The Diasporic Imaginary

Violence and Diaspora

Diaspora studies is becoming popular as a field of inquiry in both the social sciences and humanities. And not without reason; the study of diasporas opens up new points of investigation into nationalism, at the same time demanding that we rethink belonging within a global context. Its evident utility notwithstanding, however, I am concerned that the field's emphasis on an analytic model of place—central to many studies—will ultimately preempt any serious accomplishments. Place has primarily been developed to identify a diasporic people's "place of origin." This very common analytic posits that a homeland is originary and constitutive of a diaspora, and very often it supports an essentialization of origins and a fetishization of what is supposed to be found at the origin (e.g., tradition, religion, language, race). ¹ Nevertheless, for many diasporic groups, place, or place of origin, is not the primary issue.

Against the analytic model of place, I propose the diasporic imaginary (Axel 2001). My proposal has two aims. First, it is intended to foreground violence as a key means through which the features of a people are constituted. Second, it is intended to account for the creation of the diaspora, not through a definitive relation to place, but through formations of temporality, affect, and corporeality. The diasporic imaginary, then, does not act as a new kind of place of origin but indicates a process of identification generative of diasporic subjects.

This essay elaborates what is at stake in moving from a model of an originary place to that of a diasporic imaginary through instances provided by the Sikh diaspora. Sikh life over the past sixteen years has been riddled with extreme violence. Because the story of this violence is rarely told, its narration becomes an explicitly political act. From the vantage point afforded by this new model, a consideration of the Sikh case may aid in a more general rethinking of diaspora studies by situating an understanding of the role of violence in the formation of diasporas at the intersections of globalization and nation formation on the one hand and within the vicissitudes of citizenship, gender, and sexuality on the other. It is from such intersections that the features of a diasporic imaginary may emerge most powerfully.

When I write of violence and the Sikh diaspora, I have something specific in mind: the particular forms of violence emerging out of the conflicts between the Indian nation-state and Sikhs fighting to create a homeland called Khalistan (Land of the Pure). In the 1940s, anticipating the British departure, the Akali Dal, the primary Sikh political party, invented the idea of Khalistan to set against the constructions of Pakistan and India. They wished to secure an official homeland for a Sikh majority population in the northwestern part of the subcontinent. Khalistan, however, is a territory that was never created, and since Independence in 1947, the desire for a Sikh homeland has cast a dark shadow over the part of northwestern India known by the name of Punjab—itself an unstable designation over the decades.

Since the 1980s, the idea of Khalistan has been revitalized and transformed by Sikhs living in India and around the world. A cycle of violent encounters between Sikh militants and the Indian government has to date left at least 100,000 people dead. These developments have had an indelible effect on the lives of Sikhs living around the world—numbering approximately 20 million—who have delivered the idea of Khalistan into a new transnational arena of knowledge production. In Sikh—or more specifically Khalistani—discourses, however, Khalistan, the homeland, is not necessarily conceptualized as an empirical place of origin that Sikhs wish to return to and reterritorialize. Rather, Khalistan is one among many points of orientation, or organization, for a globally dispersed people—a signifier of the sovereignty of an already constituted Sikh nation (qaum). ²

My account is concerned with the period between 1983 and 1998, and this calls for a further specification...