Introduction
According to an important minority view among philosophers of religion, religious commitment does not have to be based upon belief, but only on some lesser non-doxastic pro-attitude, like faith or hope.Footnote 1 In this way, the agnostic and the sceptic too can engage in religion. I am sympathetic to this view, and in this article, it is taken for granted. The issue I want to discuss concerns the proper focus of such non-doxastic religiosity. In the view of J. L. Schellenberg, one of the most prominent defenders of non-doxastic religion, its proper focus is ‘simple ultimism’, a highly abstract view containing only the idea that there is a metaphysically, soteriologically, and axiologically ultimate reality.Footnote 2 I contest Schellenberg's claim and argue that traditional religion is to be preferred over simple ultimism insofar as it admits of religious experience.
I am not the first to challenge Schellenberg and suggest that non-doxastic religion should have another focus than simple ultimism. However, the other critics I am aware of have suggested abstract views that are equally far removed from ordinary religiosity as is simple ultimism.Footnote 3 Furthermore, the force of religious experience on a non-doxastic approach to religion has not been properly considered. I purport to show how the problems usually associated with religious experience evaporate when turning from a doxastic to a non-doxastic approach. The prima facie advantage of my view over that of Schellenberg and other proponents of abstract alternatives is obvious; it lets the sceptic partake much more wholeheartedly in traditional religious life.
The discussion to follow is kept in general terms, not presupposing any specific account of non-doxastic religion. What is presupposed is only a ‘minimal view’, consisting of two defining features that commonly recur in almost all accounts of non-doxastic attitudes: that the subject desires or positively evaluates the object of the attitude, and that it considers it an epistemic possibility. A bit more formally, for S to have a non-doxastic pro-attitude towards some state of affairs p, it is necessary that the following two conditions are met:
ND1: S desires that p or judges that p is an overall good thing.
ND2: S believes p to be epistemically possible.
Non-doxastic attitudes commonly contain a strong voluntary element. Since voluntary acts require motivation, the desire component of ND1 is needed to make a non-doxastic attitude reasonable. However, since our desires and many of our judgements are not voluntary, ND1 also delimits the voluntariness.
ND2 requires that even though one does not believe the state of affairs in question to obtain, one still regards its actuality as an epistemic possibility (and nothing more, non-doxastic attitudes preclude certainty). Schellenberg opts for a very wide sense of epistemic possibility; he holds that something is epistemically possible if ‘it is neither known nor justifiedly believed to be false’ (Schellenberg (Reference Schellenberg2009), 8). While such a wide notion of epistemic possibility might be problematic in some contexts, it suffices for present purposes and I follow Schellenberg's use of the term if not explicitly stating otherwise.
In the next section Schellenberg's notion of simple ultimism is introduced, as is my objection that the view is too abstract to allow for contact with religious reality. ‘The experience-argument . . .’ contains my main argument for preferring traditional religion over Schellenberg's simple ultimism. The section labelled ‘The aim of openness’ discusses and rejects Schellenberg's central reason for preferring simple ultimism over traditional religion. In ‘Religious experience on a non-doxastic approach’ it is shown how common objections against basing a religious view on religious experience fall flat on a non-doxastic account. A brief conclusion sums up the discussion.
Simple ultimism and its problem
Schellenberg originally developed simple ultimism as a definition of religion worthy of philosophical investigation.Footnote 4 It tries to capture, in a very generic form, the ‘core’ of religion, which according to Schellenberg is the claim that there exists a ‘triple’ ultimate, a deepest level of reality that is metaphysically, axiologically, and soteriologically ultimate. The concept of the ultimate is supposed to be general enough not to exclude any of the more detailed views of ultimate reality found in the world's great religious traditions. Most importantly, it is so general that it could be read both as a personal God (as the ultimate is usually understood in Christianity, Judaism, and Islam) and as an impersonal ultimate reality (as in many forms of Hinduism or Buddhism). This allows a distinction between generic or ‘simple’ ultimism, the abstract core of religion, and different kinds of qualified ultimism, i.e. the actual religious traditions. According to Schellenberg, it is the former rather than the latter that is the proper focus of non-doxastic religion (Schellenberg (Reference Schellenberg2009), 15–17).
The basic problem with Schellenberg's view is that it is far from clear that engaging in ultimistic religion will have any chance of getting the subject in touch with transcendent reality, if there is one. This objection should not be confused with the objection often raised against Schellenberg that simple ultimism is too stripped down and devoid of details to make any real religiosity possible.Footnote 5 Neither should it be confused with any objection about ultimism's probability of being true. As often pointed out by Schellenberg, if any qualified ultimistic view is true so is simple ultimism, because every more detailed view of the ultimate entails it (ibid., 15). The problem is rather that even if simple ultimism turns out to be true, it seems doubtful at best that it will put us in any contact with ultimate reality. And what good is a vague but true religious view if it is too abstract to let us experience religious reality?Footnote 6
To spell out the objection in more detail, let us suppose that Christianity is the true religion. In this scenario, the claims of many Christians of living in a relationship with God would be true; in other words, Christians would have contact with ultimate reality. Since Christianity is a form of qualified ultimism and as such entails the truth of simple ultimism, simple ultimism would obviously be true as well in this thought experiment. However, the important question is not whether simple ultimism is true in this scenario, but rather if its adherent, the ‘ultimist’ so to speak, would be in contact with ultimate reality. Would his commitment, confined to the abstract thesis that there exists a triple ultimate, put the ultimist in contact with God? I very much doubt that. The ultimist would be no more in a relationship with God than someone who only holds the abstract view ‘there is something on the other side of my hedge’ would be in a relationship with his neighbour. Something similar obviously holds if any other qualified version of ultimism turns out to be true. If Theravada Buddhism is true, being an ultimist will not have taken you anywhere on the eightfold path to enlightenment; if Vaishnavism is true, the ultimist will not be a devoted follower of Lord Vishnu; or, to end with a drastic example, if some conservative form of Christianity or Islam is literally true, being an ultimist will not save you on judgement day.
Note that what is claimed is only that a view that is too abstract precludes the subject from being aligned with reality, not that a detailed view is enough to bring the subject into contact with reality on its own. Clearly, it is not. Once again, consider the case of the neighbour. I might have a very detailed idea of my neighbour, say, of her being an elderly lady widowed early in life etc. and still have no contact with her. To achieve contact, I must perform some kind of action (preferably pay my neighbour a visit). The same holds for the religious case: having a detailed view is not enough to be in contact with religious reality; one must also act on this view and participate in religious life. Only then, if (and this is a big ‘if’ indeed) the detailed view one is non-doxastically committed to happens to be true, one will be in contact with religious reality.
At this point, someone might object by questioning whether actions rather than cognitive contents enable contact with reality. If so, would not one's religious actions be enough to achieve contact with religious reality, even if one cognitively commits only to an abstract view like simple ultimism? The first point to notice here is that cognitive content is what guides action. Having a view that is too abstract will not enable one to act upon it properly, since it will fail to imply a concrete course of action. Presumably, the only course of action a highly abstract view will suggest is that one should try to get more information (which is exactly what Schellenberg envisages when he claims that religious investigation will be a big part of the ultimist's religiosity).
Second, while I very much doubt that one can get in touch with religious reality (or any reality for that matter) by only making the right actions, my position does not depend on this assumption. To prefer traditional religion over simple ultimism on a non-doxastic account, we only need to acknowledge that contact with reality will be substantially better if both one's actions and one's cognitive content are aligned with it, than if one is aligned through action alone. This weaker claim seems obvious enough; in everyday contexts, action and mental content seldom come apart, but people who do lack the proper cognitive content associated with what they do tend to strike us as delusional. Consider Don Quixote fighting against windmills he believes to be giants. Surely, Quixote would have been more in touch with reality if he had started thinking ‘I am demolishing windmills’ instead of ‘I am fighting giants’ (while making a Schellenbergian move to a highly abstract claim like ‘I am doing something’ would not improve Quixote's contact with reality to any greater extent).
Another important question is: what level of detail is required to make alignment with reality possible? As a minimum, I find it reasonable to suppose that one's view must be detailed enough to postulate concrete entities of the right kind, like God or the impersonal Brahman.Footnote 7 For example, theism would constitute such a minimally detailed view.
Why is it so important to get the basics right? Consider an analogy. If you have to move around blindfolded and have no idea where you are, getting abstract but true information like ‘you are somewhere’ or ‘you are in a landscape’ will do very little to help you out. But if you get to know roughly in what kind of place you are, like ‘you are in a building’ or ‘you are at the edge of a steep cliff’, that will be sufficient to put you in some basic contact with reality and make possible the right kind of action.
Is it then enough to postulate the existence of a concrete ultimate, like God or the impersonal Brahman, or should we settle for an even more detailed view? The first thing to note is that a more detailed view is not always better. One might well reach a kind of ‘information threshold’ where additional information has no practical use, and therefore does little to improve contact with reality. If you move blindfolded in the jungle, knowing that there are lots of fallen trees ahead might be helpful, but it will make no difference to know that you are being watched by a shy monkey.
The second thing we need to acknowledge is that when it comes to religious reality (if there is one), we have simply no idea of the level of detail we need to get right to be properly aligned. I have suggested that, at the very least, we need to postulate the right kind of basic object, but it is far from certain that such a postulation is enough. So how are we to proceed? Since we are ignorant of the relevant threshold in information, of how ‘right’ of the details we must be to achieve contact with ultimate reality, I suggest we should always prefer a full-blown, fully detailed view over any abstraction. If it turns out that we only need to get the basics right, committing to a view more detailed than required could hardly be harmful. But if a more detailed view is required, committing to only the basics will not be enough. To see this, consider the following example.
Let us suppose that Shia Islam is the true view of the world. If a commitment to theism is enough for contact with reality on such a scenario, committing to bare theism would bear no advantage over a commitment to Shia Islam; both the abstract view and the more detailed get the needed basics right. However, it might also be the case that we need to be right about more than just the basic kind of entity (God in this case) to get in touch with ultimate reality. Perhaps I need to commit to Islam to be aligned with the ultimate on the scenario envisaged here, or perhaps only Shia Islam will do. Then bare theism would be too abstract a view to admit contact.
Since a detailed view (if correct) permits alignment regardless of how ‘high’ the threshold of information is, while an abstract view requires a low threshold, we should choose a fully detailed view if we want to maximize our chances of getting in touch with religious reality.Footnote 8
The experience-argument: why traditional religion is to be preferred
If the reasoning in the previous section is correct, we should prefer as an object for non-doxastic religious commitment a view that, if it turns out to be true, would have put us in contact with religious reality. But how do we know which kind of religious view to prefer, if that is our goal? As a preliminary, I would say that reports of religious experience are the best indicator we have.Footnote 9 In a tradition where veridical religious experience is thought to occur, it is part of the view that contact with religious reality is not only possible in principle, but a recurrent phenomenon. Therefore, it seems reasonable to suppose that if such a view turns out to be true, the experiences associated with it will be veridical as well.
A tradition where religious experience is thought to be frequent and considered important is in a stronger position in this regard than a tradition viewing it as a marginal phenomenon. It might well be that some religious traditions are superior to others in this regard, but it is beyond the scope of this article to try to pick out any single most ideal tradition. I will only make the modest claim that many traditional forms of religion, including the major world religions dominating the contemporary religious scene, are sufficiently rich on religious experience that it seems reasonable to suppose that if one of these traditions is true, it will put its adherents in contact with religious reality. Modest as it is, this claim is still significant for present purposes since, as we saw in the previous section, we have no reason to believe that simple ultimism would be able to put its adherents in contact with religious reality, even if the view turns out to be true.
According to Schellenberg, ultimistic religion is superior to non-ultimistic.Footnote 10 It is beyond the scope of this article to challenge that claim, and therefore I will focus solely on ultimistic religion. Since the major world religions are all different versions of ‘qualified ultimism’ according to Schellenberg, they will be my focus. I will use the label ‘traditional forms of qualified ultimism’ (or traditional religion for short) to distinguish the world religions from other forms of qualified ultimism (historical, actual, or possible). I only argue that the major world religions are better objects for non-doxastic religion than simple ultimism, leaving open the possibility that some other form of religion might ultimately be the superior choice.
With these preliminaries in place, the argument for preferring traditional religion over simple ultimism runs as follows:
Premise 1: When choosing an object for non-doxastic religious commitment, we should ceteris paribus prefer a choice that is specific and contains religious experience.
Premise 2: Simple ultimism is not specific, and does not contain religious experience.
Premise 3: Traditional forms of qualified ultimism are both specific and contain religious experience.
Conclusion: When choosing an object for non-doxastic religious commitment, we should prefer a traditional form of qualified ultimism over simple ultimism.
The argument is obviously valid, and only the first premise seems to include any controversial assumptions. The claims of premise 2 and 3, that traditional religion is both specific and contains religious experience while simple ultimism lacks both these features, should be unproblematic. The fact that simple ultimism is void of details while the views of ultimate reality in traditional religion are highly detailed is as important in Schellenberg's argument in favour of simple ultimism (see the openness aim in the upcoming section) as it is in my argument for traditional religion. The point that traditional religion contains religious experience while simple ultimism does not should be equally unproblematic. Claims about contact with a transcendent, religious reality are widespread in the major world religions. Simple ultimism, being a product of philosophical reasoning only, comes without any such claims.Footnote 11
It might be objected that since simple ultimism entails all other forms of ultimism, should not all kinds of ultimistic experience count in favour of simple ultimism? The reason this should be denied is that many different claims about the ultimate are incompatible (see the discussion of the problem of religious diversity in the section ‘Religious experience on a non-doxastic approach’), and incompatible claims about experience could not on any plausible reading be reinterpreted as claims about experiencing some abstract ‘common denominator’. If A and B look inside a freight container, and A reports seeing low-quality clothes inside, and B reports seeing an expensive, high-class automobile, we would not hold that these experiential claims taken together support the view that there is ‘something’ in the container. Rather, we would conclude that at least one report is entirely unreliable, and we would not want to base anything, not even a highly abstract view, on the unreliable informant's claim.
This leaves us with the first premise, which is not uncontroversial. The claim that we should prefer a specific view over an abstract is defended from objections in the previous section, and I will not dwell on it further. However, the premise also relies on claims about religious experience. Objections against the reliability of religious experience are common, and routinely invoked against arguments using religious experience as evidence for the existence of God. I will argue that although these objections are forceful and successful in blocking belief based on religious experience, they do not rule out the epistemic possibility of the experiences being veridical, and this epistemic possibility is all we need on a non-doxastic approach. These claims are elaborated in the section ‘Religious experience on a non-doxastic approach’.
Another way of challenging premise 1 is to advance some other objective that should guide our choice of non-doxastic religious object. One such objective is the aim of openness, which Schellenberg uses to promote simple ultimism over any qualified version. In the next section, the openness aim is assessed and rejected.
The aim of openness
Schellenberg's claim that simple ultimism is to be preferred over traditional religion relies heavily upon what he labels ‘the aim of openness’ (Schellenberg (Reference Schellenberg2009), 17–23). To understand the aim, one has to be familiar with at least two other major strands of Schellenberg's philosophy. The first is the claim that no religious belief or disbelief is justified. Belief in any of the grand claims associated with present-day religion is unwarranted, as is belief in naturalism, and anyone with a clear view of our current epistemic situation regarding religious matters should adopt scepticism.Footnote 12 The second claim is that we are at the very beginning of intellectual and cultural evolution. We are ‘early humans’ at the very beginning of civilization and of intellectual enquiry, and we should realize that there are many millions of years of evolution ahead. Even if religious belief is unjustified today, we should be open to the possibility that future discoveries in the field of religion (by us or by some other more spiritually advanced future species) might lead to very different conclusions indeed.Footnote 13
Woven together, these strands provide the rationale for the openness aim, which states that we should prefer simple ultimism over any qualified version because it makes us maximally open to any future discoveries in the field of religion. If in doubt over the reasonableness of this argument, remember that all qualified versions of ultimism entail simple ultimism, which will be true as long as any version of qualified ultimism is true (ibid., 17–29).Footnote 14
The openness aim conflicts directly with premise 1, since it holds that epistemic openness rather than the possibility of contact with religious reality should guide our choice of object for non-doxastic religion. However, the aim of openness seems highly problematic in a non-doxastic context. It seems to rest on a confusion of what is desirable for an object of belief with what makes it a reasonable object for non-doxastic attitudes. The aim of openness has truth as its main goal, and it would have us choose a view based on its probability of being true. When it comes to belief-formation, where considerations of truth are standardly given the highest priority, the openness aim seems reasonable enough.Footnote 15 However, making truth-considerations only in a non-doxastic context is bound to get things wrong. In such a setting, there are factors other than minimizing falsehoods and maximizing truths that are of equal or greater importance. These are the factors captured by ND1. Unlike when it comes to belief, what we actually desire or think would be a good thing matters for which non-doxastic attitude it is appropriate for us to have.Footnote 16
First, consider the example, much discussed by Schellenberg, of a runner who has faith that ‘I will perform well in the race’ even though the evidence tells him that his physical condition is less than optimal, and in no way warrants any corresponding belief (Schellenberg (Reference Schellenberg2005), 129–130). Surely the runner has forgotten to consider the aim of openness, when he directs his faith to such an easily falsifiable proposition! Rather, he should have faith that ‘I will perform well in this race or some other race’ or perhaps that ‘someone will perform well in the race’.
Second, consider the example of the presumed war widow. A company of a hundred men has been massacred in the war, but there are unconfirmed rumours that a single soldier might have survived. Not even the military commanders know whether these rumours are true or not. The wife of one of the soldiers clings to the desperate hope that the rumours are true and that it is her husband who has survived. Obviously, the wife has forgotten about the aim of openness! Why hope the easily falsifiable proposition ‘my husband has survived’, when there are other propositions around, much more likely to be true, such as ‘there is a survivor’?
As the examples make clear, on many occasions there will be a conflict between the openness aim and the important emotional aspect of the non-doxastic attitudes, and changing the object of our desire to fulfil the openness aim simply does not make any sense. The emotional and evaluative aspects of the non-doxastic attitudes, as expressed by ND1, are all-important and trump other considerations. Presumably, not having any non-doxastic attitude at all would be preferable to having the pointless attitudes suggested in the examples. The conflict between the openness aim and the emotional aspect is good reason to reject the openness aim.
This in no way implies that truth-considerations are irrelevant on a non-doxastic approach. The notion of epistemic possibility demands that truth be not ruled out, and this possibility of truth is vital for every non-doxastic attitude. What, if not the truth of the state of affairs in question, is the hoping subject hoping for or the faithful subject having faith in? However, when the possibility of truth is established, the emotional or evaluative factors of ND1 will have priority.
ND1 is not only a tie-breaker when several options seem equally likely. Even if we have evidence that some other option than the one we desire is the more likely, it would still not be rational to change our non-doxastic commitment.Footnote 17 Consider again the presumed war widow. Even if there was another unconfirmed rumour circulating that the survivor was a bachelor, she could still cling to her hope that the survivor is her husband. Only by learning for certain either that the first rumour is false or that the second is true does reason demand her to give up her hope.
The examples concern everyday versions of faith and hope, and their relevance for the religious case could perhaps be questioned. But why should we suppose any difference here? Are not the emotions involved in religion every bit as strong and important as those in the everyday cases? If dreary from the sufferings of this world I hope for the release of nirvana, or if the only light in my misery is my faith that I am loved by a personal God, why suppose that my emotional needs are in any way adequately addressed if instead I only hope or have faith that ‘there is a soteriologically ultimate reality’? Doing so would be every bit as awkward as only hoping that ‘some female survived the accident’ when what you really hope for is the safe return of your wife.
A defender of Schellenberg's view might complain that my examples concern a subject already having some pro-attitude towards a qualified version of ultimism, and that Schellenberg primarily envisages simple ultimism and non-doxastic religion as being for the sceptic who has lost his belief in traditional qualified ultimism or never had one (Schellenberg (Reference Schellenberg2009), 22–23). But why suppose that the sceptic who lacks religious belief also lacks desire or that he does not evaluate some qualified ultimistic claim positively? To be honest, I find it unlikely that a sceptic who does not already evaluate some qualified ultimistic claim positively would develop a positive attitude towards ultimism. However, it is not vital for present purposes to deny the possibility that some agnostic might find simple ultimism attractive enough to make a non-doxastic commitment, even though completely unimpressed by present-day religion. What matters is that such a sceptic would choose simple ultimism because he desires or positively evaluates its truth, not because of considerations of openness. It is no more reasonable to suggest that anyone should choose a view other than the one they desire than it is reasonable to suggest that the presumed widow should have hope that ‘there is a survivor’. Even if we recognize the greater probability of simple ultimism being true, this in no ways implies that we would desire or positively evaluate its truth higher than any qualified version.
A defender of Schellenberg's view might also present the more general worry that by rejecting the openness aim and arguing for a non-doxastic commitment to traditional religion, I thereby also reject the possibility of progress in religion. Is my proposal too conservative compared to Schellenberg's original account? While granting that my proposal might be less revolutionary than Schellenberg's, the turn from a believing to a non-doxastic commitment is still a major one; one that presumably will make further religious enquiry easier, since it frees the devout of dogmatically held beliefs. A non-doxastic commitment is more compatible with an investigative stance than a commitment based on belief.Footnote 18
I also think that the reliance on religious experience is a strength of my account when it comes to progress in religion. If we pay attention to how actual religious development tends to happen, claims of religious experience and alleged revelation often play an important role (together with more mundane factors). The religious landscape would surely not be what it is today, if it had not been for the work of prophets and mystics and others who claim to communicate with or experience the ultimate. If we commit non-doxastically to traditional religion, religious experience can presumably continue to play this role.
Religious experience on a non-doxastic approach
My argument for preferring traditional religion relies heavily on religious experience. In this section I address two major objections commonly raised against using religious experience to argue for religious claims: the problem of religious diversity, and the problem of naturalistic explanation.Footnote 19 My conclusion is that these objections become irrelevant on a non-doxastic approach.
The problem of religious diversity is often considered the main argument against relying on religious experience. It is based upon the observation that religious experiences from different traditions often turn out to be incompatible. And how could religious experience be a reliable indicator of truth when it produces different and incompatible results? In discussions of this problem, the emphasis is often on the incompatibility between theistic religious experiences, where ultimate reality is experienced as personal, and experiences from some non-theistic traditions, where ultimate reality is experienced as impersonal.
The discussion of religious diversity standardly centres on the question whether the problem is serious enough to make belief in the incompatible religious views irrational. What is of interest for present purposes is rather if the problem is serious enough to make the corresponding non-doxastic commitment unreasonable. As a minimum requirement for belief to be warranted, the probability of the view in question to be true must be greater than the probability of it being false.Footnote 20 Obviously, the presence of undefeated alternatives might very well make it the case that this minimum requirement fails to be satisfied. However, for a non-doxastic commitment to be warranted, all that is needed is that the view in question remains an epistemic possibility, and that is certainly possible even if there are numerous undefeated alternatives.
Could the problem of religious diversity ever be a problem on a non-doxastic approach? Can the number of undefeated alternatives be so huge that outright disbelief rather than a suspension of judgement is the only reasonable response? I am doubtful about this. It seems strange to suppose that something could cease to be an epistemic possibility simply because the number of incompatible rival epistemic possibilities gets sufficiently high. Consider the following example. When standing on the shore of a dark lake into which people have been known to dump the weirdest things, a huge number of epistemically possible scenarios of what lies beneath the surface present themselves. Since these scenarios are all equally unlikely, I cannot see how their quantity would warrant disbelief rather than a sceptical suspension of judgement. Otherwise, we would be forced to conclude that outright disbelief can be justified even in the absence of any evidence, which seems hard to swallow. But if no disbelief is justified, the different scenarios concerning the lake's content remain epistemic possibilities.
However, even if we grant, for the sake of discussion, that a huge number of undefeated alternatives might prevent us from attributing epistemic possibility, there is no reason to suppose that the number of undefeated alternatives is that high in the religious case. The discussion of the problem of religious diversity commonly supposes that there are at least three basic views of ultimate reality around: that there is a personal ultimate, that there is an impersonal ultimate, and that there is no ultimate at all. Presumably there are more. For example, different experiences of the self in relation to the impersonal ultimate in different forms of Buddhism and Hinduism seem to suggest that these traditions in fact posit different ultimate objects (experiencing that the atman is identical with the Brahman, that one is in fact part of ultimate reality, seems to presuppose a view of the ultimate that is incompatible with the Buddhist experience of oneself as a non-self).Footnote 21 However, trying to spell out exactly how many different basic and incompatible views of the ultimate there are in contemporary religion is a task well beyond the scope of this article, requiring a deep familiarity with the actual religious traditions. My point is rather that every further differentiation beyond the three basic views is significant, and is also bound to be a matter of some controversy. I find it unlikely indeed that it could be established that contemporary religion includes so many different views of ultimate reality that it could no longer be maintained that these views constitute epistemic possibilities. But that is exactly what needs to be established if the problem of religious diversity is to have any impact on a non-doxastic approach.
But even if it could be established that contemporary religion contains several hundred different views about the ultimate, all backed by experience, it could still be argued that this would only be a problem for the minor views, and not for the views embraced by billions. We standardly assume that when it comes to experience (unlike reasoning etc.), majority does matter. If there was an accident and twenty witnesses claim they saw the victim fall in front of the car while only one claims he was pushed, we do not really suspect foul play. I see no reason for treating religious experience any differently here. So, if a huge number of undefeated alternatives could prevent each other from being epistemic possibilities, and if it could be shown that contemporary religion contains a sufficiently high number of views about ultimate reality for this to be relevant (two very big ‘ifs’), that would only deprive the minor views of their epistemic possibility. The major traditions would still have to be considered epistemic possibilities.
The other major problem for religious experience is the problem of naturalistic explanation. For the sake of clarity, we might distinguish between a strong and a weak form of the argument. The strong version tries to conclude that religious experience is not caused by any ultimate reality, and that it has natural causes only. The weak version simply notes that there is a natural explanation available and questions the reasonableness of adding extra supernatural causes to the picture.
Like other contemporary philosophers arguing in the same vein, Schellenberg promotes a version of the argument that relies on processes in the brain. He draws attention to new discoveries of how something ‘phenomenologically indistinguishable’ from mystical experience can be produced artificially by stimulating certain areas of the brain (the same ones responsible for epileptic seizure). His argument is of the strong kind, since he holds that these findings indicate the non-veridicality of religious experience (Schellenberg (Reference Schellenberg2007), 188).
There are good reasons to resist Schellenberg's conclusion. The possibility of artificially inducing a kind of experience is completely irrelevant to the question whether these experiences are veridical when occurring naturally. Suppose that the taste of salt could be artificially induced by stimulating the right parts of the brain. Would we therefore say that all experiences of the taste of salt are non-veridical? Surely not! And the same holds for religious experiences. The contemporary ‘brain-version’ of the argument is perhaps the least compelling of all, since presumably all experiences can be artificially induced by the appropriate kinds of brain stimulation. If religious experiences were to be rejected for being caused by brain processes, so would all our experiences.
To make matters even worse for Schellenberg's argument, any strong version is problematic even when it does not rely on processes in the brain. To conclude with a reasonably high probability that some religious experience is non-veridical, one needs to show convincingly that the ultimate is nowhere in the causal chain leading up to that experience. Supernatural causes are not supposed to be empirical and they could coincide with natural causes in such a way that a natural explanation might look complete, when in fact supernatural causes accompanied every natural cause in it. Therefore, proving the non-existence of supernatural causes seems no less challenging than proving the non-existence of the ultimate itself.Footnote 22
These considerations suggest that any compelling version of the argument should be formulated in weak terms and that it should rely on natural explanations other than brain processes. To begin with the latter desideratum, there are many explanations available for religious experience that do not invoke brain processes, mainly from psychology and the social sciences. Explanations include deprivation, cognitive need, compensation, and society projection, to mention a few.Footnote 23
In a weak argument, it is not claimed that these explanations are enough to rule out the veridicality of religious experience, only that they are perfectly good explanations in their own right. And if there are seemingly complete and satisfying natural explanations available, why add supernatural causes?
When evaluating the weak argument on a non-doxastic approach, it is important to note that the goal of the weak version is only to establish and make reasonable the possibility that religious experience is non-veridical. This might well be enough to make belief unwarranted; if a reductive natural explanation is at least as likely as a supernatural one, we should suspend our judgement. However, it is not enough to make a non-doxastic attitude unwarranted, since it in no way denies the epistemic possibility of religious experience being veridical. To the contrary, since the whole point is to convince us that we should not adopt the veridical reading, the weak argument presupposes that the veridicality of religious experience is a relevant epistemic possibility. If such veridicality was not an epistemic possibility, why would anyone bother to offer such an argument in the first place? With these considerations in mind, it seems safe to suppose that the veridicality of religious experience continues to be an epistemic possibility even if a weak version of the argument is successful. Therefore, weak arguments have no force on a non-doxastic approach.
Conclusion
I have suggested that non-doxastic religious attitudes should be directed at traditional religion rather than at simple ultimism. The main motivation for this suggestion is that the prevalence of religious experience in traditional religion makes it an epistemic possibility that some traditional religion might put its adherents in contact with religious reality. Even though the truth of simple ultimism is entailed by traditional religion, the former seems too abstract to make such alignment possible.
A non-doxastic approach is fully capable of handling the main problems associated with religious experience: the problems of religious diversity and natural explanations. The key notion here is that of epistemic possibility. While the arguments are strong enough to block belief, they are not strong enough to stop the veridicality of religious experience from being epistemically possible.
Schellenberg's claim that the aim of openness should guide our choice between traditional religion and simple ultimism has been rejected as being an epistemic consideration with little or no force on a non-doxastic approach. At this point, a defender of simple ultimism would need to elaborate some new aims to support simple ultimism, aims important enough to make simple ultimism a viable choice even though it presumably does not admit contact with ultimate reality. It is far from clear that there are any such aims around.
