Archival Sources, Adeeb Khalid

Historical research on the subject is a small enterprise that functions under difficult conditions. Serious historical work was not possible before the last years of the Soviet period, when Mikhail Gorbachev's glasnost opened up archival access. Until then, what little was written on Islam in the Soviet Union saw Islam as a direct threat to the security of the Soviet state.Footnote ¹ This argument was based on a set of intertwined assumptions: that Uzbeks and Tajiks and other Soviet Muslims were primarily Muslim (and not national); that they could not possibly be good Soviet citizens; and that Islam was primarily a political force. In the context of the Cold War, this was a good thing, of course, and part of the calculus that underlay U.S. support for the anti-Soviet “jihad” in Afghanistan. By the time the Soviet Union collapsed, however, this calculus had already been put to question. Broader fears of Islam as a malign political force (with Iran as a prime example) had come to color thinking on Central Asia. As the states of the region emerged on the global scene in 1991 and 1992, policy thinking in the U.S. had lost much of its fascination with Islam as an anti-Soviet force. The preference now was for a secular order in which Islam would not play a central role.

This was the context in which historians entered the fray. Broader changes in our understanding of Central Asia's Soviet experience made us realize that the Soviet period had been formative in important ways, that it had shaped Central Asians’ understanding of their own identities in significant ways, and that Islam's relation to state power was quite different from what we thought. Over the quarter century since the launch of glasnost, we have come to a new appreciation of Soviet policies toward social and cultural transformation, the Soviet state's role in nation making and the creation of new identities, all of which is absolutely crucial in understanding the place of Islam in Soviet and hence post-Soviet societies.

The Soviet policy toward Islam was one of the first topics to attract historians’ attention. The story in the early period is one of devastation, as the antireligious campaigns of 1927 through 1941 destroyed the infrastructure of Islam. Afterwards, however, the state dictated a modus vivendi in which officially recognized “spiritual directorates” oversaw a limited amount of permitted religious activity. Access to archives in Uzbekistan and Russia has allowed historians to understand the fate of Islam (and Islamic education) in the 1930s.Footnote ² Scholars have also begun to explore the complexities of the religious policies of the post-Stalin era, when the Soviet state managed Islam through spiritual directorates and when Islam and Soviet Muslims acquired a new salience for Soviet foreign policy in the decolonizing world. The archival materials are plentiful but they are all (by definition) generated by the state and most of them are reports of Soviet bureaucrats speaking to each other. The challenge for historians has been to discern the viewpoints of Soviet Muslims. Some scholars, especially those without training in Islamic studies, have simply replicated the voices of Soviet bureaucrats (Ro'i Reference Ro'i2000), but more sensitive work is beginning to present a more nuanced picture (Tasar Reference Tasar2010; Dudoignon and Noack Reference Dudoignon and Noack2014).

Archival access, however, is not always to be taken for granted, and the situation in Uzbekistan has been quite problematic over the last decade. (Also, the records of the KGB and of the Politburo, where the most sensitive decisions were made, are not open.) The contribution that historical research can make to the study of Islam and security in contemporary Central Asia is to make clear that the post-Soviet world did not begin in 1991 and to point to the transformation of the Soviet era. There are many continuities with which to contend. A historically informed understanding of the question of security makes us realize that the most significant point about contemporary Central Asia is not the Islamist danger but the strength of the secular state and popular understandings of the place of Islam that emerged in the Soviet period.

Published Primary Sources, Tim Epkenhans

As a scholar of contemporary history and Islamic studies, my research is to a large extent based on written archival and published sources. Since my doctoral research dealt with modern Iran, I started researching Central Asia only as a post-doc. Compared to Iran, the conditions for researchers in Central Asia (with the exception of Turkmenistan) were much better, until the middle of the first decade of the 2000s, particularly regarding access to respondents, archives, and libraries. I am particularly interested in (Muslim) intellectual traditions, which are predominately discussed in periodicals and manuscripts. Published sources—periodical press and book production—are particularly valuable to explore the emergence of Islamist activism in Central Asia. In the late 1980s, for example, a fierce discussion took part in the periodical press about the role of religion in society. Periodicals, such as Birlik and Mustaqil Haftalik in Uzbekistan or Adobiyot va San'at and Adolat in Tajikistan, and the official government mouthpieces, such as Xalq so'zi and Jumhuriyat offer intriguing insights into the discussion, its motifs, and tropes. While the government labeled any form of Islamic activism as a manifestation of extremism or “Wahhabism,” the urban civil society in Almaty, Bishkek, Dushanbe, or Tashkent was divided over the integration of Islam in the evolving nationalist narrative (cf. for instance Mahmadnazar Reference Mahmadnazar1989).

Simultaneously, a less public debate progressed among the traditional religious elites and their activist contenders about public intervention and the question of an Islamic normativity (“Commanding Right,” see Cook Reference Cook2000). However, the debate within the religious field was published as grey literature—samizdat or perhaps more appropriately, “Islamizdat.” While only the few religious scholars from the SADUM (Spiritual Administration of the Muslims of Central Asia and Kazakhstan) had limited access to the media in the 1980s (the Soviet authorities did not encourage a public discussion on Islam), most of the unregistered mullas and Islamic activists relied on Islamizdat publications to disseminate their views. Samizdat or Islamizdat literature has only been preserved in private archives and many of these archives face an uncertain future due to the generational change and the limited appreciation of their historical value.

At the same time, researchers face increasing difficulties in finding even the officially published literature of the later Soviet Union and early independence. Smaller local libraries are often in decay due to the lack of funding, while the showcase national libraries have started to sort out “disagreeable” literature in order to control and filter the historical memory on behalf of the authoritarian governments. In any case, researchers are usually faced with the challenge of finding titles published in the transitional period between Soviet times and independence, when post-Soviet print runs were very low since authors had (and have) no access to subsidized paper or print shops, and thus many titles do not have an ISBN. Nonetheless, a careful investigation in local and larger international university libraries might unearth relevant literature, which has not been consulted by researchers previously.

An important literary genre and primary source for the trials and tribulation of Islamism in Central Asia are autobiographical texts written by representatives of the Soviet/post-Soviet intelligentsia, as well as Islamic activists. While the historical value of autobiographical texts is undisputed, their interpretation and contextualization in either collective or individual representations of the Self is discussed controversially in the relevant literature (Olick & Robbins Reference Olick and Robbins1998). Autobiographical writing is not primarily the retrospective reconstruction of the author's life or the events covered, but the self-perception, self-reflection, and construction of an identity and social role model, which also reflects flexible and shifting attitudes of personal identities, influenced by the rules and dynamics of the social fields. Greyerz furthermore points out that personal/autobiographical narratives reproduce and create discourses which are embedded in a collective context and that the reconstruction in self-narratives allows us to analyze the “specific cultural, linguistic, material and, last but not least, social embeddedness. Ultimately a majority of these texts [. . .] probably tell us more about groups than they do about individuals” (2010, 281). A good example of the extraordinary value of autobiographical sources are the recent texts published by leading representatives of Tajikistan's Islamic Revival Party (see Husaynī Reference Husaynī2013 or Roziq Reference Roziq2013), who have started to write down their version of the emergence of Islamism in the complex context of decolonization, economic transformation, and nationalist renaissance.

Elite Interviews, John Heathershaw

Elite interviews remain central to qualitative political research in general, and on Political Islam in particular. They often are essential supplements to ethnographic research in political science, international relations, and conflict studies (see Wood Reference Wood and Schatz2009, 129; Gusterson Reference Gusterson, Klotz and Prakash2009, 103). In my work, semi-structured, elite interviews complement desk-based discourse analysis and/or field-based ethnographies to explore with individuals their political place in the discursive environment; that is, how they exercise agency within the structured routines of authoritarian politics. I have long been interested in the place of the Islamic Revival Party of Tajikistan (IRPT) and have followed the increasing pressure it has faced since the early-2000s and its relentless demise under state suppression since 2010. It is here, in the study of the only legal Islamic party in the post-Soviet space, where I have used elite interviews in my research.

As a Western researcher, I was welcomed by senior figures in the party and its regional offices who sought to go out of their way to show that they were neither Islamist nor a threat inside or outside Tajikistan. In 2004, when I began making these contacts, access to the party was straightforward and senior interviewees spoke openly about their experiences of the war, the vulnerable place of the party, their plans for its growth, and their critique of the government. My religion—as a protestant Christian—was sometimes discussed but did not appear to make me more or less an object of suspicion. My research became more ethnographic as I accompanied senior party officials to conferences and seminars, had dinner with them, and put them in touch with foreign universities, international organizations and governments who invited them to speak overseas. In cultivating relationships with the IRPT, and particularly its leader Muhiddin Kabiri, whom I interviewed on several occasions, I recognized that I risked being complicit in their self-representations, and being identified by officials as an apologist for a group that was increasingly being identified as a state enemy. I knew that my IRPT participants were at greater risk and I became more and more cautious about taking meetings, eventually only meeting the party leader himself. However, I considered this problem to be a feature of all fieldwork, including interviewing, among opposition elites in authoritarian Tajikistan. In short, my precarious standing in the research field was determined not by the religious field but the political context.

This point about positionality was echoed in my interview strategy, topic guide, and analysis of findings. In order to achieve “balance” in a highly politicized domain I also spoke to secularist intellectuals and ruling party representatives. But I did not meet with state religious officials as I confined my work to politics not the general regulation of religion. Religion was not epiphenomenal in explaining the party's relationship with the state and its identification as a security threat, but neither was it primary. There was no doubt that Islam was foundational to the IRPT's identity, despite it considering removing “Islam” from its title in negotiations with officials. However, as a nationalist or democratic opposition party—and my interviewees also claimed that it was both these things—it would also have faced similar pressure in the increasingly authoritarian context. For my first round of research, in 2004–5, I used the same topic sheet for all opposition figures, Islamic and secular. For both I deployed the principle of “branching and building strategy”—where early interview responses feed in to revisions of the topic guide for later interviewees—to allow participants to collectively shape my research findings about their discourse (see Checkel Reference Checkel, Klotz and Prakash2008, 121). However, in my “ad hoc techniques” (Kvale and Brinkmann Reference Kvale and Brinkmann2009, 234) and discourse analysis, I found remarkable consistency across both Islamic and secular opposition interviewees with regard to their place in the political system and their views of the regime (see Heathershaw Reference Heathershaw2009, 105–106). It is these findings which have spurred later discourse analytic research, which has probed and problematized the representation of the IRPT, and other non-violent Muslim groups, and their putative status as a separate category of political actors (Heathershaw and Montgomery Reference Heathershaw and Montgomery2014). This analysis is consistent with other research on Tajikistan which has shown that religion, ideology, and even affiliation to the former opposition makes little difference for the survival of reintegrated elites within the post-war regime (Driscoll Reference Driscoll2015). In my methods, and in my findings, Islamic opposition elites were firstly opposition figures in an authoritarian system—with all the challenges of access, ethics, and analysis that research on such a subject group entails.

Discourse Analysis, Edward Lemon

Discourse analysis remains a vital tool for those researching the emergence and consolidation of a narrative linking the Islamic resurgence in post-Soviet Central Asia to violence, chaos, and danger. In the same way that a (post)Soviet habitus shapes the way authors of this discourse frame the relationships between Islam, security, and the state, my own background as a white, British, agnostic, male, who has spent over two years living with Central Asian Muslims who consider themselves to be pious, shapes the way in which I relate to these questions. Having spent time with a number of Tajiks whom the government has labelled “extremist”—many of whom eschew politics and violent tactics—I have become more critical of the regime's assertively secular policies. Whilst such a bias may present an insurmountable barrier for positivist, scientific analysis (King, Keohane and Verba Reference King, Keohane and Verba1994), by reflecting on our own positionality and the assumptions that we bring as researchers, we can address these challenges and still produce valid data (Bourdieu and Wacquant Reference Bourdieu and Wacquant1992).

Discourse is the rule-bounded terrain upon which meaning is produced. In Central Asia, these discourses focus on topics such as what it means to be Muslim and the “proper” relationship between religion and politics. These discourses exist in a range of official and unofficial media from videos made by Tajiks fighting with the Islamic State in Syria to memes mocking Uzbekistan's religious policies and speeches made by Kyrgyzstan's President Atambaev. With such an abundance of material, it is necessary for researchers to conduct sampling to select a range of “texts” from the population. While it would be possible to select these texts at random, purposive sampling offers a better way for the qualitative researcher to map the contours of the narrative. Such a method prioritizes finding “nodal points” within the discourse; these are seminal texts which are often referred to by others (Laclau and Mouffe Reference Laclau and Mouffe1985). Selected texts do not merely report what has happened, but script, spin and frame events in a certain way.

Once a researcher has selected a sample of texts then he or she needs to analyze each one in turn. Discourse analysis is an umbrella term encompassing a range of methods with different epistemological assumptions. At one end of the spectrum, it is possible to use quantitative content analysis to identify how many times phrases like “radical Islam” or “porous border” appear in texts. While this may provide some insights into the manifest meaning of the text (Berelson Reference Berelson1952), it does not take us further in understanding its latent meaning and answering questions of why a text is produced or what kind of assumptions about the world the author makes. The Critical Discourse Analysis (CDA) outlined by linguist Norman Fairclough and others, provides a useful tool for answering such questions (Fairclough and Wodak Reference Fairclough, Wodak and van Dijk1997). However, using this method requires an advanced command of local languages and an in-depth understanding of Central Asian culture.

CDA offers a number of advantages to scholars working on Central Asia and beyond. It allows for a longitudinal study of the emergence, consolidation and re-contextualization of discourses. In the Central Asian context, this involves researchers delving into the archives to examine the “revival” of Islam in the late-Soviet period. Many contemporary sources can be found online on the webpages of government agencies, news sites, and social media. CDA also theorizes “the way social power abuse, dominance and inequality are enacted, reproduced, and resisted within text and talk in the social and political context” (Van Dijk Reference Van Dijk, Schiffrin, Tannen and Hamilton2001, 352). Given the ways in which Central Asian regimes use discourses on religious violence to legitimize authoritarian policies, this focus is particularly pertinent for researchers working in this field. Discourse analysis is a method, but it is also a theory; it involves certain assumptions about the intersubjective construction of social reality through texts (Phillips and Jorgensen Reference Phillips and Jorgensen2002).

While many in the social sciences have been captivated by the “linguistic turn,” examining discourses alone does not help us to answer important questions about the impact the discourse of Islamic danger has on pious Central Asian Muslims (Neumann Reference Neumann2002). Are discourses instrumentalized by elites? Are Central Asians passive consumers of state discourses on the proper relationship between religion and the state? How do Central Asians negotiate and contest these discourses? CDA, with its explicit focus on social practice, can go some way in answering these questions, but researchers need to combine their findings with data generated through other qualitative methods to begin to understand relations between Islam, security, and the state in Central Asia.

Surveys and Ethnography, David W. Montgomery

One of, if not the biggest challenges of research is certainty. How does one know that he/she really understands what is happening and not simply constructing a narrative based more on the researcher's interests than what the subject is actually trying to tell the researcher? As mentioned earlier, bias is unavoidable in all social interactions and the matter of bias applies to researchers as much as it does to pundits. No matter how much we seek neutrality, our interests influence our interpretation. And this applies to both quantitative and qualitative research.

Some of the critiques of past research on Islam in Central Asia, and precisely the critique of security and journalistic coverage of Islam in the region, are related to the bias that frames the analysis. For those to whom religion is not a lived category, seeing Islam as expanding into public life can be seen as threatening and dangerous. Both personally and professionally, I identify with religion and recognize it playing an active role in varied aspects of life, and thus my general research bias is one that pushes back against secularized discomforts with religion. When the implicit argument of some is a need to protect the state from religion, I am inclined to see a need to protect religion from the state. Living a religious life can be a radical way of seeing the world, but also a moral and just way, and states often presume moral legitimacy—at least in a Weberian sense of authority—rather than earn and enact it.

That said, my methodological bias is to favor ethnographic exchange as a way of better understanding what Islam means to people in Central Asia. Ethnographic stories are quite often complex and at times contradictory, but this is reflective of how life is lived locally. Ethnographic work requires long periods of interaction in a place, building up of trust with interlocutors who see your commitment to hearing their story, and an openness to seeing the world through the eyes of others with different biases than you. In terms of researching Islam in Central Asia, this means living with people and getting to know people in the varied contexts of their lives. Religion provides a moral framework for people to understand life, to engage with and endure the joys and challenges life presents, but not all do this equally; there is a great deal of variation regarding individual and community engagement and religion is part of a process of socially navigating the world. To see this, one goes to the mosque but equally—if not more so—to the market, the school, the taxi stand, the tea house, restaurants, celebrations; any place that people go, where they congregate in public and or retreat in private, are places where religion is seen. Time in such places with people engaged in the activities and challenges of everyday life gives context to understanding Islam in Central Asia. One listens, observes, and becomes, as best as possible, part of the community. In such times, there are frequent questions that need to be asked; even when one thinks he/she knows the answer, quite often people give further insights when asked.

There is, however, a problem with ethnographic work in that one never knows exactly how representative the particular is the general. Some researchers methodologically shy away from qualitative research like ethnography, viewing it as too anecdotal or less precise than quantitative research. Surveys are useful to particular ends, especially policy ends, because they remove some of the ambiguity and messiness that individual stories create. However, with the apparent increase in generalizability comes a certain baggage. As Mollinga remarks,

A central shortcoming of survey work, at least when it comes to Islam in Central Asia, is that there is a tension between the fungibility of survey results and the certainty that often gets assigned to those results. Certainty is far easier to maintain if you lose the ethnographic context.

There is a common anecdote about surveys in Central Asia, where local surveyors respond to international researchers by saying, “Tell me the results you want and I will see that you get them.” I have seen some surveyors (a minority, I believe) in Central Asia filling out surveys in a room because they were paid by the number of surveys completed, they knew the results that were wanted, and it was easier to do that than to go door to door doing the surveys. Here again is where bias comes into play. Many of those commissioning surveys have certain goals that aim to place the local context into an incongruent international framework of understanding. Methodologically, many doing survey work begin by seeing religion and secularity in conflictual terms, looking for indicators of threat and danger to suggest how religion is creating problems for society. Thus, in a survey result that suggests 50% of Tajiks support shariʿa law, the information is mostly misleading without an ethnographic context to situate how such a question would be interpreted by respondents. Such large support for shariʿa, which the state and international organizations may view as a worrisome indicator of the rise of Islamic militancy, likely has more to do with supporting Islamic family law than it does with anti-secularity or moving to overthrow the state.

In conducting my own survey on religious practice in Kyrgyzstan, I have been questioned about the numbers of my results being skewed because respondents knew of my interest in religion and were more likely to respond positively to questions about religion. This may be true, but the opposite would certainly apply to state- or NGO-administered surveys. Neither are necessarily wrong, but both could be. The point is to make clear that quantitative data without context hides as objective in ways that can remove the nuances of local understanding and therein be subsequently manipulated to justify skewed understandings of Islamic life.

There are research challenges to doing both survey and ethnographic work on Islam in Central Asia, and the majority of these challenges are best addressed with patience and time. Religion is a personal and intimate aspect of life as much as it is communal and public. As such, understanding it requires the researcher to appreciate the fluidity and inconsistencies with which life is lived, and to acknowledge the challenges inherent in creating narratives of certainty.

Researching Religion and the State in Central Asia

Both Islam and the state profoundly impact the nature of life in Central Asia. They are at times seen as incompatible or in conflict with each other, but such views most often reflect the biases of the observer, be the observer a functionary of the state worried about religion; a religious adherent worried about the state; or a third party—scholar or international organization. Central Asia is an increasingly “difficult environment” for fieldwork on Islam and the state as many states in the region have introduced restrictive new laws which effectively limit freedom of religion (Wall and Mollinga Reference Wall and Mollinga2008). Informally, researchers are facing greater scrutiny and more suspicion when researching most topics of social scientific and humanities research in the region, as early findings from a study of fieldwork risk by the Central Eurasian Studies Society appears to show.Footnote ³ However, Central Asia has not faced the armed conflict which many predicted nor the rise of Islamic terrorism that many have feared. For these reasons, Central Asia remains a region where suppression of and accommodation with Islam by the secular state demands investigation.

To understand security and the needs of security, bias must be recognized in order to critically discern the situation in which Central Asians find themselves. To do this, we have advocated multiple approaches to researching Islam, security, and secularism in Central Asia, looking at the use of archival sources, published primary sources, methods of discourse analysis, elite interviews, surveys, and ethnographic research. All have their own challenges, yet together, a fuller picture of what is “going on” begins to appear. Where the opportunities for social scientific fieldwork in the region become more restricted, so new strategies of research will need to be found including, perhaps, fieldwork with diaspora groups and online (see, for examples, Kendzior Reference Kendzior2011; Tucker Reference Tucker2015). Such approaches do not indicate retreat from the field but following the practice of Islam, the state, and security into its new and emerging spaces.

More than ever, as this forum seeks to demonstrate, research is a collaborative endeavor where knowledge is produced between different scholars, foreign and local, senior and junior. Collectively, through the openness to multiple approaches to understanding the relationship between religion and the state, we can more clearly understand what constitutes security and how best to go about both researching it and, subsequently, achieving it.