The publication of the volume Digitally-assisted Historical English Linguistics, edited by Carolina P. Amador-Moreno, Dagmar Haumann and Arne Peters, celebrates the work of the recently retired Kevin McCafferty by presenting various studies that are in line with McCafferty’s own interests, while at the same time highlighting the influence that digitalization has had on the field of historical English linguistics. The authors of the individual chapters in this volume work with different digital and data-intensive tools in order to discuss how these have changed our understanding of historical sociolinguistics, pragmatics or dialectology. As the main aim of the volume is to celebrate McCafferty’s work, it is not surprising that the authors in this volume have either shared similar research interests with McCafferty, with several of them also having directly worked with the Corpus of Irish English Correspondence (CORIECOR), which was developed by McCafferty & Amador-Moreno in Reference McCafferty, Amador-Moreno, Migge and Chiosáin2012. The fourteen chapters of this volume have been organized into four main sections that differ in both topic and methodology.

Chapters 1–4 are part of section I ‘New methods for new questions in historical linguistics’, which focuses on the recently developed digital tools used in studies connected to McCafferty’s work on the effect of ethnicity on language change (Reference McCafferty2001), the creation of CORIECOR and on the after-perfect in Irish English (IrE) (Reference McCafferty2004).

The first chapter in this section is by Carolina P. Amador-Moreno and Karen P. Corrigan, entitled ‘Determining the impact of education and socioeconomic status on linguistic choices in the Corpus of Irish English Correspondence’ (pp. 11–33), and discusses the importance of understanding the sociohistorical context in which the data under investigation was produced. This chapter is based on the CORVIZ project, which stands for ‘CORIECOR visualized. Irish English in writing across time (a longitudinal historical perspective)’ and explains Amador-Moreno and Corrigan’s attempt at creating a multidimensional system of classification based on the occupation of the letter writers in CORIECOR/CORVIZ. The authors explore whether the social status, as determined by occupation, had an impact on the correspondents’ use of embedded inversion (EI), which is a feature often said to be typical of Irish English in both contemporary and historical dialects. Amador-Moreno and Corrigan show that EI was not salient enough, as the feature was used by the more and the less well-educated occupations alike. CORVIZ as a new visualization tool, however, can be used for further research on other IrE features and may be able to show correlations between language use and occupation or other social factors.

In chapter 2, ‘“I hope that a correspondence may still be kept up between us”: Exploring conversational dynamics through the lens of (im)politeness studies in CORIECOR’ (pp. 34–53), David Sotoca-Fernández and Nancy E. Ávila-Ledesma investigate (im)politeness strategies based on a subsection of CORIECOR, which contains letters by Irish emigrants to the United States of America. In order to show the interrelation between family talk, (im)politeness and identity, Sotoca-Fernández and Ávila-Ledesma specifically look at how the verb hope is used in requests, apologies and reproaches among family members. Hope was found to be a mitigating tool to shift the focus from what is being said to feelings and abstract notions. Additionally, the verb is being used in order to shorten the conceptual distance between the writers and the addressees to create a lower degree of imposition in requests, apologies and reproaches. Further research is currently being conducted on other letter writers in different diasporic contexts.

The third chapter in section I, ‘Traditional data sets and the question of community bilingualism: The case of perfects and further vernacular features in Irish English’ by Patricia Ronan (pp. 54–72), contains a study on data from The Schools’ Collection. More specifically, Ronan investigates data from the English–Irish bilingual locations of Roundstone in County Galway and Tralee in County Kerry, as well Saggart and Raheny in County Dublin, in which Irish has not been a community language for centuries. First, Ronan looks at general vernacular features or Irishisms in the data and finds that such features were most frequently present in the Roundstone, followed by the Tralee data, whereas the Dublin data showed the lowest density of Irishisms, with only one lexical item having been borrowed from Irish into English. In a second step, Ronan compares the use of perfects in the different locations. Here, the Tralee data set is shown to use dialectal perfects more commonly, while the Saggart and Raheny data set shows a preference for Standard English perfects, and the Roundstone data set contains both dialectal and Standard English perfects to a more balanced degree. By comparing the use of general vernacular features and perfect use, Ronan was able to find a correlation between these two features.

In chapter 4, ‘When natural language processing meets corpus linguistics: A computational approach to analyzing the Corpus of Oz Early English’ (pp. 73–88), Martin Schweinberger explores the Corpus of Oz Early English (COOEE) (Fritz Reference Fritz2007), in terms of how its Australian-based letters written between 1788 and 1900 differ from similar contemporary private letters written in Great Britain, particularly from the control corpus A Corpus of Late 18c Prose (van Bergen & Denison Reference van Bergen, David, Beal, Corrigan and Moisl2007). Additionally, Schweinberger looks at the topics present in COOEE and whether they changed over time. The data within both COOEE and the control corpus was processed in R, which was followed by a keyword analysis and topic modeling. The findings from the keyword analysis reveal a high number of idiosyncratic keywords that were used significantly more frequently in the COOEE compared to the data from the control corpus. The topic modeling reveals the six topics of family, journey, landscape, exploration, indigenous and employment. Schweinberger was also able to find a shift over time in both the keywords and the topics.

Section II ‘Old data in the new digital age’ contains chapters 5–7, which focus on older data sets and hypotheses that are being revisited with the help of new digital corpora and tools.

Jerzy Nykiel investigates the complex preposition off of in chapter 5, entitled ‘“[H]is eye went neuer off of hir”: The development of the complex preposition off of from Middle English onward’ (pp. 99–109). The main aim of Nykiel’s study is to highlight the change that off of has undergone from its origin until its present-day use. Overall, five different corpora were analyzed for this study, namely the Corpus of Historical American English (COHA), the Corpus of Middle English Prose and Verse (CMEPV), Early English Books Online (EEBO), as well as the British English (BrE) and American English (AmE) subcorpora of A Representative Corpus of Historical English Registers (ARCHER). The findings reveal that the preposition is nowadays used mainly in AmE, although it is still present in BrE. Additionally, Nykiel finds that off of had already become a complex preposition by the end of the nineteenth century in AmE as the feature was found with verbs that do not take the particle off.

Turning to chapter 6, ‘Pronouns of address in the history of Irish English’ (pp. 110–24), Raymond Hickey examines the development of the singular thou and the different plural forms that one can encounter in IrE, such as ye, yeez or youse. Hickey shows in this chapter how thou had virtually disappeared from IrE texts by the early eighteenth century due to the rise of prescriptivism during that period. In order to fill the gap that was left behind by the disappearance of thou, so that there was no longer a formal pronominal distinction between the singular and plural forms of the second-person pronoun, the different varieties of English found varying local solutions. In the case of IrE, ye appeared at the beginning of the nineteenth century, whereas the vernacular IrE yous(e) only started to appear more frequently in the early twentieth century.

The final chapter 7 in section II, by Dagmar Haumann and Kristin Killie, is entitled ‘Seriously, where do illocutionary adverbs come from? A corpus-based assessment of the main hypotheses’ (pp. 125–50) and is concerned with the development of illocutionary adverbs in English. Haumann and Killie discuss the trajectories of the illocutionary adverbs frankly, honestly and seriously. The data for their study covers the period 1550–1899 and stems from three Chadwyck-Healey fiction corpora, namely Early English Prose Fiction (EEPF), Eighteenth-Century Fiction (ECF) and Nineteenth-Century Fiction (NCF). Haumann and Killie tested two hypotheses – the reanalysis hypothesis and the ellipsis hypothesis – and found that neither of them could be empirically corroborated. They therefore suggest a new indirect ellipsis scenario in which the illocutionary feature of verbal collocates is being absorbed by the manner adverbs in question.

Section III of the volume, ‘Investigating language contact through new technologies’, covers chapters 8–10 and is concerned with language contact and its effects on English morphosyntax and lexicon.

In chapter 8, ‘Complementizer deletion in that-clauses from Old to Late Modern English: A long-term diachronic corpus-based study’ (pp. 153–78), Kristian A. Rusten investigates the deletion of complementizer that, which had previously been studied by researchers like Rissanen (Reference Rissanen, Aijmer and Altenberg1991). However, Rusten points out that, by today’s standards, Rissanen’s empirical coverage is not particularly substantial as the data is drawn from the Helsinki corpus and therefore only contains a small number of that-clauses. This chapter is meant to complement Rissanen’s study by using larger data sets and new corpus linguistic methods, as well as statistical techniques that were not used in the initial study. By consulting six modern parsed corpora of historical English, Rusten shows how complementizer-deletion (C-deletion) in Old English was less frequent than initially found by Rissanen, whereas the opposite was the case in early Middle English, where C-deletion was much more common. In Early Modern English, however, C-deletion is once again less frequent than initially indicated by Rissanen’s study. Additionally, Rusten argues that language contact between Old English and Old Norse may have had an influence on C-deletion in English.

In chapter 9, ‘Lexical evidence for the contact between Irish and Old Norse in contemporary uses of Modern Irish, Norwegian, and Irish English’ (pp. 178–200), Arne Peters and Marion Schulte consult four corpora of contemporary language use in both Ireland and Norway in order to find traces of language contact between Old Norse and Old/Middle Irish. When it comes to Old Norse loanwords in Modern Irish, Peters and Schulte find that they most often referred to categorical innovations in Ireland, such as the introduction of ships, warfare, and new goods and materials. Despite the fact that previous studies of Irish loanwords in Old Norse and its contemporary descendants found several lexical items categorized as Irish loanwords, Peters and Schulte do not identify any in their newspaper corpora, which could mainly be because the specific domains in which the loanwords appear only play a marginal role in everyday life. Peters and Schulte argue that loanwords may not persist in the long run if they are not part of semantic domains that are not socially relevant to the contemporary population.

In chapter 10, entitled ‘Sorry mine tusen skrivefeil!: Using digital language resources to assess the phrasemic and syntactic integration of the borrowed apology marker sorry’ (pp. 201–22), Gisle Andersen investigates the English apology marker sorry as a borrowed informal discourse marker or adjective in Norwegian, synchronically and diachronically. For the diachronic part of the study, Andersen consults the searchable digital text archive of the National Library of Norway, with a particular focus on books and therefore also on spoken features in fictional dialogue. As for the synchronic part, the Norwegian Web Corpus 2017 (NoTenTen17) is the basis for this study, which includes many interactive genres of the internet, such as blogs or social media. Andersen finds that before the twentieth century, sorry only appeared in English materials, and the first instances in Norwegian literature (other than in dictionaries or English books) come from a novel dated 1904 in the form of codeswitching. In the 1940s, sorry starts being used as an adjective and becomes more frequent in the 1970s. Regarding the synchronic part of the study, sorry is found in many different contexts, which either took on an English model, where all parts of a phrase were translated into Norwegian apart from sorry, or were used analogous to a Norwegian synonym for sorry. Overall, the study is a good example of the complexity of usage patterns of borrowed lexemes.

Section IV, ‘Investigating dialect in the new digital age’, is concerned with corpus-based approaches to historical dialect features, with chapters 11–14 showing how different digital methods can be used to better understand historical social networks and how traditional dialect features were present in both written and spoken language.

Chapter 11 by Dania Jovanna Bonness is entitled ‘“[…] and the Brogue their was good fun that night in Uncle James’”: A case study on a late 19th-century Ulster family network’ (pp. 225–44) and is based on the letters of Eliza Catherine Smyth from the late nineteenth/early twentieth centuries, which are documented in CORIECOR. The aim of Bonness’ study is to identify Smyth’s family network and network ties, and to analyze her linguistic features. After identifying the network members, their kinship ties and their location, Bonness presents some of Smyth’s prominent linguistic features in grammar, phonology and lexicon, which are found to be nonstandard and characteristically IrE features. The study paves the way for future studies once the Network Strength Scale initially developed by Lesley Milroy (Reference Milroy1980) has been adapted to Smyth’s family network.

In chapter 12, ‘“[T]he largest mountan in nort America”: Evidence of “Southern” Irish English consonants in Ulster before 1900 in the Corpus of Irish English Correspondence’ (pp. 245–66), Persijn M. de Rijke examines data from an Ulster-based subcorpus of CORIECOR to investigate the diachronic, geographical and social distribution of dialectal consonantal features between 1700 and 1940. Rijke discovers that correspondence written by supposedly Northern IrE speakers contains several instances of phonetic representation of consonants that are usually attributed to Southern IrE, however, not as often compared to tokens for vowels or phonological processes. Additionally, Rijke points out that the comparison of social information of the letter writers is not easily possible due to the absence of metadata. Furthermore, it can be argued that the occurrence of Southern IrE consonantal features may be explained by language contact and feature transfer between Irish and English.

Warren Maguire remarks in chapter 13, ‘A corpus of traditional South-west Tyrone English’ (pp. 266–84), that, while the linguistic landscapes of England and Lowland Scotland have been well documented through linguistic surveys in the first half of the twentieth century, this has not been the case for Ulster. Maguire looks at the rural dialect South-west Tyrone English (SwTE) with a particular focus on the use of /e/ in words belonging to the meat lexical set and explores recordings of speakers that were born in the early and mid twentieth century. It is found that an /e/-like pronunciation occurred in 43.81 percent of tokens across all speakers. Male speakers were shown to use /e/ in meat five times more often than female speakers, while male Catholic speakers used the feature significantly more often than their Protestant counterparts. This difference according to religion cannot be found in the female speakers. Additionally, the data suggests a chronological decline of the feature according to the speaker’s age.

Chapter 14 is by Kristian A. Rusten, ‘From Grimm to ngrams: English historical linguistics in the digital age’ (pp. 285–300). In this final chapter, Rusten recounts the development of historical corpus linguistics in general, from Grimm’s Deutsche Grammatik to modern digital corpora. When it comes to historical English linguistics in particular, Rusten highlights the key developments within the field over the past 35 years or so, which roughly coincides with the duration of McCafferty’s career. Rusten summarizes some of the other studies within this volume and emphasizes the importance of English historical corpora such as CORIECOR.

Overall, Digitally-assisted Historical English Linguistics is a well-rounded volume that spans over multiple time periods, varieties, methodologies, data sets and more, and corroborates the importance of the work that has been done by previous researchers such as Kevin McCafferty. As further advances are made when it comes to digitization and new digital tools are introduced regularly, this volume opens the path for future research in the field of historical English linguistics.