Date: Mon, 22 Dec 2003 10:35:10 -0000 From: Patrick Studer Subject: Historical Sociolinguistics: Language Change in ... England
Nevalainen, Terttu and Helena Raumolin-Brunberg (2003) Historical Sociolinguistics: Language Change in Tudor and Stuart England, Pearson Longman, Longman Linguistics Library.
Patrick Studer, Mary Immaculate College, University of Limerick, Ireland.
The focus of this review is an elegantly produced book which, in its attempt to apply modern analytical research methods to past stages of language, will be of interest to historical linguists, sociolinguists and corpus linguists. The study which seeks to quantitatively describe and explain correlations between language users and phenomena of change in Early Modern English raises methodological issues that need to be addressed in future research. Nevalainen and Raumolin-Brunberg's analysis is based on CEEC, a machine-readable Corpus of Early English Correspondence. The study has been published under the general editorship of Geoffrey Horrocks and David Denison in the Longman Linguistics Library series. It falls into ten chapters which divide into further sub-sections, followed by notes at the end of each chapter. Numerical information and details of CEEC, along with author and subject indices, are attached at the end of the book for reference.
Chapter 1 sets out the agenda of historical sociolinguistics as a discipline by stressing the importance of language use and users in the study of language history, linguistic innovation and change. While this research agenda may not fundamentally differ from modern Sociolinguistics, the lack of language intuition and the patchy knowledge of contemporary usage of language requires a more interdisciplinary approach involving historical linguistics, sociolinguistics and social history. The research topics that result from the interdisciplinary approach include the observation of parallel changes, gender differentiation, social status in variation, register and others.
Chapter 2 continues by outlining the sociolinguistic paradigms in the study of speaker role in past stages of language. The authors essentially distinguish between quantitative research methods (social dialectology, sociology of language) and qualitative approaches (interactional, constructivist sociolinguistics). While the authors acknowledge the complementary nature of both strands, they particularly recognise the need for quantitative baseline data (p. 20).
Chapter 3 introduces the Corpus of Early English Correspondence (CEEC), which forms the baseline data for the study. The CEEC consists of over 6000 personal letters collected from more than 700 informants, spanning the period between 1410 and 1681 (pp. 43-49). The authors find the medium of letters particularly suited for sociolinguistic analysis, arguing that letters may be placed at the oral end of the written-oral continuum and could therefore be taken as indicators of changes from below (pp. 28-29). While basically acknowledging the ''bad data'' problem in historical sociolinguistic analysis (p. 26), the authors draw attention to some advantages historical data may have over contemporary data, such as the possibility for the researcher to carry out real-time analyses (as against to apparent-time) and the temporal distance between the researcher and their data, which may lead to greater objectivity in making linguistic choices (pp. 26-28).
Chapter 4 reports the results of a time/frequency analysis of fourteen well-documented dimensions of language change. The aim of the chapter is to highlight temporal aspects of diffusion and spread of new linguistic features and to capture large-scale developments, rather than minute details, of change (p. 53). The results are put in relation to the theoretical model of the S-curve, which claims that linguistic innovation tends to follow the pattern ''slow-fast-slow''. The side effect of this procedure is that it tests the validity of the theoretical model on the basis of empirical data. The results show that the S-shape is not replicated in some changes, notably in the change from YE to YOU, which is too rapid for the S-shape, and the shift from OF-phrases into noun phrases, which progresses too slowly to fit into model. The fourteen dimensions of change that are analysed in this and the following chapters are the following:
- 'YE' vs. 'YOU' - 'MY', 'THY' vs. 'MINE', 'THINE' - POSSESSIVE 'ITS' - PROP WORD 'ONE' - OBJECT OF GERUND - NOMINATIVE SUBJECT OF GERUND - 'S' vs 'TH' SUFFIX - PERIPHRASTIC 'DO' IN AFFIRMATIVE STATEMENTS - PERIPHRASTIC 'DO' IN NEGATIVE STATEMENTS - DECLINE OF MULTIPLE NEGATION - INVERSION AFTER INITIAL ADVERBS AND NEGATORS - RELATIVE PRONOUNS - PREPOSITIONAL PHRASE vs RELATIVE ADVERB - INDEFINITE PRONOUNS WITH SINGULAR HUMAN REFERENCE
Chapter 5 discusses how the fourteen changes diffuse across age levels (apparent time model). The apparent time concept starts from the assumption that some changes in linguistic behaviour are subject to the age of the speakers, while other changes spread from one generation to the next (e.g. sound changes, morphology) or occur simultaneously in a community (lexical, syntactic changes). The apparent-time model, however, has its natural limitations for the present study: In many cases, information about the age of the speaker cannot be traced, which distorts the overall picture of the analysis. The authors again point to the problem of insufficient background information (p. 88) but at the same time find that in some changes (e.g. YE -> YOU) the age of the speakers and the linguistic choices go hand in hand. They conclude that, although there can be no unfailing correspondence, the apparent time model is a valid analytical concept. In cases where the model fails, the authors suggest further micro-level studies (p. 99).
In chapter 6, the authors shift their perspective to gender differentiation in the CEEC. They look at differences in the spread of the fourteen linguistic variables in the corpus and evaluate the correlation between the results and the diffusion of the changes. Similar restrictions regarding corpus data as in the previous chapters apply here as well, with only 20% of all letters in CEEC being written by women. Moreover, the majority of the female informants come from upper ranks so that no complete picture can be obtained with regard to the distribution and spread of language changes across the social strata. In spite of the limited access to data from women, a comparison between men and women seems to suggest that women lead the changes, with the exception of cases in which social awareness plays a role in the use of a linguistic variable (e.g. multiple negation).
Chapter 7 deals with social stratification in the corpus, testing the correlations between social order and six of the fourteen linguistic changes analysed in the study. While the authors believe that correlations exist for most changes, they express their reservations as to whether they can be recovered on the basis of their data (p. 139). Women, for example, had to be excluded from the analysis because of insufficient data (p. 137). Despite these problems, the analysis seems to show that stratification can occur at any stage of diffusion. Moreover, a further pilot study of four changes reveals that directionality of change can be assessed empirically (p. 148-50).
Chapter 8 addresses processes of supralocalisation, i.e. the horizontal diffusion and spread of changes. The authors use three levels of delicacy for their analysis and generally distinguish between London, the Court, East Anglia and the North. Although the analysis shows that most changes seem to have been led by the capital city, the results simultaneously indicate that the horizontal axis is difficult to be kept apart from the vertical one (pp. 182-83).
In chapter 9, the authors try to find hierarchies among the sociolinguistic factors studied in previous chapters. In this context, the authors are interested in finding correlations between register (especially tenor, i.e. the social relations between correspondents) with other external variables in language variation and change (p. 188; 190). For this purpose, they carry out a multivariate analysis including the following parameters: real time, region, gender and register (p. 189). The results suggest that the changes under scrutiny are either gender- or region-driven (p. 198); the weakest variable proves to be register.
Nevalainen and Raumolin-Brunberg end their comprehensive study with some conclusive remarks in chapter 10. Putting the fourteen linguistic changes into the wider picture of standard present-day English, the authors find that most changes that are completed in CEEC later became standard variants in present-day English (p. 205). They further remark that, while each linguistic change analysed in the study is socially unique and basically unpredictable (p. 209), the results seem to indicate that the driving force behind language change is the striving for socially successful interaction (p. 210). Also they stress that most changes prove to be both communal and generational. In their final note, the authors, albeit cautiously, endorse the need to consider the impact of unexpected external influences, such as e.g. Civil War effects, on processes of change.
As readers of this book, we have no doubts that Nevalainen and Raumolin-Brunberg have been most meticulous in their effort to put sociolinguistics "to the test of time" (p. 202). The authors have used more primary data than any other study I know and have carried out their analyses thoroughly, carefully and comprehensively. My critical remarks, therefore, do not concern the analytical framework, the research tools or the results of their analyses: Nevalainen and Raumolin-Brunberg have done their homework in every single detail of their study - from the preparation of data to the presentation of results. What we can and perhaps should do, however, is to question the premises on which the analyses rest and ponder, on this occasion, the generalisations the authors draw from them. In other words, we might ask ourselves: Does a 2.3 million word corpus of written text material covering 270 years of history suffice to indicate how language changes emerge, diffuse and spread socially? As much as we might hope it would, we may assume that it does not. And although we must appreciate and praise the compilers of CEEC for the labour they have put into the corpus, we cannot help noticing the severe limitations a historical corpus naturally entails for the reconstruction of the linguistic past and, in particular, of language changes below social awareness. A single-genre text corpus, although an excellent starting-point, may just not be enough for this undertaking.
The problem of representativeness becomes even more intricate if we take into account that letter writing in the Early Modern Period was a highly formal process which required more concentration than, for example, the writing of present-day e-mails or text messages. Thus, the distance between the written message and the spoken word was naturally greater than today, just as the circle of people who mastered the art of writing was much more exclusive. Interpreting sociolinguistic developments in personal letters, written by the few for the few, as potentially indicative of trends in spoken discourse makes the present research project perhaps a little too ambitious. Future research, possibly involving multi-genre corpora and more data, might be needed to analyse in greater detail the behaviour of language users in the past. While my general reservations may raise some points for debate regarding the usefulness of CEEC as a substitute for spoken data, they do by no means question the value of the book as an excellent guide to the systematic study of sociolinguistic variables in a historical environment.
By way of conclusion, we may emphasise that Nevalainen and Raumolin- Brunberg have made a substantive contribution to the study of the linguistic past. The present book shows that historical sociolinguistics shares many of the aims of its sister discipline and that quantitative surveys on historical data do yield valuable results. At the same time, important differences between the two approaches seem to become apparent: Unlike present-day sociolinguistics, historical sociolinguistics is dictated to a large extent by the availability of primary data and contextual information about these data. Given these natural constraints, historical sociolinguistics seems to be forced to walk on a tightrope between traditional historical studies, philology and strictly systematised, computer-aided research methods. This makes it a highly flexible, dynamic and interdisciplinary area of research but at the same time susceptible to fall prey to the demands of either side. The present study clearly tends towards the latter end of the scale. Perhaps a more balanced approach would make the best use of the complementary nature of both sides.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Patrick Studer teaches in Mary Immaculate College, University of Limerick, Ireland, currently finishing his PhD thesis about the development of early English media language.