Developments in English
Expanding Electronic Evidence
Edited by Irma Taavitsainen, Merja Kytö, Claudia Claridge & Jeremy Smith
Studies in English Language
Cambridge: University Press, 2015
Hardcover. xxiii+299p. ISBN 978-1107038509. £65.00
Reviewed by Laure Gardelle
École Normale Supérieure de Lyon
This collective volume, which originates in the 2010 IAUPE conference, brings together contributions from specialists of corpus linguistics and language change. Its aim is to put in perspective the use of electronic corpora and quantitative analyses in historical linguistics research: while large corpora and technological tools have the advantage of providing a strong empirical foundation for studies, making them replicable, they pose new problems. First, they require more rigour in the use of statistical methods, and secondly, they should not lead to decontextualised analyses (pragmatics is essential to the understanding of language change or variation). The book focuses primarily on methodology and on the use of electronic resources to detect language change or achieve comprehensive descriptions. It is divided into four parts. The first one investigates methodological issues in corpus linguistics; the second one presents a number of case studies on language change, both in syntax and semantics; the third one focuses on the emerging field of corpus pragmatics and its use of marginal data such as interjections or hesitators; finally, part four brings together studies of world englishes and ‘the variousness of English’ .
It should first be borne in mind that corpus linguistics is not a field of study, but rather a methodology. One question is whether the data should guide us to form hypotheses (corpus-driven approach) or check hypotheses provided by a theoretical framework (corpus-based ‘guess-and-test’ approach). No linguist begins investigating data without a theoretical bias, but this issue raises a methodological question: should corpora be annotated? According to corpus-driven approaches, they should not, so as not to introduce bias. Yet annotations are useful: even though they might force a particular paradigm of grammatical analysis, they allow for comparison between corpora and for quicker searches, especially when the aim is to draw an extensive list of occurrences rather than find a few examples. They also provide systematically tagged and/or parsed text. In the final analysis, it is important that the linguist should agree with the grammatical system underlying the tagger or parser – for instance if one wants to analyse appositions. Stefan Th. Gries adds that a good quantitative analysis, especially one that analyses frequencies, needs to use refined enough tools, so that they can take contextual factors into account. For instance, raw frequencies should always be complemented by dispersion measures. This is also a major conclusion of Baas Arts, Sean Wallis & Jill Bowie, who find that language change cannot be measured only in terms of per-million-word frequencies, but that additional comparisons are needed, as well as studies of relative changes within a given set of constructions. They also conclude that in order to get a fuller picture of a phenomenon, one needs both large corpora and smaller parsed corpora.
The case studies, which imply similar methodological conclusions, provide insights into a number of changing patterns. Minoji Akimoto studies the development of desire, wish and hope, and shows that although they are all ‘wanting’ verbs, their evolution patterns differently depending on the construction in which they are found. Hope and wish are more and more common with to and that complements (whereas desire has the opposite trend), but in parenthetical uses, it is desire and wish that pattern in tandem (hope alone shows an expansion). The author shows that such developments are not random, but depend on the specific semantic and stylistic properties of each verb. Other linguistic changes involve the connective considering (that), which in Late Modern English becomes a marginal and stylistically marked form (Matti Rissanen), and interjections, which are regularly coined over time, especially due to a functional need to express solidarity (Manfred Markus).
The third part, as indicated above, focuses on instances of marginal data, from a pragmatic standpoint. Laurel Brinton studies delocutive verbs based on interjections (e.g. to hush), and shows that these verbs arise through conversion or back formation from their gerund (-ing) form. Andreas H. Jucker finds that uh and um, two language planners in the COHA, have a different history. While the latter was the only one to serve as a planner in 19th-century texts from the COHA, uh rose in the 20th century; today, um is increasing as well. From a methodological point of view, the author shows that such uses in written texts are more salient than in corpora of spoken English, because these planners are much more uncommon (they can be connoted with hesitation or even lies), and that in written texts, they have a function of characterisation. Finally, Thomas Kohnen studies religious discourse, which thanks to its prestige and prevalence throughout history is a good source to trace linguistic phenomena. He shows differences in evolution depending on the subgenres of religious discourse (e.g. prayers, prose, letter pamphlets and prefaces). For instance, religious biographies gave up the old forms thou and present in -th earlier than private prayers. He also finds differences in the evolution and frequency of address terms (e.g. brethren, thou) or stance markers (e.g. I am sure). He then compares these evolutions with those of non-religious discourse.
The book finishes with studies of varieties of English. Susan Fitzmaurice concludes from interviews of white Zimbabweans born around 1980 that there, linguistic realisations are related to representations of identity. Andrea Sand studies the way in which in Singapore, a highly multilingual society, non-standard Singlish has had an influence on blogs, and more generally on computer-mediated communication. Raymond Hickey studies phonological mergers (e.g. hoarse / horse, which / witch, wet / vet) in the English of some former colonies, for pairs which had differing realisations during the colonial period. The mergers have spread differently, and do not all have the same causes. In particular, some are conditioned by their phonotactic environment (e.g. pre-rhotic mergers such as pour / poor), while others are non-conditioned (e.g. pen / pin). Finally, William Kretzschmar focuses on the emergence of American English. He develops the notion of ‘random interactions in a complex system’ to explain why, for instance, the Scots construction needs + past participle (e.g. needs washed) survived in western Pennsylvania but not in Appalachia, whereas the Scots verbal present plural -s survived as a variant in Appalachia but not in western Pennsylvania. Owing to the random interactions in the ‘complex system’, some variants will survive in some places and not in others. The author concludes that this emerging notion of ‘complex system’, based on complexity science, promises to be more convincing than the notion of monolithic ‘language contact’ or that of ‘cultural shift’.
As suggested by these few notes, the present volume makes very pleasant reading and provides a wealth of information on a number of specific phenomena, some of which are not very often read about. The methodological angle is also particularly valuable and informs all the contributions, so that readers can also think about their own methodology. The volume should therefore be recommended both to scholars and to students with an interest in language.
Cercles © 2015
All rights are reserved and no reproduction from this site for whatever purpose is permitted without the permission of the copyright owner.
Please contact us before using any material on this website.