Rules and Exceptions in the Acquisition of Northern East Cree Morphology

Friday, April 30, 1-2:30 p.m.

Online

This talk explores how child and child-directed speech from Northern East Cree (NE Cree) may contribute to long-standing theoretical debates about inflectional morphology. Charles Yang’s (2016) recent Tolerance Principle (TP) proposal contends that “regular” inflectional forms are generated via rules while “irregular” exceptions are lexically stored and retrieved. This formal model centers on an equation that predicts how and when rules can be extrapolated from the input and when established rules must be revised or abandoned. Yang has primarily applied the TP to account for phenomena in Indo-European languages such as English, German, Polish, Russian, and Spanish.

A member of the Algonquian language family, NE Cree poses a particular challenge for language acquisition when it comes to possessive inflection: Some noun types require a suffix -im when possessed, but all other noun types disallow the suffix—and the distribution of -im hinges on several lexical, phonological, morphological, and semantic considerations. Children must figure out from language input which noun types require/disallow -im, and they must extend patterns to new nouns as their vocabulary grows.

Using a video corpus of naturalistic data from the Chisasibi Child Language Acquisition Study, this talk considers whether the TP model can predict rules accounting for -im from the input and whether these predictions bear out in the speech of two children, Ani (from age 2;01–4;03) and Daisy (3;08–5;10). Results indicate that the TP can provide a workable account for -im from the input, predicting the eventual extrapolation of three easy-to-motivate rules related to semantics and morphophonology. This account also predicts that children will initially posit a single rule for -im but that this rule must eventually be abandoned. Child speech data are still being analyzed, and results so far indicate qualified support for these predictions. This talk will also address potential implications for language revitalization as well as challenges related to applying the TP to smaller, less dense corpora than are available for languages such as English.

https://mun.webex.com/mun/j.php?MTID=m0f2d99f092f422aa11e3eb675de0bbb7 Meeting number: 132 918 8342 Password: 5qPy4b3FS5m

Presented by Department of Linguistics

Event Listing 2021-04-30 13:00:00 2021-04-30 14:30:00 America/St_Johns Rules and Exceptions in the Acquisition of Northern East Cree Morphology This talk explores how child and child-directed speech from Northern East Cree (NE Cree) may contribute to long-standing theoretical debates about inflectional morphology. Charles Yang’s (2016) recent Tolerance Principle (TP) proposal contends that “regular” inflectional forms are generated via rules while “irregular” exceptions are lexically stored and retrieved. This formal model centers on an equation that predicts how and when rules can be extrapolated from the input and when established rules must be revised or abandoned. Yang has primarily applied the TP to account for phenomena in Indo-European languages such as English, German, Polish, Russian, and Spanish. A member of the Algonquian language family, NE Cree poses a particular challenge for language acquisition when it comes to possessive inflection: Some noun types require a suffix -im when possessed, but all other noun types disallow the suffix—and the distribution of -im hinges on several lexical, phonological, morphological, and semantic considerations. Children must figure out from language input which noun types require/disallow -im, and they must extend patterns to new nouns as their vocabulary grows. Using a video corpus of naturalistic data from the Chisasibi Child Language Acquisition Study, this talk considers whether the TP model can predict rules accounting for -im from the input and whether these predictions bear out in the speech of two children, Ani (from age 2;01–4;03) and Daisy (3;08–5;10). Results indicate that the TP can provide a workable account for -im from the input, predicting the eventual extrapolation of three easy-to-motivate rules related to semantics and morphophonology. This account also predicts that children will initially posit a single rule for -im but that this rule must eventually be abandoned. Child speech data are still being analyzed, and results so far indicate qualified support for these predictions. This talk will also address potential implications for language revitalization as well as challenges related to applying the TP to smaller, less dense corpora than are available for languages such as English. https://mun.webex.com/mun/j.php?MTID=m0f2d99f092f422aa11e3eb675de0bbb7 Meeting number: 132 918 8342 Password: 5qPy4b3FS5m Online Department of Linguistics