Lexical Strata and Phonotactic Perplexity Minimization

We present a model of gradient phonotactics that is shown to reduce overall phoneme uncertainty in language when the phonotactic grammar modularized an unsupervised fashion create more than one sub-grammar. Our recurrent neural network (Elman 1990), which, applied two separate, randomly initialized modules corpus Japanese words, learns lexical subdivisions closely correlate with main strata for (Yamato and Sino-Japanese) proposed by Ito Mester (1995). find learned model, which are based on entire prior context phoneme, reveal continuum membership, similar membership Hayes (2016) Native vs. Latinate stratification English.