1. Introduction

In the field of language attitude research a variety of methodologies has been applied since researchers started investigating the general public’s attitudes towards language varieties and their speakers in the 1930s. Regardless of whether they have employed direct or indirect, or quantitative or qualitative methods, or the matched guise or the verbal guise techniques, the common goal of these studies has usually been to measure people’s attitudes towards different dialect or accent varieties of particular languages as well as, in the majority of the cases, towards the speakers of specific language varieties.

Besides, in the past decades research in the field of language attitudes has also initiated experimenting with various language attitude measurement techniques in order to be able to measure different attributes and to draw general conclusions about the language attitudes in question. The process of experimenting has included re-testing and improving previous methods and developing new methodologies in order to uncover the ideal technique that could provide an overall picture of language attitudes being measured.

As a matter of fact, several other methods have also been incorporated in language attitude research; for example, researchers attempt to investigate language attitudes with the help of such generally applied methods as individual or group interviews that enable them to concentrate further on the context and general background of respondents’ language attitudes.

On the whole, the present paper aims, on the one hand, to summarize some key methodologies that are employed in language attitude research. On the other hand, it attempts to describe how interviews can be employed in investigating language attitudes. In particular, my goal is to argue for the need and the importance of such type of data collection, whereby I can present the result and the research outcome that has been collected by means of group interviews.

2. Background: Language attitude methodologies

The measurement techniques with the help of which language attitudes can be investigated may be categorized from different angles. For example, quantitative techniques apply statistics to be able to convey people’s attitudes in figures, while qualitative techniques might endeavor to identify the reasons behind figures, i.e. to uncover details behind figures. Apart from that, other methods might gain access to language attitudes either in a direct or in an indirect way, in other words, by asking people straightforwardly about their attitudes or requesting people to articulate their reactions to the objects of investigation in the framework of testing conditions, respectively.

Indirect methods have proved to be the most favorable and most reliable types of attitude measurement techniques throughout the past decades of language attitude research. From the beginning of indirect attitude research, speaker evaluation experiments have been carried out where listeners either have had to assess speakers’ voices after a radio broadcast or they have had to match recorded voices to potential photographs of the speakers (Mobärg 1989: 7–10). Later, in the 1960s, Lambert and his associates (Lambert et al. 1960, cited in Preston 1989:50) developed the matched-guise technique where listeners were asked to judge the same speakers talking with two different accents of the same language. The main idea of this type of technique was to control all variables beyond the concrete accents under investigation. As it occasionally seems to be complicated to find speakers who can master speaking with two different accents simultaneously, the verbal-guise technique has been introduced into language attitude studies where representatives of the accents in question are asked to provide speech samples that are later evaluated by the respondents (McKenzie 2008). Overall, these techniques are the most widely applied language attitude measurement methods. Moreover, there are procedures that are employed in individual studies, for example, the method of applying synthesized speech samples, where speech sample voices are modified to facilitate examining language attitudes towards not only accent but also phonetic differences of speakers (Levon 2006).

In the present study my general aim is not to show how one of the above described techniques function in attitude research but to discuss another technique, the interview, that has been utilized extensively in sociolinguistics and applied linguistics, accordingly in language attitude studies as well, and to observe whether and how language attitudes can be measured with its help.

3. The research procedure

In a previous study (Balogh 2008) I conducted a verbal guise language attitude study wherein I asked Hungarian university students to evaluate four speakers of different varieties of American English accents. Two speakers’ accents were Southern American English accents, and the other two speakers’ accents were different from the accent that is spoken in the Southern part of the United States. The participants in this study were required to judge the speakers on a 6-point Likert-type scale along nine character traits that can be classified into either a status or a solidarity dimension. Generally, the outcome of such
studies conducted with native speakers is that participants assess the speakers distinctively along two different dimensions, i.e. on the solidarity and on the status dimensions. Particularly, whereas standard variety speakers are frequently evaluated more positively on the status dimension and more negatively on the solidarity dimension, non-standard speakers are judged in the opposite way, that is, more negatively on the status dimension but more positively on the solidarity dimension. All in all, in the above mentioned verbal guise study, the Hungarian respondents assessed the speakers of the Southern American English accent varieties more negatively on all character traits and not along any dimensions than the two other speakers who did not speak with an accent typical in the Southern states of the U.S.

In the study reported on in this paper I conducted group interviews on the basis of the results of the above summarized study. The subjects of the interviews were Hungarian university students (N=27) majoring in English and/or American studies at the University of Szeged at the time of the data collection. Three group interviews were carried out in Hungarian, and the number of interviewees was nine in each group. The interviews started with the researcher (i.e. myself) presenting to the participants the previously described quantitative study as well as its results. Subsequently, I asked them to react and respond to what they were presented with. In addition, I requested them to attempt to explain what the reasons were behind the results, in other words, why the participants of the previous study provided the evaluations in question. Furthermore, I encouraged the interviewees to comment on the results and any topics that they believed to be related to the study or the outcome of the study. The interviews were recorded and transcribed for analysis.

4. Results and discussion

In general, the respondents’ contributions to the discussions can be assigned into different categories. First, the Hungarian students openly expressed their opinions and their knowledge about American English accent varieties concentrating, obviously, on the Southern American accent varieties, as it can be seen in Examples¹ 1, 2 and 3 below.

Example 1, Interview 1, Interviewee 1 (I1)

I1: “Southern American dialects are very different from the American standard, or they are more different, and the foreign language learners know almost only the standard.”

Example 2, Interview 1, Interviewee 2 (I2), Interviewer (Z)

Z: “What is the Southern [American English] accent like by the way?”

I2: “It’s like British English, /r/ is omitted before vowels and at the ends of the words. For me that is what is the most disturbing in it.”

Example 3, Interview 2, Interviewee 3 (I3)

I3: “And if somebody has seen a Kentucky Fried Chicken advertisement in its original language, they have already heard a Southern accent. It’s a little bit wavy; it’s a bit more intensive than the average.”

On the basis of these examples, participants’ perceptions of the American English accent varieties can be inferred. Interestingly, some of the interviewees claim that the Southern American English accent variety is different from the standard (Example 1) or from the average (Example 3) varieties; nevertheless, Interviewee 2 argues (Example 2) that the Southern American English accent is more similar to the British English accent.

Regarding specifically the Southern American English accent varieties, the respondents, as can be seen in Examples 4 and 5, characterize this accent variety as possibly scary or as being attached to negative images in the media. In other words, they state that negative or even ‘stupid’ characters might have such accents in advertisements or films, downgrading with this statement this variety and its speakers and evaluating them more negatively than other American English accent varieties and their speakers.

Example 4, Interview 2, Interviewee 3 (I3), Interviewee 4 (I4)

I3: “Well, what comes to your mind about Southern American States?”

I4: “Nothing.”  

I3: “For example such things as the Civil War, and in the Civil War why were they together? Well, because they kept slaves and then they must be some kind of slave holders, and if you’ve seen Western films in English, then “My dear Lady”. “What are you doing here you damn Yankee?”, they are always bad, evil guys, they are the ones who chase with guns the soldiers in blue uniforms, instead of helping. Or in modern films, they are the rappers. Broken cars, guns again.”

Example 5, Interview 3, Interviewee 5 (I5), Interviewer (Z)

Z: “And what kind of role do Southerners get?”

I5: “Well, stupid roles. Monkeys, animals. In American films guilty people are always black.”

In these examples participants’ lack of sociolinguistic knowledge about accents in general and about American English accent varieties in particular is also reflected. Indeed, participants substantiate the existence of the standard variety, both in the case of American and British English, which belief, in contrast, is rejected in sociolinguistics. In addition, as sociolinguistics considers all language (accent) varieties equally variable, the participants’ judgments of one or the other varieties as, for example, ‘disturbing’ or ‘wavy’ or ‘more intensive than the average’ are not supportable, thus, show the interviewees’ lack of sociolinguistic experience. Interestingly, besides specific accent features, general images of Southern Americans also contribute to the evaluations of the speakers of this variety.

Apart from the evaluations, Examples 4 and 5 also provide some explanations for the possible reasons why respondents judge Southern American English speakers more negatively. Such explanations can also be found in the additional examples (Examples 6, 7 and 8) below.

Example 6, Interview 1, Interviewee 6 (I6), Interviewee 7 (I7), Interviewer (Z)

I6: “Maybe they didn’t understand that much as…”

Z: “They read the same text.”

I6: “Maybe, even if the text is the same, they can say it differently and it’s not understandable. Maybe there is something like that in it, too.”

I7: “Maybe the one which was more understandable was more attractive at the same time.”

Example 7, Interview 2, Interviewee 8 (I8)

I8: “Then the reason is that we think a person more intelligent if they speak in a more variable, no, not variable, more articulate way.”

Example 8, Interview 3, Interviewee 9 (I9)

I9: “Americans have some kind of previous knowledge regarding the group, whereas Hungarians just put people into categories, they hear a voice and they are not able to evaluate it on the basis of dimensions. They are not able to differentiate between them on the basis of attraction.”

Apart from providing evaluations of the American English accent varieties and their speakers, the Hungarian subjects offer to justify the outcome of the previous quantitative study. On the one hand, the respondents give universal reasons for such results (Examples 6 and 7); on the other hand, they attempt to explain specifically the results of the given study (Example 8). The general reasons the participants provide are connected to the research procedure itself, namely, to the understandability of the speech samples or the articulation of the speakers who supply the speech samples. Regarding the given study, one of the participants (Example 8) attempts to differentiate between how native-speakers and non-native speakers evaluate a particular language variety. That is, Interviewee 9 claims that non-native speaker listeners might have difficulties in judging a language variety that is not their native language.

Indeed, the general explanations they provide show their language attitudes as well as linguistic ideologies that involve misbeliefs about languages and language varieties. Attitudes on the basis of the general evaluations respondents provide display, for example, that Hungarian students consider standard English varieties as the most favorable ones, the best varieties, and that they associate the standard varieties with the varieties that are taught in schools. Besides, they argue that the only version language learners learn at school and therefore know is the standard variety.

Besides attempts to provide explanations, respondents additionally seem to find solutions to how to accept non-standard accent varieties and their speakers. That is, while they themselves display negative attitudes towards non-standard American English accent variety speakers, they make an effort to find ways how such negative attitudes can be avoided or eliminated. For example, they argue for teaching more English varieties at schools or for integrating culture instructions into the language classroom; what is more, they also suggest that teachers and instructors ought to make language learners aware of the fact that all varieties of English are equally “good” and make learners experience positive feelings also in connection with non-standard varieties and their speakers.

5. Conclusion

In this paper I have presented a study wherein I aimed to examine whether and how language attitudes can be investigated with the help of group interviews. The findings of the study show that conducting interviews is a legitimate method to gain an overall picture of the interviewees’ language attitudes and that interview data might also reveal general ideologies respondents have with regard to certain accent varieties. In particular, the results of the present study correlate with the results of the quantitative study; however, the current qualitative type of attitude measurement technique can complement the former one by providing potential solutions on how to minimize the general public’s negative attitudes towards non-standard varieties.

Works cited

• Balogh, Erzsébet. 2008. Hungarian student language attitudes towards speakers of regional American English accent varieties. In: Ewa Waniek-Klimczak (ed.) Issues in accents of English. Newcastle upon Tyne: Cambridge Scholars Publishing, 150–168.
• Lambert, Wallace E., Richard C. Hodgson, Robert C. Gardner and Samuel Fillenbaum. 1960. Evaluational reactions to spoken languages. Journal of abnormal and social psychology, 60: 44–51.
• Levon, Erez. 2006. Hearing “gay”: Prosody, interpretation, and the affective judgments of men’s speech. American Speech, 81(1): 56–78.
• McKenzie, Robert. 2008. Social factors and non-native attitudes towards varieties of spoken English: A Japanese case study. International Journal of Applied Linguistics. 18(1): 63–88.
• Mobärg, Mats. 1989. English “standard” pronunciations: A study of attitudes. Götebord: Goterna, Kungälv.
• Preston, Dennis R. 1989. Sociolinguistics and second language acquisition. Oxford: Basil Blackwell.

Notes

¹ The original Hungarian utterances are translated into English by the author of the present study. ↩

Copyright (c) 2011 Erzsébet Balogh

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.