Macroparametric Space and the Learning Procedure

Thursday, Jan. 29, 2015, 12-12:50 p.m.
SN-4073

Literature critical of the concept grammatical parameter has identified significant problems--both conceptual and empirical--in the growth of parametric space necessary to accommodate small, local cross-linguistic variation, i.e. microparameters. A general solution is sketched in this talk; it is shown how a small set of initial macroparameters--potential `3rd factor' choices--may be fractured by a specific learning algorithm into a more complex parametric space on the basis of exposure to the primary linguistic data to which children are exposed. In this model, micro/nano-parametric complexity is derivative and limited, and parametric space is exponentially more constrained. Some of the initial problems and the general solution are illustrated with data involving multiple head-movement in English, Slavic, and Algonquian grammars, three language types in which this phenomena is employed to different extents. All are welcome.

 


Contact

Marketing & Communications

230 Elizabeth Ave, St. John's, NL, CANADA, A1B 3X9

Postal Address: P.O. Box 4200, St. John's, NL, CANADA, A1C 5S7

Tel: (709) 864-8000