Sticking Together? Georgia’s “Beached” Armenians Between Mobilization and Acculturation

July 9 2021
Christofer Berglund ,Marko Dragojevic  &Timothy Blauvelt
Pages 109-127 | Published online: 09 Jul 2021

As the USSR fell apart and independent countries took its place, minorities across Eurasia found themselves stranded in nationalizing states. This article focuses on one of these “beached diasporas”: Georgia’s Armenians. Through a mixed-methods approach, consisting of interviews with activists and a sociolinguistic experiment administered to adolescents (N = 529), we uncover differences among Armenians in their reactions to Georgia’s nationalization policies. Armenians from the borderland of Javakheti mobilized in defence of the in-group but their co-ethnics from the capital of Tbilisi opted for acculturation. These intragroup differences demonstrate that members of the same ethnic group can react to the same nationalization policies along disparate lines, thus adding nuance to the literature on beached diasporas in the post-Soviet space.

“An unbreakable union of free republics, Great Russia united forever,” the opening lines of the Soviet anthem proclaimed. Yet, as the USSR disintegrated in 1991, the titular nationalities (re)claimed ownership of the Soviet Socialist Republics that bore their names. Fifteen independent states took its place and adopted “nationalizing” policies to promote the dominance of the titular group.1 Left inside these countries were millions of people who, as Vladimir Putin2 later complained, “went to bed in one country and awoke in different ones, overnight becoming ethnic minorities in former Union republics.” 50 million people found themselves as “beached diasporas,” stranded in the domain of a foreign state, including 25 million Russians in the non-Russian successor states to the USSR.3

The sudden demarcation of international borders confronted these minorities with difficult choices. Some emigrated to their “historical” homeland, resulting in an “unmixing of peoples.”4 Those who remained had to decide whether to mobilize for their cause or acculturate into the existing surroundings.5 Among minorities beached outside their titular republics in the South Caucasus, mobilization often took precedence. Karabakh-Armenians rejected Azerbaijani rule and Ossetians and Abkhaz mobilized for separation from Georgia. But Georgia’s largest non-titular ethnic group, the Armenians, never coalesced around a common response.6

As late as 2011, residents in Javakheti, a Georgian region bordering Armenia, referred to by its Armenian inhabitants as Javakhk,7 pondered: How should we know where our future lies or what languages to learn? A school teacher named Sonia laid out the dilemma facing her.8 As a Soviet-citizen of Armenian origin born into the Georgian Soviet Socialist Republic, Sonia had learned Armenian and Russian during her upbringing. She still taught classes in Russian. But since her pupils did not speak Georgian – the state language – Sonia feared that she was leaving the next generation as unprepared for life in Georgia as she herself had been upon its independence in 1991.

How did Georgia’s Armenians deal with their predicament as a beached diaspora in the decades after the Soviet collapse? We explore this question through interviews with influential Armenian activists and through a sociolinguistic experiment administered to 529 Armenian adolescents residing in the urban milieu of Tbilisi and in the rural setting of Javakheti, where Georgia’s largest Armenian communities reside. Our data reveals pronounced differences in their reactions to Georgia’s nationalization policies: those from the borderland mobilized in defence of the in-group whereas those from the capital leaned toward acculturation. These intragroup differences in responses to nationalization among Georgia’s stranded Armenians have hitherto gone unnoticed.

Our findings contribute to the literature on beached diasporas in post-Soviet Eurasia. Past research has focused on Russians in the “near abroad”9 or on minorities resorting to separatism.10 Javakheti has been discussed as a potential hotspot.11 But its Armenian residents have not been compared to their co-ethnics in Tbilisi. Nor has this ethnic group been studied as a case among other “beached diasporas.” We fill this empirical gap and demonstrate that Georgia’s Armenians, despite facing the same nationalization policies, failed to stick together. We caution against overstating the boundedness of ethnic groups and argue that intragroup differences explain variations in mobilization.

In what follows, we position this article in relation to earlier research on beached diasporas and introduce both Georgia’s Armenians and the nationalizing policies facing them. Next, we trace patterns of mobilization among Armenian activists, from 1991 to 2012, and then examine attitudes toward acculturation among the first generation of Armenians born into post-Soviet Georgia. We infer that Armenians in the borderland are more prone than their peers in the capital to protest against nationalization, and conclude that several intragroup differences push them toward these divergent reactions.

Brubaker12 coined the concept of “nationalizing” states after the communist collapse and the reorganization of political space along national lines across Eurasia. He argued that scholars devoted too much attention to “state-seeking nationalisms, neglecting the ‘nationalizing’ nationalisms of existing states.” That is, elites in the Soviet successor states did not see their national project as complete with the achievement of independence. Since their states fell short of being “of and for particular ethnocultural nations,” nationalizing policies were needed to finalize the nation-state. In almost all post-communist states, officials took “remedial” actions to strengthen the titular nation through policies promoting its language, culture, demographic dominance, economic welfare, political control or some mixture thereof.

These policies often encountered opposition from diasporic minorities and their kin states. The former sought recognition of their distinctiveness and the latter asserted – albeit with different degrees of determination – their right to protect co-ethnics stranded abroad. The nationalizing state, its minorities, and their kin states were bound together in a “triadic nexus,” Brubaker13 argued. Laitin’s14 research on Russians in the near abroad occasioned a shift in focus from the states enforcing nationalizing policies to the minorities confronting them. He referred to the latter as “beached diasporas,” since these groups found themselves in foreign countries not because they had moved there, but as a result of the fluctuation of international borders.

Laitin15 speculated that Russians and Russophones in the near abroad might fuse into a distinct “Russian-speaking population.” Yet, later studies discovered little evidence to suggest “the development of a conglomerate identification among Russians” in the post-Soviet space.16 Indeed, Commercio17 finds that Russians stranded in different non-Russian successor states react along different lines.18 Context matters. But this insight can be taken further. Support for mobilization – as opposed to acculturation – might well differ even within a particular group beached in the same state and facing the same nationalizing policies.

Groupist thinking has been common among scholars of ethnic politics.19 However, “support for nationalism among members of an ethnic group is neither constant nor random.”20 Those from “mono-ethnic rural areas,” where native language upkeep is stronger, often feel more connected to their in-group and see greater appeal in protesting.21 Toft22 finds that minorities settled as “concentrated majorities” in a region are most prone to conflict – and “urbanites” residing in cities least so. Marquardt23 places less importance on settlement patterns and argues that it is subgroups not speaking the titular language, and thus facing restricted life opportunities, that mobilize.

Our investigation speaks to this debate on intragroup differences and mobilization among beached diasporas. We focus on members of the same ethnic group confronting the same predicament, Georgia’s Armenians, thus controlling for confounding factors. The “paired comparison” between Javakheti-Armenians and Tbilisi-Armenians is well-placed to gauge the importance of location and language in processes of “ethnic” mobilization.24 Mixed-methods are utilized to discern their reactions. First-hand interviews with Armenian activists and a careful reading of secondhand sources help us trace patterns of protests. We also administrated a sociolinguistic experiment to Armenian adolescents in order to uncover in-group allegiances and attitudes toward acculturation among the first generation born into post-Soviet Georgia.

Despite being rooted in the Caucasus for millennia, Armenians have often lived under “alien rule,” as Hechter25 calls it when members of one group are under the rule of members of a different group. After the fall of their kingdom in 428, some Armenians sought safe haven in the Georgian kingdoms to the north, which also struggled to preserve their independence from neighboring empires. In 1122, Georgian King David the Builder recaptured Tiflis (now Tbilisi) from the Arabs and encouraged Armenian merchants to settle in his realm, which soon came to include lands far south of modern Georgia.26 Although Georgians and Armenians came into close contact during the Middle Ages,27 their social roles became differentiated. Georgians predominated among the enserfed peasants and the landed nobles on the rural estates, whereas Armenian craftsmen and merchants dominated in the towns.28

These roles consolidated under Russian rule. In Tiflis, the center of Tsarist administration in the Caucasus, Armenians gained protection from abuse at the hands of Georgian nobles and access to markets. Armenian merchants became “honoured citizens,” entrusted with running local government in Tiflis and rose to high offices.29 “The engineers of Russian imperialism continued to hold up Armenians to other regional natives, including Georgians, as an example of proper conduct and devotion,” Riegg30 finds. As the Russo-Ottoman War ended in 1829, Armenians left Anatolia for Javakheti,31 now under Tsarist control, and local Muslims headed in the opposite direction. This population exchange enabled Russia to colonize conquered lands with loyal subjects and transformed Javakheti into an Armenian outpost.32

In the meantime, national consciousness increased among the peoples of the Caucasus. Georgian peasants and aristocrats left the rural estates and headed for the towns, where they “came face to face with a well-entrenched, financially secure, urban middle class whose [Armenian] members spoke a different language, went to a different church, and held very different values.”33 Georgian intellectuals began promoting a cultural renaissance and use of the Georgian language. Under pressure from their Georgian neighbors but also from the Russian authorities – suspicious of their “mad dream” for independence34 – Armenian nationalist intellectuals in Tiflis created the Armenian Revolutionary Federation (ARF) in 1890.35

ARF leaders were central to the declaration of the first Republic of Armenia after the collapse of the Russian Empire and then the Transcaucasian Democratic Federative Republic in 1917–18. But the founding of separate Georgian, Armenian, and Azerbaijani republics did not spell an end to inter-ethnic conflicts. The republics made competing claims on territories seen as belonging – for historical or demographic reasons – to their aspiring nation-states. Georgia and Armenia engaged in a brief war over the borderlands of Javakheti, Lori, and Borchalo. This conflagration also prompted the Georgian authorities to lash out against Armenian civilians by arresting ARF-affiliated politicians in Tiflis, closing down their newspapers, firing Armenian civil servants, and even expropriating the properties of merchants.36

From 1918 to 1921, the Armenians found themselves stranded in a Georgianizing state. The Soviet invasion did not ameliorate their problems. Nationalities policies in the USSR offered opportunities for minorities, but less so for those residing outside of their titular republic.37 In Tiflis, Armenians’ use of their mother tongue receded to the household context. Most attended Russian-language schools and learned Georgian from social interactions. Georgians had privileged access to posts in the Georgian SSR and migrated to the capital,38 becoming a plurality in the 1926 census and a majority in the 1970 census.39 In Javakheti, Armenians lived in virtual isolation from Georgians. Due to its location next to a state border, residents needed permission to travel to, or receive visitors from, other parts of the Georgian SSR. What was more, the presence of a Soviet military base in the region had a Russianizing impact on the local Armenians.

After the dissolution of the USSR and the emergence of a sovereign Georgia, Armenians again had to reassess their relationship to the state. Zviad Gamsakhurdia, who led the drive for independence and was elected president in 1991, gave voice to a zealous ethnic nationalism, reflected in the slogan “Georgia for the Georgians.” He envisaged an enclosed Georgian nation, which he claimed had a “Spiritual Mission” on account of its ancient language and adherence to the Georgian Orthodox Church. Gamsakhurdia perceived this national project to be threatened by “other nationalities which were brought here by the Kremlin, by Russia, by the empire.”40 His calls for Georgian “ownership” demoted minorities to “guests” and provoked them to rebel against or otherwise to question their role within the state.41

After conflicts erupted in the Ossetian and Abkhaz borderlands, Georgians themselves turned against Gamsakhurdia, who was ousted in a coup in late-1991. Militias ruled different corners of Georgia until Eduard Shevardnadze, chairman of the “State Council,” managed to disarm and overpower them, becoming president in 1995 and restoring a semblance of order. But state structures held together through patron-client ties, and Shevardnadze opted for a “politics of omission” to avoid frustrating both Georgian nationalists and disaffected minorities.42 All residents obtained citizenship, Soviet-era ethnic markers disappeared from Georgian passports, and multiculturalism was praised. Yet in practice Shevardnadze did little to protect minorities’ rights. Nor did he endeavor to integrate minorities into the national fabric.

After the 2003 Rose Revolution, Armenians faced nationalizing policies of a different shade.43 President Mikheil Saakashvili built a capable state apparatus and used it both to promote tolerance toward minorities and to encourage them to learn Georgian. To the first end, Saakashvili made a point of extolling historical periods known for their inter-ethnic accord. He signed the Council of Europe’s Framework Convention for the Protection of National Minorities and created institutions to execute an Action Plan for Tolerance and Civil Integration.44 Officials began to bolster anti-discrimination laws, fund TV-programs in non-Georgian languages and NGOs advancing minority cultures. Denominations other than the Georgian Orthodox Church also won the right to register as legal entities of public law.45

To the second end, Saakashvili fostered connections between Tbilisi and Armenians in Javakheti through extensive road reconstruction, halving the travel time required between the capital and the borderland. He proceeded to implement laws, ignored under Shevardnadze, demanding that public servants be capable in Georgian. Russian began to lose its role as a language of inter-ethnic communication, as fewer Georgian students learned it at school or through the media. Officials instead accelerated efforts to acculturate the next generation of minorities to the state language. They started student exchange programs between Georgian and non-Georgian schools, allotted more hours to teaching Georgian as a 2nd language in the latter, piloted multi-lingual classes in selected schools, and introduced an affirmative action scheme in order to encourage minorities to pursue education at Georgian universities.46

Armenians in Georgia are thus no stranger to alien rulers. Georgian kings, Russian tsars, and Soviet general secretaries have all offered them different life opportunities, conditional on different cultural tradeoffs. Yet, since 1991, their predicament as a beached diaspora has been acute. Could their deep roots to their territorial homeland (Georgia) offset the pull of their titular homeland (Armenia) next door and – if so – would Georgia’s Armenians mobilize against or acculturate into the nationalizing state enclosing them?

In order to ascertain their reactions, we trace the ebb and flow of mobilization from the critical juncture of independence in 1991 up until the first electoral transition of power in 2012,47 using secondhand sources and first-hand interviews with Armenian leaders spearheading the drive to protect ethnic group interests.48 Time and again, activists in Javakheti mobilized Armenians to defend their domain from the incursions of the Georgian state. In contrast, activists in Tbilisi often shunned conflict to prevent Armenians from having to choose between different dimensions of their identities.

These patterns are reflected in reactions to Gamsakhurdia’s rise. His Roundtable Coalition had little appeal in Javakheti. In its most populous district, Akhalkalaki, just half of the voters backed Georgia’s independence in 1991.49 Locals blocked ethnic Georgian prefects sent to govern the region from taking up their posts and rallied behind a nationalist movement called Javakhk. “All Armenians were prepared to fight for [regional] independence at that time,” its leader recalled.50 But Javakhk did not press the issue and stopped short at organizing elections to a Council of Representatives, which nominated an ethnic Armenian, Samvel Petrosyan, as prefect. The movement also commanded a militia comprised of veterans of the Karabakh War with arms obtained from the 147th Soviet Motor Rifle Division in Akhalkalaki. Local activists ended up taking law and order, tax collection and policing into their own hands until the mid-1990s.51

Armenians had dwindled to 12 per cent of Tbilisi’s population in 1989,52 and also found themselves in the epicenter of Georgian nationalist mobilization. Distrust toward them ran deep since their co-ethnics in Abkhazia fought on the side of the separatists to release the region from Georgian rule.53 To avoid stigmatization, some Tbilisi-Armenians changed their surnames to make them sound more Georgian.54 Others opted to emigrate, often to the Republic of Armenia. At the time of the 2002 census, Armenians’ share of the population of the capital had fallen to 7.6 per cent.55 The remaining Armenians interacted with one another through the Surb Gevork Church, the Adamian Drama Theater, and several NGOs, institutions which engaged in cultural issues rather than in political interest articulation.

In contrast, Javakheti’s Armenians mobilized to fend off Georgianization. Locals protested state-sponsored plans to settle Georgian eco-migrants in the borderland, thus forestalling a bid at demographic “colonisation.”56 The share of Armenians in Javakheti’s Akhalkalaki and Ninotsminda districts expanded from 91 to 95 per cent from 1989 to 2002.57 Locals also rallied against Shevardnadze’s “divide and rule” policies toward the region. In 1994, his government merged Akhalkalaki and Ninotsminda into the larger Samtskhe-Javakheti region, in which Armenians comprised 55 per cent of the population. Moreover, despite the formation of local self-governance bodies in 1998, appointments to important posts were still decided from the capital, thus enabling Shevardnadze to offer Armenian activists lucrative positions in return for their political allegiance.58

The Javakhk movement fell apart as its leaders took up such offers.59 Co-opted Armenian officials distributed spoils from contraband trade in their networks.60 But locals felt alienated from Georgia. They dodged military conscription and assaulted officials seen as subservient to Tbilisi.61 Due to the poor roads leading to the capital, locals seldom traveled there and instead took up jobs at, or sold produce to, Russia’s garrison in Akhalkalaki. Rubles squeezed out the Georgian Lari as the region’s main tender. In 1997, 42,000 residents signed a petition asking that Javakheti be made a separate province with its own governor. David Rstakyan, the founder of Virk, led this drive, but Vahagn Chakhalyan of the Javakheti Youth Sport Union and United Javakhk movements also roused locals.62

This push for self-determination in the borderland triggered unease among Tbilisi’s Armenians.63 The latter harbored grievances too. Churches seized under Soviet rule had not been returned to the Armenian Apostolic Church and fell into disrepair because state officials did not dare to oppose the Georgian Orthodox Church, which claimed the sites as theirs.64 Activists were also anxious about the closure of Armenian schools in the capital, since this detached pupils from their mother tongue and raised the specter of a “white genocide” through assimilation.65 The fact that Tbilisi-Armenians, despite often speaking fluent Georgian, lacked representation on the city’s legislature fed suspicions of discrimination.66 Yet, choosing sides in a confrontation would be akin to “forcing a mother to choose between her two children,” as one Tbilisi-based interviewee put it.67

Armenians in the borderland tended to dismiss their co-ethnics in the capital as “assimilated,” and some considered their plight as proof of the meager gains to be reaped from Georgianization.68 Saakashvili’s nationalization policies were thus met with resistance in Javakheti. His administration did gain a “lock on power” over the region, as Armenian officials defected to the ruling party after the Rose Revolution.69 But local protests erupted against the demand that public servants speak Georgian, against the closure of Russia’s 62nd base, against education reforms seen as a stepping-stone to assimilation, and because of fears of demographic changes.70 The degree of support behind these concerns became clear during the 2006 local election, when United Javakhk attracted over a third of the vote in Akhalkalaki.71

Even so, Chakhalyan72 stormed into the office of the district election committee, rose in hand, accusing them of fraud and declaring a revolution.73 Fearful of revolts along the state’s borders, Georgian officials and their co-opted local allies began cracking down on United Javakhk,74 ending in Chakhalyan’s arrest. Russia’s subversion efforts and admission to courting separatists and “abandoned” minorities, as part of its “small play with Georgia,” had created an atmosphere of distrust in the period leading up to the closure of the 62nd base.75 Yerevan did refrain from interfering on behalf of its “beached” co-ethnics, even preventing Russia from using its troops in northern Armenia against Georgia during the 2008 August War.76 Nonetheless, Georgian officials kept a close watch over Armenian activists in Javakheti, since fears still lingered that the borderland could serve as a bridgehead for foreign subterfuge.77

Mobilization faded as United Javakhk leaders left Georgia.78 But Saakashvili also relaxed the enforcement of language laws, enabling Armenian public servants to keep their jobs, rooted out corruption, rebuilt infrastructure, and created quotas for Armenian students to access Georgian universities. Co-opted activists praised these steps as proof that Tbilisi, “at long last,” cared for the borderland.79 Others argued that language laws were ignored not as a concession but rather to keep co-opted officials in check, since the threat of being fired for noncompliance still hung over them like a Damocles sword.80 Resistance against the center’s embrace manifested in Armenians’ reluctance to send their children to Akhalkalaki’s Georgian-language school for fear of being labeled a “traitor.”81 Moreover, soon after Saakashvili’s electoral loss in 2012, the district council tabled proposals for regional language rights.82

These patterns suggest that Georgia’s Armenians reacted to their predicament as a beached diaspora along differing lines. In Javakheti, the pursuit of self-determination came to naught after Georgian officials began co-opting activists. Armenians obtained important posts in the region, resulting in a form of “indirect rule.”83 This made it difficult for locals to protest against the capital's policies but it also made it difficult for the capital to force its policies onto the region. In Tbilisi, Armenians faced still stronger pressure for nationalization, which some skirted through emigration and others through assimilation. Those remaining sought to avoid confrontations that could worsen these splits. Even though activists in the capital had greater capabilities to mobilize,84 and some offered scathing criticism of Georgia’s nationalizing policies, urban Armenians in Tbilisi exhibited lower willingness to mobilize under this banner.

If the aforesaid conclusion is correct, then Armenians in the capital ought to see acculturation into Georgia as a less threatening prospect than their co-ethnics in the borderland. In order to investigate if this is indeed the case, we organized an experiment to capture Armenians’ attitudes toward linguistic integration and cultural assimilation. One method for doing so is the speaker evaluation paradigm, in which participants are asked to listen to audio recordings of speakers representing different ethno-linguistic identities and rate each on several evaluative traits.85 We fielded this experiment among 529 Armenian adolescents attending high schools in Javakheti (n = 426) and in Tbilisi (n = 103). Results from these samples offer us a glimpse into the sentiments toward acculturation prevailing amongst the first generation of “beached” Armenians to be born into independent Georgia after the dissolution of the USSR.

We recorded a native speaker of standard Georgian and a native speaker of standard Armenian, both females in their 20s with similar-sounding voices, reading the same text aloud in their respective language.86 These recordings were played back to respondents who then assessed the voices in terms of several traits using a Likert scale (for instance, “Is she cultured?” ranging from 1, “very little” to 6, “very much”). We told respondents that we were “interested in how people form impressions about others by hearing their voices,” but did not inform them that both recordings recurred several times. Using name tags characteristic of different ethnicities,87 the speaker in each recording was presented once as a monocultural person and once as someone posing as a member of the other group. Figure 1 plots these guises from the Armenian standpoint.

Figure 1. The experiment from the vantage point of Armenian respondents.

Display full size

After listening to each of the four “guises,” presented in a set order, respondents had a brief pause to indicate their ratings on a questionnaire. This allowed us to compare their evaluations and make inferences under controlled circumstances. On the one hand, we can learn if the same ethnic persona is seen in a different light depending on which language she speaks (vertical comparisons). On the other hand, we can learn if speakers of the same language are seen in a different light depending on their ostensible ethnic origin (horizontal comparisons). The experimental design, thus, enables us to calculate the costs or gains that our respondents associate with shifting ethno-linguistic identities.

Vertical comparisons reveal attitudes to linguistic integration. We can compare the upper left corner (eAsA) to the lower left corner (eAsG) to see if Armenians punish peers for adopting the state language. When comparing the lower right corner (eGsG) to the upper right corner (eGsA) we learn if Armenians reward Georgians for reverse integration. Horizontal comparisons reveal attitudes to cultural assimilation. We can compare the lower left corner (eAsG) to the lower right corner (eGsG) to see if Armenians punish peers for nationalizing their names. When comparing the upper right corner (eGsA) to the upper left corner (eAsA) we learn if Armenians reward Georgians for reverse assimilation.

The premise of the experiment is that a speaker’s language and name serve as a “convenient cognitive shorthand for rapidly inferring a wide range of information about a person.”88 But it presumes that respondents pick up on these cues – without noticing the ruses involved. We took multiple measures to prevent them from realizing that the same recording recurred under different names, and that different speakers voiced the same persona. Apart from recording similar-sounding speakers, we inserted distracting filler voices between those guises that were pertinent to the experiment and kept identical speakers and personas as far apart from each other as possible.89 As a result, respondents showed no signs of detecting the ruses involved.

In 2011, we administered the experiment among pupils in grades 10–12. Some findings from the resulting dataset have been reported before.90 We draw here on different parts of the same dataset, focusing on the 529 high school aged students, born between 1993 and 1996, who identified as “Armenian” at the end of the questionnaire. Since we targeted public schools, encompassing adolescents from all walks of life, we treat this sample as representative of the first generation of Armenians beached in Georgia. Yet, their profile differs depending on location. Out of the 426 respondents from Javakheti, 99 per cent were born in the region. 88 per cent chose to complete the questionnaire in Armenian and 12 per cent in Russian. 94 per cent spoke Armenian at home; 4 percent spoke both Armenian and Russian in the household. The pupils fell back on the same languages when socializing: 88 per cent spoke Armenian to all their friends; another 6 per cent each at times used Russian or Georgian for this purpose. And they rated their Armenian as much better than their Georgian skills (5.81 vs. 3.89 on a scale from 1 to 6).91

Out of the 103 respondents from Tbilisi, 83 per cent were born in the capital. 81 per cent completed the questionnaire in Russian and another 18 per cent in Georgian. Most spoke Russian (27 per cent) or Armenian (17 per cent) or both languages (31 per cent) at home. But Georgian language usage had made inroads among the remaining quarter of households. In addition, 71 per cent of respondents reported speaking Georgian to some of their friends, suggesting that acculturation has progressed further in the capital than in the borderland. Tbilisi-Armenians, in fact, rated themselves as more adept in Georgian than in Armenian (5.41 vs 3.79 on a scale from 1-6).92 54 per cent even evaluated their reading and writing abilities in Armenian as “non-existent.”

Did respondents from these localities also differ in their attitudes to linguistic integration and cultural assimilation? We first constructed an overall evaluation scale from six attributes (amusing, attractive, cultured, educated, intelligent, pleasant) in order to capture composite sentiments toward distinct ethno-linguistic identities.93 We then disaggregated Armenians attending high schools in the borderland from those attending high schools in the capital. Next, we conducted a mixed analysis of variance (ANOVA), treating speakers’ language (Armenian; Georgian) and speaker’s name (Armenian; Georgian) as within-subject factors, and respondents’ location (Javakheti; Tbilisi) as a between-subjects factor. This computation revealed several clear patterns.

We found significant main effects of language94 and name,95 and an interaction between the two.96 However, these were subsumed by a three-way interaction,97 demonstrating that the effects of speakers’ language and speaker’s name varied as a function of respondents’ location. In order to unpack this finding, we shall explore the mean evaluations of the four ethno-linguistic guises, along the 1–6 scale, among our subgroups. Table 1 presents results from the borderland (top panel) and from the capital (bottom panel). Within each panel, mean ratings for the ethnic Armenian speaking Armenian (eAsA) recur in the upper left corner. The lower left corner holds mean ratings for the ethnic Armenian speaking Georgian (eAsG). Mean ratings for the ethnic Georgian speaking Georgian (eGsG) are presented in the lower right corner. The upper right corner contains mean ratings for the ethnic Georgian speaking Armenian (eGsA).

CSVDisplay Table

When scrutinizing the standing of these guises, it becomes clear that respondents share a general preference for their in-group. Both in Javakheti and in Tbilisi, Armenians see “one of us” (eAsA) as most likeable, “one of them” (eGsG) as least likeable, and position guises posing as a member of the other group (eAsG and eGsA) in-between.

Yet, upon closer inspection, important differences emerge. In the borderland, Armenians ostracize peers for adopting the state language, since the Armenian persona is punished when switching from speaking Armenian to speaking Georgian (MeAsA = 5.02; MeAsG=4.70; p<.001). In the capital, the downgrading for linguistic integration is non-significant (MeAsA = 4.92; MeAsG = 4.79; p=.11). However, cultural assimilation is disliked regardless of geographical location. The same Georgian-language speaker is demoted on the composite scale when her Armenian name is substituted for a Georgian name, both in Javakheti (MeAsG = 4.70; MeGsG = 4.27; p<.001) and in Tbilisi (MeGsG = 4.79; MeGsG = 4.07; p < .001), although this effect is more pronounced in the capital.98

Reverse integration and reverse assimilation is rewarded among Armenians from both settings. The Georgian persona receives more positive ratings when switching from speaking Georgian to speaking Armenian, both in Javakheti (MeGsG = 4.27; MeGsA=4.75; p<.001) and in Tbilisi (MeGsG = 4.07; MeGsA=4.69; p<.001). Moreover, the same Armenian-language speaker is attributed higher mean evaluations when dropping her Georgian name in favor of an Armenian name, in the borderland (MeGsA=4.75; MeAsA=5.02; p<.001) as well as in the capital (MeGsA = 4.69; MeAsA=4.92; p<.05).

Figure 2 summarizes the results in graphical form. The left panel presents mean evaluations from Javakheti and the right panel presents those from Tbilisi. For each guise, standard errors of the mean are represented through error bars.

Figure 2. Effects of speaker’s name and language on mean evaluations across localities.

Display full size

The ethno-linguistic evaluative hierarchies in Figure 2 suggest that adolescents from the first generation of beached Armenians share primordial preferences. But the in-group is tighter-knit in the borderland. In Javakheti, both linguistic integration and cultural assimilation results in a demotion. Cultural assimilation is punished in Tbilisi too, but linguistic integration is not: “one of us posing as one of them” (eAsG) might as well be “one of us” (eAsA). The fact that urbanites are more capable in the state language seems to explain this difference.99 When rural Armenians react even against incremental acculturation it thus puts their co-ethnics in Tbilisi in an awkward position.

How did Georgia’s Armenians deal with their predicament as a beached diaspora in the decades after the Soviet collapse? As former union republics proclaimed independence, millions became ethnic minorities “beached” inside nationalizing states and had to decide whether to emigrate to the titular homeland, mobilize on behalf of the in-group, or acculturate into the existing surroundings. Scholars investigating these dilemmas often focus on Russians in the “near abroad,” or on minorities resorting to separatism, thus overlooking the chance to learn from other stranded groups. In this article, we turned our attention to the case of Georgia’s Armenians and examined the role of intragroup differences in processes of “ethnic” mobilization among beached diasporas.

Mixed-methods were utilized to answer the research question. We traced patterns of mobilization through secondary sources and original interviews with Armenian activists; and studied attitudes toward acculturation among the first generation beached in Georgia through a sociolinguistic experiment administered to Armenian adolescents. Even though Georgia’s Armenians belong to the same ethnic group and have faced the same nationalizing policies in the decades after the dissolution of the USSR, we detected differences in their reactions. Armenians in Javakheti, a rural borderland adjacent to their Armenian kin state, mobilized in defence of the in-group while Armenians in Tbilisi, the urban capital of the Georgian state, preferred acculturation.

In Javakheti, where Armenians lived in virtual isolation from Georgians, locals mobilized in defence of their domain from the outset. Activists resisted central control and established informal structures for self-governance that endured until the mid-1990s, when Georgian officials began co-opting influential Armenians into the state apparatus. This social contract generated a double-bind, making it difficult for locals to protest but also for the center to force nationalization policies onto them. As our sociolinguistic experiment demonstrated, the first generation of Javakheti-Armenians “beached” on Georgian soil harbored negative attitudes toward all forms of acculturation – linguistic integration and cultural assimilation – into the national fabric.

In Tbilisi, Armenians lived surrounded by Georgians, who came to dominate the demographics of the capital during the seven decades of Soviet rule. Nationalizing pressures hit Tbilisi-Armenians hard in the outset of the 1990s, causing some to assimilate and others to emigrate. Despite the difficulties facing those remaining, urbanites exhibited lower willingness to mobilize. Their challenge has been to avoid having to make an unpalatable choice between different dimensions of their identities. Our sociolinguistic experiment registered these leanings. As capable Georgian-speakers, Armenian adolescents in Tbilisi refrained from penalizing acculturation into the national fabric, as long as this did not spill over into complete cultural assimilation.

Taken together, this paired comparison demonstrates that Georgia’s Armenians reacted along distinct lines to their predicament as a beached diaspora. Intragroup differences prevented them from sticking together and instead set them on separate trajectories: toward mobilization in Javakheti and toward acculturation in Tbilisi. This empirical finding highlights the problem of groupist thinking that still permeates the literature on ethnic politics.100 “Nationalist leaders seek to mobilize their communities under the banner of homogenous interests,” as Caspersen101 notes, but upon closer examination such claims often turn out to be false.

Past predictions on the non-ethnic determinants behind “ethnic” mobilization alert us to the conditions pulling Georgia’s beached Armenians apart.102 Those from homogenous rural areas, where native language retention is stronger, felt more connected to the ethnic in-group. Due to their compact settlement pattern, Armenians in the borderland also experienced less vulnerability than their peers in the capital in case of conflict. On top of this, Javakheti’s Armenians lacked competence in the titular language and this exposed them, but not their Georgian-speaking co-ethnics in Tbilisi, to more restricted life opportunities.

If the above factors can explain patterns of mobilization and acculturation among Georgia’s Armenians, then intragroup differences might also account for the reactions of other beached diasporas throughout post-Soviet Eurasia. Future research should approach them with close attention to settlement patterns, language repertoires, and other factors that predispose members of the same ethnic group to differing reactions.

Thanks to Marianna Gulanian, Lusine Saghumyan, David Sichinava, Ben Sweeney, and Koba Turmanidze for expediting data collection. A stipend from Sixten Gemzéus stiftelse financed field research and a later grant from Formas (2020-00870) enabled Berglund to write-up these results. We also express our gratitude to the anonymous peer-reviewers for their constructive feedback. Remaining errors are our own.


Christofer Berglund (PhD, Uppsala University) is Assistant Professor at the Department of Global Political Studies, MalmÎ University, where he serves as principal investigator for the project: “Planning for Integration: Landscapes of Power in Borderland Governance” [InBorder]. His research focuses on ethnic politics, competitive authoritarianism, and state-building in post-Soviet Eurasia. His most recent articles have appeared in Europe-Asia Studies, Dynamics of Asymmetric Conflict, and Nations and Nationalism.

Marko Dragojevic (PhD, University of California, Santa Barbara) is an assistant professor in the Department of Communication at the University of Kentucky. His research focuses on language and interpersonal/intergroup communication.

Timothy Blauvelt is Professor of Soviet and Post-Soviet Studies at Ilia State University in Tbilisi, Georgia, and he is also Regional Director for the South Caucasus for American Councils for International Education. He has published several dozen peer reviewed articles and chapters. His co-edited volume Georgia after Stalin: Nationalism and Soviet Power (with Jeremy Smith) was published by Routledge in 2015, his co-edited volume The Transcaucasian Democratic Federative Republic of 1918 (with Adrian Brisku) has just been published by Routledge in 2021, and his monograph Clientelism and Nationalism in an Early Soviet Fiefdom: The Trials of Nestor Lakoba is forthcoming from Routledge later in 2021.

1 Rogers Brubaker, Nationalism Reframed: Nationhood and the National Question in the New Europe (Cambridge: Cambridge University Press, 1996); Lowell Barrington, After Independence: Making and Protecting the Nation in Postcolonial and Postcommunist States (Ann Arbor, MI: The University of Michigan Press, 2006).

2 Vladimir Putin, “Address by President of the Russian Federation” The Kremlin, 18 March 2014.

3 Thomas Greene, Dimensions of Migration in Russia and the Caucasus (Washington, DC: Refugee Policy Group, 1995), 1.

4 Rogers Brubaker, “Aftermaths of Empire and the Unmixing of Peoples: Historical and Comparative Perspectives,” Ethnic and Racial Studies 18, no. 2 (1995): 189–218.

5 David Laitin, Identity in Formation: The Russian-speaking Populations in the Near Abroad (Ithaca, NY: Cornell University Press, 1998), 158.

6 According to the last Soviet census in 1989, Georgia’s population consisted of 70 per cent ethnic Georgians, 8 per cent ethnic Armenians, 6 per cent ethnic Azerbaijanis, 6 per cent ethnic Russians, 3 per cent ethnic Ossetians, 2 per cent ethnic Abkhaz, and 5 per cent ethnic “Others”; Christofer Berglund, “Borders and Belonging: Nation-Building in Georgia’s Armenian and Azerbaijani Ethno-Regions, 2004–2012” (PhD dissertation, Uppsala University 2016): 57.

7 “Javakheti” denotes the Akhalkalaki and Ninotsminda districts in the Samtskhe-Javakheti region, which includes four other districts – Aspindza, Akhaltsikhe, Adigeni, Borjomi – that correspond to “Samtskhe.”

8 Interview with local resident, Akhalkalaki, 29 November 2011.

9 Pål Kolstø, “The New Russian Diaspora: An Identity of Its Own? Possible Identity Trajectories for Russians in the Former Soviet Republic,” Ethnic and Racial Studies 19, no. 3 (1996): 609–39; Laitin, Identity in Formation; Edwin Poppe and Louk Hagendoorn, “Types of Identification Among Russians in the 'Near Abroad’,” Europe-Asia Studies 53, no. 1 (2001): 57–71; Michele Commercio, Russian Minority Politics in Post-Soviet Latvia and Kyrgyzstan (Philadelphia: University of Pennsylvania Press, 2010).

10 Charles King, “The Benefits of Ethnic War: Understanding Eurasia’s Unrecognized States,” World Politics 53, no. 4 (2001): 524–52; Monica Toft, The Geography of Ethnic Violence: Identity, Interests, and the Indivisibility of Territory (Princeton, NJ: Princeton University Press, 2005); Julie George, The Politics of Ethnic Separatism in Russia and Georgia (New York, NY: Palgrave Macmillan, 2009).

11 Svante Cornell, “Autonomy as a Source of Conflict: Caucasian Conflicts in Theoretical Perspective,” World Politics 54, no. 2 (2002): 245–76; Vahram Ter-Matevosyan and Brent Currie, “A Conflict That Did Not Happen: Revisiting the Javakhk Affair in Georgia,” Nations and Nationalism 25, no. 1 (2019): 340–60.

12 Brubaker, “Aftermaths of Empire,” 9; Rogers Brubaker, “Nationalizing States in the Old ‘New Europe’ – and the New,” Ethnic and Racial Studies 19, no. 2 (1996): 411–37; Rogers Brubaker, “Nationalizing States Revisited: projects and Processes of Nationalization in post-Soviet States,” Ethnic and Racial Studies 34, no. 11 (2011): 1785–814.

13 Brubaker, “Aftermaths of Empire.”

14 Laitin, Identity in Formation, 29.

15 David Laitin, “Identity in Formation: The Russian-Speaking Nationality in the post-Soviet Diaspora,” European Journal of Sociology 36, no. 2 (1995): 281–316; Laitin, Identity in Formation, 263.

16 Poppe and Hagendoorn, Types of Identification among Russians, 68.

17 Commercio, Russian Minority Politics.

18 Among Russians in the “near abroad,” emigration became the default reaction in Central Asia but acculturation eclipsed mobilization in most successor states, apart from Moldova’s Transnistria region.

19 Andreas Wimmer and Nina Schiller, “Methodological Nationalism and Beyond: Nation–State Building, Migration and the Social Sciences,” Global Networks 2, no. 4 (2002): 301–34.

20 Dmitry Gorenburg, “Not with One Voice: An Explanation of Intragroup Variation in Nationalist Sentiment,” World Politics 53, no. 1 (2000): 115–42.

21 Dmitry Gorenburg, Minority Ethnic Mobilization in the Russian Federation (Cambridge: Cambridge University Press, 2003), 167.

22 Toft, The Geography of Ethnic Violence, 38–40.

23 Kyle Marquardt, “Identity, Social Mobility, and Ethnic Mobilization: Language and the Disintegration of the Soviet Union,” Comparative Political Studies 51, no. 7 (2018): 831–67.

24 Sidney Tarrow, “The Strategy of Paired Comparison: Toward a Theory of Practice,” Comparative Political Studies 43, no. 2 (2010): 230–59.

25 Michael Hechter, Alien Rule (Cambridge: Cambridge University Press, 2013), 2.

26 Donald Rayfield, Edge of Empires: A History of Georgia (London: Reaktion Books, 2012).

27 It is doubtful whether the groups thought about themselves in “national” or “ethnic” terms. Even though both were Christians, Georgians and Armenians adhered to different Orthodox rites (an important marker during the Middle Ages) and spoke different languages; Rusudan Amirejibi-Mullen, “Language Policy and National Identity in Georgia” (PhD dissertation, Queen Mary University of London, 2013), 156–169.

28 Ronald Grigor Suny, The Making of the Georgian Nation (Bloomington, IN: Indiana University Press, 1994), 87.

29 Ronald Grigor Suny, Looking Toward Ararat: Armenia in Modern History (Bloomington, IN: Indiana University Press, 1993), 37–41; Timothy Blauvelt and Christofer Berglund, “Armenians in the Making of Modern Georgia,” in Armenians in Post-Socialist Europe, edited by Konrad Siekierski and Stefan Troebst (Köln: Böhlau Verlag, 2016), 69–85. Laitin refers to this as “most-favoured lord” incorporation. See Laitin, Identity in Formation, 60. Tbilisi-born Armenians, such as Levon Melikov and Mikhail Loris-Melikov, served as commanders in Russian campaigns against mountaineer insurgents in the North Caucasus. The latter became Russia’s Minister of Interior in 1880.

30 Stephen Riegg, Russia’s Entangled Embrace: The Tsarist Empire and the Armenians, 1801–1904 (Ithaca, NY: Cornell University Press, 2020), 39.

31 Indeed, locals in Javakheti still speak Armenian of the non-standard Western Armenian Karin-dialect, named after “Karin” as Erzurum was known to its Armenian inhabitants. Richard Hovannisian, Armenian Karin/Erzerum (Costa Mesa, CA: Mazda Publishers, 2003), 48.

32 Stephen Jones, “Georgian-Armenian Relations in 1918–1920 and 1991–1994: A Comparison,” in Transcaucasia, Nationalism, and Social Change: Essays in the History of Armenia, Azerbaijan, and Georgia, edited by Ronald Suny (Ann Arbour, MI: University of Michigan Press, 1996), 441–460; Riegg, Russia’s Entangled Embrace, 74.

33 Suny, Making of the Georgian Nation, 115.

34 Suny, Looking toward Ararat, 45.

35 Oliver Reisner, “Ethnos und Demos in Tbilisi (Tiflis) – Armenier, Georgier und Russen in den Stadtdumawahlkämpfen 1890–1897,” in Gesellschaft als lokale Veranstaltung: Selbstverwaltung, Assoziierung und Geselligkeit in den Städten des ausgehenden Zarenreiches, edited by Guido Hausmann (Göttingen: Vandenhoeck & Ruprecht, 2002), 301–329. Reisner accounts for the confrontation between these competing ethnic nationalisms in the context of the elections for the Tiflis duma.

36 Richard Hovannisian, The Republic of Armenia, Volume I: The First Year, 1918–1919 (Berkeley: University of California Press, 1971), 122.

37 Judith Hin, “Ethnic and Civic Identity: Incompatible Loyalties? The Case of Armenians in Post-Soviet Georgia” (PhD dissertation, University of Amsterdam, 2003), 43; Laurence Broers, “Containing the Nation, Building the State: Coping with Nationalism, Minorities and Conflict in Post-Soviet Georgia” (PhD dissertation, University of London, 2004), 113; Krista Goff, Nested Nationalism: Making and Unmaking Nations in the Soviet Caucasus (Ithaca, NY: Cornell University Press, 2020), 22.

38 In parallel, some Tiflis-Armenians migrated to the Armenian SSR since Soviet housing policies led to the expropriation of the properties of the bourgeoisie.

39 Kaiser Claire, “Lived Nationality: Policy and Practice in Soviet Georgia, 1945–1978” (PhD dissertation, University of Pennsylvania, 2015), 208–209; Kaiser, Claire, “‘What Are They Doing? After All, We’re Not Germans’: Expulsion, Belonging, and Postwar Experience in the Caucasus,” in Empire and Belonging in the Eurasian Borderlands, edited by Krista Goff and Lewis Siegelbaum (Ithaca, NY: Cornell University Press, 2020), 80–94; In 1926, Tbilisi’s population consisted of 38 per cent Georgians, 34 per cent Armenians and 16 per cent Russians. In 1970, it consisted of 58 per cent Georgians, 17 per cent Armenians and 14 per cent Russians.

40 Scott Shane, “Nationalist Leader in Soviet Georgia Turns Georgians Against Minorities,” The Baltimore Sun, March 7, 1991.

41 Berglund, “Borders and Belonging,” 57.

42 Broers, “Containing the Nation, Building the State,” 211.

43 Saakashvili’s policies do not conform to one of Brubaker’s (Brubaker, “Nationalizing States Revisited,” 1786) main “nationalizing” leitmotifs: the idea of a “‘core nation’ or nationality, understood in ethnocultural terms and distinguished from the citizenry or permanent resident population of the state as a whole.”

44 More specifically, a State Ministry for Reintegration Issues, a Civil Integration and Tolerance Council under the President’s Administration, and a Council for National Minorities under the Public Defender’s Tolerance Center.

45 Christofer Berglund, “‘Forward to David the Builder!’ Georgia's (Re)Turn to Language-Centered Nationalism,” Nationalities Papers 44, no. 4 (2016): 522–42; Christofer Berglund, “Accepting Alien Rule? State-Building Nationalism in Georgia’s Azeri Borderland,” Europe-Asia Studies 72, no. 2 (2020): 263–85.

46 Christofer Berglund, “Weber's Secret Admirer in the Caucasus: Saakashvili and the Nationalisation of Georgia's Armenian and Azeri Borderlands,” Nations and Nationalism 24, no. 4 (2018): 1185–206.

47 This period covers the rule of three presidents, starting from Gamsakhurdia through Shevardnadze to Saakashvili. Soon after his electoral loss, the presidential post lost its clout due to constitutional changes.

48 15 semi-structured interviews were conducted between 2010 and 2011 with activists in Javakheti, Tbilisi, and Yerevan, who held official posts, represented non-governmental organizations, or media houses covering Armenian affairs in Georgia. Meetings took place face-to-face in English, Russian, or Armenian (in the latter case with a local translator) and without a recorder to reduce the interviewer effect.

49 Darrel Slider, “Democratization in Georgia,” in Conflict, Cleavage, and Change in Central Asia and the Caucasus, edited by Karen Dawisha and Bruce Parrott (Cambridge: Cambridge University Press, 1997), 156–198; Svante Cornell, “Autonomy and Conflict: Ethnoterritoriality and Separatism in the South Caucasus – Cases in Georgia” (PhD dissertation, Uppsala University, 2002), 163.

50 Hin, “Ethnic and Civic Identity,” 81–82.

51 Spyros Demetriou, “Rising from the Ashes? The Difficult (Re)Birth of the Georgian State,” Development and Change 33, no. 5 (2002): 859–83.

52 As late as 1922, Tbilisi consisted of 37 per cent Armenians, 35 per cent Georgians and 17 per cent Russians; Kaiser, “Lived Nationality,” 208–209. The capital became Georgianized during the seven decades of Soviet rule both in terms of demographics (Broers, “Containing the Nation, Building the State,” 256) and in terms of language usage (Hin, “Ethnic and Civic Identity,” 92).

53 Jones, “Georgian-Armenian Relations,” 455; Regnum, “Armianskii batal’on v gruzino-abkhazskoi voine,” November 4, 2011.; Interview with Abkhaz political analyst, Sukhumi, 23 July 2009.

54 Interview with activist in the Armenian Union Sayat-Nova, Tbilisi, 6 August 2010.

55 Mamuka Komakhia, Armenian Population in Georgia (Tbilisi: United Nations Association of Georgia, 2003).

56 Ghia Nodia, Ethnic-Confessional Groups and Challenges to Civic Integration in Georgia: Azeri, Javakheti Armenian and Muslim Meskhetian Communities (Tbilisi: Caucasus Institute for Peace, Democracy and Development, 2002), 26; Tom Trier and Medea Turashvili, Resettlement of Ecologically Displaced Persons Solution of a Problem or Creation of a New? Eco-Migration in Georgia 1981–2006 (Flensburg: European Centre for Minority Issues, 2007), 25–31.

57 Javakheti’s Armenian population did decline with 7.000 in absolute terms. However, their demographic share increased since other ethnicities (e.g. Dukhobors) left the region in greater numbers.

58 Jonathan Wheatley, Obstacles Impeding the Regional Integration of the Javakheti Region of Georgia (Flensburg: European Centre for Minority Issues, 2004), 12.

59 Most regional officials were Georgians. But inside Akhalkalaki and Ninotsminda, Armenians held almost all leading posts, as district administrators, police chiefs, prosecutors, tax inspectors and judges.

60 Stacy Closson, “State Weakness in Perspective: Trans-territorial Energy Networks in Georgia, 1993–2003” (PhD dissertation, London School of Economics and Political Science, 2007); Vincenc Kopecek, “Trapped in Informality? A Study of Informal Politics in Georgia's Javakheti,” Caucasus Survey 7, no. 1 (2019): 60–78.

61 Igor Rotar, “Tbilisi has only Partial Control over Georgia’s Armenian Regions,” Jamestown Foundation, 15 May 1998.

62 Voitsekh Guretski, “The Question of Javakheti,” Caucasian Regional Studies 3, no. 1 (1998); Interview with activist in Akhalkalaki Virk, 6 July 2010; and with activist in United Javakhk, Akhalkalaki, 18 August 2010.

63 Jean-Christophe Peuch, “Georgia: Tbilisi's Pro-Integrationist Armenians Uneasy As Javakheti Pushes For Autonomy (Part 2)” RFE/RL, 26 November 2002.

64 International Religious Freedom Report (Georgia, Washington D.C.: United States State Department, 2011).

65 Interview with activist in the Armenian Community of Georgia, Tbilisi, 24 August 2010.

66 Interview with activist in the European Armenian Federation for Justice and Democracy, Tbilisi, 1 August 2011.

67 Interview with activist in the National Congress of Armenians in Georgia, Tbilisi, 5 August 2010.

68 Jean-Christophe Peuch, “Georgia: Javakheti Armenians' Call For Autonomy Has Tbilisi On Guard (Part 1)” RFE/RL, 25 November 2002.; Interview with activist in Virk, Akhalkalaki, 6 July 2010; activist in the Center for Reforms and Democracy, Akhalkalaki, 10 August 2010; activist in Sardarapat movement, Yerevan, 8 September 2011.

69 Jonathan Wheatley, The Integration of National Minorities in the Samtskhe-Javakheti and Kvemo Kartli Provinces of Georgia (Flensburg: European Centre for Minority Issues, 2009), 24. Studio RE’s investigative journalism offers a glimpse into the deals used to co-opt Armenian officials. See Studio RE, Politicians’ Violations Destroying Village and its Lake (Tbilisi: Studio Monitori, 2013).

70 International Crisis Group, “Georgia’s Armenian and Azeri Minorities” (Europe Report No 178, Tbilisi/Brussels: International Crisis Group, 2006); Vahagn Chakhalyan, Political Notes from Prison (Yerevan: Amaras, 2010). News reports over these protests can be found in Civil Georgia, “Locals, Police Clash in Akhalkalaki,” 6 October 2005.; Civil Georgia, “Protesters Raid Court, University in Akhalkalaki,” 11 March 2006. article.php?id=12044; Regnum, “Armiane Samtskhe-Dzhavakhka zagovorili o shirokoi avtonomii,” 26 September 2005.; Regnum, “Naselenie Samtskhe-Dzhavakheti razgromilo tamozhni na armiano-gruzinskoi granitse,” 12 December 2005.; Regnum, “Vlasti Gruzii izbavilis’ ot armian-sudei v Dzhavakheti, na ocheredi drugie chinovniki,” 28 February 2006.

71 Hedvig Lohm, Javakheti after the Rose Revolution: progress and regress in the pursuit of national unity in Georgia (Flensburg: European Centre for Minority Issues, 2007); Due to a ban on regional parties, dating back to the 1990s, United Javakhk liaised with the Industrialists in order to field contenders. Such alliances between localized ethnic movements and national parties were rare, since most parties believed it damaged their appeal among Georgian voters.

72 Vahagn Chakhalyan, “Rose Revolution in Akhalkalak,” 2006

73 OSCE, Municipal Elections 5 October 2006: Limited Election Observation Mission Final Report (Warsaw: OSCE/ODIHR, 2006), 11, 18; OSCE election observers noted “significant procedural problems” in Akhalkalaki and criticized Georgia for drawing majoritarian constituencies that diluted the weight of Armenian votes.

74 United Javakhk activists also accused the authorities in Yerevan of abetting Georgian officials in the crackdown; Regnum, “Edinyi Dhavakhk obviniaet vlasti Gruzii i Armenii v sgovore i trebuet osvobodit? svoego lidera,” 15 October 2006. See also Tatevik Lazarian, “Armenian Editor Blames Georgia Police Official For Assault,” Radio Free Europe Radio Liberty, 1 May 2010.

75 Wikileaks, “U.S.-Russia Talks on the Caucasus.” Cable 06MOSCOW6358, 2006.; Wikileaks, “Russian Active Measures in Georgia.” Cable 07TBILISI1732, 2007.; In the absence of OSCE-monitors, Georgian officials suspected that armaments slated for withdrawal had been left in the hands of Armenian agitators. As descendants of refugees fleeing violence in the Ottoman Empire, locals clung to the base as a safeguard against Turkish aggression. Interview with analyst at the Georgian Foundation for Strategic and International Studies, Tbilisi, 1 July 2010.

76 Wikileaks, “Javakheti: An Armenian Perspective.” Cable 06YEREVAN1645, 2006.; International Crisis Group, “Georgia: The Javakheti Region’s Integration Challenges” (Europe Briefing No 63, Tbilisi/Yerevan/Brussels: International Crisis Group, 2011), 14; Yerevan’s moderation, in part, stems from its dependence on Georgia as a transit corridor to Russia. This link is vital to Armenia, since its role in the Karabakh War has led to an Azerbaijani-Turkish embargo. Interview with journalist, Akhalkalaki, 11 August 2010; official at the Ministry of Internal Affairs, Tbilisi, 3 August 2011; activist in the Javakhk Patriotic Union, Yerevan, 2 November 2011.

77 The 1921 Soviet occupation began with revolts in the Georgian-Armenian borderlands. This precedent still resonated since some activists put forth proposals for regional self-determination soon after the August 2008 War. See Sergey Markedonov, “The Big Caucasus: Consequence of the Five Day War, Threats and Political Prospects” (Xenophon Paper No 7. Athens: ICBSS, 2009), 40. In addition to the Russian threat, Georgian officials distrusted the ARF, which had a presence in Javakheti and whose platform calls for the reunification “of historic Armenian lands.” See Ter-Matevosyan and Currie, “A Conflict That Did Not Happen”; International Crisis Group, “Georgia’s Armenian and Azeri Minorities,” 19.

78 Interview with activist in United Javakhk, Akhalkalaki, 18 August 2010; and activist in Yerkir Union, Yerevan, 7 September 2011. De Varennes, The Perils of Being a National Minority: Human Rights and Chakhalyan v. Georgia (Lyon: Yerkir Union, 2012). De Varennes has chronicled the procedural violations marring the trial against Chakhalyan, which resulted in a conviction and a ten year prison-sentence.

79 Interview with Akhalkalaki’s district administrator, Akhalkalaki, 11 August 2010. His stance resonated with that of activists in Javakheti Citizens’ Forum, Akhalkalaki, 12 August 2010.

80 Interview with NGO-activist, Akhalkalaki, 11 August 2010; journalist, Akhalkalaki, 15 August 2010.

81 Wikileaks, “Georgia: A Tale of Two Ethnic Minority Cities.” Cable 09TBILISI1781, 2009.

82 Sopho Bukia, “Armenians Seek Language Rights in Georgia,” Institute for War and Peace Reporting, 22 March 2013.

83 Michael Hechter, Containing Nationalism (Oxford: Oxford University Press, 2000), 35–55.

84 “Urbanites” make more efficient mobilizers than “concentrated majorities” on account of having superior access to funds, media, and information on state policies; Toft, The Geography of Ethnic Violence, 22; Hin, (“Ethnic and Civic Identity,” 81), concurs.

85 Peter Garrett, Attitudes to Language (Cambridge: Cambridge University Press, 2010).

86 The text consisted of a short passage about Euclidean geometry. Laitin, “Identity in Formation” utilized the same passage in his experiments, but we translated it from Russian into Georgian and Armenian.

87 “Arpine Sarkisian” for the Armenian guise and “Tamar Maisuradze” for the Georgian guise. Their surnames contain characteristic ethnic suffixes (-ian and -dze). We paired these surnames with common Armenian/Georgian female first names. The name tags were proposed at a focus group session at the Caucasus Research Resource Centers in Tbilisi and vetted through a pilot trial.

88 Henry Hale, The Foundations of Ethnic Politics: Separatism of States and Nations in Eurasia and the World (Cambridge: Cambridge University Press, 2008), 243.

89 We also organized a pilot trial before fielding the experiment. This alerted us to the importance of shortening the playtime of the recordings to make particularities less noticeable. We also learnt to ensure that respondents could not go back in the questionnaire to compare their ratings for the guises.

90 Berglund, “Borders and Belonging”; Berglund, “Forward to David the Builder!”; Berglund, “Accepting Alien Rule”; Marko Dragojevic, Christofer Berglund, and Timothy Blauvelt, “Figuring out Who’s Who: The Role of Social Categorization in the Language Attitudes Process,” Journal of Language and Social Psychology 37, no. 1 (2018): 28–50.

91 t(415) = 30.50, p < .001.

92 t(101) = 9.92, p < .001.

93 Language attitudes researchers often find that respondents evaluate guises along two dimensions, distinguishing between status (e.g. cultured, educated, intelligent) and solidarity (e.g. amusing, attractive, pleasant). See, for instance, Kathryn Woolard, Double Talk: Bilingualism and the Politics of Ethnicity in Catalonia (Stanford: Stanford University Press, 1989); Marko Dragojevic, Christofer Berglund, and Timothy Blauvelt, “Attitudes toward Tbilisi- and Mingrelian-Accented Georgian among Georgian Youth: On the Road to Linguistic Homogenization?,” Journal of Language and Social Psychology 34, no. 1 (2015): 90–101. However, in prior analyses of the dataset that we draw on here, factor analytic procedures yielded a single-factor solution, suggesting that respondents did not make this distinction. See Dragojevic, Berglund and Blauvelt, “Figuring out Who?s Who.” For present purposes, we therefore averaged respondents’ ratings on the six traits to form an overall evaluation score for each guise. In each case, the composite scale exhibited good internal consistency (αeAsA = .80; αeAsG = .83; αeGsG = .77; αeGsA = .79).

94 F(1,527) = 95.75, p < .001, ηp2 = 0.15.

95 F(1,527) = 119.47, p < .001, ηp2 = 0.19.

96 F(1,527) = 20.70, p < .001, ηp2 = 0.04.

97 F(1,527) = 4.66, p < .05, ηp2 = 0.01.

98 F(1,527) = 5.68, p < .05, ηp2 = 0.01.

99 In order to test this supposition, we added respondents’ self-reported Georgian language skills as a covariate and found that the three-way interaction between speaker’s name, speaker’s language, and location became non-significant.

100 Wimmer and Schiller, “Methodological Nationalism and Beyond.”

101 Nina Caspersen, “Intragroup Divisions in Ethnic Conflicts: From Popular Grievances to Power Struggles,” Nationalism and Ethnic Politics 14, no. 2 (2008): 239.

102 Gorenburg, Minority Ethnic Mobilization, 167; Toft, The Geography of Ethnic Violence, 38–40; Marquardt, “Identity, Social Mobility, and Ethnic Mobilization.”


Emil Lazarian

“I should like to see any power of the world destroy this race, this small tribe of unimportant people, whose wars have all been fought and lost, whose structures have crumbled, literature is unread, music is unheard, and prayers are no more answered. Go ahead, destroy Armenia . See if you can do it. Send them into the desert without bread or water. Burn their homes and churches. Then see if they will not laugh, sing and pray again. For when two of them meet anywhere in the world, see if they will not create a New Armenia.” - WS