language Science Center, and Director of the Guatemala Field Station at the University of Maryland, College Park. Since 2007, she has served as Director of the Annual Heritage Language Research Institute. Her research focuses on cross-linguistic variation in syntactic phenomena and on differences between monolingual and bilingual populations, heritage speakers in particular. She has written extensively on heritage language grammars and have been active in promoting heritage language research and education worldwide.

From Rags to Riches to Resilience

In this talk, I examine recurrent properties of heritage languages with a special emphasis on what heritage speakers do well, as opposed to the usual scrutiny of what they do wrong. I present specific examples from phonetics, lexicon, and grammar, which show persistent advantage held by heritage speakers over L2 learners. Heritage speakers are extremely close to baseline speakers, or monolingual controls, in sound production and perception. In lexical production, heritage speakers stand out due to their remarkable ability to innovate; they create novel expressions without necessarily borrowing from their stronger language. Yet regardless of the ways in which heritage speakers innovate, their new structures obey established constraints of natural language design and the changes are logical. In grammar, heritage speakers show robust knowledge of tense marking and the use of determiners; in this talk, I propose a tentative explanation for these resilient features.

After reviewing areas of strength in heritage language, I discuss several well-documented areas where heritage languages fall short of reaching the native-speaker ideal. It has been noted that heritage languages have a low tolerance for optionality, a property also made manifest in a preference for one-to-one mapping between form and function. Furthermore, heritage speakers have problems with ambiguity and vagueness, which is particularly apparent in the resistance to and avoidance of material that is not perceptually salient – both in morphology and syntax. These two tendencies often lead to categorical restructuring in heritage language such that scalar categories are avoided and perceptually salient overt elements supplant light or silent elements.

The main thrust of this talk is twofold: heritage languages are languages and heritage speakers have a full-fledged grammar. As with any grammar, in order to explain complex or unusual facts, we often need to start from the most basic structures and a large body of paradigmatic facts – some of them straightforward, others very subtle. The grammar internalized by a heritage speaker may be different from the grammar of the baseline language, but the differences are not random. Rather, heritage grammars stay within the confines of what is expected of natural language. Areas of strength in heritage language reveal the bearing walls of natural language, while areas of weakness are indicative of more nuanced distinctions that may undergo change or loss.