Skip to navigation – Site map

HomeThematic Issues10A Strange Love of the Land: Ident...

A Strange Love of the Land: Identity, Poetry and Politics in the (Un)Making of South Asia

Sudipta Kaviraj


In this paper I shall try to explore how the idea of space in what we call ‘South Asia’ today got reconfigured by modernity. I shall explore three spatial conceptions of South Asia—strategic, geographic and cultural; and I shall try to explain the patterns of space-thinking underlying each one of them, show their connection with modernity, and observe why and how South Asia is still not a space that can be conceived in nationalistic or state terms as a space of belonging. Two political processes of modernity introduce a new kind of space-making—nationalism and state-formation. In the paper, I shall follow some significant points of this dual process—of the emergence of new state structures—of British colonial India—and its animation of the rise of variant forms of nationalism—in 19th century Bengal, in the Islamic imagination of the 20th century, and eventually through the institution of state-nationalisms in India, Pakistan and Bangladesh. These state forms and gestalts of affect make it impossible to think of South Asia as a space of emotional inhabitance—like India or Pakistan. Yet, despite this state-nationalist borders of consciousness, there still exist long-term historical commonalities which people spontaneously practice and enjoy—in food, material culture, literature, art, music—which have a deep and long history. This marks peoples of this region (not states) by a common intelligibility which is reflected in their easy commingling when outside their countries, and free of the pressures of state nationalism, or in artistic public spheres.

Top of page

Full text


1In this paper I shall try to explore how the idea of space in what we call ‘South Asia’ today got reconfigured by modernity. I shall explore three spatial conceptions of South Asia—strategic, geographic and cultural; and I shall try to explain the patterns of space-thinking underlying each one of them, show their connection with modernity, and discuss why and how South Asia is still not a space that can be conceived as a space of belonging. I shall run two arguments together—the first about the historical transformations of identity in modern times, and the second about the poetic construction of affect for space-related terms like India, and Pakistan, but not South Asia.

2Spaces or space-terms are constituted by specific purposes. South Asia emerged in the 1950-60s as an academic-governmental term of American coinage designating a spatial area of concern for American strategy and foreign policy. This was a typical external term: i.e. a term by which outsiders designated a territory for purposes significant to them, but devoid of any affective significance for its inhabitants. In British and European discourse, this term was relatively rare, or absent. They saw this area as primarily the space of two states, India and Pakistan, born out of a partition of British colonial India. But British colonial thinking would not have regarded this event as either surprising, or as a loss, because of their long-standing belief that nothing held India together from the inside. It was only the iron frame of the colonial state that made it a single political entity.

3Two political processes of modernity introduce a new kind of space-making—nationalism and state-formation. Though these two processes are intertwined in the rise of nation-states, for analytical purposes it is better to see them as separate. In the paper, I shall follow some significant points of this dual process of emergence of new state structures in British colonial India and their animation by various emerging forms of nationalism. A sequence of nationalist ideologies appeared in this space—in 19th century Bengal, and eventually through the institution of state-nationalisms of India, Pakistan and Bangladesh. These state forms and gestalts of affect make it impossible to think of South Asia seriously as a space of emotional inhabitance.

4Yet, despite these state-nationalist borders of consciousness, there still exist long-term historical commonalities which people spontaneously practice and enjoy—in food, material culture, literature, art, music—which have a deep and long history. This endows peoples of this region (not states) by common structures of intelligibility, reflected in their easy commingling when they are outside their countries and free of the pressures of state nationalism, or gather in artistic public spheres. There is much to be said for a revival of this pre-modern world of frontiers, where people knew how to live their lives outside the coercive containers of nation-states, which unsuccessfully mimic forms of European modernity.

5My primary argument will be that there are two ways of thinking about identities and connections. The first evolved through long gestation in pre-modern history. This pre-modern practice of identity was entirely exploded by modern political conditions and replaced by identities constructed by the boundaries of modern states; but paradoxically, these modern identities are politically more powerful yet historically more fragile. With the slow decline in the legitimacy of state-produced identities, I think we can see the persistence of the pre-modern identity of graded connections. Since modernity reconstitutes both individual and collective identities, it might be useful to start with a brief discussion of both these historical processes.

On Identity

Personal identities

6In his work on identity and violence, Amartya Sen (2006) states that individuals have many identities; but that conflicts arise when some decide to ‘absolutize’ one of their identities over others. I agree with the political direction of Sen’s argument; but I would suggest a somewhat different way of conceiving personal identity. It is better to put it in a different way—an individual does not have many identities; rather each individual’s identity is complex. My identity is constituted by the combination of characteristics or attributes which individuates me, singularizes me against all other individuals. Of course this identity of mine is a combination of many features—which I would call identity-attributes. So my identity is constructed out of attributes like being male, Indian, Bengali, non-religious, a supporter of the left in politics, of Brazil in soccer, an admirer of particular types of paintings, poetry, music and many other things. All are my qualities as an individual; but these qualities fall into at least two categories; some are strictly individual like my taste in clothes, or my bodily features, but there are some like ‘being a Hindu/Muslim’ or a Labor Party supporter—which I can bear only because there are collective groups bearing those names. Though I can be the only living individual who loves chilly flavored chocolates, for me to be a Jew, a collective body—a Jewish community—must exist. I call these characteristics identity-attributes, rather than identity, because I believe there is another aspect to personal identity that is of equal conceptual significance. I have an identity-function—a capacity to select, emphasize and mobilize into appropriate social acts the congeries of attributes that I possess at any point in time. It goes without saying that these attributes can increase or decrease in number, or intensity. An individual’s identity function or what I could call the identity-verb can foreground and fade selectively different attributes in his/her social life. To say I have many identities is confusing; this elementary distinction between attributes and function allows us to save the correct intuition behind Sen’s argument. Personal identity is a fascinating subject in itself; but I do not want to pursue that theme here.

Collective identity

7I have argued for some time that modernity deeply affects collective identities: it is not that collective identities did not exist in pre-modern times; but the character of collective identities changes as a result of two characteristically modern processes (Kaviraj 2010). The first of these are the distinctive cognitive apparatuses of modern societies; and the second are the peculiar features of modern state-power. This argument is misunderstood at times. Some suggested that I claimed pre-modern people had no collective identities, or did not have techniques for large counting. My argument was different. In pre-modern times states existed but they lacked both the incentives and the cognitive apparatuses to produce large projects of counting—like the modern census and territorial mapping. More significantly, the production of knowledge regarding spaces and populations seeped into ordinary people’s consciousness and everyday political conduct in modern times. Some people may be innumerate, but they would know of the existence of majorities and minorities, and more crucially, they would know about how to do politics with this knowledge. People do not need to possess the techniques of counting to be affected in their political and social lives by the consequences of political counting. Two other supplementary arguments were involved in the discussion. In pre-modern contexts, social agents would stress different attributes of their social selves context-dependently. In given contexts an individual might emphasize his identity as carpenter, or as a Bengali-speaker, or as a Vaisnava; and rarely would his inclusion in a particular political principality constitute a highly significant part of his identity. At least, it would not be the case that he would place this identity above all else— producing an inflexible ranking of his identity attributes and placing his state-citizenship at the top. Associated with this was another suggestion that some modern cultural processes—like the rise of print—produced a deep standardizing and homogenizing effect—erasing gradually a world of frontiers and replacing it by a world of linear boundaries. It can be pointed out, correctly, that a pre-modern world of frontiers, less categorical conceptions of political identity still lurks in the interstices of the social world; but there can be no doubt that the modern categorical identity produced by processes of enumeration enjoys a dominance in modern times. Large identities like the Muslim umma— the worldwide religious community of Muslims, or the Persianate literary culture inhabited by Persian, Indian and Central Asian literati, or sectarian identities like Vaisnavas or Saivas—of course existed in pre-modern times and were entirely intelligible to their inhabitants, but the main suggestion was that the enumeration processes initiated by the modern state converted them into more ‘objectified’ identity groups. Bernard Cohn (1996) suggested that this objectification meant fixing them permanently and clearly in the minds of their members; my concerns were more political. I stressed the fact that these identities became comprehensive, abstract and agentive. It became increasingly possible to conceive of these identities as comprehensive: when there was talk of Muslims, it more seriously included in its ambit literally all Muslims, not a local, specific, or contextually delimited group. Obviously, such comprehensive communities needed to be abstract in the specific sense in which Benedict Anderson (1991) spoke of nations as imagined communities. The most significant consideration for an understanding of politics is their agentive character. They designated not groups of people who could fall under some specific description, but groups which putatively acted as single collective actors. Under this altered optics of belonging, agency and responsibility, it becomes possible to be fluent in the deadly languages of modern hatred. Muslims are seen to have destroyed the towers at the World Trade Centre, in retaliation for indignities that Americans had heaped on the populations of the Middle East. It is pointless to show that this language uses an improbably expanded notion of political agency and moral responsibility, or that it does not allow for a slope of agency and culpability inside the hateful collectivities. It is the powerful, easy, incendiary, contagious prose of global politics; and it is a typical product of the cognitive, communicative, political conditions of modernity.

On Spatial Imaginaries


8In this outline of historical change one highly significant element is missing. In this form, the story appears to be entirely one of explicit or subtle coercion—by states or elites who capture and control political power. But this is misleading: the power of elites, at least in the colonial and postcolonial historical worlds, is not simply the power of coercion. In colonial times, power is not in an unfragmented way in their hands. They certainly wield power in society, especially in non-political sectors of social life, and in cultural spheres; but overt institutional political power is not in their control. Indigenous elites certainly absorb and deploy what Charles Taylor (2003) has called an entire range of ‘modern social imaginaries’. In colonial historical contexts in particular it is crucial to remember that political power of the indigenous elites is initially constituted through the formation of a nationalist imaginary rather than institutional power of the state. It is the emergence of the ideal of modern nationalism that produces the strange love of the land that marks modern political discourse, and creates an entirely unprecedented connection between fixed political space and a powerful emotion of inhabitance. Poetry invariably plays a critical role in creation of this affect, reflected in the urge of all modern states to set their pretensions to glory to music in the national anthem. I shall now turn to the history of the creation of this unprecedented and quite unnatural emotion in colonial India. Despite the apparent naturalness of the uplifting gestures by athletes on the Olympic podium, of soldiers saluting their flags, and the ubiquitous display of this emotion in public stages, historically there was nothing ‘natural’ about the emotion that attaches people to space.

9It follows from my previous argument that this emotion of attachment to a state could not arise in pre-modern conditions. States were often in control of spatially immense empires; but their territories were not sufficiently stable to even propose, let alone effectively create, this kind of emotional connection. Empire states, which were the common form of political organization before the European invention of the nation-state, simply could not produce a comparable emotion. Yet there were other conceptions of spaces and communities of people that were available for patterns of social action. Despite sharp sectarian divides, Hindu communities shared texts, styles of religious observance, pilgrimage circuits, which defined a recognizable space for them. Clearly, Muslim communities conceived of various kinds of shared spaces. They recognized an abstract Muslim world community confirmed by arrangements for haj pilgrimages supported by Islamic rulers, like the Mughals. Also the travels of individuals like Ibn Battuta (1304-1337) evidently showed they had a clear conception of an Islamicate world that stretched from North Africa to Java. We can find a sense of a similar ‘world’ in the sociological reflections of Ibn Khaldun (1332-1406). A Persianate cosmopolis flourished till the 18th century spanning Persia, Central Asia and North India through which authors, works and political adventurers travelled freely in search of fortune. This could not have happened if the relevant agents and social groups did not regard these as recognizable spaces where they expected to be understood, appreciated and welcomed. But the effective verb for their relationship with those spaces would be ‘to know’ rather than ‘to own’. Besides, the knownness of the spaces was related to specific functions or purposes. Sufi religious figures often travelled from Persia or Khorasan to settle down in India. Afghan military groups were involved in political warfare down to the times chronicled in the Seir Mutaqherin in the 18th century (see 2006 edition). There is however a fundamental difference between familiarity and ownership. Several features of this world are noteworthy for our analysis of spatial history. First, the spatial conceptions/pictures/images were frontiered, not bordered. One region of literary style or political control ended and another began rather approximately and tentatively; not like modern frontiers in linear boundaries across a map of objectified space. Bernier (2011 [1672]) records in his Travels his surprise at meeting ambassadors from the imperial Chinese court who were utterly vague about the precise limits of their empire and that of the Mughals. Second, this was itself partly due to the instability of political control over territory. By the fickle fortune of military conquests, territories of imperial states expanded and contracted like a concertina in very short periods. This instability denied the rulers the opportunity to establish a firmly common culture; and to the populace the time to develop intense political loyalty. In any case, the nature of pre-modern political authority precluded the possibility that people would regard this loyalty as a loyalty to their collective self—the crucial element in nationalist thinking. Ordinary communities in agricultural societies learnt the fine art of living with their heads down to allow the storms of political strife to pass overhead with minimal damage. Basically, the conditions of political life simply did not allow the kind of stable connection between bounded territories and identifiable sedentary populations required for the development of the poetry of spatial affection. Finally, and critically, the boundaries of spaces were not so final and impermeable. Boundaries of different types of spaces did not coincide—the spaces of culture, economic activity, political control, educational styles were not the same as the boundary of political power. Spaces of different activities were differently configured. Pre-modern political regimes did not produce the decisive container effect of modern sovereign states. Whatever is a significant form of social activity today, it is bounded by the borders of the sovereign state. Before modernity, societies lived in a fundamentally different configuration of spatial structures. An ancillary effect of this re-arrangement was the slow acknowledgement—not always without resentment and dissatisfaction—of the role of the state as the ultimate producer of rules and disciplines for all fields of social life. Emergence of sovereign states did not merely alter the nature of relations between states in the international arena; it was the most significant alteration of the relation between political power and the ‘internal’ spheres of social activities of all kinds. Consequently, the definition of politics changed: instead of being diverse patterns of activities by which rulers sought to keep and expand their power, and equally diverse ways in which ordinary people tried to cope with their exactions, it slowly turned into the name of the activity that dealt with a restructured form of power—with much tighter conceptions of the relations between the rulers and the ruled—keeping this container as an inescapable part of the furniture of social existence. The pre-modern configuration of non-coincident structures of space—whether the space of cultural connection, or of religious life, or economic or commercial regionality—passed irreversibly. In terms of the lives of ordinary people, it signaled the end of the pre-modern world of graded intensity of recognition, neighborliness and inhabitance.


  • 1 Despite the anachronistic poetic redeployment of Rana Pratap or Shivaji in the cause of Hindu natio (...)

10We tend to forget that living is a partly lyrical activity. It takes imagination to inhabit a space, not just a material submission to its brute-fact features. There is usually a subtle connection between the language of material necessity and the language of lyrical emotion: we should not forget again that the refinements of poetry become possible by working on distinctions that really exist. The Eskimos are said to have thirty-one different words for various kinds of snow. I realized only after coming to live in England why the English language had so many words for what we in our unrefined tropical language simply called brishti (rain). Given the historical conditions of pre-modernity it is not surprising that there were no anthems to the Mughal Empire, or a song celebrating the Persianate literary sphere, or a military tune taking pride in the very mixed soldiery of the Mughal or Sikh rulers. Soldiers laid down their lives in the service of the armies in which they fought; but it is doubtful they viewed this as sacrifice in the emotional cause of a watan, or an inviolable but imperiled motherland.1 Thus the emergence of this strange love of the land—which had no precedent in earlier history, or could not have been said to spring naturally from the human breast—is an interesting historical question. We have to provide an historical account of how something so utterly unnatural became so utterly taken for granted.

11There is an unnecessary controversy about the historical sequence of the relation between the nation and the state; partly compounded by the fact that the nation is a sentiment largely intangible, and the state is a complex of institutions that are relatively easy to date with precision. It is correspondingly harder to decide when a sentiment of nationalism arose, much easier to date a process of state formation. Ernest Gellner (2006) contributed to this controversy by pointing out that although it is more reassuring to nationalists to think that a sense of nationalism arose first and demanded and realized a state, in fact, in European history generally sovereign states were formed by powerful rulers, and within those increasingly sturdy containers, a single education system slowly prepared a singular culture which self-celebrates its community as the nation. So there is an—almost necessary—inversion between the imagined sequence of nation > state, and the historical sequence of state > nation. The controversy was fuelled by his remark that the ‘historical sense’ that accompanies nationalism is usually ‘fraudulent’, which showed a glimpse of an excessively objectivist understanding of history.

  • 2 Rabindranath Thakur, Gitabitan [Akhanda], Visvabharati, Kolkata, 1970, 818: the section is appropri (...)

12In India, historical conditions proved more complex, giving some credence to both sequences. I have argued elsewhere that the idea of India in a political sense was not a discovery as Nehru’s titled implied (2002 [1946]), but an invention of the 19th century. It required as a condition the firm control over territories that British colonial rule established over the Indian subcontinent, though the concertina effect continued in territories that lay in the North-East, Burma, and North-West of the empire. Subjection to an identical administrative and political system of sovereignty was reflected in the creation of a common education system producing similar cultural effects. British colonial discourse of course stressed the indispensability of the colonial frame for this unity, and claimed that when this iron frame was removed, ‘India’ would fall apart reverting to its pre-modern plurality of political units. But slowly the experience of common subjection to colonial rule produced a sentiment of resentment among Indians. I have argued before that Bankimchandra Chattopadhyay’s (1838-1894) literary works played a defining role in the coalescence of this sentiment. Drawing from three entirely disjointed discursive traditions, he produced the defining figure of the Motherland worshipped with a new kind of emotion of tremulous intensity in his song Vande Mataram (1882). Pre-modern culture contained discrete discourses which celebrated the beauty of nature, illustrated by Kalidasa’s (5th century C.E.) poetry, the religious traditions of worship of the figure of Shakti in the beneficent form of Durga or the avenging form of Kali; and it also had purely mundane terms of desa or ksetra which referred emotionlessly to physical space or territory reflected in Sanskrit terms like Kuruksetra (the territory of the Kurus) or Sauviradesa (the country of the Sauviras) which contained no affective surplus beyond a simple physical reference. It is essential to point out, in analyzing this mixture, that although the two other discourses had association of distinctive affects, these were separate and unrelated. Kālidāsa’s nature poetry in the Meghaduta (‘The Cloud Messenger’) or the Rtusamhara (‘A Round of Seasons’) exhibited a highly sophisticated taste for literary presentation of nature—entirely literary, and entirely secular, without any hint of religious devotion. It lacked any suggestion of people belonging to the land or the land to the people. The devotional attitude towards Shakti—though intense—had nothing to do with a sense of space, territory, human inhabitance—not to speak of any form of political belonging. Thus, it is true that these three separate discursive elements existed in earlier culture; but Bankim’s combination of these three into a combustible fusion was an entirely innovative literary act. In the original song itself there is an innovative combination of redirection of worshipfulness towards the land. Now she is re-imaged as Mother-land to whom the fierce devotion appropriate to the Mother-goddess could be re-directed. The characteristics of this Mother are not the conventional godly qualities of sustenance, protective power, kindness and affection; but natural features—being endowed with ‘good’ water, ‘good’ fruits, cooled by the gentle breezes of spring, greened by abundant crops, whitened by nights of the full moon, ornamented by flowering trees—entirely ungrammatical qualities to attribute to a figure of traditional Godliness. The traditional Goddess was—in principle—a protector of all mankind, particularly of those who have sought her protection. The new Motherland deity is a fiercely parochial and partial goddess: she bestows her affection indiscriminately to all her children; not to others. The second unusual feature is the resonating emphasis on the inclusion of the population, her children, in herself, with a cascading repetition of large numbers. This is not merely true of the original song, Vande Mataram; numerous imitative compositions, which mirrored and multiplied this new sentiment, and quoted the phrase, stressed the same numerical and aggregative impulse. A Tagore song, composed soon after, went:2

Ek-i sutre bandhiyachi sahasra-ti man

Ek-i kaje sanmpiyachi sahasra jivan

Vande Mataram Vande Mataram

We have stitched a thousand minds together in a single thread

We have devoted a thousand lives to a single (common) task

Vande Mataram, Vande Mataram.

13Clearly, the song has succeeded in working an astonishing transformation: it is in the process of converting a religious chant into a political slogan.

14After this first, founding moment, this new sentiment suddenly had acquired the initial elements of an affective language; and as political nationalism started to expand, literary productions in this language multiplied exponentially, with a wide range of poets of varying degrees of talent composing patriotic poetry. Audience response to this genre tended to not be exacting: what readers and singer wanted was more the intensity of patriotism than poetic quality. Within a surprisingly short time, this new genre developed an immense corpus. Two trends are notable in the further historical career of nationalist poetry. First, Bankim’s original composition was intense in its language, but it was not in every sense a poem of excess. It did not take long for nationalist poetry to find the path of exaggeration in celebration of the collective self. A song composed by a contemporary of Rabindranath Tagore (1861-1941), Dwijendra Lal Roy (1863-1913), declared India, accordingly, the finest place in the world—a staple of all nationalist poetic tropes:

Dhana dhanye pushpe bhara amader ei vasundhara

Ihar majhe ache desh ek sakal desher sera

Se je svapna diye tairi se desh smriti diye ghera.

In this bounteous earth, filled with wealth, food and flowers,

There is land that is the finest of all:

  • 3 Dwijendra Lal Roy also composed intense nationalist plays like Mebar Patan (Fall of Mewar).

That land is made of dreams and bounded by memory.3

  • 4 See Christopher Pinney (2004) and Patricia Uberoi (2006) on calendar art.

15Parallel to the poetic and lyrical elaboration of this image began a visual—iconic figuration of the Motherland by various schools of paintings. In paintings too we find an entirely parallel tendency towards excess and mood of suspended violence towards others. The gradual centrality of the trope of invincibility and a capacity for violence against enemies develops through an interesting process which requires separate elaboration. Initially, violence and brutality are seen as defining characteristics of the images of others—the enemies, oppressors and aggressors against the peaceable nation. Oppression requires rebellion, injustice requires requital, and these acts presuppose strength. Slowly, the Mother acquires features of violence to act against her assailants. Through the constantly transactive dynamic between the literary and visual arts, the initial visualization of the figure of the Motherland was portrayed by Abanindranath Tagore of the Bengal School. Influenced by techniques of Japanese painting, this early figure came out as ethereal, non-violent, carrying the associations of beneficence from the literary depiction by Bankim, entirely subtracting the militant, potentially violent features. As the visual image became more popular, it tended to incorporate more pronounced iconic features of Hindu martial goddesses, often riding an animal as a vahana (a deity’s vehicle or mount), and her warlike characteristics were correspondingly enhanced. In calendar art, which is a true mark of the popularization and simultaneous banalization of an image, the figure is often superimposed on a geographic map of India, and invested with features of grandeur and power—like an ostentatious crown and an array of weapons in her arms.4

16Patriotic poetry soon became common in all Indian vernaculars, with compositions from a large number of political poets. Two of Tagore’s poems later became national anthems for India and Bangladesh. Other major poets too composed patriotic songs, most famously Mohammad Iqbal’s popular Tarana-i-Hindi [Anthem of the People of Hindustan; 1904]. Not surprisingly, figures like Vinayak Damodar Savarkar (1883-1966), the founder of modern Hindu nationalism, also composed similar poems to express more intense forms of the modern emotion. What was important in this wave of poetic composition was not the quality of the poetry, but the ubiquity of the sentiment. Poetry is the specialized language of the sacred, almost a dialect of sacrality. Prose could ordinarily present mere arguments; poetry elicited emotions. The nationalist sacralization of the land, its elevation, from a mundane expanse of material territory to the lyrical space to which people could attach with an intense filial affect, was accomplished primarily by the shift to the language of poetry and music. Despite the music and the poetry, however, nationalism remained a deeply political project with all the attendant difficulties of political power. Nationalist art expressed a dream of a nation, but in the colonial context, it was a nation that dreamed of becoming free, in less lyrical translation—of becoming a state. Thus this lyricism and poetry were always destined for eventual frustration, as ultimately, when political independence arrived, nationalist politics became inescapably involved in the deadly business of borders, and the finalist, decisive nationalist stamping of people as citizens. Particularly the brutal reality of partition devastated the tranquil symmetries of the lyrical idyll, and replaced it by a ferocious scramble for land. In a supreme irony, territorial divisions were often ultimately decided by low-level British officials untouched by the poetry of emotional inhabitance. Political reality made an utter mess of earlier dreams. Iqbal’s sare jahan se accha celebrated Hindustan hamara [better than the entire world is our India], though he later advocated a separate state of Pakistan, though the song continued to resound with the lines: mazhab nahi sikhata apas me bair rakhna [faith does not teach us to bear enmity towards each other]. Tagore’s Jana gana mana [Thou Art the Ruler of the Minds of All People, 1911], eventually the Indian anthem, included Sindhu, which incongruously formed an integral part of Pakistan. Even the bucolic invocation of Bengal in sarthaka janama amar janmechi ei deshe [blessed is this life that I have been born in this land] is troubled by the division of ‘my Golden Bengal’ being cut into two states. Dreams of a nationalist future and the reality of independent states collided without any resolution. Understandably, all states in the region set to the task of creating appropriate vehicles for their new politically inviolate territories by commissioning new poetry or editing old ones. Interestingly, in both India and Pakistan, songs to extol the sacrifices of martyrs in the repeated wars were composed—but these enjoyed more transient and fragile popularity. The spontaneous connection that dreams and poetry had with the futuristic nation could hardly be established with the institutional materiality of successor states.

Vestigial presence of the pre-modern

  • 5 See Iqbal’s speech to the Muslim League conference in Allahabad in 1930.
  • 6 Iqbal (2013 [1930]).

17Even in the period of high nationalism, there existed strands of thinking that were often sharply critical of militant patriotism. Authors have already noted that two of the great figures of Indian nationalism, Tagore and Gandhi, were deeply mistrustful of nationalist sentiment, as they observed development of nationalist paranoia in Europe and its slide into the utter inhumanity of warfare. Both men warned against an imitation of European-style nationalism centered on the state. On the Islamic side, Iqbal was alarmed by the prospect of an introduction of European-style nationalism in India’s highly diverse society;5 and warned on the other side against what he called ‘Arab imperialism’.6 Alarmed by the strand of excess in Bengali patriotic poetry, Tagore began to compose some highly uncharacteristic ‘patriotic’ songs which deplored the nation’s ability to overcome deep exclusions. For example, the poem Bharat-tirtha said:

Rana-dhara bahi

jay-gan gahi unmad kalarabe

Bhedi maru-path giri-parbat jara esechila sabe

Tara mor majhe sabi biraje keha nahe nahe dur

Amar sonite rayeche dhvanite tar-i bicitra sur

All those who came down the runnels of wars,

chanting songs of victory, amidst an insane tumult,

making their way through mountains and deserts, they all exist inside me, there is not one who is distant;

My blood throbs/resonates with the rich music of their diversity] (translation by the author).

  • 7 See for the extended argument about subsumption, Sudipta Kaviraj (2003).

18In the 19th and early 20th centuries, a particular strand of nationalists began to argue that there was a long-term tendency in Indian history towards ‘unity in diversity’. This became, for understandable reasons, the state ideology of post-independent India, and degenerated inevitably into a cliché. For equally understandable reasons, initially Pakistan presented a self-image of a more united, homogeneous nation of South Asian Muslims. But the stubborn fact of the sociology of diversity against the integrationist demands of the modern state has remained an imaginative difficulty for both states. In both cases the actual politics of the everyday have deployed two kinds of political resources to deal with the problem of identity. State-nationalism everywhere is a daily referendum: so states need techniques by which they remind and exhort individuals to organize their identity-attributes in the ‘right way’, giving privilege to their attribute of citizenship, placing that identity-mark above everything. Despite the existence of explicit state institutions that emphasize citizenship, in actual social life both politicians and ordinary people depend on and return to the repertoire of graded connection and neighborliness inherited from a very long pre-modern tradition. We could probably advance a stronger argument. Through long periods of experimentation in Indian history, through a social learning process, empire states fashioned a logic of accommodative relation between political power and popular identities. Empire-states were primarily based on a logic of subsumption,7 allowing both identities and authority at a higher spatial scale to subsume those at lower levels. The Mughal Empire, the last instance of a pre-modern empire-state in South Asia, exhibited this logic in its states practices, and often, in public debates and statements, its leading figures sought to articulate a theory of this structure. Colonial history presented a fundamentally different model of the European nation-state to the imagination of local elites. Familiarity and desire for the quite different European model of sovereign nation-states instigated sections of these political elites to try to emulate that structure, based on the logic of sovereignty and cultural homogeneity. Conceptions of radical modernism in India, and of an Islamic state in Pakistan and Bangladesh have periodically mobilized that utopia of sovereignty. In actual political and social life, people depend much more on long-standing traditions of cultural pluralism—which deploys the pre-modern techniques of graded similarity and neighborliness. Social peace in the existing states of South Asia depends on practices of everyday life drawn from pre-modern times to a much greater degree than we ordinarily admit.

The ideas of South Asia

Statist conceptions

  • 8 It is common to emphasize the political differences between India and the other South Asian societi (...)

19I would like to turn briefly to the ‘idea’ of South Asia and its discontents. The idea of ‘South Asia’ exists today in two very different versions. As academics we are familiar with an academic conception of South Asia which emerged typically in the 1950s American academia, driven by the requirements of US foreign policy in the postwar world. This optic aggregated spaces from a geopolitical point of view, disregarding political and cultural differences—which may be significant for their inhabitants, but unimportant for the framing of American strategy. There is considerable literature on the history of this notion which shows how a term prompted by military calculation slowly morphed into a respectable disciplinary entity in academic scholarship. Gradually its geopolitical connotation was replaced by a different content, as postcolonial states grew and required more serious academic scrutiny. There was a concern to avoid study of the region exclusively centered on India, or one of the other post-colonial states. I am skeptical of the cognitive advantages of this move—whether study of India as part of a South Asian department substantially changes how scholars perceive the object of their enquiry. It should encourage more comparative research; but comparative analysis is plagued by an ambiguity in its self-definition. There is still a great deal of uncertainty about whether the starting point should be the historic similarity of the societies or the dissimilarities of their contemporary political systems.8

20An entirely different approach to a pragmatic conception of South Asia could be found in the tentative, intermittent attempts by the states to gingerly approach proposals of regional cooperation—on the lines of European integration. There are obvious arguments for the advantages of economic and market integration, or educational and cultural exchange, and the possibility that establishment of such connections and interdependencies might reduce political tension. These initiatives have stuttered through the recent decades, and have been plagued by anxieties about Indian dominance. Compared to the relative successes of Europe, South Asia has failed to evolve even a prosaic idea of political convenience.

Evading the states

21These disappointments are inherently related to a statist perspective on identity. We are forced to reflect on the immensity of transformation that modern politics has brought to the way people cognize and practice space. Through its long history what we now can call ‘South Asia’ was a space that was intimately recognized, but formally unnamed. It was a vast expanse which contained hundreds of political formations in which neither the states made exclusive claims on its subjects, nor did the people on their states. Technology made it hard to move easily; and oppressive social structures often made mobility impossible. But the political regimes did not act like containers or cages. Borders were shifting, ineffective and porous. It is remarkable how the power of the modern state—which can use a form of violence that is implicit, not obvious – has created containers for identity and effective circulation of political acts, of economic goods and cultural goods. State boundaries have made two cities like Lahore and Amritsar, separated by a few miles, much more distant than Kolkata and Chennai. Under modern political conditions, it is exceptionally hard to evade the state.

  • 9 Like Iqbal’s Tarana-i-Hindi (see above).

22Yet, it can be argued that, as an historical project, in some ways the containerization of singularized state identities in our region has not entirely succeeded. This is true in at least two ways. First, the creation of the states of India and Pakistan has not successfully erased the identity pluralities of previous history. States have been forced to different degrees to come to terms with internal pluralism. Second, in many respects, the pre-modern disjunction between different kinds of spatiality—of politics, economic connections, cultures and literary public spheres—has survived. Identities have remained stubbornly plural, and attempts to force them into a single stable hierarchy sanctioned by the state failed. Commerce notoriously evades political control. The difficulty of policing the Bangladesh-India border is that there is an underlying economic regionality that crosses the boundary. Laborers cross borders driven by economic incentives. The populations of these states have historical cultural commonalities. Urdu literary culture straddles the boundary between Pakistan and India; just as Bengali culture does between India and Bangladesh. Different socialities of space find ways of evading the brutal enforcement of state borders. State borders and efforts at containerization are frustrated as much by historical forces as by forces of the future. Globalization processes have added to the difficulty of maintaining the boundaries as the impermeable limits of all forms of activity. Still, there is no denying of the dominance of states and borders in contemporary political practice. It is highly unlikely that there will be a tarana9 for South Asia as a poetically felt space in the near future. But if we go back to the initial argument about two senses of space and inhabitance, we can observe a robust presence not of a modern South Asia created by the collaboration of states, but of a vestigial pre-modern South Asia present in social transactions.

23Let me illustrate this by a Tariq Ali’s interview. In an interview on Al-Jazeera television he was asked to recount his long experience of living in Britain, and the interviewer questioned him particularly about hardships on first arriving in England. He answered that he was revolted by the food served in college; and remarked to someone that ‘we do not serve food of this sort even to animals back in our place’. This sense of ‘our place’ was not merely vivid but real. It was not an act of memory; but also a reference to the present, not just a remark about culinary taste. Historians like Braudel have noted that material cultures have the longest temporal rhythms, and are the hardest to shift. It is amusing how the powers of the state have been ineffective in partitioning material life, particularly aspects like cuisine—which represents one of the fundamental ways in which humans interact with nature. Despite the inflamed and militarized frontier between our two countries, any culinarily un-cosmopolitan Indian would immediately understand what Tariq Ali meant and sympathize. Theorists have neglected to notice that food, as much as commerce, tends to seep out under the boundaries of states. Movements of culinary taste have proved impossible to control by stringent rules of immigration. When students arrive in alien cultures, they always tend to congregate along these lines of common culture. Commercial centers like Southall in London or Gare du Nord in Paris stand witness to the shared material cultures of South Asian peoples. Remarkably, when states do not intervene, cultural commonality tends to reassert itself. Isolated in a surrounding culture which ignores or devalues them, South Asians rediscover their shared relation to the natural world. Bhudev Mukhopadhyay (1827-1894), a 19th century Bengali social theorist, argued that in people who live in the same material world history produces strong filaments of samaduhkhasukhata—identical experience of pleasure and pain. Of course, this was a travesty of the term’s original meaning in the Gita, which exhorted wise people (the sthitaprajna) to consider pleasure and pain as identical. But its new meaning—in Bhudev’s hand—captures something vital in historical experience—to travelers from the subcontinent (which is our embarrassed word for invoking the continued presence of this pre-modern historical space) Southall cuisine produces delight and British food despair. Culture has ways of escaping the invigilation by states, and of creeping out of the containers of modern power. If we can overcome the comprehensive devaluation of all things pre-modern, we could become less embarrassed in celebrating these connections—which are real and deeply felt, though unlike nationalism, they lack the endorsement of magnificent poetry. Though it is possible to suggest that in the sincere appreciation of ‘classical music’, sung often through the words of medieval poets like Kabir, we still have an equally powerful poetic celebration of pre-modern neighborliness. In those words, and in the wordless expressivity of instrumental music, we may have the half-forgotten anthems of a wistful common culture of the space called South Asia.

Top of page


Anderson, Benedict (1991) Imagined Communities: Reflections on the Origin and Spread of Nationalism, London: Verso.

Bernier, François (2011) Travels in the Mogul Empire, Cambridge: Cambridge University Press, [1672].

Cohn, Bernard (1996) Colonialism and its Forms of Knowledge: the British in India, Princeton: Princeton University Press.

Gellner, Ernest (2006) Nations and Nationalism, Oxford: Blackwell Publishing Ltd, [1983].

Iqbal, Muhammad (2013) Reconstruction of Religious Thought in Islam, Stanford: Stanford University Press, [1930].

Jalal, Ayesha (1995) Democracy and Authoritarianism in South Asia, Cambridge University Press, Cambridge.

Kaviraj, Sudipta (2010) The Imaginary Institution of India: Politics and Idea, Columbia University Press, New York.

Kaviraj, Sudipta (2003) ‘The Postcolonial State in India’, in Quentin Skinner & Bo Strath (eds.), States and Citizens, Cambridge: Cambridge University Press.

Khan Tabatabai, Gholam Hossein (2006) Seir Mutaqherin: Or View of Modern Times, Being a History of India, from the year 1118 to the year 1194 A.H. (1781-82 A.D.), Lahore: Sang-e-Meel Publications.

Nehru, Jawaharlal (2002) The Discovery of India, Oxford: Oxford University Press, [1946].

Pinney, Christopher (2004) Photos of the Gods: The Printed Image and Political Struggle in India, London: Reaktion/Delhi: Oxford University Press.

Sen, Amartya (2006) Identity and Violence: the Illusion of Destiny, New Delhi: Penguin Books.

Taylor, Charles (2003) Modern Social Imaginaries, Durham: Duke University.

Uberoi, Patricia (2006) Freedom and Destiny: Gender, Family and Popular Culture in India, New Delhi: Oxford University Press.

Top of page


1 Despite the anachronistic poetic redeployment of Rana Pratap or Shivaji in the cause of Hindu nationalist history.

2 Rabindranath Thakur, Gitabitan [Akhanda], Visvabharati, Kolkata, 1970, 818: the section is appropriately called Jatiya Sangit (Songs of the Nation).

3 Dwijendra Lal Roy also composed intense nationalist plays like Mebar Patan (Fall of Mewar).

4 See Christopher Pinney (2004) and Patricia Uberoi (2006) on calendar art.

5 See Iqbal’s speech to the Muslim League conference in Allahabad in 1930.

6 Iqbal (2013 [1930]).

7 See for the extended argument about subsumption, Sudipta Kaviraj (2003).

8 It is common to emphasize the political differences between India and the other South Asian societies; but a different argument can be found in Ayesha Jalal (1995).

9 Like Iqbal’s Tarana-i-Hindi (see above).

Top of page


Electronic reference

Sudipta Kaviraj, A Strange Love of the Land: Identity, Poetry and Politics in the (Un)Making of South AsiaSouth Asia Multidisciplinary Academic Journal [Online], 10 | 2014, Online since 25 December 2014, connection on 30 May 2023. URL:; DOI:

Top of page

About the author

Sudipta Kaviraj

Professor of intellectual history and Indian politics, Middle Eastern, South Asian and African Studies, Columbia University

Top of page



Creative Commons - Attribution-NonCommercial-NoDerivatives 4.0 International - CC BY-NC-ND 4.0

Top of page
  • Logo Centre d'études de l'Inde et de l'Asie du sud
  • Logo European Association for South Asian Studies
  • DOAJ - Directory of Open Access Journals
  • Journal supported by the Institut des Sciences Humaines et Sociales (CNRS)
    CNRS - Institut national des sciences humaines et sociales
  • OpenEdition Journals
Search OpenEdition Search

You will be redirected to OpenEdition Search