Since Charles Jones referred to the eighteenth and nineteenth centuries as the ‘Cinderellas of English historical linguistic study’ (Reference Jones1989: 279), there has been a great deal of progress in research on this period, but, as Beal (Reference Beal, Markus, Iyeiri, Heuberger and Chanson2012: 22) points out, much of this has been in the fields of syntax, morphology, lexis, pragmatics, sociolinguistics and the normative tradition. Beal argues that the availability of corpora of Late Modern English texts has greatly facilitated research in these areas, but, since creating phonological corpora for periods antedating the invention of sound recording is a challenging proposition, the historical phonology of Late Modern English has benefited much less from the corpus revolution. To redress this imbalance, the editors of this issue, with technical support from the Humanities Research Institute, University of Sheffield, created the Eighteenth-Century English Phonology Database (ECEP), which is freely available at www.dhi.ac.uk/projects/ecep/

After completing the database in 2015, we wished to encourage other scholars working in the field of English historical phonology to make use of ECEP. When the call was put out for proposals for a special themed issue of English Language and Linguistics, we saw this as an ideal opportunity to invite some of these scholars to contribute articles giving accounts of research in which they used ECEP as a tool for investigating research questions of their choice. Inevitably, some scholars were too busy with other projects to contribute,Footnote ² but Raymond Hickey, Warren Maguire and Nicolas Trapateau were able to accept. This issue also includes articles from the ECEP project team which showcase the potential of the database for both research and teaching.

The article by Nuria Yáñez-Bouza which opens this special issue provides a full account of the background to the creation of ECEP and the principles behind its design, together with illustrations and examples of how scholars might use this resource.

The article by Beal, Sen, Yáñez-Bouza and Wallis considers variation with regard to yod-dropping and yod-coalescence in eighteenth-century English. It provides analyses of data compiled from ECEP supplemented with word-frequency information for the period 1700–99 from the British component of ARCHER 3.2. It sets out to determine the internal (stress, phoneme, word-position) and external (prescriptive, geographical, social) motivations for the presence or absence of yod-dropping and yod-coalescence in the eighteenth century, which underlies the variation still found today in the pronunciation of words such as mature, duke and tune.

Two articles in this issue use data from ECEP alongside other resources to investigate the history of the bath–trapFootnote ³ split in English. Wells defines the lexical set bath as ‘comprising those words whose citation form contains the stressed vowel /æ/ in Gen[eral] Am[erican], but /ɑ:/ in RP’ (Reference Wells1982: 133). Typically, northern and midland varieties of English in England have the long vowel only in Wells’ bath_c set (e.g. calf, half, can't, banana), whilst southern hemisphere varieties tend to have the long vowel, but with variation in words such as dance from Wells’ bath_b set. Evidence from ECEP shows considerable variation in the bath set in the eighteenth century, along with metalinguistic commentary on the correctness or otherwise of particular pronunciations. Ray Hickey and Nicolas Trapateau both investigate this variation. Hickey looks at the history of low vowels in English from Old English onwards from a typological perspective, noting that these have tended to move upwards and from front to back, and that they have also been subject to lengthening. Drawing on evidence from ECEP along with that from other eighteenth-century sources, notably Nares (Reference Nares1784), Hickey concludes that the variation in the bath set from the eighteenth century to the present day is complex because it is motivated by a combination of phonotactic and sociolinguistic, internal and external factors. Trapateau concentrates on the distribution of variants between the eighteenth century and the present day, setting out to determine the role played by frequency in the lexical diffusion of bath lengthening. Trapateau analyses data from ECEP together with his own electronic version of Walker's (Reference Walker1791) pronouncing dictionary, the Longman Pronunciation Dictionary (Wells Reference Wells2008) for present-day RP and General American, and the Macquarie Dictionary (2015) for Australian English. He concludes that word frequency has little influence on the diffusion of this sound change in the eighteenth century, but that phonetic environment has a much stronger effect.

Warren Maguire sets out to answer the puzzle of the anomalous appearance of pronunciations represented by spellings such as owld in some dialects of Scots. The regular reflex of Old English [ɑld] in Scots is spelt auld, but the appearance of owld-type spellings in dialects of Lowland Scots, especially in the southwest of Scotland, as well as those of the extreme northwest presents a puzzle. Maguire examines evidence from the ECEP database to demonstrate that owld-type pronunciations were common in eighteenth-century Standard English. Given that data from the From Inglis to Scots (FITS) database (Alcorn et al. Reference Alcorn, Karaiskos, Kopaczyk, Los, Maguire and Molineauxforthcoming) shows no evidence for owld spellings in Old Scots, Maguire concludes that these pronunciations are not the result of internal developments, but of contact with varieties of English from England and Ireland. Because the Standard Scottish English pronunciation of such words developed to [oːld], the owld pronunciation became interpreted as non-standard and therefore Scots. Maguire notes the importance of databases such as ECEP and FITS for investigating apparent anomalies like this.

Christine Wallis presents an account of her use of ECEP as a teaching tool. The appearance of an article dedicated to pedagogical matters in English Language and Linguistics might surprise some readers but, as Wallis points out, there has been a pedagogical ‘turn’ in linguistics publishing recently. Wallis describes her use of ECEP alongside other resources such as archive collections, Eighteenth Century Collections Online (ECCO), the Eighteenth-Century English Grammars Database (ECEG), the Old Bailey Proceedings Online (Hitchcock et al. Reference Hitchcock, Shoemaker, Emsley, Howard and McLaughlin2012) and various letter collections, to carry out research-led teaching with advanced undergraduate and master's students. She notes that combining ECEP with other resources such as ECCO enabled students not only to carry out their own research projects, but also to gain an insight into how resources such as ECEP are created from original sources.

This issue of English Language and Linguistics presents cutting-edge research in English historical phonology. We hope that these articles will encourage readers working in this field to make use of ECEP and other databases in their own research and teaching.