The role of script in Kyrgyz identity: Examining attitudes toward Kyrgyz script, Cyrillic, and Latinization

Abstract

Presently, Kyrgyzstan remains the last former Soviet Turkic-speaking Central Asian Republic to use Cyrillic script with the possibility of undergoing Latinization. Our study investigates the attitudes of Kyrgyz speakers toward the present use of Cyrillic in Kyrgyzstan and a potential transition to Latin script. To investigate, we conducted a linguistic attitude survey that draws on three themes we predicted to have the greatest impact on speaker attitudes: (1) the fit of Cyrillic for Kyrgyz phonological patterns, (2) the trajectory of de-Russification in Kyrgyzstan, and (3) Kyrgyz globalization and foreign relations. Our results indicate that de-Russification appears to be the most influential factor shaping attitudes toward both Cyrillic and Latin, which suggest that Kyrgyz speakers view written language, or script, as a factor which impacts both the strength and power of Kyrgyzstan as well as opportunities available for Kyrgyz citizens.

Introduction

This study investigates the attitudes of Kyrgyz speakers toward the present use of Cyrillic script and the potentially more long-term process of Latinization of the Kyrgyz script. Currently, Kyrgyzstan remains as one of only two Central Asian countries of the former Soviet Union still officially using Cyrillic script (Altynbayev 2019). In the past two centuries, Kyrgyz has adopted various scripts, beginning with Arabic script in the 19th century and continuing with the Latin, and now, Cyrillic scripts (Abylkasymova 1997). Each of these scripts reflects varying political, religious, and socioeconomic circumstances given that they were adopted at such different times; however, current research has yet to fully examine the ways in which script and orthographic reform, and in particular, use of Cyrillic, is viewed by the speakers today in both the public and private spheres of Kyrgyz society. That is, there is limited, if any, available research on Kyrgyz speakers’ perception of the state of Kyrgyz script today, and what information is available reveals the opinions of only a few select politicians and scholars (RFE 2017; Time News 2023; Radio Free Europe/Radio Liberty 2022). Though government, or more “official perspectives are valuable insofar as they capture the attitudes of the Kyrgyz governing body and language experts, they fail to capture perceptions of everyday speakers themselves (Banks 2020). Today, the issue of script in Kyrgyzstan sits at a crucial crossroads, as it tries to construct a uniquely Kyrgyz identity in the aftermath of the collapse of the Soviet Union, and in a society increasingly focused on hyperglobalization and alignment with the West (Everett-Heath 2003). The role of script in understanding how written language influences perceptions of nationhood, sovereignty, and identity remains relatively understudied (Sebba 2006), and Kyrgyz script and the unfolding trajectory of orthographic reform in Kyrgyzstan remains even less so. Exploring the attitudes of Kyrgyz speakers toward Kyrgyz script thus provides an important window into how written form can shape the ways in which post-Soviet states construct independent identities while remaining under Russian cultural and political influence.

The present research is particularly relevant in applied linguistics with increasing interest in how globalization and multilingualism interact to define linguistic spaces (Jaworski and Li 2020; Lee 2022), and how such spaces negotiate effects of translanguaging (Li 2018). This interest is also compounded by the increasing number of studies examining biliteracy and reading multilingual texts (Domke 2023), what different spellings can encode (Finkelstein and Netz 2023), and more generally, how written language can be codified in relation to its spoken counterpart (Matras 1999). Such literature also crucially hinges on the idea that literacy in a second language can potentially empower individuals and communities (Cook and Bassetti 2005); these themes are also relevant for Kyrgyz speakers in light of growing implications of Latinization, and by extension, de-Russification of Kyrgyz orthography and their identity. The current situation in Kyrgyzstan offers a dynamic lens through which to revisit questions of identity and language as they unfold in real-time. The motivation for this study is therefore two-fold: (1) to bring Kyrgyzstan and the Kyrgyz script into the fold of discussion on how written language and script can be deeply political and provide a key lens through which to explore speakers’ attitudes and forming identities; and (2) to investigate how such attitudes toward the use of Cyrillic and a transition to Latin script are impacted by perceptions of the relationship between Kyrgyz language, culture, and socioeconomic mobility in the post-Soviet space.

Identities do not exist in vacuums; they are shaped by a combination of external and internal factors (Malashenko 2013). This is evidenced by the way in which Soviet language policy has formed the basis for conceptions of Central Asian identity which persist today and impact how the Central Asian countries navigate their identities as sovereign states (Everett-Heath 2003; Fierman 2009). Kyrgyzstan is at once Kyrgyz, Muslim, Central Asian, and post-Soviet (Radford 2014). These identities shape each other, creating an identity that is dynamic and ever-changing. Language and script choices of the Soviet Union, in particular, enabled broad intersecting and overlapping identities, and their many constituent parts, to come to fruition (Pavlenko 2008), which suggests that Soviet script choices had an impact on identity formation in Soviet Kyrgyzstan; as such, we argue that modern script choices will also influence identity formation in an independent Kyrgyzstan. We argue that certain key themes, such as de-Russification, are at play, and that in examining the dynamics of such themes in the context of script and orthography, our findings reveal the significant role script plays in the construal of one’s identity.

The paper is organized as such: (1) in the remainder of this section (“Introduction’), we present relevant terminology and definitions, highlight key findings from previous research, and present our own hypotheses and predictions. In the next section (“Methods”), we introduce our survey used to probe attitudes toward script use, and in the final two sections (“Results” and “Discussion”), we analyze our data through a structural equation model, discuss the findings which crucially reveal important complexities and dynamic interactions in attitudes underlying the role of script in Kyrgyzstan, and present possible next steps in conclusion.

Background

Definitions and terminology

Much of the key terminology is varied and potentially contested. For our purposes, we will use the following terms: orthography and script¹. Script is widely defined as “the graphic form of a writing system” (Coulmas 1999); that is, the visual component of writing that one sees independent of the rules regarding its usage. The primary scripts that will be discussed here are Russian Cyrillic script, Kyrgyz Cyrillic Script, and Latin scripts specific to Central Asia. Orthography broadly refers to the systems that dictate how elements of the script and spoken language correspond to each other (Jaffe et al., 2012). More specifically, orthographies address the graphotactics² and grapheme-phoneme correspondence for a language (Coulmas 1999). An orthography employs a script for a particular language, for example, English orthography employs Latin script. Various orthographies can be adapted using the same script, as is the case with English and Spanish, which both use Latin script despite differences in grapheme-to-phoneme correspondence. It is worth noting that we distinguish between Kyrgyz Cyrillic and Russian Cyrillic, since despite both being Cyrillic scripts, there are differences in graphotactics and grapheme-to-phoneme correspondence of individual letters. We will use Kyrgyz orthography and Russian orthography when referring to the set of rules used by each language to govern their corresponding scripts, such as graphotactics and grapheme-to-phoneme correspondence.

Historical context and current linguistic landscape

The history of Kyrgyzstan as well as its present linguistic landscape are both important backgrounds to issues of orthography. Despite having gained independence over three decades ago, Kyrgyzstan remains a country where Russian cultural, linguistic, political, and economic influence continues to permeate every aspect of society (Kozyrina 2014). This issue stems both from an ongoing power reliance on Russia as well as the historical relationship between the two countries, which fundamentally shaped Kyrgyzstan. However, while Russian influence has been positively associated with security, opportunities for labor migration, and economic clout for Kyrgyzstan, Western influence is associated with economic, professional, and educational advantages for Kyrgyz individuals (Miller and Toritsyn 2005; DeYoung 2010). The degree to which Russian influence is present in Kyrgyz society also has implications for the status of Kyrgyz language and culture. Thus, perceptions of Russian influence in Kyrgyzstan today include both how proximity to Russia provides stable and easily accessible economic opportunities meanwhile political and cultural distance from Russia provides avenues to previously inaccessible opportunities such as access to the global market “somewhere other than as migrant laborers in Russia (DeYoung 2010)” and the bolstering of a strong Kyrgyz identity. Kyrgyz speakers are finely attuned to how proximity to Russia benefits Kyrgyzstan but also how distance from Russia brings benefits as well. Therefore, the advantages and disadvantages associated with Kyrgyzstan’s relationship with Russia make the theme of de-Russifying a focal point in shaping attitudes toward Cyrillic and Latin scripts.

Since the advent of independent Kyrgyzstan, Kyrgyz orthographic policy has alternated between pushing for de-Russification and advocating for a permanent return to the original 1953 Kyrgyz orthography implemented by the Soviet Union (Odagiri 2015). There remains a debate as to whether Kyrgyz orthography should be modified and standardized to assimilate linguistic elements of Russian origin into Kyrgyz phonological patterns or continue to allow Russian loanwords to operate in accordance with Russian orthography (Odagiri 2015; Dietrich 2005). Recently, tensions have escalated between Russia and Kyrgyzstan over the status of Cyrillic script in Kyrgyzstan. This is due to mounting pressure from Kanybek Osmonaliev, the chairman of Kyrgyzstan’s National Commission for State and Language Policies, who advocates for transitioning Kyrgyz script to Latin despite strong opposition from Kyrgyz President Sadyr Japarov (Switch to Latin-based alphabet raised in Parliament of Kyrgyzstan 2023; President Japarov reprimands Chairman of Kyrgyz Language Commission for statements about switch to Latin characters 2023. The fallout of such disputes has directly impacted Kyrgyz foreign relations, most recently in Russia’s decision to suspend the import of dairy products from Kyrgyzstan following Osmonaliev’s announcement of his enthusiasm for Kyrgyzstan to begin preparations to switch Latinize Kyrgyz (Time News 2023).

Presently, the Kyrgyz linguistic landscape displays a strong reliance on Russian, despite widespread Kyrgyz language understanding and an affinity for English. In the capital city of Bishkek, Russian dominates, even in bilingual Russian-Kyrgyz and Russian-English signage where it usually holds the primary position (McDermott 2017). Meanwhile, English, though not an official language, is viewed as prestigious and sophisticated and is increasingly used in signage, especially in business districts. By contrast, Kyrgyz signage, despite its equal status in legislation with Russian, is slightly more prevalent than English but occupies far fewer primary positions in bilingual Kyrgyz-Russian and Kyrgyz-English signage (McDermott 2017). This is due in part to Kyrgyz lacking historical dominance since its annexation to the Soviet Union and its associations with upward mobility. However, even when combined, Kyrgyz and English signage still fall behind that of Russian. Figure 1 illustrates further the linguistic landscape and exemplifies the mixed and dynamic use of scripts as previously described.

Thus, as shown by the figures, Bishkek’s linguistic landscape, while influenced by Kyrgyz and Western elements, remains overwhelmingly Russian-dominated.

As part of the linguistic landscape, it is also important to note some of challenges that arise from using the current Cyrillic orthography to write the Kyrgyz language. Vowel harmony, for example, is an important phonological process that also affects the process of suffixation in Kyrgyz, and as an agglutinative language, suffixation is key to forming new words (Abylkasymova 1997; see Lamont and Washington (2019) for consonant harmony³). The final vowel of the stem word determines the class of the subsequent vowel, and following suffixed vowels are largely determined by their preceding vowel (Washington forthcoming). Issues arise, however, in the suffixation of Russian loanwords, which constitute a significant part of the modern Kyrgyz lexicon. The use of Cyrillic presents a specific challenge with these loanwords because Kyrgyz orthography prioritizes preserving Russian spellings rather than accurately representing Kyrgyz pronunciations. This can be observed in the Russian loanword for “tractor’. In its written form (“трактор”), the presence of the orthographic rounded vowel /о/ in the second syllable of the word requires subsequent suffixes to include a rounded vowel. Despite this, however, “трактор” is pronounced with an unrounded vowel in accordance with the Russian orthographic and pronunciation rules that loanwords into Kyrgyz adhere to (Washington forthcoming). Consequently, when adding suffixes to “трактор” in spoken Kyrgyz, unrounded vowels are used instead of rounded vowels. This proves problematic because while suffixes in writing maintain rounded vowels consistent with (orthographic) vowel harmony, the spoken form does not abide by the same pattern, thus Kyrgyz orthographic rules for vowel harmony do not match those of the modern spoken language.

Previous studies

To the best of our knowledge, virtually no research has been conducted on script reform in Kyrgyzstan, and the broader literature on the relationship between script and identity-formation is limited; however, we draw on two seminal studies: Papapavlou (1997) and Kadirova (2018). Papapavlou (1997) investigates how speakers associate their language”s orthographic representation with national identity and educational experiences by eliciting attitudes toward a series of proposed orthographic changes. He presents possible modifications of the orthographic representation of Modern Greek using a 20-question survey distributed to students enrolled at the University of Cyprus. The survey was designed to measure participants” feelings toward orthographic changes and how those changes could impact (a) reading comprehension, (b) the “character” of the language, (c) the structure and semantics of the language, (d) the link between Modern and Ancient Greek, and (e) learning Greek as a foreign language. Papapavlou found that respondents’ concerns about orthographic reform were largely influenced by the perceived relationship between orthography and Greek national identity and history. His findings show that Greek speakers anticipate that orthographic reform would adversely affect the “character” of the Greek language. Speakers believed that a negative impact on the Greek language would therefore cause an estrangement between authentic Greek roots and modern Greek identity. Papapavlou (1997) argues that although a certain amount of modification is tolerated, a major change in a language’s orthographic representation invites questions about what motivated that decision and how speakers respond to and hopefully, embrace the change. Though Papapavlou’s findings point to orthography as holding a significant weight in constructing a society’s identity, they still leave questions regarding what aspects of identity are impacted by script unanswered. In the case of Kyrgyz, Cyrillic’s primary association with Soviet, rather than Kyrgyz, identity raises questions of whether a change to Latin script would impact modern Kyrgyz identity.

Kadirova (2018) analyzes the sociolinguistic attitudes of Kazakhs toward the Latin alphabet and orthographic reform in Kazakhstan using a combination of expert interviews and a survey designed for the general population. The survey is composed of five sections, two of which are demographic and three of which focus on different aspects of the linguistic landscape in Kazakhstan using a Likert scale. The demographic sections elicit sociolinguistic and socioeconomic information. The remaining three sections address attitudes toward the Latin alphabet and predictions about the future of Latinization in Kazakhstan. The results of her survey indicate that Kazakh speakers overwhelmingly support Latinization for the purposes of De-Russification and to continue their technological and scientific development through the Latin alphabet. Furthermore, her results indicate that Kazakh speakers are not wrestling with the benefits and drawbacks of Latinization; rather, they are actively looking for the most productive way to carry out this reform. Although her focus is on speakers’ attitudes toward orthographic reform given the linguistic landscape, her study invites important questions regarding how the current internal and external socioeconomic and political dynamics impact such attitudes. Kadirova’s (2018) findings are especially relevant as they illustrate not only how informative initiating discussion with the general population can be, but also crucially, that the success (or the failure) of a script is deeply intertwined and dependent on speakers’ attitudes.

Hypothesis and predictions

Building on the findings of Kadirova and Papapavlou, the present study probes comparable questions but in the context of Kyrgyzstan. In particular, we examine Kyrgyz speakers’ attitudes toward Latinization of Kyrgyz script and the current use of Cyrillic. To achieve this, we constructed a theoretical model composed of the following three themes: (1) Cyrillic Fit, (2) De-Russification, and (3) Globalization.

Cyrillic Fit addresses the attitudes toward the suitability of Cyrillic for Kyrgyz’s phonemic inventory and phonological patterns. This theme provides insight into the importance Kyrgyz speakers place on the impact of Russian on Kyrgyz orthography and the suitability of Cyrillic for Kyrgyz phonological patterns. The second theme, De-Russification, concerns the attitudes toward the influence of Russia and Russian on Kyrgyz culture and the importance of Latinization for de-Russifying Kyrgyz language and culture. Latinization has value insofar as it distances Kyrgyzstan and Kyrgyz culture from Russia and Russian culture. Thus, this theme sheds light on how culturally distanced Kyrgyz speakers see Kyrgyzstan from Russia and whether proximity to Russia is a concern to them. The third and final theme, Globalization, pertains to attitudes regarding the possible implications of Latinization or continued use of Cyrillic on Kyrgyzstan’s globalization and opportunities for Kyrgyz individuals. This set of questions encompasses topics such as whether Latinization facilitates English language learning, increases professional and educational opportunities abroad, and brings Kyrgyzstan closer to Europe and the West.

These themes were chosen based on two factors. We first drew from the primary themes identified by Kadirova (2018) as contributing to perceptions of script. She identifies De-Russification and Globalization as widely impacting attitudes toward script in Kazakhstan and introduces the idea of Cyrillic Fit as a less important but still contributing theme. We extend these themes to our study and consider how they will apply to a Kyrgyz context. Presently, the Kyrgyz government’s primary aims regarding the balance between its national identity and transnational opportunities are two-fold: (1) to promote a cohesive, strong, and independent national identity and image via some degree of De-Russification while negotiating its current economic and political dependence on Russia (Malashenko 2013), and (2) to begin expanding Kyrgyzstan’s global reach (Petric 2015). Discussions surrounding the effectiveness or suitability of Cyrillic for Kyrgyz are mediated by those two goals, from which we theorize that Globalization, De-Russification, and Cyrillic Fit (i.e. whether Cyrillic is well-suited for the Kyrgyz language) would be the main factors shaping attitudes toward Cyrillic and Latin. We argue that whether a script is well-suited for a language will be explicitly tied to identity-building.

Based on these themes, our predictions are as follows. We anticipate that the favorability toward Globalization will reflect a desire to de-Russify the Kyrgyz language and culture. We predict that certain speakers will see decreased Russian influence as a means through which Kyrgyzstan can strengthen its relationships with countries beyond the post-Soviet sphere as well as elevate the status of Kyrgyz within Kyrgyzstan. Furthermore, certain Kyrgyz speakers have also been raised and educated in a political landscape in which they have seen neighboring countries undergo the Latinization process to de-Russify their countries and cultivate an identity removed from that of the colonial oppressor. Seeing the varying levels of success and ways that these script reforms have shaped the national identities of Kazakhstan, for example, as corroborated by the findings of Kadirova (2018), will likely influence certain Kyrgyz speakers to want to proceed in a similar direction. Therefore, they will view Latinization as an important step for Kyrgyz identity development.

We also predict a negative relationship between attitudes toward Cyrillic Fit and those toward De-Russification. Speakers who view Cyrillic script as unfit for Kyrgyz will also exhibit positive attitudes toward De-Russification. Cyrillic script validates the place of Russian phonemes, orthography, and script in the Kyrgyz language (Odagiri 2015). Accordingly, positive sentiment toward De-Russification indicates an aversion toward Cyrillic altogether and consequently, Cyrillic Fit specifically. Speakers who view Cyrillic as unequipped to represent Kyrgyz are likely to view Latinization positively. In addition to potentially alleviating any orthographic problems associated with Cyrillic, Latinization is associated with the West and therefore, with Globalization. As such, there should also be a negative relationship between attitudes toward Cyrillic Fit and attitudes toward Globalization.

However, as Kyrgyzstan attempts to globalize, it must also forge an identity removed from that of Russia’s. While Kyrgyzstan maintains territorial sovereignty, its economic position, governance structures, social norms, and political representation are inextricably linked with Russia and the Soviet legacy (Ortmann 2018). The incompatibility of Cyrillic script with Kyrgyz is in itself another function of the power of Russia over Kyrgyzstan because it facilitates the inclusion of Russian loanwords and phonological processes into Kyrgyz rather than prioritize Kyrgyz words and phonological processes. Thus, the relationship between Cyrillic Fit and Globalization is entirely impacted by how Kyrgyz speakers view Russian influence as impacting their ability to globalize and how Russian influence must be addressed to globalize Kyrgyzstan.

Therefore, we predict that (H1) favorability toward Globalization will be positively correlated with a desire to de-Russify Kyrgyz language and culture; (H2) viewing Cyrillic as unfit for Kyrgyz will be correlated with a greater desire to de-Russify; (H3) Cyrillic Fit and Globalization will have a negative relationship due to the representation of Latin script as a solution to the orthoepy problems associated with Cyrillic; and (H4) the relationship between Cyrillic Fit attitudes and Globalization attitudes is fully mediated by attitudes toward De-Russification. That is, the relationship between Cyrillic Fit and Globalization is hypothesized to be an indirect effect that exists because of the influence of attitudes toward De-Russification. To summarize, we present an abstract model in Figure 2 to illustrate our predictions on how the three themes influence one another and possibly shape attitudes toward script. Lines connecting the boxes indicate predicted relationships between the themes. Our model shows that we anticipate positive attitudes toward De-Russification to mediate the relationship between Cyrillic Fit and Globalization.

Methods

To probe participants’ attitudes toward Latinization, we conducted a survey similar in structure to Kadirova (2018) and Papapavlou (1997). Eight questions regarding Globalization and the impact of Russian influence on script were adapted from Kadirova (2018) and reworded to fit a Kyrgyz context. The wording of questions regarding the integration of loanwords and the “fit” of a script was inspired by the Papapavlou (1997) questionnaire. Though the scope of Kadirova (2018) and Papapavlou (1997) were different, both studies addressed themes of national and individual identity. Thus, we used the questions concerning identity from both study questionnaires as models, taking note of key words and ideas, when formulating novel questions designed to address aspects of De-Russification, Globalization, and Cyrillic Fit that the original questions did not fully capture. Additional questions regarding Cyrillic Fit were inspired by the orthography and grapheme-specific questions from Kadirova but were designed to cover specific topics relevant to Kyrgyz orthography. This study received the relevant IRB approval and was performed in accordance with the declaration of Helsinki (World Medical Association 2001).

Materials

The survey was developed to examine the links between Globalization, De-Russification, and Cyrillic Fit. The survey was originally written in English before being translated into Kyrgyz by a bilingual Kyrgyz-English translator. The translation was then reviewed by another Kyrgyz-English translator. Any disagreements were discussed by both translators before a final translation was approved. The survey uses 14 nondemographic questions to elicit responses pertaining to the three themes discussed above. Each theme is represented by several questions distributed throughout the survey; a full breakdown of the questions and their corresponding codes can be found in the Appendix. Responses to the questions were gathered along a 5-point Likert scale, from Strongly Disagree to Strongly Agree.

Procedure

Recruitment was conducted both online and through Kyrgyz contacts in Bishkek, Kyrgyzstan. E-mails were sent to the student body at the American University of Central Asia and the Kyrgyz Russian State University to advertise the study. The study was also advertised and accessible online through social media. In Bishkek, two university students and one Kyrgyz teacher canvassed the city on foot. The recruitment team made the survey available in two formats to passersby: participants had the option of accessing the online version using link sharing or via paper copy supplied by the recruiter. The in-person recruitment team stationed themselves in micro districts 3, 4, 5, and 6 of the city to survey a broader range of demographics. Participants first provided consent before completing the survey, which took about 10–15 minutes to complete. Participants were told that participation was voluntary and that they could withdraw at any point, or skip questions they were uncomfortable answering, without consequence. Other than requiring consent to participate, no other questions required a response. This was a choice made deliberately as laws restricting free speech and expression in Kyrgyzstan, discouraging people from voicing opinions counter to those accepted by the governing body, would potentially affect participants’ willingness to partake in the study. Had we required responses to all survey questions, we would have risked not attracting sufficient participants to conduct a robust analysis. Participants were then entered into a raffle for a chance to win a gift card valued at 4000 Kyrgyz som, equivalent to 47 USD at the time (February 2023).

Participants

Data were collected from a group of 330 participants whose ages ranged from 18 to 35 and 50 to 70. Prospective participants were informed that only those within the aforementioned age ranges were eligible for the study. Prospective participants whose ages ranged from 36 to 49 were not permitted to complete the survey. We purposefully did not collect data from the in-between age group because of their potentially mixed viewpoints, having lived through a comparably politically mixed time period, whereas participants in each of the two aforementioned age groups represent two upbringings that are distinctively different in terms of Russian influence and therefore have a deeper cultural and educational divide. Perceptions of identity and culture are developed during one’s youth (Jensen, Arnett and McKenzie 2011). Thus, we expect that the starkly different upbringings experienced by both generations will have shaped their identities in such a way that will affect their attitudes toward Latinization and Cyrillic.

Participants completed the study either remotely via a shared link, or in person orally, and their answers were recorded by hand by the researcher on a printed paper survey. Participants were excluded from further analysis for skipping more than 10 items in the survey, giving a final sample of 276. The mean age of all participants who indicated their age (187 participants, 68% of the total sample) was 44.44. Both male and female participants were recruited (though only 226 participants, 82% indicated their gender), with 54% of participants identifying as female and 28% identifying as male. Participants who did not indicate their age in the survey were excluded from the age-moderated model.

Prospective participants were informed prior to commencing the survey that eligibility was restricted to Kyrgyz speakers only, this included ethnically non-Kyrgyz participants who spoke Kyrgyz.

Results

Before further statistical analysis, we manually recoded the responses to standardize for ease of analysis (e.g. assigning a numerical value of 1–5 for each response to account for discrepancies in capitalization and spelling errors due to manual coding). The data from the 276 participants remaining after data cleaning were analyzed in R (R Development Core Team 2018). The Lavaan package (Rosseel 2012) was used to conduct structural equation modeling analyses.

Table 1 outlines the demographics of our participants. The majority of participants indicated that they use Kyrgyz at home (n = 171, 63% of the 271 participants who answered this question). The participants also indicated they use Kyrgyz at university (n = 18), at work (n = 32), in public spaces (n = 34), and in public institutions (n = 33).

The distribution of participant responses for each theme item is presented in Figure 3. Participants had generally positive attitudes toward Cyrillic Fit (C1), negative attitudes toward De-Russification (C2), and mixed attitudes toward Globalization (C3). In the Globalization category, questions where a majority of the participants exhibited positive attitudes were restricted to questions pertaining to the effect of Latinization on English-speaking ability and professional and educational opportunities abroad for Kyrgyz speakers. One question in the Globalization category elicited the most negative sentiments overall, where the respondents disagreed that transitioning to the Latin script would decrease the economic and social dependence on Russia. The remaining Globalization questions did not show a particular positive or negative preference. It also appears that in general, the participants do not have particularly negative attitudes. Even for the De-Russification category, where responses skew negative, the negative attitudes do not constitute the majority of the responses.

Analysis

The descriptive statistics, reliability coefficients, and correlations among the themes are presented in Table 2. One item (C2Q3) showed a low agreement with its theme (De-Russification). Examining this item from a qualitative point of view suggested that the item was ambiguously worded and could be interpreted in both a positive or negative way. We tested the fit for this item in the theme both positive coded and reverse coded. Cronbach’s (Cronbach 1949) alpha coefficient scores for De-Russification indicated that this item should be removed from this theme (⍺_{reverse_coded-included} = 0.37, ⍺_{positive_coded-included} = 0.63, ⍺_excluded = 0.68). Therefore, this item was removed from further analysis. After removal, composite reliability measures for each theme were all above 0.7, indicating good within-theme reliability.

All correlations were significant at P < .001. Correlations between Cyrillic Fit (CF) and De-Russification (DR) and Globalization (GL) were weak (r = 0.28 and r = 0.22, respectively), whereas the correlation between De-Russification and Globalization was very strong (r = 0.80). The distribution of participant responses (Fig. 3) suggested an ordinal model approach. We fitted three structural equation models: a numeric mediation model, an ordinal mediation model, and a moderated mediation model (where the model was grouped by Age) to investigate our proposed mediation (Table 3).

The fit statistics showed the ordinal mediation model had an acceptable fit and the best fit statistics overall (χ² = 180.85, df = 62, CFI = 0.98, TLI = 0.98, AGFI = 0.96, SRMR = 0.08), whereas the numeric model and the age-moderated model did not have acceptable fit (the numeric model did not achieve adequate fit for CFI or TLI, the age-moderated model did not achieve adequate fit for SRMR). We therefore selected the ordinal mediation model for our analysis based on its superior fit.

Figure 4 depicts our full model. Factor loadings of 0.7 or greater indicate strong relationships and factor loadings ranging from 0.3 to 0.7 indicate moderate relationships. According to the results, all questions have a strong or moderate positive relationship to the theme they address, supporting the construct validity of the thematic dimensions (in Fig. 2).

The paths in the model depicted in Fig. 4 show there is a strong positive relationship of 0.80 between attitudes toward De-Russification and Globalization, a moderate positive relationship of 0.28 between Cyrillic Fit and De-Russification, and a very weak negative relationship of −0.00 between Cyrillic Fit and Globalization. These results are consistent with our prediction that attitudes toward De-Russification predict attitudes toward Kyrgyz Globalization (H1). We performed Sobel tests (Sobel 1982) to examine the mediating effect of De-Russification on the link between Cyrillic Fit and Globalization. The total effect (the link between Cyrillic Fit and Globalization, without considering De-Russification) was significant (z = 3.05, P = .002), while the direct effect (the link between Cyrillic Fit and Globalization considering the effect of De-Russification) was not significant (z = −0.55, P = .956). The mediating effect of De-Russification was significant (z = 3.42, P = .001). Therefore, our analysis indicated that De-Russification fully mediated the link between Cyrillic Fit and Globalization (H4). A graphical summary of the Sobel tests is shown in Fig. 5.

We hypothesized a positive relationship between Globalization and De-Russification (H1). This was confirmed by our results. We hypothesized negative relationships between Cyrillic Fit and De-Russification (H2) and Cyrillic Fit and Globalization (H3). These hypothesized relationships were not found in our results. In addition, the link between De-Russification and Cyrillic Fit was positive, the opposite of our prediction. When considering the total effect (Fig. 5), the link between Cyrillic Fit and Globalization was also significantly positive, the opposite of our prediction. Before further interpretation, it is important to note a possible ambiguity concerning C2Q3. This item was excluded from analysis due to potential ambiguity in the phrasing of the question, and how it was perceived by participants. The exclusion does not yield drastic changes to the output or model fit, but it does improve the Cronbach’s alpha for this particular item (from 0.63 to 0.69), which we view to be a well-motivated improvement.⁴

Discussion

In the present research, we investigated the attitudes of Kyrgyz speakers toward the Latinization of Kyrgyz script and the current use of Cyrillic script for Kyrgyz. Despite having gained independence in 1991, Kyrgyzstan remains the last former Soviet Turkic-speaking country in Central Asia to use Cyrillic script for its titular language. Given the rapid globalization of Central Asia and Russia’s intervention and influence in the region, the trajectory of Kyrgyzstan’s script policies remains uncertain, generating tension between various governing bodies and officials (AKIpress 2023). As the issue of script has become more pressing in recent years, the need to understand the attitudes of Kyrgyz speakers toward Cyrillic and Latin scripts is more important now than ever. Should Kyrgyzstan one day choose to Latinize, surveying the attitudes of Kyrgyz speakers now could be indicative of the success of the reform.

We hypothesized that three themes regarding the suitability of Cyrillic (Cyrillic Fit), de-Russification (De-Russification), and globalization (Globalization) would be key predictors for attitudes toward scripts among Kyrgyz speakers. The theme of Cyrillic Fit relates to attitudes about how suitable Cyrillic is for Kyrgyz’s phonemic inventory and phonological patterns. The theme of De-Russification concerns attitudes about the degree to which Russian influence affects Kyrgyz culture, language, public affairs, and international relations. And the theme of Globalization encompasses the perceived implications of retaining Cyrillic script or transitioning to Latin on Kyrgyzstan’s globalization trajectory and on development of economic, professional, and educational opportunities for individuals. We predicted a positive relationship between De-Russification and Globalization, and negative relationships between Cyrillic Fit and Globalization, and Cyrillic Fit and De-Russification. A positive relationship is one in which as attitudes toward one theme become increasingly positive, so too do attitudes toward the other theme. A negative relationship refers to a link where attitudes toward one theme become increasingly positive, whereas attitudes toward the other theme become increasingly negative.

Our findings suggest that speaker attitudes toward Cyrillic and Latin scripts are predominantly shaped by perceptions of De-Russification due to De-Russification’s role as the mediating variable between the three themes. Attitudes toward De-Russification affect speakers’ beliefs toward Globalization and the link between Globalization and Cyrillic Fit. De-Russification completely mediates the link between Globalization and Cyrillic Fit; in other words, any link between attitudes toward Globalization and Cyrillic fit is completely explained by attitudes toward De-Russification.

Regarding the link between De-Russification and Globalization, we had hypothesized that attitudes toward De-Russification would positively predict attitudes toward Globalization. This was borne out by our results where there was a strong positive relationship of 0.80 between De-Russification and Globalization. As explained above, the degree of Russian influence in Kyrgyzstan has implications for what international opportunities are available to Kyrgyz people. Thus, De-Russification can be seen as a path to Globalization, therefore yielding a positive link between the themes.

Furthermore, while there’s an argument to be made that overwhelming dependence on Russia threatens Kyrgyz culture, there is an equally strong argument that the dilution and possible demise of non-Western cultures is set in motion by Globalization (Ben-Rafael and Ben-Rafael 2018). In light of this discourse, the relationship between Globalization and De-Russification provides insight into how Kyrgyz speakers anticipate interconnection across borders to affect the unique and defining qualities of Kyrgyz culture and identity. Kirmse (2010) found that individual identities in Kyrgyzstan are shaped by a specific geographical and historical context and engagement with cultural Globalization. That is, identities and attitudes toward the world are constrained by the individual’s background and past and present local, regional, and national politics and institutions (Kirmse 2010).

The positive relationship between De-Russification and Globalization complements Kirmse’s findings. Attitudes toward De-Russification strongly predict attitudes toward Globalization and vice-versa because the themes shape each other. Perceptions of Globalization in Kyrgyzstan cannot be divorced from Russian influence in Kyrgyzstan. There is a battle for influence in Kyrgyzstan. As the West tries to leverage its position in Kyrgyzstan via the entrance of Western institutions, Russia is intent on staving off Western influence and cementing its place as the most influential force in the country (Malashenko 2013). Thus, participants’ understanding of Kyrgyzstan’s Globalization is naturally shaped by how they perceive Russia to be impacting the trajectory of Kyrgyz Globalization.

Concerning the link between Cyrillic Fit and De-Russification, we hypothesized that attitudes toward Cyrillic Fit would negatively predict attitudes toward De-Russification. Yet, not only did we not find evidence of this hypothesis, but we also found that attitudes to Cyrillic Fit significantly positively predicted De-Russification, as shown in Fig. 5. We would expect to see that the more a speaker feels positive about Cyrillic’s capacity to represent Kyrgyz, the less inclined they would be to de-Russify. However, our results indicate that speakers who feel positively about Cyrillic also feel positively about De-Russification. At the moment, we cannot decisively explain this result, but a possible theory emerges. It is not clear to what extent the everyday Kyrgyz speaker is aware, if at all, of how Cyrillic impacts the mechanisms of their language. Nevertheless, it is possible that participants view acknowledging the efficiency of Cyrillic as separate from supporting the use of Cyrillic.⁵ This is explained by the idea of Cyrillic Fit encompassing both the ability of Cyrillic to represent Kyrgyz suitably but also how such a script can fit within the image of a unique and independent Kyrgyz culture, identity, and nation.

Lanza and Woldemariam (2014) argue that language planning reinforces social hierarchies and establishes the linguistic practices of the dominant group as standard. Script choice, as a facet of language planning, rests within this claim. The use of Cyrillic upholds the position of Russian phonemes, orthography, and script in Kyrgyz. For example, despite constituting a large part of the Kyrgyz lexicon⁶, Russian loanwords can be challenging to write because the current orthography dictating the use of Kyrgyz Cyrillic script prioritizes Russian spellings for Russian loanwords which leads to a discrepancy between orthography and pronunciation. This would explain why it may be that speakers who are unaware of the difficulties Cyrillic can potentially present could believe that Cyrillic is suitable for Kyrgyz and still choose to support De-Russification.

In such a case, it is plausible that believing in the suitability of Cyrillic does not negate one’s beliefs that Cyrillic still represents Russian influence, and therefore, De-Russification is important, nonetheless. The positive relationship between Cyrillic Fit and De-Russification (shown in Fig. 4) may be credited in part to an unawareness of how script choice impacts language. Writing their language in a script other than the one that has been used for upwards of 70 years may be difficult for Kyrgyz speakers. In the absence of knowledge on the impact of Cyrillic on Kyrgyz, it is possible that the association of Cyrillic script with Russia factors heavily into a speaker’s consideration of Cyrillic Fit, yielding the positive link between De-Russification and Cyrillic Fit. Such a finding is also in keeping with previous scholarship attesting to the role of Cyrillic in linguistic imperialism whereby the spread of Cyrillic has been historically linked with spreading Russian imperialism (Rannut 2021; Warner 2022).

Regarding the link between Cyrillic Fit and Globalization, we hypothesized that attitudes toward Cyrillic Fit would negatively predict attitudes toward Globalization. This was not immediately evident from our data. In fact, when the mediating effect of De-Russification is not considered (as evidenced by the total effect in Fig. 5), there is a positive relationship between Globalization and Cyrillic Fit where the more positive a speaker felt toward Globalization, the more confident they felt in Cyrillic’s ability to represent Kyrgyz. This is the opposite of what we should expect. However, introducing De-Russification back into the model with Cyrillic Fit and Globalization yields a negative (albeit very weak and nonsignificant) relationship. This suggests that while a link between Globalization and Cyrillic Fit exists, it is an indirect link mediated by De-Russification.

Attitudes toward De-Russification explain the relationship between Globalization and Cyrillic Fit. Backhaus (2006) points to the linguistic landscape as being a marker of ethnic identity, language, and place. Writing declares power over a space (Spolsky and Cooper 1991). In light of this, the presence of Cyrillic as the defining feature of the Kyrgyz linguistic landscape points not only to the power Russia exerts on the country (Kozyrina 2014) but also, to a Kyrgyz identity constructed within a Russian frame of reference (McDermott 2017). That is, given the use of Cyrillic, Kyrgyz is subsequently classified within a specific class of languages that perhaps would not have been the case otherwise if not for Cyrillic. Thus, De-Russification is not only the mediating theme but also indeed the dominant theme for understanding Kyrgyz attitudes toward Kyrgyz script because the issue of Russian influence and De-Russification affects every aspect of the social, cultural, economic, educational, and political reality in Kyrgyzstan. Without factoring De-Russification into the link between Globalization and Cyrillic Fit, the relationship is disjointed.

All three themes shape a speaker’s attitude toward script by virtue of impacting each other. Our results indicate that attitudes toward De-Russification are necessary for shaping attitudes toward and relationships between Cyrillic Fit and Globalization. Most essentially, De-Russification holds the greatest weight in the formation of overall attitudes toward Latin and Cyrillic. By providing an initial understanding of such themes and how they influence each other, our study makes an important contribution: It shows that script encompasses not only identity but also feelings of economic, political, cultural, and educational freedom and opportunity. That is, our model indicates that the perceived influence of the dominating group on the subordinate group’s culture, language, and international and public affairs colors perceptions toward the suitability of a script for (1) the phonemic inventory and phonological processes of a language, and for (2) furthering the aspirations or maintaining the stability of the individual and society. In this way, attitudes toward script are also representative of both the current reality or identity of a group or individual but also the ambitions of a speaker or speaker group. These findings suggest that scripts yield a lot of power in determining the trajectory of individual and collective identity-building.

Potential limitations and directions for future research

One possible limitation in our study concerns cultural reasons surrounding censorship in Kyrgyzstan that made it difficult to implement forced responses on the questionnaire, which in turn limited data collection. A robust examination of demographic factors influencing the links between our themes was thus not possible. Nonetheless, we believe the present study offers important patterns and insights into the potential trajectory Kyrgyzstan may assume as attitudes toward script choice become increasingly apparent.

There are some additional considerations that should be taken into account for future research. While the CFA results indicated that the questions fitted into their original themes, a high correlation was found between the themes of De-Russification and Globalization. This could suggest that these two themes could be combined into one factor. Further research is necessary to explore alternative thematic interpretations and strengthen the overall validity of the constructs. For example, with a larger sample size it would be possible to split the data set and run measurement invariance tests. We tested the fit of three proposed models against standard model fit indices (Table 3). The Ordinal Mediation Model was the only model that passed all parameter checks. The Age-Moderated Mediation Model and the simplified numeric model failed on various fit parameters. The weak fit of the Age-Moderated Mediation Model may be due in part to disparities in how people might record their age. Despite the limitations of implementing a nonforced response questionnaire format, we believe this was the most appropriate method for this study which also respected the laws and norms in Kyrgyzstan. As there were many missing responses for Age, imputation techniques could not be used. Though future research in Kyrgyzstan may be similarly constrained by censorship laws, we believe a larger sample size could yield enough age responses to provide an acceptable Age Moderated Mediation Model. Another limitation of this study is that we used a cross-sectional survey design. Cross-sectional designs do not generally have diagnostic power for causal relationships (even though mediation analysis is often interpreted with causality in mind), and caution should be taken in inferring causality between the associated themes in our model. In addition, the associations between Cyrillic Fit, De-Russification, and Globalization may be influenced by residual confounding from unmeasured factors that were not examined within our survey. For example, there may be a common effect of anti-Imperialism or desire for decolonization on both De-Russification and Cyrillic Fit.

Based on the findings of this study, the next step of our research is to identify which subthemes emerge from Globalization, Cyrillic Fit, and De-Russification and how much weight each subtheme holds within its respective category. Theoretical models of language planning now include the individual and self-management as critical components of language planning policy and call to attention the impact of nonlinguistic factors, primarily colonialism and imperialism, on language planning (Spolsky 2019). By identifying the more granular factors shaping Kyrgyz attitudes toward script and their relationship to the individual and each other, we may be able to predict the trajectory of Kyrgyz language planning with increased accuracy and ascertain how these subthemes and relationships shaped current Kyrgyz policy. Our findings will therefore contribute to existing models of language planning.

Conclusion

The purpose of this study was to investigate the attitudes of Kyrgyz speakers toward the present use of Cyrillic script and a hypothetical transition to Latin script. We focused on three themes we determined to be the primary factors shaping attitudes toward Cyrillic and Latin scripts: De-Russification, Globalization, and Cyrillic Fit. De-Russification was found to have a positive relationship with Globalization and Cyrillic Fit: speakers who held positive attitudes toward De-Russification felt positively about Globalization and were confident in Cyrillic’s ability to account for Kyrgyz phonological patterns. De-Russification also explained the link between Globalization and Cyrillic Fit, indicating that the relationship is viewed through a de-Russification lens.

The findings show that De-Russification, a theme that encompasses how Russian influence affects the identity, culture, economy, and politics of Kyrgyzstan and how that influence can be reduced, is the dominant theme shaping speaker attitudes toward Cyrillic and Latin script. This suggests that script encompasses historical and present power dynamics between groups, and can express the identities and political, social, and economic conditions that an individual, group, or nation wants to shift toward. Thus, what we can conclude is that scripts yield a lot of power in determining the trajectory of individual and collective development and identity-building. This study challenges our understanding of what script encapsulates, prompting us to consider script choice and attitudes as a reflection of the politics and identities of the writers who employ it. It is not clear if or when Kyrgyzstan will formally propose a Latinization campaign. But should it ever happen, our findings provide a baseline indication of how the success or failure of the reform could unfold.

Acknowledgements

The authors would like to thank Miraida Salimova and Mirlan Ulanov for their translation work and Perizat Datkaeva, Nargiz Baudinova, and Usen Atabaev for their support with data collection. We are also grateful to Thomas Hodge, Angela Carpenter, Sabriya Fisher, Jab’ellalih Ixmata Schaaff, Emily Equihua Garnica, Audrey Yip, and Caroline Jung for their invaluable feedback and guidance.

Conflict of interest

The authors have no relevant financial or non-financial interests to disclose. The authors report there are no competing interests to declare.

Funding

This research was supported by the Jerome A. Schiff Fellowship (awarded to Sofia Rubio). The funding sources were not involved in the conduct of the research and the preparation of the article.

Materials and Code availability

The questionnaire used in the study is provided in the Appendix.

Notes on Contributors

Loren Sofia Rubio is a Speech-Language Pathology student in the Department of Speech, Language, and Hearing Sciences at California State University, East Bay. She has worked on projects relating to speech sound sequencing, language acquisition, and sociolinguistics.

Adam Roberts is a psychologist with specialization in neuro-ergonomics and is currently a Senior Researcher in the Future Resilient Systems program at the Singapore-ETH Centre. He has spent his academic career in inter-disciplinary research at the intersection of engineering and social sciences. Adam has worked on projects relating to speech and language perception in different languages, financial resilience and FinTech, environmental effects on the brain and behavior, and brain-controlled adaptive automation.

Yoolim Kim is an Assistant Professor in the Department of Cognitive Science & Linguistics at Carleton College and is also affiliated with the Cognitive & Linguistic Sciences Program at Wellesley College.

Footnotes

References