Read an Excerpt

Chapter 1
Introduction

The word exceptionalism was born of politics. In its earliest incarnation, the term was invariably prefaced by the qualifier American and used by leftist intellectuals to describe the apparently unique ability of the United States to avoid class warfare. Muslim exceptionalism, on the other hand, is a much younger term that first earned currency in political science in the 1990s (Pipes 1996). Yet it bears a conceptual pedigree that easily predates Karl Marx. The idea that something sets Muslim politics and society apart from the politics and society of everyone else is the hallmark of Orientalism, a one-way conversation started by European elites in the eighteenth century (Irwin 2006). However, following the publication of Edward Said's withering magnum opus Orientalism (1978), much self-conscious scholarship may have eschewed the idea of Muslim exceptionalism for fear of committing academic heresy or even a "thought crime" (Kramer 2006). This is clearly no longer the case.

Even before the terrorist attacks of September 11, 2001, the general belief that all Muslim societies are built on the bedrock of a shared, immutable, and alien faith influenced "dominant attitudes in academia and, with much more devastating effects, in the media" (Filali-Ansary 1999, 18). These attitudes hardened in the 1990s, when some observers noted that Muslim countries missed the "Third Wave" of democratization in the 1980s (Huntington 1991, 281). Moreover, following "the end of history," that is, the collapse of the Soviet Union (Fukuyama 1992), and the ideological bankruptcy of communism, autocratic regimes were replaced with representative governments everywhere, it seemed, except in the cradle of Islam, the Middle East (Salame 1994). In the years since 9/11, well-meaning Western proponents of interfaith tolerance—including academics, journalists, and policy makers—have tried to promote a more nuanced understanding of Islam and (to a lesser extent) the diversity of the Muslim world. Yet there remains a persistent view—on both sides of the Atlantic (Nussbaum 2012)—that the politics of all Muslims can be explained by the influence of Islam. This is evident not only in some conservative media and foreign policy circles, but also in the world of security affairs, where Muslims are perennially viewed with suspicion and alarm (Croft 2012; Kaya 2012). In short, and despite the best efforts of many civil society actors, the view that Muslims are exceptionally problematic politically is at the heart of a contentious and continuing debate (Mandaville 2013; see also, e.g., Bayat 2013; Strindberg and Warn 2011; Elshtain 2009).

Like Orientalism, the term "Muslim exceptionalism" is applied and interpreted inconsistently. One way to parse what is or is not exceptional about Muslims is to explicitly examine specific political or social variables in contrast to non-Muslims. This approach, adopted by Steven Fish, delivered a number of conclusions. In a robust comparative study, he argues Muslim societies, when compared to non-Muslim societies, are more averse to homosexuality and, to a lesser degree, averse to abortion and divorce. In addition, in societies with proportionately larger Muslim populations, both murder rates and socioeconomic inequality are lower. The caveat to the latter is that gender-based inequality is higher. Yet the most significant political finding is that democracy is "rarer" in Muslim societies (Fish 2011, 255-56). One conventional explanation for this finding, that is, that Muslims are more likely to fuse religion with political legitimacy, is empirically rejected. Instead, Fish argues there is no fundamental difference between autocrats in Muslim states and those in non-Muslim states. In other words, the same institutional impediments to democracy at work in other developing countries are essentially the same in Muslim-majority states.

Nonetheless, the view that Muslim populations are less likely to endorse democracy remains conventional even following the events of the so-called Arab Awakening beginning in 2011. Cynics point to the fact that the first elections following regime change in Tunisia and Egypt returned Islamists as the victors. Considering the dynamic flow of events across the region, the question of whether or not democracy will ultimately consolidate, whether or not peaceful transfers of power to rival parties will occur routinely following future elections, is not yet known. Thus, a more substantive response to the question of whether or not Muslim societies are resistant to democracy would benefit from observations of the Arab world in years following the Arab Awakening.

Yet despite the efforts of Fish and other scholars to determine whether Muslims are somehow exceptional, there is another critical question to which the answer remains empirically unanswered: are Muslims exceptionally resistant to ethnic nationalism? The questions are not mutually exclusive. As a doctrine of both popular sovereignty and territorial self-determination, nationalism is essential to nation-state legitimacy. By determining the criteria for who is or is not a citizen, whether determined by where one is born or to whom, a national identity enables the state to identify who is or is not a member of the demos. This identity is typically composed of social markers including, but not limited to, some combination of shared ancestry, religion, culture, and language. This identity defines a people, who in turn define a nation, which in turn describes the extent—and justifies the existence—of a nation-state. Nationalism is, in this sense, "the major form in which democratic consciousness expresses itself in the modern world" (O'Leary 1998, 79; Nodia 1994). Hence, if a state supports a shared culture that links the identity of otherwise diverse citizens, then "many of the problems that will normally appear in the effort to democratize a multinational community are simply not on the agenda" (Stepan 1998, 223). This is equally applicable to individual freedoms as "it is necessary to solve the national question before liberal rule can be possible" (Hall 1998, 13). In the case of American exceptionalism, for example, Samuel Huntington argued that a threat to the liberal creed that defines American national identity is nothing less than a threat to American democracy (Huntington 2004).

What then should be made of a Muslim national identity? Is there anything like a Muslim demos? Is there a single nation of Islam, or are there many Muslim nations? And what of the Arabs? Is (or rather was) pan-Arab nationalism a doctrine shaped around religion, ethnic chauvinism, neither, or both? And what is the impact of the Arab Spring vis-à-vis nationalism: will it ultimately lead Arab states toward a stronger shared identity but separate states, or separate identities and separate nations? These questions draw attention to a split within the discussion(s) of Muslim exceptionalism: on one side are proponents of positions that mark Arabs as a special subset of Muslims; on the other side are observers who ascribe and/or describe characteristics that apply to any society that surrenders, as it were, to Islam.

In the case of the Arab world, the Muslim exceptionalism thesis does have some leverage due to the remarkable sociolinguistic phenomenon of diglossia, a situation whereby languages function in separate registers, one Low and one High, one illiterate and one literate, one ingrained and one acquired, one vulgar and one official. Arabic diglossia means that the Low vernaculars of Arab peoples are marginalized in the public sphere in favor of High Arabic, that is, Modern Standard Arabic, which is the sole official language of the state and its institutions, most critically those of public education. This yields a population in a state but not of a state, a population part of an amorphous "Arab world" but not part of an ethnolinguistic nation as is typical outside of the Arab world. I call this phenomenon Arab dinationalism (see Chapter 4).

Outside of the Arab world, however, the perception that Muslims are exceptionally resistant to ethnic nationalism is challenged directly. To this end, six cases of Muslim-minority separatists—the Kurds of Iraq, the Uyghurs of China, the Sindhis of Pakistan, the Kashmiris of India, the Acehnese of Indonesia, and the Moros of the Philippines—were selected as examples of Muslim societies challenging the sovereignty of their state, though the nature of the challenge was not predetermined. To collect primary research material, the leadership of separatist parties and organizations were contacted personally and interviewed specifically regarding the raison d'être of their operation, whether Islamist, nationalist, both or neither. In regard to the relative strength of an ethnonational identity, a specific litmus test was the separatists' view of education language policy (see Chapter 3). This all required a great deal of fieldwork. Interviews with expatriate (or exiled) leaders required travel to Britain (Kurds), Sweden (Acehnese), and Germany (Uyghurs). However, most interviews were conducted on site, on the ground, in the separatist region itself. Hence, fieldwork was conducted in Iraq (Kurdistan), Pakistan (Sindh), India (Kashmir), Indonesia (Aceh), and the Philippines (Mindanao).

After examining and comparing all six cases, the findings of this book are as follows:
It is clear that non-Arab Muslims are in no way resistant to ethnic nationalism.
There is, however, variation among the cases. Some Muslim separatists are exclusively secular and ethnonational: this is so in Xinjiang, Sindh, and Iraqi Kurdistan. Others are influenced by Islamist politics: this is so in Kashmir, Aceh, and Mindanao.
The variation between the cases is explained by the critical role of language in a population, specifically whether or not a community is linked by a language that is (1) a mother tongue, (2) written and read, (3) printed and sold. In other words, the variation is explained by the presence or absence of what is called a print culture.
A strong print culture correlates with a strong ethnonational identity, and a strong ethnonational identity correlates with a conspicuous absence of Islamism.
Therefore, this book argues there is an inverse relationship between Islamism and secular ethnolinguistic nationalism.
Pointedly, this suggests that Islamism is not a stronger ideological force in cases of Muslim-minority communal conflict, but rather it functions as a kind of opportunistic infection: it infects citizens who are not immunized by an ethnonational bond. Thus, the prescribed "treatment" for Islamism may, in effect, be nationalism.

Definitions

A discussion engaging such broad concepts as the nation or Islam immediately risks descending into "terminological chaos" (Connor 1994, chap. 4). Such terms are typically applied, along with any number of social and political characteristics, as classifiers of the noun identity, a word that is itself beset by "definitional anarchy" (Abdelal et al. 2006). This is explicable. Identity is at once ubiquitous and ephemeral, a quantity that defies operationalization. Like water—whether in the form of ice, cloud, steam, rain, etc.—an identity has different properties depending on its environment. Hence, in this book, an identity is treated as the medium through which different political markers move, whether of religion, class, gender, culture, and so on. Put another way, an identity is like a person (or people) that has no attributes, that is, a construct that cannot exist other than in the mind. A person without any characteristics is not a person. It is the attributes, traits, and markers that are more important than the medium, and especially so in regard to political mobilization.

In specific regard to the marker nation, it is important to note that this project does not engage the extensive debate among primordialists, constructivists, instrumentalists and others over national pedigrees. Defining the term nation is a thankless job since almost any characterization is likely to offend or dissatisfy. It may be considered a contemporary manifestation of select "premodern ethnic cores" (Smith 1991, 41; 1986), a daily plebiscite of like-minded liberal citizens (Renan [1882] 1996), a woefully misguided application of Kant's metaphysics (Kedourie [1960] 1993), an incurable pathology of modernity (Nairn 1981, 359), or an "imagined political community" (Anderson 1983, 6). Defining Muslim initially appears somewhat simpler: a Muslim is an adherent of Islam. Muslim society, on the other hand, is something else entirely. Consider the analogue of "Christian society." Do Russia and the United States share the same society if they are both populated by Christians? If there is such a thing as Muslim society, is it the same thing as a Muslim civilization? Despite the mix of meanings, a working definition of Muslim society is also needed to engage the idea of Muslim exceptionalism.

Ethnicity, Nation, Nationalism

The debate over what is or is not a nation has continued for centuries because some definitions are exclusive; that is, my nation is organic, ancient, and authentic while your nation is synthetic, invented, and fake. If a nation is called bogus, then its erstwhile nationals have no claim to national self-determination and so cannot claim sovereignty over a nation-state. Hence, the definition of a nation is inextricably linked to the idea of nationalism. In the words of the most influential general theorist of nationalism, Ernest Gellner, "nations can be defined only in the age of nationalism" (Gellner 1983, 55). Because of this paradox, John Hall allows, "Gellner is quite right to insist that nation is far harder to define than is nationalism" (Hall 1993, 6).

Regardless of whether a nation is believed to be bona fide or counterfeit, a baseline definition links a self-defined people to a specific piece of real estate, that is, "Nationalism is primarily a political principle, which holds that the political and the national unit should be congruent" (Gellner 1983, 1). More often than not, a national unit defines itself along ethnic lines, for example, Japanese, Kazakh, German, and so forth, with exceptions made for settler states as so-called civic nations that do not explicitly restrict citizenship along bloodlines, for example, the United States or Australia. However, this begs the question of defining ethnicity, a term that is as contentious as nation for the same reasons of challenging authenticity.

"Ethnic politics," however, may address group identities based on a mix of shared family, region, religion, tradition, tongue, and so on (Posner 2005). This is an important point since ethnopolitical problems are a typical condition of deeply divided societies. This is not to say that ethnic diversity is an invitation to chaos—far from it—but when exclusive identities are expressed publicly and politically, the contract between state and society is called into question. In this regard, a "divided society" is one that is "both ethnically diverse and where ethnicity is a politically salient cleavage around which interests are organized for political purposes" (Reilly 2001, 4). Note, however, that a divided society is not necessarily segregated territorially. A divided society may be demographically integrated, as in parts of South Africa, so the political goals of ethnic politics need not include exclusive territorial sovereignty. However, when ethnopolitical demands for self-determination include a unique claim to a specific piece of territory, then the criterion for nationalism, that is, that the national unit and the political unit are congruent, is effectively met. In this case, because membership in the national unit is defined by ethnicity, the contest is effectively an ethnonational struggle. At the state level of party politics, identifying which elements of a multidimensional ethnic identity are politically salient is an "immense challenge," especially for quantitative methodologies that require "the collection of data that validly represents the multiple dimensions of ethnic diversity found in each country, and does so over time" (Laitin and Posner 2001, 17). Yet at the level of ethnonational politics, it would seem to matter less how the group defines its ethnicity—whether by blood, book, or belief—than that its members believe they share the same ethnicity, however perceived. Thus, ethnicity is far harder to define than ethnonational.

Muslims and Muslim Society

What is Islam and who is a Muslim? A concise answer is this: the word Islam is derived from the Arabic root for "surrender," so a Muslim is one who surrenders to God. What is a Muslim society? Theologically, there is an equally concise answer to this question: a Muslim society is an ummah, that is, a universal community of believers who declare there is no God but Allah and that Muhammad is the messenger of God. The ummah represents "the essential unity and theoretical equality of Muslims from diverse cultural and geographical settings" (Esposito 2003, 327). This de jure ideal, however, clashes immediately with any sociology that incorporates a specific culture or ethnicity into a description of a specific society. Hence, the question of whether there is a de facto global Muslim society is clearly a contentious one. This is related but distinct from the question of whether there is a Muslim civilization.

At this point, we must recall Samuel Huntington, who defined a civilization as the "highest cultural grouping of people and the broadest level of cultural identity people have short of that which distinguishes humans from other species" (Huntington and Ajami 1993, 24). In this view, if Islam were a cultural identity, then the ummah would qualify as a civilization. A cultural identity is "defined both by common objective elements, such as language, history, religion, customs, institutions, and by the subjective self-identification of people." So what happens if all the objective elements except religion are not commonly shared? Does Islam trump all? For Huntington, it did (and for his latter-day Orientalist adherents, does) because Islam is exceptional. To explain why "Islam has bloody borders," Huntington (1993, 35) cites vocal historian Bernard Lewis. It was Lewis who actually inspired the title of Huntington's famous article when he wrote—three years earlier in the Atlantic Monthly—that a clash specifically between Islam and the rest of the world is "no less than a clash of civilizations." The violence is explicable, according to Lewis, because "there is something in the religious culture of Islam which . . . in moments of upheaval and disruption, when the deeper passions are stirred . . . can give way to an explosive mixture of rage and hatred" (Lewis 1990, 60).

Note, however, that the views of Lewis and Huntington on Muslim civilization are coherent if and only if we accept the totality of a Muslim civilization defined as a "cultural entity" that is essentially a "religious culture." This exceptional religious culture is so overwhelmed by faith that other potential elements of ethnic identification, including the aforementioned "language, history, religion, customs, institutions, and by the subjective self-identification of people," are all but irrelevant. The power of this particular faith to smother all other attachments is often attributed to Islamic ideas about "the relations between government, religion, and society" (Lewis 2003, 5-6) that essentially fuse the private, the public, and the political into a single, sacred sphere.

As in the case of the imagined ummah, however, this is an ideal observed more in the breach than in practice. States with Muslim-majority populations demonstrate distinct histories, institutions, and social practices that divide the private, the public, and the political in countless ways. Accounting for the separate trajectories of Egypt, Turkey, Iran, Pakistan, and Indonesia, for example, is not possible without considering other politically salient elements—including colonial legacies, geography, and ethnonational identification—that have dramatically affected state-society relations across one-third the circumference of the Earth. In a direct response to Gellner's monolithic vision of Muslim society, Sami Zubaida retorts, "the rivals of the nation-state for solidarities and sentiments are not the universalist entities of Arabism and Islam, but more likely particularistic and factional solidarities of community and region" (Zubaida 2004, 419; 1995; 1998).

Numbering more than 1.5 billion people—or 22 percent of the planet's current population of 7 billion—the collective Muslim world "is at once very large and very diverse" and clearly bears many "marks of ethnic, linguistic and cultural diversity" (Lawrence 1998, 33). To accommodate at least some of this diversity at the substate level, it is helpful to consider a range of Muslim populations across multiple states and regions that are collectively grouped into the general classification Muslim society (which is also an effective proxy for the more general term the Muslim world). In other words, rather than using the term to describe an idealized ummah, "Muslim society" here means a great collective of diverse Muslim peoples. The multiplicity of Muslims is staggering, though it is fair to note that one generalization is indisputable, even as it is routinely ignored: most Muslims are Asian.

While half of all Muslim-majority states are in the Middle East, most Muslims are not Middle Eastern. Fewer still are Arabs: the two largest Muslim states in the region are non-Arab Turkey and Iran. The states with the four largest Muslim populations are in South and Southeast Asia: Indonesia alone has more Muslims than Morocco, Algeria, Egypt, Syria, Iraq, and Saudi Arabia combined. China has twice as many Muslims as Tunisia. Hence, any discussion of the Muslim world that fails to account for this obvious demographic dispersion is exceptionally flawed.

Muslim-Minority Separatist Conflicts

A number of large-N statistical studies have examined general patterns across states with Muslim majorities. Steven Fish, for example, shows that Muslim states are "democratic underachievers" and attributes this in part to the sorry status of women and girls in these countries (Fish 2002, 4; Pryor 2007). Jonathan Fox questions the reputation of Islam as an exceptionally contentious faith by comparing 105 ethnoreligious populations selected from the Minorities at Risk Project and finds "little evidence here to support the argument that Islam, or any religion for that matter, makes ethno-religious minorities more conflict prone" (Fox 2000, 15). In a later study using the State Failure data set, Fox shows that when Muslims battle governments, "there is some evidence that Muslim groups are more violent" than Christian, Buddhist, or other groups, but cautions that "it is not conclusive and is certainly not enough to support the stereotype of the Islamic militant" (Fox 2003, 27). Another study, by Susanna Pearce (2005), of 278 territorial conflicts in the Armed Conflict Data set of the International Peace Research Institute in Oslo (PRIO) argues unequivocally that the intensity of conflict is not correlated with any particular religion, including Islam. This finding correlates with later work at PRIO itself. Employing the 5-point Political Terror Scale, which is based on country reports from Amnesty International and the U.S. Department of State, Indra de Soysa and Ragnhild Nordås show that among 141 cases "countries with higher shares of Muslims suffer much lower levels of political repression," at least in contrast to states with significant numbers of Catholics (2007).

This research effectively challenges popular and professional notions that Islam (as a faith) and Muslims (as a people) are essentially undemocratic, contentious, militant, and bloodthirsty. But this research does not help answer the separate question of whether Muslims are exceptionally resistant to nationalism.

As for studies that examine general patterns across Muslim societies, there is much less on offer. Arguably, this is not for lack of interest, but for lack of data or (more problematically) for lack of reliable data. While this issue is in no way whatsoever limited to the general topic at hand, there are two specific problems here (Herrera, Kapur, and Tarontsi 2007). The first is "much of the research based on cross-national surveys aggregates micro-data into countrywide means and percentages," which effectively buries "other important aggregates such as cultural or ethnic groups." Even when disaggregated, there remains the second and specific problem of how "sub-national pluralism" is gauged, interpreted, and processed. In addition to the data sets mentioned above, the other major cross-national surveys include the World Values Survey, the International Social Survey Project, and the Comparative Survey of Electoral Systems project. These projects are eminently useful to some kinds of inquiry, but information about ethnicity is collected and coded "in an inconsistent way, and they have not given sufficient attention to ethnicity in the sample designs" (Dowley and Silver 2005, 226).

Case studies of specific states are better equipped to drill down into the socioeconomic, ethnolinguistic, and geopolitical strata underlying the structure of a select country or population, but the findings of even the most robust case study cannot effectively respond to the claim that the evidence is exceptional, and that Muslim national exceptionalism is the rule. Thus, the small-N comparative study suggests itself as the most promising method of exploration, though this raises the question, "a small number of what?" To show whether Muslims, when politically mobilized, are more likely to rally around their faith or their flag, what is the appropriate unit of analysis, the appropriate slice of time, and the appropriate scale: macro, meso, or micro? In other words, would it be better to compare the relevant words and deeds of specific leaders over time, such as Indonesia's Sukarno, Egypt's Gamal Abdel Nasser, Pakistan's Muhammad Ali Jinnah, and Turkey's Kemal Atatürk? Or would it be better to compare the voting records of different legislatures and the electoral outcomes of, say, nationalist versus Islamist platforms? Or perhaps conduct a content analysis of print and broadcast media including samples from Albania, Algeria, Kazakhstan, and Bangladesh? In fact, all of these projects could be illuminating and of potentially great merit. Nonetheless, it is reasonable to suggest that a project addressing the relevant role of nationalism versus Islam in the cohesion and mobilization of Muslim populations may also have some utility outside the academic world. To see why, it is useful to consider why falsifying this theory in the first place is wise. Could it, for example, help to solve any problems?

The short answer is yes. In general, answering this question will help to dispel some of the perennial conjecture about the path dependency of Islam in politics, such as whether Muslim societies, because of their faith and religious heritage, are ultimately incompatible with modernity, democracy, or liberty. Specifically, and even more important, answering this question is of significant strategic concern. If a number of strategically important Muslim states and societies are considered unstable, and a number of Muslim regions suffer from internecine conflict, appropriate responses—in terms of either domestic state policy or international foreign policy—would benefit tremendously from understanding the nature(s) of the problem(s). This is particularly the case when working out the dynamics of Washington and Baghdad, Yerevan and Baku, Moscow and Grozny, Beijing and Urumqi, New Delhi and Srinagar, Jakarta and Banda Aceh, Manila and Cotabato, Islamabad and Quetta, or Belgrade and Pristina. These, I believe, are problems to be solved (Schram and Caterino 2006; Monroe 2005; Shapiro 2005).

Hence, in the course of answering the question "are Muslims exceptionally resistant to nationalism?" this project seeks not only to falsify a claim, but also to assist those actors, agencies, and institutions trying to understand and engage cases of conflict in and around the Muslim world. The point of this project is not to prove what "caused" the conflicts in the first place. Many of these cases are burdened with extensive histories, stretching back at least decades, and in some cases centuries: the number of temporal variables alone is incalculable. For example, what is the most important factor helping to explain the rise of Abu Sayyaf in the southern Philippines: the Spanish conquest of Mindanao in the sixteenth century, the migration of Catholic settlers to the south in the twentieth, or the petrodollars of well-funded Salafis who encourage violent Muslim mobilizations in twenty-first-century conflict zones worldwide? In the case of Mindanao, it is helpful to understand that the conflict is very old but also very brittle, with fractures along cultural, religious, economic, and geographic lines.

Evidence and Cases

Because claims about the "Muslim world" apply theoretically to all Muslim societies, the cases were selected from multiple regions. In some cases, a separatist population is a minority in more than one state, for example, Kurds in Turkey and Iraq, in which case only one group was selected to maximize diversity in the set. In addition, the majority population may or may not be Muslim-majority. There are two possible scenarios:
An ethnic minority population of Muslims in a state that is home to an ethnic majority who are also Muslim
An ethnic minority population of Muslims in a state that is home to an ethnic majority who are not Muslim
Cases representing both demographic scenarios are necessarily included in this study. Muslim minorities in non-Muslim-majority states include the Uyghurs of China, the Kashmiris of India, and the Moros of the Philippines. Muslim minorities in Muslim-majority states include the Kurds of Iraq, the Sindhis of Pakistan, and the Acehnese of Indonesia. In both scenarios, the minorities speak a different mother tongue than the majority. In either, according to proponents of ethnonationalism, the minority should mobilize as a distinct people defending and promoting a distinct ethnolinguistic culture. In contrast, and also applicable in either case, proponents of Muslim national exceptionalism expect Muslim minorities to advance an Islamist political culture that eschews any ethnic identity.

Structured Interviews

Classifying the platforms of nationalist organizations is unavoidably a political act. According to the People's Republic of China, for example, the East Turkistan Information Center is a terrorist organization with purported links to al-Qaeda. Yet according to the Munich-based East Turkistan Information Center it is a secular pressure group advocating an end to the repression of Uyghurs and other Muslims in Xinjiang. Note, however, that the latter piece of information was obtained not from the Terrorism Knowledge Base funded by the Department of Homeland Security or from news reports, since the ETIC itself has not been probed directly as to the nature of its platform. Rather, I went to Germany, where exiled ethnic Uyghurs operate the organization, met the leaders personally, and asked him a number of questions. This process—and the same set of questions—was repeated in each case.

The interviews were structured but not scripted, as some groups were more sensitive than others as to the nature of my research and their willingness to discuss openly questions about Islam, ethnicity, language, and separatism. This was followed by a number of questions related to the conflict, starting with "why are you fighting?" It is important to note here that the veracity and sincerity of statements made to an American researcher are open to question. It is safe to assume that in some cases an official from a separatist Muslim-minority party or organization may self-censor (or obscure by omission) positions likely to draw fire from the international community, including governments, intergovernmental organizations, nongovernmental organizations, and the media. Nonetheless, the point here is to present on-the-record answers to precise questions that are not otherwise available to researchers of comparative politics or international relations.

As discussed in Chapter 3, language laws and language policies, including the message and medium of state education, are a critical litmus test for ethnolinguistic mobilization. Just as separatism is necessarily territorial and therefore necessarily in conflict with one or more states, ethnolinguistic nationalism is necessarily institutional. In this book, the process of language planning—including corpus and status planning—is generally described as language rationalization. This term is adopted from David Laitin's exposition of Max Weber's bureaucratic rationalization, which interprets the evolution, and inevitable expansion of state language policy toward homogenization. Although "Weber did not systematically explore language rationalization," Laitin posits "the use of state power, through administrative regulation and public education, to standardize language within the boundaries of the state is precisely what he had in mind with his concept of rationalization" (2001, 87).

Thus, the crux of each interview addressed specific positions on languages. For example, in the case of an independent Kurdistan, what would be the official language or languages of the state? Of the schools? What would be the place of Arabic, either as an important regional language or as the sacred language of an overwhelmingly Muslim population? Over the many decades of conflict with Baghdad, did separatist leaders invoke the issue of protecting or promoting the Kurdish language? If so, why? It is important to recall here that even if some of these questions appear facile, the answers are not documented, and cannot be cited as proof positive of either indifference—as expected in the case of Muslim national exceptionalism—or essentialism, that is, ethnolinguistic identification as a zero-sum condition for the existence of the group itself.

The Plan of This Book

This book offers an original response to the broader academic literature—and persistent public perception—that there is something particular and problematic about the politics of Muslim peoples. Again, this idea takes a number of shapes, but collectively the concept is called Muslim exceptionalism. The two most common interpretations of the idea argue that (1) Muslims are exceptionally resistant to democracy or that (2) Muslims are exceptionally resistant to nationalism, that is, the political allegiance of Muslims is first and foremost to Islam, rather than identifying with a unique ethnicity (nationalism), and consequently to a particular nation-state (patriotism). These positions often dovetail, since a people prone to illiberal religious fundamentalism are naturally opposed to liberal democratic politics.

It is argued here that these exceptional phenomena are evident in Arab regimes that are hamstrung by the doctrine of pan-Arab nationalism that divides populations from their patria, and erstwhile nations from states of their own. Yet this condition applies to fewer than one in five Muslims. Aside from those in Arab states, I argue Muslim societies are subject to the same "rules" of nationalism that apply to non-Muslim nation-states. However, when Muslim-minority populations mobilize as separatists, there are two paths that may be taken: one nationalist and one Islamist. The nationalists maintain a strong ethnolinguistic identity based on a strong vernacular print culture. The Islamists, on the other hand, typically lack a strong vernacular print culture and therefore sustain only a weak ethnolinguistic identity, in which case the appeal of Islamism is stronger. Therefore, there is an inverse relationship between secular ethnolinguistic nationalism and the religious identification of Islamism.

The following chapter, "Muslim Nations," explores the parallel theories of ethnonationalism and Muslim exceptionalism in detail. Chapter 3, "National Tongues," outlines and evaluates ideas about the relationship of language, ethnicity, and national identity. Chapter 4, "Modern Standard Arabs," is an extended discussion of Arabs, Arabic, and Arab states and Arab national identity. Chapters 5 through 10 consider the six case studies under examination. This is followed by a conclusion that distills the argument and, at the same time, considers how it could inform public policy in Afghanistan.