This paper offers a variationist critique of aspects of phonological theory and method, focusing on advances in descriptive methods and highlighting the problems that need to be addressed in explaining phonological variation. On the one hand, socially situated language samples which have been systematically collected and analysed constitute a legitimate – indeed often vital – source of evidence to be utilised by linguists for assessing and refining theoretical models. On the other hand, variationists cannot operate in isolation from theoretical concerns, and can benefit from an evaluation of the competing theoretical frameworks available to them.

The paper begins with a brief review of the philosophical foundations underlying the tension between ‘external’ and ‘internal’ methodology. We then focus on a particular phonological example – glottalisation in English. We demonstrate that phonological models of this can be complemented by systematic and accountable data collection and analysis of the kind associated with sociolinguistics. It is suggested that the patterns of variation produced by speakers are significantly more complex than has been indicated in the phonological literature. Consequently, these approaches can be usefully expanded and extended as theoretical models. We discuss some desiderata for extending the range of phonological models, focusing chiefly on the need to account for variability and change in language.