Using the “proper one”: language ideology in the context of Kemalism and neo-Ottomanism

Abstract

Düzgün Türkçe (proper Turkish) is an expression used to refer to well-formed linguistic structures and orthography. On Twitter, where the digital language is visible, language users, by employing the expression, comment on others’ spelling styles about what is “true” or “false.” In the context of Turkey’s ongoing conflict on the use of Latin scripts after Justice and Development Party (Adalet ve Kalkınma Partisi; AKP) policymakers’ voices over the alphabet, I argue that the expression has gained social meanings associated with two diametrically opposing ideologies: Kemalism and neo-Ottomanism. Further, I also assert that there is a semiotic contrast between these two ideologies in the context of orthography. Thus, by being aware of this contrast and operating on the semiotic resources available to them, language users deploy their language ideologies. Drawing on interactional data on Twitter, this study brings an understanding of the process of how language users deploy their language ideologies by commenting on others’ spelling styles. In explaining the outworkings of this process, the study builds on the concept of indexical order.

Introduction

The current work draws on two diametrically opposing ideologies, namely Kemalism and neo-Ottomanism, in the context of language ideology. In this context, the study tackles Twitter data to understand individuals’ linguistic practices in which they deploy their language ideologies by tweeting about the orthographical conventions adopted by others. ¹

Twitter users employ the expression düzgün Türkçe (proper Turkish) to highlight well-formed linguistic structures and spelling on the platform. The expression takes on new significance in the digital context of Twitter, where individuals rationalize and define what constitutes “good” or “esthetically acceptable” spelling. This study, adopting a social–semiotic perspective, explores the visibility of “proper Turkish” in three social contexts or social meanings. Initially, the term relates to adherence to spelling rules. Expanding on this, individuals associate knowledge of these rules with education, creating a meaning tied to “educatedness.” Finally, referencing past social meanings, where spelling rules were linked to educated individuals, the term is used to object to Justice and Development Party (Adalet ve Kalkınma Partisi; AKP) policies, aligning with a Kemalist ideology.

Utilizing interactional data on Twitter, I delve into how language ideology is conveyed through semiotic resources in tweets. Concerning this, I aim to elucidate the interconnectedness of language ideology with other ideologies, serving as a tool for conflict and legitimization. In this respect, I adopt an indexical account, specifically Michael Silverstein’s (2003) “indexical order.”

The remaining parts of the work proceed as follows. The following two sections introduce the domains (i.e. language and education in Turkey and language ideology) and the theoretical dimension (i.e. indexical order), respectively. After laying out the background, the next section draws on the data. The article continues with analysis of the data and concluding remarks.

Forging national identity and unity: the political and ideological significance of language and education reforms in Turkey

In Turkey, language and education hold a dual role as reflections of the nation’s cultural identity and as influential tools in nation-building. The early years of the Turkish Republic focused on language and education reforms to establish a national identity that valued attributes like accomplishment, civility, enlightenment, knowledge, and secularity. The current section provides the historical and political context that has shaped the country’s linguistic and educational landscape to understand modern-day Turkey’s language and education debate.

During the Ottoman period, the linguistic landscape was characterized by a dialogistic interplay marked by the dynamic coexistence of different languages and scripts. The Ottoman variety of Turkish, utilizing Perso-Arabic script, was primarily spoken and cultivated by the elites, which included rulers, the Ottoman sultans, the central administration, and the army (Strauss 2017, 116). This form of Ottoman Turkish, which incorporates elements of Persian and Arabic in its pronunciation, orthography, and lexicon, was highly valued by the Ottoman elites who sought to advance Islamic high culture (Doğançay-Aktuna 2004, 5–6; Kerslake 1998, 179–180). In contrast, the Anatolian variety of Turkish, spoken by both Muslims and Christians, remained the language of the common people, particularly the poor, uneducated, and illiterate segments of the population (Bayyurt 2010; Doğançay-Aktuna 1995, 2004). During this period, low literacy rates were prevalent alongside the widespread use of the Ottoman variety by the elites. In this context, some voices criticized the orthography, viewing it as a hindrance to efforts to reduce illiteracy (Woodhead 2012).

The Alphabet Reform of 1928, which replaced the Arabic script with the Latin alphabet, and the subsequent Language Reform of 1932, which initiated the establishment of the Turkish Language Association (Türk Dil Kurumu; TDK), marked two pivotal language-related reforms undertaken during the early Republican period. ² According to Lewis (1999), these reforms were highly interconnected in attaining the Republic’s political and cultural idealizations since the Kemalist government in Turkey saw the language reform as an essential component of nation-building, as it aimed to produce a standardized form of Turkish that could be used across the country in all areas of life (Aydıngün and Aydıngün 2004; Bingöl 2002; Doğançay-Aktuna 1995; Yüce 2019). Furthermore, according to Tachau (1964) and Eastman (1983), the language reform holds significance as it forms a borderline between the Ottoman past and the Republic. In this respect, language became a product of the politicization of the reforms in the early years of the Republic in emphasizing attributes like being accomplished, civilized, educated, secular, and modern (Çolak 2004; Cüceloğlu and Slobin 1980; Fishman 1973). This meant that the absence of these attributes became unacceptable for those who aligned with the ideals of the Republic.

Language planning refers to the organized efforts to solve language-related issues, typically undertaken at the national level (Fishman 1973). The Turkish Language Reform, with its primary objective of creating a more user-friendly language and fostering better communication between linguistic elites and the general population, focused on orthography and the lexicon (Doğançay-Aktuna 2004). Reforms in the orthography replaced Arabic script with Latin script, aiming to free Turkish from foreign-language influence and elevate its status. This shift also entailed eradicating the use of the Arabic script and Persian in schools. Reforms on the lexicon included linguistic purification or Turkification, introducing neologisms to replace foreign-origin words. This process aimed to cleanse the language of external influences, particularly Arabic and Persian while adapting the lexicon to the Anatolian spoken variety. Concerning this, policymakers recognized the need for a language that better serves as a new communicative tool for the population, aligning with the evolving communication requirements (Doğançay-Aktuna 1995).

On the other hand, the foundation of the Ministry of Education in 1920 and the Law of Unification of Education (Tevhid-i Tedrisat Kanunu) in 1924 were the leading reforms related to education in Turkey. These two were important because the Law of Unification of Education abolished religious instruction in state schools and put education under the control of the Ministry of Education.

In the early Republican era, language reform followed education reforms in an intertwined way as the new standardized language was used as a tool for modernizing, secularizing, and unifying the education system (Karapehlivan 2015, 2019; Reed 1988; Toprak 1981). Concerning this, the new orthography, together with the lexicon that displaced Arabic and Persian words, was introduced as the medium of instruction in schools and utilized to promote literacy among the general population. For example, despite its short lifespan, the establishment of the Village Institutes (Köy Enstitüleri) was one of the visible tools of this process (Karapehlivan 2015). ³ This shows that language, alongside its orthography and lexicon, gained an inherent symbolic meaning that interacts with education. This symbolic meaning, thus, became an intrinsic tool of the Kemalist nation-building process, in which language reform was seen as a means to raise the country’s literacy rate by laying out a sharp boundary between the Ottoman past and the Republic (Lewis 1999; Yılmaz 2011; Zürcher 2004).

When the conservative Democratic Party (Demokrat Parti; DP) came to power in 1950, its agenda aimed at gradually reversing the secular education reforms of the early Republican era. Notably, the establishment of more Imam and Preacher Schools (İmam Hatip Okulları) served as an alternative to the secular education system, challenging its unity (Okçabol 2005; Özgür 2012; Pak 2004). In parallel, Quran courses proliferated, offering supplementary instruction to mosque teachings, and religious courses became elective in schools during the DP period. The reopening of Imam and Preacher Schools and Quranic courses prompted a shift in language and instructional mediums, elevating the prominence of Arabic script in society. From the 1950s to the early 1980s, language choices reflected a dichotomy between progressive/leftist and conservative/rightist perspectives (İmer 2001). İmer (2001, 87), citing İmer (1990), notes that proponents of newer linguistic forms were often associated with purist, leftist, and progressive inclinations, sometimes aligning with socialist and communist ideologies. In contrast, supporters of Ottoman linguistic forms were frequently linked with rightist and conservative leanings, and this period marked a crucial juncture where lexical choices became a mechanism for discerning individuals’ political orientations.

After the 1980 military coup, compulsory education saw the introduction of mandatory religious courses under the title Religious Culture and Moral Knowledge (Din Kültürü ve Ahlak Bilgisi) by the 1982 constitution. This era witnessed a surge in İmam and Preacher Schools due to the privatization of education, leading to the establishment of diverse Islamic community schools known as cemaat (jamaat). This shift made orthography, a symbolic language resource, more accessible to broader segments of society.

Since taking office in 2002, the AKP has actively shaped Turkey’s education system towards Islamization, evidenced by eliminating laws restricting religious education and revisions to primary school curricula and textbooks (İnal 2012; Kaya 2015; Yanarocak 2022). Additionally, AKP discourse emphasizes the symbolic role of Islam, particularly the importance of teaching Turkish in the Ottoman style using Arabic scripts (Ferreira Lopes 2018). These measures align with the AKP’s ideological goal of revitalizing the Ottoman past, leveraging historical, sociocultural, and religious ties to counterbalance secular Kemalist reforms (Yavuz 2016, 2020). This ideological interplay with established Kemalist discourse, evident in language and education contexts, reflects the intricate dance between the Kemalist pursuit of modernization and the AKP’s neo-Ottoman call for a reconnection with historical roots (Ongur 2015; Yang Erdem 2017). The resulting ideologies have indelibly shaped the discourse surrounding education and language in Turkey.

To conclude, political and ideological shifts have shaped the interplay between language and education in Turkey. From the gradual reversal of secular education reforms in the early Republican period to conservative religious ideology under the AKP, language has become a symbolic resource to challenge the dominant education system. Concerning this, the medium of instruction in schools, the introduction of the Imam and Preacher Schools and religious education, and the use of Arabic script have all had significant implications for the symbolic nature of language. In this symbolic nature, language, ideology, and power are visible.

Language ideology

The concept of language ideology has been extensively studied in linguistic anthropology and sociolinguistics since the 1970s, resulting in a growing body of literature on multilingualism and language contact, highlighting language ideology as a mediator of social practice (Blommaert 2005; Gal and Woolard 2001; Irvine 1989; Lippi-Green 2012; Woolard and Schieffelin 1994).

According to Van Dijk (1998, 8), ideology refers to a framework of social views regulating how speakers interpret their and others’ practices. This definition allows language ideology, which posits that individuals employ the same framework to assert their rationalizations of language, such as what constitutes “good” or “esthetically acceptable” language use. These rationalizations are apparent in various forms, including commenting on specific languages or speech styles, embodied social practices, or visual representations (Woolard 2020, 1).

Irvine (1989, 255) defines language ideology as “the cultural system of ideas about social and linguistic relationships, together with their loadings of moral and political interests,” highlighting the social and political motivations involved in language ideological postulations. Woolard and Schieffelin (1994) add that justifying a specific language by commenting on it reflects the speaker’s identity, individuality, moral and esthetic values, and authenticity. Concerning this, how people use language is not only a matter of linguistic proficiency or correctness, but it is also related to their social and personal identities, including their gender, ethnicity, nationality, and social class. Moreover, people’s language use reflects their sense of authenticity and beliefs about what is appropriate or acceptable in different social contexts. Therefore, language ideology is not only a matter of linguistic norms or rules but is also closely related to social, cultural, and political issues.

Silverstein (1979, 1992) emphasizes the object of language ideological accounts deployed regarding language use and structure. For instance, the Turkish terms bilim adamı (literally translated as “man of science”) and bilim insanı (literally translated as “person of science”), both denoting “scientist,” were subject to ideological performance in the early 2000s. Feminist linguists, who critically analyzed language’s role in perpetuating gender biases, argued that the term bilim adamı contributed to legitimizing patriarchal ideology in society. They proposed using bilim insanı as an alternative to promote gender-neutral language. Over time, advocates against this patriarchal ideology, primarily driven by feminist linguists, began to adopt the term bilim insanı. It is worth noting that this linguistic transition reflects an ideological conflict within the language itself. However, both terms are used in the public domain today, with a visible awareness of the underlying gender ideology.

This example demonstrates that language ideological justifications, such as when someone argues for or against a particular use of language, are not one-sided or unilateral. Instead, these justifications contribute to how language ideology is formed and understood through practice. In other words, the way people use and justify language is shaped by, and also shapes, larger social and cultural ideas about language. This features the interconnectedness of language and social practices and how language ideologies are not fixed or static but are constantly evolving through use and interaction.

Indexicality and indexical order

In this study, I employ a social–semiotic approach, specifically the concept of indexical order, as an analytical tool (Silverstein 2003). The concept of indexical order derives from Peirce’s (1932) tripartite interpretation of the sign, which distinguishes between the icon, index, and symbol. In brief, a sign’s meaning can be analyzed in terms of these three categories: the icon represents its referent through physical resemblance (e.g. a picture of a person in relation to the real person), the index implies a causal relationship with its referent (e.g. smoke as the indicator of fire), and the symbol represents its referent through convention (e.g. the number “9” and its visual representation).

Peirce (1932, 172) argues that the action of indexes depends on the association by congruity, which refers to the context that hosts spatial, temporal, and causal connections between the sign and its meaning. Put differently, the meaning of an indexical sign depends on the particular context in which it is used. This concept of indexicality is essential for understanding how individuals construct and circulate new meanings based on their experiences and interactions in particular social and cultural contexts. Moreover, indexicality is also critical to understanding indexical behavior, which acquires its meaning through context-specific associations and gives rise to indexical meanings. Finally, indexical meanings are social meanings that arise within social events, reflecting the participants’ social identities, positions, and norms (Blommaert 2005).

Silverstein (2003) introduced the concept of indexical order to capture the relationship between micro-level linguistic production and macro-level ideology. Indexical order refers to ordering indexical meanings through a series of social meanings, each constructed based on the previous one(s). Each level of interpretation in indexical orders involves individuals’ ideological moves and stances, which mediate in creating new indexical meanings. This means that language users use the previous indexical meanings and construct new ones based on their stances and ideological moves. In essence, Silverstein’s concept of indexical order helps to explain how social meanings emerge from linguistic production and how they are influenced by the broader sociocultural contexts in which they are produced.

To illustrate the concept of indexical order, consider the term bilim adamı, discussed in the previous section. In the first indexical order, the phrase refers to a person who works in science, i.e. a scholar. In the second order, the term is potentially reinterpreted by society as a gendered form of production since it involves the word adam (man). Finally, in the third order, individuals may construct an identity for those who use this expression, such as someone unaware of the social situation and becoming an agent in sustaining patriarchal essentialist approaches. This example shows that each order depends on the previous level, and individuals’ stances are crucial in constructing different meanings and indexicalities (Jaffe 2009).

Data

The current work draws on interactional data from Twitter between January 1 and December 31, 2022, through the Twitter API by employing the twitteR package (Gentry 2015). The twitteR package works in the R environment and enables data collection by focusing on specific keywords, timespan, and tweet types. In obtaining the data, by using the keyword “düzgün Türkçe,” reply and quote tweets sent during the above-mentioned period were included in the dataset. ⁴ The analysis that follows this part consists of a selection of this dataset.

Düzgün Türkçe is the expression that was used to sample the reply and quote tweets. ⁵ To disambiguate this expression, I should note that it does not denote utilizing the language by excluding words of foreign origin or paying attention to the pronunciation of certain words in the context of Twitter. Instead, on Twitter, the site where individuals observe others’ spelling styles, the expression means conforming to the rules of orthography. In this respect, by using the expression “proper Turkish,” individuals refer to written instances of “problematic” linguistic forms such as the question particle mI, conjunction and discourse marker dA, and the locative case marker -DA in laying out their arguments over the language used. That is, “proper Turkish” is an expression used in a context where individuals observe “improper” spelling conventions and evaluate these productions as “good” or “bad.”

Although language ideologies are multiple within the sociocultural groups with diverse conceptualizations of language use (Kroskrity 2004, 503), in this study, I draw on the salient sociocultural group that takes part more in the meaning-making process according to their political–economic interests. This salient sociocultural group is highly mobilized to achieve and use the semiotic resources available to them. It thus produces the central discourse on how language should be used “properly.”

Furthermore, Twitter data are essential for uncovering individuals’ social and linguistic practices in deploying their language ideology because it forms a domain in which people demonstrate or feature their ideologies. This can be further argued by employing the notion of “ideological site” (Silverstein 1992, 136). Ideological sites refer to institutional domains where people can socially practice their ideological expressions. Considering this, Twitter, where the digital language is visible as a medium of communication, becomes a site where individuals observe the linguistic behaviors of others and deploy their beliefs and arguments about language and its orthographic or grammatical conventions.

Analysis and discussion

This section discusses language ideologies through three indexical orders. The first order, represented by “proper Turkish,” highlights spelling errors, establishing a preliminary social meaning as a “spelling rule.” The second order expands the social meanings of spelling rules, emphasizing its connection to “educatedness.” The third order, interconnected with the preceding two, aligns with Kemalist ideology and serves as a stance against the AKP’s neo-Ottoman idealizations on language (Figure 1).

Figure 1. Indexical order of “proper Turkish”.

In forming the indexical orders, individuals on Twitter, by pointing to the orthographic conventions, enter into a recontextualization process in which they lay out their language ideological deployments. In doing so, they create a new ideological site through which their ideologies gain visibility. Considering this, the following analysis explains this process.

First order: “proper Turkish” as an orthographic convention

Considering the first indexical order, individuals comment on the orthographical style of the individuals or institutions with many followers. This implies that individuals who comment on others’ orthographical style are very well aware of the spelling rules. Thus, this awareness forms the primary social meaning of the “proper Turkish” case: a rule or an orthographic convention. In the context of the following examples (i.e. 1 and 2), Twitter users quote the tweet sent by a Turkish singer and former model known for a writing style that deviates from the orthographical conventions, Demet Akalın, and a news source about the Council of Higher Education, respectively: ⁶

In the context of (1), a Twitter user quotes the tweet sent by Demet Akalın, and by quoting the text in which Demet Akalın shows instances of misspelled words, the user argues that the singer does not know how to write “properly.” In (2), quoting a tweet sent by an online news source that reports on the Council of Higher Education’s implementation of compulsory Turkish language examinations for foreign nationals, another Twitter user responds by arguing that even the scholars of Turkish nationality working in state universities cannot write Turkish “properly.” In both examples, individuals associate the expression “proper Turkish” with orthographical rules, e.g. in (1), the user advises the singer to learn the spelling rules, and in (2), the language user suggests Turkish scholars take a spelling test.

In the above-mentioned examples, individuals articulate their stances toward the quoted content. In the first case, the stance is developed toward the writing style employed by the singer, and in the second example, the stance occurs on the content of the message. This is a social–semiotic practice employed by individuals communicating through computer-mediated environments in that they recontextualize the quoted material to form a basis to deploy their arguments (du Bois 2007; Gruber 2017; van Leeuwen 2008). Regardless of the previous content’s central message or core topic, individuals point to the spelling style to create a context and lay out their arguments.

In both examples, individuals associate the expression “proper Turkish” with orthographical rules, giving rise to its preliminary social meaning. Even in the public sphere, orthographical rules have several non-technical namings, e.g. dahi anlamındaki de (de that means “as well”) and de da eki (the affix de da). These non-technical namings are highly popular among the internet community, and even to draw traffic, websites produce content with the titles such as dahi anlamındaki de’nin yazılışını öğrenmeyen kalmasın (Let there be no people who do not know how to write “de” meaning “as well”) alongside a website named after this public saying, e.g. dahianlamindakide.ayriyazilir.com (dethatmeansaswell.writtenseparately.com). In this respect, I argue that in such a domain as Twitter, where people communicate with and feature the digital language, the social meaning of “proper Turkish” denotes conforming to the rules of orthography.

What is more, it can also be seen in the above examples that the users operate on the semiotic resources associated with education. The texts in the reply and quote tweets involve words and expressions such as schools and examinations. In the first example, “kindergarten” denotes that the rule is essential to be learned earlier in formal instruction. Similarly, in the second example, the “spelling test” refers to an argument that spelling rules are fundamental components of measuring language proficiency. On this, I argue that these expressions form evidence for further social meanings of “proper Turkish.” This by no means bridges the first and second indexical orders or ideological moves.

Second order: “using the proper one” indexing education

I argue that in the second indexical order, “proper Turkish” acquires an additional social meaning. Twitter users, who advocate for adherence to orthographic conventions, project a persona that implies an inability to follow spelling rules due to a lack of education, employing a strategy of delegitimization by using negative labels such as “illiterate,” “unlettered,” or “poorly educated” to stigmatize the Twitter user in question. Thinking that spelling rules are acquired through formal education suggests an inherent connection between orthography and education in the social sphere.

In the second indexical order, individuals build on the first indexical order (i.e. knowing the orthographical rules) to construct an “illiterate” identity. Concerning this, the data show that Twitter users quote or reply to anyone composing a tweet that involves spelling errors. Moreover, the examples given in the second indexical order also show instances of recontextualization. For example, when language users disagree with the opinion of others or views are seen as unacceptable, they modify or recontextualize the context of the tweet to take a counter-stance. Further, as I suggest, this stance environment creates a new ideological site where individuals lay out their language ideological postulations. The following data (3 and 4) capture how individuals construct an illiterate identity by holding on to different ideologies. In the context of these examples, Twitter users reply to others in football (e.g. 3) and religious lifestyle (e.g. 4).

In both examples above, individuals enter into a recontextualization process to reject or disregard the arguments of others. What is common in the replied tweets is that they both involve spelling errors. By pointing to these spelling errors, the authors of the tweets yield their justifications. For instance, the first example (i.e. 3) occurs in a football context where individuals discuss their football teams’ presidents’ public statements, i.e. the presidents of two football clubs in Turkey, Fenerbahçe and Trabzonspor. In ending the conversation, the tweet’s author, a Trabzonspor fan, replies to another who puts forward a counterargument over Trabzonspor’s president. On this, by stating that the author of the replied tweet does not follow the spelling conventions, the Twitter user in (3) says that they do not argue with an “illiterate” person or even an “illiterate” person who is a Fenerbahçe fan.

In the second example (i.e. 4), the reply tweet occurs in a context where a Twitter user responds to another who lays out their idealizations of a Muslim clothing style. In the conversation, the tweet’s author replies to one who argues how a Muslim person chooses their clothing. In response, the language user in (4) opposes by suggesting that the author of the replied tweet learn to write Turkish “properly.”

The above-given examples illustrate that individuals make use of the semiotic tools of education by relying on two propositional contents: those who do not achieve a specific education do not know the rules and cannot write “properly,” and those who do not fulfill particular education are “illiterate.” As a result, language users construct a persona who is “illiterate” and “ignorant,” and thus, their arguments are “worthless” and “should not be taken into consideration.” By constructing an “illiterate persona,” language users, at the same time, depict themselves as an “educated” person. In this respect, I argue that the “proper Turkish” phenomenon gains new indexical meanings in the second level.

Third order: “proper Turkish” indexing Kemalist ideology

What is common in the second indexical order is that language users, when they disagree with the opinions of others, create a new context to deploy their language ideologies. In this respect, to object to the views of others, they question the educational background of the person to whom they reply or quote. That is, language users implement the semiotic tools associated with education and educatedness to recontextualize the message. The exact process, called “objection” or “disagreement,” is also visible in the third indexical order. However, typical of the third indexical order is that these objections are directed toward the AKP and its policies. By drawing on the indexical past of the “proper Turkish” in the first and second indexical orders, language users, voicing a Kemalist undertaking, exhibit their language ideologies in the third order as part of stance-taking against the AKP.

Based on this context, I present instances of reply and quote tweets sent by language users on Twitter in the following lines. The following example (i.e. 5) is an instance of a reply tweet in which a Twitter user replies to an AKP-affiliated person who comments on Ekrem İmamoğlu, the mayor of İstanbul and a member of the Republican People’s Party (Cumhuriyet Halk Partisi; CHP), about the donation of the portrait of Ottoman Sultan Suleiman I:

The example above occurs in a context where Ekrem İmamoğlu announced that the portrait of the Ottoman Sultan Suleiman I had been donated to the İstanbul Metropolitan Municipality and started to be exhibited in the Municipality Museum. In the example, a Twitter user replies to an AKP-affiliated user who comments on Ekrem İmamoğlu’s tweet and claims that the mayor does this intending to raise his voters among the opposition camp, the AKP here, in the next election. Further, the very same user also sarcastically proposes that Ekrem İmamoğlu would announce the painting of the key figures of the CHP camp, such as İsmet İnönü, who was the second president of the CHP between 1938 and 1972, to receive more votes. In the reply, the tweet’s author recontextualizes the comment by taking a stance against the author of the previous tweet. In this stance-taking environment, the Twitter user constructs a persona that portrays someone who lacks knowledge of the spelling rules and is unable to write Turkish “properly,” thereby employing a strategy to delegitimize the constructed persona’s arguments. Further, the user claims that the members of the CHP camp do not favor those who are against the Republic.

In another example, a Twitter user quotes the Interior Minister, Süleyman Soylu:

In the context of the example above, a Twitter user quotes the tweet sent by the Interior Minister. In the quoted tweet, the Interior Minister announces the newly constructed government office buildings in the Mezitli district of Mersin. In the tweet sent by the Minister, the question particle mI, which is expected to be written separately from the previous verb according to the grammar conventions, was used adjacent to the verb. Thus, the Twitter user, pointing to the written form of the question particle mI by the Interior Minister, creates a new context and employs a strategy to delegitimize the Interior Minister’s use of language, arguing that the Interior Minister does not adhere to the spelling rules. After recontextualizing the tweet by stressing the orthography, the user takes a stance toward the Interior Minister and asserts that the building is “terrible.”

In the following example (i.e. 7), a similar stance is visible:

The example given above is an instance of a quote tweet. In the tweet, a Twitter user comments on a news source that reports on the vaccination of a person who is an AKP-listed member of one of Turkey’s city councils. In contrast, the vaccination was only allowed for healthcare professionals in the country at the time of the tweet. Before drawing attention to this situation, the user recontextualizes the quoted material by pointing to the educational background of some of the AKP members, including a board member of a bank and a member of the National Education Commission. Here, it is visible that the tweet’s author operates on the semiotic tools of education, such as diplomas and “proper Turkish,” in creating the context and then associates education with the AKP. Right after establishing this new context, the Twitter user comments on the news content. Hence, the tweet in (7) illustrates another stance-taking against the AKP, not on its policies but the practices during the COVID-19 vaccination processes. ⁷

The following example (i.e. 8) features another instance that can be interpreted in establishing the third indexical order:

The tweet featured in (8) occurs in a context where a group of Twitter users comment on a tweet by a journalist. Specifically, in this reply tweet, one of the Twitter users replies to another who argues that CHP supporters pretend to be AKP members in various social environments. By recontextualizing the tweet by pointing to the “proper” language use and writing style, the tweet’s author argues that CHP supporters use the language “properly” and thus cannot pretend to be AKP supporters.

What is common in the examples illustrated in the third indexical order is that individuals subvert the message to create a new indexical meaning. In the background of this new indexical meaning, they draw on the indexical past of the “proper Turkish,” first pointing to the knowledge of spelling and grammar rules and then creating an “illiterate” and “ignorant” persona to delegitimize their arguments. Following this, they associate “ignorance” with the AKP as part of a rejection of its policies. In this process, they first create a stance against those who are featured as people or social identities that cannot use language in a “proper” way, namely the AKP-affiliated people, i.e. members and supporters, then indirectly index themselves as a social group consisting of educated people who adhere to the founding principles of the Republic. This connection between linguistic forms and social identity is indirect indexicality (Ochs 1992). In the concept of indirect indexicality, Ochs (1992) argues that the interpretation of a linguistic form depends not only on the words themselves but also on the context and social situation in which they are being used. Furthermore, stances developed by speakers are the mediators of this interpretation process that they perform by using linguistic forms between language use and the group associated with that language use (Kiesling 2011).

I argue that the third indexical order holds on the distinction between two diametrically opposing ideologies, Kemalism and neo-Ottomanism, which form a social context for the interpretation. In this context, there is a semiotic contrast between these two opposing ideologies, and orthography is one of the sources of this contrast. Concerning this, the social context that creates the third indexical order needs further explanation concerning its macro-social frames of sociocultural action. As seen in the analyses, by projecting a persona such as “an illiterate AKP member,” individuals reject and devaluate the previous argument closely associated with the AKP. I argue that language users do not solely perform this social meaning-making process only based on spelling or orthographical conventions. Instead, they perform this action by being aware of the Kemalist and neo-Ottoman opposition.

As stated earlier in this paper, the founding ideology of the Republic, Kemalism, utilized language reform as one of its tools, and orthography gained a symbolic meaning and became a crucial aspect of the Kemalist nation-building process (Lewis 1999; Yılmaz 2011; Zürcher 2004). In this process, language reform and other education reforms were intertwined to modernize, secularize, and unify the education system (Reed 1988; Toprak 1981). Language reform, which provided the education system with a medium of instruction and instrument to elaborate on the social contexts in which the standard language or variety (i.e. the İstanbul variety of Turkish) was to be used, played a prominent role in mobilizing education. In addition, the outworkings of the language reform, together with the TDK, took part in the bureaucratic running of the Ministry of Education during this mobilization process (Aytürk 2008, 276). The mobilization of education also contributed to how the economic potential yielding to the social class of the Republic was obtained. Seeing that bureaucracy was the primary sign of the class-based societal system in the Ottoman period (Keyder 2017; Mardin 2017), the mobilization process can be interpreted as a means to transform this bureaucracy-based social system into an education-based one. Thus, in the very context of Turkey, orthography equips educatedness as its primary tool in the social meaning-making process.

On the other hand, neo-Ottomanism, emerging as a foreign policy of the AKP and later becoming visible in domestic politics (Wastnidge 2019, 7), attempts the alphabet in various practices to revive its idealized Ottoman past. Neo-Ottomanism is an ideological motive for revitalizing the Ottoman past by drawing on historical, sociocultural, and religious attachments to construct and produce memories against Kemalism (Yavuz 2016, 2020). In this process, the AKP has been the central figure triggering the neo-Ottoman endeavors since 2002 (Özel Volfová 2016, 496–497). Following the former prime ministers and presidents Turgut Özal and Süleyman Demirel, who advocated economic liberalism, Recep Tayyip Erdoğan, the founder of the AKP, followed and transformed this liberal trend into a neo-Ottoman undertaking. In neo-Ottomanism, the motivation for achieving political and economic strength is linked to the exclusion of Atatürk’s secularism by drawing on religious uniformity (Fisher Onar 2009, 235–236; Özel Volfová 2016, 496–497; Taşpınar 2008, 14; Türkeş 2016, 199). In attaining this uniformity, neo-Ottomanism has been practiced in various social spheres and discourses ranging from television series to architecture (Ergin and Karakaya 2017; Kınıkoğlu 2021).

One of these practices is visible in the context of language and education (Ongur 2015, 426; Yang Erdem 2017, 714). Alongside President Erdoğan’s selection of the Ottoman Turkish lexicon and public institutions’ offering teaching the Ottoman Turkish variety as a language option (Ongur 2015) and teaching Ottoman Turkish in secondary schools (Yang Erdem 2017), there are also other practices and discourses given by the AKP-affiliated policymakers. For instance, in May 2022, a recent news source reported that an aggregate food cooperation whose partner is Bilal Erdoğan, the son of President Erdoğan, had trademarked its brand name in Ottoman scripts. ⁸ In the very same month, at the end of the commemoration of the nationalist–conservative poet Necip Fazıl Kısakürek, President Erdoğan received a board as a gift involving the poem by the poet written in the Ottoman Turkish script. ⁹ Similarly, the AKP discourse makers have raised their voices several times on the need for language courses teaching the Ottoman variety of Turkish with Arabic scripts (Ferreira Lopes 2018). These instances show that neo-Ottomanism, among its other goals, also addressed the “language issue” as part of its identity construction, aiming to override Turkish with the Latin scripts.

This opposition can be further interpreted by drawing on the political actors’ discourses of (de)legitimation of laicism and anti-laicism. Under the laicist policies, Kemalism curtailed Islam’s role in the public sphere in politics, law, and education, relegating it to private life (Anderson 2008; Ongur 2020). This repositioning was justified by logical reasoning that aimed to boost literacy rates and modernize education, manifested in adopting Latin scripts and regulating religious education. In contrast, the anti-laicist discourses of the AKP, particularly evident after the 2000s, extensively draw from neo-Ottoman idealizations. These idealizations encompass narratives of nostalgia rooted in the Ottoman past (Karakaya 2020). In language and education, this nostalgia is discernible through Erdoğan’s expressions of the desire to read Ottoman scripts and the call for teaching the Ottoman variety of Turkish. ¹⁰ Yağcıoğlu (2002) explicates the Kemalist/laicist pursuit of Ottoman legitimacy and the neo-Ottoman/anti-laicist endeavor to challenge Kemalist legitimacy in terms of their approaches to historical context. Yağcıoğlu (2002, 129) posits that the construction of historicity acquires an ideological dimension for both Kemalist and neo-Ottoman factions, emerging as an othering strategy and mechanisms for legitimizing this strategy. Specifically, Yağcıoğlu and Cem-Değer (2001) and Yağcıoğlu (2016) argue that these two groups engage in a clash of modes of expression. They contend that while the Kemalist/laicist discourse group employs evidence and arguments to advocate for secular education, it demonstrates a logos orientation. In contrast, the neo-Ottoman/anti-laicist discourse group, focused on narratives of the Ottoman past and cultural values, embodies a mythos orientation.

The Kemalism and neo-Ottomanism opposition shows a semiotic contrast in the context of orthography, the conventional spelling system of Turkish. Further, delving into the same source, they exhibit a diametrical opposition. Peirce calls this a “diametrical icon,” which features relations between opposing entities (Peirce 1955, 105). Put differently, orthography was ideologically constructed by these two ideologies as their iconic representations. Gal and Irvine (2019, 116) suggest that when an iconic source is “ideologically construed as standing in co-constitutive contrast,” the icon becomes the source of the “axis of differentiation.” That is, how Kemalism and neo-Ottomanism tackle the alphabet issue gives the alphabet’s definition, and in this definition, there is an imaginary axis. On one end, Kemalism defines the alphabet as a device in the country’s modernization. On the other end, neo-Ottomanism characterizes the alphabet as a tool to ensure a connection with the Ottoman heritage. This suggests that each of these two camps excludes the other in the domain of orthography.

Based on the analysis provided, it becomes evident that Twitter users express their Kemalist viewpoints through the deliberate utilization of specific language ideologies. This serves a double purpose: to support their stance against the AKP and to strengthen their legitimacy in the ongoing discourse. Since the idea of “proper language use” is closely tied to social and cultural contexts, individuals exhibit a keen awareness of this dynamic when critiquing the spelling practices of others and delegitimizing their arguments. They are aware of the nuances of this interaction and intentionally use certain ideologies to convey different perspectives and voices, contributing to the complex discourse.

Concluding remarks

In this work, focusing on the interactional data on Twitter, I investigated how individuals deploy their language ideologies, which are socioculturally motivated ideological manifestations, by commenting on others’ spelling or orthographical styles. In the context of the current work, the expression “proper Turkish” forms the outlet of these ideological manifestations. Concerning this, by using the expression “proper Turkish,” individuals on Twitter comment on the spelling rules employed by others.

The main goal of the study was to bring an understanding of the process in which individuals employ the semiotic resources available to them and relate this to other ideologies as part of conflict and legitimization. In explaining this process, I adopted an indexical account, namely the indexical order by Silverstein (2003).

I claim that the “proper Turkish” case has acquired a series of social (i.e. indexical) meanings in the social sphere. Further, these social meanings have been connected through overarching ideological complexes. In the first social meaning, which is the first indexical order in the current work, “proper Turkish” has the social meaning of “spelling rule.” In this regard, it refers to the rules of orthography that individuals follow in composing their tweets. In the second indexical order, this social meaning gains further indexicalities that comprise “educatedness.” About this, individuals, by using the expression “proper Turkish” and drawing on the first indexical order, comment on the spelling style of others who do not follow the orthographical rules. In this process, they construct an “illiterate” or “ignorant” identity who does not know the rules. Lastly, in the third indexical order, tackling the past social meanings of “proper Turkish,” individuals index a Kemalist ideology to object to the AKP’s policies.

Individuals also initiate their stances in the course of their language ideological postulations. In initiating those stances, they first recontextualize the initial message, which is the quoted or replied tweet in the domain of the current study, and then lay out their language ideological deployments. From the first to the second indexical order, individuals deploy their language ideologies based on the stances developed on educatedness. Similar to this but in a different direction, between the second and the third indexical orders, the recontextualization of the AKP discourse by narrowing down the topic to spelling is a tool to create another stance. In this social meaning-making process, the use of negative labels like “illiterate,” “ignorant,” or “poorly educated” are deeply rooted in the historical and ideological context of the Republic’s founding principles. These labels represent a deliberate strategy of characterizing the AKP as deviating from the idealized subject of the young nation-state, particularly concerning education and language. By associating these labels with the AKP, the individuals raising a Kemalist voice not only criticize the party’s policies but also position themselves as defenders of the founding principles, thereby delegitimizing their political opponents.

The analysis given in this work will be of interest to those who aim to understand the ongoing conflict in Turkish politics. In this respect, this study, tackling linguistic data, contributes to our understanding of political differentiation, which is also visible in the context of orthography through signs, stances, positions, historical moments, and practices.