On researching ‘Ethnic Conflict’: epistemology, politics, and a Central Asian boundary dispute

Abstract

Providing a critique of alarmist discussions of the danger of ethnic conflict in Kyrgyzstan, and the positivist epistemological assumptions and research practices that underpin them, this article develops an approach to researching ‘ethnicity’ and ‘ethnic conflict’ through the use of focus groups. Uzbeks and Kyrgyz in southern Kyrgyzstan expressed similar views about the closures of international boundaries, framed in terms of ethnicity. However, this was not an essentialist notion, but rather a concept of authentic ‘Uzbekness’ or ‘Kyrgyzness’ predicated primarily on the performance of endogenous kinship practices and Muslim/Soviet notions of class morality, nuanced by geography. These overlaps and discrepancies provide resources for those wishing to articulate visions of future social formations wider than the range of options currently propagated by ethnic entrepreneurs.

Notes

³¹This research was conducted whilst I was at Sidney Sussex College, Cambridge, under the careful supervision of the late Graham Smith and subsequently Alan Ingram, to whom I will forever be grateful. I would like to acknowledge the financial support of The Economic and Social Research Council, Sidney Sussex College, the Dudley Stamp Memorial Trust, and the Royal Geographical Society, which awarded me a Violet Cressey-Marcks Fisher Travel Scholarship. I would like to thank Matteo Fumagalli and Madeleine Reeves for carefully reading earlier drafts of this article, and two anonymous referees for their helpful comments. Finally, as ever, thanks are due to Ian Agnew for assistance with the map.

¹‘O'sh-Andijon chegarasi passajir transporti uchun yopiq nega?’, Mezon, 13 – 20 February 1999, p. 1.

²‘Despite the independence of all the countries of the Soviet Union, borders with non-CIS countries are still thought of very much as external frontiers, while those between CIS states are not much more than administrative boundaries in most cases’ (‘The thin red line?’, Labyrinth: Central Asia Quarterly, 2, 3, 1995, p. 8); see also Megoran (1996).

³This focuses on the period 1999 – 2000 because at no time before or since (until the time of writing) has a political crisis been discussed as having the same potential to fan ethnic conflict. Subsequent research in the Valley between 2004 and 2006 leads the author to conclude that the findings are still valid.

⁴For a comprehensive application of Smith's framework to Central Asia, see Patnaik (2003).

⁵For example Bennigsen and Broxup (1983), D'Encausse (1979, see footnote 14) and Imart (1987). For an acknowledgement of this error and a retrospective assessment, see Rywkin (1994).

⁶Polat (2002, p. 186). As far back as 1997, Bichel commented that, ‘the lack of violent ethnic conflict in Central Asia since independence is especially noteworthy, given the sharp decline which each of these states has suffered in its economy, severe dislocations in the workforces, reductions in patterns of national wealth and in standards of living’ (Bichel 1997, p. 148).

⁷Ulugbek Babakulov, ‘Kyrgyz – Uzbek Border Tensions’, Reporting Central Asia, 96, 3 February 2002 (London, IWPR), available at: www.iwpr.org, accessed March 2006.

⁸For example, Rubin and Lubin (1999) and Tabyshalieva (1999). For a critique, see Megoran (2000a). See also the essays in the special collection by Thompson and Heathershaw (2005).

⁹For example, Büyükakinci (2000).

¹⁰See also Tishkov (1999).

¹¹An emic category is a term of self-ascription that people use about their own lives, whereas an etic category is one devised by researchers that is not used in the same way by the research subjects themselves.

¹²The goal is not to be able to claim to have held discussions with a statistical sample of each group, but rather to explore in detail the responses of some people, whose experiences may have much in common with those of other people.

¹³In their studies Burgess ran 13 focus groups, and Holbrook and Jackson 20 groups (Burgess 1996; Holbrook & Jackson 1996).

¹⁴Space does not allow anything more than a cursory review of these issues: for a fuller discussion of conceptual, ethical, and methodological issues involved in forming and running these groups and analysing data, see N. Megoran, The Borders of Eternal Friendship? The Politics and Pain of Nationalism and Identity along the Uzbekistan – Kyrgyzstan Ferghana Valley Boundary, 1999 – 2000, PhD dissertation, Department of Geography, University of Cambridge, 2002, ch. 5, available at: http://www.megoran.org.

¹⁵‘Tübölük dostuktun baasï 5000 bölkö nan’, Res Publica, 8, 340, 16 – 122 March 1999, p. 1.

¹⁶‘Bügün Kïrgïzstan Bar: Erteng jok bolup ketishi mümkün?’, Aalam, 7, 259, 24 February 1999 – 2 March 1999, p. 1.

¹⁷‘Biz, Türkestan jamaatchïlïgï XXI kïlïmda birigishibiz kerek’, Kïrgïz Tuusu, 187 – 188, 22, 382 – 383, 3 – 6 December 1999. For a discussion of the debate generated by this article, see Megoran (2002).

¹⁸‘Jelezniy zanaves?’, Vecherniy Bishkek, 205, 7361, 22 October 1999, p. 1.

¹⁹With one or two dissenters, the Uzbekistanis Muzaffar and Elip in FG8 and FG14 respectively, and Bolot (Kyrgyz, male) in FG15.

²⁰Prayers held at the house of the deceased.

²¹ Millat and Halq can both be translated as ‘nation’.

²²According to Bohr there were 358,700 Uzbeks in Kyrgyzstan in 1997 but only 200,000 Kyrgyz in Uzbekistan (Bohr 1998, p. 153).

²³This is arguably a weakness of Liu's otherwise invaluable and rich ethnography of Osh Uzbeks (Liu 2002).

²⁴ Jumushchular. I have gendered the term as this captures the traditional class sense of a labouring male that I believe the respondent was evoking.

²⁵A comment by one participant showed the imprint of the Soviet-era discourse of the danger posed by capitalist countries. Muzaffar, an Uzbekistani Uzbek in FG8, argued that outsiders were also involved in causing trouble, pointing the finger of blame at the US and the UK. In particular, he accused the BBC's then Central Asia correspondent, Louise Hidalgo, of purposely exaggerating reports of the border issues, and implied my focus groups had a similar aim. This parallels one opinion widely held in Osh that the violence of 1990 was caused by reactionary agents from Moscow inciting instability to discredit Gorbachev's reforms. Such views are entirely understandable: the role of British imperial agents knowing local languages is attested to both in Central Asian and Western literature. See Pahlen (1964, pp. 108 – 109), Hopkirk (1990) and Sodiqov et al. (2000, pp. 53 – 60).

²⁶For example, Rubin and Lubin (1999).

²⁷The figure excludes participants of FG12 and FG13, who do not originate from the Valley but an adjacent area (Alay). If they are included, the figure rises to 88%.

²⁸This was an opinion that he professed even more emphatically at a subsequent conversation in 2004.

²⁹Anonymous, Bishkek, December 1999.

³⁰As Glenn and Akiner argue, this may be more likely due to ongoing processes of immeration and pauperisation (Akiner 2004; Glenn 2003).