Input Filtering in Syntactic Acquisition: Answers From Language Change Modeling
Lisa Pearl, Amy S. Weinberg · Language Learning and Development · 2007
We use historical change to explore whether children filter their input for language learning. Although others (e.g., Rohde & Plaut, 1999 Rohde, D. and Plaut, D. 1999. Language acquisition in the absence of explicit negative evidence: How important is starting small?. Cognition, 72: 67–109. [Crossref], [PubMed], [Web of Science ®] , [Google Scholar]) have proposed filtering based on string length, we explore two types of filters that assume richer linguistic structure. One presupposes that linguistic utterances are structurally highly ambiguous and focuses learning on unambiguous data (Dresher, 1999 Dresher, E. 1999. Charting the learning path: Cues to parameter setting. Linguistic Inquiry, 30: 27–67. [Crossref], [Web of Science ®] , [Google Scholar]; Fodor, 1998b Fodor, J. D. 1998b. Unambiguous triggers. Linguistic Inquiry, 29: 1–36. [Crossref], [Web of Science ®] , [Google Scholar]; Lightfoot, 1999 Lightfoot, D. 1999. The development of language: Acquisition, change, and evolution, Oxford, , England: Blackwell. [Google Scholar]). The second claims that children learn only from matrix clauses (Lightfoot, 1991 Lightfoot, D. 1991. How to set parameters, Cambridge, MA: MIT Press. [Google Scholar]), defining simplicity in a structural manner. We assume that certain language changes occur via mismatches during acquisition. This allows us to use patterns of change to demonstrate that filtering restrictions are necessary to model language learning. Viewing language change as a result of mismatches during learning thus constrains the learning algorithm itself.