The Wheatley Institution

Fellow Notes


Missing Our Target: Methodological Naturalism, Saving the Phenomena, & the Social Science of Religious Experience

“While careful and coherent theoretical explanation is a central aim of scientific inquiry, it cannot come at the expense of the subject matter under study.”

Ed Gantt | April 27, 2015
| Print | Google plus Google plus linkedin share button
A group of children perform a religious ceremony with their parents, demonstrating the significance of religious experiences.



Metaphysical naturalism, the philosophical position that only matter, energy, and their interactions exist, has become increasingly popular. However, there are still many in the social sciences who study religion and religious life and who reject this philosophical position. These researchers typically decline the claims of metaphysical naturalism because they feel that it pushes inquiry beyond the safe borders of legitimate science and into the dangerous territory of philosophical and/or theological speculation. This does not mean, however, that naturalism, per se, in all of its forms is something they feel must be abandoned. On the contrary, such researchers have often been quite willing to endorse methodological naturalism as a standard requirement of scientific practice because it seems to afford the privilege of studying religious phenomena in a genuinely scientific manner (i.e., empirically and objectively) while simultaneously providing an escape from having to make any ontological claims regarding whether religious phenomena might or might not reflect some supernatural or non-naturalistic reality. Thus, rather than speculate as to the possible spiritual nature, transcendent origins, or ontological reality of religious and spiritual phenomena, many social science students of religion have instead focused on their scientific methods of study, operationally defining certain hypothetical constructs and then investigating the various physically measureable conditions that are presumed to constitute reliable manifestations of those constructs.

For example, as Hood, Hill, and Spilka (2009) recently noted, the construct of “religiousness” is typically operationally defined in terms of whether persons are “members of a church or other congregation, attend religious services, read the Bible of other sacred writings, contribute money to religious causes, observe religious holidays or days of fasting, pray frequently, say grace before meals, and accept religiously based diet restrictions, among other possibilities” (p. 11). Once a key construct like religiousness has been operationalized in this manner, researchers can then employ various measuring devices (e.g., surveys, experimental manipulations, etc.) to ascertain the presence of the construct, as well as begin to formulate causal accounts of how the construct operates, what precise physical conditions gave rise to it in the first place, and how its influence might be predicted and controlled (i.e., increased or decreased) in various ways.

Unfortunately, when pursued in this manner, the social scientific study of religion and religious life becomes not so much the study of religion or spirituality as actually lived and experienced by real persons as it becomes the intense study of a variety of “stand-ins” for religious and spiritual experience – that is, it becomes the study of hypothetical constructs tied only to sets of presumably religious behaviors and particular aggregations of presumably reliable self-reports about individual spiritual beliefs. This is methodologically necessary, as Brannigan (2004) points out, because “the experimentalist has to conjure up a proxy or a shorthand artifice or substitute for the original event” (p.ix). He continues:

Rather than going to primary sources to study the phenomenon first hand, the experimentalist has to visualize a way of reducing the process to something that can be studied in a laboratory over a short period of time, whether or not this is the best method of elucidating the phenomenon. The result is not [for example] a study of genocide but a metaphor of genocide, a dramatization or allegory that enacts certain key processes that the psychologist feels are critical, though these are frequently researched in a complete empirical vacuum with respect to the original events that characterized the genocide. (p. ix-x).

This whole process is generally referred to as “methodological naturalism”

Of course, at the theoretical and explanatory level, all of this is seen to be vital because it serves the aim of providing an account of religion and religious experience in terms of their (presumably sufficient) socio-cultural, environmental, and biological (i.e., natural) causes. The assumption of such an explanatory approach is that such an account, when fully developed and confirmed, will provide adequate (i.e., sufficient) naturalistic account of religious experience. This whole process is generally referred to as “methodological (as contrasted with metaphysical) naturalism.”

While persistent questions about the adequacy of measurement in regards to both psychological and spiritual phenomena, as well as equally significant questions about how methodological naturalism may ultimately rely on and reinforce metaphysical naturalism, are certainly relevant and interesting, a discussion of such questions is beyond the scope of this short note. The important question here is whether the adoption of methodological naturalism in the social scientific study of religion does not fatally derail a genuinely scientific investigation of religious experience at the outset. That is to say, is it not the case that a methodologically naturalistic approach critically shifts research focus away from the actual phenomena of religious experience as meaningfully lived, the nature of which we most desire to know, and onto hypothetical, naturalistic constructs – the essential nature of which we already know (because we invented them, or other social scientists have used the same constructs to explain any number of other non-religious social phenomena) – and which are at best only very thin representations of what are presumed in the first place to be merely manifestations of religious phenomena and not the actual phenomena themselves? In short, then, the question being raised here is whether a methodological approach that requires us to shift our research attention away from the phenomena of actual interest is not, in actual fact, a distinctly non-empirical and, thus, non-scientific approach.

If the study of religion in psychology, sociology, and related disciplines genuinely aspires to scientific status, far more careful attention must be paid to the nature and implications of various methodological – as well as ontological – commitments that might be made, particularly if those commitments would seem to profoundly interfere with our ability to gain access to and adequately understand those phenomena which are of central investigatory interest (Gantt & Williams, 2008). While careful and coherent theoretical explanation is a central aim of scientific inquiry, it cannot come at the expense of the subject matter under study. As a number of thinkers have argued, science must “save the phenomena” (van Fraassen, 1991), or else it fails as legitimate science and devolves into little more than an ideological exercise. Indeed, as Larmer (2012) trenchantly reminds us: “Science’s task is to follow the evidence where it leads and to be open to the possibility that we may be surprised by what is found, not to insist that it either be ignored or forced into the Procrustean bed of a favored theory” (p. 148).

Maintaining an open methodological attitude, one in which saving the phenomena is of paramount importance, is vital to the genuinely scientific study of religion and religious experience. As the sociologist of religion Douglas Porpora (2006) trenchantly points out:

In any proper experience, the object of experience contributes something to the content of experience. The object, in other words, is part of what explains the content. Yet if objects of experience – whether in religion or science – are methodologically bracketed out of consideration, they are disallowed a priori from doing any explanatory work. The unavoidable implication is that there are no genuine experiences of anything so that the very category of experience dissolves. (pp. 58-59)

In other words, only by taking religious experience seriously can... a study of religion begin

In other words, only by taking religious experience seriously – that is, by treating the phenomena of religious experience on their own terms as fundamentally meaningful expressions of human social and moral experience, rather than seeking to reduce them to methodological or conceptual categories that may be quite alien to them – can a genuinely fruitful scientific study of religion begin. In order to adequately understand the behavior of people, in this case religious behavior, psychologists must first seek to sincerely and openly understand the experiential grounds of such behavior. “Minimally,” Porpora (2006) argues, “that means not to rule out tout court what people say they are experiencing” (p.59).

If social scientists genuinely wish to understand religion and religious life on their own terms – that is, in such a way as to “save the phenomena” of religious experience –they will need to seriously re-consider the widespread commitment to naturalism, as well as to singular species of theories and research practices that it naturalism requires. There are a number of reasons – scientific, philosophical, and theological – to question the adequacy of a methodologically naturalistic approach to the study of religious experience. At the very least, it seems clear that such an approach requires us as researchers to cast our scientific gaze elsewhere than toward the particular phenomena in which we are most often most interested. Rather than approaching the reality of religious experience as lived (i.e., as saturated phenomena), methodological naturalism entices us to aim at something else entirely – something readily measureable, presumably physical, and essentially reducible – and, thereby, miss our target completely. By opening ourselves up to possible alternative epistemological possibilities, alternatives that would permit us to engage religious experiences directly rather than forcing such experiences to fit into overly restrictive and distorting pre-selected conceptual boxes, perhaps then a social science of religion might begin to provide richer and more fertile accounts of religious experience – accounts that might accord more deeply with experiences of actual persons.