Taikomoji kalbotyra, 21: 62–76 eISSN 2029-8935
https://www.journals.vu.lt/taikomojikalbotyra DOI: https://doi.org/10.15388/Taikalbot.2024.21.5

Uneducated and less trustworthy. Social distance in subconscious language attitudes towards Lithuanian-Russian accent speakers

Prieglius Žuklys-Rogoża
Vilnius University
E-mail: prieglius@gmail.com

Abstract. Sociological surveys have shown that social distance towards Lithuanian-Russians (the so-called “ethnic Russian minority” in Lithuania) is rather marginal or non-existent. This paper presents a pilot study at Vilnius schools (n=151), which used the verbal guise technique to investigate subconscious language attitudes towards Lithuanian (i) with Vilnius speech traits, (ii) with a Lithuanian-Russian accent, and (iii) with a Scandinavian accent – an accent presumed unfamiliar to the study participants. Half of the participants were also informed of the speaker’s profession to assess if it can reverse their attitudes. The main findings of the study show that speakers with a Lithuanian-Russian accent are perceived as less interesting, less educated, less trustworthy, and older. These results put into question the notion of absent social distance towards Lithuanian-Russians, highlighting the need for further research on this topic.
Keywords: subconscious language attitudes, Lithuanian-Russians (“ethnic Russian minority”), social distance, Lithuanian language with an accent, unfamiliar accent

Neišsilavinęs ir mažiau patikimas. Socialinė distancija pasąmoninėse nuostatose lietuvių kalbos su rusišku akcentu atžvilgiu

Santrauka. Sociologinės apklausos rodo, kad socialinė distancija lietuvių-rusų atžvilgiu yra nedidelė arba jos nėra. (Straipsnyje vartojama sąvoka „lietuviai-rusai“ ­kaip įtraukesnis pavadinimas grupei, kuri dažnai vadinama „rusų etnine mažuma“, „Lietuvos rusais“ arba „rusakalbiais“.) Šis straipsnis pristato pilotinį tyrimą, atliktą keturiose Vilniaus gimnazijose su mokiniais (n=151). Tyrime naudotas nesuporuotos kaukės testas (angl. verbal guise test), siekiant ištirti moksleivių pasąmonines nuostatas trijų kalbinių atmainų atžvilgiu: lietuvių kalbos su a) Vilniaus kalbos savybėmis, b) rusišku akcentu ir su c) tyrimo dalyviams, tikėtina, mažai pažįstamu skandinavišku akcentu. Ankstesniems tyrimams liudijant, kad kalbėtojai lietuviškai su „slavišku“ akcentu dažniausiai siejami su žemo socialinio statuso profesijomis, šiuo tyrimu taip pat siekiama patikrinti, ar galima pakeisti nuostatas kalbėtojų atžvilgiu, jeigu tyrimo dalyviams pasakoma, kad kalbantieji su rusišku akcentu turi aukšto socialinio statuso profesiją. Pagrindiniai tyrimo rezultatai indikuoja, kad kalbėtojai lietuviškai su rusišku akcentu yra laikomi mažiau įdomiais, mažiau išsilavinusiais, mažiau patikimais ir vyresniais. Tyrimo rezultatai taip pat rodo, kad rusiško akcento indeksai pagal socialinį kontekstą (keičiamas kalbėtojų profesijas) reikšmingai nekinta. Tyrimas kvestionuoja ankstesnius teiginius, kad socialinės distancijos lietuvių-rusų atžvilgiu nėra, ir rodo tolesnių tyrimų šia tema poreikį.
Raktažodžiai: pasąmoninės kalbinės nuostatos, lietuviai-rusai („Lietuvos rusai“), socialinė distancija, lietuvių kalba su akcentu, nepažįstamas akcentas

_________

Acknowledgments. I would like to thank Loreta Vaicekauskienė for her immensely helpful comments on this article. I am also indebted to Ramunė Čičirkaitė for her useful feedback, and Jūratė Ruzaitė for her help in preparing the manuscript.

Copyright © 2024 Prieglius Žuklys-Rogoża. Published by Vilnius University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use,
distribution, and reproduction in any medium, provided the original author and source are credited.

1. Introduction

Intuitively one would think that a multi-ethnic city, where there are many contacts between various ethnic groups, would demonstrate a high level of tolerance towards different language. Vilnius, the capital of Lithuania, can be seen as a multi-ethnic city: 33 percent of inhabitants identify as a nationality other than Lithuanian (Lietuvos statistikos departamentas 2021). One of the largest ethnic minority groups is Lithuanian-Russians¹. Sociologists have used the term “social distance” (a term coined by Bogardus (1926)) to describe remoteness among various social and ethnic groups, a phenomenon that can be observed by a lack of willingness to participate in activities together, work together or live nearby. Although some sociologists claim that social distance towards Lithuanian-Russians in Lithuania is not present or marginal (Vildaitė and Žibas 2010: 124; Petrušauskaitė 2013; Blažytė et al. 2016: 114–115), this paper presents contrary evidence from the pilot study on subconscious language attitudes among Vilnius adolescents.

To study subconscious language attitudes, a verbal guise test was conducted at four schools in Vilnius where language of instruction was Lithuanian. The participants in the experiment – 151 pupils in ninth- and tenth-grade (ages 14–15) – listened to six male voice recordings. Two of them represented typical Vilnius speech, two of them were with a Lithuanian-Russian accent, and two with a Scandinavian (Danish and Norwegian) accent, which was expected to be largely unfamiliar to the study participants. With the content of the recordings and the fluency of the speakers controlled to be as similar as possible, the accent (the phonetic and prosodic features) was virtually the only variable in the experiment. The pupils were then asked to answer a combination of open- and close-ended questions related to superiority traits (e.g. intelligent, educated), dynamism traits (e.g. interesting, self-confident)² and trustworthiness. A verbal guise test by Čičirkaitė (2019) indicates that speakers with the so-called Lithuanian-Slavic accent (i.e. the accent characteristic of both Polish and Russian speakers) are less frequently described as “educated”, “successful” or “having a good job” compared to speakers with Vilnius speech traits (Čičirkaitė 2019: 139). Therefore, this study aims to compare the evaluations of Lithuanian-Russian and Scandinavian accents to assess if Čičirkaitė’s (2019) findings apply to any accent-marked Lithuanian speech or specifically to the Lithuanian-Russian accent. An unfamiliar, very rarely heard Lithuanian with a Scandinavian accent could be expected to have limited associations with stereotypes linked to a specific social group.

Bearing in mind that speakers with a Lithuanian-Russian accent tend to be associated with low-status professions (Čičirkaitė 2019: 140), this study also raises the question if these associations with the accent can be reversed by providing additional information about the speakers. A similar method has been used in other language attitude studies, where experiment participants were provided with the speakers’ names (e.g. Prikhodkine et al. 2016), photos (e.g. Holmes et al. 2001), or professions (e.g. Campbell-Kibler 2010). Campbell-Kibler’s (2010) study, for instance, examined evaluations of standard /ing/ and the non-standard /in/ pronunciations, finding that when speakers were identified as politicians, participants rated the /in/ guise as more caring than the /ing/ guise (Campbell-Kibler 2010: 218–219). In the present study, half of the participants were informed that the Lithuanian-Russian accented speakers had high-status professions, whereas the Vilnius speech and Scandinavian accented speakers held low-status professions. The other half of the participants did not receive any information on the speakers’ professions. This approach allows the study to investigate whether social interpretations of Lithuanian spoken with an accent vary from context-to-context or remain consistent across different contexts.

Thus, the main research questions for this study were as follows: (1) What subconscious language attitudes do adolescents at Vilnius schools with Lithuanian as the language of instruction hold towards Lithuanian-Russian and Scandinavian accents? (2) Do high- and low-status profession labels affect these attitudes towards the speakers? (3) Do these subconscious language attitudes confirm that social distance toward Lithuanian-Russians is absent?

2. Language attitudes and social distance

Language attitudes refer to how people perceive and evaluate a language variety (e.g. a dialect or a sociolect), an accent or a language feature, including the associations they form with them. In recent decades, sociolinguistic studies have shown that when people have attitudes towards a language variety, they do not base their opinions on the aesthetic qualities of that variety but rather on the associations they have with speakers of that language variety (Kristiansen 2009; Preston 2009; Grondelaers and van Hout 2011; among others). In other words, if one describes a language variety as “beautiful”, it is not the sounds of the variety that form such an attitude but the association with intelligent or educated individuals, for example. This relationship between a language variety and various social variables (such as social status or personal traits) is called an index (Silverstein 2003). Thus, we can say that a language variety indexes, for instance, intelligence or lack of education.

However, different attitudes may be retrieved depending on whether researchers ask study participants about language directly (such attitudes are called conscious, explicit or public language attitudes) or indirectly (such language attitudes are called subconscious, implicit or private) (Kristiansen 2009: 169–171). To elicit subconscious language attitudes, questions should be formulated in a way that avoids asking about language directly, thus avoiding any mention of language, accent, dialect or speech (Kristiansen 2011: 275). Surprisingly, conscious and subconscious attitudes towards the same language variety may differ (see Vaicekauskienė 2017 for attitudes towards different Lithuanian language varieties). When people are aware that they are asked about language, they tend to express attitudes that are socially acceptable (we observe the so-called “social desirability bias” (Garrett 2010: 44)). Subconscious language attitudes are, on the other hand, a result of automatic processing (Preston 2009). That is, when people are not aware of the topic of the study, they presumably do not control their answers. It is important to note that the distinction between conscious and subconscious language attitudes in sociolinguistics should not be seen as a claim that “subconsciousness” as a psychological phenomenon exists. Theoretically, the distinction in sociolinguistics is rather about two different value systems and awareness of one’s own attitudes; methodologically, it is about how the attitudes are being elicited (Pharao and Kristiansen 2019: 1).

In general, language attitudes are believed to be based on prejudicial beliefs that we learn from social environment (Garrett 2010: 22), i.e. how we imagine the “other” group to be. Sometimes these beliefs are so strong that we can observe social distance. To assess the existence of social distance, Bogardus (1933) asked informants which members of ethnic minority groups they would prefer not to to talk to, work with, live near, or have as family members. Social distance as a term was later expanded by sociologists and social psychologists (Parrillo and Donoghue 2005; Mather et al. 2017, among others) and was applied not only to study ethnic minority groups but also other social groups, such as queer people, religious minorities or people with mental health illnesses. Social distance is hence described as estrangement, lack of acceptance and distancing between various social (including ethnic) groups that is reflected in the lack of willingness to participate in different life situations with the members of the other group (Bogardus 1960).

3. Attitudes towards accents

Various research outside of Lithuania has shown that speech with a non-standard accent tends to be evaluated more negatively than speech with a standard accent (for an overview, see Fuertes et al. (2012)).³ Interestingly, studies have shown that even though speech with a standard accent is generally rated more favorably in comparison to speech with a non-standard accent, attitudes towards different non-standard accents can vary significantly. Some accents may have more positive subconscious language attitudes than others. For instance, Dragojevic and Goatley’s (2022) study using a verbal guise test shows that in the United States, foreign accents from non-stigmatized Western European countries are attributed more status traits (e.g. being intelligent or educated) and solidarity traits (e.g. being friendly or nice) in comparison to speakers with stigmatized accents, such as Arabic, Farsi or Vietnamese.

In Lithuania, research on accent-marked speech has been limited. Čičirkaitė (2019) researched subconscious language attitudes towards four styles of speech typical of inhabitants in Vilnius. Three of the four styles were typically used by “ethnic” Lithuanians, while the fourth was used by Lithuanian-Russian and Polish speakers. A verbal guise test in schools in Vilnius, where the language of instruction was Lithuanian, showed that adolescents allocated traits such as “uneducated”, “poor”, “unsuccessful”, “peasant” (in Lithuanian: ‘kaimietis’), and “does not have a good job” statistically significantly more often to Lithuanian-Slavic speakers than to the other three types of Vilnius speech. These speakers were also more often assigned low-status professions, such as manual laborer, cleaner, or market seller (Čičirkaitė 2019: 139). As Čičirkaitė notes, her study focused on traits that describe a speaker’s social status rather than solidarity or dynamism. Several answers in the open-ended question indicate, however, that Lithuanian--Russian and Polish speakers may be associated with a less interesting and socially less attractive personality, since some participants used adjectives like “reserved”, “weird”, “old-fashioned”, or “uninteresting” (Čičirkaitė 2019: 140).

4. Method

4.1. The verbal guise test

As a method for studying language attitudes, the verbal guise test (sometimes called “the verbal guise technique”) developed gradually. The first method to investigate language attitudes indirectly was “the matched guise test” devised by Lambert et al. (1960). In their experiment, participants in Quebec, Canada, listened to a number of audiotaped speakers (referred to as “masks”) reading the same text in English and French. Following this, participants were asked to evaluate the speakers on various traits, such as intelligence, leadership, and self-confidence. This innovative method eliminated direct questions about the participants’ thoughts on English or French speakers, thereby revealing what Lambert (1967: 94) described as “more private reactions” that might not be socially appropriate to express otherwise. What informants did not know was that one French and one English mask in the original Lambert et al.’s experiment were recorded by the same speaker, that is, “matched”. In that way, researchers could observe how participants described the same person differently when the person spoke a different language.

A matched guise test has its limitations, since the recording of text-reading differs from the natural spoken language tempo. Moreover, research on more than two language varieties becomes challenging, because there may not be many who are fluent in, for instance, four dialects or five accents. Lastly, even though Lambert et al.’s method did not ask participants directly about language, the experiment set with French and English allowed participants to realize that researchers were studying language. Therefore, Lambert et al.’s matched guise test focused on conscious language attitudes. This is why this test was later modified and developed by Kristiansen (1991, 2009) to elicit subconscious language attitudes. In the modified method, called “the verbal guise test”, the stimuli consisted of extracts of spontaneous speech from different speakers. Today, the verbal guise technique is used to study attitudes not only towards different languages but also various linguistic variables or different language varieties, including dialects, sociolects, and accents.

4.2. The stimuli

In this study, the verbal guise test consisted of six male voices, with two voices representing each of the three varieties: (1) speech typical of Vilnius (I will refer to the stimuli as VLN01 and VLN04; see Table 1), (2) Lithuanian with a Lithuanian-Russian accent (RUS02 and RUS05), and (3) Lithuanian with a Scandinavian (Danish and Norwegian) accent (SCA03 and SCA06). The audiotapes were recorded during interviews conducted specifically for the experiment. During the interviews, the men were asked to share memories they had about Vilnius and what they liked about the city.

Table 1. Experimental design

As in any experiment, stimuli must be well-controlled. In a verbal guise test, this means that voice recordings should be similar in length, content and the prominence of accents or other speech traits. In this study, the stimuli were cut and merged from the interviews into 19–24-second-long audio clips. Ideally, one would use an uncut extract from the interview, allowing the stimulus played in the experiment to represent uninterrupted, naturally sounding speech. However, the stimuli for the verbal guise test should not include any marked stylistic expressions or strongly positive or negative phrases, as these may trigger positive or negative attitudes. Therefore, the recordings in this study were edited to avoid any marked expressions while ensuring the final cut sounded as natural as possible.

Nevertheless, the audio clips used in this pilot study had several notable shortcomings, which were discovered after the experiment. Firstly, the evaluation of SCA03 was significantly more similar to those of VLN01 and VLN04 than to the other speakers (I discuss it in more detail in section 5.2). This might be due to the accent not being prominent enough for the study participants to detect; consequently, the results for SCA03 were excluded from the analysis. Secondly, in the RUS02 and SCA06 stimuli, the speakers appeared to sound less fluent than the others, likely due to editing and individual speaking styles. In contrast, both VLN01 and VLN04 can be regarded as very fluent speakers, while RUS05 is considered fluent. Thirdly, the RUS02 stimulus included the phrase “it becomes boring” (“neįdomu tampa”; see Table 2), which may have triggered the impression that the man himself was also boring. Moreover, the RUS02 stimulus exhibited prosodic features typical of the Lithuanian-Russian accent (e.g. stress), but it lacked the most prominent feature of the accent: the so-called lengthening of short vowels /i/ and /u/. It is, thus, unclear if the accent was recognizable enough as a Lithuanian-Russian accent. However, there was no statistically significant difference between the evaluations of the RUS02 and RUS05 speakers; therefore, the results of both RUS speakers were included in the analysis. Lastly, the age of the speakers varied, but I argue that it had a limited impact on the evaluation of the speakers (I discuss age in section 5.2).

Table 2. Content of the stimuli in the order they were played

To further investigate the indexes that different language varieties have, the stimuli were assigned professions. Since it was assumed that VLN and SCA would receive the most positive evaluations, they were allocated low-status professions. As Čičirkaitė (2019) demonstrates, Lithuanian-Russian speakers are less likely to be ascribed high-status professions (such as “doctor” or “businessman”) and more often associated with professions in the service sector or a blue-collar work compared to individuals with Vilnius speech traits (Čičirkaitė 2019: 140). Therefore, RUS speakers in this study were assigned high-status professions to examine if this could reverse attitudes (see Table 1). To ensure that these specific professions accurately reflected adolescents’ perceptions of professional status, surveys from the magazine “Veidas” (2011) and Liubertaitė (2021) were consulted.

4.3. Performance of the experiment

A verbal guise test was conducted in four schools in ethnically unmarked dormitory neighborhoods in Vilnius where the main language of instruction was Lithuanian. In total, 151 adolescents in 9th and 10th grades participated in the study: 68 girls, 72 boys, and 11 pupils who preferred not to disclose their gender. The study was presented as aiming to research how pupils evaluate people based solely on their voices. It was ensured that the words “accent” or “language” were not mentioned to the pupils or their teachers to elicit subconscious language attitudes. The student classes were randomly assigned to two groups: the general group and the profession group. At the beginning of the experiment, the researcher told the participants that all the speakers were male and would talk about Vilnius. Thus, the pupils needed to give their first impression and associations they got about the speakers. In both the general and the profession groups, students listened to the stimuli twice in the following order: VLN01, RUS02, SCA03, VLN04, RUS05, SCA06. They were given pauses in between the recordings to fill in the questionnaire. In the profession group, before playing the stimuli, the students were informed of the alleged profession attributed to each speaker. The profession names were also included in the questionnaire. At the end of the experiment, the researcher gathered the questionnaires and asked the classes in both groups what they thought the experiment was about. When responding to this question, none of the participants mentioned accents, suggesting that subconscious, rather than conscious, language attitudes were elicited.

4.4. The questionnaire

The questionnaire consisted of four (in the profession group) or five questions (in the general group). Although it is rather typical for verbal guise experiments to use a set of trait opposites with a Likert scale, some researchers opt for a questionnaire instead (e.g. Garrett et al. 2004; Čekuolytė 2014). It is said to make the experiment less unnatural and help to better capture the spontaneous impression the study participants get when listening to the guises. Notably, Maegaard (2005) demonstrates that using a questionnaire instead of an evaluation scale with pre-set labels can give a broader and more complex picture of the stereotypes a speaker is associated with. On the other hand, a drawback of using such a questionnaire is the bigger subjectivity of the researcher when analyzing the collected data.

The first four questions in the questionnaire were given both to the general and the profession groups:

1. How would you describe this person? What impression does he give? (‘Kaip apibūdintum šį žmogų? Koks jis tau pasirodė?’)

2. Do you think it would be worth inviting him to your school to hold a class? Explain why. (‘Kaip manai, ar būtų verta šį žmogų pakviesti į jūsų mokyklą pravesti kokios nors paskaitos? Paaiškink, kodėl.’)

3. Would you trust him if he was your neighbor? Explain why. (‘Ar pasitikėtum šiuo žmogumi, jei jis būtų jūsų kaimynas? Paaiškink, kodėl.’)

4. If he would give you and your friends a tour of Vilnius, would it be interesting for you to listen to him? Explain why. (‘Jei šis žmogus tau ir tavo draugams pravestų ekskursiją po Vilnių, ar tau būtų įdomu jo klausytis? Paaiškink, kodėl.’)

The last question was given only to the general group:

5. What could this person work as? (‘Kuo šis žmogus galėtų dirbti?’)

The first question was aimed at getting the first impression of the speakers. The second question is based on the context of a school as a formal, institutional space. Thus, the question is linked to the dimension of superiority, i.e. education and experience. The question is formulated with the phrase “would it be worth” to distinguish the second question from the fourth. While the fourth question is somewhat similar to the second, it focuses on the speaker’s evaluation in less institutionalized settings, such as a tour of the city. Here, the aim is to determine how adolescents assess a speaker in terms of dynamism, specifically whether the person is perceived as interesting and self-confident.

To assess how pupils evaluate speakers in their close environment, the third question focuses on trusting one’s neighbor. Similar questions about whether a person would be willing to live in a neighborhood with a member of a particular social or ethnic group have also been used to study social distance among adults (e.g. Blažytė et al. 2016). As adolescents are unlikely to be concerned with housing choices, the question was reformulated to be about trust. Finally, the fifth question about profession can reveal what social status each speaker is associated with.

To examine statistical significance, the data from answers to the close-ended second, third, and fourth questions were coded numerically from 1 (“no”) to 5 (“yes”). For example, the answer “maybe” was coded as 3, while “probably yes” was coded as 4. Answers to questions 1 and 5, along with additional comments following the yes/no questions 2, 3, and 4 were used as qualitative data to supplement the quantitative data.

5. Results

5.1. General group

Results from the general group show that VLN speakers are evaluated differently in terms of solidarity, dynamism and trustworthiness compared to RUS and SCA06. As observed in previous research (Vaicekauskienė 2017; Čičirkaitė 2019), men with Vilnius speech traits were ascribed both superiority and dynamism traits, resulting in the most positive evaluations in the second and fourth questions (see Table 3). It appears that both RUS and SCA06 were identified as “speakers with an accent”, leading to associations with less interesting and less intelligent individuals, presumably common indexes for any accent-marked Lithuanian language. Although all speakers were assigned many white-collar professions by the pupils, RUS and SCA06 also received more blue-collar job assignments, such as construction worker, mechanic, taxi driver, or grocery store employee.

The study participants, however, seem to distinguish a familiar Lithuanian-Russian accent from an unfamiliar Scandinavian accent, namely, by associating untrustworthiness with Lithuanian-Russian speakers. The answers to the third question show that RUS are associated with a “trustworthy” person statistically significantly less often than VLN. Even though no statistically significant difference was found between SCA06 and the other stimuli, qualitative data from the questionnaire may indicate that the index of a “less trustworthy” person may be ascribed only to RUS speakers. In the answers to the third question, the RUS speakers were described as “aggressive”, “suspicious”, or “weird”:

Such remarks occurred only several times. Nonetheless, neither VLN nor SCA06 stimuli received such comments. When commenting on why they did not trust VLN and SCA06, the participants either did not explain their reasoning or provided general explanations, such as “I don’t trust strangers”. Moreover, it is interesting to note that RUS02 and RUS05 speakers were described as “older” (‘vyresnis’), “middle-aged”, and even “old” 8 and 9 times, respectively. In contrast, VLN01 and VLN04 were described as “young” and “youthful” 21 and 9 times, respectively, even though all men whose voices were used in RUS and VLN stimuli were of the same age. This confirms similar findings in previous research (Čičirkaitė 2014: 24) and may show that pupils do not perceive RUS speakers as being of their peer age, thus distancing themselves from them. Other researchers have argued that speakers with an accent are downgraded in credibility not due to stereotypes but merely because of difficulties in processing different speech (Lev-Ari and Keysar 2010). However, the adjectives gathered from pupils’ answers do not support this claim, since existing stereotypes are directly reflected in the data. The indexes of “less trustworthy”, “aggressive”, “suspicious”, and “sounding older” can be seen as an indication of social distance towards Lithuanian-Russian speakers among Vilnius adolescents.⁵

Table 3. Answers to questions 2, 3 and 4 in the general group. Friedman test and Wilcoxon signed-rank test results. A median closer to 5 indicates that participants tend to answer the question positively. ***, **, * and / indicate statistical significance after Dunn-Bonferroni adjustment: *** mean that p < .001, ** mean that p < .01, * means that p < .05, / means that p > .05.

The attitudes pupils have towards speech with an accent also reflect the fact that Lithuania is a standard language community (Milroy 2001), where the idea of “the best language” exists. Such attitudes can be expected, considering that the Lithuanian language curriculum at school is largely based on prescriptivism (Urbonaitė 2019; Vaicekauskienė ir Urbonaitė 2019). It is also visible in pupils’ metalinguistic comments when they are asked about a lecture at school or a guided tour in the city:

In other words, both in a formal (school) and less formal context (a tour in the city), prescriptivist attitudes play a role: people speaking speech varieties that differ from the perceived standard are seen as less suitable for school and guided tour settings.

5.2. Profession group

How did telling the participants that VLN and SCA06 speakers had a low-status profession and RUS had a high-status profession change the evaluation of the speakers? Firstly, there is no statistical significance in how pupils evaluated VLN speakers in the general group and in the profession group (see Table 4). In other words, even when pupils were led to believe that VLN speakers were a security guard and a bus driver, evaluations in superiority, dynamism and trustworthiness were the same as in the general group. Interestingly, the cashier-SCA06 was evaluated statistically significantly lower than VLN speakers in terms of superiority and dynamism traits. This may be attributed to the unusualness of hearing a bus driver and a security guard – young men with Vilnius speech – speak cohesively about the city. This may explain why, compared to the general group, VLN speakers received significantly more positive adjectives such as “interesting”, “eloquent” (‘iškalbus’), and “friendly”, even with a low-status profession label. In contrast, the cashier-SCA06 did not surprise the study participants; a speaker with an accent in a way confirmed an existing stereotype. These findings further reinforce that Vilnius speech is seen as “the best language”, a de facto standard language in Vilnius (Vaicekauskienė 2017; Čičirkaitė 2019).

It is important to address the question of whether the different ages of men behind the VLN and SCA06 stimuli (see Table 1) had any impact on these results. Here, I come back to the evaluations of the Danish SCA03 stimulus, which, as mentioned previously, was excluded from the analysis. This was due to the SCA03 evaluations showing a similar pattern to the VLN evaluations: in both the general and the profession groups, SCA03 remained highly rated (with a median of 5 for questions 2, 3 and 4). The low-status profession hence did not affect its evaluations, contrary to the Norwegian SCA06. It is therefore assumed that the students did not recognize SCA03 as a speaker with an accent but regarded his speech as characteristic of an older man. Pupils described him as “older” or “middle-aged” twice as often as they did SCA06. If this assumption holds true, the older SCA03 speaker being evaluated as positively as the younger VLN speakers indicates that age had a limited impact on evaluations.

Table 4. General and profession groups: answers to questions 2, 3 and 4 in the questionnaire. Vertically: comparison of evaluations between the three linguistic varieties, i.e. Friedman test and Wilcoxon signed-rank test results. Horizontally: comparison of evaluations in the general group and the profession group, i.e. Mann–Whitney U test results. A median closer to 5 indicates that participants tend to answer the question positively. ***, **, * and / indicate statistical significance after Dunn-Bonferroni adjustment: *** mean that p < .001, ** mean that p < .01, * means that p < .05, / means that p > .05.

When it comes to speakers with a Lithuanian-Russian accent, the high-status profession labels helped RUS speakers to be more often attributed with superiority traits (in the second question). In this context, pupils explicitly mentioned the profession to justify their answer:

However, the high-status professions did not reverse evaluations in dynamism traits. The bus driver and the security guard with Vilnius speech traits were statistically significantly more often ascribed the label “an interesting person” than the doctor or the head of a company with a Lithuanian-Russian accent (in the fourth question). The main reason for positive responses seemed to be the professions the speakers were assigned:

Lastly, profession labels did not affect how participants evaluated the speakers’ trustworthiness when comparing the general group with the profession. Within the profession group, there is no statistically significant difference between VLN, RUS and SCA06.

6. Social distance in subconscious language attitudes

To summarize, the results of this pilot study suggest that Vilnius speech appears to be more often associated with an “intelligent” person (superiority traits) than accent-marked speech. This indicates that the indexes for so-called Lithuanian-Slavic speakers found in Čičirkaitė (2019) may be common to any Lithuanian language spoken with an accent, whether it is a familiar, frequently heard Lithuanian-Russian accent, or an unfamiliar Scandinavian one. This study also shows that men with Vilnius speech traits are perceived as more socially attractive, as they are more often described as “interesting” (reflecting dynamism traits) than men speaking with an accent. However, accents are not evaluated equally. Both Vilnius speech and the Scandinavian accent were associated with a “trustworthy” person, unlike the Lithuanian-Russian accent. This aligns with observations in other speech communities (Dragojevic and Goatley-Soan 2022), where stigmatized accents tend to be evaluated less positively than non-stigmatized ones.

The high-status profession label improved the evaluation of superiority traits, yet speakers’ evaluation did not change in terms of dynamism traits and trustworthiness. Strikingly, the low-status label did not reverse the evaluation of Vilnius speech. In other words, Vilnius speech continues to be indexed with an “intelligent”, “interesting” person despite the unfavorable profession label. Regarding the Scandinavian accent, the low-status profession only reinforced existing indexes. Finally, profession labels did not reverse how pupils evaluated the speakers’ trustworthiness compared to the general group. These results suggest that the social interpretation of Vilnius speech and Lithuanian-Russian accent is fairly consistent across contexts. Thus, the indexes that Vilnius speech and the Lithuanian-Russian accent have appear to be robust in shaping speaker evaluations, even when the social context (profession) should have contradicted them.

Descriptions of Lithuanian-Russian speakers as “less trustworthy”, “aggressive”, “suspicious”, and “older” put into question sociologists’ claim that social distance towards Lithuanian-Russians is absent. Taking into consideration the prominence of linguistic nationalism in the Lithuanian education system (Urbonaitė 2017, 2019), the generally negative representation of Lithuanian-Russians in media (Frėjutė-Rakauskienė 2020), and significant socio-ethnic segregation in Lithuanian cities (Burneika and Ubarevičienė 2016), this study calls for reassessment of the claim that social distance towards Lithuanian-Russians is non-existent. For a long time, sociologists have been concerned with social desirability bias, which may prevent gathering genuine attitudes towards different ethnic, racial, and social groups (Milton and Yamamoto 1998; Parrillo and Donoghue 2005, among others). Two decades of sociolinguistic research on subconscious language attitudes may provide a well-tested methodology and useful insights into the subject, paving the way for a more interdisciplinary view of social distance as a phenomenon.

Nevertheless, these results need to be interpreted with caution, as the Scandinavian accent was represented by only one speaker, and the speakers representing Vilnius speech sounded more fluent than others in the experiment. Moreover, the pilot study used only male voices as guises. Future research could build on these findings by investigating whether female speakers with an accent would be evaluated differently.

Although the question of why such social distance exists is complex and beyond the scope of this paper, I would like to draw attention to one probable explanation. Attitudes towards the Lithuanian-Russian accent do not create social distance on their own, but they can reproduce it. In doing so, these attitudes widen the existing divide between “ethnic” Lithuanians and Lithuanian-Russians, i.e. “us” and “them”. It is important to critically evaluate how Lithuanian language classes at school may contribute to this social distance. As Urbonaitė (2017, 2019) demonstrates, linguistic education in Lithuania mainly aims to teach pupils standard language norms and rarely addresses linguistic diversity. The curriculum is dominated by an ideology of linguistic nationalism (Urbonaitė 2019: 218), where the Lithuanian language is emphatically and directly associated with ethnicity (Urbonaitė 2017: 221). Such an unscientific and ethnocentric understanding of language taught at school (Urbonaitė 2019: 218) can only deepen the existing social distance rather than foster a more inclusive society. Ultimately, it is not accents that are stigmatized but the people behind them.

List of abbreviations

VLN – the Lithuanian language with Vilnius speech traits

RUS – the Lithuanian language with a Lithuanian-Russian accent

SCA – the Lithuanian language with a Scandinavian accent

References

Blažytė, G., V. Pilinkaitė-Sotirovič, K. Žibas. 2016. Visuomenės nuostatos apie socialines ir etnines grupes Lietuvoje: nepakantumo ir socialinės distancijos stebesėna in M. Taljūnaitė (ed.) Lietuvos gyventojų grupių socialinė kaita. Vilnius: Lietuvos socialinių tyrimų centras. 109–124. https://www.lituanistika.lt/content/67545

Bogardus, E. S. 1926. Social Distance in the City. Proceedings and Publications of the American Sociological Society 20, 40–46.

Bogardus, E. S. 1933. A Social Distance Scale. Sociology & Social Research 17, 265–271.

Bogardus, E. S. 1960. Trends in Social Distance in American Life. Social Science 35 (1), 10-16. https://www.jstor.org/stable/23907033

Burneika, D., R. Ubarevičienė. 2016. Socio-ethnic Segregation in the Metropolitan Areas of Lithuania. Sociologicky Casopis / Czech Sociological Review 6, 795–819. https://www.doi.org/10.13060/00380288.2016.52.6.287

Campbell-Kibler, K. 2010. The Effect of Speaker Information on Attitudes Toward (ING). Journal of Language and Social Psychology 29, 214–223. https://www.doi.org/10.1177/0261927X09359527

Čekuolytė, A. 2014. Streetwise, Active and Cool: How Do Vilnius Adolescents Perceive Their Peers’ Linguistic Identity? Taikomoji kalbotyra 6, 1–28. https://www.doi.org/10.15388/TK.2014.17488

Čičirkaitė, R. 2014. Vilnietiško balsių ilginimo socialinė reikšmė. Moksleivių pasąmoningųjų nuostatų tyrimas. Taikomoji kalbotyra 6, 1–30. https://www.doi.org/10.15388/TK.2014.17487

Čičirkaitė, R. 2019. Vilniaus kalbos fonetinių ypatybių socialinės reikšmės: stigmatizuojamas ilginimas ir santykis su žiniasklaidos kalba. PhD dissertation: Lietuvių kalbos institutas. https://www.lituanistika.lt/content/91363

Dragojevic, M., S. Goatley–Soan. 2022. Americans’ Attitudes Toward Foreign Accents: Evaluative Hierarchies and Underlying Processes. Journal of Multilingual and Multicultural Development,43, 167–181. https://www.doi.org/10.1080/01434632.2020.1735402

Frėjutė–Rakauskienė, M. 2020. Etninių mažumų ir pabėgėlių vaizdavimo tendencijos Lietuvos žiniasklaidoje 2005-2017 metais. Lietuvos Socialinė Raida. Socialinių Ir Etninių Mažumų Grupių Įtrauktis Lietuvoje 9. https://vb.lcss.lt/object/elaba:90601701/90601701.pdf

Fuertes, J. N., W. H. Gottdiener, H. Martin, T.C. Gilbert, H. Giles. 2012. A Meta-Analysis of the Effects of Speakers’ Accents on Interpersonal Evaluations. European Journal of Social Psychology 42(1), 120–133. https://www.doi.org/10.1002/ejsp.862

Garrett, P. 2010. Attitudes to language. Cambridge: Cambridge University Press. https://www.doi.org/10.1017/CBO9780511844713

Garrett, P., N. Coupland, A. Williams. 2004. Adolescents’ Lexical Repertoires of Peer Evaluation. Metalanguage: Social and Ideological Perspectives 11, 193–225. https://www.doi.org/10.1515/9783110907377.193

Grondelaers, S., R. van Hout. 2011. The Standard Language Situation in the Low Countries: Top-Down and Bottom-Up Variations on a Diaglossic Theme. Journal of Germanic Linguistics, 23, 199–243. https://www.doi.org/10.1017/S1470542711000110

Holmes, K., T. Murachver, D. Bayard. 2001. Accent, Appearance, and Ethnic Stereotypes in New Zealand. New Zealand Journal of Psychology 30 (2), 79–86.

Kristiansen, T. 1991. Sproglige normidealer på Næstvedegnen. Kvantitative sprogholdningsundersøgelser. PhD dissertation: Københavns Universitet.

Kristiansen, T. 2009. The Macro-Level Social Meanings of Late-Modern Danish Accents. Acta Linguistica Hafniensia 41, 167–192. https://www.doi.org/10.1080/03740460903364219

Kristiansen, T. 2011. Attitudes, Ideology and Awareness in R. Wodak, B. Johnstone, P. Kerswill (eds.) The SAGE Handbook of Sociolinguistics. SAGE Publications Inc. 265–278. https://www.doi.org/10.4135/9781446200957.n19

Lambert, W. E., R.C. Hodgson, R.C. Gardner, S. Fillenbaum, 1960. Evaluational Reactions to Spoken Languages. The Journal of Abnormal and Social Psychology 60, 44–51. https://doi.org/10.1037/h0044430

Lambert, Wallace E. 1967. A Social Psychology of Bilingualism. Journal of Social Issues 23 (2), 91–109. https://www.doi.org/10.1111/j.1540-4560.1967.tb00578.x

Lev-Ari, S., B. Keysar. 2010. Why Don’t We Believe Non-Native Speakers? The Influence of Accent on Credibility. https://www.doi.org/10.1016/j.jesp.2010.05.025

Lietuvos statistikos departamentas. 2021. 2021 m. gyventojų ir būstų surašymas. Retrieved 5th April, 2024 from https://osp.stat.gov.lt/gyventoju-ir-bustu-surasymai1

Liubertaitė, A. (2021). Atskleidė, kokios studijų programos šiemet populiariausios: vienas rodiklis ypač džiugina. Lrytas. Retrieved 5th April, 2024 from https://www.lrytas.lt/lietuvosdiena/svietimas/2021/08/05/news/atskleide-kokios-studiju-programos-siemet-populiariausios-viena-detale-ypac-dziugina-20333171

Maegaard, M. (2005). Language Attitudes, Norm and Gender a Presentation of The Method and Results From a Language Attitude Study. Acta Linguistica Hafniensia 37, 55–80. https://www.doi.org/10.1080/03740463.2005.10416083

Mather, D. M., S.W. Jones, S. Moats. 2017. Improving upon Bogardus: Creating a More Sensitive and Dynamic Social Distance Scale. Survey Practice 10 (4), 1–9. https://www.doi.org/10.29115/SP-2017-0026

Milroy, J. 2001. Language Ideologies and the Consequences of Standardization. Journal of Sociolinguistics 5, 530–555. https://www.doi.org/10.1111/1467-9481.00163

Milton, K., K. Yamamoto. 1998. As the World Turns: Ethno-Racial Distances After 70 Years. Social Science Journal 35, 183–190. https://www.doi.org/10.1016/S0362-3319(98)90038-6

Parrillo, V. N., C. Donoghue. (2005). Updating the Bogardus Social Distance Studies: A New National Survey. The Social Science Journal 42, 257–271. https://www.doi.org/10.1016/j.soscij.2005.03.011

Petrušauskaitė, V. 2013. Etninės ir socialinės grupės Lietuvoje: visuomenės nuostatos ir jų kaita 2013 metais. Etniškumo Studijos 2, 180–191.

Pharao, N., T. Kristiansen. 2019. Reflections on the Relation Between Direct/Indirect Methods and Explicit/Implicit Attitudes. Linguistics Vanguard, 5 (s1). https://www.doi.org/10.1515/lingvan-2018-0010

Preston, D. R. 2009. Are You Really Smart (or Stupid, or Cute, or Ugly, or Cool)? Or Do You Just Talk That Way? in P. Quist (ed.) Language attitudes, standardization and language change: perspectives on themes raised by Tore Kristiansen on the occasion of his 60th birthday. Oslo: Novus forlag.

Prikhodkine, A., D. Correa Saavedra, M. Dos Santos Mamed. 2016. “Give Me Your Name and I’ll Tell You Whether You Speak with an Accent” The Effect of Proper Names Ethnicity on Listener Expectations. CALL - Irish Journal for Culture, Arts, Literature and Language 1 (1), 1–15. https://www.doi.org/10.21427/D7D592

Silverstein, M. 2003. Indexical Order and the Dialectics of Sociolinguistic Life. Language and Communication 23, 193–229. https://www.doi.org/10.1016/S0271-5309(03)00013-2

Urbonaitė, D. 2017. Kalbos samprata kalbinio švietimo politikoje: Lietuvos ir Danijos lyginamoji analizė Daina. Taikomoji kalbotyra 9, 199–299. https://www.doi.org/10.15388/TK.2017.17453

Urbonaitė, D. 2019. Kaip apie kalbą kalba lietuvių kalbos vadovėliai gimnazistams. Taikomoji kalbotyra 12, 182–225. https://www.doi.org/10.15388/TK.2019.17237

Vaicekauskienė, L. 2017. Du sąmonės lygmenys, dvi geriausios kalbos, vienas standartas in L. Vaicekauskienė (ed.) Lietuvių kalbos idealai. Kaip keitėsi geriausios kalbos idėja. Vilnius: Naujasis Židinys-Aidai. 111–142.

Vaicekauskienė, L., D. Urbonaitė. 2019. Gimtoji kalba mokykloje: Vakarų Europos švietimo paradigmų istorija su šiandieninės Lietuvos prieskoniu. Archivum Lithuanicum 21, 233–254. https://www.lituanistika.lt/content/88759

Veidas. 2011. Vaikai nebesvajoja tapti kosmonautais – jie nori būti gydytojais ir verslininkais. Retrieved 4th April, 2024 from https://www.veidas.lt/vaikai-nebesvajoja-tapti-kosmonautais-–-jie-nori-buti-gydytojais-ir-verslininkais-2

Vildaitė, D., K. Žibas. 2010. Etninių tyrimų instituto visuomenės nuomonės tyrimų apžvalga (2005 –2010 m.). Etniškumo studijos 1, 121–137. https://www.lituanistika.lt/content/29489

Zahn, C. J., R. Hopper. 1985. Measuring Language Attitudes: The Speech Evaluation Instrument. Journal of Language and Social Psychology 4 (2), 131–157. https://www.doi.org/10.1177/0261927X8500400203

Submitted May 2024

Accepted September 2024