Religion As an Analytical-Political Category

Religion

Religion as an analytical-political category

“By itself, the utility of theory is invisible and rather irrelevant. It is in the effects of its interaction with data [...] that we see, and that our understanding is augmented. Without theory, data is just unorganized data; without data, theory is empty. But when the two come together, both are changed, and both are better understood” (Arnal 2001). I fully endorse this opinion expressed by William Arnal according to which theory and empirical data become really meaningful (and interesting) only if they throw light on each other. Relating his view to the academic theorisation of religion more specifically, Arnal contends that because “Religion” as a singular and substantial entity does not exist at the level of secular academic analysis and because for that reason the concept of religion is essentially groundless, “there cannot be a theoretical accounting for Religion, but only for discrete religious phenomena” (Arnal 2001). Accordingly, since trying to elaborate something like a “Grand Theory of Religion” is probably misconceived, “theory in religious studies [should not] refer to ‘religion’ in the abstract but rather should and must refer to the concrete human phenomena that make up the contents of this strange category” (Arnal 2001). Behind this judicious position lies the fundamental and by now widely accepted argument that religion is not a sui generis, universal anthropological phenomenon but a discursive category that has a more specifically “Western”, Christianandsecular genealogy – which notably implies that the meaning of the term “religion” has been continuously shifting in the course of its history. If there may be, at least among scholars of religious studies, acertain consensus about this point, the question of what to do with the notion of religion in “scientific” works once one has accepted that religion is not a natural domain remains debated (references): is it pertinent to use this ideologically charged term as an analytical and theoretical concept? Can it be applied to identify some phenomena in non-Western contexts? Should one not rather abandon it? Etc. These questions have generatedunending debates but are important and must be tackled because naming particular human phenomena and traditions “religion” (or not) can have significantanalytical consequences. Following Arnal’s recommendation, what my aim is not to theorise about an abstract “Religion”,because this does not exist as such, but to make my empirical dataintelligible.It is thus for the concerns of the present work thatI have tried to develop a theoretical approach whichis grounded in theobservation that religion is a discursive category which is defined and deployed differently within different socio-historical contexts and is embodied in different historical forms of life[1].

I think that a rewardingway to make sense of a particular socio-historical context begins by investigating how the peopleevolving in this context themselves try to make sense of it, that is to say by using language. This assumption reflects nothing else but the common view according to which language plays a central role in the social construction of reality. This notablyimplies that I analysediscourse as a social practice (Fairclough …). Accordingly,a theoretical perspective that can be productive in the context of the present study is not so much to conceptualise and use “religion” as a category of analysis, which always tends to be problematic, but rather to take the category of religion as an object of analysis. In other words, the approach that I adopt here is a social constructionist perspective that regards the category of religion as a popular notion which is defined and deployed or, to say this differently, put into use within historically and socially situated “emic” discourses. More concretely, what I seek to do is to investigate how the notion of religionhas come to be invested with particular meanings in the context of postcolonial India and serves, along with interrelated categories like spirituality, secularism, culture, politics, democracy, majority, Hinduism, Islam, terrorism and so on, as a (central) analytical-political category used to articulate different emic discourses in this context. This means by extension that the questions which I seek to answer to are not “What is (the function of) religion?” or “How to define religion?”, but rather “What social actor defines religion?”, “What specific nexus of powers and knowledges have enabled a social actor to define religion in a certain way?”, “How is (a particular) religion’s alleged ‘normal’ or ‘real’ domain defined in relation to and in contrastwith other domains like politics, secularism, culture, etc.?”, “Who and what is included in and excluded from a particular definition of religion?” and “What are the effects on social organisation of the authorisation and deployment of a particular normative definition of religion?”.

What is needed is thus not one more attempt to elaborate a better abstract, universal definition of religion. Whilenowadays the majority of scholars of religion may not seek to define religion’s supposed essence or functionany more, it has become relatively common (Saler 2008, ...... ), it seems to me, to define religion in order to have a heuristic concept that should be flexible enough to demarcate and describe the sets of interrelated phenomena that may be identified as religionin different socio-historical contexts, and this while avoiding “Christianising”, essentialising and homogenising the identities, knowledges, discourses and practices regarded as religious by the analyst[2]. Trying to overcome the dilemma posed by the largely Eurocentric, Christian genealogy of the category of religion, the ahistorical essentialism and reductionism of monotheticsubstantivist definitions of religion as well as the largelysimilar flaws inherent to functionalist theories of religion, such effortsto elaborate non-essentialist, unbounded (polythetic) conceptualisations and definitions of religion remain nevertheless problematic despite the laudable intention to get rid of strongly essentialist understandings ofreligious traditions and phenomena: indeed, a residual essentialism can be detected in anti-essentialist definitions in the sense that while they explicitly reject the existence of Religion as a transhistorical and transcultural substantive entity, these conceptualisations nevertheless assume that what is religious and what is not is more or lessself-evident and thus that the elaboration ofan abstract, prototypical model of religion prior to the analysis is possible (Abeysekara 2001). In short, such definitions do not lead far beyond essentialist conceptions of religion defined in substantive terms.

One can of course contend (this is the argument underlying the family resemblance approach to religion for instance) that the various sets of human phenomena observable in different historical contexts that one would typically associate with religion obviouslyresemble one another, which would justify their categorisation as religious. This is not such anintractable problem as long as one is aware that the various phenomena gathered together under the term “religion” do not share a common essence in the empirical world butthat one’s own definition of religion is nothing but a revisable heuristic deviceelaborated and used by the analyst to demarcate the phenomena which may be classified as religious for the purposes of the research. I can to a large extent accept this argument. One has almost no choice, if one hopes to be understood, but to describe, that is to say, to identify, classify and namethe data one “encounters” with the existing words of the language in which the analysis is done, and these words are either used in their prevailing folk meaning or elaborated into analytical categories from this basis. And, in fact, most of the phenomena one has to nameare named with a particular word by reason of their apparent resemblance to other phenomena that have already been named with this very word.

Even though the term “religion” as such was or is not used in non-modern and non-Western contexts in its dominant modern understanding, it remains that in these contexts there arediscursive categories and human phenomena which are definitively, if not identical, at least to some extent similar and comparable to those that are commonly identified as religious in the modern West (similarity does not mean orimply identity). Therefore, from this perspective, I do not think that naming“religion”some traditions in contexts where it is not a native term, if it is certainly problematic several respects (notably in terms of translation)[3], is necessarily a cardinal sinbecause it wouldinexorablydistort the data analysed beyond recognition and that one has to come to “the bigoted conclusion that non-modern and/or non-Western data fall outside [the] proper purview” (Arnal 2001) of scholars of religion. All the more so since arguing, as some have done for instance(Balagangadhara ..., ……), that there was nonative religion in precolonialIndia, because the “nature” of Indian traditions cannot be captured byconventional definitions of religionfor the reason that these ones are based on a particular conception of Christianityis, the pertinence of thislatter point notwithstanding,no much lessproblematic than arguing that Indian traditions are indisputablyreligions, for both positions assume that the identity of religionisself-evident. To say this differently, both positions similarlyseem toassumethat there is something like “real religion”, a kind of prototypical model of religion against which one could judge whether actual historical traditions are religions or not. Thismodel is ironicallythe same one, that which has been conceivedmore particularly, to simplify, by modern Western Christian theologians when speaking about Christianity[4]. However, as far as I know,apart from the difficulty to think “beyond” one’s own dominant cultural categories, nothingforces scholars of religion, as social scientists or historians, to accept that such a(normative)theological definitionis a valid “scientific”conceptualisation of religion as a set of actual historical phenomena (and whythis definitionrather than another?).Nothing forces them to accept that Christianity exhausts what religion is or can be and to ground their approaches to actual historical“religions”ina theological definition[5]. Despite a consequent number of arguments that are undeniably pertinent (to put it very simplistically, it is clear that Hinduism isnot Christianity: a Brahman is not just the Hindu equivalent of a Christian priest, etc.), I worry that the assumption that Hinduism is categorically not a religiontends to strengthen a culturalist understanding of humanity according to which this one would be divided into variousincommensurable cultures (orcivilisations)which would have historically evolved, before European colonialism, within self-enclosedgeographical and intellectualuniverses. Such a view neglectstheconstant interactions between these “cultures”, interactions that have contributed to make them what they are as much as “internal” developments, and this well before the imperialist expansion of Europe and of the globalisation of the narrative of modernity and its grammar of concepts[6]. I think one has no choice but to work with this tension: on the one hand the acknowledgment that words (even universalising ones) are defined differentlyin different contexts and that the meaning of these words is contested and historically altering and on the other hand the inevitability to use these samewordsto translate other conceptual universes(because one needs words and concept to think and speak!).

But what I am actually implyingis that if one has accepted, rightly in my opinion, that religion is not an anthropological universal and that it is fundamentally misconceived to search either for its essence or for its psychological or social function(s), I simply do not quite know on what basis and criteria (I mean a basis more solid and convincing than the commonsensicalidea that religion is simply “out there”) one could build up a definition of religion as a theoretical concept that would be analytically valuable and what could be the actual utility to come up with such a concept, that is, with one more universal definition of religion defined in the abstract. What I mean more concretely is that assuming for instance that Hinduism can be considered a religion and studied “like or as a religion”[7] (whatever the domain covered by the term has been abstractly delineated) because its supposedcomponents bear some resemblance to those of other traditions which are usually considered to be religions is a “finding” that has, it seems to me, hardly any interest either as a starting point andeven less as a conclusion: once it would have been determined that Hinduism may be deemed a religion because in Hinduism there are, say, some interrelated all-encompassing conceptions of the world, moral codes, ritualistic practices, beliefs in superhuman beings or whatever else, one may legitimately ask: “Ok... and what then?”.In short, I do not really see what might be the analytical and explanatory value of such universal conceptualisations of religion, which are, in this case, hardly more than selective, ahistoricalenumerations of diverse elements, for providing insights about the historical formation or the sociological mechanisms of discretereligious phenomena, which is what scholars of religionhave to explainin the first place. More importantly, when one seeks to elaborate and use an abstract, universal definition of religion as a category of analysis, even if this one is relatively flexible, one tends to miss the most crucial point aboutreligion, as TalalAsad has pointed out: “religion is a social and historical fact, which has legal dimensions, domestic and political dimensions, economic dimensions, and so on” (2002). These dimensions of religion as a fact are an integral part of it and cannot be neglected.

The critical argument underlying the assumption that religion cannot be taken as having an essence at the level of secular scholarly analysis andthat there cannot be a universal conception of religion, and therefore that seeking to define religion in the abstract is pointless (and can always be contested) is not, as Asad has argued, “because religious phenomena are infinitely varied. Nor is it the case that there is nosuch thing as religion” (2010: …). It is not “because [religion’s] constituent elements and relationships are historically specific, but because that definition is itself the historical product of discursive processes” (1993: 29). What Asad means here is that “defining [religion] is a historical act” (2010: …), that is to say, an effort to identify and demarcate a set of apparently interconnected phenomena which is always generated by and responds to specific analytical and political demands which are themselves formulated within specific orders of discourse situated in particular socio-historical conjunctures. The argument here is actually perhaps just as evident as it is fundamental: the category of religion, if and when it is part of the articulation of a historical discoursealong with other terms, iscontinuously redefined and put into use by social actors to try to make sense of life (knowledge) and to fashion it (power) – and this analytical-political act of identifying (what constitutes or should constitute) religion is what should be “deconstructed”, the aim being to understand the function of this floating signifier in the articulation of the historically altering discursive constructions of the society or tradition studied. To put Asad’s point in the simplest way possible, “[p]eopleuse particular conceptions of religion in social life” (Asad 2002), and the emic conceptions and ideological usages of the category of religion within a broader conceptual network should bea major matter of analysis.In my opinion, the task of the scholar of religion is not so much to define the category of religion (which is that of the theologian, the philosopher, the politician, the secular state, or any other “ordinary” person for that matter), but rather to analyse how religion is defined and put to work by social actors themselves.This means that the analyst has first of all to analyse not only the discursive construction of religion within a particular historical order of discourse and the analytical-political usage of this conceptualisation, but also the concrete effects, if any, of this usage in producing and sustaining particular forms of life. What can be productive is thus to start by undertaking a genealogical study of the altering and competing emic definitions of religion in a particular socio-historical context. (By “definition of religion”, I stress at this point that I do not necessarily designate a coherent or theorised definition of religionà la Clifford Geertz, but rather the continuously contested and shifting, potentially confused, knowledge about which persons, acts, practices, behaviours, institutions, things, places and so on are identified as and assumed to be religious by different social actors[8].) This requires that one begins

by asking what are the historical conditions (movements, classes, institutions, ideologies) necessary for the existence of particular religious practices and discourses. In other words, let us ask: how does power create religion? To ask this question is to seek an answer in terms of the social disciplines and social forces which come together at particular historical moments, to make particular religious discourses, practices and spaces possible. What requires systematic investigation therefore are the ways in which, in each society, social disciplines produce and authorise knowledges, the ways in which selves are required to respond to those knowledges, the ways in which knowledges are accumulated and distributed. Universal definitions of religion hinder such investigations because and to the extent that they aim at identifying essences when we should be trying to explore concrete sets of historical relations and processes. (Asad 1983: 252)

To sum up my argument, religion is a folk category whose definition is continuously produced, authorised, used and reproduced as well as contested and altered by various social actors for specific analytical and political reasons and purposes that partly structurethe socio-historical context which the analyst aims at making sense of. This notably implies that when a particular definition or conception of religion is authorised and deployed there will surely be effects on the concrete arrangement of society, on the waya normative conception of religion is concretely lived by its practitioners[9], and so on. Therefore, what should be investigated are the discursive processes through which the category of religion has come to be defined and used by social actors in a certain way in a particular context (how political and theological debates and social forces enable the production and authorisation of specific conceptions of religion) and the effects resulting from the authorisation and deployment of a normative knowledge about what constitutes “authentic” religion in producing and reproducing embodied forms of religious life (identities, practices, discourses, knowledges, and so on).