Milletic Secularism in the Balkans: Christianity, Islam, and Identity in Bulgaria

Abstract

The article introduces the concept of “milletic secularism” which invokes the Ottoman millet system to refer to divergent and competing transnational collective identities, loyalties, and frames of reference coexisting within the same nation-state. These identities are conceptualized as resembling the way religious communities functioned under the Ottoman millet system but in a reverse, upended way, as today Muslims are the minority in a pluralist society and secular state governed on the basis of non-Muslim procedures and values symbolically overarched by Orthodox Christianity. Foregrounding the case of Bulgaria, the article highlights the role of the Ottoman legacy vis-à-vis Orthodox Christian heritage for the accommodation of diversity. Milletic secularism draws on the implicit social knowledge that evokes differing antecedents and values underlying the shared identities of Christians and Muslims. Since the 1990s, after half a century dominated by the “secular religion” of Communism, the intersection of religion and politics in Bulgaria is reshaped by the reemergence of religion as a structural force. Milletic secularism has both integrative and emancipatory potential, fostered and challenged today by a variety of factors. Among them, this article foregrounds the increasingly transnational Sunnī Muslim identity and the ongoing re-Islamization in the form of Ḥanafism and Salafism.

Introduction

On Sunday, January 19, 2017, the newly-elected fifth president of the Republic of Bulgaria, Rumen Radev, took the oath of office during an official ceremony at the Parliament where he addressed the Patriarch Neophyte of the Bulgarian Orthodox Church (BOC) and other officials. Although addressing the patriarch was not unusual behavior for a Bulgarian head of state, Radev was fiercely criticized by Lyutvi Mestan, the leader of Political Party DOST. ¹ This party was established in 2016 by Mestan and other former members ² of the Movement for Rights and Freedoms (MRF) representing a large part of the Turkish and Muslim population of Bulgaria in national politics after the fall of the Communist regime in 1989. Mestan himself, before being ousted from the MRF in December 2015, was its second chair after the well-known politician, secularist Ahmed Doğan who resigned while remaining “honorary chairman.” Mestan (2017) criticized Radev because, while the president correctly “turned with due respect to all officials and institutions, including His Holiness the Patriarch of the BOC,” he “failed to show respect to the heads of other religious confessions in the Republic of Bulgaria,” notably to “the spiritual head of approximately one million Muslims in Bulgaria Dr. Mustafa Hadzhi.” In Sofia, some urban liberals, human rights activists, and journalists defended Mestan’s position, accusing the president of favoring the patriarch. ³

Yet, why did many citizens of an otherwise secular state interpret the speech of President Radev as normal and not as intolerant? Although most of the people are neither “church-goers” nor devout believers, religion remains an essential part even of many non-believers’ identity, which distinguishes the Balkans from Western patterns of religion. This symbolic emphasis on the Orthodox Church can hardly be explained only with a “politicization” or an “instrumentalization” of religion but is rather a socio-structural phenomenon which strongly influences identity construction. Is the foregrounding of a Bulgarian “constituent nation” ⁴ defined in both ethno-linguistic and religious terms something inherent to a continuous Orthodox Christian tradition from Medieval times? Or, is it rather a legacy of the more recent Ottoman experience which, despite the long “de-Ottomanization” (Lory 1985) and “de-Islamization” (Öktem 2011) of the Balkans, remained socially relevant by subtly continuing the patterns of the millet system? How do the increasingly transnational religious identities of today’s world impact this historically established pattern?

This article is an offshoot of my recent and ongoing research on the Islamic currents in the Balkans in entanglement with the Middle East. Together with written sources, it draws on ethnographic observations conducted between 2010 and early 2017, including field trips of various duration to religiously mixed or predominantly Muslim areas of Bulgaria, mostly in the Rhodope Mountains. Some of the comparative insights were stimulated also by field work in the Pomak areas of Northern Greece in 2011 and 2012, as well as in the Serbian region of Sandžak and Bosnia in 2016. I first argue for the need to discuss religion and politics in Bulgaria as indicative of why secularism in the Balkans differs from the Western ideal. I then tackle what is imagined to have been the Ottoman millet system—a historical paradigm for accommodating diversity in a heterogeneous society that had its specific continuation in the Balkan nation-states. I thus introduce the concept of “milletic secularism” as a toolkit for thinking of the structural underpinnings of secularism in the Balkans. Finally, vis-à-vis the thus described concept, I analyze the increasingly transnational Muslim identities today and the ongoing re-Islamization taking place in various forms, such as Ḥanafism (or Ḥanafī-Māturīdism) ⁵ and Salafism. ⁶

Identity, Religion, and Politics

After the fall of the Iron Curtain, religion has re-emerged as a significant element of “individual and national identity in many of the Central and Eastern European countries” (Pew Research Center 2017, 5). Although the Balkans share cultural trends with the broader Eastern European region and elsewhere, they have followed specific patterns of religion in society. Bulgaria is a predominantly Orthodox Christian country with the largest autochthonous Muslim minority within the European Union which varied from 21% in 1887 ⁷ to 10% in the early 21st century, according to the most recent census (2011). Muslims in the country are ethnically non-homogeneous encompassing “ethnic Turks,” Bulgarian Muslims (known as Pomaks), and Roma. Despite the intra-confessional diversity (0.5% are Alevis), the predominant number of Muslims (9.5% of the entire population) identify themselves as Sunnīs ⁸ vis-à-vis 76% Orthodox Christians, 0.8% Catholics, 1.1% Protestants, and 0.2% “other;” 4.7% indicate “no religious affiliation” while 7.1% do not identify themselves in terms of religion. ⁹

Why do the predominant majority of citizens agree with the “special” status of the Orthodox Church as representing Bulgaria’s “traditional religion” despite the constant claims favoring “religious tolerance,” “diversity,” and the secular character of the state? The understanding of this supposed paradox requires a more nuanced consideration of “the proper place of religion” (Asad 2006, 219) without being entrapped by some utopian models of secularism. ¹⁰ Various cultures entail different trajectories of separation, entanglement, and blurring of the lines between “religion” and “politics” the trajectories of which are contingent on two groups of factors: (1) stable and objective recurrent patterns, social rules, and conventions usually defined as “structure”; and (2) “agency” as “the socially constructed capacity to act” (Barker 2003, 427). Individuals are thereby not deterministic “products” of an overwhelming structure but neither are they unlimited “producers” of their social worlds. ¹¹

Analysis of religion in society, and especially Islam, bristle with otherwise well-intended dicta, such as the phrase that “Islam is what Muslims do.” It is conventionally claimed that identities are in constant flux and depend mainly on the subjective choices of individuals and groups, for identities “are not primordial or a fixed essence” (Eminov 2007, 1–2). However, while identities are multiple and religions—non-monolithic, as Michael Cook (2014, xviii) emphasizes, “more mainstream Muslims could easily be put on the spot with regard to disparities between what they currently believe and do and what their general commitment to Islam could be held to imply that they should believe and do.” Thus, the fear of “essentialism,” searched for even where it does not belong and turned into a standardized discourse, particularly in activist research, ¹² obscures the grasp of religion as a structural force. Insisting, then, on a proper regard for structure in the study of religion more generally, I underline the structural pattern which makes Balkan historical trajectories of secularism distinct from those in the West.

In Bulgaria, the collapse of Communism, a “secular religion” with its own “theology” and “sacred order” (Khazanov 2008), led to the crash of its totalizing discourse. New underpinnings of social cohesion were needed which reasserted religion in the social structure. After half a century of “secular religion,” another form of secularism along the lines of historical religious affiliations has been making its way. During the Communist period and thereafter, the Balkan countries have regulated and supervised the religious scene, albeit with different ends in view—to enforce homogeneity or safeguard pluralist democracy (Elbasani 2017, 17).

Signaling this post-1989 tendency, anthropologist Kristen Ghodsee (2009, 208) suggested a thorough case for the concept of “symphonic secularism” foregrounding “an Orthodox ideal of state-church relations” as the “conception of religion” distinguishing the Bulgarian pattern. According to this concept, the “traditionally” recognized religious communities in Bulgaria are free as long as they do not challenge or oppose government primacy. This makes it difficult to strictly measure “religious rights” or “religious freedoms” based on universalist discourses shaped in the West, for in Bulgaria:

Tsars, Sultans, politburos or prime ministers have consolidated their power over ethnically or linguistically diverse populations, by promoting national identity through a church (or a mosque or synagogue) that instills loyalty to the state as part of its ecumenical dogma while also granting autonomy to other religiously defined communities (Ghodsee 2009, 242).

Analyzing the Bulgarian “genealogy of secularism” as evolving primarily from an inherent “symphonic secularism” may, however, be misleading. The notion of symphoneia was developed by Justinian (d. 565) as an ideal of ecclesiastical and political hierarchies (Kapriev 2005, 14; Kalkandjieva 2011, 588–594) and was itself a utopia (Mayendorff 1987, 216), though many have believed in its longevity as a model only “modified” (or “distorted”) in Ottoman times (Louth 2009, 199). Even in Russia, which has sought to acquire the aureole of Third Rome, even though Orthodoxy had the support of the state since its adoption by Kievan Rus’ (988), “church–state relations have never strictly adhered to the symphonic nonpareil” (Knox 2005, 107). Despite the religious revival in post-Soviet Russia, there is rather a “strategic distance, an alliance at a distance; the church itself has developed a special term to describe it—sorabotnichestvo” (Agadjanian 2017, 19), that is, cooperation. Others, like Hovorun (2016, 14), go even further claiming that “the symphony as a partnership between the church and state as two separate entities [did] not exist in Byzantium.” Thus, referring to symphoneia does not suffice to explore why the Balkan trajectories of secularism diverge from those in the West.

Even as an ideal-type, “symphony” does not provide modern Orthodox Christians with a tool for accommodating religious diversity, particularly when a large minority is Muslim. Therefore, instead of viewing Bulgaria’s Muslims as having “traditionally shared the symphonic conception of religion” (Ghodsee 2009, 242), for the sake of the argument, I suggest to rather approach the country’s post-Ottoman Orthodox Christian majority as having evoked some continuous millet patterns of coexistence by resorting to them in a reverse, “reciprocal” way. After Ottoman rule and the rise of modern Bulgaria in the post-1878 era, the new rulers upended the priority of privileged subjects putting Orthodox Bulgarians at the top of the pecking order in the new state. The millet-type of relations has thus been socially transformed and mixed with modern European conceptions, such as nation-building (Katsikas 2013, 64) and secularism.

The outcome has been what I call here “milletic secularism”—a contextually shaped paradigm of social relations which, paradoxically, is based on religious affiliation but does not anymore invoke the original religious content of the term millet. Rather, this paradigm draws upon shared “implicit social knowledge” which is “what moves people without their knowing quite why or how, with what makes the real real and the normal normal, and above all with what makes ethical distinctions politically powerful” (Taussig 1984, 87).

Millet System and Milletism

Milletic secularism evokes the Ottoman millet system which designates here the Ottoman administrative system applied to Christians and Jews after the mid-18th century (Katsikas 2009, 180) and developed in more specific terms during the 19th century. The Turkish word millet deriving from Arabic milla (a religiously defined community or nation) entails various connotations.

First, millet denotes a community, a group of individuals deriving their identity from a shared religious affiliation. In this sense, millet is synonymous with the Islamic umma (millet-i islamiye), or the “ruling millet (millet-i hâkime)” to which Cevdet Paşa (1953–1967, 1: 68) once equated the Islamic millet. The non-Islamic millets along with Muslims made up the traditional four major communities (millet-i erbaa): Muslims (administratively not organized in a millet), Greek Orthodox, Armenians, and Jews. During the 19th century Latin, Catholic Armenian, Protestant, and others were recognized. Second, millet was sometimes used as an adjective designating the doctrine and practice shared by one of these religious communities, e.g., millet worship, ritual, and law, particularly the law of civil status. Third, millet referred to a formal organization of the religious community with its ecclesiastical hierarchy, clerical and judicial organs (Davison 1982, 320).

The core underpinnings of the millet pattern for accommodating diversity thus refer to milletism with its twofold potential—an integrative social force and a venue for a communal extrication, or emancipation. As Mary Neuburger (2004, 33) puts it, “milletism eventually was the ground on which non-Muslim identities would grow, divide, separate and burgeon. Although millet leaders themselves remained essentially Ottoman-loyal, the millet established a basis for rights based on lines of language and religion.” Thus, the movement for church autonomy in Bulgaria resulted in the rise of the Bulgarian Church in the Ottoman capital founded by Bishop Bosveli in 1851. Together with the activity of the elite Bulgarian intellectuals and traders competing with the Phanariots in Istanbul, these endeavors succeeded in gaining in 1871 the official status of a Bulgarian millet (eksarhhâne-i millet-i bulgar) embracing all subjects of the Ottoman state who belonged to the Bulgarian Orthodox Church (the Bulgarian Exarchate).

The historical and doctrinal origins of what became known as the Ottoman millet system are related to the concept of ahl al-dhimma (the “protected,” the “people of the Book”—i.e., Christians and Jews in Islamic law) whose legal origins are traceable to the beginnings of Islam. The centuries of Ottoman rule and the millet system, at least as we know it from the beginning of the 19th century, nurtured a strong sense of belonging defined exclusively by religion. This was reinforced by “[t]he cohesive nature of Islam as a social force emphasizing the notion of a universal Islamic community or umma” (Taji-Farouki and Poulton 1997, 3). Until the 16th century, the Ottomans applied a strategy of “accommodation” (Lowry 2003, 99) known also as istimâlet referring to the Ottoman methods of conquest as explained by Halil Inalcık (1954). These methods involved a set of attempts to persuade, gaining goodwill and coaxing “to signify the early phase of accommodation of non-Muslim populations, thereby, assisting and facilitating Ottoman conquest” (Papademetriou 2015, 54). ¹³ Under these circumstances, the Byzantine Church found itself in a new kind of church-state relations defined metaphorically by Hovorun (2016, 14–15) as a “symphony with the Ottoman state,” or “Ottoman symphony.”

Some scholars have argued that “leaders of the various millets enjoyed wide jurisdiction over their members, who were bound by their own regulations rather than the Shariat (Islamic Law)” (Poulton 2000, 47). However, such a “jurisdiction” of the non-Muslim millets has been largely overestimated, as there is no sound evidence for a legal autonomy of non-Muslim communities sanctioned by sharī‘a law in the Ottoman state despite some administrative prerogatives provided to the millets. Moreover, some Ottomanists, such as Benjamin Braude (1982), have challenged the conventional assumptions of the millet system by suggesting that before the early 19th century, the term millet as such was not officially used as referring to non-Muslim religious communities of Ottoman subjects. In that period, millet only occasionally denoted Christians and Jews but as foreign, sovereign nations—unlike ahl al-dhimma (or dhimmīs) conquered and ruled by Muslims. Hence, millet initially denoted first and foremost the Muslim community in contradistinction to ahl al-dhimma. Braude (1982, 74) thus suggests to speak of a “communal system” rather than of the “so-called millet system” as long as it was not an institution or a group of institutions but “a set of arrangements, largely local, with considerable variation over time and place.” However debatable the origins of the millet system per se might be, the millet was clearly the basic form of subjects’ categorization in the 19th century Ottoman Empire on the basis of which they were granted different privileges or tax exemptions.

The flexibility of the Ottoman Empire in the application of Islamic law across its vast territory involved some obvious exceptions from the general rule treating the communities along religious lines. Some of the core millet rules were, for example, shunned when the Ottoman state categorized the Balkan Gypsies—kepti (or kıptî, probably from “Copts”). An 1831 Ottoman census in Rumelia included them along the established millets (Lonergan 2017, 3) which indicates a highly nuanced approach to the Roma categorized for pragmatic policy reasons in millet terms regardless of their confessional affiliations. ¹⁴ Muslim Gypsies have never gained full recognition by the Ottomans as their equals. Despite Muslim Roma enjoying a higher status than “unbelievers” (kuffār), i.e., Christian Roma (Marushiakova and Popov 2001, 32–33), they remained “neither Muslims, nor zimmis” (Ginio 2004). However indicative for various aspects of the imperial practices, such exceptions were not to imply abandoning the general Ottoman rule categorizing subjects along religious lines. Even after the middle of the 19th century when the recognition of the Bulgarian millet indicated the virtual consideration of the “ethnicity factor” (Katsikas 2009, 180), religion remained the basic category for communal distinction.

When nationalism and secularist nationalist ideology penetrated Southeast Europe during the 19th century from Western and Central Europe, it came across the inextricable bond between religion and identity. This was a legacy of the millet system “issued in an entwining of nationalism and religion” (Poulton 1997, 14). From that time onward, even if reversed in the pecking order, “[e]ach ethnic and religious community in Bulgaria organizes, realizes and supports its internal integrity, which is perceived by the others as alien” (Zhelyazkova 2001, 300). Although Muslims in Bulgaria as elsewhere in the Balkans were non-homogeneous in ethnic and intra-confessional terms, nationalism, nonetheless, preserved for a long time “a frozen, unchangeable, and stultifyingly uniform image of the Muslim community, and consistently dealt with it in millet terms” (Todorova 2009, 177).

The transformation and continuation of those “millet terms” are traceable not only in Bulgaria, for they have spread in countries as different as Israel, which “utilized the old millet system in the nation-building” (Sezgin 2010, 631), and Greece. The legal status of Islam in Greece, particularly of the contemporary Turkish/Muslim minority of Thrace, even allows for using the term “neo-millet” —“a system which keeps alive pre-modern legal divisions based on religion and inserts them into the larger framework of modern citizenship” (Tsitselikis 2007, 355). In Bulgaria today, no “religious minorities” are legally recognized, but the socially implicit pattern of accommodating the various Muslim groups still preserves some millet-like elements.

Milletic Secularism and the Nation-State

Evoking the Ottoman millet system, milletic secularism refers to coexisting, divergent, and competing transnational collective identities, loyalties, and frames of reference within one and the same nation-state. What clearly distinguishes this pattern from other trajectories of secularism is that one of those identities has a dominant role serving as constituent for the state order. The model resembles the way religious communities functioned within the Ottoman millet system but in an upended way, as today Muslims are the minority in a pluralist society and the secular state governs on the basis of non-Muslim procedures and values symbolically overarched by Orthodox Christianity.

The pattern implying a structural link between religious affiliation and collective identity has been adopted by major politicians referring to Orthodoxy. For example, the current Prime Minister Boyko Borissov, once calling God “the Big Boss,” was described by political scientist Anna Krasteva (2015, 428) as “a master of political discourse.” The reverse foregrounding of the Orthodox Christian identity as the overarching national identity brings to the fore the BOC as a vault of statehood explicitly in the Constitution of the Republic of Bulgaria (1991). This does not mean that Bulgaria is not a secular state—the principle of laïcité clearly prevails: “Religious institutions shall be separated from the state” (Art. 13.2). However, while all “religious denominations shall be free” (Art. 13.1), “the traditional religion in the Republic of Bulgaria is Eastern Orthodox Christianity” (Art. 13.3). The Constitution explicitly bans “identity politics” by which “religious communities and institutions as well as faith convictions cannot be used for political ends.” The state thus differentiates the principles of secularism and pluralism from the historical and cultural underpinnings of society. This constitutional distinction has a powerful symbolic value and entails the dichotomy “traditional” versus “alien.” ¹⁵

Muslims in the post-1878 Decades: Neo-Millet and the Rise of Nuances

The foundations of the post-Ottoman accommodation of religious confessions in Bulgaria were laid with the establishment of the Third Bulgarian State through the first Tarnovo Constitution (1879). ¹⁶ Its Chapter 9, “On Faith,” pronounces the “freedom of religious confessions provided that the performance of their rituals does not violate the existing laws (Art. 40).” Structurally drawing upon Russian legal ideas, the content of that chapter was influenced by the Greek Constitution (1864), the Serbian Organic Statute (1869), and the Constitution of the Romanian Principality (1866) to privilege the Orthodox Christian faith as dominant (gospodstvuyushta vera, Art. 37). The Tarnovo Constitution enacted that the Prince and the Heir to the Crown can only be Orthodox Christian (Art. 38), although its final version allowed for an exception but only for the first prince (Battenberg), which the 1893 amendment extended for the next ruler (Ferdinand), for both were Catholics (Nedeleva 2014, 22). The exception was temporarily tolerated only for the so-called inoslavni (members of non-Orthodox Christian denominations), and not for the inovertsi (members of other, non-Christian religions) whose statute shall be enacted by specific laws (Art. 42).

The new Bulgarian state sought to achieve national homogeneity by elaborating policies toward the Muslim population ranging from educational and cultural measures to forced assimilation. Soon after 1878 an overall reduction of the Muslim population occurred in Bulgaria also due to the lower birth rate (Methodieva 2010, 45). Many Muslims, however, remained in the Bulgarian territory though some of them “declined to sign up as Bulgarian subjects and instead tried to register as Ottomans” (Methodieva 2010, 35). The existence of Muslim institutions and their legal status ensued from the recognition of rights in religion-related legislation and the international commitments of the emerging modern Bulgarian state, such as the Treaty of Berlin (1878) and the Treaty of Neuilly (1919). Muslim religious rights in Bulgaria were bilaterally guaranteed by the Istanbul Convention (1909) via the so-called Agreement on the Mufti Question settling the vakıf (waqf) and various arrangements preserving the religious autonomy of Muslims (Eminov 1997, 222).

The 1919 Statute for Spiritual Organization and Rule of Muslims in the Kingdom of Bulgaria summarized comprehensively all rights granted to the Muslim community, including the rules for spiritual administration of Muslims (1885 and 1895), the treaty obligations and bilateral agreements signed between Bulgaria and the Ottoman Empire between 1909 and 1913. The Statute defined the administration of the Chief Muftiship, the regional muftis’ offices, the Muslim confessional communities and trusteeships (Gyudurov 2013, 135). The Agreement somewhat “localized” the office of the Chief Mufti as elected by the country’s muftis – previously he was appointed by the Prince but gained his full legitimacy only after the approval by the Şeyhülislam in Istanbul. This position was strengthened by the subsequent agreement regarding the Muftis as part of the Treaty of Constantinople (1913). This Ottoman control over the Muftiship was balanced by defining the muftis as public servants who could be fired according to the secular Bulgarian legislation.

The decade of military conflicts after the Balkan wars (1912–1913) brought about new changes for a variety of reasons. Together with the unfolding national project, two sets of factors particularly influenced further the neo-millet model in Bulgaria, nuancing it in combination with other, modern categories which eventually transformed it from an explicit to a more subtle, implicit “communal system” underlying a continuous pattern for accommodating diversity. The first was related to the migration factor, including the exchange of populations between Bulgaria and the Ottoman Empire by which thousands of Muslims left Bulgaria while Christians of Eastern Thrace immigrated to the country after 1913. The second set came out of new international regulations established by the Great Powers, which required a guarantee for the rights of minority groups. Although they were not yet referred to as “minorities,” this new regulation “was important for the maintenance of Muslim populations in the new Balkan states” (Clayer and Bougarel 2017, 59).

Thus, if the period before the Balkan wars was characterized by the treatment of the Muslim community in Bulgaria as a homogenous community, more nuanced approaches to the ethnically different Muslim subgroups were gradually elaborated. The religious categories inherited from the millet system increasingly competed with other types of categorization, such as ethnicity and language. In pursuit of its national project entailing an all-embracing de-Ottomanization, ¹⁷ Bulgaria did not simply take the millet system to reverse it, with Christians now on top, but did so by gradually transforming it in entanglement with other, modern categories. As a consequence, Bulgarian state policies started to differentiate between Pomaks, ethnic Turks, and Roma among Muslims.

Pomaks, considered ethnic Bulgarians, were the subject of assimilation policies, including “reconversion” to Christianity (1912–1913) and “renaming” campaigns during World War Two, the 1960s, and the 1970s, which in the 1980s included the Turks. After World War One, the government cautiously provided the Turkish population with a minority status but did not do so with the Pomaks and Gypsies due to their “less distinct” ethnicities (Ivanova 2017, 39). Thus, when reversing the millet pattern, the Bulgarian state reaffirmed in a new way the earlier “Ottoman nuance” in treating the Roma as a separate group. Muslim Gypsies were also not fully embraced by the newly established Muftiship. The approach of the Bulgarian governments to the migration of Muslims was also differentiated. The Turks were either encouraged to emigrate or socially marginalized, while the Pomaks were prevented from leaving the country and were subjected to a set of “integration projects” (Mancheva 2001, 356). ¹⁸ The Communist policies of the so-called Revival Process ¹⁹ were the culmination of the assimilationist approaches in the 20th century.

Before World War Two, sharī‘a continued to be partially applied in Bulgaria—mostly in the areas of family and inheritance law. That situation resembles the Greek neo-millet pattern in which the functions of the Ottoman Mufti issuing Islamic legal opinions (fatwās) and the judge (qadī) “were fused into a new Mufti” (Tsitselikis 2004, 414). In Bulgaria, those rules remained in force until the establishment of the Communist regime which restricted the rights of all religious communities. Institutionally, the relations of the independent Bulgarian state with the post-Ottoman Muslims and the members of other recognized religions were arranged under the prerogatives of the Ministry of Foreign Affairs and Denominations through the Directorate of Religious Affairs. In 1950, it was transferred to the Council of Ministers to be renamed in 1954 the Committee on the Affairs of the Bulgarian Orthodox Church and the Religious Cults, which since 1957 functioned again under the Ministry of Foreign Affairs. From 1990, these activities are a prerogative of the Directorate of Religious Affairs at the Council of Ministers.

After the collapse of Communism, Turkey has been gradually regaining influence over Muslims in Bulgaria. A protocol signed in 1998 by Bulgaria and Turkey ²⁰ established the financial support of the Muftiship from Ankara through the Directorate of Religious Affairs (Diyanet İşleri Genel Müdürlüğü). More recently, the protocol has been the subject of increasing questioning in Bulgaria as the Diyanet has gained an unprecedented role in the domestic, regional, and international policies of Turkey during the governments of the Justice and Development Party (Adalet ve Kalkınma Partisi). ²¹ The protocol, supposedly suspended in 2017, arranged for some financial support of the Muftiship, including sending Turkish teachers to the Islamic religious schools and sending visiting imams and lecturers. While this model is already different from the explicit neo-millet treatment in post-1878 Bulgaria, the 1998 agreement between Sofia and Ankara, nonetheless, revived an earlier pattern shunned during the era of Communism in Bulgaria and Kemalism in Turkey.

Implicit Neo-Milletism in a Comparative Perspective

The functions of the present Directorate of Religious Affairs are regulated under the Denominations Act (Zakon za veroizpovedaniyata) in force since 2002 which reconfirmed the statute of Orthodox Christianity as the “traditional religion” represented by the BOC existing ex lege unlike all other confessions which have to register at Sofia City Court. The law was criticized by representatives of the Muslim community and the Muftiship already in the stages of its public discussion (Editorial of Myusyulmani 1999). The now DOST leader Lyutvi Mestan, then head of the MRF Parliamentary Group, resorted to the language of secularism and democracy to politically oppose the special status of the Orthodox Church together with some liberals and human rights activists who tried to prepare an alternative draft law.

However, the then leader of the MRF Ahmed Doğan had another vision of national cohesion, so his party did not officially oppose the new Denominations Act. ²² On the other hand, the Parliamentary Assembly of the Council of Europe passed a resolution regarding Bulgaria in 2004 to criticize the law though it “did not fail to note that Bulgaria’s Jews did not oppose the law” (Ghodsee 2009, 242) nor did most of the secular leaders of the Turkish-Muslim minority. In terms of milletic secularism, Doğan is thus a politician sticking to its continuous integrative power while Mestan tends to experiment with what has been categorized above as its emancipatory potential.

A comparative look shows that some historically Christian countries in Western Europe which have privileged Catholicism or Protestantism, have until recently not had any practical experience in a long-lasting coexistence with large Muslim communities on their territories similar to the experience of the Balkans. Milletic secularism thus provides an insight into why some Balkan societies took a path different from that of other European regions, though otherwise they all share a decrease in the level of religious observance. In the United Kingdom, for example, that decrease brought about what has been called “the death of Christian Britain” marking an unprecedented “religious decline” (Brown 2009, 16–18). ²³ However, in the Balkans and, more broadly, in Eastern Europe that shared decrease in devotion does not necessarily mean a shared ubiquitous retreat from religion, particularly Christianity, as a cornerstone of collective identities despite the trend toward the “decline of Christendom” ²⁴ in Western Europe.

While in Western Europe religious affiliation today is not a marker of citizenship or nationality, the “milletically” shaped communities and nations built on the combination of ethno-linguistic and religious affiliations function differently. In Greece, for example, even a non-Orthodox Christian—a Roman Catholic, a Baptist, or a member of the Jehovah Witnesses—has been looked at with suspicion and is not considered a “true” Greek citizen (Poulton 2000, 62). In Bosnia, religious identity, too, overlapped with national or ethnic identity for the three major groups: Muslims (Sunnīs), Croats (Roman Catholics), and Serbs (Orthodox Christians). Here, “religion is more than a set of beliefs. It is part of a person’s cultural identity, whether or not one is a believer” (Bringa 1993, 81). During the 2011 census the then Bosnian head (reis-ul-ulema) of the Islamic community Mustafa Cerić and his Sandžak counterpart Muamer Zukorlić issued a fatwā urging the Muslims in Montenegro to declare themselves Bosniaks during the 2011 census (Larise 2015, 208) emphasizing their distinct identity.

Unlike countries such as Britain, in the Balkans, the mass influx after 1989 of religious emissaries and evangelizers from different denominations did not lead to the realization of expectations of the emergence of a “free market of religions” (Hann 2006). The majority have instead preferred to return to “traditional” religious denominations strengthening them in terms of collective identities. Pointing out that until the end of the last century 79% of the Bulgarian population identified themselves as Orthodox Christian, Petar Kanev (2002, 76) stresses that “only a small part of the huge percentage of people calling themselves Christians are indeed religious. Orthodoxy is comprehended as a national and cultural identification, the same as with Islam.” The so-called ethnic Turks have identified themselves as Muslims even in the era of Kemalism in neighboring Turkey where nowadays Islamic identity is increasingly reasserting itself. The Turkish nation was defined “as consisting of Turkish-speaking Muslims, so the evolution of nationalism into the Turkish-Islamic Synthesis in the 1960s and the 1970s went smoothly despite the secular roots of Turkish nationalism” (Uzer 2016, 14). The case of the Bulgarian “ethnic Turks” thus indicates both the strengths and limits of milletic secularism.

Transnational Muslim Identities

Muslim identity has coexisted with various ethnicities and has sought to overarch them as a basis of collective and political identity. Under modern conditions, “despite being wrong-footed by the nationalist paradigm, Muslim identity has proved sufficiently resilient to benefit dramatically from the rapid development of communications, and this enhancement of an old Islamic conductivity has made possible some notable episodes of transnational mobilization” (Cook 2014, 52). “Transnational religious identity” signifies “personal or collective religious identities that transcend enclaves, localities, regions, nations, and states to achieve a wider unity of belief, practice, and community” (Eickelman 2015, 602). Among Muslims in the Balkans and elsewhere, “transnational Islam creates and implies the existence and legitimacy of a global public space of normative reference and debate” (Bowen 2004, 880).

From the perspective of transnational identities, the Pomaks in Bulgaria are an indicative case. The Pomaks split into three types in terms of how they imagine their ethnicity: some consider themselves Bulgarians, others self-identify as Turks, while a third part—as “Pomaks.” And while mainstream historiography asserts that the Pomaks converted to Islam during the Ottoman rule and that “the Bulgarian origin of the Pomaks is an axiomatic fact that is confirmed by the identical language and traditional folk culture” (Georgieva 2001, 304), some of the Pomaks privilege an identity separate from both Bulgarian and Turkish. In so doing, they evoke “myths of their own ancient origins” (Eminov 2007, 17) by which the non-Turkish speaking Pomaks claiming a Turkish identity try to create their own version of history different from the mainstream narrative.

More recently, some insights still awaiting their elaboration into a coherent account founded on more comprehensive empirical data, indicate an increasing number of Pomaks identifying themselves just as “Muslims.” In this sense, their Muslim identity tends to override ethic divisions serving as a basis of political identity that rises above ethnicities and national affiliations. The ethnographer Evgeniya Ivanova notes that throughout the recent years, especially after 2011, even in areas like Yakoruda and Gotse Delchev in Southwestern Bulgaria, as well as in Ribnovo, where previously the Pomaks used to prefer a Turkish consciousness “the ethnic identification as ‘Muslims’ slowly increases—on the expense of the ‘Turkish’ identity” (Ivanova 2014, 160). Although it is so far too early for a more general consideration of whether in that context, too, “Muslim solidarity has not displaced nationalism, but it has established itself as an alternative to it” (Cook 2014, 46), the indications of that clearly observable trend should not be neglected either.

The sociological survey “Attitudes of Muslims in Bulgaria—2011” ²⁵ drew upon a methodology searching for a possible “differentiation of a new ethnicity embracing not only the elites but increasing mass of people” (Ivanova 2014, 131). This externally assumed assignment of Muslim “ethnicity” to some Pomaks in order to presumably explain their self-identification first and foremost as Muslims can, however, hardly solve any major research questions of how Muslim identity interacts with ethnicity and nationalism. The fact that under certain circumstances Muslim solidarity strives to override ethnicity and nationality does not mean that Islam should be seen as an “ethnic” affiliation, whatever some politicians and observers might have claimed regarding cases such as Pakistan or Bosnia.

If Muslims have ever accentuated an ethnic or ethno-linguistic identity in exceptionally positive terms, it was the Arab affiliation which has been regarded as prestigious, for it is related to the most glorious historical moment when Islam emerged at its source. It is not by accident that part of the contemporary accounts spread among some Pomaks on their “pre-Ottoman” rise in the Balkans refer namely to an imagined Arab origin—such an “ethnic” claim could be seen as an expression of a struggle for recognition and implies that they are not only distinct but superior to both major ethnic groups in Bulgaria—Orthodox Christian Bulgarians and Muslim Turks. Despite the organization of the modern world into nation-states underscored by national identity, Muslims have not abandoned the ideal of the umma, however shunned it might have been for a certain period. When introduced to Muslim countries, the nation-state system tried to impose a dominant identity paradigm centered around an ethnicity. However, no dominant identity paradigm has overridden the Islamic identity of Muslims. In fact, in many parts of the Muslim majority world there is an ongoing revival of Islamic social and political movements. While most Muslim scholars (‘ulamā’) until today reject the concept of nationalism as undermining or totally irreconcilable with Islam, “Muslim peoples remember their ummah identity” (Ataman 2003, 89).

Nowadays, the majority of Muslims identify primarily as such in Egypt (59%), Jordan (67%), Pakistan (87%) as well is in Western Europe: 81% of the Muslim citizens of Britain, 69% in Spain and 66% in Germany. In Turkey, the part of those who foregrounded Islam on top of their identity was 51% while French Muslims split almost evenly but, still, 46% privilege Islam and 42% the affiliation to France. By contrast, the general populations in Western Europe are secular and define themselves primarily through the nation to which they belong and then through Christianity. In this complex picture, the level of Muslim identification in those Muslim majority countries compares to the level in Western Europe (Pew Global Attitudes Project 2006, 3–4).

Considering the regional variations in the Balkans, Muslims in Bulgaria seem to be no exception to a transnationally shared pattern in which Islamic identities remain multiple and non-monolithic but religion is brought to the fore for an increasing number of Muslims. In this sense, modern identities among Bulgarians, Greeks, or Serbs should be seen as constructed vis-à-vis the Muslim identities. Instead of leaving religion second to ethnicity or completely abandoning it on the model of some Western Europeans, Bulgarians, Greeks, and Serbs strengthen their shared identity by emphasizing their belonging to Orthodox Christianity as equally important along their ethnic affiliations. This provides them a sense of certainty and distinctiveness vis-à-vis Muslims who do not tend to abandon religious identity either, whatever their ethnic affiliation.

Although the official 2011 census data does not allow for directly judging which identity is central for the different social, ethnic, and religious groups, some tendencies can be discerned from the survey “Attitudes of Muslims in Bulgaria—2011.” For instance, in the Western Rhodopes and the Pirin region of Southwest Bulgaria, Pomaks, though not speaking Turkish, tend to subscribe to a Turkish identity. Thus, from a total of 3,379 inhabitants of the village of Breznitsa, 2,585 defined themselves as Turks as did, respectively, 1,234 of the 1,666 inhabitants of Kornitsa and 1,442 of the 1,602 inhabitants in Luzhnitsa. In the other Pomak areas, the percent of the “Bulgarian-speaking ‘Turks’” (Ivanova 2014, 141) is insignificant. One of the main reasons for choosing to distinguish themselves through a Turkish affiliation among the Gotse Delchev region’s Pomaks might have been the harshness of the Revival Process (Eminov 1997, 106). This contextual entwining of Islam with “Turkishness” is an important phenomenon but it seems to be part of a larger picture in which religion remains the central identity among the Pomaks, around 20% of whom indicate Islam when asked about their “ethnic” affiliation (Ivanova 2014, 176).

A recent continuation of the 2011 sociological survey (“Attitudes of Muslims in Bulgaria, 2011–2016”) ²⁶ conducted in September and October 2016 indicates some further trends for strengthening Muslim identity and an increased communal convergence around some core Islamic values despite the still prevailing diversity in various other respects. For example, 51% versus 48.6% in 2011 declared religion as “very important” in their lives. Only 33.1% versus 42% in 2011 consider “fully acceptable” if the husband shares the family household chores equally with his wife. The “belief in Allāh” remains considerably high despite the slight variation (76.9% versus 80.1% in 2011) while the trust in “foreign factors” is unambiguously headed by Turkey with 68.4%. Most telling for our case is the persistent tendency among the Pomaks for “ethnically” self-identifying as “just Muslims.” Evgenia Ivanova confirms that “already in 2011 the self-identification as ‘Muslims’ in terms of ethnicity was growing and now it continues. Earlier, these were mainly the very religious families while now ‘Muslims’ increasingly replaces ‘Turks’ and even ‘Pomaks’ among the more educated people.” ²⁷

The increasingly transnational religious identities among part of Bulgaria’s Muslims brings about the reemergence of an umma-consciousness competing with loyalty to the nation-state and even to the local Muslim community. Fostered by the ease of travel and the new media, these identities are rapidly reshaped as a result of a more direct and non-mediated access to normative Islamic teachings and leading centers of religious learning, transnationally promoting Muslim solidarity and unity. If under “classical” milletism community leaders and ‘ulamā’ played a decisive role in the process of choosing which religious imperatives shall be privileged while others are shunned, the ongoing “fragmentation of authority” (Eickelman and Piscatori 1996, 131) empowers new voices not necessarily having a formal or high institutional position to speak for “true Islam.”

Turning to those, including some formal community leaders, who try to avoid what Islam requires them to do, the influential Muslim preacher from the area of Velingrad in the Northwest Rhodopes, Hussein Hodzhda (2012), whose video lectures are extremely popular and widely shared online, firmly states:

They commit a dalavera ²⁸ acting not in accordance to what Allāh wants [them to do] but according to what they like [to do]. However, Islam is not a religion which we like but a religion as Allāh likes it. So, we have to conform to the laws of Allāh the Almighty.

Hodzha is an outstanding representative of the younger Pomak generation of revived Muslims who acquired their religious education in the Islamic University of Medina and is defined by some of his coreligionists as one of the most influential Salafīs. But a female Muslim activist from Samokov demonstrates her revived sense of religious identity in a strikingly similar vein: “However we try to imagine the light of true Islam, we will be unable to do so until we read the Admonition in order to draw a lesson. Islam is not what people do but what they should do” (Ismailova 2014, 16).

Such Muslim voices, no matter if they are Ḥanafī-Māturīdīs or Salafī-minded, pursue a re-Islamization “correcting” their local coreligionists who until recently followed supposedly superstitious and syncretic practices. Even though Islam is sociologically “localized” (Elbasani and Tošić, 2017), if we judge from certain fragments of reality, contextual diversity among stricter and more mainstream Muslims today is transformed and increasingly superseded by the “great” Islamic tradition. In social practice, Islam has never been monolithic and in this sense it is “not immune to the processes of religious individualisation and pluralisation” (Clayer and Bougarel 2017, 225). Nonetheless, as in politics, what is possible to do in religion “is generally limited by what it is possible to legitimize. What you can hope to legitimize, however, depends on what courses of action you can plausibly range under existing normative principles” (Skinner 1998, 105). Religious traditions do change but this is a gradual and time-consuming process against a “considerable inertia” (Cook 2014, xv). In the meantime, interpretations are not unlimited and fully contingent on subjective choices.

Conclusion

Milletic secularism is a socio-structural phenomenon related to the accommodation of diversity in the post-Ottoman Balkans. The concept draws on a paradox, as the millet per se remained in the Ottoman past with its original religious content which would hardly allow for secularism. However, embedded in implicit social knowledge, milletism has proved adaptive over time to survive transformed into a pattern of regulating the pecking order in entanglement with the modern European ideas of nation and secularism.

The pattern of milletic secularism entails several structural underpinnings of society. Privileging the identity of a constituent nation at the top of the pecking order, this model assumes “controlled toleration” providing a non-interference in the internal communal affairs of the “other” as long as they do not brush against the primacy of the constituent nation. This stipulates a complex and often subtle relationship between the integrative and emancipatory potential of milletic secularism. As a continuous principle for accommodating diversity, the thus described model unfolds various categorizations along the dichotomy between “traditional” versus “alien” granting the “traditional” status to those communities which have coexisted for centuries. This pattern, therefore, underlies an inherent suspicion to such “new” or “alien” transnational religious identities and loyalties that challenge the established order and social hierarchies of communal coexistence.

Encompassing a set of principles for regulating the coexistence of divergent collective identities within one and the same nation-state, milletic secularism would emerge where Islam spreads establishing “real” or “imagined” Muslim jurisdictions. Although the above analysis brought to the fore a Balkan case, the concept of milletic secularism or a version thereof seems employable beyond Southeastern Europe. Apart from other Balkan or Middle Eastern post-Ottoman contexts, the term might be useful in the analysis of societies arising from the contiguous empires of the 19th century, such as the Austro-Hungarian Empire and the Russian Empire, where the rise of the nation was combined with religious affiliation and separatist movements. Expanding the concept in such a way, however, requires a comparative study of interrelated cases by upending the pattern described here, for Islam has been present and encountered there but has not continuously dominated the social order for centuries as in the Ottoman Empire.