Theoretical linguists have traditionally relied on linguistic intuitions such as grammaticality\njudgments for their data. But the massive growth of computer-readable texts and recordings, the availability of cheaper, more powerful computers and software,\nand the development of new probabilistic models for language have now made the spontaneous use of language in natural settings a rich and easily accessible alternative\nsource of data. Surprisingly, many linguists believe that such ‘usage data’ are irrelevant to the\ntheory of grammar. Four problems are repeatedly brought up in the critiques of usage data—\n1. correlated factors seeming to support reductive theories,\n2. pooled data invalidating grammatical inference,\n3. syntactic choices reducing to lexical biases, and\n4. cross-corpus differences undermining corpus studies.\nPresenting a case study of work on the English dative alternation, we show first,that linguistic intuitions of grammaticality are deeply flawed and seriously underestimate\nthe space of grammatical possibility, and second, that the four problems in the critique of usage data are empirical issues that can be resolved by using modern\nstatistical theory and modelling strategies widely used in other fields.\nThe new models allow linguistic theory to solve more difficult problems than it has in the past, and to build convergent projects with psychology, computer science,\nand allied fields of cognitive science.
Stored only when both papers are in the corpus — 697 citations exist upstream.