1As indicated by the titles of several recent books, it has become commonplace to argue that Pakistan is ‘in search of identity’ (Ali 2009, Jaffrelot 2002, Shaikh 2009). From the secession of East Pakistan to form Bangladesh in 1971 to the on-going conflict in Balochistan, the contestation that the country has faced since its independence in 1947 indeed suggests that Pakistan has not reached a consensual conception of the nation. The two-nation theory, which postulates that South-Asian Hindus and Muslims cannot live harmoniously in the same state, has provided the main ideological foundation for Pakistani state nationalism, yet it has failed to act as a unifying concept the way it did in the years leading up to Partition.
2The books under review here by Alyssa Ayres and Farhan Hanif Siddiqui are contributions to existing scholarship on ethnic identities in Pakistan (see for instance Adeney 2007, Ahmed 1998, Amin 1993, Baloch 1987, Harrison 1981, Khan 2005, Malik 1997). These two works analyse movements that do not identify with Pakistani state nationalism but put forward localized ethnic identities, arguing for cultural recognition in a more inclusive national narrative or struggling for political autonomy. Although both books attempt to nourish the debate on pressing policy issues in Pakistan, they adopt contrasting approaches. Alyssa Ayres draws on historical analysis and visual anthropology to investigate the counter-intuitive case of Punjab from the point of view of language policy; although Punjabis constitute the ‘putative ethnic hegemon of the country’ (SLS: 5), dominating numerically the bureaucracy, the military, and the industry, in the 1980s a language movement had ‘begun to argue for official recognition of the Punjabi language and its cultural and literary history’ (SLS: 5). The first three chapters of the book examine, respectively, the construction of Urdu as a national symbol by the proponents of a separate state for Muslims, the place of Urdu in independent Pakistan, and the reactions to the state’s endorsement of Urdu as national language. The book then turns to the case of the Punjabiyat movement in two chapters that focus on ‘elite efforts’ and ‘popular culture’. In chapters 6 and 7, Alyssa Ayres retraces the construction of a unitary national narrative for Pakistan and the subsequent questioning of this narrative, which ‘open[ed] space for the idea of a nation as a composite project’ (SLS: 148). The final two chapters place the example of Pakistan in perspective with two comparative case studies, India and Indonesia. The book is adapted from Ayres’ PhD dissertation in cultural history written at the University of Chicago. Yet the author’s career extends outside academia to non-profit organizations such as Asia Society and policy circles, as she has worked for the U.S. Department of State. Consequently, Ayres’ work, as is made clear in its introduction, ‘marks an effort to engage in historiographical scholarship that can have relevance to public policy decisions’ (SLS: 4).
3Also based on a PhD dissertation, Farhan Siddiqui’s book follows a political science framework—the author is now assistant professor at the department of international relations of the University of Karachi—that relies mostly on secondary sources, several interviews and press clippings. After an introduction which briefly defines the major concepts in use throughout the work—the state, nationalism, and the politics of ethnicity—Farhan Siddiqui engages in a review of the existing literature and presents the three hypotheses that guide his study of ethnic conflict: the distinction between state and government; the disconnection of nationalism with modernity; and the need to view ethnic groups as political actors with their own internal diversity. Chapter 3 draws an historical account of ethnic conflict in Pakistan from 1971 to 1999 in relation to the state and the government and discusses the literature on the ‘identity of Pakistan’ and ethnicity. The three following chapters focus on three case studies of the ‘politics of ethnicity’ which, according to Siddiqui, have to be distinguished according to their environment: Sindhi nationalism evolving in a rural setting, Baloch nationalism in a tribal milieu, and the Mohajir movement in an urban context. In each case, the author provides an historical overview of the movement, an evaluation of the role of the state versus that of the government in stirring or appeasing conflict, and an analysis of intra-ethnic differences.
4The main issue addressed by Ayres and Siddiqui is the relation of the state of Pakistan to the cultural diversity of its population. Both authors acknowledge the state as the agent principally responsible for managing ethnic plurality in a way that maintains peace. According to Siddiqui, cultural difference does not become a cause for conflict until the state neglects or exploits it: ‘Ethnic conflict and violence (...) are dependent on the political system which serves to either attenuate or intensify feelings of ethnicity’ (PEP: 4). Siddiqui thus excludes from his study the cultural content of ethnic mobilisation and focuses rather on the role of institutions in inflaming or alleviating ethnic resentment:
It does not matter how many ethnic groups inhabit a single society. All that matters is how ruling elites co-opt different ethnic groups into the political structure of the state by empowering them with decision-making in, for example, a consociational and/or federal political system. One may conclude then by estimating that ethnic conflicts are not generated automatically nor are they necessarily related to degrees of ethnic heterogeneity which prevail within a society. Ethnic conflict is a function of political factors and it assumes importance as a response to the state and its policies (PEP: 4).
- 1 On Pakistani nationalist discourses and Islam, see Talbot (2003).
5Cultural difference becomes the basis for political mobilisation when ethnic resentment arises in response to the hegemonic project of state nationalism. Ayres convincingly explains how Pakistani state nationalism came to be defined in terms of Urdu and Islam, tracing the roots of Urdu’s association with being Muslim in South Asia to the "pre-history" of the Hindi-Urdu controversy—the foundation of Fort William College in 1800 and the 1837 colonial decision to replace Persian by vernaculars as administrative languages. The scholarship of Fort William College laid the foundation for Hindi as the language of the Hindus, to which Muslims responded with calls to ‘protect’ Urdu (SLS: 17-23). Later, as anti-colonial nationalism developed, Urdu was progressively projected as the language of Muslims and became ‘a bearer of religion’ (SLS: 20). Among all the languages spoken by South Asian Muslims, it was the only one whose defence was the object of a Muslim League resolution. ‘The year of independence, 1947, thus marked the beginning not only of a new political formation—a homeland founded on the basis of religion—but also of a new belief about the linguistic medium of a unitary culture in a large bounded territorial homeland’ (SLS: 16). Thus, as Pakistan was born, the state project seeking to ‘produce the people as Pakistani’ (SLS: 5) was in contradiction with the fact that the Pakistani nation had ‘been assumed to exist already’ (SLS: 6)—the common identity markers defining that nation being Islam and Urdu, a state nationalism later rephrased as the ‘ideology of Pakistan’. Yet, as Ayres further writes, ‘Pakistan was created, even naturalized, as the expression of a nation, but that very nation self-consciously lacked a ‘national’ culture well after its founding’ (SLS: 105). Just as Pakistan’s ‘language ideology’ (SLS: 17) ignored the ‘shallowness of Urdu’s roots’ (SLS: 31-2) in the provinces that constituted Pakistan in 1947, the stress on Islam led to a state endorsement of a Sunni, literate and reformist Islam disconnected from the religious practices of the majority of the population.1 According to Ayres, ‘Islamization of national history was an explicit focus of state planning from the very beginning’ (SLS: 14). The association of Islam and Urdu thus produced an ‘ethnicized Islam’ (Verkaaik 2004: 20-2), in opposition to which ‘regional groups resisting the authoritarian regime in Islamabad legitimized the regional basis of their protest by articulating an ethnicized Islam of their own’ (Verkaaik 2004: 21). Seeking inspiration in ‘the popular’ (Chatterjee 2006: 73), ethno-nationalist movements found in Sufism a vast reservoir for a refashioned cultural and religious identity, the clearest expression of which is perhaps found in the writings of the Sindhi leader G. M. Sayed.
- 2 Apart from having written the book to which Siddiqui refers, Aitzaz Ahsan is a renowned lawyer who (...)
6Both Ayres and Siddiqui highlight that Pakistan’s ruling elites and proponents of a unitary conception of Pakistan as one nation sought to naturalise their political claims in history and geography, as a solution to the lack of a ‘national culture’. Ayres shows that ‘much of the narrative groundwork for this national past had already been laid by the various proponents of the Pakistan Movement’ (SLS: 106). She analyses the maps drawn by Choudhary Rahmat Ali—the Cambridge student who coined the name Pakistan in 1935—that represent an imagined Pakistan in ‘geological times’. Constructing its narrative in opposition to India, Pakistani historiography insisted on the separateness of the Indus valley from the Gangetic plains, relying on a distinction between Hind and Sindh found in the writings of Persian travellers. This distinction constitutes the basis of Aitzaz Ahsan’s notion of an ‘Indus Identity’, on which Siddiqui notes that it tends to gloss over the diversity of the region to celebrate Pakistani identity: ‘by absorbing all such folk heroes into an overarching Indus (Pakistani) identity, [Aitzaz Ahsan] at once refutes their ‘local’ identification and indigenous existence’ (PEP: 30).2
7Hence, the belief that Pakistan’s future and development had to rest on a unified cultural identity led to the creation of a national narrative that stressed the unity of the (West) Pakistani region over the centuries. As Ayres rightly points out, it was the assumption of the nation as necessarily unitary that conflicted with the concrete existence of a diverse population: ‘the case of Pakistan readily reveals the disjuncture between the nation-form and its demands for a unified cultural past and present, against the lived reality of a people yet ‘unproduced’ through the nation’ (SLS: 5). Ayres’ work raises the question of the necessity for the country’s founders to think of Pakistan as one unified, centralized, nation, with one religion, one people, one language. Why did those who imagined Pakistan before 1947 and built its institutions after independence fall for the temptation of social engineering instead of conceiving the new state in a plural form? The author does not indicate whether an answer should be sought in the notion of a modular pattern of nationalism (Anderson 1983), in the worldwide spread of a shared understanding of the concept of culture (Sartori 2008), in the hegemonic nature of the state (Chatterjee 2006), or in the roots of the Pakistan movement (Jalal 1995). Nonetheless she provides two interesting examples, India and Indonesia, and compares them with the case of Pakistan. In spite of its cultural, linguistic, and religious complexity, Indonesia succeeded in making Bahasa Indonesia, a language spoken by less than 5% of its population at the time of independence, play ‘a unifying role in (…) in knitting together the bewildering cultural diversity of this archipelago’ (SLS: 171). India, conversely, managed to put an end to widespread language riots by redrawing its administrative map along linguistic lines while maintaining Hindi and English as national languages. According to the author, these comparisons ‘highlight how language ideology (…) plays a crucial role in determining whether a national language will spread easily, without protest, and how strongly it helps forge the national bond it is intended to create.’ (SLS: 172) In other words, ‘we can see how the national language emerged only as the product of official programs crafted to make national citizens learn to speak—like a state —and achieved national acceptance only when not in conflict with the local past’ (SLS: 187).
8In spite of its unitary nature, Pakistani state nationalism has nevertheless fluctuated over time, in particular during the 1970s, when Zulfiqar Ali Bhutto’s government allowed for a more inclusive narrative to be construed. ‘Bhutto allowed greater cultural expression to the regions, and it was also during his administration that cultural heritage institutions like the Lok Virsa and the Panjabi Adabi Board were founded’ (SLS: 34). This opening incorporated local histories, which were rooted in claims to an ‘ethnicized Islam’, into a Pakistani narrative. Pakistani official nationalism, far from being constant and uniform, changed over time and is perhaps better understood as a result of competing discourses in between and within institutions. Here, the distinction made by Siddiqui between state (military, bureaucracy) and government (elected officials) appears useful. According to Siddiqui:
‘ethnic identity is prone to intensify in political contexts where state rules, governs and makes policies (such as military dictatorships). Ethnic amelioration is possible if a government exists and functions on principles of consociationalism and power-sharing. (…) This, however, does not preclude the possibility of the government acting as authoritatively as military dictatorships in denying power and privilege to non-dominant ethnic groups’ (PEP: 27).
9Siddiqui concludes that ‘the history of the Pakistan polity exhibits an inexorable and incessant tension between the elected (government) and unelected (bureaucracy–military) institutions of power and authority’, a tension in which ‘governments have never been completely docile before the military’ (PEP: 32).
10But Siddiqui’s distinction between state and government remains problematic; it does not define what the state is exactly, what it seeks, and why it cannot deviate from its own positions and authoritarian behaviour. Moreover, it ignores the fact that the government abides by the rules of the state and thus serves the state even as it appeases ethnic conflict by moderating the state’s (postulated) centralizing tendency. If the government of Zulfiqar Ali Bhutto managed to mould Pakistani nationalism into a more inclusive narrative, the process went hand in hand with tighter repression of subversive discourses. Therefore, the crackdown on Baloch insurgents in 1973 seems best conceived as the necessary counterpart, for instance, to the creation of the Lok Virsa or the organisation of a major conference on Sindh’s history in 1975. The government of Zulfiqar Ali Bhutto shifted the limits between the acceptable and the subversive, but this strengthened the need to suppress the subversive. The distinction between state and government thus seems of limited heuristic value, since it neither serves as a basis for a generalization nor entails a proper sociology of the state, which would yield sharper conclusions. Siddiqui’s attempt to open the ‘black box’ of the state, in line with liberal and constructivist international relations scholarship, in fact appears to recreate monolithic objects in the state and the government.
- 3 It should be noted that the societies which, according to Anderson, developed a national model late (...)
11Both Ayres and Siddiqui find the dominant theory of nationalism (in particular, Gellner 1983 and Anderson 1983) at odds with the context of Pakistan. Alyssa Ayres questions Benedict Anderson’s concept of print-capitalism; to her, Punjab’s low literacy rate ‘poses clear limitations for the explanatory or catalytic value of print textual forms to engage this large population in a common sense of national belonging’ (SLS: 72). She adds: ‘that these three [India, Indonesia, Pakistan] independence movements, movements that envisioned nations that had never before existed, were able to convert the masses who did not actually read (…) suggests that the ‘meme’ of national consciousness can indeed coalesce through oral communication, public addresses, and other forms of non-print communication that can take place in multiple, even mixed, language forms’ (SLS: 186). That nationalism relies on means of propagation other than print3 —which is not a novel idea (see Chatterjee 1993: 73)—leads Ayres to turn to Punjabi cinema to show how Punjabi popular culture tried to reverse a sense of inferiority. The film Maula Jatt and its sequels portrayed a strong, brave and violent hero and did not shy away from using crude language, thus ‘reclaiming a Punjabi imaginaire in which Punjab would no longer be viewed as submissive’ (SLS: 97).
12Farhan Siddiqui’s critique of Ernest Gellner’s theory of nationalism concerns the relationship between industrial development and the emergence of nationalism. Treating Gellner’s thesis critically and applying it to post-colonial societies, one is bound to question his essentially modernist predilection which sees nationalism as emerging from the socio-economic context of industrialisation. (…) As the example of ethnonational movements in Pakistan demonstrates, feelings of nationalism can emerge in tribal and rural socio-economic contexts as well (PEP: 11).
- 4 Siddiqui states that ‘In many ways, the rise of nationalism in post-colonial states was unconnected (...)
13Yet Siddiqui’s categorisation of Sindh and Balochistan as, respectively, ‘rural’ and ‘tribal’ settings does not deconstruct the modern/traditional dichotomy but rather reinforces it. His assumption that ‘rural’ and ‘tribal’ societies were unconnected to industrialisation or modernity4 ignores the argument that the ‘underdevelopment’ of colonial societies was as much part of the modern imperialist project as the industrialization of the home country. The simple categorization of Sindh and Balochistan as rural and tribal overlooks the complex relations between rural and urban areas, as a result of which the changes in urban lifestyles significantly impact rural social structures. In Sindh, migrant labourers, students, as well as a wider range of people (men) who have had access to education, move far beyond a hypothetical ‘rural’ setting, thus connecting villages to cities. Incidentally, students and educated men with a rural background are often also highly politicised, and at the forefront of nationalist movements. Siddiqui’s attempt to refute the modern/traditional dichotomy misses the point: rather than asserting that nationalism can also emerge in a rural context, what needs to be examined are the socio-economic conditions that give rise to ethno-national movements, in particular the complex relationship of between the rural and the urban out of which a frustrated middle-class develops ethno-national demands (for analyses of socio-economic conditions of ethno-national movements in Pakistan, see for instance Alavi 1989 and, Ahmed 1984).
14Siddiqui’s approach does not focus on how nations are formed, but on how they behave as political actors: ‘In this sense, then, a study of nationalism is (...) about how nations conduct and evolve relevant ideologies, discourses and movements in order to achieve statehood or even provincial autonomy’ (PEP: 9). Siddiqui’s non-committal position leaves to nationalists themselves the debate over who really constitutes a nation or which group better fulfils the conditions to be one. But taking for granted that Sindh, Balochistan and Mohajirs are already formed nations, or ethnic groups, leads him to miss an important element: that nation-formation is a performative process. The ‘politics of ethnicity’ does not happen once ethnicities or nations are formed, but the definition of identity markers and boundaries is constitutive of their very existence (Barth 1969). Hence the salience of Ernest Gellner’s famous line: ‘It is nationalism which engenders nations, not the other way around’ (Gellner 1983: 55).
15Even if one admits that nationalism is a process ‘taken too literally to be political’ (Chatterjee 2006: 5), which starts much before political movements emerge, studying ‘how these politically non-dominant nations initiated their own political organisations, movements and ideologies against an authoritarian and authoritative state structure’ (PEP: 9) comes down to studying the constant process of national re-creation, provided one looks at the relation between the political elite and the population at large. Siddiqui’s study would thus have gained much by acknowledging that the ‘politics of ethnicity’ is the site of constant self-redefinition by ethnic groups, in which what is defined as representative of the group’s identity implies the rejection of what then becomes the ‘other’. Thus, understanding the politics of ethnicity requires looking into the claims to nationhood that ethno-national groups put forward on the basis of historical revisionism and the mobilisation of symbolism, a process certainly not pursued uniformly by all members of ethnic groups.
16Conscious of the risk of reifying ethnic groups as homogeneous givens, both Ayres and Siddiqui attempt to grasp internal differentiations. Siddiqui first distinguishes between the public at large and political organisations: ‘It is not ethnic groups that are involved in ethnic conflict; rather, it is the ethnic organization that claims to speak in the name of the nation and is involved in conflict with the state’ (PEP: 18). Therefore, it is not, as Siddiqui also contends, ‘ethnic groups [that] are internally stratified on the basis of the political choices that they make by allying themselves with the state or standing in opposition to it’ (PEP: 3), but the political organisations that seek to represent them. The demarcation made by Charles Kennedy between separatist, autonomist and ‘gradualist’ parties in Sindh followed a similar logic (Kennedy 1991: 947). But as Siddiqui notes, ethnic organisations and their leaders are subject to co-optation by the state, moulding the official standpoint of political organisations over time. A close study of the evolution of the stances of various ethnic political organisations is thus necessary to understand their relation to the state, as well as to show the conditions under which factions unite or fight one another. But acknowledging that these organisations only claim to speak in the name of the whole ethnic group does not inform us about how others, say persons not affiliated with any organisation, stand in relation to such parties or understand their own identity.
17Rather than focusing on the attitude of ethnic groups or their self-designated representatives towards the state, Alyssa Ayres relies on an elite/popular distinction to first examine ‘elite efforts’ for Punjabi assertion—writings of a small Lahore intelligentsia—before turning to popular culture. Her analysis of the work of elite ‘cultural entrepreneurs’ and of popular media—cartoons and cinema—highlights a Punjabi discourse of assertion, but doesn’t does not seek to explain the relationship between the elite conception of Punjab and what she presents as the popular representation of being Punjabi. Ayres overlooks the cultural hegemony that nationalism imposes on the population that nationalists claim to speak for—as well as the internal resistance that this hegemony may encounter from those it excludes. For instance, one may wonder how Siraiki speakers residing in Punjab feel regarding the Punjabiyat movement or what the reactions of people from various strata of Punjabi society may be. Ayres provides a general framework to describe how nationalism creates oppositions according to a modular pattern. ‘These divisive, ‘dichotomizing oppositions,’ in the words of Susan Gal, can recreate themselves recursively, in the manner of fractals—which replicate their forms on ‘ever-smaller social units’ (SLS: 28). She argues that the initial distinction between Hindi and Urdu, superimposed on Hindu/Muslim identities, replicated itself onto smaller entities: Bengal, Sindh, Punjab, or the Seraiki region. ‘Replicated dichotomisations’ (SLS: 29, 46, 48) thus determine ‘smaller’ nationalisms: in a type of mise en abyme, Sindhi, Baluch, Pashto or Punjabi movements must define themselves against Pakistani state nationalism, which is itself posited against Indian nationalism.
- 5 There is a vast literature on the subject, from Mancur Olson's seminal The Logic of Collective Acti (...)
18These two books provide for a stimulating discussion on ethno-nationalism and state policy towards diversity and are therefore welcome additions to the literature on ethno-nationalism in Pakistan. Siddiqui’s work offers a coherent synthesis on three of Pakistan’s major ethnic movements and their relationship with the state. Alyssa Ayres brings to the fore a little known movement and has the merit of studying new material. Yet at a time when a small fraction of the numerous new publications on Pakistan rely on empirical research, one may regret that neither Ayres nor Siddiqui uses observation to substantiate her/his reflections (although it should be noted here that Alyssa Ayres’ initial fieldwork project was held up for security reasons). Siddiqui’s work would have gained from drawing on the sociology of mobilisations and social movements.5 This would have allowed him to offer greater insights on the internal dynamics of ethno-national movements and on the lived reality of those who partake in them. It would also have allowed him to pay more attention to socio-economic conditions or to the cultural content of ethno-nationalist discourses, factors that the focus on the state and the government leads him to ignore. Ayres’ study could perhaps have gone beyond the elite/popular dichotomy by examining in their social context the strategic decisions made by cultural entrepreneurs, filmmakers or cartoonists in picturing Punjab. But in spite of my criticisms, these works provide useful perspectives on an issue that has been the source of violent conflict in Pakistan and on which Siddiqui concludes, with a rather pessimistic note, that ‘although ethnonationalist movements will not cause the break-up of Pakistan in the future, Pakistan will continue to be plagued by such movements’ (PEP: 113).