Gender-Related Variation in the Speech of English and Romanian Adolescents [1 ed.] 9781443812863, 9781443897327

Citation preview

Gender-Related Variation in the Speech of English and Romanian Adolescents

Gender-Related Variation in the Speech of English and Romanian Adolescents By

Costin-Valentin Oancea

Gender-Related Variation in the Speech of English and Romanian Adolescents By Costin-Valentin Oancea This book first published 2016 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2016 by Costin-Valentin Oancea All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-4438-9732-9 ISBN (13): 978-1-4438-9732-7

TABLE OF CONTENTS

List of Tables ............................................................................................. vii List of Figures........................................................................................... viii Acknowledgements .................................................................................... ix Introduction ................................................................................................. 1 Chapter One ................................................................................................. 7 Language and Gender: Past and Present Introduction............................................................................................ 7 1.1. From sex differences to gender variation in language use .............. 7 1.2. Men’s language versus women’s language ................................... 12 1.3. Gender-related stereotypes in English .......................................... 31 1.4. Gender-related stereotypes in Romanian ...................................... 33 1.5. Concluding remarks ...................................................................... 41 Chapter Two .............................................................................................. 43 Language Variation and Change: The Theoretical Framework Introduction.......................................................................................... 43 2.1. Variationist sociolinguistics .......................................................... 43 2.2. Communities of practice ............................................................... 67 2.3. Politeness and its relevance to the study of language and gender ... 70 2.4. Concluding remarks ...................................................................... 75 Chapter Three ............................................................................................ 76 Talkin’ or Talking? Phonological Variation in the Speech of London Teenagers Introduction.......................................................................................... 76 3.1. The (ing) variable: previous research ............................................ 77 3.2. Data collection and methodology.................................................. 83 3.3. (ing) patterns in the speech of London teenagers .......................... 87 3.4. Concluding remarks .................................................................... 104

vi

Table of Contents

Chapter Four ............................................................................................ 106 PE vs PĂ: Phonological Variation in the Speech of Romanian Adolescents Living in ConstanĠa Introduction........................................................................................ 106 4.1. (pe) in Early Modern Romanian ................................................. 107 4.2. (pe) in Modern Romanian ........................................................... 110 4.3. Data collection and methodology................................................ 111 4.4. (pe) patterns in the speech of Romanian adolescents living in ConstanĠa.................................................................................. 119 4.5. (pe) variation in the speech of Romanian adults ......................... 128 4.6. Adolescent speech vs adult speech ............................................. 134 4.7. Concluding remarks .................................................................... 136 Chapter Five ............................................................................................ 138 Adolescents as Language Innovators: Swearing, Taboo Words and Slang in English and Romanian Introduction........................................................................................ 138 5.1. Taboo language ........................................................................... 139 5.2. Bad language and sweet words ................................................... 142 5.3. Swearing, taboo language and slang in the speech of London adolescents ................................................................................... 156 5.4. Teenagers as language innovators: swearing and slang in Romanian ................................................................................. 167 5.5. Concluding remarks .................................................................... 181 Conclusions and Further Research .......................................................... 182 Bibliography ............................................................................................ 196

LIST OF TABLES

Table 1.1. Phonological differences between the speech of men and women in Chukchee Table 1.2. Phonological differences between the speech of men and women in Yukaghir Table 1.3. Phonological differences in male and female speech in Karajá Table 1.4. Morphological differences between the speech of men and women in Knjrux Table 1.5. Morphological differences between the speech of men and women in Knjrux Table 1.6. Lexical differences between the speech of men and women in traditional Japanese Table 1.7. Japanese personal pronouns Table 1.8. Gender-related stereotypes in Romanian Table 1.9. Gender-related stereotypes in Romanian across 3 age-groups Table 1.10. Gender-related stereotypes in Romanian Table 1.11. Gender-related stereotypes in Romanian Table 1.12. Gender-related stereotypes in Romanian Table 1.13. Gender-related stereotypes in Romanian Table 2.14. Preferences for [ƾ] and [n] endings, by gender Table 2.15. Gender-based differences in non-standard features of Reading English Table 3.16. Preferences for [ƾ] and [n] endings, by gender Table 3.17. Preference for [ƾ] and [n] in same gender speech Table 3.18. Percentage of [n] in men’s and women’s speech across studies Table 3.19. Information about the participants Table 3.20. The use of [n] and [ƾ] by London teenagers according to word class Table 3.21. thing compounds in the speech of London teenagers Table 4.22. Information about the Romanian speakers Table 4.23. Information about data collection Table 5.24. Types of swearing Table 2.25. Contrasting X-phemisms Table 5.26. The top ten proper slang words in COLT Table 5.27. The top ten dirty slang words in COLT Table 5.28. Slang and gender Table 5.29. Slang, gender and age Table 5.30. Information about the Romanian speakers Table 5.31. Top ten dirty slang words in the ConstanĠa corpus Table 5.32. Slang and gender in Romanian

LIST OF FIGURES

Figure 2.1. Social and regional variation Figure 2.2. Labov’s decision tree for stylistic variation in the sociolinguistic interview Figure 2.3. Social stratification of (r) in New York City Figure 2.4. Distribution of simplified clusters for variable (t,d) by gender in York, UK Figure 2.5. Gender-related differences in Sydney, Australia Figure 2.6. Percentage of backed tokens of [ԥ] Figure 2.7. Percentage of extreme raised tokens of [aj] Figure 3.8. Percent [n] for individual fraternity men across activity types for progressive forms only Figure 3.9. The number of (ing) tokens from each interview Figure 3.10. The use of [ƾ] and [n] in the speech of London adolescents Figure 3.11. The use of the [n] and [ƾ] according to the gender of the speaker Figure 3.12. Distribution of alveolar and velar variants according to the word class Figure 3.13. The use of (ing) according to word class and gender Figure 3.14. Pattern of alveolar variants of variable (ing) among nouns compared with indefinite pronouns in York and Toronto Figure 3.15. The use of thing compounds according to gender Figure 3.16. The results for reading style according to the gender of the speaker Figure 4.17. Variants [pe] and [pă] according to the gender of the speaker Figure 4.18. The use of [pe] and [pă] in the speech of adolescents from ConstanĠa Figure 4.19. The use of (pe) according to the gender of the speaker and group affiliation in the playground Figure 4.20. The distribution of [pe] and [pă] for each speaker Figure 4.21. The use of [pe] and [pă] according to the gender of the speaker Figure 4.22. The distribution of [pe] and [pă] according to gender in Hornoiu’s (2007) corpus Figure 4.23. Age and standard speech Figure 4.24. Age and standard speech in Romanian

ACKNOWLEDGEMENTS

This book is based on my PhD dissertation, defended on November 21st 2014, at the University of Bucharest, Romania. It represents the result of three long years of intensive research and wonderful opportunities which led to its current form. First and foremost, my deepest gratitude goes to my PhD supervisor, Professor Andrei Avram, University of Bucharest, who has been a mentor, role model and critic in the best of sense. He believed in me from the very beginning and persuaded me to read more about the intricate and exciting area of sociolinguistics. Professor Avram has taught me, among other things, the importance of doing research that you can truly deem your own. I am extremely grateful to Associate Professor Diana Hornoiu from Ovidius University of ConstanĠa for her support, bibliographic materials and useful suggestions. She has always been there for me with a good piece of advice, and she has encouraged me at various stages of this research project. Also, I want to thank Professor Jenny Cheshire who has welcomed me at Queen Mary’s College, University of London, even though she did not know me, allowed me to attend her courses and helped me to collect data for my English corpus. Her suggestions and expert guidance have been invaluable for me. There are also other wonderful people who have contributed one way or another to this work and to my formation as a young researcher. I would like to thank Professor Larisa Avram, Professor Ileana Baciu, and Professor Alexandra Cornilescu in particular, for the wonderful courses that they taught in the MA and PhD programs. Their passion for linguistics has been contagious. I cannot forget the guidance and encouragements that I received back at home from Associate Professor Camelia Bejan and Teaching Assistant Nicoleta Sava, who persuaded me to follow a career in linguistics. I am also indebted to my family, friends, colleagues, and my students who allowed me to record them and use their conversations in this study. There is also a very special person to whom I would like to thank for the wonderful memories that we have and for encouraging me to follow my dreams. Thank you, Cristina.

INTRODUCTION

This study represents a synchronic sociolinguistic analysis of genderrelated variation in the speech of English and Romanian adolescents. This research is motivated by the belief that variation is a characteristic of natural language, and a comprehensive understanding of language must include a grasp of the nature and function of variation. The aim is to analyse some sociolinguistic features of adolescent speech that occur in natural, spontaneous, everyday speech. This implies that the chief contribution is to the study of language in its social context. The core of this research centres on chapters 3, 4 and 5. The last section of Chapter 1 also introduces a few key ideas regarding gender stereotypes in Romanian. The issues addressed in these chapters can be roughly divided into two major groups. Chapters 3 and 4 offer a quantitative analysis of phonological variation in English and Romanian respectively. The main objective of Chapter 5 is to provide a qualitative analysis of foul language and slang in English and Romanian, which makes it a contrastive chapter. Chapter 1, “Language and gender: past and present”, offers a review of the relevant literature on language and gender research carried out in English-speaking communities, which does not claim to be exhaustive. The chapter starts by discussing the difference between sex and gender, the former being biological and the latter social. The next part of the chapter examines the idea that women are more polite than men, an idea which led Robin Lakoff (1975) to introduce the term “women’s language”. A distinction is drawn between gender-exclusive differences (see Borgoras 1922, Ekka 1972, Fasold 1990, Trudgill 2000, Bradley 2006, Meyerhoff 2006, Talbot 2010), which index gender directly, and gender-preferential differences which index gender indirectly. Section 1.2.2 sketches the most relevant theories of sex and gender developed over the years. The first one – the deficit approach – is the theory proposed by Lakoff (1975) who, albeit based on no empirical research, discusses some of the linguistic features of “women’s language”. The discussion moves to the second theory – the dominance model – associated with Zimmerman and West’s (1975, 1983) and Tannen’s (1994) analyses. Men are seen as dominant, whereas women are insubordinate and seen as an oppressed group. The overview moves swiftly to the third model – the difference approach – which is based on the work of Gumperz

2

Introduction

(1982) who stresses the idea that men and women should be seen as belonging to different subcultures. The last theory – the dynamic approach – argues for a social constructionist perspective as far as gender identity is concerned. These theories are not mutually exclusive and they can all be used to create a more unified and cohesive approach to the study of gender variation in language use. The third part of this chapter looks at gender-related stereotypes in English (Coates 1993, Crawford 1995, Holmes 1995, Fleischman 1998). Some of the gender stereotypes tackled here have been confirmed by quantitative sociolinguistic studies (e.g. Cheshire 1982, Newbrook 1982, Trudgill 1983, Romaine 1984, Eisikovits 1988, Cheshire 1997) supporting the idea that women tend to use more standard features of language than men irrespective of speech style. Starting from this hypothesis, I set out to identify gender-related stereotypes in Romanian, which constitutes the last part of this chapter. The analysis reports on the findings of two surveys which were conducted in ConstanĠa and Bucharest. The first survey focuses on M.A. students studying at Ovidius University of ConstanĠa and at the University of Bucharest. The second survey covers three age groups, carried out at one school and high school in ConstanĠa. The aim of these surveys is to understand the Romanian teenagers’ beliefs about the language they use and identify some of their linguistic choices. Chapter 2, “Language variation and change: the theoretical framework”, discusses the key concepts underlying this study and proposes a theoretical background in accordance with the chosen topic. The chapter opens with a discussion of variationist sociolinguistics (Labov 1966, Trudgill 1974, Cheshire 1982, Tagliamonte 2006, Kiesling 2011, Tagliamonte 2011), stressing the importance played by variation in the analysis of genderrelated differences in language use. The sociolinguistic interview (Labov 1966, 1972, Trudgill 1974, Shilling 2013) is discussed at length and included in this discussion are also the steps that must be followed or avoided in order to conduct a first-hand sociolinguistic research in a community. Central to this chapter is the discussion regarding the linguistic variable, which is the source of linguistic variation. I give a broad-brush picture of some of the main studies that focused on linguistic variables to analyse gender variation, and phonological variables are discussed first as variationist sociolinguistics was built on the study of phonological variation. The discussion then moves to morphological and syntactic variables which have been analysed quantitatively and presents studies focusing on the language spoken by adolescents. The concept “communities of practice”, coined by Lave and Wenger (1991), as well as

Gender-Related Variation in the Speech of English and Romanian Adolescents

3

politeness (Brown and Levinson 1987) are highlighted in relation to their relevance to the study of gender variation. The chapter aims at offering a unifying approach for the study of gender-related linguistic variation, by integrating a discussion of variationist sociolinguistics, which is the theoretical framework for Chapters 3 and 4, with a discussion of (im)politeness theory (Brown and Levinson 1987, Culpeper 1996) and of swearing and taboo language (Andersson and Trudgill 1990, Allan and Burridge 2006) for Chapter 5. Chapter 3, “Talkin’ or talking? Phonological variation in the speech of London teenagers”, is concerned with the quantitative analysis, in the variationist sociolinguistics framework. The analysis starts with a critical overview of the research on the (ing) variable in different Englishspeaking communities (Fisher 1958, Labov 1966, Shuy et al. 1967, Anshen 1969, Cofer 1972, Reid 1978, Woods 1979, Wald and Shopen 1985, Schleef et al. 2011). The reason I chose to study the (ing) is that it is a stable sociolinguistic variable, it can be analysed auditorily, and it is salient among native English speakers. The novelty of the approach consists in the fact that there are no studies, at least to my knowledge, which analyse the (ing) variable from a sociophonetic perspective. The majority of studies focusing on (ing) use programs like Goldvarb (Sankoff 1985, 1988, Paolillo 2001, Tagliamonte 2006) or the more recent Rbrul (Johnson 2009). Instead, I use PRAAT1 (Boersma 2001) for an acoustic analysis. The analysis is based on data collected at Queen Mary’s College, University of London in October 2012, as part of a research project carried out at this university under the guidance of Professor Jenny Cheshire. Eight undergraduate students (four males and four females) were interviewed, resulting in a corpus of 14,000 words. This chapter focuses on the English used by London teenagers and seeks to identify patterns of phonological variation regarding the use of (ing). In the literature there are two phonetic realisations of (ing), either as a velar nasal [ƾ] or as an alveolar nasal [n]. The former is associated with prestige while the latter with the vernacular (Trudgill 1974). The use of the velar nasal form is associated with the speech of women, as it is considered the prestigious variant whereas the alveolar nasal form is associated with the speech of men as it is deemed the vernacular variant. After reviewing the literature on (ing) the issue of data collection and methodology is addressed. This part is based on three important points: 1 I used version 5.3.84 of PRAAT. The soft program can be downloaded here (last accessed 07.07.2014).

4

Introduction

speech style, recording procedures and the subjects. In the speech style section I discuss the format of the sociolinguistic interview which contains three parts: a semi-structured questionnaire with 23 questions (casual speech style), three texts (reading style) and a list of words (word list style). The sociolinguistic interview is built in the Labovian style. The second section, recording procedures, discusses the technical equipment used to record the teenagers and the technique(s) used to interview students. The last section provides information about the subjects that took part in the research project and the fieldwork procedures. The last part of this chapter presents the results and is divided into two important parts: casual speech style and reading style. I show that the alveolar nasal variant is preferred by both sexes, which is contrary to the results obtained so far. This subsection also includes spectrograms to offer a better understanding and show the realisation of the velar and alveolar variants. Another aspect presented is the distribution of alveolar and velar forms according to word class. It is shown that the [n] variant is preferred with common nouns, progressive verb forms and gerunds, while the [ƾ] form is used in adjectives, discourse markers and pronouns. In the reading style I show that irrespective of speech style, the results do not change, and the alveolar nasal variant is the one preferred. Given the lack of sociolinguistic studies focusing on phonological variables in Romanian, chapter 4, “Pe vs Pă: Phonological variation in the speech of adolescents living in ConstanĠa”, offers a fresh and innovative perspective in this area. The chapter begins by presenting the evolution of the (pe) variable throughout centuries, starting from the form (pre) and evolving to the modern (pe). This variable is studied in the variationist sociolinguistics framework with the purpose of identifying and analysing sociolinguistic patterns. As already mentioned, the first part of the chapter tackles the (pe) variable in early modern Romanian, and provides relevant examples from 16th to 19th century textual evidence with the aim of buttressing the arguments advanced. The transition from pre to pe takes place somewhere in the 17th century, but the form pre appears in a text from the 19th century, which is peculiar. The variation between (pe) and (pă) is encountered in texts from the 18th century, thus showing that (pe) can be considered nowadays a stable sociolinguistic variable, just like (ing). The discussion continues with a review of the methodology and the methods which have been used to collect data. The research project was carried out at the Educational Center Theoretical High School in ConstanĠa. Two methods were used to collect data: the sociolinguistic interview and long-term participant observation. As far as the sociolinguistic

Gender-Related Variation in the Speech of English and Romanian Adolescents

5

interview is concerned, the analysis relies on a corpus of 15,000 words. Twelve students were interviewed using a semi-structured questionnaire which was followed by a reading task. The subjects who took part in this research are both secondary school and high school students, thus another social factor was taken into account besides gender, i.e. age. Section 4.4. is the core of this chapter as the results are presented here. There are three lines of discussion, depending on the speech style: casual speech style, reading style and the results obtained via participantobservation. The results that fall in this category were labelled “playground results”. Regarding casual speech style I show that the prestigious form [pe] is favoured in most of the cases, unlike English where the vernacular form was used more frequently. There are also agerelated differences, with high school boys using the non-standard form [pă] more often than secondary school students. It is also argued that [pă] is a marker of masculinity which shows group membership. I contend that in the playground, students are more relaxed and more prone to use non-standard features of language. The results obtained indicate that two communities of practice are formed: pops and geeks. The former are interested in showing their social and financial security through clothing (a powerful social marker), behaviour and interactions with their peers. The latter are preoccupied with school activities, gaining the respect of their teachers and parents and focusing on their career prospects. In order to fathom the variation between [pe] and [pă] the language of adults has also come under scrutiny. The analysis draws on eight hours of conversation gathered from TV talk-shows in 2013–2014. Five men and five women were recorded and 256 tokens of (pe) are interpreted. I build on the idea that [pă] is manly, tough, powerful, whereas [pe] is softer and more delicate thus paving the way to the construction of feminine identity. Moreover, (pe) could be construed as a gender marker. I also compare the results obtained with the realisation of (pe) in Hornoiu’s (2007) corpus. The last part of this chapter is a comparative overview of the use of (pe) by adolescents and adults. In the case of adolescents, the fact that they use the [pă] variant more on the playground engenders an awareness of social pressure. Adults on the other hand, especially male speakers, have a proclivity for the nonstandard form [pă]. Therefore, I stress that with age, there appears to be a rise in the use of [pă], since the willingness to be conform to societal rules decreases. Chapter 5, “Adolescents as language innovators: Swearing, taboo language and slang in English and Romanian”, analyses the use of foul language by adolescents and compares the use of such language in two different cultures: English and Romanian. The discussion starts by tackling

6

Introduction

the concept of “taboo language” and what should be deemed taboo. The subsequent section is an inquiry into bad language and sweet words, stressing the importance of euphemisms in contemporary speech. The examination continues with a scrutiny of “slanguage” as used by Stenström et al. (2002). This is an umbrella-term and includes: proper slang, dirty words, vogue words, vague words, proxy words and smallwords. Proper slang, dirty words and vogue words are discussed at length and relevant examples are provided. Section 5.3. looks at swearing and slang in the speech of London adolescents. The discussion is based on data from The Bergen Corpus of Teenage Language (COLT) and Stenström et al.’s (2002) book which is based on COLT. Included here is an analysis of gender-related differences in conversations about sex. I argue that boys and girls swear alike, the main difference lies in the fact that girls use female genitalia to refer to boys. Regarding the use of slang in COLT, many interesting facts are revealed. First, I show that proper slang and dirty words are used more frequently by boys than girls. I introduce another sociolinguistic variable, i.e. age and show that there are also age-related differences. Older girls (17–19 years old) use proper slang and dirty words more often than boys. From this finding girls’ use of foul language makes them appear more secure and powerful, thus dominating the scene. Boys’ use of dirty words can be related to the concept of toughness, usually associated with lower classes. The last part of this chapter dwells on the idea of teenagers as language innovators and focuses on their use of swearing and slang in Romanian. The study is based on a corpus of 6500 words collected in 2011–2012. The corpus consists of a three hour video recording of a group of teenagers spending New Year’s Eve in Buúteni. They are all born and raised in ConstanĠa and currently studying in this city. To this recording, several conversations in the form of commentaries from Facebook have been added, in order to obtain a more unified account of swearing and slang in both spoken and written Romanian. A significant number of slang words are discussed and several explanations accounting for their use are also provided. As far as gender-related differences are concerned, I show that boys use more slang and swear words than girls. Another interesting finding lies in the girls’ use of male genitalia when they swear, just like in English. I also argue that there are much more dirty words to refer to girls than to boys.

CHAPTER ONE LANGUAGE AND GENDER: PAST AND PRESENT

Introduction This chapter represents the starting point in my analysis of gender-related variation in the speech of English and Romanian adolescents. First I will discuss some slippery concepts, i.e. sex and gender, and another new concept derived from these two, i.e. sexuality. It is important to draw a distinction between these concepts as they are vital to this research. Next, I will present some features associated with “women’s language” and “men’s language” throughout the centuries. In the literature gender-related differences in speech are of two types: gender-exclusive and genderpreferential differences. The former are found in traditional societies whilst the latter in modern ones. In language and gender research there are four frameworks, some of them dating as far back as the early seventies. The first one, the deficit framework, states that women are weak and unassertive as language users, this framework being associated with the work of Lakoff (1975). The second one is the dominance framework according to which the language uses of men and women are viewed as enactments of male privilege or superiority, developed by Spender (1980). The difference framework is associated with the work of Gumperz (1982) and focuses on cross-cultural miscommunication. The last framework, the dynamic one, as the name suggests, looks at the dynamic aspects of conversation. The last part of this chapter is dedicated to gender-related stereotypes in the British and Romanian society.

1.1. From sex differences to gender variation in language use The study of language and gender has bloomed over the past several decades and sociolinguists have come to realize that male-female differences in language use are “by no means clear-cut and we find

8

Chapter One

similarities in speech across gender groups, as well as much diversity within groups” (Wolfram and Schilling-Estes 2008: 234). Because of the complexity of gender-based patterns of language, researchers have realized that we cannot see gender as a “given”, as a synonym for biological sex. The earliest studies which focused on men, women and language were carried out by Western Europeans and their interest was anthropological rather than linguistic. Such studies analysed phonological and lexical differences in exotic languages and noted the existence of different pronouns, affixes, nouns, used exclusively by men or by women, or to refer exclusively to a man or a woman. However, such sex-related differences are absent from European languages. In Romance languages the pronoun system is similar, and they mark sex in the third person singular and plural. On the other hand, English (a Germanic language) marks sex only in the third person singular.

1.1.1. Sex, gender, language Talbot (2010: 15) notes that one view of the relationship between language and gender – which might be called the weak one – is that language mirrors society, and the social divisions on gender grounds are mirrored in patterns of language use. A good example is that women in work settings are usually subordinate in status to men, and this fact is reflected in their greater use of politeness strategies. Another example is the existence of two traditional honorific titles for women (Miss and Mrs), in contrast with Mr for men, thus reflecting the importance society puts on women’s marital status. The stronger view suggests that language does not reflect gender division, it creates them. For example, things such as the difference in the use of politeness strategies, the asymmetry of the titles Miss and Mrs in relation to Mr and asymmetrical usage of terms of address for women and men do not just simply reflect society, but they also create and sustain inequality. The two extremes are language-as-mirror and language-asreproductive (Talbot 2010: 15). Sex, together with age, social class and ethnicity, is one of the most widely used social categories and also one of the most analysed. Holmes (1997: 195-196) states that research on language and gender has tended to follow the general development of feminist thought, moving from an essentialist paradigm where speakers were divided in terms of their biological sex through a period where the significance of the cultural concept of gender was recognized. Bucholtz (2002: 37) provides three

Language and Gender: Past and Present

9

definitions of sex2: (i) a social variable with two values: male and female; (ii) the biological differentiation of individuals into a dichotomy between female and male (in contrast to gender); (iii) the negotiable and contestable social classification of individuals into three categories of female and male based on cultural understandings of the body, especially with respect to sexuality, with attendant normative local ideologies about social, physical, cognitive, and affective practices, attributes, and capabilities. Wodak (1997:13) highlights the fact that gender is “not […] a pool of attributes “possessed” by a person, but […] something a person “does””. She further notes that “what it means to be a woman or to be a woman or a man also changes from one generation to the next and […] varies between different racialized, ethnic, and religious groups, as well as for members of different social classes”. In such a view, gender must be learned anew in each generation. On the other hand, Cameron (1998: 280-281) puts it slightly differently: Men and women…are members of cultures in which a large amount of discourse about gender is constantly circulating. They do not only learn, and then mechanically reproduce, ways of speaking “appropriate” to their own sex; they learn a much broader set of gendered meanings that attach in rather complex ways to different ways of speaking, and they produce their own behavior in the light of these meanings. Performing masculinity or femininity “appropriately” cannot mean giving exactly the same performance regardless of the circumstances. It may involve different strategies in mixed and single-sexed company, in private and public settings, in the various social positions (parent, lover, professional, friend) that someone might regularly occupy in the course of everyday life.

We cannot avoid gender because it is a part of the way in which societies are ordered around us. Gender3 is a part of our identity. 2

Butler (1990, 1993) argues that the definition of sex, like gender, is a social construct, that is, it is assigned social meanings by social beings, and the body thus semiotically indexes a host of ideologies about social and other practices and abilities. 3 In the 1970s writers and researchers did not distinguish between biological and cultural influences on the speaker. Trudgill (2000, originally published in 1974) included a chapter entitled “Language and Sex”, in which he discussed genderbased differences in language use. However, things changed towards the end of the twentieth century when more and more scholars became dissatisfied with the linking of biological sex and social behavior, and the term “gender” replaced the term “sex”. Researchers became aware that men did not prefer certain linguistic patterns or forms because they were male, from a biological point of view, but

10

Chapter One

Sex is used to refer to the physiological distinction between females and males, while gender refers to the social and cultural elaboration of the sex differences, as Cheshire (2002: 423) eloquently puts it. This process begins at birth that is why gender is the more appropriate term to use for the category than sex. In the literature both terms have been treated as binary categories: Gender differences need not map directly onto physiological sex differences, but in practice our social lives are organized around the physiological dichotomy to such an extent that a cultural connection has been forged between sex and virtually every other aspect of human experience.

West and Zimmerman (1987), on the other hand, say that gender is not something we are born with or something we have, but something we do. Is this really true? There is no clear-cut distinction between sex and gender. We do not know where sex leaves off and gender begins. Although sociolinguists have tried to separate these two terms, they are intertwined. Another distinction between sex and gender is given by Eckert and McConnel-Ginet (2003: 10) who note that “sex is a biological categorization based primarily on reproductive potential, whereas gender is the social elaboration of biological sex”. In another paper, Eckert (1989a: 246) defines sex as being a biological category that serves as a fundamental basis for the differentiation of roles, norms and expectations in all societies. These norms, roles and expectations constitute gender, which is the social construction of sex. To sum up the discussion about sex and gender, I quote Fausto-Sterling (2000: 3) who concludes that: Labeling someone a man or a woman is a social decision. We may use scientific knowledge to help us make the decision, but only our beliefs about gender – not science – can define our sex. Furthermore, our beliefs about gender affect what kinds of knowledge scientists produce about sex in the first place.

Gender is the social construct of sex and a key component of identity. Gender is something we perform every day without even knowing it. Sex is binary, gender is not. A person can be either male or female. Gender, on the other hand is socially constructed. We acquire characteristics which “because of their alignment with the norms of the culture they lived in” (Coates 2007: 66). To put it differently, people are born male or female, but their way of speaking is determined by the social and cultural influences which surround them.

Language and Gender: Past and Present

11

are considered masculine or feminine. It is possible to talk about a feminine man or a masculine woman, a woman being more feminine than another or a man more masculine than another4. Cameron (1998: 258) supports the idea of gender performance when she claims “I suspect that in conversations with their superior men use what has been regarded as women’s conversational style. The underlying issue here is likely to be hierarchy, not simply gender.” If we take her idea for granted, then this means that men are able to perform the female gender. Schilling (2011: 218) notes that gender has to do with matters like social and economic roles and relations (power relations included), conceptualizations of femininity and masculinity and with sexual orientation or sexual identity. Many textbooks have started to include chapters entitled “language and sexuality”5, as it has become a field of study in its own right. However, there has been much debate over what a study of “sexuality” should include. Bucholtz (2002: 36) argues that gendered linguistic variables have sexual associations for men though of a different kind, while Milroy (1992) notes that the language ideology that links men’s use of prestige forms to what she calls effeminacy, a term which carries ideological information not only about gender but also about sexuality. Bucholtz (2002: 37) defines sexuality as: (i) sexual orientation, based on the gender of one’s erotic attachments; (ii) one’s orientation to sexuality: sexual(ized) practices and ideologies that shape daily life, including gender. She also notes that there are two ways in which the subfield of language and gender can assist variationist sociolinguistics: it can offer a new perspective and it can help retheorize the conflicting definitions of sex and gender discussed above; and it can also demonstrate the importance of the term sexuality to variationist research. Regarding gender differences in language use, a distinction has been drawn between gender-exclusive and gender-preferential differences. In what follows I discuss them, focusing more on gender preferential differences, as they are essential to this research.

4

From a grammatical point of view, the word “masculine” has degrees of comparison: more masculine, most masculine. The word “male” does not. Forms such as *maler or *malest are incorrect. 5 There is also a book entitled “Language and Sexuality” by Deborah Cameron and Don Kulick published in 2003, as well as a “Journal of Language and Sexuality”, proving that this has become an important area of research.

12

Chapter One

1.2. Men’s language versus women’s language So far we have seen what the terms sex and gender refer to. I will now focus on the differences between the speech of men and women. According to Hornoiu (2002: 116), in the sixteenth century, English writers wrote about differences between men and women in terms of pronunciation and favoured the masculine form. In 1568, Sir Thomas Smith talks about the affected speech of women and Richard Mulcaster in his Elementarie identifies a pair of gender-linked diphthongs, implying the superiority of the masculine form, although it is the pronunciation attributed to women that has become standard in modern English: Ai [pronounced /ai/ as in fine], in the mans diphthong, and soundeth full: ei [pronounced /ei/, as in faint], the womans, and the soundeth finish in the same both sense, and use, a woman is deintie, and feinteh soon, the man feinteth not because he is nothing daintie (quoted in Hornoiu 2002: 115).

This is a good example of the androcentric view of linguistic usage that presents women’s speech as deviating from the (male) norms. Elyon in The Governer (1531) highlights that gentlemen, as the educated literate group in society, differed in their use of language from women, the former’s English being “sillable, as folisshe cleane, polite, perfectly and articulately pronounced, omittinge no letter as women often times do” (Elyon, quoted in Hornoiu 2002: 116). In 1665 the French writer Rochefort described the language of the Carib Indians, who lived in the Lesser Antilles in the West Indies. He notes: The men have a great many expressions peculiar to them, which the women understand but never pronounce for themselves. On the other hand, the women have words and phrases which the men never use, or they would be laughed to scorn. Thus it happens that in their conversations it often seems as if the women had another language than the men (quoted in Graddol and Swann 1989: 41).

Rochefort provides the following explanation for these differences: When the Caribs came to occupy the islands, these were occupied by an Arawak tribe which they exterminated completely, with the exception of the women, whom they married in order to populate the country. Now, these women kept their own language and taught it to their daughters…But though the boys understand the speech of their mothers and sisters, they nevertheless follow their fathers and brothers and conform to their speech from the age of five or six (quoted in Jespersen 1922: 237).

Language and Gender: Past and Present

13

Graddol and Swann (1989: 41-42) say that we shall never know if an invasion and subsequent slaughter of half of the population is the correct or true explanation for the linguistic differences discovered by Rochefort and other Europeans who mixed with the Carib community, but the idea that women and men might actually use different languages provoked quite a stir, and thus the Carib Indians have become a classic case in accounts of gender differences in language use. Despite this interest, it does not seem as if the Carib male and female speech varieties were actually distinct enough to count as two separate languages. Jespersen (1922) re-examined Rochefort’s data and found that distinct male and female forms accounted for only about one tenth of the vocabulary items he had recorded. Graddol and Swann (1989) further note that while the Caribs have often been seen as one of the most extreme examples of women and men using different language varieties, it is likely that some form of gender difference will be found in any language. Those differences that have been recorded, occur at all linguistic levels: for example, they include use of different words, grammatical differences and pronunciation differences. In some cases these differences are categorical – men use one form while women another. In other instances they are a matter of degree – women use some features more than men, or the other way round. According to Coates (2004: 10) commentary on gender differences in vocabulary is quite widespread in eighteenth-century writings, as demonstrated below. The following excerpt written by Richard Cambridge for The World of 12 December 1754 provides some insight into how women’s language was perceived in those times: I must beg leave…to doubt the property of joining to the fixed and permanent standard of language a vocabulary of words which perish and are forgot within the compass of a year. That we are obliged to the ladies for most of these ornaments to our language, I readily acknowledge (quoted in Coates 2004: 10).

What Richard Cambridge is actually implying is that women’s vocabulary is ephemeral and what they say is not important. Lord Chesterfield, writing in The World of 5 December 1754, makes an observation regarding women’s excessive use of certain adverbial forms: No content with enriching our language with words absolutely my fair countrywomen have gone still farther, and improved it by the application and extension of old ones to various and very different significations. They

14

Chapter One take a word and change it, like a guinea, into shillings for pocket money, to be employed in the several occasional purposes of the day. For instance, the adjective vast and it’s [sic] adverb vastly, mean anything and are the fashionable words of the most fashionable people. A fine woman…is vastly obliged, or vastly offended, vastly glad or vastly sorry. Large objects are vastly great, small ones are vastly little; and I had lately the pleasure to hear a fine woman produce, by a happy metonymy, a very small gold snuff-box that was produced in company to be vastly pretty, because it was vastly little (quoted in Coates 2004: 11).

Language commentators have little trouble in identifying what they think to be women’s words, though their lists are usually impressionistic and have little validity. An anonymous contributor to The World (6 May 1756) complains of women’s excessive use of certain adverbial forms: Such is the pomp of utterance of our present women of fashion; which, though it may tend to spoil many a pretty mouth, can never recommend an indifferent one. And hence it is that there is so great a scarcity of originals, and that the ear is such a daily sufferer from an identity of phrase, whether it be vastly, horridly, abominably, immensely, or excessively, which, with three or four more calculated for the same Swiss-like service, make up the whole scale or gamut of modern female conversation (quoted in Coates 2004: 11).

This characteristic of women’s excessive use of adverbial forms is also found in 19th century literature. Jane Austen mocks it in her novel Northanger Abbey (1813), in the speech of Isabella Thorpe: ‘My attachments are always excessively strong.’ ‘I must confess there is something amazingly insipid about her.’ (Jane Austen, Northanger Abbey, Ch. 6)

The use of adverbial forms of this type was very fashionable in those times, and was evidently associated with women’s speech. Furfey (1944: 222), in an early review of women’s and men’s language, argues that the very existence of sex-differentiated forms implies: some consciousness of men and women as different categories of human beings. Furthermore, at least at some period in the history of language, this distinction must have been regarded as being of a certain consequence; for it would seem to be a general truth that the great categories of grammar are not based on distinctions regarded by the speakers as trivial.

Language and Gender: Past and Present

15

Furfey (1944: 223) further notes that “language sometimes serves as a tool of sex dominance”. Beyond this very general level, few satisfactory explanations were offered for sex differentiated forms in language. The view that women have their own vocabulary has been held over more than three centuries. The following list provides a sample of words that have been ascribed to women (Hornoiu 2002: 117): 18th century – ah!, oh!, such, so, somehow, fine, pronominal one, ruck, flirtation, vast, vastly, frightful. 19th century – implicit, splendid, pretty, horrible, unpleasant, thousands, ‘any number greater than two’. 20th century – person, ‘woman’, nice, perfectly, lovely, darling, sweet, horrid, mean, dear, just-too-sweet, poor thing, minx, cat, just, so, too, adorable, precious, cunning, cute, stunning, itsy bitsy, terribly, awfully frightfully, sweetie, honey, doll, all rightie, beige, mauve.

1.2.1. Gender-exclusive differences Meyerhoff (2006: 202) points out that the so-called exclusive features are those used only by (or to) speakers of a particular sex. She further notes that in MƗori (the Polynesian language spoken in New Zealand), the words for siblings provide information about both the referent and the speaker. For example, the word teina tells us that the speaker is referring to a younger sibling that is the same sex as the speaker is (younger brother for a male speaker, younger sister for a female speaker). If a man wants to refer to his sister, he would use a completely different word, tuahine, and this could refer to a younger or older sister. Ochs (quoted in Meyerhoff 2006: 203) has described words like those above as a direct index of gender. Direct index means that a word has a semantic feature [+female] or [+male] as part of its basic meaning. Personal pronouns like he or she directly index gender. Meyerhoff (2006: 204-205) writes that there is one region where it seems that in a community women and men do use different languages, and this is the Vaupés6, an area between Colombia, Peru and Brazil. The Vaupés is an 6

Sorensen (1967: 671) first introduced the anthropological and linguistic communities to the fascinating sociolinguistic situation encountered in the Vaupés river basin of Brazil and Columbia, which he described as “a large, culturally homogenous area where multilingualism – and polylingualism in the individual – is the cultural norm”. Stenzel (2005: 3) notes that multilingualism as it is encountered in the Vaupés system is the result of several complementary factors. A person’s social identity is established by patrilineal descent and has language

Chapter One

16

area of great linguistic diversity, and according to the tradition one must marry outside the father’s home language group. 1.2.1.1. Phonological differences Phonological differences between the speech of men and women have been noted in a variety of languages. According to Coates (2004: 29), the Chukchee language, spoken in Eastern Siberia, varies phonologically depending on the gender of the speaker. Women use /Ȓ/ where men use /ȶ/ or /r/. For example, the word ‘people’ is pronounced by women [ȒamkǹȒǹn] while men pronounce it [ramkǹȶ ȶǹn]. In his analysis of Chukchee, Borgoras (1922: 665) notes that women generally substitute /Ȓ/ for /tȒ/ and /r/, particularly after weak vowels. They also substitute /Ȓ/ for /rk/ and /tȒ/. The sounds /tȒ/ and /r/ are quite frequent so that the speech of women, with its ever-recurring /Ȓ/ sounds quite peculiar, and is not easily understood by an inexperienced ear. Women can pronounce /tȒ/ and /r/ and when quoting the words of a man – for example in tales – they use these sounds. In ordinary conversation, however, the pronunciation of men is considered as unbecoming a woman. Men’s pronunciation

Women’s pronunciation

ra’mkitȒhin

Ȓa’mkiȒȒin

Gloss ‘people’

tȒǎmña’ta Pa’rkala

Ȓǎmña’ta

‘by a buck’

Pa’ȒȒala

‘by a Parkal’

TȒaivu’urgin

Ȓaivu’uȒȒin

(a name)

Table 1.1. Phonological differences between the speech of men and women in Chukchee (Source: Borgoras 1922) Borgoras further adds that men, particularly in the Kolyma district, drop intervocalic consonants, especially /n/ and /t/. In this case the two group affiliation as its primary marker. Sorenson (1967: 677) explains “A woman invariably uses the language of the longhouse – her husband’s language – when talking directly with her children. But she is usually not the only woman from her tribe in a longhouse. In a longhouse of any size there are usually several women from other tribes; and during the course of a day, these several groups of women usually find occasion to converse with each other in their own original languages.”

Language and Gender: Past and Present

17

adjoining vowels are assimilated. Women say [nitvaqenat] while men pronounce it [nitvaqaat]. It would seem that this process of elimination of intervocalic consonants has been very important in the development of the present form of Chukchee. Wardhaugh (2009: 318) notes that in Bengali, an Indo-European language spoken in India, men often substitute /l/ for initial /n/; women, children, and the uneducated do not do this. He further adds that in Yukaghir, a northeast Asian language, both women and children have /ts/ and /dz/ where men have / tj / and / dj /. Old people of both genders have a corresponding /tȒj/ and /jj/. This proves that the difference is not only gender-related, but also age-graded, meaning that it is specific to a certain age. These differences are set out in the table below: MALE CHILD /ts/ /dz/

ADULT /tj/ j

/d /

/ȶ /

FEMALE CHILD ADULT /ts/ /ts/

/ȶj/

/ȳj/

/dz/

/ȳj/

OLD j

/dz/

OLD

Table 1.2. Phonological differences between the speech of men and women in Yukaghir (Source: Wardhaugh 2009) Another example of phonological differences7 is highlighted by Trudgill (2000: 68) in Darkhat Mongolian, a language spoken in Asia. The back rounded vowels /u/ and /o/ in men’s speech correspond to the mid vowels /Ș/ and /ø/ in women’s speech, whereas male /Ș/ and /ø/ correspond to female /y/ and /ø/ - front rounded vowels. Although female speakers do not use /Ș/ and /ø/ where male speakers use them, there is no taboo prohibition to prevent them from using these sounds in other cases. Talbot (2010:5-6) notes that in Brazil there is a tribe called Karajá, whose language has more differences between male and female speech than any other language. In Karajá, the gender of the speaker is marked phonologically. There are systematic sound differences between male and female forms of words, even occurring in loan words from Portuguese. Some examples are provided in Table 1.3 below:

7

There are also some extremely important phonological and morphological differences between the speech of men and women in American Indian languages, such as Koasati, Gros Ventre, Yana. For more details see Campbell (1997), Haas (1944), Mithun (2006), Taylor (1982), among others.

Chapter One

18

Male speech heto out bisileta nobiotxu

Female speech hetoku kotu bisikreta nobikutxu

Portuguese

bicicleta domingo

Gloss ‘house’ ‘turtle’ ‘bicycle’ ‘Sunday’

Table 1.3. Phonological differences in male and female speech in Karajá (Source: Fortune and Fortune 1987, quoted in Talbot 2010) 1.2.1.2. Morphological differences Fasold (1990: 89-90) states that there are languages where the sex of both the speaker and the hearer is important. A woman might use a different form when she is talking to another woman compared with when she is talking to a man, while a man might use a third form, with the exact meaning as the first two, irrespective of to whom he is addressing. Such a language is Knjrux8, a small-group Dravidian language spoken in India. In Knjrux, there are several morphological forms used by women only when addressing another woman; they are not used by men or by women to address men. Some representative forms are given below in table 1.4. TWO-WAY CONTRASTING FORMS BY GENDER IN KNjRUX Man speaking, any Woman speaking addressee; or woman Woman addressee Gloss speaking, man addressee bardan bar’en ‘I come’ bardam bar’em ‘We (my associates and I, but not you) come’ barckan barc’an ‘I came’ barckam barc’am ‘We (my associates and I, but not you) came’ xaddar xadday ‘children’ Table 1.4. Morphological differences between the speech of men and women in Knjrux (Source: Ekka 1972) 8

For a more detailed analysis of the differences in the speech of men and women in Knjrux, see Ekka (1972).

Language and Gender: Past and Present

19

These forms are the first-person singular and first-person plural exclusive verb paradigms, and the noun “children” in the plural. Fasold further notes that “verb morphology in the second-person singular is even more sensitive to sex”. There is one form used by either men or by women when they are talking to men. When women are addressed there are two separate forms depending on the sex of the speaker. A man would use a different form to woman from what a woman would use to another woman, as illustrated below in table 1.5: Man or woman speaker, man addressee barday

Woman speaker, woman addressee

Man speaker, woman addressee

Gloss

bardin

bardi

barckay

barckin

barcki

‘you come’ ‘you came’

Table 1.5. Morphological differences between the speech of men and women in Knjrux (Source: Ekka 1972 and Meyerhoff 2006) Taylor (1951, quoted in Fasold 1990) illustrates another example, from Island Carib from the Caribbean nation, Dominica. In this language, there is a tendency for men to use the names of qualities, states, and actions as if they carried feminine gender while women treat them like masculine gender nouns. The expression “the other day” is ligira buga if a woman says it, but tugura buga if uttered by a man. Interestingly, Taylor reports that “perhaps a minority of men” regularly use feminine forms for nonconcrete nouns but that “all women resort to this trick” when they are quoting conversations between men. Another language in which there are morphological differences between the speech of men and women is Yanyuwa, an aboriginal language spoken in Australia. Here men and women really speak two different dialects. In his analysis of Yanyuwa, Bradley (2006: 14) says that the most common statement given by the Yanyuwa people in relation to their language is that “Men speak one way, women speak another, that’s just the way it is!” Other people believe that they speak two different languages to show respect for the opposite sex. One individual claims that: I don’t really know, but I was thinking that men and women have to respect each other, so we talk different ways and so we show respect for each other, just like ceremony; you know men have their ceremony and

Chapter One

20

their language well same way women have their own ceremony and their own language (Bradley 2006: 14).

Bradley further notes that the younger generations no longer speak the language so it is almost impossible to find out the way in which Yanyuwa language was acquired by children. However, it seems that in very early childhood children spoke a form of neutral Yanyuwa, meaning that the dialectal markers were deleted from words, so that ‘at or with the fire’ became Ø buyuka-la rather than the correct ji-buyuka-la for women and ki-buyuka-la for men. In adolescence around the age of 12 boys are initiated through a series of rituals which include circumcision, after which they are considered men and from that moment on they are supposed to speak the men’s dialect (Bradley 2006: 15). When a young male uses Yanyuwa he often speaks the women’s dialect, for which he is severely criticised. The following excerpt is part of a conversation between mother and son: (1) Son: Mother:

Son:

Mother:

Mum, did you buy ni-warnnyi [meat]? Hey! Are you a man or a woman? Man got to talk nawarnnyi not ni-warnnyi that’s women’s talk, you got to talk properly, you not little kid now. Hey look you complain because young people don’t talk language and when we do you got to laugh at us, man may as well not even bother. Well, you just got to learn to talk proper way just like we did. (Source: Bradley 2006: 16)

It is difficult for boys to start using the men’s dialect because when they were born they acquired the women’s dialect and after their initiation they have to forget that dialect and start speaking a completely different one. In Yanyuwa, differences go beyond sounds and words and include pronouns, grammatical affixes (as illustrated above) and other parts of speech. Bradley (2006: 17) provides further evidence from songs cycles, where there are also female dialectal markers on common nouns and a number of verb stems from the female dialect. Consider the following examples:

Language and Gender: Past and Present

(2) Song verse Manankurra kiya-alarri Manankurra Place name

21

‘At Manankurra He (a Shark Dreaming) stood’ kiya-alarri he: stand

The prefix kiya- in the second line of the above verse is a women’s dialect prefix, while in the men’s dialect it is ka-. (3) Song verse Warriyangalayani ni-mambul ni-ngurru Warriyangalayani Hammerhead shark

‘The Hammerhead Shark’ makes spray with its nose’ ni-mambul ni-ngurru its: spray its: nose

The prefix ni- in the second line represents the female masculine form. In the male dialect it would be na-. Holmes (2008: 158) claims that in traditional and conservative styles of Japanese, forms of nouns considered appropriate for women are frequently prefixed by o-, a marker of polite and formal style. 1.2.1.3. Syntactic differences Meyerhoff (2006: 205) highlights that in AnejomѺ , a language spoken in the Republic of Vanuatu, Oceania, speakers refer to a same-sex sibling with a possessive structure known as “direct possession”, for example, etwa-k ‘same.sex.sibling-my’, and an opposite-sex sibling with a subordinate construction, for example, nataheñ erak ‘sister-my’, nataüañ erak ‘brother-my’. Direct possession constructions are generally used with things like body parts (‘my hand’), or things over which we cannot control (‘my spirit’). Subordinate constructions are used with things that can be removed (‘my blood’, ‘its lid’). However, there is an asymmetry in how speakers refer to a spouse. A man uses the direct possession construction to refer to his wife, ega-k ‘wife-my’, but a woman uses a third construction, which is called “active possession”, to refer to her husband, nataüñ uñak ‘husband-my’. 1.2.1.4. Lexical differences Holmes (2008: 158) writes that in some languages there are also differences between the vocabulary items used by women and men, though these are never very extensive. Traditional Japanese is a case in point:

Chapter One

22

Men’s form oyaji hara umai kuu

Women’s form otǀsan onaka oishii taberu

Gloss ‘father’ ‘stomach’ ‘delicious’ ‘eat’

Table 1.6. Lexical differences between the speech of men and women in traditional Japanese (Source: Holmes 2008) Holmes further adds that in modern Japanese, these distinctions are more a matter of degrees of formality or politeness than gender; so that the men’s forms are restricted to casual contexts and are considered to be macho or coarse, while the women’s forms are used by everyone in public contexts. In most languages, the pronoun system marks gender distinction in the third person singular (he/she). According to Coates (2004) and Holmes (2008), Talbot (2010), in Japanese there are a number of words for the personal pronoun ‘I’ varying primarily in formality, but women are generally restricted to the more formal variants. So, ore is used only by men in casual contexts and boku, another casual form is used almost entirely by men, while women are traditionally expected to use only the more formal forms, such as atashi and watashi, and the most formal one watakushi. There are also a number of words for the personal pronoun ‘you’. These forms, as well as the forms for the first person pronoun ‘I’ are given in table 1.7 below: Men’s speech

Women’s speech

Plain

watakushi watashi boku

Deprecatory

ore

watakushi atakushi watashi atashi Ø

FIRST PERSON Formal

SECOND PERSON Formal Plain Deprecatory

anata kimi anta omae kisama

anata anata anta Ø

Table 1.7. Japanese personal pronouns (Source: Coates 2004)

Language and Gender: Past and Present

23

It is noticeable that certain forms are exclusive to men, for example boku (first person pronoun) and kimi (second person pronoun). The deprecatory pronouns ore (first person) and omae and kisama (second person) are also used exclusively by men. Women have no deprecatory forms. Another difference is that the pronoun watashi is formal for men, but plain for women. However, Talbot (2010: 5) claims that Japanese high school girls say that they also use the first person pronoun boku, because if they use atashi they cannot compete with boys. Feminists have been reported using the form boke to refer to themselves. Coates (2004: 31) notes that for the Trobriand islanders9 the kinship terms are organised on the basis of two criteria: (i) same/different gender as the speaker, (ii) older/younger than the speaker. For the word sister, the terms will vary if the speaker is male or female and if the speaker is younger or older than the sibling. In the case of the relationship we call sister, the Trobrianders have three different terms (luguta, tuwagu, bwadagu). So, they make no distinction between a man’s sister and a woman’s brother (the term used is luguta), nor between a man’s brother and a woman’s sister if the age is the same in both cases (tuwagu or bwadagu). So far I have shown that gender exclusive differences are typical of tribal societies. I will now discuss gender preferential differences and see what this term, gender-preferential, really means.

1.2.2. Gender-preferential differences A preferential feature is one that is distributed across speakers or groups, but it is used more frequently by some speakers than by others. According to Meyerhoff (2006: 289) gender preferential features only indirectly index gender. They are generally associated with several other social meanings, for example, casualness and vernacularity with masculinity. Due to the fact that these other factors help to constitute what it means to be ‘male’ the index between vernacular variants and male speakers/masculinity is indirect. Gender preferential differences have been analysed quantitatively (Tagliamonte 2006, Kiesling 2011, Tagliamonte 2011). In this subchapter

9

The Trobriand Islands (officially known as the Kiriwina Islands) are an archipelago off the eastern coast of New Guinea. They are situated in Milne Bay Province in Papua New Guinea. The language of the Trobriand peoples is Kilivila, an Austronesian language, though various different dialects of it are spoken by each different tribe.

24

Chapter One

I give an overview of some of the studies which combined a social variable (gender) with a linguistic variable. 1.2.2.1. The deficit approach to language and gender One of the earliest and most significant studies on language and gender is Lakoff (1975), an attempt at identifying the features of what she called “women’s language”. Lakoff examined linguistic disparities in two facets of language: “the way women are expected to speak” (their use of language) and “the ways in which women are spoken of” (how they are represented in the language). Lakoff’s assertions have been associated with a deficit model of women’s language use. She suggested that the way women speak is inadequate in several respects. Her claims were based on informal observations and her own opinions and intuitions about language use. No empirical research was carried out to uphold her claims. Lakoff identified noticeable differences between the language use and speech style of men and women. The American sociolinguist wrote that, as an experiment, one might present native speakers of standard American English with pairs of sentences, identical from a syntactic point of view and in terms of referential lexical items, the only difference being in the choice of “meaningless” particles, and ask them which one was uttered by a man and which by a woman. Consider the following example: (4) a. Oh dear, I’ve put the peanut butter in the refrigerator again. b. Shit, I’ve put the peanut butter in the refrigerator again. She claims that one can easily predict that (4a) was uttered by a woman, while (4b) was said by a man. This assertion, however, is not based on a study, but it is rather impressionistic. This might have been possible in the late 1970s, but nowadays this claim is no longer valid. In what follows some of the characteristics of “women’s language” will be given, as proposed by Lakoff (1975): (i) colour terms Lakoff suggests that women can distinguish more colours than men, words like ecru, aquamarine, lavender, mauve, etc. are unremarkable in a woman’s active vocabulary, but absent from that of most men. I do agree with this claim and to support this I quote the following conversation between husband and wife taken from a British sitcom, entitled My Family:

Language and Gender: Past and Present

Susan: Ben: Susan: Ben: Susan:

Ben: Susan: Ben:

25

So, what do you think? Oatmeal or barley white? Oh, no, no, no. You’re not dragging me into this. Why did you change your mind? I’ve got to drag you into this. If you don’t help me choose the colour you’ll complain afterwards. I’m complaining now. Why did you change your mind? Come on, come on, come on, come on. We’re down to the last five: oatmeal, barley white, taupe, cream or ecru? Susan, there is no discernable difference between any of them. Maybe not to your eye. Look. They’ve just paid someone to come up with five different names for beige. (My Family, 2nd season, 1st episode, 2001)

I consider this to be a very good example in support of Lakoff’s (1975) claim that women can distinguish between different shades of the same colour and more colours than men. (ii) adjectives Men and women use a different set of adjectives to convey their opinion on matters. Some of these adjectives are neuter as to the sex of the speaker: either men or women may use them. But another set seems to be largely confined to women’s speech. Compare, for example: (5)

Neutral a. great b. terrific c. cool d. neat

Women only adorable charming sweet lovely, divine

(iii) expletives Women tend to avoid speaking in a way that conveys strong emotions and they usually use “weaker” expletives than men, as was illustrated above (e.g. oh dear as opposed to shit). (iv) tag questions Apparently women use more tag questions than men. A tag, in its usage as well as its syntactic shape (in English) is midway between an outright

Chapter One

26

statement and a yes-no question: it is less assertive than the former, but more confident than the latter. (6)

John is home, isn’t he?

Meyerhoff (2006: 225) provides the following explanation: “of course men use tag questions, too, but the frequency with which they use them and the contexts in which they use them are constrained by the different social expectations of men and the attributes and characteristics associated with being a successful or good man”. Dubois and Crouch (1975, quoted in Fasold 1990: 103) found out that “only men used tag questions in the question and comment segment of a professional meeting”. While discussing this finding, they mention that tag questions have a usage in British English which is not common in American English. For example, in British English a speaker can use a tag question, with falling contour, appended to an assertion that the addressee may not agree with. A speaker of British English might say: (7)

I’ve got to leave now, haven’t I?

The tag question in example (7) is considered to be verbally aggressive. (v) intensifiers Lakoff (1975) suggests that the intensive so seems more characteristic of women’s language than of men’s, though it is found in the latter, particularly in the speech of male academics: (8)

a. I feel so happy! b. That movie made me so sick!

Again, when making this assertion, Lakoff does not provide any empirical evidence to support her claim. (vi) Women use hypercorrect grammar and more superpolite forms than men. (vii) Women use minimal responses (e. g. yeah, aha, mhm). (viii) Women use emphatic stress: (9)

It was a BRILLIANT performance.

Language and Gender: Past and Present

27

Although Lakoff’s work marked a turning point in the analysis of gender related differences in language use, nowadays her work seems out-dated and out-of-tune with modern attitudes. Her book has been subjected to severe criticism. 1.2.2.2. The dominance model The second approach to looking at language and gender differentials is the dominance approach which is connected with Zimmerman and West’s (1975, 1983) analyses, as well as Tannen (1994). Their approach differed from that of Lakoff (1975) in that it was based on an empirical study of conversation. They claimed that women are seen as an oppressed group, women are subordinated, and men are dominant. Men always interrupted women, and by doing so they denied women’s equal status as conversational partners. Zimmerman and West (1975: 105) suggested that “there are definite and patterned ways in which the power and dominance enjoyed by men in other contexts are exercised in their conversational interaction with women”. Interruptions are viewed as “a way of doing power in face-to-face interaction” (West and Zimmerman 1983: 111). James and Clarke (1993) review research on gender and interruption and find that there is no clear pattern of males interrupting females. Tannen (1994: 35) argues that “interruption” or “overlap”, a term preferred by Tannen, is “a paradigm case of the ambiguity of power and solidarity”10. Hornoiu (2008: 54) notes that male dominance manifests in language in certain ways: terminology is male-oriented, women are usually excluded when the generic use of the personal pronoun he or the noun man are used. When they get married, women adopt their husband’s family name as this is the social convention. Another argument is that women have more labels to describe themselves, but these terms often gain a negative connotation. I have labeled them “brave new words” (Oancea 2011a, 2013). English, as we already know, makes certain gender-based distinctions. For example: actor – actress, bachelor – spinster, master – mistress, lord – lady, etc. There are, however, nouns which are not marked for gender, for example: writer, doctor, editor. Romaine (2001: 157) points out that men have the right to be referred to as ‘writers’ or ‘doctors’. Women who 10

Tannen analyses a two and a half hour Thanksgiving dinner conversation (Tannen 1984) and her analysis reveals that some speakers consider talking along with others to be a sign of enthusiasm, of solidarity, of creating connections, whilst others agree that only one voice should be heard at a time. For them overlapping is seen as interruption.

28

Chapter One

occupy these professions are frequently marked with special titles such as ‘lady/woman doctor’ or ‘female/woman writer’. In the British National Corpus (BNC) lady doctor appears 125 times, whilst woman doctor appears only 20 times. In the Corpus of Contemporary American English (COCA)11 lady doctor appears only 25 times, whereas woman doctor 43 times. Fowler (1927, quoted in Romaine 2001) wanted to revive certain –ess forms which were no longer used, for example editress (3 occurrences in BNC and only 1 in COCA) and inspectress. However, the suffix –ess has been attached to other nouns, so new words arise: doctress (4 occurrences in COCA) and manageress (only 3 occurrences). So, the female term is the marked one. The only two cases in English where the male terms are the marked ones are that of bridegroom, bride being the basic term, and widower, widow being the basic term to which the suffix –er is added to form the male form. A noun such as nurse has come to be used differently. A distinction is drawn between male nurse (20 instances in BNC, and 78 in COCA) and female nurse (1 occurrence in BNC and 16 in COCA). The same goes for secretary. The Penguin Dictionary of American English Usage and Style (2002) contains a rather peculiar entry about the distinction between bachelor and spinster. A movie reviewer said: “William Hurt plays Graham Holt, a male spinster who shocks neighbours when he decides to adopt a 10-yearold” (italics mine, C.O.). Male spinster is a contradictory form just as female bachelor. A spinster is a female by definition. Romaine (2001: 159) notes that the term spinster “originally meant a woman engaged in spinning. Because these women were often unmarried, this connotation eventually ousted the original meaning and became the primary sense of the word in the 17th century.” Returning to the dictionary entry, the actor described in the movie review plays the role of a bachelor. Many reliable dictionaries define bachelor as an ‘unmarried man’. This word usually implies that the man (i) is of the usual age for marrying, or beyond, and (ii) has never been married. Two dictionaries recognize the term bachelorette (176 instances in COCA) and the synonymous bachelor girl. In the US the term spinster has become archaic and is no longer used. However, the term is still preserved and used in British English. There are also many words referring to women which have lost their original meaning and have undergone pejoration. As mentioned above, English draws a distinction between lord and lady. Schultz (1975, quoted 11

I have looked up all the words written in italics in the Corpus of Contemporary American English (COCA) and then compared the findings with Romaine’s (2001) findings from the BNC.

Language and Gender: Past and Present

29

in Romaine 2001) claims that the noun lord still preserves its original meaning, while lady is no longer used to describe a woman of high rank. In the 17th century lady became a synonym for a prostitute. The same happened with the word courtesan, which originally described a female member of the court. Sir is still used to show respect and as a title, while a Madam is one who runs a brothel. Also, the word master has not lost its original meaning, whereas mistress has started to have sexual connotation, denoting a woman with whom a man cheats on his wife. In an engaging article examining marked gender terms like pronouns and nouns in four corpora of British English, Baker (2010) shows that while there has been a decrease in the use of marked male terms like Mr or he, there has been no increase in explicitly marked female terms. He also looked at gender neutral terms (police officer, spokesperson) and revealed that they appear to be increasing. Researching gender differences using the dominance theoretical framework, Fishman (1978) suggests that a clear power structure underlies marital relationships and everyday interaction represents the epitome of male dominance over women. The idea that men and women have different life experiences which lead them to form different language communities was stressed by linguists like Lakoff (1975), Fishman (1978), Zimmerman and West (1975). This view leads to a new model in language and gender difference, namely the different culture approach. 1.2.2.3. The difference approach to language and gender The third approach, the difference approach, which accounts for patterns of language use, has its origins in Gumperz (1982) and expresses the idea that men and women belong to different subcultures. The difference model depends on a “two cultures” account of male and female socialisation. This approach is often offered as an alternative to the dominance model as an explanation for patterns of language use. Hornoiu (2008: 56) quotes some studies on gender differences which stress the idea that women are more willing and likely to: (i) inquire about upsetting situations; (ii) use comforting messages that explicitly acknowledge, elaborate and legitimize the feelings of others; (iii) use supportive strategies that confront emotions; (iv) provide emotional support; (v) seek support; (vi) feel confident about their ability to provide emotional support; (vii) value support-giving skills.

30

Chapter One

This idea that men and women are different and that they should be regarded as members of different cultures is also used in popular press. Best-sellers like Men are from Mars, Women are from Venus by John Gray (1992), You Just Don’t Understand: Women and Men in Conversation by Deborah Tannen (1990) supported the idea of different cultures. They became extremely popular because of the advice on how to talk to your husband/wife, how to behave in a marriage or how men should talk to women or the other way round. Many scholars from different fields claimed that men and women live in different gender cultures: It is clear that men and women do come from different cultures, and the crucial difference between those cultures is that men come from a culture that emphasizes status and power, whereas women come from a culture that emphasizes relative closeness rather than relative power (Noller 1993:148).

In a very interesting paper, Maltz and Borker (1982) note that girls learn to do three things with words: (i) create and maintain relationships of closeness and equality; (ii) criticize others in acceptable ways; (iii) interpret accurately the speech of other girls. Boys, on the other hand, learn to do the following when they speak: (i) assert a position of dominance; (ii) attract and maintain an audience; (iii) assert themselves when another speaker has the floor. The conclusion that can be drawn is that the boys’ peer groups are organized and competitive with dominance and assertiveness playing a vital role. Women, on the other hand, are interested in equality and emotional closeness. 1.2.2.4. The dynamic approach The most recent approach is usually called the dynamic approach because it emphasizes the dynamic aspects of interaction. Coates (2004: 6) says that researchers who adopt this model take a social constructionist perspective. Gender identity is seen as a social construct rather than as a “given” social category, hence the distinction between sex (which is

Language and Gender: Past and Present

31

biological) and gender (which is social). Hornoiu (2008: 60)12 argues that gender varies across time and space, indicating that gender is conceptualized in different ways depending on the culture. Holmes (1998) provides a list of what she calls “sociolinguistic universal tendencies”. This list comprises five claims: (i) Women develop different patterns of language use; (ii) Women tend to focus on the affective functions of an interaction more often than men; (iii) Women tend to use linguistic devices that stress solidarity more often than men do; (iv) Women tend to interact in ways which will maintain and increase solidarity, while (especially in formal contexts) men tend to interact in ways which will maintain and increase their power and status; (v) Women are stylistically more flexible than men. It has been shown that gender differences between the speech of men and women do exist and there are many theories and models of analysis. In what follows, I discuss gender stereotypes in English and Romanian.

1.3. Gender-related stereotypes in English There are many stereotypes13 about the language used by men and women14: women are more polite than men; girls are verbally more precocious than boys, etc. In the literature there are four generalizations regarding such stereotypes:

12 She also provides a list of communication-related differences between males and females that have been reported in the literature: (i) male childhood and adolescent communication is characterized by interruptions, powerful assertions and more straightforward judgments than female childhood and adolescent communication; (ii) girls are interested in maintaining a relationship based on intimacy with their friends or partners; (iii) boys are more interested in shared activities than girls. 13 Brannon (2005: 160) argues that “a gender stereotype consists of beliefs about the psychological traits and characteristics of, as well as the activities appropriate to, men or women. Gender roles are defined by behaviours, but gender stereotypes are beliefs and attitudes about masculinity and femininity.” 14 Lakoff’s (1975) book Language and Woman’s Place is famous for creating a stereotype of its own.

Chapter One

32

(i)

in mixed-gender settings, women speak considerably less than men (Crawford 1995); (ii) women are generally more focused on the personal/interactional aspects of conversation men tend to be more interested in conveying information (Holmes 1995); (iii) women tend to be more tentative than men in their use of language, both in conversation (Coates 1993) and in some forms of writing (Rubin and Greene (1992)), tending to use more hedges, possibility modals; (iv) women’s discourse is lower in the use of persuasive strategies, tending to emphasize narrative strategies more (Fleischman 1998). Some gender stereotypes have been confirmed by empirical research. For example, the stereotype that women use more grammatically correct sentences is confirmed by quantitative sociolinguistic studies carried out in different English-speaking communities. Such studies reveal an important trait of female speech. Irrespective of their social class or age, women have the tendency to use more standard forms than men both in formal and informal styles of speech (Cheshire 1982, Newbrook 1982, Trudgill 1983, Romaine 1984, Eisikovits 1988, Cheshire 1997). According to Hornoiu (2008: 39), it is important to conduct variation studies on urban communities because they provide extremely valuable insights into the external and internal forces which govern language change and draw attention to the central role of gender differences in these processes. The sociolinguistic studies made in the 1960s and 1970s claimed to establish an intriguing difference between the language used by women and by men: namely that across social classes, “women consistently tend to use more of the features associated with the prestige ‘standard’ variety of a language than men do” (Talbot 2010:19). So, two new concepts arise: “prestige” and “stigma”. Coates (2004: 47) points out that prestige is said to represent those linguistic forms normally used by the social group with the highest social status. Prestige is then associated with RP. The use of the standard variety in the major institutions of society – education, broadcasting, media, the law – perpetuates this prestige. On the other hand, stigma is attached to non-standard forms, in other words to the vernacular. A very good example to illustrate this situation is Trudgill’s 1974 study of the (ing) variable in Norwich. The analysis of the data he collected in Norwich, including speaker’s gender as well as social class and contextual style, shows that scores for male and female speakers are quite different. It was revealed that women, regardless of style and social class, tend to use the prestige variant [ƾ] significantly more than men.

Language and Gender: Past and Present

33

The studies which I present focus on morpho-syntactic variation and explore the everyday interaction of adolescents. A case in point is Cheshire’s (2006) study which analyzes the grammatical usage of young people in the city of Reading, England. Her research revealed significant gender differences. Boys use more often the non-standard or the vernacular form than girls, and the boys’ consistent use of the vernacular forms is not so much a gender marker than a marker of their adherence to the vernacular culture. Cheshire’s (2006) results were corroborated by Eisikovits’s (2006) study, which focused on the speech of adolescents from inner-Sydney working class suburbs. Eisikovits (2006) found out that while boys use a higher proportion of non-standard forms, women’s speech is closer to the norms of Standard English. This is just a brief overview of quantitative studies which have confirmed or disconfirmed certain gender-based stereotypes in Englishspeaking communities. I will discuss them more thoroughly in Chapter 2.

1.4. Gender-related stereotypes in Romanian The first linguist to have ever studied gender-related differences in Romania is, to my knowledge, Hornoiu (2002, 2008) who analyzes women’s and men’s speech in English and Romanian, focusing more on Romanian. This section reports on the findings of a survey whose aim was to establish the existence of gender-related stereotypes in the speech of Romanian men and women.

1.4.1 Data collection In order to establish whether there are gender differences in Romania, a pilot project was devised and conducted among Romanian MA students studying at Ovidius University of ConstanĠa and the University of Bucharest. The sample consisted of 30 students, balanced for gender (15 males and 15 females), aged 22–26. I devised a set of 16 sentences, incorporating some of the linguistic variables proposed by Lakoff (1975) and Holmes (1995). I was particularly interested in morpho-syntactic and lexical variation in the speech of men and women. The people interviewed were BA and MA students, as they were all available and willing to cooperate.

Chapter One

34

1.4.2 Methodology The informants were asked to choose between M (for those sentences they think are uttered by a man), F (for those sentences they think are uttered by a woman) and M/F (for the sentences they think are uttered by both a man and a woman). All the results were counted and reduced to percentages since this makes comparison much easier. The list of sentences devised for this survey is the following: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

Aútept să mă suni mîine. ‘I will wait for you to call me tomorrow.’ Mi-am cumpărat o pereche de pantofi negri úi o cămaúă albă. ‘I bought a pair of black shoes and a white shirt.’ Mi s-a stricat maúina úi am rămas în cîmp. ‘My car broke and I was stranded in the middle of the road.’ Păpuúă, poĠi să îmi aduci o scrumieră? ‘Doll, can you bring me an ashtray?’ Este o persoană adorabilă, nu crezi? ‘She is adorable, isn’t she?’ Este o persoană drăguĠă. ‘She/He is a nice person.’ Aú vrea să plec în vacanĠă vreo două săptămîni. ‘I wish I could go on holiday for two weeks.’ Ce dracu’ se întîmplă cu tine? ‘What the hell is the matter with you?’ Sacoul tău verde-măr se asortează cu pantalonii ăia de velură. ‘Your apple-green jacket goes with those trousers.’ Mă simt incredibil de bine. ‘I feel incredibly well.’ Mă simt bine. ‘I am okay.’ Îmi place modul tău de a acĠiona. ‘I like the way you act.’ Mă doare-n fund de ce zice. ‘I don’t give a damn what he says.’ Nu Ġi se pare genială ideea sa? ‘Don’t you think his idea is brilliant?’ Aoleu, am uitat să cumpăr ciocolată. ‘Oh dear, I forgot to buy chocolate.’ Este aúa de frumos! ‘It is so beautiful!’

Language and Gender: Past and Present

35

1.4.3 Findings When asked how they decided which sex to attribute, the students said that they thought of the language they use, as well as of the language used by their parents, relatives, friends, ordinary people in the street, etc. The results are given in Table 1.8. When the informants were asked to account for their answers, some stereotypes were identified. M 1. Aútept să mă suni mîine. 2. Mi-am cumpărat o pereche de pantofi negri úi o cămaúă albă. 3. Mi s-a stricat maúina úi am rămas în cîmp. 4. Păpuúă, poĠi să îmi aduci o scrumieră? 5. Este o persoană adorabilă, nu crezi? 6. Este o persoană drăguĠă. 7. Aú vrea să plec în vacanĠă vreo două săptămîni. 8. Ce dracu’ se întâmplă cu tine? 9. Sacoul tău verde-măr se asortează cu pantalonii ăia de velură. 10. Mă simt incredibil de bine. 11. Mă simt bine. 12. Îmi place modul tău de a acĠiona. 13. Mă doare-n fund de ce zice. 14. Nu Ġi se pare genială ideea sa? 15. Aoleu, am uitat să cumpăr ciocolată. 16. Este aúa de frumos!

0% 20%

F 60% 20%

M/F 40% 60%

6,6%

26,6%

66,6%

86,6% 6,6% 20% 0%

6,6% 86,6% 26,6% 13,3%

6,6% 6,6% 53,3% 86,6%

20% 6,6%

26,6% 80%

53,3% 13,3%

0% 40% 20% 6,6% 0% 6,6% 0%

66,6% 0% 20% 20% 73,3% 73,3% 93,3%

33,3% 60% 60% 73,3% 26,6% 20% 6,6%

Table 1.8. Gender-related stereotypes in Romanian The figures reveal that Romanian women are polite, sensitive, and they pay more attention to details. Some of Lakoff’s (1975) claims concerning women’s language are confirmed by the Romanian data. It appears that women have a preference for what Lakoff (1975) calls “empty” adjectives (adorabilă ‘adorable’, genială ‘brilliant’), colour terms (verde-măr ‘apple green’), tag questions, emphatic stress, intensifiers (aúa ‘so’). However, Romanian women also tend to use taboo words, as sentence (13) points out. This is very interesting, as this is not a feature associated with “women’s language”. Girls use taboo words and swear because they want

36

Chapter One

to compete with boys and be cool. By using foul language, girls gain power and acquire a certain “prestige” in high-school. In the next section I discuss this phenomenon more thoroughly and elucidate when girls start to swear and how they swear. Unlike women, men chose to be either polite or vulgar. Non-standard linguistic forms, swearing, taboo language and concise sentences are considered to be features of male speech, features also pointed out in Hornoiu (2008). In addition, as already stated, women’s language is no longer seen as weak, unassertive as it used to be. The above pilot study revealed some important differences between the speech of men and women. I replicated this survey and added four more sentences to the list and made some minor changes regarding the other sentences. The survey was replicated and extended in order to cover other age groups: 150 students were interviewed, distributed across 3 age groups: 50 students were secondary school pupils (age-group 12–14); 50 students were high-school students (age group 15–18) and 50 were 1st year students at Ovidius University of ConstanĠa (age group 19–21). All the students interviewed live and study in Constanta. I randomly chose two schools and two highschools where the interviews were conducted. In each age-group the number of males and females interviewed was equal (i. e. 25 males and 25 females). In one of the high-schools where I conducted the research, the English teacher whose students I interviewed, found some of the sentences in the questionnaire inappropriate. The sentences in question were (4) and (15). The teacher considered the words căcat ‘shit’ and fund ‘bottom’ to be taboo. She even suggested replacing the words. I tried to convince her that the questionnaire was for a scientific paper and that I was interested in analysing the language used by men and women. She finally accepted to interview her students.

Language and Gender: Past and Present Sentence

1.Aútept să mă suni mîine. 2. Mi-am cumpărat o pereche de pantofi negri úi o cămaúă albă. 3. Mi s-a stricat maúina úi am rămas în cîmp. 4. Ce căcat faci acolo? 5. Păpuúă, poĠi să îmi aduci o scrumieră? 6. Este o persoană adorabilă, nu crezi? 7. Este o persoană drăguĠă. 8. Aú vrea să plec în vacanĠă vreo două săptămîni. 9. Ce dracu’ se întâmplă cu tine? 10. Sacoul tău verde-măr se asortează cu pantalonii ăia de velură

Age-group 12 - 14 M F M/F

Age-group 15 - 18 M F M/F

37 Age-group 19 - 21 M F M/ F 0% 40 60 % % 24% 38 38 % %

20%

26%

54%

8%

48%

55%

54%

20%

26%

52%

20%

28%

66%

12%

22%

60%

20%

20%

56%

22 %

22 %

58%

20%

22%

46%

22%

32%

38%

78%

16%

6%

96%

2%

2%

96%

24 % 2%

38 % 2 %

10%

70%

20%

12%

64%

24%

10%

74 %

16 %

16%

52%

32%

22%

38%

40%

20%

38 %

42 %

18%

26%

56%

6%

24%

70%

14%

14 %

72 %

48%

22%

30%

52%

16%

32%

46%

16 %

38 %

8%

74%

18%

10%

70%

20%

14%

70 %

16 %

Chapter One

38 11. Mă simt incredibil de bine. 12. Mă simt bine. 13. Cizmele alea sunt belea. 14. Îmi place modul tău de a acĠiona. 15. Mă doaren fund de ce zice. 16. Este o idee nemaipomenit ă, nu-i aúa? 17. Tre’ să mă duc în oraú. 18. Aoleu, am uitat să cumpăr ciocolată. 19. Este aúa de frumos! 20. Este genială ideea, nu Ġi se pare?

22%

28%

50%

8%

46%

46%

16%

34 %

50 %

22%

28%

50%

20%

16%

64%

32%

0%

80%

20%

18%

78%

4%

10%

12 % 70 %

56 % 20 %

54%

16%

30%

18%

32%

50%

24%

28 %

48 %

44%

16%

40%

48%

8%

44%

32%

12 %

56 %

18%

44%

38%

16%

48%

36%

10%

48 %

42 %

20%

34%

46%

40%

12%

48%

46%

2%

24%

42%

34%

18%

52%

30%

8%

72 %

52 % 20 %

8%

72%

20%

2%

72%

26%

0%

12%

60%

28%

2%

64%

34%

0%

82 % 76 %

18 % 24 %

Table 1.9. Gender-related stereotypes in Romanian across 3 agegroups A similar situation occurred with secondary-school pupils. When I gave them the questionnaire and they started reading the sentences almost all of them began to giggle. Their English teacher told them to stop laughing because they use far worse words in their conversation (i. e. swearing and taboo). I noticed that Romanian women also tend to use swearing and taboo words and also some non-standard forms, therefore I wanted to see whether women really use these forms, why they use them, and to what extent. Table 9 offers a clear picture of gender variation in Romanian. It should be pointed out that expanding the number of informants to 150 has not affected the results previously obtained. With regard to women’s language, Lakoff’s (1975) claims seem to apply to the Romanian society as well. Romanian women seem to be specialists when it comes to precise colour terms/shades, empty adjectives, tag questions, intensifiers,

Language and Gender: Past and Present

39

euphemisms. Sentence (6) is definitely uttered by a female, according to my informants. I will present some of the detailed results and comment upon them. M F 12–14 years old 10 % 70 % 15–18 years old 12 % 64 % 19–21 years old 10 % 74 %

M/F 20 % 24 % 16 %

Table 1.10. Gender-related stereotypes in Romanian It is a fact that women prefer empty adjectives and tag questions, to men. Women seem to prefer tag questions which make them appear insecure, unassertive, and weak. This is also revealed by sentences (16) and (20). In Romanian, women tend to use the standard forms and they are concerned with their public image. Adjectives like adorabil/ă ‘adorable’ were rarely used by girls. Instead they preferred adjectives like miúto ‘cool’, naúpa ‘sucks’ and others, which are borrowed from Romani. In this case, girls over-reported using a sentence like (6), when they do not use it. Boys do not use such adjectives (as the one in (6)) at all. If they used such adjectives they would be laughed at or considered gay or they would use it to make fun of girls. The results obtained for sentence number (10) are illustrated below: M F M/F 12–14 years old 8 % 74 % 18 % 15–18 years old 10 % 70 % 20 % 19–21 years old 14 % 70 % 16 % Table 1.11. Gender-related stereotypes in Romanian Precise colour terms/shades appear to be women’s speciality. Over 70 % of my informants believe that a sentence like (10) is uttered by a woman. The fact that women use these precise colour terms shows that they pay more attention to details. Some of the males interviewed confessed that if they use such colour terms they are likely to be considered gay. Another explanation for this usage would be that girls read fashion magazines or watch TV programs related to fashion and this is how they learn and use all the colours and shades. Men, on the other hand, except for designers and homosexuals, do not usually read fashion magazines or watch fashion shows. Of course, this is just one explanation, but there could be other factors involved. Women also tend to use intensifiers, as the results with

40

Chapter One

respect to sentence (19) indicate. Over 60 % of the persons interviewed think that this sentence is used by a woman. Women tend to use intensifiers when they want to emphasize something or draw attention. They use foul language and the way they use language apparently has no impact on their public image. Some of the secondary-school girls whom I interviewed told me that boys “are allowed” to use swearing or taboo words in conversation. Sentence number (4) is one of those newly-added sentences. I deliberately inserted a taboo word and the purpose was to see whether only men use this type of words, or women also tend to use them. The results for sentence number (4) presented below: M F 12–14 years old 58 % 20 % 15–18 years old 46 % 22 % 19–21 years old 38 % 24 %

M/F 22 % 32 % 38 %

Table 1.12. Gender-related stereotypes in Romanian The percentages obtained reveal that by taking age into account there appears to be a decrease in the usage of taboo words by men. Almost 60 % of the secondary school pupils interviewed believe that sentence (4) was uttered by a man. And the use of taboo words seems to decrease as people grow older. At least this is what the percentages indicate. What is interesting is that while there is a decrease in the use of taboo words by men, there is an increase in the use of such terms by women. One explanation could be that, with age, women become more confident, and they gain more power. Some of the high-school girls whom I interviewed claimed that they use taboo words and swearing because they are competing with boys and they want to be cool and in-tune with modern attitudes. “This is how we teenagers talk”, said two of them. The percentages obtained for sentence number (9) also supports this view. An interesting view regarding taboo language is that of Allan and Burridge (2006) who note that it contains “dirty words”. They provide a list of terms which fit into the category “dirty words” (Allan and Burridge 2006: 41): a. bodily organs concealed by bikinis and swimming trunks – because they are organs of sexual desire, stimulation, and gratification and/or used for micturition and defecation; b. activities involving these SMD (sex, micturition, defecation) organs; c. bodily effluvia issuing from these SMD organs.

Language and Gender: Past and Present

41

Many of the high school girls whom I interviewed used certain body parts which they do not possess, when swearing. An example to illustrate this is the example provided in (7), taken from my field notes: (7)

Ce mă enervează căcatu’ ală. Dă-l în pula mea. ‘I hate that fuck’ ‘Fuck him’

The sentence in (7) was uttered by many girls who participated in this survey. A sentence like the one above is extremely common in the speech of high school girls. The sentence we are interested in, is the second one Dă-l în pula mea ‘Fuck him’, particularly the use of the word pula, which is a vulgar word for penis. Romanian teenage girls use it frequently when swearing, probably to appear more powerful and masculine. Sentence number (5) reveals that men tend to use all kinds of words to refer to a woman. In this case the term păpuúă ‘doll’ is used to refer to a woman. The percentages obtained for sentence (5) are given below: M F 12–14 years old 78 % 16 % 15–18 years old 96 % 2 % 19–21 years old 96 % 2 %

M/F 6% 2% 2%

Table 1.13. Gender-related stereotypes in Romanian It is worth mentioning that men use other words to describe or to refer to a woman, not only in Romanian, but in English and other languages as well, words such as honey, sugar, sweetheart, darling, pumpkin, sweetie, etc. However, they also use more offensive words such as fanny, bitch, tart, cunt, whore, slut, wench, bimbo, hoe, floozy, sleaze, skunk, etc. These words are predominantly used in colloquial speech. More than 80 % of my informants believe that this is typical of men and that women seldom use such words. Romanian teenage girls, in same sex conversations, use words like fă or păsărică ‘pussy’ to refer to other girls. Although the word fă was mainly used in the countryside to denote a form of address, nowadays it is used by high school girls in cities.

1.5. Concluding remarks The aim of this chapter was to discuss the theories and models used in the analysis of language and gender differences. In the first part of this chapter gender differences in different languages spoken all over the world have

42

Chapter One

been presented. The purpose has been to provide an overview of the literature in the field and to establish that men and women do speak differently. A distinction has been drawn between sex and gender, the former being biological and the latter social. I have also discussed genderexclusive and gender-preferential differences in language use, and I pointed out that the first are typical of traditional societies whereas the second can be found in modern societies. In last part of this chapter I have focused on different approaches to language and gender: the deficit, the dominance, the different culture and the dynamic approaches. I also set out to identify gender-related stereotypes in the speech of Romanian adolescents and I have also discussed gender stereotypes in the English culture.

CHAPTER TWO LANGUAGE VARIATION AND CHANGE: THE THEORETICAL FRAMEWORK

Introduction The first chapter provided a thorough analysis of language and gender differences and a departure point on which to begin the study of genderrelated linguistic variation. The different case studies presented enable us a better understanding of this phenomenon as well as the different approaches and models to languages and gender. This chapter proposes a framework that draws on variationist sociolinguistics, an approach used for a quantitative analysis (Labov 1966, Trudgill 1974, Cheshire 1982, among others) and (im)politeness theory as developed by Brown and Levinson (1987) and Culpeper (1996), with a special focus on swearing and taboo language (Allan and Burridge 2006), for the qualitative analysis. I start by discussing variationist sociolinguistics, why it is relevant for this study as well as the central tenets of this framework. The linguistic variable is discussed at length with a special emphasis on phonological and morphological variables. I chose to blend these two frameworks in order to obtain an exhaustive insight into gender-related variation. I start with the framework for the quantitative analysis which is applied in Chapters 3 and 4, and then continue with the framework for the qualitative analysis which represents the topic of Chapter 5.

2.1. Variationist sociolinguistics In recent years a plethora of research has been done in the variationist sociolinguistics framework. Before embarking on the discussion proper, we must first define two important concepts: sociolinguistics15and variation. 15

A distinction should also be drawn between sociolinguistics, and the sociology of language to enable us a better understanding of these concepts. Sociolinguistics

44

Chapter Two

Sociolinguistics has its roots in dialectology, historical linguistics, and language contact with influences from sociology and psychology (Koerner 1991:65) and studies the connection and interaction between language and society. Many believe that sociolinguistics studies language variation and its aim is to discover what variation tells us about language and speakers’ knowledge of language, as many of us have different views and opinions about language. These opinions are known as “folklinguistic beliefs”.16 In this study we will adopt the definition given by Chambers (2003: ix) according to whom sociolinguistics is “the correlation of dependent linguistic variables with independent social variables.” ChiĠoran and Ionescu-Ruxăndoiu (1975: 15) define sociolinguistics as the study of linguistic variation. Variation is a key concept in sociolinguistics and implicitly in Language Variation and Change (LVC) research. Labov’s (1969: 728) claim that variation is an inherent part of language represents the foundation of the LVC approach, which we adopt in this study. But what is variation? If there is variation then it means that something varies (i.e. we can say the same thing in at least two ways). For example, if we have a verb like watching, the alternation between watching with [ƾ] and watchin' with [n] in spoken English indicates that there are two ways of pronouncing the {ing} morpheme without changing the meaning of the word (denoting the ability to watch). Variation analysis is an important part of sociolinguistics, as it represents the branch of linguistics that starts has been defined in many ways but one of the most comprehensive definitions is given by Chambers (2002: 3) who defines it as “the study of the social uses of language, and the most productive studies in the four decades of sociolinguistic research have emanated from determining the social evaluation of linguistic variants. These are the areas most susceptible to scientific methods such as hypothesis-formation, logical inference, and statistical testing.” On the other hand, the sociology of language is “the study of society in relation to language” Hudson (1980: 5), while Labov (1970: 30) considers that “it deals with large-scale social factors, and their mutual interaction with languages and dialects. There are many open questions, and many practical problems associated with the decay and assimilation of minority languages, the development of stable bilingualism, the standardization of languages and the planning of language development in newly emerging nations. The linguistic input for such studies is primarily that a given person or group uses language X in a social context or domain Y.”One of the main representatives of the sociology of language is Joshua Fishman (1972). The label sociology of language and not sociolinguistics signifies a greater concern with sociology rather than linguistics. 16 For a more detailed discussion see Hudson (1980), Wardhaugh (2005), Holmes (2008).

Language Variation and Change: The Theoretical Framework

45

from the rules of grammar and investigates the contact between these rules and society (Tagliamonte 2006). Variationist sociolinguistics has developed considerably in the last decades and one of the reasons was to come up with a model of language which could explain and account for the paradoxes of language variation and change. It is difficult to give an all-encompassing definition of variationist sociolinguistics, however Tagliamonte succeeds and describes it as follows: The branch of linguistics which studies the foremost characteristics of language in balance with each other – linguistic structure and social structure; grammatical meaning and social meaning – those properties of language which require reference to both external (social) and internal (systematic) factors in their explanation (Tagliamonte 2006: 5).

The central tenets of variationist sociolinguistics are considered to be, firstly, the notion of orderly heterogeneity17, secondly, the fact that language is a living organism and always changes, and thirdly, that language provides important non-linguistic information18. A variationist study is in fact a quantitative analysis and this type of research is deductive, i.e. we develop hypotheses based on certain theories and then we try to prove or disprove them, in the course of an empirical investigation (Rasinger 2008: 11). Other important features of variationist sociolinguistics are the data collection techniques: the sociolinguistic interview (Labov 1969, 1972) and the methods of analysis: multivariate analysis – Varbrul, Goldvarb and later versions (Cedergren and Sankoff 1974, Tagliamonte 2006). A fundamental concept in the variationist sociolinguistics approach is the “principle of accountability” (Labov 1966: 49). Let’s take, for example, the relative pronoun which. The principle of accountability clearly states that besides examining which, all the other potential variants within the relative pronoun system must be considered. In the analysis we have to include all the relevant forms in the subsystem of grammar that we have set out for investigation (Tagliamonte 2011: 10). An accountable study means an exhaustive analysis for every case in which “a variable element occurs out of the total number of environments where the variable could have occurred” (Tagliamonte 2006: 13), or as Labov (1972c: 72) puts it “report values for every case where the variable element occurs in the relevant environments”. 17

The concept was used by Weinreich et al. (1968: 100) and describes the fact that language varies, and speakers can say the same thing in more than one way. 18 For a detailed discussion see Kiesling (2011) and Tagliamonte (2006).

46

Chapter Two

In what follows I will briefly discuss language variation and change and see why language varies and changes.

2.1.1. Language variation and change Why does language vary and why does it change? We have to see language as something alive and in a perpetual change. For example, we do not speak the same Romanian that was spoken 300 years ago, and the British do not speak the English used by Chaucer or Shakespeare. A sentence like I don’t need no money is considered ungrammatical in contemporary Standard British English, as it contains double negation. In earlier times it was considered correct. The wh-words (what, when, why, etc) in modern English were known as hw- words in Old English, and the examples do not stop here. It is a fact that “all languages change through time and we do not really know why this is, but it is a characteristic of all human languages.” (Trudgill 2003: 7) The most important thing about variation in language is that it occurs in the vernacular of everyday life. An adolescent says I was like, whatever dude while a 70-year old would say something like You was always workin’ in them days19. Are such utterances considered slang or mistakes or even part of a dialect? I dare say that they are not. We should try to understand why people speak like this and what factors are involved. The vernacular was first defined by Labov (1972: 208) as “the style in which the minimum attention is given to the monitoring of speech” and later analyses of the vernacular showed that its target of investigation should be “every day speech” (Sankoff 1980: 54), “real language in use” (Milroy 1992: 66) and “spontaneous speech reserved for intimate or casual situations” (Poplack 1993: 252). But how do we gain access to the vernacular? If we want to analyse it we must infiltrate in a speech community, both as an observer and a participant. An excellent example is the study carried out by Cheshire (1982) in Reading, England. She joined a gang and after being accepted by the members of the gang (both girls and boys) she started recording them in the same setting (the adventure playgrounds in the town). As she was interested in linguistic variation in nonstandard English, she chose speakers from the lower end of the social scale, who use nonstandard linguistic forms more often. The language used by such speakers is often known as the vernacular.

19

Examples taken from Tagliamonte (2011: 2)

Language Variation and Change: The Theoretical Framework

47

Figure 2.1 Social and regional variation (Source: Trudgill, 1975: 21)

The term ‘vernacular’ can be used with another meaning, referring to the effect of speech style on linguistic variation (Cheshire 1982: 6). If we follow the traditional Labovian framework of analysis, speech style is seen as forming a linear continuum, reflecting the attention paid by speakers to their speech. It is believed that in formal situations people control the way they speak and, as a result they tend to use more socially prestigious forms and correct grammar. In informal situations, on the other hand, they are more relaxed and they pay less attention to the language they use, hence the usage of nonstandard forms and also swearing. Labov (1972a: 112) writes that the most consistent patterns of variation are encountered in the vernacular speech style. However, analysing the vernacular speech is sometimes problematic as it is difficult to record it.

2.1.2. The sociolinguistic interview One of the tools of variationist sociolinguistics is precisely the method of data collection. The most widely used method is the sociolinguistic interview. It differs from the participant observation in that it involves elicitation and not the observation of language in everyday use. The former method is used to carry out a quantitative analysis while the latter is used for a qualitative analysis. There have been many discussions in the literature (Schilling-Estes 2002, Schilling 2013, among others) which have questioned the value of the sociolinguistic interview in the Labovian sense. The main aim of the

48

Chapter Two

sociolinguistic interview20 is to elicit natural speech and record it. The speaker is encouraged to speak as much as he can and give long answers to the questions he is asked. Another key issue in a successful sociolinguistic interview lies in its structure. Questions should be grouped into “modules focusing on particular topics and the modules can be re-arranged as the interview progresses to approximate the flow of natural conversation” (Schilling 2013: 93). The main idea is to make the speakers feel comfortable and encourage them to talk for as long as they like on a topic of their interest. Before designing the interview it is advisable to learn a few things about the community the speakers live in. Although questions focus on things believed to be universal, it would be inadequate to ask an adolescent to talk about politics or university life, or to ask someone living in Alaska to talk about hot weather and the tropical rainforest. If the interviewees are asked both general and community-specific questions this will make them feel confident and relaxed, as many enjoy telling stories or describe (humorous) past events from their lives, and when they do so, they produce truly vernacular speech. A very interesting example of a flourishing sociolinguistic interview is Wolfram et al.’s (1999) study of the rural island community of Ocracoke, North Carolina. The questions that were used in the questionnaire included topics of general interest which would give access to vernacular speech: games and free time, danger of death, etc. In what follows I give an example of the danger-of-death-module: Ocracoke danger-of-death module iv. Weather/danger of death (1) What was the worst storm you’ve ever been in? (a) What was that like? (b) Did it do much damage? (c) Are the roads OK? (d) Do the roads get washed out a lot? (e) What do you think should be done? 20

“While critics of this method have claimed that the sociolinguistic interview is not a natural speech event (Wolfson 1976), it is not clear what constitutes a “natural” speech event, and what is natural about it. If we assume speakers have a repertoire of speech events that they regularly engage in, and that other events are so strange as to elicit unnatural speech, we are working on a static view of linguistic practice. Conversations, perhaps particularly with strangers, are inventions” (Eckert 2000, quoted in Schilling 2013: 116).

Language Variation and Change: The Theoretical Framework

49

(2) Do you think you’re getting more storms now than you used to? How come? (3) Have you ever been in a hurricane? (a) What was that like? (4) How about shipwrecks? (a) Do you remember any shipwreck stories? (5) Have you ever been in a situation where you thought, “This is it, I’m going to die”? (a) What was that like? Tell me about it. (Source: Schilling 2013: 95)

The traditional Labovian sociolinguistic interview contained several subsections, with the purpose of obtaining a range of different speech styles. The conversational part would yield casual speech, including the vernacular, depending on the questions the interviewees were asked. The interview also contained a reading passage that the participants were asked to read, which would produce careful speech, a word list and a list of minimal pairs (Labov 1972b: 70-109). Labov (1966, 1984, 2001) states that no matter how good the conversational part of the sociolinguistic interview is designed, an important part of the conversation will still be considered careful. In order to draw a line between careful speech and casual speech, which is of interest, Labov (2001: 94) devised the decision tree which encompasses eight contexts usually found in a sociolinguistic interview.

Figure 2.2. Labov’s decision tree for stylistic variation in the sociolinguistic interview (Source: Labov 2001: 94)

50

Chapter Two

In a nutshell, the main goal of the sociolinguistic interview, in the traditional sense, is to elicit and “be able to identify each interviewee’s most casual, least selfconscious style – the vernacular” (Schilling 2013: 98). This is found in Labov’s vernacular principle which states that “the style which is most regular in its structure and in its relation to the evolution of language is the vernacular, in which the minimum attention is paid to speech” (Labov 1972c: 112).

2.1.3. The Observer’s Paradox Studies of language in its social context aim to discover how people talk in everyday life. It is well-known that, when they are recorded, people talk differently, trying to use standard language. Unfortunately we cannot analyse everyday language without making tape recordings first. Many people refuse being recorded and if they agree, they strive to sound intelligent, especially if they are on television. As mentioned above, the data collection technique used in variationist sociolinguistics is the sociolinguistic interview. The problem is that the interview is a formal situation, and we must assume that however relaxed and friendly speakers may appear to be, they still use a more formal speech style because they know that they are being recorded. So, how can we avoid this and get access to the vernacular of everyday life? There have been many attempts to develop a methodology that will solve the problem of the Observer’s Paradox, by working as “a friend of a friend” within social networks21, or by asking speakers to record themselves so that the linguist need not be present (Cheshire 1982: 7). There are, however, some drawbacks to these methods. It is difficult to use them in a variationist study. First of all, the investigator must be present during the sociolinguistic interview so that he can elicit the variable(s) he/she is interested in. Secondly, large amounts of data are necessary in order to guarantee an adequate number of occurrences of the variable(s) and this means that the investigator must go back to the interviewees several times to make the recordings. The questions which form the sociolinguistic interview must be carefully chosen. When speakers are emotionally involved (angry, excited, happy) they are more concerned with what they say rather than how they say it (Milroy and Gordon 2003: 65). Labov’s (1972a: 93) “danger of death” question is a case in point. This question asks subjects about situations in which they feared for their lives. Despite his success with this technique, attempts by others with this question have not been so 21

For a more detailed discussion see Milroy (1980), among others.

Language Variation and Change: The Theoretical Framework

51

successful. I used it when I collected data from London teenagers and many of my informants answered “No” or “I can’t think of a situation right now”. Trudgill faced the same problem in his Norwich study and he suggested that perhaps the people from Norwich have led less eventful lives than people from New York City. This does not stand for London. However, one reason which could explain the lack of success of this type of question is that I used it with adolescents who do not have a lot of experience, although they are quite adventurous. One of the questions which was really successful was “What kind of memories do you have about being taught to ride a bike?” This question triggered all kind of happy and funny memories and my informants told me different anecdotes and stories about the time when they learned to ride a bike. Another method meant to overcome the Observer’s Paradox is longterm participant observation. In this case the fieldworker becomes the member of a group of speakers for a period of time, allowing them to gain the group’s trust and record them. This method was successfully used by Labov et al.’s (1968) study in Harlem. A house was rented for an academic year and the teenagers were taken in a ‘bugged’ minibus for recording sessions at the university. This method proved efficient in obtaining natural spontaneous speech. The same method was used by Cheshire (1982) in her study of the English spoken in Reading, England. In what follows we turn our attention to the linguistic variable and discuss phonological, morphological and syntactic variables, which are relevant to our study.

2.1.4. The linguistic variable Speakers always make choices when they speak. For example, a teenager would say What’s up? to a friend and How are you? to his grandmother, teacher, neighbour, etc. The meaning of the question is the same, but the degree of formality is different. In the literature, one of the most basic definitions of the linguistic variable is two or more ways of saying the same thing. One of the earliest studies which focused on a linguistic variable was carried out by Fisher in 1958 in rural New England. He investigated the (ing) variable22 and found 22

The (ing) variable is considered to be a stable sociolinguistic variable, meaning that it is not undergoing linguistic change and a correlation with age is not expected (Tagliamonte 2011: 187). However, Labov’s extensive study in New York City showed that [ƾ] is sensitive to age and Horvath’s (1985: 97) research in Australia proved that [n] was a feature found in the speech of teenagers. Schleef (2011) states that (ing) is a good variable to take as a starting point and provides

Chapter Two

52

out that a “model boy”23 used the [ƾ] more often while a “typical boy”24 favoured the [n] form. He also discovered that in formal situations, boys used more frequently the [n] form than girls, as shown in Table 2.14:

Boys Girls

-ing > -in’ 5 10

-ing < -in’ 7 2

Table 2.14. Preferences for [ƾ] and [n] endings, by gender (Source: Fisher 1958: 48) Fisher (1958) was the first who set out to investigate a linguistic variable, quantitatively. It is important to decide which type of linguistic variable will be the focus of your investigation, as this can sometimes prove to be a daunting task. Below I give an overview of the most important phonological, morphological and syntactic variables studied in different communities. 2.1.4.1. Phonological variables At this stage it is important to mention that variationist sociolinguistics was built on the study of phonological variation. It is almost impossible to discuss the massive coverage of this phenomenon, and look at all the phonological variables studied. Instead, I will look at some of the most important variables that have been studied, namely postvocalic (r) and (t,d), but I also refer briefly to other variables as well. I will not discuss the (ing) variable at this stage, as Chapter 3 focuses entirely on it. Foulkes and Docherty (2006: 412) discuss different types of phonological and phonetic variables and in each case the variation also has complex social correlates:

five reasons for studying it: (i) it occurs fairly frequently in conversation, (ii) it is salient among native speakers of English, (iii) it is stable (and has remained so for at least fifty years), (iv) it can be analysed auditorily and (v) it has been studied in a number of varieties of English since the 1950s. 23 The model boy was described as being popular, industrious in school and considerate. 24 The typical boy was described as being strong, mischievous and not afraid of doing forbidden things.

Language Variation and Change: The Theoretical Framework

53

(i) in the Glasgow dialect /x/ and /ȓ/ are receding in frequency and they are replaced by /k/ and /w/. This frequency is indexical of age as older speakers use them more frequently than younger speakers. Middle class children use more /x/ than working class children do (Lawson and Stuart-Smith 1999); (ii) variation regarding the phonotactic distribution of phonemes, the English /r/ being a case in point. The distribution of /r/ differs across varieties, marking a division between rhotic and non-rhotic accents (Wells 1982); (iii) lexical distribution of phonemes. In England, both northern and southern accents contrast /a/ and /Ǘ:/ but the difference lies in which vowel is used in a word. The short vowel is preferred by the Northern varieties in words like path, class, bath, while the long vowel in used in the south; (iv) allophonic realization, which is the most common type of phonetic variables. In Newcastle upon Tyne, UK, stops are variably realised as plain oral plosives but can also be largynealized and glottal (Stuart-Smith 1999). Hughes and Trudgill (1987: 32) note that the majority of English accents25 permit /r/ where it occurs before a vowel (e.g. rat, trap). However, they differ when /r/ occurs after a vowel (postvocalic /r/), in words like car, bar, etc. RP, for example, does not have postvocalic /r/, while Scottish and Irish accents, as well as the majority of North American accents do have /r/ in this position. The most important study analysing postvocalic /r/ is Labov (1966) carried out in New York City. Labov chose three department stores in Manhattan: Sacks Fifth Avenue (a high-class store near the centre of the high-fashion district), Macy’s (a store for the middle class with middle prices) and Klein’s (a store which sold cheap items for poor customers). He pretended to be a customer and asked at what floor he could find shoes, knowing that the answer had to be “fourth floor”. He pretended to have hearing problems and asked the customers or salespersons to repeat, thus eliciting more tokens in more careful style. Labov managed to gather over 1,000 tokens of the variable (r) and showed that in New York City /r/ can either be realised as a rhotic approximant or it can have a zero realization, 25

Here, the notion of accent refers to varieties characterized by a specific phonetics and phonology whereas dialect describes varieties additionally distinguished from each other by differences of grammar and vocabulary.

54

Chapter Two

with rates of [ǿ] production correlating with social class. Higher social group use more [ǿ] than lower social groups. The occurrence of postvocalic /r/ may index social class. Interestingly, in England things are the other way round. A high rate of postvocalic /r/ production indicates that the speaker has a low social status (Wells 1982: 35).

Figure 2.3. Social stratification of (r) in New York City (Source: Labov 1966: 151)

Labov’s (1966) study of New York English revealed two important aspects of sociolinguistic stratification: linguistic differentiation and social evaluation. Regarding the linguistic differentiation, the patterning of (r) in Figure 3 above shows the following tendencies: (i) the difference between the social classes is not categorical, meaning that no class is characterised by the complete presence or absence of postvocalic /r/; (ii) all social classes show an increase when moving from informal to more formal styles. This reveals that the variable marks not only status but also style;

Language Variation and Change: The Theoretical Framework

55

(iii) as we follow the progression towards more formal styles, the LMC shows a greater increase in the use of /r/, until in word list and minimal pair styles they overtake the UMC averages. Another well-studied variable in the variationist sociolinguistics framework is the (t,d) variable26. The studies which focused on this variable show that it is a stable sociolinguistic variable and it qualifies as a linguistic marker because it can be correlated with style or formality but not with social class or apparent time (Tagliamonte 2011: 180). Tagliamonte (2005) analysed the (t,d) variable in York English and found an interaction between speaker age and sex. Among the adult York speakers there is gender variation in the sense that males tend to use simplified clusters more than females.

Figure 2.4. Distribution of simplified clusters for variable (t,d) by gender in York, UK (Source: Tagliamonte 2011: 184)

Tagliamonte’s data reveal a correlation between gender and age. As far as middle-aged York speakers are concerned, there is a visible difference in that males tend to use simplified clusters more than females. Another finding which is of interest to us is that there are significant gender differences among adolescents (the 17–19 year olds). The same thing cannot be said about 20 year olds or the older generations (50+) where there are almost no differences. 26

In the literature it is claimed that variable (t,d) is conditioned by the following linguistic factors: (i) the preceding and following phonological contexts, (ii) the morphological structure of the word.

56

Chapter Two

An interesting study analysing the replacement of oral stops with glottal stops in Newcastle upon Tyne was carried out by Milroy et al. (1994). They found that male and female speakers focused on different variants of /t/, the women preferring the supra-local glottal variant, which is spreading rapidly in a number of other British urban dialects, and the men using the glottalised variant. In their work on dialect levelling in the cities of Milton Keynes, Hull and Reading, Kerswill and Williams (1997) found evidence that the replacement of /t/ with a glottal stop is one of the few phonological changes in progress in British English that teenagers know of. The following two comments were made by a teenage girl (8a) and boy (8b) from Milton Keynes: (8) a. My mum takes the Micky if I say bu’er. She’ll say butter. b. My parents don’t like me missing letters out, like if I say wa’er. (Source: Kerswill and Williams 1997: 165) The two comments show that the teenagers are aware of the change, and they also know that the form is considered to be a non-standard one. They are immediately corrected by their parents who also view the spread of [ȣ] a non-standard form. In many varieties of English, there is laxing of /i/ and /u/ before /l/. Laxing means that there is a merger between [ܼ] and [i] and between [‫]ݜ‬ and [u] so that “still” and “steel” become homophones (Meyerhoff 2006: 214). Bailey’s (1993) study in Texas showed that the laxing of [u] to [‫]ݜ‬ was used more frequently by younger women than by younger men. Eckert’s (2000) pioneering work on language variation and change among Detroit teenagers revealed yet another interesting phenomenon. She discovered that the central vowel [‫ ]ݞ‬was backing in some speakers (bus sounding more like boss). This change in progress was confined to the speech of the group of adolescents she called burnouts, and within this group it was more advanced in the speech of girls than of the boys. 2.1.4.2. Morphological variables Kiesling (2011: 14) argues that the variable context of morphological variables is identified by the grammatical function of the morpheme, for example tense and aspect marking on verbs, or plural marking on nouns. One of the examples discussed by Tagliamonte (2011) is verbal (s). In Standard English, the verb is marked in the third person singular (e.g. sings, eats, etc). However, variable verbal (s) means that the -s suffix can be absent in third person singular.

Language Variation and Change: The Theoretical Framework

57

(9) a. She always phones me here and reverseø the charges to me. b. He comes every three times a week he comeø. (Source: Tagliamonte 2011: 208) The –s suffix can be present in third person plural, as the examples in (10) illustrate: (10) a. Well, dreams comes true. Lots of dreams comes true. b.’Cos people comeø along and they comes in with the kiddies. (Source: Tagliamonte 2011: 208) In her study on the English spoken in Reading, England, Cheshire (1982) found out that the suffix -s occurs with other subjects, and she provides the following examples: (11) a. I starts Monday, so shut your face. b. You knows my sister, the one who’s small. c. They calls me all the names under the sun, don’t they? (Source: Cheshire 1982: 31) How can we account for this type of variation? One explanation would be that in the case of the (s) variable there is a standard and a nonstandard form, depending on the linguistic context: in Standard English the -s suffix is attached to the verb form in the third person singular, while in nonstandard English it can be attached to the verb form not only in the third person singular, as the examples above show. Another explanation provided by Cheshire (1982: 31) is that the occurrence of the -s suffix in the 1st, and 2nd persons singular and plural and 3rd person plural can be attributed “to the previous influence of Northern varieties of English...the Northumbrian dialect of Old English had an -s suffix throughout the present tense paradigm, and this pattern was extended in the Middle English period to Midland areas”. The verbal (s) variable was studied in African American Vernacular English (AAVE) from a quantitative perspective and it was discovered that it was highly variable across individuals and conditioned by extralinguistic factors, i.e. it did not occur in formal styles or in middle-class adult speech. This is why verbal (s) was labelled “irregular and unsystematic” (Labov et al. 1968: 167) and a case of free variation. The variable inflection of present tense verbs irrespective of the grammatical person or number of the subject is one of the best features documented in AAVE. Poplack and Tagliamonte (2004) provide the following examples:

58

Chapter Two

(12) First person singular a. I forgets about it. b. I forget the place where it is. (13) Second person singular a. You speaks fine French. b. When you speak with me, fast I don’t...know what you tell me. (14) Third person singular When she come out she goes and she takes her children. When she’s on vacation well, she remain in the home. (15) First person plural We call her Virgie. (Interviewer: Why?) ’Cause that’s the name we calls her. That’s her nickname. (16) Third person plural They speak the same English. The English people talks with grammar. (Source: Poplack and Tagliamonte 2004: 203-204) Apart from AAVE, Poplack and Tagliamonte (2004) discuss verbal -s variation in Devon English, and claims that “Devon is the ideal place to study verbal -s variation, as it represents one of the very few contemporary varieties of English in which nonconcord -s is not simply a remnant, but remains productive (and variable) in all grammatical persons”. This is illustrated below: (17) First person singular I forgets now how long I stayed there. Quite a good while I think I stayed with’em. (18) Second person singular a. You goes up that lane and you goes down another road. b. You pack up for a fortnight, put your kit on your back, you go out, and have a beautiful time. (19) Third person singular Nice maid, her. But her likes more the bloody old boy than her do the maid.

Language Variation and Change: The Theoretical Framework

59

(20) First person plural We get’s a lot of trips, don’t we? Once a month we go. (21) Third person plural a. Yeah they drives’em...They help out b. The cattle all goes to, to the big markets, these days...they go straight to the slaughter house. (Source: Poplack and Tagliamonte 2004: 209) This morphological variable, which occurred with all persons in Old and Middle English, has been preserved in Devon English. As far as Romanian in concerned, there is a general tendency of masculinising the feminine. (22) a. RăspundeĠi voi invitaĠiei lui Mădălina. ‘You answer Mădălina’s invitation’ b. RăspundeĠi voi invitaĠiei Mădălinei. (Source: Zafiu 2010a: 37) The correct form is the one in (22b) but few people use it. The more popular variant is the one in (22a). In the example provided in (22a) the noun Mădălina is in the genitive case, and the correct genitive form is Mădălinei. The preposed article lui is used before masculine names (lui Andrei, lui Mircea, lui Costin, etc) as well as before feminine names of foreign origin (lui Elisabeth, lui Cosette, etc), and names ending in a consonant (lui Carmen) or in the vowels -o, -i (lui Teo, lui Mari). I do not know whether there are gender-related differences regarding this morphological variable, but it is worth looking into it. This is also encountered in common nouns or demonstrative pronouns, as the following examples (Zafiu 2010a: 38) illustrate: (23) a. să-i ia lu’ doamna pachetul de Ġigări. ‘to buy the lady a pack of cigarettes’ b. mâna lu’ fata ‘the girl’s hand’ c. din cauza lu’ băiatu’ ‘because of the boy’ d. împotriva lui ăsta votez ‘I vote against him’

Chapter Two

60

2.1.4.3. Syntactic variables Syntactic variables are difficult to define, as it is not always clear if two variants are equivalent or not at some abstract level. In pro-drop languages, such as Romance languages (Romanian, Spanish, Italian, Portuguese, etc), the use of on overt pronoun is optional: (24) a. Vorbesc româneúte. (I) speak Romanian b. Eu vorbesc româneúte. I speak Romanian The sentences under (24) are both grammatically correct, and express the same thing. Another variable which received considerable attention is the English deontic modality system. Deontic modals express obligation or necessity. We use them in order to exert pressure on ourselves or on someone else to do certain things. The forms included in this analysis (Tagliamonte 2011) are: must, have (got) to and got to: (25) a. Next time I’m in the doctor’s I must ask to see the physio. b. They have to keep up with the Jones’ now. c. You’ve got to have a vice of some kind. (Source: Tagliamonte 2011: 228) Tagliamonte further argues that must is the oldest form, dating back to the Old English period. The deontic have to is considered to have emerged in Middle English, have got to in the nineteenth century and got to and gotta in the twentieth century27. There have also been studies showing that there are gender differences in the use of subordinate clauses, some of them dating back to the beginning of the twentieth century. The problem was that these studies were not based on empirical research. Jespersen (1922: 252) claimed that “a male period is often like a set of Chinese boxes, one within another, while a feminine period is like a set of pearls joined together on a string of ‘ands’ and similar words...In learned terminology we may say that men are fond of hypotaxis and women of parataxis”. Mondorf (2002) looked for gender differences in the use of subordinate clauses in the London-Lund 27

For a more detailed analysis see Tagliamonte (2004), Tagliamonte and Smith (2006).

Language Variation and Change: The Theoretical Framework

61

Corpus and found that women tend to use postposed adverbial clauses more than men do. She also stated that men tend to use finite adverbial clauses to express high commitment to the truth of the propositions expressed, while women use them to the opposite effect (Mondorf 2002: 166). So far we have seen that variationist sociolinguistics is an empirical method which is data-driven. An equally important fact is that the method of variationist sociolinguistics is based on correlating a linguistic variable, with other factors, such as extralinguistic factors (social class, gender, age, ethnicity, etc). This will be the focus of the next subchapter, in which we will discuss gender, as a social variable.

2.1.5. Variationist studies: quantifying gender Many studies of language variation and change, including the studies carried out in the sixties and seventies (the urban stratification studies28) claimed to have established an intriguing difference between the language used by men and women, namely that across social classes “women consistently tend to use more of the features associated with the prestige ‘standard’ variety of a language than men do” (Talbot 2010: 19). In the case of stratification studies, gender differences represent a statistical tendency. We don’t find distinct “male” and “female” forms: women tend to use more prestige forms than men. In Britain, one of the most important studies analysing the speech of adolescents was conducted by Cheshire (1997) in Reading. She looked at nine non-standard features of Reading English: (26) a. the present tense suffix with non 3rd person singular subjects We goes shopping on Saturday. b. has with non 3rd person singular subjects We has a little fire, keeps us warm. c. was with plural subjects (and singular you) You was outside. d. multiple negation I’m not going nowhere. e. negative past tense never, used for standard English didn’t I never done it, it was him. 28

Labov’s (1966) study of New York City and Trudgill’s (1974) study of Norwich.

Chapter Two

62

f. what used for standard English who, whom, which, that Are you the boy what’s just come? g. auxiliary do with 3rd person singular subjects How much do he want for it? h. past tense come I come down here yesterday. i. ain’t, used for negative present tense forms of be and have, with all subjects I ain’t going. I ain’t got any. Some of these features are markers of vernacular loyalty for adolescent boys in Reading but not for adolescent girls. Although these features characterize the speech of both boys and girls, the frequency with which they are used is significant in some cases, as pointed out in Table 15: FEATURE Present tense –s non 3rd p. sg. subjects Has non-3rd person singular Was singular you, plural subjects Multiple negation Past tense negator never Relative pronoun what Auxiliary do 3rd person singular Past tense come Ain’t = auxiliary have Ain’t = auxiliary be Ain’t = copula

FREQUENCY INDICES Boys Girls 53.16 52.04 54.74 51.61 88.15 73.58 88.33 51.85 46.84 40.00 36.36 14.58 57.69 78.95 100.00 75.33 92.00 64.58 74.19 42.11 85.83 61.18

Table 2.15. Gender-based differences in non-standard features of Reading English (Source: Cheshire 1997) Cheshire (1997) notes that the boys’ consistent use of vernacular forms is not really a gender marker but a marker of their adherence to the vernacular culture. Another interesting study focusing on adolescents was carried out by Eisikovits (1988, republished in 2006) in Sydney, Australia. She focused on three variables: (i)

Non-standard past tense forms such as seen and done: He woke up and seen something.

Language Variation and Change: The Theoretical Framework

63

(ii) Multiple negation: They don’t say nothing. (iii) Invariable don’t: Mum don’t have to do nothing. The results showed that girls used the standard forms more frequently than boys, who preferred the vernacular forms. 60 50 40 30

Male

20

Female

10 0 Non-standard past tense

Multiple negation

Invariable don't

Figure 2.5. Gender-related differences in Sydney, Australia (Source: Eisikovits 1988/2006) A score of 0 per cent suggests consistent use of the standard forms while a score of 100 per cent shows consistent use of non-standard forms. Her study corroborates the results obtained by others, namely that there is a preference among women for the standard forms of language. Labov (1966: 214-15) points out that social class and ethnicity have a greater effect on language variation than does gender. His sociolinguistic study of the English spoken in New York City abandoned the idea than any one person could be representative of a complex urban area. Instead, Labov relied on speech samples collected from a random sample of 103 men and women representative of different social backgrounds, ethnicities, and age groups. Labov was interested in finding out whether phonological variables such as postvocalic /r/ in words such as floor, cart, barn, car, is pronounced or not. Although he was not interested in sex-based patterns Labov noted that women’s extreme range of stylistic variation can be explained in terms of their linguistic insecurity, an insecurity rooted in the adoption of standards of correctness of a higher social class group. Romaine (2003: 99) notes that if you count variants of different kinds in

64

Chapter Two

tape-recorded interviews and compare their occurrence across different groups of speakers, you can make some generalizations about the relationship between linguistic variables and society. Labov (1966) and Wolfram (1969) show that the patterning of malefemale language differences differs from a social class to another, and male-female differences in usage levels for prestige forms are greatest in social class groups locate in the middle of the hierarchy than at either end. Labov goes further and claims that male-female differences differ from a speech style to the other, and women sometimes show higher levels for prestigious forms in formal styles, indicating a greater stylistic range overall than men. Labov (1990) formulates three principles regarding the relationship between gender and language: Principle I: In stable sociolinguistic stratification, men use a higher frequency of nonstandard forms than women; Principle Ia: In change from above29, women favour the incoming prestige forms more than men; Principle II: In change from below30, women are most often the innovators. Principle II, however, contradicts Jespersen’s (1922) claim that men actually introduce “new and fresh expressions” and not women but men are “the chief innovators of language” (Jespersen 1922: 247). Dubois and Horvath (1999) note that Principle I and Ia refer to language spread, while Principle II refers to innovation – a change that takes place within a speech community. They claim that “Principle I is like a lull in the tug-of-war game; for some reasons the process has halted. Principle Ia captures the game becoming more active once again” (Dubois and Horvath 1999: 309) whereas Cheshire (2004: 426) argues that Principle Ia and II directly relate to language change, while Principle I represents a more stable state of affairs, possibly a temporary one. Women’s tendency to use a higher proportion of the standard forms than men, illustrated by Labov’s Principle I, has been of very wide general 29 Change from above: forms introduced by the dominant social class (not necessarily the highest one). They usually contain borrowings from higher-prestige speech communities. It appears in careful speech style. 30 Change from below: forms not driven by extra-linguistic (social) factors. They contain systematic changes that appear first in the vernacular, and represent the operation of internal, linguistic factors. They may be introduced by any social class.

Language Variation and Change: The Theoretical Framework

65

interest, so much that it has become a textbook case, illustrating a fundamental tenet of sociolinguistics (Cheshire 2002: 426-7). Schilling-Estes (2002: 124) highlights the fact that women are both more and less conservative than men, as they show higher usage levels for standard, conservative forms as well as for linguistic innovations. This contradiction is the foundation of what is referred to as Labov’s (2001: 67) gender paradox, which states that “women conform more closely than men to sociolinguistic norms that are overtly prescribed, but conform less than men when they are not”. Kiesling (2011: 79) mentions that in patterns of stratification, some sort of social prestige is attached to one variant and some kind of stigma to the other. Regarding accommodation31 patterns, speakers take into consideration not only the prestige or stigma of a variant, they also express solidarity with other speakers in their networks. Among the studies carried out in the United States of America, which focused on the speech of adolescents, we mention Eckert’s (1989, 1998, 2000) research carried out in Detroit at Belten High. The students under investigation belonged to two dominant groups in the school: jocks32 and burnouts33. Eckert was interested in phonological variation, and the variables under investigation were [ԥ] as in fun, cuff, but and [aj] as in line, file, fight. These sounds are in flux in the local variety of (white) American English used by the students: the vowel in words such as but is moving back (but will be pronounced more like bought), while the first element [a] of the diphthong in words like file is being raised and pronounced more like foil. The majority of students use the full range of pronunciations but they vary in the frequency with which they use the more conservative and more innovative pronunciations (Coates 2004: 59-60). The results are presented below:

31

Accommodation describes the process of adapting our linguistic behaviour to conform to the linguistic behaviour of our interlocutor and show solidarity towards them. 32 The term jock describes students who participate enthusiastically in school culture and plan to go to university. 33 The term burnout represents the students who reject the idea of the school as central to their lives, and who are interested in extracurricular activities. They represent a working-class culture, whereas jocks a middle-class culture.

Chapter Two

66

70 60 50 40

girls

30

boys

20 10 0 jocks

burnouts

Figure 2.6. Percentage of backed tokens of [ԥ] (Source: Coates 2004: 60)

80 70 60 50 40 30 20 10 0

girls boys

jocks

burnouts

Figure 2.7. Percentage of extreme raised tokens of [aj] (Source: Coates 2004: 60)

These variables are not only gender markers – gender and social category are intertwined. The burnout girls are the most advanced speakers in terms of new vernacular forms, while the jock girls use the more conservative forms. This means that the girls’ usage is more polarised than the boys’: “the jock and the burnout girls’ values for both [aj] and [ԥ] backing constitute the linguistic extremes for the community” (Eckert and McConnel-Ginet 1999: 195).

Language Variation and Change: The Theoretical Framework

67

Eckert (2012) notes that as far as variation analysis is concerned, one can identify three waves of study: (i) First wave studies: which investigate quantitative correlations between demographic factors (e.g. social class, gender, ethnicity, age) and patterns of language variation (Labov 1966 – the study of the Lower East Side of New York City); (ii) Second wave studies: social categories and the potential social meanings of linguistic variables are no longer imposed from above but from below by looking at the locally important social and linguistic meaning with the help of long-term participant observation (Eckert’s (2000) study of a high school from Detroit); (iii) Third wave studies: these studies focus on the social meaning of linguistic variants rather than on correlating variants and social groups, whether in terms of speakers’ social group memberships or those of their audiences. The concepts and studies discussed so far represent the starting point in the analysis of gender variation in English and Romanian. It is important to stress out that the studies presented so far reveal that there are no genderexclusive differences in the speech of males and females, in other words, none of the phonological, morphological or syntactic variables discussed above are an exclusive gender marker. What is inferred is that there is a general tendency among women irrespective of social class, age, ethnicity to use standard forms.

2.2. Communities of practice So far we have seen that in any variationist study, the variable and the sampling methods play a pivotal role. However, things are not as clear-cut as they might seem. This study focuses on adolescents as aforementioned and in order to have a complete picture of the gender-related variation in their speech, we must also look at the language they use in the community34 to which they belong. The language of any community includes certain specialised uses of language which have different

34

In this study we use the term community as a social unit – larger and looser than a family, smaller and more cohesive than a society – whose activities are based on past shared experience, common interests and language-based ways of thinking together.

68

Chapter Two

functions within that community. Once we have decided to study a community, there are certain things that must be taken into account: (i) background information (e.g. settlement history, general description of the area); (ii) material artifacts (e.g. architecture, signs, instruments of communication like telephones, radios, books, television sets, computers and drums); (iii) social organisation (e.g. community institutions, ethnic and class relations, social stratification, residential and association patterns); (iv) legal information (e.g. language-related laws); (v) artistic data (e.g. literary sources: newspapers, pamphlets, drama and other genres of verbal performance, web logs, etc.); (vi) common knowledge (e.g. formulas such as “Everybody knows...”, “As they say...”); (vii) beliefs about language use (e.g. taboos, language attitudes); (viii) data on the linguistic code (e.g. the first thing on which sociolinguistics will focus on: units of lexicon, phonology and grammar). (Source: Saville-Troike 2003: 92-95) The community of practice should not be confused with speech community35. For years sociolinguists have talked about the speech communities that they studied, but the exact definition of a speech community has been the subject of much debate. For Labov (2007: 347) the speech community has three important dimensions: (i) well-defined limits; (ii) a common structural base; and (iii) a unified set of sociolinguistic norms. Other radically different definitions of the speech community include Patrick’s “socially-based unit of linguistic analysis (2002: 577) and Bucholtz’s “language-based unit of social analysis” (1999: 203). Labov (1972b: 158) also states that “a speech community cannot be perceived as a group of speakers who all use the same forms; it is best defined as a group of people who share the same norms in regard to language. 35

The term speech community was first used by Gumperz (1968: 381-6) to refer to any human aggregate characterized by regular and frequent interaction by means of a shared body of verbal signs and set off from similar aggregates by significant differences in language usage. Years later, Labov (1972: 120-121) noted that “the speech community is not defined by any marked agreement in the use of language elements, so much as by participation in a set of shared norms. These norms may be observed in overt types of evaluative behavior, and by the uniformity of abstract patterns of variation which are invariant in respect to particular levels of usage.”

Language Variation and Change: The Theoretical Framework

69

Closely related to the concept of speech community is that of community of practice. This term has been coined by Lave and Wenger (1991) to refer to groups which are united by common purposes and who engage in joint activity. The term was later used by Eckert (2000) in her study of a Detroit high-school. Eckert and McConnel-Ginet (2006: 490) define a community of practice as: An aggregate of people who come together around mutual engagement in some common endeavour. Ways of doing things, ways of talking, beliefs, values, power relations – in short, practices – emerge in the course of their joint activity around that endeavour. A community of practice is different as a social construct from the traditional notion of community, primarily because it is defined simultaneously by its membership and by the practice in which that membership engages. Indeed, it is the practices of the community and members’ differentiated participation in them that structures the community socially.

In this study, the community of practice is represented by the university/faculty where the research project for the (ing) variable was carried out, namely Queen Mary’s College, University of London (London, England) and Educational Center Theoretical High School (ConstanĠa, Romania) where the research project for the (pe) variable took place. I chose to include the community of practice in my analysis of gender-related variation in the speech of adolescents because it provides a framework for understanding the social as well as the linguistic facets of sociolinguistic variation, thus offering a more thorough picture of the phenomenon under investigation. Wenger (1998: 77-85) spells out three criteria which must be met in order to talk about a community of practice. These criteria are also discussed by Meyerhoff and Strycharz (2013: 429) and they illustrate them with some case studies. The first criterion is that there must be mutual engagement between the members. Wenger, however, points out that mutual engagement may be harmonious or conflictual. This means that a CofP is not necessarily made up of a group of friends or allies. The second criterion for a CofP is that members share some jointly negotiated enterprise. The members of the CofP get together for a reason and this reason is defined through their pursuit of it. Last, but not least, a CofP is characterized by the members’ shared repertoire. We have seen that the CofP is an aggregate of individuals negotiating and learning practices that amount to the realisation of a common goal. Wenger (1998: 125-6) notes that a CofP must also be characterised by:

70

Chapter Two

(i) the rapid flow of information and propagation of innovation; (ii) absence of introductory preambles and very quick setup of a problem to be discussed; (iii) substantial overlap in participants’ descriptions of who belongs and mutually defining identities; (iv) specific tools, representations, and other artifacts, shared stories and inside jokes; (v) jargon and shortcuts to communication; (vi) a shared discourse that reflects a certain perspective on the world. A range of linguistic behaviours will be characteristic of any given community of practice as long as these linguistic behaviours function in slightly different ways from one community of practice to another.

2.3. Politeness and its relevance to the study of language and gender In what follows we discuss the theoretical framework used for the qualitative analysis, which is a fusion between politeness theory (Brown and Levinson 1987), and swearing and taboo language (Allan and Burridge 2006). We start by discussing politeness36 and its relevance to the study of language and gender. Brown and Levinson (1987)37 present a cohesive, fully-fledged theory of politeness in which linguistic devices are viewed as realisations of specific politeness strategies. The concept of face is central in their theory, which is derived from that of Goffman (1967) and from the English folk term which ties face up with the notion of feeling embarrassed or humiliated, hence ‘losing one’s face’. According to Brown and Levinson (1987: 61) face represents “the public self-image that every member wants to claim for himself” and is of two types: negative face and positive face. Negative face refers to the desire to be unimpeded in one’s actions, whereas positive face describes the desire to be approved of. Many communicative activities presuppose imposition on the face of both or either of the participants: that is, they are face threatening acts 36

Fraser and Nolan (1981: 96) claim that “no sentence is inherently polite or impolite. We often take certain expressions to be impolite, but it is not the expressions themselves but the conditions under which they are used that determine the judgement of politeness.” 37 The book was written ten years after their extensive essay ‘Universals in language usage: politeness phenomena’ (Brown and Levinson 1987).

Language Variation and Change: The Theoretical Framework

71

(FTAs). Any act that impinges upon a person’s face to some extent (typically insults, criticisms) is an FTA. People are generally motivated to avoid FTAs and are willing to make efforts, take pains or incur emotional costs in order to save face. Brown and Levinson (1987) distinguish between two types of politeness: positive politeness, which has to do with the expression of warmth and friendliness towards others; and negative politeness, which has to do with not imposing on others, or threatening their face. Brown and Levinson (1987) proposed five superstrategies for performing an FTA and they are systematically related to the degree of face threat: (i) Bald on record: the FTA is performed “in the most direct, clear, unambiguous and concise way possible” (Brown and Levinson 1987: 69); (ii) Positive politeness: the use of strategies designed to redress the addressee’s positive face wants; (iii) Negative politeness: the use of strategies designed to redress the addressee’s negative wants; (iv) Off-record: the FTA is performed in such a way that “there is more than one unambiguously attributable intention so that the actor cannot be held to have committed himself to one particular intent” (Brown and Levinson, 1987: 69). To put it differently, perform the FTA by means of an implicature (Grice 1975). One of the earliest and most influential accounts of gender and politeness comes from Brown (1980) who carried out her research in Tenejapa, a Mayan community in Mexico using the politeness theory she developed with Levinson in 1978. Her research led her to conclude that women in the Tenejapan community used the extremes of positive and negative politeness, while men spoke more matter-of-factly. Women also had certain characteristic styles of politeness and so did men, such as sexy joking and a preaching and declaiming style. Based on the politeness theory model proposed by Brown and Levinson (1987), Ionescu-Ruxăndoiu (1999) looked at the Romanian spoken in rural communities. She found out that in rural communities, negative politeness usually occurs between insiders and outsiders or between insiders who do not have the same status. The examples provided by her come from a discussion between a man who comes to the house of a woman on a cold winter’s night and asks her brother to give him shelter for the night, using a conditional clause.

72

Chapter Two

(27) Dacă ieúti aúa de bun, zic, să mă primeú úi pă mine sâ...găzduiesc acolo... ‘If you’re so kind, say [I] to receive me to... stay there...’ This utterance is extremely polite. The girl’s brother must give his consent for the marriage, so he becomes powerful and therefore a social distance appears – which is practically inexistent as the interlocutors are both peasants. The proof is the usage of the term domnule ‘mister’ which is not common among villagers: (28) Domnule [...], zic, vino puĠîn pîn la poartă-ncuaci! ‘Mister say come a little to the gate here’ The fact that he abandons the local system in favour of the prestigious one represents a way of captatio benevolentiae (Ionescu-Ruxăndoiu 1999: 110). Positive politeness, attempts to minimize the distance between interlocutors by expressing friendliness and solid interest in the hearer’s need to be respected, in other words to minimize the FTA. One piece of advice given by a mother-in-law to her daughter-in-law is a good example of positive politeness. Talking about the way in which one must put the clothes to dry, the woman says: (29) Le scuturi frumos, le-ntinzi úi puni-le pă culme [...] Să le placă úi la altu dă rufili tăle. ‘You give them a good shake, you put them to dry So that others like your laundry’ Turning to the English language, a study on language use in New Zealand initiated by Holmes (1995) was also based on politeness theory. Holmes claimed that “women’ utterances show evidence of concern for the feelings of the people they are talking to more often and more explicitly than men’s do” (Holmes 1995: 6). Her study focused more on hedges, including tag questions. Closely related to politeness theory is the notion of impoliteness developed by Culpeper (1996). Impoliteness refers to the use of strategies whose aim is to have the opposite effect – that of social disruption. These strategies are oriented towards attacking face, an emotionally sensitive concept of the self (Goffman 1967, Brown and Levinson 1987). The politeness superstrategies mentioned above have opposite impoliteness

Language Variation and Change: The Theoretical Framework

73

superstrategies. They are opposite regarding their orientation to face, meaning that they attack face: (i) Bald on record impoliteness – the FTA is performed directly, clearly, unambiguously; (ii) Positive impoliteness – the use of strategies designed to damage the addressee’s positive face wants; (iii) Negative impoliteness – the use of strategies designed to damage the addressee’s negative face wants; (iv) Sarcasm or mock politeness – the FTA is performed with the use of politeness strategies that are clearly insincere, and thus remain surface realisations.38 (Source: Culpeper 1996: 356) In terms of politeness, women have been said to be more polite than men (Lakoff 1975, Coates 2004) and that men use swearing and taboo language more often than women do. This will be discussed more thoroughly in what follows. Swearing is a form of linguistic expression often referred to as bad language. Some people find swearing to be rude and disrespectful, while others use it every day. Allan and Burridge (2006) write that swearing, like slang, is found in the colloquial style and it includes a wealth of obscenities taken from the pool of dirty words. Slang, however, does not necessarily include swearing, but it usually does. Swearing can also act as an in-group solidarity marker within a shared colloquial style. Boys usually greet themselves using swear words and this feature has started to be used by girls also. Men use all kind of words to describe women: honey, sugar, pumpkin, sweetheart, darling, sweetie, etc, but also more offensive words like: fanny, tart, bitch, skank, cunt, whore, slut, wench, bimbo, hoe, floozy, sleaze, etc. Allan and Burridge further note that it is the use or non-use of swearing that marks the in-group. To this we may add the forms of jargon, slang and lots of abbreviation to increase the efficiency of communication. The idea that swearing is used by males and abhorred by girls or women has been proven to be false by research carried out in America (Risch 38 This is closely connected to Leech’s (1983: 144) Banter Principle which states that “in order to show solidarity with H, say something which is (i) obviously untrue, and (ii) obviously impolite to H” [and this will give rise to an interpretation such that] “what S says is impolite to H and is clearly untrue. Therefore what S really means is polite to H and true.

74

Chapter Two

1987), Great Britain (Hughes 1992) and South Africa (de Klerk 1992). Speakers usually use swearing in the presence of members of the same gender. However, men and women swear and use dirty words differently: The word ass was used by females to denote either a social deviation or a body part; it was used mainly as a body part by males. Cock, cunt, and dick appeared as body parts in males’ data but were not recorded for females. Similarly, neither tits nor pussy were used by females. For males, tits was a body part and pussy referred to a social deviation. Piss referred to anger for females but was more likely to mean a process for males. Balls, fuck, shit, and suck were used more or less by both males and females (Jay 1992: 139).

Taboo words have been divided by the British anthropologist Edmund Leach (quoted in Andersson and Trudgill 1990: 15) into three major groups: a. Dirty words having to do with sex and excretion (e.g. bugger, shit); b. Words that have to do with the Christian religion such as Christ and Jesus; c. Words which are used in ‘animal abuse’ (calling a person by the name of an animal) (e.g. bitch, cow). The word fuck is “one of the most interesting and colourful words in the English language today” (Andersson and Trudgill 1990: 60) which can be used in many different situations besides its sexual meaning: Fraud: I got fucked by my insurance agent. Dismay: Oh, fuck it! Trouble: I guess I’m fucked now. Aggression: Fuck you! Passive: Fuck me. Confusion: What the fuck? Difficulty: I can’t understand this fucking business. Despair: Fucked again. Philosophical: Who gives a fuck? Incompetence: He’s all fucked up. Laziness: He’s a fuck-off. Displeasure: What the fuck is going on? Rebellion: Oh, fuck-off!

Language Variation and Change: The Theoretical Framework

75

The word ‘fuck’ is also used in auxiliary swearing (i.e. not aimed at someone directly), for example: This fucking exam is giving me nightmares. For Kiesling (1998: 88) the word fuckin’, as a profanity, is associated with the vernacular, working class and physical power. Building on these theories and analyses I try to show that men and women do swear differently, and there are gender differences in this area. Dooling (1996: 5) claims that men swear because they are “uncouth warthogs by nature” and they feel manly in a violent way. He further states that women react violently to swearing and suggests that a man being harassed would tell his harasser to fuck himself while a woman would file a formal complaint. Swearing for men is a substitute for a “good, long cry”, since men are incapable of indulging in crying.

2.4. Concluding remarks The aim of this chapter was to build up a methodological and theoretical framework for a better understanding of gender-related linguistic variation in the speech of adolescents. For the quantitative analysis, a variationist approach is extremely useful to the study of gender-related preferences in the speech of adolescents, as the focus is on linguistic variables (phonological, morphological and syntactic). We correlate these linguistic variables with social factors (gender) and account for the differences. Research carried out within the framework of variationist sociolinguistics in English-speaking communities has revealed important gender differences between the speech of men and women and represent an excellent starting point in this research. The fact that men prefer the vernacular forms and women the prestige forms provides invaluable insight. The quantitative analysis must be followed by a qualitative one that is why I proposed, as a theoretical framework, politeness theory including the study of swearing and taboo language. It was shown that women are more polite than men and also that men use more swearing than women do and that they do swear differently.

CHAPTER THREE TALKIN’ OR TALKING? PHONOLOGICAL VARIATION IN THE SPEECH OF LONDON TEENAGERS

Introduction London is one of the most important cities in the world, a landmark in the heart of England. Millions of students from all over the world come to study here at university, students of different ethnic and social backgrounds. This analysis is a variationist one and it examines phonological variation in the English spoken by teenagers of British origin studying in London, seeking to answer the following questions: (i) are there any gender differences regarding the use of (ing); (ii) is the use of the [n] or [ƾ] variant influenced by other social factors besides gender. Based on previous research (Fisher 1958, Labov 1972c, Trudgill 1974) I hypothesise that: (i) boys might use the [n] variant more frequently than girls; (ii) there might be variation in the usage of (ing) depending on the speech style; (iii) girls might use the [ƾ] variant more frequently than boys; (iv) there might be variation between the [n] and [ƾ] variants in the speech of the same speaker. Before embarking on the discussion proper, I begin with some background information about the (ing) variable and previous studies which analysed this variable in different English-speaking communities. Then I will analyse the (ing)39 variable (the variants of (ing) differ from community to community, but usually there is an alternation between a velar nasal, 39

In some studies the variants of the (ing) variable are written as -ing and -in’, or G and N (Cofer 1972) and Houston (1985), but I find them somewhat complicated and I prefer the [ƾ] and [n] variants.

Talkin’ or Talking? Speech of London Teenagers

77

which I will refer to as [ƾ] and an alveolar nasal, which I will refer to as [n], e.g. talking [tǣ:kǹƾ] and [tǣ:kǹn]40) in the speech of London teenagers. This chapter reports on the findings of a research project carried out in October 2012 at Queen Mary’s College, University of London (QMUL).

3.1. The (ing) variable: previous research There is a vast body of literature on methods and findings regarding the (ing) variable, so here I give only a broad-brush picture of some of the main findings. It is the first variable to have been studied quantitatively (Fisher 1958). It is considered to be a stable sociolinguistic variable, meaning that it is not undergoing linguistic change and a correlation with age is not expected (Tagliamonte 2011: 187). However, there have been some studies which showed that a correlation with age can be expected – Labov’s (1966) famous study of New York City speech revealed that [ƾ] is sensitive to age, while Horvath (1985: 97) showed that [n] was a feature of Australian teenagers’ speech. Schleef et al. (2011) provide five reasons for studying the (ing) variable: (i) (ii) (iii) (iv) (v)

it occurs frequently in conversation; it is salient among native speakers of English; it is stable (and has remained so for at least fifty years); it has been studied in a number of varieties of English since the 1950s; it can be analysed auditorily

The linguistic variable (ing) has been correlated with social class and gender. As mentioned above, Fisher (1958) was the first to analyse the (ing) variable quantitatively and the community he chose was rural New England. He discovered that boys used more frequently the [n] form and girls preferred the [ƾ] form.

40

The alternation between [t‫ܧ‬:kܼƾ] and [t‫ܧ‬:kܼn] in most spoken English shows that there are (at least) two phonetic realisations of the (ing) morpheme without changing the meaning of the word as referring to the act of talking.

Chapter Three

78

-ing > -in’ -in’ Boys Girls

5 10

-ing < 7 2

Table 3.16. Preferences for [ƾ] and [n] endings, by gender (Source: Fisher 1958: 484) Fisher concludes that in this community [ƾ] is considered to symbolise female speakers while [n] is associated with male speech. He also found variation in the speech of same gender speakers. For example, he distinguished between a “model boy”41 who, almost exclusively used the [ƾ] ending, while the [n] ending was used with higher frequency by a “typical boy”42. This is illustrated in Table 3.17:

Model boy Typical boy

-ing 38 10

-in 1 12

Table 3.17. Preference for [ƾ] and [n] in same gender speech (Source: Fisher 1958: 484) There has been a plethora of research concerned with the (ing) variable, subsequent to Fisher’s. The (ing) variable has been studied in the United States: New York (Labov 1966), Detroit (Shuy et al. 1967), North Carolina (Anshen 1969), Philadelphia (Cofer 1972, Labov 2001), Los Angeles (Wald and Shopen 1985); in Canada (Ottawa (Woods 1979); in Britain: Norwich (Trudgill 1974), Edinburgh (Reid 1978), London and Edinburgh (Schleef et al. 2011). It has been identified in all varieties of English and it is one of the most analysed features of the English-speaking world (Chambers 2002: 120). Besides the [ƾ] and [n] variants, a third variant [ƾk] was found in Canberra (Shopen 1978, Wald and Shopen 1985). This is also corroborated by Hughes and Trudgill (1987: 45) who note that in thing words, e.g. nothing, something, (ing) may be phonetically realised as [ƾk].

41

A model boy describes a boy who did his school work well, was popular among his peers, and considerate. 42 A “typical boy” is physically strong, dominating, full of mischief, and disarmingly frank about his transgressions.

Talkin’ or Talking? Speech of London Teenagers

79

Labov (1966: 394) considers the (ing) variable “a case of stigmatization without change”. He further notes that it has been in use since the 19th century with almost the same variants and social associations. Consider the following quotations from writers such as Sir Arthur Conan Doyle (in England) and Mark Twain (in the United States of America): (30) “You must not mind my bein’ just a little short wi’ you at first, for I’m guyed at...” Sir Arthur Conan Doyle – The Sign of the Four (2007: 128) (31) “’What’s de user makin’ up de camp-fire to cook strawbries en sich truck? But you got a gun, hain’t you? Den we kin git sumfn’ better den strawbries.’” Mark Twain – The Adventures of Huckleberry Finn (1994: 48-49) These two examples (in written English) reveal that even in the 19th century writers noted differences in terms of pronunciation. The [n] variant was, of course, associated with the lower class. In (30) the person who speaks is an oar man who lives in the East End. In (31) the sentences were uttered by an African American slave, Jim. So, here, the use of the [n] variant is associated with social class and ethnicity. In many studies, class is a key feature of the distribution of (ing). Campbell-Kibler (2006: 26) pinpoints the fact that the [n] variant is frequent in the speech of people lower on the class hierarchies while the occurrence of the [ƾ] variant increases in the speech of those higher up. In the literature it is noted that female speakers favour the [ƾ] form while male speakers favour the [n] variant. Tagliamonte (2011: 189) states that the [ƾ] variant is correlated with formality in speaking, while [n] is associated with the working class and less education, as indicated in the two excerpts in (30) and (31) above. In his epoch-making study of the English spoken in Norwich, Trudgill (1974) correlated the (ing) variable with social class (MMC = middle middle class, LMC = lower middle class, UWC = upper working class, MWC = middle working class, LWC = lower working class) and analysed it in five different speech styles (WLS = word list style, RPS = reading passage style, FS = formal speech, CS = casual speech). Trudgill found that male speakers used a higher percentage of the [n] forms than female ones. He also noted that women are more status-conscious than men, and are more aware of the social significance of linguistic variables (Trudgill 1974: 94). He suggested that speakers associated working class with

80

Chapter Three

masculinity, causing men to favour the [n] variant and women the [ƾ] form. This relationship between the (ing) variable and masculinity is discussed more thoroughly by Kiesling (1998) who studied the variable in a college fraternity in Virginia. His analysis revealed that some members frequently used the [n] form during fraternity meetings as compared to social times. The fact that some members use [n] as a masculine identity marker does not contrast with the female patterns of speech, but with the masculinities of other members, who constantly use [n] in social situations and [ƾ] in their fraternity meetings. This is shown in Figure 8:

Figure 3.8. Percent [n] for individual fraternity men across activity types for progressive forms only. (Source: from Kiesling 1998: 85, Figure 8)

Another important study focusing on adolescents, this time, is Reid (1978) carried out in Edinburgh. He found that adolescents used more frequently the [n] variant when the situation was less focused on language and less oriented towards adult topics. The [ƾ] form was favoured in the reading passage, and the [n] variant was used more frequently in the formal interview, displaying an even higher rate in the adventure playground. Many studies correlated the (ing) variable with gender. Consistent findings for this variable are that females prefer the [ƾ] variant while males favour the [n] form. The [ƾ] variant is also associated with formality in speaking: in careful speech in probably all standard accents of English the

Talkin’ or Talking? Speech of London Teenagers

81

variable (ing) is phonetically realised as [ƾ] (Chambers 2003: 121-126). Table 3.18 summarizes the results of studies which found gender differences: Study

Location

Labov (1966) Shuy et al. (1967) Shopen (1978) Houston (1985), 35yo Wald and Shopen (1985)

New York Detroit Canberra Britain Britain Canberra

Men 36 62 24 88 78 23

Women 21 16 72 76 24

Both 31 -

Table 3.18. Percentage of [n] in men’s and women’s speech across studies (Source: Campbell-Kibler 2011: 35) In these studies the influence of gender interacts with social class as well as the context. Labov’s (1966) data revealed more [n] use for men within the lower class speakers and also showed that as far as the middle class is concerned, men used less [n] forms than women. Although Anshen’s (1969) study is not included in the table above, it is worth discussing it at this stage. Even though Anshen did not find any gender-related differences regarding the use of the (ing) variable, he did discover an interaction between the gender of the speaker and the race of the interviewer who was male. The interviewees were all black and they used the [n] variant more with a black interviewer than with a white one and this was more frequent among the men than the women. The study made by Wald and Shopen (1985) is of particular interest because of the method used to collect data, but also because of the results obtained. They collected their data via note taking, a method that is more susceptible to researcher bias than the methods which presuppose recording. The stylistic situations that they noted are based on the speaker/addressee relationship (family, friends, etc). Their Canberra data reveals higher [n] usage among friends than with family or other situations. They also note that [n] occurs more frequently in common words used every day, whereas [ƾ] is favoured in longer, more sophisticated words. Campbell-Kibler (2006: 23) notes that the phonological influences on (ing) are relatively few and straightforward. In the so-called thing words, the variable is phonetically realised as [ƾ] in most varieties of English. However, Hughes and Trudgill (1987: 46) note that in nothing, something,

Chapter Three

82

etc., (ing) may be [ǹƾk], a variant also found by Shopen (1978), Wald and Shopen (1985) in Canberra. On the other hand, the thing words have an important influence on the rates of (ing) use and they are divided into: something and nothing which favour [n], and everything and anything which exhibit almost exclusively [ƾ] (Campbell-Kibler 2006: 23). Cofer (1972) and Houston (1985) offer an explanation regarding variation in thing words, suggesting that the latter category (i.e. everything, anything) has a secondary stress on (ing). Another example is the phrase going to which in some varieties of English is pronounced either [gԥșǹntԥ] or [gǡnԥ] (especially in American English) while in others it is variable, but the former variant is strongly favoured. Let us consider the following excerpt taken from a sociolinguistic interview: .

(32) “So she said, ‘Well, I want something on my desk by five-oclock.’ You-see, well. ‘You’ve got no chance.’ ‘Well when can I see it?’ So I said ‘Don’t worry, there’ll be somethin’ on your desk by nine o’clock tomorrow.’ Put the phone down.” (York English Corpus, male, aged 40, source: Tagliamonte 2006:79) This illustrates the occurrence of intra-speaker variation in the phonetic realization of the word something. The situation in which the word occurs is practically the same, so it is difficult to explain why it varies. In a recent study, Schleef et al. (2011) analysed the (ing) variable in the speech of Polish migrants living in Edinburgh and London. This study was motivated by the belief that migrants (Polish) acquire patterns of variation typical of their local peer group. 3389 tokens of (ing) were analysed and it was revealed that the [ƾ] form is used more frequently by the nativespeaker adolescents in Edinburgh than in London. On the other hand, the Polish teenagers living in Edinburgh use the apical variant more often than native-born Londoners: In London, the constraint which operates most consistently across both speaker groups is an effect of progressive dissimilation. That is, where the phonological segment immediately preceding (ing) has an apical place of articulation (e.g. spending up), (ing) is more likely to be realised as a velar [ƾ]; when the phonological segment immediately preceding the variable (ing) has a velar or glottal place of articulation (e.g. speaking out), the (ing) variable is more likely to be realised as an apical [n] (Schleef et al. 2011: 218).

Another interesting finding was that Polish boys strongly favour the velar realisation of (ing), i.e. the [ƾ] variant, while Polish girls prefer the apical

Talkin’ or Talking? Speech of London Teenagers

83

realisation, i.e. the [n] variant. This is contrary to Fisher (1958), Trudgill (1972, 1974), Newbrook (1982) and Trudgill (1983). Trudgil’s (1974, 1983b) survey conducted in Norwich revealed that women, irrespective of the style and social class use the [ƾ] variant more often than men. Newbrook’s (1982) study of West Wirral corroborated his findings: his female informants used almost twice as many instances of the prestigious variant as the men.

3.2. Data collection and methodology 3.2.1. Speech style As mentioned in Chapters 1 and 2, the aim of this study is to discover gender-related variation in the speech of adolescents in English and Romanian so it makes sense to choose speakers of different social backgrounds. For Labov (2001), the speech style forms a linear continuum, which mirrors the amount of attention that speakers give to their speech. In formal situations they will monitor their linguistic behaviour and focus on how they say/speak. However, given the fact that this study focuses on teenagers, they do not really pay attention to their speech, so we can have access to what we are really interested in: the vernacular and non-standard forms. The sociolinguistic interview consisted of three parts: a semi-structured questionnaire, three texts (reading style) and a list of words (word-style). The reason why I included the texts and the list of words was to see whether there are differences in pronunciation from one type of speech to the other. The crux of the matter is that when reading aloud, people tend to be more conscious of their speech and use more standard language, while in casual conversation they focus more on what they want to say rather than the way they say it, allowing their vernacular to emerge (CampbellKibler 2006: 28). The interviews lasted between 25 and 30 minutes, giving me about four hours of speech for analysis. It is a truth universally acknowledged that the language of teenagers is of particular interest because of the important innovations and changes in language use (Labov 1972c, Cheshire 2005). Teenagers are, indeed, responsible for the linguistic innovations and changes, some of which are incorporated into the general structure of the language over time. This particularly applies to the lexical level, as teenagers are creative in their use of language and constantly invent new words or borrow lexical items from other languages.

84

Chapter Three

3.2.2 Recording procedures The data examined here come from eight sociolinguistic interviews with British teenagers (four boys, four girls), who live and study in London. The interviews took place at Queen Mary, University of London (QMUL), in October 2012. The recordings took place inside the campus (building A1) to avoid the noise and different disturbances. Before the interview, each participant signed an informed consent form, thus allowing me to use the recording (the data) for academic purposes43. Entering a speech community may prove to be a daunting task, if you are a foreigner (as I was). The best way to enter the community is with the assistance of school officials (Schilling 2013: 185), which I did. I contacted Professor Jenny Cheshire, and after I told her about my project she kindly accepted to help me. She allowed me to attend one of her course on “Variationist Sociolinguistics” and introduced me to her students. She also e-mailed her colleagues from QMUL who introduced me to their students. This friend-of-a-friend technique was successful and I was able to interview the students. Jenny Cheshire also suggested approaching students in the cafeteria, which I did. This was not successful at all, as I was an outsider from another country and they seemed reluctant. Many of them said that they do not have the time, others that they have a lecture in about ten minutes while others simply refused to be recorded. The fact that I entered the speech community via university officials allowed me to conduct my research. The professors asked me to tell something about my research to the students, which I did (omitting, of course, that I was interested in the (ing) variable). I told them that I was also a student, conducting research for my PhD dissertation. Some of the students agreed to get involved in the research and I made appointments. The boys were more reluctant to participate so I had to persuade them. The girls were more enthusiastic and sympathetic.

43

For Tagliamonte (2006: 33) the most important ethical guidelines for collecting interviews are: (i) written consent for audio-recording; (ii) guaranteed anonymity; (iii) voluntary participation; (iv) access to researcher and research findings. When collecting the data, all these criteria have been met.

Talkin’ or Talking? Speech of London Teenagers

85

As far as the equipment used to record the interviews, I used a Sony ICD-PX312M audio recorded, with a microphone which produced high quality recordings. I transcribed the audio interviews in Microsoft Office Word resulting in a corpus of around 14 000 words. Also, I used MP3 Cutter to cut the audio recordings and extract the words containing the (ing) variable. The audio format of the recordings is WAV (Wave Sound). I was careful when choosing the recorder and made sure that the format of the recordings will be WAV. I was interested in this particular audio format as it is more common and it is compatible with acoustic analysis softwares such as the software programme PRAAT (Boersma 2001) which I used. Acoustic analysis together with variationist analysis were used to offer detailed descriptions of variants and to corroborate auditory judgements. I had a set of questions which I used, but with some speakers I also had to use other questions to elicit more information on a certain topic. For example, one of the questions was “Would you consider moving to another country”. Some speakers simply answered with “Yes” or “No”, so I had to use questions like “Ok, then could you describe for me your favourite country?” to elicit as much data as I could. The speakers were left to speak without interruption and at times I used backchannels like mhm, yeah so that the speaker knew that I was paying attention to what he/she was saying. The interview was structured around certain topics of conversation (e.g. family, living in London, friends, university life, childhood), but the conversation was not limited to these topics and the participants were encouraged to talk freely on other topics as well. They were also recorded performing a reading task (3 texts of about 15 lines each, and a list of 70 words).

3.2.3. The speakers All of the teenagers come from English-speaking families and now they live and study in London, at QMUL (Mile End Road). They belong to one age group (19–20), as revealed below:

Chapter Three

86

Namee

Age

P Place of birth h

Livess in

Amannda Suzannne Joannna Mary Peter Mark Tom Josh

19 19 20 20 19 19 20 19

Central Londo C on (Camden T Town) I of Wight Isle N North London n (Barnet) K Kings N North London n (Barnet) L London (City of London) L London (City of London) N Norfolk (Norw wich)

London London London London London London London London

Table 3.19. Information about the pa articipants In order to hhave a clear picture p of the places p where the participan nts live in London (wiith a few excceptions), I included a ma map showing London’s L boroughs:

w tes.com) Map 1. The bboroughs of Lonndon (Source: www.anglonaut

Talkin’ or Talking? Speech of London Teenagers

87

Age is a significant social factor, so I tried to choose speakers who were 19 or 20. I also tried to record some high-school students, but the situation was more complicated. They were still under age, so the school principal did not allow me to record them without their parents’ written consent, even though the high school students agreed to take part in the research project. The fieldwork procedures used here meant that the sociological variables could not be controlled, since speakers were not pre-selected, but they volunteered to take part in the research. I interviewed all of them, without any type of discrimination. Some of them were Muslims, some Catholics, others Protestants. I could not ask them detailed personal information about their incomes or their parents’ incomes and level of education because after all I was a stranger and they might have refused to participate in the interview. Although the concept of social class is very important, the participants were all university students, so this criterion was met.

3.3. (ing) patterns in the speech of London teenagers Each token of (ing) was coded as alveolar [n] or velar [ƾ]. Each token was also coded for the independent variables of speaker, speech style (casual, reading, word-list) and grammatical category. The dependent and independent variables used in this analysis and the factors coded in for each are the following: Dependent variable: Ɣ (ing): realised as velar nasal [ƾ] and as alveolar nasal [n]. There is, however, a third variant [ƾk] which appears in the speech of only one participant and it was excluded from the analysis. I will refer to this case later on. Independent variables: Ɣ word class: common noun (e.g. wedding), progressive verb form (e.g. she’s singing, was doing), gerund (e.g. we like eating bread), adjective (e.g. interesting), discourse marker (e.g. or something), pronoun (e.g. I do know something about it); Ɣ number of syllables in the word: 2, 3, 4 (something, interesting, developing). Mono-morphemic words (e.g. thing, ring, king, sing, etc) have been excluded from the analysis; Ɣ speech style: speaking, reading; Ɣ gender: male, female.

Chapter Three

88

Unclear tokens of (ing) – such as the reduced form gonna of going to were excluded from the analysis. The [ƾk] variant which occurred in the word something and which was used only by Mark, was treated separately as it could not be coded as [n] or [ƾ]. This left 342 tokens of (ing) from our London data obtained from speech (the interview) to which we add another 60 tokens of (ing) from reading style, giving us a total of 402 tokens of (ing). Below I give the number of tokens for each participant in the interview (speech style):

Number of ING tokens 80 60 40 20

Number of ING tokens

0

Figure 3.9. The number of (ing) tokens from each interview

I used MP3 Cutter to extract from the recordings the words containing the (ing) variable and I analysed all the tokens using PRAAT and ELAN. Then, I counted all the [ƾ] and [n] tokens for each teenager. The results for speech style, 342 tokens of (ing), are presented in Figure 10:

Talkin’ or Talking? Speech of London Teenagers

89

60 50 40 30

[n]

20

[ż]

10 0

Figure 3.10. The use of [ƾ] and [n] in the speech of London adolescents

The studies carried out so far in English-speaking communities revealed that there is gender-related variation regarding the use of the (ing) variable.

3.3.1. Casual speech style: results The data presented in Graph 8 shows the distribution of the variants of (ing) among the teenagers recorded at QMUL for interview (speech style) only. To give an idea of the distribution of (ing) depending on the gender of the speaker, I grouped the use of the alveolar nasal and the velar nasal according to the gender of the speaker, presented in Figure 3.11: 150 100 [n] 50

[ż]

0 Boys

Girls

Figure 3.11. The use of the [n] and [ƾ] according to the gender of the speaker

90

Chapter Three

It seems that London teenagers prefer the alveolar variant to the velar one. However, the situation is not as simple as it may seem. There are other social factors, besides gender, that must be taken into account. Note that Amanda shows a high usage of the [n] variant. She admits using Cockney (although she was born in Camden Town) and says that life was quite rough for her. Cockney is associated with the lower class (Hughes and Trudgill 1987) and has very low social prestige (Altendorf 2003: 38). As shown by Trudgill (1974) the [n] variant is used more frequently by people lower on the social scale irrespective of their gender. This could explain Amanda’s higher use of the alveolar variant. Figure 11 also indicates that the girls use more frequently the [n] variant, with the exception of Joanna. On the other hand, all the boys prefer the alveolar realisation of (ing) which was not a surprise, as this accords with what other studies have found. It was interesting to find that girls prefer the alveolar variant and not the velar, as has been suggested in the literature so far (Fisher 1958, Labov 1972, Trudgill 1974). My results corroborated those obtained by Schleef et al. (2011: 227) regarding the use of the (ing) variable by Polish migrants living in London and Edinburgh: “Polish boys strongly favour the velar realisation of (ing) and Polish girls favour the apical realisation, the exact opposite of what we would expect based on previous research”. As already mentioned, for the analysis of (ing) I used PRAAT. The two spectrograms below show the realisation of [n] and [ƾ] in the speech of Amanda. The first spectrogram shows the occurrence of the velar in the phonetic realization of the word being, which can also be seen in more detail in the second spectrogram. The circled area represents the velar [ƾ], and also included is the pitch. Spectrogram number 3 shows the realisation of the same word (i.e. being), but this time exhibiting an alveolar. I also magnified it to highlight the realisation of (ing) as [n].

Talkin’ or Talking? Speech of London Teenagers

Spectogram 1. The velar realisation [ƾ] by Amanda of (ing) in being

Spectrogram 2. The velar realisation [ƾ] by Amanda of (ing) in being

91

92

Chapter Three

Spectrogram 3. The alveolar realisation [n] by Amanda of (ing) in being

In spectrogram 3, I used brackets to highlight the realisation of (ing) in “being”, as an alveolar, and there is a noticeable decrease in the pitch. This also shows that there is intra-speaker variation for the same word, and Amanda is not the only speaker where I came across this phenomenon. In what follows I give some examples of intra-speaker variation from the English corpus: (33) “Goin’ to school? Uhmm uni or just school? Yes, I used to like going to school but at uni...” (Joanna) (34) “...was writing it and I was havin’ this argument with like the regional south England manager for somethin’ and then he turned up one night at about three in the mornin’ when I was workin’ there umm and asked what’s the meaning of this...” (Mary) (35) “...some people are probably better off having just the basics and then stayin’ in their village and or just havin’ a quiet life...” (Suzanne) (36) “Umm I’ve had one incident while while I was sleepwalking and when I woke up everyone was tellin’ me that I just peed at the bottom of my parent’s bed...” (Amanda) (37) “...it would be very nice seeing my grandparents, my cousins, aunts, uncles because I never got to see’em many months apart and it is a very enjoyable experience and also speakin’ Italian...” (Tom) (38) “...that I fault or I fault or somethin’ or someone might’ve like play acted or something and stuff like that.” (Mark)

Talkin’ or Talking? Speech of London Teenagers

93

(39) “...I do enjoy spending time with people but there’s a lot more that I prefer doin’ that just I don’t have to think about things for example playin’ video games I prefer by myself, watching films...” (Josh) (40) “...go around the world doin’ quite a little gigs in pubs or somethin’ but just travellin’ and doin’ something I’d love to do...” (Peter) These excerpts show that there is intra-speaker variation and that speakers prefer the non-standard variant [n]. From the tokens of (ing) gathered from the interview with Amanda, she used the alveolar form in 90% of the cases, which shows a preference for the non-standard form. The one who used the prestige form in about 85% of the cases was Joanna. Another independent variable taken into consideration for the analysis of (ing) was the word class. I divided the (ing) tokens obtained from the interview into: common nouns, progressive verb forms, gerunds, adjectives, discourse markers and pronouns. In Figure 3.12, I give the number of (ing) tokens for each grammatical category indicating which variant was favoured: the velar or the alveolar. 120 100 80 60 40 20 0

alveolar velar

Figure 3.12. Distribution of alveolar and velar variants according to the word class

As expected, there is a higher use of progressive verb forms and gerunds in contrast to the other word classes where variation occurs. Teenagers tend to use the alveolar nasal more frequently in common nouns, progressive verb forms and gerunds. On the other hand, they tend to use the velar nasal in adjectives, discourse markers and pronouns. The thing compounds have been grouped into two word classes, depending on the

94

Chapter Three

word class in which they occurred: discourse markers and pronouns, hence the preference for the velar form. Above, I presented the realisation of (ing) according to the word class in the speech of each teenager interviewed to enable a better understanding of gender-related variation. I grouped the pronouns and discourse markers into one umbrella-term, i.e. THING compounds. The THING compounds will also be discussed separately, as they require special attention.

[ƾ] Mary 8/4 10/2 7/2 2/0 1/3 28/11

[n] Joanna 2/6 2/2

1/10 0/3 0/7 5/28

16/2 1/5 0/10 21/19

Suzanne 4/2 0/0 28/1 4/0 2/3 52/5

Amanda 5/1 13/0 28/6 2/0 1/1 38/11

Tom 1/2 6/2

Table 3.20. The use of [n] and [ƾ] by London teenagers according to word class

Common nouns Progressive verb forms Gerunds Adjectives THING compounds Total

Word class

Talkin’ or Talking? Speech of London Teenagers

8/5 0/0 5/5 13/10

Mark 0/0 0/0 34/8 0/2 0/3 43/17

Josh 2/2 7/2 29/0 0/0 2/5 37/5

Peter 1/0 5/0

95

96

Chapter Three

Schleef et al. (2011: 219) note that the most important linguistic constraint on (ing) in English is that it is affected by the nominal-verbal continuum (Labov 2001: 88). However, the teenagers that participated in my research project favoured the alveolar variant also with common nouns, not to mention the gerunds (which have properties of both verbs and nouns) where there is a high use of the alveolar variant. From Table 20 it is clear that, with some minor exceptions, the apical variant, i.e. [n], is used more frequently, which gives us an important insight into the word class containing (ing) and which variant is preferred. The next question that must be answered is “Why?” In doing so, I first present a graph which shows the use of (ing) according to two independent variables: word class and gender, since gender variation is what I am interested in: 120 100 80 60 40 20 0

M [ż] M [n] F [ż] F [n]

Figure 3.13. The use of (ing) according to word class and gender

Abramowicz (2007: 30) states that (ing) is subject to little or no phonological conditioning and Labov (2001) observed that this variable is more of a morphological alternation than a case of phonological reduction. Abramowicz further notes that it is also subject to fine grammatical conditioning, meaning that the closer (ing) is to being a verb, the greater the chance for opting for the [n] variant. In order words, the usage of the velar and apical variants follows the nominal-verbal continuum (Houston 1985, Labov 2001) mentioned above. With respect to thing compounds (discourse markers and pronouns) the velar form was the one preferred by both genders, as stated above. There were 48 thing compounds in the corpus and there was variation between the [n] variant and the [ƾ] variant only in the compound something. A third

Talkin’ or Talking? Speech of London Teenagers

97

variant [ƾk] occurred only in the speech of Mark, but I did not include it in the analysis as it was an isolated case and he was the only one to use it. The other thing compounds contained a velar realisation. Houston (1985) points out that in American English thing compounds like anything and everything (which have 3 syllables) carry secondary stress on (ing) and favour a velar realisation, which is also confirmed by our data. Tagliamonte (2011) examined the (ing) variable in York and Toronto and she concluded that with pronouns like something and nothing there are higher rates of the alveolar variant in Toronto than full noun phrases. In York on the other hand, something and full noun phrases act the same, as opposed to the others.

Figure 3.14. Pattern of alveolar variants of variable (ing) among nouns compared with indefinite pronouns in York and Toronto (Source: Tagliamonte 2011: 194).

In what follows we look at the use of the thing compounds in more detail, as they were used by each student interviewed, for a better understanding of the phenomenon.

nothing everything something somethin’

everything something nothing something something something something

everything everything everything something anything anything something something anything anything

Suzanne something something something somethin’ somethin’

Amanda

Tom somethin’ nothing

Chapter Three

Table 3.21. thing compounds in the speech of London teenagers

Mary

Joanna

98

somethin’ something somethin’ something somethin’ something something somethin’ anything somethin’

Mark

Peter something something anything anything anythin’ somethin’ something

Josh nothing something something

Talkin’ or Talking? Speech of London Teenagers

99

As can be noticed, variation occurs only in the case of something. There is also a case of variation regarding anything, but it can be considered an isolated case, as it occurred only once in the speech of Peter. In the literature, the thing compounds have been divided into something and nothing, which favour the apical realisation of (ing) and everything and anything in which (ing) has a velar realisation. Recall that these words carry secondary stress on (ing). At this point it is also necessary to include spectrograms in the analysis of these compounds to show variation, but keeping in mind the fact that variation occurred almost exclusively in the case of something, at QMUL. It is still debatable whether these differences are linguistic or social, but the analysis proves how important it is to code these nominal types separately. It is hard to identify the particular constraints that affect the use of the velar or alveolar variant, as apparently there are no such constraints. A possible explanation is offered by Cofer (1972) and Houston (1985) who suggest that this is due to the syllable structure, and there is no variation with words like everything and anything because they carry secondary stress on (ing).

Spectogram 4. Alveolar realisation [n] by Mark of (ing) in something

100

Chapter Three

Spectogram 5. Velar realisation [ƾ] by Mark of (ing) in something

The two spectrograms above show the realisation of (ing) in something as either alveolar or velar nasal in the speech of Mark, as in his case intraspeaker variation was more pervasive. Between the red parenthesis is the realisation of (ing) as [n] in spectrogram 4 and as [ƾ] in spectrogram 5. I also include spectrograms of the thing compounds where there is no variation in order to show the lack of it.

Spectogram 6. Velar realisation [ƾ] by Suzanne of (ing) in everything

Talkin’ or Talking? Speech of London Teenagers

101

Spectogram 7. Velar realisation [ƾ] by Suzanne of (ing) in anything

The realisation of (ing) as a velar nasal is indicated in the red brackets. The tokens were chosen from the speech of Suzanne as in her case the use of the thing compounds was more consistent, just like in the case of Mark. There is also gender variation regarding the use of thing compounds. As expected, girls prefer the velar variant almost exclusively. Although male speech is associated with the vernacular forms, in this case the boys also preferred the prestige variant, i.e. the velar one, as shown in Figure 3.15 below: 15 10 [ż] 5

[n]

0 Boys

Girls

Figure 3.15. The use of thing compounds according to gender

Chapter Three

102

Since there was no variation with the other thing compounds, only with something, these results do not come as a surprise. Our finding is also corroborated by other studies which focused on the (ing) variable44. So far I have discussed the findings from the interview, i.e. speech style. I have shown that in the casual speech style girls tend to use the alveolar variant more often than the velar one, just like boys do. With regard to common nouns, girls tend to use the [n] variant more often than boys, who prefer the [ƾ] variant. Another gender-related difference is in the use of thing compounds, but variation occurs only in the case of something. Another intriguing example regarding the (ing) variable is provided by Holmes (2008: 167): (41) Knocker: Jim: Knocker:

Jim: Knocker:

Comin’ down the club Jim? Not friggin’ likely. It’s rubbish that club. It ain’t that bad. Music’s cool. I seen a couple of sharp judies there too. If we plays our cards right...Anyways you was keen enough last week. The music’s last Knocker. I’m off down the Pier’ ead if there ain’t nothin’ better on offer. Bleedin’ rozzers crawlin’ round down there. Come down ours instead. (Source: Holmes 2008: 167)

Holmes (2008) considers that the vernacular forms express machismo. This idea is not a new one, as there have been many voices who advocated it45. In what follows I analyse the results obtained from reading style and word-list style.

44

See Houston (1985), Labov (1989), and more recently Schleef (2011). After each interview conducted at QMUL each participant asked me the purpose of this research project and what I was interested in. After I told them that I was interested in the (ing) variable, Amanda, Mary and Suzanne told me that they never pronounce the final “g” and that it is incorrect to pronounce it. “That’s bad English” said Suzanne. This statement was quite intriguing as the girls used more of the standard form than the boys. The fact that London girls tended to claim that they use more the vernacular form than they actually did supports the claim that these forms carry connotations of toughness and masculinity and girls want to be identified with such traits. 45

Talkin’ or Talking? Speech of London Teenagers

103

3.3.2. Reading style: results After finishing the first part of the sociolinguistic interview (the semistructured questionnaire), the students were asked to read three texts of about fifteen lines each. The aim was to gain insight into styles that are more formal than the casual speech style and to obtain information on a wider range of styles. This is the most formal style in the interview, which is referred to as “reading style” (Trudgill 1974: 47). The informants were asked to read the texts as naturally as possible, as they would do it in everyday life. The texts contained 37 tokens of (ing). I did not expect too much variation as this situation was more formal than the questionnaire, and when reading you pay attention to the pronunciation, especially when you are being recorded. The results were, however, beyond our expectations, as the girls, almost exclusively, used the alveolar nasal and not the velar nasal. The results are given below: 35 30 25 20

[n]

15

[ż]

10 5 0 Boys

Girls

Figure 3.16. The results for reading style according to the gender of the speaker

The girls used the [n] form more often than the [ƾ] form, even in the reading style. They were not interested in the way they read the text; they just read it without any regard for the intonation or punctuation. The boys, on the other hand, paid attention to the text, some of them even asked me to give them a moment to have a look at the text first, before reading it. This indicated that they paid attention to the pronunciation and acknowledged that they were being recorded. They strove to sound as close to RP as possible, and after the interview some of them admitted

104

Chapter Three

doing so. This provides us with valuable information regarding the importance they assign to the way they are perceived in society. The next stage in the interview is known as the “word list style”. The interviewees were asked to read a list of 68 words. The list included 23 tokens of (ing) and the rest of the words were distracters. The informants were told to read aloud the words, at a normal speed and as naturally as they could. The word list test is an artificial one, as Trudgill (1974: 48) admits it, but it is useful in eliciting a formal style, thus providing information about the community’s linguistic norms. Although the informants were asked to use their normal everyday pronunciation, they are not likely to do this when reading a list of words. The results obtained for the word list style are almost identical with the ones obtained for reading style. It is debatable whether reading styles are on the same plane as spoken styles, since people can have specialized reading registers (Schilling 2013: 104) that differ in a number of ways from spoken speech, not just the degree of carefulness (Macaulay 1977, Milroy 1987, Romaine 1978, 1980).

3.4. Concluding remarks This analysis has shown that there is gender-related variation in the speech of London teenagers. The research project carried out at QMUL was intended to look at this variable and analyse it. When conducting a sociolinguistic interview one is faced with the problem of overcoming the observer’s paradox “to obtain the data most important for linguistic theory, we have to observe how people speak when they are not being observed” (Labov 1972c: 113), which can be done, especially when one is working with adolescents, as they do not always pay attention to the way they speak. The results obtained appear to debunk the myth that it is always the case that women use the prestige forms and men the vernacular ones. The vernacular is the style that we are interested in as it is “the style which is most regular in its structure and in its relation to the evolution of language, in which the minimum attention is paid to speech” (Labov 1972c: 112). The fact that girls preferred the alveolar nasal to the velar nasal indicates that they do not pay the same attention to speech as they used to. It is difficult to account for this variation, as there are so many factors which influence it. An elegant solution is suggested by Cheshire et al. (2011) who say that “it is likely that young people’s orientation towards Caribbean and African American youth culture, especially hip-hop and

Talkin’ or Talking? Speech of London Teenagers

105

rap, makes the use of ‘black’ features attractive” (Cheshire et al. 2011: 164). This is a valid explanation, as many of the interviewees said that they listen to hip hop and rap, some of them even dressed like rap singers. Another possible explanation is that girls are competing with boys, and this is connected to the concept of “coolness”. Being cool is fashionable nowadays, and this is reflected not only in the way they dress, but also in the way they speak. And since all the most important singers usually use the [n] variant, teenagers tend to use this form also.

CHAPTER FOUR PE VS PĂ: PHONOLOGICAL VARIATION IN THE SPEECH OF ROMANIAN ADOLESCENTS LIVING IN CONSTANğA

Introduction This chapter offers a variationist analysis of phonological variation in spoken Romanian. I start from the hypothesis that adolescents are language innovators, therefore the use of non-standard forms is likely to be found in their speech and I set out to investigate: (i) phonological variation in the speech of adolescents; (ii) gender-related differences regarding the use of the (pe) variable; (iii) age-related differences in the use of standard and non-standard language. Given the lack of such an empirical study on spoken Romanian, based on the results obtained from research projects carried out in English speaking communities, I hypothesize that: (a) The use of non-standard [pă] is consistent in the speech of adolescents; (b) There are gender-related differences regarding this linguistic variable; (c) Boys tend to use the vernacular form [pă] whereas the standard form [pe] is used by girls; (d) There might be intra-speaker variation regarding the use of (pe). The chapter starts with a discussion of the (pe) variable in early modern Romanian, to be used as a starting point in my analysis, and see whether this variable has undergone any linguistic change throughout the

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

107

centuries. Then I analyze the preposition pe in different linguistic contexts and provide relevant examples from the Romanian literature to show that there is also variation in written Romanian. So, I employ quantitative variationist methodology to identify and analyse sociolinguistic patterns, i.e. a correlation between a linguistic variable and external social factors like gender and age. This chapter reports on the findings of a research project the aim of which was to identify patterns of phonological variation in the speech of adolescents. The project was carried out in 2013 and 2014 at the Educational Center Theoretical High School in ConstanĠa (ECTH). The research methods used were the sociolinguistic interview and the ethnographic method (long-participant observation) in order to offer a more unified and cohesive account of language and gender variation. The results obtained are then compared with those found in the speech of adult speakers.

4.1. (pe) in Early Modern Romanian The oldest document written in Romanian46 is considered to be Scrisoarea lui Neacúu (Neacúu’s Letter) from Câmpulung, addressed to Johannes Benkner. It was established that the letter was written in 1521. In the 16th century documents started to be written in Romanian, the original texts being relatively short and consisting of letters and different types of documents (donations, wills, receipts, diplomatic documents, etc.). The religious texts are translated into Romanian, especially parts from the Bible (The Gospels, the Psalms). Between 1556 and 1581 Coresi and his apprentices started publishing religious texts in Romanian, in the city of Braúov.47 To these texts historical writings are added, written by 46 The birth of a language is never a temporary phenomenon, occurring at a given point in time, but a long-lasting one. It involves different stages of evolution and their succession. The same thing happened with Romanian, which is defined as being “the Latin language spoken uninterruptedly [...] from the moment Latin started being used in these provinces to present day” (Rosetti 1968: 77). Common Romanian is a fundamental stage in the development of the history of our language (Cătănescu 1996: 72). Common Romanian was “the language spoken by the ancestors of Daco-Romanians, Aromanians (also called Macedo-Romanians), Megleno-Romanians and Istro-Romanians, before any connection between them was interrupted” (Puúcariu 1974: 58). 47 These texts include: Catehism (Întrebare Creútinească, 1559/1560), Pravila (c. 1560-1562), Apostolul (1563), Cazania I and Molitvenicul (c. 1567), Psaltire and Liturghier (1570?), Psaltirea slavo-română (1576-1578), Psaltirea slavo-română (1577), Cazania II (1581), as presented in Cătănescu (1996: 93).

108

Chapter Four

Moldavian chroniclers (Grigore Ureche, Miron Costin, Dimitrie Cantemir). It is worth looking at the texts mentioned above, as insights into how the preposition pe evolved throughout relevant quotations from these texts to see the contexts preposition was used as well as its form.

Ion Neculce, they provide time. I offer in which the

(42) ùi au ales 4 evangheliúti den limba ovreiască pre limba grecească, de-au scris Evanghelia. (Coresi – Catehism 1559/1560) ‘And they chose four gospellers from the Jewish language the Greek language to write the Gospel’. (43) ùi voiu lăsa pre voi limbi pogîne úi vor vărsa sîngele vostru. (Legenda Duminicii, în Manuscrisul de la Ieud, 1621-1633, p. 153) ‘And I shall let to you pagan languages and they will spill your blood’. (44) Toate cîte-s, pre tine? (Miron Costin - ViiaĠa Lumii, 1671-1673, pp. 162-164) ‘How many are there on you?’ (45) întriabă pe ce vremi am scris úi cît am scris [...] eram pre atuncea la úcoală la Bar, în Podoliia, pre cale fiind de la sat la oraú.” (Miron Costin – LetopiseĠul ğărâi Moldovei, 1675, pp. 128-130) ‘ask in what times I wrote and how much I wrote [...] I was in school back then in Bar, in Podoliia, on the way from the village to the city’. (46) va a păzi pe un cetăĠean úi pământean [...] găsindu-úi pe un neam ce-i zicea alani (Stolnicul Constantin Cantacuzino – Istoriia ğării Rumâneúti sfârúitul secolului al XVII-lea – începutul secolului al XVIII-lea, pp. 176-178) ‘shall guard a citizen and an earth dweller [...] finding a people called alani’. (47) cununi de dăruit úi pe cela ce fără îndoinĠă (Dimitrie Cantemir – Divanul, 1698, pp. 35-36) ‘crowns to give and that person without doubt’. (48) din toĠi boierii Ġărîi mai de treabă la voroavi pe Miron Costîn, care au fostu mai pe urmă úi logofăt mare. (Ion Neculce – LetopiseĠul ğării Moldovei, 1733-1743, pp. 45-46) ‘from all the country’s boyars who were more kind spoke about Miron Costîn, who was also a great logothete’ (Source: Cătănescu 1996: 133-193)

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

109

The examples given above (42–48) provide information about the use of the preposition pe in the Romanian used in the 16th, 17th and 18th centuries. In the early texts the form of the preposition is pre, as shown in the examples (42)–(45). In the 17th century, there was variation between pre and pe, which appears in e.g. Miron Costin’s chronicle, LetopiseĠul ğărâi Moldovei. Towards the end of the 17th century and the beginning of the 18th century the form pe was used almost exclusively in written Romanian. The period between 1780 and 1840 represents a very important stage in the development of the Romanian language and literature. This stage is also known as the premodern period, and it is associated with the writings of the Văcăreúti Poets. IenăchiĠă Văcărescu is also the author of one of the first grammars of the Romanian language, entitled ObsevaĠii sau băgări de seamă asupra regulelor úi orânduelelor gramaticii româneúti [=Observations or notes on the rules and norms of the Romanian grammar] published in 1787. It is also in this period that we find the first attestations of the variant (pă), in the poems of Alecu Văcărescu (1767– 1799), IenăchiĠă Văcărescu’s eldest son: (49) În ei cată să te vezi/ Întocma pă cât luminezi ‘Try to look in them/Exactly how much you glow’. (50) ùi cusur să nu rămâie/ Pă cât simte să-l mângâie ‘And in order not to be a flaw/ To caress him as she feels like’. (Source: Boldea 2009) The examples provided in (49) and (50) demonstrate that the (pă) variant has existed for four centuries and it is still being used in present day Romanian. Another interesting example is found in the poems of Nicolae Văcărescu (1786–1825), IenăchiĠă’s second son: (51) Că-l Ġineam tot pe cărare, / Pe bere úi pe mâncare / Pre potecă făr’ de soare Pă-l cu inima vitează...” ‘I was keeping him on track/ On beer and food / On the road without sun/ The one with a brave heart’ (Source: Boldea 2009) It seems that three variants of the (pe) variable were used simultaneously without any difference in meaning. There are no constraints which might indicate the use of one variant instead of another. A thorough discussion of the (pe) variable in early modern Romanian is beyond the scope of this chapter. I have only given a very broad-brush picture of the (pe) variable

Chapter Four

110

in the first and most significant written documents in Romanian to be used as a starting point in my analysis.

4.2. (pe) in modern Romanian The preposition pe appears in different syntactic contexts and has two special uses: classical and special (Mardale 2006). For the classical usage we have the following examples: (52) a. b. (53)

Maria a pus cartea pe birou. ‘Mary put the book on the desk’. Ion a venit din tabără pe 15 iulie. ‘John came from camp on 15 of July’. Diana contează pe tine. ‘Diana is counting on you’.

For the special usage we have examples such as: (54) a. b.

Băiatul l-a cunoscut pe director. ‘The boy met the principal’. L-au botezat pe el. ‘They baptized him’.

The preposition pe is used in Romanian to mark the accusative case, as in the examples above. In (52a), for example, pe birou ‘on the desk’ has the syntactic function of direct object. Since I am interested in the variants of (pe), I considered it relevant to provide some examples from the Romanian literature, to offer a better understanding of this phenomenon. There is variation between (pe) and (pă) in the 19th century Romanian literary masterpieces. Consider the following example: (55) Dar când să ies de pă maidan cu dumnealui, tocmai trecea jupînul de la vale; am lăsat pe persoana în chestie să m-aútepte pe maidan úi eu am sărit ‘But just as I wanted to get off the street with him, the master was passing; I left the person in question to wait for me and I jumped’ (I.L.Caragiale – O noapte furtunoasă, 1983: 35) In (55) there is intra-speaker variation, as the two variants of the variable (pe) are used in the same context. The speaker first uses the [pă] variant

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

111

and then the [pe] variant. This is relevant as it shows once again that there are no constraints which govern the use of the (pe) variable. In Caragiale’s play, the male character, Spiridon, is a poor boy pertaining to the lower class and his employer is a wealthy man. He is the only character in the play to use the form [pă] which indicates that this variant is associated with the lower class and lacks prestige. Another interesting example comes from the story Alexandru Lăpuúneanu written by Costache Negruzzi and published in 1840 in Dacia Literară: (56) De mă voi scula pre mulĠi am să popesc úi eu. ‘If I rise I shall punish many myself’. The use of the variant (pre) in (56) is quite interesting and intriguing at the same time as Alexandru Lăpuúneanu is one of the only 19th century texts in which I found this variant. Towards the 20th century, the only variant which occurs in literary writings is (pe). [Pă]48 is only used to indicate social class, and considered to be a popular form of (pe). In example (55) the vernacular form [pă] is used by a male, thus becoming a marker of masculinity.

4.3. Data collection and methodology 4.3.1. The sociolinguistic interviews The data used in this chapter emerged from research on the social and linguistic practices of students at the Educational Center Theoretical High School (ECTH), a large, culturally diverse high school located in the city of ConstanĠa. Being a teacher at this high school, it was easy to get access to my speakers and befriend them. I spent all my breaks outside with them earning their trust and observing them on the playground. I also had a discussion with the principal and asked for permission to conduct the interviews at school. She turned out to be very sensitive to my research and gave me free access to all the school’s facilities and talked to some of the parents about my research. All the students were told about the project and those who wanted to participate in the research project volunteered, some out of curiosity, others just because they wanted to help. They also received a form with a few details about the research project and they were 48

The form (pă) appears in many Romanian traditional songs, among which we mention Liviu Vasilica’s song Pă la noi pă Teleorman or Andreea Voica’s song Argeleana pă picior.

112

Chapter Four

asked whether they agree to have the interviews audio-recorded. A form was also sent to their parents, as all the students were minors, and they signed and returned them to me. Another important thing about the school where I conducted my research is that it has the following structure: kindergarten, primary school, secondary school and high school. This offered a wonderful opportunity to include students both from secondary school as well as high school, thus also taking the social variable of age into consideration, beside gender. The sociolinguistic interview followed the Labovian structure and included two components: the questionnaire and a reading text. The aim was to obtain two different types of speech styles: casual and careful. Tagliamonte (2006: 39) argues that the ideal way to start a sociolinguistic interview is to begin with questions about demographics, community, neighbourhood, etc. and then move on to more personal questions. This is exactly how I structured my questionnaire. I asked the students to tell me about the place where they were born, to describe the city they live in and to tell me whether they like the city as well as the country they live in. Even though I had a standard set of questions, in many cases I used alternative questions as well, to yield maximal talk from the interviewees. I also brought sweets with me to the recording sessions as I tried to make them feel relaxed by creating a casual and friendly atmosphere. I avoided conducting the interviews in classrooms, because they were considered a formal place. Instead, I used the cafeteria when there was nobody there or the playground, as they were more informal. The recording procedures are the same as the ones mentioned in Chapter 3. The interview contained 22 questions, and each interview lasted between 20 to 25 minutes. Twelve adolescents participated in this project (six boys and six girls), giving us approximately five hours of speech. After I devised the questionnaire I piloted it with two students to see if the questions are well formulated and whether the students gave long and complex answers, as the purpose was to elicit as much data as possible. I also avoided yes/no questions, and instead I used questions like Tell me about the place where you were born. The most successful questions were the ones in which they had to describe their best friend and recount experiences from their childhood.

4.3.2. Adolescents as a source of data Working with adolescents is a wonderful experience, but can soon turn into a daunting task, because of their moody personality. It is traditional to view adolescents in our society as sloppy (they leave their clothes on the floor), rebellious (they don’t do what they’re told), and

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

113

irresponsible (they forget their pencils). This view of adolescents is visited on their language, which is judged sloppy in its imprecision, rebellious in its supposed use of slang and profanity, and irresponsible in its greater use of non-standard grammar (Eckert 2004: 362).

In his 1965 study, Labov develops a model for the acquisition of Standard English, which portrays the peer group as having a powerful influence on speakers between the ages of 5 to 12, at which time it is seen as an important influence on speech. His data confirm the fact that after the age of 14 adolescents in New York City become aware of the social significance of different varieties of English. His conclusion is that the speech of adolescents is less prone to style-shifting and there is a greater chance of obtaining recordings of vernacular speech if the speech of adolescents is used for analysis as well as a source of data. Another important advantage of studying the speech of adolescents is provided by Cheshire (1982: 9) who notes that it is easy for an outsider to become part of the group. Given that the research project was carried out at a school, I did not have problems in becoming part of their group. I also taught English and Spanish to some of the students interviewed, but, even though I was very serious and strict in the classroom, during the breaks and in the cafeteria or on the playground I was very relaxed, I joked with them and made them feel comfortable in my presence. By adopting this method outside the classes, I managed to earn their trust and they talked freely when I was present. This was the only way I could have access to the vernacular, by using the long-term participant-observation method (Cheshire 1982, Eckert 1989, 2000, Tagliamonte 2006, 2011, Schilling 2013). There are also some disadvantages of a study that is confined to adolescents. First of all, the analysis will only refer to the language used by a small group in society and this might present a limited view of language, as stressed out by Cheshire (1982: 10). Secondly, a study which covers only one age group can only speculate on the role of language variation and yield a distorted view on language. In order to overcome this problem and present a more unified and cohesive analysis of phonological variation in Romanian, I also included a brief analysis of the (pe) variable in the speech of adults. Due to the lack of time, I did not interview adults. Instead, I recorded some talk-shows from different TV channels, and I used them to complete my analysis of the (pe) variable in Romanian. Section 4.5 focuses entirely on adults and also presents the results in contrast with those that I found for adolescents.

Chapter Four

114

4.3.3. The speakers and the school The speakers that participated in this project are studying at the Educational Center Theoretical High School (ECTH) in ConstanĠa. They are all native speakers and come from Romanian native speaker families. They belong to one age group 13–17 and they are grades 6–11, as revealed in Table 4.22. Name

Age

Alexandra Amalia Andreea Mara Ionela Teodora Bogdan Cosmin Mario Daniel Tibi Vlad

14 16 14 15 13 17 14 15 13 14 17 15

Grade 8 10 8 9 6 11 8 9 6 8 11 9

Lives in ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa

Table 4.22. Information about the Romanian speakers The students all come from good families with a good socioeconomic status. I chose to conduct my study in ConstanĠa first of all because it is the city where I live and work, but above all because it is a multicultural city. The interviewees include Orthodox Christians and Catholics as well as Muslims. ConstanĠa is the biggest city in Dobrogea and it is situated on the Black Sea coast. In ancient times it was known as Tomis and it was and still is one of the most important cities in Romania due to its geographical location as well as resources.

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

115

Map 2. Map of Romania

I started my research immediately as I was employed by the school in April 2013. It took me a while to get acquainted with the staff and the students, but once I achieved that, things became easier. My aim in this work was to be relatively low-keyed and informal and convince the students to take part in the research. The school has a dress code, and students and staff must wear uniform at all time. The uniform consists of a blue costume, white shirt and red tie for the male teachers and blue skirt, blue jacket, white or blue shirt and red scarf for the female teachers. The students’ uniform is grey and includes: grey trousers for the boys and grey skirts for the girls; a white shirt and grey vest. The school’s policy is that students must wear the uniform everyday to school, and if they do not wear the uniform they are not allowed in the school. The girls resented wearing the skirt over trousers or jeans, and they usually did not wear it. They used to keep it in their schoolbag or their desk and wore it only when there was an inspection. The uniform is not obligatory for high school students, so nobody wears it. I will return to this later on. The school contains two buildings. The main building is where primary and secondary school students learn, and it is also the place where the offices of the school’s staff are situated. The new building (as it started being used in September 2013) has five floors plus the basement where the gym is situated. The first and second floor is where the kindergarten in

116

Chapter Four

located. The third floor is known as the playground, where kindergarten and primary school students go to play one hour each day. The fourth floor is where the cafeteria is and students eat their lunch there from 13:00 to 13:50. The last floor is where the high school students study (grades 9-12). The school also has a library and a Cambridge Examination Room for Cambridge exams. The class schedule outlines a routine for each student, organising encounters with other students only during breaks and lunch time, as they do not have to move from one class to another. Each grade has its own class where all the lessons take place, with the exception of Biology, Physical Education, Physics, Chemistry and Computer, which have special laboratories. The school day starts at 8:30 and is divided into 50-minute classes from 8:30 to 10:10, then they have a 20-minute break. Lessons start again at 10:30 until 13:00. From 13:00 until 13:50 they go to the cafeteria to eat their lunch and after that they play outside. From 13:50 until 14:40 they have lessons and after that they can either go home or stay at school until 16:30 to do their homework. The administration also imposed a “closed campus” policy, forcing students to remain in the school building during school hours except for authorised trips on different occasions. Even though students are not allowed to leave the building during lunch, they are allowed to gather in the courtyard and play football, chat, run, etc. The school is video-monitored 24 hours and it also has 3 guardians in different places of the school. The teachers are also on duty during breaks, either in the school or outside on the courtyard, maintaining the order and making sure that students do not hurt each other or fight. Eckert (1989: 62) astutely outlines several key issues in realising a thorough analysis of the speech of adolescents. She stresses out the idea that clothing is a powerful social marker and it signals economic status as well as specific group identity. The main problem is that this was something I could not take into consideration as the school uniform was mandatory. Luckily, on a few occasions, the school organised different extra-curricular activities which included a trip to Tropaeum Traiani, a trip to Bucharest to visit the Antipa Museum and a one week camp to Predeal. In these trips the school uniform was not mandatory so nobody wore it. This presented a very good opportunity to see them casually dressed, and behaving differently. Dressing style is also a marker of social identity and it also indicated the affiliation to a group. Following Eckert’s (1989) groundbreaking study of the English used by jocks and burnouts at Belten High, I also grouped the students from ECTH into two groups: geeks and pops (short for popular). The factors

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

117

that I took into consideration when I divided them into these two groups were the language they used, behaviour and clothing. In the Oxford English Dictionary online, a geek is defined as an unfashionable or socially inept person. At ECTH this category includes the students who are more reclusive, who are interested only in obtaining high grades and are very industrious. They often criticize the pops and the way they speak or behave. Most of them do not go outside during breaks, but stay inside the school and revise for the next class or just sit at the window and speak badly about their colleagues. Some of them even wander the corridors copybook in hand pretending to study when they are actually trying to hear the latest gossip. Those who do go outside during breaks have a special place in the school’s courtyard where they gather. The term pops, on the other hand, describes the students who are interested in the way they are perceived by their peers, are very popular in school and fashion is everything to them. Cool girls use slang words, as well as the vernacular forms used by boys as they imitate boys’ linguistic style just to be considered their equals and members of their group. Language is used as a power tool by adolescents to be accepted in a particular group. They pay attention to the way they dress this being also valid for cool boys. Pops, particularly high school girls, wear a good number of outfits and consciously avoid using the same item of clothing twice a week. This is in contrast to geek girls who tend to wear the same outfit two or three days in a row. The majority of cool girls shop in groups in the most important shopping malls in ConstanĠa and some of them order their clothes from different American websites just to make sure that their clothes are unique. Many pops’ enhanced socioeconomic status is reflected in their passion to follow the latest fashion and their desire to wear designer clothes or very expensive clothes designed by important fashion houses like Gucci, Calvin Klein, Zara, Prada, etc. From a cultural point of view, geek girls and cool girls study at the same school, live in the same neighbourhood, attend the same courses, but their gender styles are completely different, as well as their linguistic ones. Nerdiness is not just about avoiding coolness but about resisting gender hegemony. It allows girls to opt out of the heterosexual market altogether or to enter it when they feel ready and on their own terms. Nerd girls remove themselves from coolness and its attendant gender obligations as much through their untrendy vowels as through their deliberately unsexy, unsophisticated clothing. The popular ideological association of advanced variants of these vowels with airheaded beach bimbos may have something to do with this aversion as well (Bucholtz 2002: 39).

118

Chapter Four

The school can be seen as a big speech community, in which there are two communities of practice (Lave and Wenger 1991). Eckert (2003: 112) argues that modern education has somehow insulated adolescents from adult society and eliminated them from the workforce, thus confining them to age-homogenous institutions (Coleman et al. 1974). By doing so, adolescents created a world of their own, being responsible for their social and linguistic practices.

4.3.4. Long-term participant observation In Chapter 2 we have seen that in conducting a successful variationist study, the sociolinguistic interview is a key factor. Another angle from which sociolinguists study smaller groups is in terms of their adherence to a social group. In this section further consideration is given to another important method used by sociolinguists, which I also used, namely longterm participant observation. I used this method in order to overcome the Observer’s Paradox discussed in Chapter 2. In the sociolinguistic interviews the students, albeit relaxed, strove to sound as intelligent as possible, thus using the standard forms to the vernacular ones. They used the vernacular in the courtyard, when talking to their peers where attention to the language they used was at a minimum. This mode of inquiry lends a fresh perspective to my research goals. As an insider I gained access to the vernacular forms they used, slang and swear words included, while preserving an outsider detachment through long-term involvement. So, beside my role as a researcher, I also became a participant in their activities. Richards (2003) argues that ethnographers should focus on four important areas of social interaction: (i) the physical setting of events; (ii) the systems and procedures that are followed at these events; (iii) the people taking part in these events; (iv) the social practices (language included) that are noticed at these events. Being a scientific research, the aim of observation is to identify systematic patterns of language that can be correlated to external factors (Levon 2013: 75).

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

119

Tagliamonte (2006: 20) is an advocate of the participant observation49 method as it is “fundamental to variation analysis”. The ethnographic approach “consists of the intensive involvement of the researcher in a given social setting in order to describe and identify, through the use of a variety of complementary research techniques, the cultural patterns and regularities that structure and perpetuate a society” (Poplack 1979: 60). To put it differently, the long-term participant observation approach aids the sociolinguist to understand the cultural context of the speech community he is investigating. Knowing the cultural context can provide insight into what is important to analyse. One of the most famous participant observation study is Eckert’s (2000) carried out in a Detroit high school. The valuable data that come from such studies give the researcher extremely valuable information regarding some of the most important questions in the study of language variation (Tagliamonte 2006: 21). Immersion in the speech community under study is vital in order to come up with a valid explanation for the linguistic choices of that community. The main goal of this chapter is to offer a detailed analysis of the (pe) variable in the speech of Romanian adolescents living in ConstanĠa. In providing an empirical study of this variable, I immersed myself in the community studied, but I never lost my objective ability to analyse it. Outside in the courtyard I never used an audio recorder, because I would have scared the students away and because of the technical difficulties that could have presented. Instead, I always carried a notebook with me and I wrote everything that I observed in it.

4.4. (pe) patterns in the speech of Romanian adolescents living in ConstanĠa The audio interviews were transcribed in word, resulting in a corpus of 15,000 words. I counted all the tokens of (pe) and the total number is 113. The independent variables taken into account for the analysis of the dependant variable (pe) are speech style (casual speech and reading style) and the gender of the speaker. Another important fact is that the (pe) 49

Participant observation has become widely used by researchers conducting an ethnographic study, but also by sociolinguists carrying out quantitative or qualitative research. In her analysis of the lives of college students, Cathy Small (Nathan 2006) moved for a year to a college dorm and frequented university classes. Another example is found in MacLeod (1987) who got involved in a community youth project as he was researching the aspirations of young people in a poor neighbourhood. In these cases, participant observation requires “close, longterm contact with the people under study” (Fetterman 1989: 47).

Chapter Four

120

variable is a stable sociolinguistic variable and has not been subject to any type of change for at least two centuries.

4.4.1. Casual speech: results As far as the analysis of this variable is concerned, unclear tokens such as the reduced form p-ormă instead of “pe urmă” were excluded from the analysis. More tokens of (pe) were found in the speech of boys than in that of girls, as revealed below in Figure 4.17. In the analysis of (pe) I used ELAN, as in the previous chapter. I chose not to use PRAAT as this variable can easily be analysed auditorily, but I also checked the tokens with three native speakers. 60 50 40 30

Pe

20

Pă

10 0 Boys

Girls

Figure 4.17. Variants [pe] and [pă] according to the gender of the speaker

The results obtained were surprising as the non-standard form (pă) was not preferred almost at all by both genders. They used the standard variant in 95% of the cases. On a closer look, I noticed that the [pe] variant was used exclusively by 6 graders as well as by some of the high school girls. In the sociolinguistic interviews, with a few exceptions, I could not find any gender-related differences, regarding the use of (pe). I present the variants for each speaker in Figure 4.18.

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

121

25 20 15 10

Pe Pă

5 0

Figure 4.18. The use of [pe] and [pă] in the speech of adolescents from ConstanĠa

The [pe] variant was used more consistently by Amalia and Vlad while [pă] was preferred by Cosmin, although he used the [pe] variant more often. It appears that the standard variant is preferred by both secondary school as well as high school students. The Observer’s Paradox did not influence the results as the students were relaxed, they recounted many episodes from their childhood and laughed while telling them. The fact that school and schooling is central to adolescents makes the dichotomy standard-vernacular language ground material for their linguistic choices (Eckert 2003: 113). School is by definition associated with education, standard language, homogeneity, thus explaining the students’ choice for the standard language. Nonstandard language is associated with toughness, the working class, anti-rules, local innovation and gangs. A difference that I discovered is related to age. If in secondary school the [pe] variant was almost exclusively used by the students I interviewed, the situation was somehow different in high school, where some of the boys also used the [pă] variant. Let us consider the following examples taken from the interviews:

122

Chapter Four

(57) a. úi io-l vedeam pă el úi-l luam în braĠe (Cosmin) ‘and I saw him and hugged him’ b. Da eram mergeam pe stradă úi eu mergeam pe trotuar cu niúte prieteni úi pretenu’ meu a făcut o glumă m-a împins m-am dus pă stradă... (Cosmin) ‘I was walking down the street I was walking on the sidewalk with some friends and my friend made a joke and pushed me I went in the street’ c. dau drumu’ la muzică mănânc pă urmă iară dau drumu (Tibi) ‘I turn on the music I eat then I turn on again’ d. úi am decis să ne întoarcem pă jos... (Vlad) ‘and we decided to return on foot’ How can we account for this type of variation? In high school students start to rebel, they become more independent, this being reflected not only in the way they behave and dress, but also in the way they speak. I also checked in Hornoiu’s (2007) corpus of spoken Romanian for this type of variation and I could not find any use of the [pă] variant in the speech of adolescents. [Pă] however was used more consistently by adults. I will address this issue in section 4.5. The nonstandard and the standard variants of (pe) co-occur variably. This variation did not, however, appear to be governed by any linguistic constraints. A possible explanation is that, being at school, students tend to be corrected all the time by their teachers (seen as a figure of authority), whenever they talk “incorrectly”50. So, they strive to speak as correctly as possible and it is at this stage that their speech (i.e. the way they speak) is influenced by those around them. Another possible explanation which might account for the lack of variation is that secondary school students are raised by their mothers. They spend a significant amount of time with their mothers as they are the ones who nurture them. Therefore, it is actually expected that they use the language their mothers use and the preference for the standard form should not come as a surprise51. One of the most frequently repeated claims about male and female speech is that women’s speech is more conservative than that of men’s 50

This situation is reminiscent of George Bernard Shaw’s play Pygmalion, where Higgins is trying to teach young Eliza to speak like a lady. 51 This situation is also encountered in traditional societies. Bradley (2006) reported that in Yanyuwa men speak one language and women another language, as mentioned in Chapter 1. Children are nurtured by their mothers and grandmothers, thus in a predominantly female atmosphere, so the language the children use, irrespective of their sex, is that of their mothers.

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

123

(Coates 2006: 66). The patterns of phonological variation are somehow heterogeneous. Women tend to use the prestige forms whereas men prefer the vernacular ones. This hypothesis has been verified for English (Labov 1966, Trudgill 1974, Lakoff 1975, Holmes 1995, Coates 2004, Eisikovits 2006). It appears that in Romanian society, just like in British, American and Australian societies, women are viewed as using the standard forms of language whereas men have a preference for the vernacular ones. Another important factor in people’s use of variation stems from their relation to linguistic markets. The concept of symbolic market was coined by Bourdieu and Boltanski (1975) and it was borrowed and adapted to the study of variation by Sankoff and Laberge (1978). From this perspective, males and females belong to different linguistic markets. Trudgill’s (1972) analysis of gender variation in the English spoken in Norwich found that women were more conservative in their use of almost all variables that he studied. One of his speculations was that this was the result of women’s exclusion from the workplace. The majority of students’ mothers from ECTH are housewives and those who do work have white-collar jobs, the kind of jobs referred to as “technicians of language” by Sankoff et al. (1989). This notion encompasses jobs like academics, directors, secretaries, etc. This explains the preference for standard language, something that children acquire from their mothers. [Pă] is a marker of masculinity not only in the speech of men, but also in the speech of high school adolescents. It has low prestige, thus being associated with the working class. Students used it more in same-gender conversations and they tended to use the standard form in mixed-gender conversations. Their speech style joins with other aspects of style which include clothing (Eckert 1980, 2000), makeup (Mendoza-Denton 1996), musical taste, territory, interest in the same activities (Eckert 1989). Standard language is a very important asset in gaining membership or “a powerful tool of membership in the halls and homes of global power” (Coates 2006: 369). In contrast, the vernacular is a flavour of local or regional differences, something acquired in the neighbourhood. Those who have strong ties to the neighbourhood where they grew up are prone to use or embrace this form of language, as an expression of their local identity and affiliation to a certain group. The high school is seen as a globalizing institution dominating the life of adolescents (Coates 2006: 369) and the adolescents’ use of the standard or the vernacular forms of language is influenced by the way they view that institution.

124

Chapter Four

4.4.2. Reading style: results For the reading style the interviewees were supposed to read a text of half a page out loud. Although this speech style is an artificial one, I included it to elicit more types of speech and see where variation occurred. As expected, this being a more formal situation, the students paid attention to what they read and there was no variation whatsoever as far as the (pe) variable is concerned. They all used the standard form irrespective of gender or age. This gives us valuable insight into their sociolinguistic competence as it proves that they are aware of the language they use and how they use it. They become aware not only of the fact that there are differences in the way they speak but also of the fact that some forms are “considered bad or improper while others are correct and those are the one that they must use”, as some of them confessed. School exerts a great influence on the language they use. They come in contact with the written as well as with the spoken word. Nonstandard forms are not tolerated in school and they are constantly reminded to use proper and standard language in all situations. In secondary school they tend to obey their teachers and strive to sound as intelligent and as correct as possible. In the next section I discuss the findings from the playground where the students were not recorded, only observed, and compare the results with the ones from the sociolinguistic interviews.

4.4.3. The playground: results In this section I focus on the data that I gathered as a participant-observer and account for the variation that occurred. At this stage it is important to mention that using this method (i.e. participant-observation) it was really hard to follow the speakers that participated in the interviews. The focus was on them and also on the group of students they were friends with. Some of them belonged to the geek group mentioned above others were seen as pops. In the playground (the school’s courtyard) students got involved in different activities. Some of them were playing football, others were playing hide and seek, while others were just talking about different things. Some of the high school students used to gather in a special place and play truth or dare. The following examples are taken from my field notes. The examples are from the speech of the same students as I interviewed to facilitate the comparison between these findings and the ones from the interview.

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

125

(58) a. Să vii mâine pă la mine p-acasă (Alexandra) ‘Come tomorrow to my house’ b. Du-te pă la magazine bă (Cosmin) ‘Go to the shops, dude’ c. de Revelion când pusese mama pă masă mâncare (Tibi) ‘on New Year’s Eve when my mother had put food on the table’ d. Cheam-o fată pă Teo să vină afară (Amalia) ‘Girl tell/call Teo to come outside’ It is noteworthy that in the playground the secondary school students, except for Alexandra, did not use the [pă] variant at all. Could this be an indication that they are more status conscious than high school boys? Is there a correlation with age? There are no differences which could be accounted for in terms of age grading, as the (pe) variable is a stable sociolinguistic variable, but it does seems that with age there is an increase in the use of it. High school boys used it quite frequently especially when they were playing sports or when they were swearing. Forms like pă mă-ta instead of pe mă-ta were extremely used by boys. Girls, on the other hand, used to imitate boys as if this appeared to be the rule of the group. 70 60 50

[pe] pops

40

[pă] pops

30

[pe] geeks

20

[pă] geeks

10 0 boys

girls

Figure 4.19. The use of (pe) according to the gender of the speaker and group affiliation in the playground

Intra-speaker variation occurred in the case of some speakers, but I could not identify any linguistic constraints that might explain this phenomenon. What is worth commenting upon is that girls only used the [pă] form in certain contexts. For example, when they were outside in the playground

Chapter Four

126

and the boys from their group were around they used the non-standard form more consistently. In same gender gatherings they used the standard form. This did not occur with boys. In their case the usage of the vernacular form was constant at all times. Some of the high school boys even had nicknames (Cosmin’s nickname was Adonis, Tibi’s nickname was Basu while Vlad was called Lache by his friends)52. Interestingly enough, even some of the girls had nicknames (Andreea’s nickname was Puca while Alexandra’s was Bubu). They insisted on being called like this, as if these nicknames offered them a new identity. Andreea and Alexandra are part of the pops group. One can easily tell because they are very noisy, they would do anything to attract attention, they constantly use non-standard forms and this can also be seen in the interviews as they used the [pă] form two or three times. Alexandra used the (pe) variable five times in sociolinguistic interview. She used the standard form three times and the non-standard form twice. As mentioned earlier in this chapter, the concept of coolness is associated with the use of non-standard [pă]. Following Cheshire (1982) I also used a toughness scale to see whether it can be correlated with the use of nonstandard linguistic features. In the sociolinguistic interview I included the question Do you practice any sports? as I already knew that the majority of students take up different sports and participate in different competitions. Besides tennis, swimming and football, some students, like Cosmin for example, practice kickbox all style. Consider the following answer from the sociolinguistic interview: (59) Interviewer: Cosmin:

Practici vreun sport? ‘Do you practice any sports?’ Da practic kickbox all style úi înotu’ ‘Yes, I practice kickbox and swimming’

Cosmin was one of the students who used the [pă] variant consistently in the playground. He is also considered to be one of the school’s bullies as he teases his colleagues as well as the secondary school students. Other students like Mario, Bogdan or Daniel were given a zero score on the toughness scale while Cosmin is an 8 out of 10. Cosmin is a skilled fighter, often demonstrating his fighting abilities in the playground. I consider toughness to be an indicator of adherence to the vernacular 52

This is reminiscent of Labov’s (1973) Harlem study where the groups he studied had names like ‘the Jets’ and ‘the Cobras’. Another interesting fact is that they had a hierarchical structure with leaders, things that usually define a gang and not a group. At ECTH I did not encounter this situation.

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

127

culture53. There is gender differentiation here as none of the girls took part in such activities, although they were drawn to them. They admired the skilled fighters and mocked or teased the ones that were not good at all. Analysing the use of AAVE features by white kids in Northern California, Bucholtz (1999) interprets this as a way of laying claim to coolness. In a study of the development of English among a group of adolescents living in Northern California it was proven conclusively and through meticulous analysis by Kuwahara (1998) that the students who were interested in school activities developed standard English while in the speech of those who pertained to the street culture the AAVE features were more predominant. In my analysis the high school geeks represent a school-oriented culture. They are planning to attend university, they get involved in many extra-curricular activities, hence the preference for the standard form. The other group, on the other hand, the high school pops, reject school as their social base. They rely on the fact that their parents have money and they do not need to study or work really hard, instead they can enjoy their life and do whatever they like. I considered the school a big speech community where I distinguished two communities of practice: geeks and pops. The members of each community of practice align themselves with one another because they believe that they have shared interests with one another. The geeks did not engage in any type of violence (another marker of masculinity) be it physical or verbal. They were the ones who were interested in school activities, participated in different competitions and used almost exclusively the standard forms. At times they used the non-standard forms with the sole purpose of belittling the pops. They established an excellent relationship among them based on mutual respect and friendship. The pops on the other hand were much more aggressive and vocal. An early psychological study on gender differences carried out by Maccoby and Jacklin (1974) suggested that aggression made a clear cut distinction between males and females. In a more recent paper, Bjorkqvist et al. (1992: 55) focusing on Finnish children, found that although boys are more physically aggressive than girls, boys and girls are very much alike in the use of verbal aggression. At ECTH aggression was the clearest way in which the two groups were differentiated. In the pops group verbal aggression was not a gender marker as girls were as verbally aggressive as boys. The only difference lies in that girls do not use as many swear words as boys and they are not as physically aggressive as them. Girls, however, 53 Cheshire (1982: 97) notes that “adherence to the vernacular culture should be seen as a temporary phase in the lives of the members of the peer groups”.

128

Chapter Four

often criticised their peers behind their back rather than in a face-to-face encounter. To conclude, the [pă] variant occurred more frequently in the playground than in the sociolinguistic interview, and it was used more by boys belonging to the pops group. When students are in school they use fewer nonstandard forms and more of the corresponding prestige form, i.e. the [pe] variant. There are also other linguistic variables that were used by boys and not by girls and they occurred both in the sociolinguistic interviews as well as in the playground. I will discuss them in the last part of this chapter. All in all, the long-term participant observation method revealed that in different gatherings the girls and the boys use the [pă] variant more frequently.

4.5. (pe) variation in the speech of Romanian adults Given the results that I obtained from examining adolescent speech, I set out to investigate the use of (pe) in the speech of Romanian adults. I kept hearing people on TV, on the bus, in the street using the [pă] variant so I wanted to compare the two different types of speech: adolescent versus adult and see the results. The analysis is based on eight hours of conversations gathered from TV talk shows.

4.5.1. Data collection and methodology The data collected come from different Romanian TV talk shows, recorded between 2013 and 2014. It consists of eight hours of spoken interactions between the host of the respective talk show and their guests. Due to the lack of time, I did not transcribe the conversations from the talk shows into word. I used ELAN54 to analyse the conversations and identify the tokens of (pe). In what follows I present first the talk shows that I recorded and the people who engaged in conversation and then I give the number of (pe) tokens for each person.

54

The software is available here http://tla.mpi.nl/tools/tla-tools/elan/.

PE vs PĂ: Romanian Adolescents Living in ConstanĠa Talk show

TV Channel

Person recorded

În Gura Presei

Antena 3

Mircea Badea

50 de minute cu Gabriel Liiceanu úi Andrei Pleúu Gândul Live

TVR Cultural

Andrei Pleúu

Gândul Live

Happy Hour

PRO TV

Cristian T Popescu Corneliu V Tudor

Lumea lui Banciu

B1 TV

Radu Banciu

Teo Show

Kanal D

Teo Trandafir

Un Show Păcătos

Antena 1

Oana Zăvoranu

Acces Direct

Antena 1

Simona Gherghe

Mireasă pt fiul meu

Antena 1

Mirela B. Vaida

Cireaúa de pe tort

Prima TV

Simona Florescu

Total

129

(pe) tokens Time 52 36:16 20 51:47 29 25:58 23 33:54 30 24:04 22 60:46 12 23:07 36 1:48:05 19 52:20 13 44:16 256

Table 4.23. Information about data collection In total there are 256 tokens of (pe) to be analysed. I omitted from the analysis unclear tokens like p-ormă or other similar contractions. The reason I chose talk shows is that the atmosphere is relaxed; the host as well as the guests speak casually even though they are on TV. At times there were heated discussions among the guests and in such situations they forget that they are on TV. Some of them even used swear words, given the nature of the shows. Although the data was not collected following an interview, I think that we can consider the language used in the aforementioned talk shows as casual.

130

Chapter Four

50 40 30

[pe]

20

[pă]

10 0 MB AP CTP CVT RB

TT

OZ SG MBV SF

Figure 4.20. The distribution of [pe] and [pă] for each speaker

In Figure 4.20 I used the initials of the persons recorded, the first five are the males and the last five are the females.

4.5.2. Findings The use of the nonstandard form is more consistent in the speech of males than in the speech of females, thus revealing that there is gender-related differentiation. Another difference lies in the topics of discussion. While men were discussing politics or commenting upon Romanian culture, women were talking about the latest gossips, plastic and cosmetic surgery, fashion and cooking. The fact that they focused mostly on personal aspects indicates that they interact with each other in order to maintain and increase solidarity. The fact that men use the [pă] variant more frequently indicates that they are not interested in the way they are perceived by society and they are more interested in conveying information. They also used other nonstandard forms like dă instead of de, swear words or taboo language. Let us first look at the figures for [pe] and [pă] respectively, and then comment upon the differences.

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

100 90 80 70 60 50 40 30 20 10 0

131

[pe] [pă]

Men

Women

Figure 4.21. The use of [pe] and [pă] according to the gender of the speaker

The hypothesis that men prefer the vernacular form is verified for Romanian as well. By replicating the sociolinguistic survey and including adults into the equation I have identified gender-related variation concerning the (pe) variable. According to the quantitative studies carried out in English speaking communities since the pioneering works of Labov (1966, 1972) and Trudgill (1974), stylistic variation is often more predominant in women’s speech than men’s. In the case of adult men (pă) is a marker of masculinity and plays an important role in the construction of masculine identity. Consider the following examples from the recordings: (60) domnu’ Varujan pă blog citesc dă pă hotnews...iete d-aia n-avem io pulovăr pă gât. (Mircea Badea, În gura presei, 14.10.2013) ‘Mister Varujan on his blog I’m reading from hotnews...that’s why I didn’t have a pullover on my neck’ (61) Frica a devenit un divertisment úi pă mine mă enervează chestia asta (Andrei Pleúu, 50 de minute cu Pleúu úi Liiceanu – Despre frică, 4. 06. 2013) ‘Fear has become a sort of entertainment and I am very annoyed by this thing’

132

Chapter Four

(62) A te supăra pe poporul tău e ca úi cum te-ai supăra pă tatăl tău... (Corneliu Vadim Tudor, Happy Hour, 19.01.2014) ‘Being angry on your people/country is like being angry on your father’ (63) Nici io nu stau pă al meu d-aia suntem singure amândouă (Teo Trandafir, Teo Show, 12. 06. 2014) I am not sitting on my own that’s why we’re both single (64) Da’ toĠi bărbaĠii mă iubesc mai mult pă mine decât îi iubesc io pe ei (Oana Zăvoranu, Un Show Păcătos, 7. 03. 2014) ‘All men love me more than I love them’ (65) úi pe naúa aĠi invitat-o? (Simona Gherghe, Acces Direct, 3. 06. 2013) ‘did you invite the godmother?’ The male and female speakers recorded do not belong to the working class, as they are intellectuals. In the literature the vernacular is associated to the working class, so in the case of (pă) we might speculate that working class features of speech are becoming attractive and such patterns surface in upper class speech. This fact provides valuable insights into the internal as well as the external forces governing language variation and change and it also draws attention to the pivotal role of gender differences in these situations. The concept of toughness appears again on the menu, as men exert their power and masculinity through language. They use the nonstandard form to appear strong, secure and powerful. [Pă] is manly whereas [pe] is softer and more delicate, thus the latter is used in the construction of feminine identity. Femininity and masculinity are not biological traits (like male and female), but they are constructed through the process of socialisation55, in other words they are culturally constructed (Hornoiu 2008: 70). The examples provided above in (60)– (65) show that the (pe) variable can be considered a gender marker. In order to corroborate this claim, I have also looked up the use of the (pe) variable in Hornoiu’s (2007) corpus of spoken Romanian. I counted all the (pe) tokens found in her corpus and the total number was 290. I excluded from the analysis unclear tokens. The results are given below in Figure 4.22:

55 For adolescents the process of socialisation takes place at school, among their peers or in their family whereas adults socialize in the workplace.

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

160 140 120 100 80 60 40 20 0

133

[pe] [pă]

Romanian adult Romanian adult women men

Romanian adolescent women

Figure 4.22. The distribution of [pe] and [pă] according to gender in Hornoiu’s (2007) corpus

The standard variant is preferred by adult as well as adolescent women. Adult men also use the standard form [pe] in 87% of the cases. However, in Hornoiu’s corpus the focus is more on conversations between women. Another aspect is that there are no mixed-gender conversations. It would have been interesting to see men and women’s linguistic choices in mixgender settings. These findings corroborate the results obtained from the recorded interviews focusing on adults as well as the results from the sociolinguistic interviews. Women, irrespective of their age tend to use the [pe] variant. There is a change when it comes to men. If in secondary school boys tend to use the standard form almost exclusively, the situation changes gradually. In high school they start using the non-standard form, i.e. [pă], sometimes exclusively in same-gender conversations. Adult men have a preference for the [pă] form, the use of the vernacular being interpreted as a marker of masculinity. Closely connected to masculinity is the concept of power. By using the vernacular men are seen as powerful and dominant. Speech style becomes an important tool for both genders. Steffen and Eagly (1985) acknowledged that high-status persons used a more direct and less polite style while lower-status persons were more interested in face-saving. Therefore the higher the status, the more direct and less polite the style of talk was considered to be (Thimm et al. 2003: 531). If we follow this direction, two approaches arise: the “sex dialect hypothesis” or “genderlect” or “female register hypothesis” (Thimm 1995)

134

Chapter Four

and the second, “the sex stereotype hypothesis” (Thimm et al. 2003). The former assumes that the evaluation of communication between males and females is based on language performance differences. The latter hypothesis suggests that linguistic judgements are determined by stereotypical expectations. Therefore, men are expected and allowed to use a powerful style but in the case of women such behaviour is not approved. This could also explain men’s higher use of the [pă] variant. The higher their social status the more they use the vernacular. In the case of Mircea Badea, Andrei Pleúu and Corneliu Vadim Tudor this hypothesis is verified. They are high-status men, at the top of their respective fields, and the use of [pă] is consistent in their speech. Their speech style is also direct and impolite at times. I consider their use of the vernacular a form of popularity, of adhering to the masses. By identifying themselves with the working class (through the language they use) they become popular among the people who watch them.

4.6. Adolescent speech versus adult speech My original hypothesis in this chapter was that adolescents are considered linguistic innovators and patterns of variation are likely to appear more frequently in their speech. For Coulmas (2013: 70) adolescent speech is characterized by “the use of substandard, dialectal and vernacular forms, slang and innovative”. These features associated with adolescent speech serve three functions: (i) to make the language suitable for the speaker’s own purposes; (ii) to show group membership and construct a different identity; (iii) to indicate that the speaker does not want to conform to societal norms. However, adolescent speech is also influenced by many extralinguistic factors, as mentioned above. Adolescents are known to break cultural and societal barriers and behave differently from adults, but when it comes to language use, the factor that must be taken into account is their group membership. We have seen that adolescents speak differently depending on their group membership. If the group is involved in different illegal activities, then the use of non-standard forms increases. A more suitable word to describe such a group would be gang (Cheshire 1982). Interestingly, the use of the vernacular in Romanian adolescent speech is connected to group affiliation.

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

135

Figure 4.23. Age and standard speech (adapted from Downes 1984: 191)

Figure 4.23 presents the correlation between the use of standard or vernacular language and age. Thus, in adolescent speech the use of the vernacular is highly consistent, according to Downes (1984), whereas the use of standard forms is preferred by middle-aged speakers, followed by a rise in the use of the nonstandard features in the speech of elderly speakers. In other words, this is a typical manifestation of the phenomenon of age grading. I strongly believe that, although adolescents are language innovators, their use of the vernacular is somehow restricted and influenced by various factors. The fact that they used the [pă] variant more often in the playground than in the sociolinguistic interview indicates that they are aware of social pressure and speak differently in distinct situations. On the whole, the [pă] variant occurs consistently in the speech of high school boys and it is also used by high school girls consistently only in mixedgender settings. Figure 4.24 below summarizes my findings and draws a comparison between adolescent speech and adult speech.

136

Chapter Four

Figure 4.24. Age and standard speech in Romanian

Non-standard forms of (pe) are more consistent in the speech of Romanian adults. How can we account for this? An explanation offered by Coulmas (2013: 71) is that with age the pressure of societal norms or even the willingness to conform to such rules decreases. That could account for the rise in the use of nonstandard [pă]. However, this applies to males only as females use the nonstandard form in certain contexts only. The expectation was that adults were supposed to use the standard form, as proposed by Downes (1984). However, the results that I obtained disconfirm this account. If we were to include women in the equation, things would change as their preference for the prestige form has been repeatedly demonstrated. To sum up, there appears to be a change with age in relation to the use of the (pe) variable. Not only does the use of [pă] increase with age, but it is considered a marker of masculinity and closely connected to this is the concept of power. Adult men as well as high school boys use the vernacular form to exert their masculinity and toughness. High school girls use the standard form in same gender conversations and the vernacular in mixed gender conversations, in order to compete with boys and appear cool. By imitating boys’ way of speech they are seen as their equals. Adult women prefer the standard form in all situations, irrespective of the gender of their interlocutors. This indicates that they are status-conscious and lay emphasis on the way they are perceived by society. These differences are not gender-exclusive, they only show a preference for one form or another.

4.7. Concluding remarks This chapter has addressed the main findings concerning variation in the speech of Romanian adolescents living in ConstanĠa. The analysis of the

PE vs PĂ: Romanian Adolescents Living in ConstanĠa

137

(pe) variable has shown that in secondary school adolescents tend to use the standard variant to the vernacular one. A change occurs in high school where the [pă] variable starts to be used. However, caution needs to be exercised when making such claims, as language choices are influenced by a number of external factors. For example, at ECTH I have identified two communities of practice: the geeks and the pops. In showing their allegiance to the group they use standard or nonstandard features more often. In the presence of their peers the pops’ use of nonstandard forms increases. The geeks use the prestige forms irrespective of the situation they are in and irrespective of their interlocutors’ gender. I have also shown that [pă] can be considered a marker of masculinity, of exerting power and toughness. I also included in the analysis a discussion of adult speech and compared it with adolescent speech. The comparison yielded the following results: (i) the use of [pă] is more consistent in the speech of Romanian adult males than females, (ii) and also more frequently used compared to adolescents. This indicates that with age, there is a preference for the vernacular among male speakers. A conclusion that we can draw is that this is in line with the empirical research conducted in English speaking communities, and with the findings for the (ing) variable presented in Chapter 3. Another interesting aspect is that the use of [pă] is an important factor in the construction of masculine identity whereas [pe] is used in the construction of feminine identity. However, this claim is valid for high school adolescents and adults only. In secondary school students prefer the prestige form as they are nurtured by their mothers or by a female figure so they borrow their linguistic features. Two research methods have been employed to offer a more detailed analysis of phonological variation among adolescents. The sociolinguistic interview has revealed almost insignificant results about the use of the (pe) variable but the long-term participant observation method offered a more in-depth analysis of linguistic variation. The two communities of practice identified differ in their linguistic as well as social practices: geeks opting for the standard features of language, although this made them look unpopular among their peers and, pops choosing the nonstandard form which made them look tough, powerful and cool in the school.

CHAPTER FIVE ADOLESCENTS AS LANGUAGE INNOVATORS: SWEARING, TABOO WORDS AND SLANG IN ENGLISH AND ROMANIAN

Introduction This chapter offers a qualitative analysis of swearing, taboo words and slang in the speech of adolescents. It is a contrastive chapter, as it focuses both on English and Romanian and sets out to identify similarities and differences between the two cultures. The chapter begins with a discussion of taboo language and what should be considered taboo. Then the discussion moves to bad language, as viewed by Allan and Burridge (2006) and McEnery (2009). The role of dirty words (orthophemisms) and sweet words (euphemisms) is discussed and relevant examples are provided. The English analysis draws on the Bergen Corpus of London Teenage Language and on Stenström et al.’s (2002) findings. It will be shown that there are interesting gender-related differences in the English spoken by London adolescents, but also age-related ones. The Romanian analysis is based on a corpus of 6500 words and it discusses the use of slang and swearing by teenagers living in ConstanĠa. Slang words, as they are used by teenagers, are discussed at length and the etymology of these words is given together with different plausible explanations which might account for their use. The chapter ends with a discussion of gender variation in spoken and written Romanian.

Adolescents as Language Innovators

139

5.1. Taboo language The word taboo56 is usually used to describe something that cannot be said because “it is either ineffably sacred, like the name of God, or unspeakably vile, like cannibalism or incest” (Hughes 2006: 462). The English word taboo, as we know it, is derived from the Tongan word tabu, which started being used towards the end of the eighteenth century (Allan and Burridge 2006: 2). An interesting account of what is considered to be taboo is given by Captain James Cook, who, on his first voyage to Tahiti (1768-1771) observed the following about the Tahitians: The women never upon any account eat with the men, but always by themselves. What can be the reason of so unusual a custom, ‘tis hard to say, especially as they are a people in every other instance, fond of Society, and much so of their Women. They were often Asked the reason, but they never gave no other Answer, but that they did it because it was right, and Express’d much dislike at the Custom of Men and Women Eating together of the same Victuals. We have often used all the intreatys we were Masters of to invite the Women to partake of our Victuals at our Tables, but there never was an instance of one of them doing it in publick, but they would Often goe 5 or 6 together into the Servants apartments, and there eat heartily of whatever they could find, nor were they in the least disturbed if any of us came in while they were dining; and it hath sometimes hapned that when a woman was alone in our company she would eat with us, but always took care that her own people should not know what she had donn, so that whatever may be the reasons for this custom, it certainly affects their outward manners more than their Principle (Cook 1893: 91, quoted in Allan and Burridge 2006: 3).

Cook does not use describe this custom as taboo. However, he does use the word in a later expedition in an entry for 17 July 1777: Taboo as I have before observed is a word of extensive signification; Human Sacrifices are called Tangata Taboo, and when any thing is forbid to be eaten, or made use of they say such a thing is Taboo; they say that if the King should happen to go into a house belonging to a subject, that house would be Taboo and never more inhabited by the owner; so that when ever he travels there are houses for his reception (Cook 1967: 176, quoted in Allan and Burridge 2006: 3).

56

Freud (1950: 18) writes that “taboo is a Polynesian word, the translation of which provides difficulties for us because we no longer possess the idea which it connotes”.

140

Chapter Five

In these entries from his logbook, Cook observes what is considered to be taboo in terms of behaviour in the Tahitian society. Other types of taboos include: (i) Food taboos (e.g. Muslims are not allowed to eat pork); (ii) Name taboos (e.g. the name of one of the plays written by Shakespeare is referred to as “the Scottish play” in the theatre and not as Macbeth); Name taboos are unfortunately under-researched. An account of this type of taboos is provided by Allan and Burridge (2006: 128) who discuss a case of name taboo encountered among the Nguni peoples of South Africa, who practice hlonipha, i.e. respect shown by avoiding the use of personal names of a husband’s father and all the males from that family. A woman is strictly forbidden to use her father-in-law’s name when she is talking to him. Given the low status of women among the Nguni peoples, they receive orders not only from their husband but also from the husband’s brothers. When she is spoken to, her personal name is not used. Instead she is referred to as the daughter of her father. After she has a child she is named as the mother of that infant. Having sexual intercourse with a minor is deemed to be taboo. Incestuous marriages are also taboo, but many kings did this with or without their knowledge. The first example that comes to mind is that of Oedipus, who fulfilled his destiny, i.e. he killed his father and married his mother. This, in my opinion, could be viewed as the first example of taboo. Another example is that of the French writer Voltaire (1694-1778) who had a sexual relationship with his niece, to whom he wrote things like: My child, I shall adore you until I’m in my grave...I would like to be the only one to have had the happiness of fucking you, and I now wish I had slept with no-one but you, and had never come but with you. I have a hard on as I write to you and I kiss a thousand times your beautiful breasts and beautiful arse (quoted in Allan and Burridge 2006: 10).

Not your average uncle, is he? The same thing was done by Adolf Hitler who is rumoured to have had a secret relationship with his niece, Geil Raubal, who is said to have committed suicide precisely because of this incestuous relationship. And the examples go on. What is important to highlight is that such practices are considered taboo and are condemned by society.

Adolescents as Language Innovators

141

From a linguistic point of view, taboo is rooted in what Hughes (2006: 462) calls “word magic”, particularly in the belief that certain names or creatures or even forces must not be named. A good example is the Third Commandment, which states that people “should not take the Lord’s name in vain” (Exodus 20: 7). Many examples come from literature. Let us consider the following example: (66) Jim was monstrous proud about it and he got so he wouldn’t hardly notice the other niggers. Niggers would come miles to hear Jim tell about it and he was more looked up to than any nigger in that country. (Mark Twain – The Adventures of Huckleberry Finn, 1994: 15-16) (bold mine) The word nigger57 is considered to be an offensive word (Macmillan English Dictionary) used to refer to a black person. I do consider it to be a taboo word, as in many American communities the use of such word is forbidden. The fact that Twain used this word in his novel caused many discussions and people were appalled. The same happened when Margaret Mitchell published her novel Gone with the Wind (which eventually turned into a movie) and the expression Frankly my dear, I don’t give a damn was used. The use of the word “damn” was quite controversial. During the Communist regimes it was considered taboo to say something against the regime or Party officials. So, the term taboo has many connotations and encompasses a wide number of meanings. The notion taboo language describes offensive emotional language (Jay 2009, Pinker 2007). It is noteworthy that, albeit scarce, research on taboo language has claimed the existence of socio-cultural differences (Jay 1980, McEnery 2006, Jay 2009, among others). For example, although female speakers have been reported as dealing with and expressing emotions more frequently than men (Barrett et al. 1998), male speakers have been found to know and use taboo words much more frequently (Kutner and Brogan 1974, Lakoff 1975). There have been studies which argued that depending on the pragmatic context, male or female speakers justify the use of taboo words (Locker and Watts 2005). Pilotti et al. (2012: 17) write that “support for the notion that the speaker’s pragmatic context is relevant to emotional expression comes from the finding that sex differences in self-reported intensity of emotional experience are more likely to emerge from interactions that involve the opposite sex than the 57 It is important to underline the fact that the word nigger is confined to American English and used as an insult to the black population.

142

Chapter Five

same sex”. So, it appears that women are reluctant to use taboo words when talking to the same gender, but have no problem in using them with the opposite gender. In this chapter, following Allan and Burridge (2006: 1) I include under the heading of taboo language the following: (i) Bodies and their effluvia; (ii) The organs and acts of sex, micturition and defecation; (iii) Naming, addressing, touching and viewing persons and sacred beings, objects and places. Many people are still reluctant to use the word fuck in public. Is this word still considered to be taboo? Although the question is simple, the answer is complex. It depends on the community in which it is being used. The word fuck is considered even today to be “the most powerfully taboo term for copulation over several centuries and it is still regarded as unmentionable by the vast majority of middle-class people” (Hughes 2006: 188).

5.2. Bad language and sweet words What does the expression bad language really mean? When we hear “bad language” we might be tempted to think about swearing or slang. Is it bad language when we curse because something bad happened to us? When the f-word is used on TV or in a novel is it considered to be bad language? If we were to cut ourselves we would say something like Damn it or Fuck or Shit. Is this bad language? McEnery (2009: 565-566) defines bad language as “language which, when used in polite company, would be likely to cause offence”. Imagine if I were to talk about bad language at a conference and I started discussing words like cunt, dick, fuck, masturbation, porn, shit, etc., everybody in the audience would feel very uncomfortable. This is because of the social pressure that society puts on these words. They should not be used in a conversation as they are considered inappropriate. Let us consider the word masturbation. Allan and Burridge (2006: 145) provide four reasons why the word masturbation is taboo: (i) Masturbation involves stimulation of the genitals, not for procreation but for pure pleasure – it is seen by some as an unnatural act because it goes against God’s plan for procreation; (ii) Masturbation is supposedly addictive;

Adolescents as Language Innovators

143

(iii) Masturbation typically involves manipulation of the genital organs, giving rise to the emission of effluvium; (iv) It was long believed that seminal fluid and its counterpart in women, vaginal secretion, was refined blood that bore generative seed; wasting such precious fluid was supposedly even more debilitating than loss of blood. Bad language may come in different forms, e.g. pronunciation, poor grammar, slang, and of course swearing (Burridge 2004: 92). Let us consider the following examples: (67) I knew you was lying to me. (68) Where’s the dough? (69) Fuck that shit! Are the sentences provided in (67)–(69) examples of bad language? Although we might be tempted to say yes, I believe that these are not examples of bad language, at least sentences (67) and (68). Cockney is an accent which is rated negatively even by those who speak with such an accent. The main function of language is to permit communication. Language is also used to identify ourselves with a certain geographical space and is also part of our identity. The way we speak (i.e. our accent, vocabulary, grammar) shows our ethnicity, gender, race, religion, sexual orientation, social class, etc. If we take accent into account, if that respective accent has low prestige, should we consider it bad language? We can all agree that some accents, like RP, for instance, are associated with the upper class and prestige and education. Accents like RP are considered to have overt prestige (Burridge 2004: 93). By contrast, bad language has covert prestige: “Swearing and an imposing accent are associated with toughness and strength and they can be viewed as highly valued qualities” (Burridge 2004: 93). I think that there are many good things in bad language, otherwise it would not survive the test of time. In 2010 the movie The King’s Speech was released. The film tells the story of King George VI of Great Britain and Northern Ireland, his ascension to the throne and his speech difficulties. Let us consider a conversation between King George VI (portrayed by Colin Firth) and his speech therapist Lionel Logue (played by Geoffrey Rush):

144

Chapter Five

(70) King George VI: All that work down the drain. My own brother, I couldn’t say a single word to him in reply. Lionel Logue: Why do you stammer so much more with David than you ever do with me? King George VI: ‘Cos you’re b...bloody well paid to listen. Lionel Logue: Bertie, I’m not a geisha girl. King George VI: St...stop trying to be so bloody clever. Lionel Logue: What is it about David that stops you speaking? King George VI: What is it about you that bloody well makes you want to go on about it the whole bloody time? Lionel Logue: Vulgar, but fluent; you don’t stammer when you swear. King George VI: Oh, bugger off! Lionel Logue: Is that the best you can do? King George VI: Well...bloody bugger to you, you beastly bastard. Lionel Logue: Oh, a public school prig could do better than that. King George VI: Shit. Shit, shit, shit, shit, shit, shit, shit, shit, shit, shit, shit, shit! Lionel Logue: Yes! King George VI: Shit! Lionel Logue: Defecation flows trippingly from the tongue! King George VI: Because I’m angry! Lionel Logue: Do you know the f-word? King George VI: F...f...fornication? Lionel Logue: Oh, Bertie. King George VI: Fuck. Fuck! Fuck, fuck, fuck and fuck! Fuck, fuck and bugger! Bugger, bugger, buggerty buggerty buggerty, fuck, fuck, arse! Lionel Logue: Yes... King George VI: Balls, balls... Lionel Logue: You see, not a hesitation! King George VI: ...fuckity, shit, shit, fuck and willy. Willy, shit and fuck and...tits. This is an interesting example of the use of swear words. They are considered taboo as they refer to different parts of the body or to coitus. Later on in the movie King George VI confesses that he does not usually use these words as he is not allowed to speak like this, at least not in public. This is a good example of the social pressure that society puts on the use of swear words in public. When Lionel Logue asks the King whether he knows the f-words, the latter hesitates a little and his first

Adolescents as Language Innovators

145

impulse is to use the word fornication and not fuck. Fornication is not as powerful and taboo as fuck. It is vital to keep an open mind if we study bad language. But, what is bad language used for? Table 5.24 below provides a detailed account of the uses of bad language, to be used as a starting point in my analysis: Category Predicative negative adjective Adverbial booster Cursing expletive Destinational usage Emphatic adverb/adjective Figurative extension of literal meaning General expletive Idiomatic ‘set phrase’ Literal usage denoting taboo referent Imagery based on literal meaning Premodifying intensifying negative adjective Pronominal form with undefined referent Personal insult referring to defined entity Reclaimed usage – no negative intent Religious oaths used for emphasis

Example The movie is shit Fucking great Fuck you! Fuck off! He fucking did it/in the fucking car To fuck about (Oh) Fuck! Fuck all/give a fuck We fucked Kick shit out of The fucking idiot Got shit to do You fuck! Niggers/Niggaz as used by African American rappers by God

Table 5.24. Types of swearing (Source: McEnery 2009: 569) A discussion of taboo and bad language inevitably leads to a discussion of politeness and impoliteness. In doing so, we must also take into discussion the following terms: (i) euphemism (sweet talking) (ii) dysphemism (offensive talk) (iii) orthophemism (straight talk) In order to illustrate their meaning, consider Table 5.25:

Chapter Five

146

Orthophemism faeces toilet menstruate vagina Jesus

Euphemism poo loo period bits Lord

Dysphemism shit shithouse bleed cunt Christ!

Table 5.25. Contrasting X-phemisms (Source: Allan and Burridge 2006: 32) In Table 5.25, the words labelled as dysphemistic are taboo and considered to be impolite choices. I consider orthophemisms and euphemisms to be sweet words, as they represent an alternative to something that we do not want to say. The difference between orthophemisms and euphemisms is that the former are formal while the latter are more colloquial and figurative. The occurrence of euphemisms can be seen as a reaction to the harshness of taboo words or to the words that were considered inappropriate. The use of euphemisms helps avoid bad words and thus, rudeness is neutralised. Hornby (2005) views euphemisms as words or expressions which refer to something unpleasant, rude, or embarrassing and make them acceptable to be used in society, while Holder (2008) describes euphemisms as milder or uncertain terms which are used to replace unsuitable expressions. For Ljung (2011: 11) euphemistic replacements have many forms. McArthur (1992: 661) identifies two types of such replacements which he names “minced oaths”: (i) creating a nonsense equivalent of a swear word (e.g. Gosh for God), (ii) substituting an everyday expression of similar sound and length (e.g. damn for darn). At this point a remark is in order. These euphemistic replacements work only in these contexts, i.e. in interjections to show surprise, anger, happiness, etc. (71) I believe in God. In sentence (71) above God cannot be substituted for Gosh as it would not make sense. A sentence like I believe in Gosh is ambiguous, at least.

5.2.1. Swearing, insult and dirty words There is a folklinguistic belief according to which swear words are the first words that we learn in a new language. If swearing is so condemned by

Adolescents as Language Innovators

147

society how can this be true? Andersson and Trudgill (1990: 35) argue that “no word is in itself bad. It is bad only in the eyes of those who evaluate and look at the language”. Andersson and Trudgill (1990: 53-55) further claim that swearing is bad and that taboo words are words that we should not say. On a similar note, Jay (1992: 13) writes that “words that are inhibited or taboo are not necessarily obscene and whether a word is taboo or not fluctuates as a function of the speaker-listener relation”. Jay gives as an example the word fuck. Such a word is unsuitable in the presence of a child but can be used without any restriction among friends. A few years ago, while an undergraduate student, I overheard the dean of the Faculty of History saluting a fellow colleague by asking him Ce faci, băi pulă? ‘How are you, dick?’ The word pulă ‘dick’ in Romanian is considered to be taboo and it is usually avoided. However in this context the interlocutor was not offended at all. Such words might well be used among friends without causing bad feelings. In the literature different types of swearing have been identified. Ljung (1984: 14-15) draws a line between aggressive swearing which shows the speaker’s emotional and social swearing used to strengthen group affinity. McEnery (2009: 572) discusses the strength of swear words and groups them into: (i) Very mild: bird, bloody, crap, damn, God, hell, hussy, idiot, pig, pillock, sod, son-of-a-bitch, tart; (ii) Mild: arse, balls, bitch, bugger, Christ, cow, dickhead, git, Jesus, Jew, moron, pissed off, screw, shit, slag, slut, tit, tits, tosser; (iii) Moderate: arsehole, bastard, bollocks, gay, nigger, piss, Paki, poofter, prick, shag, spastic, twat, wanker, whore; (iv) Strong: fuck; (v) Very strong: cunt, motherfucker. Closely connected to the concept of swearing are two other concepts: cursing and insult. Allan and Burridge (2006: 76) note that “the dysphemistic connotations of swearing lead to its being associated with cursing ‘imprecating malevolent fate’”. (72) a. b.

God damn it! Fir-ar al dracului!

The sentences provided under (72) are examples of cursing in English and Romanian. Interestingly, cursing involves the name of a deity or God. In this case, in English the name of God is invoked while in Romanian the

Chapter Five

148

name of the Devil is used. Both sentences have the same meaning, and (72a) is translated into Romanian as (72b). Jay (1999: 9) identifies ten myths about cursing including: “swearing, obscenity, profanity, blasphemy, name calling, insulting, verbal aggression, taboo speech, ethnic-racial slurs, vulgarity, slang and scatology”. The following myths are relevant to this analysis, as they are formulated in relation to teenage talk (Jay 1999: 257-259): Myth 6: Myth 7: Myth 8: Myth 10:

Cursing is mainly a problem for the teenage years; Cursing is a habit of the undereducated and lower classes; People are cursing more than ever before; Cursing is due to an impoverished lexicon and laziness.

Cursing depends on the context and I do not believe that it can be associated with age. Everybody curses, from teenagers to the elderly. Jay (1999: 257) admits that “teenagers may do more cursing in public” which I consider valid, at least for Romanian. Myth 7 can be easily dismissed as cursing is not influenced by social class. Consider the following extract from a conversation between Corneliu Vadim Tudor (member of the European Parliament) and ex-senator and Marioara Zăvoranu58: (73) M.Z: C.V.T: M.Z: C.V.T: M.Z:

C.V.T: M.Z.:

58

Unde e domnu’ Vadim? ‘Where is Mister Vadim?’ Lasă-mă-n pace că te plesnesc de... ‘Leave me alone or I’ll smack you...’ Du-te domne’ de aicea ‘Leave me alone’ Du-te-n pizda mătii de javră ce eúti. Hai sictir de aici ‘Go fuck yourself you piece of shit. Go to hell’ Du-te-n pula mea de nenorocit ce eúti. De ce te iei tu dă mine mă? ‘Fuck you, you motherfucker. Why are you picking on me?’ Du-te dracului ‘Go to hell’ Dacă mai se ia de fiica mea nu útiu ce-i fac...îi sparg capu’ fir-ar mă-sa care l-a făcut. El face circ cu fata mea úi cu mine? Când am curvăsărit io?

The entire conversation can also be seen here (last accessed July 29, 2014).

Adolescents as Language Innovators

C.V.T:

C.V.T:

M.Z: C.V.T:

149

‘If he picks on my daughter I won’t answer for myself...I will crack his head piece of shit. He makes fun of my daughter and me? When was I a whore?’ Cine e, mă asta? Du-te-n morĠii mă-tii. Tu nu te vezi cum arăĠi? ‘Who is this? Go fuck yourself. Don’t you look at yourself?’ Băga-mi-aú pula-n gura ta. Vrei să te iau ca pă Rahova? ‘Fuck you. Do you want me to speak to you as in Rahova?’ Dă mă ceasu’ ‘Give me the watch’ Ăsta-i ceasu’ tău? Vrei ceas? Da o muie nu vrei tu? ‘Is this your watch? Do you want a watch? How about a blowjob?’

This is a perfect example why Myth 7 is not valid, at least not for the Romanian society. Cursing and swearing are classless. Another conclusion that we might draw from this mixed-gender conversation between two adults is that men and women swear and curse alike. I will return to this issue later on in this chapter. It is difficult to confirm or infirm Myth 8 as we cannot verify what occurred in the past. As far as Myth 10 is concerned, Jay (1999: 259) states that “curse words are not substitutes when speakers cannot find the ‘right’ word but caused by neurological, psychological, and sociocultural forces”. On the other hand, an insult represents a verbal abuse by using “contemptuous, perhaps insolent language” (Allan and Burridge 2006: 79). Allan and Burridge further note that verbal insults can be found in all the styles of language and the main purpose of insults is to offend or wound the addressee. The words used in verbal insults are entirely dysphemistic and they are subject to censorship. When such insults appear on TV they are always censored with a beep. Dysphemistic words used in insults have been grouped into the following categories: (i) Name of animals identified with people: (e.g. ape, bitch, bull, cat, cow, dog, monkey, mule, pig, swine, etc); (ii) Words derived from tabooed bodily organs (e.g. asshole, prick), bodily effluvia (e.g. shit, piss), sexual behaviours (e.g. cocksucker, dipshit, fucker, slut, slapper, wanker, whore, etc.);

Chapter Five

150

(iii) Dysphemistic words that pick on real physical characteristics treated as if they are abnormalities (e.g. Fatty!, Baldy!, Foureyes!, Short-arse!); (iv) Words describing mental subnormality or derangement (e.g. Airhead!, Dickhead!, Moron!, Retard!, Shithead!, etc.) (Source: Allan and Burridge 2006: 79-82) Another angle from which these words can be studied and their usage accounted for is language and gender. So far I have presented different concepts and claims without taking into account the social factor that I am interested in, i.e. gender. With regard to the use of swearing by male and female speakers in conversation, and whether there are gender-related differences, this area has been under-researched. However, there are certain claims (Jespersen 1922, Lakoff 1975, Hughes 1992, Coates 2004) that men swear more than women. Lakoff (1975: 10) provides an example of ‘male’ and ‘female speech’: (74) a. b.

Oh dear, I’ve put the peanut butter in the refrigerator again. Shit, I’ve put the peanut butter in the refrigerator again.

The American linguist argues that the sentence in (74a) is uttered by a woman whereas the example provided in (74b) is typical of male language. Lakoff also writes that: A girl is damned is she does, damned if she doesn’t. If she refuses to talk like a lady, she is ridiculed and subjected to criticism as unfeminine; if she does learn, she is ridiculed as unable to think clearly, unable to take part in a serious discussion: in some sense, as less than fully human (Lakoff 1975: 6).

Lakoff’s claim is not only out-of-date but also based on no empirical research. The suggestion that the use of swear words is typical of male speakers and that this type of language is avoided by female speakers has been shown to be false by research studies carried out in the United States of America (Risch 1987), the United Kingdom (Hughes 1992), and South Africa (de Klerk 1992). Everybody swears, even unintentionally, and even if they did not they would still have passive knowledge of almost all swear words (Allan and Burridge 2006: 89). In the conversation fragment provided in (73) we notice that even though the discussion between the two persons (of different genders) is very heated they do not refrain themselves from using swear words, cursing each other, something one

Adolescents as Language Innovators

151

would not normally hear in a mixed-gender conversation. The fact that Marioara Zăvoranu uses words describing male genitalia indicates that she is trying to appear powerful and not in an inferior position. Corneliu Vadim Tudor exerts his dominant position by using all kinds of swear words, he even calls his interlocutor javră ‘bitch’. By doing this he secures himself a position of power and the reason is to make the addressee uncomfortable and attack her face (Brown and Levinson 1987). The concept of face as devised by Brown and Levinson (1987: 61) in their politeness theory states that it is “the public self-image that every member wants to claim for himself”. Social interaction is usually oriented towards saving one’s face and in the case of Marioara Zăvoranu and Corneliu Vadim Tudor they lose their faces.

5.2.2. Slanguage The term slanguage was coined by Stenström et al. (2002) and refers to taboo language. Their analysis was based on COLT (The Bergen Corpus of London Teenage Language). The authors provide the following definitions: (i) Proper slang words: words which are considered slang by dictionaries. In this category two types are distinguished: general slang words which are not part of a group or trend (e.g. booze, fag) and specific slang words which are part of a group or trend (e.g. speed, spliff); (ii) Dirty words: words labelled offensive which can be divided into slang words and swear words (e.g. piss somebody off, bugger off, fucked-off, motherfucker); (iii) Vogue words: words that already exist which are fashionable for a short period of time (e.g. massive, reckon) or which acquire a new sense (e.g. wicked); (iv) Vague words: words which have a very general meaning and which replace adequate words (e.g. thingy) or are used as set markers (e.g. and that lot); (v) Proxy words: substitutes for quotation verbs (e.g. be like); (vi) Smallwords: this category encompasses ‘tags’ (innit, yeah), ‘hedges’ (e.g. just, like, sort of) and ‘empathizers’ (e.g. you know). (Source: Stenström et al. 2002: 64-65) Providing a definition for slang appears to be a herculean task. For Lighter (1978: 14-16) slang is “a taboo term in ordinary discourse with persons of

152

Chapter Five

higher social status or greater responsibility” while Quirk et al. (1985: 27) as well as Andersson and Trudgill (1990: 69) avoid defining slang by writing “what slang is and what it is not”. In this study I consider slang the words and expression which dictionaries label as slang. The Macmillan English Dictionary (2002) defines slang as “words or expressions that are very informal and are not considered suitable for more formal situations. Some slang is used only by a particular group of people”. Oxford English dictionary online describes slang as “a type of language consisting of words and phrases that are regarded as very informal, are more common in speech than writing, and are typically restricted to a particular context or group of people”. Stenström (2014: 13) notes that what is considered or viewed as slang changes through time. These words are used for a limited period of time and then they are replaced by new ones. I remember when I was in high school that the word marfă as in Ce marfă e tricoul tău ‘Your t-shirt is so cool’ was very fashionable and everybody used it. It was a vogue word in Stenström et al.’s (2002) terminology. Nowadays this slang word is no longer in use, i.e. it is no longer used with the meaning of ‘cool’. It was replaced by words like miúto, úmecher which mean ‘cool’. Regarding the origin of slang words, Andersson and Trudgill (1990: 82) provide three sources: new expressions, new use of standard expressions and borrowings from other languages. As far as the last source is concerned, in Romanian teenage talk some slang words are borrowed from English, as expected, given the influence the American culture has on the Romanian teenage culture, but there are also many slang words borrowed from Rromani. This may come as a surprise given that the Rroma people and their language are not easily tolerated by Romanians. This assertion is based on no empirical research, just on my observations in different situations where Romanians and Rroma were involved. I also recorded a conversation on Facebook, between two Romanians, arguing about Rroma59: (75) Mihai: ùtii cui ar trebui să-i spui? Ġiganilor care erau cât pe ce să mă taie acum o saptamână... ‘You know who you should tell? The Gypsies that were about to cut me about a week ago...’ Mihai: rasismul ăsta nu există degeaba, au oamenii niúte motive să fie rasiúti. ‘Racism is not in vain, people have reasons to be racist’ 59

The conversation is reproduced with the original orthography and punctuation.

Adolescents as Language Innovators

153

Cosmin: http://www.yousign.org/ro/tricolor Cosmin: Zarvă mare la o úcoală din Covasna după ce o elevă de clasa a IX-a s-a afiúat la úcoală cu o bentiĠă cu tricolor. ‘Lots of trouble at a school in Covasna after a 9th grade student came at school wearing a bandeau in the colours of the Romanian flag’ Mihai: cosmin nu mia distribui asta, discriminezi o minoritate care se piúă pe tine. ‘cosmin stop distributing this, you’re discriminating a minority that pisses on you’ Cosmin: bn măh ‘ok dude’ Meriem: ba fix degeaba există. Crezi că aia de voiau să te taie o făceau că sunt Ġigani? sau mai bine zis, nu există úi români,sau orice altă naĠie „dispuúi” la acelaúi comportament? nu e corect să pui o purtare greúită pe seama unei întregi comunităĠi, cum aceeaúi purtare nu găseúti doar la o comunitate ‘it exists for no reason. You think that those who wanted to cut you were doing it because they were gypsies? Or to put it differently, aren’t there Romanians or any other nation “inclined” to the same behaviour? It’s not right to label an entire community because of one misbehaviour as you won’t find the same behaviour in just a single community’ Mihai: ete pula mea viicu 1000 de întrebări idioate, ce pula mea te tot ascunzi după degete? De când mă útiu niciodată nu am fost aemniĠat sau furat de un român, întotdeauna de Ġigani, iar rasismul nu s-a născut aúa deodată s-au gândit toĠi „hai să fim rasiúti”, e replică – răspunsul la comportamentul unei naĠii, oameni care NU MAI SUNT DEMULT MINORITARI. ‘Fuck you, you come up with one thousand idiotic questions, what the fuck are you hiding beneath your fingers? For as long as I know I was never threatened or mugged by a Romanian, always by gypsies, and racism was not born all of a sudden everybody thought “let’s be racist”, it’s a replica – the answer to the behaviour of a nation, people who are no longer a minority’ Mihai: eu am priteni Ġigani cu care am împărĠit úi o pâine dar în acelaúi timp îmi feream buzunarul, hai să nu ne mai

154

Chapter Five

prefacem atât că sutem buni când útim cu toĠii care-i treaba. ‘I have Gypsy friends with whom I shared bread but at the same time I was guarding my pocket, let’s not pretend to be good when we know how things work’ Meriem: păi înseamnă că ei degeaba erau prieteni cu tine, că tu nu erai cu ei. ‘well it means that they were friends with you for nothing, because you were not their friend’ Mihai: euzic să nu vorbeúti ca o proastă fără să cunoúti. le-am dat din hainele mele, i-am ajutat mereu când au avut nevoie deci nu mai vorbi ca proasta, idioato. Problewma e la tine te prefaci aici ca fiind mare păsătoare de minorităĠi úi de fapt freci cucu că n-ai intrat prea des în contact cu ei altfel útiai despre ce e vorba. Sunt cei mai naúpeĠi căcaĠi. ‘I think that you should not speak like an idiot without knowing things. I gave them some of my clothes, I always helped them so stop talking like you are stupid, you idiot. The problem with you is that you pretend to care a lot about the minorities when you are actually talking shit as you didn’t have much contact with them otherwise you would have known what this is about. They are the worst pieces of shit’ Mihai: eu nu eram prieten cu ei pentru că mă furau? handicapată eúti. ‘I wasn’t their friend because they were stealing from me? You’re such a dumbass’ Mihai: rău de tot. ‘really bad’ Cosmin: bă da’ nu poĠi să îĠi exprimi punctul de vedere fără să fii ciobârdău? chiar dacă e o diferenĠă în ideologia voastră, n-ar trebui să te exprimi aúa, Ġigănesc. ‘dude, can’t you express your point of you without being an ass? Even if there is a difference in your ideology, you shouldn’t express yourself like that, like a Gypsy’ Meriem: :)))))))))))))))))))))) hai că eúti amuzant, rămâi tu cu fascismul tău úi lasă-mă pe mine în universul meu în care toĠi oamenii sunt egali, fii tu ăla mai deútept.

Adolescents as Language Innovators

155

‘you are really funny, stay with your fascism and leave me to my universe where all people are equal, you be the smart one’ This conversation was collected from Facebook on the 21st of March 2013, at 12:30 pm60. Meriem (18 years old, female) posted a photograph on her wall which contained an antiracist message. The other two participants in the conversation were (Mihai, 18 years old, male) and Cosmin (19 years old). I included this conversation in the analysis for two reasons: (i) because it corroborates my assertion that the Rroma people are not easily tolerated by Romanians and (ii) because despite this ethnicity issue, teenagers borrow and use slang words from Rromani language. According to Mihai, the Rromani are to be blamed for many things, especially violence, yet he still uses slang words borrowed from their language, i.e. naúpet ‘not cool’. He also insults his addressees thus exerting his masculinity and toughness. The crux of the matter is that language he uses (swear words, cursing, taboo) can be seen as aggressive and extremely offensive, especially when talking to a girl, in this case Meriem. Cosmin even blames Mihai for using Rromani language, thus associating taboo language, swearing and cursing with Rromani, which is actually quite interesting given the fact that he considers himself neutral, i.e. with no ethnical prejudices. The word naúpa deserves special attention, since it is the most frequently used slang word by adolescents. Zafiu (2001) discusses the etymology as well as the use of this word. The Romanian linguist argues that naúpa is the reduced form of naúparliu, which is considered to come from the Rromani adjective nasvaló, meaning ‘sick’. It is also believed that nasvaló might be the source of the older (adjective, noun, adverb) nasol ‘bad’, which is no longer considered slang. However, there is another hypothesis according to which nasol comes from the Rromani word nasul which means ‘bad’, ‘ugly’. Zafiu also notes that nasvalo (or nasfalo) transformed into nasfarliu which later became naúparliu. The antonym of naúpa appears to be miúto ‘cool’ also very frequently used by adolescents. In another paper, Zafiu (2009) argues that miúto is, without doubt, a fundamental word in Romanian slang, as it expresses a positive appreciation. The word was borrowed from Rromani and has been in the Romanian language quite some time now, despite Graur’s (1934) assertion that the word is about to get out of use and be replaced “actuellement, 60 More information about the Romanian data collection for this chapter is discussed in section 5.4.

156

Chapter Five

pourtant, il semble que le mot n’est plus en faveur, sa place étant prise par úucar” (Graur 1934, quoted in Zafiu 2009). Returning to the notion of slang61, Eble (1996) puts emphasis on the social aspect and claims that slang is: an ever changing set of colloquial words and phrases that the speakers use to establish or reinforce social identity or cohesiveness within a group or with a trend of fashion in society at large (Eble 1996, quoted in Stenström 2014: 13).

To put it differently, slang serves a double role, that of establishing group membership or showing affiliation to a group or as a limit to other social groups. Partridge (1933) identifies several reasons for using slang, some of which are related to adolescent talk: (i) to be different; (ii) to create an atmosphere of friendliness or intimacy; (iii) to show group membership (e.g. the slang used in a certain school); (iv) just for fun As mentioned above, adolescent slang can be divided into two groups: specific slang, which is used in a particular place or by a particular group or even a community of practice, to show affiliation and to identify to a group (school slang) and general slang words which are used by the entire community, without any particular value.

5.3. Swearing, taboo language and slang in the speech of London adolescents This subchapter is based on conversations found in The Bergen Corpus of London Teenage Language (COLT). COLT is the first important English Corpus with a focus on the speech of adolescents. The corpus was compiled in 1993, consisting of spoken English gathered from 13 to 17year-old adolescents living in different parts of London. The corpus has half a million words and it is part of the British National Corpus. Also, the conversations as well as the analysis for London adolescents, as far as 61

Jay (1992: 6) considers slang to be a subcategory of ‘dirty words’, and provides the following definition of slang “an informal nonstandard vocabulary composed typically of coinages, arbitrary changed words, and extravagant, forced, or facetious figures of speech”.

Adolescents as Language Innovators

157

swearing is concerned, are taken from Stenström et al. (2002) whose analysis is based on COLT.

5.3.1. The Bergen Corpus of London Teenage Language COLT is based on naturally-occurring conversations between adolescents. They were not disturbed by the presence of a tape recorder and were willing to participate. The research team had an advantageous status, just like I did when I collected my corpus for the (ing) variable), that of foreigners. They were Norwegians, from Bergen University. However, the team also consisted of six other well-known linguists with experience in the field and in corpus-building. The adolescents that were recorded come from different social and ethnic backgrounds. The conversations from COLT also show that teenagers are interested in the social differences that exist between them. Consider the following examples: (76) Brett: Come on. Nick a few. Nicolas: No. We’re upper class teenagers Brett. (77) Anne: You common types wouldn’t know what intellectual people wear. Carola: Camden town residents. (78) Now for some fucking dirty swear! Woooooh! You fucking bitch! You Irish bastard. (Source: Stenström et al. 2002: 14-15) These examples reveal that adolescents are aware of the social differences (social class, ethnicity) that exist between them, and they are able to recognise and appreciate the linguistic features that are associated with such differences. The adolescents that were recorded come from all the London boroughs with ages between 13 and 17 (Stenström et al. 2002).

5.3.2. Adolescents talking about sex: gender-related differences One of the most frequent topics for discussion among adolescents was sex. In such conversations taboo words are expected to be used as well as slang.

Chapter Five

158

(79) Vic:

Your Mum’s got so many hairs on her fanny Tarzan don’t know which one to swing on! Wyatt: I’ve he=, I’ve heard that one before, but like...it’s just like this one innit? Your Mum’s fucking fanny...is so wrinkly...the crabs...have got walking sticks. (...) Wyatt: Go tell your Mum to to=, stop changing her lipstick cos she’s making my cock look like a rainbow. Will: Yeah, tell your Mum to give me...the twelve pence, er, er the two P cos erm...it wasn’t, it wasn’t worth last night. Wyatt: Tell your Mum to give my money back. (Source: Stenström et al. 2002: 38)

In this conversation Wyatt (13 years old) and his friends engage in insulting each other about their mother. It is interesting that from such an early age they talk about coitus and male as well as female genitalia with such ease. Insulting each other’s mother could also be viewed as something taboo since mothers are considered to be sacred. Girls, on the other hand, also discuss about sex, but they do it somewhat differently: (80) Sharon: What, oral sex? Allie: I wouldn’t mind provided I had flavoured condoms

laugh Susie: I don’t think there’s nothing the matter with oral sex [as long as,] Allie: [There is nothing the matter it’s just another form of sex] Susie: as long as he’s got a [clean penis!] Allie: [It’s, it’s] safer! It’s sa= safer sex, you know. Susie: No, I don’t think there’s anything the [matter with oral sex.] Kate: Would you ever toss someone off. Toss your boyfriend off? Susie: What’s that? Allie: Wanking for him. Susie: I’ve already done that (Source: Stenström et al. 2002: 40-41)

Adolescents as Language Innovators

159

Allie (14 years old) and her friends are discussing about oral sex and masturbation (considered taboo by Allan and Burridge 2006) and share their experiences. There is no insult or swearing involved in the conversation, which shows the girls’ closeness and their solidarity towards each other, given the very intimate and delicate topic of their conversation. If we were to compare the two conversations given in (64) and (65), it is clear that girls are more polite than boys, at least at first glance, as they refrain themselves from using dirty words62, whereas boys do not. Regarding male genitalia, boys refer to it using the dysphemistic word cock, while girls prefer the orthophemistic word penis. At some point, Kate uses the expression toss your boyfriend off, which is slang for masturbation, something that Susie does not understand, so Allie joins the conversation and explains it for her by providing another slang expression, wanking for him. A closer look at the boys’ conversation reveals that they are actually competing against each other in the use of insults. They view it as a competition in which the one that uses the toughest insult or profanity wins. Talking about someone’s mother’s genitalia and comparing it with different animals can be viewed as a profanity. So, the fact that that boys and girls use different words for male genitalia can be considered a gender difference and another conclusion that can be drawn at this early stage is that girls appear to be more polite than boys, who view everything as a competition and in participating in this competition they pay no attention whatsoever to the words they use, or how they insult their interlocutor as long as they are going to win. These are cogent arguments in favour of the hypothesis that girls and boys use taboo words as well as slang differently.

5.3.3. London adolescents and the use of slang Any discussion of the language used by adolescents must include slang. As previously mentioned, slang is an umbrella term which encompasses

62

Allan and Burridge (2006: 40-41) note that the part of taboo language labelled “dirty words” refers to terms describing: “(i) bodily organs concealed by bikinis and swimming trunks – because they are organs of sexual desire, stimulation and gratification and/or used for micturition and defecation; (ii) activities involving these SMD (sex, micturition, defecation) organs; (iii) bodily effluvia issuing from these SMD organs”. Another interesting view is that of Read (1977: 9) who states that “the ordinary reaction to a display of filth and vulgarity should be a neutral one or else disgust; but the reaction to certain words connected with excrement and sex is neither of these, but a titillating thrill of scandalized perturbation”.

Chapter Five

160

many subcategories. In this section I will refer to proper slang and dirty words. In COLT the top ten proper slang words are: A closer look at the top ten proper slang words reveals that the majority of words are nouns, and not adjectives, verbs or adverbs. The majority of proper slang words in this list are used to refer to a male speaker (e.g. man, mate, bloke, guy) which should not come as a surprise given that terminology is male-oriented and women are excluded when the generic use of man or bloke or even the personal pronoun he are used. As far as dirty slang words are concerned, in COLT the top ten dirty words are: Word man sad wicked mate bloke guy cool massive rough quid Total

Frequency 358 140 132 84 84 77 75 58 56 55 1119

Table 5.26. The top ten proper slang words in COLT (Source: Stenström et al. 2002: 70) Word crap arse dick bastard bitch take the piss fuck wanker suck cunt Total

Frequency 96 65 62 61 51 50 47 40 39 36 547

Table 5.27. The top ten dirty slang words in COLT (Source: Stenström et al. 2002: 71)

Adolescents as Language Innovators

161

Just like with proper slang words, the majority of dirty slang words are nouns, and there are three words referring to a male (dick, bastard, wanker) and only two referring to a female (bitch and cunt). They can also be viewed as taboo words related to sex, describing male and female genitalia (dick, cunt), sexual intercourse (fuck) and promiscuity (bitch, wanker). It was also mentioned that boys and girls use swear words and slang differently. Now let us see how they talk about each other: (81) Boys about boys But the guy is an arsehole He’s a gay bent bastard Unbelievably crap person and sad Acts like a complete dick to try and get attention Miguel’s a sad faggot He’s a knob Oh what a penis He’s so shit He’s a fat wanker (Source: Stenström et al. 2002: 72) Words referring to the male genitalia are preferred by boys when talking about boys. It is noteworthy that the orthophemistic word penis appears to be used, as well as the dysphemistic word dick. As far as girls are concerned, they refer to boys as: (82) Girls about boys He’s a complete arsehole Rick’s a cunt He’s the biggest knob out, he is a real cock He’s a right little shit He’s a slag. (Source: Stenström et al. 2002: 72) Interestingly enough, girls refer to boys as a cunt or a slag, words associated with females. One explanation for using such words is to make them appear powerless and weak. By doing so the power balance inclines in the girls’ favour. Hornoiu (2008: 55) argues that language can also be used as an indication of the status of a relationship and power imbalances, which might account for the fact that words such as whore, bitch, slut,

162

Chapter Five

slag, old maid have acquired more negative connotations that their corresponding male variants stud, bastard, bachelor. (83) Boys about girls She’s a bitch bitch She’s a stupid cow She’s a dick She’s a right fuck She’s a tart (84) Girls about girls Kelly’s a little bitch She was such a soppy little cow She is such a rough tart She’s one dirty whore (Source: Stenström et al. 2002: 73) By comparing the way boys talk about girls and girls about girls, we notice that there are actually no differences, as they use almost the same words (e.g. bitch, cow, tart). These findings, however, run counter to McEnery’s (2006: 571) analysis who claims that male and female speakers swear differently. His study is based on spoken data from the British National Corpus, consisting of ten million words. He argues that words like fucking, fuck, Jesus, cunt and fucker tend to be used more by male speakers while female speakers use words like God, bloody, pig, hell, bugger, bitch, pissed, arsed, shit and pissy. A closer examination at the words reveals that male speakers prefer stronger swear words then female speakers. A remark is in order, as McEnery’s (2006) study does not study the language of London adolescents and it does not focus only on adolescents but also on older speaker. He argues that with age, there appears to be a decrease in the use of swear words, speakers aged 60+ showing a low rate of swear words. Regarding London adolescents, it appears that girls talk about boys using strong swear words by calling them cunt, cock, shit, slag, etc. Boys, on the other hand, refer to girls as bitch, cow, dick, tart, etc. It seems that boys are called using female genitalia and girls the other way round which is quite interesting considering that these slang words are really strong. In a similar vein, Risch (1987) is surprised to find in her English data the use of male terms for female and vice versa. She notes that these terms are losing their semantic features [+male] and [+female], which could explain the use of male terms to describe females and the other way round in COLT.

Adolescents as Language Innovators

163

To sum up this discussion, we can highlight that if we take two sociolinguistic variables into account, i.e. gender and age, the use of slang is a complex phenomenon, and reveals differences as well as similarities in the use of slang words, be they proper or dirty.

5.3.4. Gender variation in COLT So far it was revealed that girls swear as much as boys do, and they have no problem in using taboo words. They are aware of the fact that they swear, but since they were guaranteed anonymity they have no problem in swearing as much as possible, even if they are recorded. Let us consider the following conversations: (85) Regina: Tommy I want you to piss off, because I have an exam tomorrow Tommy: But I don’t care...She don’t know what to say. laugh. You’re allowed, fuck off, you’re allowed to say that. You’re allowed to swear as much as you like. Regina: Are you? Tommy: Yeah. Regina: Fuck fuck fuck fuck! laugh I’m gonna fail GCSE tomorrow gonna fail! (86) Bradley: It’s fucking brilliant Johnny: The one who’s [down here as well?] Bradley: [Shit they didn’t record] that did they? Nicolas: It doesn’t matter. Bradley: Well I said a rude word. Kerry: laugh Nicolas: [Well no it doesn’t matter er] Brett: [Anonymity guaranteed.] Kerry: it doesn’t matter [nobody’s nobody’s got ano= anonymity] Nicolas: [Anonymity] many: laugh pissing hell Johnny: Pissing hell! Brett: Anona=anona= Nicolas: anonymity Kerry: Yeah.

164

Chapter Five

Nicolas: is guaranteed so it doesn’t really matter. (Source: Stenström et al. 2002: 79) The COLT teenagers have no problem whatsoever in using foul language and are not offended by it. They have no problem in using swear words even though they are aware of the fact that they are recorded. Take the conversation in (85), for instance. Regina is a bit hesitant when it comes to swearing as she does not know if she is allowed to use it, but Tommy convinces her that she can, by using the f-word and by saying that she can swear as much as she wants. It is interesting that upon hearing this Regina uses the f-word four times, as if she is rejoicing that she can swear. Boys are a bit reluctant at first also, but when they are certain that they are under the protection of anonymity they have the same reaction as girls. What is noteworthy here is the fact that COLT adolescents have internalized the idea that they are responsible for their linguistic choices and they are also interested in the way they are perceived by society. Stenström et al. (2002) analyse slang in relation to gender as well as age and they come up with the following results: Gender male female

Proper slang 1724 916

Dirty slang 796 523

Slang words 2520 1439

Table 5.28. Slang and gender (Source: Stenström et al. 2002: 73) Male speakers use slang more frequently that female speakers. While the difference is quite consistent as far as proper slang words are concerned, when it comes to the use of dirty slang words the difference is less pronounced. Allen (1998: 882) proposes an elegant solution to account for this phenomenon, and states that the taboo expressions that “belonged to the restricted speech of male culture are today losing their exclusiveness and isolation”. Let us now see whether the results change if we add another sociolinguistic variable, besides gender, i.e. age.

Adolescents as Language Innovators Age group 10 –13 14 –16 17 –19 Total

Proper slang Male Female 513 102 1009 574 113 194 1625 870

165

Dirty slang Male Female 303 34 437 388 28 83 769 506

Table 5.29. Slang, gender and age (Source: Stenström et al. 2002: 74) Both genders use proper slang more frequently than dirty slang. In the 10– 13 age group male speakers use proper slang five times more often than female speakers. The situation changes when it comes to dirty slang, boys using almost ten times more frequently dirty slang than girls. This reveals that, in this age group, girls are more polite, more status-conscious than boys, or they have not been exposed to dirty slang as much as boys were, since at such an early age their parents usually impose a strict curfew on them. Given that slang is also known as street language, due to the lack of exposure, girls do not use it as frequently as boys do. The situation, however, changes in the next age group, 14–16, with girls gaining “territory” and using dirty slang almost as often as boys do. Male speakers in this age group use dirty slang in 2.0 % of the cases while girls in 1.7%. So, it appears that, with age, girls become more aggressive, compete with boys and are no longer interested in the way they are perceived by society. Another explanation for the rise in the use of dirty slang by girls could be explained by their adherence to a certain group/gang or to the “street culture”. As far as proper slang is concerned, the male speakers in this age category are the ones who use it more often than any other age category. In the third age group, 17–19, there is a change in the use of both proper slang and dirty slang. If in the other age groups male speakers were the ones leading in the use of both categories of slang, in the last age group girls use it more frequently than boys. If we look at the female speakers in the three age categories, we notice that the 17–19 year-olds dominate in the use of both proper slang and dirty slang. Also, girls use dirty slang almost three times more often than boys do. Stenström et al. (2002: 74) note that with age there appears to be a decrease in the use of slang, especially from the age of 20 onwards. Apart from the gender differences that were identified, it seems that there are also age-related differences in the use of slang, with younger girls (10–13) using proper slang as well as dirty slang seldom and teenage girls (17–19) dominating the scene, and using slang more frequently than boys. A plausible explanation to account

Chapter Five

166

for this is connected to the concept of power. By using dirty words female speakers gain more power and status. Girls also discuss the importance of dirty slang and what they consider to be the strongest one. Consider the following conversation: (87) Susie:

We come to an agreement and that’s our biggest word, fuck. I’ve got everyone’s, oi, Abdullah! What’s your favourite swear word? ... Caryl: Bollocks! Susie: Mine’s like, cunt and things like that. Allie: Usually shit and fuck, I think are my vocabulary. Susie: Yeah, or oh shit, you cunt! Allie: Or shite is also another one Susie: Yeah, or, or, or, budging hell (Source: Stenström et al. 2002: 83-84)

For Susie the strongest dirty word is fuck. She also inquires the girls’ favourite swear word and they come to a mutual conclusion: shit, fuck and cunt. These words can be viewed as strong expletives in Bailey and Timm’s (1976) and de Klerk’s (1991) terminology. Under the label ‘strong expletives’ we find words such as: asshole, bastard, crap, fuck, shit while words like blast it, darn, holy moly, jeez, nuts, rats are considered ‘weak expletives’ (Bailey and Timm 1976: 440-441). As a conclusion, there are gender-related differences as far as foul language is concerned. I discussed the concept of “slanguage”, introduced by Stenström et al. (2002) encompassing proper slang, dirty words (swearing and taboo language), vogue words, etc. It is vital to keep an open mind regarding the conversations from COLT. The fact that the data from the corpus dates back from 1993, i.e. twenty years ago, means that there are also other slang words that appeared from 1993 onwards and are not found in this corpus. Stenström et al. (2002: 105), who were the principal investigators in the COLT corpus, claim that “the dirty slang words, we assume, are largely the same in the year 2002 as in 1993”. Proper slang appears to be used more often than dirty slang, which reaches the peak in the 17–19 age group. Regarding gender, boys were found to use more slang than girls, dirty slang included, although in the 17–19 age group girls used dirty words more than boys. However, swearing was more common in the speech of boys belonging to the 10–13 age group.

Adolescents as Language Innovators

167

5.4. Teenagers as language innovators: swearing and slang in Romanian DicĠionarul Explicativ al Limbii Române63 defines slang as a conventional language, used especially by homeless people or wrongdoers so that they are not understood by the rest of the society. This subchapter argues that Romanian adolescents should be treated as real language innovators since they are responsible for the creation of new slang words. I discussed the use of slang words such as naúpa and miúto earlier in this chapter and now I focus more on dirty slang words and swearing.

5.4.1. Data collection and methodology This analysis is based on a corpus of 6500 words, collected using two different methods. The first part of the corpus consists of three hours of conversations which were video recorded in December 2011. The recording focuses on a group of teenagers spending New Year’s Eve in Buúteni. The group consists of eight teenagers (four boys and four girls) belonging to one age group 17–19. A villa was rented by the teenagers from December 28, 2011 until January 2, 2012 for the New Year’s Eve party and they were recorded every day. This data collection method is reminiscent of Labov et al.’s (1968) study in Harlem where a house was rented for an academic year and teenagers were taken in a ‘bugged bus’ for recording sessions at the university. For the video recording a Sony DCR-HC44E Digital Video Camera Recorder was used, thus obtaining a high-quality recording. I transcribed the conversations in Microsoft Word. The second part of the corpus is made up of conversations gathered from Facebook in March–April 2013. I only managed to collect ten such conversations as it is really difficult to identify long conversations. On Facebook the conversations consist of comments given by different people on different posts. For example, someone posts something on their wall and their friends comment on that post. I chose the comments that were given by adolescents only, with ages between 13 and 19. The reason I chose to include these comments from Facebook in the corpus was to have a more detailed account of the use of slang and swearing by adolescents in both spoken and written Romanian. The analysis, however, draws mainly on the data from the video recording.

63

I used the online version www.dexonline.ro. (The Romanian Dictionary)

Chapter Five

168

5.4.2. The speakers in the ConstanĠa corpus As mentioned above, the speakers belong to one age group 17-19 and they live and study in ConstanĠa. They come from families with a good socioeconomic situation and they all go to good schools. Four of them are grade 12 in high school and the other four are first year undergraduate students at Ovidius University of ConstanĠa. More information about them is given below in Table 5.30: Name Adi

Age/Gender 17/M

Ana

18/F

Andreea

19/F

Cosmin

18/M

Cristian

19/M

Mihaela

19/F

Roberta

17/F

ùtefan

19/M

School Traian High School Traian High School Ovidius University Mircea High School Ovidius University Ovidius University Decebal High School Ovidius University

Lives in ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa ConstanĠa

Table 5.30. Information about the Romanian speakers When the video recording was made the purpose of it was not an academic one, but just to have some good memories from that trip to the mountains. About 85% of the recording depicts the teenagers playing a game – mine. So, given that the video camera was put in a corner of the room at about one metre from the speakers, they forgot that it was there and it was recording and they used swear words quite frequently. Getting teenagers to swear in front of a camera or on tape is a difficult undertaking. Since I knew all the teenagers that were recorded I told them about my PhD thesis and asked them to allow me to use and analyse their conversations. They signed a consent form which, among other things, granted anonymity.

Adolescents as Language Innovators

169

In the conversations some of the teenagers appear to be annoyed by the presence of the video camera. Consider the following examples: (88) Roberta: Hai mă ce dracu’ mă filmezi ‘Why the hell are you filming me’ (89) Mihaela: Hai nu mă mai filma aúa ‘Come on stop filming me like this’ Only girls appear to be disturbed by the video camera, but this situation occurs in the first three minutes of the recording. This was the only time they acknowledged the presence of the camera and seemed to be disturbed by it. Another explanation is that girls are interested in the way they are perceived by society and if they were not properly dressed or not wearing make-up could explain their refusal to be recorded.

5.4.3. Dirty words: results Regarding the use of foul language in the ConstanĠa corpus the dirty words that occur most frequently are as follows: Word pula căcat muie/muist găoz(ar) fut suge lindic cur labă păsă Total

Frequency 73 12 5 3 3 3 2 2 1 1 105

Table 5.31. Top ten dirty slang words in the ConstanĠa corpus The dysphemistic word pula64 ‘cock’ is the most frequently used dirty slang word in the ConstanĠa corpus. Let us look at some of the contexts in which this word occurs: 64 Etymologically, the noun pula, which is a vulgar term for penis, comes from the Megleno-Romanian and Aromanian word pulă, from the Latin word pulla,

170

Chapter Five

(90) Cristian: Unde sunt mă sticlele alea-n pula mea, în care cutii erau? ‘Where the fuck are those bottles, in which boxes were they?’ (91) Adi: Sunt beĠi...a căzut lemnu’ ăla băga-mi-aú pula-n el ‘They’re drunk...that piece of wood fell fuck it’ (92) Ana: Eúti puternic în pula mea hai ‘You’re strong fuck it come on’ (93) Adi: Pulă puĠă penis ‘Cock dick penis’ (94) Cristian: Să-mi sugi pula ‘Suck my dick’ (95) ùtefan: Cu cuvinte d-astea normal că nu ghiceúti cum pula mea? ‘You can’t guess with this kind of words how the fuck?’ (96) Roberta: Îmi bag pula când Ġi-am zis să mimezi pipă ai zis eeeeee ‘Fuck it when I told you to mime a pipe you said eh’ (97) Cristian: Are gust de pulă bă muistule ‘It tastes of cock you cocksucker’ (98) Cristian: Cum mă juma’ de minut du-te-n pula mea de aici ‘Half a minute? Fuck you’ These are some of the contexts in which the word pulă ‘cock’ is used. The meaning of sentence (90) is that of frustration as Cristian is unable to find some bottles. The message conveyed does not refer exactly to the male genitalia, or to the sexual act but it expresses annoyance, just like in sentence (91). In (92) in intended meaning is that of ‘I don’t know’ or ‘whatever’ so here it can be viewed as a linguistic form which strengthens the force of the utterance, with the risk of sounding impolite. However, in this group, due to the frequency with which the word pula is used, the dysphemistic word loses its force, its vulgarity, and it is used as a substitute to express different feelings: annoyance, frustration, disdain, impatience. In sentence (94), on the other hand, the word is used with its intended meaning, referring to the act of oral sex. It is used as a swearing but, again, due to the familiarity between the speakers it loses it powerful meaning. It expresses the hatred or annoyance of the speaker because his interlocutor feminine of pullus (=chicken). The fact that pullus means ‘chicken’ (pui in Romanian) could account for expressions like Ce faci, pui? ‘How are you, dear?’ or Du-te-n puii mei ‘Damn you’ (Source: www.dexonline.ro).

Adolescents as Language Innovators

171

is not paying attention to what he is saying. This expression is also considered to be taboo as oral sex is deemed inappropriate by society. In (98) the expression du-te-n pula mea ‘fuck you’ is used as a rebuke and it is addressed to a friend who is accused of dishonesty or is blamed for his lack of solidarity. The second most frequently used dirty word in the ConstanĠa corpus is căcat ‘shit, crap’. The difference in frequency between the two dirty words is quite impressive, which shows that pula is a much stronger swear word that căcat. In COLT the situation was different, with crap being used more often than dick. Let us look at the following occurrences: (99) Roberta: (100)Ana: (101)ùtefan: (102)Cristian: (103)Roberta:

Ce căcat e mă? ‘What the fuck is it?’ Ce căcat o fi? ‘What the fuck is it?’ Nu útiu ce căcat avea ‘I don’t what the fuck was wrong’ La fel când avea ăla te căcai pe tine ‘When we had that you were boasting’ Nu mai mânca căcat ‘Stop eating shit’

A closer look at the examples (99)–(103) reveals that in none of them the word shit is used with its intended meaning, i.e. bowel movement. In (99) the meaning is ‘What the hell do you want’ the word shit is used to add force to the inquiry, as it is the case in (100) and (101). It expresses exasperation, annoyance and even anger. In (102) te căcai pe tine is a very informal expression meaning to boast, so again the word shit is used for a different purpose. In (103) a mânca căcat is again an expression meaning to tell lies. So, to sum up, in none of these examples the word shit is used with its original meaning. Given its status as a dirty word, it adds power to the meaning of the sentence, to what the speaker is trying to convey.

5.4.4. Proper slang: results Proper slang words were not used as frequently as dirty words. The majority of slang words used by the teenagers have to do with the activity of stealing or the person who performs this activity. Let us consider the following conversation:

Chapter Five

172

(104)Mihaela:

Cosmin: ùtefan: Cristian: ùtefan: Andreea:

Cosmin: Mihaela:

Cosmin: Roberta: Cristian: ùtefan: Roberta:

Dar nu mori nu poĠi fi ucis nemuritor invincibil ‘But you don’t die you cannot be killed immortal invincible’ ùterpelitoare ‘Pickpocket’ HoĠ hoĠ de buzunar ‘Pickpocket’ Ciorditor ‘Pickpocket’ Manglă ‘Stealing’ Roberta m-aúteptam să mimezi altfel ‘Roberta I was expecting you to mime differently’ HoĠ de buzunar ‘Pickpocket’ Hai Cosmin că m-aúteptam la mai mult de la tine ‘Come on Cosmin I was expecting more from you’ E neologism ceva? ‘Is it a neologism?’ Cam e ‘Kind of’ Ciorditor pula mea ‘Pickpocket fuck’ Ciordilaú ‘Pickpocket’ Bă nu d-astea ciordilaú căcat ‘Not words like pickpocket fuck’

All these words designate either a thief or the activity of stealing. The words denoting the person who is involved in such activities end in -or for masculine or -oare for feminine (e.g. ciorditor, ciorditoare, úterpelitor, úterpelitoare). The verb a ciordi ‘to steal’ is borrowed from the Rromani üoráw ‘to steal’ (Sarău 1992, quoted in Zafiu 2010: 152-153). Zafiu (2010b: 153) notes that in Rromani the verb üoráw has a rich word family which includes the word üor (written also as cior) ‘thief’. The word is believed to be the source for the pejorative meaning of cioară, describing the Rroma people. The word ciordilaú, however, is not found among the forms listed by Zafiu to come from a ciordi, therefore we could consider it

Adolescents as Language Innovators

173

a newly formed word. The word manglă comes from the verb a mangli which is used almost exclusively with the meaning of ‘to steal’ so it is a synonym for a ciordi. Zafiu (2010b: 231) provides the following examples with the word manglă ‘the act of stealing’: Nu útie dacă este o “manglă” sau nu ‘He doesn’t know whether it is something stolen or not’. These proper slang words are used mainly by boys, as it can be noticed in the examples provided. In (104), Roberta does not approve the use of the word ciordilaú and she even condemns ùtefan for using it. To reinforce this idea she uses a strong expletive (i.e. căcat ‘shit’), to show her discontent. The other slang words that are used by teenagers are those denoting a crazy person: (105)ùtefan: Cosmin: Cristian:

Tâmpit, sandilău, zăpăcit ‘Crazy’ Grăbit, aiurit, ciumpalac ‘Hurried, scatterbrain, crazy’ Bă Popescule nu fi mă bulangiu ‘Popescule don’t be such an arsehole’

The word sandilău is listed in the Romanian dictionary as belonging to the teenagers’ slang. These words are used quite frequently by teenagers. A plausible explanation to account for the use of such words is that they avoid using words like nebun ‘crazy’ which can prove to be quite offensive. Unlike nebun, sandilău, does not mean a person who has neurological problems and suffers from a malady, but has a deviant behaviour, meaning that he is adventurous. It acquires a rather positive meaning, just like the word ciumpalac. However, it is noteworthy that the word nebun has recently acquired a connotative meaning. A fi nebun după cineva ‘to be crazy about someone’ has a positive meaning, and does not refer to the mental illness, to be demented, but to love someone. Also, a question like Eúti nebun ‘Are you crazy’ is used to show surprise, disapproval or even admiration towards someone’s deeds or assertions. A face pe nebunul ‘Act like a crazy person’ refers to a haughty person, who is looking down on others or it is used to describe a person who is not satisfied with something. The adjective diliu, dilie ‘crazy’, or the verb form a dili is worth mentioning, although they do not occur in the corpus. However, the verb form is attested in the data collected from Facebook: V-aĠi dilit. The word

174

Chapter Five

is frequently used in different contexts65. The word is borrowed from Rromani: diló, dilí ‘crazy’. It is used euphemistically instead of nebun: Eúti diliu la cap? (Zafiu 2010b: 171), and it is included in all the recent Romanian slang dictionaries (Tandin 1993, Croitoru Bobârniche 1996, A. and G. Volceanov 1998). Zafiu (2010b: 171) argues that from the form diliu different amusing forms have been derived: dilimache, diliman, dilimandros and lately the form dilău started being used, with the same meaning (úofer dilău merge pe contrasens ‘crazy driver drives on the opposite side of the road’). We could argue that the word sandilău is an improved and new version of dilău, given the remarkable resemblance between the two words, dilău being incorporated into sandilău. Zafiu (2010b: 172) highlights that insanity is one of the most productive areas regarding word formation, especially as far as slang is concerned. It is actually expected that in a language of in-group solidarity the negative label to be that of insanity, i.e. a behaviour which is deviant from the group’s rules and regulations. Another interesting slang word that is attested in the ConstanĠa corpus is bididiu. According to Zafiu (2010b: 112) in a small glossary which appeared in the Dilema magazine (6-12.07.2001) the noun bididiu was listed as meaning ‘sucker’ or ‘cool’. The denotative meaning of the word bididiu provided by the Romanian Dictionary is “young horse, agile and beautiful”. The word is of Turkish origin, from bedevi. The word is now used to describe a young person, who is usually small and even cute, E un băiat bididiu, E o fată bididie. As far as the word bulangiu ‘arsehole’ is concerned, this does not denote a crazy person, but a person who is not willing to perform an activity. It is used as a reproach, and by referring to a person with a slang word, bulangiu in this case, the emotional bond between the participants in a discussion or dialogue is reinforced. Another relevant example is muist ‘cocksucker’. Consider the following examples: 65

Zafiu (2010b: 171) notes that the verb a dili has other meanings besides ‘to be crazy, to become crazy’. In the 1930s it was used with the meaning of ‘punching or hitting someone’, thus being a transitive verb: Mi-a dilit unu o labă peste muie, de nu mai vedeam ‘A bloke punched me in the face and I couldn’t see anything’. A third meaning identified by Zafiu is connected to the IT jargon, dilit, from the English word ‘to delete’. To this list of meanings, we could add a new one, which is not given by Zafiu, that of something or someone which/who smells really bad: Dileúte peútele ăla de nu poĠi să stai în bucătărie ‘That fish smells so bad that you cannot stay in the kitchen’ (Source: www.facebook.com). A fifth meaning identified is that of ‘being broken’: S-a dilit telefonu’ meu ‘My phone is broken’. In this case the verb is reflexive. (Source: www.2in.ro).

Adolescents as Language Innovators

(106) Adi: Roberta:

175

Nu fi mă muist ‘Don’t be a cocksucker’ I-ai zis muistule ‘You told him you cocksucker’

Here, the word muist ‘cocksucker’ does not refer specifically to the person who engages in oral sex but it is used to describe a person who refuses to do something or has already done something that he was not meant to. The form muist is derived from the noun muie which has several meanings: (i) describing a person’s face; (ii) oral sex; (iii) a person. (107)Ce muie avea ăla de la magazin. ‘Did you see that guy’s face’ (108)N-ai vrea să-mi faci o muie? ‘Would you like to give me a blowjob? (109)Muie, hai cu mine să iau niúte Ġigări ‘Dude, come with me to buy some cigarettes’ (Source: www.peteava.ro) These were the proper slang words that were used by the teenagers in the ConstanĠa corpus. Hornoiu (2008: 122) also provides some examples of slang terms used by adolescents living in ConstanĠa, and she astutely outlines that the frequent use of slang terms by teenagers makes them a marker of age rather than gender: (110)Raluca: am stat prin cartier la o caterincă ‘we hung around the neighbourhood chewing the rag’ (111)Raluca: mamă úi m-am ofticat ( ) dacă eúti bulangiu ‘man it ate me ( ) if you’re an arsehole (112)Raluca: aĠi văzut voi două ciumece ‘I’ve told you smart ones’ (113) I: úi gagica-i de gaúcă ‘and the chick is really cool’ (Source: Hornoiu 2008: 122) The noun caterincă means irony, mocking something. Zafiu (2010b: 142) notes that the term caterincă ‘barrel organ’ is borrowed from Ukrainian, where ɤɚɬɟɪɢɧɚ acquired this meaning from a well-known song, interpreted by barrel organs: the German song Charmante Katharine. Zafiu argues that caterincă first became a metaphor for ‘mouth’, and then through analogy (‘mouth’ – ‘noise’) it acquired a new meaning. A similar

Chapter Five

176

scenario is suggested by Iordan (1975: 316), according to whom flaúnetă ‘barrel organ’ was recorded as a slang metaphor for ‘mouth’. This would support Zafiu’s hypotheis i.e. ‘mouth’ – ‘noise’ – ‘the treacherous things that you say using the mouth’. The word ciumece ‘cool’ the plural word of ciumec (also known as ciumeg) should be considered a vogue word as it was extremely used for a limited period of time and then it was gradually replaced by miúto, marfă, calumea, beton. To sum up this discussion, it is worth mentioning Brown and Levinson (1987) for whom the use of slang and jargon is seen as a device which shows in-group identity.

5.4.5. Swear words: results The group of teenagers used swear words quite often, but not with the purpose of causing offence, but as a solidarity marker. Consider the following examples: (114)

Cristian:

(115)

Cristian:

(116)

Cristian:

(117)

Adi:

(118)

ùtefan:

(119)

Andreea:

(120)

Adi:

(121)

Roberta:

Bă fir-ar mă-ta a dreacu’ ‘Damn you’ futu-Ġi morĠii mă-tii băi ‘Fuck you’ Voi sînteĠi chiar tâmpiĠi să moară mama ‘You’re really dumb damn it’ MorĠii tăi de bou ‘Fuck you you arsehole’ Uită-te-n gâtu’ mă-tii aicea ‘Look here God damn it’ Ce dracu’ mănânc eu ‘What the hell do I eat’ Futu-te-n gură cu pomelnicu’ tău ‘Fuck you with your long string’ Da mă boule creion dintr-ăla care scria cu mov ‘Yes you arsehole, that pencil which wrote in purple’

In Brown and Levinson’s (1978) politeness theory the use of swear words in the examples (114)-(121) can be viewed as a positive politeness strategy. Under different circumstances they could count as Face Threatening Acts (FTA) but given the relationship status between the members of the group, face is not lost. So, in these examples swear words

Adolescents as Language Innovators

177

function as solidarity markers. Girls have absolutely no problem with using swear words, although not as frequently as boys do, probably just to ensure their group membership.

5.4.6. Gender variation in the ConstanĠa corpus As far as COLT adolescents are concerned, there are gender-related differences in the use of slang and swearing, so the expectation is that there are also such differences in the ConstanĠa corpus. There are genderrelated differences regarding the use of dirty words as well as proper slang terms, as presented in Table 5.32, below: Gender male female

Proper slang 17 5

Dirty words 88 22

Total slang words 105 27

Table 5.32. Slang and gender in Romanian Dirty words are used more frequently than proper slang, which is contrary to the situation found in English. The ConstanĠa corpus is, indeed, much smaller that COLT, but despite its limitations it still suggest certain clear linguistic preferences among teenagers. Male speakers tend to use both proper slang and dirty words more frequently than female speakers. The speakers from the ConstanĠa corpus belong to one age-group (17–19). In COLT in this age group female speakers were leading in the use of both dirty slang and proper slang, but it seems that in Romanian the situation is vice versa. So, male teenagers dominate the scene as far as slang is concerned. It should also be taken into account that the dirty words used by teenagers had different linguistic functions, not only that of causing anger, resentment, offence or displeasure. Another interesting gender-related finding concerns the use of the word(s) for male genitalia by girls also. This situation is exactly the same as the one found in COLT. However, girls, with one exception, never used the word(s) for female genitalia. The only word used was păsă66 ‘pussy’, as in the following example: (122)Roberta:

66

La mulĠi ani păsă ‘Happy birthday pussy’

A truncated form of păsărică ‘pussy’.

Chapter Five

178

Here păsă is used as a term of endearment, to show appreciation and, why not, female solidarity. In the data that I collected from Facebook there were also other situations alike: (123)Adrian: (124)Adrian: (125)Cătălin: (126)Iulian: Adrian:

:)) muie ce faci măh? ‘Dude what’s up?’ o ai strâmbă de la atâta labă coae! ‘it’s curved from so much handjob dude’ Nu mai are niciun farmec coiĠă...:)) ‘It’s not funny anymore dude’ pokemonule ‘pokemon’ să îmi iei Pokemonu între gingii Iulian ‘Take my Pokemon between your gums Iulian’

The fact that girls used the male genitalia when swearing is intriguing. In mixed conversations girls always swore using the male genitalia but the situation changed in same-gender conversations. In the latter situation the word(s) describing male genitalia was/were used rarely, only when the speaker was extremely irritated by something and wanted to express her anger. So, in this case it is used as a strong expletive. Boys, on the other hand, used the word(s) irrespective of their interlocutor’s gender, so both in mixed as well as same-gender conversations. An explanation for the girl’s use of such strong expletives only in mixed-gender conversation is again connected to prestige and power. They want to be accepted as members of the group, despite their gender, and by doing so they borrow the words used by boys, in different contexts, for specific purposes. Weak expletives such as naiba ‘damn’ were found to be part of “women’s language”, as predicted by Lakoff (1975). (127)Mihaela:

Ce naiba mă? ‘What the heck?’

Boys never used the word naiba to express annoyance, instead they preferred stronger expletives like dracu’ ‘hell’ or the even more frequent dirty word căcat ‘shit’. So male speakers swear more frequently, the insult each other and are very competitive in conversation. Even though the girls in the ConstanĠa corpus used swear words they did not use them as frequently as boys did. In the Romanian culture there are other things that we might consider “women’s language”. For example, boys frequently used mă and bă to

Adolescents as Language Innovators

179

refer to each other. Girls also used it quite often. Girls also used another form of address but only in same-gender conversations, i.e. păsă ‘pussy’. They never used it to refer to a male speaker, unlike English adolescents who called boys cunts. There is also another form, fă or făi which occurs in the data collected from Facebook. Consider the following conversation: (128)Denisa: Maryy: Mirela:

Anna Maria: VictoriĠa: Mirela:

Anna Maria:

Eúti o bunăciune păsă! ùi mi dor de tine! ‘You’re a hottie pussy! And I miss you’ frumii fok ‘so beautiful’ vai...vai...nu te mai suporĠi făi ‘oh dear...oh dear...you can’t stand yourself girl’ Mai taci făi sarcastico ‘Shut up you sarcastic girl’ sunt în spatele tău ‘I’m behind you’ aúa te-a învăĠat tanti să vorbeúti?...ruúinică păsărică ‘this is how Mrs taught you to speak?...shame on you pussy’ Dacă nu te opreúti cu fleanca aia a ta plină de sarcasm :)) cum ai vrea să-Ġi vorbesc făi :)) ‘If you don’t stop with your maw full of sarcasm :)) how would you like me to speak to you girl :))’

The girls that engaged in this conversation belong to the same age group (18–20). In same-sex conversations girls use this form of address without causing offence, just like boys used forms like coaie ‘balls’, muie ‘blowjob’. These words are considered dirty words and have a different effect, they function as a solidarity marker, something which shows the boys’ membership to a group. Of course, such words do not cause offence when they are used among friends. In other contexts they could be considered as extremely impolite. If a boy were to call a girl using fă or făi this could be deemed inappropriate as it is an FTA. Another strategy used more often by girls is that of paying compliments to one another, this being considered as positive politeness, in Brown and Levinson’s (1987) terminology. The use of compliments among girls is a sign of solidarity and friendship with the purpose of establishing and maintaining a close relationship as well as bonding.

Chapter Five

180

Another gender-related difference lies in the fact that there are much more dirty words used to describe a girl than a boy. (129)Marius:

Virgil:

ce panaramă! O curvă răsuflată a ajuns moderatoare! ce vocabular de căcat are zdreanĠa asta ‘what a floozy! A godforsaken slut has become a moderator! What shitty vocabulary does this bitch has’ penală e! târfa dreacu’ :)) ‘she’s such a skank! fucking whore’

So, when it comes to nouns that refer to a female speaker, they abound. There are also words to describe a male speaker but the ones for girls prevail. The meaning is also different, as these dirty words used for female speakers have a negative connotation whereas the ones used for male speakers have acquired a rather positive one. At this stage it is also worth mentioning two highly colloquial words that are used to refer to a man and a woman: cocalar and piĠipoancă. They are used as derogatory terms, with the purpose of criticising people. Cocalar as well as piĠipoancă were initially used as insults but the terms have become very fashionable and enjoy a great success among teenagers. The words are now used to describe and mock a certain social class (Zafiu 2010b). Cocalar, borrowed from Rromani (Sarău 1992), is used with the meaning of mitocan ‘arsehole’, a man who is rude and makes inappropriate remarks usually to girls. The word is also used to describe a man who is not cool, who is poorly dressed, a Gypsy or a manelist ‘a person who listens to manele, a type of music usually sang by Gypsies’. So, from this point of view cocalar is highly pejorative and has a negative connotation. The word piĠipoancă is used to describe teenage girls or adult women who are considered to be sluts. They are recognised by the way they dress and the way they speak. The difference between the two terms is that piĠipoancă does not necessarily have the meaning [+rude] attached to it.

5.5. Concluding remarks The purpose of this chapter has been to discuss the role of swearing and slang in the speech of London and ConstanĠa adolescents. The chapter started with a discussion of taboo language and what is considered to be taboo in different societies. Then, the role of bad language was presented

Adolescents as Language Innovators

181

and what should be labelled as “bad language”. The concepts of swearing, insult and dirty words followed and here the focus was on the concept of “slanguage” introduced by Stenström et al. (2002). The third part of the chapter has been devoted to the use of swearing and slang by London adolescents. The analysis draws on the COLT corpus, collected and compiled by a group of researchers from Bergen University, Norway and discussed in Stenström et al. (2002), who were actually the Norwegian researchers. Among the gender-related differences identified, it is worth mentioning that boys use proper slang as well as dirty slang more often than girls. When another sociolinguistic variable was added (i.e. age) the situation changed, and this revealed that besides gender there are also age-related differences. For example, younger girls do not use dirty words or proper slang as often as boys do, but the situation changes with older girls (aged 17–19) using foul language more often than the male speakers. This could be explained by the fact that they become independent and maybe to secure their position in the group. The last part of this chapter has been devoted to the use of swearing and slang by Romanian teenagers living in ConstanĠa. The analysis draws on a corpus collected in 2011– 2012, focusing on a group of teenagers. To this a number of conversations have been added from Facebook, collected in March–April 2013. It was shown that teenagers are indeed real language innovators and much of the slang is created and used by them. Interestingly, the majority of slang words used by Romanian teenagers are borrowed from Rromani. As far as gender differences are concerned, it has been argued that slang words (proper slang as well as dirty slang) are more frequently used by boys than by girls. Another interesting finding was that swearing, insults and dirty words are used not only to cause offence but also as a solidarity marker, by both genders. Lastly, the use of words to refer to a female and male speaker was discussed and it has been shown that there are many words to describe girls than boys, often used with a negative and degrading meaning.

CONCLUSIONS AND FURTHER RESEARCH

This study has been a comparative sociolinguistic analysis of genderrelated variation in the speech of English and Romanian adolescents. The aim has been to identify patterns of linguistic variation in the speech of adolescents and compare the two cultures. In doing so, I have relied on data collected in different ways. For the English analysis, I have based my analysis and arguments on data collected at Queen Mary’s College, University of London (QMUL), in 2012. Regarding the Romanian analysis, the research project was conducted at Educational Center Theoretical High School (ECTH), in 2013–2014. There are several issues that others might consider to be problematic in this dissertation. The number of subjects interviewed for the English and Romanian corpora might be seen as too small to draw the conclusions that I have drawn. Each speaker received a fair chance to get selected, since both research projects were conducted at educational institutions. Whoever wanted to participate in the project was given a chance and was interviewed. The analyses reflect the language used by adolescents living in London and ConstanĠa, meaning that they do not provide national generalizations. The linguistic patterns that I have identified are confined only to the speech communities that were under investigation and to their speakers. There may be other linguistic and social practices outside these communities, but this dissertation could not have covered all of them. Chapter 1, “Language and gender: past and present”, has offered a relevant review of the literature on language and gender research. I have outlined the most important concepts and theories which structured my argument. I have discussed the difference between sex and gender, thus accounting for the use of the word gender in the title of this study and not sex. It was shown that sex is a biological trait while gender represents the social construction of sex. Gender is vital in establishing an individual’s identity. I also introduced some key ideas regarding the construction of gender identity through language and discussed the differences between genderexclusive and gender-preferential differences in language use. The former are found in traditional societies where women do indeed speak differently than men, and the examples provided bear testimony. The latter are typical of modern societies and express a tendency for the use of certain forms of

Gender-Related Variation in the Speech of English and Romanian Adolescents

183

language. Jespersen (1922) was among the first linguists to discuss this phenomenon and even included a chapter entitled “The Woman” and Lakoff (1975) was the first to introduce the concept of “women’s language”. Several of Lakoff’s claims, albeit based on no empirical research, have been confirmed by quantitative as well as qualitative analyses. Section 1.2.2. has been devoted to gender-preferential differences, the type of differences that I was interested in. They index gender indirectly and they are distributed across speakers or groups, but the general tendency is that they are used more by some speakers than by others. Throughout the years several models have been proposed to account for gender differences in language use. The first one, the deficit approach, was proposed by Lakoff (1975) to identify the linguistic practices of women. She referred to these practices as “women’s language” and provided a list of linguistic features associated with women. The key idea was that women were seen as more polite than men, and they used the standard forms of language while men preferred the non-standard ones. The second theory discussed was the dominance model, associated with the work of Zimmerman and West (1975, 1983) as well as Tannen (1994). Their theory is based on the fact that women are seen as an oppressed group and they live in a male dominated society. Included here was also a discussion of “brave new words” (Oancea 2013). So far terminology was maleoriented as well as the use of the masculine pronoun he and the noun man with a generic meaning. In recent years, however, many feminine counterparts have appeared for words which were so far considered masculine, e.g. female doctor, inspectress, manageress, doctress, etc. The difference approach, which was the third theory discussed, has its roots in the work of Gumperz (1982). According to this theory men and women belong to different cultures as far as socialisation is concerned. The last theory discussed was the dynamic approach which offered a social constructionist perspective on language and gender research. I did not use a particular approach in my dissertation, but I combined these approaches and integrated them into a cohesive model that approached language from both a quantitative and qualitative perspective. In the next part of this chapter I gave a broad-brush picture of some gender-related stereotypes in English, some of which have been confirmed by quantitative analyses. From all of these studies it resulted that women have a tendency to use the standard forms of language, associated with prestige in both formal and informal styles of speech, while men prefer the vernacular forms (Cheshire 1982, Newbrook 1982, Trudgill 1983, Romaine 1984, Eisikovits 1988, Cheshire 1997). Such studies conducted

184

Conclusions and Further Research

in urban environments and on urban communities provided important information on the internal and external factors that govern language variation and change and highlighted the pivotal role of gender differences in these processes. The last part of this chapter tackled gender-related stereotypes in Romanian. Given the sparseness of studies which approach gender differences in Romanian, my intention was to identify certain gender stereotypes among speakers and I subsequently raised the question whether they are aware of the existence of such stereotypes. Another reason was to determine the role played by language in the construction of gender identity. The analysis of the aforementioned stereotypes was based on two research projects. The first one was conducted among MA students studying at Ovidius University of ConstanĠa and the University of Bucharest. Thirty students were interviewed (fifteen girls and fifteen boys), aged 22–26. A set of sixteen sentences were devised, which contained some of the linguistic variables proposed by Lakoff (1975) and Holmes (1995). The results obtained confirmed the hypothesis that women are more status conscious, more polite than men, and also use taboo words. This latter finding required a more in-depth analysis, which constituted the basis for Chapter 5. In order to obtain a more thorough understanding of the existence of such gender-related stereotypes, I replicated the survey, and introduced another sociolinguistic variable besides gender, i.e. age and also added four more sentences to the questionnaire. One hundred fifty students were interviewed, covering three age groups: fifty students were secondary school pupils (age group 12–14), fifty students were high school students (age group 15–18), and fifty students were first year undergraduate students from Ovidius University of ConstanĠa (age group 19–21). There was also a gender balance in each age group category, twenty five girls and twenty five boys. All the students that took part in the project lived and studied in ConstanĠa. The school and high school have been randomly selected and each student was offered a chance to be interviewed. The results remained the same, even though the number of interviewees was extended considerably. The results obtained have confirmed some of Lakoff’s (1975) claims, regarding the use of precise colour terms, intensifiers, and empty adjectives by women. Men appeared to be more straightforward, impolite at times, and preferred the non-standard forms. It has also been revealed that women also use taboo words, something which is associated to male language. Regarding foul language, it has been found that with age there seems to be a decrease in the use of such language by men and an increase in the use of taboo words by women. This showed

Gender-Related Variation in the Speech of English and Romanian Adolescents

185

that women become more confident and more powerful. The two projects pinpointed that gender-related stereotypes in Romanian society are very similar to the ones found in English-speaking communities. Chapter 2, “Language variation and change: the theoretical framework”, has aimed at providing a suitable theoretical framework to be used in the analysis of gender variation in English and Romanian. The chapter has been roughly divided into two important parts. The first part has discussed variationist sociolinguistics as a conceptual framework to be used in the analysis of the phonological variables discussed in Chapters 3 and 4, while the second part of the chapter focused on the politeness phenomena (Brown and Levinson 1987) and swearing and taboo language (Andersson and Trudgill 1990, Allan and Burridge 2006). Given the plethora of research in the variationist framework, only the relevant studies have been presented and used as starting points in the analysis. The chapter opened with a discussion of variationist sociolinguistics, stressing the importance of two key concepts: sociolinguistics and variation. My first undertaking was to get a firm grasp of these concepts as they are the pillars of variationist sociolinguistics. The central tenets of variationist sociolinguistics have also been highlighted: orderly heterogeneity, language as a living and ever-changing organism and language as a provider of significant non-linguistic information. A variationist study is actually based on a quantitative analysis, which sets out to confirm or disconfirm a hypothesis that is grounded on empirical research. Given the importance of variation and change, a discussion of these notions was imperative. The essential thing about variation and change is that they occur in the vernacular of everyday life. Gaining access to the vernacular might be deemed a daunting task. In formal situations speakers have the tendency to control their speech and strive to sound as intelligent as possible, especially if they are recorded. I have also argued that in order to conduct a first-hand research of a linguistic variable, the sociolinguistic interview plays a crucial role. This is the most widely used data collection method in the variationist sociolinguistics framework. In section 2.1.3. I have tackled the relevance of the Observer’s Paradox and presented ways in which it can be avoided. The sociolinguistic interview is deemed a formal situation and speakers might not use the nonstandard features of language that we, as researchers, are interested in. Several methods to overcome the Observer’s Paradox have been used over the years, such as Labov’s (1972) danger of death questions or Milroy and Gordon’s (2003) proposal that people do not pay much attention to the language they use when they feel happy, excited or angry. Another method

186

Conclusions and Further Research

is to create a casual, informal and friendly atmosphere when conducting the research, gain the speaker’s confidence and make them feel comfortable and willing to share their stories, experiences, etc. The main aim of the interview is to elicit as much data as possible. In the next section I have examined the linguistic variable, and discussed phonological, morphological and syntactic variables. Given the multitude of studies it was impossible to discuss all the phonological variables studied. I have stressed the idea that variationist sociolinguistics has its roots in the analysis of phonological variation. Two important phonological variables have been considered in this section, namely postvocalic (r) (Labov 1966) and (t,d) (Tagliamonte 2005). The morphological variables under scrutiny were verbal (s) (Cheshire 1982, Poplack and Tagliamonte 2004, Tagliamonte 2011) and the masculinisation of the feminine in spoken Romanian (the use of the preposed masculine article lui with feminine common nouns, certain proper nouns and demonstrative pronouns). I have also briefly discussed syntactic variables, as they are much more difficult to define, and looked at the use of the overt pronoun in pro-drop languages (Romance languages) as well as at the use of the English deontic modality system (Tagliamonte 2011). I have also mentioned some of the most important variationist studies which have quantified gender. Using the so-called urban stratification studies carried out in the sixties and seventies as a departure point, I have highlighted some of the most important findings and then presented quantitative studies focusing on morphological (Cheshire 1982, 1997; Eisikovits 1988) and phonological variables (Eckert 1990, 1998). Also mentioned here are Labov’s (1990) principles concerning the relationship between gender and language. The second part of this chapter has been devoted to the concept of “communities of practice” coined by Lave and Wenger (1991) and subsequently used by Eckert (2000) in her ground-breaking study of a Detroit high-school. I have included this notion in my analysis of genderrelated variation in the speech of adolescents because it has offered a suitable framework for the understanding of the social as well as the linguistic facets of sociolinguistic variation. This notion proved extremely helpful to account for the gender differences in the speech of Romanian adolescents, presented in Chapter 4. In the last part of this chapter I have focused on the politeness phenomena (Brown and Levinson 1987, Culpeper 1996), as a framework for the qualitative analysis of this study, concentrated in Chapter 5. I have discussed the concept of face and Face Threatening Acts in Brown and

Gender-Related Variation in the Speech of English and Romanian Adolescents

187

Levinson’s (1987) terminology and delved into the notions of swearing and taboo language (Andersson and Trudgill 1990, Allan and Burridge 2006). Chapter 3, “Talkin’ or talking? Phonological variation in the speech of London adolescents”, has forayed into the analysis of phonological variation in English and it has represented the first personal contribution to this dissertation. The analysis has been based on a research project carried out at Queen Mary, University of London in October 2012, which resulted in a corpus of 14000 words. The chapter started with a review of the studies that focused on the (ing) variable, starting with Fisher’s (1958) analysis of (ing) in rural New England and ending with Schleef et al.’s (2011) study conducted in London and Edinburgh among Polish immigrants. The (ing) variable is one of the most studied phonological variables, and has been the first one studied quantitatively, by Fisher (1958). The variable has also been studied in the United States: New York (Labov 1966), Detroit (Shuy et al. 1967), North Carolina (Anshen 1969), Philadelphia (Cofer 1972, Labov 2001), Los Angeles (Wald and Shopen 1985); in Canada (Ottawa (Woods 1979); in Britain: Norwich (Trudgill 1974), Edinburgh (Reid 1978). I have chosen to study this variable because it can be analysed auditorily, it is salient among native speakers of English, and it is a stable sociolinguistic variable. The (ing) variable usually has two phonetic realisations: as a velar nasal or as an alveolar nasal. Many studies have correlated (ing) with gender and the result was that women preferred the velar nasal, considered to be the prestige variant, while men tended to use the alveolar nasal, the vernacular variant. The more recent studies have used Goldvarb or Rbrul (Boersma 2001, Paolillo 2001) to analyse (ing), these programs being the most frequently used in variationist research. The second part of this chapter has presented the data collection and methodology used. I have included relevant details about the speech style, more importantly how I have devised my questionnaire. The analysis was based on audio data collected from eight speakers (four girls and four boys) studying at QMUL. Each interview has lasted for twenty five to thirty minutes, thus offering around five hours of speech which has been analysed. The sociolinguistic interview included three parts: a semistructured questionnaire (casual speech), three reading texts (reading style) and a list of words (word list style). The recordings were made at QMUL. The results obtained were discussed in the last part of this chapter. I have used PRAAT and chose to do an acoustic analysis of the (ing) variable. Each token of (ing) has been verified using PRAAT, and several spectrograms have been included in the chapter to show gender variation

188

Conclusions and Further Research

but also intra-speaker variation between the velar nasal and the alveolar nasal. The independent variables that I have taken into account were: word class, the number of syllables of the word containing (ing), speech style and gender. 342 tokens of (ing) have been analysed and a preference for the alveolar forms has been found for both genders. An important finding as well as intra-speaker variation was quite consistent with some speakers. Regarding the independent variable word class, I have grouped the words thusly: common nouns, progressive verb forms, gerunds, adjective, discourse markers and pronouns. The last two groups have been later united into one umbrella-term called THING compounds. It has been shown that the (ing) variable occurred more frequently with progressives, gerunds and common nouns, where the alveolar variant was preferred. The velar nasal variant was used more frequently with discourse markers, pronouns and adjectives. Labov (2001: 88) argued that the most important linguistic constraint on the (ing) variable is that it is affected by the nominal–verbal continuum, but the teenagers in my analysis showed a strong usage of the alveolar variant with common nouns, but also with gerunds (which have both verbal and nominal properties). Among the gender-related differences identified here it is worth mentioning that, at times, girls used the non-standard variant more frequently than boys, something which was not attested in the other studies focusing on (ing). This provided valuable insight into the external factors which influenced language variation and change, as they do not appear to be as statusconscious as they once were. The use of the alveolar variant was more consistent in the speech of boys, and decreases, to a certain extent, in the speech of girls. The same phenomenon occurred with the velar nasal variant. It increased with gender, i.e. male speakers seldom used it, while girls used it more often. The analysis has also focused on the THING compounds. There have been forty eight compounds in the corpus, and there was variation between [n] and [ƾ], but only with something. A third variant [ƾk] also occurred, but it was excluded from the analysis as it only occurred in the speech of Mark and it was treated as an isolated case. Relevant spectrograms have also been included, which showed the realisation of (ing) as either a velar or an alveolar in the speech of Mark and Suzanne. There was also gender variation, with girls preferring the velar nasal form almost exclusively, unlike boys who, despite a tendency to use the velar nasal, showed a higher use in the alveolar variant than girls did. These were the results obtained for the casual speech style elicited in the first part of the sociolinguistic interview.

Gender-Related Variation in the Speech of English and Romanian Adolescents

189

The next stage in the analysis of the (ing) variable has been to identify variation patterns in the reading style. For this part I have used three different texts taken from various English textbooks. Students were asked to read them out loud as naturally as possible. The aim of this part was to obtain information on a wider range of styles and compare them. Thirty seven tokens of (ing) have been coded and analysed. Given that this is a much more formal situation than the questionnaire, the expectation was that the students would use the velar form. The results, however, have been quite surprising and unexpected. Girls used the alveolar form almost exclusively in this part of the interview, while boys used the velar form. Interestingly enough, there has been a change in the accent that girls put on way they are perceived in society. Boys were more careful in using the standard form than girls, and they acknowledged that they were being recorded. The last part in the analysis of (ing) has been the word list style, in which the students were asked to read a list of sixty eight words. In the list there were twenty three tokens of (ing) and the rest were distracters. The main finding corroborated the one found in the reading speech style, thus reinforcing the claim that I have made, i.e. even though this is an artificial test, boys acknowledged that and strove to used the standard form, whereas girls did not pay much attention to it and used the vernacular form. This chapter has debunked the myth that women tend to use the standard forms of language while men the vernacular one. The use of the non-standard form of (ing) by individual speakers provided valuable insight into how speakers are able to exploit such sociolinguistic resources. The idea that language use is an important tool in the construction of a person’s identity is well known, but what is interesting is that girls chose to identify themselves with the vernacular culture, by using the alveolar variant. Cheshire et al. (2011: 164) puts this on the attractiveness of certain features pertaining to the Carribean and African American youth culture, which made the use of such “black features” attractive. The degree of “coolness” was another important factor in youth culture, given the importance that teenagers put on being fashionable and accepted by the community. Also, in many songs the alveolar variant is preferred, and this might have an important impact on the speakers. Many American rap and pop bands use the non-standard form and given that music plays a central role in the lives of teenagers, this might also account for the tendency to use this form, especially by girls who are trying to imitate the boys’ style.

190

Conclusions and Further Research

The next stage in my analysis of gender-related variation in the speech of adolescents has been to investigate and identify patterns of phonological variation in Romanian. The results were presented in Chapter 4, “Pe vs Pă: Phonological variation in the speech of adolescents living in ConstanĠa”. Phonological variation in spoken Romanian was an underresearched area, therefore I had to build up my own methodology. I have used the results obtained in Chapter 3 as well as the ones from different English-speaking communities as a starting point in my analysis and set out to investigate whether the (pe) variable is subject to phonological variation and whether this variation is influenced by social factors such as gender and age. In the first part of the chapter I have given a diachronic account of the (pe) variable, as used in Early Modern Romanian and followed the evolution of the variable up to present day Romanian. In the 16th and 17th century texts, the form of the variable under scrutiny was (pre), which gradually changed towards the 18th century texts to (pe). In both 19th and 20th century texts the variable was used with two variants: (pe) and (pă). The rest of the chapter has attempted to account for this type of variation and proposed a theory. Section 4.3. has tackled the data collection and methodology employed in this undertaking. The sociolinguistic interviews have been conducted at the Educational Center Theoretical High School in ConstanĠa in 2013/2014. The interviews followed the same structure of the one used for the (ing) variable except for the word list style. It contained twenty-two questions and each interview lasted between twenty and twenty five minutes. The speakers have been selected on a volunteering basis, i.e. all of them were informed and only those who were willing to participate were interviewed. Twelve students (six boys and six girls) took part in the project, some of whom were secondary school students and others high school students. In this analysis I followed Eckert’s (1989) groundbreaking study of the English used by jocks and burnouts at Belten High. I grouped the students from ECTH into geeks and pops. The former were the ones who took a real interest in school activities and in their academic trajectory, thus obtaining high grades. The latter were socialites, very popular in school and clothes were an important factor for them. The next stage in the analysis has involved a discussion of the longterm participant observation method, to overcome the Observer’s Paradox and to give a more detailed account of the language used by adolescents in their natural habitat, i.e. in the playground. This mode of inquiry has added a fresh perspective to the research by gaining access to their vernacular

Gender-Related Variation in the Speech of English and Romanian Adolescents

191

and becoming involved in their activities, but only as an observer. My observations in the playground were noted in a special notebook that I carried with me at all times. An audio recorder could not be used for practical reasonsm but also because the students would have refrained from using the vernacular forms at the sight of such a device. In section 4.4. I have pinpointed the results of my analysis of (pe) in spoken Romanian. I have transcribed the audio interviews into Microsoft Word which resulted in an aligned corpus of 15000 words, with a total number of one hundred thirteen tokens of (pe). The independent variables that have been taken into account for the analysis of (pe) were speech style (casual and reading style) and the gender of the speaker. The (pe) variable has been selected for analysis because it is a stable sociolinguistic variable and has not been subject to any change for at least two centuries. In the casual speech style, the nonstandard form [pă] was not used almost at all. Both genders preferred the standard form, especially the secondary school students. The fact that the standard form was preferred showed the central role played by school in the life of adolescents. The younger students were more careful with the language they used then high school students. With age, they become more interested in the way they are perceived by society and by their peers. I have also shown that the standard and nonstandard variants of (pe) co-occur variably, and this was not governed by any linguistic constraint. The variant [pă] has been interpreted as a marker of masculinity not only in the speech of boys, but also in the speech of girls. For the reading style, the students had to read a one-page text out loud. In this case, unlike English where the nonstandard variant was preferred, the students used the standard variant almost exclusively, thus corroborating the results obtained for the casual speech style. The students’ linguistic choice highlighted once more the important influence of the school. Given that they were corrected at all times by their teachers and they were reprehended when they used nonstandard forms has been reflected in the language they used. The most important and surprising findings have been made using the long-term participant observation method (Schilling 2013). Two communities of practice have been identified in the school: the geeks and the pops. I have noticed that with age, there appears to be a rise in the use of the [pă] variant. Secondary school students did not use the nonstandard form almost at all, except for Alexandra who pertained to the pops group. In high school pops became more assertive, and at times even violent. Swearing was also extremely used by pops boys in the playground. I also used a toughness scale (Cheshire 1982) and correlated it with the use of

192

Conclusions and Further Research

nonstandard linguistic features. This proved extremely helpful as it indicated that toughness was an indicator of the students’ adherence to the vernacular culture. The pops’ preference for verbal aggressiveness (shouting, name-calling, swearing) contributed to the formation of their identity as dominant, cool and assertive. It also showed their membership to the group. However, verbal aggressiveness in the pops group was not interpreted as a gender marker as both genders used it. The only exception was the use of swearing which was confined to boys. The geeks, on the other hand, used the standard forms at all times. There were also times when they used the nonstandard forms, but with the purpose of making fun of or imitating the pops. The next stage in the analysis of the (pe) variable has been to analyse it in the speech of Romanian adults. For this I recorded different talk shows on TV from 2013 and 2014. The analysis has been based on eight hours of spoken interaction. The recordings have not been transcribed into Word, instead they were analysed using ELAN. In total two hundred fifty six tokens of (pe) have been investigated. One might argue that when they are on TV people are careful with the language they use, but in these talk shows the atmosphere was very relaxed and everything quite informal. After a thorough analysis I have found that male speakers preferred the nonstandard form [pă] while female speakers showed a tendency for the [pe] variant. Another gender-related difference resides in the topics of discussion. Female speakers were interested in fashion, the latest gossip, plastic and cosmetic surgery and cooking, while men were discussing politics or talking about Romanian culture. Women’s interest in more personal details was perceived as a solidarity marker, whereas men focused more on conveying information. The consistent use of [pă] by male speakers has also been interpreted as a masculinity marker used in the construction of masculine identity. Butler (1999: xv) postulates that gender is performative and is culturally constructed through repeated stylised acts. To add more validity to this claim, I used Hornoiu’s (2007) corpus for control, and looked at the occurrences of the (pe) variable. They corroborated the claims that I have made, i.e. women prefer the prestige form and men the vernacular one. The last part of this chapter has been devoted to a comparison between adolescent speech and adult speech. I have argued that adolescents are language innovators but their use of vernacular forms is rather restricted and influenced by a multitude of extralinguistic factors. The students’ tendency to use the nonstandard forms in the playground showed their awareness of the social pressure and the fact that they speak differently depending on the context. In adult speech, especially in the speech of men,

Gender-Related Variation in the Speech of English and Romanian Adolescents

193

the vernacular form, i.e. [pă] was more consistent. This has been interpreted as a decrease in the willingness to conform to societal norms and this type of pressure fades away. These findings, however, were not gender-exclusive, but they showed a preference for one form or another, be it the standard or the nonstandard one. The last chapter of this book, “Adolescents as language innovators: Swearing, taboo words and slang in English and Romanian”, has been concerned with offering a qualitative analysis of swearing and slang in adolescent speech. It has been a contrastive chapter, which aimed at comparing the two societies. In the first part of the chapter I have discussed the notion of “taboo words” (Allan and Burridge 2006) and I have highlighted different types of taboos. Section 5.2. has provided a discussion of bad language and sweet words and what should or should not be deemed bad words. I have also included a conversation taken from the movie The King’s Speech, between King George VI and his speech therapist, Lionel Logue, to highlight the social pressure exerted by society on the use of swearing in public. Orthophemisms, euphemisms, and dysphemisms were also under scrutiny and relevant examples to illustrate the difference between them have been provided. I have also tackled the concepts of “swearing”, “dirty words”, and “slanguage”, and used them as a departure point in my analysis. The notion of “slanguage” has been coined by Stenström et al. (2002) and encompasses the following: proper slang, dirty words, vogue words, vague words, proxy words and smallwords. The analysis of swearing and slang in the speech of London adolescents has been dealt with in section 5.3. The analysis has been based on the COLT corpus, compiled in 1993 and on the book based on this corpus, published by Stenström et al. (2002), who were actually the researchers who collected the data. The adolescents that were recorded came from different social and ethnic backgrounds, thus adding diversity to the corpus. As shown in 5.3.2. sex was among the preferred topics of conversation of London adolescents. It was pointed out that the boys’ conversation was actually a competition in the use of insults against each other. The girls’ conversation did not contain swear words or insults, which showed the girls’ closeness and solidarity to one another. I have also presented the top ten most frequent slang words in the COLT corpus as well as the top ten most frequent dirty words. It was also shown that boys and girls refer to each other using either male or female genitalia. Boys used terms denoting male genitalia when they referred to other boys, while girls used different offensive words to describe other girls. Interestingly enough, girls used feminine words to describe boys,

194

Conclusions and Further Research

and boys used the same feminine words to refer to girls. Section 5.3.4. presented the gender differences identified in the COLT corpus. Male speakers used proper slang and dirty words more often than girls did. The difference between the two genders was quite consistent in the use of proper slang words, and it was less poignant concerning dirty words. I have also taken the social variable of age into account, besides gender, and identified some differences. There were three age groups (10–13, 14–16, 17–19) under investigation and the analysis has yielded the following results: (i) the use of proper slang was more frequent with male speakers in the 10–13 age group, whereas the use of dirty slang in the same age group category was ten times more consistent in male speech; (ii) in the 14–16 age group the situation changed, and girls started using proper slang as well as dirty slang more often than the girls in the 10–13 age group; (iii) in the 17–19 age group girls exceeded boys in the use of both proper slang and dirty words. These findings have shown that younger girls are more polite, more status conscious, weak and unassertive, but with age they become more confident, more powerful and start to dominate the scene due to their adherence to the street culture or their membership to a group or gang. Boys in the 17–19 age group used proper slang words and dirty words quite infrequently, compared to the boys from the other age groups. The last part of this chapter has been devoted to the analysis of swearing in slang in Romanian. The analysis has drawn on a corpus of 6500 words, compiled in 2011. The study is based on a three hour video recording which followed a group of teenagers from ConstanĠa who spent New Year’s Eve in a villa in Buúteni. To this several conversations have been added in the form of comments collected from Facebook in March– April 2013. This enabled me to have a more unified account of the use of slang and swearing by Romanian adolescents. In a subsequent subsection I have provided detailed information about the speakers that took part in the research and then presented the results. I have shown the top ten most frequently used dirty words in the ConstanĠa corpus and provided different contexts which showed the use of these words. A special emphasis has been placed on the use of slang words and their etymology and different examples have been given. The use of swear words has been discussed in relation to politeness theory, as advanced by Brown and Levinson (1987). I have argued that the use of swear words can be seen as a positive politeness strategy, given the close relationship and familiarity between the speakers. The last part of the analysis of swearing and slang in Romanian has tackled gender-related differences. As shown in 5.4.6., male speakers used

Gender-Related Variation in the Speech of English and Romanian Adolescents

195

proper slang and dirty words more often than girls, and despite the corpus’s limitations, it did portray several linguistic choices among teenagers. It was also mentioned that girls also refer to each other using female genitalia and they swear using male genitalia. Regarding the use of slang, many of the slang words used by Romanian teenagers were borrowed from Rromani, as discussed in section 5.4.4. Another claim that this chapter has made was that swearing and slang were not only used to cause offence or to lose one’s face but also as a solidarity marker, by both genders. This study could be continued in many ways. First of all, the number of subjects for both the English and Romanian analyses could be extended to cover other age groups, and see whether the results change in any way, or if the results change significantly when the number of speakers increases. Another interesting continuation would be to look at other variables, for example morphological, syntactic or even lexical, analyse them quantitatively and look for gender differences. In Romanian such topics are under researched and it would be a good opportunity to delve into this type of research and investigate both the internal and external factors that influence this type of variation. Lastly, the speech of adults is also worth looking at, especially older speakers, where variation (be it phonological, morphological or syntactic) is even more consistent.

BIBLIOGRAPHY

Abramowicz, L. 2007. Sociolinguistics meets exemplar theory: Frequency and recency effects in (ing). University of Pennsylvania Working Papers in Linguistics 13: 27-37. Allan, K., Burridge, K. 2006. Forbidden Words: Taboo and the Censoring of Language. Cambridge: Cambridge University Press. Allen, I. L. 1998. Slang: Sociology. In J. May (ed.) Concise Encyclopedia of Pragmatics, 878-883. Amsterdam: Elsevier. Altendorf, U. 2003. Estuary English: Levelling at the Interface of RP and South-Eastern British English. Munich: Gunter Narr Verlag. Andersson, L.-G., Trudgill, P.1990. Bad Language. Oxford: Blackwell. Anshen, F. S. 1969. Speech Variation among Negroes in a Small Southern Community. PhD dissertation, New York University. Bailey, G., Wilke, T., Sand, L. 1993. Some patterns of linguistic diffusion. Language Variation and Change 5 (3): 359-390. Bailey, L. A., Timm, L. A. 1976. More on women’s and men’s expletives. Anthropological Linguistics 18 (9): 438-449. Baker, P. 2010. Will Ms ever be as frequent as Mr? A corpus based comparison of gendered terms across four diachronic corpora of British English. Gender and Language 4 (1): 125-149. Bjorkqvist, K., Lagerspetz, M. M. J., Kaukiainen, A. 1992. Do girls manipulate and boys fight? Developmental trends in regard to direct and indirect aggression. Aggressive Behaviour 18: 117-127. Boersma, P. 2001. Praat, a system for doing phonetics by computer. Glot International 5 (9/10): 341-345. Boldea, I. 2009. Lirica românească premodernă. PoeĠii Văcăreúti. Limba Română XIX (5-6): 1-5. Borgoras, W. 1922. Chukchee. In F. Boas (ed.), Handbook of American Indian Languages 631-903, Part I and II. Washington, DC: Bureau of American Ethnology. Bourdieu, P., Boltanski, L. 1975. Le fétichisme de la langue. Actes de la recherche en sciences sociales 1 (4): 2-32. Bradley, J. 2006. Yanyuwa: ‘Men speak one way, women speak another’. In J. Coates (ed.), Language and Gender. A Reader, 13-20. Oxford: Blackwell Publishing.

Gender-Related Variation in the Speech of English and Romanian Adolescents

197

Brannon, L. 2005. Gender: Psychological Perspectives. Boston, MA: Ally and Bacon. Brown, P. 1980. How and why are women more polite?: Some evidence from a Mayan community. In S. McConnel-Ginet, R. Borker and N. Furman (eds.), Women and Language in Literature and Society. New York: Praeger. Reprinted in J. Coates (ed.). 1988. Language and Gender: A Reader, 81-99. Oxford: Blackwell Publishers. Brown, P., Levinson, S. 1978. Universals in language usage: politeness phenomena. In E. N. Goody (ed.), Questions and Politeness: Strategies in Social Interaction, 56-310. Cambridge: Cambridge University Press. Brown, P, Levinson, S. 1987. Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Bucholtz, M. 1999. Why be normal? Language and identity practices in a community of nerd girls. Language in Society 28 (2): 203-223. Bucholtz, M. 2002. From sex differences to gender variation in sociolinguistics. University of Pennsylvania Working Papers in Linguistics 8 (3): 33-45. Burridge, K. 2004. Blooming English: Observations on the Roots, Cultivation and Hybrids of the English Language. Cambridge: Cambridge University Press. Butler, J. 1990. Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge. Butler, J. 1993. Bodies that Matter: On the Discursive Limits of “Sex”. New York: Routledge. Cameron, D. 1998. Gender, language and discourse: A review essay. Signs: Journal of Women in Culture and Society 23: 945-973. Campbell, L. 1997. American Indian Languages: The Historical Linguistics of Native America. New York: Oxford University Press. Campbell-Kibler, K. 2006. Listeners Perceptions of Sociolinguistic Variables: The case of (ING). PhD dissertation, University of Stanford. Cătănescu, C. M. 1996. Limba română: Origini úi dezvoltare. Studiu, antologie de texte româneúti vechi, explicaĠii, glosar úi bibliografie. Bucharest: Humanitas. Cedergren, H. J., Sankoff, D. 1974. Variable rules: Performance as a statistical reflection of competence. Language 50 (2): 333-355. Chambers, J. K. 2002. Patterns of variation including change. In J. K. Chambers, P. Trudgill and N. Schilling-Estes (eds.), The Handbook of Language Variation and Change, 349-372. Malden, MA and Oxford: Blackwell Publishing. Chambers, J. K. 2003. Sociolinguistic Theory: Linguistic Variation and Its Social Significance. Malden, MA and Oxford: Blackwell.

198

Bibliography

Cheshire, J. 1982. Variation in an English Dialect. Cambridge: Cambridge University Press. Cheshire, J. 1997. Linguistic variation and social function. In N. Coupland and A. Jaworski (eds.), Sociolinguistics. A Reader and Coursebook, 185-198. London: Macmillan. Cheshire, J. 2002. Sex and gender in variationist research. In J. K. Chambers, P. Trudgill, N. Schilling-Estes (eds.), The Handbook of Language Variation and Change, 423-443. Malden, MA and Oxford: Blackwell Publishing. Cheshire, J. 2006. Linguistic variation and social function. In J. Coates (ed.), Language and Gender. A Reader, 29-41. Oxford: Blackwell Publishing. Cheshire, J., Kerswill, P., Fox, S., Torgersen, E. 2011. Contact, the feature pool and the speech community: The emergence of Multicultural London English. Journal of Sociolinguistics 15 (2): 151-196. ChiĠoran, D., Ionescu-Ruxăndoiu, L. (eds.). 1975. Sociolingvistică. Orientări actuale. Bucharest: Editura Didactică úi Pedagogică. Coates, J. 1993. Women, Men and Language. London and New York: Longman. Coates, J. 2004. Women, Men and Language, 3rd edition, Harlow: Longman. Coates, J. 2007. Gender. In C. Llamas, L. Mullany, P. Stockwell (eds.), The Routledge Companion to Sociolinguistics, 62-69. New York: Routledge. Coates, R. 2006. Properhood. Language 82 (2): 356-382. Cofer, T. 1972. Linguistic Variability in a Philadelphia Community. PhD dissertation, University of Philadelphia. Coleman, J. C. 1974. Relationships in Adolescence. London: Routledge and Kegan Paul. Coulmas, F. 2013. Sociolinguistics: The Study of Speakers’ Choices, 2nd edition. Cambridge: Cambridge University Press. Crawford, M. 1995. Talking Differently: On Gender and Language. London: Sage. Croitoru Bobârniche, N. 1996. DicĠionar de argou al limbii române. Slobozia: Arnina. Culpeper, J. 1996. Towards an anatomy of impoliteness. Journal of Pragmatics 25 (3): 349-367. Dooling, R. 1996. Blue Streak. New York: Random House. Downes, W. 1984. Language and Society. London: Fontana Press.

Gender-Related Variation in the Speech of English and Romanian Adolescents

199

Dubois, S., Horvath, B. M. 1999. When the music changes, you change too: Gender and language change in Cajun English. Language Variation and Change 11 (3): 287-314. Ebble, C. 1996. Slang and Sociability. Chapel Hill and London: The University of North Carolina Press. Eckert, P. 1980. Clothing and geography in a suburban high school. In C. P. Kottak (ed.), Researching American Culture, 139-145. Ann Arbor, MI: University of Michigan Press. Eckert, P. 1989. Jocks and Burnouts: Social Identity in the High School. New York: Teachers College Press. Eckert, P. 1990. The whole woman: Sex and gender differences in variation. Language Variation and Change 1 (3): 245-267. Eckert, P. 1998. Gender and sociolinguistic variation. In J. Coates (ed.), Language and Gender: A Reader, 64-75. Oxford: Blackwell. Eckert, P. 2000. Linguistic Variation as Social Practice. Oxford: Blackwell. Eckert, P. 2003. Language and gender in Adolescence. In J. Holmes and M. Meyerhoff (eds.), The Handbook of Language and Gender, 381400. Oxford: Blackwell. Eckert, P. 2004. Adolescent language. In E. Finegan and J. Rickford (eds.), Language in the USA, 361-374. Cambridge and New York: Cambridge University Press. Eckert, P. 2012. Three waves of variation study: The emergence of meaning in the study of variation. Annual Review of Anthropology 41: 87-100. Eckert, P., McConnel-Ginet, S. 1999. New generalisations and explanations in language and gender research. Language in Society 28 (2): 185-201. Eckert, P., and McConnel-Ginet, S. 2003. Language and Gender. Cambridge: Cambridge University Press. Eckert, P., McConnel-Ginet, S. 2006. Communities of practice: Where language, gender, and power all live. In J. Coates (ed.), Language and Gender: A Reader, 484-494. Oxford: Blackwell Publishing. Eisikovits, E. 1988. Girl-talk/Boy-talk: Sex differences in adolescent speech. In P. Collins and D. Blair (eds.), Australian English, 45-58. Sydney: Australian Professional Publication. Eisikovits, E. 2006. Girl-talk/Boy-talk: sex differences in adolescent speech. In J. Coates (ed.), Language and Gender. A Reader, 42-54. Oxford: Blackwell Publishing. Ekka, F. 1972. Men’s and women’s speech in Kurux. Linguistics 81 (1): 21-31.

200

Bibliography

Fasold, R. 1990. The Sociolinguistics of Language. Oxford: Blackwell Publishers. Fausto-Sterling, A. 2000. Sexing the Body: Gender Politics and the Construction of Sexuality. New York: Basic Books. Fetterman, M. D. 1989. Ethnography: Step by Step. Newbury Park, CA: Sage Publications. Fisher, J. L. 1958. Social influences on the choice of a linguistic variant. Word 14 (1): 47-56. Fishman, J. 1972. The Sociology of Language: An Interdisciplinary Social Science Approach to Language in Society. Rowley, Ma: Newbury House. Fishman, P. 1978. What do couples talk about when they’re alone? In D. Burthurff and E. L. Epstein (eds.), Women’s Language and Style, 1122. Akron, OH: L & S Books. Fleischman, S. 1998. Gender, the personal, and the voice of scholarship: A viewpoint. Signs: Journal of Women in Culture and Society 23: 9781016. Foulkes, P., Docherty, G. 2006. The social life of phonetics and phonology. Journal of Phonetics 34 (4): 409-438. Fraser, B., Nolan, W. 1981. The association of deference with linguistic form. International Journal of the Sociology of Language 27 (1): 93109. Freud, S. 1950. Totem and Taboo, transl. James Strachey. London: Routledge and Kegan Paul. Furfey, P. H. 1944. Men’s and women’s language. The American Catholic Sociological Review 5: 218-223. Goffman, E. 1967. Interaction Ritual. Essays on Face-to-Face Behaviour. New York: Pantheon Books. Graddol, D. and Swan, J. 1989. Gender Voices. Oxford: Basil Blackwell. Gray, J. 1992. Men Are from Mars, Women Are from Venus: A Practical Guide to Improving Communication and Getting what You Want in Your Relationships. New York: Harper Collins. Grice, P. 1975. Logic and conversation. In P. Cole and J. Morgan (eds.), Syntax and Semantics, vol. 3, Speech Acts, 41-58. New York: Academic Press. Gumperz, J. J. 1968. The speech community. In International Encyclopedia of the Social Sciences, 381-386. New York: Macmillan. Gumperz, J. 1982. Discourse Strategies. Cambridge: Cambridge University Press. Haas, M. 1944. Men’s and women’s speech in Koasati. Language 20 (1): 142-149.

Gender-Related Variation in the Speech of English and Romanian Adolescents

201

Holder, B. 2008. Dictionary of Euphemisms. New York: Oxford University Press. Holmes, J. 1995. Women, Men and Politeness. New York: Longman. Holmes, J. 1997. Women, language and identity. Journal of Sociolinguistics 1 (2): 195-223. Holmes, J.1998.Women’s talk: The question of sociolinguistic universals. In: J. Coates (ed.), Language and Gender. A Reader, 461-483. Oxford: Blackwell. Holmes, J. 2008. An Introduction to Sociolinguistics. Harlow: Longman. Hornby, A. S. 2005. Oxford Advanced Learner’s Dictionary of Current English. Cambridge: Cambridge University Press. Hornoiu, D. 2002. Gendered language. Analele UniversităĠii Ovidius. Seria filologie XIII: 115-134. Hornoiu, D. 2007. Language and Gender. An Analysis of Conversational Discourse in English and Romanian. PhD dissertation, University of Bucharest. Hornoiu, D. 2008. Language and Gender. An Analysis of Conversational Discourse in English and Romanian. ConstanĠa: Ovidius University Press. Horvath, B. M. 1985. Variation in Australian English: The Sociolects of Sydney. Cambridge: Cambridge University Press. Houston, A. 1985. Continuity and Change in English Morphology. The Variable (ing). PhD dissertation, University of Pennsylvania. Hudson, R. A.1980. Sociolinguistics. Cambridge: Cambridge University Press. Hughes, A. and Trudgill, P. 1987. English Accents and Dialects An Introduction to Social and Regional Varieties of British English. London: Edward Arnold. Hughes, G. 2006. An Encyclopedia of Swearing. The Social History of Oaths, Profanity, Foul Language and Ethnic Slurs in the Englishspeaking World. New York: M. E. Sharpe. Hughes, S. E. 1992. Expletives of lower working-class women. Language in Society 21 (2): 291-303. Ionescu-Ruxăndoiu, L. 1999. ConversaĠia. Structuri úi strategii. Sugestii pentru o pragmatică a românei vorbite. Bucharest: All. Iordan, I. 1975. Stilistica limbii române, 2nd edition. Bucharest: Editura ùtiinĠifică. James, D., Clarke, S. 1993. Women, men and interruptions: A critical review. In D. Tannen (ed.), Gender and Conversational Interaction, 231-280. New York and Oxford: Oxford University Press.

202

Bibliography

Jay, T. 1980. Sex differences and dirty word usage: A review of the literature and reply to Haas. Psychological Bulletin 88: 614-621. Jay, T. 1992. Cursing in America. Philadelphia: John Benjamins. Jay, T. 1999. Why we Curse. Amsterdam/Philadelphia: John Benjamins. Jay, T. 2009. The utility and ubiquity of taboo words. Perspectives on Psychological Science 4: 153-161. Jespersen, O. 1922. Language: Its Nature, Development and Origin. London: George Allen & Unwin. Johnson, D. E. 2009. Getting off the GoldVarb standard: Introducing Rbrul for mixed effects variable rule analysis. Language and Linguistics Compass 3 (1): 359-383. Kerswill, P., and Williams, A. 1997. Investigating social and linguistic identity in three British schools. In U.-B. Kotsinas, A.-B. Stenström and A.-M. Malin (eds.), Ungdomsspråk i Norden. Föredrag från ett forskarsymposium, 159-176. Stockholm: University of Stockholm, Department of Nordic Languages and Literature. Kiesling, S. F. 1998. Men’s identities and sociolinguistic variation: The case of fraternity men. Journal of Sociolinguistics 2 (1): 69-99. Kiesling, S. F. 2011. Linguistic Variation and Change. Edinburgh: Edinburgh University Press. de Klerk, V. 1991. Expletives: Men only? Communication Monographs 58: 156-169. de Klerk, V. 1992. How taboo are taboo words for girls? Language in Society 21 (2): 277-89. Koerner, K. 1991. Toward a history of modern sociolinguistics. American Speech 66 (1): 57-70. Kutner, N. G., Brogan, D. 1974. An investigation of sex-related slang vocabulary and sex role orientation among male and female university students. Journal of Marriage and the Family 36: 474-484. Kuwahara, Y. L. 1998. Interactions of Identity: Inner-city Immigrant and Refugee Youths, Language Use, and Schooling. PhD dissertation, Stanford University. Labov, W. 1965. Stages in the acquisition of Standard English. In E. R. Shuy (ed.), Social Dialects and Language Learning, 77-103. Champaign, IL: National Council of Teachers of English. Labov, W. 1966. The Social Stratification of English in New York City. Washington DC: Center for Applied Linguistics. Labov, W. 1969. Contraction, deletion, and inherent variability of the English copula. Language 45 (4): 715-762. Labov, W. 1970. The study of language in its social context. Studium Generale 23 (1): 30-87.

Gender-Related Variation in the Speech of English and Romanian Adolescents

203

Labov, W. 1972a. Language in the Inner City. Philadelphia, PA: University of Pennsylvania Press. Labov, W. 1972b. Sociolinguistic Patterns. Philadelphia, PA: University of Pennsylvania Press. Labov, W. 1972c. Some principles of linguistic methodology. Language in Society 1 (1): 97-120. Labov, W. 1973. The boundaries of words and their meanings. In J. Fishman (ed.), New Ways of Analyzing Variation in English, 340-373. Washington, DC: Georgetown University Press. Labov, W. 1989. The child as linguistic historian. Language Variation and Change (1): 85-98. Labov, W. 1990. The intersection of sex and social class in the course of linguistic change. Language Variation and Change 2 (2): 205-254. Labov, W. 2001. Principles of Linguistic Change, vol. 2: Social Factors. Malden, MA and Oxford: Blackwell. Labov, W. 2007. Transmission and diffusion. Language 83 (2): 344-387. Labov, W., Cohen, P., Robins, C., Lewis, J. 1968. A Study of Nonstandard English of Negro and Puerto-Rican Speakers in New York City. Final Report, Co-operative Research Project 3288. Washington, DC: U.S. Office of Health, Education and Welfare. Lakoff, R. 1975. Language and Woman’s Place. New York: Harper & Row. Lave, J., Wenger, E. 1991. Situated Learning: Legitimate Peripheral Participation. Cambridge: Cambridge University Press. Lawson, E and Stuart-Smith J. 1999. A sociophonetic investigation of the ‘Scottish’ consonants (/x/ and /hw/) in the speech of Glaswegian children. In J. J. Ohala, Y. Hasegawa, M. Ohala, D. Granville and A. C. Bailey (eds.) Proceedings of the XIVth International Congress of Phonetic Sciences, 2541-2544. Berkeley, CA: University of California. Leech, G. 1983. Principles of Pragmatics. London: Longman. Levon, E. 2013. Ethnographic fieldwork. In C. Mallinson, B. Childs, and G. Van Herk (eds.), Data Collection in Sociolinguistics: Methods and Applications, 69-79. New York: Routledge. Lighter, J. E. (ed). 1978. Introduction. Historical Dictionary of American Slang, vol. 1, New York: Random House. Ljung, M. 1984. Swearing. Studier i modern språkvetenskap utgivna i samverkan med nyfilologiska sällskapet i Stockholm. Stockholm: Almqvist och Wiksell International. Ljung, M. 2011. Swearing: A Cross-Cultural Linguistic Study. Basingstoke: Palgrave Macmillan.

204

Bibliography

Locker, M. A., Watts, R. J. 2005. Politeness theory and relational work. Journal of Politeness Research 1 (1): 9-33. Macauley, R. K. S. 1977. Language, Social Class, and Education: A Glasgow Study. Edinburgh: Edinburgh University Press. Maccoby, E. M., Jacklin, C. N. 1974. The Psychology of Sex Differences. Stanford, CA: Stanford University Press. MacLeod, J. 1987. Ain’t no Making it. Leveled Aspirations in a LowIncome Neighborhood. Colorado, CO: Westview Press. Macmillan English Dictionary. 2002. Oxford: Macmillan. Maltz, D. N. and Borker, R. A. 1982. A cultural approach to male-female miscommunication. In J. Gumperz (ed.), Language and Social Identity, 196-216. Cambridge: Cambridge University Press. Mardale, A. 2006. Categorii lexicale versus categorii funcĠionale via categorii semilexicale. Cazul prepoziĠiilor a, de, la úi pe din limba română. Available at < http://www.llf.cnrs.fr/Gens/Mardale/mardale_categ.lex_categ.funct200 6.pdf> (last accessed 18.04. 2014). McArthur, T. 1992. The Oxford Companion to the English Language: Oxford: Oxford University Press. McEnery, T. 2006. Swearing in English. Bad Language, Purity and Power from 1586 to the Present. London and New York: Routledge. McEnery, T. 2009. Bad language. In J. Culpeper, F. Katamba, P. Kerswill, R. Wodak, T. McEnery (eds.), English Language. Description, Variation and Context, 564-575. Basingstoke: Palgrave Macmillan. Mendoza-Denton, N. 1996. ‘Muy macha’: Gender and ideology in ganggirls’ discourse about makeup. ETHNOS 61 (1-2): 47-63. Meyerhoff, M. 2006. Introducing Sociolinguistics. London and New York: Routledge. Meyerhoff, M., Strycharz, A. 2013. Communities of practice. In J. K. Chambers and N. Schilling-Estes (eds.), The Handbook of Language Variation and Change, 2nd edition, 428-447. London and New York: John Wiley and Sons. Milroy, J. 1992. Linguistic Variation and Change. Oxford: Blackwell. Milroy, J., Milroy, L., Hartley, S. 1994. Local and supra-local change in British English: The case of glottalisation. English World-Wide 15 (1): 1-33. Milroy, L. 1980. Language and Social Networks. Baltimore, MD: University Park Press. Milroy, L. 1987. Language and Social Networks. Oxford: Wiley.

Gender-Related Variation in the Speech of English and Romanian Adolescents

205

Milroy, L. 1992. New perspectives in the analysis of sex differentiation in language. In K. Bolton and H. Kwok (eds.), Sociolinguistics today: International Perspectives, 162-179. London: Routledge. Milroy, L., Gordon, M. 2003. Sociolinguistics: Method and Interpretation. Malden, MA and Oxford: Blackwell. Mondorf, B. 2002. Gender differences in English syntax. Journal of English Linguistics 30 (2): 158-180. Mithun, M. 2006. The Languages of Native North America. New York: Cambridge University Press. Nathan, R, 1996. My Freshman Year: What a Professor Learned by Becoming a Student. New York: Penguin. Newbrook, M. 1982. Sociolinguistic Reflexes of Dialect Interference in West Wirral. PhD dissertation, Reading University. Noller, P. and Fitzpatrick, M. A. 1993. Communication in Family Relationships. Englewood Cliffs, NJ: Prentice-Hall. Oancea, C. 2011a. Some notes on language and gender. Revista Română de Studii Eurasiatice 7 (1-2): 53-62. Oancea, C. 2011b. Gender exclusive differences in language use. In Analele UniversităĠii Ovidius. Seria filologie XXII (1): 187-199. Oancea, C. 2011c. Language and wo(men)’s place in 21st century Romania. Ethnologia Balkanica 15 (1): 295-310. Oancea, C. 2013. Language and Gender-related Variation in English and Romanian. Saarbrüchen: Lambert Academic Publishing. Partridge, E. 1933. Slang Today and Yesterday. London: Routledge and Kegan Paul. Paolillo, J. C. 2001. Analyzing linguistic variation: Statistical models and methods. Stanford, CA: CSLI. Patrick, P. L. 2002. The speech community. In J. K. Chambers, P. Trudgill, and N. Schilling-Estes (eds.), The Handbook of Language Variation and Change, 573-597. Malden, MA: Blackwell. Pilotti, M., Almand, J., Mahamane, S., Martinez, M. 2012. Taboo words in expressive language: Do sex and primary language matter? American International Journal of Contemporary Research 2 (2): 17-26. Pinker, S. 2007. The Stuff of Thought: Language as a Window into Human Nature. New York: Viking Press. Poplack, S. 1979. Function and Process in a Variable Phonology. PhD dissertation, University of Pennsylvania. Poplack, S. 1993. Variation theory and language contact. In D. Preston (ed.), American Dialect research: An Anthology Celebrating the 100th Anniversary of the American Dialect Society, 251-286. Amsterdam: Benjamins.

206

Bibliography

Poplack, S., Tagliamonte, S. 2004. Back to the present: Verbal -s in the (African American) diaspora. In R. Hickey (ed.), Legacies of Colonial English, 203-223. Cambridge: Cambridge University Press. PRAAT, version 5.3.84, available here (last accessed 07.07.2014). Puúcariu, S. 1974. Cercetări úi studii. Bucharest: Editura Minerva. Quirk, R., Greenbaum, S., Leech, G., Svartvik, J. 1985. A Comprehensive Grammar of the English Language. London: Longman. Rasinger, S. 2008. Quantitative Research in Linguistics. London: Continuum. Read, A. W. 1977. Classic American Graffiti: Lexical Evidence from Folk Epigraphy in Western North America. Waukesha, WI: Maledicta Press. Reid, E. 1978. Social and stylistic variation in the speech of children: Some evidence from Edinburgh. In P. Trudgill (ed.), Sociolinguistic Patterns in British English, 158-171. London: Edward Arnold. Richards, K. 2003. Qualitative Inquiry in TESOL. Basingstoke: Palgrave Macmillan. Risch, B. 1987. Women’s derogatory terms for men: That’s right, ‘dirty’ words. Language in Society 16 (3): 353-358. Romaine, S. 1978. Post-vocalic /r/ in Scottish English: Sound change in progress. In P. Trudgill (ed.), Sociolinguistic Patterns in British English, 144-157. London: Edward Arnold. Romaine, S. 1980. A critical overview of the methodology of urban British sociolinguistics. English World-Wide 1 (2): 163-198. Romaine, S. 1984. The Language of Children and Adolescents: The Acquisition of Communicative Competence. Oxford: Basil Blackwell. Romaine, S. 2001. A corpus-based view of gender in British and American English. In M. Hellinger, H. Bußmann (eds.), Gender across Languages, vol. 1, 113-135. Amsterdam/Philadelphia, PA: John Benjamins. Romaine, S. 2003. Variation in language and gender. In J. Holmes and M. Meyerhoff (eds.), The Handbook of Language and Gender, 98-118. Oxford: Blackwell. Rosetti, A. 1968. Istoria limbii române de la origini până în secolul XVIIlea. Bucharest: Editura pentru literatură. Rubin, D. L., and Greene, K. G. 1992. Gender-typical style in written language. Research in the Teaching of English 26: 7-40. Sankoff, D. 1985. Statistics in linguistics. New York Encyclopedia of the Statistical Sciences 5, 71-81. Malden, MA: Wiley. Sankoff, D. 1988. Variable rules. In U. Ammon, N. Dittmar, K. J. Mathheier (eds.), Berlin Sociolinguistics: An International Handbook

Gender-Related Variation in the Speech of English and Romanian Adolescents

207

of the Science of Language and Society, vol. 2, 984-997. Berlin: Walter de Gruyter. Sankoff, D, Cedergren, H. J., Kemp, W., Thibault, P., Vincent, D. 1989. Montreal French: Language, class and ideology. In R. W. Fasold and D. Schiffrin (eds.), Language Change and Variation, 107-118. Amsterdam/Philadelphia, PA: John Benjamins. Sankoff, D., Laberge, S. 1978. The linguistic market and the statistical explanation of variability. In D. Sankoff (ed.), Linguistic Variation: Models and Methods, 239-250. New York: Academic Press. Sankoff, G. 1980. The Social Life of Language. Philadelphia, PA: University of Philadelphia. Sarău, G. 1992. Mic dicĠionar rom-român. Bucharest: Kriterion. Saville-Troike, M. 2003. The Ethnography of Communication. 3rd edition. Malden, MA: Blackwell. Schilling-Estes, N. 2002. American English social dialect variation and gender. Journal of English Linguistics 30 (2): 122: 137. Schilling, N. 2011. Language, gender, and sexuality. In R. Mesthrie (ed.), The Cambridge Handbook of Sociolinguistics. Cambridge: Cambridge University Press. Schilling, N. 2013. Sociolinguistic Fieldwork. Cambridge: Cambridge University Press. Schleef, E., Meyerhoff, M., Clark, L. 2011. Teenager’s acquisition of variation. A comparison of locally-born and migrant teens’ realisation of English (ing) in Edinburgh and London. English World Wide 32 (2): 206-236. Shopen, T. 1978. Research on the variable (ING) in Canberra, Australia. Talanya 5: 42-52. Shuy, R., Wolfram, W., Riley, W. K. 1967. Linguistic Correlates of Social Stratification in Detroit Speech. USOE Final Report No. 6-1347. Sorensen, A. 1967. Multilingualism in the Northwest Amazon. American Anthropologist 69 (6): 670-684. Spender. D. 1980. Man Made Language. London: Routledge. Steffen, V. J., Eagly, A. H. 1985. Implicit theories about influence of style: The effects of status and sex. Personality and Social Psychology Bulletin 11 (2): 191-205. Stenström, A. B. 2014. Teenage Talk: From General Characteristics to the Use of Pragmatic Markers in a Contrastive Perspective. Basingstoke: Palgrave Macmillan. Stenström, A. B., Andersen, G., Hasund, I. K. 2002. Trends in Teenage Talk. Corpus Compilation, Analysis and Findings. Amsterdam/ Philadelphia, PA: John Benjamins.

208

Bibliography

Stenzel, K. 2005. Multilingualism in the Northwest Amazon, revisited. Memorias del Congreso de Idiomas Indígenas de Latinoamérica-II, University of Texas at Austin, available here:

(last accessed: 10. 05. 2013). Stuart-Smith, J. 1999. Glasgow: Accent and voice quality. In P. Foulkes and G. Docherty (eds.), Urban Voices: Accent Studies in the British Isles, 201-222, Leeds: Arnold. Tagliamonte, S. 2004. Have to, gotta, must: Grammaticalization, variation and specialization in English deontic modality. In H. Lindquist and C. Mair (eds.), Corpus Research on Grammaticalization in English, 3755. Amsterdam/Philadelphia: John Benjamins. Tagliamonte, S. 2005. So who? Like how? Just what? Discourse markers in the conversations of young Canadians. Journal of Pragmatics 37 (11): 1896-1915. Tagliamonte, S. 2006. Analysing Sociolinguistic Variation. Cambridge: Cambridge University Press. Tagliamonte, S. 2011. Variationist Sociolinguistics. Change, Observation, Interpretation. Malden, MA and Oxford: Wiley-Blackwell. Tagliamonte, S., Smith, J. 2006. Layering, change and a twist of fate: Deontic modality in dialects of English. Diachronica 23 (2): 341-380. Talbot, M. 2010. Language and Gender, 2nd edition. Cambridge: Polity Press. Tandin, T. 1993. Limbajul infractorilor. Bucharest: Paco. Tannen, D. 1984. Conversational Style: Analyzing Talk Among Friends. Norwood, NJ: Ablex. Tannen, D. 1990. You just Don’t Understand: Women and Men in Conversation. New York: Oxford University Press. Tannen, D. 1994. Gender and Discourse. New York and Oxford: Oxford University Press. Taylor, A. 1982. ‘Male’ and ‘female’ speech in Gros Ventre. Anthropological Linguistics 24 (3): 301-307. The Penguin Dictionary of American English Usage and Style. 2002. New York: Penguin. Thimm, C. 1995. Durchsetzungsstrategien von Frauen und Männern: Sprachliche Unterschiede oder stereotype Erwartungen? In C. M. Heilmann (ed.), Frauensprechen- Männersprechen. Geschlechtsspezifisches Sprechverhalten, 120-129. Munich and Basel: Reinhard. Thimm, C., Koch, S. C., Schey, S. 2003. Communicating gendered professional identity: competence, and conflict in the workplace. In J.

Gender-Related Variation in the Speech of English and Romanian Adolescents

209

Holmes and M. Meyerhoff (eds.), The Handbook of Language and Gender, 528-549. Oxford: Blackwell. Trudgill, P. 1972. Sex, covert prestige and linguistic change in urban British English. Language in Society 1 (2): 179-195. Trudgill, P. 1974. The Social Differentiation of English in Norwich. Cambridge: Cambridge University Press. Trudgill, P. 1975. Accent, Dialect and the School. London: Edward Arnold. Trudgill, P. 1983a. Sociolinguistics: An Introduction to Language and Society, 2nd edition. Harmondsworth: Penguin. Trudgill, P. 1983b. Acts of conflicting identity: The sociolinguistics of British pop song pronunciation. In P. Trudgill, On Dialect: Social and Geographical Perspectives, 141-160. Oxford: Blackwell. Trudgill, P. 2000. Sociolinguistics: An Introduction to Language and Society, 4th edition. London: Penguin. Trudgill, P. 2003. The Norfolk Dialect. Cromer, England: Poppyland Publishing. Volceanov, A., Volceanov, G. 1998. DicĠionar de argou úi expresii familiar ale limbii române. Bucharest: Livpress. Wald, B. Shopen, T. 1985. A researcher’s guide to the sociolinguistic variable (ING). In V. Clark, P. Escholtz, A. Rosa (eds.), Language: Introductory Readings, 515-542. New York: St. Martin’s Press. Wardhaugh, R. 2005. An Introduction to Sociolinguistics, 5th edition. Oxford: Blackwell. Wardhaugh, R. 2009. An Introduction to Sociolinguistics, 6th edition. Oxford: Blackwell. Weinreich, U., Labov, W., Herzog, M. 1968. Empirical foundations for a theory of language change. In W. P. Lehmann and Y. Malkiel (eds.), Directions for Historical Linguistics, 95-188. Austin: University of Texas Press. Wells, J. C. 1982. Accents of English: The British Isles. Cambridge: Cambridge University Press. Wenger, E. 1998. Communities of Practice. Cambridge: Cambridge University Press. West, C., and Zimmerman D. 1983. Mall insults: A study of interruptions in cross-sex conversation between unacquainted persons. In B. Thorne, C. Kramarae, N. Henley (eds.), Language, Gender and Society, 102117. Rowley, MA: Newbury House. West, C., and Zimmerman D. 1987. Doing gender. Gender & Society 1 (2): 125-51. Wodak, R. (ed). 1997. Gender and Discourse. London: Sage.

210

Bibliography

Wolfram, W. 1969. A Sociolinguistic Description of Detroit Negro Speech. Washington, DC: Center for Applied Linguistics. Wolfram, W, K. Hazen and N. Schilling-Estes. 1999. Dialect Change and Maintenance in Outer Banks English. Tuscaloosa, AL: University of Alabama Press. Wolfram, W and Schilling-Estes, N. 2008. American English. Dialects and Variation, 2nd edition. Oxford: Blackwell Publishing. Wolfson, N. 1976. Speech events and natural speech: Some implications for sociolinguistic methodology. Language in Society 5 (2): 189-209. Woods, H. 1979. A Socio-Dialectal Survey of the English Spoken in Ottawa: a Study of Sociolinguistic and Stylistic Variation. PhD dissertation, University of Ottawa. Zafiu, R. 2001. Naúpa. România Literară online, 36, available at < http://www.romlit.ro/napa> (last accessed July, 20, 2014). Zafiu, R. 2009. Miúto úi legenda bastonului. România Literară online, 6, available at < http://www.romlit.ro/mito_i_legenda_bastonului> (last accessed July, 20, 2014). Zafiu, R. (ed.) 2010a. Eúti COOL úi dacă vorbeúti corect. Bucureúti: Univers Enciclopedic Gold. Zafiu, R. 2010b. 101 cuvinte argotice. Bucharest: Humanitas. Zimmerman, D., West. C. 1975. Sex roles, interruptions and silences in conversation. In B. Thorne, N. Henley (eds.), Language and Sex: Difference and Dominance, 105-129. Rowley, MA: Newbury House.