Methodological Atheism and Agnosticism in the Study of Religion

Academic study focuses on the world of cause and effect, on evidence and probability. And so there has been much discussion of methodological naturalism in the sciences, at least among those who are disgruntled by the fact that the sciences don’t make room for God or provide evidence for God.

The same discussion can be had in relation to the study of religion – and indeed, it might be wiser to speak of methodological naturalism in that context, too. But a recent discussion between Zeba Crook and Carl Stoneham prefers the terminology of atheism and agnosticism. Here is a quote from Stoneham to whet your appetite:

The problem with MAtheism is that it assumes that at least one plausible answer to the question cannot be an answer to the question but it cannot prove that said answer can’t be an answer. As such, it runs afoul of the very same thing that it accuses confessional scholars of doing: proceeding from a position it cannot support through an appeal to a certain set of “this-worldly” critical tools. To be perhaps a bit too flippant, MAtheism justifies itself because… MAtheism. I think MAgnosticism takes a more honest position insofar as it acknowledges that the religion scholar is not equipped with the tools to adjudicate that sort of truth claim, so it brackets them and looks for information in other areas. MAtheism doesn’t bracket the question, but answers it quite clearly.

‪I’m not suggesting that because the theological claim cannot be disproved, it *has* to be acknowledged as a possibility. Instead, I just want to be more careful about how we treat that claim. Methodologically, atheism does not seem to be the appropriate response, especially insofar as we might one day have the tools to adjudicate these claims. Methodological agnosticism seems to be more “future proof,” if you will.

Click through for the rest of a fascinating discussion about the appropriate starting stance for those engaged in the academic study of religion to adopt!

Religion Bulletin

Stay in touch! Like Religion Prof on Facebook:
  • Cecil Bagpuss

    Methodological naturalism is often a matter of faith. Consider the question of why humans have large brains. The answer, presumably, is that our large brains are the result of natural selection. But there is a problem with this. Big brains are expensive; so the benefit of a big brain would have to outweigh the cost. For most animals, including our closest relatives, the cost of having a very large brain seems to outweigh any potential benefit.

    This must mean that our ancestors were subject to very particular evolutionary forces. The trouble is that we can only speculate as to what these forces were. We can never hope to reconstruct what happened during this period. And we can never replay the tape of human evolution to see whether these forces do reliably produce large-brained creatures.

    So when we say that our large brains are the result of natural selection, we are engaging in an act of faith. There is some justification for this faith. If there is a choice between believing that our brains are the result of natural selection and believing that they are the result of divine intervention, there is good reason to opt for the former.

    Natural selection is a force that is known for certain to operate, whereas divine intervention is not. But if we define faith as believing in something that can never be tested, then it applies in this case.

    • arcseconds

      If you park your car somewhere, and you come back and find it’s gone, is it a faith-based claim to say someone’s taken it?

      I mean, you’re reliant on a well-attested principle — that large, solid, material objects don’t simply vanish — which may not be testable in this particular case (you may not ever get your car back or find out what happened to it).

      • James

        Theists often equivocate on the meaning of “faith” – sometimes it means something like “evidence-based trust” and when convenient it seems to be mean something like “trust even in the face of evidence to the contrary.” We have evidence-based reasons to know that the most common reason a car would disappear is that it has been stolen by other people; no faith required. That doesn’t mean there isn’t another possible explanation – perhaps it was towed, or a spouse who also has a key took it. The idea that it just dematerialized would be an extraordinary claim – i.e. highly implausible – and that seems to be what theists very often mean by faith.

    • Gerald Moore

      The existence of evolution is an observed fact. The Theory of evolution helps explain things in a coherent manner. Not everything we observe has yet been explained in detail and is why Evolution is still being studied. The Equivocation fallacy is when someone uses a word with two or more meanings as if it had only one. Religious faith is belief in something without evidence or reason. Faith in established science is high confidence based upon both evidence and reason and a mountain of observational data.

      You don’t get to re-define “faith” as acceptance of something that cannot be directly tested, especially since evolution can be directly tested and indirectly tested and is falsifiable unlike the god hypothesis.

      • Cecil Bagpuss

        Common descent is a fact but, as Richard Dawkins has pointed out, the role that natural selection plays in evolutionary change is much less securely established. Certainly, natural selection has been demonstrated, but it is very difficult to carry out large-scale studies of natural selection in the wild. It is also extremely difficult to isolate the factors which are involved in the selection. And in the case of which factors may have influenced the evolution of something like the human brain, we can only speculate.

        That isn’t to say that such speculation is unwarranted, but it is still speculation and can never be anything else.

        • Gerald Moore

          no, and I told you why already.

        • arcseconds

          It’s true that we can’t definitively rule out other processes, and in fact one other is known about already: random drift.

          But random drift won’t do for human brains, and you’ve pinpointed the reason yourself: the energy cost.

          Thinking it’s very probably the only viable mechanism that we know of that did the work seems the only reasonable conclusion to make under the circumstances.

          If ‘faith’ covers everything from thinking that the best available scientific explanation probably also applies in this case to believing in things contrary to all evidence due to ideological commitments, then doesn’t it just mean belief, reasonable or otherwise?

          In which case we can drop the word ‘faith’ from our vocabulary, as it is redundant and has awkward connotations :-)

          • Cecil Bagpuss

            I’m not trying to “rule in” other processes. Natural selection is the only explanation we have for the existence of something like the human brain. But building a brain isn’t like stealing a car. Stealing a car is a well-understood process – in general, that is; I don’t know much about it myself.

            Presumably, there is a series of mutations connecting Australopithecus and Homo sapiens. I imagine it would be difficult to list them in order and explain how each one provided enough of a selective advantage to become established in the gene pool.

            So there is faith in the sense of believing that we know what kind of explanation is the correct one even if we can never know the details.

          • arcseconds

            I’m talking about your situation, not the situation of humanity in general. That someone else knows about the finer details of stealing doesn’t mean that your epistemic situation is improved. And in fact of course such a person is presumably quite ignorant of the fact that your car is even missing, unless they happen to be the thief themselves.

            Also note that I didn’t say ‘stolen’, that is a further assumption, one that’s less warranted than ‘taken’. ‘Taken’ can cover: your partner driving it away; someone happening to have the same make, model and a key that happens to work in the lock and has driven it away by accident; being towed; being stolen.

            Even if you understood all of those processes in detail, you don’t know which of them actually occurred. Even if you’re pretty sure it was stolen, you don’t know exactly how they got in, where they drove it to, etc.

            So again there’s ‘faith’ in believing that you know what kind of explanation is the correct one, and you may never know the details.

            But it seems very strange to call thinking you have a general handle on a specific case on the basis of things you know from other, related circumstances ‘faith’.

            This seems to mean that all sorts of totally humdrum beliefs held for excellent reasons count as faith, which both seems to deviate from the usual understanding of the meaning of the word, and extend the term so far that it becomes kind of meaningless.

          • Cecil Bagpuss

            If there is an unexplained case of a missing car, we can reasonably assume that it follows the pattern of other cases of missing cars whose details are known. If there was a case of macroevolutionary change whose details were unknown, we might assume that it follows the pattern of cases whose details are known. The problem is that there aren’t any cases of macroevolutionary change whose details are known.

            That doesn’t mean that there isn’t a natural explanation or that we don’t have a very general idea of how it happened. But when our knowledge is limited to the extent that it is here, it seems that there is an element of faith involved.

          • arcseconds

            I don’t think that’s true at all. There are lots of well-documented cases of macroevolutionary change where there’s a clear route to selection, because the advantage is obvious, and the genes involved are well known.

            The classic example of the moths that changed from predominantly white to predominantly brown is a simple and well-known example.

            The evolution of eyes is something that has been studied in depth: there are extant examples of eyes of every stage of sophistication, the genes and proteins are well understood, and they have known origins in proteins used for other things, and there are pretty clear pathways by which every kind of eye could have its origin in simpler eyes.

            There are doubtless other, better examples where the exact mutations are known. Biologists are constantly tracing things to a mutation in the HOX gene about half-a-million years ago and suchlike.

            Of course you can insist that we don’t know all the details, down to exactly which individuals were struck by what cosmic rays or whatever, but this just seems obfuscatory: we always lack this level of detail in everything we claim to know: no-one knows exactly what synapses fired in the brain of the car-thief, for example, but that doesn’t stop us knowing what is going on at an appropriate level of detail.

            Plus, as you acknowledge yourself, it just seems like the only epistemically warranted thing to do.

            So I’m really not sure what you mean by ‘an element of faith’ when you’re applying it to believing in the only thing that is warranted to just the extent it is warranted. You seem to think this is saying something other than ‘believing just what the evidence suggests’, but what is that, exactly?

          • Cecil Bagpuss

            The case of the peppered moth is a good example of natural selection in action, but I wouldn’t regard a difference in a single gene as macroevolution.

            There are plausible intermediate forms of the eye but I don’t think there is a good understanding of the genetic changes that would turn the eye of something like the Nautilus into an octopus eye. But even if there was a good understanding of the changes, we would never be able to prove that natural selection could favour every step in the series.

            I think that we are justified in believing that natural selection is responsible for change of this kind but when we have to extrapolate to the extent that we do in this case, there is an element of faith.

          • arcseconds

            I have already dealt with these points.

            You seem to agree that in the case of a missing car, it’s not faith to say that someone took it.

            But we don’t know the details in this case, either. If the car is just gone, we would never be able to prove that a person drove it away.

            And in fact we don’t know the details of most human actions to the level that you seem to be requiring of evolution: that every step of the way we have to conclusively rule out unknown mechanisms and happenings at every step of the way before we can say “we know how this happened”. Maybe an event that was videotaped from beginning to end… but if we don’t have this, we can’t conclusively prove there weren’t aliens, gremlins, and psychic powers but just ordinary human actions at work everywhere there’s the slightest gap in our account.

            And once again, I’m going to ask you what you mean by ‘faith’ here. Are you saying that we don’t have epistemic warrant to think it’s natural selection? It doesn’t seem so. So what work is faith doing over epistemic warrant? and why do we need the term?

            is it faith to think there probably aren’t gremlins and aliens and mysterious processes lurking everywhere we don’t know absolutely everything about?

  • arcseconds

    OK, finally got through the article and the followup discussion. It’s all very interesting, and there’s some interesting side-issues and references to follow up.

    However, I find myself quite confused by all of this. Of course, this may be my own limitations, but at the moment I’m thinking it’s because the discussion is confused. The fact that Crook and Stoneham also don’t seem too clear about what MAthiesm and MAgnosticism are supports this contention.

    I’m not clear on any of the following:

    — what MAtheism is
    — what MAgnosticism is
    — what the distinction is

    — what they do over normal commitments to, say, believing that a well attested to process is more likely, ceteris paribus, than a totally novel process involving new theoretical entities.
    — what it would be to not employ one or other of these two
    — what subjects we are talking about

    It seems to me that we actually want different attitudes to the views of the subjects of our investigation depending on what we’re trying to investigate.

    For example, if we are doing cultural anthropology, I take it that a central goal of the discipline is to understand the culture from the inside, as it were, and of course to interpret the culture in terms that outsiders (the scholarly community) can understand (it does anthropology no good if the anthropologist ‘goes native’ completely and forgets being a Western scholar altogether).

    Here, surely, we want the scholar to put to one side (to a great extent at least) basically all their beliefs that differ from the members of the culture. If we want to understand their ideas of how rain forms, for example, insisting on contemporary meteorology is of little use. Insisting that they’re wrong and primitive or whatever is likely to be at best a distraction. Probably few people feel very strongly about meteorology, but many people do feel strongly about other things (sexual practices! but also morality more generally and, yes, religion) and there a researcher has more reason to train themselves to not let their own attitudes interfere with their research.

    The details of how to achieve this perhaps is up for discussion. So things like methodological ludism might be suggestions of this nature. Anthropologists don’t generally pretend to be anything other than what they are, and if it’s relevant they do sometimes discuss this in their work.

    MAtheism doesn’t really seem appropriate here. We could call this MAgnosticism if we wanted to, but

    In this case, I agree with Cook that no-one denies that people have religious experiences, and have beliefs about their experiences, and have a discourse relating to those experiences, and this is all that’s necessary to deal with in this kind of case.

    (I’m kind of having difficulty believing Cantrell is presenting Berger, or MAtheism fairly here: surely no-one thinks it’s appropriate for religious anthropologists to go around wagging fingers in their subjects’ faces and saying “your religious experiences are purely psychological! There is no reality behind them!” or to insist this in their work describing a culture.)

    Another extreme is where we don’t care at all about people’s beliefs, but are just trying to recover (what we take to be) facts. One example would be trying to ascertain the who/when/where of actual migrations. Here while we might pay attention to the culture’s narratives, those will be just one thing we factor alongside physical evidence, genetic evidence, narratives from other cultures, the plausibility of the migration account , etc. If we find the narration account is contradicted by all the other evidence, then we don’t think the events happened as that account recorded.

    Here we want to apply all our knowledge to the problem, because we’re after a new piece of knowledge that we can assent to.

    And our knowledge surely includes things like the fact that people don’t normally migrate carrying huge statues of gods, and even if this does happen from time to time, these
    statues do not whisper directions to their carriers

    This seems more than methodological agnosticism: it’s not that we’re not too sure about whether whispering statues are a good explanation for how a migration occurred: we know it is not.

    So is this methodological atheism? Well, I suppose in the sense that divine intervention of whatever sort is given a very low probability, we could call it that, but it’s surely no more methodologically atheist than looking for explanations of the weather in terms of high pressure regions and Kelvin-Helmholtz instabilities, rather than the activities of Thor or Ouranos.

    Someone who insists on some hitherto-unknown phenomenon that affects migration has to defend that appropriately, probably in an academic context, before people can appeal to that phenomenon without further proof of their own. Suggesting divine guidance for migrations needs to meet at least the same level of proof as proposing people migrate to enjoy the nicer scenery to a greater extent than previously thought, for example.

    There are things between these two extremes, for example if we want to know how a culture’s beliefs help them to survive, we need to understand their beliefs, but we also want to know how they really survive, and not how they think they survive.

    There’s another category where I suppose it’s appropriate to indulge yourself in your own religious beliefs, and that’s if you’re actually doing theology, or religious philosophy. Someone doing ‘self-anthropology’ might fall into this category, and this might potentially be where someone’s experience of the sacred or the ‘numinous’ might fit.

    So, horses for courses, in other words…