A metaphysics for the study of religion: A critical reading of Russell McCutcheon

The fork in the road for the critical study of religion

McCutcheon is right that I identify as a “critical realist.” The label refers in this case to the movement begun by the philosopher of science, Roy Bhaskar (1944–2014). Bhaskar’s first book concerned the natural sciences, but he then applied his approach to the study of the more complicated world of human behavior, and today it is often sociologists who are championing his work. ² The primary value of critical realism, in my judgment, is that unlike one-sided views of the sciences that exaggerate the role of either the knower or the known, this meta-theory provides a coherent reconciliation of the epistemic and the ontic aspects of knowledge. Bhaskar argues that, on the epistemic (or “critical”) side, when a group of people conduct any inquiry, the result of that inquiry

is a social product much like any other, … no more independent of its production and the [people] who produce it than motor cars, armchairs or books, which has its own craftsmen, technicians, publicists, standards and skills and which is no less subject to change than any other commodity. (2008, 21)

Though McCutcheon raises some concerns with my work, he ends his chapter on a generous note, suggesting that he and I might disregard our differences “in service of the common sled we can pull, at least for a ways” (2018, 118). The common sled we both seek to pull is our shared focus on the constructed nature of the concept religion. But how great are the differences we would have to overcome?

Given my critical realism, McCutcheon and I are not now on the same page. As mentioned, I see “religion” as referring to a particular social structure that operates in some cultures, so I do not agree that scholars should discard the use of the term (11n1). I do not agree that scholars have seen “the collapse of religion as an analytic category” (28). I do not agree that the word “religion” has no reference (39; also 104). I do not agree that scholars should only “study ‘religion’ and not religion” (18; also 23). In fact, the very title of Fabricating Religion reflects McCutcheon’s rejection of the realism I want. As he explains, the process of human beings coming to have a world has two stages: first, people pick out some aspect of their otherwise undifferentiated environment and slice it off from the rest, creating it as a “something,” but, second, they can fail to recognize their own contribution to the thing’s “reality” (57). Because the verb to fabricate can mean both “to build” and “to lie,” this book’s title (unlike that of Manufacturing Religion [1997]) implies both of the steps in this creative but self-deceived process (McCutcheon 2018, 2–4, 9, 42). The aim of the critical nonrealist study of religion is then to denaturalize the discourse about religion so that one can recognize that the statement “religions exist,” like every claim about reality, is a fabrication.

McCutcheon is exactly right when he says that the difference between his nonrealist version of the critical study of religion and my realist version turns on my belief that some things exist that are observer-independent (110). He is right that I wish not to relinquish what he dismissively labels the “commonsense assumption” that there is an “exterior world” (117–8; also 104n15). A realism like mine permits one to see that human conceptualizing does invent some realities, like British traffic laws, NBA championships, marriages, private property, and U.S. dollars. None of these social realities exist independent of collective agreements between people. They are therefore not natural kinds but social or human kinds (Khalidi 2013; Mallon 2015); they require a “historical ontology” that studies culturally emergent ways of being (Hacking 2002, esp. ch. 1). But I also hold that other realities (like carbon, stars, volcanoes, or cell nuclei) existed long before human beings arrived to name them. McCutcheon does not recognize this distinction between objects that do and do not depend on human conceptualizing for their existence, because “when we talk about objects in the world … we create them” (McCutcheon 2018, 64n4, also 175). From his perspective, both cultural and natural realities are socially created (111). He holds that reality is not “already packaged” into cells or molecules and so on, since it has no structure before human conceptualizing imposes one (74, also 4, 38–9). He even criticizes the claim that language or culture “mediates” the world, because this verb implies that there is “some sort of a real, exterior world” that is not completely created by human concepts (4). The beliefs that we consider true are therefore no more anchored in a “real world” than their rivals (87). In a previous paper (Schilbrack 2019, 66), I argued that if a scholar adopts this view that there is no reality apart from language, then the scholar would never be able to claim that their scholarship is “accurate.” I meant this as a reductio ad absurdum, but McCutcheon agrees, “None is therefore any more accurate than another” (McCutcheon 2018, 40). I think that it similarly follows from this nonrealism that scholars cannot accuse rival views of “distorting” things or of providing a “flawed” description. Again, McCutcheon agrees that, without a reality to be accurate about, scholars should stop using such terms (105, also 107). There is no “world that we can describe in better or worse ways” (104n15, also 41).

Given these two options, critical realism and critical non-realism, scholars engaged in the critical study of religion who come to see that the concepts with which they think were invented by particular people, at a particular historical location, for a particular purpose, face a fork in the road. The difference between these two approaches does not have to do with whether one is willing to interrogate the concept religion; both sides want that. It also does not have to do with whether concepts are inventions; both sides hold that view. Rather, McCutcheon and I disagree about a question that is, in the end, metaphysical: I hold that human beings have invented concepts as tools in order to cope with a world with a structure that exists whether or not people know it, and McCutcheon thinks that human concepts create that structure. In traditional terms, I nest the study of religion within a realist metaphysics. Since McCutcheon argues that all the things of the world depend for their existence on the conceptual activities of people, his approach to the study of religion is built upon an idealist metaphysics. ³

Arguments against realism in the critical study of religion

The difference between realism(s) and idealism(s) has been a deep division in the history of philosophy, and it is unlikely to get settled in this essay. But scholars involved in the critical study of religion should be clear about what genealogizing or deconstructing one’s concepts does and does not imply about how those concepts work. I would therefore like to consider three of the arguments that McCutcheon gives for his deflationary view and show that none of them give critical scholars a good reason to adopt nonrealism about “religion.”

One tack that McCutcheon takes in this book is to argue that the concept religious is co-constituted as part of a binary pair with the concept secular, “with neither appearing first nor one exclusively anchoring the other” (13). Recognizing the codependence of “religious” and “secular” helps one see that the meaning of each term in this binary is internally linked to that of the other, just as with, for example, the binary pair of left/right. Concepts do not get their meaning by an invisible thread that attaches them to some real thing in the world (what has sometimes been called the “Fido”/Fido theory of meaning). Instead, the meaning of concepts emerges as people use them, and the intelligibility of one side of a binary parasitically depends upon that of the other as its foil: one cannot have one without the other. If one cannot have one side of paired concepts without the other, what sense does it make for realists to claim that religion existed “before” secularity emerged? We can call this “the codependent meaning argument.”

The codependent meaning argument does not support a non-realist conclusion. The fact that concepts depend for their meaning on relations to other concepts does not undermine their ability separately to refer to things in the world; on the contrary, it makes such reference possible. One can see this clearly with the left/right pair. Take the description of someone’s driving that “She made a right turn at the intersection,” or the description of someone’s politics that “He is working for a party on the left.” Even though left and right are co-constituted as a binary pair, with neither appearing first nor one exclusively anchoring the other, it does not follow that left and right collapse as analytic categories, nor that we should discard the use of the terms, nor that one should only study those who use these labels. Moreover, the fact that people at a certain time cannot understand one side of a binary without presupposing the other conceptually does not imply that one could not exist without the other historically. The claim that human societies at one point in time were religious-not-secular before the binary emerged is just as coherent as the claim that at one point they were all hunter-gatherer-not-agricultural even though that binary did not emerge until centuries later. The codependence of the concepts left and right does not imply that “right turn” and “left-wing party” do not refer to things in the world, and the same is true with religious and secular: the conceptual link between these terms enables them to name different structures in the world. In fact, because the concepts oppose each other for their meaning, the referents of these concepts will also be opposites. Just as a left turn is the opposite of a right turn, a monastery seized by a secularizing government and made into a post office has become an opposite kind of building.

A second argument that McCutcheon deploys in the book is that it is inappropriate for realists to label a social structure as “religion” if those so labeled did not think of themselves that way. As several scholars in the critical study of religion have pointed out, there is no term in pre-modern Japan, India, or China, nor in Rome, Greece, Israel, or Egypt, that matches the semantic range of the modern English term “religion.” Is it therefore not illegitimate to retroject the concept into cultures where it was not known? We can call this “the anachronism argument,” and it is a running theme in Fabricating Religion. McCutcheon critiques the use of the label “religion” when scholars are describing “the actions of people for whom the term continues to be an alien import” (16). He points out that before the term emerged in the early modern West, people were not “naturally” or “necessarily” thinking that they are religious (5). In a paragraph that takes up almost an entire page, he argues that it is unlikely that those who are said to be in a religion perceive that they are in a religion (16–7). Scholars should not presume that participants they study see the realness of their own religion (17). Those in the past whom scholars allege to have practiced religion were not “walking around talking about their religion” (19). He asks, “what sense are we making of them if this modern word is just ours?” (34, emphasis in original).

The anachronism argument also does not support a non-realist conclusion. It is true that realists about “religion” will use the term anachronistically. Because realism is the view that there are structures in the natural and social environments operating whether or not we recognize them, realists may apply labels that those so described do not know. Although there will be ethical and political reasons to be cautious about the labels one uses for people, there is nothing conceptually illegitimate about this commonplace practice. Robert Segal (2006) is a realist about religion who defends this use of labels, arguing that scholars in the study of human culture are like medical doctors who identify the features of a culture and their causes, and so they are expected to and in fact sometimes must use etic terms that the participants do not know. For Segal, just as a doctor can tell a patient that he has melanoma, and the doctor can be correct whether the patient knows that term or not, the same is true of the term “religion.” McCutcheon assumes that realism claims that religion “is indeed really in the so-called believer’s mind” (McCutcheon 2018, 16, cf. 5), but this is off target. For realist scholars of religion, it is irrelevant whether people engaging in religious practices have the concept religion themselves. ⁴

McCutcheon provides an analogy meant to illustrate the realist’s mistake. He argues that someone absorbed in following the rules that outsiders might call a religion is like someone absorbed in following the rules of reading a piece of written English: the latter person is likely not aware of “being literate” (16). McCutcheon is surely right that the literate person, immersed in a good book, is not likely to be simultaneously thinking, “I am being literate right now.” There will also be plenty of people who can read who, even on reflection, do not know this label. But the realist view of literacy is that the concept refers not to someone’s present state of consciousness but rather to social practices that give powers of communication and memory not possessed by those who cannot read. Literacy, like religion, is precisely a historically emergent social structure that exists with effects on societies and psyches, even in cases when those so described do not know the term. McCutcheon’s reading example therefore actually undermines the anachronism argument and supports a realist understanding of social labels.

A third argument McCutcheon gives for nonrealism is that what people call religion is actually indistinguishable from culture. As he repeatedly says, one should not see the religious features of a culture “as being in distinction from the things not categorized in this manner” (13). Religion is not “a distinct thing in the world” (12); the word does not name “something distinctive” (13). Now, as Leibniz argued, if two things are indistinguishable, then they are identical. If religious things cannot be distinguished from non-religious things, it would not follow that the word “religion” has no reference, but it would follow that it would have no analytic value not already captured by “culture.” ⁵ I argued above that even if religion is a codependent concept and an anachronism, these arguments do nothing to undermine realism. But if this “indistinguishable-from-culture argument” is correct, then McCutcheon is right that critical scholars of religion can discard the term.

The indistinguishable-from-culture argument is not correct. The key here is to be clear whether a critic of “religion” is arguing, as I do, that religion is not separable from the rest of culture, or the scholar is arguing more radically, as McCutcheon does, that religion is not even distinguishable from the rest of culture. ⁶ Realists can and should say that religion is never separate from the rest of culture. Religion is necessarily an aspect of culture. ⁷ Realists then treat “culture” as an umbrella term and religion as one of the forms that some particular cultures take. Whether or not any extra-cultural realities like Allah or the Dao or Devas exist, scholars of religion study texts, artworks, wars, monarchs, migrations, techniques of the body, etc.—which is to say, human culture. Even when one studies mental phenomena like religious beliefs or experiences, these will always be shaped by the interests of a specific time and place in history. However, if the concept religion is to have any analytic value, then realists must be able to distinguish this category from others. This notion that one can disentangle two inseparable concepts is sound and ordinary. The claim above that religion is one form of culture among others already distinguishes the two conceptually without separating them; analogously, even though one cannot have an apple without having fruit, “apple” and “fruit” are still analytically distinct. ⁸ One cultural form can also be distinguished from other cultural forms, even when they pervade or are coextensive with each other. Just as a film is both an artistic and a financial venture, and its very existence depends on its finances, it does not follow that the film’s artistic merit is indistinguishable from its cost, nor that “art” has lost its analytic function as a category. Analogously, even if “religion” and “politics,” say, are coextensive, it does not follow that they are identical.

In this essay, I have argued that critical reflection on our technical terms need not lead to the view that concepts for religion do not refer to real entities. As one recognizes that the critical study of religion includes both a nonrealist and a realist option, one may be surprised by the large number of scholars on this path from whom McCutcheon distances himself. For instance, Talal Asad’s massively influential genealogical approach to the concept of “religion” uncovers the way that the concept has been taken to refer to autonomous cultural systems, and Asad argues that these cultural systems shape the disciplinary mechanisms and thereby the subjectivities in societies, so they are not at all autonomous. But McCutcheon objects that Asad assumes not only that religions exist as cultural systems but also that they can be hard to manage (13). Brent Nongbri agrees with McCutcheon that religion did not exist before the modern emergence of the concept. But McCutcheon objects that Nongbri also claims that historians can have more or less distorted views of ancient realities, and this claim assumes that social structures of the past are not wholly constituted by our present redescriptions but rather exist independent of them (19, also 22, 41). Jason Ānanda Josephson analyzes the way that the use of the category of “religion” in Japan carved society into new categories that had not existed before, but McCutcheon says that his “critique has not gone far enough” (22). Josephson holds that when one uses a Western concept like “Buddhism,” one takes a system of texts and practices and social roles that were connected to each other in some way, and re-invents them as a “religion.” But McCutcheon objects that Josephson assumes that so-called “Buddhist” practices, texts, and monasteries were in some way linked to each other, and that these links had a history and influence in Japan before Europeans arrived, and Josephson does not see that the use of the western term “Buddhism” invents a link between an otherwise disparate collection of claims, actions, artifacts, and institutions (22–3). Carlin Barton and Daniel Boyarin argue that historians should imagine no religion in antiquity, but McCutcheon objects to their assumption that the job of scholars is to “get things right” about the past (ch. 2, esp. 35–6). Guy Stroumsa seeks to undermine monolithic views of religions and he argues that scholars should be careful not to reproduce the ideologies of religious elites as if they speak for all the group’s members, but McCutcheon objects that to call such scholarship “subversive” is to take sides between the groups one studies (71). William Cavanaugh argues that the concept of “religious violence” is a rhetorical invention, deployed by those who wish to consolidate the power of the secular state, but McCutcheon objects that Cavanaugh does not draw the conclusion that all distinctions between religion and non-religion are mere rhetoric and so Cavanaugh “only goes part way” (27). David Chidester highlights the way that the term “religion” was invented and deployed as a tool for colonialism in Africa, but McCutcheon objects to Chidester’s view that colonializing powers misunderstood and misrepresented African religions (17, 67). Ann Taves also argues that, since no experience is inherently religious, scholars of religion should realize that what they study is always what some people “deem” religious, a move that rejects the notion that religion is sui generis and treats religion instead as simply one example of a larger social process of marking some experiences as special. But McCutcheon (2012, 238) objects that she does not then restrict the study of “religion” to this deeming process but is willing to look at the characteristics of the experiences so labeled, not realizing that apart from our ideas, reality is nothing but “white noise.”

This is a remarkable list. Asad, Nongbri, Josephson, Barton and Boyarin, Stroumsa, Cavanaugh, Chidester, and Taves are not scholars who refuse to reflect on the politically invested origins and colonialist uses of “religion.” They do not assume religion is universal. They do not think that religion is apolitical or sui generis. They do not think of religion as a natural kind. To a significant extent, these scholars are the critical study of religion. Nevertheless, McCutcheon says that none of them goes far enough because, in each case, they fail to see the world or the past as completely created by the way we describe it. Each in some way takes the realist view that the study of religion should have a double focus that includes both “religion” and religion, both concept and entity. And McCutcheon’s list should actually be longer. J. Z. Smith (1998, 269) was committed to what he called a “double archeology” that investigates both the representations of history and the social reality that is represented—a commitment that led him to say that religion did exist even where the word was not known—and so he should be on that list, too. Critical historians Bruce Lincoln and Aaron Hughes also treat the social structures of the past as more than the retrojection of contemporary speakers. Neither abandons the idea that a historical description can be distorting or flawed. Ivan Strenski and Robert Segal also reflect on the construction of “religion,” as do Michael Stausberg and Steven Engler, but none of them accept McCutcheon’s view that reality apart from human concepts is unstructured white noise. This paper opened with the claim that the critical study of religion need not lead to nonrealism because a realist critical study was possible, but the truth is that almost all critical scholars of religion are realists. Fabricating Religion should therefore be read not as representing “the” critical study of religion, but rather as a radical approach shared by very few who are engaged in that project. McCutcheon’s view is radical not because it reflects on the categories that scholars usually take for granted, but rather because his version of critical study denies that the world has any structure before human beings arrived.

Materialism in the critical study of religion

Given this metaphysical fork in the theoretical road, what is the critical scholar of religion to do? Clearly, I think that critical scholars who are idealists should admit the existence of a world not created by human conceptualizing; they should become realists. Until that day, however, I want to return to McCutcheon’s offer of cooperation and see whether we can find common ground. Despite our divergent metaphysical views, both McCutcheon and I hold that we want a materialist account of social structure.

McCutcheon says that, like many of us, he was initially inspired to reflect critically on the concept of religion by J. Z. Smith’s Imagining Religion (1982). Also influenced by Noam Chomsky’s Manufacturing Consent (1988), however, he deliberately veered away from the idealism implied by the word “imagination” and gave his first book a title designed to take into account the role of political and economic power. The combo title of Manufacturing Religion (1997) reflects what McCutcheon calls his “more materialist bent” (2018, 2). One also sees McCutcheon’s materialist bent in his long-standing interest in social formation (e.g., 2001), that is, the process of how people struggle over how their collective life should be organized, and how the winners institutionalize and seek to legitimate those arrangements. Crucial to this process is the idea that people’s actions generate social structures, “material conditions beyond the scope or control of the individual” (2001, 27), that in different ways enable and constrain the subjectivity of those who live within those structures. On this materialist account, the thoughts and actions of individuals, their sense of identity and their very sense of what a person is, are always conditioned by the structures operating in the social worlds they inhabit (McCutcheon 2018, 63). Those social structures were generated through previous acts of human agency, and that agency was itself generated though previous social structures, so there is a dialectical or “recursive” relationship between social structure and agency (McCutcheon 2001, 28). Since McCutcheon’s model of the relationship between structure and agency distinguishes between intentional human actions and the non-intentional material conditions that make them possible, it resembles Bhaskar’s “transformational model of social activity” according to which “[s]ociety is both the ever-present condition (material cause) and the continually reproduced outcome of human agency. And praxis is both work, that is conscious production, and (normally unconscious) reproduction of the conditions of production, that is society” (Bhaskar [1979] 2015, 34–5). ⁹ In both cases, even when the structures are the product of deliberate actions, they are material or objective in the sense that, as Karl Marx said, one does not create one’s own circumstances. In short, both McCutcheon and the critical realists want to include full appreciation of the role of material structures in the study of human behavior.

Let’s talk more about these material structures. From the critical realist perspective, a social structure is a set of relations among social positions and social constructs. ¹⁰ A set of relations would be invisible, so what does it mean to say that it exists? This is a fascinating metaphysical question because a social structure might be composed of nothing but people, but the realist argues that such a structure explains the difference between a random set of people who lack internal relations to each other, who are just an aggregate, and a group of people (like a team, a band, a business, or a religion) who have those relations. In some cases, the social structure requires those in them to understand and recognize their relations to each other (e.g., one cannot get married or be a quarterback without knowing it). In these cases, people deliberately create and maintain the social structure. In other cases, however, the social structure does not require those in them to recognize or understand the relations at work (e.g., one can be privileged or exploited without knowing it). These latter cases are created and maintained as side-effects of people’s actions. As with structures of opportunity given changes in the market, or structures of power given the transformation of traditional authorities, these social arrangements are not deliberate products of human agency and they may even have not yet been discovered. In both cases, however, and whether or not the people participating in these structures realize it, the social structures have top-down effects in that they make some of the participants’ actions possible and disallow others; they empower some of the participants and disempower others.

Material conditions of human behavior do not exist only in the natural world. I argued above that a critical realist will distinguish between things that do not depend for their existence on human concepts (like carbon and stars) and those that do (like traffic laws and private property). Within this latter category, a critical realist account of social structures will further distinguish between concept-dependent realities that are merely subjective and those that have become objective. The former include my intention to have enchiladas for dinner, my belief that Kansas City is in Kansas, or my judgment that Memento is the best movie I have seen. Subjective realities like these intentions, beliefs, and judgments can shape an individual’s behavior. They are relatively fragile, however, and if the individual drops them, they simply cease to exist. But when multiple people come to share a commitment to an intention, a belief, or a judgment, it becomes much less fragile. Think of a commitment by two groups of people to treat a line in the sand as the border between their lands, a commitment that in time becomes entrenched, institutionalized, and policed. Entrenched commitments like this underlie the concept-dependent truths that Texas is now part of the United States, that tenants pay rent to landlords, and that Wednesday is the day after Tuesday. Even if they come to be taken for granted, these social facts are concept-dependent and so they remain historically contingent and contestable truths. Nevertheless, to the extent that people come to accept political, economic, and calendrical systems, there are objective answers to questions like “Have we crossed the border?,” “Do we have enough money for rent this month?,” or “Am I supposed to go to work today?” ¹¹ In this way, what began as a way of thinking or acting can become a (concept-dependent) entity in the world, a social structure, with sometimes powerful effects. ¹² Those who argue that a border is merely a concept are not wrong, from one perspective. But this response is unlikely to have much of an influence on border guards.

As we have seen, McCutcheon wants to make a distinction between human subjectivity and the social structures that are the material conditions that make that subjectivity possible. ¹³ A distinction like this is necessary for any materialist account of social formation. When McCutcheon actually analyzes social structures, however, he reverts back to treating them and their effects on the world simply as ideas. For example, he writes that we think that nations like Canada or Sweden are real entities “only because we all think and act as if there is such a thing as a Canada and Switzerland” (2018, 61, emphasis in original). The alleged legislative, economic, and military powers of nations are also just ideas: nations are “thought to have a variety of effects in the world” and they are “thought to exert influence” (61, emphasis added). From the critical realist perspective, this is half right: nations are the product of collective human efforts, including collective mental efforts, and nations do depend for their existence on concepts. If all human beings ceased to believe in Canada and Switzerland, then those nations would cease to exist. However, a purely idealist or “as if” approach fails to recognize that the ideas of “Canada” and “Switzerland” have become institutionalized to the degree that they no longer depend on what any individual thinks. The laws, economies, and military force of these nations are more than mere concepts. They have achieved an objective or material status. By treating nations as nothing but ideas, McCutcheon’s analysis subjectivizes social structures and returns to the language of imagination. This is the opposite of a materialist account of power. ¹⁴ I judge that there is no way to give a materialist account of social formation without recognizing the distinction between the concept-dependent realities that are subjective and those that have become objective in this way.

A materialist account should treat “religion” in this same way. As thoughts and actions predicated on alleged superempirical realities become shared practices, as those practices generate social roles, and as those roles get institutionalized, a concept-dependent social structure emerges. The set of relations in those practices, roles, and institutions—for example, the relations between renunciant and householder or those between shaman and patient—will shape the agency and subjectivity of the participants. A social structure like this will have these effects whether or not the participants or scholars call this religion, and these effects are exactly the criterion for saying that a religion is “an anthropological and psychological reality.” Unfortunately, however, that religions exist as material structures like this is the claim that McCutcheon (2018, 15–6) repeatedly rejects on the grounds that religion is not there “before we start talking about it” (15). McCutcheon’s treatment of religions, like his treatment of nations, does not examine how religious social structures operate as the material conditions of social formation but instead defends the idealist view that people think and act as if there are religions that are thought to have effects. ¹⁵

In order to provide a materialist account of how religious social structures cause effects in the world, scholars of religion should endorse a realist metaphysics. As we have seen, McCutcheon dismisses claims of a “some sort of a real, exterior world” (4), and there are many scholars today who are wary of any claims that one has access to a reality not constructed by human concepts. But the realism of critical realism is not, as Kant’s realism was, a gesture at some unknown noumenal reality, an unknowable “X” that we have to posit to make sense of appearances. Instead, “reality” for critical realists refers to the causal mechanisms that operate in natural structures such as ions, carbon dioxide, ribosomes, or ecosystems and in human-built structures such as ghettoes, electoral colleges, public schools, or equity-based derivatives. “Realism” here means that the way we conceive, articulate, and model these mechanisms may fail to do them justice. On this account, even though human beings invented the concepts of genes, germs, the French Revolution, white privilege, and so on, these concepts are not like virtual reality glasses that trick one into thinking that there is a world independent of one’s mind, but rather like barbecue tongs: flawed tools designed to grasp real things.

Since metaphysics by definition concerns the most abstract ideas, metaphysical discussions can be fairly distant from the concerns of social scientists. What difference does it make whether a scholar is a realist or idealist? In closing, let me suggest two answers that should appeal to critical scholars of religion who may find themselves tempted by idealism.

The first value of a realist metaphysics has to do with the goal of reversing the disciplinary isolation of religious studies in the university. This essay has pointed toward a nonreductive materialism according to which human conceptualizing does not float free of its roots in the material world. According to this layered or stratified ontology, human beings are material entities whose behavior is both enabled and constrained by the laws of nature, by biochemical process in their bodies, and by the history of natural selection. As one’s attention moves from the atomic to the molecular to the cellular strata, new properties emerge, and even more emerge as some animals develop the capacity to reflect on their intentions, as they create groups and not just aggregates, and as the patterns of their movement generate habits, cultures, and social structures. This metaphysics provides a conceptually coherent map of the relation of the disciplines in the university. Given a nonreductive materialism like this, the tasks of understanding, explaining, and evaluating human behavior require one to take into account the way human beings are constituted by material forces and scholars who study religious phenomena from this perspective do not insulate their objects of study but instead nest them within the insights of the other research disciplines in the university.

The second value of a realist metaphysics concerns the explanatory task of the academic study of religion. A nonrealism like McCutcheon’s incapacitates the project of identifying the materialist conditions that help us explain social behavior. The most accomplished critical realist today, I expect, is sociologist Margaret Archer, and at a conference I once asked her why theorists in the social sciences seemed increasingly to be turning from the focus on language that dominated theorizing in the 1990s to the burgeoning number of realist proposals today. Her answer was that as people struggle with and seek to address problems in their lives, they are required to find the real mechanisms that generated the problems. She gave the examples of global warming and wealth inequality. People are not satisfied by the answer that global warming or inequality is merely a rhetorical or linguistic convention, that they are a product of one’s “final vocabulary,” or that the source of the problem is that people think as if global warming or wealth inequality are real. To address problems in the world often requires scholars to go beyond human conceptualizing to locate the physical or social structures that have caused those problems. The same is true for scholars of religion. If some of the material conditions that enable and constrain human behavior are religious social structures, then scholars of religion will have to adopt a realist framework if they hope to identify the casual mechanisms that explain that behavior. ¹⁶