This item has the following downloads:
1 YOUTH LANGUAGE MEETS POLITICAL DISCOURSE: USAGE OF FRENCH PRINTEMPS RABLE By ROBYN ALEXIS NOVAK A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT O F THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2013
2 2013 Robyn Alexis Novak
3 To the students of Quebec and all others fighting to make education available to and affordable for a ll To my Mom, for her unwavering support
4 ACKNOWLEDGMENTS I thank my professors, who have given me the priceless gift of knowledge and members, Dr. Hlne Blondeau and Dr. Theresa Antes, for their guidance in the development and writing of this study. In addition, I would like to thank the faculty and staff of the Department of Languages, Literatures, and Cultures at the University of Florida for their assistance. I also thank all of my classmates and colleagues in the department. I would also like to thank my family and friends for their support and encouragement throughout the process of writing this thesis and throughout my graduate education.
5 TABLE OF CONTENTS ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF ABBREVIATIONS ................................ ................................ ............................. 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 11 Youth Language ................................ ................................ ................................ ..... 13 The Context: Le Printemps rable ................................ ................................ .......... 15 2 LITERATURE REVIEW ................................ ................................ .......................... 18 Age in Sociolinguistics ................................ ................................ ............................ 18 Discourse Markers in Pragmatics and Sociolingui stics ................................ ........... 21 Public and Political Discourse in Sociolinguistics ................................ .................... 30 Conclusion ................................ ................................ ................................ .............. 31 3 STUDY DESIGN ................................ ................................ ................................ ..... 33 The Methodological Framework ................................ ................................ .............. 33 The Data ................................ ................................ ................................ ................. 33 The Spokespeople ................................ ................................ ........................... 34 The Data Collection ................................ ................................ .......................... 35 Conclusion ................................ ................................ ................................ .............. 36 4 FUNCTIONS OF THE DISCOURSE MARKERS ................................ .................... 38 Introduction: Form and Function ................................ ................................ ............. 38 Classification of Data ................................ ................................ .............................. 39 Agreement Marker ................................ ................................ ............................ 39 Interactive Marker ................................ ................................ ............................. 40 Progression Marker ................................ ................................ .......................... 41 Filler or Hesitation Marker ................................ ................................ ................ 43 Causation Marker ................................ ................................ ............................. 44 Contradiction Marker ................................ ................................ ........................ 45 Command Softener ................................ ................................ .......................... 46 Discourse Marker Usage ................................ ................................ ........................ 46 Conclusion ................................ ................................ ................................ .............. 48 5 INTER INDIVIDUAL VARIATION ................................ ................................ ........... 50
6 Chapter Introduction ................................ ................................ ............................... 50 Speakers ................................ ................................ ................................ ................. 50 Gabriel Nadeau Dubois ................................ ................................ .................... 50 Lo Bureau Blouin ................................ ................................ ............................ 51 Martine Desjardins ................................ ................................ ........................... 52 Conclusion ................................ ................................ ................................ .............. 54 6 DISCUSSION AND GENERAL CONCLUSIONS ................................ .................... 55 APPENDIX: CORPUS S OURCES ................................ ................................ ................ 60 LIST OF REFERENCES ................................ ................................ ............................... 61 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 65
7 LIST OF TABLES Table page 3 1 Discourse marker forms by speaker ................................ ................................ ... 37 4 1 Discourse marker forms by function ................................ ................................ ... 49 5 1 Discourse marker function by speaker ................................ ............................... 54
8 LIST OF ABBREVIATIONS CLASSE tudiante: one of the three main student organizations involved in the strikes. FECQ Fdration tudiante collgiale du Qubec: one of the three main student organizations involved in the strikes. FEUQ Fdration tudiante universitaire du Qubec: one of the three main student organizations involved in the strikes. GND Ga briel Nadeau Dubois: co spokesperson of the CLASSE. LBB Lo Bureau Blouin: spokesperson of the FECQ. MD Martine Desjardins: spokesperson of the FEUQ.
9 Abstract of Thesis Presented to the Graduate School of the University of Florida in P artial Fulfillment of the Requirements for the Degree of Master of Arts YOUTH LANGUAGE MEETS POLITICAL DISCOURSE: USAGE OF FRENCH PRINTEMPS RABLE By Robyn Alexis Novak August 2013 Chair: Hlne Blondeau Major: French and Francophone Studies This study analyzes discourse marker usage by three young student group Printemps rable of 2012, a social movement that was initiated by students protesting university tuition hikes. The corpus used was ga thered from Internet sources of interviews or conference lectures involving one or more of the three speakers studied. The main goal of the study was to examine how the usually higher register of political discourse manifests itself in young speakers, whos e language is generally analyzed in a more informal context. The discussion focuses on discourse marker forms and functions attested in the corpus and how they relate to each other, as well as observations on intra s discourse marker usage. The most prominent of the functions studied in the corpus were the progression marker and the filler. The most common forms attested in the corpus were donc en fait l and puis This analysis showed that Gabriel Nadeau Dubois d isplayed the most diverse usage of discourse markers in terms of form and function, while Lo Bureau Blouin demonstrated a relatively restricted usage. The markers comme genre and ( a) fait que which are usually not associated with more formal or stand ard speech, were
10 not attested in this corpus, suggesting that these speakers used a higher register when engaging in the political discourse seen in these interviews.
11 CHAPTER 1 INTRODUCTION In society as a whole, youth language as a vernacular is often stigmatized by standard language. Studies of language attitudes and representations (see Gadet 2007 and Faygal 2010, among others) have shown evidence of the marginalizatio n and stigmatization of the youth language vernacular by older speakers of the language. In linguistic studies, and especially in sociolinguistics, youth language vernacular, particularly adolescent speech, has been shown by some studies to be a catalyst f or linguistic change that is eventually adopted by the general population (See Labov 1972, Tagliamonte 1999). In the Francophone world, attitudes towards youth language have evolved over nalized (Gadet 2007) and in some cases such as youth language in the banlieues of Paris, where youth language has increasingly become characterized by mixing with immigrant languages, such as Arabic youth language has even come to be stigmatized (Fagyal toward youth language in French have evolved from regarding the vernacular as an elite examined prosodic elements of youth language in the banlieues of Paris, discusses the stigmatization of the youth language vernacular, which is viewed by some as a detriment to the French language and national identity. Linguistic studies with age grouping s as a social factor have shown variation in the speech of young speakers in several areas, including the usage of discourse
12 markers (see Heisler 1996, Tagliamonte and Hudson 1999, Thibault and Daveluy 1989, and Sankoff et al. 1997). Thibault and Daveluy ( 1989), for example, studied the difference in usage of five discourse markers based on age (interviewing the same speakers at different times in their lives), finding both differences in usage as a speaker ages (generational differences) and evidence of li nguistic change in progress. As Sankoff et al. (1997) stated, discourse markers are not taught in school and therefore family, friends, and other members of the language co mmunity. In this way, discourse markers are unique: they are a part of discourse that is picked up from other speakers of the language and not covered in grammar books or explicitly taught in school. For this reason, variation in usage of discourse markers based on individual speaker differences, age, and register can give insight into how the discourse marker lexicon is developed. Do young speakers add or delete certain discourse markers to their lexicon as they get older? What other social factors (e.g. g ender, socioeconomic background, education level, etc.) may influence which discourse markers are or are not used? Does discourse marker usage also vary based on the social context or register? The present study will focus on the speech of relatively young speakers and analyze the discourse marker forms and functions that they use in a very specific situation, that is, political discourse related to the student strikes known as the Printemps rable in Quebec during the spring of 2012. As public political di scourse generally belongs to a higher register of speech (Martel 2008), it is expected that the speech of the young speakers studied in this corpus will represent a higher register of youth language than would that of a young person just speaking with frie nds or family.
13 How will the young speakers studied, who are being interviewed in a political context where they are playing a critical leadership role, use discourse markers, in terms of both form and function? Youth Language In recent decades, several stu dies (see Gadet 2007, Fagyal 2010, and Thibault 1997, among others) have focused on variation in youth language and linguistic attitudes towards the youth vernacular in French. As will be discussed in Chapter 2, youth language is often stigmatized but, at the same time, it can serve as a catalyst for linguistic change. (Fagyal 2010; Thibault 1997). Perhaps partially due to varying attitudes towards it, youth language has recently come to be viewed as a sociolect in its own right, garnering attention from co untless sociolinguistic studies. Examples of variation in French youth language at the lexical level include borrowing, truncation of words, and verlanization, or the rearrangement of syllables to form new words (Gadet 2007). These types of variation can a lso, as Gadet notes, be combined, such as in the verlanization of a word borrowed from Arabic with an English suffix added. Lexical variation is just one aspect of youth language that creates a plethora of variables available for linguistic analysis. Other studies (e.g. Heisler 1996, and Sankoff et al. 1997) have examined the usage of discourse markers by young speakers of French. Heisler (1996) did not concentrate on young speakers as a group for his entire study, but rather examined age groups as one soci al factor influencing the usage of OK as a discourse marker. Sankoff et al. (1997) studied discourse markers of young Anglophone L2 French speakers because, as it is not taught in school, discourse markers usage is an indication of how well an L2 speaker h as integrated socially into the Francophone community.
14 Although many studies have examined youth language as a whole, we will also see variation between individuals in this age group. These differences may be based on factors such as education, the region in which the speaker and his/her parents were born and grew up, socio economic background, and others. Since the present study focuses on the speech of only three individual speakers, we will mostly discuss individual differences between these speakers rat her than general tendencies in youth narrow context, we will discuss their usage of discourse markers in this context only. The corpus used in the present study was gathered from television journalistic interviews of three relatively young speakers: Gabriel Nadeau Dubois (21 years old), Lo Bureau Blouin (20 years old) and Martine Desjardins (30 years old). These speakers were chosen for the corpus because, as spokespeople for the three major student groups of Quebec, they were interviewed many times by journalists. Their discourse in the context of these interviews was also very political in nature, since they were representing the student population in negotiations with polit icians. The fact that all of these interviews were filmed for television or the internet also adds a formality to the context of the interviews, as the speakers knew that they were being filmed and watched by countless viewers. Their speech in these inter views, therefore, represents a higher register of youth language, as they ultimately wanted to act in a professional manner and be taken seriously so that their agenda would be taken seriously. The interviews were conducted in 2012 during the Printemps ra ble in Quebec, which will be discussed in the following section.
15 The Context: Le Printemps rable The 2012 Quebec student strikes began as a student protest against tuition hikes announced by the government in 2010, but led to a larger social movement kno wn as le Printemps rable (Sorochan 2012). The tuition hikes being protested, announced by premier Jean Charest and the Parti Libral du Qubec (PLQ) in early 2010, were intended to raise tuition by 75% over five years beginning in 2012 (Sorochan 2012). Th e students of Quebec first responded with petitions and small demonstrations and of students voted in favor of a general unlimited strike in mid February 2012 (Soroch an 2012). Three student groups were largely involved in the organization of the strike and represented the students in negotiations with government officials: the Fdration tudiante collgiale du Qubec (FECQ); the Fdration tudiante universitaire du Q ubec (FEUQ); the Association pour une solidarit syndicale tudiante (ASS), specifically its temporary coalition formed for the 2012 student strikes, the Coalition deman d of the strike was the rescinding of the tuition increase, but more radical demands called for movement to a tuition free system; this goal was particularly sought after by the CLASSE, the most radical of the student groups, which holds in high esteem the ideal of free post secondary education (Sorochan 2012). Not surprisingly, the involvement of these student groups and the high profile nature of the strikes led to a high amount of public discourse on the topic of the strikes, especially by the relatively young spokespeople of these three groups. During the strikes, they were interviewed by countless journalists and gave press conferences and motivational speeches to the student population. The controversial nature of the topics
16 discussed in these intervie ws and the somewhat diverse social backgrounds of the spokespeople make this an important area of study for the sociolinguistic aspects of public discourse. Specifically, this study will analyze the discourse markers used by the spokespeople of these stude nt groups during interviews with the press, examining their functions in the context of public political discourse. Examining discourse marker usage is a clear way to analyze public discourse in these interviews and compare the results with previous studie s on similar forms and functions of discourse markers. During the student strikes, the spokespeople for the student groups took on the student groups in negotiati ons with political leaders and participated in debates with politicians and pundits, which likely made their language use more politician like than student like. Youth language vernacular often includes the use of less formal lexical items, such as slang, or modification of existing lexical items, effectively making them seem less formal, such as borrowing, verlanization, and truncation. Since youth language is generally considered to be a less formal vernacular, in some cases to the point of stigmatization its combination with political discourse, which usually takes on a more formal register, promises to be a fascinating topic of study. Whereas youth language has largely been studied in a vernacular context (see Gadet 2007, Fagyal 2010), observations of y outh speakers in a political discourse context will show another register of youth language that has not yet been well documented. Examining the younger speakers vary their s peech stylistically from their usual less formal vernacular
17 to match the more formal register expected from a political figure, by using a different register than they would use in conversation with family or friends. riation discusses variation based on context and interlocutor. According to Gadet, speakers of every language change various aspects of their speech based on the person to whom they are speaking. In French, there are four levels of language, which are, fro m least to most formal: populaire, familier, standard and soutenu It is expected that observations made in this study will show one of the higher levels of language, due to the fact that the speakers are being filmed and participating in public political discourse.
18 CHAPTER 2 LITERATURE REVIEW Age in Sociolinguistics According to Gadet (2007), young speakers of any given language in any given culture set themselves apart from adults through increased usage of vernacular language and unique codes, known in French as verlan (Gadet 120). Gadet asserts that youth language has only recently come to be considered a sociolect by sociolinguists, primarily due to a prolonging of adolescence due to economic dependence, unemployment or difficulty finding work, or, on the other hand, increase in young presented by the fact that youth language is simultaneously stigmatized and considered a source of linguistic change, as it is instable and in con stant evolution (Gadet 121). The traits of youth language that Gadet discusses at the lexical level include borrowings (from Arabic, English, and African languages), truncation of words (e.g. ouette for cachoute ), reduplication (e.g. zonzon for prison ), m etaphor or metonymy, and verlan (rearrangement of words, e.g. cfran for franais ) (Gadet 124). As in most vernacular forms of languages, especially oral forms, the usage of these traits is highly variable and the language is subject to combinations of the se variations, such as the verlanization of words borrowed from Arabic, followed by an English suffix (Gadet 124). spoken in the Parisian suburbs and the effects of contact wit h immigrant languages, she also makes some general comments on the study of youth language. In the 1980s, youth language was considered a style reserved for adolescents under 20; there was a o its members and had
19 to either be explained to or guarded as a secret from their parents (Fagyal 28). During these years, the youth language was viewed by the press as a hip code, much like that of a secret society, that represented a new, better version of French whose use was to be restricted to only young people and forbidden to their parents (Fagyal 28 29). In the mid 1990s, the primary characteristic of youth language changed: instead of being a variety mainly used by speakers of a certain age group, what was considered youth language became a vernacular mainly used by speakers in the suburbs (Fagyal 29). marginalized and more secretive, which caused objections from spe akers of the variety, who complained that they were not understood due to the obscurity of their code (Fagyal 29). By the end of the 21 st century, attitudes toward suburban French youth language had become rather negative; what had once been considered an elitist secret franais mutil of the language and even of the nation (Fagyal 32;34). 1985 study on the p ronunciation of the /r/ in Montreal French. Until the 1950s, an apical /r/ was the traditional norm of the area, but Sankoff recordings of native Montreal French speakers, when compared with interviews from the 1940s, showed an increase in the tendency to favor the uvular /r/ (Thibault 23). An examination of these corpora showed that the adolescents of the 1940s began the trend was shown i n the 1971 corpus, suggesting to the researchers that these changes were passed down via the first source of language acquisition: the family (Thibault 23). The
20 if they d id not instigate this linguistic change, at the very least played a key role in the diffusion of the uvular /r/, especially by passing it to their children (Thibault 23). inter generational differences. If a characteristic of youth language is carried into adulthood, it is most likely a linguistic change; however, if this trait fades as the individual becomes an adult, the variation is more likely to be an inter generationa l difference that is only seen when the individual is younger (Thibault 24). language in the suburbs of Paris, which suggest a marginalization of youth language, survey of studies seems to point to variation in youth language as a significant source of linguistic change in certain situations, such as those observed with the pronunciation of /r/ in Montreal in Sankoff observations on yo uth language present two opposing sides of linguistic attitudes towards adolescent vernacular, one being positive and the other negative. These conflicting views on youth language may be one reason that, according to Gadet (2007), linguists have recently b ecome interested in studying it as its own sociolect. As previously discussed, Gadet presents other various reasons for the interest in youth language, including the influence of adolescents on consumerism; however, the linguistic variation and the attitud also demonstrate compelling reasons for linguistic study of youth language, both in variation and in attitudes.
21 Discourse Markers in Pragmatics and Sociolinguistics Butler and King (2008) refer to Sch this article, they discuss the discourse marker mais dame which first appeared in France in the 19 th century but whos e usage has diminished except for in one variety of French spoken in Newfoundland on the Port au Port peninsula (Butler and King 63). They study the distribution and discourse functions of mais dame both in early literary usage and contemporary vernacula r usage (Butler and King 63). After describing the history of this discourse marker, Butler and King examine its usage in 19 th century literature, using the Grand Robert, and in contemporary vernacular in Newfoundland, nse Canards. They found that varieties of French in which this discourse marker is not attested use other items for the same discursive effects, such as bien oui, et pourtant, en fait etc. (Butler and King 80). Both the literary sources and the contem porary oral data showed mais dame reinforcing a stance, denying a challenge to the stance, and returning to a prior concern (Butler and King 81). They also concluded that in Newfoundland French, mais and mais dame are both discrete discour se markers with similar but separate functions that cannot be used interchangeably (Butler and King 81). Mais dame is an archaic form that seems to be restricted to Newfoundland French and whose usage serves to increase discourse coherence (Butler and Kin g 81). Sankoff, Thibault, Nagy, Blondeau, Fonollosa, and Gagnon (1997) studied spontaneous speech production in young Anglophones living in Montreal in interviews conducted in French. They were particularly interested in discourse markers due to the fact that they are not taught in school, but are an indication of whether the speaker has
22 been socially integrated into the community of the L2; only L2 speakers with significant contact with native speakers will master the use of discourse markers (Sankoff et al. 193). Their study examined the rate of usage and the choice of markers, as well as the social factors that may have influenced these choices (Sankoff et al. 193 4). In this study, discourse markers are defined as having the following characteristics: 1) do not enter into the construction syntactically with other elements of the sentence; 2) are not crucial to the propositional meaning of the sentence; 3) are subject to semantic bleaching compared to their source forms; 4) undergo greater phonological reduction than their source forms; and 5) are articulated as part of the smoothly flowing speech production (Sankoff et al. 195, 197). The discourse markers found in this corpus were categorized into three types, dependent on their relationship to English markers: French markers with an English equivalent ( comme tu sais ), French markers without an English equivalent ( bon l ), and those without a one to one semantic equivalent in English, but with a common function ( fait que ) (Sankoff et al. 198 9). Comm e is English equivalent, like ; this usage has been attested in Quebec French from over 60 years before the publication of this article (Sankoff et al. 205). The most freq uent use of comme however, was that of a desemanticized punctor, similar to like in English (Sankoff et al. 205). The results of this study showed that the speakers used discourse markers about three times more frequently in their L1 than in their L2 and that the least fluent speakers used them the least (Sankoff et al. 213). Those speakers with early exposure to the language outside of school were more like to use markers such as l with no English equivalent (Sankoff et al. 214). In general, speakers who were exposed
23 to the L2 in early life in the form of social interaction with native speakers were most likely to use discourse markers competently and, therefore, considered more fluent (Sankoff et al. 214). Gold and Tremblay (2006) assert that eh and hein are very similar in their discourse particle functions, but diverge in their function as identity markers; eh is part of the Canadian English dialect and the Canadian national identity, whereas hein has no identity marker properties for French Canadia ns (Gold and Tremblay 237). In their study, they found that hein in French is reported as more frequently used than eh in English and that francophone attitudes toward hein are more positive than Anglophone attitudes toward eh (Gold and Tremblay 248). Th ey studied the use of these in ten contexts: opinion, fact, command, exclamation, question, request for repetition, fixed expression, insult, accusation, and narration (Gold and Tremblay 249). The study was based on surveys completed by French speakers (U niversit Laval) or English speakers (University of Toronto) who were born in Canada and under the age of 30; the survey gave sentences and asked if the respondent had heard it, if they use it themselves, and their attitude toward the usage (Gold and Tremb lay 248). Beaulieu et al. (2007) studied the use of comme in Quebec French as a marker of comparison and asserted that the approximation and exemplification usages of comme are not unique to Quebec French, but are attested in other languages, notably Engli sh and Chinese (Beaulieu et al. 27). They note that these new utilizations of comme represent a broadening of its original meaning to include a distancing quality (Beaulieu et al. 27). The corpus that they used includes literary examples, some oral speec h samples from BDTS (Banque de donnes textuelles de Sherbrooke), and some
24 fabricated examples (Beaulieu et al. 27). The authors suspect that the non standard use of comme similar to that of like in English, developed as a result of the influence of Engl ish on the language (Beaulieu et al. 27). The article also discusses the metadiscursive function of comme ; metalanguage meaning a part of the sentence that has no semantic bearing on what is being said, but rather explains something about the usage of the code or language in the sentence (Beaulieu et al. 33). It is important to note that the meaning of comme (exemplification, approximation, etc.) has no effect on the use of comme in a metadiscursive context (Beaulieu et al. 34). The article also addresse s the discourse marker function of comme using the examples of comme on dit and both of which add nothing, grammatically or semantically, to the sentence in which they take part (Beaulieu et al. 36). They note that Susanne Fleischman mentions a similarity between genre in hexagonal French, comme in Quebec French, and like in English; genre and comme have similar usages to like in English, including approximation, hedging, and indirect relating of discourse (Beaulieu et al. 37). The au thors conclude that these new usages of comme in Quebec French discourse show a semantic shift of the term from the standard comparison meaning to multiple qualities that we can group under the general category of distancing (Beaulieu et al. 40). This sem antic shift has been accompanied by a syntactic shift, which has transformed comme into a discourse marker and given it an additional metadiscursive function (Beaulieu et al. 40). Forget (1989) studies the use of the particle l in Quebec French in its var ious usages, including as a discourse marker. She discusses how l has undergone a shift in function from deictic, an element of the sentence whose referential quality is
25 dependent upon the context of the statement, to discursive, where its primary purpos e is to establish or maintain a connection with the listener (Forget 57 9). As proof of the non systematic nature of l Forget cites the fact that it can be inserted at almost any point in the statement (Forget 60). Forget analyzes l in terms of its pr imary usages: identification, detachment from the subject, and reinforcement of the action (Forget 63 6). She concludes that the particular usage of l in Quebec French shows that it is necessary to overlook the assumption of lexical meaninglessness in or der to examine the semantic and pragmatic qualities that its use in discourse demonstrates; it is therefore important to recognize the connection between the discourse markers and the situation in which they are used (Forget 80). Heisler (1996) examined t he use of OK as a discourse marker in Montreal French using the Sankoff/Cedergren corpus of 1971 and the 1984 Montreal corpus, both corpora of sociolinguistic interviews in Montreal French (Heisler 293). The study describes three discourse functions that O K performs in the corpus agreement marker, interactive marker, and structural marker which all address interactional problems linked to the discourse produced (Heisler 293 294). Heisler defines the agreement marker function as marking an agreement with the preceding discourse and notes that, with this utterance, the speaker indicates that the exchange can be closed with regard to the preceding discourse and thus opens a new exchange, continuing the linear progression of the conversation (Heisler 295). A n interactive marker is defined as establishing, maintaining, prolonging, or breaking the punctual relationship between the is to link the interlocutor to the discour se of the speaker, favoring the communication
26 between the speakers and the linear progression of the conversation (Heisler 296). This function is also sub divided into four sub functions: attention getter, back channel signal, command softener, and progres sion check (Heisler 296). As a structural marker, Heisler notes that OK marks one of two things: the onset of a structural problem in the discourse or the return to normal from the structural problem, which is determined by the level of discourse at which it occurs: the act, the move, or the exchange (Heisler 298). Heisler quotes Sinclair and Coulthard (1992) on the definition of these terms: factor and determined that men utter OK in all three functions, more frequently than women (Heisler 302). Heisler offers the possible explanation that, according to Chambers (1995), women are more sensit ive to the social significance of linguistic variables and OK is often thought of as an Americanism, not a standard French form (Heisler 303). He also suggests that women may experience fewer of the problems that the usage of OK addresses (Heisler 303 304) The study then analyzed social class as a factor, demonstrating that the middle class used OK in all three functions more frequently than the upper and working class (Heisler 304). The upper and middle class were close to equal in frequency for two of th e functions, but the working class used the third function, the structural marker, much more frequently than the upper class (Heisler 304). Heisler found, however, that age was the most significant sociolinguistic factor in the frequency of OK usage as a d iscourse marker (Heisler 307). Younger speakers (aged 15 33) used OK as a discourse marker the most frequently, with 67% of tokens coming from this age group (Heisler 307). Heisler concludes that the influence of
27 English, from both the rest of Canada and t he United States, in the form of journalism, the media, music, movies, etc. is the most obvious explanation for this change in the usage of OK ; however, examining the use of OK in other languages, researchers may be able to more firmly conclude whether con tact with English is the real cause of the increase in frequency of this usage of OK as a discourse marker (Heisler 310). analyzing discourse markers: instead of isolating one m arker and discussing its various functions, Tagliamonte and Hudson discuss one function of discourse markers found in their corpus and how certain markers found in the corpus fit into this function. The article examines the use of quotative constructions, which introduce reported speech, in the speech of British and Canadian youth (Tagliamonte and Hudson 147). They used two corpora, one collected in York, England in 1996 and the other collected in Ottawa, Canada in 1995, both of which consisted of recorded narratives of personal experiences by university students ranging in age from 18 to 28 (Tagliamonte and Hudson 154). Most of the quotatives found in their analysis of the corpora used one of four quotative constructions: say go think and be like (Taglia monte and Hudson 155). Their multivariate analysis of these data attempted to determine how certain factors, both linguistic (grammatical person and content of the quote) and social (sex of the speaker), influenced which quotative construction was used in each token (Tagliamonte and distance, the usage of be like by British and Canadian youth is following the same say that it has come to be used in a similar fashion to other quotatives, but that the choice of quotative also varies
28 depending on individual narrative styles (Tagliamonte and Hudson 165; 168). This study is an example of a different theoretical approach than that used in studies such as speech, and discussed the different forms used to achieve this function and which Alt hough not specifically on the subject of discourse markers, Poplack and induced grammatical change fail to prove either that the change has occurred and/or that it is caused by contact and not internal evolution of the language (Poplack and Levey 391). They explain that contact based explanations of change are problematic because alternate explanations that are equally likely may be present and because it is often difficult to de termine whether the change is community wide or just idiosyncratic (Poplack and Levey 394). Variability, for example, can be confounded with change because it is a necessary condition of change, although it does not necessarily indicate it (Poplack and Lev ey 394). Poplack and Levey describe the following criteria for contact induced change: 1) absent in a pre contact or non contact variety or, if present, not conditioned in the same way as in the source and 2) can also be shown to parallel, in a non trivial way, the behavior of a counterpart construction in the source language (Poplack and Levey 398). They also specify that evidence of these criteria can only be obtained through systematic quantitative comparisons of the construction in question with a pre c ontact or non contact variety (Poplack and Levey 398). They present the an example, concluding that, although normally attributed to contact, this phenomenon
29 is actually u nlikely to be a contact induced change, due to the fact that the conditions leading to preposition stranding differ from those observed in English (Poplack and Levey 405). Their conclusion states that contact not possi and that it is especially important, given the recent advances in empirical linguistics, that we use these quantitative means to support claims of contact induced change, rather than r elying on anecdotal observations (Poplack and Levey 412). These previous studies examined specific discourse markers in Canadian varities of French, especially Quebec French, including their frequency and their various functions within the discourse. I wil l refer back to these observations when discussing the functions of the discourse markers that are being analyzed in the present study. As for criteria on determining which items are discourse markers, I will use the following working definition: a discour se marker is an item in an utterance that does not add to nor subtract from the meaning or grammaticality of the utterance. The meaning referred to in this definition refers to lexical meaning; that is, the discourse markers examined in this study may have a pragmatic function, but, in most cases, have not retained their lexical meaning. This definition does not include some of the criteria discussed in previous studies (such as semantic bleaching, discussed in Sankoff et al. 1997) because a broader working definition of the term will allow for a greater amount of data to be analyzed. The last article discussed in this review (Poplack and Levey 2010) described an empirical methodology for determining whether a grammatical change could be attributed to a lan guage contact situation. Although it seems that certain discourse
30 markers in Quebec French may be related to contact with English, due to seemingly analogous structures in English, we must be careful not to assume that this is the case without careful empi rical analysis of the evidence. In order for a construction to be considered a contact induced change, one must have access to comparable data in a pre contact or non contact variety of that language. Public and Political Discourse in Sociolinguistics Mart el (2008) studied the argumentation strategies, and effectiveness thereof, of politicians in televised political debates during the Canadian federal election campaign participants of different ages, genders, and political affiliations and asked them to 2). Martel discusses three aspects of the way in which the politicians conduct themselves during important media appearances: their media communication style, which depends on the type of media on which they are appearing; their professional identity, or p ublic persona; and their personal identity (Martel 4 5). According to Martel, these three of the Quebec provincial elections of 2003, the discourse analysis showed that only Jean Charest, the leader of the party that opposed the incumbent, conducted himself in keeping with his social, or political, identity by debating his opponents in the man ner expected given the interactional context; the other party leaders sacrificed their professional identity and the combative nature of a debate in order to preserve a more sympathetic image of their personal identity (Martel 5).
31 ses on argumentation used in online political discussions on Internet forums in English and French. Her corpus consisted of posts by both native and non native speakers on forums hosted by five periodicals ( The Financial Times, The Guardian, Le Monde, Le F igaro and Le Nouvel Observateur ) on three political topics related to current affairs: European Union politics, Middle Eastern politics, and globalization (Lewis 1802). Lewis notes a surprising heterogeneity of the messages in length, language style, stru cture, and topic (Lewis 1808). The main components of the message structure that she discussed were: message openings, which often included discourse markers; and position statements, which can take the form of question and answer, concession and counterar gument, or claim and evidence, among others (Lewis 1808 1809). She also discusses what she refers to as concessive structures as a way of linking the message opening with the position statement: the idered the concession aspect, but follows it with a contradicting idea (Lewis 1811). The two markers that she found to fulfill this concession function in her corpus were of course for English and bien sr for French (Lewis 1812 1813). Conclusion This Ch apter has given an overview of previous studies on age as a sociolinguistic factor and discourse markers as studied in pragmatics and sociolinguistics, as well as studies on political discourse in sociolinguistics. These studies will later be referred to d uring the discussion of the forms and functions of discourse markers examined in the present study. Chapter 3 will discuss the design of
32 the study presented, including the methodological framework, the corpus, and biographical background on the speakers ex amined in the study.
33 CHAPTER 3 STUDY DESIGN The Methodological Framework Two possible theoretical frameworks for analyzing discourse markers have been discussed: one analyzes the different possible functions of one or more discourse markers (as the in stu dies of Heisler 1996 and Sankoff et al. 1997, among others) while the other examines one function of discourse markers and which markers fall into that category (as in the previously discussed study of Tagliamonte and Hudson in 1999). The former is more co mmon in linguistic studies of discourse markers, perhaps due to the possibility of these functions overlapping. The present study will use the second framework, analyzing each function and the markers used to fulfill those functions, specifically using the OK which was adapted marker functions that he examines in his study; however, the sub functions identified in our data will be different from those that Heisler used in his description of OK These sub functions will be discussed in further detail in Chapter 4. The Data The study used a corpus of spoken Quebec French gathered by the author from interviews involving one or more of the three young spokespeople for the main student groups: CLASSE, FECQ, and FEUQ. The corpus was built by the author using videos gathered from the internet in February and March of 2013. The links for the sources of the corpus can be found in Appendix A. This corpus enabled an analysis of young people who are in particularly leadership oriented and often political roles. Examining their speech will therefore provide insight into the discourse of young speakers in a
34 political context. Biographical backgroun d information on the spokespeople whose discourse was examined in this study Martine Desjardins (FEUQ), Lo Bureau Blouin (FECQ), and Gabriel Nadeau Dubois (CLASSE) will now be discussed. The Spokespeople Lo Bureau Blouin was born in Montreal, Quebec in December 1991 (Assemble Nationale de Qubec 2013). His father is a director for an exposition center and his mother works for an artistic group (Gervais 2012B). Although born in Montreal, he grew up in Saint Hyacinthe and also attended cgep there (L essard 2012). He earned a in social sciences and administration at Cgep de Saint Hyacinthe in 2010 (ANQ 2013). Bureau Montral, but began a political career before finishing the de gree (Lessard 2012). After serving as the president of the Fdration tudiante collgiale du Qubec (FECQ) from 2010 to 2012, he was elected as member of the Assemble Nationale for Laval des Rapides during the general election of September 2012 (ANQ 2013 ). He is affiliated with the Parti qubcois (ANQ 2013). Gabriel Nadeau Dubois was born in Montreal, Quebec in May 1990 (Gervais 2012A). His parents are from Thetford Mines, a town in the east of Quebec, and they met while they were both participating in the International Movement of Catholic Students (Gervais 2012B ). He enrolled at the Universit de Qubec Montral (UQAM) in the fall of 2009 to begin working on a baccalaureate in History (Gervais 2012A). He was elected spokesperson of the Association p our une solidarit syndicale tudiante (ASS), a group of which he had been a member since fall 2007, in April 2010 (Gervais 2012A). In fall 2011, he became the co spokesperson of the Coalition (CLASSE) (Gervais
35 2012A). CLASSE was a temporary coalition of the ASS formed in December 2011 whose specific goal was to organize and carry out protests in the spring of 2013 (ASS). Marine Desjardins was 30 years old during the student strikes and was b orn in Montreal (Gervais 2012B). Her father works as an aeronautical technician and her mother is a retired hospital consultant (Gervais 2012B). She was elected president of the Fdration tudiante universitaire du Qubec (FEUQ) in 2011 and was re elected to continue as president in 2012 (FEUQ). Before becoming president of the FEUQ, she was also involved the Association des tudiantes et tudiants en sciences de representative in 2009 before becoming president of the ADEESE in 2010 (Gervais 2012B). Desjardins is a PhD student in educational sciences at UQAM and would like to teach at the university level, although she has not ruled out a career in politics (Gervais 2012B). The Data Collection This study was conducted empirically, using data collected from publicly available Internet sources. The tokens used were collected from seventeen interviews and press conferences with student group spokespeople Gabriel Nadeau Dubois, Marti ne Desjardins, and Lo Bureau Blouin. The videos were found using the Google search engine. All videos came from You Tube or TVA Nouvelles and their durations ranged from about six to thirteen minutes. Some interviews were split into multiple videos due to file size limitations imposed by the hosting site. Only occurrences uttered by the three spokespeople were transcribed, since the study focuses on their usage of the markers; therefore, markers used by journalists or other interviewees were not
36 counted. O ccurrences of the selected discourse markers were extracted from each interview, transcribed in context into a Microsoft Excel spreadsheet, and coded for discourse marker form, speaker, discourse marker function, and, where applicable, sub function. Gene rally, studies on discourse marker frequency discuss the number of each discourse marker used per 10,000 words (see Beeching 2007, among others). Since the videos examined in this corpus were not completely transcribed and only the tokens of discourse mark ers were transcribed, the number of total words is not available; therefore, the frequency of discourse markers in the corpus per 10,000 words cannot be determined. Since the data were gathered in this way, with only the discourse marker tokens being extra cted from video recordings and the entirety of the videos not being transcribed, this study will not focus on frequency of discourse markers, but rather on the discourse marker forms and functions attested in the corpus. Table 3 1, below, shows the discou rse markers collected from the corpus by form (rows) and by speaker (columns). Totals for each form attested in the corpus are given as well as totals for each speaker. The grand total of discourse marker tokens extracted from the corpus was 165. Conclusio n The information presented in Chapter 3 has outlined the methodology for the study. The study was performed empirically, with the tokens collected from seventeen videos of television interviews or conference lectures involving the three student group spok espeople. The data were classified based on speaker, form, function, and sub function, where applicable. The discourse marker functions and sub functions used in present st udy; however, the methodological framework differs from that of Heisler, who
37 chose one discourse marker form and studied its various functions. Instead, this study analyzes each function attested in the corpus and the discourse marker forms that do or do n ot fulfill them. The discussion of the results will also focus on discourse marker form and function instead of on frequency. Chapters 5 and 6 will analyze the data collected from two different perspectives. Chapter 5 will analyze the data in a quantitativ e manner, focusing on the discourse marker functions and the forms that are attested for each in the corpus. Chapter 6 will discuss the data more qualitatively and focus on variation between the three speakers and what impact their biographical backgrounds discussed earlier in Chapter 3, may have on their speech. Table 3 1. Discourse marker forms by speaker. Marker GND LBB MD Total Alors 4 0 0 4 (2%) Ben 2 9 6 17 (10%) Bon 2 2 0 4 (2%) Donc 22 13 26 61 (37%) En fait 5 1 21 27 (16%) L 12 8 9 29 (18 %) OK 1 0 0 1 (<1%) Puis 13 7 2 22 (13%) Total 61 40 64 165
38 CHAPTER 4 FUNCTIONS OF THE DISCOURSE MARKERS Introduction: Form and Function Chapter 4 will focus mainly on analyzing the functions of the discourse markers collected from the corpus, inclu ding which discourse marker forms do and do not fulfill each one. Most of the previous studies on discourse markers discussed in Chapter 2 examined both form and function of the discourse markers that they studied. In Butler mais dame in Newfoundland French, the main functions of the marker were reinforcing a stance, denying a challenge to that stance, and returning to a prior concern. Beaulieu et al. (2007) focused on comme as a marker of comparison but also discussed its approxi mation and exemplification uses, while Forget (1989) studied l and found its main function to be establishing and maintaining a connection analytical approach: the authors chose one function (reported speech) and examined various structures that fulfilled this function. Regardless of how their analysis was structured, however, these previous studies on discourse markers discussed the relationship between form and function. Often, one marker is predominantly, although not completely, associated with one primary function, such as mais dame l (1989) study. In other cases, a function that is fulfilled by certain discourse marke rs cannot be fulfilled by others. For example, the marker OK (1993) study due to its inherent semantic properties. Therefore, although He isler used
39 marker he studied did not fulfill it. In the present study, examining the attested discourse markers according to both form and function will allow us to obs erve similar patterns where one or more forms are predominant in a function or one or more forms cannot (or does not, in this corpus) fulfill a certain function. Classification of Data Once the data were collected, the tokens were first coded according to function. describe the usage of OK in Quebec French: agreement marker, interactive marker, and structural marker. Most markers taken from the corpus fell into the interac tive marker category for function (162 of 165 total tokens), with only three tokens fulfilling the agreement marker function and no tokens fulfilling the structural marker function. It is not surprising that the interactive marker is dominant in this corpu s, due to the nature of n the context of the strikes. to describe the data; however, since the vast majority of tokens fell into the interactive marker category, sub functions of this function were id entified in order to further analyze the data. The sub functions of the interactive marker will be discussed below. Agreement Marker Heisler (1996) defines the agreement marker as indicating that the speaker agrees with a previous statement and is indicati ng a change in the topic or direction of the discussion, thus furthering the linear progression of the conversation (Heisler 295). Therefore, a token was coded as an agreement marker if it indicated that the speaker
40 was agreeing with the previous statement and moving the conversation in another direction or to another topic entirely, as in example (1). (1) ben news (LCN)) In example (1), the marker ben indicates that the speaker acknowledg es the previous point that the law will apply in this case and is moving the conversation along to a different, although related, point. The discourse markers found to fulfill the agreement marker function in this corpus were ben, bon, and OK Other marker s not attested in this function category such as en fait would not logically be able to fulfill this function. The term en fait for example, for which the English equivalent is in fact has a contradictory sense that would generally not allow it to be u sed in this context. Interestingly, the only instance of OK the marker studied by Heisler, found in the corpus fulfilled this function. This token is shown in example (2). (2) OK mais coute il fait des trs bons films (GND; 2/26/12; Tout le monde en par le ) In this utterance, the speaker is conceding the point made by the speaker to whom he is responding and moving the conversation along in a different direction. It is significant to note that this utterance was not a part of the interview about the stri kes, but rather a digression from the political discourse that comprises the majority of the corpus. This may explain why the agreement marker is almost absent from the corpus; due to the political and often heated nature of the debate, there is not much n eed for a discourse marker that acts to concede a point or agree with the previous statement. Interactive Marker The interactive marker establishes, maintains, prolongs, or breaks the punctual relationship between the interlocutors (Vincent 1993) and its main purpose is to link the
41 interlocutor to the discourse of the speaker and promote communication between speakers and the linear progression of the discourse (Heisler 296). Except for the three agreement marker tokens, all tokens were coded as interacti ve markers, as there were no instances of structural markers in the corpus. The sub functions used to classify the interactive marker tokens were progression marker (Heisler 1996), filler (similar to tradiction marker (both types of progression marker), and command softener (Heisler 1996). Progression Marker The progression marker sub function used in the present study was adapted ctive marker sub functions examined in his study. The progression marker tokens serve the purpose of indicating a transition between arguments that the speaker is asserting, as seen in example (3). (3) donc ; news interview; LCN) In example (2), the marker donc to the next one and to indicate to the listener that he is transitioning to another point. The progression marker is therefore similar to the agreement marker in marking the transition to the next topic, but differs from it due to a lack of agreement with the previous statement. In other words, the progression marker simply shows progression to the next statement and does not indicate agreement with the p revious one. The progression marker may also serve to prepare the listener for the next point that the speaker makes, as in examples (4) and (5). (4) euh donc hors de la question de finance publique (GND; 7/31/12; conference at the Center for Research on G lobalization)
42 (5) puis on voit aussi une transition de gouvernement (MD; 3/30/12; interview on LCN news) The markers donc and puis represent the majority of the tokens with the progression marker sub function, with 55 (63%) and 21 (24%) tokens (respectiv ely) fulfilling this sub function. Progression marker is also the most frequent sub function for both donc and puis : 90% of tokens of donc are progression markers and 95% of tokens of puis are progression markers. Other discourse markers that were found to function as progression markers in the corpus were alors (4 tokens), bon (1 token), and l (6 tokens). Again, the marker en fait would not be a logical marker for the progression marker category due to its contradictory semantic properties. The low number of instances of alors (fours tokens) in the corpus also coincides and Daveluy found an increased number of tokens of alors used by older speakers when compared with a lower frequency of alors in the same speakers at a younger age. Therefore, the low number of tokens of the marker found in this corpus suggests a similar decrease in the usage of alors It cannot be determined from this result, however, whether this marke r is decreasing in usage in the general population. It is also important to note that this marker was only uttered by one of the speakers studied and that his usage of the marker is quite possibly an effect of intra individual variation rather than a gener ational difference, as was seen in Thibault and Daveluy (1989). That the progression marker was the most widely used by all three speakers on discourse is largely political in nature; that is, the purposes of their discourse is argumentative rather than narrative. The progression marker is used to form a liaison
43 between the arguments that the speaker is making and ties into the main function of the interactive marker, which is t Filler or Hesitation Marker The filler function was adapted from the hesitation marker function examined in implicatio moment when it was uttered. Instead, these tokens indicate breaks in the progression of generally came at the beginning of a sentence or when the speaker first started answering a question. Some, however, were used to place emphasis on a particular part of the discourse, as shown in example (6). (6) ces entreprises de bnficier en fait e (MD; 10/30/12; lecture at SPEP conference) The marker en fait in example (6) is clearly not marking the progression of the discourse to the next statement, since it is inserted in the middle of an idea, but is instead adding emphasis to the word bnfi cier marker does not, however, always follow the word to which it adds emphasis; it can also precede the emphasized word, as seen in example (7). donc refinancer les universities (GND ; 7/31/12; conference lecture at the Center for Research on Globalization) In example (7), the marker donc precedes the word to which it is drawing attention: refinancer When the listener hears this marker interrupting the normal syntax of the utterance, his/her attention is drawn to what the speaker says next, thus reinforcing his argument.
44 Tokens of l ben and en fait comprise the majority of the discourse markers in the corpus that fulfill the filler function, with 22 (39%), 16 (28%), and 11 (19%) t okens respectively. Ben is used almost exclusively as a filler, with only one token fulfilling the agreement marker function. L also functions mainly as a filler, with 22 out of 27 tokens (81%) fulfilling this function. The marker OK would not logically b e fit into the filler category due to its semantic properties as an agreement marker. Puis is also unlikely to fulfill this function, as it generally marks a transition to the next point in an argumentative discourse or to the next event in a narrative dis course. Causation Marker subset of the progression marker used in his study. Like the progression marker, it functions to show a connection between the two statements but unlike the progression marker, it indicates a cause and effect relationship between the two statements or ideas, as in example (8). donc street with multiple journalists) In example (8), the marker do nc shows a direct relationship between the first statement, ambiguous causation marker; for ambiguou s cases where it was unclear whether there was a cause and effect relationship, the tokens were categorized as progression markers. It is also interesting to note that the only token of this marker shows similar syntactical properties to those seen in the filler sub function; the marker is inserted right before the relative pronoun que and therefore breaks up the syntactical progression of the sentence.
45 Contradiction Marker The contradiction marker, similar to progression and agreement markers, shows a con nection between two statements. Like the causation marker, it is a subset of the It is the opposite of the agreement marker, however, as it functions to contradict a previous statement that was either made by another speaker or repeated by the present speaker for the purposes of refutation. In question to the speaker or a statement from an opposing side repeated by the jo urnalist (9) en fait il faut comprendre que le problme l (MD; 3/30/12; news interview; LCN network) In example (9), the marker en fait the questio n in order to address a misconception on the part of the journalist or the audience and to indicate that she is correcting that misconception. The discourse marker en fait dominates the contradiction marker sub function, with all 16 tokens being occurrence s of this marker; this is not surprising since the phrase is the equivalent of en fait is distributed between the filler (11 tokens or 41%) and contradiction (16 tokens or 59%) functions. As with the causation marker function, ambiguous cases where it was unclear whether the relationship between two statements was a contradictory one, the token was classified in the progression marker category. The contradiction marker function is similar to that of the progression marker in that it links two statements either made by or responded to by the speaker; however, these two functions differ in that the progression marker indicates a positive link
46 between the two statements, meaning that he/she believes them both to be true, whereas the contradiction marker indicates a negative link between the two. In other words, the speaker believes only one of the statements linked together to be true (usually the second one, or the counterargument). Command Softener OK as a sub function of the interactive marker. The command softener is added to a command as a means of dampening the harshness of the command, as in example (7). (10) puis un l Penelope McQuade (talk show); Radio Canada) Example (10), which is the only instance of this sub function found in the corpus, is used as part of a quotation, so it is important to note that this exampl e does not represent a marker that the speaker used in a spontaneous manner. It is nevertheless significant to note the amount of markers used in this utterance, as there is also a filler marker ( l ) right before the quotation and a progression marker ( pui s ) at the beginning of the statement. Also, it is possible that this marker is not a part of the quoted speech and was, rather, added by the speaker. We transcribed the utterance in this way because we believe the marker to be part of the quotation; howeve r, since the tokens were extracted from a corpus of oral speech, it is impossible to know for sure whether the bolded l in example (10) should be inside or outside the quotation. Discourse Marker Usage In Table 3 1, the raw numbers for discourse marker f unction according to speaker are shown, with percentages in parentheses. Table 5 1 shows the raw
47 numbers for each discourse marker form according to the speaker, with percentages in parentheses. A detailed analysis of the inter individual variation seen i n these tables will be provided in Chapter 5. As seen in Table 5 1, Nadeau Dubois was the only speaker who used the marker alors and also used puis significantly more frequently than the others, although his most used marker was donc Bureau Blouin also u sed donc more than the other seven markers, but also uttered bon relatively frequently; in fact, he uttered this marker more donc and en fait She uttered these markers more than the other two speakers and was responsible for 21 of the 27 tokens (78%) of en fait Desjardins also demonstrated fairly frequent use of ben and l but attested the lowest frequency of puis of the three speakers. Table 4 1, below, shows the raw numbers for discourse marker form relative frequency according to function category, with the percentages in parentheses. Donc is mostly used as a progression marker, with a significant number of tokens of puis also fulfilling this function. Puis is almos t exclusively used as a progression marker, with only one token fulfilling the filler function. The marker ben is almost exclusively used as a filler, as is l ; these two markers are the most frequent in the corpus for the filler function, with en fait hav ing a significant number of tokens in this category as well. The use of en fait is split between the filler and contradiction marker functions, with 11 and 17 tokens in each category, respectively. All four tokens of alors in the corpus are used as progres sion markers, while bon agreement marker (1 token), and progression marker (1 token) functions. The three
48 agreement marker tokens are comprised of one token each of ben bon and OK This token of OK i s the only instance of this marker attested in the corpus. Discourse Marker Forms Absent from the Corpus : The markers comme, genre, and (a) fait que, which were previously discussed in Chapter 2, were not attested at all in this corpus, despite their pres ence in previous studies on discourse markers in Quebec French (see Beaulieu et al. 2007 and Sankoff et al. 1997). One possible explanation for the absence of these markers is that they are more frequent in vernacular speech or lower registers, whereas the corpus for this study was taken from journalistic interviews, in which one would expect a more formal or higher register. These markers, especially comme and genre, also tend to be used more in narrative discourse or reported speech (as in the spontaneous speech corpus studied by Sankoff et al. 1997) than in argumentative or political discourse, which was the focus of the present study. Also, since this corpus was relatively small with only 165 tokens in 17 sources, it is possible that these markers may ha ve been found with a larger and more varied number of corpus sources. Conclusion Chapter 4 discussed the results of the study in terms of the forms and functions of discourse markers attested in the corpus. The functions used most in the corpus were the pr ogression marker and filler functions, while the discourse marker forms used most were donc, en fait, l and puis The absence of the markers comme genre and (a) fait que was also discussed. The discussion in Chapter 5 will focus on individual variatio n between the speakers studied in the corpus.
49 Table 4 1. Discourse marker forms by function. Marker Agreement Filler Progress Causation Contradiction Soften Total Alors 0 0 4 0 0 0 4 Ben 1 16 0 0 0 0 17 Bon 1 2 1 0 0 0 4 Donc 0 5 55 1 0 0 61 En fait 0 11 0 0 16 0 27 L 0 22 6 0 0 1 29 OK 1 0 0 0 0 0 1 Puis 0 1 21 0 0 0 22 Total 3 (2%) 57 (35%) 87 (53%) 1 (<1%) 16 (10%) 1 (<1%) 165
50 CHAPTER 5 INTER INDIVIDUAL VARIATION Chapter Introduction In Chapter 4, we presented the data and discussed the res ults according to the discourse markers and their functions as attested in the corpus. The discussion in markers, according to both form and function. Due to the fact that t his study concentrates only on the speech of three specific speakers, general linguistic tendencies of certain groups according to social factors cannot be accurately assessed; however, differences between the individual speakers will be discussed. It is a lso important to note that, although the number of discourse markers used by each speaker relative to the other two will be mentioned, the discussion will focus mostly on the relative frequency of forms and functions used by each speaker. This is due to th e nature of the corpus used in this study; as the corpus was collected from online oral sources for which transcriptions were not available, frequency of the discourse markers per 10,000 words cannot be accurately assessed. The discussion will therefore co rather than on how many markers total were used by each speaker. Table 5 1 (below) shows the distribution of discourse marker functions by speaker. Speakers Gabriel Nadeau Du bois Of the 165 tokens of discourse markers in the corpus, Gabriel Nadeau Dubois uttered more than a third (37%) of them, which was approximately the same amount as Martine Desjardins; however, his usage does differ from hers in both marker form and
51 functi on. In terms of function, Nadeau varied. He utilized five out of the six functions studied at least once, but he used the progression marker function most (36 out of 61 total tokens). His other markers fulfille d the functions of agreement marker (2 tokens), filler (16 tokens), causation marker (one token), and contradiction marker (5 tokens). Clearly, Nadeau usage according to function is quite diverse, although the amount of tokens for each function used is not necessarily very high, except in the case of the progression marker. In terms of form, Nadeau Dubois used donc l and puis most frequently and, while he used alors only four times, he was the only one of the three speakers to us e that particular marker in the corpus. He also uttered the only instance of OK found in the corpus. Nadeau Dubois used the markers ben and en fait however, much less frequently than the other two speakers. It is also interesting to note that Nadeau Duboi s was the only one of the three speakers to use all eight forms of discourse marker attested in the corpus at least once. Overall, Gabriel Nadeau Dubois displayed the most diversity in both form and function of discourse markers of the three speakers exami ned in this study. Although some forms and functions were used more than others, his speech displays a significant variety of markers that differs from the usage of the other two speakers. Lo Bureau Blouin Lo Bureau Blouin displayed both the most restri cted usage of markers in terms of functions and the lowest amount of marker tokens of the three speakers. His speech accounts for less than a quarter (24%) of the total discourse marker tokens in the corpus and he used only three out of the six functions, one of which contained only one token, which was an instance of reported speech. The two discourse marker functions
52 that Bureau Blouin utilized in tokens that represented spontaneous production were progression marker and filler. The tokens were almost exa ctly evenly split between the two categories, with 18 tokens in the filler category and 21 tokens in the progression marker category. Bureau Blouin used the marker donc most frequently of the eight forms attested. He did not use the forms alors or OK at al l and used en fait only once in the corpus. Ben was his second most frequently used marker and he uttered over half the tokens of this marker (9 tokens out of 17). Overall, Bureau much less varied than the other two did use the marker ben more than the other two speakers and this difference was especially noticeable when compared to Nadeau Dubois, who used the marker only twice in the corpus. The differences in Bureau B different area (Saint Hyacinthe) than the other two speakers, who were both raised in Montreal. A comparison study with o ther speakers from this region could provide more insight into whether the tendencies observed in this study are typical of the Saint Hyacinthe region. Martine Desjardins As previously mentioned, Martine Desjardins used about as many discourse markers in t he corpus as Gabriel Nadeau marker usage, however, was more evenly distributed among the filler and progression marker categories (23 and 29 tokens, respectively) and she used the contradiction marker function over twice as frequently (11 tokens) as did Nadeau Dubois (only 5
53 tokens). The only other function she used in the corpus was one token of the agreement marker. Like the other two speakers, Desjardins used the marker donc most in the corpus, with 26 tokens of that marker; however, her other most used form, en fait was tokens of this marker in the corpus, whereas Nadeau Dubois used the form five times and Bureau Blouin us ed it only once. About half of the tokens of en fait used by Desjardins (11 tokens) were in the contradiction marker category, while the others were in the filler category. In contrast, Nadeau en fait were all in the contradiction marker category. Her increased usage of the form en fait correlates with the high frequency of the contradiction marker in her speech, since en fait was the only marker attested in the contradiction marker category. Desjardins also used ben l, and puis but to discourse marker usage was concentrated in two functions (filler and progression marker) and two forms ( donc and en fait ). partially because, although she is also a student, she is a graduate student and is about ten years older than Bureau Blouin and Nadeau Dubois. Although age grouping for what is considered youth language can vary based on the study 30 years old is generally no longer considered to be in the category for youth language, which usually focuses on adolescents to speakers in their early 20s.
54 Conclusion Chapter 5 has discussed observations on differences in discourse marker usage betw een the individual speakers whose speech was examined in this study. Gabriel Nadeau Dubois showed a more varied usage of both forms and functions, including the use of two discourse marker forms that were only uttered by him in the whole corpus. Lo Bureau Blouin, however, displayed a more restricted usage of discourse markers in terms of function, using only three of the six functions attested in the corpus. Chapter 6 will summarize the thesis as a whole and provide the general conclusions drawn from the o bservations made in this study. Limitations of the study and possibilities for future studies will also be discussed. Table 5 1. Discourse marker function by speaker. Speaker Agreement Filler Progress Causation Contradiction Soften Total GND 2 16 36 1 5 0 61 (37%) LBB 0 18 21 0 0 1 40 (24%) MD 1 23 29 0 11 0 64 (39%) Total 3 57 87 1 16 1 165
55 CHAPTER 6 DISCUSSION AND GENERAL CONCLUSIONS The primary goal of this study was to observe a higher register of discourse within youth language, which would b e expected when young speakers are involved in leadership roles in their communities and engage in political discourse with other members of the community. It was hypothesized that the young speakers whose speech was studied would use a more formal level o f discourse than is generally associated with youth language, as they are acting in more professional roles and want to be taken seriously. This study focused on the usage of discourse markers in order to make these observations because this was a simple w ay to empirically examine the Discourse markers are also unique in that they are not taught in schools or grammar s discourse is dependent on the marker usage will vary between individuals based on several different social factors. The corpus from this study was gathered from jo urnalistic interviews and conference lectures, all of which were in the context of public political discourse, featuring three relatively young speakers: Gabriel Nadeau Dubois (21), Lo Bureau Blouin (20), and Martine Desjardins (30). The discourse marker tokens were extracted sub function. The most prominent functions in the corpus were filler and progression marker, while agreement marker, causation marker, and command softener were only seen in a few tokens each. The most common forms were donc en fait l and puis with only a handful of tokens each of alors ben bon and OK As well as analyzing the
56 form and function of the discourse marker tokens, we also discusse d, in Chapter 5, the intra individual variation observed among the subjects. Nadeau Dubois seemed to have the most diverse usage of the different functions, using all but one of the six functions, while Bureau Blouin used only three out of the six function s, only two of which represented spontaneous speech, as the command softener token was reported while Bureau f discourse markers was also more evenly distributed between her two most used functions, filler and progression marker, than that of Nadeau Dubois and Bureau Blouin. It is interesting the note that, in the general election of September 2012, Bureau Blouin was elected as a member of the Assemble Nationale and has therefore begun a career in politics at a very young age. The main distinction made between youth language and political discourse in French in the previous studies discussed was the formality of the language; that is, youth language is considered a less formal vernacular form, whereas political discourse would be more formal and closer to standard French. In an informal context, which is often associated with youth language, we would also expect, in the case of discourse markers, a higher frequency of forms or constructions either borrowed from other languages or with an equivalent form in another language, such as comme or tu sais as discussed in Sankoff et al. (1997). OK as a form that is usual ly thought of as an Anglicism and not a standard French form (Heisler 1996), would be expected to have a high frequency as well among young speakers, as Heisler found in his study on the marker (Heisler 307). As previously discussed, there are no instances of comme genre
57 or ( a) fait que attested in this corpus and only one instance of the marker OK which was part of a dialogue that was unrelated to the strikes and in the context of a somewhat less formal interview on a talk show. As discussed in Chapte r 2, Poplack and Levey (2010) describe the need for concrete evidence of effects of language contact. Although Heisler (1996) described OK as an Anglicism and not a standard French discourse marker form and Sankoff et al. (1997) discussed the similarities of the usages of comme and genre in French to those of like in English, the single instance of OK found in this corpus is obviously not sufficient evidence to warrant a discussion of contact induced effects. Contact was therefore not discussed as a factor of discourse marker usage in this study. The absence or extremely low frequency (in the case of OK ) of these more informal discourse markers suggests that the speech of the three student spokespeople studied is more representative of a more formal politica l discourse rather than of less formal youth language. The tendencies displayed by these young speakers to use more traditional forms of French discourse markers suggest that, when faced with a situation where their discourse is more closely watched by the public and where they are playing an important political role, young speakers will make an effort to self monitor and use more standard forms of the language, in contrast with the borrowing, verlanization, truncation, slang, and other lexical modification s that are often associated with the vernacular of youth language. Limitations of the study and possible future directions : Although this study presents some pertinent observations on the subject of youth language discourse, its relatively small scope unfo rtunately makes it difficult to draw significant and definitive
58 could only make observations on the differences between the speakers rather than discuss larger generalized tendencies of youth speakers. Although the results do suggest a tendency to omit less formal discourse markers such as comme, genre, or (a) fait que, a larger sampling of young speakers would be needed to draw truly definitive conclusions from these obse speech in a more informal setting, such as a sociolinguistic interview or a similar recording of spontaneous speech, could provide results from a different register with which to compare the results of the pre sent study. Unfortunately, the data gathering for this study was limited to Internet sources and adding a corpus of spontaneous informal speech from the speakers was not possible. Future studies, however, could broaden the scope of the corpus by gathering more informal and spontaneous speech from the speakers or include the speech of more young speakers in political roles in order to provide a larger sampling of the population. Additionally, future studies on this subject could focus on one observation in p articular made in this corpus that was not discussed in the results, specifically that the tone of the interview and friendliness of the interviewer seemed to significantly affect the amount of markers that Bureau Blouin used. For example, in one of the vi deos in the corpus, taken from a show called Franchement Martineau the host, Richard Martineau, was very argumentative and combative towards Bureau Blouin while discussing the situation of the strikes. During this video, Bureau Blouin attested only two to kens of discourse markers, both the form donc and both in the progression marker category. Conversely, in a video of similar length from the talk show Pnlope McQuade the host
59 was more affable and Bureau Blouin used a wider variety of markers and functio ns. In this interview, he used the filler, progression marker, and command softener functions. This interview was also different because part of it focused on his personal life and asked him about his childhood; therefore, this part of the video demonstrat ed narrative rather than political discourse. Although potentially significant, this observation was not explicitly discussed in the results, since the trend was only seen in these two sources with this speaker and therefore cannot be compared with reactio ns of other speakers to these different tones or moods. Future studies could analyze more sources in which the toward the interviewee has an effect on the discourse marker usage of the speaker. Despite these limitations, the current study has nevertheless provided some important insight into the usage of discourse markers by young speakers in the specific context of political discourse. The prominence of the interacti ve marker function, which accounted for all but two tokens in the corpus, confirms the importance of the link between the speaker and the listener(s) in the context of political discourse. The low number of instances of the marker alors also coincides with the tendency noted in of alors increases with the age of the speaker (Thibault and Daveluy 44). Finally, the absence of the markers comme genre and (a) fait que whi ch tend to be used in less formal contexts, suggests that in the context of these interviews, the young speakers used a more formal register. The observations made based on the results of this study therefore show a more formal register that is used by you ng speakers in leadership roles in the context of public political discourse.
60 APPENDIX CORPUS SOURCES 1. http://www.youtube.com/watch?v=qIVTZmiZTXQ 2. http://www.youtube.com/watch?v=_aufCDG5oE4 3. http://www.youtube.com/watch?v=oYkI2qF1tc8 4. http://www.youtube.com/watch?v=uIFQt M Jb5w 5. http://www.youtube.com/watch?v=uAoxKP_f7NU 6. http://www.youtube.com/watch?v= EeSK3ZxgL8 7. http://www.youtube.com/watch?v=Gzv19kywYDQ&feature=youtu.be 8. http://www.youtube.com/watch?v=efeAKPnEc4Q 9. http://tvanouvelles.ca/video/1537767921001 10. http://www.youtube.com/watch?v=bCIkllodmEc 11. http://www.youtube.com/watch?v=VDCf JrBIRsY 12. http://www.youtube.com/watch?v=i1UdIpGB6pQ 13. http://tvanouvelles.ca/video/1557193644001 14. http://tvanouvelles.ca/video/1653278759001 15. http://www.youtube.com/watch?v=YZlY0IWjby4 16. http://www.youtube.com /watch?v=6yQ_DjCMOMs 17. http://www.youtube.com/watch?v=abrRw8imm o
61 LIST OF REFERENCES Appel tous. Dubois lance un appel tous. Online video clip. YouTube. YouTube, 2 Nov. 2012 Web. 11 Mar. 2013. Assemble Nationale de Qubec. 2013. Biography of Lo Bureau Blouin. http://www.assnat.qc.ca/en/deputes/bureau blouin leo 12175/biographie.html (Accessed 4/13/2013). pour une solidarit syndicale tudiant (ASS). 2013. Historique. http://www.asse solidarite.qc.ca/asse/historique/ (Accessed 4/13/2013). Beaulieu Mas son, Anne, Melanie Charpentier, Lisannae Lanciault, and Xiaoyan Liu. 2007. Comme en franais qubcois. Communication, Lettres Et Sciences Du Langage 1(1): 27 41. Beeching, Kate. 2007. La co variance des marqueurs discursifs dire, enfin, hein quand meme, quoi et si vous voulez ? Langue franaise 154 : 78 93. bl1ndedbyfear. Tout le monde en parle Gabriel Nadeau Dubois et Arielle Grenier partie 1 YouTube YouTube, 26 Feb. 2012. Web. 11 Mar. 2013. bl1ndedbyfear. Tout le monde en parle Gabriel Nadeau Dubois et Arielle Grenier partie 2 YouTube YouTube, 26 Feb. 2012. Web. 11 Mar. 2013. Butler, Gary R. and Ruth King. 2008. The French Discourse Marker Mais Dame : Past and Present Functions. Canadian Journal of Linguistics 53 (1): 63 82. Blouin sur la loi et la crise tudiante (LCN 24 05 2012 YouTube YouTube, 24 May 2012. Web. 11 Mar. 2013. Du bois vs Martin Pouliot video clip. YouTube YouTube, 1 Apr. 2011. Web. 11 Mar. 2013. Blouin Online video clip. YouTube YouTube, 6 Jun. 2012. Web. 11 Mar. 2013. tineau et Lo Bureau Blouin: dmission des leaders Online video clip. YouTube YouTube, 15 May 2012. Web. 11 Mar 2013. Fagyal, Zsuzsanna. 2010. Accents de banlieue: Aspects prosodiques du franais igration
62 Fdration tudiante universitaire du Qubec. 2013. http://feuq.qc.ca/a propos/equipes anterieures/ (Accessed 4/13/2013). Forget, Danielle. 1989. L: un marque ur de pertinence discursive. Revue qubcoise de la linguistique 18 (1): 57 82. Gadet, Franoise. 2007. La variation sociale en franais Editions Ophrys. Paris. Online video clip. Y ouTube YouTube, 3 May 2012. Web. 11 Mar. 2013. Gervais, Lisa Marie. March 9, 2012 (A). Point chaud Un printemps tudiant Le Devoir. http://www.ledevoir.c om/societe/education/345404/point chaud un printemps etudiant (Accessed 4/13/2013). Gervais, Lisa Marie. April 16, 2012 (B). Qui sont les leaders du mouvement tudiant? http://www.lactualite.com/societe/qui sont les leaders du mouvement etudiant/ (Accessed 4/13/2013). GlobalResearchTV. Dubois: Lutte tudiante et enjeux conomiques YouTube YouTube, 31 Jul. 2 012. Web. 11 Mar. 2013. Dubois: Lutte tudiante et enjeux conomiques YouTube YouTube, 31 Jul. 2012. Web. 11 Mar. 2013. Dubois: Lutte tudiante et enjeux conom iques Online video clip. YouTube YouTube, 31 Jul. 2012. Web. 11 Mar. 2013. Heisler, Troy. 1996. OK A dynamic discourse marker in Montreal French. In Sociolinguistic Variation Data, Theory, and Analysis: Selected Papers from NWAV 23 at Stanford Ar nold, Jennifer, Brad Davidson, Scott Schwenter, and Julie Solomon (Eds). Stanford: Center for the Study of Language and Information. 1996. pp. 293 312. Gold, Elaine and Mireille Tremblay. 2006. Eh? and Hein?: Discourse Particles or National Icons? Canadia n Journal of Linguistics 21 (2/3): 247 263. Labov, William. 1972. Sociolinguistic patterns. Vol. 4. Philadelphia : University of Pennsylvania Press. Lessard, Denis. 2012. Lo Bureau Blouin candidat du PQ. La Presse. http://www.lapresse.ca/actualites/politique quebecoise/201207/23/01 4558296 leo bureau blouin candidat du pq.php Accessed 5/1/2013.
63 Lewis, Diana. 2005. Arguing in English and French asynchronous online discussion. Journal of Pragmatics 37: 1801 1818. performance communicationnelles: des strategies argumentatives pour le dbat politique tlvis. Argumentation et Analys e du Discours 1: 2 12. Dubois Martine Desjardins RDI 24 Online video clip. YouTube YouTube, 13 Feb. 2013. Web. 11 Mar. 2013. Poplack, Shana and Stephen Levey. 2010. Contact induced grammatical change: A cautionary ta le. In Language and Space: An International Handbook of Linguistic Variation, Volume 1: Theories and Methods Auer, Peter and Jrgen Erich Schmidt (Eds). De Gruyter Mouton. 2010. pp. 391 419. Sankoff, Gillian, Pierrette Thibault, Naomi Nagy, Hlne Blonde au, Marie Odile Fonollosa, and Lucie Gagnon. 1997. Variation in the use of discourse markers in a language contact situation. Language Variation and Change 9: 191 217. Sorochan, Cayley. 2012. The Quebec Student Strike A Chronology. Theory & Event 15(3): 2012 Supplement. Online video clip. YouTube YouTube, 30 Oct. 2012. Web. 11 Mar. 2013. Tagliamonte, Sali and Rachel Hudson. 1999. Be like et al. beyond America: The quotative system in Britis h and Canadian youth. Journal of Sociolinguistics 3 (2): 147 172. Thibault, Pierrette. 1997. Age. In Sociolinguistique: les concepts de base Marie Louise Moreau (Ed.). Mardaga. 20 26. Thibault, Pierrette and Michelle Daveluy. 1989. Quelques traces du pa ssage du temps dans le parler des Montralais,1971 1984. Language Variation and Change 1 (1) : 19 45. TVA Nouvelles TVA Nouvelles, 30 Mar. 2012. Web. 11 Mar. 2013. TVA Nouv clip. TVA Nouvelles TVA Nouvelles, 11 Apr. 2012. Web. 11 Mar. 2013. TVA Nouvelles TVA Nouv elles, 23 May 2012. Web. 11 Mar. 2013.
64 Vincent, Diane. 1993. Les Ponctuants de la langue et autres mots du discours Nuit Blache Editeur. YouTube YouTube, 30 Apr. 2012. Web. 11 Mar. 2013.
65 BIOGRAPHICAL SKETCH The author obtained her Bachelor of Arts degree with a major in French and Francophone studies from Earlham College in 2007. She then spent a few years working before beginning the Master of Arts program in French and Francophone Stud ies, with a specialization in French linguistics, at the University of Florida in 2011. Her research interests include sociolinguistics, syntax, and psycholinguistics, especially child language acquisition.