UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Prestige and standard in Canadian English : Prestige and standard in Canadian English : Richards, Donna Jean 1988

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1988_A1 R53.pdf [ 23.91MB ]
Metadata
JSON: 831-1.0098220.json
JSON-LD: 831-1.0098220-ld.json
RDF/XML (Pretty): 831-1.0098220-rdf.xml
RDF/JSON: 831-1.0098220-rdf.json
Turtle: 831-1.0098220-turtle.txt
N-Triples: 831-1.0098220-rdf-ntriples.txt
Original Record: 831-1.0098220-source.json
Full Text
831-1.0098220-fulltext.txt
Citation
831-1.0098220.ris

Full Text

PRESTIGE AND STANDARD IN CANADIAN ENGLISH: EVIDENCE FROM THE SURVEY OF VANCOUVER ENGLISH By DONNA JEAN RICHARDS B.A., The University of Briiish Columbia, 1981 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Department of English) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1988 © Donna Jean Richards, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of F.ngl i sh  The University of British Columbia Vancouver, Canada Date October 14. 1988 DE-6 (2/88) i i ABSTRACT A survey of the use of standard and prestige in general descriptions of English, and of Canadian English in particular, reveals terminological confusion caused by the similarity of the two concepts and by cultural differences among the national dialects being discussed. This work argues, however, that these concepts can and should be distinguished. Once working definitions for both terms are formulated, they are tested against data from the Survey of Vancouver English. Vancouver English reveals little or no evidence of prestige, defined as "that variety (or those forms) used by the highest socio-economic group and emulated by others." The absence of a highest socio-economic group sufficiently well established to provide forms for others to emulate may explain this result, since, in Vancouver, social homogeneity seems to complement the geographical homogeneity that typifies Canadian English. While Vancouver English does reveal evidence of standard, defined as "that variety used by the majority of speakers and typified by correctness," the evidence also suggests that the notion of standard may need to be refined. The effect of various social factors on correctness is analyzed in order to provide a more precise notion of what "correctness" reflects, and education is found to contribute significantly to correctness. Furthermore, consideration of the four processes of standardization—selection, codification, elaboration of function and acceptance—in Canadian English confirms the importance of education to standardization and suggests not only that a standard exists in Canadian English but also that Canadian English is a standard variety distinct from other varieties of English. Standard is thus redefined to reflect more directly the role of correctness and the centrality of the four processes to standardization. The study concludes with a brief reconsideration iii of standard and prestige in light of these Canadian findings and suggests directions for further research. i v Table of Contents Abstract ii List of Tables.... vi List of Figures vii Chapter 1. Introduction: Prestige and Standard 1 1.1 Study of Standard and Prestige 1 1.1.1 Purpose of the Study 3 1.1.2 Scope and Method of the Study 3 1.2 Nature of the Problem 4 1.2.1 Terminological Confusion 4 1.2.2 Existence of Prestige and Standard 9 1.2.3 Social and Regional Dialects and the Development of Prestige and Standard 10 1.2.4 Linguistic Levels as Delineators of Prestige and Standard 14 1.3 Background of the Problem 16 1.3.1 Semantic Similarity 16 1.3.2 Cultural Considerations 17 1.3.2.1 Linguistic Level 20 1.3.2.2 Terminological Inconsistency 21 1.3.2.3 Existence of Standard and Prestige 24 1.3.2.4 Development of Prestige and Standard 25 Chapter 2. Prestige and Standard in Discussions of Canadian English 29 2.1 Use of the Terms in Descriptions of Canadian English 29 2.1.1 Terminological Confusion 30 2.1.2 Existence of Standard and Prestige in Canadian English 33 2.1.3 Social and Regional Dialects and the Development of Prestige and Standard in Canadian English 40 2.1.4 Linguistic Levels as Delineators of Prestige and Standard in Canadian English 43 2.2 Definitions of the Terms 45 2.2.1 Essential Distinctions between Terms 45 2.2.2 Working Definitions 56 Chapter 3. Prestige in the Survey of Vancouver English 58 3.1 The Survey of Vancouver English 58 3.1.1 Description of the Survey 58 3.1.2 Reliability of Self-Reporting 65 3.2 Prestige in the Survey of Vancouver English 69 3.2.1 Forms used by the Highest Socio-economic Group 71 3.2.1.1 General 72 3.2.1.2 Lexical Variables 77 3.2.1.3 Grammatical Variables 78 3.2.1.4 Phonological Variables '. 81 3.2.1.5 Special Profile Items 84 3.2.2 Forms used by the Highest Socio-economic Group and Emulated by Others 86 3.2.2.1 Phonological Items 88 3.2.2.2 Special Profile Items 93 3.3 Development of Prestige in Vancouver English 101 V 3.4 Conclusion 104 3.4.1 The Definition 104 3.4.2 Prestige in Vancouver English 105 Chapter 4. Standard in the Survey of Vancouver English 108 4.1 Variety Spoken by the Majority of Speakers 108 4.2 Variety Typified by Correctness Ill 4.2.1 Notion of Correctness 113 4.2.2 Conventional Correctness 118 4.2.3 Social Factors and Correctness 131 4.2.3.1 Gender 132 4.2.3.2 Age 133 4.2.3.3 Socio-economic Status 136 4.2.3.4 Level of Education 138 4.2.3.5 Teachers 139 4.3 Processes of Standardization in Canadian English 142 4.3.1 Selection 143 4.3.2 Codification 146 4.3.2.1 Dictionaries 147 4.3.2.2 Grammar Books 150 4.3.3 Elaboration of Function 154 4.3.4 Acceptance 156 4.3.4.1 Development of Canadian English 159 4.3.4.2 Influences on Canadian English 162 4.3.4.3 Acceptance of Canadian English 166 4.4 Conclusion 169 4.4.1 The Definition 169 4.4.2 Canadian English 172 Chapter 5. Conclusion: Reconsideration of Standard and Prestige 174 5.1 The Terms 174 5.2 Prestige 175 5.3 Standard 177 5.4 Canadian English 179 5.5 Future Studies 179 5.5.1 Canadian Studies 181 5.5.2 General Studies 186 5.6 General Conclusion 187 Bibliography 188 Appendices A: Outline of Contents of SVEN Questionnaire 209 B: Reading Passage 210 C: Visual-Aural Prompts 213 D: Forms Used by Most Speakers in All Socio-economic Groups 229 E: Forms Used by Majority Versus Those Used by Each Socio-economic Group 251 F: "Correct" Grammatical Forms 273 G: Dictionaries and Handbooks Consulted 274 v i List of Tables Table 3.1 Distribution of Reliable and Unreliable Reporting Across Socio-economic Groups 68 Table 3.2 Pattern of Emulation of Prestige 87 Table 3.3 Phonological Variation Across Speech Styles: Socio-economic Group IV and One other Group Uniform 89 Table 3.4 Phonological Variation Across Speech Styles: Socio-economic Group IV least Consistent Across Styles 91 Table 3.5 Special Profile Variation Across Speech Styles: Socio-economic Group IV Uniform 94 Table 3.6 Special Profile Variation Across Speech Styles: Socio-economic Group IV Not the Only Uniform Group 97 Table 3.7 Special Profile Variation Across Speech Styles: Socio-economic Group IV the Group to Differ 98 Table 3.8 Special Profile Variation Across Speech Styles: Socio-economic Group IV and One Other Group Differ 99 vii List of Figures Figure 3.1 Uniformity 68 Figure 3.2 Variant Different from Majority-Overall Pattern 87 Figure 3.3 Variant Different from Majority-Lexicon 89 Figure 3.4 Variant Different from Majority-Grammatical Variables ....91 Figure 3.5 Variant Different from Majority-Phonological Items 94 Figure 3.6 Variant Different from Majority-Special Profile Items 97 Figure 3.7 Phonological Variation by Socio-economic Groups Across Speech Styles 98 Figure 3.8 Special Profile Variation by Socio-economic Groups Across Speech Styles 99 To the One through whom all things are possible. And to those who tried to make it impossible. Also to my family and friends, who, at different times, did both. 1 CHAPTER 1 PRESTIGE AND STANDARD The concepts of prestige and standard are central to sociolinguistic explanations of language variation and change. Labov's exemplary work on the "Social Stratification of English in New York City" (1966) demonstrates the correlation between social stratification and linguistic variation and identifies the importance of sociolinguistic prestige in language variation and change. Subsequent works, such as those done by Trudgill in Norwich (1974), Pringle, Dale and Padolsky in Ottawa (1985), Clarke in Newfoundland (1981) and Gregg in Vancouver (1984), corroborate Labov's findings and establish prestige as an integral part of socio-dialect studies. Standard is also an essential consideration or subject of investigation in socio-dialect studies. Wells' definition of standard as an accent that "enjoys overt prestige" (1982:104) and Trudgill's recognition that RP, the prestige accent in Britain, "only occurs together with Standard English" (1974:18) are just two indications of the extent to which prestige and standard are bound. However, in general discussions of language structure, variation or history, the terms prestige and standard are often neither clearly defined nor consistently used. 1.1 Study of Standard and Prestige The theoretical confusion of definition between standard and prestige has serious, practical consequences. Theories of language variation and change depend on clarification of these essential terms; similarly, descriptions of language variation that are based on those underlying theories depend on clarification of these essential terms. However, definitions of these terms should be motivated by empirical analyses of object language data. As Labov 2 points out, it is important to quantify sociolinguistic variation, especially in order to test and better understand theories: If this term [sociolinguistics] refers to the use of data from the speech community to solve problems of linguistic theory, then I would agree that it applies to the research described here. But sociolinguistics is more frequently used to suggest a new inter-disciplinary field--the comprehensive description of the relations of language and society. This seems to me an unfortunate notion, foreshadowing a long series of purely descriptive studies with little bearing on the central theoretical problems of linguistics or of sociology. (Labov 1966:v) Clearly, both descriptive studies and interpretation of the data gathered in the studies are needed to "solve problems of linguistic theory." Both are also needed to solve problems in sociolinguistic theory, not in the sense of "the comprehensive description of the relations of language and society," but rather in the sense of that part of linguistic theory that accounts for variation that is socially determined. The results of the studies done must be used not only to add to the body of sociolinguistic description but also to reassess the theory on which the studies were based, including the definitions of essential terms. It is also important to consider data gathered in different areas of the English speaking world. As Baugh, writing about standard, points out, "We must recognize that in the last two hundred years English has become a cosmopolitan tongue and cultivate a cosmopolitan attitude toward its various standard forms" (Baugh 1957:381). Not only has the attitude become much more cosmopolitan since the late 1950's, but also, following the pioneer work of Labov in the late 1960's, most dialect studies have been what are more accurately called socio-dialect studies and the body of available data has increased immensely. It is possible to use these studies, especially those from different countries, to broaden the scope of reassessment of sociolinguistic 3 theory. Although the number of Canadian studies lags behind those done in Britain and the U.S., the available Canadian data, especially as they differ from American and British data, are necessary both to extend the cosmopolitan attitude toward the standard forms of English as Baugh suggests and to reassess sociolinguistic theory as Labov suggests. 1.1.1 Purpose of the Study The purpose of the study is twofold: first, to expand the description of Canadian English by analyzing and reporting different aspects of the data that were gathered in the Survey of Vancouver English (SVEN) and, second, to show that the terms standard and prestige can and should be distinct. 1.1.2 Scope and Method of the Study The study will begin with a survey of the terms standard and prestige in the general literature; this survey will reveal the background and the nature of the terminological inconsistency and confusion (Chapter 1). The study will then survey the terms in descriptions and discussions of Canadian English; this consideration will provide a means of better understanding the contextual aspects of the disagreement about standard and prestige and will lead to the construction of working definitions of the terms standard and prestige (Chapter 2). These working definitions of the terms (or aspects of the definitions) will be provisional ones and will then be tested empirically against data from the Survey of Vancouver English. Claims for both the existence and the development of prestige in Canadian English will be tested (Chapter 3). Claims for the existence of Standard Canadian English and the 4 role of grammatical correctness as an aspect of Standard English will be analyzed; the definitions will be refined in light of the findings (Chapter 4). The work will then conclude with a reconsideration of prestige and standard in the Canadian context and recommendations for future studies in light of the findings (Chapter 5). 1.2 Nature of the Problem Wyld, writing in 1914, claims that "the facts [about Standard English] are so patent that they have merely to be stated to command assent by all who consider questions of this kind" (Wyld 1963 [1914]:150). However, the facts are far from clear; a survey of histories, grammars and dialectal and sociolinguistic descriptions of English reveals that there is some terminological confusion, especially between uses of standard and prestige, and little agreement about the actual existence of either standard or prestige, about the development of standard and prestige forms and about linguistic levels as delineators of standard and prestige. 1.2.1 Terminological Confusion Prestige has been defined by Gregg as "those [forms] used by the top socio-economic group who have the highest level of education" (Gregg 1984:A3). Labov's finding that "in general, a variant that is used by most New Yorkers in formal styles is also the variant that is used most often in all styles by speakers who are ranked higher on an objective socio-economic scale" (Labov 1966:405) provides the further qualification in the generally accepted definition of a prestige form or variety as those forms used by the highest socio-5 economic groups and emulated by others. Such definitions, based on empirical evidence, identify concrete referents for the term prestige. However, there is a somewhat more general sense of prestige evident in the literature as well. A prestige variety may be described simply as a dialect that is admired by many speakers; for example, McDavid describes prestige as "the speech of some people or groups of people [which] is considered better than that of others, and hence more worth imitating" (McDavid 1969:80). Prestige, then, may refer specifically to the dialect of a particular group of speakers or it may refer more generally to a positively evaluated dialect; in most of its uses it does have clear referents. On the other hand, standard is used to refer to concepts or objects that are difficult to delineate because they are much broader than the referents of prestige. When delimiting a variety of a particular language such as English, linguists use standard in its general sense of an approved model. For example, Trudgill defines Standard English as "the dialect that is normally used in the writing of English throughout the English-speaking world (Trudgill 1983:186). MacLeish omits reference to writing and identifies Standard  Engl ish as "a term applied to an item of usage or a dialect of English that is acceptable to educated, cultivated speakers of the language. Standard English is the English used to carry on the daily business of the nation" (MacLeish 1972:115). Although both these referents for Standard English are, in principle, identifiable, they are nonetheless difficult to isolate; how does one determine the parameters of a dialect "normally used in writing...throughout the English-speaking world" or "used to carry on the daily business of the nation"? The referents are, to say the least, vague or diffuse. Thus, though 6 standard can, in some sense, be defined, the referents of the term are less specific than those of prestige. Despite the possibility of distinct senses, however, there is considerable terminological confusion in descriptions of English. Some writers use standard to refer to prestige while others use it to refer to a composite of standard and prestige and still others keep the two separate. De Saussure, in his lectures given between 1906 and 1911, introduces the notion of language having only dialects and describes the emergence of one of the dialects, the "privileged" dialect and its promotion "to the rank of official and standard language;" he says that this dialect "seldom remains the same as it was before. It acquires dialectal elements from other regions and becomes more and more composite" (de Saussure 1966:195). Like others that follow him, de Saussure uses standard to refer to the imposed or selected dialect that is then common in that it "serves the whole community" (deSaussure 1966:195). However, his recognition of such a dialect as being "privileged" and "promoted" suggests prestige as well as standard. Subsequent scholars also identify the standard as both common and a prestige dialect. Francis defines a prestige dialect as "a dialect admired and emulated by the speakers of other dialects" and goes on to say that "a dialect generally admitted by the majority of speakers to be superior to all other dialects in its language is the standard dialect of that language, or simply the standard language" (Francis 1954:48). Thus, Francis merges the two; for both de Saussure and Francis standard and prestige are not clearly distinguishable but rather are elements of the same variety. Others simply use standard to refer to prestige. Wells uses standard to refer to an accent that "enjoys overt prestige" (Wells 1982:104) thereby using standard 7 to refer to what others call prestige. Similarly, Bloomfield writes, "Children who are born into homes of privilege, in the way of wealth, tradition, or education, become native speakers of what is popularly known as 'good' English; the linguist prefers to give it the non-committal name of standard English" (Bloomfield 1933:4s). 1 Standard, then, has been used to refer to prestige as well as to standard and to a composite consisting of both standard and prestige. On the other hand, Wyld seems to distinguish standard and prestige. He defines Received Standard as "that form which all would probably agree in considering the best, that form which has the widest currency and is heard with practically no variation among speakers of the better class all over the country," and Modified Standard as "Standard English, modified, altered, differentiated by various influences, regional and social...[with] no uniformity, and no single form of it heard outside a particular class or a particular area" (Wyld 1963:150). Wyld's terminology is unfortunate; he uses standard to describe both prestige (his "Received Standard") and standard (his "Modified Standard"). Confusion is inevitable when the same term, whether modified or not, may refer to both the "best" English and to "the vehicle of everything that affects the nation as a whole" (de Saussure 1966:195), but Wyld 1 Note that though linguists intend standard in a strictly descriptive sense, having no evaluative dimension and replacing such value laden terms as prestige, proper and correct, the attempt has been largely unsuccessful; just as the complementary term, nonstandard, is value laden, so is standard. The attempt has led to some confusion in terminology but has not successfully provided a term which is purely descriptive in import. 8 makes a distinction, probably between standard and prestige. 2 It could be argued that Wyld is not making a real distinction between standard and prestige, but that he is referring to degrees of the same thing; however, his identifying Received Standard as the best and as that used by those of the better class allows or leads to distinction between the two. Samuels, in his discussion of motivation of language variation and change, makes the distinction; he recognizes that standard and prestige are similar but not the same. He writes: Where a standard language exists, this too may provide a model to speakers of lower-class dialects, especially through the medium of broadcasting. But the dialect of national broadcasters may not necessarily coincide with the regional prestige dialect of a particular city; both are likely to influence . other dialects in the same general direction, but they cannot be entirely equated. (Samuels 1972:103) Some writers, then, use standard to refer to something having the features of both standard and prestige; others use it to refer to something having the features of prestige rather than standard. Still others use the term standard to refer to distinct referents including those which would more precisely be 2 Wyld's notion of Modified Standard includes regional and social varieties of Standard English, presumably as opposed to nonstandard English. The distinction between standard and nonstandard varieties remains unclear, though classification of varieties as either standard or nonstandard flourished with the movement from prescription to description, especially after the early American structuralists and transformationalists' and generativists* insistence on neutral (i.e. not normative or judgemental) terminology; because the term correct had to be eliminated from grammatical descriptions, sentences became either grammatical or ungrammatical, and these, in turn, were either acceptable or unacceptable in a context of use. Moreover, dialects became either "standard" or "nonstandard" and more terminological confusion arose. Not only is there confusion between standard and prestige, then, but also between standard as a variety of English and standard as a kind of English, as determined by judgement of grammaticality. 9 called prestige and correct as well as to the standard itself. With standard used in so many different senses, confusion is inevitable. 1.2.2 Existence of Prestige and Standard There is also little agreement among scholars about the empirical status of a standard or a prestige in English. Wyld asserts that "the existence of Standard is a reality" (Wyld 1963:150), and Labov argues for the existence of prestige: The notion of "prestige" must be defined in terms of the people using it and the situation in which it is used; that is, brought out of the area of speculation and made the focus of empirical investigation. (Labov 1972:308) His work in New York City (1966) reveals that a prestige, when defined as forms used by speakers of a particular socio-economic group, does exist. Labov identifies overt prestige (i.e. that which is above the level of consciousness and is generally associated with higher socio-economic groups) and speculates that covert prestige (i.e. that which is below the level of consciousness and is generally associated with lower socio-economic groups) also exerts pressure on language variation. His speculations about covert prestige are confirmed by Trudgill, who finds empirical evidence for covert prestige and concludes that he has "established that covert prestige does in fact exist" (Trudgill 1983:179). On the other hand, Jones writes simply that "it cannot be said that any standard exists" (Jones 1950:3). Similarly, Francis argues that "there is no such thing as a single standard form of American English especially in pronunciation" (Francis 1963:247). It seems that the disagreement about the existence of standard and prestige is a matter, in part, of how the particular terms are used by the writers in question; therefore, 10 this point of disagreement will be considered in detail later in this work when definitions of the terms are being constructed. However, it is clear from the survey of the literature that there is little agreement about the existence of standard and prestige in English. 1.2.3 Social and Regional Dialects and the Development of Prestige and Standard There is also little agreement among linguists about the development of prestige and standard, particularly as related to social and regional varieties. Linguists generally agree that there are two kinds of dialects—regional dialects and class, or social, dialects. Regional dialects are those whose particular features are peculiar to a limited geographical region. Dialect geography may be approached in one of two ways: either an area is selected and the dialect spoken within that area studied for distinctive linguistic characteristics, or linguistic variants are studied and the area in which those variants are found is determined. Using either method, a dialect geographer may describe and contrast dialects from areas of varying size, such as specifying Canadian English as opposed to British or American English, British Columbian English as opposed to Newfoundland English, Vancouver English as it differs from Ottawa English or from a rural British Columbian variety, and even, some claim, Kitsilano English as it differs from Shaughnessy English within Vancouver English. However, because areas within cities are determined more by social than geographical criteria, distinctions between varieties within an urban area are more likely to be determined by class than geographical location and, hence, indicate sociolects rather than dialects. Like regional dialects, class dialects may be studied either by determining the socio-economic class (usually grossly categorized as Upper, Middle and 11 Working Classes) of the speakers and identifying the linguistic variants common to speakers within the class or by studying and recording linguistic features and identifying the class of the speakers using the features. Clearly, the two kinds of varieties intersect extensively. Although class structure may be explicit or merely tacit, each designated geographical area may be broken down by class; some social hierarchy, however informal, develops in each region. Thus each speaker of a language may speak both a regional and a class variety. Furthermore, as was the case in the early development of English, a regional dialect may develop into a class dialect by a process Kurath calls "spreading," which involves interaction of social and regional elements: spreading [of various phonetic features] takes place on social levels, cultured speakers being first to adopt features of a neighbouring prestige dialect. The middle class probably adopts these innovations from the cultured in their own communities. (Kurath 1964:142) Hence a regional dialect may become a prestige or social dialect in a dialect area different from that of its origins. The considerable overlap between social and regional dialects and the possibility of one becoming another contributes to there being little agreement among linguists about the relationship among social and regional dialects and standard and prestige. Wyld distinguishes regional and social dialects and identifies Standard English as "chief among the latter," defining it as "the best and most refined type of English, that which in one form or another has long been usurping the place of the old Regional dialects" (Wyld 1963 [1914]: 14). Wyld's statement that "everyone who does not speak a Regional dialect, speaks a Class dialect" (Wyld 1963:148) suggests that the two could be mutually exclusive. Such a division is as complex and controversial as 12 de Saussure's insistence on dichotomizing synchronic and diachronic was and is best recognized as extreme and perhaps more ideal and theoretical than real and practical. Halliday, who describes "'RP,' standing for 'received' (that is, generally accepted) pronunciation," as a "regionally neutral variety of English," recognizes the interaction of social and regional factors in language and dialect development; he writes, "Our dialects and accents are no longer simply regional: they are regional and social, or 'socioregional'" (Halliday 1973:18). The practice of including both social and geographical factors in modern studies of language variation (i.e. socio-dialectal studies) indicates the extent to which theorists have come to recognize the interrelation between social and regional elements in variation. It seems, then, more important to recognize the vitality of interaction between regional and social elements in variation than it is to maintain or assert a distinction between the two. There is also little agreement about the pattern of development of standard varieties in relation to regional dialects. Sturtevant claims that the standard is a dialect "that imposes itself upon the speakers of other dialects" (Sturtevant 1917:157). Similarly, Trudgill calls Standard English a superposed variety of English because there is a general consensus among educated people, and in particular amongst those who hold powerful and influential positions as to what is standard English and what is not — standard English is, as it were, imposed from above over the range of regional dialects. (Trudgill 1974:18) Thus he views standard, a class dialect, as being superposed on regional varieties. Similarly, Labov discusses Standard in relation to both social and regional dialects. He defines dialect in what he calls a neutral sense as a subvariety of a language and recognizes standard as a dialect by that 13 definition. However, he goes on to give Standard English a different status from other dialects for three reasons: first, because the standard is clearly superordinate in a hierarchy of prestige or appropriateness for formal speech; second, because the standard has a technical vocabulary and a literary syntax which are largely missing with other dialects; and third, because it has widespread distribution through the mass media, with less geographic differentiation than we find at the colloquial or vernacular level. (Labov 1972b: 191) Again, Standard, defined more by class than by regional criteria, is considered superordinate to or superposed on regional dialects. On the other hand, Sapir, discussing the future of regional dialects, writes that "the modern mind insists on having the process of standardization take the form of a democratic rather than an aristocratic process" (Sapir 1931:126), thereby suggesting that there is no impositon of a standard. Similarly, Jones writes that "if the public wants a standard...some appropriate standard will evolve itself" (Jones 1950:3), suggesting that there be no imposition but rather a development. They seem to think, then, that, if there is a standard variety, it will emerge or develop within a speech community but cannot or ought not be imposed on speakers. To some extent, Labov and Trudgill are classifying or describing kinds of varieties of English rather than accounting for the way in which a standard variety might develop in a speech community as Sapir and Jones do. The assumptions underlying the classification and the account of development differ; hence, what seems like a disagreement about the relationship between social and regional varieties and standard or prestige arises from the different assumptions made respecting the heirarchic/democratic variable. On the one hand, classification of standard in relation to regional and social dialects 14 assumes that the relationship between standard and regional varieties is hierarchically motivated, Standard being highly rated; on the other hand, explanations of the development of standard assume democratic (and not hierarchic) motivation. These, then, are incompatible assumptions rather than disagreement about the relationship between social and regional varieties and standard or prestige. Thus, disagreement about the development of standard and prestige reflects the complexity of processes of development. If the relationship between social or regional varieties and standard or prestige is important to development, it is likely that the essential consideration is whether standard and prestige are derived from social or regional varieties rather than whether standard and prestige are standard or regional varieties. However, the means of development of standard and prestige remains unclear. 1.2.4 Linguistic Levels as Delineators of Prestige and Standard There also seems to be little agreement about the relationship between linguistic level and prestige and standard, especially as such a relationship may be used as a means of distinguishing the standard from prestige; some linguists use accent as a means of distinguishing prestige from standard, while others do not. Strang, using the general definition of standard as "that variety of English...used for all public occasions" claims that "by and large, with rather trivial exceptions, [Standard English] is the same wherever English is used, except in one area of its organization—the accent, or mode of pronunciation" (Strang 1970:18, emphasis mine). By identifying one linguistic level, phonetic, as varying and the other two, lexical and 15 grammatical, as being more uniform, Strang allows for standard being delineated by linguistic level. Trudgill notes regional variances within Standard English by citing pronunciation differences rather than grammatical or lexical; thus he, too, uses linguistic criteria to delineate Standard. He writes, "There is no universally acknowledged standard accent for English, and it is, at least in theory, possible to speak standard English with any regional or social accent." He notes, however, that some localized accents do not usually occur with Standard English and that "there is also one accent which only occurs together with Standard English. This is the British English accent... known to linguists as RP," thus distinguishing standard, which has a "widely accepted and codified grammar and vocabulary" (Trudgill 1974:18) from prestige which refers to the pronunciation of RP. On the other hand, Francis claims that all social levels of English ... educated or standard English... uneducated English, and between them ... the vernacular... have in common the larger part of their grammar, pronunciation and basic vocabulary but are marked by significant differences in all three areas. (Francis 1963:246) Labov and linguists who follow him confirm Francis' statement, finding variation on all linguistic levels. Thus, there seems to be little agreement about the relation between social variation and linguistic level and about using the levels as criteria for delineating prestige and standard. It seems from the survey of the literature, then, that the facts about standard and prestige are not, in Wyld's terms, "patent" (Wyld 1963:150). The terms standard and prestige are used loosely, even inconsistently, and there seems to be little agreement among linguists about what constitutes standard and prest ige. 16 1.3 Background of the Problem 1.3.1 Semantic Similarity To some extent, the confusion about standard and prestige results from the similarity between the two concepts; the two terms share semantic features that are essential in sociolinguistic explanations of language variation and change. Prestige, in the general sense of something that has been positively evaluated, is motivational in language variation and change. As Arlotto writes, The importance of social prestige in accounting for the acceptance and spread of new linguistic forms cannot be overestimated. In the last analysis, people will tend to adopt linguistic.habits different from their own only when they feel that in some way these other habits are "better." As with most human institutions, what constitutes a better item is defined by the higher social prestige of the individuals who use that item. Thus, the prestige factor might be looked upon primarily as a cause or reason for linguistic borrowing, whether across language or across dialect boundaries. (Arlotto 1972:204) Similarly, McDavid writes, In any locality the speech of some people or groups of people is considered better than that of others, and hence more worth imitating. The last force—prestige or lack of it—is the essential cause of the development of social dialects. (McDavid 1969:80) Both writers, then, use prestige in the sense of forms or varieties that are positively evaluated. Both Standard English and prestige forms, or varieties such as RP, are evaluated positively and may act as models for speakers of other varieties; therefore standard and prestige share the feature [+POSITIVELY 17 E V A L U A T E D ] . 3 In this sense, then, both may be motivational in language variation and change. This shared feature facilitates most, if not all, of the terminological inconsistency and confusion; the terms standard and prestige are used interchangeably when the feature [+POSITIVE EVALUATION] is taken to be central and the main point of reference. That is, the terms seem to be used interchangeably in a general sense of prestige rather than in any technical sense. Thus, the terminological inconsistency is explainable, even understandable; I will argue, however, that the terms can and should be distinguished in spite of this important area of overlap. 1.3.2 Cultural Considerations Another partial explanation for the terminological confusion between standard and prestige is the different contexts, or extended contexts, in which the terms are used. An understanding of any term can be fixed only when the word is considered in its context. Although usually that context refers to the co-text, i.e. the sentence, clause, paragraph or passage in which the word occurs, in the case of prestige and standard it is necessary to consider an extended context, i.e. the situation and culture in which the term is being used. Linguistically, the British and North American contexts differ in two 3 [+PRESTIGE] would also express the important overlap; however, I use [+POSITIVELY EVALUATED] to avoid confusion between prestige in its technical, sociolinguistic sense and in its general sense. I would prefer to use the feature [ -STIGMATIZED] here to be certain to avoid confusion. However, the features [-STIGMATIZED] and [+PRESTIGE] are not complementaries; rather they are antonyms—poles of a set of graded terms; assertion of one implies denial of the other but denial of one does not imply assertion of the other. Thus, though [ -STIGMATIZED] is essential to the concepts of standard and prestige, the feature does not exactly express what is needed here. (See 2.3 of this work for a more thorough discussion of these problems.) 18 important ways: first, the settlement periods and patterns are different and, therefore, the dialect patterns in the two locations are different; second, usage can be different because of the dialect differences. English has been used and developing in Britain since as early as the fifth century; British dialects, then, developed at a time when there was no widespread or rapid travel or general public education of the masses and in a culture in which there was an explicit aristocratic social hierarchy. In North America, on the other hand, English has been used and developing since it arrived with British colonists in the seventeenth century; much of the development of the language has taken place during a time of widespread public education, of widespread travel and in nations in which social hierarchy is implicit, democratic and developing rather than explicit and aristocratic.4 As a result, in Britain there are more identifiable distinct regional varieties of English in relation to the size of the 4 Bloomfield's early identification of the social differences, though somewhat extreme, is not without grounding; he writes "In England there are similar regional types, but they are not granted equal value...In England, but scarcely in the United States, provincial colorings of standard English are tied up with differences of social level" (Bloomfield 1933:49). More recent sociolinguistic work such as that done by Labov (1966) reveals that there is also a correlation between social status and language in the U.S.; however, I think the explicitness of the British aristocracy as opposed to the implicitness of an emerging elite (or a more democratically developed elite) in America does affect the extent to which and the rate at which nations are sociolinguistically stratified. I believe, therefore, that the differences between the two cultures ought not be ignored. 19 land mass than there are in North America,^ and in Britain, a clear prestige is established whereas in North America the emergence of a prestige is less clear or only still in progress. I believe these cultural differences account for much of the terminological confusion and inconsistency and for much of the disagreement about the importance of linguistic level as a delineator of standard and prestige, partially account for the disagreement about existence of standard and prestige and give some insight into the disagreement about the development of standard. For example, those writers who are describing British English or varieties of World English rather than varieties of North American English are more likely to distinguish standard from prestige because they must include in their descriptions the prestige accent, RP. On the other hand, those writers describing varieties of North American English for which there has not yet been identified a prestige accent comparable to RP, may use the terms more loosely without seeming confusion. Considering each of the points of disagreement in light of the extended contexts in which the terms are used will reveal that much of the confusion and disagreement can be explained; 5 It can be argued that there are as many regional varieties in the U.S. as there are in Britain, but I think to do so is to reveal a particular focus rather than a fact. Focusing on distinctions, one finds many varieties whereas focusing on similarities, one finds a national variety. The point I am making here is the very general one, which is evident to even the most untrained people: variety is wider and greater per square mile in Britain than it is in North America. If 10 people were gathered from across Britain and brought together in a room, one would hear many differences in speech; if 10 people from across the U.S. were gathered together, one would hear differences, though not as many; if 10 people from across Canada were gathered, one would hear even fewer differences—one might even argue that most of the speakers sounded the same. It is in this general sense that I make reference here. 20 furthermore, I believe that the same consideration will support my argument that the two terms be distinguished. 1.3.2.1 Linguistic Level Much of the disagreement about the importance of linguistic level to the delineation of prestige arises because of the differences in the contexts. The prototypical prestige variety of English is the British accent known as RP (Received Pronunciation), that accent used by the well-educated upper socio-economic groups in Britain and emulated by those aspiring to approach similar positions. RP can be, and has been, described6 in detail and is clearly distinguished from other British and nonBritish accents. In Britain, then, there is a clearly identifiable referent for prestige. Because the referent is an accent, it is reasonable for those scholars who are describing British English or varieties of English including British English, to argue that linguistic level is essential in distinguishing prestige from standard. There is, however, no comparable prestige accent among the North American varieties of English; as Francis writes of the U.S., for example, "the nearest thing to [a standard] is...network English...Because of its nationwide use, network English is an acceptable standard form everywhere. But it is not a prestige dialect" (Francis 1963:247).7 Though there have been many 6 See, for example, Wells (1982) and note that such scholars as Sweet (1890), Jones (1950[1909]) and Wyld (1963[1914]) have chosen to describe RP because it has been, for example, the most "widely understood pronunciation" (Jones 1950:3). 7 Once again there is disagreement among scholars about the status and even existence of network English or General American. Some, such as Bailey 21 descriptions of American varieties, including Standard American, there is no single accent or dialect that can be identified as a distinct prestige in the sense that RP is in Britain. Labov's work (1966, 1972) identifies some prestige forms, but there is not as yet any identification of a prestige accent or variety as a whole. Similarly, in Canada, works such as those done by Gregg (1984), Pringle (1985) and Clarke (1981) identify some prestige forms, but there has not been any variety or accent identified that would be a prestige Canadian accent comparable to RP in Britain. In the North American context, then, there are no clear grounds for considering linguistic level as a viable determiner of prestige. 1.3.2.2 Terminological Inconsistency The same cultural differences that account for what appears to be disagreement about the importance of linguistic level to distinguishing between standard and prestige, contributes to the terminological inconsistency in uses of standard and prestige. The fact that there is an identifiable prestige accent in Britain encourages distinct use of the terms standard and prestige by those who are describing or discussing either British English or English in general; linguists describing English including RP, then, must maintain a distinction between prestige, which refers to RP, and (1982) disagree with Francis, claiming that no such variety exists or is used. I think, however, that this, too, is a question of focus; one can look either at the similarities of the speech of many Americans from across the nation or at the differences. In the former case, one would be able to identify something that could be called General American (or Network English); in the latter, one would have to argue that there is no such variety but rather many smaller regional varieties (or even idiolects). 22 standard, which might refer to the generally used variety of English, particularly the grammar and lexicon of that variety, that may or may not be accompanied by the prestige accent. However, the absence of a clear prestige in North American contexts leads to less precision in the use the two terms. Because there is no clear referent for prestige in North America and it is not yet clear to what extent standard is comparable in North America to prestige in Britain, there has been a tendency to use standard, the more general term, frequently and loosely in the North American context; in the absence of a clear referent for prestige there has been less demand for terminological consistency outside of Britain. The differences in contexts thus go a long way toward explaining the origins of the terminological inconsistency. To a very great extent, these arise from dialectal pressures exerted not only by the language under investigation but also by the intrusive effect of the dialect spoken by the investigator. The same differences between British and North American cultures that have affected the development of language varieties and prestige, have affected the use of the pertinent terms. Even the term dialect** has been used differently in the two nations, a difference which may reflect a distinction between what has become technical linguistic usage and the common or lay definition of "an inferior speech variety;" McDavid defines dialect in the American sense as "simply any habitual variety of a language, regional or social" (McDavid 1969:80) but he clarifies that "in other parts of the world, scholars would not 8 There is little agreement about the ways in which dialect and language might be best defined; for helpful discussions and redefinition of the terms, see Hudson (1980) and Chambers and Trudgill (1980). 23 use dialect for the speech of educated men and women...In short, for them dialect means a mode of speech quaint and uneducated, found in isolated communities" (McDavid 1969:79).9 These dialect differences reflect the effect of cultural differences on dialect development and study. McDavid identifies the effect of the cultural differences on the assessment of language varieties: Until recently, there were two typical relationships between standard and nonstandard dialects. In England there was one standard variety, many local nonstandard ones. In the United States, there have been several regional standard varieties, each with corresponding nonstandard ones. (McDavid 1969:87) The tendency now is for linguists to allow that there may be variation within standard (there may be varieties of standard) and to attempt to adopt neutral terminology and a neutral approach to language description; however, cultural differences clearly affected early language description. The differences also indicate the extent to which variation can affect communication; not only can the nuances of terms be lost in different cultures, but also the referent can change with the context. In Britain, the specific referent for prestige (RP) acts to filter out some of the features which 9 Linguists are now careful to distinguish senses of dialect; as Chambers and Trudgill write: "What exactly is a dialect? In common usage, of course, a dialect is a substandard, low status, often rustic form of language, generally associated with the peasantry, the working class, or other groups lacking in prestige. DIALECT is also a term which is often applied to forms of language, particularly those spoken in more isolated parts of the world, which have no written form. And dialects are also often regarded as some kind of (often erroneous) deviation from a norm—as aberrations of a correct of standard form of language...we shall not adopt any of these points of view. We will, on the contrary, accept the notion that all speakers are speakers of at least one dialect—that standard English, for example, is just as much a dialect as any other form of English—and that it does not make any kind of sense to suppose that any one dialect is in any way linguistically superior to any other" (Chambers & Trudgill 1980:3). Thus they are careful to address dialectal differences in use of terms. 24 would otherwise be assigned to standard thereby limiting the referent for standard. In America, there is no comparable referent for prestige: therefore, standard may be used to refer to something that is comparable to the standard in Britain but has the added features of prestige. One must, therefore, be clear about the context in which the terms are used in order to fully understand both the nuances and referents of terms. 1.3.2.3 Existence of Standard and Prestige To some extent, the cultural differences could also contribute to lack of agreement about the existence of prestige; if one identifies RP as the prestige accent and can identify speakers of RP, then one is not likely to argue against the existence of such an accent. However, if one cannot identify a prestige accent in a community, then one may well question the existence of such a thing. It would seem likely, then, that those writing of British English would be more likely to argue for the existence of prestige than those writing of North American English. Most of the comments on existence and nonexistence, however, have been on standard rather than prestige. This disagreement cannot be explained by a difference between British and North American cul tures. 1 0 Because standard is common to both cultures, the lack of agreement about its existence is not affected by differences between British and North American contexts. It is more likely that the lack of agreement about the existence of standard arises because of the broad scope of the referent for standard than because of cultural differences. Thus, cultural 1 0 Recall that Jones (1950) writing of English (mostly British) argues against the existence of a standard. On the other hand, Baker (1986) and Kinloch (1986) writing of Canadian English, argue for the existence of a standard. 25 differences could contribute to disagreement about the existence of prestige but are much less likely to contribute to disagreement about the existence of standard. However, the variant uses of the terms standard and prestige that arise from cultural differences may mask agreement about the existence of s t a n d a r d . 1 1 Thus, it is important to keep the differences in mind when considering any discussion of the existence of standard and prestige. 1.3.2.4 Development of Prestige and Standard Similarly, the effect of cultural differences on the disagreement about the means of development of standard and prestige is relatively small and only partially explanatory. The hierarchical class structure and the extended time English has been the language of the land may account, in part, for the clearly identifiable prestige accent, RP, in Britain; the lack of the same two factors in North American countries may have contributed to the absence of a similarly identifiable prestige accent in Canada or the U.S. It may even be the case that a prestige accent will not develop in North American countries as it has in Britain, in part, because of the differences between established 1 1 It is important to note, for example, that the lack of existence of something (a prestige accent, for example) in one culture does not indicate a lack of the existence of that thing in general or in other cultures; rather, the absence of something in a particular culture raises questions about the culture and perhaps about the appropriateness or the applicability of the concept in that culture rather than about the existence of the thing. Pringle's dismissal of the existence of standard but consideration of a concept of standard, for example, could suggest either the philosophical difference between existence of things and of concepts or that the concept of standard may be valid because it has been identified in some communities but that such a thing does not exist in the community which he is describing (see 2.1.2 and Pringle 1986). It is, however, reasonable to accept a concept of, say, standard, as valid, though not appropriate in a particular context, only if there is evidence of the existence of standard in some other community; there must be some evidence to validate the concept before it can be accepted either valid or existing. 26 aristocracies and the New World democracies. Until there is a clearly established, distinct and identifiable upper class, there will not be a distinct prestige accent such as RP is in Britain. However, the cultural differences have less effect on standard; as standard is common to both North American and British cultures and, in a general sense, is less dependent on socio-economic classes than prestige is, the development of standard is less clearly affected by the differences between aristocracy and democracy and by time than the development of prestige is likely to be. In many ways, standard is more closely linked to written language 1 2 than it is to simple socio-economic status, though there is, of course, standard speech as well as standard writing. Samuels explains the relationship between standard and written language: Once a standard has come into being, from whatever diverse origins, the most potent force in its acceptance, propagation and preservation is the written language. Written Standard English has not changed radically since it first emerged as the vehicle of off icial business and administration around 1430...In its pronunciation, change and dialectal variation continued on a large scale till 1750, and on a smaller scale to the present day. But the basis of standard throughout has been its written form, which provides a codification for what otherwise would be far more fluid. (Samuels 1972:109) Although historically in Britain socio-economic status had a profound effect on one's exposure to literary language, the same is not true today and has not been true to the same extent in North American countries. Thus, there is no longer necessarily a connection between socio-economic status and exposure to literary language through education as there once was in Britain; taking standard as being closely related to literary language, one can see that it is no 1 2 Note that de Saussure uses the terms standard English and literary English interchangeably in Course in General Linguistics ; see, for example, his discussion of "Literary Language and Local Idiom" (de Saussure 1966: 195). 27 longer bound to socio-economic status as prestige is. Hence, the differences between British and American cultures do not have the same effect on either the development of standard or an understanding of the use of the term standard as they do on the development of prestige and an understanding of the use of the term prestige. The inconsistent use of standard and prestige, then, arises from the considerable overlap in referents of the two terms and from differences between the cultures whose language varieties are being described. Cultural differences also account for disagreement about the validity of distinguishing standard from prestige on the basis of linguistic level and partially account for disagreement about the existence and development of prestige and standard. It is important, therefore, to consider both context and usage in order to clarify the distinction between standard and prestige. As national context seems to be a factor in the confusion between standard and prestige, it is particularly important to extend the contexts in which the terms are studied and used in order to clarify the problem. The important differences seem to be those between Britain and North America; clearly, Canada and the U.S. have many more cultural similarities than do Canada and Britain or Canada and the U.S.13 However, there may be some difference either between the sociolinguistic environments of Canada and the U.S. or 1 3 Of course, Canada and the U.S. also have much in common with Britain, and politically, Canada is more like Britain than it is like the U.S. However, in those cultural matters that affect language variation and development, there are more similarities between Canada and the U.S. than between either of them and Britain. 28 between the study of socio-dialects in the two nations that provide insight into standard and prestige; at any rate, considering the work that has been done in Canada as well as that done in the U.S. may help clarify the problems. I will, therefore, use data from the Canadian context in order to examine the differences between standard and prestige. The study, then, will consider both the terms and the concepts in the Canadian context in order to determine the distinctions between standard and prestige and to refine definitions of the terms. 29 CHAPTER 2 PRESTIGE AND STANDARD IN DISCUSSIONS OF CANADIAN ENGLISH Before constructing working definitions of standard and prestige and testing them against the Canadian data, it is important to consider the use of the two terms in descriptions and discussions of Canadian English. Only if the problems identified in the general context are also evident in the Canadian context, will it be useful to test the adequacy of definitions against Canadian data; therefore, a survey of Canadian descriptions is essential before undertaking a test and generalizing from the findings. Furthermore, examining the Canadian context specifically may confirm the hypothesis that much of the disagreement about prestige and standard arises because of cultural differences. It will then be possible to more clearly identify the essential features of each term and to construct working definitions. Therefore, use of standard and prestige in descriptions of Canadian English will be considered in order to expand an understanding of the background and nature of the problem and to construct provisional definitions of the terms in light of that understanding. 2.1 Use of the terms in Descriptions of Canadian English Generally speaking, standard is used more frequently than prestige is when Canadian English is being discussed; however, the senses of the two terms are often unclear. These terminological confusions are similar to but do not parallel exactly those discoverd in Chapter 1, where the general linguistic uses are surveyed. They coincide in one respect and diverge in another. Both in general discussions and in Canadian discussion, there is little agreement about 30 the existence and development of prestige and standard. However, unlike in general discussions, one finds little disagreement within Canadian studies about the importance of linguistic levels as delineators of prestige and standard. 2.1.1 Terminological Confusion Although in the Canadian literature it is usual to discuss standard more frequently and extensively than prestige, both terms are used. Both seem to be used to refer to a concept that shares features of standard and prestige, especially in the general senses of that variety or those forms which are used by most speakers and that variety or those forms which are highly evaluated. Avis seems not to distinguish between standard and prestige in Canadian English; he writes: "General Canadian...that variety of English heard generally from Ontario westward...with roots in...Toronto area of influence,...is the prestige dialect throughout most of Canada" (Avis 1975:118n). His recognition of General Canadian (or Standard Canadian) as the prestige variety in Canada suggests that he is using prestige in its most general sense of positively evaluated and that Standard Canadian functions as a prestige variety does in, say, Britain (i.e. it is used by speakers of the highest socio-economic groups and its forms are emulated by speakers of other varieties).1 He allows, then, for the concepts of both standard and prestige, but implies that they exist in Canada in the same variety of English. What is unusual about Avis' usage in the Canadian context is that he uses prestige: it is more common 1 The possibility that Standard Canadian English is considered the prestige in Canada and, hence that referents for standard and prestige are identical will be discussed later in this work (See 2.2.2). 31 among descriptions of Canadian English to use standard to refer to prestige or to a composite of standard and prestige. Southerland also suggests a concept that combines the senses of prestige and standard but use standard where Avis used prestige: he writes, "In most speech communitites, there is a single speech variety called the s t a n d a r d , which is perceived by members of the community to be higher in status than the others" (Southerland 1987:346). Similarly, McConnell claims that "a standard dialect of any language...carries the prestige of its speakers" (McConnell 1979:97), thus implying that elements of standard and prestige are combined. Like Avis, Southerland and McConnell indicate that the concepts of standard and prestige overlap; they, however, use standard to refer to something with features of both whereas Avis uses prestige for the same purpose. Woods uses standard in the sense more of prestige than of standard; he defines standard "in sociolinguistic terms, i.e., the standard is a description and, by implication, a prescription, of how the upper classes and upper middle class of a nation or region use the language" (Woods 1986:139). Thus, it is evident that standard is used to refer to prestige and to a concept that combines the characteristics of standard and prestige. Others distinguish between prestige and standard but use the term standard to refer to both. Chambers bases his discussion on Strevens' (1985) three definitions of standard—language standards2 or "the norms or canons of 2 Strevens1 distinction between standard and standards is an important one which makes explicit the difference between a variety and guidelines or recommendations for usage that is, usually, common to that variety. Such a distinction allows for comparison of the adviseability of providing such standards (guidelines) with the adviseability of describing a standard variety; it also allows for considering overlap between the two. Unfortunately, although I am sure most linguists make the distinction between standard and standards mentally, they do not always reflect the distinction in their use of terms; hence, standard is sometimes used to refer to standard and sometimes 32 generally-accepted language usage," standard language, which further subdivides into standard dialect, or "the grammar and core vocabulary of educated usage," and standard accent, or "the pronunciation, [that is] the sounds, stress, rhythm and intonation of people whose speech is not highly localized" (Chambers 1986:2). On this basis, he uses standard in two senses; he distinguishes standard in an evaluative sense in which something is imposed on people by some authority (a sense of prestige) and in a neutral sense of "an average or conventional property" (Chambers 1986:3). Although Chambers rejects the former sense of standard and urges study of the latter only, he does use the same term for both standard and prestige. Similarly, Kinloch uses standard in both senses, defining it as "that which [is] popular" and as "that which [is] preferable" (Kinloch 1986:40). Pringle uses standard in more than two senses. He defines one "sociolinguistically oriented concept of standard." which he says is "structured covariation of linguistic forms with other sociological markers" (Pringle 1986:21). This first concept is the general sense of standard. Pringle's second concept of standard is "an overarching norm...an overarching awareness on the part of all groups that there is some societal norm different, to some extent, from their common practice, a norm which, in specifiable circumstances, it is appropriate to use (or at least try to use) in preference to one's usual practice" (Pringle 1986:24). This second concept would be more aptly referred to as prestige. Pringle identifies a third concept of standard as "the purist's sense...purely a matter of belief; it is used to refer to standards. Thus standard can refer to a variety of English, a kind of English (as in standard versus nonstandard English—see note page 5) or norms of usage. therefore impervious to empirical investigation...it cannot tolerate variability; part of its essence is that, whenever two forms seem to be fulfilling the same roles, one is to be proscribed" (Pringle 1986:25), a sense of standard for which Pringle would "prefer another term, namely correct" (Pringle 1986:35). Pringle, then, uses standard to refer to standard, prestige and correct. For Chambers, Kinloch and Pringle, then, the concepts of standard and prestige are distinct; however, they use the term standard to refer to each concept. As is the case in the general literature, standard and prestige are used imprecisely and inconsistently; either term may refer to a concept that combines features of both standard and prestige and standard may be used to refer to individual concepts of standard, prestige and correct. However, in the Canadian literature, it is more clearly evident that the terms may be used to refer to something that is both standard and prestigious or, perhaps, the prestige; it is also the case that standard is more likely to be used than prestige. 2.1.2 Existence of Prestige and Standard in Canadian English Just as there is disagreement about the existence of prestige and standard in the general context, there is disagreement among Canadian scholars as well. Kinloch argues that "this standard [a Canadian standard] [does] in fact exist and all educated users of the language [strive], consciously or unconsciously and with greater or lesser success, to achieve it" (Kinloch 1986:40). Baker also has no doubt that Standard English exists in Canada, having, "amongst other things, observed an absence of solecisms among the speakers at the 34 conference, 'In Search of the Standard in Canadian English,' in spite of the diversity of origins, backgrounds, and professions" (Baker 1986:153). On the other hand, Pringle argues against the existence of standard: the notion that an invariable standard of 'correct' Canadian English exists seems...deeply at odds with the nature of Canadian society. Moreover, it takes no account of the extremely complex, patterned covariation which always seems to exist in society's use of language between linguistic parameters and such other parameters as socioeconomic status, sex, race, age, situation and content. (Pringle 1986:35) Note, however, that here Pringle is using standard in the sense of correctness; he is objecting more to prescription of correct (or standard) Canadian English than he is arguing that standard does not exist at all. He goes on, however, to make a point which reveals that, to some extent, the disagreement about existence of a standard is a philosophical one. In a paper given at the Conference entitled "In Search of the Standard in Canadian English," Pringle points out that the title "implies that Canadian English has a standard, which can therefore be found." He argues that the real question is "whether or not there is a concept of a standard which is relevant to Canadian English" (Pringle 1986:21). Pringle rejects the existence of a standard in Canadian English but accepts the existence of a concept of standard. It may be that Pringle's notion of the search for a concept of a prestige or standard is more appropriate than a search for a linguistic realization of that concept.3 It is not 3 Speakers may have a notion of standard or prestige whether or not such a thing has been or can be quantified and delineated; in such a case, it may be more appropriate to consider the existence of a concept of standard rather than the existence of quantification of a variety that is standard. The possibility that language variation is motivated as much by notions or concepts as by actual prestige or standard forms or varieties will be considered further in Chapters 4 and 5 of this work. 35 clear, however, that discussion of existence of a thing as opposed to existence of a concept would move beyond the Platonic shadow on the wall of the cave and be of significance to sociolinguistic analysis or theory; the philosophical aspects of the discussion about existence and nonexistence of prestige and standard need not inhibit studies which describe standard varieties of English. However, other aspects of the disagreement about the existence of standard and prestige must be considered. It is clear that standard is being used loosely and in both the sense of a standard (i.e. standard language) and the sense of standards (i.e. norms or rules). This terminological looseness obscures an important distinction: that between standard language and the source of the standards or norms. Jones argues that "it is thought by many that there ought to exist a standard, and one can see from several points of view that a standard speech would have its uses...But though attempts have been made to devise and recommend standards, it cannot be said that any standard exists" (Jones 1950:3). Here, he distinguishes between standard (i.e. standard language) and standards (i.e. rules or norms). Chambers makes a similar distinction when arguing that a standard can be found in a variety such as Canadian English. He writes, "it will not do to interpret whatever standards we may find as conferring any special prestige on those who maintain them" (Chambers 1986:4). The argument that no standard exists but that standards (i.e. the rules which, if applied, would produce such a standard) do exist suggests that standards are concrete and identifable but that the end result of the application of those standards, the standard, is an abstraction not a reality. In that case, standards precede but do not necessarily produce standard English. What, then, is the source of the standards? If the standards exist, seemingly 36 independently from the standard, then it is possible that the standards may be constructed artificially. This possibility points to the crux of a problem for linguists, a problem that may act as a constraint. Chambers' warning against assigning any prestige to those speakers who maintain standards provides a clue to the constraint; to delineate standards, especially if one assigns prestige to those who maintain them, is to seem to prescribe rather than to describe, yet modern linguistics is a descriptive rather than a prescriptive study. In general, linguists no longer concern themselves with the distinction between prescription and description; few would see the development of descriptive linguistics as a problem. However, in some areas of study, avoidance of prescription may constrain the description. Linguists are in the difficult position of having to be neutrally descriptive, avoiding all prescription, even when describing something that is not socially neutral. As Sankoff puts it, Perhaps the major task of sociolinguistics is to reconcile the essentially neutral, or arbitrary, nature of linguistic difference and of linguistic change, with the social stratification of languages and levels of speech unmistakeable in any complex speech community. (Sankoff 1976, cited in Hudson 1980:199) While scientific methodology and developing technology are doing much to facilitate such reconciliation, paradoxically, the emphasis in modern linguistics on scientific objectivity makes this task of the linguist difficult through obscuring some problems; in those areas which are the least neutral, such as the identification of standard and prestige, it is still unclear, in some instances, whether linguists are finding that something which could be called standard or prestige does not exist or whether they are simply constrained by the unacceptability of prescription in recognizing or identifying these. 37 The association between language standards and prescriptivism is clear, as Warkentyne explains: promulgation of language standards is inevitably associated with some form of prescriptivism. Linguists, especially in North America, are understandably chary of dealing with prescriptive norms largely because of the stigma they attach to non-standard speech. Imposing linguistic norms may also be seen as an attempt to interfere with the natural evolution of language, hence slogans like 'leave your language alone.1 Furthermore, linguists have tended to believe that prescriptive norms have no significant impact upon language usage in the community. However recent sociolinguistic studies, on hypercorrection for example, show that language norms have a strong influence on general usage and should therefore not be lightly dismissed. (Warkentyne 1986a: 170) Here, then, is further difficulty for the sociolinguist; the association of language standards with prescriptivism constrains the consideration and description of standard English. Although some linguists (Pringle [1986], for example) rigourously refuse to identify language standards out of some sense of democracy and neutrality, others attempt to discover acceptable ways to describe and delineate the standard. Gregg goes so far as to suggest that: we cannot advise a speaker to leave his/her language alone. Our advice might result in putting barriers in the way of his/her advancement. The lesson of Pygmalion (or My Fair Lady ) is surely that, in order to be accepted as an educated person or to move freely at a higher social level, Eliza Doolittle had to modify her dialect very considerably. (Gregg 1986a: 167) However, he makes the suggestion based on his argument that there is now an acceptable means of determining the standards. He claims that linguists must meet "the task of elaborating a method for deciding which variable should be chosen as the standard or, if we wish to recommend more than one, we must 38 determine how the chosen variants should be arranged in the priority list" (Gregg 1986a: 163) and recommends "precise statistical breakdowns" as the only means or basis of "speak[ing] with authority on the prevalence of a given construction in different segments of the Canadian population. Gut reaction and other intuitions are not a good enough base on which to build an acceptable Standard" (Gregg 1986a: 168). Gregg recommends "a series of urban surveys along socio-linguistic lines—linguistic probes we might call them— whose aim is to elicit the specific information required for making judgments about usage in various parts of the country" and argues that: only on the basis of precise information from such 'probing' surveys...will it ultimately be possible to reach valid conclusions about...preferred forms (based on the consensus of overall..majority), as well as prestige forms (used by most of the top status group), over against stigmatized forms (to be eliminated from their speech by those concerned with upward social mobility). (Gregg 1986a: 167) What Gregg is suggesting, then, is that the specification (and, by extension, prescription) of the norms be made scientific and, by implication, neutrally descriptive. He recommends that linguists analyze and describe language use, including standard usage, to produce the guidelines which delineate the bounds of acceptability for a given situation. Gregg's recommendation does much to diffuse some of the reaction against earlier prescriptivism, that which simply imposed rules on speakers; it is less offensive to describe language as people use it and produce a set of guidelines on that basis than it is to produce a set of guidelines based on some preconceived notion of standard language. However, the recommendation does not eliminate the problem of there being no necessary connection between the standard and standards; there is a kind of circularity in the 39 recommendation to identify the standard delineating the norms that produce such a standard without considering just how the standard might be identified. In the case of prestige (in the sense of that variety used by the highest socio-economic groups and emulated by others), a careful sociolinguistic description would produce objective guidelines; the referent of prestige identifies the source of that prestige. However, in the case of standard, which does not have such a clearly defined referent that can be analyzed, a careful analysis of language in use in order to produce guidelines is circular because the source of the norms would remain unclear. Gregg's suggestion, then, is valid for prestige and certainly does much to reduce the reaction against prescriptivism; however, unless standard can be equated with a particular socio-economic group or some equally identifiable variable, the suggestion is inadequate for delineating standards. A good beginning has been made, but the gap between standards (which can be specified) and standard (which may only be a concept) has not actually been closed because the source of the standards remains unspecified or vague. It seems, then, that the difficulty of being neutral and avoiding prescription when attempting to describe standard and prestige affects both the description and the discussion of these matters. One must be careful to consider, for example, whether or not a statement argues against the existence of a prestige or standard or is an attempt to be neutral, or avoid prescription. The constraint of prescriptivism, then, can obscure the investigation and identification of standard and prestige; arguments about the existence of standard and prestige may also be constrained. Not all of the disagreement 40 among linguists about the existence of standard and prestige, then, is necessarily as clear cut as it may seem to be on first consideration. 2.1.3 Social and Regional Dialects and the Development of Standard and Prestige in Canadian English There is less discussion by Canadian scholars about the relationship between standard and regional varieties and about the development of standard than there is by scholars in general, perhaps because of the unusual homogeneity in Canadian English4. However, there is some comment but little agreement. Chambers identifies "the standard accent [as] the one that is regionless rather than regional" (Chambers 1986:11), thus implying that the standard accent is social rather than regional. Other linguists recognize the similarities and convergence of the two, though the relationship between regional and social dialects is disputed. Gregg (1986) regards standard as being nonregional (or perhaps trans-regional), distinguishing the shared elements, the standard, but allowing for different regional manifestations of that generally shared base. Thus, for Gregg, there are regional varieties of standard; he uses prestige to refer to a class dialect (i.e. that used by upper class speakers) thus keeping the senses of standard and prestige more or less explicitly distinct but allowing for interaction of social and regional elements. As in the general context, the adoption of a socio-dialectal model in Canadian dialect studies indicates the extent to which Canadian scholars recognize the interrelation between social and regional elements in variation. 4 Canadian English is characterized by a degree of homogeneity across much of the nation unsurpassed in other nations. See note 7, this chapter. 41 The Canadian discussion of (or assumptions about) the development of standard or prestige also differs slightly from the general discussion; there is less comment and disagreement about this development. It is the norm among Canadians to reflect Sapir's observation that "the process of standardization take the form of a democratic rather than an aristocratic process" (Sapir 1931:126). Chambers, working in the Canadian context, argues that standard in its marked sense in which something is imposed on people by some authority, is "alien to—perhaps even repugnant to—our national character" (Chambers 1986:3) and Pringle agrees, claiming that even the existence of a standard is "deeply at odds with the nature of Canadian society" (Pringle 1986:35). However, once again, there is a sense in which the discussion is more theoretical or definitional than an actual consideration of how standard or prestige develops. Chambers and Pringle are objecting to prescription or imposition of a standard; they are not necessarily arguing that standard is not imposed but rather that it ought not be imposed. On the other hand, definitions such as that given by Southerland suggest that some kind of hierarchic order or even imposition of a standard is possible; he writes: In modern, developed societies there is one language variety that stands above the others. This superposed variety is usually called the standard language; it is the variety that is employed by the government and communications media, is used and taught in educational institutions, and is the main or only written language...It is to the written standard that prescriptivists, those who seek to regulate how others use language, usually appeal when they condemn some usage as incorrect, improper, or even barbarous. (Southerland 1987:355) It is not clear, then, whether the disagreement is about how a standard actually develops (or has developed) or about the means such development ought to take. It seems that some Canadian scholars are reacting more to 42 prescriptivism than their counterparts in other nations. Perhaps the recent establishment of the Strathy Language Unit at Queen's University^ has prompted a reaction against prescriptivism and is colouring discussion of dialect development in Canada; Strathy's request that the unit publish a guide to written and spoken communication seems to have been taken by some scholars as an attempt to impose standards or to interfere with language development; there is, therefore, a small resurgence of opposition to prescriptivism among some Canadian scholars. This unusual situation in Canada brings to light an effect of the difficulty of reconciling the essentially neutral nature of language with the hierarchic nature of society and language use. Recognition that reactions against prescriptivism can affect discussions of the development of standard reveals that the consideration of whether standard is imposed or emergent is complex and is a delicate topic for linguists; it seems that more consideration is necessary, particularly of the nature and the source of standard as they may address the question of how a standard evolves, is imposed or both. It is clear that the question of how a standard develops is far from settled and that prescriptivism (or reactions against it) is something to be reckoned with albeit differently than it once was.6 5 "The Strathy Language Unit was founded in 1981 by a bequest from John Richard Strathy. Mr . Strathy left to Queen's University the annual income from a trust; he specified in his wi l l that the income was to be used to establish in the English Department of Queen's University a unit for the continuing study of English usage within Canada. That Unit was to stimulate interest in Canadian English usage and to publish successive editions of a guide to written and spoken communication" (Lougheed 1986:190). 6 Whereas early discussions of and reactions against prescriptivism focussed on the need for descriptivism, current discussions must consider the possibility that reactions against prescriptivism may masquerade as 43 2.1.4 Linguistic Levels as Delineators of Prestige and Standard in Canadian English Canadian dialectologists recognize that there is variation on all linguistic levels and tend not to use linguistic level as a delineator of prestige and standard; however, some Canadians do suggest that there is some relationship among linguistic level and standard and prestige. Chambers (1986) identifies accent as being more varied than what he calls dialect (i.e. grammar and lexicon) and Kinloch (1986) argues that although there is little likelihood of there being a standard pronunciation, there is more agreement than disagreement among educated speakers about other aspects of language (morphology, syntax and vocabulary). Chambers and Kinloch, however, are focussing on differences in pronunciation within Canadian English; when the focus is changed to describing the ways in which Canadian English can be seen to be distinct from other varieties such as British and American English, then it becomes clear that Canadian English is characterized by a degree of homogeneity not found in other national varieties of English.7 Such arguments about the development and existence of standard and prestige. Hence, the effects of reactions against prescriptivism cannot be ignored in any consideration of standard and prestige, especially as they may form assumptions underlying any argument. 7 Two examples of comments on and descriptions of the homogeneity of Canadian English are the following: Wells says that "Canadian pronunciation is extremely homogeneous, considering the vastness of the territory over which it extends and the varied history of settlement. From Ottawa to Vancouver is more than 3000 kilometres; their accents are virtually the same. Not only is there very little geographical variation, but what social variation there is in accent seems to be confined to trivial details of lexical incidence"(Wells 1982:491) and Avis identifies General Canadian as being the dialect spoken from Toronto west (Avis 1975). 44 homogeneity accounts for the tendency of Canadians not to use linguistic level as a distinguisher of standard and prestige when other linguists do. It is evident, then, that the context in which something is described determines, to some extent, how terms are used and how distinctions are made. In general, the discussions of prestige and standard in the Canadian context reveal the same areas of confusion and disagreement as those identified in the general discussion; it is, therefore, a significant beginning to use Canadian data for testing definitions of these terms. Generalization from the findings is possible only to the extent of proposing hypotheses for further studies on other varieties. Information gleaned from Canadian discussions of standard and prestige expands understanding of the nature of the problems in prestige and standard. In particular, the Canadian context reveals that there are two considerations that cannot be overlooked while studying standard and prestige or while reconsidering those areas of disagreement that have not been fully explained; first, that attitudes toward prescriptivism are factors in descriptions and discussions of standard and prestige, and, second, that, to some extent, the difference between searching for a concept of something and searching for the linguistic realization of that concept may be important in the study of standard and prestige. These two points, then, must be considered along with cultural differences as possibly contributing to confusion about standard and prestige. While identifying these issues, the consideration of standard and prestige in the Canadian context, for the most part, confirms that cultural differences account for variant uses of the terms standard and prestige, account for disagreement about the validity of distinguishing standard from prestige on the basis of linguistic level and partially account for disagreement about the existence and development of prestige and standard. Although the cultural differences cannot fully explain the latter two areas of disagreement, those areas are not as central to defining standard and prestige as the former two are. It is, of course, important that possible cultural differences be kept in mind while defining the terms and that those areas of disagreement that have not been fully explained be reconsidered after analysis of available data; however, it is possible, on the basis of both the survey of the use of terms and the understanding provided by the consideration of cultural differences as they affect use of the terms, to construct working definitions of the terms and thus proceed with a consideration of prestige and standard. 2.2 Definitions of the Terms Further consideration of standard and prestige requires some analysis of available data in order that a practical understanding can be added to the discussion. However, before an analysis of standard and prestige in a particular context is possible, it is necessary to formulate working definitions of the terms. This, in turn, requires determining the essential distinctions among several terms. 2.2.1 Essential Distinctions between Terms The consideration of the use of the terms standard and prestige in the literature reveals that they interrelate with several senses of other terms. These must be dinstinguished to clarify central issues in sociolinguistic research. Standard must be distinguished from prestige, and standard from 46 standards: prestige must be distinguished from prestigious, and norm from norms. Moreover, attention must be paid to some of the differences between British and North American contexts in order to make these various distinctions. Clearly, standard and prestige share some important features; both are approved models, neither is stigmatized. These shared features account for much of the use of standard in the place of prestige: if there is no clear prestige variety or accent such as RP is in Britain, then it is possible that the standard functions in some ways as RP does in Britain. Furthermore, because both standard and prestige are approved models, both carry some kind of prestige in its most general sense. Prestige is essential in language variation and change. As McDavid writes; "The last force—prestige or lack of it—is the essential cause of the development of social dialects" (McDavid 1969:80). Labov's work, which illustrates Sturtevant's claim that "before a phoneme can spread from word to word...it is necessary that one of the two rivals shall acquire some sort of prestige" (Sturtevant 1947:74, cited Labov 1972:3), demonstrates clearly that prestige (either overt or covert) is an essential motivator in language variation and change (see Labov 1966, 1972 passim ). The claim, however, is not that a prestige accent such as RP must exist and be the motivator of variation and change, but rather, that prestige be assigned to particular forms such that variation and change is motivated. It may well be the case that the standard forms have been assigned the necessary prestige to motivate variation and change in some (or even many) instances. In these cases, then, standard would be functioning as the prestige (RP) does. Thus, standard and prestige share a function in language variation and 47 development. Both are also widely used by speakers of the highest socio-economic group; however, standard is widely used by many other socio-economic groups as well. However, it soon becomes clear that standard and prestige, especially if defined carefully, are not the same; it seems that they have fewer shared features than non-shared features. One cannot ignore the identification of RP as the model for prestige varieties or accents; taking RP as the prototypical prestige accent, one must define prestige as it is distinct from standard, determining the semantic features which delineate prestige. Prestige can be correlated with the upper socio-economic group and can be said to be emulated by speakers of lower socio-economic groups. Because of the correlation between prestige and the highest socio-economic groups, prestige can be said to be used to distinguish membership in the highest socio-economic group; thus, it indicates an elite group membership. Descriptions of prestige as the indicator of an elite group membership focus on the ways in which it is distinctive from the speech of other groups. Its association with the highest socio-economic group also restricts the number of native speakers of RP considerably. Because RP is an accent, it can also be said that prestige is associated more with accent than with grammar or lexicon; it can also be said that it is a feature of speech more than of writing. Furthermore, because members of lower socio-economic groups emulate the prestige speech of the upper socio-economic groups, prestige can be said to be a motivator of language variation and change. 48 On the other hand, standard is used by educated speakers of most socio-economic classes. Perhaps the clearest model of standard is Standard Written English; its spoken counterpart is learned and used by people from across the socio-economic spectrum. The use of standard distinguishes general rather than elite group membership and descriptions of the standard focus on the common or shared aspects of the language rather than on the ways in which that variety may be distinct from others. As Gregg writes, all regional varieties of English must share a great deal in common. It is these shared elements which contribute to what might be called the Standard for World English. Likewise shared elements within the sub-regional varieties can contribute to separate standards for each of the regional varieties. This regional manifestation of a generally shared base within English-speaking Canada is the goal of our endeavours in the search for the Standard in Canadian English. (Gregg 1986a: 161) Because of the commonality of standard English, both in terms of its features and its use across the socio-economic stratum, it has many native speakers. Its close association with Standard Written English, indeed its development from what has been called literary English, makes it more a matter of grammar and lexicon than of pronunciation. It is, however, common to both writing and speech. Standard Written English is learned rather than acquired or emulated; its spoken counterpart, when not acquired, may be learned along with the written standard. The close association with an established, relatively fixed literary language makes standard English a control of change rather than a motivator of change; as people learn the standard, their other (native) varieties are controlled from developing further and further away from the standard. Thus, whereas prestige serves to motivate change by distinguishing a group of people whom others emulate, standard serves to control change by embodying those aspects of varieties that are shared;8 learning Standard Written English unites speakers in a common ground rather than distinguishes some from the others. The semantic features, then, that are essential to distinguishing standard and prestige are as follows: +INDICATES GENERAL GROUP MEMBERSHIP +INDICATES ELITE GROUP MEMBERSHIP It is important to keep in mind that, rather than being a complete list of the semantic features of standard and prestige, the above includes only those features that the preceding discussion has determined to be necessary in accounting for the inconsistencies and disagreement found in the literature. The features are based on accepting RP as the model for prestige and Standard Written English as the model for standard and are identified in order to clarify the distinctions between standard and prestige. Therefore, such things as the possibility of standard being emulated in similar ways to prestige are not included in this list of features; rather, the most distinct instances of standard and prestige are delineated here. Standard +APPROVED MODEL -STIGMATIZED +SHARED FEATURES +GRAMMAR AND LEXICON +WRITING AND SPEECH +SPEAKERS OF MOST CLASSES +FOR GENERAL USE Prestige +APPROVED MODEL -STIGMATIZED +DISTINCT FEATURES +ACCENT +SPEECH +SPEAKERS OF ONLY THE TOP CLASS +FOR SPECIAL USE +MANY NATIVE SPEAKERS +CONTROLS CHANGE +LEARNED +FEW NATIVE SPEAKERS +MOTIVATES CHANGE +EMULATED 8 See Kroch (1978) for a good discussion of the role of standard in inhibiting language change. 50 The consideration of the similarities and differences between standard and prestige reveals the importance of distinguishing prestige from prestigious. There are two main reasons for distinguishing prestige from prestigious. First, as the above discussion suggests, the prestige variety or forms (as in those used by the highest socio-economic group) are not the only ones that are approved or positively evaluated by speakers (i.e. are prestigious). For example, standard English, specially as it is distinct from nonstandard English, is positively evaluated by many speakers; therefore, it carries prestige in the general sense, or is prestigious. It is not clear, however, that standard is emulated per se because standard is taught; learning a variety or form is not quite the same thing as emulating someone else's speech or variety. Thus standard, though prestigious, is not necessarily emulated in the way that the prestige (as in RP) is thought to be. Other varieties of English may also be prestigious in particular communities or situations. Prestige in its general sense is more relative than absolute; there may be degrees of prestigiousness within any speech community. The extent to which a prestigious form or variety is evaluated positively may determine, in part,9 whether or not that form will be emulated. It could even be the case that a prestigious form would not necessarily be emulated at all, though it is clear that a form must be prestigious before it will be emulated. Thus, it is clear that prestige in its general sense of positively evaluated is not the same as prestige in its specific 9 Other factors such as the appropriateness of the prestigious form to a particular situation, the speaker's role in the situation and the speaker's own intention also affect emulation. See, for example, Labov's identification of overt and covert prestige (1966, 1972), Brown and Fraser's identification of the importance of situation to variation (in Scherer and Giles 1979) and Ryan and Giles' identification of the relative importance of marking individuality or group membership in different situations (Ryan and Giles 1982), all of which reveal the complexity of motivation for language variation and change. 51 sense of that variety used by speakers of the highest socio-economic group and emulated by others. The former, which I would prefer to call prestigious, is a relative sense of prestige, is determined by the evaluation of forms by speakers and does not necessarily prompt emulation; the latter is an absolute sense, determined by the socio-economic status of the speaker10 and does prompt emulation. Keeping the two senses distinct (or using prestige and prestigious more precisely) may lead to a more complete understanding of the function of prestige in language variation. Second, it may be the case that the most highly evaluated variety in a speech community does not correspond to the variety used by the highest socio-economic group, as seems to be the case in the Canadian context. Gregg (1984) Pringle (1985) and Warkentyne (1986) have all found that British English is highly evaluated by Canadian speakers; their works reveal that British English is more prestigious than either Canadian or American English. On the basis of these findings, British English is even referred to as the prestige dialect in Canada. However, British English is not the variety used by the highest socio-economic group in Canada, nor is its use limited by socio-economic status. Furthermore, as Warkentyne notes, "the prestige dialect [BE] 1 0 Note that there can be some discrepancy between the linguists' use of socio-economic status as a delineator of prestige speech and speakers' evaluation of forms as being prestigious; it generally follows that the speech of the highest socio-economic group provides the model for the speech of others (i.e. is the prestige variety) but that is not necessarily the case. Thus, in contexts in which the speech of the highest socio-economic group is not necessarily distinct from that of other groups or is not necessarily the most highly evaluated, it might seem that linguists' use of socio-economic status as a determiner of prestige variety is at odds with the speakers' evaluation of the language. However, the linguists' method does hold for most speech communities and in most situations; hence it is generally suitable for sociolinguistic evaluation. 52 is so distant and vaguely understood that it is not a force for linguistic change in Canada" (Warkentyne 1986a:172). He correctly recognizes that Canadians do not generally emulate British English, however prestigious that variety may be. Thus, on two counts the positively evaluated variety, British English, does not serve as the prestige in Canada: first, its use does not correlate with the highest socio-economic group in Canada, and second, it is not emulated by Canadian speakers. It is necessary, then, to distinguish prestige from prestigious to adequately account for language variation and change. Such a distinction allows for the absolute case of the prototypical prestige variety (and, consequently, the use of socio-economic status and emulation as delimiters of prestige) and the relative sense of prestigiousness, both of which are evident in the complexity of language variation and change. Prestige, then, is the specific and practical term, essential in describing and analyzing one aspect of sociolinguistic variation; prestigious is the more general term, essential in accounting for the complexity of language variation and of various speech communities. Another distinction that is essential to a clear understanding and explanation of language variation and change is that between standard and standards. As mentioned earlier, standard generally refers to a variety of language that is used in the writing of English and in conducting the business of the nation; it is more common than uncommon and, as Baker points out, it is characterized by being "a form of English that does not distract from the message...does not draw attention to misspelling, faulty diction, regionalisms, slang, etc., rather than to the message... does not...create noise" (Baker 1986:153). Clearly, the 53 notion of standard is quite imprecise. However, as Hudson points out, it is the case that standard will have passed through the processes of selection, codification, elaboration of function and acceptance (Hudson 1980:32). Selection refers to a process by which a variety emerges to be developed into the generally agreed upon model. Codification refers to the production of grammar books and dictionaries as references for the identification of the correct "standard" forms. Elaboration of function refers to the development of the standard for use in conducting the business of the nation; for example, legal and technical vocabularies must be developed so that the standard may function as the vehicle of legislation. Acceptance refers to the recognition of the variety as being the agreed-upon standard for the relevant speech community. Although there are several points of disagreement surrounding the development of standard,11 the four processes Hudson identifies do provide a means of delineating standard language in a more precise way than other definitions have done. What becomes evident is that establishment and development of a standard variety involves deliberate intervention (some would suggest interference) by speakers of the language (usually educators) in such a way that forms are fixed, that variation and change within the standard is minimized. Bounds of acceptability are specified, usually in terms 1 1 For example, not all linguists agree that standard should extend beyond grammar and lexicon to include phonology and there is little agreement about the value and adviseability of standardization. As has been illustrated in this chapter, there is also little agreement about how a standard might develop and about whether or not prestige should be attached to the standard. 54 of correctness.12 What emerges is a set of rules that acount for standard language; these rules are often called language standards or norms. Standards, then, refers to the set of rules that, if applied, will produce those forms that typify a standard variety. Standardization, that process by which forms or varieties are selected, codified and accepted as the standard, is a strong force in language variation and change. It is likely the case that only a select few (such as educators, publishers, writers and, perhaps, legislators, for example) concern themselves with standardization and with both identifying the standard of a particular community and specifying the standards that produce that standard;13 nonetheless, the process of standardization, particularly in nations in which there is widespread public education, exerts a strong controlling force on language variation and change. One cannot, therefore, disregard either standard or standards in language variation and change. However, one ought to maintain a distinction between the two, recognizing that, as interdependent as the two are, they are not the same. 1 2 There have, of course, been developments in this area, particularly in the last few decades, as linguists have come to recognize that correctness is, to some extent, culture and situation specific. As a result more than correctness is taken into account when acceptability is specified and it is recognized that there is some variation within standard. However, correctness, or some notion of correctness, continues to be considered an attribute of standard language. 1 3 Note that the conference "In Search of the Standard in Canadian English" gathered together teachers, linguists and editors, all of whom are concerned about delineating the standard for Canadian English; it is not clear, however, whether they are determined to describe or quantify Standard Canadian English or they are determined to provide standards for use in Canada (See Lougheed 1986). I suspect that there were elements of both being considered, with the emphasis on standard or standards being determined by the particular interests of the individual. 55 Another term for the rules governing speech is norms: according to Hudson: Skill in speaking depends on a variety of factors, including a knowledge of the relevant rules governing speech...We shall call such rules N O R M S because they define normal behaviour for the society concerned, without being associated with any specific sanctions against those who do not follow them. (Hudson 1980:116) His choice of terms is clear and indicates that standard speech has a normative function, that it does exert a controlling force in language variation and change. However, his use of norms also points to the importance of distinguishing norms from norm. As he points out, the term norms is used because standard is seen as "normal" usage, presumably in a community as a whole. However, there are subcommunities with their own varieties quite distinct from standard; in these subcommunities, something other than standard is "normal" or forms the norm. N o r m , then, refers to that which is typical of, or most common in, a speech community; Gregg calls these preferred f o r m s 1 4 or "those used by the majority of speakers" (Gregg 1984:A3). Only in a community in which standard is the most widely used variety of the language will the norm be that standard. Norms (standards), then, only produce the norm in communities which use standard English; in other communities the norm may be quite different from standard. Norms, then, must not be confused with the norm or be expected to produce that which is commonly used (or normal) in all speech communities. The use of terms suggestive of normal is unfortunate when describing something as 1 4 Note that care must also be taken with this term as others may use it in the sense of prestige or those forms thought to be better than others. 56 diverse as language variation; it is clear, therefore, that care must be taken with the terms. Thus, it is important to distinguish between norms and norm as well as between standards and standard. It is also helpful to distinguish prestige from prestigious and is essential to distinguish standard from prestige. Without identifying and maintaining such distinctions, one cannot clearly assess such things as the existence and development of prestige and standard in speech communities 2.2.2 Working Definitions The survey of the literature and consideration of various terminiological distinctions provide important criteria which can now be included in working definitions for an investigation of prestige and standard in language variety. The discussion reveals that the essential distinctions between prestige and standard pertain to socio-economic status of the speakers of each, emulation of prestige forms as opposed to learning standard forms, and codification, which imparts a notion of correctness to standard. To test the applicability and centrality of these features of standard and prestige, the following definitions will be adopted and tested against the data collected in the Survey of Vancouver English: prestige - that variety (or those forms) used by the highest socio-economic group and emulated by others. standard - that variety (or those forms) used by the majority of speakers and embodying notions of correctness. In the discussion of the findings in the Survey of Vancouver English, prestigious will be used to refer to forms or varieties that are positively 57 evaluated by speakers but which are not necessarily used by the highest socio-economic group or necessarily emulated. Standards will be used to refer to the set of rules (or norms) which produce standard English; no distinction will be made between standards which have been explicitly specified in grammar books or other reference works and those which are implicitly derived from the standard but have not necessarily been set down. Norms, as a synonym of standards, will be avoided; norm, as those forms which typify a particular speech community, will also be avoided where possible. These definitions are neither exhaustive nor necessarily the most appropriate in all situations and for all purposes; however, they (and the distinctions implicit in them) have been motivated from discussions of previous work and are appropriate for this Canadian context. They will facilitate both assessment of the definitions of standard and prestige and expansion of the description of Canadian English. They will, therefore, be used throughout the remainder of this work. 58 CHAPTER 3 PRESTIGE IN THE SURVEY OF VANCOUVER ENGLISH Analyzing the database of the Survey of Vancouver English for the existence of prestige as it has been defined here allows both testing of the definition and further quantification of the linguistic behaviour of Vancouver speakers. The two aspects of the definition must be considered separately; first, the forms used by the highest socio-economic group will be compared with those used by the majority of speakers to test for the existence of prestige in the sense of "those forms used by speakers of the highest socio-economic group;" second, changes in forms across speech styles will be examined for evidence of emulation by others of the forms used by speakers of the highest socio-economic group. To determine whether or not prestige forms may be developing or emerging in Vancouver English, the relevant data will also be tested using an analysis of change in apparent time. The results of the tests can then be used to reassess both the definition of prestige as "that variety or those forms used by speakers of the highest socio-economic group and emulated by others" and the existence of prestige in Vancouver English. 3.1 The Survey of Vancouver English 3.1.1 Description of the Survey Before testing for prestige, it is essential to consider what the available database comprises and how it can be used for the purpose of analyzing prestige. The Survey of Vancouver English (SVEN) was conducted by R.J. 59 Gregg from 1976 to 1984 using a questionnaire1 comprising a battery of over one thousand questions aimed at determining phonological, lexical and syntactic variables that characterize Vancouver English. The questionnaire was designed to evoke a full range of speech styles and was administered to three hundred informants from the Greater Vancouver area including North and West Vancouver as well as Vancouver itself, and extending east to Lynn Valley and Coquitlam, south to White Rock, and north west to Lion's Bay. The informants were all native English speakers and comprised equal numbers of male and female speakers, equal numbers in each of three age brackets—old, middle, and young-and equal numbers in each of four socio-economic classes-upper and lower working class and upper and lower middle class as calculated on the Blishen scale. Because the overall goal was to assess Canadian English, informants had to be Canadian; 43% of the informants were third generation Canadian or more, 26% were second generation, 24% were first and only 7.1% were born outside of Canada.2 Gregg used a standard sociolinguistic model, fashioned after that used by Labov in New York City in 1966 and refined by Trudgill in Norwich in 1974; however, unlike Labov, whose survey was conducted in the Inner City and included ethnic variables, and unlike Trudgill, whose survey included only a small group of informants, Gregg included 300 Canadian informants from a large urban area in his survey. 1 See Appendix A for an outline of the contents of the SVEN questionnaire. 2 There are two methods of identifying first generation Canadians: either the generation to immigrate is considered to be first generation Canadian or the offspring of the generation to immigrate are considered to be the first generation Canadians. The Vancouver Survey uses the latter method of calculating; thus, the first generation born in Canada rather than the first to arrive in Canada is considered to be first generation Canadian. 60 After Gregg and his team designed the questionnaire, his fieldworker interviewed sixty of the informants—the Pilot group—from 1976 to 1978. During that time, Howard Woods, a graduate student of Gregg's, was working on a similar study in Ottawa. Gregg and Woods attempted to make their surveys easily comparable so as to contribute more fully to Canadian dialect information, especially so as to establish the degree of uniformity that exists within a wide region such as Canada and to discover smaller (provincial) regional preferences as well as urban preferences. To quantify language attitudes and awareness and to complement Woods' study, Gregg added a section, "Subjective Attitudes and Language Awareness," to his pilot study before proceeding. After designing the new section and slightly modifying the original questionnaire, Gregg's fieldworker interviewed the other 240 informants. Gregg's team then transcribed the recordings of the interviews, coded the information for the computer, using the SPSSX programme, and stored the entire database in the computer at The University of British Columbia. Because SVEN was designed to test variation on all linguistic levels, the data can be divided into lexical, grammatical and phonological groups with further subdivision of the phonological group into three categories: those "involving a systematic change of a given sound in a particular position in words generally" as in the voicing of intervocalic dental stops such as /t/ -> /d/ in butter, "potential rhyme words" or "what was potentially the same sound in two neighbouring words" such as the diphthongs in without a doubt and the consonant clusters in dental centre, and "special profile words" or "words in 61 which specific sounds may vary freely, and independently, creating a number of patterns" such as garage, which was found to have eighteen possible versions (Gregg 1984:Dl-3). For the purposes of checking the prestige forms, the first two categories of phonological variants (i.e. the systematic sound changes and the potential rhyme words) need not be separated because there are no significant differences in the patterns of the two; I will refer to those two subcategories combined as "phonological variants." The third subcategory, the special profile words, must be considered separately; hence, there are two phonological categories, the phonological variants and the special profile words. Both the phonological items and the special profile items appear in the data in five possible speech styles—minimal contrast, word list, series, visual-aural prompting and reading passage—the minimal contrast being the most formal or careful style and the reading passage being the least formal or most casual style. Although there was a more casual style elicited, that of spontaneous speech, I do not have access to that data in a form in which it can be analyzed and cannot, therefore, consider the patterns that appear in that style; however, because of the subject matter of the reading passage,^  the style elicited was considerably less formal than both what is usually expected in reading style and what was elicited by the other four methods, thus providing an adequate range of speech styles. Of the 53 phonological items 3 See Appendix B. The reading passage included very casual dialogue between characters in very ordinary situations and did not, therefore, prompt the careful style elicited by more formal passages. There has been some examination of the ways in which the method used in eliciting speech styles influences the results; for example, Woods 1979 and Milroy 1980 point to the problem of using reading tasks to elicit various stylistic levels. However, it is beyond the scope of this study to assess the effect of the methodology used to elicit speach styles. Here I can only acknowledge that the SVEN reading passage produced a less formal style than usual. For a consideration of Reading Style in SVEN see Rodman 1986. 62 considered, 5 had to be rejected because of computer programming errors leaving 48 items that could occur in each of the five styles, making a possible 240 occurrences. However, each phonological item did not occur in each speech style; the 48 items actually occurred in some of the five speech styles a total of 170 times. . The largest phonological group, the special profile words, included 84 items, 5 of which had to be rejected because there was so little agreement about the possible forms that the category "other" predominated. Since "other" does not reveal anything beyond the lack of agreement, there is little point in considering it further. The remaining 79 items occurred in some of the five different speech styles a total of 156 times. In an attempt to test all possibilities, occurrences of each phonological and special profile item in each style can be considered separately; hence the total number of occurrences of phonological items that can be considered is 326. The lexical and grammatical items were elicited most often by direct question, independently of speech style. The lexical items were gathered by two sections of the questionnaire, visual-aural prompting and local words. In visual-aural prompting, pictures 4 were shown to each informant, who was then asked to name an item or items in the picture. Although in general this method of eliciting responses is good, occasionally something about the picture may bias the responses slightly. For example, one picture (No. 282) is of a drawing similar to those that appear in school readers and children's books; it seems that the respondents reacted to the informality of the context in which the picture would be appropriate rather than to the item itself, a reaction which See Appendix C for copies of some samples of the visual-aural prompts. 63 indicates the need for consideration of appropriateness and congruity in all assessments of language use and language attitudes. Just as the content of the Reading Passage affected the style used for reading the passage, the context and particular medium depicted in the picture seems to have affected the responses elicited by the photographs.^ However, for the most part, the pictures or visual prompts used elicited a direct, spontaneous response. The local words section of the questionnaire, on the other hand, is less spontaneous and elicits information about words rather than documents the use of words. Here the informants were given words and asked what each term meant, whether they had ever heard anyone say the word and whether they themselves used it. The responses to the latter two questions are self-reporting only and may not be reliable. Unfortunately, there is nothing in the data that can be used to check the reliability of the self-reporting of familiarity with and use of the "local terms" examined. However, for the purposes of amassing a description of the English spoken by residents of the Lower Mainland of B.C. , the method of eliciting data is adequate. Although my particular purpose here is to test a relationship between socio-economic status and actual use rather than reported use, the patterns of responses to the local words section of the questionnaire will be considered as there is no reason to think that there would be any connection between socio-economic status and reliability of self-report ing; 6 all 51 lexical items will be considered with the proviso that the 12 5 The evidence revealing the importance of content to Reading Style and of media congruity in visual prompting aids is a significant, though inadvertant, contribution of S V E N . The consequent indication of the importance of appropriateness and congruity are important bonuses for advancement in sociolinguistic methodology. 6 See test of reliability of self-reporting of grammatical items below; no significance of socio-economic status is found. which are assessed by self-reporting be considered primarily for general patterns. Similarly, many of the responses to the grammatical items in the questionnaire must be regarded with caution. Of the 108 grammatical questions, 84 were elicited by questions asked in pairs, the first asking what the informant thought the "correct" form was, the other asking whether or not the informant used any form or forms other than the one judged "correct." Not only did all socio-economic groups tend to respond negatively to the second question in each pair, but also, the rate of denial of using any but forms they identified as being correct was exceptionally high, ranging from 72.4% to 100% (average - 90.7%). Such a high rate of agreement calls into question the reliability of self-reporting and indicates the need for further testing (see below). A further 18 variables were elicited through questions that required self-assessment; 9 sentences containing the word gotten and 9 sentences containing the utterance eh? were read and the informants were asked in which sentences they would use the utterances. For those sentences containing gotten to which the respondents answered "no," they were asked to specify whether they would say "got" or some other word instead of gotten. Informants were also asked to make comments about &h. Although it is very difficult to test the reliability of self-reporting, some test is necessary because it is unlikely that, for example, though most informants identify lay as the correct form in "Yesterday, he lay in the sun for 3 hours.", they "never" use laid in the same environment, or though most informants identify fewer as the correct form in the sentence "There are fewer people here today than yesterday.", they "never" use less in the same environment. Yet the relative 65 frequency of responding "never" to the question of whether people use forms other than fewer is 88.9% and the rate for Iay_ is 95.4%. 3.1.2 Reliability of Self-Reporting One would think that reliability of self-reporting would vary not only from speaker to speaker but also within a single speakers' own reported usage. To test reliability, I randomly selected 24 informants, being sure to have equal representation among socio-economic, age and gender groups. I then reviewed the grammatical items that were elicited by self-reporting to determine which were likely to occur in spontaneous conversation or speech within the interviews. Because many of the grammatical forms tested were past participial forms (has drunk, has got, for example) or simple past forms (dived, sneaked, for example) and the question used to elicit spontaneous speech (usually "If you won a lottery, what would you do with the money?") required either narrative present or future tense, many of the variables did not occur and could not be tested. It seemed most likely that the forms "there  are eight..." "She's a real nice..." "I don't have any money," "to Tom and me" and eh., which was also considered in the study as a grammatical item, would occur. I then listened to the spontaneous narrative, the comments about Vancouver and the personal information sections of the tapes of each of the twenty-four informants I had selected, listening for occurrences of the five grammatical variables. One difficulty of testing reliability is that for cases in which people report not using a particular form, negative evidence (i.e. the absence of those forms) is inconclusive. Hence, only unreliable reporting of non-usage can be proven. 66 One might think, however, that there would be plenty of evidence of use of e_h by Canadians (if it is the marker it is supposed to be) and that, since most people reported never using eh. or not using it in certain structures, occurrences would indicate unreliable reporting. However, eh. very rarely occurred in the interviews, probably because its most frequent use seems to be in eliciting information or encouraging a response rather than in reporting or responding. Hence, though the interviewer used eh quite frequently, the informants rarely did. However, there were 3 occurrences of eh. that showed reliability of self-reporting. One speaker who claimed to use eh. in all of the structures considered did use £h. in a structure like "Nice day, eh?" Another who claimed to use eh in four of the structures, also used it in a structure like "Nice day, eh?" A third who claimed to use eh. in the structure "It goes over here, eh?" did not use eh. in the interview proper, but in a phone conversation which interrupted the interview said "You'll be home tonight I guess, eh?" Thus, there were three examples of reliable reporting of the use of eh.. There was, however, one example (or 25 examples, depending on how one counts) of unreliable reporting of the use of e_h. One informant who claims that he would not use eh. in any of the structures given and that he avoids it used eji 25 times. His most frequent use was in the function of a rhetorical check for approval of the statement made (as in "Nice day, eh?"), but he also used it between sentences as a kind of link or filler. Thus the reliability of self-reporting of the use of eh. is difficult to test and varies from speaker to speaker, though here there is more evidence of reliable reporters than unreliable. There were 45 occurrences of the other grammatical forms used to test reliability; for those 45 occurrences, 7 speakers reported reliably, 3 reported 67 unreliably and 6 reported reliably in some instances and unreliably in others.7 Of the 6 who were sometimes reliable and sometimes not, 4 had an equal number of examples of unreliable and reliable reporting, but 2 had more reliable than unreliable; thus, there is indication that there is slightly more reliable than unreliable reporting. Though initial expectations that reliability varies both by speaker and by variable are confirmed, there is evidence of more reliability than unreliability in self-reporting. The advantages of using a questionnaire and self-reporting for eliciting particular forms in a database are also evident; many of the forms thought to be significant to a particular variety do not occur frequently in general conversation. Not only did only 45 instances of the four grammatical items occur among the 24 speakers tested, but also they occurred sporadically or not at all in individual speakers; 8 speakers did not use any of the forms, 7 used only 1 and none of the speakers used all 4 forms. Thus, not only is the questionnaire the only practical means of eliciting the particular items being investigated but also the self-reporting that questionnaires necessitate seems to be reasonably reliable. Furthermore, although there may be a connection between socio-economic status and the reliability of self-reporting, the Vancouver database does not seem to reveal such a connection. Table 3.1 shows the distribution by socio-economic status of the incidents of reliable and unreliable self-reporting. 7 There were 8 informants for whom there weren't any occurrences of the other variables tested, thus the total number of informants tested is 24. 68 SES I SES II SES III SES IV 8 re l iable 3 2 2 2 unre l i ab le 2 0 2 0 mixed 0 3 0 3 no data 1 1 2 1 Table 3.1 - Distribution of Reliable and Unreliable Reporting across Socio-economic Groups Although for the results to be conclusive more than 24 informants should be tested, the early indication is that reliability is not significantly affected by socio-economic status; there is no evidence of tendencies of reliability to increase or decrease with socio-economic status. Further testing of the reliability of self-reporting is indicated and should be undertaken when the spontaneous narrative section of the data becomes available in a form which will allow detailed analysis and cross-tabulations. However, this early evidence of self-reporting being reasonably reliable and not significantly affected by socio-economic status indicates that the data elicited by self-reporting can be included in the test of the existence of prestige forms in Vancouver English. Therefore, the 42 questions that ask whether or not an informant uses grammatical forms other than those that he perceives as being "correct" and the 19 items that involve self-reporting of the use of gotten and eh. will be included in the analysis to determine prestige. Four other grammatical items will be retained, though the method of eliciting the responses is unusual; these four items (anywayls l . out of. to/for, into') were included in the Reading Passage. It was thought that speakers might substitute the forms they normally use for the printed forms, especially since the tone of the passage 8 SES I is the lowest socio-economic group tested; SES IV, the highest. 69 was very casual. One other item can be retained as it is likely to be an accurate response to a Visual-Aural item (265, a picture of a cat "lying/laying" on a bed). Hence, the 42 questions about which form the informants think is "correct" and the Visual-Aural item make 43 items which are likely to be accurate responses, and a further 65 items that are of questionable accuracy but that can be retained because the unreliability of the responses affects all socio-economic groups equally make a total of 108 grammatical items. All together, then, 485 items (108 grammatical, 51 lexical and 326 phonological) can be used to assess the existence of prestige forms in the speech of Vancouverites. 3.2 Prestige in the Survey of Vancouver English To test the appropriateness of the definition of prestige as "the forms used by the highest socio-economic group and emulated by others" and to compare the findings in Vancouver with those in other studies, two aspects of prestige must be tested. First, the usage of the highest socio-economic group must be compared with usage of the other groups or with the usage of the majority of the speakers. Gregg, in his report to the Social Sciences and Humanities Research Council, suggests that the "prestige" forms, which he defines as "those [forms] used by the top socio-economic group who have the highest level of education" and the overall "preferred"9 forms, which he defines as 9To prevent terminological confusion, I will avoid the term preferred. I will use majority rather than preferred to refer to the forms chosen by the majority of speakers, leaving preferred to refer, when necessary, to the forms that are chosen by each socio-economic group. For example, if more of the Class IV speakers choose dived rather than dove, then I will call dived the form preferred by Class IV speakers. The term preferred is ambiguous, not only if it is used in the sense that Gregg gives it but also because others use it to refer to 70 "those [forms] used by the majority of the SVEN population," be compared (Gregg 1984:A3). If there is a clear prestige in Vancouver and it is that spoken by the top socio-economic group as has been seen to be the case in Britain and, slightly differently, in the U.S., then comparing the forms used by Class IV speakers with those used by the majority of the speakers should reveal more differences than similarities between the two groups. Second, the linguistic behaviour of lower socio-economic groups in different speech styles must be compared with the behaviour of the highest socio-economic group in order to determine whether or not lower groups are emulating the highest group. Labov, in his investigation of the "social stratification of English in New York City," based his evaluation of the variables on his finding that "in general, a variant that is used by most New Yorkers in formal styles is also the variant that is used most often in all styles by speakers who are ranked higher on an objective socio-economic scale" (Labov 1966:405). Because he found that the highest socio-economic group usually used one form in all speech styles and that the other socio-economic groups tended to use different forms in different styles, he could conclude that the highest socio-economic group was using and perhaps setting the prestige forms. Labov found further that there was some tendency for the second highest socio-economic group (the lower middle class) to hypercorrect in their attempt to emulate the style of the higher, prestige group, a tendency which he attributes to "profound linguistic insecurity...[which] is perhaps an inevitable accompaniment of a prestige or correct form. Here, where necessary, I aim to use it in a neutral and general sense, i.e. the form that most speakers of any given group choose, rather than in its marked sense of prestige or a specific sense of one group (i.e. the majority) of speakers. 71 social mobility and the development of upward social aspirations in terms of the socio-economic hierarchy" (Labov 1966:475). Clearly, then, from his findings it was logical for Labov to conclude that the prestige form is generally that used by the highest socio-economic group and emulated, often unsuccessfully especially in casual speech, by the lower socio-economic groups, and to develop explanations of what he termed "hypercorrection" on the basis of identifying such a prestige. To determine whether or not there is a comparable prestige evident in Vancouver English and to examine whether or not emulation is a central feature of prestige the terms used by the highest socio-economic group must be compared with those used by the majority and variation must be compared in different styles of speech. 3.2.1 Forms Used by the Highest Socio-economic Group To compare the responses of the highest socio-economic group with those used by the majority of speakers, I first calculated simple frequencies to determine which of the variants is that used by the majority of speakers. The frequencies for grammatical and lexical items simply reflect the percentage of speakers who use a particular form, whereas frequencies for phonological items reflect the percentage of possible occurrences of the form. Because there were several variants possible for most of the items tested, it may be the case that the highest rate of usage of any one of the possible variants of an item is low. For example, people identified eight possible meanings for the term slough: with so many possibilities, the figures are thinly spread and the rate at which the majority thinks slough refers to a man-made, run-off ditch is only 22.0%. However, as 22.0% is a higher rate of agreement than that for any of the other given meanings of the term s lough. "a man-made, run-off 72 ditch" is taken to be the sense of slough that is understood by the majority of speakers. Major i ty is, then, used in a relative rather than an absolute sense. What is being compared here are the forms preferred by the majority of the speakers and those used by each socio-economic group; hence the rate of use of a term is less important than which term is used. I then sorted the responses by socio-economic groups, with group I being the lowest and group IV being the highest on the Blishen scale, and calculated the frequency for each group. Thus, the forms preferred by the majority could be compared with those preferred by each socio-economic group. 3.2.1.1 General The comparison of forms used by the highest socio-economic group and those used by the majority of speakers suggests that there is no clear prestige among Vancouver speakers. The results of comparing the forms selected by each socio-economic group show that there is much more uniformity across the socio-economic groups, particularly for lexical, grammatical and phonological variants, than there would be if there were prestige forms or a prestige variety in the Canadian context (see Figure 3.1). 73 90 80 70 60 Rate of o u Uniformity 4 0 30 20 10 0 Lexicon Grammar Phonology Special Profile Figure 3.1 - Uniformity For 82.4% of the lexical items, 84.3% of the grammatical items and 89.4% of the phonological items, all four socio-economic groups and the majority of the respondents selected the same forms. 1 0 For the special profile words, the agreement among socio-economic groups about the forms used drops to 61.5%, which, though lower than for the other three kinds of variables, is still more than 50% uniformity among socio-economic groups. A partial cause of the difference in rates of uniformity between the special profile words and the other three groups is the nature of what is being tested. A socio-dialectal survey generally sets out to identify or quantify those items which typify a given dialect or sociolect; hence, the survey focuses on those items for which there is thought to be general or a high rate of agreement. In the case of S V E N , the items for which there is a high rate of agreement are 1 0 See Appendix D for lists of the forms chosen by most of the speakers in all socio-economic groups. 74 those in the lexical, grammatical and phonological categories; the special profile words are those which are noted for their high rate of variation in the geographical area being surveyed and are, therefore, those for which one would expect less general agreement. Furthermore, whereas phonological variables are isolated by one particular sound, such as intervocalic voicing in butter or the pronunciation of a low back vowel in father, the special profile words include several variables; hence, the possible combinations of variations increase and the general agreement decreases. However, with uniformity across classes ranging from 61.5% to 89.4%, both the existence of a prestige in Canadian English and the appropriateness of defining prestige as "those [forms] used by the top socio-economic group who have the highest level of education" in the Canadian context are questionable. However, to examine the possibility of some evidence for prestige forms existing or developing in Canada or, alternatively, for stigmatization of some forms being the trend in Canada, the patterns of deviance for those variables for which the socio-economic groups do not respond with any uniformity must be considered. To test for patterns, I isolated the items for which the preferred variants were not uniform across classes. I then identified which socio-economic group or groups deviated from the majority for each of the items and calculated the frequency that each group or groups deviated from the m a j o r i t y . 1 1 As Figure 3.2 reveals, overall, of the responses that show 1 1 Note that all grouping here is done on the basis of each variant tested; i.e. for some variables, a single socio-economic group used a form different from that used by the majority; for others two socio-economic groups used a form different from that used by the majority. Thus, calculations include the frequency that each single group deviated and the frequency that some pairs of socio-economic groups deviated. 75 differentiation by socio-economic group, the highest group (IV) is different more often than the other groups, deviating in 24.0% of the cases; the lowest socio-economic group follows, deviating in 19.2% of the cases and the lowest two socio-economic groups (I & II) are third, deviating together in 18.3% of the cases.12 Rate of Deviation Socio-economic Group Figure 3.2 - Variant Different from Majority-Overall Pattern To determine all possible trends, it is important to consider totals of and divisions among socio-economic groups as well as the individual patterns; hence, combinations must also be calculated. For example, to determine whether or not the highest two socio-economic groups behave differently 1 2 Figure 3.2 may be read, for example, "group IV chose a different form from that used by the majority and other groups in 24% of the cases for which there was no uniformity...groups II & I together chose a different form from that used by the majority and other groups in 18.3% of the cases." Figures 3.3-3.6 may be read in an analogous way. (The category "other" on these figures includes odd combinations of scoio-economic groups, such as I & IV or II & IV, for example, as well as several items for which there was no discernable pattern.) from the lower two groups, the rates at which group IV, group III and the two together chose a different form from the majority must be added and compared with the total of the rates at which group II, group I and the two together chose a different form from the majority. For 37.5% of the cases, either group IV, groups IV and III or group III (i.e. the highest two socio-economic groups) deviate, but for 45.2% of the non-uniform cases, either group I, groups I and II or group II (i.e. the lowest two socio-economic groups) deviate. (For 2.9% of the cases, there were no similarities in the responses of any of the socio-economic groups—usually because responses were more or less evenly divided between two favoured forms—and for 14.4% of the cases, groups IV and II or I and III or II and III or I and IV or II, III and IV behaved similarly; these are reflected in the category "other" in Figure 3.2.) Hence, it appears not only that there is some division between the highest and the lowest two socio-economic groups but also that, although the highest socio-economic group responds differently from the others and from the "majority" slightly more frequently than any other single group, the lower two groups are more often likely to respond differently from the "majority" than the higher two groups are. Thus, it seems, initially, that, although there is some evidence for a prestige form used by the highest socio-economic group in a small percentage of the responses, there is more evidence for different forms used by the lower two socio-economic groups, forms which may, therefore, be stigmatized. The patterns, however, can be examined more carefully by considering each of the linguistic levels separately. 77 3.2.1.2 Lexical Variables For lexical items there were only 9 forms (17.7%) for which the four socio-economic groups were not in agreement;13 speakers did not consistently agree about skookum (has used). 10:30. slough, drapes/curtains, squamish. skookum (meaning), verandah, train tracks, or dinner. Of the 9, both the highest and the lowest two socio-economic groups deviated from the "majority" in 44.4% of the cases. As Figure 3.3 indicates, group IV had the highest rate of deviation for a single group, deviating in 33.3% of the items for which there is no uniformity across styles. 4 0 1 Rate of Deviation IV Ml II I IV & III III & II II & I Other Socio-Economic Group Figure 3.3 - Variant Different from Majority-Lexicon However, it is clear from the figure that although the highest socio-economic group deviates from the other groups more often than any of the other groups deviates, it accounts for deviation in only one third of the items. Between 1 3 See Appendix E for a list of the forms preferred by the majority and those preferred by each socio-economic group. 78 them, the top two socio-economic groups use a form different from the other groups and the majority in their use of four lexical items. Similarly, the bottom two groups use a form different from the other groups and the majority in their use of four lexical items. The other grouping of socio-economic groups I and III is odd. Thus, for the lexical items, the clearest division is between the top and bottom two socio-economic groups. 3.2.1.3 Grammatical Variables For only 17 (15.7%) of the grammatical items was there variation among the four socio-economic groups in their choice of forms; speakers did not agree about the forms has drunk, sneaked/snuck. rea l lTyl . between you and me/I. m e  (object'), to whofml. gotltenl to try. Still here, eh?, gotltenl an answer. lend/ loan . gotTtenl braced, if...was/were warmer. It's me/I. these kindls l of  apples, between John and me/I. Have you got?, or gotTtenl k i l l e d . 1 4 The groups that differed from the majority most often in their choice of grammatical forms were the lowest two; as Figure 3.4 shows, groups I and II both differed 52.9% of the time, II differed 5.9% of the time and I differed 11.8% of the time, making a total of 70.6%, leaving only 29.3% of the cases for which the highest socio-economic group could conceivably be distinct from the majority or the other groups. On its own, the highest group differed in only 11.8% of the cases. 1 4 See Appendix E for a list of the forms chosen for each of these grammatical va r i ab l e s . 79 Rate of Deviation 60 50 40 30 20 II I IV & III II & II II & I Other Socio-economic Group Figure 3.4 - Variant Different from Majority-Grammatical Variables Clearly, the highest socio-economic group is not more likely than the other groups to be significantly different from the majority of speakers in their choice of grammatical forms. Group IV selected a form different from the majority of speakers for only 11.8% of the items and, together with group III for another 5.8%; oddly, it also differed with group II for 11.8% of the forms, making the total number of forms for which group IV, with or without another group, selected a different form just 29.3%. With groups I and II, together and individually, selecting forms different from the majority of speakers for 70.6% of the items, the trend here is clearly toward a division between the upper and lower two groups, with the lower groups differing more frequently than the upper. Thus, rather than prestige being indicated, stigma may be developing here. The fact that the lower two groups are most often different here may reflect the importance of education to grammatical usage. If nothing else, education 80 makes people more aware of "correct" grammatical forms. Martin Meissner, a sociologist at U.B.C., was using the S V E N data to investigate the effect of education on grammatical usage, hypothesizing that any connection between usage and some social factor is most likely to be through an education factor. His initial impressions upon beginning the analysis, however, were that there was no clearly significant pattern by education l eve l . 1 5 Because the socio-economic grouping of the informants includes an education factor, it may be the case that there is little difference between the results of analysis by socio-economic factor and by educational level. Meissner's observation was not that there was no pattern, but that the significance factor was so low as to render the analysis questionable. 1 6 In light of the logic of explaining the pattern by education factor rather than other social factors and of others' findings that education is significant to grammatical variation (see, for example Warkentyne 1986a), Meissner's dismissal of the relationship between Personal communication. 1 6 There is disagreement between sociologists and linguists about the necessity and value of tests of statistical significance. Labov, Trudgill and others do not rely on significance tests in their work; Labov writes: "It is immediately obvious to the sophisticated statistician that tests of significance are irrelevant...even if a particular case were below the level of significance, the convergence of so many independent events carries us to a level of confidence which is unknown in most social or psychological research" (1970). In keeping with the sociolinguistic tradition, I am putting less emphasis on tests of statistical significance than Meissner and other sociologists do; therefore, I use significant in the general sense unless specifically noted. (For those who are interested, the statistical significance of all analyses in this section of the work is minimal or even below the level of significance; this could be taken to support my observation that, in general, there is no clearly established prestige form in Vancouver English. However, it cannot, as Meissner claims, be taken to indicate that there is no relation between language and social status or education; there is too much evidence of correlation between language and social status in sociolinguistic studies and in general to deny the connection.) 81 education and grammatical variation is questionable and further analysis of the effect of education on grammatical variation is indicated and will be taken up in Chapter 4 of this work. At any rate, it cannot be concluded from either the grammatical or the lexical variables that the highest socio-economic group behaves significantly differently from the majority of speakers or from the other socio-economic groups. Rather, there is evidence that the lower two socio-economic groups are more likely to use grammatical forms that are different from those used by the majority of speakers.17 3.2.1.4 Phonological Variables All socio-economic groups did not choose the same phonological items for the following variants: /VtV/ (in butter, tomato...') /aerV/ (in married, vary...') fo / (in tomato, vase), /nju/, /tju/, /dju/ (in newspaper, student, duke...). <alm> (in palm, almond...). tot (in garage, album...). /V9r/ (intrusive 3 in near, mere. our...), /ns/ (in once, sense...) and /eri/ ( in raspberry, library...).18 Because each speech style is being considered separately here, deviance of variants in each style is considered as a single item. As speakers did not agree about several of the above phonological items in each of several different speech styles, the total number of phonological items for which all socio-economic groups did not choose the same form is 18 (10.6%). Of the 18, the highest socio-1 7 The difference in grammatical choices of the upper and lower two classes may also reflect the general sociolinguistic observation that grammatical variables tend to display a sharper socio-economic profile than do other types of variables, particularly phonological; however, the relationship between linguistic level and sharpness of the socio-economic profile neither explains the pattern nor provides evidence that the upper socio-economic group behaves differently than the other groups. 1 8 See Appendix E for a complete list of the variants chosen. economic group differs more frequently than the other groups, as Figure 3.5 reveals . Rate of Deviation IV III II I IV & III III & II II & I Other Socio-economic Group Figure 3.5 - Variant Different from Majority-Phonological Items For 33.3% of the cases, group IV chose differently from the majority and the other socio-economic groups; for a further 11.1% of the cases groups IV and III differed from the majority and for 5.6% group III differed from the majority. Thus, the highest two socio-economic groups deviated from the majority in 50% of the items for which there is socio-economic variation. However, the lowest two groups did not differ from the majority for the other half of the cases; for 5.6% of the cases, group I deviated and for a further 22.2%, groups I and II deviated. For the remaining 22.2% of the cases of phonological variation in which the four socio-economic groups did not agree, there were unusual combinations of socio-economic groups. For example, groups IV & I both preferred a different form from that preferred by the majority for 11.1% of the items, and groups IV and II both preferred a different form for 5.6% of the items. For another 5.6% of the items, group II was the only group to prefer the same form as that preferred by the majority. 83 It is evident, then, that although the highest socio-economic group differs from the majority in more of the cases than any other group and that the highest two groups differ for 50% of the cases, the upper two and lower two groups do not divide more or less evenly as they do for the lexical items. Nor do the lower two groups differ more often than the upper groups as they do for the grammatical items. Hence, it seems that linguistic level is a significant factor in variation by socio-economic status; the variation patterns are not the same for each linguistic level. Furthermore, there are combinations of socio-economic groups that indicate that the socio-economic groups do not behave as if there is a hierarchy intact. If the social hierarchy were clearly established in Canada, groups IV and I or IV and II would not be likely to demonstrate linguistic behaviour different from the majority. The evidence that the social hierarchy is somewhat unsettled in Canada could indicate that the role of socio-economic status may be over-rated as an influential factor in Canadian English; however, evidence of the highest socio-economic group (or two groups) establishing some forms that could be considered to be prestige forms indicates that both a social hierarchy and some kind of prestige forms of English may be developing in Canada. Thus, in spite of the fact that the rate of uniformity indicates that there is no clear prestige form in Canada, there is some evidence here that prestige may be developing, that is, that the lack of prestige in Canadian English is not necessarily indicative of a permanent peculiarity of the variety but rather that Canadian English is in an early developmental stage. 84 3.2.1.5 Special Profile Items However, the patterns of variation in the special profile items for which there is no general agreement among the socio-economic groups also indicate some lack of establishment in the social hierarchy; again there is considerable unusual conjoining of socio-economic groups. But for the special profile items, there is less evidence that the highest socio-economic group could be establishing a prestige form, as Figure 3.6 shows. Rate of Deviation Socio-economic Group Figure 3.6 - Variant Different from Majority-Special Profile Items Group IV differs from the majority for 23.3% of the 60 (38.5%) items, but group I differs for 25.0%; the higher two groups, either singly or in combination, differ for a total of 38.3% of the cases, whereas the lower two groups differ for 43.3%. Hence, once again there is more of a division between the upper two and the lower two socio-economic groups (with the lower two groups differing slightly more often than the upper two) than of a high group behaving significantly differently from the majority. It seems, then, that people's use of special profile items is more similar to their use of grammatical items and, to 85 some extent, to lexical items than to their use of other phonological items. It may be the case that consciousness of variation is a factor in usage; just as grammatical items are brought to speakers' attention, particularly by teachers, special profile items are brought to speakers' attention, perhaps by teachers, by the media and by other speakers. Increased awareness of the possible prestige or stigma of particular forms may affect usage of those forms. Thus, the analysis of the special profile items indicates that, though socio-economic status may be significant to variation, other factors such as awareness of variants and their stigma or prestigiousness may also influence variation. Thus, the simple comparison of forms used by the majority of speakers and those used by each of the socio-economic groups reveals several interesting things about prestige in Vancouver English. "The forms used by the top socio-economic group who have the highest level of education" are not generally different from the forms used by other groups; there is uniformity among socio-economic groups for an average of 79.4% of the items in the survey. Hence, there seems to be no clear prestige in Vancouver English. However, where there is no uniformity, there is some evidence that a prestige could be developing; socio-economic group IV differs from the majority more often than the other groups do. Furthermore, because the two extreme socio-economic groups differ more frequently than the others, there is some indication that both prestige and stigmatized forms are developing. The division found between the highest and the lowest two groups indicates that education may be a factor, especially in grammatical variation, and inconsistency across linguistic levels of analysis indicates that linguistic level 86 is a significant factor in variation by socio-economic status. However, before reconsidering the definition of prestige as those forms used by the highest socio-economic group and examining the issues raised by the analysis of prestige in Vancouver English, including testing the possibility that a prestige is developing in Vancouver English, it is important to consider the other aspect of the definition of prestige, that the forms used by the highest socio-economic groups are emulated by others. 3.2.2 Forms used by the Highest Socio-economic Group and Emulated by Others The other element of prestige to be considered is the emulation of the speech of speakers of the highest class by those of the lower classes. Although in the absence of a clear prestige there is nothing distinctive for speakers of the lower socio-economic groups to emulate, it may be the case that patterns typical of emulation of prestige forms may be emerging in the Vancouver area. Labov found that speakers of the upper class were more uniform across speech styles than those of the lower classes, and that those of the lower classes used the same forms as upper class speakers in their careful speech but used different variants in their more casual speech. His observation was that lower class speakers emulated the prestige forms of the upper class but were inconsistent or unsuccessful when speaking casually. To examine whether or not "a variant that is used by most [Vancouver speakers] in formal styles is also the variant that is used most often in all styles by speakers who are ranked higher on an objective socio-economic scale" (Labov 1966:405), I eliminated those variables that occur in only one style to consider the stylistic consistency of the highest socio-economic group, particularly as it contrasts with inconsistency of the lower three groups. The grammatical and lexical variables and some of each of the phonological and special profile variables are thus eliminated; what remains are 45 phonological items occurring 167 times and 49 special profile items occurring 124 times. I used simple frequencies to calculate which variant is that preferred by each socio-economic group in each speech style and then compared the preferred forms looking for both uniformity across styles for the highest socio-economic group and variation across styles for the lower groups. Table 3.2 shows a pattern that might be expected if there were emulation of the sort described by Labov in his 1966 work. M i n i m a l Word Series Visual Read ing Contrast List A u r a l Passage IV X X X X X III X X X Y Y II X X Y Y Y I X Y Y Y Y Table 3.2 - Pattern of Emulation of Prestige On the above table, X and Y denote hypothetical variants and minimal contrast is the most careful speech style; the styles decrease in formality from left to right on this table, with the reading passage thought to be most casual. Although it is unlikely that a pattern as regular as that shown on the table would emerge, the basic pattern of uniformity across the styles for class IV speakers and across classes for the most formal speech (i.e. across the top row and down the first column of the table) should hold, with some variations in the other classes and styles. 88 3.2.2.1 Phonological Items For 46.8% of the phonological items that occurred in more than one style, A L L socio-economic groups use the same forms in each speech style; hence there is no contrast between the highest socio-economic group and the other three groups for almost half of the cases. For a further 31.1% of the phonological items, A L L socio-economic groups shift forms uniformly across styles, usually using a different form in either the most formal or the most casual speech style. Again, there is no contrast between the highest socio-economic group and the other three groups; hence for 77.9% of the cases, what Labov found to be a fundamental distinction in New York seems not to be a distinction in Vancouver. Furthermore, for the phonological items for which the four socio-economic groups do not behave uniformly, group IV is no more or less uniform than any other class. For 6.6% of the cases, socio-economic group IV uses the same phonological variant in each style, but it is never the only group to do s o ; 1 9 for item 17C (low back vowel in khaki and garagel all socio-economic groups use / a / in the word list, the visual-aural prompting and the reading passage, except group III, which uses fx/ in the reading passage and group I, which uses /ae/ in the word list; for item 35 (mid vowel in the penultimate syllable of words such as l ibrary. February and raspberry'), all groups use It / in the word list, the reading passage and visual-aural prompting, except III and I, which use /9/ in visual-aural prompting; for item 14 (/ju/ following n, t or d in words such as news, student and duke1) socio-economic groups IV and III use /j u / in all five styles while group I uses /u / in i 9 Al l transcriptions were done by the Gregg team; I have taken them directly from the code books that identify each item stored in the computer. 89 all five styles and group II uses /u/ in all but the reading passage where they use / j u / , as Table 3.3 shows. 20 M i n i m a l Word Series V i s u a l R e a d i n g Contrast List A u r a l Passage 17C. v/a (in khaki and garage") I V la / la / /a / I I I la/ la/ / * / I I la / la / /a / I 1^1 la/ la/&/xl 35. - e r i Cin l ibrarv . February , raspberry.. I V It / It / /c / I I I It/ / a / It/ I I It / It / It / I It/ /9/ It/ 14. n/t/d:ju fin news, student, duke.. .) I V /ju/ /ju/ /ju/ /ju/ l\ u/ I I I /j u/ /ju/ /ju/ /ju/ /j u/ I I /u / / u / / u / /u / /ju/ I /u/ /u/ /u/ /u/ /u/ Table 3.3 - Phonological Variation Across Speech Styles: Socio-economic Group IV and One other Group Uniform For a further 6.6%, socio-economic group IV is the one to be the least consistent across speech styles and for 2.2% of the cases, socio-economic group IV together with group III is not consistent across speech styles; for item 17B (low back vowel in words such as vase, and tomato), all groups use la/ in minimal contrast, all use /e i / in the reading passage but whereas the majority and groups I, II and III use /e i / in the word list and visual-aural prompting, 2 0 In Table 3.3, the uniformity across styles is indicated by boldface charac te r s . 90 group IV uses la/; for item 1 (intervocalic voicing) all socio-economic groups use /th/ in the minimal contrast items and [d] in the reading passage, but whereas group IV uses /th/ in the word list, series and visual-aural prompting, the other groups use [d] in those styles; for item 16 (low front vowel followed by r and a vowel in words such as marry, carry and Harold'), all groups use It/ in the minimal contrast, word list, visual-aural and reading passage, except group IV, which uses lm / in the minimal contrast and the word list items; for 25 (the low vowel in <alm>), all socio-economic groups use la/ in the minimal contrast style and /o / in the word list and reading passage, but in the visual-aural prompting whereas the majority and groups II and I use la/, groups IV and III use /o/, as Table 3.4 shows.21 Minimal Word Series Visual Reading Contrast List Aural Passage 17B. a/t>/ei (in vase, tomato) 1. VtV IV la/ / a / la / / e i / III la/ / e i / / e i / / e i / II la/ / e i / / e i / / e i / I la/ / e i / / e i / / e i / IV /th/ / t h / / t h / / t h / /d/ III /th/ /d/ /d/ /d/ /<JV II /th/ /d/ /d/ /d/ /d/ I /th/ /d/ /d/ /d/ /d/ 2 1 In Table 3.4, those forms which show variation rather than uniformity are indicated by boldface characters. 91 Minimal Word Series Visual Reading Contrast List Aural Passage 16. srV (in m r^xz,c^ nxH r^jald....) IV /© / lm i It/ It/ III It/ It/ It/ It/ II It/ It/ It/ It/ I It/ It/ It/ It/ 25. <alm> IV la/ lo/ lo i lo/ III la/ 10/ lo / 10/ II la/ lo/ la/ lo/ I la/ lo/ la/ lo/ Table 3.4 - Phonological Variation Across Speech Styles: Socio-economic Group IV least Consistent Across Styles For the remaining 4.4% of the phonological variables, there is no clear pattern by socio-economic group. For the phonological variables in the Vancouver survey, then, the pattern found by Labov in New York is not revealed. As Figure 3.7 shows, not only is there a high rate of uniformity among socio-economic groups but also what variation there is is not what would be expected if people of the lower socio-economic groups were emulating the speech of those in the highest group, especially in careful speech; not only is the highest socio-economic group neither consistent across speech styles nor consistently different from the other socio-economic groups but also it is inconsistent more often than the other groups are.22 2 2 Note that, in spite of the possible methodological complication, particularly in eliciting speech styles through reading as noted on page 61, footnote 3, these data indicate that the inconsistency of the highest socio-economic group across speech styles is evident even if the Reading Style is omitted. 92 % of Cases Uniform IV & Other- IV Least IV & III No Pattern Same Consistent Inconsistent Socio-economic Group Behaviour Figure 3.7 - Phonological Variation by Socio-economic Groups Across Speech Styles Furthermore, for at least two of the phonological variables, the highest socio-economic group seems to hypercorrect, a phenomenon that other surveys have shown to be typical of lower middle class speakers; because in Canadian English /t/ is generally aspirated only in syllable initial position, an aspirated [th] between vowels is the hypercorrect form and although all socio-economic groups use the aspirated [th] in the most careful speech style and use [d] in the most casual, the highest socio-economic group maintains the aspirated [th] in the intermediate styles, dropping it only in the reading passage, which indicates hypercorrection on their part; similarly the low front vowels used before /r/ + V show hypercorrection by the highest group. Whereas all groups use the higher of the two vowels (ItI) in words such as marry, carry and Harold in the visual-aural prompting and the reading passage and groups I, II and III use the same vowel in the word list and the minimal contrast items, 93 group IV uses the /ae/ in both the two most careful styles, which suggests hypercorrection by the highest group. Thus socio-economic analysis of the phonological variants of the Vancouver survey indicates both that there are few similarities between the findings in Vancouver and those in New York and that there is no prestige form evident in the Vancouver survey. 3.2.2.2 Special Profile Items The analysis of the special profile w o r d s 2 3 confirms the findings of the phonological data. For 20.4% of the special profile items, A L L socio-economic groups use the same variant in all styles; for a further 12.3% of the items, all classes shift uniformly in some styles. Thus for 32.7% of the special profile items, there is no contrast between the highest socio-economic group and the other three groups. In only one case (94. Saturday) is the highest socio-economic group the only one to use the same form across styles. As Table 3 . 5 2 4 indicates, however, the pattern is not one that would be expected for emulation of a prestige form in careful speech. 2 3 As noted on page 54, special profile items are those "in which specific sounds may vary freely, and independently, creating a number of patterns" (Gregg 1984:Dl-3); the possibility of there being several different combinatorial variants of the item makes these different from simple phonological items. For example, whereas for phonological items, there are several variants of a single phoneme in any of a number of words, for special profile items there are several combinations of phonological variants of more than one phoneme in a single item. Thus, variants of Tuesday include all the possible combinations of /ju/, /u / or their variants in the initial syllable with /I/, / e l / or their variants in the final syllable. Thus, although one could consider the first syllable of Tuesday as an environment for the variants of /u/, the S V E N database isolates those items, "special profile items," in which more than one phonological variable appear in several combinations. 2 4 In Table 3.5, the uniformity is indicated by boldface characters. 94 Minimal Contrast Word List Series Visual Aural 94. Saturday IV III II I /seethardei/ /sethardei/ /sethardei/ /sedardei/ /sedardei/ /sedardei/ /sethardei/ /sedardei/ Reading Passage /sethardei/ /sethardei/ /sethardei/ /sethardei/ Table 3.5 - Special Profile Variation Across Speech Styles: Socio-economic Group IV Uniform Note that socio-economic group II does not use the same form as the other groups in careful speech and that all socio-economic groups use the "prestige" form in the most casual speech.25 This, however, is the closest the distribution of any variants comes to that of a pattern typical of the kind of emulation of prestige Labov found in his work. Although for an additional 22.5% of the special profile items the highest socio-economic group uses the same form in all styles while some other socio-economic group or groups use a different form in at least one of the styles, the highest socio-economic group is never the only one to be consistent across styles, as Table 3.626 indicates. 1 5 Use of the prestige form in the reading passage could indicate that the style was not as casual as the Gregg team has generally found; it seems to me that the passage being a casual dialogue would prompt role playing, including adopting particular speech forms and style, rather than consistent casual speech. The elicitation of Saturday includes collocation with your father rather than dad or some informal term; hence, the readers may have processed this string in a more formal manner than in other parts of the dialogue. The reliability of assuming the reading style is consistently informal, then, is questionable. Further testing using the spontaneous narrative data when it becomes available is indicated. 2 6 In Tables 3.6 - 3.8, those forms which show variation rather than uniformity are indicated by boldface characters. 95 M i n i m a l Contrast Word List Series V isual A u r a l Read ing Passage 42. Garage IV III II I 43. Groceries IV III II I 58. Often IV III II I 62. Leisure IV III II I /garag/ /garag/ /garag/ /goreedg/ / g r o s r i z / / g r o s r i z / / g r o s r i z / / g r o s r i z / /ofan/ /ofan/ /afan/ /ofan/ /teg.../ /teg.../ /teg.../ / t i g . . . / /garag/ /garag/ /garag/ other 2 7 /garag/ /garag/ /garag/ other / g r o s r i z / / g r o s r i z / /gro Jriz/ /gro Jriz/ /afan/ /ofan/ /ofan/ /ofan/ /ofan/ /ofan/ /afthan/ /afan/ /teg.../ /teg.../ /teg.../ / * i g . . . / 2 7 The category, "other," groups together pronunciations that are different from those found to be most common; for example, seven of the possible pronunciations of garage were found to be common to many speakers, whereas the other eleven pronunciations occurred only in isolated incidents. Thus, the eleven uncommon pronunciations form the group "other." It is, however, unusual for there to be as many as eleven variations grouped under "other." 96 71. Without IV III II I Minimal Contrast Word List / V l 8 A U t / /VlSAUt/ Series Visual Aural Reading Passage /VlSAUt/ / V l 8 A U t / / V l 8 A U t / /VlSAUt/ / m 8 A U t / ZWIQAUX/ / m 8 A u t / /WIBAXIX/ 90. Tuesday IV III II I 91. Wednesday IV III II I /venzdei/ /venzdei/ /wenzdei/ /venzdei/ /tjuzdei/ /tjuzdei/ /tjuzdei/& /t j u zd i / /tjuzdi/& / t u z d i / /wtnzdei/ /venzdei/ / v e n z d i / / v e n z d i / /tjuzdei/ /tjuzdei/ /tjuzdei/ /tj u z d i / /venzdei/ /wenzdei/ /venzdei/ /venzdei/ 96. 98. Khaki IV III II I Newspapers IV III II I 105. Recognize IV III II /karki/ /karki/ /karki/ /k©ki/ /njus.../ /njus.../ /nus.../ /nus.../ /rekagnaiz/ /rekagnaiz/ /reksgnaiz/ /rekagnaLz/ /karki/ /karki/ /karki/ other word /njus.../ /njus.../ /njus.../ /nus.../ /rsksgnatz/ /reksgnaLz/ /rekagnaiz/ /rekgnaiz/ 97 120. M i n i m a l Contrast Library IV III II I Word List / i a i b r e r i / / t a i b r e r i / / t a i b r e r i / / t a i b r e r i / Series V isual A u r a l Read ing Passage / t a i b r e r i / / t a i b r e r i / / t a i b r e r i / / t a i b r e r i / / i a i b r e r i / / t a i b r e r i / /taibrsri/ /Jraibreri/ Table 3.6 - Special Profile Variation Across Speech Styles: Socio-economic Group IV Not the Only Uniform Group Hence for 55.2% of the special profile items, socio-economic group IV does not use forms differently from the other three socio-economic groups. Furthermore, not only is the highest socio-economic group the only one to be consistent across styles for only 2% of the items, but also for 18.4% of the items, it is the group that differs from the others, as Table 3.7 indicates. 45. Schedule IV III II I 46. Fifth IV III II I Beautiful IV III II I 57. M i n i m a l Word Series Contrast List other /skedg©*/ /skedgo*/ /skedgo*/ / f i f e / / f i f e / / f i f e / / f i f e / /bjuthaf©*/ /bjudafoi/ /bjudafo*/ /bjudafo*/ Visual A u r a l Read ing Passage /skedz/a*/ /skedzuaiy /skedgo*/ /skedgua*/ /skedz©*/ /skedguat/ /skcdg©*/ /skedzua*/ m e / / f i f e / / f t f e / / f i f e / / f i f e / / f i f e / / f i f e / / f i f e / /bjudafo*/ /bjudafo*/ /bjudafot/ /bjudafo*/ /bjudafo*/ /bjudafot/ /bjudgfo*/ /bjudaf©*/ 98 M i n i m a l Word Series V isual Reading Contrast List A u r a l Passage 74. Either IV / icfer/ /aicFar/ III / i 5 a r / /ifer/ II / icfar/ / icfar/ I / i 3 a r / / icfar/ 82. Kitsilano IV / k i t s a i a t n o u / / k i t s a * a i . n o u / / k i t s a * © n o / III / k U s a * a i . n o u / /kLtsa*a i .nou/ / k i t s a * a i n o u / II / k i t s a i a i n o u / / k L t s a * a i n o u / / k i t s a * a i n o u / I / k i t s a * a i n o u / / k L t s a * a i n o u / / k i t s a * a i . n o u / 97. Caramel IV other / k a r m a * / III / k a r m a * / / k a r m a * / II / k a r m a * / / k a r m a * / I / k a r m a * / / k a r m a * / 102. Tomato IV other / t ame idou / / t ame idou / III / tameidou/ / t ameidou / / t ame idou / II / tameidou/ / t ameidou / / t ame idou / I / tameidou/ / t ame idou / / t ame idou / 122. Mountain IV / m a o n t h n / / m a o n t h n / / m a o ? n / III / m a o n t ' n / / m a o n t ' n / / m a o ? n / i i i II / m a o n t ' n / / m a o n t ' n / / m a o ? n / • i i I / m a o n t ' n / / m a o n t ' n / / m a o ? n / • i i Table 3.7 - Special Profile Variation Across Speech Styles: Socio-economic Group IV the Group to Differ In a further 8.2% of the analyses of the special profile items by socio-economic status, the highest group in combination with either the second or third highest group is the one to differ rather than remain constant, as Table 3.8 shows. 99 M i n i m a l Contrast Word List Series Visual A u r a l 108. Read ing Passage 40. 104. 48. Potato IV III II I Sandwiches IV III II I Apricots IV III II I Arctic IV III II I 117. Dual IV III II I /patheithou/ /patheidou/ /patheidou/ /patheidou/ / s c n d v i d g a z / / s s m v i d g a z / / s c n w i d g s z / / s s m v i d g a z / /ei. . . / /©.../ /ei. . . / /©.../ /arktluk/ /arktluk/ /arthik/ /arthik/ / d j u a * / / d j u s * / / d j u s * / / d j u a t / /dua*/ /dua*/ /dua*/ /dua*/ /patheidou/ /pateidou/ /patheidou/ /pateido/ /patheidou/ /pateidou/ /patheidou/ /pateido/ / sendv tdgaz / /sefevidgaz/ /scmvidgaz/ /s&mdgaz/ /stemvidgaz/ /sevidgaz/ /ssemvidgaz/ /ssevidgaz/ /©. . . / /©. . . / /ei. . . / /« . . . / /arthik/ /arthik/ /arthik/ /arthik/ both 2 8 both /ei.../ Table 3.8 - Special Profile Variation Across Speech Styles: Socio-economic Group IV and One Other Group Differ 2 8 Socio-economic groups IV and III used both / © . . . / and /ei.../. There is little agreement about the pronunciation of apricot: not only are groups IV and III evenly divided between the two pronunciations in reading style, but also all figures cluster around the 50% mark. 100 Thus it is evident that for 26.6% of the special profile items, the highest socio-economic group, either singly or with one other group, chooses a form different from that chosen by the other socio-economic groups, making the total rate for which the highest socio-economic group does not behave as Labov found the same group in New York to behave 81.8%. In only 2% of the cases is anything similar to the patern of emulation found. For the remaining 16.3% of the special profile items there is little pattern in the analysis by socio-economic status. As Figure 3.8 reveals, emulation of a prestige is not clearly evident for special profile items in the Vancouver survey. Prestige Other- Consistent Other Not Pattern Same Consistent Socio-economic Group Behaviour Figure 3 . 8 - Special Profile Variation by Socio-economic Groups Across Speech Styles Clearly, then, Labov's findings are not duplicated in the Vancouver survey; not only do all socio-economic groups behave uniformly for more than 77% of the phonological variants and over 32% of the special profile items but also the highest socio-economic group does not use the same forms consistently across speech styles. There does not, therefore, seem to be emulation of prestige forms by speakers of lower socio-economic groups. 101 3.3 The Development of Prestige in Vancouver English Both the comparison of the forms used by the highest socio-economic group with those used by the majority of speakers and the analysis of differences in the forms used in different speech styles by each of the socio-economic groups indicate that there is no prestige form, no form used by the highest socio-economic group and emulated by others, in Vancouver English. Before considering whether the definition needs to be reformulated or there is something unusual about prestige in Vancouver English, the possibility that prestige forms are developing in Vancouver English must be tested. Given the recent date of SVEN, it is too early to do another survey and compare the findings of SVEN with those of a new survey to determine whether or not there has been change in real time; hence, a test of change in apparent time must be used. To do so, the forms for which the highest socio-economic group does differ from the majority of speakers must be identified and cross-tabulated by age. Those forms that might be considered here include 3 lexical items, 2 grammatical items, 3 phonological items and 12 special profile items; they are: skookum (has used), 10:30 (second choice), slough (meaning), It's  me/I. these kind(s) of apples, intervocalic /t/, low front vowel before /r/ plus a vowel, low back vowel in tomato and vase, fifth, beautiful, either. Kitsilano/Capilano. caramel, tomatoes, potatoes, mountain, sandwiches. schedule. Saturday, and lieutenant.2^ If there is a prestige form developing, 29 See Appendix E for the forms chosen by Socio-economic Group IV and those chosen by the majority and the other groups. 102 the oldest age group of the highest socio-economic status will use forms different from the majority with less frequency than the youngest group. 3 0 Cross-tabulating these 20 items by socio-economic status and age indicates that there is no prestige developing in Vancouver English; there is no evidence of the youngest group of the highest socio-economic status using a form that is different from that used by the majority more frequently than the middle or oldest age groups do. For 12 of these items, cross-tabulation by age and socio-economic status shows nothing significant; 10:30 (second choice), s lough (meaning), these kind(s) of apples, fifth, beautiful. K i ts i lano /Cap i lano . caramel, tomatoes, potatoes, mountain, sandwiches, and schedule show neither patterning nor significance by age. For the other 8 items, cross-tabulation by age reveals not the development of prestige but a general change in progress; for skookum (has used), It's me/I. intervocalic /t/, low front vowel before /r/ plus a vowel, low back vowel in tomato and vase. Saturday, either and lieutenant the oldest age group in all socio-economic status is retaining older forms. For example, in the minimal contrast style where socio-economic group IV differed most noticeably from the majority of speakers in their 3 0 Note that in tests for simple change in progress using apparent time, the oldest group would have the highest rate of usage of the older form while the youngest group would have the lowest rate of that form (or, the youngest group would have the highest rate of usage of the newer form, while the oldest group would have the lowest rate of that form). However, what is being checked here is the development of a form that is used by the highest socio-economic group but not by the majority of speakers; the indication of such development is a higher rate of usage by the youngest group of the top socio-economic status than by the oldest group of the same status. The rate of difference between socio-economic group IV and the majority is the key factor here; the rate should increase as age decreases if a prestige form is developing. 103 treatment of /t/ intervocalically, [d] was used by the youngest age group of the lowest socio-economic status in 64% of the occurrences compared with [d] being used by the oldest age group of the highest socio-economic status in only 5% of the occurrences. The figures for occurrence of [d] are stratified by age and socio-economic status decreasing in frequency with increased age and socio-economic status. Inversely, an aspirated [t] is found most frequently among the oldest age group and the highest socio-economic status; the oldest group of the highest socio-economic status realized intervocalic ft/ as an aspirated [t] in 93% of the occurrences whereas the youngest group of the lowest socio-economic group realized aspirated [t] in only 33% of the occurrences. Once again the figures are stratified by age and socio-economic status. Not only is the predicted pattern of a lower rate of use of a different form (in this case a form other than [d]) by the oldest group not evident here but also all socio-economic groups change in apparent time; thus a change in progress is evident rather than a developing prestige. Although the figures are not as striking or as clearly stratified for the other 7 variables, it is clear that there is a change in progress evident rather than a developing prestige in skookum (has used), It's me/I. low front vowel before /r/ plus a vowel, low back vowel in tomato and vase. Saturday, either and lieutenant as well as in intervocalic /t/.3 1 3 1 It could, of course, be the case that the change in progress itself suggests that a new prestige model is exerting its linguistic influence on younger speakers, in which case one would assume that the older speakers of the highest socio-economic group maintain an older prestige variant (aspirated t) while younger speakers are establishing a new prestige variant (the voiced variant). However, the SVEN data are inconclusive in this matter; at best, such a suggestion is highly speculative. 104 Because there is no evidence of the development of a prestige in Vancouver English in these forms that show some evidence that Socio-economic Group IV differs from the majority of speakers in some instances, there is also no evidence of developing emulation of prestige forms. Saturday, the only item for which Socio-economic Group IV was the only group to use the same form in all speech styles, shows a pattern typical of a change in progress rather than a developing prestige or developing emulation; there is, therefore, nothing to test for developing of emulation of the speech of the highest socio-economic group by speakers of lower groups. It seems, therefore, that there is no evidence of either a distinct prestige (in the sense of those forms used by the speakers of the highest socio-economic group) developing or a tendency for speakers of lower socio-economic groups to emulate speech of the highest group. The evidence indicates that Vancouver English neither has nor is developing a prestige as it is defined here. 3.4 Conclusion The absence of a prestige form in Vancouver English indicates either that the definition of prestige as those forms used by the highest socio-economic group and emulated by others is inadequate or that the Vancouver speech community is unusual. Both possibilities must be considered. 3.4.1 The Definition Given that the definition of prestige used here adequately delineates prestige as it is found in Britain and the U.S., two of the major English speaking nations in the world, it is unlikely that the definition is misguided or somehow 105 inadequate. What typifies the Vancouver speech community is widespread homogeneity across socio-economic groups; on the basis of this finding, should one redefine prestige as, for example, those forms used by the majority of speakers? To do so would be to disallow a prestige accent such as RP is in Britain and the emulation found in both Britain and the U.S. Or the term would then have distinct referents in Canada and in Britain and the U.S. From the point of view of a coherent overall theory of dialect variation, neither of these possibilities is desirable. There is nothing in the Vancouver data that suggests that the definition is incorrect or inadequate; none of the evidence reveals some aspect of prestige that may be excluded by that definition or may be inappropriate in the Canadian context. Rather, the data reveal only negative evidence, i.e. a lack of prestige by the definition given, and negative evidence is inconclusive. It is, therefore, much more reasonable and practical to conclude that defining prestige as that variety or those forms used by the highest socio-economic group and emulated by others is adequate but that prestige does not exist in Vancouver English. 3.4.2 Prestige in Vancouver English It is necessary, therefore, to consider the absence of prestige in Vancouver English. The absence of a prestige form in Vancouver does not support the argument that prestige does not exist; rather, it suggests that the cultural differences between Canada, Britain and the U.S. are significant. It is likely that Vancouver, which celebrated its centenary in 1986, is too young to have developed either a socio-economic group that is securely in a position above the others or a prestige form of a dialect. Several things in the SVEN data indicate the absence of a socio-economic group sufficiently secure to set 106 targets for or impose rules on the groups beneath; the clearest indication of an absence of a clear upper socio-economic group is the hypercorrection by the highest group indicated in the preceding analysis and the lack of a cross-over pattern (i.e. the tendency of a middle socio-economic group, in their striving to be correct or to get ahead, to have a higher rate of use of some forms than the group above them on the socio-economic scale) such as the one between the highest two classes found by Labov in New York (Labov 1966: passim ) and that between the working classes found by Trudgill in Norwich (Trudgill 1972:90-132 passim ). The S V E N database shows no cross-over pattern in either the attitudes section of the data or in the quantification of the language used by Vancouver speakers. Although Hasebe-Ludt is using the spontaneous narrative data for the purposes of analyzing phonological patterns rather than prestige, her impression from what she has done is that there is no evidence of a cross-over pattern between socio-economic groups in the spontaneous narrat ive; 3 2 hence no such pattern is found in any of the data collected in the Vancouver Survey. The lack of such a pattern, especially when coupled with the evidence that the highest group itself tends to hypercorrect, indicates that all four socio-economic groups in the Vancouver area are, or consider themselves to be, upwardly mobile; there is no clearly established group that behaves as upper classes, those identifiable in older, more established countries such as Britain and the U.S. behave. Since there seems not to be a clear upper class toward which to aspire in the area of the Vancouver Survey, it follows that there is also not a clear prestige form within that dialect, as the survey indicates. Personal communication Thus, it is important to view the Vancouver sociolinguistic community, if not the Canadian one, as a developing community and to consider what, in the absence of a clear prestige, functions as prestige is thought to function in language variation and change; what do Vancouver speakers aspire to in the absence of a prestige? Given the striking homogeneity among Vancouver speakers, it is possible that different aspects of the standard in Vancouver English may function in language variation and change as prestige does in other communities. It is important, therefore, to go on to consider standard in Vancouver English. 108 CHAPTER 4 STANDARD IN THE SURVEY OF VANCOUVER ENGLISH Standard has been defined here as that variety (or those forms) used by the majority of speakers and embodying notions of correctness. The first part of the definition is taken from the generally held view that standard serves as a medium for the general communication of a community or nation; the second part is taken from the popular view that there are correct forms and that correctness is a criterion of standardization. Both parts of the definition will be tested here using the S V E N database. 4.1 Variety Spoken by the Majority of Speakers In the examination of prestige in Vancouver English (see Chapter 3), it was sufficient to consider relative majority, since what was being considered was whether or not the highest socio-economic group differed from the majority of speakers and from other socio-economic groups. However, for a consideration of standard, it is absolute majority which is relevant, since what is being considered is which forms are most frequently used. Therefore, to consider the possibility that a standard might be that variety or those forms used by the majority of speakers, simple usage frequencies must be tabulated to determine for what percentage of the items more than 50% of the speakers agree on a single variant. If standard can be defined in part as those forms used by the majority of speakers, then there should be agreement by more than 50% of the speakers for more than 50% of the items tested in the survey. 109 Overall, for the items tested in S V E N there is more than 50% agreement among speakers about a particular variant; in fact, there is clear majority usage for 78.6% of the items tested. The rate of agreement ranges from 50.7% to 100%; for example, 50.7% of the informants thought that "between you and me" is the correct form and 100% of the informants claim not to use any form other than "I didn't see any money in the box." There are, however, very few responses for which the rate of agreement is near either 50% or 100%; the bulk of the responses range from the high 50's to the low 90's, with, as one might expect, the average in the mid 70's. There is also a considerable range across linguistic levels in the percentage of items for which there is a clear majority. There is majority usage for 93.3% of the grammatical items, 85.1% of the phonological items, 78.4% of the lexical items and 57.4% of the special profile i tems. 1 A low rate of agreement about special profile items is typically expected. Since these are items for which most variation is possible, 2 one would expect lack of agreement on any one form. This expectation is not borne out however. There is clear majority of agreement for more than 50% of even these items. This high rate of agreement on special profile items markedly supports the stronger findings for the more settled items. 3 It is important to note, further, that dialect studies test the unsettled not the settled 1 See Appendices D and E for lists of the forms chosen by the majority of speakers (both relative and absolute majority). 2 For example, there is variation in the pronunciation of the initial syllable, the medial consonant and the final vowel of schedule: thus, rather than one variable being tested, three different variables are considered in this single item. There are, therefore, many possible variant pronunciations of schedule: all special profile items contain at least two variables within the item; therefore, there is little agreement about any one pronunciation of these items. For example, S V E N records 18 different pronuciations of garage. 3 Note that the overall trend to majority usage for the grammatical, phonological and lexical items holds in 85.6% of the cases. 110 forms; one does not test, for example, use of basic function words or common, settled terms such as item, note, test or example. The figures here, then, represent majority use of the least agreed upon items in the variety tested. Thus, the figures for items used by the majority over all of Canadian English would be much higher than the 78.6% found for the variable items. It is clear, then, that for most items in SVEN, there is one variant which is used i by the majority of speakers; furthermore, even for more than 50% of those items for which there is thought to be little agreement, there is one variant which is used by the majority of speakers. Taking these findings in light of the fact that only variable items are tested in dialect surveys, it can be concluded that "those forms used by the majority of speakers" is an essential feature of Vancouver English. The high rate of clear majority usage for items in SVEN is consistent with the argument that Canadian English is typified by a degree of homogeneity not found elsewhere. Generally, the homogeneity described is geographical; the fact that Vancouver English is so similar to the English of, say, Ottawa or Toronto, is the general focus of statements about homogeneity. However, the findings in Chapter 3 of this work indicate that there is also an unusual level of sociolinguistic homogeneity in the SVEN database. Recall that for 79.4% of all items in SVEN there is no difference in forms chosen by different socio-economic groups; there is, therefore, not only majority agreement about 78.6% of the items in SVEN but also no differentiation by socio-economic group for 79.4% of the items. The most significant variation in the Vancouver Survey, then, is that across speech style; only 33.6% of the items do NOT I l l i change forms across speech style. However, that variation is also atypical; one would expect differences between casual speech and the most formal speech, especially among the lower socio-economic groups. Here, however, as well as those items which do not change forms across speech styles in any socio-economic group, there are items for which all socio-economic groups rather than just the lowest change forms; for a further 21.7% of the variables, ALL socio-economic groups shift across styles. Thus, even for variation across speech style, there is some kind of uniformity for 55.3% of the items. It seems, then, that the language used by Vancouver speakers is typified by sociolinguistic as well as stylistic homogeneity. This homogeneity is in keeping with the high rate of items for which one variant is used by the majority of speakers. It seems clear, therefore, that majority usage is a central feature to Vancouver English and constitutes a standard. On the basis of the SVEN findings, then, it can be concluded that standard is viably defined by the majority usage; what remains to be tested is whether or not standard also embodies notions of correctness. 4.2 Variety Typified by Correctness There is less general agreement that correctness is central to standard than that standard consists of forms used by the majority of speakers; however, both speakers and scholars seem to attach some notion of correctness to standard. As Wells observes: Certain accents have a special position in that they are regarded, whether tacitly or explicitly, as standard. In 112 England it is RP which enjoys this status, in the United States the range of accents known collectively as GenAm. A standard accent is the one which, at a given time and place, is generally considered correct: it is held up as a model of how one ought to speak, it is encouraged in the classroom, it is widely regarded as the most desirable accent for a person in a high-status profession to have. (Wells 1982:34, second emphasis mine) Wells is, of course, quick to add: a standard accent is regarded as a standard (or 'norm') not because of any intrinsic qualities it may possess, but because of an arbitrary attitude adopted towards it by society, and reflecting the attitude the community implicitly holds towards its speakers. (Wells 1982:34) While there is some disagreement about the extent to which correctness is a feature of standard, there is general agreement among scholars that the adoption of forms as standard is arbitrary and results from attitudes toward the variety or form rather than from intrinsic value. Standards of correctness are established on arbitrary grounds of either a practical or an attitudinal sort. Moreover the two types are quite distinct. On practical grounds, standard forms may be established because grammarians have singled them out and considered them to be correct; speakers then follow this advice for practical reasons. In this case, correctness is an attribute of the form itself. Sometimes, however, speakers themselves attribute correctness to a form as reflective of their positive attitude toward it. This attitudinal attribution is made quite irrespective of whether grammarians have identified this form as being a standard (and hence correct) form. In this latter case, it is the attitude itself that leads to the adoption of or acceptance as standard. The question, then, is not simply whether or not standard forms are traditionally correct (i.e. in the practical sense), but rather, whether or not there is a notion of correctness associated with standard (i.e. in the attitudinal sense). Either or both of these 113 are possible; hence, both correctness and notions of correctness will be tested here using the grammatical section4 of the Survey of Vancouver English. 4.2.1 Notion of Correctness It is clear from the data that there is a notion of correctness among Vancouver speakers; not only do they rarely respond "I don't know" to the question "Which do you think is the correct form?" but also there is some evidence that they expect there to be some correct form of Canadian English. Vancouver speakers unhesitantly identify "correct" grammatical forms. SVEN informants were asked to identify the correct variant of 42 grammatical items;5 of these 42, only 12 elicited any "Not sure or don't know" responses. For only 28.6% of the items, then, was there any indication of uncertainty; furthermore, the rate of uncertain response for those 12 items ranged from .4% to 2.7%, with the average being 1.6%. Thus the rate of uncertain response is very low; Vancouverites seem to have a strong notion of correctness. It has been suggested that there may be evidence in SVEN that Canadian speakers may be unusually insecure about grammatical correctness for speakers in a standard speech community;6 if that were the case, there would 4 There are two reasons for limiting this test to the grammatical section of SVEN: first, and most important, is that the grammatical section is the only one in which respondents were asked specifically about correctness; second, it is generally agreed that correctness is associated more directly with grammar than with phonology or lexicon or, perhaps, that correct, when modifying grammatical items, is thought to be quite literal but, when modifying phonological and lexical items, is thought to be metaphoric. It is, then, most useful to limit this analysis to the grammatical section of SVEN. 5 See Appendix F for the forms deemed correct. 6 Bernard Saint-Jacques, personal communication. 114 be a high rate of "Not sure or don't know" responses and Canadians would be uncertain about grammatical forms for which there is no disagreement or uncertainty among other speakers such as Americans. Not only do the above figures indicate that such is not the case but also there is no indication that the uncertainty that was recorded is peculiar to Canadian speakers. Most of the items for which there is little agreement among Canadians are given special attention in American Handbooks and are discussed in such works as Fowler's Modern English Usage; 7 such attention indicates that these forms are unsettled for speakers of varieties other than Canadian as well as for Canadians. The following few grammatical items elicited the little amount of uncertainty that SVEN informants expressed:8 drank/drunk .7% proven/proved 2.1% lend/loan (verb) .4% If I was/were you .4% If it was/were warmer 1.3% between you and me/I 2.7% me/I (as object) 1.3% it is me/I 1.3% anyway(s) .8% lie/lay down .4% laid/lain (It has laid/lain/lied there all night.) 1.7% forgot/forgotten .7% Of these, only drank/drunk in the simple past is surprising; what discussion there is of drank/drunk in most handbooks is of the past participial form, for which there is uncertainty among speakers other than Canadians. For proved/proven, lend/loan. anvwavCs). lie/lay, laid/lain and forgot/forgotten there is also uncertainty in all standard speech communities, as is evidenced 7 See Appendix G for the Dictionaries and Handbooks used for and cited in this chapter of the work. 8 The percentages given are the rate of "Don't know or Not sure" responses for each item. 115 by their inclusion in the "Glossary of Usage" sections in most handbooks.9 Many of these are also included in Fowler's Modern English Usage, still thought by many to be the authority on modern usage. Several of the other 5 items also appear in the "Glossary of Usage" sections or get special mention in the appropriate sections of handbooks; confusion about I and me as the object, especially in the sentence "It's I/me!", is found in all varieties of English. The use of the subjunctive is so uncertain in modern usage that some think it to be passing out of English. ^  Thus, there is no indication in the SVEN database that Canadians are more insecure about grammatical correctness than Americans or Britons are; rather, there is evidence that they have a strong notion of correctness. Unusual responses to two of the questions in SVEN indicate in a less direct way that Canadians have a notion of correct and standard English. Question 64 asks "Would you consider the CBC announcers' language an example of good 1 Canadian English?"; question 1050 requires informants to agree or disagree with the following: "The language of CBC announcers should be the standard for spoken Canadian English." I believe that respondents had some notion of correctness and of standards, in mind when they responded to those questions and that what is indicated is that there is a notion of correctness and of some desire for guidelines for "correct" Canadian English. My conclusion follows 9 See for, example, Baker (1972) and Crews (1977), both American handbooks, H.R. Fowler (1980), a British work, Millward & Flick (1984), the Canadian edition of an American handbook, and Messenger & de Bruyn (1980), a Canadian handbook. 10 Fowler suggests the subjunctive is "dying" and limits the use to a few clear cases (Fowler 1965:595). I have even seen University professors mark standard use of the subjunctive in those clear cases as an error. 116 from the fact that speakers responded to these questions despite considerable lack of familiarity with CBC English. In the personal information section of the questionnaire, the informants were asked how much TV they watch and radio they listen to and which stations and channels they tune in to. Only 17% listen to CBC radio, though 22% listen to CBC radio news, and only 3.3% watch CBC TV, though 33% watch CBC TV news. I don't know how to account for the fact that more people listen to the CBC news than listen to the CBC radio or for the similar pattern in the responses to TV viewing other than to suggest that some people treat hearing or seeing the news as entirely separate from general radio listening or TV viewing; nor can I account for the fact that 3% of the people who claim to watch or listen to CBC news also claim to have no exposure to the language used by CBC announcers. I suspect that many of the informants are so much less familar with the CBC than the researchers assumed that the questions themselves were somehow unclear. While it is true that many, or even most, academics listen to CBC, the same is not true for other communities. The breakdown by social status indicates that 49% of the people who listen to the CBC are in the highest socio-economic group and a further 24% are in the third group. Of those who listen to the CBC radio news, 52% are in the highest socio-economic group and a further 23% are in the third group. Similarly, of those who watch the CBC TV news, 37% are in the highest group and 33% are in the third group. Only 19% of the people in the lowest two socio-economic groups listen to the CBC. It is a mistake, therefore, for researchers to assume general familiarity with the CBC. Even if the 22% who listen to CBC radio news are taken to be different informants from the 33% who watch CBC TV news, the total percentage of informants who may be familiar with CBC English is 52% (a figure which is high: because 40% of those who listen to the 117 CBC radio news also watch the CBC TV news, the figure is likely to be closer to 45%, at most). Despite this confusion and the low rate of familiarity with CBC English, most informants answered the questions about CBC language; only 19% admitted to having no exposure to the CBC when asked, "Would you consider the CBC announcers' language an example of good Canadian English?". 48% of the informants responded positively; because this figure is higher than the exposure to CBC language seems to be, it seems as if people are responding to what they think ought to be rather than what they have observed to be the case. People think CBC English should be good; they have a notion that there is correct English and that it is likely to be that spoken by CBC announcers. This notion is further reflected in the attitudes question, "The language of CBC announcers should be the standard for spoken Canadian English," a statement which was ambiguous for many informants. Some informants took it to mean that Canadians should adopt the language of the CBC announcers as standard Canadian, and others took it to mean that the CBC should consciously and intentionally provide a standard for, if not impose a standard on, Canadians. The ambiguity is reflected in the high rate of neutral (or don't know) responses; 25% of the informants indicated neutrality or uncertainty. The other responses are questionable because of the ambiguity and the unusual reporting of familiarity with the CBC; however, the fact that 42% of •I informants agreed that "The language of CBC announcers should be the standard for spoken Canadian English" and 33% disagreed despite the ambiguity and the lack of familiarity with the CBC seems to indicate a notion 118 that there is something that must be standard Canadian English. It may also indicate a desire for guidelines for that standard. Although this evidence is indirect, it confirms the strong notion of correctness and of standard that the grammatical responses reveal. There seems to be no doubt that people have some notion of correctness. That is not to say, however, that the notion of correctness necessarily corresponds to what is identified by grammarians as conventionally and traditionally correct. It is important, therefore, to consider how well the Vancouver respondents' identification of correctness corresponds to what has traditionally been regarded as standard. 4.2.2 Conventional Correctness To assess conventional grammatical correctness, it is necessary to consider both the conventions for Standard English in general and for the Canadian variety of Standard in particular; therefore, the forms identified as correct by the Vancouver speakers must be compared with "correct" forms given in both general and Canadian handbooks. Generally, Vancouver speakers have identified correct grammatical forms quite conventionally, and with an absolute rather than a relative majority;11 for only 9.5% of the grammatical items was the highest rate of agreement less than 50%. For a further 19% the majority agreement was between 50 and 55%; for the remaining 71.5% the majority agreement was over 55%, with the average being 80.7%. Once again, 1 1 See Appendix F for the "correct" grammatical forms, with rate of selection. See also 4.1 of this work for absolute majority usage over all linguistic levels. 119 this is strong evidence that the majority of Vancouver speakers have at least a notion of correctness and do not appear grammatically insecure. Those items for which there is not a clear absolute majority selection are either unsettled items or items for which there are so many variants that the responses are spread quite thinly among them. For example, there were 8 options given for the conditionals, "If we had seen you, we would have spoken," and "If I had gone home, I would have found it." Rate of agreement on these two is very low. The others for which there is a low rate of agreement are those which are unsettled in general 1 2 or those for which there is a strictly correct form in formal usage and a different form frequently heard in informal speech. 1 3 Agreement about grammatical correctness, then, is generally high; what remains to be seen is the extent to which the Canadian informants identify standard forms as correct. Dialectologists generally test the unusual rather than the usual; hence, most of the forms tested are unsettled. The Survey of Vancouver English was designed to test variables that are unsettled and that may be either changing or typical of Canadian English; hence, there is likely to be little agreement about some items and there may be some for which there are differences between what Canadians think is correct and what others think is correct. However, the variables being considered for correctness are grammatical and part of a variety of standard English; they are, therefore, less subject to 1 2 See discussion below. 1 3 It is likely that these responses are for casual speech; they were elicited through speech and the tone of the interviews was very informal. Where there are differences between speech and writing, then, it may be that the "correct" form given here is that deemed acceptable in speech. 120 variation than the other linguistic levels are. There are some differences but there are also many similarities between what Canadians report in SVEN and what seems to be the case in other speech communities. I cannot here draw direct comparisons between what SVEN speakers deem correct and what, say New York speakers or London speakers deem correct; complete contrastive studies would be needed for that. I can, however, determine the extent to which SVEN informants agree with standard usage as recorded (or prescribed) by grammarians. To assess the extent to which the forms identified as being "correct" by Vancouver speakers are those traditionally considered correct in Standard English, I have used the Oxford English Dictionary, Fowler's Dictionary of Modern English Usage, Random House Dictionary of the English Language, the Gage Canadian Dictionary and the handbooks listed in Appendix G of this work. Given the findings, I have divided the variables into four groups: those that are the least controversial and for which there is a high degree of agreement, those for which there is little agreement among speakers, those for which there is little agreement among or comment by scholars and those for which there seems to be a Canadian form. I will also discuss those forms that SVEN informants have deemed correct that may be considered traditionally incorrect.14 Generally, Canadians have identified general standard English forms as being correct; for 83.3% of the grammatical items, the forms deemed "correct" by 1 4 These forms remain in the four categories as well, where they are noted with an asterisk, as it is not clear that Canadians would be the only ones to get these wrong. 121 Vancouver speakers are traditionally considered standard forms. 47.6% of the forms are in the first category: I LEAST CONTROVERSIAL: MOST AGREEMENT 669 Yesterday he drank/drunk 3 glasses... d rank 98.3% 677 Boughtfen) bread isn't as tastv as... bought 87.4% 685 I've often brought(en) them home... b rought 96.0% 713 There are less/fewer people here than... fewer 71.3% 737 I didn't see anv/no money in the box. a n y 99.7% 773 You should have/of/shoulda seen him... should have 83.3% 789 I saw/seen him vesterday. saw 96.3% 805 After we ate/had eaten dinner, we plaved...had eaten 65.7% 681 There is/are 8 eggs in the fridge. there are 69.2% 733 Where are mv gloves? where are 80.4% 749 If I was/were vou. I'd vote against it. were 93.8% 697 Helen and I are going shopping. Helen and I 97.9% 725 He doesn't/don't bother me. doesn't 96.0% 753 He doesn't like anvthing. a n y t h i n g 99.2% 661 He was standing behind me. b e h i n d 91.3% 665 The car is behind the garage. b e h i n d 74.6% 761 Well, anvwav(s). that's over. anyway 86.3% 705 Yesterdav he lav/ laid in the sun. lay 68.3% 813 L ie / lav down right awav! l ie 78.7% 769 He must have forgot(ten) mv name. forgotten 94.0% Here there is little disagreement; it is, perhaps, surprising that so many informants recognize the subjunctive were (749) and the past tense lay (705) as being correct. There is generally considerable confusion about these forms; though traditional grammar books are clear about the correct forms and use of these, many speakers confuse them. 1^ However, here Vancouver speakers indicate that they know what the correct forms are (which is, of course, not to say that they use them!). 15 Note that lie/lav appear in virtually all Glossaries of Usage and are identified as troublesome; Fowler writes that "confusion even between the words lav and He themselves is common in talk" (Fowler 1965:327). The subjunctive, as mentioned in 4.2.1, is also a difficulty; Fowler says the subjunctive "is dying" (Fowler 1965:595). 122 28.6% of the grammatical terms fall into the second group: TT LITTLE AGREEMENT AMONG SPEAKERS 673 Today he has alreadv drank/drunk. . . d r u n k 47.7% 729 She's a reality") nice lady. rea l ly 61.7% 793 If it was/were warmer, we could go for.. . were 54.2% 689 Marv is sitting between John and I/me. I* 53.7% 693 Just between vou and me/I. I think... m e 50.7% 717 He gave it to Tom and me/I. m e 52.3% 785 To whofm) did you give the book? to whom 43.9% 809 It's me/I! me * 66.7% 709 Mv house is verv different from vours. f r o m 60.5% 817 It has la id / la in there all night. l a i n * 54.8% 745 If we had seen vou. we would have... had seen 50.4% 777 If I had gone home. I would have found it. would have gone* 45.9% Not only are there several forms here for which the S V E N informants attribute correctness to a form traditionally deemed incorrect, but also there is a low rate of agreement about just what the correct forms may be. For three of these forms, the disagreement among speakers reflects some disagreement among scholars; d rank /drunk , real /real ly and different from/than/to are unsettled forms. Although drunk is generally given as the past participle of drink, most resource books give drank as an alternative form: Gage, for example, writes, "Although drunk is the usual past participle, many educated North Americans use drank in speech...In writing drunk should be used, except perhaps when writing dialogue" (Gage 1983:360). Fowler identifies have drank as "formerly not unusual, but now [a] blunder or conspicuous archaism" (Fowler 1965:140). Similarly, real as an adverb receives attention from many grammarians; for example, Fowler identifies real meaning "very" as an Americanism and suggests speakers avoid it. Gage is less quick to * Indicates forms which may be considered incorrect. 123 condemn: "In formal and most written English, real is used only as an adjective...In non-standard and some informal use, it is also used as an adverb meaning "really" or "very" (Gage 1983:936). Baker writes, "Do not use Ireall for very" (Baker 1972:369) and Millward and Flick concur.16 Americans, Britons and Canadians, then, all condemn use of real as an adverb in formal and written English at least. There is also general agreement among scholars from all three countries about the preposition following different: from is recommended most strongly with than and to following. The comment in Gage In formal English, the standard idiom is different from. Informal usage is divided; using from, occasionally to (which is a common British idiom), and more often than. Different than is becoming more common when the object is a clause: The house was a good deal different than he remembered it. (Gage 1983:329) Fowler accepts from and to but suggests avoiding than, which he says should be left for the comparative (Fowler 1965:535). Baker is stronger: Never use rthanl. Things differ from each other. Only in comparing differences could than be used: 'All three of his copies differ from the original, but his last one is more different than the others.' But here than is controlled by more, not by different. (Baker 1972:360) Thus, though the strength of the recommendations vary, there is agreement, and Vancouver speakers have identified the form generally thought to be "correct." 1 6 Recall that Gage is Canadian, Fowler is British, Baker is American and "Millward and Flick" refers to the Canadian edition of an American work. 124 The other forms for which speakers have a low rate of agreement are forms about which scholars agree; these include the objective pronouns, forms of the verbs He. and lay (see note above) and the subjunctive in conditional clauses. Here speakers either know the rules and disregard them, especially for speech, or they haven't learned the rules; there also may be some slow change in progress, a point which will be considered briefly in 4.2.3.2 For these 12 items, then, speakers are dealing with variables which are unsettled; there is agreement among scholars about what the "correct" forms are, but speakers are not necessarily sure about those forms. For a further 16.7% of the grammatical items, there is either little comment by scholars or there is little agreement among them: TTT LITTLE AGREEMENT AMONG OR COMMENT BY SCHOLARS 757 Let's not/Don't let's take the bus. Let's not 97.1% 701 I'm always aren't I/am I not? aren't I? 67.9% 765 We used not/didn't use to go there. never used 75.8% 657 Have you got/Do vou have enough...? Do you have 38.1% 781 Do you like these kind(s) of apples? these kind* 56.9% 721 They sneaked/snuck into the fridge. sneaked 52.1% 821 You'll find it out back/in the back. in the back 52.9% 801 Can you lend/loan me $5 lend 57.7% That 97.1% of speakers think let's not is correct is not surprising in light of the fact that there is no comment about such a phrase in dictionaries or handbooks; perhaps there is no need for comment. There is also no comment about aren't I as a tag question (though there are comments about avoiding ain't). There is little comment about never used other than Fowler's comment that never is acceptable as an emphatic negative (Fowler 1965:388); he does, however, say that used not is "the proper negative form" (Fowler 1965:670), but 125 this is as compared with a present tense form rather than with never. Fowler explains what he considers a difference between British and American (perhaps, North American?) ways of asking whether a person has something. His explanation is worth quoting at length: in the U.S. Do you have a match? 1 don't have a match are idiomatic where our own idiom requires Have you (got) a match? I haven't (got) a match. In British idiom do have is permissible only subject to two limitations: that have is used in a sense different from that of physical possession, and that habit or repetition or general practice is implied. Do you have coffee? means for us is it your habit to drink it; have you coffee? means is there any to make the drink with....The American idiom is...encroaching on ours. (Fowler 1965:136) Quoting C.S. L e w i s , 1 7 Fowler claims the loss of the distinction between the two is regrettable. Fowler also comments on inclusion of got in have you got...?: he acknowledges the condemnation of including got in the phrase but, along with Ballard, argues that "it is not a real error but a counterfeit invented by schoolmasters" (Fowler 1965:227). That North Americans don't comment may indicate that the distinction is not acknowledged at all; with such little agreement among speakers, 1 8 it seems that some discussion or teaching is indicated; consideration of whether the early British distinction is known by many would also be interesting. 1 7 "The language which can with the greatest ease make the finest and most numerous distinctions of meaning is the best....It was better to have the older English distinction between 'I haven't got indigestion' (I am not suffering from it at the moment) and "I don't have indigestion" (I am not a dyspeptic) than to level both, as America has now taught most Englishmen to do, under 'I don't have'" (Lewis in Fowler 1965:136). 1 8 Have you was the second most popular choice, with 35.1% and Have you got followed with 20.1%; 6.7% said both Have you got and do you have are correct . 126 There is also very little comment about these kinds of apples; Millward and Flick say, "kind, sort, type: these words are singular and should be modified by this or that, not by these or those. In the phrase this kind (sort, type) of a the 2. is redundant and should be omitted" (Millward & Flick 1985:469). Others do not comment. There is also very little mention of snuck as the past tense of sneak. The Oxford dictionaries do not include snuck. The Random House Dictionary of English Language lists snuck. identifying it as a "chiefly dialectal" past tense and past participle of sneak. The Gage Canadian Dictionary gives snuck as an alternative informal past form; the handbooks do not include it in either their glossaries of usage or their unusual verb forms. I assume that this is a North American form (snuck is used in informal speech in Canada and 45.8% of the S V E N speakers consider it "correct"). The lack of comment may reflect the low rate of use of snuck in writing. Similarly, out back may also receive no comment because it is more a spoken than a written form. There is considerable comment on lend/loan but there is little agreement. A l l acknowledge that loan is sometimes used as a verb; however, some recommend that such use be kept to informal speech only (Gage, Millward and Flick), some accept it (Messenger and de Bruyn, Random House Dictionary)1^ and others 1 9 Winston Canadian Dictionary for Schools and the OED consider loan as a verb to be an acceptable Americanism; Fowler also identifies it as an A m e r i c a n i s m . 127 condemn it as a verb entirely (Baker, Fowler20). It is small wonder, then, that only 57.7% of the SVEN informants identify lend as the correct form. There remain two items about which there is considerable comment and for which there seem to be Canadian forms. IV CANADIAN FORMS 741 He dived/dove into the water. dove* 73.6% 797 It's already been proved/proven. proven* 75.4% Most reference works comment on proved and proven, generally agreeing that proved is the "correct" past participial form but giving proven as an alternative form. Most, however, claim that proven is "archaic" (OED ), "best left alone" (Fowler) or "an adjective only" (Baker). Proven is recognized as being appropriate "if quoting from Scots law" (Fowler 490) and some think that proven can be used only in a legal sense or legal context. Baker at least implies that proven, as an adjective, refers to "tested by time," whereas proved is simply "successfully tested or demonstrated" (Baker 1972:369). The Canadians, however, simply give proven as an alternative form for proved: Gage does not comment or condemn, neither does Flick. 75.4% of the SVEN informants identify proven as the "correct" form. Neither the Canadian scholars nor the speakers condemn this form; thus, I think has proven may be a Canadianism; whether or not has proven originates in Scots law, in Canada it is a common and acceptable form.21 2 0 Fowler considers that loan, the verb, "was expelled from idiomatic English by lend But it survived in U.S., and has now returned to provide us with a NEEDLESS VARIANT" (Fowler 1965:341, emphasis his). 2 1 It has been my own experience and that of several of my Canadian colleagues that Canadians must learn that proved is an acceptable form; I suspect that many Canadians who do not need to pay particular attention to 128 There is less agreement about dived and dove as the past participle of dive. Most scholars acknowledge both possibilities and recommend dived as the acceptable form. Perhaps in part to attempt to account for both forms existing even while recommending one, scholars have identified some distinctions in use and meaning of the two forms. In Old English there was a strong intransitive verb from which we get dove: there was also a weak transitive verb from which we get d ived. This particular transitive-intransitive distinction has disappeared in Modern English; however, some speakers maintain a distinction between animate and inanimate subjects with dived and dove. Few people maintain the distinction; those who do use dived with inanimate and dove with animate (The submarine dived; The boy dove). (See Scargill 1974:21 for a brief discussion.) The Oxford English Dictionary makes a different distinction, however; it says that dove is used in the sense of "submerged", whereas dived is used in the sense of jumping into or breaking the surface of the water. I am not sure that these distinctions in meaning are known to or maintained by many modern speakers. It does seem, however, that Canadians do not agree with the assessment of dived as the correct form; 73.6% of the S V E N informants claim dove is the correct form. Canadian sources seem to support this assessment, at least to some extent; Millward and Flick's comment is: dived is the preferred past tense of dive, but dove is also acceptable" (Millward and Flick 1985:463). Similarly, Gage claims that "both forms are used in Canadian English for the past tense, though dived seems to be more widely preferred in writing and in formal English. However, dove is language do not realize that proved is correct. Though anecdotal, this experience supports the findings that has proven is a Canadian form. 129 the standard form for many people" (Gage 1983:345). According to The Survey of Canadian English22 slightly more speakers use dove than dived with both human and inanimate subjects. All 73.6% of the SVEN informants who think that dove is the correct form also claim to use dove rather than dived. Thus, it seems that dove is a Canadian form, though it is used elsewhere as well. It seems, though, that Canadians too have to justify the use of dove: on a recent production of CBC Radio's "Gabereau,"23 Vicki Gabereau responded to a recent barrage of criticism about sports commentators' English.24 Gabereau's guest claimed that the distinction between dived and dove is one of rank, that, whereas dove would be used for a non-professional or a non-competitive dive, dived would be used for competitive and professional dives. Thus, speaking of the neighbourhood kids at the lake or the pool, one would say, "Yesterday, they dove into the lake" but, speaking of O l y m p i c athletes, one would say, "Yesterday, they dived into the pool." This explanation (whether valid or not) is at least in keeping with some sense of formal to informal usage; dived, the traditionally approved form in standard English, is reserved for the most formal case, while dove, the traditionally censured form, is reserved for a less formal case. However, it seems to me that dove is simply the form used by Canadians.2^ 2 2 See Scargill (1974:21-23). 2 3 June 18, 1988. 2 4 I believe this arose from the recent coverage of the selection of athletes for the Summer Olympic Canadian teams. 25 As was the case with proved, some of my Canadian colleagues and I had to learn to accept dived. 130 Of all of the above grammatical forms, then, Canadians could be said to have mistakenly identified "correct" forms for only 16.6% of the items; for the following, the form deemed correct is not, strictly speaking, traditionally correct: V CANADIANS "INCORRECT" 741 He dived/dove into the water. dove 73.6% 797 It's already been proved /proven . p r o v e n 75.4% 781 Do you like these kind(s') of apples? these kind 56.9% 689 Mary is sitting between John and I/me. I 53.7% 809 It's me/I! me 66.7% 817 It has la id / la in there all night. l a i n 54.8% 777 If I had gone home, I would have found it. would have gone 45.9% Of these, proven and dove are acceptable Canadian forms and are not, therefore, necessarily "incorrect" in Canadian English. These kind is not clear to me; it seems odd that the Survey of Vancouver English did not test this kind, which I think is a common, standard Canadian form, along with the alternatives these kinds and these kind. Of course, these kind is incorrect, but I am not sure that only Canadians would confuse the forms; I think more testing of these forms needs to be done. The other errors, I believe, are not at all peculiar to Canadians; these are commonly found (in speech, at least) in most speech communities. I think only the most die-hard purists would not accept It's me! in speech in most speech communities (and where would it occur in formal writing?); many would agree that It's I! sounds pretentious. At any rate, I don't think that any of these is a mistake that non-Canadians would not make as well; I do, however, think that proven and dove, might be considered as Canadian rather than deemed "incorrect." 131 What is clear, however, is that Vancouver speakers identify standard forms as being "correct" for 83.4% of the items and that Canadians are no more or less "correct" than any other speakers of English; it is also clear that for many items there is both variation among speakers and lack of agreement among scholars about just what might constitute "correctness." Nor is there reliable evidence26 here that speakers who do identify the correct forms necessarily use those forms or use them consistently. These data, then, do not necessarily indicate that absolute or traditional correctness is essential to Vancouver English; they do support the claim that a notion of correctness is an essential or central feature of Vancouver English. It may be the arbitrary nature of traditional grammatical "correctness" that prevents its being a central feature of standard: that does not, however, prevent there being a notion of correctness that is associated with standard varieties. 4.2.3 Social Factors and Correctness Because a notion of correctness seems central to standard, it is important to understand in greater detail the factors that might contribute to that notion of correctness. Social factors such as gender, age and socio-economic status are generally found to be relevant to linguistic variation and may also affect notions of correctness; because of an association between notions of correctness and education, level of education may also be significant to correctness. Therefore, in order to better understand the nature of notions of correctness as an aspect of standard, I will analyze the responses about 2 6 Most speakers claim to use only those forms that they deem correct; however, the reliability of this particular self-reporting is extremely questionable. 132 correctness by the relevant social factors. To do so, I cross-tabulated the grammatical responses with the factors, gender, age, socio-economic status and level of education. Because 16% of the S V E N informants were teachers, I also considered the relationship between notions of correctness and whether or not a person were a teacher. Statistical significance was measured by the Chi-square test of independence using a 5% significance level. Al l of the social factors but gender are statistically significant in at least one third of the grammatical items. Some of the findings are particularly revealing of the nature of correctness, others of the variables for which there is little agreement. The following report organizes these results by social factor. 4.2.3.1 Gender Gender is generally not statistically significant to correctness, although there may be some relation between gender and notions of correctness. Of the 35 "Don't Know/Not sure" responses, 10 were from men and 25 from women. It is generally thought that women are more conscious of correctness and appropriateness in language than men are . 2 7 These findings that women are more likely to admit uncertainty lend support to that bel ief 2 8 by indicating not so much insecurity, but awareness that there may be more than one possibility; if women are more sensitive to linguistic appropriateness or correctness than men are, then they are likely to be more aware of the 2 7 See, for example, Labov (1966) Trudgill (1974) and Hudson (1980). 2 8 It seems to me that the term gender is too broad here; it is more likely that such things as role are hidden factors in gender that would more accurately or precisely define the correlation. However, it is not possible at this time to study such a likelihood; therefore, I use the term gender here with the proviso that it be considered as a social factor that conflates more precise relat ions. 133 possibilities, including the possibility of error. However, 35 responses are too few to be considered solid evidence; more testing is indicated. The only other indication of gender as a significant factor in correctness is in the cross-tabulations of the variables real/really and between you and me. For these, more women than men recognize the traditionally correct form: Males Females rea l ly 49.2% 74.2% rea l 42.5% 20.8% between you and me 44.4% 57.0% 2 9 between you and I 53.0% 39.6% Whereas 74.2% of females identify really as the correct form, only 49.2% of men do the same; similarly, whereas 57.0% of females identify me as the correct objective form, only 44.4% of the males agree. Although these findings are interesting, one cannot conclude that gender is significant to correctness on the basis of its significant effect on only 2 of 42 variables. Rather, it seems that gender may be significant to notions of correctness but is not clearly significant to assessment of correctness. 4.2.3.2 Age Age, on the other hand, seems to be significant to the assessment of correctness but not to notions of correctness. 3 0 For 15 (35.7%) of the 2 9 Note that Labov (1966), Trudgill (1972) and others have found hypercorrection such as "between you and I" to be more typical of women than men. These results seem to contradict their findings; however, this is an assessment of correctness rather than an indication of usage. It may well be that Vancouver women hypercorrect more than men in speech as others have found; this particular test simply is not a test of usage and, hence, of h y p e r c o r r e c t i o n . 3 0 Though 35 cases are too few to determine statistical significance of a 3 group factor, the distribution of those cases across the age groups seems to 134 grammatical items, age is a clearly significant factor in assignation of correctness; for 12 of those there is gradation by age, indicating a change in progress, and for 3, the youngest group behaves differently from the older two groups. The forms and rates of assessed correctness are as follows: Young Middle Old Do you have 51.4% 32.6% 29.3% Have you 21.0% 40.0% 45.5% real 23.8% 38.8% 32.5% really 72.5% 57.5% 55.0% anyways 17.5% 7.5% 5.0% anyway 73.8% 92.5% 92.5% has drank 51.4% 42.7% 36.4% has drunk 38.1% 46.9%... 58.6% Aren't I? 77.9% 72.6% 52.6% Am I not? 18.3% 21.6% 43.3% sneaked 16.3% 51.3% 88.8% snuck 82.5% 43.8% 11.3% dived3 1 6.3% 17.5% 32.5% dove 93.7% 73.8% 53.8% if it was warmer. 58.8% 42.5% 30.0% if it were warmer. 37.5% 55.0% 70.0% between John and m e 28.6% 39.6% 56.6% between John and I 65.7% 57.3% 37.4% between you and m e 41.9% 49.0% 61.6% between you and I 54.3% 50.0% 34.3% gave it to Tom and m e 32.9% 50.0% 73.8% gave it to Tom and I 67.1% 50.0% 22.5% indicate that age is not significant; 12 of the 35 responses were in the youngest age group; 8 were in the middle group and 15 were in the oldest group. 3 1 Note that dived seems to be passing out of Canadian English; dove as the popular past tense form of dive is becoming more common. 135 It's me! It's I! 82.5% 13.8% 76.3% 23.8% 41.3% 52.5% to whom who to 38.8% 48.8% 30.4% 40.5% 62.5% 15.0% lay down lie down 30.0% 68.8% 21.3% 76.3% 8.9% 91.9% it has laid there all night it has lain there all night 47.5% 35.0% 41.3% 55.0% 22.8% 74.7% The first three, do you have, real(ly) and anyway(s) show either gradation with a noticeable division between the lowest age and the other two or just a division between the youngest group and the others. The others show gradation by age. There is indication, then, that people's assessment of correctness is changing, with older people being more conservative than younger. There is clear indication here that the traditionally correct objective pronouns forms are being replaced as is the subjunctive; 3 2 young people are more likely to deem the traditionally unacceptable past tense forms dove, has drank, and snuck correct and young people are more likely than the old to confuse the principle parts of He. and lay. The regularity with which gradation is found in these 15 items is striking; one might expect to find such gradation for some unsettled items, but not for most. Thus, the evidence suggests that what uncertainty there is about the correctness of these 15 unsettled items is likely to be because of change in progress; the indication, then, is not that Canadians are unsure about correctness but that there is change in progress. Thus, though age is significant to only 35.7% of the grammatical items, it is clearly an important concern for correctness; there is 3 2 This finding confirms Fowler's assessment of the subjunctive as a dying form (Fowler 1965:595). 136 a clear relation between age and identification of traditionally correct forms, with evidence of change in progress for 15 of the unsettled items. 4.2.3.3 Socio-economic Status Socio-economic status is also significant to assessment of correctness for 38% of the items and, for most items, the significance is stratified.33 Socio-economic status is significant in the following forms: SES I SES II SES III SES IV34 Have you got? 24.7% 28.6% 14.9% 9.4% Do you have? 35.1% 34.5% 48.6% 34.4% has drank 62.3% 58.3% 36.5% 10.8% has drunk 27.3% 38.1% 55.4% 75.4% less 31.7% 33.3% 20.0% 10.0% fewer 56.7% 66.7% 76.7% 85.0% lend 45.0% 53.0% 59.3% 73.3% loan 48.3% 41.7% 32.2% 20.0% between you and m e 35.1% 45.2% 55.4% 70.8% between you and I 62.3% 52.4% 40.5% 26.2% It's me! 81.7% 70.0% 71.7% 43.3% It's I! 15.0% 25.0% 23.3% 56.7% to whom 23.3% 39.0% 45.0% 68.3% who to 46.7% 40.7% 35.0% 16.7% behind 56.7% 70.0% 81.7% 90.0% in back of 10.0% 5.0% 1.7% 3.3% Yesterday he laid in the sun. 45.0% 26.7% 20.0% 15.0% Yesterday he lay in the sun. 50.0% 66.7% 73.3% 83.3% 3 3 The importance of socio-economic status to notions of correctness as indicated by "Not sure/don't know" responses cannot be assessed here; there are so few occurrences of the responses that their spread across the four socio-economic groups is inconclusive. 3 4 Recall that SES I is the lowest group and SES IV the highest. 137 lie down lay down 66.7% 28.3% 79.7% 20.3% 75.0% 25.0% 93.3% 6.7% It has laid there all night. It has lain there all night. 36.7% 48.3% 44.1% 45.8% 45.0% 50.0% 23.3% 75.0% there's there are 29.9% 48.1% 8.4% 75.9% 4.1% 75.7% 3.1% 78.5% real really 51.7% 35.0% 51.7% 45.0% 16.7% 75.0% 6.7% 91.7% these kinds these kind 13.5% 79.7% 28.6% 65.5% 44.0% 47.3% 66.2% 30.8% between John and I between John and m e 72.7% 22.1% 69.0% 25.0% 47.3% 48.6% 18.5% 76.9% give it to Tom and I give it to Tom and m e 61.7% 36.7% 66.7% 33.3% 40.7% 59.3% 16.7% 80.0% Only the first of these, Have vou got/Do vou have?, is not clearly stratified by socio-economic status; for this item, the second highest socio-economic group has a higher rate of identification of Do vou have as the "correct" form than the highest group does. For the other forms there is either clear socio-economic gradation or, as for the last 5 items, clear gradation with a noticeable break between either the lowest socio-economic group and the other three (there's/there are) or the lowest and the highest two socio-economic groups.3 •> It could be that education, as a factor in the calculation of socio-economic status, is salient in this variation. It is clear, however, that for these items, 35 Most of the forms for which socio-economic status is significant are those for which there was little agreement among Vancouver speakers about correctness. However, it is interesting to note that behind/in back of is added here; Scargill explains that in back of is common in the U.S., whereas behind is the British form. He suggests that this is a social distinction here (Scargill 1974:36); these findings confirm his suggestion. 138 socio-economic status is significant to correctness and that the relationship between them is regularly stratified. Because of the likelihood that education is significant to grammatical correctness,36 it is worthwhile to test education apart from its role in socio-economic status. There are four tests possible using the SVEN data; an education factor (+. PSE, i.e. plus or minus post-secondary education) can be used to test the significance to both uncertainty and correctness, and a teacher factor can be used to do the same. They will be considered in turn. 4.2.3.4 Level of Education Of the 35 "Don't know/not sure" responses, 19 were made by people without post- secondary education; although this does not represent a sufficient database on which to draw substantial conclusions, the fact that over half of the people who are unsure have less formal education than others does indicate that level of education may be a factor in notions of correctness. The results do indicate that education ought not be disregarded as a factor in correctness or grammaticality as some have suggested. It is the case that for 38% of the grammatical items tested in SVEN, + PSE is a statistically significant factor; the following forms are affected by level of education: 3 6 Traditional grammatical correctness is arbitrary and conventional and, therefore, learned rather than acquired; hence, there is likely to be a direct connection between education and grammaticality. 139 +PSE -PSE has drunk 62.0% 34.8% fewer 82.6% 61.8% rea l ly 80.7% 45.8% these kinds 54.2% 21.3% lend 70.6% 46.9% between John and m e 59.9% 24.7% between you and m e 64.8% 38.0% give it to Tom and m e 66.7% 40.5% to whom 59.3% 31.3% It's I! 40.4% 21.4% b e h i n d 83.5% 67.2% Yesterday he lay in the sun. 78.9% 59.5% d i v e d 3 7 25.0% 13.7% there are 73.2% 65.6% different from 70.2% 51.5% anyway 91.7% 81.7% For these items, those with post-secondary education are more likely to select traditionally correct forms than those without post-secondary education. Although 38% is relatively few of the grammatical variables, the differences between these two groups for these items cannot be ignored; the degree of significance makes it clear that level of education is a factor in correctness. 4.2.3.5 Teachers It seems that teachers are more certain of grammatical correctness than others and that they identify traditionally correct forms as the correct ones at a higher rate than non-teachers do. Only 4 of the 35 uncertain responses were given by teachers; whether or not teachers know or use correct forms, they 3 7 N o t e that for d ived , the majority of neither group choses the traditional form. 66.7% of those with post-secondary eduation chose dove as did 79.4% of those without; thus, though education level seems to be a factor, especially in choosing the traditional form, the traditional does not win out for the past tense of dive. Education does affect assessment of this form, but not to the extent that dove be considered incorrect in Canadian English. 140 are certainly less hestitant about them than others are. They, of all groups considered here, clearly have a notion of correctness. The teacher-factor is also statistically significant to 33% of the grammatical items; the following are those items: Teacher Non- teacher has drunk 70.2% 43.5% fewer 92.1% 67.3% rea l ly 84.2% 57.4% dived 34.2% 15.9% these kinds 72.3% 30.4% between John and m e 80.9% 34.0% between you and me. 83.0% 44.7% give it to Tom and me. 86.5% 46.0% It's I! 44.7% 27.2% lend 76.3% 54.2% there are 83.0% 66.7% to whom 60.5% 40.8% different from 76.6% 57.5% anyway 100.0% 83.7% Again, the rate of significance is noteworthy here; surely, though there is significance for only 33% of the items, the extent of the difference between responses of teachers and of non-teachers indicates that the significance to correctness cannot be ignored. One might well ask why teachers respond any differently than non-teachers. Of course, teachers have post-secondary education; hence they fall into the better-educated group already shown to identify more traditional forms than those with less education do. However, that can only be a partial explanation or the results would be more similar to those of the analysis by education factor; more teachers than others with post-secondary education correctly identify traditional forms. It is common knowledge that one learns best by teaching; however, there is no indication that these are teachers of grammar or even of English. However, it is possible that teaching, regardless of the subject matter, makes one more conscious of 141 language and correctness; teachers are responsible for communicating clearly and for correcting the communication skills of their students regardless of the subject matter being taught. Given that even the youngest age group of SVEN would have completed high school before the emphasis shifted from essay-writing to multiple choice examinations (preferably computer-marked!) and from correctness to creativity, it could be that the SVEN teacher-respondents have both learned traditional correctness and, by teaching and marking, gained a level of certainty about correctness that their non-teacher counterparts do not have. At any rate, the finding that teachers respond differently to questions about grammar than non-teachers do indicates teaching and education are significant to both correctness and notions of correctness. If correctness is associated with standard and if education is important to correctness, as these findings reveal, then the role of education in standard, both in its development or maintenance and in its role in language variation, cannot be ignored. Future studies are needed to investigate in more depth the nature of the relationship among education, correctness and standard. However, it is clear from the above tests that a notion of correctness is associated with standard English. Therefore, the tests of the definition of standard as that variety spoken by the majority of speakers and typified by correctness indicate that both parts of the definition are corroborated, though "correctness" must be qualified. Before attempting such qualification, however, it is necessary to consider the processes through which a variety must have passed for it to have become standard, both for what such a consideration may reveal about standard, correctness and 142 education and about the extent to which Canadian English is a distinct variety of Standard English. 4.3 Processes of Standardization in Canadian English The test of standard using the SVEN data indicates that the definition given here is acceptable with this set of data; however, it is important to determine whether or not the standard identified here is a distinct one. One means of determining whether or not a variety is a standard one or is a distinct one is to assess whether or not it has passed through (or, perhaps, is in) the four stages essential to the establishment of a standard variety. Hudson (1980) identifies four processes in the development of a standard: selection, codification, elaboration of function and acceptance. Testing Canadian English for evidence of the four processes is made complicated by there being two levels to consider—Standard English and Canadian English as a variety of Standard. Both questions that have been raised about Canadian English (i.e. Is Canadian English standard? and Is there a standard Canadian English?) can be addressed by considering evidence of the completion of the four processes: if Canadian English is a variety of Standard English, it will have passed through the four processes just as other varieties of Standard have done; if there is a standard variety of Canadian English, there will be evidence that this variety of Canadian English has passed through the four processes. I will argue in the following sections that there is some evidence in Canada for both. 143 4.3.1 Selection For a language or a variety to have become standard, it must have completed a process of selection, i.e. it will be a variety that has been selected as the one to be developed into the standard. The criteria for selection of a variety as standard vary, but are generally extra-linguistic. Perhaps the best documented standard variety in English is Standard Written English, which is shown to have developed from London English and to have been accepted by about 1430 as the language of official business. 3 8 There is evidence that the early London dialect was an Essex type with some correspondence with Middlesex and Surrey as well; by about 1258 the London dialect had changed in character from a Southern to a more Midland type of dialect. As a Midland dialect, London English represented a middle position between the extremes of the North and South; it represented a kind of compromise. The district was also the most heavily populated of the major dialect areas and was prominent politically and economically. The district was also the location of the two universities; thus, it was prominent in education as well. It follows, then, that a Midland dialect would be selected as the standard for written English. Because London was the capital of England, London English was the variety selected as the standard, both as the basis of Standard Written English and subsequently as the basis of standard spoken English. In the ensuing period, there were no radical changes in Standard Written English, since writing tends to fix forms rather than encourage variation or change. Thus, though English has spread widely from the British Isles, Standard Written English today shows relatively little variation from nation to nation. Canadians, like 3 8 See, for example, Baugh (1957), Samuels (1972) and Williams (1975). 144 Americans and the people of other British colonies, have adopted Standard Written English as the vehicle of official business. This adoption is, in a sense, another level of selection, one which contributes to Standard Written English transcending regional dialect barriers. Thus, it is clear that Standard Written English in Canada has completed a process of selection. The development and selection of standard spoken rather than written English is less clear because spoken varieties are more diverse and less fixed than written ones. It is possible, particularly in accent, to have several varieties of a standard form. Furthermore, it is difficult to recognize what varieties may serve as vehicles of official business or, in de Saussure's words, "serve the whole community" (de Saussure 1966:195). One test of a standard spoken variety is whether or not it is accepted or endorsed by public schools and institutions of higher education; if students are encouraged to adopt a particular accent as being appropriate to the school or to the career for which they are preparing, then that accent has been selected as a standard. Thus, in Britain, speakers of a variety such as Cockney, for example, would be encouraged to leave their accent behind and adopt Standard British English when they pursue higher education; similarly, in the U.S. speakers of strong regional varieties such as those in the deep south may be encouraged to adopt an accent akin to Network English as they pursue higher education.39 It is 3 9 The advisability of teaching (or, as some put it, imposing) Standard English in schools has been discussed at length by educators (see, for example, Trudgill [1984] and Edwards [1979]). Although many are loathe to admit, especially in a democratic country such as the U.S., that there are varieties of language which may be more socially acceptable than others, the continuing controversy about teaching standard English at the expense of regional varieties clearly indicates that some varieties have been selected as standard and others have not. It is not important for my purposes here to identify precisely what those varieties are; I merely show that such a selection process 145 evident, then, that some accents of English in Britain and the U.S. have been selected and developed as standard spoken varieties. In Canada, there is less clear or specific evidence of the preference of some speech over others because of the homogeneity of the Canadian English accent; nonetheless, there has been selection of standard speech within Canada. Although General Canadian (i.e. "that variety spoken from Toronto westward" [Avis 1975:118]) is used by so many speakers in Canada that it is easy to neglect that it is a selected standard form, the censuring of Newfoundland English by the media, some educators and employers,40 for example, indicates that General Canadian has indeed been selected as standard. As Avis points out (1973a), General Canadian is likely to have developed from a dialect of old Upper Canada, specifically in Toronto, the area of most influence in the development of Canada; Toronto English, then, has been selected and developed as the standard, now General Canadian English. Thus, there is evidence in Canada that there has been selection of varieties to become both written and spoken standards; a previously selected form (Standard Written English) has been adopted as the Canadian standard written form, and Toronto English was selected and has developed into General Canadian English, the standard Canadian spoken form. The first of the has taken place. Hence, I refer readers to the discussion elsewhere and to dialect descriptions in each country if they wish to pursue the details of the discussion or of the selected varieties. 4 0 Despite linguistic tolerance in Canada, Newfoundland English is not heard widely beyond Newfoundland itself. Furthermore, it is censured by some; for example, as recently as 1980, a major retail outlet in Vancouver, replaced its telephone receptionist who had a strong Newfoundland accent because her accent did not project the image the firm wanted to project to the general public. 146 processes essential to development of a standard, then, has been completed in Canada. 4.3.2 Codification The second process a standard variety will have gone through is what Hudson calls codification. After a variety has been selected, it will be delineated and described in dictionaries and grammar books; in this way, the variety is fixed and the agreed-upon forms are given. Speakers, then, can learn to use the forms identified as typical of or correct in that variety. Codification generally refers to or is useful for the written standard rather than the spoken, though pronunciation keys in dictionaries and some descriptions of speech also serve to provide the kind of guidelines that are inherent in codification.41 Given the relationship between codification and written standard, however, it is necessary to consider both codification of Standard Written English and of such varieties as British, American and Canadian; once again, if there is evidence of codification of Canadian English as distinct from Standard English in general, then it can be argued that Canadian English is a variety of Standard or, perhaps, that some variety of Canadian is Standard Canadian English. 4 1 Note that there is inevitably some sense of prescriptivism in even the most descriptive dictionaries and grammar books; as soon as speakers seek the conventional form, the work referred to is used prescriptively however the writers may have intended it. I think in the case of dictionaries and grammar books, it is pure folly to pretend that prescriptivism can be avoided; speakers seek answers to the question "What is the correct form?" and they expect lexicographers and grammarians to provide those answers. I will not, therefore, in this discussion of codification, attempt to avoid sounding prescriptive; to do so is not appropriate in this context. 147 4.3.2.1 Dictionaries The earliest complete English dictionary is that of Samuel Johnson, published in 1775; his codification of English did much to fix spelling and usage, especially in Britain. In 1828, Noah Webster's two volume dictionary was published in the U.S. Webster undertook his work, including an earlier, shorter dictionary, because he wanted to set up an American standard of good usage in contrast to the British standard set by Johnson. Thus, both British and American English had dictionaries by the early part of the nineteenth century. Johnson's dictionary was succeeded by the Oxford English Dictionary, published in twelve volumes between 1884 and 1928, with a supplement published in 1933. It is generally agreed that no other dictionary approaches the OED for historical detail and, for many, it remains the final authority. There are newer and shorter dictionaries (many of them Oxford editions) that make the codification of British English readily accessible to many speakers. Similarly, other American dictionaries have been produced. Webster's dictionary, or rather various editions of it, continues to be a popular resource for speakers of American English, while others such as Random House Dictionary of the English Language and Funk and Wagnall's Standard College Dictionary are used as well. Recent editions of the American Heritage Dictionary include notes on usage (the opinions of a Panel on Usage) under entries for which there is disagreement of variant forms; some of these include specific reference to differences between American and British usage, thus recognizing differences between Standard British and Standard American E n g l i s h . 148 Dictionaries of Canadian English are a much more recent development than those of British and American English, and their publication has not been without difficulty. Until quite recently, Canadians used either British or American dictionaries or they had recourse to Canadian editions of American dictionaries. For several generations, Canadian public schools have used the Canadian edition of Winston's school dictionary; there is also a Canadian edition of Funk and Wagnall's college dictionary. However, these Canadian editions differ very little from their American counterparts, generally doing little more than making mention of some spelling differences in the introduction. While these Canadian editions do indicate a need for, perhaps even a market for, a Canadian dictionary proper, they are only a partial or temporary solution to the demand for Canadian works and indicate only the very beginning of codification. Negative attitudes toward the study and description of Canadian English have delayed production of a Canadian dictionary; some thought Canadian English did not warrant its own dictionary, while others thought the audience for a Canadian English dictionary was not extensive enough to make such an undertaking financially viable.42 The following excerpt from Avis' article, "Canadian English," reprinted in the introduction to the Gage Canadian Dictionary , indicates some of the critical attitudes that Canadian lexicographers had to deal with before gaining approval of their undertaking: It has been argued in these pages that there is such a thing as a distinctive variety of Canadian English; yet it should be observed that this distinctive variety is referred to as "Canadian English" and not as "the Canadian language." The fact is that Canadians share one language 4 2 See Avis 1957, 1966a and 1973 for discussions of the problems of justifying and producing a Canadian dictionary. 149 with Britons, Americans, Australians, and a host of other people, both inside the Commonwealth and beyond it. To claim that there is a Canadian language, or, as many Americans do, an American language, is to distort the meaning of the word language for nationalistic purposes. On the other hand, it is a form of blindness to insist, as many do, that "English is English" and that only fools "dignify the slang and dialect" of Canada by giving it serious attention. (Avis in Gage 1985:xxiii) In the 1960's Canadian English began to receive the serious attention Avis and others thought it should have; Gage published a series of Canadian dictionaries. The 1983 edition of the Gage Canadian Dictionary is a revision of the earlier Canadian Senior Dictionaries ; it incorporates material from the Dictionary of Canadianisms on Historical Principles , which was published by Gage in 1967. There are also Canadian school dictionaries for both juniors (i.e. elementary school) and intermediates (i.e. high schools). The recent edition of the Gage Canadian Dictionary includes usage notes, some of which indicate British or American preferences, thus recognizing differences between Canadian, British and American Standards. This codification of Canadian English, then, indicates that another of the processes through which a standard must pass has at least begun in Canada. There are also now small dictionaries of at least two regional varieties of Canadian English; the Dictionary of Prince Edward Island English and the Dictionary of Newfoundland English have both been published in Canada. Though scholars would not argue that either of those varieties is to be seen as the standard for Canadians in general, publication of these dictionaries does indicate that varieties of Canadian English are now being identified and contrasted; rather than Canadian English being compared with only British 150 and American English, different varieties within Canada are now being compared. Thus, this codification provides evidence that Canadian English has gained a stature similar to that of American and British English; since one can describe varieties within Canadian English, it is evident Canadian English is not being dismissed as a variety of either British or American English but is being accorded its own position. The evidence of dictionaries of Canadian English and varieties of Canadian English, then, suggests that Canadian English both is and has a standard variety. 4.3.2.2 Grammar Books There have, of course, been many English grammar books written in the last two centuries; Jespersen's Modern English Grammar, Zandvoort's A Handbook of English Grammar and Curme's English Grammar are only three of many. Most grammar books describe or prescribe (or both) Standard Written English, which has been adopted as the standard written form in most English speaking nations; therefore, this form of codification is evident for most written English. There are also, however, many handbooks for writers, a type of grammar book compiled specifically to help in composition rather than to describe the grammar of a language or to teach grammar; these are a kind of applied grammar book. In the last decade or two, there has been a proliferation of handbooks of English, especially in the U.S.; there is a large market for handbooks that can be used in English composition classes at colleges and universities throughout the U.S. Most major publishing companies address this market; hence, Harcourt Brace, Random House, Little Brown and Prentice-Hall, for example, all have handbooks of English on the 151 market. This evidence of codification is extensive, especially for American English. As was the case for Canadian English Dictionaries, the production of handbooks has been much slower in Canada and has been preceded by a period of production of Canadian editions of American handbooks. As in the U.S., the demand for handbooks in Canada is high; as Peter Desbarats writes in his introduction to the second edition of the Harbrace College Handbook for Canadian Writers : In Canada, the publication of this revised edition could hardly be more timely. In the study and teaching of -English, as in many other academic, disciplines, educators and parents are scrambling back to "the basics" after several decades of "progressive" education. Many of us have felt an entire generation—our children—have suffered from an excess of "doing your own thing" amid exaggerated claims that literacy in print was about to join the study of Latin and Greek among the discarded antiques in the attics of academia. Low levels of literacy in university students confirmed this. Now, with a renewed interest in such basic subjects as English grammar, this book should be welcomed not only by students and educators but by the many adults who feel the need for a "refresher course" in grammar, and for a quick reference to the basic rules of good Canadian English. (vi) This increased demand for handbooks has coincided with the increased recognition of Canadian English as an independent variety of English; therefore, there has been a demand in Canada for Canadian handbooks. Jane Flick in her introduction to the Canadian edition of Celia Millward's Handbook for Writers states the rationale behind producing Canadian editions rather than writing Canadian handbooks: two considerations lessen the desire for all-Canadian post-secondary texts. One is, pragmatically but importantly, the 152 economics of producing books for a relatively small market. The other is the proven excellence of the structure and basic content of certain American textbooks. Happily, some publishers have found a creative solution: bringing out Canadian editions of American texts, special editions that are Canadianized not merely in their place of publication and their cosmetics but in their examples, their references, their spelling, their content, and their emphasis. (Millward and Flick 1985:x) The first stage in this kind of Canadian codification, then, has been the production of Canadian editions. The extent to which a Canadian edition is Canadianized ranges widely; some do as little as add a preface to the Canadian edition and a list of Canadian resource material; others add to those two things some Canadian examples (or, more accurately, substitute Canadian place names in examples or substitute a Canadian topic [such as "The Riel Rebellion"] in a sample research paper), and adopt metric units and "Canadianized" spelling.43 Many also add a discussion of spelling differences, including what is preferred in Canada, in their introduction to spelling.44 The most Canadianized editions include comments throughout (and especially in the "Glossary of Usage") on differences between American, British and Canadian usage where relevant. Some will also include a list of Canadian spelling variants, usually giving the preferred Canadian version first.4-* Thus, there is a range of Canadianization of handbooks. At best, I think, the publication of (and increase in) Canadian editions of American handbooks indicates a beginning of this kind of codification of Canadian English; it is far from complete. 4 3 See, for example, Baker, Ledbetter and Gamache's The Canadian Practical Stylist with Readings. 4 4 See, for example, Hodges and Whitten's Harbrace College Handbook For Canadian Writers. 4^ See, for example, Millward and Flick's Handbook for Writers. 153 There are, however, some original Canadian handbooks in publication now. Perhaps, the best known is Messenger and de Bruyn's the Canadian Writier's Handbook published in 1980. Their aim is to provide not a rhetoric but a handbook or reference book covering the important elements of correct writing: the conventions of grammar, punctuation, mechanics, spelling, and usage that prevail in Canada today. (xix) In doing so, they do not belabour discussion of Canadian English as it differs from others; they do point out usage that is peculiar to Canadian English or is thought to typify Canadian English, particularly in their discussion of spelling. For the most part, they simply present a handbook of Canadian English. This, then, is a central example of codification of Canadian English. Joanne Buckley has recently published Fit to Print: The Canadian Student's Guide to Essay Writing, which also simply presents Canadian conventions without qualifying them. The Department of the Secretary of State of Canada has also produced a Canadian handbook, The Canadian Style: A Guide to Writing and Editing. It does include a discussion of American, British and Canadian spelling differences and makes one reference to American usage in "words commonly misused or confused"46 and is more self-consciously Canadian than the other two Canadian manuals mentioned here, perhaps because it is a government publication whose purpose is, in part, to contribute "to the Department's role of supporting and fostering the use and development of Canada's official languages both within the federal public service and in other sectors of Canadian society" (Williams and Bayerl 1985:21). This publication is further evidence of a growing body of handbooks of Canadian English. 4 6 '"Indian reserve' in Canada; 'Indian reservation' in the United States." (222) 154 There is, then, evidence of codification of Canadian English, in the production of both Canadian dictionaries and Canadian handbooks or grammar books. Thus, two of the four processes through which a standard must pass are evident for Canadian English. 4.3.3 Elaboration of Function The third process which a variety must have undergone in order for it to be a standard variety is what Hudson calls elaboration of function: it must be possible to use the selected variety in all functions associated with central government and with writing, for example in parliament and law courts, in bureaucratic, educational and scientific documents of all kinds, and of course in various forms of literature. This may require extra linguistic items to be added to the variety, especially technical words, but it is also necessary to develop new conventions for using existing forms—how to formulate examination questions, how to write formal letters, and so on. (Hudson 1980:33) To some extent, elaboration of function was complete when Canadians adopted Standard Written English as their written form. Standard Written English has been elaborated to serve all the purposes of the language of nations; legal terms, expressions and precedents have been set, scientific and medical terminology has been developed, and an extensive corpus of literature has been written in English. However, there is also evidence of Canadian English undergoing elaboration of function itself; the most concrete evidence is codification of particular uses of Canadian English. For example, the most significant evidence of elaboration of function within Canadian English is the recent publication of Yogis' Canadian Law Dictionary and of Sanigan's 155 Canadian Words and Phrases—Legal Maxims ; one might think that in a legal system based on precedent and originating in the laws of England, legal language would transcend such national divisions as those between Canada and Britain. However, when standard language is used for a particular function (such as legal), then the particular use interpretation supercedes the general;47 in a legal context, then, one interprets a term in its specific legal sense rather than in its general sense. Similarly, when language is used in a particular situation, then it must be interpreted in a way appropriate to that situation; for example, if a term has one meaning in England and one in Canada, then one must interpret appropriately to the situation in which the term occurs. In Canadian law, just as a legal precedent set by a Canadian judge supercedes a precedent by an American or a British judge, the meaning of a legal term specific to Canadian English is accepted rather than the meaning of the legal term in the U.S. or Britain. That there are enough differences in legal language among Canada, the U.S. and Britain to warrant compilation of Canadian legal dictionaries48 indicates that Canadian English is sufficiently different to justify its study and description. The publication of dictionaries of Canadian legal terms (along with Canadian guides for legislative draftsmen) 4 7 For example, in general usage, infant refers to a very young child or baby; in Canadian legal usage, however, the term refers to people below the legal age of independence (19 in B.C.). In legal documents, then,"The infant children of..." refers to children under the age of 19 rather than to babies only as it would in general usage. It would be inappropriate to interpret infant in its general sense when reading a legal document; the legal use interpretation must supercede the general. 4 8 Note that Words and Phrases—Legal Maxims is continually updated; the most recent addition was published in 1986. 156 also indicates the extent to which Canadian English has undergone its own elaboration of function for legal matters. There is also evidence of Canadian English having been elaborated for use in government, writing, education and the media. For example, Canadian handbooks, such as those mentioned above, are evidence of the development of particular conventions for using forms in different kinds of writing and documentation; the increase in publication of Canadian handbooks and their use in education systems indicate elaboration of function for education. The publication by the government of Canada, The Canadian Style: A Guide to Writing and Editing, is particularly indicative of elaboration of function for bureaucratic purposes. There are also the CBC's guides to Canadian English, which delineate that form of English which is appropriate to public broadcasting in Canada. There is also, of course, a substantial body of Canadian Literature;49 the publication of the Oxford Companion to Canadian Literature is evidence that the Canadian literary corpus is both extensive and recognized as unique. Thus, Canadian English, as distinct from Standard Written English, has been elaborated for most functions typical of a standard. 4.3.4 Acceptance Hudson's description of acceptance is as follows: the variety has to be accepted by the relevant population as the variety of the community—usually, in fact, as the national language. Once this has happened, the standard language serves as a strong unifying force for the state, as a symbol of its independence of other states (assuming that its standard is unique and not shared with others), and as a 4 9 See Klinck's Literary History of Canada: Canadian Literature in English. 157 marker of its difference from other states. (Hudson 1980:33) Acceptance is somewhat more difficult to assess than the other processes, especially for a variety of Standard English; Hudson's qualifying assumption (i.e. "that its standard is unique and not shared with others") encapsulates some of the difficulty. Strictly speaking, since unique is an absolute term, if one could discover even a single difference between Standard American English and Standard Canadian English, for example, then the qualification would hold; however, scarcely anyone would wish to distinguish one variety from another on the basis of a single form. One must consider, rather, whether there are enough differences to justify identifying a variety as unique. It is, I believe, in this area that scholars have disagreed about Canadian English; some have thought that there are not enough differences between Canadian and American English to warrant identifying Canadian English as unique; others have thought that there are. The problem is intensified when discussing standard varieties of English, especially written varieties, because they are fixed, made more conventional through institutionalization, and consequently, are less subject to variation and change than regional varieties; thus, they will be more similar than dissimilar. It is, therefore, difficult to determine both which versions of Standard English are unique varieties and which of those has been accepted. The problem for Canadian English is complicated by the reluctance among scholars to accept this variety as a legitimate subject of study; the extent to which researchers have been inhibited by the resistance to serious treatment of Canadian English is evident even in SVEN. In Section XII of the Survey, attitudes toward Canadian, American and British English are tested; 158 informants were asked to rate their agreement or disagreement with such statements as "A British accent is more pleasant to listen to than an American accent" and "Since the American spelling of some words is much more reasonable than the British, Canadians should use the American rather than the British spelling of these words." Though these statements do not directly elicit language attitudes in the way that a matched guise test would, both are worded positively and both assume that American and British English are extant varieties and that Canadians recognize them. However, the question meant to assess attitudes toward Canadian English is not only worded negatively, indicating doubt or defensiveness in the minds of the investigators, but also is a statement about the existence or recognition of Canadian English rather than about the value of it; respondents had to agree or disagree with the statement, "It makes no sense to speak of a uniquely Canadian English, differing from both 'American English' and 'British English.'" Such a statement does not assume that Canadian English exists or that Canadians would recognize their own variety. Thus, even those researchers most dedicated to describing Canadian English are inhibited by the continuing negative attitude toward such study. This difficulty has arisen, in part, at least, because of disagreement about the origins and development of Canadian English and about the effect of American and British influences on Canadian English. These points of contention are peculiar to scholars rather than to speakers of Canadian English; therefore, they may reflect a seeming lack of acceptance by scholars rather than by speakers. However, each must be considered before acceptance of Canadian English by speakers themselves can be assessed. 159 4.3.4.1 Development of Canadian English There is little agreement among scholars about the source of Canadian English; some, like Scargill (1975), think it developed directly from British English, while others, like M. Bloomfield (1975), think it developed from American English through the Loyalists. Although I will argue that the origins of Canadian English are incidental in determining whether or not Canadian English is a variety of standard English, this lack of agreement about those origins both explains the questioning of Canadian English as a variety of standard English and reflects the uncertain status of Canadian English. There are several events and periods in the early settlement of Canada that are thought to be important to the development of Canadian English. Newfoundland was colonized by the English as early as 1583; in 1713, the Treaty of Utrecht gave Acadia (the Maritime Provinces) to England. Between 1717 and 1763 both British and American settlers moved into Nova Scotia. In 1749, the city of Halifax was founded; at that time there were several thousand British, about one thousand American and one thousand five hundred German or Swiss inhabitants in Halifax. After the Treaty of Paris in 1763, which gave the British control over Acadia, Canada, Newfoundland and Cape Breton, there was very little immigration until after American Independence in 1776; between 1776 and 1812 there was a heavy influx of United Empire Loyalists, raising the population of Upper Canada to about 100,000, approximately 66% of which was American. However, after the War of 1812, immigration from Britain was encouraged; about 300,000 British immigrants arrived between 1820 and 1850. Thus, during the period between about 1783 and 1820, American 160 English was predominant in Central Canada, but after 1820, heavy and steady immigration from the British Isles re-established at least a balance between British and American immigrants. 5 0 Those who argue that Canadian English is a direct descendent of British English cite the very early settlement of Eastern Canada, the early possession of Canada by Britain and several immigrant surges as evidence for their theory; they claim that the rapid increase in British immigration after 1812 must have affected the dialect and that, as institutionalization was undertaken by British rather than American immigrants, 5 1 it is more appropriate to describe Canadian English as a descendent of British English than as a branch of American English. Those who support the Loyalist theory, however, argue that the period during which there were more American than British immigrants in Canada (late eighteenth and early nineteenth century) was a crucial one in the development of a dialect; for example, one claims that "Canadian English...is basically eighteenth century American English modified by other influences, notable among which are Southern Standard English and the English taught by Scots school teachers" (Bloomfield 1975:6). To support their claim, they minimize both the extent of immigration from Britain before and after the influx of United Empire Loyalists and the prestige that would be attached to military, clergy and teachers, most of whom were British. However, one cannot deny that there are many similarities between 5 0 By the mid 1800's, roughly half of Canadian settlers had immigrated from Britain, compared with the two-thirds American immigrants in the early 1800's. 5 1 The military and most of the clergy were British, and schools were established by British on the British system. 161 American and Canadian English that could indicate American origins; Canada is, afterall, a part of North America. Both theories, however, have merit because of the many similarities between Canadian English and both British English and American English. Proponents of the theory of British origins cite the many similarities between Canadian and Northern British English varieties as evidence for their argument; they also point to the strong Canadian r (also a feature in Northern British dialects) as compared with the American northern and southern r-less varieties. On the other hand, proponents of the theory of Loyalist origins argue that Canadian pronunciation resembles American more than British English and that even the distinctively Canadian feature known as "Canadian raising" is also found in Martha's Vineyard . While an argument can clearly be made for either origin of Canadian English, both theories ignore the possibility that the origins of Canadian English, like the variety itself, are simply mixed. Both theories also minimize the possibility that Canadian English has developed in ways of its own, or that some of its development parallels that of American English without necessarily having derived from the American; Canadian and American English may contain independent but parallel innovations. Furthermore, neither theory pays enough attention to the fact that American English itself has its origins in Britain; whether the English that is spoken in Canada came directly from Britain or from Britain through America is in some ways immaterial. Rather, the important point is that all three (British, American and Canadian) are gross regional varieties of a common language—English. Thus, it seems to me that the origins of Canadian English, particularly as being either British or 162 American, are not pertinent to consideration of whether or not Canadian English is standard or is accepted by the people as such. The disagreement is, however, indicative of scholars' indecision about the status of Canadian English and does, therefore, contribute to there being two questions rather than one about standard in Canadian English and to the seeming lack of acceptance of Canadian English as standard by scholars. 4.3.4.2 Influences on Canadian English Similarly, the assumptions about American and British influences on the development of Canadian English contribute to there being two questions rather than one about standard in Canadian English and indicate some lack of acceptance by scholars of Canadian English as standard. Most scholars think that British and American English have always exerted strong influences on Canadian English; studies of Canadian English classify forms as either British or American and discuss the effects of British and American English on Canadian, including consideration of whether Canadian English is more British than American or is becoming more American than British (see, for example, Avis 1957, Scargill 1972, Allen 1975, Gregg 1975). Both the presentation of the material on Canadian English and the material itself indicate that Canadian English is developing in an unusual situation which Scargill calls "the unique situation of Canadian English:" Because of the course of its settlement and because of its geographical situation, Canada has close ties not only with Britain but also with the United States. Canadian English shows divided usage in certain areas of spelling, pronunciation, vocabulary and syntax. (Scargill 1972:13) 163 It is assumed, then, that the unique context affects the development of Canadian English. It cannot be denied that Canadian English has a unique context and that both American and British influences are evident. Not only is there a great deal of contact between Canadians and both Americans and Britons but also the emphasis on what is called "the Canadian mosaic" contributes to the uniqueness of the Canadian context. It is typical of Canadian culture that immigrants retain not only their own language but also their own language variety; though Canadians are expected to learn one of the two national languages, there are many work places in which bilingual foremen are hired to facilitate communication between managers who speak only English and labourers who speak only Italian, German or other European or Asian languages. In such a tolerant linguistic environment, there is even less pressure for speakers of other dialects of English to adjust their accents to more closely resemble Canadian. Furthermore, teachers in the Canadian public school system, some of whom either originate from or have been trained in British systems and use and teach British forms, conventions and variants and some of whom either originate from or have been trained in American systems and use and teach American forms, conventions and variants, bring the two influences into the immediate learning environment of young children; the influences of British and American on Canadian are integrated in the Canadian culture. Thus the two influences on Canadian English that are thought to be very powerful, though foreign, work w i t h i n the Canadian speech community. The Canadian mosaic, then, moves what once might have been foreign influences into the immediate speech community; 164 the cultural tolerance demanded by the mosaic heightens the uniqueness of the Canadian context and could increase the strength of American and British influences on Canadian English. It is not surprising, then, that Canadian English is described as it compares and contrasts with both British and American English; both its origins and development are related to these two national varieties. However, there are some ways in which the consideration of the two influences is distorted or exaggerated. Keeping in mind that Canadians do speak Canadian English,5 2 not British or American, one must question the emphasis on the two foreign influences. To focus on such questions as "Is Canadian English more American than British?" and "Is Canadian English becoming more and more like American English?" is to ignore Canadian English as it is and to oversimplify the two influences thought to exert force on Canadian. Although it is clear that neither British nor American English can be ignored or discounted as influences on Canadian, it is not likely that their influences are the same. On the one hand, however one thinks English found its way to Canada, one can see British English as the roots or the ancestral language of both Canadian and American English; hence, British English may prompt attitudes of loyalty and respect.53 On the other hand, American English may be seen as a competitor, 5 2 Such studies as those done by Allen (1975), Avis (1957 & 1966a), Chambers (1975), Clarke and Hollet (1986), Gregg (1975), Pringle, Dale and Padolsky (1986), Scargill and Warkentyne (1972) and Woods (1979) adequately quantify a distinctive variety of English used by Canadians. 5 3 People's language attitudes are central to language variation and change; positive evaluation of language varieties will motivate adoption of forms of those varieties and negative evaluation of varieties will inhibit adoption of forms from those varieties. See, for example, Ryan and Giles 1982 and Lambert 1972. 165 as a variety from which Canadian English must be distinguished rather than one to which it should be connected; hence, American English may prompt attitudes of disregard and rejection.54 Furthermore, British English, as an origin of Canadian English, has its strongest effect or influence in the past, in the history of Canadian English; geographically the two nations are more distinct than Canada and the U.S. are and there is less direct communication and travel between Britain and Canada than there is between the U.S. and Canada. American English, rather than being primarily the source of Canadian English, is a continual influence through both intermingling of the people of the two nations and widespread, extensive communication, mostly from the U.S. to Canada. Thus, the two influences are different in nature, in frequency and in intensity; Canadian English, then, is not caught between two equally powerful influences. Rather, American and British influences, in different ways and at different times, operate along with the other influences such as age, gender, socio-economic status and function found to be significant to language variation and change. Thus, though the Canadian context is unique, the American and British influences are simply two factors in the development of Canadian English, not the only or the main motivators of that development. While the tendency to describe Canadian English as it compares and contrasts with British and American English is natural, a full description must move on to look at the variety itself and at differences within the variety. 5 4 Section XII of the SVEN database shows that Canadians do evaluate British English positively and American English negatively; these findings are confirmed by Woods (1979) and Warkentyne (1983). See also Pringle (1985). 166 Investigation of Canadian English is relatively new and is only now developing beyond the early stages; recent studies of varieties within Canada55 indicate the beginning of scholars' acceptance of Canadian English. Thus, though it is clear that the early tendency to consider Canadian English primarily as it differs from or is similar to British and American English has contributed to uncertainty about the status of Canadian English as a variety of Standard English, many scholars now accept Canadian English as a distinct variety of Standard. Thus, the lack of agreement about the origins of Canadian English and the focus on (and, perhaps, exaggeration of) British and American influences on Canadian English both lead to the possibility of asking whether Canadian English is a variety of Standard English (or even, some might suggest, a variety in its own right). Although it seems clear that the question is unjustified (or accusatory), it is evident that those who insist that Canadian English is merely a variety of American English, for example, are challenging the acceptance of Canadian English as a standard variety. However, those who have gone on to study both Canadian English as it is distinct and varieties within Canada clearly accept Canadian English. 4.3.4.3 Acceptance of Canadian English The reluctance of some scholars to accept Canadian English is not reflected in the speakers' acceptance and use of the variety; not only do Canadians 5 5 See, for example, those done by Clarke (1981) in Newfoundland, Gregg (1984a) in Vancouver and Pringle, Dale and Padolsky (1986) in the Ottawa Valley. 167 continue to speak Canadian English in spite of American and British influences right within the culture, but also they identify and even defend their own variety. In spite of the bias inherent in the question about Canadian English in S V E N , the informants indicated an awareness, and even a defence, of Canadian English. 58% of the informants disagree with the statement (i.e. think there IS a variety that could be called Canadian English), 14% are neutral and 28% agree. However, the responses are stronger than they first appear; the tendency in rating scales is for responses to cluster around the neutral category; for this question, however, responses clustered at the extremes. Of those who disagree, 38% do so strongly and only 19% do so moderately; of those who agree, 18% do so strongly and only 10% do so moderately. Thus, despite the question being worded negatively, it is clear that most speakers accept and even defend Canadian E n g l i s h . 5 6 This, then, is further evidence that Canadian English is accepted. Furthermore, there is now enough quantification of Canadian English as it is distinct from American and British English to identify it as a unique variety of English; the findings specify the differences that have made people intuitively recognize Canadian speakers as being neither British nor American. Both the usage of Canadians and the recognition by others that the 5 6 Results of tests are, of course, only as strong as the weakest part of the test. In the case of the attitudes and awareness section of S V E N , the weakness is the methodology used; self-reporting using rating scales is not the best way to assess language attitudes and the questions were not worded clearly. Thus the figures may not accurately reflect people's attitudes; however, as these results are not invalid, only weak, and they confirm what one observes in usage and confirm findings in other, more reliable studies (see Warkentyne 1986), it is reasonable to use them as an indication of acceptance of Canadian English as standard. 168 speakers are Canadian (or, in some places, depending on how far from either Britain or the U.S. the speaker is travelling, that the speakers are not American, not British or not either) are evidence of acceptance of Canadian English. Furthermore, the use of General Canadian rather than, say, Newfoundland English in the media, legislature, schools and in the general business of the nation, indicates acceptance of one variety of Canadian as standard. Although in an age of science and technology people assume that the only proof of something is statistical and is found through controlled experimentation, to discredit what one knows and sees evey day because it has not been tested in a lab and the results published in a respectable journal is not to be encouraged. One must keep in mind that scientific experimentation might confirm or specify what one intuitively knows is the case but it is never to replace or to confound common observation. Rather, we must accept what is commonly observed as well as that which has been confirmed in experimentation. Speakers' use and identification of General Canadian, then, are evidence of acceptance; Canadian English has attained a level of acceptance that is typical of standard. Thus, there is evidence of completion of selection and acceptance in Canadian English and a growing body of evidence of codification and elaboration of function. The evidence indicates clearly that Canadian English is a variety of standard English; there is also some evidence that there is a standard variety within Canadian English. Clearly, then, the standard identified in this analysis of the SVEN database is distinct. 169 4.4 Conclusion This consideration of the processes of standardization in Canadian English and the tests of standard using Canadian data, then, reveal a good deal about both the definition of standard and the status and nature of Canadian English. Not only do they clarify the extent to which Canadian English has undergone the processes of standardization but also they provide information that is helpful for reconsidering the definition of standard. 4.4.1 The Definition In light of the findings, the definition of standard as that variety used by the majority of speakers and typified by correctness can be refined. The tests show that the majority of speakers agree on most of the correct forms; however, "used by the majority" on its own does not distinguish standard. If one were describing the language of, for example, a counterculture that defined itself in part by the use of a particular variety of a language, the majority of the speakers in that community would agree on particular forms; however, the forms would not necessarily be standard—they would be typical of that community and used by the majority, but not standard beyond that community. There must, therefore, be further specification of standard such that it would not mean simply normal or typical of any single speech community. To some extent, addition of "typified by correctness" provides the necessary specification; however, it is neither accurate nor precise. Although there is clearly a notion of correctness associated with standard, it is not clear that standard is necessarily typified by traditional correctness. At best, I think "typified by correctness" may be a peripheral 170 feature of standard, while "associated with a notion of correctness" may be a central feature; therefore, a more accurate qualification may be "and associated with a notion of correctness" or "and thought to be correct." However, it may be the case that even "notion of correctness" is imprecise. The tests show that education is a statistically significant factor in correctness; as education is also a central aspect of the codification and the elaboration of function of language that are seen to be essential to standardization, it may be the case that a notion of correctness is an important feature more because it implies education, codification and elaboration of function than because it is itself essential to standard. Traditional correctness is established or fixed by codification and is then taught or perpetuated by education systems; notions of correctness arise, in part, at least, from both the establishment of correctness and the adoption of and instruction in that "correct" variety by such institutions as schools. Thus, both correctness and notions of correctness are bound to education as a part of (or instrumental in) the processes of establishment of standard varieties. Precisely what about correctness or even a notion of correctness might be essential to definition of standard remains unclear; thus, more tests and consideration of the nature of correctness or notions of correctness as delineators of standard are indicated. However, the preliminary indication based on the analysis of Canadian data is that refinement of the definition may take into account the processes of standardization: selection, codification, elaboration of function and acceptance. Since these processes are central in the establishment of a standard variety and can be used to determine the status of a variety as being 171 standard, they might be taken as essential features of standard and, therefore, possible elements in an acceptable definition of the term. It is possible that the very general definition of standard as "the approved model" is more accurate than "associated with notions of correctness" as the qualifier of "used by the majority" because "approved model" alludes to the process of acceptance. However, "approved model" is too general in that there is no indication of who or what has given the approval; just as "used by the majority of speakers" could describe a nonstandard variety used by the majority of speakers in a counter-cultural group, "approved model" could be approved by the leaders of such a community. Furthermore, "approved model" only alludes to the processes, whereas a notion of correctness is included in the process of codification. Hence, "approved model" is not only less accurate but also less precise than "associated with a notion of correctness" is. The preliminary indication that education is significant to "correctness" and the connection among education, correctness and the processes of establishment of a standard variety suggest that a more precise definition might include the processes of establishment rather than one factor that develops from (or with) those processes. Hence, a more precise and accurate definition of standard might be "that codified variety used by the majority of speakers for conducting the business of the nation." I assume here that without selection having taken place, codification would not be possible, that "used by the majority of speakers" indicates the variety has been accepted and that "conducting the business of the nation" encompasses all aspects of elaboration of function; thus, this definition embodies all four processes of establishment of a standard and respects those aspects of standard that were tested against the S V E N data and found essential to Vancouver English. 172 Until such time that tests have been conducted which reveal more clearly the relationship among education, correctness and the processes of establishing standard varieties, particularly the relative salience of each as aspects of standard varieties, I believe this refined definition is more useful and precise than the earlier one; it delineates standard and it eliminates the emphasis on the little understood element "correctness." It also includes those processes through which a standard is established. Thus, though more tests of and work on standard are clearly indicated for better understanding and possible further refinement of the definition, these tests using Canadian English indicate that the original observational definition of standard as "that variety used by the majority of speakers and typified by correctness" should, for accuracy, at least be revised to "that variety used by the majority of speakers and associated with notions of correctness." Even more precise is the following: that codified variety used by the majority of speakers for conducting the business of the nation. 4.4.2 Canadian English The above consideration of standard in Canadian English provides evidence both that Canadian English is a variety of Standard English and that there is a standard variety of Canadian English. The extent to which General Canadian English has been codified, elaborated for specific functions and accepted as the variety of Anglo-Canada removes doubt about its status and about the validity of using Canadian data as a basis for generalizing about English. The evidence of the extent to which the majority of speakers select standard forms 173 (or the same forms) supports the observation and quantification of the unique homogeneity of Canadian English; the geographical and social homogeneity is complemented by majority agreement for almost 80% of the SVEN variables by an average of more than 75% of speakers. Canadian English, then, is a variety of Standard English, has its own standard variety (General Canadian) and is typified by a unique kind and degree of homogeneity. On the one hand, the similarities between Canadian English and other varieties of Standard English are sufficiently great to allow legitimate testing for those varietal properties that distinguish standard and prestige. On the other hand, the differences between Canadian English and other varieties of Standard English are particularly illuminating when helping us to formulate maximally viable definitions of prestige and standard. This work will, therefore, conclude with a brief reconsideration of prestige and standard in light of these findings. C H A P T E R 5 C O N C L U S I O N : R E C O N S I D E R A T I O N O F S T A N D A R D A N D PRESTIGE The initial two surveys of use, one general, the other specific to the Canadian context, reveal a certain confusion about the meaning of the terms standard and prestige as these have been employed in sociolinguistic and dialect theory and studies. In discussing these various definitions, we were able to locate possible sources which contribute causally to the terminological disarray. Subsequent tests on Canadian data using provisional definitions of standard and prestige were intended to further clarify their meanings. These tests have revealed important theoretical points about the terms and their use, about prestige, about standard and about Canadian English. I will summarize these and then conclude with suggestions for further research. 5.1 The Terms The terms standard and prestige have often been used interchangeably; they can and should, however, be kept distinct. Prestige has here been defined as that variety spoken by the highest socio-economic group and emulated by others. Although there is little or no evidence in Canadian English of the realization of such a concept, evidence in British English indicates that the definition is acceptable. Standard has been redefined here as that codified variety used by the majority of speakers for conducting the business of the nation. This definition is appropriate for General Canadian, the standard variety for Canadians. It is likely that it is also appropriate for other standard varieties, but future studies will be necessary to confirm this as well as to substantiate the redefinition. However, it is clear that standard and prestige can be distinguished; whereas a prestige 175 variety is used by the highest socio-economic group, a standard variety is used by the majority of speakers. Dialectally and contextually, there have been some differences in the use of the two terms; some scholars, having concrete evidence of both a standard variety and a prestige accent, have used the terms precisely and distinctly. Others, having no concrete evidence of linguistic realization of a prestige variety, have used both terms loosely. However, sociolinguistic and dialect studies, especially of standard varieties, are necessarily contrastive; therefore, it is essential for clarity to use the terms precisely and distinctly even where no prestige variety has been realized linguistically or socio-linguistically in a particular speech community. 5.2 Prestige The absence of a clear prestige in Canadian English raises questions about the mechanism of variation and change. Prestige is thought to be a main motivator in sociolinguistic variation and, subsequently, in change. However, since one observes variation and change in such varieties as Canadian English in which there is no evidence of a distinguishable prestige, it seems likely that the motivator for variation is a concept or notion of prestige rather than a concrete, identifiable variety or set of forms realizing that concept. One indication of prestige as a motivator of language variation is hypercorrection (i.e. the over-generalization of a rule by speakers of lower classes, especially in careful speech, in their attempt to emulate the prestige group). Both Labov (1966 passim ) and Trudgill (1974:90-132 passim ) found hypercorrection in their surveys of New York and Norwich respectively. In their surveys hypercorrection was indicated by cross-over patterns (i.e. the tendency of a middle socio-economic group, in their striving to be correct or to get ahead, to 176 have a higher rate of use of some forms than the group above them on the socio-economic scale); Labov found a cross-over pattern between the highest two socio-economic groups and Trudgill found one between the working classes. In the Vancouver Survey, no cross-over patterns indicate hypercorrection in the lower groups as they attempt to emulate the prestige group; rather, one finds some hypercorrection in the highest socio-economic group. For example, the highest socio-economic group adopts an aspirated III intervocalically in the most careful speech; since the rule in Canadian English is that ft/ is aspirated syllable initially rather than intervocalically, this is evidence of hypercorrection on their part. The highest socio-economic group, then, behaves as if there were prestige forms other than their own to emulate; this is evidence both of the lack of an established upper socio-economic group and of motivation for language variation that is typical in communities in which there is a prestige group or variety. Vancouver speakers apparently have some notion of prestige, though they do not have a realization of that prestige to imitate. It could also be the case that speakers simply avoid stigmatized forms in the absence of clear prestige forms. It is often the case that people know or recognize what they do not like or will not accept before determining what they do like or will accept. Thus, the absence of an established socio-economic hierarchy and the absence of a clear prestige form in Canadian English may indicate a stage of dialect development during which speakers avoid stigma rather than emulate prestige. An effect of avoiding stigma, or rather, of over-avoiding stigma, would bear striking and perhaps indistinguishable resemblances to hypercorrection; only the motivation and direction of over-t. 177 avoidance would be different from those of hypercorrection. Thus, the effect on language use of avoiding stigma would closely resemble that of emulating prestige, especially in cases where there are two complementary forms, one desirable, the other not. It is not clear, then, that anything must function as the prestige even when one finds evidence of "hypercorrection;" in early stages of dialect development, stigmatization may produce a similar effect. However, it is also possible that stigmatization serves to confirm a notion of prestige as a motivator of variation or that the two may work together to influence language variation. It is important, therefore, to recognize that prestige as a motivator of variation and change does not function in any simple way nor need it be concretely realized to be influential. The concept of prestige (either alone or in conjunction with identifiable stigmatized forms) may motivate variation; a linguistic realization of that concept is not necessary as a motivator. 5.3 Standard While the concept of prestige alone may serve as a motivator of variation and change without being linguistically realized, the tendency is to look for something concrete, a realized concept, that serves to motivate variation and change. It is sometimes thought that standard serves instead as a motivator of variation and change, since neither prestige nor standard is stigmatized. The findings here indicate that it would be more accurate to say that, in the absence of a clear prestige such as RP, standard may provide the notion of prestige that is central to variation and change. In Canadian English, the standard variety, General Canadian, seems not to be the most prestigious variety; tests such as those done by Warkentyne (1983), Gregg (1984a) and 178 Woods (1979)1 show that British English is the most highly evaluated variety in Canada; Canadian English follows, with American English being the least highly evaluated. However, work such as that of Clarke (1982) suggests that, although speakers of British English may be most highly evaluated by Canadians on status dimensions, such is not the case with respect to the solidarity or personal attractiveness dimensions. In other words, British speakers do not appear to be uniformly positively evaluated by Canadians. Thus, British English seems to be more prestigious than Canadian English but it does not serve as the prestige variety; it is neither spoken by the highest socio-economic group nor emulated by other groups. As Warkentyne claims, British English "is so distant and vaguely understood that it is not a force for change in Canada" (Warkentyne 1986a:172). Given the distance between British English and Canadian English, the lack of uniformity of the evaluation of British English and the lack of emulation of British English by Canadians, I would argue that, though British English is prestigious, it does not provide a model for even the concept of prestige in Canadian English.2 Rather, the approval and acceptance that are a part of the standard (General Canadian) provide the notion of prestige necessary to variation and change; in this sense only, then, standard may function in Canadian English as prestige does in British English; the two are not, however, synonymous. 1 See also Pringle (1985). 2 See Trudgill (1986a) for a discussion of the essential nature of face-to-face interaction in language change. 179 5.4 Canadian English It is clear from this study that Canadian English is a variety of Standard English; it is also evident that there is a standard variety (General Canadian) within Canadian English. The study also confirms the commonly held view that Canadian English is typified by a unique homogeneity; it is a distinct variety of standard and, as such, is valuable for studying differences as well as similarities. These findings reveal that the homogeneity of Canadian English extends beyond geographical to include social and, to some extent, stylistic parameters; thus, even while the study illustrates that Canadian English is standard and as such can be seen as representative of standard, it also confirms the uniqueness that distinguishes Canadian English from other varieties. Therefore, Canadian English is shown here to be useful for studying both concepts thought to be common to standard varieties and those thought to typify Canadian itself. 5.5 Future Studies The evidence here that our study and understanding of standard have been too general indicates a need for further study; furthermore, these findings suggest that future studies be more specific than those conducted in the past. Early sociolinguistic studies have been immeasurably more specific and concrete than any work done before Labov's innovative study in New York City; however, the sociolinguistic factors that he identified are very broad. Although it has been well established that socio-economic status, age and gender are significant to sociolinguistic variation, this study indicates that those general factors must be broken down to reveal and test their more 180 specific elements. For example, future socio-dialect studies in all contexts should include an education variable, distinct from the education factor implicit in the socio-economic status of the informants; such inclusion will facilitate testing the effect of education on sociolinguistic variation. The indication here that education is significant to sociolinguistic variation suggests, then, that socio-economic status is too general and must be broken into its components and each component tested for salience in language variation and change. Similarly, age and gender must, in some instances, be considered as general rather than specific; gender may mask such things as roles in situations and relations and age may mask such things as loyalty to or familiarity with other varieties of English.3 The identification of education as a factor in correctness and grammaticality, then, leads to the recognition that the time has come for sociolinguistic studies to undertake identification of the salient aspects of the traditional social variables and to conduct studies using the more specific social variables. Only as studies move from the general to the specific, will our understanding of the mechanism of sociolinguistic variation increase; it is, therefore, important that future sociolinguistic studies in Canada and elsewhere test variables more specific than age, gender and socio-economic status. The evidence in this work suggests beginning with a consideration of education (income, occupation and place of residence) rather than general socio-economic status. 3 This is particularly important in studies of Canadian English as American and British influences are thought to be significant to variation within Canadian English; in this context, then, what appears to be age may more accurately be national loyalty, for example. Given the history and development of Canadian English, these factors are likely to be as important in Canadian English variation as a simple age factor is. 181 5.5.1 Canadian Studies Given the absence of a clearly established upper socio-economic group in Western Canada, it is, perhaps, too soon to conduct further studies of prestige in Canadian English, at least in the west. However, studies of standard are recommended; Canadian studies are needed to further investigate the role of education in standardization, perpetuation of the unique blend that constitutes Canadian English and in establishment of the homogeneity that typifies Canadian English. It seems that neglecting direct studies of the effect and role of education on language variation and development has contributed to there being little concrete and specific understanding of the relation among education, variation and standardization. There are more vague notions that there is a connection than there is evidence of how direct and how strong that connection might be. I recommend, therefore, that studies begin by testing the effect of particular trends in public education on such concrete categories as traditional grammatical correctness. For example, there was a trend in Canada (and, I believe, in the U.S.) beginning in the early 1970's and extending to the mid 1980's to de-emphasize correctness and to emphasize creativity. Not only was traditional grammar not taught in public schools during that period, but also teachers did not correct spelling, grammar or punctuation at all; they gave positive feedback on the creativity or on the content of students' work. Thus, there is a group of people who have not had traditional education in grammar and the rudiments of English. One could, then, study the effect of that particular education and the lack of it on language variation by comparing a group of people whose education did not include traditional training with those whose education did. Such a study could provide specification of the effect of one aspect of education on 182 language development; in doing so, it could provide the basis both for further study of the role of education in a more general sense and for the expansion and elaboration of an understanding of the role of education in establishment and development of standard varieties. Studying the effect of one specific, directly related aspect of education, such as teaching grammar, on language variation and the development of standard would provide a solid basis on which to develop a more complete consideration of the effect of general education. Although it is clear that the blend that constitutes Canadian English arises out of the origin and development of Canadian English, it is also likely that the blend is perpetuated because of the influences of American and British within Canada, especially within the education system. Because there are both Americans and Britons teaching in Canadian schools, the blending of forms that typifies Canadian English is ongoing. Canadian children are, probably unintentionally, being taught both American and British forms, not Canadian English and the blend that typifies it, but both American and British forms; a teacher trained in Britain may insist that a British form is correct, another trained in the U.S. may insist that an American form is correct. It is not just that Canadian students, then, cleverly learn to use, for example, -our when writing for one instructor and -or when writing for another (though they do that), but that they may be instructed to use, say, -ize rather than -ise. but -re rather than -er. In this way they are taught different forms that blend to comprise Canadian English without knowing that it is a blend or that there are alternatives. The blend that is Canadian English, then, is perpetuated quite unselfconsciously. It is important, therefore, to test the influence of education 183 on the perpetuation of Canadian English as a blend of British and American forms; rather than assuming that the influences lie mostly in the origins and development of Canadian English and in exposure through media to American and British forms, the extent to which the influence within the public school system perpetuates the blend should be examined. Testing the effect of education on the blend that constitutes Canadian English could be incorporated in a study of the effect of education on standardization; analyzing the extent to which the mix is perpetuated through education could begin as a component of the larger study. Not only should the group that has been taught traditional grammatical correctness (either directly or by being expected to write and speak correctly) be studied for the effect of that teaching on usage, but also it should be analyzed for whether it has been taught a mixture of American and British forms, how implicit or explicit that teaching has been and how effective it has been. One study could thus provide inroads to a better understanding of the role of education on the development of standard English and on the perpetuation of the mix that typifies Canadian English. It is also important to consider the extent to which education has been a factor in establishing the homogeneity that typifies Canadian English. Education is thought to be one factor in dialect levelling; as students attend school and learn standard English along with their own varieties, the learned standard variety exerts control on their native spoken varieties. As a result, their spoken varieties become more like the standard and less distinctive; in some cases, the standard learned at school replaces the spoken variety. Thus, education has a levelling effect on dialect variation. In North America, where public education has been available to a wide range of the socio-economic 184 spectrum since very early in the countries' settlements, there is less dramatic dialect variation than there is in Britain, where education was available to only the elite in the early settlement period when English dialects were developing. Therefore, the diversity that typifies dialect variation in Britain is not as striking in North American countries where education has been a right for all since early in the nations' developments. Although the connection between education and dialect levelling seems obvious, the effect of education on dialect levelling has not been directly studied. To some extent, the effect of education on dialect levelling has been eclipsed by the belief that mass communication has been instrumental in levelling. However, it seems to me that, to some extent, the effect of television and radio on dialect development has been exaggerated. Although the aural media transmit general (and standard) varieties to speakers who would otherwise be isolated and in the process of developing their own varieties and in doing so may provide a model for speakers to emulate, there is no interaction between the hearer and speaker that would make the media variety seem to be vital or practical to the hearer. Furthermore this aural medium does not provide an environment in which the hearer (when he took on the role of speaker in a conversation) might actually imitate the speaker he hears in the media. Hence, rather than being instrumental in language development and change, mass communication is more likely to exert control. Because the language varieties generally used on radio and television are standard varieties, and standard varieties exert control on the amount of change or divergence there is likely to be in regional varieties, at most it is likely that extensive television viewing and radio listening would inhibit change away from the standard, but would not necessarily initiate or motivate change toward that variety. 185 Furthermore, many dialects other than the general and standard are transmitted through the aural media; many popular programmes, especially both American and British soap operas that are viewed each week by thousands of Canadians,4 transmit regional and social variation that would not otherwise be heard by many Canadians. Hence, though mass communication may support or contribute to dialect levelling, it is not likely that it would be the cause of widespread levelling toward the standard. As Trudgill argues "linguistic accommodation to salient linguistic features in face-to-face interaction is crucial in the geographical diffusion of linguistic innovations" (Trudgill 1986a:82); thus, it is more likely that only direct exposure to standard varieties would contribute to levelling. Exposure through education is direct and active; there is interaction and intercommunication between speakers of standard and those learning it or learning to use it more effectively. Hence, it is likely that education is much more powerful in dialect levelling than mass communication is. Therefore, the effect of education on dialect levelling should be studied. If the homogeneity that typifies Canadian English is an extreme in dialect levelling, then the Canadian context ought to be studied as it has contributed to that homogeneity. It is an over-simplification to explain the homogeneity in terms simply of the settlement pattern of Canada; of course that pattern accounts for much of the homogeneity, but the role of education at least as a part of that settlement pattern must be considered also. 4 The SVEN questionnaire included questions about TV viewing and radio listening; the data gathered indicates that people spend more of their TV viewing hours watching programmes that include very casual dialogue, much of it in regional varieties rather than general or standard English, than they do watching the news and documentaries that are broadcast in general and standard English. One need only flip from channel to channel during prime time viewing hours to sample the many varieties that are now being transmitted to viewers. 186 It is important, therefore, to consider the role of education in both the development of the homogeneity of Canadian English and the development of a blend of British and American forms that typifies Canadian English. 5.5.2 General Studies Studies of a general nature, or in contexts other than the Canadian, are also indicated. Some contrastive studies are necessary not only to test the applicability of findings from one context in another but also to increase understanding of the concepts themselves. For example, comparison of the importance of education to correctness and language development in Canada and the U.S. is necessary if we are to understand the effect of education on dialect levelling. These two North American countries have similar (though not the same) development patterns and periods and yet the dialect variation in the two countries is quite different. Contrastive studies are needed, therefore, to determine the difference that is significant to language variation and development in each country. For example, it could be that the same influence that contributes to Canadian English continuing as a blend of American and British forms may be significant to the extent of homogeneity that typifies Canadian English; contrasting similar American and Canadian speech communities may reveal the absence of a similar influence in the U.S. Just as Canadian and American English have been contrasted, so too must the influences on variation in the two contexts be contrasted if we are to more fully understand such things as standardization and levelling. These early tests using Canadian data indicate several possibilities for consideration and confirmation in other contexts. The definitions themselves 187 should be tested against data from different contexts; the role of education in correctness should also be tested in other contexts (as well as in Canada). Studies beyond Canada are also needed to expand the contexts in which something may be said to be the case. Testing definitions of standard and prestige using Canadian data, for example, is a good beginning. Similar tests in other communities are necessary, however, to extend the contexts and generalize from the findings. 5.6 General Conclusion Thus, this consideration of standard and prestige in general and in Canadian studies illustrates that, though our understanding and analysis of these concepts have increased sharply since the advent of sociolinguistic and socio-dialect studies in the 1960's, more study and analyses are necessary for refinement of definitions of the terms and for understanding of the development and role of the concepts in language variation and change. This work reveals that Canadian studies are both acceptable and important for expanding that understanding, and that contrastive studies are needed. Furthermore, it indicates that future studies must include more specific social variables than those established as significant by early sociolinguistic work; if our understanding of the concepts of standard and prestige, especially as aspects of the mechanism of sociolinguistic variation and change, is to be refined, the model for testing socio-dialect variation must also be refined. Future socio-dialect studies must move beyond the general variables, age, gender and socio-economic status, and consider their more specific components such as familiarity or loyalty, role in a speech situation and education. 188 Bibliography Abercrombie, D. et al., editors. 1964. In Honour of Daniel Jones. London: Longmans-Green. Alatis, J.E. and G.R. Tucker. 1979. Language in Public Life (GURT 79). Washington, D.C: Georgetown University Press. Alexander, Henry. 1939. "Charting Canadian Speech." Journal of Education (Nova Scotia), 4th Series, 10, 457-458. . 1940. "Linguistic Geography." Queen's Quarterly 47, 38-47. . 1941. "Collecting Canadian Speech." Queen's Review 15, 45-47. . 1951. "The English Language in Canada: An Essay Prepared for The Royal Commission on National Development in the Arts and Sciences" ("Massey Report") In Royal Commission Studies. Ottawa: King's Printer. 13-24. . 1955. "Is There a Canadian Language?" CBC Times, February 27. 2-3. Allen, Harold B. 1973-6. The Linguistic Atlas of the Upper Midwest. Minneapolis: University of Minnesota Press. and Gary N. Underwood, editors. 1971. Readings in American Dialectology. New York: Appleton Century Crofts . Anshen, Frank. 1978. Statistics for Linguists. Rowley, MA: Newbury House Publishers. Argyle, M. 1969. Social Interaction. London: Methuen. Arlotto, Anthony. 1972. Introduction to Historical Linguistics. New York: University Press of America. Austin, J.L. 1962. How to Do Things With Words. Cambridge: Harvard University Press. Avis, Walter S. 1957. "Canadian English Merits a Dictionary." Culture 18, 245-256. . 1965. A Bibliography of Writings on Canadian English: 1857-1965. Toronto: W.J. Gage. . 1965a. "Problems in the Study of Canadian English." In Communications et Rapports du Premier Congres International de Dialectologie Genirale. Louvain. 181-191. . 1966. "Canadian Spoken Here." In M.H. Scargill and P.G. Penner, editors, Looking at Language. Toronto: Gage. 17-39. 189 . 1966a. "Why a Canadian Dictionary?" Speaking of Dictionaries 1. Toronto: Gage. 1-4. . 1969. "Canadian English, Part II of Linguistica Canadiana, edited by J-P Vinay." Canadian Journal of Linguistics 15, 70-73. . 1972. "So eh? is Canadian, eh?" Canadian Journal of Linguistics 17, 89-104. . 1973. "Problems in Editing a Canadian Dictionary: Phonology." In Annals of the New York Academy of Science (Lexicography in English) 211, 110-114. . 1973a. "The English Language in Canada: A Report." Current Trends in Linguistics 10, 40-74. . 1975. "The Phonemic Segments of an Edmonton Idiolect." In J.K. Chambers, editor, Canadian English: Origins and Structures. Toronto: Methuen. 118-132. and A.M. Kinloch. 1978. Writings on Canadian English: 1792-1975. Toronto: Fitzhenry and Whiteside. Bahr, Dieter. 1977. A Bibliography of Writings on the English Language in Canada from 1857 - 1976. Heidelberg: Carl Winter UniversitStsverlag. Bailey, Richard W. and Manfred Gorlach. 1982. English as a World Language. Ann Arbor: University of Michigan Press. and Jay L. Robinson. 1973. Varieties of Present-Day English. New York: Macmillan . Bain, B. 1982. The Sociogenesis of Language. New York: Plenum. Baker, R.J. 1986. "A Response to H.B. Woods." In W.C. Lougheed, editor, In Search of the Standard in Canadian English. Occasional Papers Number 1, Strathy Language Unit. Kingston, Ont: Queen's University. 152-156. Bates, E. 1976. Language and Context: The Acquisition of Pragmatics. New York: Academic Press. Baugh, Albert C. 1957. A History of the English Language. 2nd edition. Englewood Cliffs, NJ: Prentice-Hall. Bernstein, B. 1971. Class, Codes and Control. Vol. 1: Theoretical Studies Towards a Sociology of Language. London: Routledge and Kegan Paul. Berry, J.W., R. Kalin and D.M. Taylor. 1977. Multiculturalism and Ethnic Attitudes in Canada. Ottawa: Minister of Supply and Services, Canada. 190 Blishen, B.R. 1958. "The Construction and Use of an Occupational Class Scale." Canadian Journal of Economics and Political Science 24, 519-531. Bliss, A. 1979. "The Emergence of Modern English Dialects in Ireland." In D. O'Muirithe, editor, The English Language In Ireland. Dublin: Mercier. Bloomfield, Leonard. 1933. Language. New York: Holt, Rinehart and Winston. Bloomfield, Morton W. 1975. "Canadian English and its Relation to Eighteenth Century American Speech." In J.K. Chambers, editor Canadian English: Origins and Structures. Toronto: Methuen. 3-11. Brown, P. and C. Fraser. 1979. "Speech as a Marker of Situation." In K. Scherer and H. Giles, editors, Social Markers in Speech. Cambridge: Cambridge University Press. 33-62. . and S. Levinson. 1979. "Social Structure, Groups and Interaction." In K. Scherer and H. Giles, editors, Social Markers in Speech. Cambridge: Cambridge University Press. 291-342. Bynon, Theodora. 1977. Historical Linguistics. Cambridge: Cambridge University Press. Chamberlain, Alexander F. 1890. "Dialect Research in Canada." Dialect Notes 1, 43-56. Chambers, J.K. 1975. Canadian English: Origins and Structures. Toronto: Methuen. 1986. "Three Kinds of Standard in Canadian English." In W.C. Lougheed, editor, In Search of the Standard in Canadian English. Occasional Papers Number 1, Strathy Language Unit. Kingston, Ontario: Queen's University. 1-15. and Peter Trudgill. 1980. Dialectology. Cambridge: Cambridge University Press. Chomsky, Noam. 1965. Aspects of the Theory of Syntax. Cambridge, MA: M.I.T. Press Clarke, S. 1981. "Speech Stereotypes in Newfoundland: An Initial Investigation." Unpublished Manuscript, Memorial University of Newfoundland. . 1986. "Sociolinguistic patterning in a new-world dialect of Hiberno-English: The speech of St. John's, Newfoundland." In J. Harris, D. Little and D. Singleton, editors, Perspectives on the English Language in Ireland. Proceedings of the First Symposium on Hiberno-English. Dublin: Trinity College. 191 and Robert Hollett. 1986. "Sociolinguistic Variations in a Small Urban Context: The Saint John's Survey." In H.J. Warkentyne, editor, Methods V - Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria. Colombo, John Robert. 1974. Colombo's Canadian Quotations. Edmonton: Hurtig. . 1976. Colombo's Canadian References. Toronto: Oxford University Press. Cronin, Fergus. 1965. "Do You Speak Canadian?" Canadian Weekly, February 27. Cullen, Connie. 1971. "Dialect Research on Prince Edward Island." English Quarterly 4:3, 51-53. Curme, George O. 1947. English Grammar. New York: Barnes and Noble. d'Anglejan, A. and G.R. Tucker. 1973. "Sociolinguistic Correlates of Speech Style in Quebec." In R.W. Shuy and R.W. Fasold, editors, Language Attitudes: Current Trends and Prospects. Washington, D.C: Georgetown University Press. 1-27. Darnell, Regna, editor. 1971. Linguistic Diversity in Canadian Society. Edmonton: Linguistic Research, Inc. • ., editor. 1973. Canadian Languages in Their Social Context. Edmonton: Linguistic Research, Inc. ., editor. 1976. Language Use in Canada. Edmonton: Linguistic Research, Inc. Davis, A.L., editor. 1972. Culture, Class and Language Variety. Urbana, IL: National Council of Teachers of English. and L.M. Davis. 1969. "Recording of Standard English Questionnaire." Orbis 18, 385-404. Davis, Lawrence M., editor. 1972. Studies in Linguistics: In Honour of Raven I. McDavid, Jr. University, AL: University of Alabama Press. De Camp, D. 1958-9. "The pronunciation of English in San Francisco." Orbis 7, 372-391, Orbis 8, 54-77. de Saussure, F. 1966. Course in General Linguistics. Edited by Charles Bally and Albert Sechehaye in collaboration with Albert Riedlinger. Translated by Wade Baskin. New York: McGraw-Hill Paperbacks. 192 deVries, John and Frank G. Vallee. 1975. "Data Books on Aspects of Language Demography in Canada." For Conference on the Individual, Language and Society, Kingston, November 1975. Carleton University. . 1980. Language Use in Canada. Ottawa: Statistics Canada. Minister of Supply and Services, Canada. Authority of President of Treasury Board. deWolf, Gaelan. 1984. "A Comparison of Phonetically-Ordered Phonological Variables in Two Major Canadian Urban Surveys." Journal of the International Phonetic Association. . 1986. "Methods in Statistical Analysis of Compatible Data from Two Major Canadian Urban Sociolinguistic Surveys." In H.J. Warkentyne, editor, Methods V - Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria. Dittmar, N. 1976. Sociolinguistics: A Critical Survey of Theory and Application. London: Edward Arnold. Drysdale, P.D. 1959. "A First Approach to Newfoundland Phonemics." Canadian Journal of Linguistics 5, 25-34. . 1971. "Her Parkee, Made of Caribou...Using a Dictionary as Source Book." English Quarterly 4, 47-50. Duncan, H.D. 1962. Communication and Social Order. New York: The Bedminster Press. Eckman, F.R. 1977. Current Themes in Linguistics. New York: Hemisphere. Edwards, J.R. 1979. Language and Disadvantage. London: Edward Arnold. England, George Allen. 1925. "Newfoundland Dialect Items." Dialect Notes 5, 322-346. Ervin-Tripp, S. 1964. "An Analysis of the Interaction of Language, Topic and Listener." In J.J. Gumperz and D. Hymes, editors, The Ethnography of Communication. American Anthropologist Special Publication. Menasha, Wl: American Anthropological Association. 86-102. Evans, Mary S. 1936. "Terms from the Labrador Coast." American Speech 6, 56-58. Fasold, Ralph W. and Roger W. Shuy. 1972. Analyzing Variation in Language. Washington, D.C: Georgetown University Press. Ferguson, C. 1959. "Diglossia." Word 15, 325-340. 193 Fishman, J.A. editor. 1968. Readings in the Sociology of Language. The Hague: Mouton. . 1971. Sociolinguistics: A Brief Introduction. Rowley, MA.: Newbury House. , R.L. Cooper and A.W. Conrad, editors. 1977. The Spread of English. Rowley, MA: Newbury House. Foster, Brian. 1968. The Changing English Language. New York: Macmillan. Fowke, Edith. 1976. Folklore of Canada. Toronto: McClelland and Stewart. Fowler, H.W. 1965. Dictionary of Modern English Usage. 2nd edition. Revised by Ernest Gowers. Oxford: Oxford University Press. Fowler, R.B., B. Hodge, G. Kress, T. Trew. 1975. Language and Control. London: Routledge. Francis, W. Nelson. 1954. The Structure of American English. New York: The Ronald Press. . 1963. The English Language. New York: W.W. Norton. Fraser, C. and K.R. Scherer, editors. 1982. Advances in the Social Psychology of Language. Cambridge: Cambridge University Press. Gardner, R.C. and R. Kalin, editors. 1981. A Canadian Social Psychology of Ethnic Relations. Toronto: Methuen. , D.M. Kirby, R.Y. Pablo and E.S. Castillo. 1975. "Ethnic Stereotypes: The Role of Language." Journal of Social Psychology 96, 3-9. Geikie, A.S. 1857. "Canadian English." Canadian Journal 2, 344-355. Giglioli, Pier Paolo, editor. 1972. Language and Social Context. Penguin Modern Sociology Readings. London: Penguin. Giles, H. 1977. "The Social Context of Speech: A Social Psychological Perspective." ITL: A Review of Applied Linguistics 35, 27-42. and R.Y. Bourhis. 1976. "Methodological Issues in Dialect Perception: Some Social Psychological Perspectives." Anthropological Linguistics 18, 294-304. R.Y. Bourhis and A. Davies. 1979. "Prestige Speech Styles: The Imposed Norm and Inherent Value Hypotheses." In W. McCormack and S. Wurm, editors, Language in Anthrolopology IV: Language in Many Ways. The Hague: Mouton. 589-596. R. Bourhis and D. Taylor. 1977. "Toward a Theory of Language in Ethnic Group Relations." In H. Giles, editor, Language, Ethnicity and Intergroup Relations. London: Academic Press. 194 R.Y. Bourhis, P. Trudgill and A. Lewis. 1974. "The Imposed Norm Hypothesis: A Validation." Quarterly Journal of Speech 60, 405-410. and P. F. Powesland, editors. 1975. Speech Style and Social Evaluation. London: Academic Press. W.P. Robinson and P. Smith, editors. 1979. Social Psychology and Language. London: Pergamon Press. W.P. Robinson and P.M. Smith, editors. 1980. Language: Social Psychological Perspectives. Oxford: Pergamon Press. and B. Saint-Jacques, editors. 1979. Language and Ethnic Relations. Oxford: Pergamon Press. K. Scherer and D. Taylor. 1979. "Speech Markers in Social Interaction." In K. Scherer and H. Giles, editors, Social Markers in Speech. Cambridge: Cambridge University Press. 343-381. and P. Smith. 1979. "Accommodation Theory: Optimal Levels of Convergence." In H. Giles and R. St. Clair, editors, Language and Social Psychology. Oxford: Blackwell. P. Smith, B. Ford, S. Condor and J. Thakerar. 1980. "Speech Styles and the Fluctuating Salience of Sex." Language Sciences 2, 260-282. and R.N. St. Clair, editors. 1979. Language and Social Psychology. Oxford: Blackwell. D.M. Taylor and R.Y. Bourhis. 1973. "Towards a Theory of Interpersonal Accommodation Through Language: Some Canadian Data." Language in Society 2, 177-192. Goody, N. 1978. Questions and Politeness: Strategies in Social Interaction. Cambridge: Cambridge University Press. Graham, Robert Somerville. 1955. "The Anglicization of German Family Names in Western Canada." American Speech 30, 260-264. . 1957. "The Transition From German to English in the German Settlements of Saskatchewan." Journal of the Canadian Linguistic Association 3, 9-13. Greenbaum, S. 1976. "Contextual Influence on Acceptability Judgements." International Journal of Psycholinguistics 6, 5-11. . 1977. "Judgements of Syntactic Acceptability and Frequency." Studia Linguistica 31, 83-105. 195 . 1977a. "The Linguist as Experimenter." In F.R. Eckman, editor, Current Themes in Linguistics. New York: Hemisphere. 124-144. . editor. 1985. The English Language Today. Oxford: Pergamon. Gregg, Robert J. 1973. "The Diphthongs /si/ and /al/ in Scottish, Scotch-Irish and Canadian English." Canadian Journal of Linguistics 18, 136-145. . 1973. "The Linguistic Survey of British Columbia: The Kootenay Region." In Regna Darnell, editor, Canadian Languages in Their Social Context. Edmonton: Linguistic Research Inc. . 1981. "General Background to the Survey of Vancouver English (SVEN)" in H.J. Warkentyne, editor, Methods IV - Papers from the Fourth International Conference on Methods in Dialectology. University of Victoria. . 1983. "Local Lexical Items in the Sociodialectal Survey of Vancouver English." Canadian Journal of Linguistics 28, 17-23. . 1984. "Canadian English" English Around the World. Vol. 2. Tokyo: Kenkyusha Ltd. . 1984a. "Final Report to the Social Sciences and Humanities Research Council of Canada on An Urban Dialect Survey of the English . Spoken in Vancouver." Unpublished manuscript. . 1984b. "The Survival of Local Lexical Items as Specific Markers in Vancouver English." Orbis. . 1986. "New Methods Developed in SVEN for the Investigation of Grammatical Variability and for the Analysis of Resultant Data." In H.J. Warkentyne, editor, Methods V - Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria. . 1986a. "The Standard and Where We Stand Now." In W.C. Lougheed, editor, In Search of the Standard in Canadian English. Occasional Papers Number 1, Strathy Language Unit. Kingston, Ont: Queen's University. 157-168. Gumperz, J. and E. Herasimchuk. 1975. "The Conversational Analysis of Social Meaning: A Study of Classroom Interaction." In M. Sanches and B. Blount, editors, Socio-cultural Dimensions of Language Use. New York: Academic Press. and D. Hymes, editors. 1964. The Ethnography of Communication. American Anthropoligist Special Publication. Menasha, Wl: American Anthropological Association. 196 and D. Hymes. 1972. Directions in Sociolinguistics. New York: Holt, Rinehart and Winston. Gzowski, Peter. 1985. The Morningside Papers. Toronto: McClelland and Stewart. Hale, Kenneth. 1965. "On the Use of Informants in Field-work." Canadian Journal of Linguistics 10, 108-119. Halliday, M.A.K. , Angus Mcintosh and Peter Strevens. 1973. "The Users and Uses of Language." In R.W. Bailey and J.L. Robinson, editors, Varieties of Present-Day English. New York: Macmillan . Hamilton, Donald E. 1964. "Standard Canadian English: Pronunciation." In Horace G. Lunt, editor, Proceedings of the Ninth International Congress of Linguists. The Hague: Mouton. 456-459. Harris, Barbara P. 1986. "Self-image, Perception of Stereotypes and Language Variation." In H.J. Warkentyne, editor, Methods V - Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria. and Joseph K. Kess. 1975. "Salmon Fishing Terms in British Columbia." Names 23:2, 61-66. Hasebe-Ludt, E. 1981. "Aspects of Spontaneous Speech in the Urban Dialect Study of Vancouver English." In H.J. Warkentyne, editor, Methods IV - Papers from the Fourth International Conference on Methods in Dialectology. University of Victoria. . 1986. "Methodology of Spontaneous Speech Analysis." In H.J. Warkentyne, editor, Methods V - Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria. Hatch, Evelyn and Hossein Farhady. 1982. Research Design and Statistics for Applied Linguistics. Rowley: Newbury House Publishers. Hewett, N. 1971. "Reactions of Prospective English Teachers Toward Speakers of a Nonstandard Dialect." Language Learning 21, 205-212. Hewstone, M. 1985. "The Role of Language in Attribution Processes." In J. Jaspars, F. Fincham and M. Hewstone, editors, Attribution Theory: Essays and Experiments. London: Academic Press. Higgins, E.T., C P . Herman and M.P. Zanna. 1980. Social Cognition: The Ontario Symposium. Hillside: Lawrence Erlbaum Associates. Higinbotham, John D. 1962. "Western Vernacular." Alberta Historical Review 10, 9-17. Hill, Archibald, editor. 1969. Linguistics Today. New York: Basic Books. 197 Howay, F.W. 1942. "Origins of the Chinook Jargon." British Columbia Historical Quarterly 6, 225-250. Hudson, R.A. 1980. Sociolinguistics. Cambridge: Cambridge University Press. Hughes, A. and P. Trudgill. 1979. English Accents and Dialects: An Introduction to Social and Regional Varieties of British English. London: Edward Arnold. Hymes, D. 1964. Language in Culture and Society. New York: Harper. . 1974. Foundations in Sociolinguistics: An Ethnographic Approach. Philadelphia: University of Pennsylvania Press. Innes, J.M. and C. Fraser. 1971. "Experimenter Bias and Other Possible Biases in Psychological Research." European Journal of Social Psychology 1, 297-310. Irwin, R. 1977. "Judgements of Vocal Quality, Speech Fluency and Confidence of Black and White Speakers." Language and Speech 20, 261-266. Jespersen, Otto. 1956. A Modern English Grammar on Historical Principles. 7 vols. London: Allen and Unwin. Jones, Daniel. 1940. An English Pronouncing Dictionary. London: Dent . . 1950. The Pronunciation of English. 3rd edition. Cambridge: Cambridge University Press. Joy, R.J. 1972. Languages in Conflict. Toronto: McClelland and Stewart. Kachru, B. 1982. The Other Tongue: English Across Cultures. Urbana: University of Illinois Press. Kinloch, A.M. 1971. "The Survey of Canadian English: Possible Evidence for Pronunciation." English Quarterly 4:4, 59-65. . 1973. "The Survey of Canadian English: A First Look at New Brunswick Results." English Quarterly 5:4, 41-49. . 1986. "A Response to I.W.V. Pringle" In W.C. Lougheed, editor, In Search of the Standard in Canadian English. Occasional Papers Number 1, Strathy Language Unit. Kingston, Ont: Queen's University. 39-40. Kirwin, William. 1960. "Labrador, St. John's and Newfoundland: Some Pronunciations." Journal of the Canadian Linguistic Association 6:2, 115-116. . 1965. "Lines, Coves and Squares in Newfoundland [Street] Names." American Speech 40, 163-170. 198 . 1968. "Bibliography of Writings on Newfoundland English." Regional Language Studies - Newfoundland 1, 4-7. . 1974. "Newfoundland Usage in the 'Survey of Canadian English'." Regional Language Studies - Newfoundland 5, 9-14. Kleiven, J. 1979. "Social Stereotypes Elicited by Linguistic Differences." In R. Rommetveit and R. Blakar, editors, Studies of Language, Thought and Verbal Communication. London: Academic Press. 401-407. Klinck, Carl F. 1965. Literary History of Canada: Canadian Literature in English Vols 1-3. 2nd edition (1976). Toronto: University of Toronto Press. Klymasz, Robert. 1963. "The Canadianization of Slavic Surnames: A Study in Language Contact." Names 11,81-105, 182-195, 229-253. Kroch, Anthony S. 1978. "Toward a Theory of Social Dialect Variation." In Language in Society 7, 17-36. Kurath, Hans. 1949. Word Geography of the Eastern United States. Ann Arbor: University of Michigan Press. .. 1964. "British Sources of Selected Features of American Pronunciation: Problems and Methods." In D. Abercrombie, et al, editors, In Honour of Daniel Jones. London: Longmans-Green. 146-255. 1964a. "Interrelations Between Regional and Social Dialects." In Horace G. Lunt, editor, Proceedings from the Ninth International Congress of Linguists. The Hague: Mouton. 135-143. [Reprinted in Harold B. Allen and Gary N. Underwood, editors, Readings in American Dialectology. New York: Appleton Century Crofts. 365-373.] 1965. "Some Aspects of Atlantic Seaboard English Considered in their Connections with British English." In Communications et rapports du premier congres international de dialectologie ginirale. Louvain. 239-40. 1970. "English Sources of Some American Words." American Speech 45, 60-68. 1972. "Relics of English Folk Speech in American English." In Lawrence M. Davis, editor, Studies in Linguistics: In Honour of Raven I. McDavid, Jr. University, AL: University of Alabama Press. 367-375. and B. Bloch. 1939. Handbook of the Linguistic Geography of New England. Brown University Press. and R.I. McDavid, Jr. 1961. The Pronunciation of English in the Atlantic States. Ann Arbor: University of Michigan Press. 199 M. Hanley, B. Bloch and G.S. Lownan Jr. 1939-43. Linguistic Atlas of New England. Brown University Press. Labov, William. 1963. "The Social Motivation of a Sound Change." Word 19, 273-309. . 1964. "Stages in the Acquisition of Standard English." In Roger W. Shuy, editor, Social Dialects and Language Learning: Proceedings of the Bloomington, Indiana, Conference 1964. Champaign, IL: National Council of Teachers of English. . 1965. "On the Mechanism of Linguistic Change." Georgetown Monographs in Language and Linguistics 18, 91-114. . 1966. Social Stratification of English in New York City. Washington, D.C: Centre for Applied Linguistics. . 1972. Language in the Inner City. Philadelphia: University of Pennsylvania Press. . 1972a. Sociolinguistic Patterns. Philadelphia: University of Pennsylvania Press. . 1972b. "Variation in Language." In Carroll E. Read, editor, The Learning of Language. N.C.T.E. New York: Appleton Century Crofts Educational Division, Meridith Corporation. Lambert, Wallace. 1972. Language, Psychology and Culture. Stanford: Stanford University Press. . 1981. "Social Influences on the Child's Development of an Identity." In R. Gardner and R. Kalin, editors, Ethnic Relations. Toronto: Methuen. Laver, J. and P. Trudgill. 1979. "Phonetic and Linguistic Markers in Speech." In K. Scherer and H. Giles, editors, Social Markers in Speech. Cambridge: Cambridge University Press. 1-32. Lee, D.J. 1979. Emerging Ethnic Boundaries. Ottawa: University of Ottawa Press. Lee, R. 1971. "Dialect Perception: A Critical Review and Re-evaluation." Quarterly Journal of Speech 57, 410-417. Lehmann, Winfred P. 1962. Historical Linguistics: An Introduction. Second edition. New York: Holt, Rinehart and Winston . Leon, P.R., and P. Martin. 1976. Toronto English: Studies in Phonetics and Phonology. Unpublished Manuscript. Lieberson, S. 1970. Language and Ethnic Relations in Canada. New York: Wiley. 200 Light, R.Y., D.P. Richard and P. Bell. 1978. "Development of Children's Attitudes Toward Speakers of Standard and Nonstandard English." Child Study Journal 8, 253-265. Lougheed, W.C., editor. 1986. In Search of the Standard in Canadian English. Occasional Papers Number 1; Strathy Language Unit. Kingston, Ontario: Queen's University. Lovell, Charles, J. 1955. "Lexicographical Challenges of Canadian English." Journal of the Canadian Linguistic Association 1:1, 2-5. . 1955-6. "Whys and Hows of Collecting for the Dictionary of Canadian English." Journal of the Canadian Linguistic Association 1:2, 3-8, 2:1, 23-32. . 1958. "A Sampling of Material for a Dictionary of Canadian English Based on Historical Principles." Journal of the Canadian Linguistic Association 4:1, 7-33. Lunt, Horace G., editor. 1964. Proceedings of the Ninth International Congress of Linguists. The Hague: Mouton. Lyons, John, editor. 1970. New Horizons in Linguistics. Harmondsworth: Penguin. MacKey, W.F. and J. Orastein, editors. 1979. Sociolinguistic Studies in Language Contact. The Hague: Mouton. Marckwardt, Albert, H. 1971. "The Concept of Standard English." The Discovery of English. Urbana, IL: National Council of Teachers of English. and Randolph Quirk. 1964. A Common Language: British and American English. London: BBC and the United States Government. Marks, CT. 1976. "Policy and Attitudes Towards the Teaching of Standard Dialect: Great Britain, France, West Germany." Comparative Education 12, 199-218. McConnell, R.E. 1979. Our Own Voice: Canadian English and How It Is Studied. Vancouver: Gage. McCormack, W. and S. Wurm, editors. 1979. Language in Anthropology IV: Language in Many Ways. The Hague: Mouton. McDavid, Raven I., Jr. 1951. "Midland and Canadian Words in Upstate New York." American Speech 26, 248-256. . 1954. "Linguistic Geography in Canada: An Introduction." Journal of the Canadian Linguistic Association 1:1, 3-8. 201 . 1960. "The Second Round in Dialectology of North American English." Journal of the Canadian Linguistic Association 6:2, 108-115. . 1969. "Dialects: British and American Standard and Nonstandard." In Archibald Hill, editor, Linguistics Today. New York: Basic Books. 79-88. . 1972. "Some Notes on Acadian English." In A.L. Davis, ediitor, Culture, Class and Language Variety. Urbana, IL: National Council of Teachers of English. Milroy, L. 1980. Language and Social Networks. Oxford: Blackwell. Moon, Barbara. 1962. "Does Anyone Here Speak Canadian?" MacLean's, June 2. Murdoch, M. 1981. "Visual-Aural Prompting in the Vancouver Survey Questionnaire." In H.J. Warkentyne, editor, Methods IV - Papers from the Fourth International Conference on Methods in Dialectology. University of Victoria. . 1985. "A Proposal for Standardization of Computer-Coding Systems in Linguistics Surveys." In H.J. Warkentyne, editor, Methods V -Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria. Noseworthy, Ronald G. 1974. "Fishing Supplement: Linguistic Atlas of Newfoundland Dialect Questionnaire." Regional Language Studies - Newfoundland 5, 18-21. O'Grady, W. and M. Dobrovolsky, editors. 1987. Contemporary Linguistic Analysis: An Introduction. Toronto: Copp Clark Pitman. O'Kane, D. 1977. "Overt and Covert Prestige in Belfast Vernacular Speakers: The Results of Self-Report Tests." Belfast Working Papers in Language and Linguistics 2, 3. Orkin, M.M. 1970. Speaking Canadian English: An Informal Account of the English Language In Canada . Toronto: General. . 1973. Canajun, eh? Don Mills, Ont.: General Publishing. Orton, H. 1962. Survey of English Dialects I-IV. London: Edward Arnold. , S. Sanderson and J. Widdows. 1978. The Linguistic Atlas of England. London: Croom Helm. Osgood, CE. and P.H. Tannenbaum. 1955. "The Principle of Congruity in the Prediction of Attitude Change." Psychological Review 62,42-55. 202 Owens, Thompson. 1986. "Testing for Linguistic Insecurity in Language Variation Surveys." In H.J. Warkentyne, editor, Methods V -Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria. Padolsky, Enoch and Ian Pringle. 1977. "Reflexes of M.E. Vowels before /r/ in Ottawa Valley Dialects of Hiberno-English Types." Unpublished Paper presented to Canadian Linguistic Association, Fredericton. Partridge, Eric and John W. Clark, editors. 1968. British and American English since 1900. New York: Greenwood. Patterson, G. 1895-7. "Notes on the Dialect of the People of Newfoundland." Journal of American Folklore 8, 27-40; 9, 19-37; 10, 203-13. Peacock, F.W. 1974. "Languages in Contact in Labrador." Regional Language Studies - Newfoundland 5, 1-3. Perlin, A.B., editor. 1959. The Story of Newfoundland. St. John's: Guardian Limited. Poison, James. 1969. "A Linguistic Questionnaire for British Columbia: A Plan for a Postal Survey of Dialectal Variation of B.C., with an Account of Recent Research." Unpublished MA Thesis. University of British Columbia. Porter, Bernard H. 1963. "A Newfoundland Vocabulary." American Speech 38, 297-301. Powesland, P. and H. Giles. 1975. "Persuasiveness and Accent-Message Incompatibility." Human Relations 28, 85-93. Pratt, Mary-Louise. 1977. Toward a Speech Act Theory of Literary Discourse. Bloomington: Indiana University Press. Priestly, F.E.L. 1957. "English Language" in Encyclopedia Canadiana. . 1968. "Canadian English." In Eric Partridge and John W. Clark, editors, British and American English since 1900. New York: Greenwood. Pringle, Ian. 1985. "Attitudes to Canadian English" In Sidney Greenbaum, editor, The English Language Today. Oxford: Pergamon. 183-205. . 1986. "The Complexity of the Concept of Standard." In W.C. Lougheed, editor, In Search of the Standard in Canadian English. Occasional Papers Number 1, Strathy Language Unit. Kingston, Ontario: Queen's University. 20-38. , Ian Dale, and Enoch Padolsky. 1986. "Linguistic Survey of the Ottawa Valley." In H.J. Warkentyne, editor, Methods V - Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria. 203 Quirk, R. and J. Swartik. 1966. Investigating Linguistic Acceptability. T h e Hague: Mouton. Reed, Carroll E. 1961. "The Pronunciation of English in the Pacific Northwest." In Harold B. Allen and Gary N. Underwood, editors, Readings in American Dialectology. New York: Appleton Century Crofts . Robinson, W.P. 1972. Language and Social Behaviour. Harmondsworth: P e n g u i n . Rodman, Lilita. 1974-5. "Characteristics of B.C. English." English Quarterly 7:4, 49-82. . 1986. "Reading Style." In H J . Warkentyne, editor, Methods V -Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria Press. Rogers, P. W. 1970. "Unlocking the Canadian Word Hoard." Queen's Quarterly 77, 111-123. Rommetveit, R. and R. Blakar, editors. 1979. Studies of Language, Thought and Verbal Communication. London: Academic Press. Ryan, E.B. and J.T. Cacioppo. 1981. "Review of 'Social Markers in Speech', edited by K.R. Scherer and H. Giles." Language in Society 10, 443-447. Ryan, E.B., M.A. Carranza and R.W. Moffie. 1977. "Reactions Toward Varying Degrees of Accentedness in the Speech of Spanish-English Bilinguals." Language and Speech 20, 267-273. Ryan, E.B. and H. Giles, editors. 1982. Attitudes Towards Language Variation: Social and Applied Contexts. London: Edward Arnold. . 1982a. "Prolegomena for Developing a Social-Psychological Theory of Language Attitudes." In E.B. Ryan and H. Giles, editors, Attitudes Toward Language Variation: Social and Applied Contexts. London: Edward Arnold. 208-223. Ryan, E.B., H. Giles and J. Sebastian. 1982. "An Integrative Perspective for the Study of Attitudes Toward Language Variation." In E.B. Ryan and H. Giles, editors, Attitudes Towards Language Variation: Social and Applied Contexts. London: Edward Arnold. 1-19. Samarin, William I. 1967. Field Linguistics: A Guide To Linguistic Field Work. New York: Holt, Rinehart and Winston. Samuels, M.L. 1972. Linguistic Evolution with special reference to English. Cambridge: Cambridge University Press. Sanches, Mary and Ben G. Blount. 1975. Sociocultural Dimensions of Language Use. New York: Academic Press. 204 Sandilands, John. 1912. Western Canadian Dictionary and Phrase-Book: Things a Newcomer Wants to Know. Winnipeg: Telegram Job Printers. [Reprinted, Edmonton: University of Alberta Press, 1977.] Sanigan, Gerald D. 1986. Canadian Words and Phrases—Legal Maxims. 3 vols. Don Mills: Richard DeBoo. Sankoff, D. 1978. Linguistic Variation: Models and Methods. New York: Academic Press. Sapir, Edward. 1921. Language, an Introduction to the Study of Speech. New York: Harcourt, Brace and Company. . 1931. "Dialect." In E.R.A. Seligman, editor, Encyclopedia of the Social Sciences. Vol. V. New York: Macmillan. 123-126. . 1933. "Language." In E.R.A. Seligman, editor, Encyclopedia of the Social Sciences. Vol. IX. New York: Macmillan. 155-168. Scargill, M.H. 1954. "A Pilot Study of Alberta Speech: Vocabulary." Journal of Canadian Linguistic Association 1:1, 21-22. . 1955. "Canadian English and Canadian Culture in Alberta." Journal of Canadian Linguistic Association 1:1,26-29. . 1956. "Eighteenth Century English in Nova Scotia." Journal of Canadian Linguistic Association. 2:1, 3. . 1965. "The Growth of Canadian English." In Carl F. Klinck, editor, Literary History of Canada: Canadian Literature in English. Vol. 1. Toronto: University of Toronto Press. 265-273. . 1968. "Canadianisms from Western Canada, with Special Reference to British Columbia." Transactions of the Royal Society of Canada, Vol. 6, Series 4, 181-185. . 1973. "Is Riz, Some Hot, Clumb, and Other Canadianisms." English Quarterly 6:2, 115-121. . 1973a. "Using the Historical Dictionary: A Concise Dictionary of Canadianisms." Speaking of Dictionaries. No. 2. Toronto: Gage. . 1974. Modern Canadian English Usage: Linguistic Change and Reconstruction. Toronto: McClelland and Stewart. . 1977. A Short History of Canadian English. Victoria: Sono Nis Press. and P.G. Penner. 1966. Looking at Language. Toronto: Gage. and H.J. Warkentyne. 1972. "The Survey of Canadian English: A Report." English Quarterly 5:3, 47-104. 205 Scherer, K. and H. Giles, editors. 1979. Social Markers in Speech. Cambridge: Cambridge University Press. Searle, John R. 1969. Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. Seary, E.R. 1958. "The French Element in Newfoundland Place Names." Journal of Canadian Linguistic Association 4:2, 63-69. . 1962. "Linguistic Variety in the Place Names of Newfoundland." Canadian Geographical Journal 65, 146-155. . 1971. Place Names of the Avalon Peninsula of the Island of Newfoundland. Memorial University Series 2. Toronto: University of Toronto Press. , G.M. Story and W.J. Kirwin. 1968. The Avalon Penninsula of Newfoundland: An Ethno-linguistic Survey. National Museum of Canada, Bulletin No. 219. Anthropological Series 81. Ottawa: Queen's Printer. Sebeok, Thomas A. 1973. Current Trends in Linguistics 10: Linguistics in North America. 2 vols. The Hague: Mouton. Sebeok, T., editor. 1960. Style in Language. New York: Wiley. Seligman, C, G.R. Tucker and W.E. Lambert. 1972. "The Effects of Speech Style and Other Attributes on Teachers' Attitudes Toward Pupils." Language in Society 1, 131-142. Shuy, Roger W., editor. 1964. Social Dialects and Language Learning: Proceedings of the Bloomington, Indiana, Conference 1964. Champaign, IL: National Council of Teachers of English. . 1971. Sociolinguistics: A Cross-Disciplinary Perspective. Washington, D.C: Center for Applied Linguistics. and R.W. Fasold, editors. 1973. Language Attitudes: Current Trends and Prospects. Washington, D.C: Georgetown University Press. Shuy, Roger W., Walter Wolfrom and William Riley. 1968. Field Techniques in an Urban Language Study. Washington, D.C: Georgetown University Press. Sledd, James. 1973. "Bidialectalism: A New Book and Some Old Issues." American Speech 48:3-4, 258-269. Smith, E.M., editor. 1972. Studies in Linguistics in Honor of George L. Trager. The Hague: Mouton. 206 Snyder, M. 1980. "Seek and Ye Shall Find: Testing Hypotheses about other People." In E.T. Higgins, CP Herman and M.P. Zanna, editors, Social Cognition: The Ontario Symposium. Hillside: Lawrence Erlbaum Associates. 277-303. Southerland, Ronald H. 1987. "Language in Social Contexts." In O'Grady and Dobrovolsky, editors, Contemporary Linguistic Analysis: An Introduction. Toronto: Copp Clark Pitman. Spilka, I.V. 1970. Force Study of Diglossia in French Canada. Montreal: Mimeo University de Montreal. St. Clair, R.N. and H. Giles. 1980. The Social and Psychological Contexts of Language. Hillside: Lawrence Erlbaum Associates. Stevenson, Roberta C. 1977. "The Pronunciation of English in British Columbia: An Analysis of the Responses to the Phonological Section of the Gregg-Poison Postal Questionnaire for B.C., with Suggestions for Further Dialect Investigation in the Province." Unpublished MA Thesis. University of British Columbia. Story, G.M. 1957. "Newfoundland English Usage." In Encyclopedia Canadiana 7, 321-322. . 1957. "Research in the Language and Place-Names of Newfoundland." Journal of Canadian Linguistic Association 3, 47-55. . 1959. "Newfoundland Dialect." In A.B. Perlin, editor, The Story of Newfoundland. St. John's: Guardian Limited. 68-70. Strang, B. 1970. A History of English. London: Methuen. Strevens, Peter. 1985. "Standards and the Standard Language." English Today 2 (April), 5-8. Sturtevant, E. 1917. Linguistic Change: An Introduction to the Historical Study of Language. Chicago: Phoenix Books, University of Chicago Press. . 1947. An Introduction to Linguistic Science. New Haven: Yale University Press. Sweet, Henry. 1890. A New English Grammar, Logical and Historical. Oxford: Oxford University Press. Toye, William. 1983. The Oxford Companion to Canadian Literature. Toronto: Oxford. Trudgill, Peter. 1974. The Social Differentiation of English in Norwich. Cambridge: Cambridge University Press. 207 1974a. Sociolinguistics: An Introduction to Language and Society. Harmondsworth: Penguin. 1978. Sociolinguistic Patterns in British English. London: Edward Arnold. 1981. "Linguistic Accommodation: Sociolinguistic Observations on a Socio-Psychological Theory." In Proceedings of the Chicago Linguistic Circle. Chicago: University of Chicago Press. 1983. On Dialect: Social and Geographical Perspectives. London: New York University Press. 1984. Applied Sociolinguistics. London: Academic Press. 1986. "Why 'Canadian Raising' is not Canadian: Dialect Mixture and the Analysis of Colonial Dialects." In HJ. Warkentyne, editor. Methods V- Papers from the Fifth International Conference on Methods in Dialectology. University of Vicoria. 1986a. Dialects in Contact. Oxford: Blackwell. and Jean Hannah. 1985. International English: A Guide to Varieties of Standard English 2nd edition. London: Edward Arnold. Turner, J.C. and H. Giles. 1981. Inter group Behaviour. Oxford: Blackwell. Ullrich, H.E. 1971. "Linguistic Aspects of Antiquity: A Dialect." Anthropological Linguistics 13, 106-113. Van Eerde, J. 1979. "Facets of the Breton Problem." Language Problems and Language Planning 3, 1-8. Veltman, CJ. 1979. "New Opportunities for the Study of Language Shift: The Anglicization of New England Language Minorities." Language Problems and Language Planning 3,65-75. Wanamaker, M.G. 1959. "Canadian English: Whence? Whither?" Journal of Education (Nova Scotia) 5th series, 9, 22-26. . 1976-77. "Who Controls Writing Standards?" English Quarterly 9:4, 45-52. Warkentyne, HJ. 1971. "Contemporary Canadian Usage: A Report on the Survey of Canadian English." American Speech 46:3-4, 193-199. ., editor. 1981. Methods IV - Papers from the Fourth International Conference on Methods in Dialectology. University of Victoria. . 1983. "Attitudes and Language Behaviour." Canadian Journal of Linguistics 28:1, 71-76. 208 ., editor 1986. Methods V - Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria. . 1986a. "A Response to RJ. Gregg." In W.C. Lougheed, editor, In Search of the Standard in Canadian English. Occasional Papers Number 1, Strathy Language Unit. Kingston, Ont: Queen's University. 169-177. Weinreich, U. 1963. Languages in Contact. The Hague: Mouton. Wells, John C. 1982. Accents of English 1-3. Cambridge: Cambridge University Press. . 1986. "Sound Changes in British and North American English and their Synchronic Consequences." In HJ. Warkentyne, editor, Methods V - Papers from the Fifth International Conference on Methods in Dialectology. University of Victoria. Williams, F. 1970. Language and Poverty. Chicago: Markham. Williams, Joseph M. 1975. Origins of the English Language: A Social and Linguistic History. New York: The Free Press. Willinsky, J. 1986. "Computing the Standard." In W.C. Lougheed, editor, In Search of the Standard in Canadian English. Occasional Papers Number 1, Strathy Language Unit. Kingston, Ont: Queen's University. 101-111. Wilson, H. Rex. 1973. "Dialect Literature: A Two-Way Street?" Canadian Journal of Linguistics 18, 157-162. Wolfram, W. and R.W. Fasold. 1974. The Study of Social Dialects in American English. Englewood Cliffs, NJ: Prentice-Hall. Woods, Howard. 1979. "Socio-dialectal Survey of the English Spoken in Ottawa." Unpublished PhD Dissertation. University of British Columbia. Wyld, H.C. 1963. A Short History of English. 3rd edition. London: John Murray. Yogis, John. 1984. Canadian Law Dictionary. Toronto: Barrons Educational. Zandvoort, R.W. 1957. A Handbook of English Grammar. London: Longmans. 209 APPENDIX A: OUTLINE OF CONTENTS OF SVEN QUESTIONNAIRE The sections of the questionnaire used to conduct the "Urban Dialect Survey of the English Spoken in Vancouver" include: Section I Background Information Section II Visual-Aural Prompting Section III Word List Section IV Picture Description and Narrative Section V Syntactic Variation Section VI Local Words Section VII Reading Passage Section VIII Questions about Vancouver Section IX Spontaneous Narrative Section X Series Section XI Word Pairs Section XII A Subjective Attitudes and Language Awareness Section XII B Language Rating Scales 210 APPENDIX B: READING PASSAGE "Hi, Mum!" Barry shouted running into the house. "What's for supper tonight? This morning at the library I saw Mary Wilson, a girl I knew in high school. She's been working for the past year and is just beginning at UBC. I want to call her now and I'd sure like to ask her to supper. How about it, eh?" His mother was in the living-room putting up a new mirror she'd just bought at Eatons. "That'll be all right, dear," she answered in her usual calm and leisurely way. I've invited Aunt Dorothy and Uncle Harold also. We're going to have the trout your father caught last Saturday, with potato salad, mixed vegetables, and sliced tomatoes. Apricot whip for dessert with almond cookies. A real hot weather meal, nice and easy to prepare!" "That's fantastic, Mum! Mary and I wanted to see a Peter Sellers picture at The Vancouver East Cultural Centre. It's supposed to be very good, so we'd better be there by a quarter past eight to be sure of getting seats. Any chance we might eat a little early?" "I'm sorry, son, we can't possibly eat for about an hour and a half. If I'd known, we could have eaten sooner. But I was late getting things started because the electricity went off this afternoon. You'll have plenty of time, anyway. Why don't you ask Mary to come over right away, so we can start eating once it's ready? I'm sure it won't bother your father to eat early, as he has a heavy schedule of work for this evening." "Okay, Mum. Say, can I have something to eat now? I only ate egg sandwiches for lunch and an orange, so I'm starving. What's in the fridge?" "Help yourself to anything, Barry, except for the makings of tonight's meal, naturally. There's bread and butter and malted milk, or whatever else you like. When you're finished, would you carry that carton of empty bottles 211 out to the garage? Mike's been out there nearly all day working on that old Pontiac he got on Tuesday. The brakes are faulty and there also seems to be something wrong with the steering wheel, so it's important to get all that fixed before driving anywhere. He's talking about going to Eastern Washington for Saturday and Sunday. While you're out there, you can remind him we're having company for supper. Tel l him there's plenty of hot water when he wants to wash up. He's probably very dirty!" A little later Barry went out to pick Mary up in his father's Falcon. After she'd been introduced to the Walters family, they noticed she was wearing a cast on her right leg, and begged her to tell them the details of what had happened. "Last Wednesday I was riding my bike down Balsam Street towards Kitsilano Beach when I accidentally ran the front wheel through some oi l and skidded. After a futile attempt to keep my balance I fell. It was right in front of a dentist's office, and somebody who was looking out of the window rushed out and drove me down to the Vancouver General Hospital without waiting for an ambulance. Unfortunately I had fractured my leg. Actually I had a narrow escape, as it could have been much worse. I hate not being mobile, particularly as I always enjoy dancing so much. I can't go swimming either. But wasn't it interesting my meeting Barry again today? We were in a lot of the same groups our last year in school, probably because both our last names begin with W. But I almost didn't recognice him." "What have you been doing since high school, Mary?" asked Barry's grandmother, who had come from Edmonton to visit her son and daughter-in-law. 212 "During the winter I went to business college, and since February I've been working as a secretary at City Hall. From our office building there's a wonderful view of the North Shore mountains. It was a very interesting job, especially as I often got to attend council meetings. I gained a lot of valuable experience, but I'm really enjoying being a student and attending lectures. My parents live in the United States, you know, south of Bellingham, but I've been living in Vancouver with my aunt on Fifth Avenue since I was thirteen. I wish they were nearer, of course, but we see each other regularly." "Do you really find living in Canada preferable to the U.S.A.?" asked Mrs. Walters. "I certainly do," answered Mary. "Someday I'd like to become a writer and settle down here in this beautiful province of British Columbia, but right now I'm trying to decide whether or not to become a Canadian citizen. Without that I couldn't get the job I want in Ottawa after I've graduated." After dinner, everybody congratulated Barry's father for catching such delicious fish, and his mother for being such a good cook. Then they all sat out on the balcony until 7:30, when it was time for Barry and Mary to leave for the theatre. 213 APPENDIX C: VISUAL-AURAL PROMPTS I have included here photocopies of only the Visual-Aural prompts for some of the items to which I refer in this work. Most prompts used were glossy, colour photos taken from catalogues and magazines; they were mounted on large cardboard cards and shown to the informants by the interviewer, who indicated which picture the informant was to respond to. Although the pictures photocopy very poorly, the attached give an idea of what was included and may clarify some confusion, particularly about lexical items. 214 256. Rug/Carpet 215 256/257. Rug/Carpet 257. Rug/Carpet 217 258. Rug/Carpet 218 258. Rug/Carpet 219 259. Can/Tin 260. Canned/Tinned 259/260 220 4 4 ^ * « ? 7 I r * , i « J 4 * 265. L y i n g / L a y i n g 265. L y i n g / L a y i n g 221 268/269. Train/Railway Station 270. Train/Railway 222 270. Train/Railway Tracks 277. Stove/Range 224 J A N U A R Y 275. Frying Pan/Skillet 214. Tulips 277. Stove/Range 226 279. Curtains/Drapes 227 228 285. Verandah/Porch 229 APPENDIX D: FORMS USED BY MOST SPEAKERS IN ALL SOCIO-ECONOMIC GROUPS LEXICAL VARIABLES1 ITEM 256. Carpet/Rug2 258. Rug/Carpet3 259. Can/Tin 260. Canned/Tinned Fruit 262. Mow Lawn/Cut Grass 263. Sofa/Chesterfield (1st choice) 263A. Sofa/Chesterfield (2nd choice) 264. Tap/Faucet 265. Lying/Laying4 267. Cutlery/Flatware/Silver 269. Train Station/Railway Station 271. Television/TV 275. Fryfing] Pan/Skillet 277. Stove/Range RESPONSE OVERALL RATE Carpet 66.0% Rug 66.3% Can 61.0% Canned 91.7% Cut Grass 48.5% Chesterfield 70.7% Sofa 39.5% Tap 62.1% Lying 73.7% Cutlery 46.4% Train Station 43.7% TV 94.6% Frying Pan 81.7% Stove 69.0% ^he first 25 items listed here are responses to visual-aural prompting; hence the responses are the forms used to refer to pictures shown. The remaining 17 items, beginning with 842 "Saltchuck - meaning" are responses to questions asked about local terms. Respondents were asked what each term meant, whether they had ever heard anyone say the word and whether they themselves use it; hence the responses to the last two questions are self-reporting only and cannot be checked for accuracy. 2 Picture of wall-to-wall shag "carpet" with several pieces of furniture on it. 3 Picture of an area covering with some design and a fringe around the edge. There is an item of furniture partly on the "rug" and several large figures on the floor near the "rug." 4 Picture of a cat on a bed. 230 ITEM RESPONSE OVERALL RATE 278. Blind[s]/Shade[s] 5 Bl ind 67.5% 281. Verandah/Porch l 6 V e r a n d a h 70.4% 283. Verandah/Porch 3 7 P o r c h 51.9% 284. Verandah/Porch 4 8 P o r c h 69.8% 285. Verandah/Porch 5 9 V e r a n d a h 57.3% 286. 2:45 (1st choice) Quarter to Three 90.0% 287. 2:45 (2nd choice) Two Forty-five 50.9% 290. 11:15 (1st choice) Quarter After Eleven 44.7% 291. 11:15 (2nd choice) Eleven Fifteen 45.5% 292. 10:30 (1st choice) Ten- th i r ty 84.3% 297. Z - Zee/Zed Zed 76.9% 842. Saltchuck - Meaning Sea, Ocean, Seawater 58.7% 843. Saltchuck - Heard Yes 91.7% 844. Saltchuck - Has Used Yes 55.0% 846. Skookum - Heard Yes 86.0% 848. Siwash - Meaning Don't Know 44.2% 849. Siwash - Heard Yes 52.2% 5 Picture of a window covering that is attached to a roller at the top of the window opening and pulls down to form a solid cover for the entire window; there is a fringe at the bottom and a cord for pulling the blind down or letting it up. 6 Stately Victorian home with turret and gables and a covered "verandah" across the front and extending partly down the sides of the house. 7 Small two storey house with steps going up to the front door and a mall gabled roof above the door. The starway and the roof are not connected except by the wall of the house. 8 Old three story house with stairs leading up to a small "porch" at the front door. The "porch" has stone posts and wood pillars leading up to a small gabled roof which covers the area. 9 Old farm house with a covered "verandah" extending across the front of the house. ITEM RESPONSE OVERALL RATE 850. Siwash - Has Used No 85.2% 850A. Siwash - Manner of Use Derogatory 72.2% 850B. Siwash - Heard Recently No 86.9% 851. Squamish - Meaning Don't Know 50.9% 853. Squamish - Has Used No 68.8% 854. Oolichan - Meaning [Small] Fish 56.0% 855. Oolichan - Heard Yes 84.0% 856. Oolichan - Has Used Yes 65.2% 857A. Saskie - Meaning Don't Know 95.6% 857D. Saskabush - Meaning Don't Know 61.3% 860. Slough - Spelling S lough 85.1% 232 G R A M M A T I C A L V A R I A B L E S 1 0 ITEM RESPONSE OVERALL RATE 658. Have you got? (Other used) N e v e r 1 1 72.4% 661. Behind/in back of me (Correct) B e h i n d 91.3% 662. Behind/in back of me (Other) Never 97.9% 665. Behind/in back o f 1 2 (Correct) B e h i n d 74.6% 666. Behind/in back of (Other Used) Never 96.0% 669. Drank/Drunk 1 3 (Correct) Drank 98.3% 670. Drank/Drunk (Other used) Never 98.7% 674. Has Drank/Drunk 1 4 (Other used) Never 93.2% 677. Bought[en] 1 5 (Correct) Bought 87.5% 678. Bought[en] (Other used) Never 95.2% 681. There is/are 8 eggs...(Correct) There Are 69.2% 682. There is/are (Other Used) Never 82.8% 685. Brought[en] /Brung 1 6 (Correct) Brought 96.0% 1 0 86 of the questions concerning grammatical variables were asked in pairs, one asking what the informant thought the "correct" form was, the other asking whether or not the informant used any form or forms other than the one judged "correct." Al l socio-eonomic groups had exceptionally high rates of negative response to the second question, that involving self-reporting of the forms they themselves use. Al l groups had rates ranging from 72.39% to 100.00% (average 90.71%) claiming that they "never" use forms other than those they identified as being correct. Although such responses are suspect and indicate the need for further testing to assess the accuracy of self-reporting, they are included in this appendix to show both the consistency of the four socio-economic groups and the unusually high rates of response. 1 1 The informants reported that they "never" use a form other than the one they deemed "correct." l 2 I n "The car is behind/in back of the garage." 1 3 I n "Yesterday he drank/drunk 3 glasses of milk." 1 4 I n "Today he has already drank[en]/drunk[en] 2 glasses of milk." 1 5 I n "Bought[en] bread isn't as tasty as homemade." 1 6 I n "I've often brought[en]/brung them home with me. 233 ITEM RESPONSE OVERALL RATE 686. Brought (other used) Never 98.0% 690. Between John and me/I (Other) Never 85.4% 694. Between You and me/I (Other) Never 87.7% 697. Helen and I/me 1 7 (Correct) Helen and I 97.9% 698. Helen and I/me (Other Used) Never 93.8% 701. Aren't I?/Am I n o t ? 1 8 (Correct) Aren't I? 67.9% 702. Aren't I? (Other used) Never 81.6% 705. L a y / L a i d 1 9 (Correct) L a y 68.3% 706. Lay/Laid (Other Used) Never 95.4% 709. Different from/to/than(Correct) F rom 60.5% 710. Different from/to/than (Other) Never 95.2% 713. Less/Fewer People (Correct) Fewer 71.3% 714. Less/Fewer (Other Used) Never 88.9% 718. Me - Object 2 0 (Other Used) Never 90.8% 722. Sneaked/Snuck 2 1 (Other Used) Never 86.6% 725. Doesn't /Don't 2 2 (Correct) Doesn't 96.0% 726. Doesn't/Don't (Other Used) Never 97.7% 730. Real/Really 2 3 (Other Used) Never 86.6% 733. Where are/Where's 2 4 (Correct) Where Are 80.4% 1 7 I n "Helen and I are going shopping." 1 8 In "I'm always short of money, aren't I/am I not/amn't I/ain't I? 1 9 I n "Yesterday he lay /laid/lied in the sun for 3 hours." 2 0 I n "He gave it to Tom and me." 2 1 In "They sneaked/snuck into the theatre." 2 2 I n "He doesn't/don't bother me." 2 3 I n "She's a real[ly] nice lady." 2 4 I n "Where's/'re/are my gloves?" 234 ITEM RESPONSE OVERALL RATE 734. Where are (Other Used) Never 90.4% 737. Any/No Money (Correct) A n y 99.7% 738. Any/No (Other Used) Never 100.0% 741. D ived /Dove 2 5 (Correct) Dove 73.6% 742. Dive/Doved (Other Used) Never 90.0% 745. Had seen/saw 2 6 (Cor rec t ) Had Seen 50.4% 746. Had seen/saw (Other used) Never 77.4% 749. If I was/were you (Correct) Were 93.8% 750. If I was/were (Other Used) Never 98.8% 753. Doesn't/Don't 2 7 (Correct) Doesn't like anything. 99.2% 754. Doesn't/Don't (Other Used) Never 99.6% 757. Let's not/Don't let 's 2 8 (Correct) Let's Not 97.1% 758. Let's not (Other Used) Never 98.3% 761. Anyway[s] (Correct) A n y w a y 86.3% 762. Anywayfs] (Other Used) Never 79.6% 765. Used not/Didn't u s e 2 9 (Correct) Never used to 75.8% 766. Used not (Other Used) Never 90.1% 769. Forgot[ten] 3 0 (Correct) Forgotten 94.0% 770. Forgot[ten] (Other Used) Never 98.3% 2 5 i n "He dived/dove into the water." 2 6 I n "If we had/had of/ hadda/would of/would have seen/ we saw you, we would have spoken to you." 2 7 I n "He doesn't/don't like anything/nothing." 2 8 I n "Don't let's/Let's not/ Let's don't take the bus." 2 9 I n "We used not/didn't use/usedn't/never used to go there." 3 0 I n "He must have forgot(ten) my name." 235 ITEM RESPONSE OVERALL RATE 773. Should Have31 (Correct) Should Have 83.3% 774. Should Have (Other Used) Never 93.9% 777. Had gone/went32 (Correct) Had gone 45.9% 778. Had gone/went (Other Used) Never 75.7% 782. These Kinds of Apples (Other) Never 96.0% 786. To who[m] (Other Used) Never 75.9% 789. Saw/Seen33 (Correct) Saw 96.3% 790. Saw/Seen (Other Used) Never 100.0% 794. If I was/were warmer (Other) Never 93.7% 797. Proved/Proven34 (Correct) Proven 75.4% 798. Proved/Proven (Other Used) Never 89.2% 802. Lend/Loan35 (Other Used) Never 86.6% 805. Ate/Had Eaten36 (Correct) Had Eaten 65.7% 806. Ate/Had Eaten (Other Used) Never 83.8% 810. It's me/I 3 7 (Other Used) Never 79.1% 813. Lie/lay down (Correct) Lie 78.7% 814. Lie/lay down (Other Used) Never 97.1% 817. Laid/Lain38 (Correct) Lain 54.8% 3 1 In "You shoulda/should of/should have seen him run." 3 2In "If I went/hadda gone/had of gone/would of gone/would have gone/had gone/had of went home, I would have found it." 3 3In "I saw/seen him yesterday." 3 4In "It has already been proven/proved." 35In "Can you lend/loan me $5?" 36In "After we ate/had eaten dinner, we played cards." 37In "(You knock on the door. In answer to the question "Who's there?", you answer:)" 3 8In "It has laid/lain/lied there all night." 236 ITEM RESPONSE OVERALL RATE 818. Laid//lain (Other used) Never 93.7% 821. Out back/in the back39 (Correct) In the back 52.9% 822. Out back/in the back (Other Used) Never 94.4% 825. Nice day, eh?40 No 47.9% 826. It goes over here, eh? No 61.8% 828. Think about it, eh? No 57.2% 829. What a game, eh! No 43.4% 830. What are they trying to do, eh? No 79.0% 831. This girl is up on the 27th floor, eh, there she gets out on the ledge, eh,... No 84.8% 832. Eh, what did you say? No 79.1% 832A. Thanks, eh? No 90.7% 832B. Comments about eh. Tries to avoid it/ Doesn't like it 42.3% 833. Gotten over cold41 ("Gotten" used) Yes - "Gotten" 45.2% 834. Gotten cooler42 ("Gotten" used) No - "Got" 39.9% 836. Gotten there43 ("Gotten" used) No - "Other" 49.5% 837. Gotten fixed44 ("Gotten" used) No - "Got" 37.7% 39In "(Someone asks you for something that is in the backyard, behind the house). You'll find it..." 40Nine sentences containing e_h. were given to the respondents and they were asked to tell which ones they would say. 4 1 In this and the following 4 items, respondents were asked not which form was correct, but in which sentence they would use "gotten". The sentence for this was "I haven't gotten over my cold yet." 4 2In "It had gotten cooler." 43In "She was supposed to have gotten there at 3." 4 4In "He had just gotten the refrigerator working when the power failed." 237 839. Gotten up45 ("Gotten" used) Yes - "Gotten" 42.6% ITEM RESPONSE OVERALL RATE 1065. Into/to/in46 Into 98.8% 1096. To/for47 To 83.0% 1226. Anyway[s]4 8 Anyway 98.3% 1435. Out of49 Of 77.7% 265. Lying/Laying50 Lying 73.7% 4 5In "You should have gotten up earlier." 46ln the Reading Passage 4 7 In the Reading Passage 4 8In Reading Passage ("You'll have plenty of time, anyway.") 4 9In the Reading Passage, (in "somebody who was looking out of the window") testing whether or not the "of is lexicalized and, if so, how it is realized. 5uIn Visual-Aural prompting; picture of cat on a bed. 238 PHONOLOGICAL VARIABLES ITEM AND STYLE RESPONSE OVERALL RATE 1. /VtV/ (butterful, tomato, thermometer.water..) Minimal Contrast /th/ 67.8% Reading Passage /d/f/ 76.1% 2A. /VtVn/ (button, Eatons, eaten) Minimal Contrast A ' / ? / 78.8% Word List A ' / 7 / 86.3% Visual-Aural A ' / ? / 91.7% Reading Passage A ' / ? / 90.2% 2B. /rtVn/ (curtain, carton, important) Minimal Contrast /rt'/r-V 83.8% Word List /rt'/r ?/ 83.6% Visual-Aural /rt'/rV 91.2% Reading Passage /rt'/r ?/ 96.5% 2C. /ntVn/ (mountain, sentence) Minimal Contrast /nth/ 61.7% Word List /nth/ 58.3% Visual-Aural /nth/ 62.5% Reading Passage /nt/ 51.9% 2D. /ttVn/ (sultana) Visual-Aural /*t»/*V 77.5% 3. /ntV/ (painting, mounted, centre, dentist...) Minimal Contrast /nth/ 91.4% Word List /nth/ 76.5% Series /n0/ 56.8% Visual-Aural /nth/ 69.2% Reading Passage /nt/ 47.7% 4. /rtV/ (Quarter, Alberta, thirty, shorter...) Minimal Contrast /rth/ 82.3% Word List /rth/ 57.3% Series /rd/rr/ 66.1% Visual-Aural /rd/rr/ 60.3% Reading Passage /rd/rr/ 65.8% 239 ITEM AND STYLE RESPONSE OVERALL RATE 5. /tiV/ (salted, melting, altar, insulted...) Minimal Contrast /* th/ 96.7% Word List /* th/ 88.2% Visual-Aural /*th/ 64.2% Reading Passage Iiii 88.1% 6. /3iC-v/ (light, bicycle, bike, rice...) Minimal Contrast /si/ 98.5% Word List /si/ 95.9% Visual-Aural /si/ 94.8% Reading Passage /si/ 97.4% 7. /s i tV/ (typewriter, united, lighter, invited...) Minimal Contrast /si/ 94.0% Word List /si/ 90.4% Visual-Aural /si/ 89.6% Reading Passage /si/ 83.6% 8. /AUC -V/ (house, mouth, south,, outside...) Minimal Contrast / A U / 93.5% Word List / A U / 92.6% Visual-Aural / A U / 89.8% Reading Passage / A U / 95.9% 9. /AUtV/ (outer, shouting, about a, out of) Minimal Contrast / A U / 91.5% Word List / A U / 95.8% Visual-Aural / A U / 91.3% Reading Passage / A U / 95.6% 10. An/ (Vbl) (painting, lying, working...) Minimal Contrast An/ 72.9% Word List An/ 68.6% Visual-Aural An/ 67.3% Reading Passage An./ 78.1% 240 ITEM AND STYLE RESPONSE OVERALL RATE 11. Aq/ (Nom) (Ceiling, awning, building...) Minimal Contrast Ar)/ 80.0% Word List A q / 72.2% V i s u a l - A u r a l A q / 78.1% Reading Passage A q / 71.9% 12. A q / (Pron) (something, anything, nothing) Word List A q / 87.4% V i s u a l - A u r a l A q / 84.6% Reading Passage A q / 89.7% 13. Aq / (other) (during, entering...) Minimal Contrast A q / 61.7% Word List A q / 66.7% Reading Passage A q / 67.0% 15. Ihv/f (why, wheelbarrow, what, where...) Minimal Contrast / w / 71.4% Word List / w / 78.5% V i s u a l - A u r a l / w / 79.9% Reading Passage / w / 77.3% 16. /aerV/ (married, carrot, barrow, marry...) V i s u a l - A u r a l It / 63.4% Reading Passage It/ 74.8% 17B. /t>/ (vase) Minimal Contrast la/ 50.8% Reading Passage / e i / 81.6% 17C. / W (khaki, garage) V i s u a l - A u r a l la/ 47.9% 17D. Ix>/ (aunt, half) Reading Passage 92.4% 241 ITEM AND STYLE RESPONSE OVERALL RATE 17E. tol (palm, calm, almond) Minimal Contrast lr> i 59.9% Word List /at 72.6% Visual-Aural lat 54.3% Reading Passage lat 58.5% 17F. tol (because, whatever, what else, want to) Minimal Contrast / A / 95.0% Word List / A / 75.8% Visual-Aural la/ 61.9% Reading Passage / A / 75.0% 17G. tol (balcony, album, alcohol) Word List / « / 53.0% Visual-Aural /© / 69.4% Reading Passage / 41.6% 18A. /u/ (good, cookie, books, foot, could...) Minimal Contrast lo t 91.9% Word List la t 76.5% Visual-Aural lo t 74.0% 18D. /u/ (your, you're, sure...) Minimal Contrast I u I 99.1% Word List In I 98.3% Reading Passage / u / 79.0% 19. Ar / (mirror, ear, earring, nearer, here...) Minimal Contrast lit 84.1% Word List lit 76.8% Visual-Aural lit 81.3% Reading Passage HJ 92.2% 242 ITEM AND STYLE RESPONSE OVERALL RATE 20. /eg/ (legs, veggies, pegs, begged...) Minimal Contrast / e i / e i / 65.3% Word List It / 48.2% V i s u a l - A u r a l / e i / e i / 64.2% Reading Passage It I 61.5% 21. laiil (sterile, futile, missile, mobile) Minimal Contrast /a i / 89.2% Word List / a i / 59.6% V i s u a l - A u r a l /a / 72.2% Reading Passage /at / 86.6% 22. /Va r / (intrusive a) (near, mere, our, here ear...) Word List mi 97.9% V i s u a l - A u r a l /a / 51.9% Reading Passage 10/ 98.2% 22A. / V a l / (intrusive a) (file, futile) Minimal Contrast 10/ 71.3% Word List / a / 60.9% V i s u a l - A u r a l /a / 61.8% Reading Passage /a / 54.9% 22C. / V a n / (intrusive a) (been, known) Minimal Contrast 10/ 97.2% Word List 10/ 88.8% Reading Passage / a / 67.8% 23. /o rV / (orange, borrow, foreign sorry...) Minimal Contrast lot 89.8% Word List /o / 86.3% V i s u a l - A u r a l /o / 78.1% Reading Passage /o / 85.9% 243 ITEM AND STYLE RESPONSE OVERALL RATE 24. Imi (Canada, that, cast, Vancouver, balance...) Minimal Contrast /ffi / 70.0% Word List /ffi / 77.5% Visual-Aural / 66.7% Reading Passage /ffi / 88.8% 25. <alm> (aim, balm, palm, psalm, almond, calm) Minimal Contrast Iv I 52.7% Word List /o / 55.9% Reading Passage /o / 54.4% 26. #and# Word List /nd/ 90.1% Visual-Aural /nd/ 97.9% Reading Passage /nd/ 93.4% 28. /ns/ (sense, dancing, once, sentence...) Reading Passage /nts/ 66.7% 29. lis! (false, pulse, else, wilson...) Minimal Contrast lis/ 75.3% Word List Its/ 81.9% Visual-Aural Us/ 80.4% Reading Passage lis/ 89.9% 30. /t # j/ (front yard, that yard, not yet...) Word List Aj/tjj/ 45.9% Visual-Aural P\t 36.6% Reading Passage P\/ 46.9% 31. /tj/ (natural, culture, temperature...) Word List AJ/ 99.2% Visual-Aural A | / 99.6% Reading Passage /tj/ 99.0% 32. / 9 r a / (temperature, naturally, cultural...) o Word List I0r9/ 52.3% Visual-Aural /srtf/ 56.1% Reading Passage lars/ 36.3% 244 ITEM AND STYLE RESPONSE OVERALL RATE 33. /C-alvju/ (regular, vocabulary, particular...) Word List / j u / j o / 73.7% V i s u a l - A u r a l / j u / j a / 94.8% Reading Passage / j s / 61.1% 34. /h-stress/ (her, his, him) Reading Passage / h / 76.5% 35. /c r i / (raspberry, library, Calgary...) Word List It / 71.8% Reading Passage It / 75.3% 37. / C i / (Prefix) (reduce, because, remind...) Minimal Contrast HI 87.7% Word List HI 89.5% Reading Passage /9 / 58.2% 38. <oun> (mountain, council) Word List / a o / 77.1% V i s u a l - A u r a l / a o / 87.9% Reading Passage /aco/ 98.5% 39. /nts#/ (chants, cents, prints, students) Minimal Contrast I nisi 69.9% V i s u a l - A u r a l / n t s / 79.6% SPECIAL PROFILE WORDS ITEM AND STYLE RESPONSE OVERALL RATE 40. Sandwiches Reading Passage 41. Vase Minimal Contrast Word List Visual-Aural 42. Garage Visual-Aural 43. Groceries Word List 45. Schedule Visual-Aural Reading Passage 46. Fifth Word List Reading Passage 47. February Reading Passage 48. Arctic Visual-Aural 49. British Columbia Word List Visual-Aural Reading Passage /seevidgaz/ 36.7% /vaz/ 50.4% /vaz/ 53.0% /vaz/ 65.6% /garag/ 32.6% /grosriz/ 45.3% /skedga*/ 32.5% /skedgua*/ 39.2% /fife/ 61.9% /fife/ 87.9% /febjuveri/ 36.0% /ortluk/ 45.1% /bridij ka*Ambia/ 46.0% /bndij ktAmbia/ 40.8% /bndij kaiAmbia/ 75.4% 246 ITEM AND STYLE 50. Quebec Word List V i s u a l - A u r a l 52. Vancouver Word List V i s u a l - A u r a l Reading Passage 53. Ottawa Reading Passage 55. W V i s u a l - A u r a l Reading Passage 56. Z V i s u a l - A u r a l 57. Beautiful V i s u a l - A u r a l Reading Passage 58. Often Word List Reading Passage 59. Again Word List 63. Milk V i s u a l - A u r a l Reading Passage RESPONSE /kvabek/ /kvabek/ /vsnkuvsr/ /vsenkuvar/ /vferjkuvar/ /dAb9*ju/ /dAbsju/ /zed/ /bjudsfo*/ /bjudsfo*/ / a f s n / / a f s n / /9gein/ /nutk/ /mUk/ OVERALL RATE 48.0% 54.0% 29.3% 36.3% 58.4% 55.9% 61.5% 69.3% 76.8% 71.2% 73.3% 40.7% 68.6% 57.0% 67.9% 74.7% 64. Congratulate Word List Reading Passage /...et/.../ /...©tj.../ 71.3% 61.2% 247 ITEM AND STYLE 65. Luxury Word List 66. Anything Word List Reading Passage 67. Anyway[s] Reading Passage 68. Anywhere Reading Passage 70. With Word List Reading Passage 71. Without Visual-Aural 72. Nuclear Word List 73. Always Word List Reading Passage 74. Either Word List 75. Neither Word List 76. Been Minimal Contrast Word List Reading Passage RESPONSE A..kj9.../ /eni6irj/ /eniSirj/ /enivei/ /eniver/ / V L 8 / / v i e / /VlSAUt/ /nuktiar/ /...weiz/ /...wiz/ /icfsr/ /nicfer/ / b i n / / b i n / / b i n / OVERALL RATE 83.9% 86.7% 69.7% 84.6% 73.3% 88.2% 73.0% 56.7% 50.4% 56.9% 97.9% 67.5% 67.1% 82.1% 88.2% 55.8% 248 ITEM AND STYLE 78. Were Minimal Contrast Word List Reading Passage 81. Oregon Minimal Contrast Word List 82. Kitsilano Word List Local Terms 84. Mary Reading Passage 85. Mom Reading Passage 86. eh Reading Passage 87. Prestige Word List 88. Sunday Reading Passage 91. Wednesday Word List Reading Passage 93. Friday Series RESPONSE /war/ /war/ /war/ /oragin/ /oragin/ /k i tsatainou/ /k i tsgfainou/ /mcri/ /mAm / /ei/ /prsstig/ /...dei/ /vcnadei/ /wen2dei/ /...di/ OVERALL RATE 83.9% 92.9% 99.8% 57.9% 42.5% 34.6% 41.7% 80.4% 91.6% 84.5% 66.0% 61.5% 61.2% 49.2% 55.3% 94. Saturday Reading Passage /SfEthardei/ 46.4% ITEM AND STYLE 95. Oolichan Local Terms 97. Caramel Visual-Aural 99. Vegetables Reading Passage 100. Pumpkin Visual-Aural 101. Something Word List Series Reading Passage 102. Tomato Visual-Aural Reading Passage 103. Iron Word List Visual-Aural 105. Recognize Word List 107. Hundred Series 108. Potato Visual-Aural RESPONSE OVERALL RATE /u*9k9n/ 54.5% /karma*/ 42.3% /vedztebtz/ 73.0% / . . .mpk.../ 88.3% /sAmpSirj/ 60.3% /SAmpSirj/ 55.9% /SAmp8ir j / 85.4% Aameidou/ 49.7% /t9meidou/ 59.2% /atgrn/ 77.3% /aL9rn/ 82.3% /rekggnaia/ 80.3% /hAndrgd/ 58.3% /pgtheidou/ 55.8% 109. Oranges Word List Reading Passage /orinj / /ornj / 48.0% 41.3% 250 ITEM AND STYLE 110. Unfortunately Reading Passage 111. Details Reading Passage 112. North Shore Reading Passage 114. Canadian Reading Passage 115. Our Minimal Contrast Word List Reading Passage 116. Guarantee Word List 118. Wheelbarrow Word List Visual-Aural RESPONSE /An.../ OVERALL RATE 100.0% stress syllable 1 98.7% /nor jor/ /k9neidisn/ /ar/ /aar/ /geeronti/ /wiiberou/ /viiberou/ 67.9% 97.7% 66.7% 73.5% 93.5% 43.8% 27.3% 35.0% 119. Going to Reading Passage 120. Library Word List Reading Passage other / t a i b r e r i / / i a i b r e r i / 32.0% 61.3% 43.7% 122. Mountain Reading Passage /macb'n/ 61.0% 251 APPENDIX E: FORMS USED BY THE MAJORITY VERSUS THOSE USED BY EACH SOCIO-ECONOMIC GROUP Reported in the following table are the forms which vary across socio-economic groups. Note that "majority" refers to relative rather than absolute majority; i.e. what is reported is the form used by the most speakers. In cases for which there were several possible variants, the relative majority may be as low as 22%; nonetheless, if that is the percentage given in the table, it represents the highest rate of agreement. What was being considered here was merely whether or not socio-economic group IV differed from the majority and from other socio-economic groups; hence the rate of use of a particular variant is not significant to the analysis. However, as some readers may be interested in which variants were preferred as well as the rate of agreement among speakers of a particular socio-economic group, I have included the tables in this appendix. Each linguistic level is considered separately, and tables are arranged by socio-economic group which differs from the majority. Lexical Items Socio-economic group IV 847. Skookum (Has Used) Majority 2) 55.2% IV 1) 50.8% 1) Yes 1 III 2) 52.7% 2) No II 2) 64.3% I 2) 52.6% 1 For each item, the socio-economic group is given (in Roman Numerals), then the code number (in arabic numerals) of the form chosen and the rate at which the form was chosen; the legend for the codes is given at the right. Hence the figure may be read 55.2% of the majority chose "no"; 50.8% of the fourth (highest) socio-economic group chose "yes", etc. 252 293. 10:30 (2nd choice)2 858. Majority 4) 52.1% IV 2) 46.2% 1) Ten-thirty III 4) 61.1% 2) Half past ten II 4) 53.2% 4) Does not use other I 4) 59.5% form Slough (Meaning)3 Majority 7) 22.0% 7) Run-off ditch-IV 8) 23.1% man-made III 7) 25.7% II 7) 25.0% 8) Dirty, dead, low-I 7) 23.4% lying water •economic groups IV and III Drapes/Curtains4 Majority 2) 55.3% IV 1) 52.3% 1) Curtains III 1) 46.0% 2) Drapes 2) 46.0% 3) Others II 2) 69.1% I 2) 64.9% Socio-economic group I 852. Squamish (Heard) Majority 1) 50.9% IV 1) 72.4% 1) Yes III 1) 54.2% 2) No II 1) 53.5% 3) Not Sure I 2) 76.3% 2 The first choice was 1) Ten-thirty. 3 The rates for this group are low because people identified eight possible meanings for the term, including 2) water channel, dried creek, 3) pond, small lake, 4) backwater, arm of river, 5) tidal pools, flats, inlet, 6) swamp, marsh, 7) run-off ditch—man-made, 8) dirty, dead, low-lying water, 9) backwater and tidal flats. With so many possibilities, the figures are thinly spread. 4 A picture showing solid coloured fabric hanging straight down either side of a small window. 253 845. Skookum (Meaning)5 Majority IV III II I 5) 5) 5) 5) 1) 25.3% 29.2% 28.4% 23.8% 27.3% 5) (very) good, fine, okay 1) Don't know Socio-economic group II 282. Verandah6 Majority IV III II I 2) 2) 2) 1) 2) 46.4% 39.0% 56.7% 50.0% 48.3% 1) Verandah 2) Porch 3) Others Socio-economic groups I & II 270. Train Tracks Majority IV III II I Socio-economic groups I & III 2) 2) 2) 3) 3) 37.7% 46.9% 44.6% 40.5% 52.0% 1) Train Tracks 2) Railway Tracks 3) Railroad Tracks 4) Other 298. Dinner7 Majority IV III II 4) 29.3% 4) 55.4% 1) 31.1% 4) 25.0% 5) 28.6% 1) Only Dinner 2) Only Supper 3) Dinner/Supper 4) Usually Dinner 5) Usually Supper 5 Rates are low because people gave five different definitions (2) big, 3) strong, 4) sturdy, husky, hefty, 5) (very) good, okay, fine, 6) other) and some said they didn't know. 6 A picture of the lower portion of what appears to be a drawing of a two-storey house. Two steps lead up to a covered "porch" extending across the front of the house. The "porch" has a railing supported by spindles, and spindled posts support the "porch's" roof. Four people, a man, woman and two children, sitting on the porch are drawn in the style associated with children's books and are depicted in fashions from the early twentieth century. 7 Responses to the question, "What do you call your evening meal?" Grammatical Items Socio-economic groups I and II 673. has drunk (correct)8 721. 729. Majori ty 2) 47.7% IV 2) 75.4% 1) d rank III 2) 55.4% 2) d r u n k II 1) 58.3% I 1) 62.3% sneaked/snuck9 (correct) Majori ty 1) 52.1% IV 1) 65.0% 1) sneaked III 1) 51.7% 2) snuck II 2) 53.3% I 2) 51.7% Real [ l y ] 1 0 (correct) Majori ty 2) 61.7% IV 2) 91.7% 1) rea l III 2) 75.0% 2) rea l ly II 1) 51.7% I 1) 51.7% 693. Between you and me/I (correct) Majori ty 1) 50.7% IV 1) 70.8% 1) me III 1) 55.4% 2) I II 2) 52.4% I 2) 62.3% 717. me - object 1 1 (correct) Majori ty 2) 52.3% IV 2) 80.0% 1) I III 2) 59.3% 2) me II 1) 66.7% I 1) 61.7% 8 In "Today he has already drank[en]/drunk[en] 2 glasses of milk." 9 In "They sneaked/snuck into the theatre." 1 0 In "She's a real[ly] nice lady." 1 1 In "He gave it to Tom and I/me." 785. To Who[m] (correct)12 Majority IV III II I 841. Got[ten] to try. (used) Majority IV III II 2) 2) 2) 1) 1) 43.9% 68.3% 45.0% 40.7% 46.7% 3) 29.5% 3) 45.2% 3) 41.8% 1) 27.3% 2) 27.3% 1) 36.2% 1) who...to 2) to whom 1) 3) 2) 4) yes no, +other no, +got reject 827. Still here, eh? (used) Majority 1) 44.2% IV 1) 50.8% III 1) 48.5% II 2) 49.3% I 2) 47.7% 1) 2) yes n o Socio-economic groups I and II (with some deviance) 840. Got[ten] an answer. Majority 3) 28.9% 2) 28.9% IV 3) 43.6% 1) Yes III 3) 32.4% 2) No, +got II 1) 29.6% 3) No, +other I 1) 32.9% 2) 32.9% Socio-economic group I 801. Lend/loan (correct)13 Majority 1) 57.7% IV 1) 73.3% 1) lend III 1) 59.3% 2) loan II 1) 53.3% I 2) 48.3% 1 2 In "To whom did you give the book?/Who did you give the book to?" 1 3 In "Can you lend/loan me $5?" 838. Got[ten] braced Majority 3) 33.3% IV 3) 51.6% III 3) 37.7% II 3) 31.6% I 1) 49.3% Socio-economic erouD II 793. If...was/were warmer. (correct) Majority 2) 54.2% IV 2) 60.0% III 2) 53.3% II 1) 53.3% I 2) 56.7% Socio-economic group IV 809. It's me/I (correct) Majority 1) 66.7% IV 2) 56.7% III 1) 71.7% II 1) 70.0% I 1) 81.7% 781. These kind[s] of apples (correct) Majority 2) 56.9% IV 1) 66.2% III 2) 47.3% II 2) 65.5% I 2) 79.7% 1) yes 3) no, -i-other 4) other - reject 2) no, +got 1) was 2) were 1) Me 2) I 1) these kinds 2) these kind Socio-economic groups IV and III 689. Between John and me/I (correct) Majority 1) 53.7% IV 2) 76.9% 1) I III 2) 48.7% 2) me II 1) 69.1% I 1) 72.7% Socio-economic groups IV and II 657. Have you got? (correct) Majority 2) 38.1% IV 3) 42.2% 2) Do you have.. HI 2) 48.7% 3) Have you...? II 3) 35.7% 1) Have you got. I 2) 35.1% 257 835. Got[ten] killed (used) Majority 1) 36.1% IV 3) 48.4% 1) yes III 1) 37.1% 3) no, +other II 3) 31.6% 2) no, +got I 1) 44.9% 258 Phonological Items Socio-economic group IV 1. /V tV / - (butter, tomato, writer, united...) Word List Majori ty 2) 49.3% IV 3) 62.6% 2) d / r III 2) 48.9% 3) th II 2) 55.0% I 2) 57.1% Series Majori ty 2) 54.2% IV 3) 55.8% 2) d / r III 2) 56.7% 3) th II 2) 62.5% I 2) 58.3% Visual-Aural Majori ty 2) 60.5% IV 3) 49.8% 2) d / r III 2) 59.5% 3) th II 2) 67.2% I 2) 66.9% /aerV/ (married, vary, narrow, carry...) Minimal Contrast Majori ty 2) 46.6% IV 1) 60.0% 1) £ III 2) 48.3% 2) C II 2) 57.8% I 2) 49.2% Word List Majori ty 2) 54.2% IV 1) 49.2% 1) £ III 2) 55.8% 2) C II 2) 60.8% I 2) 52.5% Ivi (tomato, vase...) V i s u a l - A u r a l Major i ty 5) 46.4% IV 2) 48.3% 2) III 5) 50.0% 5) ei II 5) 50.0% I 5) 54.2% 259 Socio-economic groups IV and III 14. /nju/, /tju/, /dju/ (newspaper, student, duke. Minimal Contrast Majority 2) 45.9% IV 1) 56.4% 1) ju/j© III 1) 47.7% 2) u/o II 2) 51.2% I 2) 56.2% 25. <alm> (Palm, almond, calm...) Visual-Aural Majority 1) 39.9% IV 3) 44.2% 1) x> III 3) 42.1% 3) 0 II 1) 39.8% I 1) 41.5% Socio-economic group III 17C A>/ (garage, album...) Reading Passage Majority 2) 38.1% IV 2) 35.6% 2) a III 4) 43.3% 4) 33 II 2) 46.7% I 2) 40.4% Socio-economic Group I 22. /Var/ (intrusive 8) (near, mere , our...) Minimal Contrast Majority 1) 52.9% IV 1) 54.4% 1) a III 1) 53.3% 2) 8 II 1) 59.9% I 2) 55.9% Socio-economic groups I and II 14./nju/, /tju/, /dju/. (newspaper, Tuesday, duke...) Series Majority 1) 43.5% IV 1) 68.3% 1) ju/jo III 1) 48.3% 2) u/o II 2) 31.7% I 2) 50.8% 260 Visual-Aural Majority 1) 43.3% IV 1) 58.3% 1) ju/j© III 1) 43.1% 2) u/o II 2) 47.6% I 2) 49.4% Word List Majority 1) 44.6% IV 1) 54.3% 1) ju/jtt III 1) 48.2% 2) u/o II 2) 49.1% I 2) 45.8% 28. /ns/ (once, sense...) Visual-Aural Majority 2) 50.4% IV 2) 55.3% 1) n S III 2) 55.2% 2) nts II 1) 50.5% I 1) 51.0% Socio-economic groups IV and I 28. /ns/ (sense, dancing, once...) Minimal Contrast Majority 2) 50.3% IV 1) 53.9% 1) n s III 2) 52.8% 2) nts II 2) 54.2% I 1) 50.3% Word List Majority 2) 49.4% IV 1) 50.8% 1) n s III 2) 51.9% 2) nts II 2) 52.0% I 1) 52.5% Socio-economic groups IV and II 35. /cri/ (raspberry, library...) Visual-Aural Majority 2) 38.9% IV 1) 40.0% 1) c III 2) 42.2% 2) II 1) 40.8% I 2) 43.8% 261 Socio-economic groups IV. Ill, and II 14. /nju/, /tju/, /dju/. (news, student, duke) Reading Passage Majority 2) 46.3% IV 1) 46.8% 1) ju/jo III 1) 49.1% 2) u/o II 1) 49.7% I 2) 54.7% Special Profile Items Socio-economic group IV 46. Fifth Visual-Aural 57. 74. 82. 83. Majority 1) 50.0% IV 2) 48.3% 1) flf8 III 1) 53.3% 2) fl8 II 1) 56.7% I 1) 48.3% Beautiful Word List Majority 1) 50.0% IV 2) 43.1% 1) bjudgfo* III 1) 51.4% 2) bjuthgfot II 1) 47.6% I 1) 61.0% Either Reading Passage Majority 2) 64.3% IV 1) 51.7% 2) III 2) 70.0% 1) aicfer II 2) 67.2% I 2) 71.7% Kitsilano Reading Passage Majority 1) 30.4% IV 6) 18.3% 1) kits9lainou III 1) 35.0% 6) kitsaleeno II 1) 38.3% I 1) 31.7% Capilano Word List Majority 1) 37.9% IV 3) 38.3% 1) kaepgiamou III 1) 45.0% 3) keepgfanou II 1) 45.0% I 1) 45.0% 263 97. Caramel Word List 102. Tomatoes Word List 108. Potatoes Word List 122. Mountain Word List Visual-Aural 40. Sandwiches Visual-Aural Majority 1) 31.9% IV 33) 30.8% 1) karma* III 1) 28.4% 33) other II 1) 34.1% I 1) 44.2% Majority 1) 35.0% IV 33) 46.2% 1) tameidou III 1) 33.8% 33) other II 1) 39.3% I 1) 41.6% Majority 1) 36.3% IV 2) 52.3% 1) patheidou III 1) 39.2% 2) patheithou II 1) 31.0% I 1) 45.5% Majority 1) 45.0% IV 2) 35.0% 1) macont'n III 1) 53.3% 2) maanthn II 1) 58.3% I 1) 45.0% Majority 1) 43.3% IV 2) 38.3% 1) maont'n III 1) 38.3% 2) maonthn II 1) 50.0% I 1) 56.7% Majority 1) 28.7% IV 3) 28.3% 1) s©mvidgaz III 1) 27.2% 3) s-Bndmdzaa II 1) 46.7% I 1) 45.0% 45. Schedule Word List Majority 1) 26.7% IV 33) 27.7% 1) skedgo* III 1) 23.0% 33) i other II 1) 29.8% I 1) 37.7% 94. Saturday Series Majority 2) 30.5% IV 1) 27.7% 2) seedsrdei III 2) 30.1% 1) seethgrdei II 2) 36.1% I 2) 31.2% 60. Lieutenant Word List Majority 2) 56.0% IV 1) 61.5% 2) tut... III 2) 50.0% 1) tcft... II 2) 60.7% I 2) 72.7% Socio-economic groups IV and III 48. Arctic Word List Majority 1) 32.9% IV 2) 54.7% 1) arthik III 2) 35.6% 2) arktluk II 1) 38.6% I 1) 37.3% 117. Dual Minimal Contrast Majority 1) 42.7% IV 2) 49.2% 1) dust III 2) 37.7% 2) djU9* II 1) 53.2% I 1) 50.7% 265 117. Dual Word List Majority IV III II I Socio-economic group III 109. Oranges Visual-Aural Majority IV III II I 121. Mirror Minimal Contrast Majority IV III II I 1) 2) 2) 1) 1) 1) 1) 2) 1) 1) 1) 1) 2) 1) 1) 31.3% 27.7% 35.1% 39.0% 35.5% 47.2% 46.2% 45.9% 54.2% 50.6% 40.5% 52.3% 44.6% 42.2% 40.3% 1) dU9* 2) djuat 1) ornj 2) orinj 1) mir: 2) our: 121. Mirror Visual-Aural 88. Sunday Series Majority IV III II I Majority IV III II I 1) 1) 2) 1) 1) 2) 2) 1) 2) 2) 41.5% 56.9% 36.5% 41.7% 38.2% 53.4% 50.8% 50.7% 52.4% 61.8% 1) mir: 2) our: 2) ...di 1) ...dei 89. Monday Series 92. Thursday Series Majori ty 2) 51.8% IV 2) 54.7% III II I Majori ty IV III II I 1) 54.1% 2) 51.2% 2) 57.1% 2) 2) 1) 2) 2) 53.3% 50.8% 50.0% 54.8% 58.4% 2) 1) 2) 1) .di .dei .di .dei Socio-economic group II 71. Without Word List 59. Again 121. Mirror Reading 94. Saturday Majori ty 1) 55.0% IV 1) 55.4% 1) VlSAUt III 1) 64.9% 2) VlcfAUt II 2) 52.4% I 1) 57.1% Passage Majori ty 1) 56.8% IV 1) 66.2% 1) sgen III 1) 62.2% 2) 9gein II 2) 50.6% I 1) 51.3% Passage Major i ty 4) 36.7% IV 4) 47.7% 4) m i r s r III 4) 43.2% 1) mir : II 1) 29.8% I 4) 36.4% t Majori ty 1) 38.3% IV 1) 50.0% 1) ssethardei III 1) 41.7% 2) s s d s r d e i II 2) 46.7% I 1) 33.3% 69. Sherbet Word List Majority IV III II I 104. Apricots Visual-Aural Majority IV III II I Socio-economic groups I and II 43. Groceries Visual-Aural Majority IV III II I 90. Tuesday Series Majority IV III II I 91. Wednesday Series Majority IV III II I 1) 54.7% 1) 67.7% 1) ...8t 1) 66.2% 2) ...art 2) 58.3% 1) 49.4% 5) 51.0% 5) 50.0% 5) ei... 5) 55.0% 1) ee... 1) 52.5% 5) 53.3% 1) 54.2% 1) 73.3% 1) grosriz 1) 66.7% 2) grojriz 2) 45.0% 2) 55.0% 1) 23.6% 1) 35.4% 1) tjuzdei 1) 30.6% 2) tuzdei 3) 17.9% 3) tjuzdi 1) 17.9% 4) tuzdi 2) 23.7% 4) 23.7% 1) 53.5% 1) 58.5% 1) venzdei 1) 60.3% 2) venzdi 2) 49.4% 2) 48.7% 1) 48.7% 80. New Westminster Word List 113. Preferable Reading Passage Socio-economic group I 58. Often Visual-Aural 105. Recognize Reading Passage 120. Library Visual-Aural 62. Leisure[ly] Word List Majority 1) 26.3% IV 1) 33.8% 1) njuvesminstar III 1) 31.1% 33) other II 33) 27.4% I 33) 29.9% Majority 1) 59.6% IV 1) 83.3% 1) stress syllable 1 III 1) 63.3% 2) stress syllable 2 II 2) 53.3% I 2) 55.0% Majority 5) 41.3% IV 5) 53.3% 5) of an III 5) 38.3% 3) ctfthan II 5) 47.4% I 3) 36.2% Majority 1) 61.7% IV 1) 81.5% 1) rekagnaiz III 1) 71.6% 3) rekanaiz II 1) 53.6% I 3) 49.4% Majority 1) 42.5% IV 1) 56.7% 1) •Jraibreri III 1) 36.7% 2) taibrari II 1) 46.7% I 2) 48.3% Majority 2) 52.5% IV 2) 66.2% 2) III 2) 62.2% 1) II 2) 50.6% I 1) 66.2% Reading Passage Without Reading Passage Khaki Word List Khaki Visual-Aural Newspapers Word List Visual-Aural Majority 2) 56.9% IV 2) 70.8% 2) III 2) 62.2% 1) ttj... II 2) 57.8% I 1) 61.0% Majority 1) 58.7% IV 1) 68.3% 1) Vl6AUt III 1) 60.0% 2) VL^AUt II 1) 58.3% I 2) 51.7% Majority 1) 42.0% IV 1) 46.9% 1) karki III 1) 50.0% 2) keki II 1) 39.0% I 2) 53.8% Majority 1) 42.4% IV 1) 53.4% 1) karki III 1) 41.7% 6) other colour II 1) 40.0% substituted I 6) 36.7% Majority 1) 30.7% IV 1) 44.6% 1) njus... III 1) 33.8% 2) nus... II 1) 27.4% 2) 27.4% I 2) 42.9% Majority 1) 29.7% IV 1) 33.3% 1) njus... III 1) 35.0% 2) nus... II 1) 26.7% I 2) 32.3% 42. Garage Word List Majori ty 1) 35.0% IV 1) 43.1% 1) ggrag III 1) 39.2% 2) gsreedg II 1) 36.9% I 2) 26.0% 99. Vegetables Word List Majori ty 1) 44.2% IV 1) 50.0% 1) vedgthabtz III 1) 43.3% 2) vedzdab^ z II 1) 50.0% I 2) 45.0% 47. February V i s u a l - A u r a l Majori ty 1) 20.1% IV 1) 30.0% 1) februeri III 1) 16.9% 2) febjugri II 1) 23.3% I 2) 33.3% 90. Tuesday Reading Passage Majori ty 1) 35.3% IV 1) 39.0% 1) tjuzdei III 1) 40.7% 2) tuzdei II 1) 40.0% I 2) 30.0% 61. Genuine Word List Majori ty 4) 53.0% IV 4) 76.9% 4) ...in III 4) 52.7% 1) ...ain II 4) 52.4% II 1) 62.3% Socio-economic groups II and III 63. Milk Word List Majori ty 2) 48.3% IV 2) 50.0% 2) mi3*k III 1) 43.3% 1) miik II 1) 46.7% I 2) 60.0% 121. Mirror Word List Majority IV III II I 53. Ottawa Visual-Aural Majority IV III II I Socio-economic groups IV and II 40. Sandwiches Word List Majority IV III II I 104. Apricots Word List Majority IV III II I Socio-economic groups I and III 42. Garage Reading Passage Majority IV III II I 1) 49.2% 1) 61.7% 1) mir: * 2) 50.0% 2) ffltr: 2) 43.3% 1) 50.0% 2) 11.7% 2) 16.7% 2) 8) 15.0% 7) ad8V9 7) 16.7% 8) 2) 13.3% 1) 24.7% 3) 24.6% 1) seemvidzaz 1) 21.6% 2) ssenvidzsz 2) 23.8% 3) seendwidzsz 1) 36.4% 5) 52.7% 1) 49.2% 5) ei... 5) 54.1% 1) ee... 1) 48.8% 5) 61.0% 1) 22.0% 1) 29.7% 1) gsrag 33) 21.6% 33) other 1) 25.0% 33) 24.3% 99. 108. Vegetables V i s u a l - A u r a l Majori ty IV III II I Potatoes Reading Passage Major i ty IV III II I 1) 1) 2) 1) 2) 3) 3) 5) 3) 5) 47.9% 66.7% 43.3% 46.7% 40.0% 32.0% 38.5% 36.5% 34.5% 26.0% 1) vedgthsbfz 2) vedzdab^z 3) psteidou 5) psteido No discernable pattern by socio-economic group 47. February Word List Major i ty IV III II 53. Ottawa Word List 4) 15.4% 5) 15.4% 5) 21.7% 33) 20.0% 3) 23.3% I 4) 20.0% Majori ty IV III II I 8) 2) 8) 12.7% 16.9% 14.9% 33) 17.6% 1) 15.6% Even distribution between two forms 104. Apricots Reading Passage Majori ty IV III II I 1) 1) 5) 1) 5) 1) 5) 50.4% 50.0% 50.0% 50.0% 50.0% 58.3% 56.7% 3) feb juver i 4) febjueri 5) f c b r u v e r i 33) o ther 1) adswa 2) pths-wa 8) u t h g v a 33) o ther 1) 5) ei... m... 273 APPENDIX F: "CORRECT" GRAMMATICAL FORMS 657 1 Have vou got enough money? Do you have 38.1% 669 Yesterday he drank/drunk 3 glasses... d r a n k 98.3% 673 TnHay he has alreadv drank/drunk. . . d r u n k 47.7% 677 Boughtfen") bread isn't as tasty as... bought 87.4% 685 T've often broughtCen") them home... b rought 96.0% 701 T'm always short of money, aren't I? Aren't I? 67.9% 713 There are less/fewer people here than... fewer 71.3% 721 They sneaked/snuck into the fridge. sneaked 52.1% 729 She's a reaKly 1) nice ladv. rea l ly 61.7% 737 T didn't see any/no monev in the box. a n y 99.7% 741 He dived/dove into the water. dove 73.6% 757 Let's not take the bus. Let's not 97.1% 765 We used not to go there. never used 75.8% 773 You should have/of/shoulda seen him... should have 83.3% 781 Do you like these kind(s) of apples? these kind 56.9% 789 I saw/seen him vesterdav. saw 96.3% 797 Tt's already been proved /proven . p r o v e n 75.4% 801 Can vou lend/loan me $5 lend 57.7% 805 After we ate/had eaten dinner, we played...had eaten 65.7% 681 There is/are 8 eggs in the fridge. there are 69.2% 733 Where are mv gloves? where are 80.4% 749 If I was/were vou. I'd vote against it. were 93.8% 793 If it was/were warmer, we could go for... were 54.2% 689 Mary is sitting between John and I/me. I 53.7% 693 Just between vou and me/I. I think... m e 50.7% 717 He gave it to Tom and me/I. m e 52.3% 785 To who(m) did vou give the book? to whom 43.9% 697 Helen and I are going shopping. Helen and I 97.9% 809 It's me/I! m e 66.7% 725 He doesn't/don't bother me. doesn't 96.0% 753 He doesn't like anything. a n y t h i n g 99.2% 661 He. was standing behind me. b e h i n d 91.3% 665 The car is behind the garage. b e h i n d 74.6% 709 My house is very different from yours. f r o m 60.5% 821 You'll find it out back. in the back 52.9% 761 Well. anvwavCs'). that's over. anyway 86.3% 705 Yesterdav he lav/ la id in the sun. lay 68.3% 813 L ie / lav down right away! l ie 78.7% 817 It has la id / la in there all night. l a i n 54.8% 745 If we had seen vou. we would have... had seen 50.4% 777 If I had gone home. I would have... would have gone 45.9% 769 He must have forgotCten1) mv name. forgotten 94.0% Numbers correspond to those in the questionnaire. APPENDIX G: DICTIONARIES AND HANDBOOKS CONSULTED Dictionaries American Heritage Dictionary of the English Language, New College Edition. Boston: Houghton Mifflin, 1978. Fowler, H.W. 1965. Dictionary of Modern English Usage, 2nd edition. Revised by Ernest Gowers. Oxford: Oxford University Press. Funk and Wagnalls Standard College Dictionary, Canadian Edition. Toronto: Fitzhenry and Whiteside, 1974. Gage Canadian Dictionary . Toronto: Gage, 1983. Oxford Advanced Learner's Dictionary of Current English. London: Oxford University Press, 1974. Oxford English Dictionary. Compact Edition. 2 vols. London: Oxford University Press, 1971. Random House Dictionary of the English Language: College Edition. New York: Random House, 1968. Sanigan, Gerald D. 1986. Canadian Words and Phrases—Legal Maxims. 3 vols. Don Mills: Richard DeBoo. Webster's New Collegiate Dictionary. 8th ed. Springfield, MA: G. & C. Merriam, 1976. Winston Canadian Dictionary for Schools. Toronto: Holt, Rinehat and Winston of Canada, 1974. Yogis, John. 1984. Canadian Law Dictionary. Toronto: Barrons Educational. Handbooks Baker, Sheridan. 1972. The Complete Stylist. 2nd edition. New York: Thomas Crowell. Buckley, Joanne. 1987. Fit to Print: The Canadian Student's Guide to Essay Writing. Toronto: Harcourt Brace Jovanovich. Crews, Frederick B. 1977. The Random House Handbook, 2nd edition. New York: Random House. 275 Department of the Secretary of State of Canada. 1985. The Canadian Style: A Guide to Writing and Editing. Malcolm Williams and Frank Bayerl, editors. Toronto: Dundurn Press. Elsbree, Langdon, Nell G. Altizer and Paul V. Kelly. 1981. The Heath Handbook of Composition. 10th edition. Lexington, MA: D.C. Heath Fowler, H.R. 1980. The Little, Brown Handbook. Boston: Little, Brown. Hodges, John C. and Mary E. Whitten. 1986. Harbrace College Handbook For Canadian Writers. 2nd edition. (Editorial consultant, Bruce R. Lundgren, University of Western Ontario. Foreword by Peter Desbarats, Dean, Graduate School of Journalism, Western Ontario.) Toronto: Harcourt Brace Jovanovich. Messenger, William E. and Jan de Bruyn. 1980. The Canadian Writer's Handbook. Scarborough: Prentice-Hall. Millward, Celia and Jane Flick. 1985. Handbook for Writers: Canadian Edition. Toronto: Holt, Rinehart and Winston of Canada. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0098220/manifest

Comment

Related Items