During the decade I have been teaching undergraduates to think about theory in the study of religion, I have noticed a pattern to how they evaluate different definitions. A student will often say something like “the problem with this definition is that it doesn't take into account the experiential aspects of religion.” But religion, in (for example) Hervieu-Léger's (2000: 87) definition, is an “invocation of the authority of a tradition … which serves as support for the act of believing … One designates as ‘religious’ all forms of believing which justify themselves, first and foremost, upon the claim of their inscription within a heritage of belief.” Experience doesn't come into it. But the student was starting with an implicit model of religion, and assessing how well the definition matched it.
The broader point that I was trying (and apparently failing) to make is that definitions and theories are two sides of the same coin: the definition encircles the data we use. This is indeed the origin of the word ‘definition’—dēfīnīre, to set bounds to, to limit or confine. So if we continue to use Hervieu-Léger's definition, we would have to include sports, nationalisms, and Marxism in the category ‘religion’ (while noting that Hervieu-Léger herself seems to implicitly distinguish between religious and secular ‘belief’). If, on the other hand, we use Martin Stringer's (2008: 113) definition of religion as “intimate relationships with the non-empirical other,” then we would have to exclude these, along with Confucianism and much of Theravada Buddhism. The data set derives from the definition, not the other way around. The definition is only ‘wrong’ insomuch as it doesn't do what we want it to do.
Perhaps it was never realistic to expect undergraduates to grasp this meta-theoretical point. However, as I want to argue in this article, I think that this hints at a larger issue in category formation in the study of religion, and perhaps in the social sciences more broadly. To demonstrate this, I want to contrast and compare two case studies from my own research—Gnosticism and conspiracy theories—to show how our concepts shape our empirical findings and what we conceive of as the stuff for analysis. I don't propose that the issue is restricted to these cases, however, nor even that they necessarily make the best argument. Rather, it was while working on them that the significance of the following issues became clear to me.
First, in both cases, the category is privileged over the data—although in strikingly different ways. Second, ideas of what counts as ‘religion’ (typically implicit) come into play in demarcating their boundaries. Essentially, these decisions are made by privileging (certain kinds of) knowledge (or ‘belief’) as legitimate and others as deviant. This means that, third, terms from a specific emic discourse are treated as though they are comparative analytic categories.
Robert Baird (1971) identifies three types of definition in the study of religion, which I will use throughout this article:
Stipulative (or functional)1—a definition that sets out that, for the purposes given, a term will be taken to mean what is stipulated. Definitions of this type are therefore semi-arbitrary2 and heuristic, indicating what we will do with the data. They lend precision to an argument by removing ambiguity, especially where a term has more than one meaning. As they do not posit any relationship to hypotheses, however, they have no truth value—in other words, they cannot be true or false. They can only be more or less useful within the specific context in which they are established (ibid.: 6–10).
Lexical—a definition that asserts what a term's meaning was to a specific person or group at a specific place and time. It is therefore a historical statement, and, as such, it does have a truth value. It may not or may not be true that ‘religion’ means ‘a belief in spiritual beings’, but it is true that E. B. Tylor asserted that it did. It is such definitions that the critical study of religion is predicated upon (ibid.: 10–11).
Real—a real definition is a true statement about reality. As such, it has a truth value, given that the statements can be verified with proof—that is, with empirical data of whatever sort. In the study of religion, this is often a search for an ‘essence’, and in more metaphysical or theological versions, data will not be readily forthcoming. Sometimes this is an attempt to unify the various stipulative definitions included in an ambiguous term; other times, it is a more normative project, which insists that the reader take a stipulative definition as a real definition (ibid.: 11–14).
The students in question, then, were confusing stipulative definitions with real definitions, and using what Baird named the “essential-intuitional method” (ibid.: 2):
By intuitional I mean a method in which the historian of religions does not recognize a need to begin his work with a definition of ‘religion’, thereby marking the limits and extent of his study. This method assumes that we all know what is meant by the word, and that, given room for accidental differences, ‘religion’ is essentially unambiguous. This introduces the other aspect of this method: essentialism or realism. It means (by implication and method) that religion is a something out there whose ‘essence’ can be apprehended by the historian of religions.
More than that, this approach implies that the lack of an agreed-upon definition of a term such as ‘religion’ stems not from the fact that it has been used to mean many different things, but rather from the difficulty in putting such a complex (and perhaps even ineffable) thing into words. It is as though ‘religion’ is outside the data as defined. Our job as scholars, then, is to somehow put into words what we all already know. As an aside, the confusion between these types of definition is behind Jonathan Z. Smith's (1998: 281) famous and playful remark:
It was once a tactic of students of religion to cite the appendix of James H. Leuba's Psychological Study of Religion (1912), which lists more than fifty definitions of religion, to demonstrate that “the effort clearly to define religion in short compass is a hopeless task” (King 1954). Not at all! The moral of Leuba is not that religion cannot be defined, but that it can be defined, with greater or lesser success, more than fifty ways.
What makes this issue significant is that it evidences the developing ‘post-theoretical shift’ in the study of religion more broadly. For McCutcheon (2017: 4), scholars working in this paradigm are positing that “they can somehow get behind or beyond theory, to the so-called empirics of the matter … thereby acquiring some sort of impeded access to the world.” These claims often come with a rejection of explicit theorizing. In choosing not to explicitly theorize one's categories, however, one simply accepts the implicit category, which is often a real definition presented as a stipulative definition. Not explicitly addressing the criteria for the data one is using makes it easy to then manipulate that data to fit the category, or sometimes vice versa, as we will now see.
Gnosticism
My first example is one with a long heritage in the study of religion. Indeed, Gnosticism was well-established even before the emergence of religious studies as a discipline separate from theology. A full explanation of the life and times of Gnosticism, and the so-called Gnostics, is beyond the scope of this article. To learn more, readers may seek out my Gnosticism and the History of Religions (Robertson 2021a), but for our purposes here, a brief summary will suffice. First, there has never been a consensus on the definition of Gnosticism, and, second, almost everything people know about it is little more than speculation.
The catalogue of groups confessing so-called gnosis assembled by Irenaeus in the second century ce was concerned only with identifying heresy. The issue with these heretics was, for Irenaeus, that they claimed knowledge (i.e., gnosis) falsely—that is to say, illegitimately. They weren't a specific group or movement, and they didn't all preach the same heresies. Nor, it should be added, was gnosis considered a special or unique kind of knowledge. It was a term widely used at the time, and Irenaeus's problem was only the illegitimacy of these particular claims.
The narrative of the heresy hunters got taken up in the polemic of Protestant theologians in the eighteenth century, who cast Catholic ritualism and iconography as gnostic barbarism. This then transformed in the mid-nineteenth century, when these Gnostics became proto-Protestants—purveyors of a ‘pure’ Christianity, free of later Greco-Roman accretions (in other words, Catholicism). This is all still drawn from the heresiological accounts, along with some Mandaean and Manichaean texts from Iran and Persia that we now know had no substantive connection with Irenaeus's groups. Nevertheless, the motif of the survival of this pure Christianity through the ages and into modernity got picked up by Theosophists, psychologists, and existentialists in the late nineteenth and early twentieth centuries, and has currency even today.
Then, in 1945, something quite remarkable happened—a body of sectarian texts from the fourth to the fifth century ce, mostly previously unknown, was discovered in the Egyptian desert near Nag Hammadi. Together with the Dead Sea Scrolls, discovered in 1947, the Nag Hammadi texts revolutionized our understanding of the complex sectarian make-up of the early Christian period. They were immediately identified as Gnostic, and due to a lack of scholarly access to the texts, this belief continued until the mid-1970s when the Nag Hammadi material was made widely available. It then became clear that the primary sources were at odds with the category as it had long been established in the study of religion. In biblical studies today, the category has lost much of its currency for this reason, after a heated debate instigated by Michael Williams's (1996) Rethinking “Gnosticism.” In the aftermath of these ‘Gnostic wars’, some scholars maintain that Gnosticism remains useful as a heuristic category (notably the so-called Yale School, particularly David Brakke and Bentley Layton), but few argue that Gnosticism existed as a historical religious movement, or even a world religion, as earlier scholars including Hans Jonas and Gilles Quispel maintained.
Yet most popular and scholarly ideas of what Gnosticism was (and, often, is) come from these earlier scholars, who were extrapolating from hereseological accounts and supposed transmissions to later religions. In fact, what we see is a category so firmly entrenched that some scholars rejected the new data because they did not fit the category, rather than the other way around. Like ‘religion’ in the students’ essays, the problem becomes finding a definition for a very complex but nevertheless intuitively real thing, rather than acknowledging that the category itself is the problem.
For example, April DeConick (2016: 6) simply dismisses critical work on Gnosticism, lamenting that “because this perspective has become so dominant, since the 1980s definitions in academia have become impossible to maintain.” Similarly, Roelof van den Broek (2011: 2) argues:
Michael Williams has convincingly shown that the popular view of Gnosticism as a monolithic religious movement is untenable. He concluded that it would be better to dismantle the whole category of “Gnosticism” altogether, and that terms such as “gnosis” or “Gnosticism,” “gnostic,” and “gnostic religion” are so vague that they have lost any specific meaning and should be avoided as well. Elsewhere I have argued that while Williams’ analysis is convincing, his solution is too radical. The terms “gnosis” and “gnostic” are perfectly applicable to all ideas and currents, from Antiquity to the present day, that emphasize the idea of a revealed secret gnosis (spiritual knowledge) as a gift that illuminates and liberates man's inner self. The term “Gnosticism,” however, should better be restricted to the mythological gnostic systems of the first centuries.
Van den Broek claims to be ‘convinced’ by Williams's argument, but intends nevertheless to go on using the terms ‘gnosis’, ‘gnostic’, and ‘Gnosticism’ just as he always has—that is, with ‘gnosis’ as a category of special knowledge identified phenomenologically by the scholar, and ‘Gnosticism’ indicating a gnostic religion of the early Christian era. Williams's point is not about which signifiers we use, however, but that no such signified ‘movement’ ever existed. Van den Broek's use of “emphasize” here is significant also, as he includes many groups that do not claim gnosis for themselves, including the majority of the ‘Western esoteric’ sources he is writing about. Their inclusion is van den Broek's decision, based apparently upon his assessment of whether their claims of special knowledge are identical with gnosis. Thus, this is a real definition—it is predicated upon the existence of a special kind of knowledge, found throughout history, and intuitively identifiable by the scholar.
Another aspect of this is shown by the many contemporary scholars who do not acknowledge the groups who do actually identify as gnostics today, including some that are rather large, like the Samael Aun Weor groups in Brazil and throughout South and Central America, and Ecclesia Gnostica and the Apostolic Johannite Church in the United States. These groups are not mentioned in van den Broek's work or in the work of most other scholars using this terminology in the Western esotericism framework, such as Wouter Hanegraaff's (1996) New Age Religion and Western Culture. DeConick's (2016) The Gnostic New Age argues that the roots of the New Age movement lie in the ‘countercultural spirituality’ of the gnostics of Late Antiquity, but makes no mention of the Aun Weor groups, even though Aun Weor explicitly identified himself as the avatar of the New Age. In these cases, contemporary Gnosticism is essentially rejected because it doesn't fit the scholar's opinion of what the essence of Gnosticism is.
The issue is that what gnosis is and who are real gnostics are first-order insider claims. Gnosticism is by definition a second-order category, a categorical abstraction created by scholars for their own purposes, but it remains reliant entirely upon first-order ontological claims. Thus, although it presents itself as a stipulative definition, it is in fact a real definition—albeit one presented without evidence.
As Baird (1971: 5) notes, the tendency to “proceed with the study of religions without an explicit definition seems to have been an attempt to avoid an unwanted a priori.” This a priori, as Fitzgerald (1997: 93) argues, is inescapably theological, despite the arguments of scholars that our data are based exclusively in historical reality:
When we talk about “religion” in a non-theological way, we are fundamentally talking about culture in the sense of institutions imbued with symbolic meaning through collective recognition. Further, I suggest that the proposal made by some writers that religion, while part of culture, is a distinct sub-category of culture, fails. In that case, I argue that the word “religion”, with its theological and supernaturalist resonances, is analytically redundant. It picks out nothing distinctive and it clarifies nothing. It merely distorts the field.
Thus, what is at stake when engaging with Gnosticism, and indeed religion, critically, is this theological or supernatural or metaphysical a priori that the category serves to protect and mystify.
Conspiracy Theories
Unlike Gnosticism, conspiracy theories did not become a significant area of academic research until the late 1990s, first in American history and later in social psychology and philosophy. Such theories have only appeared in religious studies since about 2010, where they have increasingly revolved around the idea of an emergent worldview called alternatively ‘conspiracism’ or ‘conspirituality’. In it, conspiracy theories are presented as either combining with religious and spiritual ideas in different ways, or as constituting new religious worldviews in and of themselves. Despite this, conspiracy theories have been compared and contrasted with religion since their earliest appearances—notably, Karl Popper's ([1945] 2013: 306) observation that they come from “abandoning God and then asking: ‘Who is in his place?’” So while some examples here are not from religious studies, how religion is defined (usually implicitly) is nevertheless central.
Those interested in a full history and critique of the category ‘conspiracy theory’ can read my UFOs, Conspiracy Theories and the New Age (Robertson 2016). For this article, suffice to say that, like Gnosticism, conspiracy theory is a highly contested category and one that changes significantly over time. There have been conspiracies throughout political history, and there have probably been theories that posit conspiracies for just as long. There are a number of examples where what we would now think of as conspiracy theories were simply accepted as fact by governments, the media, and the vast majority of the population—in other words, the exact opposite of how conspiracy theories are usually understood. These include the claims of secret societies driving the French Revolution, the Communist sleeper cells of Joe McCarthy and the House Un-American Activities Committee, and the Satanic Panic that spread across the United States in the 1980s and 1990s. None of these were, in fact, true.
But ‘conspiracy’ is not the same thing as ‘conspiracy theory’. The category as we know it today emerged during the Cold War, although it only became prominent in popular culture during the 1990s. Initially, conspiracy theory was associated with totalitarianism specifically, but broadened out to a concern with ‘extreme’ political ideologies in the wake of McCarthyism (Thalmann 2019: 25–63). An important point in this evolution toward the modern popular meaning is Richard Hofstadter's (1964) famous article, “The Paranoid Style in American Politics,” in which he challenges what he saw as a rhetorical style based on a polarized Manichaean worldview and an entrenched ideological position, which we would today recognize as conspiracist.
There have been a number of attempts to establish a real definition of conspiracy theory. Social psychological studies have attempted to quantify belief in conspiracy theories according to a ‘Conspiracy Mentality Scale’ (Bruder et al. 2014), or to ascribe a ‘crippled epistemology’ (Sunstein and Vermeule 2008) or ‘schizotypal’ tendencies (Barron et al. 2014), literalizing the metaphorical paranoia of Hofstadter's article. Like all quantitative research, there are issues about how, in reducing the field to binary questions, such data tend to reflect the concerns of those doing the study. If we start with the assumption that conspiracism is necessarily paranoid and/or incorrect, and then base a survey on that assumption, the results of that survey are highly likely to reinforce the assumption.
Philosophers, however, have questioned the assumption that anything can be identified that renders conspiracy theories in general a priori unjustified (Dentith 2018). They are not inherently illogical, nor are they always wrong. While it is often claimed that conspiracy theories are non-falsifiable (and it is certainly true that many conspiracy theories do not stand up to scientific standards of proof), they are by no means unique in that respect—a point we will return to shortly.
For our purposes here, it is enough to say that there is no agreed factor or set of factors that make a certain belief a ‘conspiracy theory’. Rather, like ‘Gnosticism’, definitions are generally derived post hoc from an agreed data set—but in this case, the category's function is maintained by changing the data set. For example, Sunstein and Vermeule (2008: 3) connect conspiracy theories explicitly with terrorism, and state that they stem “from a ‘crippled epistemology,’ in the form of a sharply limited number of (relevant) informational sources.” Yet while they accept that some conspiracy theories have turned out to be entirely correct (they mention Watergate, MKULTRA, and Operation Northwoods), their solution is to focus only “on false conspiracy theories, not true ones” (ibid.: 4). Moreover, they further remove everything except “potentially harmful theories” (ibid.: 5). So to prove that conspiracy theories are irrational and dangerous, they chose to look only at irrational and dangerous conspiracy theories. Contradictory data are removed from the data set, and the category is maintained.
Where the data set is derived from a real or stipulative definition, this is not a problem, and indeed is the very demarcation that an analytic category is intended to perform. But when the category is derived from the data set, a circular logic emerges: data are included in the category ‘conspiracy theories’ due to being irrational and incorrect, so when a specific claim is proven to be correct, it can therefore no longer be a conspiracy theory. The category is maintained over the data.
In this case, such circular logic can have real-world effects. The category course-corrects so as to include only things that are presently regarded as irrational or at least unwarranted—but then the category is applied as a marker of irrationality. For example, at present, governments are working hard to identify anti-vaxxers as driven by conspiracy theories. But it is not fair to say that all of these objections are irrational per se—absence of evidence is, after all, not evidence of absence. If it indeed turns out that COVID-19 did in fact originate from a pharmaceutical laboratory, then this claim will simply no longer be a conspiracy theory, so the claim that conspiracy theories are irrational and inherently dangerous can be maintained.3
Like Gnosticism, conspiracy theory is treated like a stipulative definition when it is actually a real definition based on claims about reality. To be a stipulative definition, specific features, essential ideas, or common functions would need to be identified from which the data set would be drawn. What we find in practice is that when this is done, there will be significant exceptions—and as importantly, other things with those features will not be included. While these analyses of irrationality, paranoia, leaps of logic, and motivated reasoning certainly fit (some of) the data for conspiracy theories (some of the time at least), the same analyses are not applied to other groups.
We have already discussed how ideas that are promoted by power brokers are excluded (such as McCarthyism or Satanic ritual abuse), serving to ensure that “conspiracy theory” applies only to the marginal. One of the most significant ways in which this is achieved is by excluding religious beliefs from equivalent scrutiny, even though the connection with belief in the supernatural is often stressed. For one example, van Prooijen et al. (2018: 320) begin their article with the statement that “people often hold irrational beliefs, which we broadly define here as unfounded, unscientific, and illogical assumptions about the world. Although many irrational beliefs exist, belief in conspiracy theories and belief in the supernatural are particularly prevalent among ordinary, nonpathological citizens.” They define supernatural beliefs as “beliefs that violate scientifically founded principles of nature, including superstition, belief in the paranormal, horoscopes, and telepathy” (ibid.: emphasis added).4 The belief in the Virgin Birth or transubstantiation or any of Jesus's miracles would fit this description, so we should expect to find religious belief included in such analyses. So why don't we? Perhaps because the authors want to present conspiracy theories as inherently dangerous (ibid.):
Such irrational beliefs are not necessarily harmless. Belief in conspiracy theories predicts maladaptive perceptions and behaviors such as withdrawal from politics, decreased civic virtue, hostility, and radicalization … Supernatural beliefs may lead people to consult spiritual healers instead of qualified medical specialists to treat dangerous illnesses, or to base important life decisions (e.g., whether to buy a house, or get a divorce) on information derived from horoscopes or a random draw of tarot cards.
It is apparently taken as self-evident that these are not issues that might be the result of religious belief, even though there is a vast body of research suggesting otherwise. Alternatively, it may be that the researchers were aware of this issue but chose not to include it because it would make their conclusions less clear-cut, or to avoid the implications if these critiques were to apply to the 80 percent or so of the American population who identify as Christian. Whichever is correct, one would assume that if the authors conclude that “beliefs that violate scientifically founded principles of nature” can be harmful, surely they would want to address all such cases, especially those that are most widespread.
So whereas the category Gnosticism is used to construct an elite type of religion, conspiracy theories rather exclude religion from a category of ‘illegitimate knowledge’. Conspiracy theory is a category that sets up and defends a hard line between legitimate and illegitimate knowledge—that is, it is about establishing ‘Truth’. As such, conspiracy theories are not stipulative or lexical definitions, but real definitions—the problems identified above notwithstanding.
Category Formation, Critical Theory, and Decolonization
Anthropology has long since moved on from pathologizing the beliefs and practices of communities in exotic parts of the globe as ‘hysteria’, ‘syndromes’, or ‘madness’, shifting to a more respectful position that attempts to explain the inner logic of such thinking, even when certain aspects are somewhat offensive to Western sensibilities (witchcraft, cannibalism, circumcision, and so forth). Yet most academic work on conspiracy theories continues to pathologize them, ascribing irrationality, paranoia, and schizophrenic tendencies. On the other hand, religious claims (including Gnosticism) are familiar parts of the familiar order of knowledge, and as such they are ideas you are permitted to think. That scholars have completely internalized some irrational narratives but ridicule other equally irrational narratives is, quite simply, hegemony—and the deconstruction of such hegemony is the purpose of the critical project.
When scholars look at conspiracy theories without a critical lens, in many respects we are doing what Tylor, Muller, Frazer, and other armchair anthropologists did in the nineteenth century when they encountered the Other—constructing colonial rationality. We describe the primitive minds of the periphery back to the sophisticated, discerning minds of the colonial center. We explain that people like them have confirmation bias, but people like us do not. We seek to explain their ideas symbolically, or explain them away as a result of deprivation or the lack of critical ability or cognitive faculty. Above all, we differentiate their ‘belief’ from real ‘knowledge’. To use Latour's (2013: 173) pithy phrasing: “We believe that we know. We know that the others believe.”
The critical turn requires us to examine why only certain versions of rationality are set apart and denied further investigation by scholars. If we fail to address why some problematic epistemological positions—like Gnosticism—get a ‘free pass’ in scholarship, while others—like conspiracy theories—do not, in whose interests are we acting?
It is for this reason that I have been advocating an epistemic turn in the study of religion, and indeed more broadly in the social sciences. Fitzgerald (1997: 95) suggests:
The way forward for those scholars working within religion departments who do not have a theological agenda, but who recognize the phenomena usually described as religion as being fundamentally located within the arena of culture and its symbolic systems, is to redescribe and rerepresent their subject matter as the study of institutionalized values in different societies and the relation of those values to power and its legitimation.
I do not think we can do this, however, without taking the epistemic question into account. Categories such as conspiracy theories, Gnosticism, and religion are central to how such ‘institutionalized values’ are legitimized, and how their relationship to power is mystified. This is, at its core, a question of which knowledge claims are presented as sacred, central to human identity, and protected by international law, and which are presented as irrational, dangerous, and demanding legal intervention.
I argue that we should make these epistemic contestations explicit, and that such claims ought to be central to our analysis. Religious studies scholars are uniquely equipped for dealing with claims of knowledge from different epistemologies—indeed, the bracketing of such claims through ‘methodological agnosticism’ might be the closest thing to a distinct method the subject can claim. Moreover, this shift could potentially free us of the framework inherited from Christianity and colonialism, without losing a distinct object of inquiry and the unique insights afforded by scholars specifically trained in the study of religion. Our concern would broaden to how knowledge claims—including but not limited to religious ones—are mobilized in the particular episteme of different groups, societies, and cultures.5
This is why critical reassessment of how we engage with knowledge claims is vital to the decolonization of academia. ‘Making space around the table’ for marginalized groups will not take us very far, so long as it is the Master's table. Which marginalized claims are permitted is still subject to colonial ideas about religion. True decolonization is only possible if we simultaneously develop a new awareness of how such claims operate, both in society at large and within our own discipline.
Conclusion
In summary, then, Gnosticism constructs a special ‘religious’ knowledge and defends its place in academia. In many cases, the category takes precedence over the data, relying instead on first-order insider claims of special knowledge. Such work confuses a stipulative definition with a real definition, mystifying the nature of its epistemological claim. Conspiracy theory, on the other hand, seeks to defend (or construct) an inviolable rationality against (particular kinds of) special knowledge. It lionizes a particular epistemic position, subtly defending elite Europeans (and their religion) above ‘primitive’ beliefs. To do so, however, it is forced to ignore or downplay some ‘irrational’ claims of special knowledge—again, prioritizing the category over the data set.
Both, then, in different ways, construct ‘religion’ against ‘merely’ irrational belief. This is what is at stake in the analytical work we all do when engaging religion. When we use these and other categories, we are stepping into a colonial episteme that does not ‘merely’ or ‘simply’ describe self-evident phenomena. Instead, it constructs, defends, and mystifies certain knowledge claims and the power structures they support.
But I argue that more is at stake. The post-theoretical turn is not only a rejection of the need for theorizing; it is also a rejection of any attempt to consciously consider our theoretical models and, in doing so, potentially change them. It is a rejection of challenges to business as usual in the study of religion, because it is this business as usual—legitimizing the claims of special knowledge by certain groups—that these scholars are interested in.
I think that in part this is a result of the methodological weakness of religious studies compared to other disciplines. Religious studies has long been regarded as poly-methodological, an umbrella term for a field of study rather than a distinct discipline per se. It is worth noting, however, that there have been attempts to establish a distinct methodology, most notably phenomenology in the 1950s and 1960s. However, as the phenomenology of religion has clearly demonstrated, that metaphysical a priori is also in part something baked in, something that many in our field want. A field split down the middle along such fundamental lines—neither one thing nor the other—is doomed to fail. But, as Baird (1971: 15) insightfully notes:
If whatever implicit definition he operates under is treated as a ‘real’ definition’, i.e. as a statement (implicit or explicit, vague or clear, defined or intuited) that gives the ‘essence’ of the thing religion, then one is … positing a ‘real definition’ under an essential-intuitive method, whether stated or not … [I]t is difficult to see how such understanding could be achieved until our use of terms is clarified. Anything less takes the historian of religions out of the academic and makes his work part of a religious experience.
In academia more broadly, and in governance and the media, religion as a category acts as a legitimizer of certain kinds of special knowledge while demonizing others. Religious studies scholars are uniquely equipped to critique this role, but too often they are content to continue to obscure it. In so doing, we risk defending, rather than critiquing, the dominant order of knowledge, and thus mystifying, rather than revealing, hegemonic power. Categories like Gnosticism and conspiracism operate as gatekeepers of knowledge, and rethinking how we engage with them—and religious studies itself—as gatekeepers for special knowledge must take place before we can hope to truly decolonize.
Notes
Baird uses both, tending to favor functional, but I am favoring functional here because it is less similar to other terms used in this article, and so is less likely to confuse.
Semi-arbitrary only, because while one could define a term as anything at all, in practice this is generally not the case. Rather, the functional definition is one of a range of recognizable positions, taken from common or academic usage. So, for example, one might stipulatively define ‘religion’ as ‘belief in spiritual beings’ or ‘relationship to the sacred’ or so on.
Now, one could argue, quite reasonably, that this is simply how stipulative definitions work. If we define ‘conspiracy theories’ as meaning ‘false theories’, then no problem. But this is not how Sunstein and Vermeule (2008) define it. If it were, it wouldn't apply to most of what it popularly does, and it would apply to some things that it doesn't popularly apply to. In short, were this the case, it would be a different category.
As an aside, the category ‘paranormal’ is itself fraught, being used more or less pejoratively in a number of different disciplines, and with religious (and especially Christian) beliefs typically excluded, except when there is some normative point to be made from their inclusion (see Robertson 2014: 58–62).
I detail this approach at greater length in Robertson (2021b), but see also Fuller (2002, 2020) and McCarthy (1996).
References
Baird, Robert D. 1971. Category Formation and the History of Religions. Berlin: De Gruyter.
Barron, David, Kevin Morgan, Tony Towell, Boris Altemeyer, and Viren Swami. 2014. “Associations between Schizotypy and Belief in Conspiracist Ideation.” Personality and Individual Differences 70: 156–159.
Bruder, Martin, Peter Haffke, Nick Neave, Nina Nouripanah, and Roland Imhoff. 2013. “Measuring Individual Differences in Generic Beliefs in Conspiracy Theories across Cultures: Conspiracy Mentality Questionnaire.” Frontiers in Psychology 4. https://doi.org/10.3389/fpsyg.2013.00225.
DeConick, April D. 2016. The Gnostic New Age: How a Countercultural Spirituality Revolutionized Religion from Antiquity to Today. New York: Columbia University Press.
Dentith, Matthew R. X., ed. 2018. Taking Conspiracy Theories Seriously. London: Rowman & Littlefield.
Fitzgerald, Timothy. 1997. “A Critique of ‘Religion’ as a Cross-Cultural Category.” Method & Theory in the Study of Religion 9 (2): 91–110.
Fuller, Steve. 2002. Social Epistemology. 2nd ed. Bloomington: Indiana University Press.
Fuller, Steve. 2020. A Player's Guide to the Post-Truth Condition: The Name of the Game. London: Anthem Press.
Hanegraaff, Wouter J. 1996. New Age Religion and Western Culture: Esotericism in the Mirror of Secular Thought. Leiden: Brill.
Hervieu-Léger, Danièle. 2000. Religion as a Chain of Memory. Trans. Simon Lee. Cambridge: Polity Press.
Hofstadter, Richard. 1964. “The Paranoid Style in American Politics.” Harper's Magazine (November): 77–86.
Latour, Bruno. 2013. An Inquiry into Modes of Existence: An Anthropology of the Moderns. Trans. Catherine Porter. Cambridge, MA: Harvard University Press.
McCarthy, E. Doyle. 1996. Knowledge as Culture: The New Sociology of Knowledge. London: Routledge.
McCutcheon, Russell T. 2017. “Beyond Cynicism: A Sampling of Current Work in the Swiss Study of Religion.” Bulletin for the Study of Religion 46 (1): 3–6.
Popper, Karl R. (1945) 2013. The Open Society and Its Enemies: New One-Volume Edition. Princeton, NJ: Princeton University Press.
Robertson, David G. 2021a. Gnosticism and the History of Religions. London: Bloomsbury.
Robertson, David G. 2021b. “Legitimizing Claims of Special Knowledge: Towards an Epistemic Turn in Religious Studies.” Temenos 57 (1): 17–34.
Robertson, David G. 2016. UFOs, Conspiracy Theories and the New Age: Millennial Conspiracism. London: Bloomsbury.
Robertson, David G. 2014. “Transformation: Whitley Strieber's Paranormal Gnosis.” Nova Religio 18 (1): 58–78.
Smith, Jonathan Z. 1998. “Religion, Religions, Religious.” In Critical Terms for Religious Studies, ed. Mark C. Taylor, 269–284. Chicago: University of Chicago Press.
Stringer, Martin D. 2008. Contemporary Western Ethnography and the Definition of Religion. London: Continuum.
Sunstein, Cass R., and Adrian Vermeule. 2008. “Conspiracy Theories.” Law & Psychology eJournal. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1084585.
Thalmann, Katharina. 2019. The Stigmatization of Conspiracy Theory since the 1950s: “A Plot to Make Us Look Foolish.” London: Routledge.
van den Broek, Roelof. 2011. “Sexuality and Sexual Symbolism in Hermetic and Gnostic Thought and Practice (Second–Fourth Centuries).” In Hidden Intercourse: Eros and Sexuality in the History of Western Esotericism, ed. Wouter J. Hanegraaff and Jeffrey J. Kripal, 1–22. New York: Fordham University Press.
van Prooijen, Jan-Willem, Karen M. Douglas, and Clara De Inocencio. 2018. “Connecting the Dots: Illusory Pattern Perception Predicts Belief in Conspiracies and the Supernatural.” European Journal of Social Psychology 48 (3): 320–335.
Williams, Michael Allen. 1996. Rethinking “Gnosticism”: An Argument for Dismantling a Dubious Category. Princeton, NJ: Princeton University Press.