Methodological Issues in Studying Derivation
Rochelle Lieber, Pavol Štekauer, Rochelle Lieber · Oxford University Press eBooks · 2014
This chapter looks at the ways in which derivation may be studied, concentrating on what constitutes adequate data on which to base theoretical claims. The author assesses the pros and cons of various ways of generating data, including self-generated data, data from dictionaries and grammars, data obtained by fieldwork, data mined from large-scale corpora, and data generated by experimental methods. It is argued that there are perils in relying solely on intuition to generate data and determine possible words. The author suggests that data from large-scale corpora often call into question such intuitions and the theoretical proposals based on them.