Keywords

1 Introduction

The link between language policy and planning, and state security, in both political and public discourse, is not new in Estonia (Kudus 2002). In the typology of language planning for security, Liddicoat (2008) classified assimilation of the potential enemy group, aimed at removing linguistic cleavage, as a typical measure for conflict prevention and management (between language groups). Assimilation was also the essence of the post-Soviet language planning that became a central tool for normalizing the situation in Estonia after decades of Soviet occupation and Russification of the public space (Smith 2003). Not only ethnolinguistic differences but also the threat of separatism in areas with a high share of Russian-speakers informed the language and citizenship policies (Feldman 2005). Brubaker (1998) has described the nation-building policies of Estonia and a number of other post-communist states as “nationalizing”, where the degree of nationalization and perception of inter-group stability depend on to what extent the core nation senses its position as being (extremely) unstable or vulnerable and continuously in need of protection (Verkuyten and Reijerse 2008).

The topos of threat, which Feldman (2005) describes as a distinctive endangered disposition of the Estonian identity, and Kalmus (2003) calls an “almost minoritized” position of the endangered majority, has been deployed as a powerful silencing tool in political debates about language in Estonia. Ehala has analyzed identity construction through confrontation and threats between the two language groups, concluding that this is a zero-sum game, since more positive self-esteem for Russophones can only be achieved by some lowering of the status of Estonian self-esteem (Ehala 2009, 155). Koreinik (2011, 36) found that the themes of endangerment present in academic discourse, which are reflected in public and media discussions, have obvious parallels in the post-Cold-War security debates, making use of discursive mechanisms to effectively suppress contesting views on language that use arguments of threat to the nation-state and combine them with the survival of the idealized and symbolic national language norm.

The normalization policies were mainly intended for the Russian-speakers who had settled in Estonia during the Soviet occupation, as the result of Soviet inner-immigration policy (Matthews 1993), and live now in concentrated enclaves in the capital or in the former industrial towns in Ida-Viru county, close to the Russian border, in the northeastern corner of Estonia. After Estonia re-gained its independence in 1991, the borders were literally “moved across people” (Ryazanova-Clarke 2014a). The Russian-speakers found themselves suddenly physically cut off from their usual routes and relations and living in isolation in terms of their interethnic socialization and communication with Estonians. This division continued in the Estonian Republic because of the two powerful Soviet legacies: separate Estonian- and Russian-medium schools and separate Estonian- and Russian-language mass media.

The normalization of the “asymmetrical bilingualism” (Ozolins 1999) inherited from the Soviet times was based on the logic of “language first”, where the Estonian language was seen as the key to integration into the Estonian society and information space. Liddicoat (2008) considers this belief that improved language capacities are seen as necessarily leading to enhanced communication an oversimplified isomorphism, since a shared language does not guarantee mutual comprehension. Indeed, while the older Russian-speakers were left out of the national communication flows due to their limited national language competences and limited contacts with native Estonian-speakers, the younger Russian-speaking generations, whose Estonian language competence was steadily improving, making them more frequent Estonian media users, reported feeling increasing distrust and alienation. In the Estonian-language media, they were confronted with depersonalized pictures of Russian-speakers, categorized as a homogenous group of “them”, or “Russians”, often depicted in negative terms, such as “civil occupiers” (Vihalemm and Hogan-Brun 2013, Karasawa 2011).

Twenty years after regaining independence, several reports (Estonian Human Development Yearbook 2010–2011, Integration monitoring 2012, and Bilingual learning in Russian-medium schools 2013) concluded that language-centered normalization tools had not been sufficient to create a coherent civic society, and that there was an increasing need for dialogue and the building of mutual trust between the two language groups. The reports were soon followed by an increasing focus on a dialogue in the political discourse, as political parties attempted to attract the votes of Estonian Russian-speakers, which reached its peak during the electoral debates preceding elections to local municipalities in October 2013 (Schneider and Cheung 2015). By the beginning of 2015 the political discourse had changed drastically from dialogue to distrust. The opening parliamentary election debate was on the changed security situation in Estonia after Russia’s annexation of Crimea and featured the leaders of the political parties. During the debate that took place on Estonian Public Broadcasting (ETV) on February 20, 2015, language political issues as the transition to Estonian-medium education in Russian schools and the general integration of Russian-speakers into Estonian society were conceptualized as central security topics.

This chapter analyzes the changes in language political debates in the Estonian online media news portal Delfi before and after Russia’s annexation of Crimea in March 2014 by applying a multimodal methodological framework of corpus-assisted discourse studies (CADS), suggested for the analysis of political discourses by Barker et al. (2008) and Ädel (2010). Our analysis is corpus-driven (Tognini-Bonelli 2001), meaning that the media texts studied are first analyzed with corpus linguistics tools, and the results of statistical computed calculations are interpreted with tools from qualitative discourse analysis. CADS combines a quantitative approach with qualitative analyses and helps to demonstrate how “concurrent and conflicting language ideologies are historically situated discursive constructs” and include “(re)construction and negotiation of language, national and ethnic identities unfolding in time” (Hult and Pietikäinen 2014, 2). The CADS definition of discourse is based on Foucault’s interpretation: discourse is a representation of reality in a way that has consequences for power relations (involving social class, ethnicity, gender etc.), either by preserving or challenging these relations (Baker & McEnery 2015). Before outlining the theoretical and methodological basis of our study, we will offer a short description of the media situation in present-day Estonia.

2 Russian-Speaking Minority and Media

Creating a common space of information in Estonian for all inhabitants was one of the tasks of the process of normalizing post-Soviet policies in Estonia. However, this attempt ran into changes in the global media and information environment as global connections grew and transnational networks challenged traditional national hierarchies of decision-making (Lievrouw and Livingstone 2009). These changes transformed both societies and information networks, as national and sub-national forms of social, political and economic inclusion and exclusion were reconfigured by the increasing reliance on borderless information and communication technologies. Since citizens’ media choices influence the development of their political attitudes, especially in countries undergoing political transitions (Loveless 2010), new media habits and transnational information networks meant that citizens’ political preferences were no longer “controllable” by national media. One other factor has influenced media consumption in post-Soviet Estonia: language competences.

During the Soviet occupation, the Estonian language served as a compensatory divide for the removed state border between the two language groups (Siiner 2006). The symbolic borders of the Estonian language nurtured a common free space for Estonian culture, where anti-Soviet ideas proliferated. Russian-speakers willingly stayed out of this space since the Russian, not the Estonian language was the language of prestige and a prerequisite for social mobility in the Soviet society. After regaining independence and in spite of the remedial language policies, which aimed to create a monolingual common space in the Estonian language, separate information spaces continued to exist partly due to inertia but also because of need and a lack of desire on Estonians’ part to open up their Estonian language information space to Russian-speakers.

Russia, on the other hand, quickly claimed ownership of the Soviet legacy of “the common information space” and continued producing Russian-medium broadcasting for the CIS and for former Soviet countries, including Estonia (Ryazanova-Clarke 2014b, Luhn 2015). This was necessary, according to Putin, since “the Russian nation has become one of the biggest ethnic groups divided by state borders” (as cited in Wilson 2014, 162). While many post-Soviet nation-states, including Estonia, promoted one nation-one language ideologies, in which language, culture and identity are bound together in one territory, historical continuity and a homogenous state (Ryazanova-Clarke 2014b), the Russian language continued to be a pluricentric language, not tied to one center or territory (Berdicevskis 2014). The Russian-speakers abroad were continuously addressed by the Russian media and politicians as compatriots (sootechestvenniky): people still concerned about the state and future of the Russian culture and language. In 2007 Putin launched a foundation called Russki Mir (Russian World, www.russkiymir.ru), which sought to dissolve physical national borders, create more porous, virtual borders of identity and promote a more positive image of Russia abroad. The Russian-language press in Estonia did not take a clear stance on what sort of space, and in common with whom, it would like to create, and the Estonian government did not see the Russian-language press as an important partner in getting their message through either (Vihalemm and Hogan-Brun 2013). The common media space in Estonian was thus for most of the 1990s defined only by Estonian-speakers, as the Estonian-language media channels also faced the risk of upsetting their mainstream consumers who were not ready to allow more “ethnic others” into their everyday information space (Vihalemm and Hogan Brun 2013).

Rather than merging into the Estonian-language media space as was expected by Estonian politicians, Russian-speakers turned away from over-politicized dailies, where they were treated as objects (Vihalemm and Hogan-Brun 2013). The most frequently and widely used mass-media platform among Estonian Russian-speakers is the Riga-based Russian language First Baltic Channel, often referred to as the “Russian state channel”, which mixes originally produced information about local affairs and entertainment (talk shows etc.) with news imported from Russian producers, offering Russian-speakers positive self-identification (ibid, 72). This engaging content makes the Russian media consumers’ motivation to switch to the alternative Estonian-language media channels for everyday opinion formation particularly unlikely. However, there is an exception in e-media, especially the freely accessible online news portal Delfi. Recent media consumption analysis shows that 23% of Russian-speakers have posted comments or questions on such internet news portals as Delfi on topics ranging from culture and politics to products and services, thus not only practicing consumption but also contributing in a modest way to content production (ibid.). As language ideological debates are moments when the civil society enters into policy making (Blommaert 1999), Delfi as the primary news portal for media consumers in Estonia has created the basis for data collection to map the topics and agents of public language ideological debates.

3 Theoretical and Methodological Considerations

3.1 State Security and Language Planning

State security and language planning and policy are closely connected in language policy theory, as both areas are of strategic interest to any nation and hence are sensitive to the geopolitics of the region where the country is located (Rajagopalan 2008). Furthermore, as argued by Ricento and Hornberger (1996), major political issues, such as the preservation of power or independence and opposition to state enemies, are often the principal impulses behind state involvement in language matters. Considering that the relationship between languages and national security is not new, language planning and policy studies have dealt surprisingly little with national security-related language planning and policy in a conceptually systematic and sound way. This may be related to the problem at the heart of classical language planning theory: the conceptualization of the nature of the language problems that official policies are designed to solve (Lo Bianco 2004). In his analysis of post-September 11 national language policy debates in the US, Lo Bianco (2008) criticizes technical and rational analysis of language policy issues that fail to capture the discursive dimensions of language policy making. In cases where language planning involves compounding the difficulties of trust and loyalty of a minority language group, defining the problem that has to be solved by a language policy is not at all straightforward (Lo Bianco 2008). In these cases, the problems become essentially ideological artifacts, reinforcements of ideologies used to justify the chosen harsh policy interventions (Edelman 1988).

Liddicoat states that there is an orientation in language planning which is not directly related to conflict management or prevention, but operates indirectly by creating a policy context in which issues of security and responses to threats are decided and enacted (Liddicoat 2008). This discursive practice in language planning is important in the context of the present chapter, since here the creation of perceptions of threat, i.e. the discursive construction of security as language use- and status-related, authorizes or precludes particular courses of action and constructs groups as oppositional on the basis of different languages or as allies on the basis of a shared language. The aim of the present chapter is to focus on this orientation in language planning, i.e. how language planning as a security issue enters into the discourse and into “existence” in language ideological debates in media. In the next sections, we will discuss methodological approaches to analyzing discourses in large data sets.

3.2 Corpus-Assisted Discourse Studies

Corpus linguistics-inspired CADS (Partington et al. 2013) has become increasingly popular in the analysis of presentations of certain issues in media text corpora (Taylor 2014), as it makes it possible to conduct the quantitative analysis of language use in large corpora of electronically stored (usage-based) texts with qualitative in-depth analysis of the lexical patterns that appear. Baker and McEnery (2015) argue that, due to the fact that corpus analysis is a computer-assisted method of analyzing naturally occurring examples of language use, it offers the researcher a reasonably high degree of objectivity. However, the subjective researcher’s input cannot be entirely avoided in building up the corpus and deciding which computer-detected linguistic patterns are to be analyzed and how. There is, therefore, a large variety of corpus linguistics methodologies (Barker et al. 2008), depending on what is analyzed and what corpora are assembled.

Corpus-assisted methodologies can be especially useful in studies of political discourse, i.e. how power is enacted through discourse in news reports (Ädel 2010, 592). Since the computer-assisted calculations of word frequencies and lexical patterns can be used to detect markedness of discourses (Baker 2010) and central agents and topics in texts in terms of who is mentioned most often, how much space an issue gets etc., CADS also makes it possible to determine the keyness of a text or corpus compared to another corpus. Keyness indicates the “aboutness” of a text or a corpus and leads to the statistically significantly higher frequency of particular words or nodes in the corpus under analysis in comparison with another comparable specialized corpus (Ädel 2010; Barker et al. 2008). Keyness analysis can be used to compare both synchronic and diachronic corpora (Partington 2014). These portions of texts revealed by corpus-based processes are then interpreted “by hand” to detect patterns of meaning making, which are related to the broader, extra-linguistic context to make sense of them (Wodak 2007).

3.3 Discourse Analysis of Language Ideological Debates

Language ideologies are, according to Spolsky (2009), historically situated discursive constructs embedded in daily language practices, and they carry articulations and beliefs about the nature, value and function of languages that are occasionally expressed in media. Blommaert (1999) adopts a processual perspective on language ideologies, demonstrating that these values and beliefs are not just expressed but negotiated in what he calls “language ideological debates”. In these debates, often occurring in media, interrelationships between discourses and their networked characteristics are seen in connection with broader historical, social, economic and political practices inside and outside the society where the debates occur (Hult and Pietikäinen 2014). Language ideological debates are thus slowly unfolding processes of discursive exchange that take place when polity enters into policy-making (Blommaert 1999, 5).

Blommaert developed his theory further, suggesting that, in order to give realistic accounts of objects under study that are bound to remain unstable, there is an increasing need for complex, rather than linear models that explain rather than predict the changes that societal developments cause for languages and language values (Blommaert 2015). While debates may become salient at particular moments of political crises, neither the debates nor the ideologies themselves emerge ex nihilo from the crisis point. They are, rather, contemporary instantiations and re-contextualisations of historically situated discourses (Blommaert 1999). Blommaert calls this discursive contextualization tool a jump-to-the-past for a chronotope. Chronotopes are chunks of history that are invoked to organize the indexical order of discourses, i.e. to have their parameters expanded or limited, which, in turn, has an effect on how those discourses are foregrounded and deployed in subsequent debates (Hult and Pietikäinen 2014). Inspired by Bakhtin’s key notion of heteroglossia, Blommaert explains how the introduction of chronotopes leads to a historical-sociological analysis of different “voices” within the social stratigraphy of the language of that moment. Understanding history means evaluating it from one’s own specific position in the sociolinguistic system (Blommaert 2015). We will elaborate further on the usefulness of chronotopes in our analysis section.

4 Corpus-Assisted Analysis of Language Ideological Debates on the Delfi News Portal, August 2013 – February 2015

4.1 Building the Corpora

We would like to emphasize that a corpus-assisted study can only tell us about the language in the corpus one is employing and, therefore, the composition of the corpus will necessarily affect the conclusions reached (Partington et al. 2013). However, in contrast to qualitative discourse analysis, where the starting point depends on the researcher”s standpoint and understanding of the meaning of the discourse, CADS, even in cases like ours where the researchers have to build corpora themselves, does not have to deal with the issue of biased research to the same extent, since the selection of texts in the early stages of the analysis is computerized and therefore free of the researcher’s prejudices (Baker and McEnery 2015, 6). Since there is no common and up-to-date corpus of Estonian media texts, we had to create a sample corpus (also known as DIY – do it yourself – corpus, Fitzsimmons-Doolan 2015) for the purpose of our analysis. We chose to create corpora based on texts published on the largest online news portal, Delfi. Delfi has a broad platform and it ranks as the most popular Estonian news website among Estonian internet users, based on an analysis of internet traffic (GemiusAudience) and surveys of self-reported media trust use (Vihalemm 2011). There is no universal measure of the representativeness of a corpus, although a corpus is considered balanced if it consists of a variety of texts, which our corpus does given the variety of texts published on DelfiFootnote 1 (our corpora contain news reports, readers’ letters, interviews, features and editorials) (Fitzsimmons-Doolan 2015). Two specialized corpora, based on predetermined content-focus words (Fitzsimmons-Doolan 2015), were located through the archival research of online publications on delfi.ee for articles in Estonian, and delfi.ru.ee for articles in Russian. The search parameters focused specifically on the period from August 2013 to March 2014, giving us seven months before the annexation of Crimea, in March 2014, and from March 2014 to the parliamentary elections in Estonia on March 1, 2015. A higher number of language ideological debate articles in the months before the elections for local governments in October 2013 and for the European Parliament in May 2014 was noticed early in our research, and thus we wanted to test whether this also held true for the parliamentary elections on March 1, 2015. The chosen texts related primarily to the debate about the integration of the Russian-speaking minority in Estonia and were identified through keyword list searches of the repositories. All of the archival searches and cross-checks were based on the following phrases: “Russian/Estonian language” “Estonian Russians”, “integration”, “citizenship” and “Russian school”. All of the selected articles had to mention “Russian language” and/or “Estonian language”.

The search resulted in 210 articles in Russian and 289 articles in Estonian, which were saved separately as .txt files organized in sub-directories by the corresponding months and then included in two parallel corpora of approximately 200,000 and 300,000 tokens. The collected corpus can thus be characterized as a small, specialized corpus (Barker et al. 2008) that can be processed by computer corpus software in a preliminary way, and the evidence of which can be examined manually and individually, while important features of the context of the production of the texts may become lost in a large corpus (Clark 2007). Since we wanted to detect changes in language ideological discourse patterns before and after the annexation of Crimea in March 2014, our criteria for corpus compilation were broader context, such as events in Ukraine and Estonia, and the narrower context of individual articles, such language policy-related topics as integration, language status, acquisition and citizenship issues. The two corpora, in our evaluation, were both sufficiently representative for our study and large enough to justify a corpus-assisted approach.

In spite of the fact that our analysis of changing language ideological debates in media was cross-linguistic, we did not have to face the challenges common to cross-linguistic CADS, such as finding comparable search terms for collocation analysis (Taylor 2014). Our comparative interest was largely cultural, social and historical, and not linguistic, since we analyzed Estonian and Russian corpora separately. However, since all of the corpus analysis was carried out in the original language, we faced the challenge of literal vs. functional translation and thus could not present the key-word-in-context (KWIC) concordance lines, which are central for illustrating language patterns, as the word order and even the node can get lost in translation (Taylor 2014). Furthermore, since both Estonian and Russian are languages with synthetic-inflectional structures, where nouns and adjectives are declined in cases, automated functions such as “keyness” could only be used tentatively, and the results had to be verified by concordance and collocation analysis.

4.2 Electoral Debates and Priming

The chart (Fig. 1) depicting the number of articles on language policy-related issues in Estonian and Russian (in delfi.ee and delfi.ru, respectively) reveals a remarkable increase in articles on LP in both Estonian and Russian Delfi in March 2014 (with 33 and 21 articles, respectively). Another surge in Estonian Delfi occurred in August 2014 (20 articles), in the period of the Day of Restoration of Independence and the Russian military occupation of eastern Ukraine, and the corresponding reaction in the Russian version of the portal (19 articles in September, 2014). The figures also demonstrate an increase in the number of articles in October 2013 and in February 2015, the pre-election periods of the local government elections (in 2013) and parliamentary elections (in 2015). As this indicates a connection between language ideological debates and electoral debates, it is necessary to outline the latter before we proceed, as it is closely tied to the discursive tool of priming: how media play a central role in what issues dominate during an electoral campaign.

Fig. 1
figure 1

Language policy-related articles in delfi.ee and delfi.ru.ee

Full size image

In the articles published on delfi.ee in October 2013, the tokens “Russians” occurred 52 times and “Russian-speakers” 31 times, having the nodes “Central Party” with “Savisaar” (leader of the party), “media” and “Tallinn” as the most frequent collocations (most often co-occurring words with the node, where the co-occurrence exceeded chance; Baker et al. 2008). The main topics were (number of concordance lines): What do Russian-speakers want/need (25)? Why did Russian-speaking voters vote for the Central PartyFootnote 2 (24)? Where do Russian-speakers get their information (21)? Supplied with discussions on a need for a TV channel in Russian, where could politicians enter into dialogue with Russian-speakers (20)? How often the media chose to cover a topic in their news stories played a key role in making an issue politically salient, by priming voters regarding it (Mutz 1992), thus being consequential for public opinion formation (Camaj 2014).

Priming theory, particularly psychologically, links agenda-setting effects to the formation of political judgment by offering a comprehensive explanation of how citizens formulate their political attitudes as a consequence of media content they consume (Nowak 2012). According to the priming theory, news media call attention to some issues or problems and ignore others in order to provide audience members with specific political knowledge they tend to use when forming political judgments (Nowak 2012). Russians/Russian-speakers were primed as a homogenous group of voters who unanimously voted for the Central Party. Dialogue with them (for example through an Estonian Russian-language TV channel) and an interest in their needs were of instrumental importance in finding out how to make them vote for other parties than the Central Party. Priming literature emphasizes two separate aspects of media priming mechanisms: issue priming, i.e. increased media salience (frequency) of certain issues, and attribute priming, in which “certain issue attributes emphasized in the media will become significant dimensions of issue evaluation among the public” (McCombs 2002). We will now proceed to our model of analysis.

4.3 Corpus-Assisted Analysis of the Corpora

For statistical analysis of the data, the AntConc concordance tool for Windows, version 3.4.4 from 2014 (Available from http://www.laurenceanthony.net/), was used to create a word list for the most frequent words in the chosen (parts of the) corpora. As frequency in itself is not sufficient to account for the statistical significance of a word, it was supplemented with a collocation tool and a keyness tool (Barker et al. 2008), which was used tentatively. The keyness tool makes it possible to compare two corpora to see how they differ. We compared articles published after the annexation of Crimea (March 2014–February 2015) with articles published before (August 2013–February 2014).

4.3.1 The Tentative Keyness of the Corpora March 2014 – February 2015

According to Partington (2014, 130), keyness can also be defined as the measure of both relative presence and absence of items in one data set compared to another, and the absence can be both relative (lesser use of previously dominating words) and absolute absence and presence (when new words appear and old disappear). He applied a comparison of diachronic corpora to detecting when new words enter and old words exit common usage. We applied a similar method and compared two diachronic corpora in both languages: one from before and one from after the Crimean annexation (the high frequency of articles in Fig. 1).

Keeping in mind the limitations of the keyness analysis mentioned in the previous section, what becomes clear from Table 1 is that both the Russian and Estonian corpora contain the words (in different cases) “Nato”, “Ukraine”, “Russia”, “Putin” and “Crimea” with high keyness factors and high frequency. “Russia” appears alone in the nominative case 177 times and “Ukraine” 89 times in the Estonian corpus, while in the Russian corpus “Russia” appears 117 times and “Ukraina/Ukrainye/Ukrainy” 145 times. For a more precise analysis, we examined concordance lines to see how Ukraine was related to the language ideological debate (Table 1).

Table 1 Key words in the Estonian (delfi.ee) and Russian (delfi.ru.ee) corpora of texts published in March 2014–Feb2015, as compared to the corpora of texts published in Aug13–Feb14
Full size table

In the Estonian Delfi, almost all occurrences of “Ukraine” from March 2014 (all together 334) were about how events in Ukraine had changed the security situation in Estonia, since a) Estonia, too, had a high concentration of Russian-speakers in its northeastern corner and b) these Russian-speakers, largely due to failed integration, consume Russian-language (and Russian-produced) media. One example from an article published in delfi.ee on April 2, 2014: “Events in Ukraine remind us again about the danger of Russian-speakers’ overwhelming consumption of Russia’s media”.Footnote 3 The verb “reminds” is here used as a scaling tool (Hult and Pietikäinen 2014), linking the perceived danger to events in Ukraine and to the occupation of Estonia by Russia.

The debate was similar in delfi.ru.ee, where mainly Estonian politicians and journalists referred to the separate language and information spaces that Russian-speakers in Ukraine and Estonia are living in, and explained why a group of a quarter million, occupying a vast territory in Russian-dominated information space in Estonia, could present a latent conflict potential and a security threat: “Ukraine first, Estonia next”. While delfi.ee articles published in March 2014 focused mainly on security issues, delfi.ru.ee published numerous articles during the summer of 2014 describing the Estonian government’s efforts to inform the residents in Ida-Virumaa about its firm interest in and attention to the region’s problems. The problems listed were cultural, socio-economic and educational; they also discussed access to judicial information in Russian and, towards the end of 2014, a simplified procedure for obtaining Estonian citizenship and the adjustments needed for educational reform in Russian schools. These are topics that local figures, such as politicians, teachers and school headmasters, had desired to discuss for years (Metslang et al. 2013), but finally were taken in account after the events in Ukraine, when “the voice of the Kremlin” entered into the debate (i.e. the high keyness of “Putin”, “Kremlin” and “Russia” in both corpora). The Estonian politicians’ voices were intended to drown out voices from Russia.

While the massive presence of Ukraine-related topics appearing in the language ideological debates in March 2014 (cf. their high keyness factor) was not surprising, what caught our attention was the high keyness factor of the words “border” and “Narva” in delfi.ee and “Latvia” and “Baltics” in delfi.ru.ee. We will take a close look at these key words by investigating the concordance lines (or key-words-in-context, KWIC).

4.3.2 Concordancing Narva

As has been stressed for years, the problems related to the linguistic integration of Russian-speakers are regional in nature, related to areas with high concentrations of Russian-speakers, such as the Tallinn area of Lasnamäe, Ida-Viru County and its biggest towns, Narva and Jõhvi. Experts have pointed to the need for regionally sensitive solutions for linguistic integration (Uus and Kaldur 2013) and have analysed the complexity of this task in a country with a high concentration of minoritie sin certain areas (Siiner 2014). The concordance lines where “Narva” appeared (in all cases of the word) in the delfi.ee corpus were qualitatively analyzed by hands-on examination. Table 2 demonstrates the main circulating topics NARVA was associated with. For comparison, we have also given the results for delfi.ru.ee.

Table 2 Concordance lines for NARVA and the main topics March14–Feb 15 (the number of concordance lines in the corpora for Aug 13–Feb 14 are in parenthesis)
Full size table

Table 2 demonstrates that in delfi.ee “Narva” occurred almost twice as often in the corpus after March 2014 as before and was mainly related to the border and Russian-speakers’ loyalty/separatism issues, and Narva residents’ Estonian language competences. The large number of concordances related to life in Narva was directly connected to Katri Raik, who at that time was the director of Narva College. Raik sees as her mission “bringing Estonia to Narva” (30.04.14), and says she often gets the feeling that her real job is being “an Estonian in Narva” (20.12.14). She is often interviewed in the Estonian media about Narva and so her mission seems also to be “bringing Narva to Estonians”. The few other proper names mentioned several times were two consecutive Ministers of Education, Aaviksoo and Ossinovski, whose names appeared in relation to the issue of the language of instruction in Russian schools in Narva. In addition, a lot of Estonian and foreign ministers started to arrive in Narva and talk to Narva residents after the events in Crimea. There was, however, a suspicious absence of “local voices”, the voices of actual Narva residents, such as ordinary Russian-speakers, school headmasters, teachers and local politicians. There were a few anonymous voices of people on the streets of Narva (they may have remained anonymous for their own safety) who were asked about their loyalty to the Estonian state vs. loyalty to Russia, and whether they would prefer living in Russia or Estonia. In an article in delfi.ee from April 17, an anonymous Narva resident said:

We do not need protection from either the USA or Russia. But we do not like all the negative stories about Russia. How would you feel if somebody criticized your mother? Russia is our mother. And we support Russia in the Ukrainian conflict, without doubt. On the other hand, we love Estonia as well; our families and friends live here. This is our home. And nobody here would vote to be annexed by Russia.

Narva had a prominent place in the language ideological debate, but not the Narva residents themselves. Narva was primed as a chronotope, a (potential) Estonian Crimea, a synonym for “border” and the impossibility of dealing with the heteroglossia. The “historical bodies” (Blommaert 2015) of the anonymous Russian-speakers cited affiliated them with both sides, “I am loyal to Estonia AND I support Putin” (from March 2014), while in the Estonian national narrative (closely tied to the chronotope of threat from Russia) Russian-speakers” loyalty was a zero-sum game. In delfi.ee, the chronotope of “border” and the alienation of Narva residents were conceptualized as something unavoidable: “Narva is a buffer zone. People living here are friendly to both Estonia and Russia” (Katri Raik, 17.04.14), and “Narva is geographically a perfect danger zone: in no other place in Europe is Russia so close” (Raik 30.12.14). The concordance lines, furthermore, reveal that the increased focus on life in Narva strengthened the view of Narva as an exotic town somewhere far away (and possibly dangerous).

The new historical reality established in March of 2014 dramatically affected the chronotopic organization of language ideological debates, invoking and deploying the chronotopes of Soviet occupation, mistrust, and almost quarter-century-old conflicts, thus changing the indexical orders of the same discourses and giving those debates new ranges and understandability (Blommaert 2015). For example, considering the intensity of language debates after March 2014 in the Russian language Delfi (over three times as many articles in delfi.ru.ee), Ida-Virumaa, which before March 2014 was strongly connected with language issues (cf. the regionality of the language problems mentioned), almost disappeared and was replaced by the node Narva. Hands-on examination of ‘Narva’ concordance lines showed that though still implicitly related to language, the attention to the region had shifted to the overall loyalty of the Russian-speaking population to the Estonian Republic (21 vs. 0 before March 2014) and integration (12 vs. 1). The quadrupled attention of Estonian and EU politicians (26 vs. 1) and media to the region’s life conditions (24 vs. 2) may have been related to the (re)contextualization of language problems and the depersonalization of Russian-speakers, making them again an exotic, homogenous group of aliens (Zazubovich 2014). While the central question in the pre-electoral debate in October 2013 was “what do Russian-speakers want?”, the priming of Russian-speakers in the pre-electoral debate in February 2015 was “The issue of the integration of Russian-speakers is a security issue in the sense that there are people in Estonia and politicians in Estonia who support the Kremlin’s politics” (Prime Minister Taavi Rõivas in the electoral debate on Delfi TV on February 20).

The impossibility of constructing a hyphenated Russian-Estonian identity (as one is based on opposition to the other) became clear during the Bronze Soldier conflict in April 2007, when Russian-speakers had to choose either the Estonian or the Russian version of the causes, course and impact/aftermath of World War II (Ehala 2009). In their analysis of the Bronze Night events Smith and Burch wrote, “By focusing wholly on the external dimension to this crisis [in 2007], the Estonian government also denied the possibility that its local opponents might have their own voice and subjectivity independent of Russia” (Smith and Burch 2012, 420). The unease that arose from having to deal with the dilemma of the plurality of voices (different accounts of what happened in WWII) was successfully avoided by leaving out the personalized voices of the Russian-speakers. The topic has been discussed earlier by Laitin (1996) suggesting that a Baltic-Russian identity was much more probable than an Estonian-Russian or Latvian-Russian identity. The impossibility of this choice for both Russians and Estonians has not been discussed in the media due to the relatively few voices of Russian-speakers who are Estonianized. This discussion is furthermore effectively silenced by calling the Russian-speakers “occupiers” or supporters of the Kremlin’s politics. The high occurrence of the token “Latvia” after and before March 2014 (89 lines vs. 1) and the high frequency of the issue of Russian-speakers in Latvia (in 31 lines) – citizenship issues, integration, ethnic crisis, and discrimination against Russian-speakers in Latvia – reveal the deficit of narratives linking this double-identity to the context of a nation-state/Estonia, and that it may be easier in the broader context of post-Soviet/Baltic states. This echoes similar tendencies in Russian-speakers’ media consumption. Since the national television presentations of the “historical homeland” or the “new homeland” do not fully meet the social needs of transnational immigrant consumers because they do not reflect the dilemmas and contrasts these consumers have to deal with, the most popular Russian media channel among Estonian Russian-speakers, the First Russian Channel, has successfully overcome shortages of such phenomena by mixing imported (from Russia) program elements with locally produced news (Vihalemm and Hogan-Brun 2013, 80).

5 Discussion

Linguistic ideology is not something that is only discussed in the media. The influence of the media on contemporary societies, including changes in individuals’ ways of speaking (Coupland 2007), is such that it is arguably the single most frequently studied institutional domain of discourse use (including political communication) in sociolinguistics and the (critical) analysis of discourse (Cotter 2001). In this article, we have carried out a corpus-assisted discourse analysis of the priming of language policy-related issues as issues of security in the Estonian online news portal Delfi, before and after the annexation of Crimea. Our aim was to shed more light on the rarely investigated role of language policy and planning in conflicts and in peacemaking: how language and language ideologies play a fundamental role in conflicts and can be used as a basis for differentiation, allowing one group to be identified as an enemy and maintaining separateness by preventing communication between groups (Chilton 1998). We found that the change was mainly due to a shift in focus from local problems (in delfi.ru.ee) and discourses about the need for dialogue between the language groups (in delfi.ee) to the alienation of Russian-speakers (the anonymous voices of Russian-speakers from Narva) and the priming of Russian-speakers as a security risk. The dominant focus on Narva in post-Crimean delfi.ee was not the recognition of the localness of language-related problems but rather a focus on Narva as a border town, where the fate of Estonian indivisibility once more will be decided (with reference both to events in Crimea and to struggles for independence in WW I and WW II). In delfi.ru.ee, language-related problems (citizenship, Russian-medium schools and unemployment) related to Ida-Virumaa, which were actively discussed in delfi.ru.ee before March 2014, were replaced after Crimea’s annexation by issues of loyalty (if one had failed to integrate (i.e. speak Estonian) that meant one was disloyal to the Estonian state). The absence of personalized local Russian-speakers’ voices in the texts, and the massive presence of “Russia”, “Kremlin” and “Putin” raise the question of which agents’ influence on Estonian-language ideological debates was most prominent: the people whose language problems had to be solved, or Putin, Lavrov and other Russian politicians, who were repeatedly cited as claiming to have the right to intrude in the life and fate of compatriots living in the “near-abroad”? Asking Cooper’s (1989) seminal LPP question “Who plans what for whom and how?”, as a way to explore the interplay between policy orientations and how policies are negotiated by different actors to different ends, Estonian Russian-speaking polity has, through discursive means, avoided entering into language ideological debates. One example of attribute priming is labeling the few Russian-speaking politicians’ criticism of political decisions as separatism (“Ossinovsky is like a separatist in the government who is firing out criticism of educational reform “, Delfi 20.02.2015). In Estonia, LP has, similar to in Ukraine, been an important part of nation-building and has been driven mainly by the political elite (Polese 2011, 37). Little if any space has been left for what Polese calls spontaneous nation-building by the polity (ibid.), especially in the post-Crimean language ideological debates. The post-Crimean language ideological debate that linked language policy issues to security issues contributed to the preservation of the language-as-problem orientation in the Estonian society (Ruiz 1984, 16). The dialogue taking place between the Estonian state and the Kremlin, rather that between the Estonian state and Estonian inhabitants, has once more moved the focus away from what, according to the human rights researcher Henry Steiner, is the basic conflict-avoiding task of the state: contributing to the creation of a civic and collective “us” identity (Steiner 1999).