Credo in unam credentiam: religious beliefs are standard beliefs

Abstract

Does religious belief differ in any interesting way from other kinds of belief? For now, take ‘belief' to mean how one takes the world to be, on the basis of which they act. Call beliefs like this ‘ordinary beliefs'. There are also more complicated, or abstract, beliefs. Call such beliefs ‘non-ordinary beliefs’. Are religious beliefs different in any significant or interesting way from what we call ‘standard belief’? Our analysis shows that they are not. Although the content of religious belief is different, and sometimes so is the function, religious beliefs are still mental, dispositional attitudes that aim to provide a correct representation of the world. Therefore, religious beliefs are best understood as a sub-category of standard beliefs, alongside ordinary and non-ordinary beliefs. This account of religious belief provides further insight into the meaning of belief and supports a revision of a strict demarcation between standard and non-standard beliefs.

1 On belief

Does religious belief differ in any interesting way from other kinds of belief? For now, take ‘belief’ to mean how one takes the world to be, on the basis of which they act (Crane, 2013, p. 3). For instance, if one has the mental state ‘I believe it is raining’ then they are committed to the truth that it is in fact raining, and are likely to take actions informed by this information, such as carrying an umbrella. Drawing upon Willard Van Orman Quine, we will call beliefs like this ‘ordinary beliefs’ (Van Orman Quine, 1995, p. 49). There are also more complicated or abstract beliefs, for instance, beliefs about universal instantiation or Platonism about mathematical entities. Call such beliefs ‘non-ordinary beliefs’. How do ordinary and non-ordinary beliefs, both of which are a kind of what we can call ‘standard beliefs’, differ from religious beliefs? Are religious beliefs different in any significant or interesting way? Perhaps they are a kind of standard belief? Most of the literature on the matter says that religious beliefs are very distinct from standard beliefs. They are not, for although the content of religious belief is different (as the content of any two distinct beliefs is different), and sometimes so too is the function, religious beliefs are still mental, dispositional attitudes, that aim to provide a correct representation of the world. Therefore, religious beliefs are best understood as a sub-category of standard beliefs, alongside ordinary and non-ordinary beliefs.

The case for religious beliefs as standard beliefs will be made as follows. In Sect. 2, we will explain in further detail what belief is. This will include a more detailed explication of standard beliefs and the distinction between ordinary and non-ordinary beliefs, and then at Sect. 3 a definition of ‘religious’. At Sect. 4, we will present the case for religious beliefs being very different from standard beliefs, a position we dub hard disparatism. At Sect. 5, we analyse arguments for soft disparatism, the view that religious beliefs are moderately different from standard beliefs. In Sect. 6, we make the case for non-disparatism, the view that religious beliefs are a kind of standard belief. Lastly, at Sect. 7, we will analyse the concept of faith and consider whether or not it is a unique aspect of religious belief that excludes it from standard beliefs, and conclude that it does not sufficiently differentiate religious beliefs to the extent that they can be excluded from standard beliefs. By this analysis we can achieve two additional objectives. First, we can show that an appropriate understanding of belief ought to quantify over a broad range of kinds of beliefs, including religious beliefs, rather than concerning itself with only a narrow scope of kinds of beliefs that exclude non-standard cases. Second, we can push back against a dogmatic view in philosophy that holds that there is a stark and stable division between rational, naturalistic, standard beliefs on the one hand, and irrational, non-naturalistic, non-standard beliefs on the other.

2 What is belief?

According to Tim Crane, beliefs are mental, dispositional attitudes that aim to provide true or correct representations of the world (Crane, 2013, p. 3). Drawing upon Bernard Williams (Williams, 1973), Crane takes beliefs to be a way of guiding an agent through the world, like a kind of map (Crane & Farkas, 2022).

This is a foundation, yet we can say a few more things about beliefs. First, beliefs are often taken to be credences because we believe certain things more strongly than others. After all, humans have doubts, convictions, speculations, hypotheses, and so on. We can talk about a degree of confidence in one’s belief about a proposition P (Papineau, 2012, pp. 89–90), therefore, we can measure the correctness or accuracy of a belief, at least in principle, when we say that it aims at truth. Second, beliefs are manifested in actions that express preferences among various outcomes. Generally, the greater degree of belief an agent attaches to P, the more inclined said agent will be to act in such a way as if P were true (Papineau, 2012, pp. 89–90; Tenenbaum et al., 2011, p. 1280). These probabilistic calculations are usually taken to be inaccessible to conscious reasoning processes at the intentional level (Block, 2002; Dennett, 1978, pp. 304– 5).

On this description, beliefs are dispositional attitudes, so for a belief to cause action it must be connected to other mental states, such as desires. This means that, third, to act on a belief one must have a motivating desire. This account is a kind of functionalism about belief and entails that not all beliefs are conscious, although they can be brought to consciousness in episodes of conscious thinking. Here ‘conscious thinking’ refers to when the unconscious mental state, the belief, gives rise to phenomenally conscious events. This unconscious representation of the world serves an adaptive function: to model an agent’s environment so as to allow them to make optimal predictions and accordingly take action (Armstrong, 1973, p. 3; Crane, 2017b, p. 9; Ramsey, 1931, p. 238; Theriault et al., 2021, pp. 103–104). On this account beliefs are unconscious (Crane, 2017b), functional (Pietraszewski & Wertz, 2021), and occur at a sub-personal level. Thus, fourth, beliefs themselves are never conscious, but are brought to consciousness as thoughts expressed by assertions (Hawthorne et al., 2016); the belief itself is an unconscious disposition to act. In this sense, belief is largely an involuntary, unconscious, and non-linguistic phenomenon. The account of belief we have just outlined is referred to as dispositionalism.

Another popular position is representationalism about belief. Representationalists describe central cases of belief as the state of having in one’s mind or head a representation with the same propositional content as the belief (Schwitzgebel, 2023). What this means is that beliefs are about and/or somehow depend upon propositional content that is determined by representations; information about the world, such as a map or language, that presents the world to an individual; hence re-presenting. As these beliefs can be accessed even when the aspect of the world that was represented is not present, it is understood that the propositional content determined by a representation is stored in memory. This characterization of belief has been challenged by some philosophers, but it is probably fair to say that the majority of contemporary philosophers of mind accept the bulk of this picture, which embodies the core ideas of the representational approach to belief, according to which central cases of belief involve someone’s having in their head or mind a representation with the same propositional content as the belief.

Proponents of representationalism about belief include Burge, 2010; Cummins, 1996; Dretske, 1988; Fodor, 1968; Fodor, 1975; Fodor, 1981; Fodor, 1987; Fodor, 1990; Millikan, 1984; Quilty-Dunn & Mandelbaum, 2018; Mandelbaum & Porot, 2023; and Zimmerman, 2018. Representationalists have raised concerns with dispositionalist accounts of belief. For example, Jake Quilty-Dunn and Eric Mandelbaum argue that a dispositionalist about belief must adopt a functionalist account to accommodate the complexity of psychological states—as discussed above—yet there is a concern that functionalism undermines the explanatory power of dispositionalism (Quilty-Dunn & Mandelbaum, 2018). Much of this has to do with the multiple-realisability of behaviours.

It is not our objective to resolve the dispositionalist-representationalist dispute here. The dispositionalist account is adopted for its explanatory power, and due to the following two concerns: (i) on representationalism there is a commitment to all beliefs being stored in memory, but this means one can never assent to a belief without having previously assented to it, however, we sometimes do have beliefs about propositions we have never before considered and which are thus not yet stored in memory; and (ii) representationalism commits one to storing an implausibly high number of beliefs in memory to explain beliefs that might have never been brought to assent before, or perhaps only once per lifetime (Crane, 2017b, p. 4). Having said this, as much of the debate centres around how and where beliefs are stored in the mind or head, and how they are brought to consciousness, and because this debate does not provide specific insight into the nature of religious belief, it ought not be our focus. For reasons of parsimony and preference, we will address a dispositionalist account, unless stated otherwise, however, the analysis is also applicable to a representationalist account because both representationalists and functionalists agree that beliefs influence how one takes the world to be, upon which they base their actions.

To maintain a focused and manageable analysis, we will proceed with a dispositionalist account: take ‘belief’ to mean ‘how one takes the world to be, on the basis of which they act’ (Crane, 2013). On this account, the beliefs themselves are not dispositions, but attitudes that dispose one to act in conjunction with desires. Recall our earlier example about rain as an illustration: if one has the mental state ‘I believe it is raining’ then they are committed to the truth that it is in fact raining and are likely to take actions informed by this information, such as carrying an umbrella. Call this example umbrella. Umbrella is a paradigm example of how belief informs action.

Let us make some further things clear about our account of belief. For our purposes, we can point to three particularly interesting aspects of belief: (a) content; (b) function; and (c) truth-aim. A belief has content C that in conjunction with desires D gives rise to a function φ, the objective of which is to provide a true representation of the world to guide action. In umbrella, the content of the belief is something like < It is raining >¹ and the function is to guide behaviour so as to avoid getting wet. The belief helps the believer achieve this objective by aiming at a true representation of the world: the fact that water is actually falling from the sky, and that by holding an umbrella over one’s head, one will thereby remain dry.

When it is said that one kind of belief is different from other kinds of beliefs, it appears what is meant by this is that there is a difference in (a) content, (b) function, and (c) truth-aim, or any combination of these three. These three categories are not equal when it comes to distinguishing kinds of belief. Almost every belief will have different content: one’s belief that it is raining is different from one’s belief that it is sunny outside, so a difference in content is hardly grounds for claiming that something like religious belief is very different from other kinds of beliefs. Function, on the other hand, is less clear. In a scenario, heaven or hell, one can have a belief that heaven exists and one’s actions will determine whether they will one day go there, likewise for hell, or they might be agnostic about the existence of one or both. If the function in umbrella is to keep one dry, and the function in heaven or hell is to determine if there is a heaven or hell and get (presumably) to heaven, they may not be comparable because one is ‘this-worldly’ and immediate, and the other is ‘other-worldly’ and less immediate (one hopes) (Nietzsche, 2006, p. 6). On the other hand, both might have a function to ‘keep one safe’ by ensuring one takes the correct course of action. The status of different function is less clear than the status of different content, as regards delineating kinds of beliefs. Finally, truth-aim is the most promising way to distinguish beliefs. Truth-aim is the intended action that follows from a belief, what a belief in conjunction with a desire directs one to do. Perhaps the truth-aim of umbrella is to determine if it is in fact raining and avoid becoming wet, while heaven or hell has no truth-aim: it is a musing about the theoretical notion of an afterlife, not a commitment to how reality is, to thereby guide action. Of course, as we will discuss, heaven or hell might in fact be aiming at belonging to a community or aiming at the truth like umbrella to direct the believer to heaven.

Equipped with these three aspects of belief, we can distinguish between two different kinds of beliefs. It is not a sharp division but a spectral delineation. Nor is it the only possible subdivision of kinds of beliefs. On the one hand, there are beliefs like umbrella. Such beliefs comprise the majority of beliefs one has. For this reason, drawing upon Quine, we call such beliefs ordinary beliefs (Van Orman Quine, 1995, p. 49). On the other hand, there are beliefs that are less quotidian, often with more abstract content and/or prima facie less evident functional implications. For example, metaphysical beliefs, such as a belief in universal instantiation or Platonism about mathematical entities. By way of contrast, we call these kinds of beliefs non-ordinary beliefs. Although the content of these two kinds of beliefs differ, neither is what we might think of as kinds of specifically religious beliefs. As we are considering if religious beliefs differ in interesting ways, we call all non-religious beliefs—ordinary or non-ordinary—standard beliefs. Religious beliefs, therefore, are what we can also call non-standard beliefs.

Now we should consider if this distinction is justifiable. To determine this, we must answer the following question: do religious beliefs differ in any interesting way from standard beliefs?

Naturally, one might wonder why it matters which beliefs are standard or non-standard, ordinary or non-ordinary? Why do we need such definitions to understand belief? There are two philosophically important reasons to analyse belief in such a way.

The first reason is that if we wish to better understand what beliefs are, as many philosophers seem to, then there are two plausible approaches; the first is to start with a narrow class of beliefs and get clear on what they are and how they work, and then expand from there. For instance one could analyse cases like umbrella and extrapolate from this to understand less ordinary cases like heaven or hell. The second approach is to take all different kinds of beliefs and analyse them together because neither historically nor conceptually precedes the other. For example, perhaps umbrella and heaven or hell cannot be understood in isolation and neither is more representative of belief than the other because we have all kinds of different beliefs. The first approach is common in the literature on belief, the latter less so. The first approach can be characterised as a Procrustean bed method: one first gets clear on representative cases of belief and then cuts outlying kinds of beliefs down to size to fit the representative cases, or else discards them. We suggest this approach misunderstands belief because it fails to properly account for all kinds of beliefs, including some of our most deeply held, informative, and meaningful beliefs. Indeed, if non-standard beliefs do not fit the agreed account of belief then one may accede to the odd conclusion that these things are not beliefs at all.² We are going to need a bigger bed. Hence why we ought to employ the latter approach because to better understand what belief is we have to understand how people use the word, without prejudice. We can gain greater insight into the nature of belief by engaging with all types of beliefs, including religious beliefs.

The second reason is to put to rest the implacable spectre of one aspect of logical positivism that still lingers as an implicit, anti-metaphysical attitude in philosophy. There is, at times, a distaste in philosophy for things which are not verifiable on a naturalist stance. In particular, all things religious, metaphysical, or thus far physically inexplicable, are often pushed into their own category because they are deemed peculiar, incredulous, or even irrational. Yet if the last century of analytic philosophy has taught us anything, it is that such a stark division has not held. As we shall see, there are few good reasons to claim that beliefs about cricket are rational but beliefs about Buddha are not or that beliefs about the opening of British parliament have nothing in common with beliefs about attending a funeral. Therefore, by showing the division does not hold in the case of belief—between standard and non-standard—we can shake off a dogmatic approach to belief and thereby make progress in our understanding of what belief is. The lesson might even be expanded to inform a general philosophical perspective, that we ought to challenge philosophical dogmatism in all its forms, and be receptive to alternative philosophical approaches. For when our conclusions ignore the motivating concepts, problems, and data that generated the initial philosophical puzzle, then we have done the dialectic under consideration a disservice.

3 What is religious belief?

We ought now say something about religion. ‘Religion’ is a slippery term; William James put it best in The Varieties of Religious Experience when he said that religion has no single principle or essence (James, 1902). It is difficult to demonstrate commonalities between Abrahamic faiths, religions such as Daoism, and folk religions like the worldview of the Batak (Smith, 1982, p. xi). The Oxford dictionary provides the general “belief in and worship of a super-human controlling power...” and “A pursuit or interest followed with great devotion”. Problematically, the God of Abraham, the fatalism of the Norse, and the telos of the arc of Whiggish history are all superhuman, controlling powers, and the entailed moral claims are pursued with equal devotion. Additionally, this definition of ‘religion’ might apply to ideologies, even explicitly atheistic ideologies. Indeed, this accusation is often levelled at totalitarian politics or political philosophies, for example juche sasang (주체사상) (see Goodrick-Clarke, 1985; Ling, 1980; McCarraher, 2019; Omer & Springs, 2013).

Friedrich Nietzsche famously said that only that which has no history can be defined (Nietzsche, 1999).³ Religion has a long and complex history that ought not be simplified, for when one presents complex history as simple one does not thereby show history to be simple and oneself complex, rather, history remains complex but one shows oneself to be simple. Given this complexity and the polysemous nature of the term ‘religion’, an airtight definition may elude us (Griffiths, 2000, p. 30). Yet it is difficult to deny that there is what Ludwig Wittgenstein called a Familienähnlichkeit, or family resemblance, between the many things we call ‘religious’ (Wittgenstein, 2001).⁴

Let us try to be a little more specific about what is allegedly being resembled. We begin with sociologist Émile Durkheim (1995, p. 35) who says that:

Religion is a unified system of beliefs and practices relative to sacred things, that is to say, things set apart and forbidden.

Now we must unpack what the ‘sacred’ is. Rudolf Otto’s (Otto, 1923) proposal of the concept of the ‘numinous’ captures the notion of the sacred. ‘Numinous’ is a derivative of the Latin ‘numen’, meaning ‘spirit’. ‘Numen’ is etymologically derived from the concept of divine will and represented by a ‘nodding’ of the head; a kind of intentional aspect to the sacred that appears lacking in a purely mechanistic view of the universe. We will take ‘numinous’ as akin, but different to, ‘numinal’, because while they share an etymology, the latter implies a strong sense of intentionality; something suggesting a divine will as, for example, is apparent in theism.⁵ Otto says that what the numinous refers to evades precise formulation in words and that it is more similar to the beauty of a musical composition, and thus must be discussed in symbolic terms (Otto, 1923). This being said, here is C. S. Lewis’ articulation of the numinous (Lewis, 2009, pp. 5–6):

You would feel wonder and a certain shrinking—a sense of inadequacy to cope with such a visitant and of prostration before it—an emotion which might be expressed in Shakespeare’s words “Under it my genius is rebuked.” This feeling may be described as awe, and the object which excites it as the Numinous.

Thus, as a starting point, we can say that religion is about the sacred or numinous (or numinal). We can add further criteria to make the definition more robust. Important classifications have been presented by Edward Herbert, Edward Burnett Tylor, William James, Rodney Needham, Martin Southwold, Rem Edwards, and J. Z. Smith, to name a few. Here are Tim Crane’s (Crane, 2017a, p. 7) requirements for religion:

Religion[. . . ]is a systematic and practical attempt by human beings to find meaning in the world and their place in it, in terms of their relationship to something transcendent. This description has four essential elements: first, religion is systematic; second, it is practical; third, it is an attempt to find meaning; and fourth, it appeals to the transcendent.

Crane systematises these into the following criteria for religious belief (Crane, 2017a, pp. 7–9). Religion is:

• 1. Systematic: it essentially involves a collection of ideas and practices that are designed to fit together.

• 2. Practical: it involves believing in certain propositions or doctrines, knowing certain stories, but also acting in a certain way.

• 3. Meaningful: it attempts to find the meaning of life as a whole, what Karen Armstrong (Armstrong, 2014, p. 3) has called the “investment of everything with ultimate meaning”.

• 4. Transcendent: religious belief includes an awareness of the transcendent, and it is in the transcendent that the search for religious meaning ends.

Note that there is an important distinction between spirituality and religion, which we analyse at Sect. 5. For now, take ‘spirituality’ to refer to kinds of beliefs with content about the numinous but with no, or weak, commitment to a community. Of course, as we will discuss, many spiritual people might contend that spirituality has a strong connection to community, and likewise religious people might contend that their beliefs have little to do with their community, but we must find a delineation between spirituality and religion, and this seems the most plausible, albeit we acknowledge that there is not a sharp divide.⁶ Crane’s criteria appear to capture the notion of religion noscitur a sociis, whether theistic or no. Therefore, we will continue with our investigation by employing this holistic account, acknowledging that we are engaging in a bounded polythetic approach to the concept (Laurence & Margolis, 1999; Schilbrack, 2022).

4 Religious belief is very different: hard disparatism

Do religious beliefs differ in content, function, or truth-aim, in any interesting way from standard beliefs? The first view answers with an emphatic ‘yes’. Obviously, religious beliefs differ in content: they are about religious things. Umbrella is not about God, unless one worships the rain. Yet it might also be the case that religious beliefs differ in function and/or truth-aim. Call the view that religious belief is very different from standard belief hard disparatism: ‘disparate’ because these theorists take religious belief to be different in kind to standard beliefs, ‘hard’ because they take religious belief to be very different. What is indicated by the adverb ‘very’ is that religious beliefs do not sufficiently overlap with standard beliefs according to the criteria of function and truth-aim, and additionally, they violate the spirit of naturalism. The logical positivists of the Wiener Kreis, or Vienna Circle, capture the alleged chasm that divides standard beliefs, in this case ordinary beliefs, from religious beliefs (Hahn et al., 1929, p. 10):

The metaphysician and the theologian believe, thereby misunderstanding themselves, that their statements say something, or that they denote a state of affairs. Analysis, however, shows that these statements say nothing but merely express a certain mood and spirit. To express such feelings for life can be a significant task. But the proper medium for doing so is art, for instance lyric poetry or music. It is dangerous to choose the linguistic garb of a theory instead: a theoretical content is simulated where none exists. If a metaphysician or theologian wants to retain the usual medium of language, then he must himself realise and bring out clearly that he is giving not description but expression, not theory or communication of knowledge, but poetry or myth. If a mystic asserts that he has experiences that lie above and beyond all concepts, one cannot deny this. But the mystic cannot talk about it, for talking implies capture by concepts and reduction to scientifically classifiable states of affairs.

Note that they also include non-ordinary standard beliefs (metaphysical beliefs) in this category, which we will address at Sect. 6. Philosopher and logical positivist sympathiser A.J. Ayer in Language, Truth, and Logic even went so far as to argue that religious claims (and their denial) are without cognitive content (Ayer, 1936). This hard disparatist attitude to religious belief is still prevalent. Consider a case of two contemporary philosophers that advocate for what we are calling hard disparatism, Neil Van Leeuwen and Tania Lombrozo, who are pluralists about belief (Van Leeuwen & Lombrozo, 2023). According to pluralism, there are importantly different ways to “believe” an idea. For instance, they claim that most standard beliefs help serve the function of tracking the likely state of the world, but other kinds of beliefs, such as religious beliefs, serve interpersonal functions like generating signals of group membership (Cruz, 2020; Kahan, 2015), and intrapersonal functions, such as finding a sense of peace (Luhrmann, 2020).

The literature on why religious beliefs allegedly differ dramatically from standard beliefs is vast. Here are thirteen additional proposals that capture many of the claims of the hard disparatists. Religious beliefs are, allegedly:

• 1. Distinct in ritual, material, doctrinal and philosophical, experiential and emotional, social and institutional, narrative, and mythic, matters (Geertz, 2016; Smart, 2000).

• 2. Symbolic, not literal (Evans-Pritchard, 1976).

• 3. Group beliefs that can contradict individual beliefs (Astuti & Harris, 2008).

• 4. Contextual (Astuti, 2015; Barrett & Kiel, 1996).

• 5. Characterised by conviction, often involving an initiatory declaration of orthodoxy (Ruel, 1982, p. 109).

• 6. Opaque: they remain mysteries, and rely on ritual deference (Crane, 2017a).

• 7. Half-understood ideas (Naumescu, 2011).

• 8. Unreflectively held without attention to reasons to believe (Sperber, 1997).

• 9. Minimally counter-intuitive: they violate our intuitive naturalist ontology (Boyer, 2001). They are ‘superempirical’: things that are not the product of empirical things nor need be understood empirically (Schilbrack, 2013).

• 10. ‘[A]ll explicit and implicit notions and ideas, accepted as true, which relate to a reality which cannot be verified empirically’ (Wiegers & Platvoet, 2002).

• 11. Irrational (Kunin, 2003) or ‘irrational’ imitation (Naumescu, 2011): the property alleged to cause an individual’s behaviour is empirically unsupported.

• 12. Often contradictory (Astuti & Harris, 2008, p. 144).

• 13. A set of required actions (Smith, 2017; Vásquez, 2011).

What exactly binds all these descriptions together? Our tripartite division into content, function, and truth-aim best explains this. (1)–(5) are distinguishing religious belief based on function; they are saying that the function of religious belief is not to navigate the world but to foster social cohesion. (1)–(13) appeal to the truth-aim to explain that the aim is not to get the world right on a naturalistic perspective but to aim at fostering social cohesion (1)–(5), or to contemplate a concept (6)–(7). (9)–(13) suggest that religious beliefs either do not aim at truth in the same way as standard beliefs, or if they do, they fail due to irrationality or contradiction.

These criteria also imply that religious beliefs violate naturalism. What these theorists mean by naturalism is not clear. We take them to be talking about a worldview that puts stock in empiricism, the natural sciences, consistent nomological laws, and a general kind of physicalism: the thesis that the only fundamental properties in the universe are physical, and that all other properties are derived from physical properties. This talk of naturalism is important, as although hard disparatists might agree that all standard beliefs are naturalist beliefs, others might think that beliefs about logic, or Platonic forms, are not naturalistic, but still standard.

In summary, hard disparatists claim that the functional implication of a religious belief is not always about naturalistically navigating the world, and religious believers sometimes have no interest in the truth-aim of their beliefs or are ambivalent or agnostic about said aim. Sometimes it is even claimed that there is no truth-aim present at all, as aptly captured by Gareth Moore (1988, p. 287):

We may say: People do not discover religious truths, they make them.

Let’s return to Neil van Leeuwen who is representative of the thinking of hard disparatists that religious belief is alien to standard beliefs (Van Leeuwen, 2014, 2017a, 2023). Van Leeuwen argues that many religious and supernatural “beliefs” are not factual beliefs, but a kind of ‘secondary attitude’, similar to imaginings, hypotheses, or assumptions for the sake of argument: one might hypothesize that P, imagine that P, assume that P, and so on. He explains that factual beliefs (Van Leeuwen, 2014, p. 1):

(i) are practical setting independent, (ii) cognitively govern other attitudes, and (iii) are evidentially vulnerable. By way of contrast, religious credences (a) have perceived normative orientation, (b) are susceptible to free elaboration, and (c) are vulnerable to special authority.

Van Leeuwen provides the example of the Maya-speaking Itza claim that humans sometimes transform into animals, yet they also do not worry that meat-eating is potentially cannibalistic (Atran, 2002, pp. 84–86). This example does not really work, however, as the Itza position is logically and naturalistically consistent; if these people are no longer human animals, then eating them is not cannibalism. Van Leeuwen also argues that ‘factual’ belief is not the same as religious belief because despite employing the same words, one’s attitudes toward the given propositions differ (Van Leeuwen, 2014). Further, Van Leeuwen thinks religious beliefs are somehow incorrigible in a way that standard beliefs are not (Van Leeuwen, 2014, p. 707). He provides an example: at the end of the twentieth century many feared an event called Y2K. It was alleged that due to the way data about dates was handled by digital computers, said systems would crash on January 1, 2000. This event did not manifest, and consequently most people discarded their belief in Y2K. According to Van Leeuwen, Y2K is a different kind of belief to religious belief because if a religious cult predicted a global catastrophe on January 1, 2000, and it failed to materialize, then there would not be a rejection of the belief. This is because the standard belief was held factually and the religious belief is (somehow) immune to evidence. Van Leeuwen (2014, 2017a, 2023) concludes that religious beliefs are credences, but not factual beliefs.

Van Leeuwen even goes so far as to say that religious beliefs often fail to respond to evidence (Van Leeuwen, 2017a, p. 20). He posits that:

Most religious people just don’t seem to care about evidence for their credences. The Evidence Game is played mostly by intellectuals and apologists.

Van Leuwen argues that for religious beliefs, everything is potentially relevant, and thus they are immunized from falsification. Thus, only the cleverest religious folk engage in rational thinking and the rest employ a permissive, non-evidentiary system, immune from falsification. He even equates religious belief with children’s make-believe, and refers to such beliefs as the ‘Playground’ (Van Leeuwen 2017b, p. 8; Van Leeuwen 2023). He argues that because religious belief is make-believe, or what he calls a ‘fictional imagining’, it does not guide action. This captures the hard disparate position well: religious beliefs inform function differently because they are not about the world, and they do not aim at truth.

We will make the positive case for non-disparatism at Sect. 6 and will refute Van Leeuwen’s position, but we would be remiss not to immediately point out that Van Leeuwen’s terminology does a disservice to the dialectic. Not only is the term ‘playground’ an ad hominem, it commits a petitio principii, because Van Leeuwen argues that religious belief is a ‘fictional imagining’ precisely because it allegedly does not conform to naturalism (Van Leeuwen 2017b, p. 8). In other words, he begs the question in favour of standard beliefs as necessarily naturalistic, and states that necessarily religious beliefs cannot be naturalistic, and what is not naturalistic is not rational. He begins with the division between standard and religious beliefs already cemented. Of course, on this reasoning, this would rule out non-ordinary standard beliefs, including some beliefs about universality or metaphysics generally, as the logical positivists pointed out. For instance, on Van Leeuwen’s position, ontological beliefs about nominalism or Platonism come out as being unconcerned with evidence on the grounds that the ‘evidence’ here quantifies over entities with an ontological net larger than naturalism. One is compelled to ask: is the ontology from which Van Leeuwen derives his own epistemological judgements and naturalistic commitments thereby also unconcerned with evidence? In other words, Van Leeuwen begs the question against religious beliefs by claiming that only their worldview is too broad to be unrevisable, while a naturalist-physicalist worldview should not be revised, although both kinds of beliefs quantify over the same domain.

Note also that in Sect. 2 we discussed how many beliefs, standard or religious, can come in degrees. Therefore, religious beliefs as credences does not distinguish them in any manner from standard beliefs. We investigate this further at Sect. 6, where we also counter Van Leeuwen’s notion that standard beliefs are factual and religious beliefs are not.

The upshot of Van Leeuwen’s position is that due to the allegedly non-naturalist, non-factive, and incorrigible nature of religious beliefs, they are distinct in both function and truth-aim from standard beliefs. Both umbrella and Y2K involve a function to successfully navigate a scenario, and aim at truth to manifest success. Conversely, Van Leeuwen might suggest that heaven or hell involves a function to contemplate an idea not to guide behaviour, and it aims more at imagination or contemplation than a commitment to truth and successful navigation of the world.

Keeping in mind Crane’s definition, the thirteen listed criteria, Van Leeuwen’s arguments, and the logical positivist-informed naturalist stance, we can see that hard disparatists take religious belief to be very distinct from standard beliefs. Hard disparatists are not arealists about religious belief: they still think there are psychological states and propositions with religious content. What they do think, however, is that there is something ‘off’ with the function following from religious beliefs, and the way they aim at truth. In other words, the function is not always connected with the content, and religious believers sometimes have no interest in the truth-aim of their beliefs, or are ambivalent or agnostic about said aim. Sometimes it is even claimed that there is no truth-aim present at all.

5 Religious belief is somewhat different: soft disparatism

One might accept somey of the above characterisations of religious belief but not commit to hard disparatism. Call this position soft disparatism. Soft disparatists agree that the function served by religious belief is different, however, they usually think that religious beliefs aim at truth, albeit in a different way.

Here are two ways religious beliefs may aim at truth differently from standard beliefs:

• i. Praxis (πρᾶξις): religious beliefs aim at helping one connect with a community, satisfying a religious impulse, or according one’s behaviour with religious practices. The aim is not about determining if a proposition is correct in its characterization of the world.

• ii. Theoria (θεωρία) or contemplation: religious beliefs aim at helping a believer to contemplate the propositional content under consideration, not to definitively determine if the proposition is correct in its characterization of the world. Hence, when a notion or idea is contemplated, there is no commitment as to whether or not the proposition is true or false because the aim of the belief is to intellectually investigate the meaning of the proposition and not to definitively commit to whether or not the content accurately maps the world (Phillips, 1976, p. 181; Phillips, 1999). Soft disparatists might acknowledge that there are some religious beliefs where the aim is directed at what the content of a proposition is about, like standard beliefs, but they will consider these a small fraction of cases.

Armstrong (2010) argues that there is a modern preoccupation for religion to provide answers in the same way that naturalism does. What she, and others we deem ‘soft disparatists’ like Ellis (2017), mean by naturalism is not clear, just as it was not with the hard disparatists. Therefore, we also take them to be talking about a worldview that puts stock in empiricism, the natural sciences, consistent nomological laws, and a general kind of physicalism. Armstrong’s position is that religion does not aim at truth in the same way as naturalism, but that this does not entail that it does not aim at truth at all. We can enhance our understanding of this by reference to Crane’s idea in ‘The Unity of Unconsciousness’ that belief is a ‘worldview’ (Crane, 2017a, p. 9). On this perspective, individual ascriptions of belief (e.g. via propositions) model aspects of this worldview. The worldview is the means by which an agent navigates their environment. A worldview that aims at truth will be more apt for navigating the environment, thus, an agent that seeks to successfully navigate their environment will put stock in aiming at truth.

For Armstrong, the religious worldview is about praxis: the object and content of religious thought is about relating to religious content in practical instead of theoretical terms (Ellis, 2017). Religious beliefs functionally help one to connect with one’s communities and to inculcate a sense of belonging (Tillich, 1957). Indeed, some philosophers, such as Philip Goff, even go so far as to say that this sense of belonging and providing meaning is the only objective of religious belief. Goff refers to this as ‘religious fictionalism’ (Goff, 2022). Crane takes a similar line when he talks of one aim of religious belief being to aid a believer in belonging to a community, and another about satisfying the religious impulse (Crane, 2017a, p. 97). On this understanding, religious beliefs give rise to a function, and one can be more or less successful at implementing this function, depending on the accuracy of the truth-aim. For instance, if the aim of one’s belief is to belong to a safe, prosperous, and accepting community in eighth-century Norway then one’s belief in the importance of the rituals of say Scandinavian Heathenism, and one’s commitment to the worship of Thor, will likely help one navigate the world. This is more than mere ‘make-believe’ as Van Leeuwen puts it, because it has a factual goal: survival, belonging, cultivation of mental fortitude, and so on. Hence, this lends support to soft disparatism because there is still some kind of aim and a rational function.

The praxis-oriented account maps well theories of religion in the Wittgensteinian and pragmatist traditions. Wittgenstein noted that the meaning of language is not found in referential fidelity but in its use; what he called the ‘forms of life’ (Wittgenstein, 2001). Wittgenstein-inspired philosophers such as Phillips (1966, 1976), B. R. Tilghman (Tilghman, 1994), and Wettstein (2012) employ their interpretation of this Wittgensteinian insight to religion.⁷ Much of their analysis appears to support a kind of soft disparatism, for while they deny that religious belief is irrational or has any kind of truth-aim, they contend that the function of religious belief and language is to help an agent navigate social circumstances. Talk about God, prayer, the soul, and so on, has the function of achieving solidarity with, influence over, or connection to, other speakers. Howard Robinson summarises this interpretation of Wittgenstein (the ‘Swansea school’) thus (Robinson, 2003, p. 354):

Because of the acceptance of Hume and Kant’s critique of metaphysics, no religious statements can be accepted as descriptively true, on a par with descriptions of the physical world and the statements of science. Rather they have an entirely different function, being, roughly, expressions of value and attitude towards the world.

On such a Wittgensteinian account, the hard disparatist assertion that religious belief is irrational somewhat misses the mark because religious speakers are playing a language game, and to judge them by the standards of scientific or metaphysical frameworks is to misunderstand their function (Phillips, 1992), for a function they retain (Robinson, 2003, p. 356):

A positivist treats metaphysics as a form of discourse which is meant to be literally true but is in fact meaningless. The Wittgensteinian thinks it was never meant to be factual.

This is a clear example of a praxis-focused approach to religious belief. Of course, an alternative reading of this so-called ‘Wittgensteinian Fideist’ tradition could take the position that the ‘form of life’ of religious language is so distinct from standard beliefs that very little criteria is common between them. In a similar vein, pragmatic theories of truth propose that truth be defined in terms of utility, and are connected with the works of C. S. Peirce and William James. As regards religion, pragmatists reject the ‘metaphysical monster’ (James, 1902, p. 447) that is theological attempts at explaining religion as something that aims at truth based on evidence. The utility or function of religious belief could be construed as praxis: navigating social roles and aiming at belonging. Again, one might debate whether the aim is so distinct as to exclude religious beliefs from standard beliefs. However, both Wittgensteinians and pragmatists are ascribing this-worldly functions to religious beliefs that conforms well with praxis.

One might suggest that the communal or praxis aspect of religious belief is sufficient but not necessary, as the contemplative aspect is also sufficient. For instance, one’s belief in the importance of committing to a ritualistic personal hobby practiced alone on a mountaintop, will likely not help one navigate the social world nor satisfy their religious impulse and desire to belong to a community. This raises a question, how important is the aim at communal belonging for religion? Is not an Anglo-Saxon Christian, lost and alone in eighth-century Norway—as prisoner, missionary, or shipwreck survivor— religious? Crane would likely answer that the Christian missionary is still engaged in a communal practice, but it so happens that his community is on the other side of the North Sea. Also, the Anglo-Saxon perhaps wants to bring pagan Norwegians into his community. Yet then consider the case of the desert hermits and stylites (Doran and Harvey, 1992) who sought solitude precisely to enhance their religious experience. In German these people are called einsidelære (colonies of one); people who establish themselves in isolation whether that be geographically, socially, or mentally. They seek to be both religious and yet completely isolated from community, at least any community more tangible than an apostolic church.

This raises an important distinction between religion and spirituality. Both concern the numinous and numinal, but Crane contends that while the function of religion is related to community, the function of spirituality is not. An example of spirituality might be ietsism; uncommitted beliefs in an undetermined transcendent reality without community. ‘Religion’ captures those beliefs and practices that (largely) involve some communal aspect while ‘spirituality’ refers to beliefs and practices that are (largely) individual. Thus, even though both sorts of beliefs may have transcendental content or content about the numinous, soft disparatists contend that they come apart in terms of function.⁸ We might say the Norwegian pagan is religious but the desert hermit is spiritual. Alternatively, we might say that the Norwegian pagan is praxis-religious, and the hermit is theoria-religious.

We have just considered religious belief as having a functional aspect described as praxis, but as the hermit case shows, religious belief may also be distinct in that it does not aim at truth, but rather is a form of theoria. So, another way soft disparatists divide religious belief from standard belief is by claiming that religious believers do not always seek to determine the truth or falsity of a religious proposition. Again, this is different from hard disparatists who hold that religious belief is simply wrong, contradictory, or not able to aim at truth. Religious belief, on the theoria view, is a contemplation of a notion or impulse, rather than a theory to be proved. Crane says that the content of religious belief might importantly be about an indeterminate thing—about something that people cannot fully understand. Perhaps it is even essential that one does not fully understand it (Crane, 2017a). For example, people sometimes go through rituals even if they do not understand them. Religious belief on this view may be characterised as a kind of ‘alief’ (Gendler, 2008). Tamar Gendler (Gendler, 2008) provides examples of many such cases where we say we believe x but refuse to φ in accordance with that belief. For instance, one might affirm that they believe that a glass walkway overlooking the Grand Canyon is safe, and yet still fear to walk on it. Likewise, one might make a sign of the cross sans a belief that it will protect them, but they perform it regardless, just in case. The theoria account is different from the praxis account because the aim is not a practical navigation of the world predicated on truth-aim, but a theoretical function. Soft disparatists still contend that religious belief serves a function but the function is a contemplative one. According to this account, naturalism does not have a monopoly on reality and explanation, and the objective of religious belief is not to understand the world in the same way as natural science (Cottingham, 2014). Ellis suggests that divine properties are in their own sub-category, and just because naturalism does not analyse the content of religious belief, that does not mean that religious beliefs are completely different. This deftly avoids Van Leeuwen’s petitio principii. This is also suggestive of Quine’s (Van Orman Quine, 1964, p. 43) claim that we can distinguish between observational statements and theoretical statements.⁹ A soft disparatist might say that religious beliefs are theoretical but standard beliefs are observational, yet both aim at truth in their respective ways. In this manner, religious beliefs come out as somewhat different to standard beliefs because although there is a function and truth-aim present, the aim is divergent. We can appreciate this distinction with Wittgenstein’s example of ordering five apples, when he explains language games (Wittgenstein, 2001, PI, 1.1). Imagine, he says, that one goes into a store and orders ‘five red apples’. It is obvious that the referents of ‘red’ and ‘apples’ are clear. One might ask: but what of the meaning of the word ‘five’? Wittgenstein says that this is not the proper question, what is important is only how the word ‘five’ is used in the context (Wittgenstein, 2001, PI, 1.1):

And one has to say this in many cases where the question arises “Is this an appropriate description or not?” The answer is: “Yes, it is appropriate, but only for this narrowly circumscribed region, not for the whole of what you were claiming to describe.”

Maybe talk of numbers aims at truth in a different way to talk of apples? Maybe religious talk is like number talk? As Ellis puts it, religious belief is a certain mode or manner of understanding the world, and it has both a practical and theoretical component (Ellis, 2017).¹⁰ Practically speaking, religious belief satisfies one’s religious impulses and fosters a sense of community, and theoretically it seeks to explain value and the numinous through contemplation.

The case for praxis and theoria separating religious beliefs from standard beliefs has some force. Yet there are two reasons not to adopt a soft disparatist position. The first is that on a dispositionalist account one does not contemplate a belief; one contemplates a theory or idea, but not a belief. Indeed, Williams (Williams, 1973) says that we do not decide if a belief is true, we simply believe the world is a certain way and act accordingly. There is nothing voluntary about this: we do not determine that it is raining, navigate the world in response to this, and then decide that we believe that it is raining. The belief is the disposition that it is raining in conjunction with our desire to remain dry that directs us to grab an umbrella. Of course, once one becomes sceptical of one’s beliefs then one can contemplate the idea and perhaps decide to give up the belief, but contemplation is not itself a belief. Thus, when dispositionalists talk about the content of a belief, it is a tool to ascribe a mental state to someone else so as to make sense of the messy mental reality that informs how that person is navigating the world. This lends support to the idea that a good theory of belief will include religious beliefs in the picture because one cannot make sense of someone’s mental reality to predict and makes sense of their behaviour if the theory of belief does not make room for the correct, behaviour-guiding role of religious belief. Hence, a belief is a commitment that comes before the content, and this is not what is happening with contemplation.

The second reason not to adopt soft disparatism is that one can contemplate standard ideas that interact with one’s beliefs as well. Van Leeuwen says that we hypothesize that P, imagine that P, assume that P, and to this we can add contemplate P, entertain that P, strive to understand P, and so on. There is nothing uniquely religious about these functions. Thus, while praxis and theoria are relevant for some religious beliefs, they are not relevant to all religious beliefs nor only to religious beliefs.

We have shown that there are different ways of understanding soft disparatism: one can think that that the function served by religious belief is to belong to a community, praxis, or to contemplate ideas, theoria. This is the line Armstrong and Crane take. Ellis thinks that religious belief aims at truth in a contemplative manner, but not in a naturalist way. What these people share in common is a commitment to religious belief being different from standard beliefs, but not so very alien to standard beliefs as the hard disparatists suggest.

6 Religious belief is not different: non-disparatism

We have seen that the praxis-theoria divide lends some support to soft disparatism, however, the preferable view is non-disparatism: religious beliefs serve a similar function, and aim at truth in the same way as many standard beliefs, therefore, they do not differ from standard beliefs in any interesting way. We will present two lines of argument for non-disparatism. Firstly, we will argue for theological realism: the idea that the things that religious beliefs are about can exist independently of human beings (where appropriate), can be known, and can be spoken about truthfully. Secondly, we will argue that standard beliefs meet many of the criteria the hard-disparatists proposed as definitive of religious belief, and thereby dissolve the strict distinction.

The first line of argument is that many religious believers are theological realists. Theological realism is the view that the things that religious beliefs are about are entities, objects and properties outside of the mind, be they things like social properties and relations, physical objects, or non-physical laws or powers (excluding obvious mind-dependent cases). According to realism, what the content of religious beliefs are about depend on the way the world is externally (Stenmark, 2015; Wynn, 2020). This means that religious sentences have truth-conditions that determine their propositional content just like ordinary sentences.

Consider three theological traditions that are all realist: evidentialism, natural theology, and reformed epistemology. Reference to the kind of claims made by religious philosophers quickly shows that many of them take religious beliefs to have propositional content, the correctness or accuracy of which can be measured, that aim at truth.

Evidentialism is a view that for a person to be justified in some belief, that person must have some awareness of the total evidence available for the belief. The confidence a believer has in the given belief will correspond to the evidence (Taliaferro et al., 2012). Evidentialism implies that religious beliefs are justified only if there is conclusive evidence for the content of the beliefs (Forrest, 1997). Of course, there are different kinds of evidence, such as evidence of the senses, what is self-evident, introspective evidence, moral evidence, or recollected evidence from memory. Some might even count religious experience as evidence. Putting aside whether or not evidentialism is a correct thesis about justification for belief, it is the case that it is committed to a truth-aim that is descriptively about how reality is and there are theist evidentialists, such as theist evidentialist philosopher Richard Swinburne. Although many argue against theist evidentialism, or against evidentialism altogether, this does not detract from the fact that many religious believers, including philosophers of religion, take religious beliefs to be aiming at a committed truth about how reality descriptively is.

Natural theology is a philosophical approach that seeks to demonstrate God’s existence (Aquinas 1265-74; Braine, 1988; Craig, 1979; Miller, 1991). To show that a belief in God’s existence is justifiable, according to natural theology, it is sufficient to have a deductively valid argument with justifiable premises (with objections considered and defeated). Deductively valid arguments from justifiable premises are also taken as support for standard beliefs. Additionally, probabilistic arguments are also employed (Swinburne, 1979). Despite different approaches, all natural theology appeals to evidence and logical reasoning to conclude the truth of how the world is, and thereby determine whether or not a belief is justifiable. Even apopathatic accounts of the numinous take themselves to be aiming at truth in that they take the things that they are talking about to be descriptively true or false.¹¹

A third example of theological realism is found in reformed epistemology, a philosophical position that commits itself to the view that belief in God may be as properly basic as one’s ordinary beliefs about other persons (Plantinga & Bergmann, 2016). Thus, beliefs about God are warranted provided they are grounded—in whatever is taken to be the appropriate grounds—and defended against known objections. Indeed, reformed epistemologist Alvin Plantinga even argues that religion is more natural than secular naturalism (Plantinga, 2011, p. 60; Plantinga, 1993).

Sebastian Gäb (Sebastian, 2014) proposes two theses in support of a particular kind of theological realism: (1) metaphors are irreducible in religious language; (2) their meaning is based on certain experiences. According to (1), metaphorical sentences are not true in a different way to literal sentences. This is because we usually do not regard utterances as devoid of meaning just because there is no way to explicate their content other than by repeating them. Here are two examples (Sebastian, 2014):

• 1. His voice sounds like Sinatra.

• 2. This fruit tastes like lemon.

How does one grasp the proposition expressed in these two sentences? According to the causal theory of reference, some experiential context is relevant for fixing the reference of terms like ‘water’ and not a set of descriptions (Devitt, 1981, p. 66). Likewise, the reference of (1) and (2) can be traced back to some experience, even though the metaphor is irreducible. Thus, although some content is metaphorical, it still refers to an experiential context that fixes the reference: in the case of (1) Ol’ Blue Eyes and (2) a lemon.

The same reasoning is applicable to much religious language. According to Gäb, religious experience is the starting point for understanding a lot of religious language and the same reasoning—that metaphors are irreducible in religious language and their meaning is based on certain experiences—is applicable. Religious language is replete with such irreducible metaphors, for example when Christ says ‘I am the door’ (Douay-Rheims Bible, 2011, John 10:9) or ‘I am the true vine’ (Douay-Rheims Bible, 2011, John 15:1), and these metaphors are intended to convey something true about the world. ‘By me, if any man enter in, he shall be saved’ (Douay-Rheims Bible, 2011, John 10:9); what could be more real than a claim about the elimination or salvation of one’s soul; one’s personal identity?

Consider the case of the testimony of Dr. José Maria de Almeida Garrett who was present at a notable religious event at Fatima, Portugal on October 13, 1917 (Martins, 1984):

The sun, whirling wildly, seemed all at once to loosen itself from the firmament and, blood red, advance threateningly upon the earth as if to crush us with its huge and fiery weight. The sensation during those moments was truly terrible.

The people present that day took this strange occurrence to be a religious experience, hence, their beliefs about the phenomena were religious beliefs. Dr Garrett adds (Martins, 1984):

All the phenomena which I have described were observed by me in a calm and serene state of mind without any emotional disturbance. It is for others to interpret and explain them. Finally, I must declare that never, before or after October 13 [1917], have I observed similar atmospheric or solar phenomena.

We can appreciate that Dr Garrett’s beliefs about the events of October 13, 1917 are religious but also about the truth of the world; how to make sense of it and navigate it. His beliefs are about an external event and object, the operations of the sun, and about the religious aspect of the event and object, even if the meaning of the truth-conditions that determine their propositional content is only reducible to the events at Fatima.

Gäb holds that metaphors can be irreducible in religious language and their meaning is based on certain religious experiences. Therefore, theological realism postulates the existence of objective truths about religious entities, but not that we necessarily have linguistic access to them. This is the strongest case for non-disparatism: that religious beliefs aim at truth, even if many of those truths are complex or abstract in such a way that the language is metaphorical, unverifiable, or imprecise. Certainly, God, Vishnu, Allah, and Buddha, are all real entities to which many religious believers intend to refer to when they talk about them, even if they appreciate the limits of language to describe concepts outside of space–time or that perhaps elude human cognitive capabilities. The same is true of less precise terms such as ‘heaven’ or ‘saṃsāra’ (संसार). This means that religious beliefs are in fact corrigible. If one pushes a harder line and thinks all metaphorical language is suspect or incorrigible, then as we shall see, this applies to language about standard beliefs, hence, even if this argument holds, it prevents a clean delineation between standard and non-standard beliefs. It also leaves us laying in the narrowest of procrustean beds.

The second line of argument is that standard beliefs and religious beliefs satisfy many of the same criteria, and are not definitively distinct. Recall criteria (1)–(13) for the definition of religious belief provided by the hard disparatists, in section three:

• 1. Distinct in ritual, material, doctrinal and philosophical, experiential and emotional, social and institutional, narrative, and mythic, matters (Geertz, 2016; Smart, 2000).

• 2. Symbolic, not literal (Evans-Pritchard, 1976).

• 3. Group beliefs that can contradict individual beliefs (Astuti & Harris, 2008).

• 4. Contextual (Astuti, 2015; Barrett & Kiel, 1996).

• 5. Characterised by conviction, often involving an initiatory declaration of orthodoxy (Ruel, 1982, p. 109).

• 6. Opaque: they remain mysteries, and rely on ritual deference (Crane, 2017a).

• 7. Half-understood ideas (Naumescu, 2011).

• 8. Unreflectively held without attention to reasons to believe (Sperber, 1997).

• 9. Minimally counter-intuitive: they violate our intuitive naturalist ontology (Boyer, 2001). They are ‘superempirical’: things that are not the product of empirical things nor need be understood empirically (Schilbrack, 2013).

• 10. ‘[A]ll explicit and implicit notions and ideas, accepted as true, which relate to a reality which cannot be verified empirically’ (Wiegers & Platvoet, 2002).

• 11. Irrational (Kunin, 2003) or ‘irrational’ imitation (Naumescu, 2011): the property alleged to cause an individual’s behaviour is empirically unsupported.

• 12. Often contradictory (Astuti & Harris, 2008, p. 144).

• 13. A set of required actions (C. Smith, 2017; Vásquez 2011).

Note that not all these criteria cohere with one another, so a religious belief should satisfy only one or more, but not all of them. This conforms with our polythetic approach. Upon consideration it becomes apparent that there are applicable standard belief cases for each, hence the above features fail to delineate religious beliefs from standard beliefs in an interesting way. Distinctions (1), (2), (3), (5), (10), (11), are all applicable to archetypical cases of non-religious beliefs. Beliefs that align with (1) apply as readily to standard beliefs about government as they do to religion, as anyone who has attended a citizenship-awarding ceremony can attest. Standards beliefs can also be characterised by conviction and declaration of orthodoxy, say to a shared jurisprudential justification for a legal system or ‘our democracy’, per (5). As for (2), (10), (11), and (12), standard beliefs can hold even if unsupported. One example is the current debates about the so-called adoption problem in logic which is worth exploring. Saul Kripke concludes from Lewis Carroll’s paper, What the Tortoise Said to Achilles (Carroll, 1895) that certain basic principles of valid inference, such as universal instantiation and modus ponens could not possibly guide our inferential practices, because if one already infers in accordance with them, then no adoption is needed, but if one does not infer in accordance with them, then no adoption is possible (Boyd, 2022; Devitt & Roberts, 2024; Padro, 2015p. 42). Unlike Quine, who holds that logical beliefs can be empirically supported, Kripke is claiming that logical beliefs—non-ordinary standard beliefs—are held a priori and are not empirically supported. Unsupported standard beliefs would also include contradictory beliefs, for instance a belief about the liar paradox (Pleitz, 2018). Even if one has religious beliefs without understanding them, as Crane suggests, this is not unique. For instance, standard beliefs such as obsessive compulsive disorder-informed beliefs may also be irrational or believed even if the subject does not fully understand why. Concerning (2), (4), (6), and (10), many standard beliefs only appear rational in their context (Spiro, 1966, p. 98), such as a coronation or the rules of cricket. Regarding (8) and (9), it is unclear why a belief conforming with global intuitions makes it more correct: firstly, intuitions are not always reliable (consider folk intuitions about gravity), and religious beliefs are no less minimally counter-intuitive than say Platonist beliefs about mathematical entities or fundamental metaphysical laws; secondly, even if conformation with intuitions is evidence for correctness, religious intuitions, such as a belief in a kind of ‘higher power’, are historically and statistically near-universal and conform more with intuitions than many standard beliefs—including naturalist beliefs—do. Further, intuitions are also not consistent across cultures, even for standard beliefs. Notably, Machery, Mallon, Nichols and Stich (Machery et al., 2004) presented a story modeled on the Gödel case from Naming and Necessity in a study of undergraduates’ intuitions, and proposed that their respondents’ answers showed that Westerners are prone to rely on intuitions that match the causal-historical picture of proper names, proposed by Donnellan and Kripke, while East Asians appear to be guided more by classical descriptivism. Finally, (13) aptly applies to moral, legal, and social beliefs.

One might counter that while there are non-religious beliefs that meet some of the above criteria, it is only religious beliefs that meet a critical cluster of them. For example, consider the religious belief ‘Jesus is the son of God’; it could be argued that relative to a given believer, all or nearly all thirteen criteria for distinguishing religious beliefs could be met by this belief. Yet there are examples of unambiguously non-religious beliefs that meet just as many of the criteria as stereotypical religious beliefs. The best cases are found in politics, such as opening sessions of Commonwealth parliaments, pseudo-scientific theories like Lysenkoism, or aspects of totalitarian ideologies such as juche sasang (주체사상). If religious beliefs do not have a monopoly on a particular cluster of these criteria, then they do not differ in any significant manner from standard beliefs. In the case of juche sasang which is an explicitly atheist worldview, one might interestingly have beliefs which count as religious even if the believer vehemently denies any religious content! There are also non-religious moral and metaethical beliefs, and these also meet Crane’s criteria at the beginning of investing life with ultimate meaning (Brown, 2024).

Another counter could be Van Leeuwen’s hard-line position that religious beliefs are essentially irrational. Yet as we have discussed, many religions are explicitly concerned with factual, corrigible, beliefs that functionally serve to aim at truth. One example is from Aztec, or Nahuatl, religion. James Maffie (Maffie, 2014) characterises the chief problematic of Nahuatl religion as the following question: how can humans maintain their balance on the slippery earth? For the Nahua, earthly life is filled with pain because we live upon tlaltic-pac, the ‘point or summit of the earth’ (Burkhart, 1989, p. 58). The Nahuatl proverb, ‘Tlaalahui, tlapetzcahui in tlalticpac,’ ‘It is slippery, it is slick on the earth,’ demonstrates how morally, epistemologically, and aesthetically, human behaviour is defined in terms of maintaining harmony in a world in flux. This captures a functional desire to maintain a truth-dependent balance. Thus, the function and aim of Aztec philosophy’s guiding question is not irrational, but thoroughly pragmatic and ‘this-worldly’.

A second problem for Van Leeuwen is that his arguments apply to standard beliefs. According to Eric Mandelbaum and Nicholas Porot (Mandelbaum & Porot, 2023, p. 77), a Van Leeuwen-style account encounters the issue that factual beliefs also seem to have problems updating (Anderson, 1983; Mandelbaum, 2019), hence, the argument that religious beliefs are incorrigible and standard beliefs are not, falters. Take Van Leeuwen’s own example. People who denounced their belief in the apocalyptic consequences of Y2K did not also denounce their belief in the power of computer programming and perceived societal dependence on digital technology, they simply updated their belief about the fine-grained causal relationship between computers and digital calendars post Jan 1, 2000. Likewise, a religious group can update specific unrealised eschatological predictions—beliefs not applicable to all religions—without changing their beliefs regarding the power of God to end the world, or a sacred text to be generally prophetic. Neither standard nor religious beliefs are completely incorrigible nor corrigible; political predictions regularly go awry without swaying many peoples’ political loyalties, likewise, the point of no return for action on climate change has been adjusted since the early 1970s without scientific consensus disavowing climate science as a discipline. Standard beliefs appear as corrigible or incorrigible as religious beliefs.

We also return to our earlier problem, that per (7)–(10), hard disparatists claim that the only way to have a non-contradictory, rational, and intuitive belief is if one’s foundational theory of truth is a kind of empirical-naturalism. This ironic faith in empirical-naturalism begs the question by carving religious and standard beliefs at an arbitrary joint, a joint that applies to other standard beliefs about a priori subjects. The difficulty justifying religious beliefs mirrors the difficulty justifying standard, metaphysical beliefs, as seen with the adoption problem (Geertz, 2016, p. 98). As we discussed, this was admitted by the hard disparatists par excellence, the logical positivists (Hahn et al., 1929, p. 10):

The metaphysician and the theologian believe, thereby misunderstanding themselves, that their statements say something, or that they denote a state of affairs. Analysis, however, shows that these statements say nothing but merely express a certain mood and spirit.

A final argument against ejecting religious beliefs from the standard belief camp, is that they both can employ identical reasoning but be classified differently due to context (Taliaferro, 2017). A Hindu might be a convinced physicalist who takes a physically-closed universe to be the most powerful, entity: Brahman (God) (Ram-Prasad, 2001). Likewise, an anti-theist might take the whole physically-closed universe to be the most powerful possible entity, but they don’t like using the G-word. In this case, the Hindu and non-Hindu follow the same reasoning and hold near identical beliefs—perhaps even with the same teleological commitments about fine-tuning and initial conditions—but one is religious and the other is not. Undoubtedly, this distinction is important, but both beliefs serve the same function and truth-aim, and both equally fall under the barrel of (1–13).

We are not disputing the distinctions people like Van Leeuwen make between religious and non-religious belief; there is distinct content, and praxis and theoria might be more or less present in one of these categories. Rather, what we are doing is pointing out that whichever way hard disparatists cut beliefs, standard beliefs can also be carved along these same lines. So let us carve along the grain, that is the superior line of division between ordinary and non-ordinary. Of course, we are not denying their reading of empirical data, nor their analysis of different kinds of beliefs. Yet as standard beliefs can operate in functionally identical ways, and can also aim at truth in the same manner as religious beliefs, the only way to neatly separate religious belief into its own category is by committing a petitio principii. There is also an additional moral here: it is important to listen to how religious people characterise their beliefs, rather than approaching the issue with a presupposition that these beliefs are incompatible with naturalism, and that naturalism is true. The latter approach is reminiscent of the famous event at the Battle of Copenhagen when Admiral Horatio Nelson held up a telescope to his blind eye and declared: “I really do not see the signal”. In fact, there are very many signals for those with eyes to see; there are instances where religious people and philosophies are explicit about their realist claims. If we are to properly grasp what religious beliefs are, let us look upon them with an unjaded and seeing eye. In this way, we can come to a more accurate understanding of belief generally, by informing the concept with all cases and kinds of beliefs. In this manner, we thereby also challenge the hangover from logical positivism that non-naturalist beliefs are irrational or strongly distinguishable from naturalistic beliefs.

7 On faith

Is the term ‘belief’ homonymous or polysemous? Homonymous expressions are words with no relation beyond identical phonology, e.g. a ‘Pole’ refers to a Polish person, and sounds identical to a ‘pole’ which flies a flag, but there is no shared etymology or meaning. Polysemous expressions are single words with multiple meanings, e.g. a ‘mouth’ can belong to a person, or a cave, but both uses share the same etymology: the latter originated from conventional use associated with the former (Amaral, 2008; Devitt, 2004). On the one hand, ‘belief’ could be homonymous: when a Muslim says they believe that ‘There is no deity but God’ they mean something completely different from someone who says ‘I believe that exercise is good for my health’. Similarly, when person A says ‘The bank is wet’ they mean a riverbank, but when person B says ‘The bank is wet’ they mean the floor of a building. On the other hand, ‘belief’ might be polysemous: when a Hindu says they believe that Lord Krishna appeared to Arjuna (Vyasa, 2000) they are using the same word in a slightly different way than when they say ‘I believe that it is raining’. The homonymous view lends support to disparatism: religious beliefs are non-standard beliefs because the use of the word ‘belief’ has almost nothing in common with standard beliefs. The polysemous view can support soft disparatism: religious beliefs may be non-standard beliefs because the word ‘belief’ has a different meaning in religious contexts. Yet the polysemous view can also support the position that religious beliefs are non-ordinary standard beliefs. For example, someone might spend their whole life up to time t1 believing that certain kinds of yellow minerals are ‘gold’, only later at time t2 to discover that some are ‘gold’, aurum Au79, but others are ‘fool’s gold’, pyrite FeS2: ordinary gold and non-ordinary gold. In both cases they mean a certain kind of yellow mineral, but after time t2 they can be distinguished at a more fine-grained level by reference to their chemical composition. So it may be that religious beliefs are different to ordinary beliefs, yet are still standard beliefs, because belief is a term that is polysemous between various ordinary and non-ordinary meanings.

One way to test this would be to translate ‘belief’ into different languages (Bach, 2004; Kripke, 1977). For instance, Kent Bach says that if the definite article ‘the’ is ambiguous in regards to referential or attributive use, one could suspect it would be ambiguous in other (non-English) languages too, and in exactly the same way (Bach, 2004, pp. 226–267):

It would be a remarkable fact that an ambiguous word (‘the’ in this case) in one language should have translations in numerous other languages that are ambiguous in precisely the same way.

Does ‘belief’ mean the same thing in Vedic traditions as it does in Semitic traditions? Perhaps sceptics might be willing to accept that some religious beliefs are standard, for instance, they might assent that Vedic scriptures are more akin to philosophical approaches to carving the world at its joints, given the tradition’s conception of itself as committed to a kind of proto-naturalist explanatory mission. However, said sceptics might still differentiate other, more incredulous religious beliefs from standard beliefs. If we translate ‘belief’ into Sanskrit we must distinguish between ‘मत’ which we can translate as ‘opinion’ or ‘(standard) belief’ and ‘विश्वास’ which means something closer to ‘faith’ or ‘trust’. This Vedic account of ‘belief’ is distinct from a translation of ‘belief’ into Hebrew ‘אמונה’ (emunah) which translates to ‘faith’ or ‘faith in God’. This translation exercise shows that perhaps some religious beliefs are distinctive because they are faith-based or at least have faith as a load-bearing criteria. Translation may reveal that it is these kinds of religious beliefs that are interestingly different from standard beliefs.

Not all religious believes involve an aspect of faith, however, it could be the case that those religious beliefs which do are not standard beliefs. For many people ‘faith-based’ is synonymous with unreasonable, and ‘non-faith-based’ is synonymous with reasonable. On this dichotomous account the categories of belief will be interestingly distinct. However, if we take a more nuanced approach to faith, the distinction is not so obvious. It is true that for some people and groups, such as rare fundamentalist Protestant Christian denominations, faith can mean something like ‘trust in something despite, independent of, or in opposition to, reason’. Perry Miller in The New England Mind provides the following description of a Puritan understanding of God which comes close to this notion (Miller, 1939, p. 10):

The Puritan God is entirely incomprehensible to man. The Puritan system rests, in the final analysis, upon something that cannot be systematized at all, upon an unchained force, an incalculable power[. . . ]He is a realm of mystery, in whom we may be sure that all dilemmas and contradictions are resolved, though just how we shall never in this world even remotely fathom. He is the reason of all things, and though men can “explain” the behavior of things, they cannot pretend to expound the reason of reasons.

Fideists and Fideist-adjacent philosophers such as Søren Kierkeegaard (Kierkegaard, 1843) provide an apt case. Call this kind of faith F-faith.

Faith, from Latin fides which means trust, can be as much trust in something because of evidence as it is despite evidence. Thus, for other groups, for example Roman Catholics, faith means something more akin to an inference to the best explanation (Swinburne, 1996). Pope Pius X (Pius X 1910) defines it thus:¹²

I hold with certainty and sincerely confess that faith is not a blind sentiment of religion welling up from the depths of the subconscious under the impulse of the heart and the motion of a will trained to morality; but faith is a genuine assent of the intellect to truth received by hearing from an external source.

Call this kind of faith I-faith. On this definition, faith is more like trust in the whole being coherent even if bridging information is missing, like someone trusting that a puzzle can be solved even if the solution eludes her. At least some appeals to faith, I-faith, are forms of abductive reasoning. Abductive reasoning is (by most accounts) rational.¹³ Therefore, an appeal to faith alone does not explain why religious belief is unique, provided standard beliefs also occasionally appeal to abductive reasoning. We also here re-encounter the lesson of letting the religious describe their own beliefs rather than ascribing false definitions to them, such as ascribing all religious belief as involving F-faith.

A third kind of faith is as a covenant or contract (see Miller, 1939, Ch.XIV), C-faith. Consider God’s various covenants with the children of Abraham throughout the Old Testament/Torah. The contract is a conditional relationship between God and the Israelites: if they behave correctly (praxis) then they earn His favour, and if not, His wrath. To have faith on this account is to agree to the contract, to functionally aim to fulfill the contract, and to trust that the contract will hold. This covenant explanation is evident in standard-beliefs, for instance legal contracts, trusts, fiduciary duties, and even social contract theories informing the legitimacy of governments.

This appeal to faith does not successfully show that religious beliefs are not standard beliefs, for not all religious beliefs are faith-based, and not all standard beliefs are non-faith based. One might have religious beliefs for a host of non-faith reasons, such as acceptance of a cosmological argument. Conversely, many standard beliefs can rely on F-faith, I-faith, or C-faith. Some standard beliefs can be classified as employing F-faith, such as a parent’s denial of their murderous child’s guilt despite overwhelming contrary evidence or a lover’s unconditional trust that their serially unfaithful and deceptive spouse will remain true in the future despite all past evidence contradicting such a supposition. Some standard beliefs employ I-Faith, such as one’s faith that an aeroplane will safely land, despite one’s insufficient knowledge about aerodynamics.¹⁴ And of course, many philosophers suggest that we daily operate under the auspices of social contracts, involving C-faith. We can conclude that faith is an aspect that can distinguish types of ordinary and non-ordinary standard beliefs and religious beliefs, but it is of little use as a tool for separating standard from religious belief.

8 Religious beliefs are standard beliefs

Is there any interesting difference between standard and religious beliefs? We analysed three different answers to this question: (1) religious belief is very different (hard disparatism); (2) religious belief is somewhat different (soft disparatism); and (3) religious belief is not different (non-disparatism). We concluded that religious beliefs are not different from standard beliefs in any significant manner; they are mental, dispositional attitudes that aim to provide a correct representation of the world. Therefore, religious beliefs are best characterised as non-ordinary, standard beliefs.

Aside from presenting a more accurate account of religious belief, two additional insights follow from this analysis. First, we learned that if we wish to better understand what beliefs are then we ought to take all different kinds of beliefs, including religious beliefs, and analyse them together. In this way we can account for all kinds of beliefs, including some of our most deeply held, informative, and meaningful beliefs, and gain better understanding of how people employ the concept. Second, we learned that we ought not posit a stark division between standard and non-standard beliefs because there is insufficient reason to support such a distinction. By rejecting this dichotomous approach, we can thereby reject a dogmatic classification of religious beliefs (and some metaphysical beliefs) as peculiar, incredulous or even irrational. In these two ways, we have made progress in our understanding of what belief is, and may now dare to say: credo in unam credentiam.¹⁵