While McEnery and Wilson (1996: 16) argue for a combination of introspection and corpus data, many linguists still only draw on either of the two data sources. In this article I will show that treating the two types of data as corroborating evidence allows a far more detailed analysis of the categorical and variable constraints governing syntactic variation than the two types of data would have allowed individually. To illustrate this point, I will present the results of a case study on preposition placement in English relative clauses which combines a statistical corpus analysis and a magnitude estimation grammaticality judgement experiment.
Stored only when both papers are in the corpus — 70 citations exist upstream.
Imported from OpenAlex (CC0) — nothing here is generated.