Date: Wed, 18 May 2005 12:48:08 -0500 (CDT) From: Richard Cameron Subject: Talk That Counts: Age, Gender, and Social Class Differences in Discourse
Author: Macaulay, Ronald K. Title: Talk That Counts Subtitle: Age, Gender, and Social Class Differences in Discourse Publisher: Oxford University Press Year: 2005
Richard Cameron, Joint Appointment in the Department of English and in the Department of Spanish, French, Italian, and Portuguese, University of Illinois at Chicago.
DESCRIPTION OF THE BOOK
In this new work by the distinguished sociolinguist Ronald Macaulay, he formulates and presents a new approach to variation at the level of discourse across a range of speakers of English from Glasgow and Ayr, Scotland. Written primary for a professional audience, Macaulay explores a remarkably wide range of discourse features in the speech of 33 adults and adolescents, balanced by two categories of age, social class, and gender.
Although the book consists of 15 chapters, the chapters may be divided thematically into three broad types. Chapters 1 through 5 provide introduction to the discourse features of study, methodology, the data base, background assumptions and theory. Chapters 6 through 12 present various in-depth studies of the social and discourse patterning of numerous discourse features. Chapters 13 through 15 provide summary, discussion, and generalization. I will selectively review the chapters here in greater detail, though no review can match the level of detail of this book.
Chapter 1 on Discourse Variation provides the reader with a brief overview of variationist or quantitative treatments of Social Class, Gender, and Age. Other extralinguistic factors such as personality, ethnicity, network, or ambition are also mentioned though not discussed. Here, Macaulay seeks both to align his current project with the variationist agenda and to distinguish his project from previous work both in discourse analysis and quantitative sociolinguistics. Perhaps most fundamental to this chapter are the following three points:
First, in contrast to research which takes social class as an element in the analysis of language change in progress, his focus is on the stable correlates of class and language use within Scottish English as spoken in Ayr and Glasgow.
Second, despite the pervasive rejection of Basil Bernstein's early pronouncements on social class and language, Macaulay finds Bernstein useful as a basis for formulating testable predictions, most of which he then goes on to falsify throughout the book. Indeed, Macaulay draws on and reconsiders Bernstein in ways reminiscent of early research into language and gender which drew on Robin Lakoff's (1975) important work.
Third, and to my estimation most importantly for his agenda, Macaulay addresses the vexing issue of how to quantify discourse features. In quantitative sociolinguistics, a sociolinguistic variable may be understood, broadly, as two or more ways of saying or accomplishing the same thing. Thus, the analyst quantifies the differing variants of a variable over the total number of occurrences of the variable . For instance, if we were to study Puerto Rican Spanish word final (s) in a stretch of recorded speech, we would first identify the variants. In this case, there are three:[s], [h], and a deleted variant . In turn, we quantify the total number of times that each variant of the variable does occur. Then, we derive the frequency of occurrence of each variant relative to the overall number of possible occurrences of the variable (s). Using simple fractions as a model, the quantities of each individual variant would be a numerator over the total number which would serve as the denominator. But, in order to study discourse features such as the use of discourse markers or adverbs or articles, there is no denominator because the discourse features which Macaulay has chosen to investigate cannot be conceived of as variants of a variable. His solution is to measure the number of times a given discourse feature appears within a stretch of 1000 words. With 1000 as the denominator, Macaulay is then able to derive relative frequencies of occurrence and test them for significance using the Mann-Whitney nonparametric test in the speech of individuals classed according to three social categories of identity: Adolescent vs. Adult, Middle vs. Working Class, and Female vs. Male.
Chapter 2 addresses important specifics of how Macaulay gathered spoken samples of speech and the assumptions which accompany his methods. It should be stated that although Macaulay did the interviewing of speakers from Ayr, all of Macaulay's work on Glasgow in this book relies on the samples gathered through the innovative work of Jan Stuart-Smith (1999, 2003). Stuart-Smith arranged for 1/2 hour open-ended tape-recorded conversations among friends in the absence of the researcher. In total there were 33 adults and adolescents, balanced by two categories of age, social class, and gender. In Chapter 2 Macaulay also gives a few preliminary illustrations of the social distributions of Minimal Responses to give the reader a sense of the utility of his approach and results. Chapter 3 provides further details of the speakers from Ayr and Glasgow, the overall word totals per individual and group, the presence of narrative, and brief discussions of transcription and statistics.
Chapters 4 and 5, which treat Social Class and Bernstein's sociology of class and language use, clearly show the imprint of a sociolinguist of experience, hard-won practicality, and honesty. Though brief, Macaulay's discussion of Class revisits the tensions between sociological and sociolinguistic approaches to class as found in Labov and Trudgill. For the uninitiated, his review of Bernstein is even-handed, thorough, and critical. In particular, he criticizes the wide-spread acceptance of Bernstein's views in educational circles despite the vagueness of Bernstein's claims and the slender-to-nonexistent empirical support for these clams. One should note, however, that Macaulay does not provide similar discussions of Age and Gender, a point we will return to in the Critical Evaluation.
Chapters 6 through 12 turn away from method and theory to an extensive catalogue of findings. Chapter 6 reviews the social and discourse patterning in the Glasgow data of Minimal Responses, Question Asking, Imperatives, and the English Discourse Markers of Oh and Well. Chapter 7 adds more on Oh, Well, You know, I mean, and Like. Chapters 8 through 12 follow similar formats of presentation for such features as Coordinate and Because Clauses, Passives, Clauses of Movement such as Left and Right Dislocation, Modals, Adverbs and Adjectives, Articles, Personal and Impersonal Pronouns, Relative Pronouns, and finally, Direct Quotations and Quotatives within Narrative. Each of these chapters is packed with details of textual occurrence and statistical representation of social distributions. In total, Macaulay subjects 42 discourse features to the Mann-Whitney nonparametric test of significance. This is a remarkable range.
The remaining three chapters, 13 through 15, summarize and explore the implications of the findings reported in the preceding chapters. In Chapter 13, he notes that out of the total number of tests of significance, 46 sets of significant differences emerge. Overall, the smallest number of significant differences were associated with Social Class. Age differences, between Adolescents and Adults, provided the largest set of differences. Gender differences provided the second largest set. Thus, Age more than Gender and Gender more than Social Class correlate significantly with frequencies of discourse feature use. Chapter 15, entitled Discourse Sociolinguistics, serves to summarize and revisit the methods and main assumptions that Macaulay initially laid out in the first 5 chapters. Chapter 14, however, sets out to do more than summarize.
Entitled Discourse Styles, Macaulay begins Chapter 14 by distinguishing discourse style from style as conceived and debated within variationist research. Though he provides no clear definition, it appears that Discourse Style may be conceived of as ways of talking that are constituted by sets of discourse features which significantly distinguish one social group from another or, more specifically, which significantly distinguish the peer-based interactions within one social group from the peer-based interactions in other relationally defined social groups. He begins with Adolescents whose initial set of distinguishing features would include the use 'like' as well as taboo words. However, such items as these show little connection to the quality or nature of routine interaction between adolescents. For example, Macaulay found less use of certain discourse markers in adolescents yet significantly more use of questions and imperatives. The higher frequencies of questions and imperatives seem to result from a more demanding manner of interaction among adolescents when compared to adults. This notion of demanding style also shows up in his analysis of Working-Class boys who tease and playfully insult one another in a consistent fashion. Similar intersecting or interacting factors emerge in his discussion of Male and Female discourse styles. For example, because females overall told more stories with embedded dialogues than males, and because these stories tended to be about other women, female speakers relative to males showed a higher use of personal pronouns and coordinate clauses. Likewise, when comparing Middle Class with Working Class speakers, even as both groups clearly share the same grammar, they do show differing frequencies of adverbs and evaluative adjectives. Specifically, Middle Class speakers, relative to Working Class speakers, when telling narratives tend to add more evaluative adjectives. Because the overt expression of evaluation may entail a particular ideology as does the relative lack of overt evaluation, Macaulay cautiously infers a class based ideology that we may summarize in broad strokes as this: If Middle Class, express your attitudes and evaluations of events. If Working class, let your descriptions of and quotations from the events do the narrative work. All of this, then, leads to certain generalizations which I will paraphrase here, again, in broad strokes. A preference for certain topics or for certain modes of expression has implications such that those who prefer these topics or modes will end up using more of some discourse features and less of others depending on how these features are needed for the topics or modes. Because preferred topics or modes of expression may be functions of ways of living, one begins to see how the statistical structuring of discourse differences patterns across such categories of Gender, Age, or Class. After all, these categories are not essentialist qualities of individuals. They are cover terms for different circumstances and ways of living.
In reading this book, I was continually impressed by the attention to the details of linguistic analysis, the care with which results were presented, and the cautiousness with which generalizations were made. Moreover, the range of research that he draws upon when discussing results is top-notch. I also suspect that it will be or could be a foundational text for a new strand of sociolinguistic research that focuses on sociolinguistic variation with aspects of language structure in use that cannot be captured with the tool of the sociolinguistic variable. In particular, given his findings, future researchers in English will want to look more closely at such features as adverbs and evaluative adjectives. Likewise, his work on quotatives contributes to active and on-going research in this area. At the same time, his work indicates certain features that may not be worth further investigation by those interested in social patterning of language. I think here of his chapter 9 on Modals and Modality.
Yet, with all research, one may wish that certain aspects of theory or findings could have been pursued in different ways. For example, as I noted in the summary of the book's contents, in chapters 4 and 5, Macaulay provides us with concise commentary on issues of measuring class and the ideas of Bernstein. However, judging from the title of the book, I would have expected an equal amount of background reporting on gender and on the differences between adolescent and adult speech. Given that Age and Gender contribute a larger number of significant differences in the use of discourse features than Class, the lack of theoretical framing and prediction for Age and Gender leaves me wanting more. I suspect that his discussion of Discourse Style actually could have benefited from a close reading of such foundational work as Susan Harding's 1975 study on "Women and words in a Spanish village." Harding makes points that are quite similar to some of his ideas about the relationship between discourse features and ways of life. Also, Macaulay reports various instances where Age, and Gender and Class interacted. Such findings clearly support an concept prevalent in current gender research that Gender is not easily separated from other categories of social experience, like Age or Class. Macaulay does note these interactions. For example, on page 167 he observes that "the Glasgow adolescent conversations underline the need to consider social class as well as gender." Yet, I would have liked to have seen these findings somehow framed relative to theory and prediction based on research in Gender or Age in addition to the discussions of Class.
Overall, this is a book to which I will return again and again. It provides us with a wealth of data, a model for future research, and a basis for future theory building.
Harding, Susan. 1975. Women and words in a Spanish village. In R. Reiter (ed.) Toward an Anthropology of Women. New York: Monthly Review Press. Pp. 283-308.
Lakoff, Robin. 1975. Language and woman's place. New York: Harper and Row.
Stuart-Smith, Jane. 1999. Glasgow. In P. Foulkes & G. Docherty (eds.). Urban voices: Variation and change in British accents. London: Arnold. Pp. 203-222.
Stuart-Smith, Jane. 2003. The phonology of modern urban Scots. In J. Corbett, D. McClure & J. Stuart-Smith (eds.) The Edinburgh companion to Scots. London-Arnold. Pp. 101-37.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Richard Cameron teaches Sociolinguistics and Discourse Analysis in the TESOL Program of the Department of English and in the Hispanic Linguistics Program in the Department of Spanish, French, Italian, and Portuguese at the University of Illinois at Chicago. In Sociolinguistics, he pursues quantitative dialect research with the goal of applying or testing linguistic and social theory. His current research interests include the interactions of aging and gender in Puerto Rican Spanish and in Chicago Englishes.