How might evidence of language use - writing and speech - be used as a way of studying language? Corpus linguistics is the study of linguistic data from a particular language or set of languages. It is a fast-moving approach to studying language, and there is still a degree of divergence in how research questions are approached using corpus data. This book uses a framework, based on the work of Karl Popper, to explore a number of fundamental issues in corpus linguistics. It critically evaluates how these issues are tackled, and proposes a set of best practices for future research. It spells out why using corpus data is valuable, what we can learn from using it, and how we may most effectively progress our understanding of language by using such data. It is essential reading for researchers and students of language in general, and of applied linguistics and English language in particular.