‘IN RUSSIA WE ARE MUSLIMS, IN TURKEY WE ARE GYAVURS’: FLUID IDENTITIES OF THE ARMENIAN- SPEAKING

The study examines the process of re/formation of social, ethnic, and religious identities in the Caucasian Black Sea frontier. It resulted in empirical validation of constructivist paradigm. The author tries to elucidate the creative process of ethnic identity formation, which is, according to the presented empirical data, directly linked to socio-economic surrounding of the group. Admittedly, the Hemshils serve as a vivid example of fluidity and flexibility of social identity. The study tries to show how Hemshils’ ethnicity has been shaped by their historical destinies and how they represent it in their everyday life. Living in the borderlands, subsequent bitter experience of deportation, legal disabilities and social deprivation in the receiving societies have predetermined unstable and situational quality of Hemshils’ ethnic identity. The last few decades have been crucial for survival of Hemshils’ communities. New social conditions set the stage for the rearticulation of their ethnic self-identification. Actually, each member of the group may virtually choose between the habitual Turkish option, ‘domestic’ Hemshil or ‘lost’ one, as they have suddenly realized, and some of them regained their Christian-Armenian identity. The versatile dramatic experience tends to foster fluid type of identities redefining and reiterating them repetitively. The aim of the research is the ethnographic description of the discoursive contexts of those unstable changes. The research addresses social scientists and students as well as anybody who cares of how social identity is molded.


INTRODUCTION
The basics for this article was a research describing and analysing the processes of ethnic self-identification among representatives of Hemshil communities, living relatively densely in Krasnodar Territory's Apsheronsk and Belorechensk districts (Russian Federation) 1 . The fundamental questions are as follows: what significance does ethnic identity have under the conditions of radical social change, and what do behaviour strategies ensue; what social conditions and factors determine community's ethnic (re)orientation?
The article is based primarily on field research materials gathered in the places where Hemshils reside (the villages of Vpered, Erik, Kim, Kalinin, and stanitsa Kubanskaya, all in Apsheronsk district, Krasnodar Territory, Russian Federation; the villages of Sarpi, Akhalsopeli, Kakhaberi, Gonio, Dzharnali, Feriya and Urekhi in Khelvachaur district, Adzharia, Georgia). Qualitative, 'soft' research methods (biographical narratives) in the context of the 'cultural-analytical' tradition were used. Explanatory models in the research are given in the key of a constructivist theoretical paradigm (Berger & Luckmann, 1995;Anderson, 2001;Hobsbawm, 1998;Hobsbawm & Ranger, 1983;. It is normally taken that Hemshils (they call themselves Homshetsi) are Armenian-speaking Turks with an 'non-precise', 'migrating' ethnic identity. The majority of researchers tend to consider Hemshils to be descendants of Armenians from Hamshen region on the Eastern Anatolian coast of the Black Sea who were subjected to forcible islamization (Kuznetsov, 1995, p. 22-25;Megavorian, 1904, p. 365).
Owing to the fragmentary nature of surviving written sources it is possible to construct the history of the islamization of Hemshils in the Ottoman Empire, with all the resulting consequences (for the greater part stabi-lising and vital). Alongside with this, apart from fragmentary reports that the Turkish Hemshils secretly performed Christian as well as Muslim rites, in contemporary daily life Hemshils perform 'strange' rituals or customs that suggest that in the past they may have followed Christian faith 2 . Survivals of Christianity could still be found among the Hemshils at the beginning of the 19 th Century and in some districts, for example Riza, according to academician V. Gordlevskii, even later (Gordlevskii, 1962, p. 329). At the same time researchers noted the development of unbridled zeal on the part of the newly converted, 'neighbouring towns around Trebizond --Of, Siurmene --later supplied Mullahs whose fanaticism was known throughout Asia Minor; they brought to Islam the intolerant fervour of the con-vert…'(p. 144). A strict adherence to Islam was already observed in the Soviet period even among the Laz neighbours of the Hemshils in Adzharia 2 Sources that testify to Hemshil crypto-Christianity include: 'Khamsheny deliatsa na dve chasti. Mnogie iz nikh obrashcheny /v islam/, odnako sobliudaiut khristianskie obychai. Oni ne skupiatsa v prinosheniyakh i /razdache/ milostyni. Pochti vse v den preobrazhenia i uspeniya bogoroditsy otpraviaiutsa v tserkov, zazhigaiut svechi i prinosiat zhertvy za spasenie dush svoikh predkov' (Melikset-Bekov, 1950, p. 166 A.P. Megavorian in "Towards the issue of ethnographic conditions for development of the peoples of the Chorokhsk basin" (1904, p. 367), writes of the islamization of the 'Hamshinli' as accepted fact and discusses the consequences of this event for the 'national character'; 'mass conversion to a new religion and particularly the religion of the dominant political tribe, as in the case of Christian peoples turning to Islam, offering to a disenfranchised infidel all rights, and even privileges, must have produced fundamental changes in their morals, customs and occupations…' From a range of rituals performed unconsciously by Hemshils: using a knife to scratch a cross above the door jamb when accompanying the bride from the house (Kuznetsov, 2000, pp. 240-241); using the edges of two knives to cut a cross shape on the upper part of the door jamb when the bride arrives in the house of the groom (the village of Urekhi, informant Durie Karaibragim ogly, 'to this day we ask the old people, why a cross? No one has given us an answer…'); a type of diagonal cross made by the Hemshils with the edge of the palm when baking bread (interview in the village of Akhalsopeli, hamlet Kakhaberi, village of Feriia, village of Urekhi, Adzharia, Georgia).
Informants explain this in the following way, 'we and our forebears always did it this way, just because it makes it easier to divide the bread into parts'.
Informant Ayvazov Levan, born in 1934, Batumi, 'All our local Muslims stored corn bread (tchabi) in a stone pan and always made a diagonal cross with the hand on the bread, but not an upright cross. No one was able to explain why…' (village of Sarpi) 3 . According to I. Kuznetsov, 'islamization for the Armenians did not only mean wholesale deformation of culture, from domestic and agricultural to one of social organization, but loss of ethnic identity' (Kuznetsov, 1995, p. 22).
The Georgian researcher Tsate Batsashi claims new combinations of Khemsil identity and provenance, stating that among the Hemshils of Lazistan, Laz identity was dominant. He writes that the Hemshins of Turkey give children 'domestic' pet names and nicknames that are clearly Laz, both in meaning and in sound. Apart from that, referring to data of N. Marr, he speaks of Hemshin family names that 'are also Laz both in form and meaning' (Batsashi, 1988, pp. 116, 117) Other versions are put forward in articles by the English researcher Anthony Bryer. In one of the articles he suggests that the Hemshins may, through their cultural and historical origins, be linked to the people known as 'Chan', or in the Greek pronunciation 'dzani' (Bryer renders this as 'Tzan'). In general he tries closely to define the dividing line between the Laz and the Chan. Bryer writes that there is a certain tradition of identifying the 'Chan' with the 'Laz', or more accurately, seeing the 'Chan' as a subgroup of the 'Laz'. He calls this tradition 'Russian' and 'Georgian'. But at the same time he begins his analysis from the standpoint that for ancient and medieval authors these two peoples were manifestly different (pp174-175). The 'Chan', unlike the 'Laz', have disappeared completely as an ethnic group by the Middle Ages. Bryer merely suggests that this group was assimilated in the frontier zone of the Trebizond Empire. Perhaps the Hemshils are one of the ethnic groups who continue an agricultural style of life ascribed to the 'Chan' in various sources. Bryer shows that from the late Middle Ages the 'Chan' are being identified with the 'Laz' in the region (at that time the former having already disappeared as a group with its own distinct cultural, linguistic or other identity, the latter beginning to be called 'Chant' or 'Chan'). Meanwhile the whole Pontos at certain periods in the late Middle Ages appears in Armenian, Byzantine and Turkish sources under the name of Djanik (later, and up until now, the region around Samsun in the West Pontos bears this name), which derives not from 'Chan' (Tzan) but 'Laz' (Chan). Since for the contemporaries outside the Pontos all this territory was seen as being inhabited by people that were culturally very different from the remaining population of Anatolia, the regional name was linked with the ethnonym of a small group from the Eastern Pontos. More-over, the ethnonym is not autochthonous for the Laz but was borrowed, or rather was transferred to them following the disappearance of another 'picturesque Pontic group' of 'Chan' (Tzan). Bryer also indicates that the ethnonym 'Chani' is not a name used by the Laz themselves, but in contemporary linguistics is an external 'Georgian' designation for the Georgian [Kartveli]-speaking population of the Eastern Pontos (p. 174) Such is the case to date, when the whole population of the Pontos in Anatolian Turkey is called 'Laz' (lazlar), and Pontic Greeks who migrated to Greece in 1923 also continue to be called there Laz .
Clearly, the issue of the origins of the Hemshils is fairly complex and disputed. However, in the framework of this article I deal with the concrete mechanisms at play in the change of identity among representatives of the Hemshil communities. The historical discourse is relevant here in view of the opinion that history allows one to uncover the reasons, sources and conditions for contemporary societal changes (Nisbet, 1969, pp. 302, 303).

FACTORS THAT CAUSE MARGINALISATION OF THE HEMSHILS
It has come about that Hemshils are the bearers of more than one identity simultaneously. Which factors can be considered to have influenced that process? Ethnic identity has very frequently (one can say, with a certain cyclical tendency to repetition) played a fatal role in the history of the Hemshils. National concepts, both among their forebears and contemporary Hemshils, emerged from difficult personal experiences in contemporary culture, where national identities led and continue to lead to really disastrous situations, constituting threat to physical existence. Complex peripetea in the history of the Hemshils were directly linked, to all appearances, with the fact that they resided in a frontier area (Turkey, Georgia, the border with the USSR). That is, they were marginal by definition. It is precisely their location on the border, and the existence of family members in Turkey that were the ostensible reasons for the Hemshils' deportation in 1944, under the so-called 'preventive measure during wartime that rendered essential the 'desirability' and 'loyalty' of the border populations' (Nekrich, 1978, p. 104). This is confirmed by informants. Fate having left them within the borders of the USSR, in theory the Hemshils could reckon on relative stability, in so far as 'regardless of the existence of the "fifth pointnationality" in various documents, ethnic identity was in practice of little significance. Under the 'levelling' Soviet model of socialization, nationality formed part of the background of general daily 'Soviet' culture' (Voronkov & Osvald, 1998, p. 6). However, this concerned everyone except the Hemshils and several other 'undesirable' peoples (including Germans, Greeks, Kurds, Georgian Muslims, Turks, Chechens, Ingush, Karachays, Crimean Tatars and others) who were exiled by the Soviet authorities in the difficult days during the war with Germany. In this instance the Turkish identity of the Hemshils played, rather, a fatal role (in contrast with the reality of the Pontic-Anatolian context), because of the threat posed by political relations with the Turkish Republic during the Second World War.
To this one can also add the factor of the political and economic interests of the higher administration and ordinary inhabitants of Adzharia (in the Georgian Soviet Socialist Republic). These factors were not decisive, but in the long term the strategies that they dictated played a negative role, impeding the Hemshils' return to the places where they formerly lived, as well as their access to compensation for loss of property. Here we have in mind the local population (Georgians, Laz) who secretly had an interest both in ethnic 'cleansing' (the Georgian administration), and in acquiring the property of the hardworking, usually prosperous Hemshils (some ordinary citizens) 5 . On 25 November 1944, in accordance with decree No. 6279 of the Committee of State Security (KGB) of the USSR, dated 31 June 1944, the Hemshils were deported as an ethnic group in the extrajudicial, administrative, absolute manner of the time 6 .
Having lived through the 'Special Settlement' regime, and recovering after the rehabilitation in 1956 that gave them equal rights with all other Soviet citizens and removed the stigma of 'traitors to the homeland' 7 the Hemshils were faced with new destructive problems linked with the socalled 'parade of sovereignty', the disintegration of the USSR and ethnic conflicts in the 'independent' states on the periphery of Russia. Under pressure from nationalist bandit groups in Kyrgyzstan, the Hemshils were again forced to abandon their homes and moved to Krasnodar Territory 8 . 5 Although the fact remains that, according to informant testimony, the process of settling direct neighbours in the Hemshils' homes was neither completely simple nor smooth. The local inhabitants, bound to the newly exiled Hemshils by ties of friendship and neighbourliness, often refused to move into the Hemshils' empty houses. 'The Georgian government was forced to disseminate propaganda and invite young families from the mountainous village of Khulo, who also were not completely willing to be taken from their large patriarchal families (informants from the villages of Akhalsopeli and Feria). 6 According to the data of N. Tsetskhladze, the deportees from the Khaliachaur district comprised 152 Hemshin families, the total being 1087 people. In November 1944, 304 Khenshin families were deported from Adzharia, including 231 families from the Batumi district, 6 from Batumi itself, 34 families from Kobuleti district, 5 from the Keda district, 28 from the Khulo district (Mgeladze & Tunadze, 2003, p. 272). 7 A. Ossipov (unpublished manuscript), 'In 1958 seven families tried to return independently to the village of Avga, but the Adzharian authorities did not allow them to do so. A little later others repeated the attempt, also without success. In June 1960, thirty Hemshil families who had settled, after all their hardships, in the Lanchkhut district of Georgia, were forced out of the republic. For the next almost 30 years the Hemshils were forced to remain in the places to which they had been deported, Kyrgyzstan and Kazakhstan.' 8 Concerning the theoretical (statistical and census data) and real (figures gathered by researchers and community activists) number of Hemshils, the following constructions exist: Hemshils lived within the borders of the Adzhar ASSR, right up to 1944, and according to the All-State Population Census of 1926 numbered 627. In November 1944, at the moment of deportation to Kyrgyzstan and Kazakhstan they numbered 1,400. Subsequent censuses did not allocate them a separate 'nationality'. In 1983, 12 Hemshil families came to Krasnodar Territory. According to A. Ossipov's data (contained in an unpublished manuscript), after the Uzbek-Kyrgyz conflict in 1989 another 150 families came to Krasnodar Territory. By the beginning of the 1990s, for various reasons, above all deterioration in inter-ethnic relations in Krasnodar Territory, the Hemshil migration had basically come to an end.

PARADIGMS OF HEMSHIL IDENTITY, THEIR NUMBERS AND SITUATIONAL CHARACTER
From the beginning ('as far as we ourselves can remember'informants) the Hemshils are carriers of a double identity, Turkish (based, above all, on religion) and Hemshil (based on language). Hemshils speak in two domestic languages, Turkish and Hemshil, although thanks to their knowledge of Uzbek, Kyrgyz, Kazakh and Russian the group is multilingual. Also worthy of note is the hypertrophied Muslim identity among many Hemshils, above all among older people (Sunni Muslim in character), whose reinforcement can most likely be attributed to contact with the Muslim cultures of Central Asia during the period of exile (1944 to 1989). At the same time it is well known that unsuccessful attempts were made to establish contact with Armenia with the aim of facilitating the removal of several groups of Hemshil intellectuals from Kyrgyzstan 9 . As regards Turkish identity, this is fairly robust and tends to be in the foreground, above all in social situations. According to field research by V. Kyrylev, two young Hemshils (both born in 1960) introduced themselves upon meeting as Turks, and only later added that according to their passports they were Hemshils (Kurylev, 1992, p. 29). With regard to the 1980s migration to Krasnodar Territory 10 , this group's sense of identity is even more complex, revealing nomadic forms of identification. This is above all linked to the fact that after many centuries the Hemshils again found themselves intertwined with their original historic (Pontic) context. To the great surprise of the Hemshils themselves 11 , they discovered that their 'original' language that they called Hemshin (homshesma) was completely comprehensible and used by the neighbouring 'amshen' Armenians. However, as confessional identity among the Hemshils is strong, one observes a tendency among ordinary Hemshils to further strengthen a robust Turkish (Muslim) identity and to distance themselves from their Armenian neighbours. It is likely that the Hemshils experience all the 'unbelievable paradoxality' of the phenomenon of the 'Armenian Muslim' given the connotations of Armenian history relayed by their neighbours in the course of everyday life. Nevertheless, informants also expressed themselves as follows, 'Our native tongue is homshesma, we were Armenians but became Muslims, but I have no idea how that came about…' 'Our native tongue is Hemshil, it's like Armenian' (Kim hamlet). Turkish identity and the degree to which it is fixed depends also, according to interview data, on profession and on the degree of intensity of people's links with the Turkish economy and with other Turks. However several issues arise. 'In Turkey people often say we are not real Turks, and ask who we are. I usually answer, by nationality I am a Turk, my people are the Hemshils. I am a Muslim and so is everyone there, and that's that!' 'I do business with Turkey. My colleague [in Turkey] says 'you aren't a Turk'. I answer that I am an Ottoman, so there!' 12 (Informant Karaibragimov Mavlud, born in 1971, city of Novorossisk, Russian Federation). There was an interesting incident in Apsheron market. In the middle of an interview with the traders in the 'Turkish' section of the meat market 13 I was called aside by one of the Hemshil meat traders, who begged me 'please, for the sake of the Turks, don't write anything, it will just be 11 The Informants spoke of the situations when they had accidentally discovered for themselves the fact that their language could be understood by Armenians. Informant Salikh ogly Said, born in 1959, in the village of Vpered, tells: 'It was in a train. I was saying to my aunt in Hemshil that she would have to take the upper bunk, because the old lady who was with us in the compartment was too old and sick. She turned out to be Armenian and understood every word. She hugged me and thanked me. I was amazed, how could she understand Hemshil? Because I thought that no one except us Hemshils could understand our language.' Service in the Soviet Army also threw up examples. Informant Salikh ogly Batyr, born in 1968, village of Erik, tells: 'I served in Sevastopol and somehow or other got leave. I went to the restaurant and suddenly got all excited. I heard a guy playing on the piano and singing in Homshesma. I went up to him and said, brother are you a Hemshil? He said he didn't know what I was talking about. I said, 'but I heard you singing in Hemshil'. He said, 'no brother, I was singing in Armenian'. Afterwards I wondered how it was that I could understand Armenian.' 12 In the Ottoman Empire, right up the beginning of the 20th Century, only peasants from the remote parts of Asia Minor were called Turks. Members of the ruling circles called themselves Ottomans. But with the rise of the national movement at the end of the 19th Century the word 'Turk' began to be attached to all categories of the Turkish population (Starchenkov, 1990, p. 85). 13 In the central market in the town of Apsheronsk, in Spring 2002, incidents erupted as a result of which the 'black' meat traders were excluded from the 'Cossack' market as 'aliens', on the crest of a wave of xenophobia. The situation was resolved in the following way. A 'Turkish' businessman built a new 'Turkish' pavilion for the meat traders who had been left with nowhere to sell their meat --Meskhetian Turks, Hemshils, Yezids , Kurds. It is common in Krasnodar Territory for economic competition to take on a racist tinge. worse, we have no one to protect us…' In general the market, and relations at the market, are a very interesting focus for a study of nomadic identity. It is no secret that, at least in the south of Russia, economic activities are carried out on the basis of personal contacts, sympathies and ethnic networks. In such an atmosphere ethnicity becomes a resource. 'Turkish' informants (Meskhetians) bitterly complained that 'Hemshil meat traders pretend to be Armenians and speak in Armenian to Armenian customers and a minute later switch to Turkish and call themselves Turks if a Turkish customer approaches his stall.' (Shamanadze Khalida, stanitsa Kubanskaya) Concerning matrimonial strategies, unions between Hemshils and Meskhetian Turks are rare but do occur (the marriages occurred exclusively in Central Asia, and in Krasnodar Territory they take place no longer); unions between Hemshils and Armeniansnever. The most common mixed marriages take place with Kurds, but they are 'unusual' and not with all Kurds, only so-called Hemshil Kurds (they also call themselves Kurd-Hemshil and live in the town of Khadyzhensk, Apsheronsky district of Krasnodar Territory) 14 . In general, the group follows endogamous practices, often between relatives because of the small size of the community. If one considers marriage strategies to be an identity marker, then such practices in their turn are an indication of a strongly separatist nature of Hemshil identity in private life (but do not necessarily signify that the community's intentions and tendencies are isolationist).
Muslim spirituality and religious practices fill the daily life of the older generation in many families (prayer five times a day (namaz), frequent washing (abdest), annual observance of the month-long fast (ramazan). These rituals imbue every sphere of daily life, and in particular the wedding cycle, the wedding itself (the compulsory nikah or nikoh ritual that is the Muslim equivalent of the Christian wedding ceremony), sexual practices (the compulsory ghusur, ablutions 'from the ends of the hair to the heel' immediately after intercourse). Young Hemshils are either completely indifferent to religion or observe the laws of Islam purely formally. However, and here the ubiquitous Weberian principle comes into play, 'Social agents follow the rules when the advantage of submitting to them is greater than that of breaking them'. The question is, under what conditions is the rule most often applied and are regulatory practices observed? Additionally, if the advantage coincides with submission to the rule, the agent derives a double advantage, a symbolic and a practical or economic one (P. Bourdieu). For example, a Hemshil woman who wishes to protect her son from the dangers of service in the Russian army invests the maximum effort in sending him to study to become a mullah in Turkey. Some Hemshils react negatively to the least allusion to former Christianity or Armenian origins. As a rule, this is again those who derive some advantage from articulating this rather than another identity. Such was the case at a wedding in July 2001 in khutor Kalinin. A young man started an argument about a newspaper article that spoke of the possible Armenian origins of the Hemshils (the newspaper of Armenians in the South of Russia Erkramas, 2001, p. 8) and behaved rather aggressively. His behaviour was provoked by my answer to his question concerning my own nationality. 'We are not Armenians, we were never Armenians and even less Christians! Leave us in peace! You come her telling your tall tales and then we have problems, our business is harmed. You give us no peace…' (Aydin, stanitsa Pshekhskaya).
A degree of marginalisation is often reflected by informants. A Hemshil woman from Batumi complained that a jealous Muslim Hemshil in Kemal-Pasha was still called Ziya Gyavur (gyavur means faithless, an infidel). 'We belong nowhere, that's what is so awful. In Russia we are Muslims, in Turkey gyavurs' (Informant Feyz ogly Pakise, born in 1961 in Batumi). Often it is precisely these spontaneous reflexes, thinking aloud, that reveal the direct process of constructing ethnic identity, 'What is a homeland? Where you are born and grow up. But we don't have that. In this century we are migrants, from Turkey to Batumi, from Batumi to Kyrgyzstan, Central Asia, and here. A homeland is somewhere where people accept you, where they know you, respect you, value you. We don't have that here. We are the smallest people, on the brink of extinction, we either have to attach ourselves to some people or other, or…? Even if we fight to preserve our people, there is very little guarantee of success. And our culture has completely degraded, illiteracy, we have no lawyers, no professors, there are doctors, but… [We] are a people who have lost a lot, but who is interested in us? No one. I think that to preserve a nation you need a homeland, but where can we find one? Where is my homeland now? I was born in Central Asia, my grandfather in Turkey, my father in Batumi, and my son was born here. How can I say where my or my son's or my father's homeland is? It is already impossible to say which homeland is the main one. Adzharia? As for me, I think one has to be attached to one spot, but that will never happen. Already we can't seek our past because our past is in different places. Some say that our homeland is where we were born, let's go there. Fine, but what about our children? They already speak a dif-ferent language, they have been brought up differently' (Informant Salikh ogly Rovshan, born in 1963 in the village of Vpered).
So the formulation of Hemshil identity, the Hemshil ethnic code, is based on an all-embracing marginalisation, the community's exclusion from social networks in the host culture. These divisions have a huge social significance at the micro level. People need to be able to present themselves in society; this influences what sort of niche they can occupy in that society and whether they can occupy one at all, filling the role of outcasts. Formulating this code has a direct relation to real history, which is specifically constructed on 'fate', mobilising collective identity on the principle of 'if they don't accept us in the new society we will create our own internal solidarity'. The result is a closed community. In this sense the choice is predetermined. Social development in the host society is made exceedingly difficult, while social life outside one's own community is almost nonexistent.
It is therefore important, it seems, to differentiate the types of cultures and socio-political systems oriented towards (1) survival and (2) development/improvement. These aspirations are directly related to the level of economic development, and to the political or ideological leanings in the society being examined. In the first case, in the issue of ethnic selfidentification there can be no freedom and abstraction from the daily context. The ethnicity that has given rise to such a quantity of real problems cannot, in principle, be a 'private' ethnic identity, an identity 'for yourself' (unlike the second case), because the social milieu plays too active, not to say repressive, a role. We can therefore say that formation of an ethnic identity is a dynamic process, but in no way an act of free choice. Given that social conditions exert too strong a pressure on communities creating at times insurmountable barriers to integration and assimilation, it is the real contemporary situation prevailing in the Russian provinces.

CONTEMPORARY ASPECTS OF THE PROBLEM: MIGRATING BARRIERS OF IDENTITY AND MARKING OF A NEW MARGINALISATION
The ambiguous situation in which the Hemshils of Krasnodar Territory find themselves has driven the Hemshils themselves into huge confusion and misunderstanding, along with the local population and above all local official structures (which cannot distinguish who is who). In the multiethnic context I have described, where there is a multiplicity of self identities, people are put in a situation where they no longer understand who they are. The Hemshil leaders appear impotent and cannot make a choice that would bring their community 'the optimal economic, symbolic and cultural capital' (Bourdieu, 1994, p. 106). They reside in tortuous vacillation concerning their final choice and legitimization of their ethnic status, a status that would provide their community with stability and prosperity. For this reason there has still been no indication that opinion is unified among rank and file Hemshils. The scales tilt one moment towards strengthening Turkish identity and uniting forces with the Meskhetians with the aim of survival, one moment towards insistence upon the unique nature of Hemshil cultural roots, one moment towards reconstruction, the renaissance of a 'lost' Armenian identity. Alongside this, they are extremely careful in adopting, even less institutionalising, Armenian identity. It is possible that this is because were they to adopt this identity, the Hemshil guilt complex of the potential traitors would increase. To the experience of deportation with the label 'undesirables' and 'enemies of the people', the stamp of confessional inconsistency and 'disloyalty' would be added. Recognising oneself as Armenian would mean accepting that one's ancestors 'betrayed' Christianity and, later, Soviet ideals. All the more so, the fact of a Christian past is hard to reconcile with the Islamic discourse that dominates their daily lives. Intermingled with this is the complex mixture of feelings linked with general Caucasian concepts of masculinity, although this issue is the theme for a separate research.

PUBLIC DISCOURSE THAT FORMS REALITY. PHOBIAS
The Hemshils, as well as the Meskhetian Turks, Kurds and other ethnic groups in Krasnodar Territory fall into the category of one of the most disenfranchised population groups. They are excluded from the status of refugee and from the social protection mechanisms that derive from that status, and deprived of residence registration. The issue of residence registration and citizenship is a central one. Resolving this issue would automatically resolve most of their other problems. However the Krasnodar authorities, true to the categories and concepts of the essentialist paradigm, unambiguously and at a variety of levels refuse legalisation to deprived groups (including the Hemshils). Hemshils, like the Meskhetian Turks and other ethnic groups, are considered to constitute a 'destabilising factor' by regional officials, on the basis of ideological myths and spy-mania. The Governor and his supporters throw out threats of deportation camps which will be the basis for 'expelling migrants'. The fears and difficulties have increased in recent years to such a degree that Hemshils (in particular the older people who know from experience the destructive power that the state machine can unleash) are ready to use any opportunity to demonstrate their loyalty to the local 'host' community. One of the Hemshil elders finished his welcome speech aimed at the young people at a wedding with the fol-lowing words, 'Three cheers for the Cossack ataman Gromov!' (with great enthusiasm). The next speech, also pronounced by a grey-haired elder, was shouted loudly into the microphone, like a mantra, 'We all remember, as we will always remember, that the Russian nation is the greatest in the world and is always in the vanguard.' The infamous racist speech of Governor of Krasnodar Territory A. Tkachev made on central television agitated all the ethnic minorities, but in particular the Hemshils and Meskhetian Turks, because it was they who bore the family names that ended with the syllables pronounced as 'outside the law' (The Newspaper, 2002). Family names, in the context of contemporary public discourse, create a particularly blatant dependency or marker of formal 'ethnic' status. The spiritual leader of the Hemshils stated, 'I am changing my family name and dropping the 'ogly'. I went to the public records office and made an application to Kyrgyzstan on behalf of my children, I have already received the response…' At the level of daily intercourse one's family name, external appearance, accent and other markers give rise to and reinforce inequality. Changing one's family name means ridding oneself of the stigma, at least for children born here, whom it will already be impossible to pick out on the basis of differences in intonation and 'odd' turns of phrase, borrowed from 'domestic' language. Of course marginalisation and a 'rich' experience of social inequality create a need to artificially correct one's identity.
Xenophobia in public discourse and explosions of racism has made the problem of Hemshil identity a contemporary one, forcing them to resolution and a more defined self-identification. The Hemshil leaders initiated a written request to the Russian Academy of Sciences Institute of Ethnology and Anthropology named after N.N. Miklukho-Maklay for a historical certificate defining the status of their ethnic group. 'We, the representatives of the Hemshil people (who call themselves Khomshetsi) are requesting that you assist us in obtaining a historical certificate which will confirm our ethnic origin as a separate Hemshil nation (Khomshetsi). At the moment, for various subjective and objective reasons, our people are designated in a range of official documents Hemshily, Turks, Georgians and so on. The same situation exists with our family names. Some have Turkish endings 'ogly', some Georgian '-dze', some Russian '-ov', '-ev' and so on. The historical certificate is essential to protect our ethnic identity and for submitting, when required, to various state bodies.' (Of course the very fact of the appeal to an authoritative academic society is proof of the current social reality in Krasnodar Territory). The response signed by Doctor of History S.A. Arutiunov from the Caucasus Department of the Institute of Ethnology and Anthropology was as follows, 'The Hemshils, who originated in the Black Sea districts of Asia Minor, in particular Trebizond, must be seen as a distinct people, ethnically close to the Armenians, who converted to Islam in the Middle Ages. They speak their own form of the Amshen dialect of the Armenian language… According to international legal norms the Russian Federation is obliged to recognise the right of the Hemshils to Russian citizenship, to live in any part of Russia, to receive legal and social protection from the Russian federal authorities. …Hemshil family names may have a range of suffixes ('-ogly', '-dze', '-ev', 'ian' and so on) depending on the traditions and history of each concrete family.' (The document is dated 20.06.2002.) However, in the political situation that prevails in Krasnodar Territory, being an Armenian is 'unprofitable' and dangerous (as evidenced by the destruction of Armenian tombstones all over the Territory, for example in Krasnodar and Korenovsk; the unpunished destruction of commercial buildings belonging mainly to ethnic Armenians in Slavianskna-Kubani), as is being a Meskhetian Turk. In an attempt to resolve their ethnic identity, the Hemshil leaders originally chose a strategy of 'independence'. It is very likely that continuing political practices have laid the ground for (or rather, practically created) a new, fully fledged Ethnos' . However, subsequent developments have revealed a powerful thrust towards Turkish identity. This is linked with some activities of the USA State Department in developing a programme under which the Meskhetian Turks of Krasnodar Territory could be received as refugees. Confronted by this situation the Hemshils have again designated themselves as a sub-group of the Meskhetian Turks, 'Hemshil-Turks'. This tactic could also have a result of the keen interest that international human rights organisations have taken in the Meskhetian Turk question. Hemshils have clearly taken this initiative in the hope of quickly receiving status within or outside the Russian Federation, based on a calculation that they will receive the maximum resonance among the international community. This situation offers one the opportunity to see, practically before one's own eyes, how the discriminatory discourse and policies of the Krasnodar authorities, in setting the interests of different ethnic groups against each other, becomes a significant factor in forming an ethnic group identity, including Hemshil identity, that is 'unstable' or 'nomadic'.

HEMSHIL SOCIAL ORGANISATION
In my opinion, the development of an independent Hemshil identity is linked with attempts to set up Hemshil organisations, in essence, an institutional core. In 1994 the idea had not taken hold, as the Hemshil community felt that the problems they were facing could be better resolved alongside the Meskhetian Turks in one organisation 'Vatan'. In 2001 the idea revived and began to come to the fore. All the necessary documents were submitted to the relevant department of the Ministry of Justice in Krasnodar Territory. However, because of a change in the procedure for registering organisations with the Ministry of Justice, the matter remains 'on the back burner' to this day. The problem is not merely one of bureaucratic delays 15 , but also of vacillation on the part of the Hemshils themselves. This vacillation appears to be linked to changes in the current situation, that is, the offer of political asylum that may be issued by the USA to Meskhetian Turks from Krasnodar Territory. The commotion caused in the Territory by the news that 'America is going to take the Turks', evoked a retranslation of the discourse myth about the 'indigenous population' (by which they mean the Cossack population, Russians and occasionally some mention is made of Adygi-shapsugi), and 'non-indigenous' (everyone else). Responding to the news, the Governor of Krasnodar Territory A.N. Tkachev spoke on television and in the press in the following vein, 'This information rather surprises me as no one has approached us officially. If such is the case, the local authority will not object. It is a private matter between the Meskhetian Turks and the Americans. Moreover I want to say that the Meskhetian Turks are not natives of the Kuban. They can take these decisions independently. We will have no objections. On the contrary, we will support this process.' This official reaction could not have been clearer, and one can only imagine the degree of agitation among the Hemshil leaders, who for a period were left with no trustworthy, direct, firsthand information. In such a heightened atmosphere a telephone call from the administration of the stanitsa Kubanskaya (Apsheronsk district) brought the prevailing fear and confusion to its apogee. The member of the Apsheron district council who has responsibility for nationality issues was demanding that lists be compiled within one week of those who wished to emigrate. The Hemshil leaders were above all afraid that they were being misled, that the whole 'show' could be merely a loyalty test 'it occurred to me that this could be a test, someone was dangling bait before us like in 1937. Immediately I was thrown into confusion. I was afraid for myself, my children, for the young people, most of whom want to leave. We were driven into a corner politically.' (Hasan Salikhogly, Hemshil leader) Clearly, the historical parallels were too obvious. The community leaders did not provide the lists, insisting on a meeting where all three sides (the American representatives, Hem-shils and local authorities) could discuss the American proposal. However the lists were not needed as the American initiative, for some unknown reason, was not carried through. And the Hemshil leaders are again talking about registering their organisation… The metamorphoses in Hemshil identity that I have described above appear to have peaked, but the process is not yet at an end. This research vividly demonstrates the extreme flexibility and suppleness of ethnic identity, continually reacting to social change and particularly to social disruption that threatens disaster for the ethnic community. The conclusion is not a novel one. The research illustrates the theses of constructivist researchers, 'Identity changes when for some reason it becomes a problem. …Radical changes in the social structure can be a factor when accompanied by changes in the prevailing psychological reality' (Berger & Luckmann, 1995, p. 289). It is precisely these processes that have been at play and that continue to prevail in the case of the Hemshils.