Scientism and the value of scientific evidence for religious belief

Abstract

This article presents a novel argument against an application of evidential scientism to religious belief. In particular, our target is those arguments at whose core lies the claim that it ought to be the case that, if one holds religious beliefs, then those beliefs are based on the best scientific evidence. Moreover, rather than focussing on the philosophical puzzles that usually fall within the purview of philosophers of religion, we are interested in the mundane beliefs of ordinary believers about their everyday interactions with God. Our argument combines recent work on epistemic partiality in close personal relationships with insights from analytic theology on the personal nature of believer's relationships with God. We argue that it's inappropriate for believers who take themselves to have a personal relationship with God to base their religious beliefs about God's nature on scientific evidence. In particular, it's precisely because these believers are in a personal relationship with God that it's sometimes inappropriate for them to form their beliefs about God's nature on the basis of scientific evidence.

Introduction

This article presents a novel argument against an application of evidential scientism to religious belief. In particular, our target is those arguments at whose core lies the claim that it ought to be the case that, if one holds religious beliefs, then those beliefs are based on the best scientific evidence. Moreover, rather than focusing on the philosophical puzzles that usually fall within the purview of philosophers of religion, we're interested in the mundane beliefs of ordinary believers about their everyday interactions with God.

What, then, is evidential scientism? Evidential scientism follows from premises which, taken in isolation, may be plausible to some:

1. 1. Contemporary empirical science is the gold standard for evidence.

2. 2. It ought to be the case that: if one believes that p, then one's belief is based on the best available evidence.

3. 3. It ought to be the case that: if one believes that p, then one's belief is based on the scientific evidence (1, 2).

In this article, we will present an original argument against the application of this way of thinking to everyday religious belief. We will argue that it's not the case that it ought to be the case that: if one holds religious beliefs, then one's religious beliefs are based on the scientific evidence. Our argument combines recent work on epistemic partiality in close personal relationships with insights from analytic theology on the personal nature of believer's relationships with God.

Philosophers of religion have tended to focus on the questions of (1) whether the belief that God exists is reasonable, justified, or enjoys some other positive epistemic status, and (2) whether beliefs about God's nature are reasonable, justified, or enjoy some other positive epistemic status.¹ We needn't answer these questions the same way. Here, focusing on reasonableness, we argue that it's inappropriate for believers who take themselves to have a personal relationship with God to base their religious beliefs about God's nature on scientific evidence. In particular, it's precisely because these believers are in a personal relationship with God that it's sometimes inappropriate for them to form their beliefs about God's nature on the basis of scientific evidence. For instance, if the believer holds the belief that God will protect them because of their relationship with God, then it would be appropriate for them not to seek scientific evidence in support of this claim, regardless of the probative value that such evidence might (or might not) have, or so we argue in this article.

What is scientism?

Many readers will be familiar with the term ‘scientism’. In the last decade or so, a handful of prominent public intellectuals and science communicators have been criticized on the grounds that certain of their proclamations betray a scientistic outlook. The proclamations that provoke accusations of scientism tend to be criticisms of religion and philosophy based on the failure of these disciplines to stand up to scientific scrutiny; the tenor of the retort is that these criticisms are – in one way or another – over-reliant on the (undeniable) virtues of the scientific method. Now, familiar as we may be with the term, it's not clear exactly what it refers to. There is no one banner under which all those accused of scientism march; there is no single methodological doctrine or ethical code to which they all subscribe.

Before proceeding, then, it's important to specify what we understand by the term ‘scientism’. Now, we are far from the first to comment on scientism. According to Alexander Rosenberg – who endorses the position – scientism ‘is the conviction that the methods of science are the only reliable ways to secure knowledge of anything’ (Rosenberg 2011, 6). In a 2012 article, Susan Haack – who is critical of these ways of thinking – has identified the following six signs of scientism:

1. 1. Using the words ‘science’, ‘scientific’, ‘scientifically’, ‘scientist’, etc., honorifically, as generic terms of epistemic praise.

2. 2. Adopting the manners, the trappings, the technical terminology, etc., of the sciences, irrespective of their real usefulness.

3. 3. A preoccupation with demarcation, that is, with drawing a sharp line between genuine science, the real thing, and ‘pseudo-scientific’ imposters.

4. 4. A corresponding preoccupation with identifying the ‘scientific method’, presumed to explain how the sciences have been so successful.

5. 5. Looking to the sciences for answers to questions beyond their scope.

6. 6. Denying or denigrating the legitimacy or the worth of other kinds of inquiry besides the scientific, or the value of human activities other than inquiry, such as poetry or art. (Haack 2012, 77–78)

These six signs of scientism aren't intended to be understood as a list of necessary or sufficient conditions for scientism – although we suspect that having one of these features is sufficient to make one vulnerable to accusations of the same. Some writers have pointed out that scientism is widespread in society nowadays (White 2013; Williams and Robinson 2015), although it's often implicitly adopted and is often not explicitly defended or argued for. That being said, some bestselling and highly regarded authors and popularisers of science seem to advocate for a scientistic worldview (Dawkins 2006; Harris 2010). Moreover, some philosophers have done so, albeit defending much more sophisticated positions than the one that we are discussing here.²

Some scientists have argued that the empirical sciences are sufficiently well equipped for answering ethical questions. Massimo Pigliucci (2018) offers the example of a prominent study in the American Psychologist, in which it was argued that the scientific evidence supports the hypothesis that a ratio of 2.9013 positive to negative experiences was necessary for human flourishing (Fredrickson and Losada 2005). These claims were rebutted in a subsequent article (Brown et al. 2013). The strongest objections emphasized the methodological weakness of applying the mathematics of fluid dynamics to questions of human flourishing. There are also legitimate philosophical objections that can be made to a theory of well-being that describes well-being as a function of positive experiences. This serves as an amusing but informative illustration of scientism in practice. We take issue with a specific version of scientism that naturally follows from Haack's sixth sign of scientism: ‘Denying or denigrating the legitimacy or the worth of other kinds of inquiry besides the scientific.’ We will discuss this and how it's applied to religious belief in the following section.

‘Evidential scientism’ and epistemological criticism of religious belief

This article presents a novel argument against the use of evidential scientism to mount epistemological criticisms of everyday religious belief. So, how has this kind of thinking been used to mount such criticisms? Most prominently, perhaps, these kinds of criticisms have been advanced by Richard Dawkins in The God Delusion (2006), Sam Harris in The End of Faith (2004), and Christopher Hitchens in God Is Not Great (2007). The theme that unites these works is that evidence – in particular, scientific evidence – excludes religious belief on pain of epistemic impropriety, and that ordinary religious believers are therefore deserving of criticism for their religious beliefs. For a clear statement of the scientistic mindset, consider the following words by Steven Pinker, writing for the periodical New Republic:

To begin with, the findings of science entail that the belief systems of all the world's traditional religions and cultures – their theories of the origins of life, humans, and societies – are factually mistaken. We know, but our ancestors did not, that humans belong to a single species of African primate that developed agriculture, government, and writing late in its history. We know that our species is a tiny twig of a genealogical tree that embraces all living things and that emerged from prebiotic chemicals almost four billion years ago . . . We know that the laws governing the physical world (including accidents, disease, and other misfortunes) have no goals that pertain to human well-being. There is no such thing as fate, providence, karma, spells, curses, augury, divine retribution, or answered prayers – though the discrepancy between the laws of probability and the workings of cognition may explain why people believe there are. (Pinker 2013, para. 15)

It's important to note that the criticisms that interest us here aren't typically made within the philosophy and theology departments of research universities. Rather, these kinds of criticisms have mainly been made by public intellectuals and science communicators through more accessible media such as TV documentaries, podcasts, and books aimed at a general readership. Even though some of these figures do have backgrounds in the academy, their comments about religion aren't intended for an academic audience. It's not clear to us whether this should require methodological changes on our part. As far as we are concerned, we can take these criticisms at face-value unless we have a good reason to think otherwise, and we are yet to come across a good reason to do so. Furthermore, given the ubiquity of such criticisms and the influence they presumably have on the public understanding of science (and philosophy and religion), we dare say that we ought to take them at face value.

What exactly is the logic of the application of evidential scientism to religious belief, then? On our view, the scientistic criticism of religious belief that interests us can be formalized in the form of the following argument:

1. 1. Contemporary empirical science is the gold standard for evidence.

2. 2. It ought to be the case that: if one believes that p, then one's belief is based on the best available evidence.

3. 3. It ought to be the case that: if one believes that p, then one's belief is based on the scientific evidence (from 1, 2).

4. 4. It ought to be the case that if one holds religious beliefs, then those beliefs are based on the best available evidence.

5. 5. Conclusion: it ought to be the case that, if one holds religious beliefs, then those beliefs are based on the best available scientific evidence (from 3, 4).

The first premise is a strong claim about the probative value of a certain kind of evidence. The thought is that scientific evidence is the best kind of evidence with respect to its probative value. For any inquiry whether p, scientific evidence is most likely to get you the correct answer, when compared with other kinds of evidence, such as introspection. This may be true, for certain subject matters. If you want to find out how quickly the ice caps are melting or how to create sustainable energy from nuclear fusion, the scientific method will serve you better than introspection, casual observation, or informal deliberation ever could. Of course, this premise is problematic. We will talk more about its weaknesses in the following section.

The second premise, while it closely resembles the evidentialist norm proposed by Connee and Feldman (2004), is better understood as an epistemic norm about inquiry.³ As such, it tells us how we ought to conduct our inquiries. Admittedly, we must concede, it does not necessarily tell us what epistemic/normative status the resulting beliefs will have – that would depend on one's other epistemological commitments. While, for instance, adherence to this norm may be sufficient for justified belief within an evidentialist framework, it might not be sufficient for justified belief within a virtue epistemological framework. Nevertheless, it might fit within such a framework, if it's the case that basing your beliefs on the best available evidence is epistemically virtuous. In any case, we will assume that epistemic virtues do indeed play a role in determining the epistemic status of belief (and other doxastic states). Moreover, this norm is appealing to the extent that it captures the prima facie plausible notion that optimizing the reasons we have for our beliefs is a good thing, and that only epistemic reasons are suitable for that task.

The third premise follows from the first and second premises, yielding a more specific epistemic norm about inquiry. The thought is that if scientific evidence is the best evidence, and our beliefs ought to be based on the best evidence, then our beliefs ought to be based on the scientific evidence. This notion certainly seems to guide the thinking of those who make the kinds of epistemological criticisms of religion mentioned above. (To be fair, there might be other arguments to support this conclusion other than that which we are discussing here. But since we are most interested in the legitimacy of applying this norm to religious belief, we won't be addressing those arguments here.)

The fourth and fifth premises can be taken together. They represent the application of the above to the case of religious belief. What we are left with is a norm that says that it ought to be the case that: if one holds religious beliefs, then those beliefs are based on the best available scientific evidence. This is the kind of thinking that leads to the position that we seek to reject in the rest of this article. There is a matter of scope that is worth mentioning at this point. The pronoun ‘one’, unless otherwise specified, has a wide scope. A legitimate reading of ‘one’ in this argument is as meaning something like ‘all human persons’. Granted the wide-scope logic of epistemic norms, the conclusion of this argument, so construed, would apply to all human persons. But we question the generality of this conclusion. Not all religious believers are alike, of course, and we think that the kind of religious believer one is makes a difference to what epistemic norms apply. We will explore this further in the following section.

Responding to evidential scientism on the behalf of religious belief

In this section, we will present a new defence against the application of evidential scientism to religious belief. Before presenting this defence, though, we will briefly sketch out some existing responses to scientism. This is by no means a review of the literature – we just mention these by way of keeping our inquiry on track.⁴

It's not the case that science is the gold standard for evidence

The first defence starts by denying the generality of the claim that contemporary empirical science is the gold standard for evidence. This is a simple point. The generalization that scientific evidence is the best evidence in term of probative value is implausible. If you want to know whether you prefer beer or wine, you can know the answer simply by introspection. A scientist might be able to tell you the same: perhaps there are cases in which our own mental states are inaccessible to us via introspection. But there are clearly cases where this is not the case. More importantly, though, scientific evidence is not the best evidence (in the sense pertaining to its probative value) for religious beliefs – or rather, we shouldn't take that for granted, and those who criticize religious belief because it fails to live up to the standards of evidential scientism certainly shouldn't, on pain of begging the question. Not to mention centuries of theology and philosophy, religious life itself appears to offer all manner of putative evidential resources to believers: introspection, divine revelation, testimony, and so on.

Now, natural theology may seem to be vulnerable to this point, at least if natural theology is understood as entailing a commitment to a broadly naturalistic, empirical outlook. Roughly, natural theology is the view that good arguments about the existence and nature of God can (and perhaps should) be made on the basis of reason and ordinary experience. Think of arguments such as the fine-tuning argument, according to which the best explanation for the apparent suitability of the universe for life on Earth (i.e. the laws of physics and the starting conditions) is the existence of a God-creator who set it up that way. Natural theology is by no means a uniquely Christian tradition. This sort of approach has been also endorsed within Islam, Judaism, and Hinduism at different times in their histories.⁵ Certainly, particular groups of followers of each of these religions has, at times, considered the pursuit of the scientific method (of the time) compatible with and even complementary to the religious life.⁶

In any case, scientism faces a challenge from some philosophers of religion on this very point. According to reformed epistemology, the support of scientific evidence is not necessary for the epistemic rationality of religious belief. If the reformed epistemologists are right, then it follows that even if science is the gold standard for evidence sometimes, it's not always, and crucially, it's not when it comes to the central claims of religion. As Alvin Plantinga has famously argued (2000), we cannot decisively answer the de jure question about what it's rational to believe until we have decisively answered the de re question about whether God exists. This is because, according to Plantinga, whether God exists makes a difference to the very faculties with which we are equipped and, moreover, to what kinds of things we can know and how we can know them. In short: if God does not exist, then the epistemological criticisms with which we began this article might have more traction. But we cannot assume that God does not exist when determining whether it's reasonable to believe that God exists. In any case, as we have already suggested, there are other interesting questions in the epistemology of religion than those about the existence of God. It's perfectly legitimate to assume that God exists in order to establish what else could be rationally believed (or perhaps known), if theism were true.

The norm is far too demanding (P3)

The other major problem with the scientistic outlook is that it's far too demanding. This suggests that it's not the case that it ought to be the case that: if one believes that p, then one's belief is based on the best available evidence. If this norm were true, then a great deal of our beliefs would violate it. Here we can adapt a point made persuasively by van Inwagen (1996) in a reflection on William Clifford's ‘The Ethics of Belief’, namely, that, if it's wrong always, everywhere, and for anyone, to believe anything on the basis of unscientific evidence, then we should expect proponents of scientism to make (or at least to be willing to endorse) similar criticisms of all manner of beliefs or ‘schools of thought’. In other words, if we adopt the norms of scientism and implement them fully, we will be forced to reduce our confidence drastically in very many of our beliefs.

In fairness, it has recently been argued that norms that cannot be satisfied aren't necessarily false: in other words, ought does not imply can.⁷ The thought is that unsatisfiable norms can still guide our behaviour in overall positive ways, providing ‘robust action guidance’. More to the point, Buckwalter and Turri (2020) have argued that empirical evidence shows that people tend not to think that ‘ought’ really implies ‘can’.⁸ If that is correct, then it might help to explain the intuitive appeal of highly demanding norms, be they epistemic or ethical. The thought is that while there is no logical relation between them, the notion of unfeasible obligations chimes with a notion of supererogatory conduct. But this is not sufficient, on our view, to motivate the stronger norm that we ought always to have the best possible evidence, let alone when it comes to religious belief. After all, we are interested specifically in moving beyond philosophical questions about whether God exists, to more everyday matters about ordinary believers' beliefs about their interactions with God.

In the following section, then, we proceed to a different way of thinking about scientism and its discontents, namely, that it fails to appreciate the nature of everyday religious belief.

Friendship, partiality, and epistemic norms

With those preliminaries set aside, we will go about introducing another possible line of response to the sorts of epistemological criticisms that have been sketched out above. In the remainder of this section, we will argue that it's not the case that it ought to be the case that, if one holds religious beliefs, then one's religious beliefs are based on the scientific evidence. Crucially, some aspects of religious life are overlooked in contemporary debates about the epistemology of religious belief. Philosophers of religion have tended to focus on the closely related questions of (1) whether God exists and (2) whether it's (ever) epistemically rational to believe that God exists. But there are further valuable epistemological questions to be asked about what believers can rationally believe about God. One question that we aren't going to explore in this article is the matter of God's existence. A great many religious people believe that they have a personal relationship with God, and – plausibly – this relationship is a source of at least some of what they believe about God. Taking this seriously grants us a new perspective on the epistemology of religious belief.

Ultimately, we will argue that it's inappropriate for religious people who take themselves to have a personal relationship with God to base some of their religious beliefs on scientific evidence. To give an example: if a religious person believes that God has their best interests in mind when he apparently answers (or declines to answer) their prayer, then it would be inappropriate for them to seek scientific evidence in support of this belief. We will mention one such case now. The following testimony is attributed to Peggy, quoted in Froese and Bader's 2010 book on religious life in America, and is supposed to represent a fairly common view – at least within the United States – of how God interacts with the world:

He flushed my toilet one time. I was living overseas and I didn't have a garbage disposal or a sink. My bathroom was where I would wash my dishes and I had left some food, some rice, in a Tupperware container too long in the fridge, and so I had to throw it out. I was going to flush it. Well, it had become hard and kind of compacted in the shape and so it got clogged in the toilet in that shape and I filled the toilet bowl to the rim with water and it just sat there. Nothing happened. I didn't know what I was going to do. I was so embarrassed. I didn't want to call the janitor and have to explain to him that a stupid American had blocked her toilet with rice. And so I just started to pray and as soon as I started to pray it just went . . . flushed. (Froese and Bader 2010, 22)

There are lots of ways we might respond to such testimony. Even some believers might recoil at the thought that God works in such earthly ways. A scientistic response to this avowal of divine intervention would be to point out that the laws of physics are incompatible with the possibility of special divine action. The flushing of the toilet can only have had a physical cause – there is simply no room in the scientific world view for the possibility that a supernatural God intervened on the natural world and flushed the blocked toilet.

In order to understand what is going wrong when we apply this norm to religious belief, we need to pay closer attention to the nature of religious life. The particular feature of religious life that is relevant for this argument is the second-personal nature of the relationship between the religious believer and God. In short, there are good reasons to believe that the nature of the relationship between believers and God is personal.

Many religious believers, especially Christians, think that they have a personal relationship with God. There is some sociological evidence which supports this point. In America's Four Gods, a study of beliefs about God in the United States of America, for instance, it's claimed that around 55 per cent of Americans believe that God is highly engaged with them personally and the world more generally, although within this group there is disagreement about how judgemental God is. Quite what form this relationship takes seems to vary. Just over half of those who believe in a highly engaged God believe that God is judgemental (Froese and Bader 2010, 26). We might speculate that those who believe that God is highly engaged and highly judgemental conceive of their relationship with God as analogous to the relationship between child and parent, and that those who conceive of God as highly engaged but less judgemental conceive of their relationship with God more like a friendship between peers. The key point is that it many believers think that they have a personal relationship with God.⁹

There is some biblical evidence to support the notion that Christian life involves a relationship with God too: when Jesus bids farewell to the disciples in the Gospel according to John, he addresses them as friends, with these words:

This is my commandment, that you love one another as I have loved you. No one has greater love than this, to lay down one's life for one's friends. You are my friends if you do what I command you. I do not call you servants any longer, because the servant does not know what the master is doing; but I have called you friends, because I have made known to you everything that I have heard from my Father. (John 15.7-15)

Whether believers actually enjoy a genuine personal relationship of friendship with God may be debated, but it suggests that the notion that Christian life involves friendship with God is by no means a radical departure from orthodoxy.

Returning to the main point: how should believers form and maintain the beliefs they have about God, if not on the basis of scientific evidence? One place to start to look for an answer to this question is in the ways we form and maintain beliefs about friends and close personal relations. Recent work in the epistemology of friendship has explored the question of whether friendship demands differential epistemic treatment. This differential treatment has come to be known as epistemic partiality. Now, some proponents of epistemic partiality in friendship have argued that friendship can demand us to form and maintain our beliefs in ways that amount to epistemic irrationality – in short, that friendship can be a moral obstacle to knowing. Sarah Stroud, for example, has argued in favour of the radical view that being a good friend requires one to set aside the norms of epistemic propriety and respond to other considerations in the ways that one forms, maintains, and shares one's beliefs. Stroud writes that ‘Friendship positively demands epistemic bias, understood as an epistemically unjustified departure from epistemic objectivity. Doxastic dispositions which violate the standards promulgated by mainstream epistemological theories are a constitutive feature of friendship. Or, to put the point as succinctly – and brutally – as possible, friendship requires epistemic irrationality’ (Stroud 2006, 518). Now, we have no need to adopt such a strong position on epistemic partiality in friendship as Stroud's. Simon Keller has argued in favour of a softer view, according to which‘when good friends form beliefs about each other, they sometimes respond to considerations that have to do with the needs and interests of their friends, not with aiming at the truth, and that's part of what makes them good friends’ (Keller 2004, 330).

What does epistemic partiality like this entail? As Stroud writes, it seems that we ‘owe [our] friends something other than an impartial and disinterested review of the evidence where they are concerned’ (Stroud 2006, 504). Stroud considers how we would respond if we heard a rumour about a friend's misconduct. If a third party told us that our friend had acted badly, and we had no independent reason to doubt this third-party report, should we believe that our friend acted badly? Well, epistemic propriety seems demand that we do – if someone tells you that p and you lack good reasons to doubt that they are telling the truth, then you should believe that p. But friendship's demands seem to pull us in another direction. A good friend does not believe damaging reports about their friends unless the evidence is overwhelming. A few options are open to a good friend. Before believing such reports, we should at least speak to our friend and see what they have to say for themselves. Imagine if we did not. Our friend would surely be hurt to find out what we believed about her. And our relationship with the friend makes a difference here: it's not just a matter of what we believe about her, but also one of what we believe about her. There may be a sense of betrayal in the friendship case that is not present in other cases. Interestingly, our friend would be hurt even if the rumours were true. If there is a rumour about your friend, it's plausible that we should speak to our friend about it before we reach a judgement, even if we are amply furnished with evidence.

What could justify such a radical position? Stroud (2006) defends epistemic partiality in friendship by appealing to reflections on cases – a sort of phenomenological argument. Keller (2004) has argued that we believe for others' sake, to borrow his words, out of concern for their well-being.¹⁰ Very roughly, our beliefs can harm and benefit people whose interests matter to us, and that makes a difference to how we form and maintain our beliefs, and therefore, indirectly, to what we should believe. But one might argue that we cannot harm God. If that is the case, it seems plausible that this strategy won't work here. All is not lost. As argued elsewhere (Efird and Warman forthcoming), Marušić and White's insights about how our beliefs can wrong other people can provide us with a promising defence of epistemic partiality towards God. Marušić and White argue that one person ‘wrongs another person when one's beliefs and judgments fall short of the regard the other is entitled to expect from one’ (Marušić and White 2018, 101). Their view is based on the Strawsonian insight that in personal relationships, we are morally entitled to be treated from the participant stance, that is, from the perspective that sees us as a fellow agent (Strawson 1962). This stance contrasts with what Strawson calls the objective stance. This is the position towards people when we treat them as objects rather than as persons, such as when we make predictions about them based on behavioural psychology. When you fail to respect someone's entitlement to be treated from the participant stance, you wrong them.

Departing from Stroud's view, the aspect of epistemic partiality that interests us here is that which occurs at the level of evidence gathering. Whenever we deliberately go about inquiring whether p, we have choices to make about what kind of evidence we look for and the way we gather it. Let us return to friendships with non-divine persons. A plausible extension of the general notion that we should treat our friend with epistemic partiality is that we shouldn't gather scientific evidence about our friends. How come? Well, in practice, the epistemic norms of inquiry, and those concerning evidence gathering in particular, are influenced by conventional as well as non-purely epistemic considerations. What evidence you gather and how you gather it are shaped, not only by the nature of your question, but also your resources, including time, equipment, and expertise. Note that this has nothing to do with practical reasons for belief. Imagine that you want to find out whether your shirt is made of cotton before you put it in the laundry. You could simply read the label, or you could examine the physical structure of the individual fibres under a microscope. Both approaches would get you an answer, but despite its potentially higher probative value, the latter is an unacceptable method because it's highly impracticable: it's an expensive and time-consuming process for which you aren't properly trained. If you employ the second method, your inquiry is open to criticism that it employs an inappropriate evidence-gathering process. This is just to illustrate that non-purely epistemic considerations (such as logistical considerations) are relevant when considering whether an inquiry is a good inquiry. To reiterate, this is a different point from the one made by supporters of pragmatic encroachment and related views.

We can apply a similar sort of thinking to the ways in which we form and maintain our beliefs about our friends. Considerations of friendship can play a similar role in our evaluations of inquiries as logistical ones. While they aren't purely epistemic considerations, they make a difference to how we should plan and execute certain kinds of inquiry, and they are therefore relevant when evaluating those inquiries. Let us consider an example: suppose you wonder whether your friend intends to break up with their romantic partner, in the aftermath of a serious argument. There are several ways you could gather evidence that is relevant to your inquiry. You could find out their intentions by making observations about their behaviour and, using the insights of behavioural psychology, determine whether they do, in fact, have the intentions that you suspect they might. This might get you an accurate answer about your friend's intentions: after all, it can be hard to know our own intentions by introspection, and people are sometimes uncomfortable about sharing sensitive plans. But there is nevertheless something rather strange about this way of forming beliefs about a friend. If your friend found out that you formed beliefs about them in this way, it wouldn't be surprising if they resented you for it. Why didn't you just ask them whether they intended to end their relationship?

The upshot of this slightly far-fetched example is that close personal relationships make demands of the ways we gather evidence. People with whom we share a personal relationship justifiably expect to be treated from the participant stance; that is, they expect not to be treated as one human being among many about whom scientific evidence can be gathered and predictions can be made. To fail to take this seriously in your interactions with someone is to jeopardize your relationship with them. More to the point, your conduct wrongs them.

How does this fit into the dialectic here? The reflections outlined in this section put pressure on the second premise of the argument presented above, namely, that it ought to be the case that: if one believes that p, then one's belief is based on the best available evidence. We have presented friendship as a counterexample to this position. If we ought to treat our friends with epistemic partiality, then it's not the case that our beliefs ought to be based on the best available evidence.

Application to religious belief (P5)

Now we will explain what this means for religious belief and how it allows us to push back against the application of evidential scientism to religious belief. After all, we have not shown explicitly what epistemic partiality has to do with a believer's relationship with God and their beliefs about their interactions with God. We will turn to that now.

Some religious believers hold that they are in a personal relationship with God. This relationship is sometimes characterized as a friendship. At other times it's characterized as parent–child relationship, where God takes the parent-role. If friendship requires epistemic partiality, then perhaps parent–child relationships do too. But in any case, it seems that for some believers, there is a friendship-related requirement to treat God with epistemic partiality. As argued in a previous section, there are good philosophical reasons for thinking that our epistemic behaviour – how we form and maintain our beliefs – can harm and benefit our friends. If that is the case, then, given the requirement that friends treat each other from the participant stance, it follows that believers should treat God with epistemic partiality.

What would this look like in practice? Earlier we considered the curious case of the blocked toilet. In short, someone blocked a toilet and prayed for God's help to unblock it. The toilet suddenly flushed. And the person who prayed to God for help explains this sudden unblocking as a miraculous intervention by God, to save her from an embarrassing quandary. We were interested in the question of whether it was acceptable, from an epistemological point of view, for the person to believe this account of events. After all, according to a scientistic outlook – so the thought goes – it's not. Given the inviolability of the laws of physics, there must be a better explanation, a naturalistic one, of the miraculous-seeming unblocking. The person who believed that God flushed her blocked toilet is, on this view, deserving of epistemic criticism because they hold a belief that is not based on the best available evidence. If we assume that Peggy enjoyed a friendship with God, then we can see that this criticism is misguided.

The concept of epistemic partiality, applied to the relationship that believers take themselves to enjoy with God, allows us to understand a little better why it might be wrong for the scientistic-minded critic to criticize them for giving God the credit for unblocking the toilet and ignoring the scientific evidence. Even if, for us outsiders, what she says seems completely irrational, it makes sense from her perspective. From her point of view, it's obvious that God flushed her blocked toilet in response to her prayer. And more to the point, given her relationship with God, it would be wrong for her to consider alternative, scientific explanations. She may be mistaken, but failing to scrutinize her interaction with God with scientific rigour is not a good enough reason to criticize her epistemic conduct in this case. After all, when you ask a friend to do a favour and what you wanted gets done, you would be wronging them if you looked for other, less benevolent explanations of their behaviour or questioned whether they really answered your request, and it was not just lucky that their intentions coincided with yours.

Concluding remarks

The view that we have proposed in this article can help us to explain some interesting data about how some religious beliefs interact with scientific beliefs. On the one hand, just as some religious beliefs seem to respond to scientific evidence, some don't. This view can help explain why some religious beliefs don't respond to scientific evidence. Making room for the epistemic norms governing everyday religious beliefs, and in particular the evidence gathering norms to form and maintain them, to be more expansive, allows for the explanation that some religious believers who have a friendship with God needn't consider the available scientific evidence when forming and maintaining certain religious beliefs. We don't suggest, as some do (van Leeuwen 2014, 2017), that religious beliefs are actually not responsive to scientific evidence and so aren't properly beliefs, and in those cases when they seem to be, the believers in fact engage in some sort of make-believe. The proposed view doesn't exclude religious belief from the category of belief. Excluding them doesn't allow us to assess them rationally (if we understand this rationality as a normative property possessed by a belief) and so to expose them as irrational when appropriate.¹¹ Moreover, the unresponsiveness to evidence isn't, as some have suggested (Pritchard 2017), because some religious beliefs have some doxastic centrality in the religious believers' psychological economy or are ‘hinge commitments’, which aren't subject to rational evaluation,¹² but because they are the product of a friendship with God. The proposed view then has the further advantage that it's not in tension with the fact that, as we will next see, some religious believers think their fundamental commitments are rational. This view holds that the religious beliefs that aren't sensitive to scientific evidence and are the product of a friendship with God are nonetheless rational. On the other hand, this view can also be part of a more comprehensive explanation as to why some religious believers (as well as non-believers) think that science and religion are in conflict (Pew 2015). After all, regarding those religious beliefs that are the product of friendship with God, there certainly can be a tension between the religious beliefs and the available scientific evidence (cf. Principe 2015).¹³

This view also helps us to understand the apparent offensiveness of some complaints about the (alleged) epistemic irrationality of religious belief. Presumably, some religious believers wouldn't agree that their religious beliefs are irrational. The view here proposed can help us explain how this can be so even if the available scientific evidence is ignored. Given the nature of their relationship with God such evidence ought not to be considered. This in turn can help us avoid attributing a massive error on the part of the religious believers; consequently, being more charitable to them and avoiding the need for an error-theory. Moreover, religious beliefs tend to be psychologically stable beliefs and this resilience can be explained not by means of the believer's irrational stubbornness but by their rationality.

We have argued that, at least, some epistemic norms governing religious beliefs can differ from those governing other beliefs that aren't associated with personal relationships. If this is so, then we should be more careful when charging religious believers, like Peggy, with irrationality. In fact, if what we have argued is correct, then the epistemic conduct and thoughts of some religious believers can be charitably understood as epistemically rational.