1 Varieties in focus

In the first chapter ‘The emergence of new varieties of English in North America: Complex systems’ (pp. 21–36), William A. Kretzschmar describes how complexity science can explain the scalar notion of emerging dialects of a language. This scale translates into a hierarchy of dialects with different ‘scales’ of complexity. At the high-end of the scale, we can compare differing levels of complexity. At this level, we have fully developed languages, such as British and American English (p. 21). In contrast, earlier stages would include different types of language, existing, for instance, in colonies, or localities. To illustrate how these systems describe different types of linguistic variation, Kretzschmar (pp. 22–4) first of all analyzes lexical examples from the Linguistic Atlas of the Middle and South Atlantic States (LAMSAS). One prompt in this survey asked participants what expression they would use to convey the meaning of ‘the big hand is on the nine’. The responses resulted in a continuum with ‘quarter to’ (480 responses), ‘quarter till’ (392) and ‘quarter of’ (353) being most frequent, and expressions such as ‘15 until’ (21), ‘three quarters past’ (3) and ‘quarter before’ (1) being much less frequent.

The chapter ends with two sections comparing how complexity science better explains the development of new varieties than more traditional explanations. Kretzschmar first of all critiques Schneider's (Reference Schneider2007) Dynamic Model, which posits five stages of development that lead to the formation of a new variety. The stages initially range from the ‘foundation of the colony’, followed by intermediate stages, including such factors as ‘nativization’, ‘formation of an internal norm’ and ‘diversification’. A second section contains a comparison of a number of varying views of how North American systems of English developed, and a section on the development of three dialect regions in the Eastern United States and how complexity science explains their development. Traditionally, this region was divided into ‘three broad bands’ (p. 32): the Northern, the North Midland and the South, which can be further broken down into ‘fifteen subdivisions’ (p. 34). What Kretzschmar argues is that these are merely choices, and depending upon the particular analysis, one can focus on any one of the subdivisions.

In ‘Coordination in the courtroom: The uses of and in the records of the Salem witchcraft trials’ (pp. 37–64), Merja Kytö opens her discussion noting that while considerable research has been conducted on the use of and in conversational English, much less is known of the role that it played in earlier forms of English. She analyzes the use of the conjunction and in the Corpus of English Dialogues 1560–1760 (CED) and in work done by Smitterberg (Reference Smitterberg and Hundt2014: 321–8) on CED. These two corpora were then compared with the text categories (both spoken and written) in the Salem witchcraft trials (p. 39). As a prelude to her discussion, Kytö (pp. 40–1) notes that the only records surviving the trials were the pretrial records. While this may seem like a relatively small amount of data upon which to conduct her analysis, she notes that she was able to analyze a number of samples of speech-based writing released in Records of the Salem Witch-Hunt (Rosenthal Reference Rosenthal2009), a source that enabled the study of a number of speech-based registers, including depositions, examinations and indictments (p. 40). Nevertheless, despite certain limitations present in some of the texts (e.g. texts containing notes or shorthand rather than quotes limited their value for analysis) (p. 41), a variety of different texts remained suitable for analysis, including Examinations and Depositions as well as Indictments and Warrants (p. 42).

The remainder of the chapter contains a number of tables and bar charts providing statistical information on the structure of the various text types that were examined. One of the tables, table 3 ‘Clausal and phrasal and in the Salem data across the text categories’, contains statistical information on the occurrence of clausal and phrasal and in the various texts. For instance, while examinations contained many different examples of clausal and, warrants contained relatively few examples, largely because and is more likely to occur in spoken contexts such as examinations rather than warrants.

In ‘Dialect in early African American plays: A qualitative assessment’ (pp. 65–86), Alexander Kautzsch discusses and critiques the various sources that have been used to reconstruct early African American English (AAE). He then focuses on the use of AAE in an entirely new source, namely early plays (1858–1938), a text-type that enabled the examination of what he terms ‘imagined instances of AAE speech’ (p. 66). One problem with this kind of data, he notes, is that in previous literature on AAE many treatments of AAE using data of this type have been characterized as invalid and not accurately representative of the structure of AAE in earlier periods. For instance, he quotes Holton (Reference Holton1984: 57), who states that ‘any literary writing of dialect must be regarded as suggestion rather than as authentic representation of the speech of a particular group of speakers’ (p. 66). Kautzsch concludes from this attitude that many scholars feel that data of this type is only suitable for ‘qualitative’ analyses (p. 66).

To provide a more ‘reasonable assessment’ (p. 67) of the value that these plays can have in revealing linguistic structures of this period, he describes the findings made available in comparable corpora of the period: the World Atlas of Varieties in English (WAVE, Mouton), including WAVE features in a corpus of Earlier AAE transcribed speech (CEAAE-TS) and WAVE features in A Corpus of Early African American English Plays (CEAAE-P). CEAAE-TS provides ‘235 non-standard morpho-syntactic features in many varieties of English around the globe’ (p. 67), and CEAAE-P contains thirteen plays written by African American playwrights whose work occurred between 1858 and 1938 (pp. 69–70). Comparing these corpora, he was able to quantify the two forms in three contexts: (i) 78 features were found in both corpora, for instance pronouns, noun phrases, adverbs and prepositions (p. 78); (ii) a number of features were found only in CEAAE-TS, including certain kinds of pronouns lacking gender distinctions for singular and third-person pronouns. He found considerable differences between the two corpora. For instance, he was able to isolate 227 forms (table 5, p. 78) that occurred in both corpora (e.g. negation and pronouns). In the remainder of the chapter, he describes other features and their distributions in the corpora.

2 Towards General American English

Lieselotte Anderwald's chapter ‘Historical retention, progressive nation, or the eye of the beholder? The evolution of morphological Americanisms’ (pp. 89–122) is concerned with ‘the time frame of development’ of three verb forms regarded as ‘typical Americanisms’ (p. 89): gotten, which is expressed as got in British English; strong verbs associated with American English, such as dove rather than British English dived; and verbs that have become regular verbs in American English (burned) as opposed to irregular verbs in British English (burnt). Commenting on these usages, Anderwald lists a number of attitudes that have developed towards the forms in American English. For instance, she notes that these usages have led to claims that American English is in the ‘vanguard of change’ with British English using the more conservative forms as a type of ‘anti-Americanization’ (p. 90). To explore these attitudes as well as document the origin and development of the three verbs in American English, she draws upon corpus data, treating each verb form individually as case studies in three separate sections.

The analysis of the past participle gotten provides a representative example of the approach taken with all of the case studies described in the chapter. Anderwald notes that the usage of gotten, the original form, remains in American English, leading some to conclude that American English is much more conservative than British English, which uses got. Because of the lack of work on the historical development of gotten, she examines its usage in the Corpus of Historical American English (COHA). She discovers that its development from 1810 to 2000 does not match the traditional notion of its prominence in American English. In fact, its rise in usage does not begin until the 1930s when its overall frequency is slightly over five occurrences per million words. Its usage rises considerably in the twentieth century when, interestingly, it begins to compete with have got. The remainder of this section of the chapter tracks the development of similar types of usages, such as burn/burnt, smell/smelt and learn/learnt. Burnt, for instance, did not regularize to burned until well into the twentieth century. In contrast, dreamt (which is in the same class as burnt) regularized to dreamed much sooner (1810s).

The chapter concludes with a discussion of Americanisms emerging in the nineteenth century, a pivotal century as at this time English in the United States and Canada had evolved to the point that it was recognizably distinct from the English spoken in the ‘homeland’ (p. 101). In this section, evidence is provided from COHA to demonstrate that certain usages only occurred in American grammars. For instance, by 1870, grammars of American English specifically stated that usages such as dove were clearly preferred over dived.

Jukka Tyrkkö and Juhani Rudanko's chapter is titled ‘Grammar, text type, and diachrony as factors influencing complement choice in historical American English’ (pp. 123–46) and focuses on two types of complement structures illustrated in these sentences: (a) …he was reluctant to leave Mrs. Nash (1990, FIC) and (b) Frank was unused to eating in restaurants at all (1993, FIC) (p. 124). The first example contains the infinitive to before the verb leave, the second example a gerundial pattern following unused. These particular examples were selected from the historical corpus COHA and the goal of this study was to examine a claim made in Rohdenburg (Reference Rohdenburg, Dalton-Puffer, Kastovsky, Ritt and Schendl2006) that the gerundial forms are shifting to to-infinitive forms.

Various tables provide information on texts contained in COHA relevant to the current study (e.g. fiction, magazines, newspapers, non-fiction) and the decades that they cover (1900–50). The genre of non-fiction was selected for analysis because of the ‘diversity of the range of topics’ it covers, ranging from the social sciences to world history (p. 126). A series of tables document changes in the verb accustomed over time and in the four genres. The reason this verb was selected is that it is quite frequent and ‘it has selected both types of attentional complements in recent times’ (p. 127): (a) He was accustomed to weigh the danger of perilous alternatives (1921, FIC) and (b) He was not accustomed to discussing his affairs with household helpers (1921, FIC).

Other tables provide frequency information on the occurrence of the two forms in various contexts; for instance, figure 1 (p. 126) lists the frequency of the two forms by decade in the period 1900–50. One trend discovered was that between 1920 and 1940 both forms occurred with equal frequency. However, by 1950 ‘the gerundial patterns have become predominant, by a ratio of 4 to 1’. Figure 4 provides an interesting graphic portrayal (Mosaic plot) of the frequency of the two complement types (to+infinitive or to+ing in the registers in COHA), broadening the scope of registers discussed previously in the chapter to include fiction, magazines, movies and non-fiction.

The remainder of the chapter is more theoretical with a focus on various more technical analyses. In addition, statistical analyses are conducted to verify whether the generalizations that are made are based on statistically significant claims.

3 Ideology and beyond

The process of how American English became a recognized variety of English has been extensively discussed. The chapter by Ingrid Paulsen, ‘Enregisterment processes of American English in nineteenth-century U.S. newspapers’ (pp. 149–82), describes how one theory, enregisterment, explains the process that American English went through before being recognized as a distinct language.

One source of disagreement is the extent to which ‘social factors’ played a role in English emerging as a language. Some feel such factors were crucial. Others, such as Paulsen, disagree with this view. Instead, she proposes what she terms enregisterment processes – processes that make a clear distinction between structural and discursive levels of language. Her goal in the chapter is to ‘identify enregisterment processes of American English in a specific time period: the nineteenth century’ (p. 155), and to conduct this study, she focuses on five forms that were highly stigmatized in the nineteenth century (p. 159): /h/-dropping and /h/-insertion, the vowel in the lexical set bath, the distribution and realization of /r/, and the lexical variants pants and trousers.

These constructions were studied in databases containing Nineteenth Century U.S. Newspapers and America's Historical Newspapers. The results are described in various tables reporting the frequencies of the various structures listed above. One interesting comment on the final databases that she created is that while they are quite balanced by region, ‘most articles come from newspapers in Massachusetts and New York’ (p. 158). This results in some overlap, though ‘the overlap is rather small’. In terms of absolute frequencies (table 1), the search terms occurred in both databases, though with varying frequencies: deah, fellah and dawnce more frequently than hinglish and trousers. Figure 4 plots the frequency per million and decade. Viewed from this perspective, hinglish had a different trajectory from the other words: it was quite frequent in the 1820s, while the other words ‘show an enormous increase in the 1880s’ (p. 163).

The next chapter is ‘“Gems of elocution and humour”: Ideology, prescription and (self-)educational materials’ by Marina Dossena (pp. 183–202). Although there were many sources of information in early American English for prescriptive comments on the appropriate use of English, a neglected area of study is other more popular sources such as ‘textbooks, usage guides, “readers” and even newspapers and popular magazine articles’ (p. 184). To fill the gap that Dossena identifies, she examines the relationship between prescriptive norms as they appear in the texts listed above and the role that such texts play in what she terms ‘(self-)education’ (p. 185): individuals who wish to find information in usage guides rather than in, for instance, some kind of educational context, such as a class.

As a prelude to her discussion, she discusses the ideological role that both linguistic description and prescription play in how language is evaluated. Currently, evaluations of language have been reduced to ‘stereotypical concepts of what constitutes “bad language” and what is instead to be considered a “standard” model of usage’ (p. 186). Language prescriptions can even be found on Facebook, with advice on how to distinguish affect and effect. In earlier periods, American English was often described as ‘rustic’, ‘provincial’ or ‘vulgar’ (p. 186). Humor was also often used to characterize American usages ‘to mitigate’ the harshness of the terms used to describe English, particularly in textbooks (p. 186).

To provide more empirical information on the status of English usage during this period, Dossena analyzed the usage of Americanism(s) in COHA, finding increasing frequencies between 1810 and 1920, though she cautions against any strong conclusions that can be drawn from the data. Other sources she consulted for information on American English included textbooks and self-help books, specifically the title pages in these sources that provide specific references to American English. For instance, on the title pages in the Nietz collection of nineteenth-century textbooks, she found that ‘23 textbooks mentioned “the qualifier ‘American’” in the title pages while 17 mentioned the USA, the latter generally in contexts in which history and/government or geography are considered’ (p. 191).

4 Beyond the borders: Canadian English

The first chapter in this part is Stefan Dollinger's ‘Canadian English lexis and semantics: A historical-comparative resource in contrastive, real-time perspective, 1683–2016’ (pp. 205–30). Historically, as Dollinger notes, English language dictionaries have focused primarily on British and American English, to the exclusion of Canadian English, a bias that continued up to the 1950s. At that time, dictionaries of ‘non-dominant’ (p. 206) varieties began to appear, including not just Canadian English but Jamaican English. To explore one such dictionary, Dollinger focuses his discussion on a recently published version of the Dictionary of Canadianisms on Historical Principles (DCHP-2), which is an upgrade of DCHP-1.

The discussion of this dictionary is divided into a series of sections. For instance, section 2 ‘Introduction to the Dictionary of Canadianisms’ provides an overview of the dictionary. Section 3 ‘Theory and postcolonial lexicography: Canadianisms and other -isms’ focuses on the problems faced when creating ‘exclusive dictionaries’ (p. 208) rather than dictionaries of dominant varieties, such as American or British English. The remaining sections (4 and 5) provide extensive information about the corpus. For instance, the dictionary was carefully planned and based on five principles, including a quotation file that contained ‘diachronic data’ integrated with ‘synchronic studies of Canadian English’ (p. 210). The dictionary would also be designed to appeal to a wide audience, not just scholars but the general public. Evidence would be provided that would explain why each entry was a Canadianism in the form of a ‘word story’ that included, for example, additional information about the word, such as its ‘historical development’.

The rest of Dollinger's contribution contains an extensive description of various features of the dictionary with occasional references to differences between DCHP-1 and DCHP-2. For instance, entries for the dictionary are organized by ‘meaning’ with each meaning labeled as ‘one of six types of Canadianisms or as Non-Canadian, and in some cases more than one’ (p. 211). Included with each entry is a ‘Word Story’, which provides the reasoning for the classification given to the entry. Other features covered in the dictionary include how a particular lexical item can be classified as Canadian, and the semantic domains that can be used to classify particular words. Table 3 (p. 222) lists the top thirteen domains, with food and drink the most frequent domain (11.59 percent) and outdoors 2.77 percent.

In ‘Dialects as a mirror of historical trajectories: Canadian English across Ontario (North America)’ (pp. 231–58), Sali A. Tagliamonte discusses the role that synchronic variation in current dialects plays to help explain the ‘diachronic processes’ (p. 231) that led to this variation. To study these processes, she and her research team did fieldwork (sociolinguistic interviews) in Ontario in a range of ‘cities, towns, and villages’ (p. 231). A variety of different individuals from each locale were interviewed with the goal of representing ‘the main demographic and social characteristics of the place’ (p. 231). Such representation is important, she notes, ‘because linguistic innovations tend to originate in areas of cultural or political dominance. It is not until later that changes reach locations that are more ‘geographically remote or socially isolated’ (p. 232).

The research she discusses in this chapter was conducted as part of the Ontario Dialects Project. This project was conducted in thirteen communities, ranging from Toronto to Burnt River and represented different types of communities (p. 235): Mid-Large Urban, e.g. Toronto (population 2,481,494); Small Urban & Rural, e.g. Burnt River (population 250); Rural, e.g. Almonte (population 4,752). To illustrate the kind of research that can be done in the project, she discusses three case studies. One study involved the usage of come rather than the standard form for the past tense, came. After describing the history of past tense come, she includes a table presenting the usage of this form by community and age. Its prevalence is definitely stratified by age with older speakers using it much more frequently than younger speakers. The usage also varies by community with the form used quite frequently among older speakers residing in North Bay and only rarely (eight cases) in all the communities studied. The other two case studies concerned the use of ‘adjectives of smallness’ (e.g. little and teeny) stratified by date of birth, ranging from individuals born in the 1900s to those born in the 1990s. For instance, figure 3 (p. 244) plots variations of the use of small by individuals’ date of birth. Interestingly, little was the overwhelming choice by both groups. The only exceptions included were wee (chosen by older speakers) and tiny by younger speakers.

5 Evaluation

The co-editors chose a very interesting topic to explore, the rise of early North American English. They solicited nine chapters from ten scholars with expertise in this area. The essays were grouped into four different thematically based parts: Varieties in Focus, Towards General American English, Ideology and Beyond, Beyond the Borders: Canadian English. These groupings added real texture to the book, something that would have been lost if the chapters had simply been added randomly. The individual chapters were very well researched and clearly written, and tables (where applicable) contained useful statistical information, information that was effectively integrated into empirical findings in subsequent discussions.

While the authors provide extensive information on how American English became a full-fledged variety of English, one of the more interesting facets of the book is how the authors gathered empirical information upon which they based their analyses. Five of the chapters, as noted earlier, consulted one of the corpora listed below (note: the corpus descriptions below were provided on the websites listed for each corpus):

• – Corpus of English Dialogues 1560–1760. CED was compiled as a tool for the study of the language of the Early Modern period; the focus was placed on dialogues because interactive face-to-face communication is known to be an important factor in language change. The corpus was designed to offer easy access to a substantial quantity of data for variationist studies and research into historical pragmatics, as well as the study of speech presentation: it was compiled with particular variables in mind, such as text type, time, gender, and social rank. (https://varieng.helsinki.fi/CoRD/corpora/CED/)

• – Corpus of Historical American English. COHA is 475 million words in length and contains a variety of different texts covering the decades 1820–2010. It permits various kinds of searches, ranging from basic lexical items to more complex structures. For instance, it is possible to search for be passives (the book was stolen by a thief) as well as get passives (the thief got arrested by the police). (www.english-corpora.org/coha/)

• – Corpus of Regional African American Language. CORAAL is the first public corpus of AAL data. It features recorded speech from regional varieties of AAL and includes the audio recordings along with time-aligned orthographic transcription from over 220 sociolinguistic interviews from speakers born between 1888 and 2005. (https://oraal.uoregon.edu/coraal)

• – A Representative Corpus of Historical English Registers. ARCHER is a multi-genre corpus of British and American English covering the period 1600–1999, first constructed by Douglas Biber and Edward Finegan in the 1990s. It is managed as an ongoing project by a consortium of participants at fourteen universities in seven countries. (www.projects.alc.manchester.ac.uk/archer/)