|Trans||Internet-Zeitschrift für Kulturwissenschaften||15. Nr.||Juni 2004|
6.1. Standardvariationen und
Sprachauffassungen in verschiedenen Sprachkulturen | Standard
Variations and Conceptions of Language in Various Language Cultures
Meryem Sen [BIO] & Nazli Baykal (Kocaeli University & The Middle East Technical University, Turkey)
This study is concerned with the social and linguistic evaluations of standard accented speakers related to non-standard accented speech and speakers. Two groups consisting of 80 university students, half male and half female, participated in the study. One group of students attending the Department of English Language and Literature was identified as linguistically aware, the other group of students attending the Department of Food Engineering as linguistically unaware. The recorded comments of the subjects about accented speech and speakers were analysed and cumulated in the categories of knowledge, emotion and linguistic behaviour. While the results showed a more critical perspective on the part of linguistically aware females towards accented speech and speakers, they, on the part of linguistically unaware males and unaware females, pointed to more stringent standards of acceptability of accented speech and of some personal traits of accented speakers. On the other hand, the evaluations of linguistically unaware males tended to become more positive towards accented speech and speakers in many respects.
Language attitude research, despite the fact that it is a main concern of social psychology, has become a major part of sociolinguistics constituting an important source of data since 1970. It has provided us with social evaluations of language use in society including systematic patterns about language maintenance, shift, planning, adoption, death, and birth. Peter Garrett (2001: 630), in Dialogue, draws the attention towards the revealing aspects of language attitude research:
"Language attitudes research in sociolinguistic communities can reveal the dynamic identificational and relational forces at work within them. These include prejudices held against (or in favour of) regional or social varieties. They also include allegiances and affiliate feelings towards one's own or other groups' speech norms. ... So, in addition to sociolinguistic processes at the level of the social group, social evaluative studies can access local processes of interpersonal attraction and distancing and help anticipate the character of social relationships ... And since explanations of socio-linguistic phenomena are most likely to be found in social psychological processes, language attitudes are a key component of sociolinguistic theory-building."
Language attitude studies reaching as far back as 1930's are concerned with numerous topics in the context of regional and standard language, and language behaviour (for reviews Giles et. al 1987; Bradac 1990). The vast majority of these studies have focused mainly on profiling people's evaluative reactions to a host of accents and languages. For example, in Wilson and Bayard's (1992) study New Zealand listeners' evaluations of New Zealand-, Canadian-, Australian-, British- accented English were examined in the sense of their linguistic perceptions. Their results indicated that, among other things, the Australian-accented speaker was perceived to be the most intelligent and most likeable, followed by the Canadian-, and then the New Zealand-accented speakers. The British-accented speaker was perceived to be one of the most intelligent but one of the least likeable of all speakers. Thus, it seems to be possible to be informed through language attitude studies about listeners' both cognitive and emotive evaluations towards non-accented speakers (Wilson & Bayard 1992; Cargile & Giles 1997).
Studies that examine individual language attitudes towards regional dialects and accent variations, tend to view ' attitude' as a reaction to a linguistic stimulus (Cargile & Giles 1997). In the studies conducted by Demirci and Kleiner (1999), and Demirci (2000) regarding the perception of Turkish dialects by Turkish respondents it is shown that several significant, systematic, gender and age-based patterns emerge from the evaluation of the regional dialects, and that many of the phenomena reported in findings for U.S society are also found in Turkish society. That is, respondents favoured their own dialects. Similarly, it has been observed that listeners' own linguistic abilities influence perception in the context of group solidarity. For instance in Luhman's study (1990), Appalachian English speakers did not accept the stereotypical beliefs that Appalachian English has a lower status in comparison to standard English.
However, as suggested by Cargile and Giles (1997) the need exists to move language attitudes research beyond the traditional paradigm of simply charting evaluative profiles, which could be possible by taking listeners' affective reactions to speakers and to their social identities into consideration. Because there are some other factors which affect listeners' perception such as listeners' social or ethnic identity, vocal features, situational characteristics and the context in which verbal interaction occurs (Bradac 1990). Scholars have recognised that social identities are critically important in human interactions (Tajfel & Turner 1986; Giles & Johnson, 1987; Deaux, 1993), yet the study of language attitudes has failed to account fully for the ways in which these identities may influence the nature and expression of these studies.
Furthermore, though there are some (e.g., Labov 1972; Trudgill 1972, 1974; Milroy 1980), but not very much sociolinguistic research has been carried out to compare the attitudes and evaluations of men and women of the linguistically arware-group towards accented speech and regional varieties. Hence, in response to this need, the present study was designed and undertaken. Our purpose was in conducting this study to explore the effect of speakers' linguistic awareness in evaluating non-standard accented speakers and speech. Specifically, we are in search of understanding linguistically aware and unaware speakers' affective reactions towards accented speakers and accented speech. We begin discussion of this study with a brief review of the relevant literature.
Research on language perception generally shows that standard language is preferred more favourably than non-standard language. In many cases, using the matched-guise technique developed by Lambert, Hudson, Gardner and Fillenbaum (1960), standard and non-standard speakers' characteristics related to status, attractiveness and dynamism have been displayed. In Ryan, Hewstone and Giles' (1984) study standard accented speakers had more positive reactions than non-standard accented speakers in status characteristics (for example intelligence and education etc.) . On the other hand, in the context of attractiveness (for example friendliness, likeability) nonstandard accented speakers are found more likeable and friendly than standard accented speakers. A contrary result to this case was obtained in the studies related to speakers who spoke Japanese- (Cargile & Giles 1998) and British accented English (Stewart, Ryan & Giles 1985). In those studies speakers using standard accented English were found to be less attractive in the context of attractiveness, whereas they had got positive reactions concerning with their status characteristics. It is obvious that Americans have stereotypical beliefs about the Japanese and the British that they are reserved, but competitive and have higher status.
Accent variations are the chunks of informative data enabling listeners to make social evaluations about speakers. Lippi - Green (2001:30) emphasizes that we perceive the variations in others' speech and use this information in the construction of personality. As it is clearly indicated in Lippi- Green's words, listeners, as interactive participants, are in a position of receiving messages from speakers, analyzing and evaluating them during the process of interaction. The literature of social psychology supports the view that accents are the paralinguistic sources of knowledge (Riches & Foddy 1989). They convey information in the context of social attractiveness and social power (Giles 1973; Luhman 1990). Social attractiveness is a well-known aspect of the sociolinguistics literature and deals with human characteristics such as trustworthiness, dependability and friendliness (Bayard 1995; Podberesky, Deluty & Feldstein 1990). Generally, English speaking countries evaluate their own accents higher on social attractiveness and perceive native accented speakers more trustable, friendly and dependable (Giles 1973; Luhman 1990). Accents rated higher on social attractiveness have been also found more persuasive (Giles 1973).
Accents are indicative about the power of the speaker. Generally, in studies related to various English accents, it is shown that the accented speaker is perceived less powerful when the difference between the standard accented speaker and the non-standard accented speaker becomes greater (Callan & Forbes 1983). Besides, standard language and standard accent have been found more powerful and respectable in various British accents (Stewart, Ryan & Giles 1985; Luhman 1990). In Giles's (1973) study, socially attractive speakers have been found more persuasive than socially powerful speakers.
As mentioned above, this study is mainly concerned with standard accented speakers' perceptions and evaluations of non-standard accented speakers and accented speech in terms of knowledge, emotion and linguistic behaviour. It is also the concern of the study that how standard accented speakers' linguistic awareness affects their perceptions in evaluating accented speech and accented speakers. Hence, the aim of this study is of two sided in this manner: first, to unfold language attitudes of standard accented speakers towards accented speakers and speech; second to expore the effects of speakers' linguistic awareness on their evaluations of accented speakers and speech. Specifically, we are looking for answers to the following questions:
Q1. What sociolinguistic impressions does non-standard accent make on the standard accented speaker?
Q2. Is there a linear relationship between accented speech and accented speakers' personality attributions?
Q3. What attitudinal differences exist between linguistically aware and aware speakers in evaluating non-standard accented speech and speakers?
In view of the above presentation and reasoning this study has been conducted in order to examine the validity of the following hypotheses:
H1. Linguistically aware and unaware speakers differ in their linguistic attitudes towards accented speakers and accented speech.
H2. The attitudes of linguistically aware female speakers are negative towards accented speech but positive towards accented speakers.
H3. The attitudes of linguistically aware male speakers are positive towards accented speech and speakers.
H4. Both linguistically aware male and female speakers are not willing to communicate with accented speakers.
H5. Both linguistically unaware male and female speakers display positive emotional attitudes towards accented speech and speakers.
H6. Both linguistically unaware male and female speakers are willing to communicate with accented speakers.
Numerous studies, using matched-guise technique based on stimulus-response theory (Lambert 1967), tried to explain language attitudes of listeners towards the language and accent of the speakers (for review Giles et.al 1987). However, studies using this technique are deprived of revealing language attitudes as they neglect listeners' social identities, emotions and awareness (for review Sebastian et.al 1980; Giles et.al 1995), which are important in human communication and interaction (Tajfel & Turner 1986; Giles & Johnson 1987). Therefore, a hermeneutic approach supported by in-depth interviews is employed in this study to explain gender related linguistically aware or unaware language attitudes towards accented speech and speakers.
The data base for this study was composed of in-depth interviews made with 80 students, 40 of whom have been attending Kocaeli University, English Language and Literature Department, and the other 40 attending The Middle East Technical University, Food Engineering Department. Each group of subjects were equal in number of female and male participants. Those attending the Department of English Language and Literature were identified as linguistically aware, and the others attending The Department of Food Engineering as linguistically unaware. Their educational variation was taken as the criteria in labelling groups as linguistically aware and unaware.
Subjects were asked to give their comments and opinions about accented speech and accented speakers that they met or had verbal interactions or exposed to in their daily lives. The researchers' guidance during the interviews was limited with supporting the fluency of the conversation through asking questions when the speaker could not proceed further. Those questions were very general in nature and did not include any prejudices ( e.g., what is your opinion about the strength of accented speech?). All the interviews were recorded and transcribed later.
The transcribed data were analysed, and speakers' comments about accented speech and speakers were grouped into three main categories: knowledge, emotion and linguistic behaviour. Their subcategories are displayed in figures 1, 2 and 3. Speakers' comments presenting knowledge about accented speech and speakers were included in the category of Knowledge (Fig. 1); comments related to speakers' emotional perceptions of accented speakers' individual and personal identity traits, (such as their likeability, intimacy, status, educational level etc.) took place in the category of Emotion (Fig. 2); and comments about speakers' perception of the relationship between accented speech and the identity of the accented speaker, and speakers' willingness for communicating with accented speakers were included in the category of Linguistic Behaviour (Fig. 3).
Fig.1. Subcategories in the category of Knowledge.
Fig. 2. Subcategories in the category of Emotion.
Fig. 3. Subcategories in the category of Linguistic Behaviour.
7.1. Evaluation of Accented Speech in the category of Knowledge
Female and male subjects' evaluations of accented speech are displayed in Table 1. Female subjects attending the Department of English Language and Literature (LF hereafter) commented on accented speech positively in respect to its message transmission (60%), fluency (50%), attractiveness (60%) and contextual appropriateness (50%). On the other hand, they did not find accented speech grammatically appropriate (0%) and acceptable (20%). Male subjects attending the Department of English Language and Literature (LM hereafter) displayed negative attitudes to accented speech in all categories except for the category of attractiveness ( 60%).
L: Students attending the Department of English Language and Literature, linguistically aware.
Table 1. Evaluation of Accented Speech in the category of Knowledge (%) Female Male Female Male L E L E L E L E Message Transmission 60 40 30 90 + - - + Grammatical Structure 0 10 10 10 - - - - Fluency 50 40 30 40 + - - - Attractiveness 60 50 60 70 + + + + Acceptability 20 30 10 20 - - - - Contextual Appropriateness 50 40 20 50 + - - +
E: Students attending the Department of Food Engineering, linguistically unaware.
The second group of female students, attending the Department of Food Engineering (EF hereafter), commented positively on the attractiveness of accented speech (50%) and negatively on the other characteristics; however, male engineering students (EM hereafter) evaluated accented speech positively regarding its message content (90%), attractiveness (70%) and contextual appropriateness (50%).
7.2. Evaluation of Accented Speakers in the Category of Knowledge
Table 2. Evaluation of Accented Speakers in the category of Knowledge Female Male Female Male L E L E L E L E Language Competence 10 20 30 40 - - - - Language Performance 30 40 20 30 - - - -
Table 2 shows the attitudes of standard accented female and male subjects towards accented speakers in terms of knowledge category. It is obvious that all subjects were not satisfied with accented speakers' language competence and performance.
7.3. Evaluation of Accented Speakers in the category of Emotion
Table 3. Emotional Evaluation of Accented Speaker's Individual Personality. (%). Female Male Female Male L E L E L E L E Intelligence 50 20 50 30 + - + - Intimacy 50 70 70 90 + + + + Likeability 30 50 30 60 - + - + Dynamism 20 20 40 40 - - - - Talkativeness 40 30 60 20 - - + - Self-confidence 40 60 40 70 - + - +
Table 3 shows that LF and LM subjects evaluated accented speakers favourably in the category of intelligence (50%, 50% females and males respectively), intimacy (50%, 70% females and males respectively) and evaluated negatively in the categories of likeability (30%, 30% females and males respectively), dynamism (20%, 40% females and males respectively), and self-confidence (40%,40% females and males respectively). On the other hand, while LF subjects were evaluating accented speakers' talkativeness negatively (40%), LM subjects evaluated them positively (60%) in this respect. Both EF and EM subjects' positive evaluations of accented speakers were observed in the category of intimacy (70%, 90% females and males respectively), likeability (50%, 60% females and males respectively), and self-confidence (60%, 70% females and males respectively).
7.4. Evaluation of Accented Speakers in the Social Context (education)
Table 4A displays females' and males' evaluation categories of accented speakers in the social context. According to the table LF subjects seem to be more creative than LM subjects in creating categories related to the educational level of accented speakers such as primary, secondary school graduate or uneducated. Both EF and EM subjects' categorisation of accented speakers remained within two, as uneducated and not educated enough by female respondents, and not educated enough and not university educated by male respondents.
Table 4A. Evaluation of Accented Speaker in the Social Context Education Education Female Male L E L E Primary School 20 - 10 - Secondary School 10 - - - University 20 - - - Uneducated 20 40 - - Not educated enough - 30 30 50 Not University educated 20 - 30 10
7.5. Evaluation of Accented Speakers in the Social Context (Class)
Table 4B. Evaluation of Accented Speaker in relation to Social Context - Class (%) CLASS Female Male L E L E Lower 10 10 10 40 Middle 50 40 50 10 High 20 - 20 -
In table 4B, it is observed that both LF and LM subjects' evaluation of accented speakers seemed to be dispersed within different categories, but cumulated mainly around middle class ( 50%). EF and EM subjects differed in this respect: while the females' evaluation (40%) was middle class, the males' (40%) was lower class (40%).
7.6. The category of Linguistic Behaviour (Accent + Identity Relationship)
In Table 5, all EF subjects (100%) and majority of EM subjects, LF subjects (60%) and LM subjects (40%) claimed a connection between accented speech and social identity. However, 50% of LM subjects were not supportive of accent and personality relationship.
Table 5. Accent + Identity Relationship (%). Female Male Female Male L E L E L E L E Yes 60 100 40 80 + + - + No 10 - 50 20 - - + -
7.7. The category of Linguistic Behaviour (Willingness for Verbal Interaction
All subjects expressed their willingness to communicate with accented speakers (Table 6). While EF and EM subjects' willingness were observed over average (90%, 80% females and males respectively), LF and LM subjects displayed a moderate behaviour in this respect (60%).
Table 6. Willingness to interact with Accented Speakers (%) Female Male L E L E Yes 60 90 60 80 No 20 10 30 10 Yes / No 20 10 10
7.8. Categorisation of Accented Speakers
The results in Table 7 show that all subjects create categories in perceiving accented speakers. However, the percentages of negative responses also indicate that the subjects remained hesitant in this respect.
Table 7. Categorising the Accented Speaker Female Male L E L E Yes 60 50 40 70 No 40 40 40 30
7.9. Subcategorisation of Accented Speaker
LF subjects subcategorised accented speakers as educated / uneducated, asocial, illiterate, cultured / uncultured (not intellectual) and ethnic oriented (Table 8). As opposed to LF subjects, LM subjects produced only cultured and uncultured subcategories. While EF subjects did not create any subcategory in this respect, EM subjects identified the category of the accented speaker as uncultured and ethnic originated.
Table 8. The categories in which the Accented Speaker is perceived female male L E L E Educated / Uneducated + - - - Asocial + - - - Illiterate + - - - Cultured / Not cultured + - + + Ethnic originated + - - +
Linguistically aware female subjects evaluated accented speech positively in the categories of message transmission, fluency, attractiveness and contextual appropriateness and negatively in the categories of grammatical structure and acceptability. Linguistically aware male subjects, on the other hand, favoured accented speech only in the category of attractiveness, but negatively in the categories of message content, fluency, contextual appropriateness, grammatical structure and acceptability. Although linguistically aware females displayed some positive evaluations towards accented speech in some respects, their attitude became similar in evaluating accented speech as grammatically inappropriate and unacceptable in comparison to standard speech.
Linguistically aware subjects did not differ in their evaluations of accented speakers: they both evaluated them positively in the categories of intelligence and intimacy, but negatively in the categories of likeability, dynamism and self-confidence. Thus, linguistically aware female and male subjects demonstrated common negative language attitudes in the evaluation of accented speakers: they found accented speakers not likeable, dynamic, and self-confident. The unique difference between linguistically aware females and males is that males evaluated them as talkative, too.
These findings seem to contradict the results of many studies in the sociolinguistic literature as it is a well-known phenomenon that females, compared to males, value standard language over non-standard language (e.g., Trudgill 1974; Fasold 1990; Romaine 1994). While women tend to use standard variety, men are more inclined to the covert prestige of non-standard varieties of language use. However, in the present study linguistically aware female subjects display more tolerant attitude in comparison to males in the evaluation of accented speech in the categories of message transmission, fluency, attractiveness, and contextual appropriateness. On the contrary, linguistically aware male subjects' evaluation of accented speech negatively in all categories except for the category of attractiveness, and accented speakers negatively apart from the categories of intelligence, intimacy and talkativeness contradicts findings that males are more positive than females towards accented speech and regional dialects. In this case our hypothesis that linguistically aware males would be more positive towards both accented speech and speakers was not completely confirmed. On the other hand, the hypothesis that linguistically aware females would be negative towards accented speech, but positive towards accented speakers was partially confirmed.
Linguistically unaware female subjects did not favour accented speech except for its attractiveness, thereby they have displayed a more radical language attitude than linguistically aware females. As opposed to them linguistically unaware male subjects evaluated accented speech positively in terms of message transmission, attractiveness and contextual appropriateness, which confirms partially our related hypothesis that both linguistically unaware subjects' attitudes towards accented speech would be positive. However, it is significant that they did not favour accented speech grammatically appropriate and acceptable. Interestingly, the linguistic attitude of linguistically unaware males towards accented speech has become much alike to that of linguistically aware females. They both evaluated accented speech positively in terms of its message content, attractiveness and contextual appropriateness, which indicates a more relaxed standpoint on the part of linguistically unaware males and linguistically aware females. Another striking resemblance in the linguistic attitudes towards accented speech has occurred between linguistically unaware females and linguistically aware males. They appreciated accented speech only in terms of attractiveness, which means a more critical perspective on their part. However, the one common point is that all subjects, linguistically aware and unaware, do not regard accented speech grammatically appropriate and linguistically acceptable. This common attitude towards accented speech points clearly to the stringent standards of acceptability in their awareness..
In the evaluation of accented speakers linguistically unaware subjects, males and females, described them as intimate, likeable and self-confident, but not as intelligent, dynamic and talkative. In this way, their evaluation has become much alike with linguistically aware subjects' evaluation of accented speakers. However, they differed in the categories of their evaluations: while linguistically unaware subjects describe accented speakers with solidarity dimensions as intimate, likeable and self-confident, linguistically aware subjects describe them as intelligent and intimate. The reverse of this explanation is that linguistically unaware subjects do not find accented speech grammatically appropriate and acceptable (compared to standard language), and they also do not find accented speakers intelligent, dynamic and talkative. Thus, it seems plausible to claim that there is a partial linear relationship between accented speech and personality traits. The same tendency exist between linguistically aware subjects' evaluations of accented speech and accented speakers. However, both groups which partially appreciate accented speakers do not display a firm linear relationship between accented speech and personality traits.
When the group members are compared to each other in terms of their linguistic behaviour, however, some significant and interesting results were obtained. Linguistically unaware males, in comparison to linguistically aware males, perceive accented variations in a linear relationship. That is, they perceive accented speech meaningful, attractive and contextually suitable, and perceive accented speakers in the same manner as intimate, likeable, and self-confident. However, we cannot talk about the same sort of linearity in the evaluation of linguistically unaware females. They are, like linguistically aware males, rather more critical about accented speech than linguistically aware females as the only evaluative strategy to which they ascribe accented speech is its attractiveness. This is a more critical perspective on the part of linguistically unaware females and linguistically aware males.
One common evaluation between linguistically aware and unaware subjects is about the relationship of accent and identity. They all agree that the accent reflects the identity of the speaker. And apart from linguistically aware males they also agree that they categorise accented speakers. Among the subjects linguistically aware females displayed more detailed categories than other subjects in relation to accented speakers' social identity in terms of status, education and class, and categories in which they perceive and evaluate accented speech, which points to a biased attitude as well. Because they define accented speakers as low/middle class members, who can be uneducated, asocial, illiterate, and ethnic originated. It is clear that they regard accented speech as semiosis (Garret et al. 1999) of low culture, low/middle class and status, and uneducatedness in the context of social identity. However, the term ' 'bias' here reflects 'social reality', not psychological one, which indicates a more critical perspective on their part. In comparison to them linguistically unaware females did not mention any category in this respect, but linguistically unaware males two (uncultured and ethnic originated) and linguistically aware males one (uncultured). This case shows that linguistically aware females have more concrete readily applicable thoughts (Tajfel & Forgas 1982) in their awareness.
Nonetheless, both groups of female and male subjects are observed as willing to have linguistic interaction with accented speakers, which is inconsistent with our hypothesis related to the linguistically aware subjects that they would be unwilling to communicate with accented speakers. On the other hand, the hypothesis related to linguistically unaware subjects' willingness for interaction with accented speakers has been confirmed.
Consequently, the findings of the present study show that non-standard accented speech and speakers do not hold a favourable impact in the perception and evaluation of the standard accented speaker. Yet, we cannot talk about an absolute stigmatisation towards accented speech and speakers since both linguistically aware and unaware subjects are all willing for verbal interaction with them, and they display their positive attitudes towards some aspects of accented speech and the accented speaker's individual identity. For instance, approving the message transmission and attractiveness of accented speech, and perceiving accented speakers as intimate, intelligent and so on. Hence it seems to be plausible to talk about positive attitudes towards accented people and accented speech in some respects, but not about the grammatical appropriateness and acceptability of accented speech. The results do not point to a significant discrimination between linguistically aware and unaware Turkish speakers' attitudes towards accented speech and speakers. However, it seems to be important to explore the role of linguistic awareness in linguistic attitudes of people for better language policies. Thus, it is promising in future to conduct research evaluating the effects of linguistic awareness in language attitudes with more respondents.
© Meryem Sen & Nazli Baykal (Kocaeli University & The Middle East Technical University, Turkey)
6.1. Standardvariationen und Sprachauffassungen in verschiedenen Sprachkulturen | Standard Variations and Conceptions of Language in Various Language Cultures
Sektionsgruppen | Section Groups | Groupes de sections
Inhalt | Table of Contents | Contenu 15 Nr.