Based on the title of this book, one might expect a discussion of Anglo-Saxon or Proto-Germanic, but ‘roots’ here refers to the dialectal sources of modern English as it is spoken around the world in the twenty-first century. In less than 500 years, a group of dialects spoken by a few hundred thousand people on a northern European island has become the most widely spoken language in human history. As they spread, these dialects served as input to the development of new regional and national dialects as well as to more restructured varieties and to pidgins and creoles. Those of us engaged in research on these new forms of English would like to know where their features came from: are they the legacies of the dialects that served as input, or did they arise from contact with other dialects and languages and/or independent innovation? Such questions underlie debates about the formation of new dialects and English-based creoles and about the prior creolization of varieties such as African American Vernacular English (Poplack Reference Poplack2000; Poplack & Tagliamonte Reference Poplack and Tagliamonte2001; Schneider Reference Schneider2007; Trudgill Reference Trudgill2004). This book represents an attempt to address these questions by contributing new evidence to these debates. As Tagliamonte points out in her Introduction (chapter 1, pp. 1–5), the original dialects spoken by the emigrants from Britain and Ireland who spread English around the world are no longer extant and the little direct evidence that we have of them is insufficient to answer our questions. However, the descendants of other speakers of those dialects who never left can still be found in the more isolated regions of the British Isles, and their English might provide further insight into the origins of many of the features of newer English dialects.
The use of English speakers from such ‘relic areas’ as evidence in the development of new dialects is justified in chapter 2, ‘Dialects as a window on the past’ (pp. 6–26). Any debate about the origins of features in North American English dialects is handicapped not only by the extensive dialect mixture that took place at the early stages of English-speaking settlement but also by the lack of direct historical and comparative evidence. Invoking the Uniformitarian Principle (‘the idea that “knowledge of processes that operated in the past can be inferred by observing ongoing processes in the present” (Christy Reference Christy1983: ix)’ (p. 7)), Tagliamonte argues that we can assume that contemporary dialects of English in communities that have remained isolated will tend to preserve features of earlier stages of the language. By comparing the quantitative patterning and linguistic conditioning of such features across these relic-area dialects, we can establish which of these features have been retained and lost and whether they share a common source (Poplack & Tagliamonte Reference Poplack and Tagliamonte2001; Tagliamonte Reference Tagliamonte, Chambers, Trudgill and Schilling-Estes2002). Using these criteria, she constituted the ‘Roots Archive’ (RA), a corpus of recorded sociolinguistic interviews with 110 relic-area speakers, explored further in chapter 3, ‘The Roots Archive’ (pp. 27–48). The research sites in England (Maryport, Cumbria), Scotland (Cumnock, East Ayrshire) and Northern Ireland (Cullyback and Portavogie) represent the primary sources for emigration to North America. The English spoken in these communities not only differs markedly from Standard English but is also consistent with studies of other northern British dialects. The RA is supplemented with the ‘British Dialects Archive’ (BA), sociolinguistic interviews with 182 speakers of various ages in locales in Scotland (Buckie) and England (Wheatly Hill, York, Wincanton, Tiverton and Henfield) also recorded by Tagliamonte and her graduate students. The speakers in the RA and the older speakers in the BA retain many features of English that no longer exist in current mainstream dialects.
Chapter 4, ‘Methods of analysis’ (pp. 49–63), sets out the methodological assumptions employed in the analyses of subsequent chapters. The historical comparative method locates systematic correspondences in related language varieties to reconstruct prior stages of linguistic history. Tagliamonte combines this method with the key principles of the variationist method (Labov Reference Labov1972): inherent variability, which recognizes variation as an integral part of language and linguistic analysis; accountability, which requires calculation of relative frequencies to take into consideration not only the form or feature of interest to us but also the other forms or features with which it co-varies; and diagnosticity, which allows us to adduce a common origin for different varieties when they share conditioning of the variation by language-internal factors. The common origin of features across different dialects may be attributed to the ‘normal’ transmission of the linguistic system within a speech community (from adults to children), to the diffusion of language change within or across communities (from adults to adults), or otherwise to ‘drift’ (Sapir Reference Sapir1921), the independent and parallel innovation in related language varieties (Labov Reference Labov2007). Another source of variability within and across dialects is grammaticalization (e.g., Hopper & Traugott Reference Poplack and Tagliamonte2003), the adaptation of lexical material for grammatical purposes. Successive processes of grammaticalization produce ‘layering’ in the linguistic system: the co-occurrence of alternate ways of saying the same thing that derive from different lexical sources. For example, two exponents of reference to the future in English represent the grammaticalization of different constructions at different points in history: will from a verb of desire (Old English willan ‘to want’) and going to from a verb of motion. Thus, differences in rates or linguistic conditioning across dialects may be used to infer differences in stages of grammaticalization. A competing theory of language change, the Constant Rate Effect (CRE; Kroch Reference Kroch1989), argues that language change occurs abruptly and its apparent gradualness reflects propagation through the speech community. According to the CRE, overall rates may differ across dialects but the conditioning by language-internal factors will remain the same.
Each of the following four chapters deals with a different sector of the grammatical systems of these relic-area varieties. Chapter 5, ‘Word endings’ (pp. 64–93), concerns morphological variation (verbal -s, adverbial -ly and -n’t). Chapter 6, ‘Joining sentences’ (pp. 94–120), examines variation in the forms used to link clauses (relative pronouns, complementizers and purposive for-to clauses). Chapter 7, ‘Time, necessity and possession’ (pp. 121–60), deals with grammatical distinctions of tense (futurity), deontic modality (obligation or necessity) and possession (have ~ have got ~ got). Chapter 8, ‘Expressions’ (pp. 161–84), analyses variation in discourse strategies, such as fixed expressions and general extenders (… and stuff). For some of these features, earlier reports of a north–south divide in British English dialects are confirmed. For example, auxiliary contraction and possessive have seem to be more frequent in northern dialects than they are in the south. Other features seem to reflect different degrees of proximity to mainstream usage or the ‘overlay’ of prescriptive influence, as in the case of the relatively recent introduction of wh-relatives (who, which, etc.) and the use of complementizers. The patterning for some features conflicts with historical studies, perhaps reflecting differences in genre rather than language change. For example, Tagliamonte finds little regional difference in the use of zero complementizer, which has historically been very sensitive to genre (cf. Torres Cacoullos & Walker Reference Torres Cacoullos and Walker2009: 5). This consideration raises the question of whether historical changes that have been inferred using written data really reflect changes in the language or simply changes in the nature of particular genres. Finally, some features appear to show progressive ‘colloquialization’. For example, the younger speakers in the BA show much higher rates of future going to and deontic (have) got to/gotta. In chapter 9, ‘Comparative sociolinguistics’ (pp. 185–94), Tagliamonte compares the overall distribution of features across the communities. There is a contrast between the Northern Irish (Ulster Scots) varieties with the other varieties and, within Britain, between the Scottish and English varieties.
The final chapter, ‘The legacy of British and Irish dialects’ (pp. 195–213), provides a summary of the analyses presented in earlier chapters. Despite the highly conservative nature of the RA dialects, many of the older features of interest in debates over new-dialect formation and restructuring are already disappearing or absent. For example, purposive for to is rarely used and, oddly, ain’t is virtually absent from these data! Nevertheless, Tagliamonte argues that the study of grammatical change can be profitably conducted using a cross-variety comparison.
One disappointing aspect of this book is its reliance on simple percentages to compare dialects, rather than the more sophisticated statistical methods we have come to expect (Tagliamonte Reference Tagliamonte2006; Tagliamonte & Baayen Reference Tagliamonte and Baayen2012). While percentages may give us some idea as to how communities differ in their linguistic behaviour (whether in terms of overall rates of use of particular features or their conditioning by language-internal factors), we cannot be certain which of these differences are meaningful without recourse to correlational statistics and tests of statistical significance.
Tagliamonte's final verdict on language change remains a bit unclear. Some of the results appear to support the CRE, in that linguistic conditioning remains constant across dialects regardless of the overall rates of use. However, in other cases we find the changes in conditioning across dialects that the theory of grammaticalization would predict (cf. Poplack & Malvar Reference Poplack and Malvar2007). Do such differences reflect different types of variable or language change? Tagliamonte suggests that changes imposed from outside (such as the introduction of wh-pronouns into the relative clause) may proceed differently in mainstream and peripheral communities (p. 105). She conjectures that ‘language change has probably always proceeded at different rates across communities as well as according to key components of the social and cultural situation’ (p. 210). If these suggestions are true, we would like to explore in more depth differences among variables and these ‘key components’.
Despite these limitations, the book will serve as a useful resource not only for researchers on variation and change in dialects of English but also for historical linguists working on the reconstruction of language history. Tagliamonte's coupling of the historical comparative method with variationist analysis is an exciting exemplification of what might be called ‘historical sociolinguistics’.
