The Post-Colonial Method versus the Sociological Method in the Study of Religion: Top-down or Bottom-up?

In the current study of “religion,” two different methods have developed that are often in tension with each other.  The first is the Post-Colonial Method (PCM)[1] with scholars like Talal Asad, Edward Said, Timothy Fitzgerald, and Gayatri Spivak, and the second is the Sociological Method (SM) with scholars like Steve Bruce, Phil Zuckerman, and Barry A. Kosmin.  Each of these methods has their own histories, assumptions, and trajectories for the study of religion.  The tensions between the two arise about issues of defining or employing the term “religion,” the problems with religious or non-religious self-identification, and the usefulness or irrelevance of the social scientific study of religion.  This article is going to address some of these tensions while examining specific issues in these two methods.  I will argue that there is a balance to be struck between a top-down (PCM) and a bottom-up (SM) approach to the study of religion.[2]  I will primarily be using Talal Asad and Steve Bruce as my interlocutors for discussing issues surrounding “religion” in these two methods.

Let me begin by briefly explaining what I mean by the PCM being “top-down” and the SM being “bottom-up.”  What I mean by this is that the PCM tends to focus on grand historical and political narratives, rightly challenging concepts and power-relations that flow from the top-down (i.e. colonialism).  Religion, defined in universal and essentialist ways, is just such a concept.  This does not mean, however, that the PCM is not concerned with people on the “ground level.”  The subalterns, the “others,” and the oppressed people of the world are taken very seriously.  I do see this method as distinct from the SM.  The SM, in my opinion, is not as much an enterprise of examining historical and political narratives, as it is a focus on present-day people who identify in certain ways.  The sociology of religion, for instance, deals with statistics and demographics whose locus is people reporting on their own lives.  In this sense, the SM is bottom-up.  One could argue that the SM is simply imposing Euro-centric notions onto large western populations, and as such, is actually “top-down.”  Although I find some of these arguments powerful, I nonetheless see these two methods as primarily leaning towards one trajectory over another.

It is clear from a cursory reading of Asad that his focus, which resembles many post-colonial sentiments, is historical, political, and anthropological.  As the title of his book Genealogy of Religion points out, Asad wishes to trace the genealogy of the term “religion.”  Bruce, on the other hand, primarily uses the social sciences, statistics, and demographics to argue his points.  What are some of the problems with these approaches to religion?  Does the post-colonial method miss the fact that people self-identify as religious or as being part of a religion, or the fact that we need to go to actual people, not cultural concepts and political history, to discover what “religion” amounts to?  Or does the sociological method often ignore the biased history and politically charged terminology that is commonplace in the social sciences?  Does the sociology of religion rely upon antiquated, colonial, Euro-centric, Christian notions of “religion?”

In critiquing Clifford Geertz’ famous definition of religion, Asad sums up one of his main points in Genealogy of Religion: “My argument is that there cannot be a universal definition of religion, not only because its constituent elements and relationships are historically specific, but because that definition is itself the historical product of discursive processes.”[3]  These “discursive processes” are products of European and Enlightenment thought, are often centered around some form of Christianity, and often juxtapose the rational, doxastic, and cognitive elements of “true religion” with the irrationality, practical, and non-cognitive nature of primitive idolatry.  According to Asad, “The demand that received practices must affirm something about the fundamental nature of reality, that it should therefore always be possible to state meanings for them which are not plain nonsense, is the first condition for determining whether they belong to ‘religion.’”[4]  William E. Connolly, while discussing Asad’s critique of Wilfred Cantrell Smith, says something similar: “many secularists, ministers, theologians, anthropologists, philosophers, and social scientists place such practices [religious devotional practices] within a cognitive framework that either ignores the embedded character of embodied faith, diminishes its importance, or reduces it to modes of cultural manipulation to be transcended by cognitively pure belief.”[5]

Asad’s trajectory is to dismantle any notion that “religion,” as a specific and distinguishable category and field of study, is simply a “given” handed down from on high.  Along with Tomoko Masuzawa,[6] Asad sees the Euro-hegemonic origins and applications of the term “religion” and problematizes what we come to think of as “clear-cut” differences between politics, ideology, and religion.  Asad also points out that our commonplace definitions of religion, handed down from the early sociologists (Emile Durkheim, Max Weber, Herbert Spencer, and August Comte), are a product of certain Christian influences.  As Asad says,

“Thus, what appears to anthropologists today to be self-evident, namely that religion is essentially a matter of symbolic meaning linked to ideas of general order…is in fact a view that has a specific Christian history.  From being a concrete set of practical rules attached to specific processes of power and knowledge, religion has come to be abstracted and universalized.”[7]

A crucial point Asad is making here is the fact that what people in the West have come to understand as “religion,” was birthed from the womb of European Christianity.  From the inception of this cognitively focused idea of religion, European scholars “universalized” religion and created a taxonomy to which certain traditions had to comport.

The SM, as is seen in the work of Steve Bruce, Phil Zuckerman, and Barry A. Kosmin, has somewhat of a different trajectory.  The goal of these scholars is to discover the statistical and demographical self-identification of religious and non-religious persons.  “Religion,” as something different from “non-religion” is taken for granted in the SM.  This is because certain people identify as being part of a “religion” or as participating more or less in “religious” activities.  The SM primarily focuses on society—such as, social groups and patterns, the social construction of reality, limitations placed upon individuals by their location and family, and deviance and non-conformity—and how “religion” plays a role in it.

Steve Bruce has taken some time to argue not only that secularization has occurred in the West,[8] but also that the terms “religion” and “secular” are not arbitrary titles.[9]  For instance, Bruce says, “I see no great difficulty in defining religion as ‘beliefs, actions, and institutions that assume the existence of supernatural entities with powers of action, or impersonal powers or processes possessed of moral purpose.’”[10]  In response to criticisms like Edward Said’s, Bruce says,  “I see nothing in this definition that confines it to the West or to theistic salvation religions.”[11]  Bruce has no qualms with this substantive definition of religion and its universal applicability.  Bruce thinks that there is “nothing particularly wrong with a very broad conceptualization of religion provided it is accompanied by a well-informed sense of the variety it can take.”[12]  He also thinks that, “A detailed universally applicable definition of religion is clearly necessary if one aims to produce a universally applicable theory of the origins and purposes of religion.”[13]

Bruce has plenty to say in response to Asad’s PCM as well.  “The most common criticism of substantive definitions of religion,” Bruce says, “is a watered-down version of Asad’s assertion about the historical specificity of such definitions: there is just too much variety.”[14]  Bruce responds to some points in Asad’s Genealogy of Religion rather explicitly:

“First, the origins and development of a concept have no necessary bearing on the reality it purports to comprehend because discovery is not the same as invention.  Newton’s discovery of gravity was a ‘historical product of discursive processes’ but prior to its discovery people did not have trouble adhering to the earth’s surface.  Second, the uses to which some idea is put does not exhaust the idea and even demonstrating that an idea has been used for bad ends does not of itself demonstrate that the idea is badly conceived and should be disregarded.  Third, the post-modern critique of the idea of religion applies equally well to every other concept and definition we ever use, including those used in the construction of genealogies and in discussing alternatives.”[15]  

Bruce’s distinction between “discovery” and “invention” is key to the tensions between the PCM and the SM.  The PCM prefers the latter, while the SM prefers the former.  Bruce’s critique of Asad involves the idea that even before “religion” could be used by the powers-that-be as a conceptual tool, people had religions and acted religiously (“people did not have trouble adhering to the earth’s surface”), just as before the modern idea of family was defined, humans behaved in filial ways.

Bruce also worries that Asad and Fitzgerald conflate a concept in-itself with a concept as-it-is-used.  Religion may very well have been used by European powers to justify all kinds of heinous acts, but that is not the fault of a definition that needs to be jettisoned, rather, it is the fault of those who utilized the term as a justification for committing evil acts.  This does not, however, mean that some definitions of religion are not biased; it simply means that sociology of religion should aim for a definition that is as least biased as possible.  Along the same lines, Bruce thinks the post-modern critique of “names,” “definitions,” and “labels” can be applied equally to those who employ the PCM.  Flawed definitions, Bruce thinks, can be handled with the tools available to the social sciences:  “What is often not appreciated by the theorists who dismiss empirical social research on the grounds that it operates with flawed definitions is that we possess a range of tools for assessing the relative validity of different conceptualizations of our subject matter.”[16]

In light of these two methods for studying religion, I am going to provide what I consider to be three positive and three negative aspects to these methods.  The hope is that both can learn from each other, try to find some common ground, and aid in our understanding of what “religion” truly amounts to.  Although I am more inclined to defend the SM over the PCM, I will nonetheless try my best to learn from those I sometimes disagree with.

The Post-Colonial Method (PCM):

(A) Three Positives

1. Challenges the status quo of the social sciences and religious studies.
2. Stirs people to think from the side of the oppressed and colonized.
3. Focuses on the development of concepts that have been used to marginalize.

(B) Three Negatives

1. Often focuses too much on concepts and theory and not direct political action.
2. Confuses content with the application of concepts because knowledge reduces to power.[17]
3. Most of it is written using obscure and vague language that is far removed from common-place nomenclature, especially of “subalterns.”

The Sociological Method (SM)

(A) Three Positives

1. Works to help people better understand how human beings identify themselves.
2. Focuses on statistical and demographical facts.
3. Has strong explanatory power when it comes to understanding religion.

(B) Three Negatives

1. Assumes substantive definitions of religion that have historical, political, and ideological baggage.
2. Fits people into molds that do not necessarily represent their individual views.
3. Thinks of itself as objective and value-free and not, at least in most sociologists minds, an extension of colonialism.

The PCM “Often focuses too much on concepts and theory and not direct political action.”  While I think that the PCM does incorporate “real life people” into their theories and criticisms, I cannot see what its actual political goals (e.g. realpolitik) are.  Does challenging the power inherent in discourse, in definitions, and in classifications actually accomplish anything on the ground level?  Is the goal to de-colonize academic discourse, real-world politics, or both?  If the goal is both, and if the PCM is primarily Marxist, it is safe to assume that any real political end would be brought about through Marxist means (e.g. violent proletariat revolution and/or State redistribution).  Terry Eagleton highlights some of what he considers the overreaching political rhetoric of those who use the PCM:

“There are some kinds of criticism – Orwell’s would do as an example – which are a good deal more politically radical than their bluffly commonsensical style would suggest. For all his dyspepsia about shockheaded Marxists, not to speak of his apparent willingness to shop Communists to the state, Orwell’s politics are much more far-reaching than his conventionally-minded prose would suggest. With much post-colonial writing, the situation is just the reverse. Its flamboyant theoretical avant-gardism conceals a rather modest political agenda. Where it ventures political proposals at all, which is rare enough, they hardly have the revolutionary élan of its scandalous speculations on desire or the death of Man or the end of History. This is a feature shared by Derrida, Foucault and others like them, who veer between a cult of theoretical ‘madness’ or ‘monstrosity’ and a more restrained, reformist sort of politics, retreating from the one front to the other depending on the direction of the critical fire.”[18]

The SM, not off the pressure-cooker, has some of its own flaws as well.  In my opinion, Asad’s critique of anthropology can equally be applied to sociology.  Just replace “sociology” for “anthropology” in the following quote: “Thus the scientific definition of anthropology as a disinterested (objective, value-free) study of “other cultures” helped to mark off the anthropologist’s enterprise from that of colonial Europeans (the trader, the missionary, the administrator, and other men of practical affairs); but did it not render him unable to envisage and argue for a radically different political future for the subordinate people he objectified and thus serve to merge the enterprise in effect with that of the dominant status-quo Europeans?”[19]  An argument can be made that the early European and Anglophone sociologists had just as much a part in colonialism as their anthropological counterparts, although, Asad’s criticism here is not just about aiding the “subordinate people” but also the notion that many people consider the social sciences “value-free” and “objective.”

The three criticisms I have against the SM are very close to what many who utilize the PCM would say.  Classifications and categories of “religions” often meant that many people who fill out sociological surveys have limited choices to choose from.  The justification from the SM for this issue is that a good questionnaire or survey has an “other” box that can be checked and a place where one could write in what they consider their own “religion.”  This is why, say for the American Religious Identification Survey (ARIS) of 2001, over 100 different “religious groups” were reported.  As Kosmin and Keysar point out, “By casting a wide-net self-reporting catches a vast variety of self-descriptions.”[20]  Here is an instance of what Bruce thinks of as extra “tools” that the SM has to work out some of its exclusionary kinks.

The best alternative I can think of between the SM and the PCM is to create what I call “post-colonial sociology” (PCS) or the “sociological method with qualifications” (SMQ).  This kind of SM would be cognizant of the criticisms from the PCM, and continually qualify its statements about “religion” or “non-religion.”  They may still disagree on certain issues of knowledge and power, the relationship between “invention” and “discovery,” and other such disputes, but this does not mean that they cannot compliment each other when used carefully.  There are obvious merits to both methods, and each ought to learn from one another, and to simply isolate the disciplines and methods from one another does more damage than good.

However, given the strong historical and ideological disagreements between those who use the SM and those who use the PCM, it seems like a very large chasm to bridge.  Much like the gulfs between analytic and continental philosophy, between the “West” and the “East,” and within the social sciences themselves, much of the disagreement comes from not being willing to see the merits and problems with both sides of each “position.”  The pragmatist in me wants to bring the best of both methods in the understanding of what “religion” does or does not amount to.  Let’s admit our various biases and basic assumptions, acknowledge the strengths and weaknesses in both of these methods (or any method for that matter), and do our best to move forward.  This may seem like a modest proposal, but I think it has far-reaching consequences for both of these methods.