The costs of recognition: global politics, religion, and the colonial history of South Asia

Abstract

In this article I consider religion in international political scholarship and suggest a study of its epistemological politics and conceptual history. I argue that scholarship which strives to ‘engage’ or ‘recognize’ religion in global politics remain ignorant of the costs involved. Building on this argument, I ask if the troubles with recognizing religion reflect more basic qualities of recognition scholarship. Following the work by Jacques Rancière, Patchen Markell, Elizabeth Povinelli, and Jens Bartelson I argue that recognition has two faces and that along with its frequently acknowledged empowering aspect, it also comes with costs. In order to assess the costs of recognition I propose a study of its conditions of possibility, that is, a study of the ways in which the subjects of recognition become recognizable as such. In the final section of the paper, I apply this to the example of religion in global politics and the formation of the Muslim subject in the lead-up to the partition of British India and the founding of Pakistan.

Introduction

Throughout the last three decades religion has increasingly become part of a public vocabulary to make sense of international conflict dynamics while also describing the contours of cultural, epistemic, and ontological diversity.¹ Governments, international organizations, and universities have invested vast resources in assessing ‘religious conflicts’ and in engaging with ‘religious’ institutions and actors, their ideas, claims and visions of the world. While the place and role of religion in global politics continue to be disputed, international policymakers and scholars alike agree that one can no longer address world affairs without it: religion demands recognition.² However, as decades of scholarship in religious studies has shown, religion – just like sovereignty, democracy, or freedom – is an unstable concept encompassing a shifting order, range and scope of social relations, actors, institutions, ideas, and practices. If the object to be engaged with – religion – is continuously shifting, then what exactly should scholars engage with, recognize, or include in our studies, policies, and predictions?

In this paper I argue that there are costs involved in recognizing religion in global politics that are neither sufficiently understood nor appropriately evaluated. These stem, among other things, from the fact that recognition processes presuppose the prior identification of entities and subjects ready to be recognized.³ Acts of recognition build upon an assumed ‘fact’ of recognizability and so, in order to become recognized, a subject must be or become recognizable to a regime of knowledge. Arguments that we should recognize religion in global politics ignore, therefore, the conceptual, epistemological, and historiographical politics of recognition, instead conceiving of religion as intelligible prior to recognition as something that can be included or excluded, governed, managed, and engaged with.⁴

Building on this argument, I ask if the troubles with recognizing religion in the study of International Relations (IR) reflect more basic qualities of recognition. After an introduction to the arguments for the recognition of religion in global politics, I study recent critical debates about recognition in relation to individuals and groups, minorities, nations, empires, and states. Building on the work of Jens Bartelson, Patchen Markell, Elizabeth Povinelli, James Tully, and Jacques Rancière, I argue that recognition has two faces and that along with the frequently acknowledged empowering and emancipatory aspect of recognition, the conditions for its possibility are also bound up with costs. One of these conditions is that arguments for recognition both presuppose and reproduce a differentiated social logic; that is to say, a logic which assumes an identifiable and differentiated subject. This subject acts as a benchmark for understanding whether or not a process of recognition has been successful. It depends on a prior establishment of the criteria of recognizability that one can either successfully meet or fail to meet. By studying the processes by which a subject becomes recognizable we can start understanding the costs that accompany recognition.

I end the paper by giving an indication of what such a study might look like. I examine the government of British India and the subsequent establishment of the Islamic Republic of Pakistan. I show how the quest for statehood, the role of minorities, political representation, and international border-making both shaped, and were shaped by the concepts, agents and identities associated with ‘religion’ that broke through the threshold of political recognition to establish themselves as taken-for-granted political entities on the global stage. I illustrate how claims for the recognition of religion are intimately connected to the colonial epistemological politics through which such processes played out and which costs they carried.

Religion in global politics: narratives of exclusion and arguments for recognition

In this section I look at arguments for the recognition of religion in global politics as illustrated through narratives of its exclusion and arguments for its rehabilitation. I point out that these arguments for recognition have costs as they both depend on and reproduce an intelligible subject of religion as already available for recognition. This is significant because scholars and policy makers alike use the conceptual apparatus of religion to make sense of the dynamics of global political order, including violent conflicts and peace building, state building processes, minority protection claims and violations, and the development of democratic institutions and aid programs. I end the section by suggesting that this is not exclusively a problem of religion in IR but is inherent in the grammar of recognition in a broader sense.

During the early 2000s interest in religion and politics grew exponentially in academia, as well as in policy circles and public discourse. With it a debate moved center stage regarding the secular foundations of the liberal political (international) order.⁵ The debate outlined and critiqued various secularist assumptions about political order and legitimate authority which explicitly saw religion on the outside, often as defining the limits of the political and the public.⁶ According to those critiques, religion was not as easily separated from politics as had been assumed and liberal secularity, as we know it today, was argued to be the result of a particular settlement in a particular historical period.⁷ The settlement was, in this sense, neither universal nor neutral, and the fact that its idea had been near to universalized via the institution of the liberal state did not make it a necessary condition of liberal democracy. According to this position, the secularism(s) underlying the Western liberal political order had a history, a genealogy, and was to be studied in detail rather than used as a measurement for democratic development.⁸

This secularism debate continued into the following decade and cleared space for a new form of engagement with religion, both in scholarship and in policy circles.⁹ I focus here on the former. In an attempt to re-read international political history and theory after the critique of secularism, Daniel Philpott saw religion as constitutive of the current Westphalian international order of sovereign states, since this order would never have emerged, were it not for the way in which the Reformation and Protestant ideas of political authority shaped the states' interests in sovereign statehood.¹⁰ ‘No Reformation, no Westphalia’ is the short version of the argument.¹¹ Scott Thomas continued to argue that assumptions about a secular Westphalian international system that had relegated religious conflict to the inner life of states skewed the understanding of the ‘resurgence of religion’ in the Global South, regions where religion had never lost its political and social salience.¹² In this sense when scholars retained Westphalia as a symbol for secularized IR the global resurgence of religion came to look like an internationalization of a private matter, threatening the international order.¹³ Following Thomas in an effort to de-securitize religion, William Cavanaugh argued against what he called the ‘secular myth of religious violence’.¹⁴ He pointed to the ways in which the emerging early modern state established a discourse of religious violence and the perceived necessity to remove religion from the public in order to shift loyalties from the religiously constructed identity of a community to the new territorial claim to power and authority of the state. Elizabeth Shakman Hurd argued in Politics of Secularism in International Relations that secularism cannot be written off as a fixed and final achievement of European modernity, showing rather that secularism needs to be considered a socially constructed and historically contingent form of religious governance which has consequences for domestic and global politics. Once secularism is seen as a mode of political governance rather than a stable and neutral solution to the problem of religion in public, it is possible to start looking at the evolution of different forms of secularism and their multi-faceted trajectories, histories, and political consequences.¹⁵

Following the deconstruction of the secular ontology of IR, other scholars from a variety of disciplinary backgrounds saw religion as potentially embodying the move to a cultural and epistemic diversification in the field. In undoing the knowledge-power regime of secularism IR scholars expected to broaden the range of actors, arguments, ideas, institutions, and forms of knowledge, opening up to ‘otherwise neglected forms of being, becoming and knowing’.¹⁶ Expanding the ‘hermeneutic register’ to reach localized cultural markers or vernaculars¹⁷ or establish an ‘open hermeneutic margin’ for non-Western actors to widen the resources of negotiation,¹⁸ to pursue an ‘ethos of engagement’ with a plurality of previously untapped moral resources,¹⁹ or alternative ontologies²⁰ was considered a value in and of itself. The value of diversity and plurality was particularly emphasized in relation to the decolonization or provincialization of an otherwise Eurocentric framework of global politics. ‘Taking views of other worlds seriously’, Erin Wilson writes, ‘requires that we do not attempt to relate different ontologies to one another through language and concepts that belong to only one of them’.²¹ Following the postsecular account of Jürgen Habermas, Mariano Barbato argued, that the semantic figures of religious communities might offer ‘resources for fueling deliberation processes with notions of arguing beyond narrow concepts of self-interest’ broadening the range of ethical possibilities that the instrumental reason of a secular system of thought would be unable to respond to.²²

This movement toward exploring previously excluded aspects of international life, however, not only identified new perspectives, and increased the range of relevant actors, institutions, or epistemologies but was also productive in defining who and what would count as belonging to the category of religion and thereby as worthy of recognition. In various ways but in a similar manner both the narratives of exclusion tracing what had been lost due to the ‘secularist’ marginalization of religion as well as restorative arguments and policies regarding what needed to be ‘brought back in’ relied on particular notions of what counts as ‘religious’ and what does not.²³ In these cases ‘religion’ referred to difference – cultural, ontological, or epistemic – which may not be fixed in substance but known when seen.²⁴ Even scholarship that pursued a more relational approach tended to treat religious identities, ideas, norms, and practices as independently constituted, putting pre-given actors, structures, and interest at the center of their analyses focusing instead on what Jonathan Agensky has called their ‘interdependence and strategic accommodation’.²⁵

The productive power of exclusion narratives and recognition arguments

If global politics are to be governed through the category of religion – aiming to solve ‘religious’ conflicts, engage with ‘religious’ actors, and protect ‘religious’ freedom – this will incentivize association accordingly.²⁶ As Hurd puts it in relation to the politics of religious freedom, ‘(g)overning religion as a right funnels individuals into discrete faith communities, empowers those communities and their spokepersons, and marginalizes other modes of solidarity’.²⁷ The risk of emphasizing the ‘role of religion’, then, is that boundaries between groups are made more salient and that this creates new forms of social friction defined by religious difference.²⁸ Responding to and governing of religious difference further ‘puts pressure on nonestablished, unorthodox, [or] nonconforming’ forms of religious life to yield to the recognizable versions thereof or risk being rendered invisible.²⁹ The burden of normalization is thereby thrown onto those who find themselves outside the realms of the intelligible, compelled to shape themselves in a manner that does not render them imperceptible to the various available forms of recognition and empowerment. Anthropologist Elizabeth Povinelli writes in this vein regarding the recognition of the indigenous Australian population, noting that they are ‘called on to perform an authentic difference’ and to ‘transport [their] ancient prenational meanings and practices to the present in whatever language and moral framework prevails at the time of enunciation’.³⁰ In order to be recognized, in other words, one has to be recognizable and if this is not the case, one needs to become so. To accept a process of recognition, as political theorist Sara Ahmed puts it, is to ‘value those who can “be heard and act” under its name’.³¹ Subjects in global politics, therefore, do not only exist by virtue of being recognized, but do so in a prior sense, by having become recognizable.³² International organizations, governmental foreign policies, local administration, and international law have shaped a new global politics of religion that defines the criteria of what it means to be recognizably religious and have, as Hurd shows, created new categories of actors in world politics.³³ Recognition as an actor, then, depends on a prior establishment of the criteria of recognizability that one can either successfully meet or fail to attain.³⁴ In this way both the new and never-conceived become ‘tied to that which is already cognized’.³⁵

In a paper published at the height of international scholarship's engagement with religion, Robert M. Bosco described this move to ‘capture the “global resurgence” of religion’ as a persistent Orientalist discourse inhabiting the ability to ‘seamlessly appropriate new phenomena into received representations’ while continuously overlooking the politics of the definition of religion.³⁶ The assumption that it is possible to engage with ‘religions’ in a neutral manner, however, masks the epistemic politics interwoven into the concept and its history.³⁷ While agreeing with Bosco that the integration of ‘religion’ into IR scholarship has been defined by the appropriation of pre-determined conceptions of what ‘religion’ is supposed to be and do, I argue that the engagement with ‘religion’ did not simply reproduce the existing conceptions of the term and the political orders and hierarchies it served but reflects a more basic problem with recognition. The point that I want to make is that these are not random examples of the productive and reifying power entailed in engaging religion in global politics and beyond. Rather, I see the problems of recognizing religion as reflecting a more general issue with the grammar of recognition, namely that recognition presupposes the prior identification of entities and subjects ready to be recognized.³⁸ The act of recognition, then, depends on an established ‘fact’ of recognizability. It assumes a subject that can be known and differentiated from that which it is not prior to the moment of recognition. In order to understand how religion became available as an object of knowledge that was possible to recognize and the marker of differentiation of subjects claiming recognition, we must know more about how it became recognizable in different circumstances as distinct and identifiable, and somehow different from politics, law, or culture. Once we have a better grasp on how religion became recognizable, we will be able to assess what costs accompanied those processes and how these were distributed. As I will show in detail below, the costs of recognizing religion in IR are on the one hand side epistemic in the assumption that there is an identifiable and differentiable subject that can be extended recognition to, that is, religious minority, religious argument, institution, or organization, a religious doctrine, idea, norms, or identity. Thereby, those outside this legibility scheme are either forced to claw their way back into intelligibility by the means at their disposal or risk remain unrecognizable. The costs are also context dependent and particular to each case, referring, for example to the marginalization of unrecognizable groups, the reification of social and legal structures and institutions, the heightening of conflict fault lines, and so on. I will return to the question of the costs of recognition in the next section and continue to illustrate examples of specific costs in the last section examining state-building efforts in South Asia.

Two faces of recognition

In his book Bound by Recognition, Patchen Markell illustrates the double nature of recognition by drawing on an analysis of the emancipation of European Jews in the late 18^th and early 19^th centuries. The emancipation had aspired for inclusion into the ‘civic’ nation, which meant access to citizenship rights, education, and the removal of discriminatory laws. However, the assimilation of Jews into the broader structures of European political communities did not simply imply that Jewishness was deemed irrelevant to political membership. Emancipation was, instead, both an effort to remove restrictions on ‘Jewish life’ but also, ‘an active effort to reshape Jewish … identity’.³⁹ While emancipation had provided recognition and empowerment to the Jewish population, it had also conditioned that empowerment by means of integration into a certain form of structure that encompassed political institutions, an educational system, and legal norms. In Enlightenment in the Colony Aamir Mufti continues that, as a recognized minority European Jews became vulnerable in novel ways. They became identifiable, quantifiable and, in a word, governable, as the contours of their communities were made more easily known and detectable.⁴⁰

As Markell puts it, by making the ‘distribution of resources and the institutionalization of rights dependent upon one's recognizability as the bearer of an identity’, the politics of recognition ‘risks subjecting the very people whose agency it strived to enhance to powerful forces of normalization’.⁴¹ Processes of recognition may give shape to the existing political order but they are also making it governable, helping to ‘create the world that [they] purport … to oversee’.⁴² The conditions of empowerment are thereby also the conditions of control.

Another example is the case of British Indian Muslims, to whom I will return below. They gained political influence as colonial subjects of the British Empire to the degree to which they were accountable and enumerable through, among other instruments, the colonial census and the governmental logics that structured it. Recognizing religious difference and the subjectivities tied to it empowered those who were identifiable within the logics of the census. The recognition of religious and cultural difference was, in this way, productive of the subjects that it named while also binding them to existing systems and structures of power – extending rights, legitimacy, and authority at the same time as implicating those subjects in its hierarchical structure.⁴³ A deeper understanding of these two sides of recognition – empowering and subjecting – allows us to account for the costs that accompany it. Focusing on recognition or misrecognition as a ‘fact of rightly or wrongly cognizing and respecting an already-existing identity’ distracts from the fact that this form of acknowledgment partakes in the constitution and government of the subjects in place.⁴⁴

In the next section I look at recognition theories in IR, arguing that several approaches incorporate and reflect this problematic assumption regarding the nature of the subject of recognition. I contrast the accounts of recognition with an alternative which echoes the shift from a struggle for recognition to a struggle over recognition gestured to by James Tully in his later work. This means that instead of arguing for the recognition of certain subjects – individuals, communities, or states – my focus is on the struggle over recognition's conditions of possibility. I then proceed to illustrate what that looks like in practice linking it back to the question of religion and the claims for recognition in the cases of British India and Pakistan.

Actualizing the potential: the productive and performative power of international recognition

During the 1990s, questions of recognition increasingly came into focus in political and international political discourse in the wake of a flourishing discussion of multiculturalism and identity politics at international and domestic levels.⁴⁵ In international political scholarship, this took the form of analyses of identity-based conflicts, used to explain why actors sometimes seemed to act against their material interests in pursuit of an identity-based goal.⁴⁶ In contrast to prevailing realist and liberalist approaches in IR, the focus on recognition emphasized the importance of identities and subjectivities alongside the more traditional quest for power or wealth and modes of coordination and cooperation beyond the state.⁴⁷ Recognition was considered, however, as not simply one goal among others, but a basic need.⁴⁸ Unlike other basic needs, such as shelter or food, that must be met to ensure the survival of the individual or collective subjects, the need for recognition was considered different – its very fulfillment served to constitute the individual or collective as a particular kind of actor. Recognition mattered in IR, because it described the process through which actors came to exist as actors within the international system and to take on a particular role and place within that system.⁴⁹ This constitutive theory added the productive, relational, and social power of recognition to declaratory or cognitive theories which relied much more heavily on an objective set of criteria necessary to obtain the sought for recognition. In this sense, recognition was not only cognitive of existing actors, but elevated into actorhood for those international entities that were recognized by others.⁵⁰

Recognition was considered in these accounts not only as part of the process constituting the (collective) identities of international actors but also as an intrinsic part of the dynamics of international politics and conflicts, and as a motor of change.⁵¹ By struggling for recognition, actors were understood as challenging prevailing political structures to establish a new order that better met their needs. This changing order would be continuously challenged by others to include an ever-growing community of recognized actors, perpetually developing toward ever-increasing inclusion. One example is IR scholar Christian Reus-Smit's book Individual Rights and the Making of the International System, in which he draws on Axel Honneth's Hegelian study of struggles for recognition and argues that the relational striving for, and realization of, individual rights has repeatedly driven large-scale change in the international system.⁵² Reus-Smit's argument is one example of a longer tradition in IR of tracing international dynamics back to the realization of a potential, imagined reality, or an as yet unrealized subjectivity that is made realizable through struggles for recognition.⁵³ Shannon Brincat also draws on Honneth's story of the subject's journey toward a progressively successful ‘self-actualization’, reshaping it as a quest for the cultivation of cosmopolitan social relations and an ethical life which ‘increases possibilities for social freedom’.⁵⁴ A decade earlier Alexander Wendt had told a similar, teleological story about self-realization and the historical evolution of the international state system from one of differentiated entities into a single overarching Self in the form of the World State.⁵⁵ Transcending the particular and developing into a universal entity, as Wendt envisioned, however, assumes a starting point of clearly distinguishable, bounded subjects – states in this case – that insist on being recognized in their particularity or difference. In other words, his teleology depends upon a preexisting logic of differentiation. The central point here is that for these authors there is, and needs to be, a difference that can be recognized and, subsequently, overcome for the process to move forward. There needs to be a recognizable difference between ‘Self’ and ‘Other’ before one can insist on the recognition thereof as a step toward reconciliation as a higher level ‘We’. Without such difference there would be no intersubjectivity but rather a form of radical fusion.⁵⁶

While the teleological aspect of the approach adopted by Wendt and others can be viewed as problematic in its own right, that is not my focus here. The important point for my argument, rather, is that such approaches require and presuppose a preexisting, latent subject, without which the struggle for recognition would not be possible. They require a differentiated logic, a differentiated social ontology, in order for the mechanisms of recognition to gain a foothold. With respect to scholarship positing recognition as a source of change and a means through which to actualize a potential self, the question remains: from where does the potential subject emerge on behalf of which recognition can be claimed (or denied)? How does that which is potential differentiate itself from that which it is not? How does one traverse the threshold of recognition? There is a crucial step missing in the recognition-based arguments considered above – namely an explanation for how the subject on behalf of which recognition is claimed, granted, or denied can be assumed.

As we saw above, in much of the scholarship on recognition in international politics, the game of recognition – demanding, extending, or receiving it – is conceived of as part of the constant political activity through which agency is acquired and identities formed. However, this is achieved via a vision of ‘identity as the always already settled criterion of proper intersubjective relations’.⁵⁷ An identity is asserted and assumed that grounds and guides its carrier's actions and, here I agree with Jens Bartelson, such theories of recognition ‘assume that the entities to be … recognized already are given, be they national communities or cultures’.⁵⁸ That is, in order to extend recognition to the multiple varieties of different states, communities, or cultures that demand and deserve it, they need to be distinguishable and ready to be recognized. To put it succinctly, they need to be recognizable. As Bartelson warns us, it is tempting, therefore, to argue that such subjects need to have been politically recognized – making them legible and intelligible – before they can be morally recognized as part of a larger diverse order. In Bartelson's words, it is ‘tempting to argue that moral recognition actually depends on prior political recognition of these entities, constituting them as bounded containers of distinct cultures’.⁵⁹

While theories of international political recognition view acts of mutual political recognition as constitutive – of, for example, the status of statehood, as we saw above in Honneth's reference to the ‘potential’ becoming ‘actual’ – they nonetheless also assume an actor that can enter into these games of recognition. That is, political recognition presumes the ‘existence of pre-constituted actors as a baseline for [its] explanatory endeavour’.⁶⁰ Or, as Oliver Kessler and Benjamin Herboth put it, to ‘frame politics in terms of recognition is to presuppose a world a priori divided into a multiplicity of distinct and separated collectivities’.⁶¹

The benchmark of misrecognition and the assumption of the subject

Recognition also can fail, according to the scholars that follow its dynamics in the global realm.⁶² The need for recognition is universal and so are the consequences of it failing.⁶³ However, despite their attempt to avoid essentializing the identities that were recognized, these scholars remained dependent on identity as a benchmark by which to judge the various available recognitive structures.⁶⁴ It is on the basis of identity that the failure of recognition is measured; misrecognition is only intelligible if recognition itself is a matter of the cognition of an identity that is in some sense independent of the uncertainties of human interaction. Because, if identities were not independent in this way ‘they could not serve as reliable benchmarks by which to judge the adequacy of particular recognitive act of structures’.⁶⁵ For Taylor, Honneth, Brincat, Reus-Smit, Wendt, and the large majority of political and international political scholarship, misrecognition can be read as the failure of recognition.⁶⁶ The conventional idea of misrecognition as a failure to adequately recognize others is further reflected in Axel Honneth's argument, considered above, about how recognition indicates the transition from a ‘potential’ to an ‘actual’ identity.⁶⁷ The pre-existence of a ‘potential’ identity in this case is necessary as a criterion for judging the adequacy of the act of recognition, which would, in turn, lead into an ‘actualized’ identity.⁶⁸ That is, in the interdependency of the development of a ‘Self’ vis-á-vis an ‘Other’, this constructive theory of recognition still depends on the existence of a ‘potential’ entity, identity, or subjectivity. Absent the potential subject, there are no grounds on which to know if recognition has succeeded or failed. Misrecognition assumes an existing identity as a benchmark for failed recognition. When Honneth argues that the ‘path for civilizing international relations primarily lies in sustained efforts at conveying respect and esteem for the collective identities of other countries’,⁶⁹ he is referring to the actualization of the potential collective identities of these other countries. Or, as he argued, against Patchen Markell, a few years earlier, ‘the struggle for recognition represents a struggle for the social articulation of pre-existent knowledge’.⁷⁰ The ‘pre-existing knowledge’ of a ‘potential’ subject carries the assumption of a pre-differentiated entity awaiting recognition, a subject that is harmed should recognition fail.

I have argued that attempts to analyze international dynamics through the lens of recognition assume a differentiated and recognizable subject. However, one might suggest that IR's constructive arguments of recognition – building on Hegel and Honneth's reading of him – do not assume a sovereign subject at all, to the contrary. However, as I will show below, my point is not to say that IR-scholars assume a sovereign subject but that they assume that sovereignty in knowledge about the subject is possible and necessary in order to assess whether or not recognition had succeeded or failed. They must assume a recognizable subject in order to proceed the analysis of the dynamics of recognition struggles. That does not mean that the grammar of recognition is stuck with fixed or reified subjectivities, but rather that it is dependent on an assumption that sovereignty in knowledge is possible. If this were not to be the case, again, there would be no way of knowing whether the suggested recognition of a known subject – substantive or enacted – would have been successful or not.

Patchen Markell and James Tully: non-sovereign knowledge and non-essential recognition

The pursuit of recognition in IR scholarship functions in a framework where knowable and differentiable subjects are the conditions of possibility for intersubjective relations. But through its aspiration and desire for epistemological sovereignty the politics of recognition misses the constitutive vulnerability of subjectivity.⁷¹ Hegelian-inspired scholarship on recognition in IR often assumes the possibility of knowing who we and others are; it assumes sovereignty in knowledge of identity, and the recognition of this knowledge by others.⁷² It assumes that this identity will guide and ground our actions and, moreover, that knowing the identity brings knowledge of the actions. Thus, if we were to know who an actor is, it would be possible to know what the actor would do. What this perspective elides is that identity does not simply ground action but is constituted by it, including the performative claims to recognition themselves. Further, since human action can never rid itself of its non-sovereign character, neither can the identity and agency that is formed by such action ever be fully sovereign. There is a limit to what we can expect our knowledge of ourselves and of others to do for us.

IR scholarship on recognition often remains blind to the possibility of the ‘non-sovereign’ character of human action and knowledge.⁷³ Since action will always be partly unpredictable the knowledge and subjectivities that emerge from it will remain non-sovereign. In this sense, the subject is not lacking sovereignty due to its dependency on an ‘Other’ but because it is performatively enacted and action will inevitably remain partly unpredictable.⁷⁴ Acknowledging the contingency and chance of human action means acknowledging the vulnerability and practical limits ‘imposed upon us by the openness and unpredictability of the future’.⁷⁵ Politics, Markell writes, ‘is in part a response to the experience of vulnerability’ to the fact that our identities are shaped in part through the unpredictable responses of other people. This is ‘what makes being recognized by others seem so acutely important in the first place’.⁷⁶ It is in reaction to the troubling unpredictability of the responses of others that the politics of recognition demands ‘that others recognize us as who we already really are’ be it an individual, a community, or a state.⁷⁷ It is because subjectivities are vulnerable and instable that the recognition of them appears critical. Recognition depends upon the impossible assumption of sovereignty in knowledge over that which already is.⁷⁸ The desire for sovereignty performs the function of making the ever-ongoing project of identity seem to be a stable and a steady ground for action.

I began this article by claiming that recognizing religion in global politics has costs. This follows from the fact that recognition has costs, both ontological – in the misrepresentation of the non-sovereign nature of human knowledge and subjectivity – and epistemological – by obscuring those subjects and forms of knowledge that fall outside of the currently recognizable, a point I will return to below. One might contend that international political scholars arguing for the importance of recognition do not assume this kind of sovereignty in a subject at all. In fact, a point might be made saying that the very grounds for political claims for recognition is that a subject is not (yet) recognized; if there were a recognized subject, there would be nothing to struggle for. As we saw above, recognition, in such arguments, is the force of change both for agents and epistemic systems.⁷⁹ My argument is, however, not about whether a new political order is realized as a result of subjects' struggles for recognition. My argument is that even absent its recognition, the subject is assumed to be there – the process of recognition assumes a subject that can be known and differentiated from that which it is not prior to the process of recognition. Such an approach begs the question: if claims or struggles for recognition are fueled by the fact that the subject is as yet unrealized, on behalf of what or whom are these claims made? How did they become recognizable in the first place?

To return to the empirical example introduced above of Jews in 18^th and 19^th century Europe; if recognition realized and constituted a European Jewish political subject – through the emancipation of Jewish minorities into citizens – how could a political subject be successfully recognized which did not yet exist in this manner?⁸⁰ The obvious answer to that question seems to be the performative enactment of Jewish subjectivities by individuals and groups claiming rights in the name of a Jewish community, the Honnethian ‘potential’ subject. Similarly, British Indian Muslims were vibrant communities prior to the allocation of rights and forms of representation by the British colonial regime. Does the argument of the costs of recognition break down once the subjectivities are not assumed as reified identities, but rather conceptualized as contingent and changing? In other words, what about non-essentialist recognition? In proceeding to the second part of my argument which leads us to the building blocks for thinking differently about recognition in international scholarship, I turn to Canadian political theorist James Tully. By identifying a shift in his work from what I conceptualize as an analysis of struggles for recognition to struggles over recognition I show how I see the conversation on recognition in IR moving forward.

From a struggle for recognition to a struggle over recognition

James Tully has increasingly featured in IR scholarship throughout the last decade, mainly in reference to his agonistic approach to social and political theory but also, prior to that, as a proponent for a ‘post-imperial cultural diversity’ defined by mutual recognition.⁸¹ I will not be able to treat Tully's work with the detail it deserves.⁸² What I will do, however, is to use a shift in Tully's scholarship to show how an approach to recognition which does not depend on reified identities but instead emphasizes non-essential subjectivities remain subject to the costs of recognition I outlined above. What Tully helps me to do, then, is to move from a framework focusing on the struggles for recognition – where non-essentialist recognition still assumes a differentiable subject tied to the existing categories of legibility where costs are entailed – to a framework focusing on the struggles over recognition, that is, the struggles in defining the conditions of possibility for recognition and its epistemic frameworks of legibility. First, two words on Tully's initial position.

Published in 1995 back-to-back with Charles Taylor's work on The Politics of Recognition, James Tully's book Strange Multiplicity takes a different approach than its fellow Canadian counterpart. Here, the politics of recognition is described as an ongoing process which values not the end-state of recognized subjects, but the shared political activity of subjects that are both the authors of and constituted by the engagement.⁸³ However, when reframing the question of recognition as dealing with performatively enacted subjects, as Tully does in his examples from North American indigenous peoples, he is still aiming at a more than just a form of successful recognition of subjectivities as they are.⁸⁴ While Tully remains sensitive to the risks of domination and effects of various power hierarchies on the course of negotiations, he also slips back into a cognitive understanding of recognition, that is, a position where the construction of the subject is performative but where that performativity is still limited to the available categories of intelligibility.⁸⁵ Non-essentialist recognition is not by definition free from costs, and Tully's earlier account echoes the dependency on an intelligible subject – even if it is ‘multiple, overlapping and contested’⁸⁶ – and the reproductive power entailed therein. If recognition is tied to the currently recognizable – reproducing the epistemic structures that enable it – what are the prospects for change?

In later work Tully enacts a useful shift in perspective from struggles for recognition, as in the case of peoples or nations, to the struggle over recognition. This implies a new focus on challenges to the norms underpinning recognition practices as well as the pathways to arriving at a place of legibility, of becoming recognizable. In his two-volume Public Philosophy in a New Key he picks up where he left off in Strange Multiplicity but highlights the contingent nature of the norms underpinning the processes, practices, and claims for recognition within and between nations and states. This contingency and the struggle to define the norms, subjectivities, and epistemic frameworks is, however, not a problem. Rather, the struggles over intersubjective norms of mutual recognition are necessary not only for the development of just social and political relations but also for the prevailing legitimacy of social and political order at large.⁸⁷ It is, for Tully, therefore not enough for an existing legislation, set of minority rights, or international covenants to recognize minorities or nations. Instead, he points out, these social, political, and legal institutions need to contain the conditions of possibility for change in the norms underpinning that legislation, the minority rights, or covenants.⁸⁸ If we take this agonistic idea of the centrality of the struggle over recognition with us, emphasizing the importance of the possibility for change – both in the norms, the subjectivities, and in the epistemic framework that holds them – how is that different from the approach to recognition in IR outlined above? For this, I turn to Jacques Rancière and Axel Honneth.

Axel Honneth, Jacques Rancière, and the possibility of epistemic change

Throughout the paper I have argued that the costs of recognition can be found, among other things, in the ontological misrepresentation of the non-sovereign nature of human knowledge and subjectivity. I have also gestured toward a second argument to which I will now turn. Here, I show that while recognition depends on a recognizable subject the framework of knowledge within which the subject has become recognizable is also due to be reproduced through the recognition of that recognizable subject. In other words, I am suggesting the contrary to IR scholars who see recognition as a motor of change and emancipation; that recognition has a reifying and potentially repressive element.⁸⁹

In order to forward this argument, I turn to a notable debate between Axel Honneth, whose reading of Hegel informs a large part, if not most, work on recognition in IR, and the French philosopher Jacques Rancière. The book Recognition or Disagreement was published in 2016 and captured a conversation between Honneth and Rancière at the Institute for Social Research in Frankfurt am Main a few years earlier. In Honneth's description of the struggle for recognition, he presents previously excluded subjects as challenging the prevailing social and political order, including the knowledge system that underpins it. Such challenges undermine and change the epistemic range of whom can be included in a particular social and political order, broaden its range of intelligibility – that is, they extend the range of that or who is available for recognition.⁹⁰ The inclusion of a new type of nation, for example, changes the meaning of nationhood itself. Rancière's critique holds that the inclusion of new subjects will not change the range of the intelligible in any way, since it is only possible to include into the new order those who are recognizable as subjects to begin with.⁹¹ The established framework of knowledge, what Rancière calls the ‘common sense’, will not change through the inclusion of that or those who were previously excluded from it. A change in the framework of knowledge, Rancière argues, does not come through the acknowledgment of the previously excluded but, rather, through the enactment of a different reality altogether.⁹²

Epistemological change – what Rancière calls a change in the ‘distribution of the sensible’ – stands in stark contrast to theories of the transformative power of recognition, which take place within a certain normative and cognitive framework. It is epistemological change that holds the possibility for transforming the workings of such normative and cognitive frameworks,⁹³ going ‘beyond given systems’ and proposing a ‘new system of norms’.⁹⁴ A change to the distribution of the sensible is not a broadening to include that or those who were previously excluded, nor is it the self-realization of a group becoming ‘aware’ of itself, finding its voice or imposing its weight on society.⁹⁵ It is not the acknowledgment of the grievances or suffering of those on the wrong side of a hierarchical order that will change the range of the recognizable; the wrong by which politics occurs is not ‘some flaw calling for reparation’.⁹⁶ A change in the foundations of knowledge comes, rather, through the interruption of the common sense by those and that which were invisible to it. This is what Rancière calls the introduction of an ‘incommensurable at the heart of the distribution of speaking bodies’.⁹⁷

The range of what or who can count as a subject does not change through the inclusion of those who are considered rightly worthy of a place, but through the enactment of a different reality all together – acting as if one were a subject.⁹⁸ Such an action is an intervention into the conditions of knowledge of what a subject is and who is eligible to become one. A redistribution of the sensible means a change in the framework of the intelligible – it makes ‘what was unseen visible’.⁹⁹ For Rancière, the struggle for recognition in Honneth's terms aims to change the range of the recognizable by convincing or pressuring the arbiters of the boundaries of that knowledge to expand them. By contrast, a change in recognizability, in Rancière's understanding, is about claimed space regardless of the arbiters or boundaries – it is the change of terms altogether.¹⁰⁰

As political theorist Hanna Pitkin noted in her analysis of Niccolò Machiavelli, in order to communicate an alternative, one ‘wants not to convey new information to (one's audience), but rather to change the terms, the conceptual framework through which they presently organize their information’.¹⁰¹ Referring this back to the discussion of recognition the question is not ‘how does a subject become recognized?’, but ‘how does it become recognizable?’. In the previous case of the Jewish or Muslim subjects referenced below, how did those subjects become intelligible as distinct from their surroundings, as something available for recognition? How was this subjectivity stabilized and made coherent? What were the consequences as to what or who could be included in the categories of ‘Muslim’ or ‘Jewish’ as those categories became linked to the minority, the nation, and the state?

Through Rancière's critique we see that Honneth's struggle for and achievement of recognition is trapped within its own referential bubble. Requiring the assumption of a subject or, in Hegelian terms, a consciousness for the relational process of recognition between a Self and an Other, Honneth is unable to account for forms of subjectivity other than those that are expected, those that are intelligible as subjects from the start. Recognition might change the number of players in the recognition game, but it will not be able to change the game itself. Rather, recognition will reproduce the common sense, the established social and political order.

Turning back to the question of religion and the recognition of religious difference, a process that follow's Honneth's logic will extend acknowledgment and empowerment to that or those who are already recognizable as religious, thereby confirming powerful actors' claims to represent religion, religious groups, and religious truths. Such a process marginalizes those who are outside of the spectrum of the recognizably religious, writing out of the picture an entire range of possibilities while those who remain unrecognized struggle to achieve a place on the public register of recognition. The requirement of an already-recognizable subject strengthens the voices of conventional and established religion, and further hardens the boundary between those included in that category and those who are not.¹⁰² It is, therefore, not merely that the recognitive structures of social and political orders assume and reproduce a knowable and differentiated subject, they are also risk reifying the structures and hierarchies of a given prevailing order.

If it is true then, as Axel Honneth put it, that ‘the struggle for recognition represents a struggle for the social articulation of pre-existent knowledge’ – that is, an articulation of that which is, already, known – the question emerges as to how that knowledge comes about? Returning to the example of the Jewish and Muslim subjects, how does one know what or who is Jewish or Muslim, and who is not? What are the processes that make the contours of a subject identifiable and how is it differentiated from other subjects? How does it change regarding that which is possible to recognize, and what are the social and historical forces affecting that change? What are the consequences? And who pays – who carries the costs? By examining in detail the conditions of possibility of recognition we gain a better grasp of the costs that come with it. In the next section I illuminate the conditions of possibility for the recognition of ‘religion’, and the recognition of ‘Muslim’ during the establishment of the independent state of Pakistan. The line between religion and politics is, as Hussein Agrama points out, ‘historically connected with modern state sovereignty and its constitutive indeterminacies’.¹⁰³ Studying the birth of a state claimed and recognized, in part, along the lines of ‘religious difference’ provides us with important resources for understanding the conditions and the costs of that constitutive relationship between religion, state sovereignty, and the system in which they are embedded.

Recognizing religion and postcolonial independence: India and Pakistan

In this final section I illustrate how ‘religion’ and a ‘Muslim’ nation came to form part of a transnational vocabulary describing minorities, borders, and conflicts see also: Devji 2013; Dubnov 2019; Kattan 2008. As I have shown elsewhere, the British, Indian, and Pakistani state-building projects were grounded in a shared assumption that a political and epistemic mastery of ‘religion’ was essential to the working of political and legal governance.¹⁰⁴ My approach to religion here builds on and develops recent scholarship in political theory, anthropology, and sociology.¹⁰⁵ Religion is considered a ‘moving target’ shaped by, while simultaneously informing, broader social formations, nationalist agendas, and political structures.¹⁰⁶ Here, I study the road to recognizability of a Muslim subject and ask: how did the conceptual apparatus of religion develop to describe minorities, borders, and conflicts? And how was that apparatus put to use in negotiating claims for independence and various visions of statehood? How did it structure the analysis of conflicts and expectations about how those conflicts would develop? And how did it feature in struggles about authority, legitimacy, and political order? What were the consequences and costs of employing the vocabulary of religion and how were they distributed?

In August of 1947, India and Pakistan became sovereign states, independent from both the British Empire and from each other. The partition of South Asia served as a resource for Middle Eastern state builders not only with respect to seeing partition as a solution to the problems of increasingly intense conflicts, the administration of refugees and absentee property, but also in terms of the terminology used to make sense of it.¹⁰⁷ An example is ‘religion’ and a ‘Muslim subject’. The partition of Bengal in 1905 set the precedent. In that case, the colonial government had aimed to splinter and weaken political resistance to British power by dividing the province into an eastern part, where Muslims formed a majority, and a western part, where non-Bengali Hindus dominated numerically. As Penny Sinanoglou has argued, from the vantage point of later events in Palestine, the ‘partition of Bengal is significant because it manifested the concept of dividing territory and political representation along religious lines’.¹⁰⁸

Although the partition of Bengal was reversed in 1911 in response to large-scale protests, the principle of representation by religion had become an integral part of Indian politics and law. The efforts to govern Bengal also further laid the groundwork for the British understanding of Muslim governance and the ‘manner in which Islam might be handled by the British Empire’ elsewhere.¹⁰⁹ The party of the All-India Muslim League was founded a year after the Bengali partition, while in 1909 the Indian Councils Act (also known as the Morley–Minto reforms) created exclusively Muslim electorates in provincial and central legislative councils. The provisions, defined as protections for religious minorities, were developed and embedded in the Indian political system through both the Montagu–Chelmsford reforms of 1919 and the new constitutional framework developed in the 1935 Government of India Act.¹¹⁰ The rationale behind these ‘marked’ electorates was to increase Muslim representation in the system of elective local government.¹¹¹ At the same time, however, they also established a link between religion and political representation, power, and patronage.

Connecting political power to one's recognizability as a religious minority cuts both ways; as Iza Hussin notes, ‘[a]t the level of legal and political discourse, the ability to make claims on behalf of a Muslim interest required an acceptance of the space within which Islam had been assigned, thereby often undermining the logic and power of the claim itself’.¹¹² In The Politics of Islamic Law, Hussin shows how the realm of Islamic law became intelligible as a separate category of law in Muslim societies under British imperial rule. Colonial encounters both subjugated Islamic law – detaching it from public politics and restricting it to the private realm of family law – but also, through its reification, codification, and textualization, gave Muslims (in courts and in larger society) concrete and legible references for litigation, appeal, and reform. This made Islamic law a powerful tool in articulating alternatives to and challenges to the authority of the colonial state.¹¹³ By compartmentalizing religion into various domains, and ceding those domains to the local elite, ‘(t)he colonial state in fact helped constitute the private domain that Indian nationalists later cherished as free from colonial interference’.¹¹⁴

The processes of minoritization such as giving a ‘Muslim minority’ exclusive electorates brought both access to power and subjugation to it. It granted access to power by making the members of a population recognizable as political actors. Yet, the condition of that access to power was subjugation or subjection to the particular form of knowledge that grants access – in this example becoming legible to the apparatus of the state. Returning to the terminology of the discussion above, recognition, in this sense, had a double face, making the conditions of empowerment simultaneously the conditions of control.

My point in drawing a parallel between the recognition of Islamic law and minority recognition is to draw attention to the two faces of recognition. Recognition brings with it access to power but also brings costs. The British colonial state separated Islamic law, and the Muslim subjectivities connected to it, from public political power. At the same time, the unification, codification, and reification of Islamic law opened up space for anti-colonial actors to draw on those highly specified codes and rules in order to articulate an alternative to British rule. Similarly, the recognition of British Indian Muslims as a minority created a unified actor with political access. At the same time, it limited that access to those who were recognizably Muslim in the eyes of the colonial state and the technologies of categorization it used to visualize its subjects (such as the census or the law).¹¹⁵ The form of power expressed through the regimes of recognition thus both established the subject and governed it at the same time.¹¹⁶

The same is true of the conditions of resistance to colonial rule. In order to be recognizable to the state as opposing a certain policy on Islamic law or identity, it was necessary to identify as an ‘Indian Muslim’ – an act which ‘brought [the category] into being as a political and public reality’.¹¹⁷ That being the case, alternatives to the colonial state still depended on them remaining legible to it; those who resisted the state still had to align themselves along the unified categories of race and religion in order to be recognizable to and recognized by the state. Even the act of resistance against local and colonial elites helped, therefore, to reinforce the idea of a unified ‘Muslim’ subject.¹¹⁸ At the same time, however, the second role played by the bounded category of religion was the creation of this unified Muslim subject, whose codification in law and representation in politics would become a node in the anti-colonial resistance and claims to independence that would follow. The imperial recognition of the Indian Muslim was, then, Janus-faced, working both as a condition for government and power and as a resource for challenging and opposing them.

Conclusion

The last decade has seen critical accounts add to the debate about recognition in global politics.¹¹⁹ Jens Bartelson has argued that ‘international relations are mute when it comes to the conditions of epistemic recognition, implying that the classificatory grounds for recognizing an entity as an actor of a certain kind already are given or at least sufficiently unproblematic’.¹²⁰ Janis Grzybowski has added that recognition is reproductive of the ontology of the state and the state system, and ignores ‘the prior question of what states actually are and how they can be identified as such’.¹²¹ Ayse Zarakol, meanwhile, has joined Charlotte Epstein, Ole Jacob Sending, and Thomas Lindemann in reversing the recognition ontology to one of misrecognition, where global political subjects strive for, but can never achieve, sovereign agency.¹²² Such perspectives hint at the futility of the strive and struggle for recognition in global politics, suggesting that its effect is perpetuating order and hierarchies, rather than challenging them.

In this paper, I have gone beyond the critique of recognition to argue for a study of its costs. These are not the result of ‘recognition gone wrong’, or its ‘instrumentalization’ in more intricate power games but can only be understood through an examination of recognition's conditions of possibility. I have argued that recognition depends on and reproduces a knowable and differentiable subject, without which it would be impossible to evaluate the success or failure of recognition in the first place. By understanding how this subject became recognizable, it becomes possible to examine the costs bound up with it. That means studying the vulnerability and governability bound up with recognition – making visible the forgetting of certain histories, the abandonment of certain practices, the silencing of certain voices, and the privileging of certain forms or regimes of difference over others. Regarding the question of religion, it is by studying the path to recognizability, that is, the global epistemological politics of religion that we can begin to understand the costs of its recognition.

Over the last 20 years an impressive body scholarship has taken shape in the fields of anthropology, history, sociology, law, and political theory, tracking these productive processes and their implications in detail. Winnifred Sullivan, Nadia Marzouki, Iza Hussin, Benjamin Schonthal, Noah Salomon, Saba Mahmood, Elizabeth Shakman Hurd, Ben Crosson, Amélie Barras, Joseph Blankholm, Justin Richland, Meadhbh McIvor, Brannon Ingram, Teren Sevea, and many others have analyzed the dynamics of the co-constitutive relationship between modern national and international politics and religion illustrating in detail the ways in which the productive and epistemic power of political and legal orders shape modern political, religious, and legal subjects.¹²³ This scholarship on religion and politics shares the critical theoretical, conceptual, and methodological groundwork of scholarship in International Political Sociology, Historical IR, Postcolonial Global History and Decolonial Theory, and any variety of Feminist approaches which have remained reluctant to engage with scholarship on religion in IR since, as we saw above, a majority of this scholarship remains locked into the epistemological trap of rehabilitating religion and its accompanying hierarchies. IR does not need to recognize religion. Rather, international scholarship needs to engage with the costs of recognition by studying the histories and global epistemological politics of religion that produces recognizable religious and political subjects.