AbstractLinguists of historical English, of traditional dialects and present-day varieties of English, generally rely on written texts, now often available in the form of corpora. However, the historical development of English, including its regional dialects, was naturally rooted in the spoken vernacular, rather than the literary standard. This paper, based onEDD Online(3.0), therefore, argues that the wealth of sources as used by Wright in his comprehensiveEnglish Dialect Dictionary(EDD) should no longer be disregarded, given that no better information is available. After a critical assessment of the widespread scepticism towards theEDDsources and of the different motivation of scholars not primarily concerned with traditional dialects (such asOEDlexicographers), the paper first provides a survey of the different types of sources used by theEDDand presented in different lists and tables inEDDOnline, and then focuses on the unpublished sources. The subsequent section shows that part of the problem of spoken sources results from the unjustified insistence of many scholars on phonetics to be the level of linguistic interest. In answer to theOED’sscepticism towards Wright’s sources as expressed in a paper by Durkin (2010a), the final section provides an analysis of Northamptonshire dialect words as a test case, with various linguistic issues beyond theOED’sfocus on the temporal frame of reference.