How to Measure the Firmness of a Belief?

Abstract

One of the more well-known of Wittgenstein’s thoughts about the nature of religious beliefs is that we go wrong if we try to vindicate or refute religious beliefs in the same way as we do in the sciences. This may make it seem as if Wittgenstein held a view where the world can be divided into two separate spheres, one hard, objective, world of facts where beliefs are held because we have proof for them, and another subjective, softer, vaguer, where our beliefs cannot be proven and are held for completely different reasons. Religious beliefs would thus fall into the second category. In this text, I will argue (1) that even though it is true that Wittgenstein did not think that religious beliefs were on a par with scientific beliefs (held for similar reasons, vindicated in similar ways), he nevertheless did not divide the world into two (in the above mentioned way); and (2) that Wittgenstein’s reflections on the nature of religious beliefs tells us something important about what it means to hold a belief (in general) that challenges several predominant theoretical views about beliefs. I will, with some help from C.S. Lewis, try to show that thinking about the differences in beliefs according to the predominant model—where the “beliefs” are fundamentally different in a scientific and a religious idioms, which leads us to think that one of them has to be endorsing the right, true, belief; or that they are incommensurable—is a model that misrepresents the “conflict.” The matter may not be as intellectual as one may be prone to think—given that the concept of “belief” is at the center—but may rather be best understood (and, hence, the difficulties most efficiently overcome) if we learn to exercise other features of our experience. In particular, we need to learn how to listen and look at things that sound and look strange. A self-critical training of one’s ears is what is needed. (And for these reasons, the article starts in a different register than one might expect.)

A young man meets a girl. The whole world looks different

when he sees her.

(C. S. Lewis)

1. Dissonances

It is part of the grammar of surprise that it constitutes a breach with the anticipated. It’s part of the grammar of anticipation that things are repeated. The moment of surprise is thus a testament to both repetition and the possibility of traveling down new tracks. Resisting, or rejecting, or being unable to hear, or see, the breach is thus a silent way of expressing comfort in what one is used to (do, think, believe).

Ludwig Wittgenstein’s remarks about the shortcomings of Mendelssohn’s music strike a note here. Mendelssohn’s music is, as Wittgenstein describes it, “not a peak, but a plateau” (Wittgenstein 1998, p. 4). What is missing in Mendelssohn is not knowledge or talent. It is the fact that everything is in its right place. There are no surprises. “What is lacking in Mendelssohn’s music? A ‘courageous’ melody?” (Wittgenstein 1998, p. 40). A “‘courageous’ melody” is something that allows for surprise; that is, it is not afraid of the breach. Musically, this often means that some notes or chords are allowed to travel outside the key or expand the scale in unexpected ways. When this is done well, the breach will, after a moment or two, feel like an absolutely natural part of the song. What initially rubbed the wrong way, all of a sudden, not only fits, but heightens. “Within all great art there is a WILD animal: tamed” (Wittgenstein 1998, p. 43). Mendelssohn never allowed himself to let the wild animal in. “If one wanted to characterize the essence of Mendelssohn’s music one could do it by saying that there is perhaps no music by Mendelssohn that is hard to understand” (Wittgenstein 1998, p. 27), and that makes Mendelssohn a “‘reproductive’ artist” (Wittgenstein 1998, p. 43). Since every single note is anticipated, there is, in a sense, only repetition.

In more contemporary, popular (or semi-popular) registers, Radiohead is one of those bands that stand out as being non-Mendelssohnian. One of the reasons that one can argue that they are a good band is precisely their capability to produce melodies and chords that break our expectations, often go off key, and then, after a while, start to feel like the right notes, and that all the expected notes would have been wrong (though “theoretically” more “correct”). There are things to say about this that could excite music theorists—about diminished chords, key changes, dissonance, and so on—but one does not need theory to experience the unexpected. These features do not necessarily make music good or bad (you can go off-key and still make very bad music, and you can make very good music that doesn’t), and the recognition and acknowledgment of the things that are happening does not necessarily make you like it. Dissonance is not pleasing to all ears. But Wittgenstein’s comments reach far beyond “like” or “not like.” His problems with Mendelssohn and the lack of courage in his music—that it is merely reproductive—is that it thereby also lacks what Wittgenstein calls depth (Wittgenstein 1998, p. 43). On the side of the listener, being courageous means that one may have to dive into the disharmonies, welcome them, and seek to understand that which constitutes the breach with our anticipations in order to discern the depth that they harbor. Radiohead’s “How to disappear completely” from Kid A (Radiohead 2000) may, for example, sound perplexing and disturbing to many. The song starts with a rather high-pitched, noisy chord/stringsound, and when the more conventional chord progression emerges it is in a completely different key, which constructs a very disturbing, eerie, dissonance. This dissonance continues (messing with our heads) throughout the first verse and disappears only when the song reaches the chorus for the first time. If one looks at this dissonance “objectively” it is, one may say, simply wrong. Musically, theoretically, this high-pitched sound does not fit.1

But if one “reads” the dissonance as an expression of the lyrics of the song, as giving the words another voice, things change drastically. The song is about the effort to disassociate from that in the world which causes harm, or the overwhelming nature of our present reality; and the dissonance fades away when the lyrics reach the chorus: “I’m not here/This isn’t happening/I’m not here/I’m not here.” It may not sound “good” (meaning “pleasant” in a banal sense), and one may not necessarily have to like it, but the dissonance helps to create the depth of the song.

There are three main reasons why I begin a text about what it means to hold a belief (and a religious belief in particular) with this short excursion to Wittgenstein’s remarks about Mendelssohn’s music. The first concerns the grammar of surprise and anticipation; the second concerns depth and seriousness; and the third is related to the fact that some expressions need to be “heard” in the right way to be understood. What is important here is not the possibility of theoretical elucidations of musically surprising notes, but the fact that they appear in this way to all who listen to (and not merely hear) the songs. When our expectations are denied, what we took for granted is revealed. Thus, anyone who listens to this kind of music will have his or her background assumptions unearthed. The revelation of what is not anticipated brings our pre-judgments (prejudices) into view, and it also reveals the level of our readiness to allow them to be questioned. Mendelssohn lacks, as Wittgenstein argued, courage (a concept that wears morality on its sleeve) because he was too afraid to move anywhere outside his own inherited expectations, and he thus also exhibits a reluctance to self-scrutiny. These things are all crucial for understanding Wittgenstein’s reflections on the nature of religious beliefs.

I want to show that the “silently assumed,” or “the inherited background” (Wittgenstein 1969, §94; cf. Murdoch 2003, p. 260) to our thinking, are things that are of utmost importance (yet ever so hard to bring into view) for a sound philosophizing about the nature of beliefs (religious and non-religious). Background beliefs, as the most fundamental, as unwarily “accepted” (scare quotes are necessary here), are not beliefs that we hold for a number of “reasons”—but they are nevertheless at work in the background, as the background, and the more deliberated and rationally adopted beliefs depend on these. This means that it is a sign of intellectual courage to be ready to address, if not welcome, the unanticipated when it confronts us, and do so in a self-critical manner. Furthermore, the effort to simply reject the unanticipated because it does not square with how I want things to be, or how I have been taught to order the world (Mendelssohn), is a sign of lack of depth. Understanding (the other) and self-understanding are thus intimately intertwined. Understanding as self-understanding is also, importantly, moral in nature (or, at the very least, not “outside ethics,” wherever that is supposed to be).

This brings us to Wittgenstein’s frustrated “response” to Frazer’s anthropology.

2. Savages

Frazer is far more savage than most of his savages, for these savages will not be as far removed from an understanding of spiritual matters as an Englishman of the twentieth century. His explanations of primitive practices are much cruder than the meaning of these practices themselves.

(Wittgenstein 2020a, p. 44)

Frazer is, as Wittgenstein sees it, blocked because he is unable to approach the unexpected in the spirit of seeking to understand, which includes being ready and willing to see one’s own prejudices or pre-judgments, the background to one’s thinking, as prejudices (or pre-judgments) that guide us unwarily. What is required is this moral effort of allowing oneself to question that which one took for granted, that has now been exposed. There is thus a sense in which Wittgenstein’s complaint against Frazer is that he never allows himself, his view of the world, to be exposed, let alone (self-)criticized. The (loosely speaking) “colonial” attitude is, in that respect, connected to a form of laziness or lack of interest in self-reflection.

Wittgenstein’s more precise diagnosis is that “Frazer’s representation of human magical and religious notions is unsatisfactory: it makes these notions appear as mistakes” (Wittgenstein 2020a, p. 32). Frazer is, one may say, too comfortable within a form of Western (British) culture, a fairly scientistic worldview, and that makes it difficult for him to hear religious or magical expressions in their right register. For him, religious and magical expressions only appear as dissonance, as discords that do not fit with “everything we know”—and “know” is here given a very specific meaning. He takes religious beliefs to be similar to scientific hypotheses. The critical point here is not that Frazer is wrong to think that science is a good and fruitful way of explicating features of our reality; it is that he is unable to hear that he is trying to sing in a different register from the ones he is “observing.” Their notes do not fit his tune. To return to the original analogy: Frazer sees/hears dissonance and proclaims “error!”; but he is unable to think that maybe the breach with his expectations actually meant something, actually might have carried genuine weight.

Therefore, the magical rituals he encounters all appear as mistakes or errors of a specific kind; they look like endorsements of faulty statements or hypotheses, as if their function were to give explanations of natural phenomena. In response to this attitude Wittgenstein, perhaps surprisingly, stresses that explanations are “too uncertain” (Wittgenstein 2020a, p. 34) in this context. I’ll get back to this in Section 4 below, but the short answer to why he sees explanations as too uncertain is precisely that they are hypothetical. As such, they are possible to test. We also have a clear idea about how to perform these tests, and what would count as evidence and results. Wittgenstein suggests that this is a fundamental misunderstanding of what a religious belief is, and the possibility of genuine understanding is thus lost already at the descriptive level. There is, in his view, nothing hypothetical about religious convictions.

But why shouldn’t the fact that some religious beliefs seem to contradict everything science teaches us, teach us that the religious belief is, plain and simple, a falsehood?

A first way to tackle this problem is to start thinking a little bit about the concept of “miracle.” Frazer, in Wittgenstein’s understanding, dismisses the magical practices of cultures foreign to him because they endorse false hypotheses about our physical reality—and this is a line of thinking that I think we have all encountered in both everyday conversations and contemporary philosophical settings. But if we think a little bit about the roots of a concept such as “miracle,” it quickly starts to look strange to say that people who believe in miracles do not believe in, say, natural laws. Indeed, as C. S. Lewis pointed out, the concept of miracle feeds on a belief in a natural order of things.

We must not say ‘They believed in miracles because they did not know the Laws of Nature.’ This is nonsense. When St Joseph discovered that his bride was pregnant, he was ‘minded to put her away’. He knew enough biology for that. Otherwise, of course, he would not have regarded pregnancy as a proof of infidelity. When he accepted the Christian explanation, he regarded it a miracle precisely because he knew enough of the Laws of Nature to know that this was a suspension of them. When the disciples saw Christ walking on the water, they were frightened: they would not have been frightened unless they had known the laws of Nature and known that this was an exception.

(Lewis 2014b, p. 8f.)

Thus, there is an important link between a belief in the natural order of things and, say, miracles. (You can’t have anything supernatural without having something natural.) Lewis was, just like Frazer, an “Englishman of the twentieth century” (Wittgenstein 2020a, p. 44). I do not know of Frazer’s personal relation to religion, but that does not matter much. The point here is to mark out the fact that miracles and religious rituals may only look like “bad science” if one hears them in the wrong register, and to note that thinking of the religious as simply out of tune depends on a failure to see that what rings false appears dissonant in relation to everything else. There’s no dissonant note without another next to it. Frazer, and the person who insists on saying that religious beliefs and magical rituals are simply bad science, are thus revealing themselves as thinking that anything that is off-key is wrong. Wittgenstein responds by saying that maybe the dissonance is the point! Frazer is Mendelssohn; Wittgenstein is Radiohead.

3. Believers

In Wittgenstein’s view, religious beliefs are different from scientific beliefs, at the very least, in the way we vindicate or disprove them. The famous discussion of believing in a Last Judgment from the beginning of “Lectures on Religious Belief,” points in that direction.

Suppose someone were a believer and said: “I believe in a Last Judgement,” and I said: “Well, I’m not so sure. Possibly.” You would say that there is an enormous gulf between us. If he said “There is a German aeroplane overhead,” and I said “Possibly I’m not so sure,” you’d say we were fairly near.

It isn’t a question of my being anywhere near him, but on an entirely different plane, which you could express by saying: “You mean something altogether different, Wittgenstein”.

(Wittgenstein 2007, p. 53)

We will come back to the idea of an “enormous gulf,” but we need to understand the ways in which the notion of “belief” works here first, for I think it will be a misunderstanding to say that the person who believes in the Last Judgment and the person who does not (based on all scientific evidence) holds different opinions about the truth of the same belief. This idea assumes that, or makes it look as if, there’s one object with one propositional content—Judgment Day—that people disagree about. Not in the sense that they disagree about what it is, but that they disagree about whether “it” is true or not. Seen in this way, the “Western” science-oriented mindset holds the belief that the sentence “The Last Judgment will come” is false, whereas the religious person holds that the same sentence (with the same propositional content) is true.

In some superficial sense, this analysis has some truth to it. But what kind of truth that is, is not as easy to survey as it initially may appear. We are here approaching the difference between the believer and the non-believer as a difference in judgment about one sentence, and we are not thinking about that sentence as a personal expression, as an utterance made by someone for some reason in some context. And this is one of the core problems that Wittgenstein here brings into view: the belief that each sentence has one propositional content, or refers to one singular thought, is what enables us to say that science and religion simply hold the opposite views when it comes to the truth of this proposition. (For a very helpful discussion of these themes, see Diamond 2005.) Wittgenstein is suggesting that such an analysis misses the target. One can say it, of course, and one can see why it may make sense to think so, but it nevertheless misses the point. Part of the problem here is that such analysis is (at least a partial) rehearsal of the kind of tone-deafness that Wittgenstein saw as central to Frazer’s analysis of foreign cultures.

As often is the case in philosophy, our thinking goes astray for some reason, and only rarely because we simply have reasoned poorly. For example, the idea of life after death works in a similar way to miracles. You must know what (natural, biological) death is in order to give flesh to the religious idea of a resurrection or an afterlife. This point is hinted at in Wittgenstein’s claim that “If you ask me whether or not I believe in a Judgement Day, in the sense in which religious people have belief in it, I wouldn’t say: ‘No. I don’t believe there will be such a thing’. It would seem to me utterly crazy to say this” (Wittgenstein 2007, p. 55). Indeed, Wittgenstein claims that he “can’t contradict” the person who believes in a Last Judgment. So, the person who does not believe in the Last Judgment does not hold the opposite belief to the person who does. Contradiction requires that two opposing judgments are passed about one propositional content. But, “I believe in a Last Judgment” (said by a religious person) and “I don’t believe in a Last Judgment” (said by an adamant atheist) is not translatable to “x or not-x.”

We have seen that Wittgenstein argues that we will misunderstand these kinds of disagreements if we approach them in terms of evidence—as if the “x or not-x” could be determined by means of reference. The disagreements that we supposedly find here are very different from what he calls “normal controversies” (Wittgenstein 2007, p. 56). Wittgenstein suggests that in religious language, concepts such as evidence, hypothesis, probability, or opinion are simply out of place outside “normal” (often scientific) practices (Wittgenstein 2007, p. 56f.). Our procedures of justifying beliefs are completely different when it comes to religious life. In fact, Wittgenstein goes so far as to say that “if there were evidence, this would in fact destroy the whole business” (Wittgenstein 2007, p. 56).

This is partly why one would be reluctant to say: “These people rigorously hold the opinion (or view) that there is a Last Judgement”. “Opinion” sounds queer.

It is for this reason that different words are used: ‘dogma’, ‘faith’.

We don’t talk about hypothesis, or about high probability. Nor about knowing.

In a religious discourse we use such expressions as: “I believe that so and so will happen,” and use them differently to the way in which we use them in science.

Although, there is a great temptation to think we do. Because we do talk of evidence, and do talk of evidence by experience.

(Wittgenstein 2007, p. 57)

This is the point where it may start to look as if the problems regarding the differences between the normal scientific and the religious discourses are such that we must talk about a divided world.

It is clear that Wittgenstein held that we will go wrong if we try to vindicate or refute religious beliefs in the same way as we do in the sciences. But you can be a scientist, or a philosopher who thinks that everything we do in philosophy must be in concordance with the (natural) sciences, and still be religious. For such a person, there are two separate areas of life. This may make it seem as if Wittgenstein held a view where the world can be divided into two separate spheres, one hard, objective, world of facts, where beliefs are held because we have proof for them, and another subjective, softer, mushier, vaguer, where our beliefs cannot be proven, and are held for completely different reasons (cf. Forsberg and Conant 2013). Religious beliefs would thus fall into the second category. This, I will argue, is too simplistic.

I want to say that the idea of a doubled world is too simplistic, not because religious expressions are, after all, partly scientific (they are not, at least not beyond the sense that they sometimes take the form they do precisely by constituting a breach with our perception of the natural order); but because religious language and scientific language (and indeed ordinary language and literary language too) are not completely different languages (as if there were enclosed spheres, each with its own syntax and semantics). The ways in which they are interwoven are important to understand—one of the central points of Section 4 below. We need to understand some of the effects of the differences that we can find between a religious belief and a scientific one first. The view of two radically different spheres, controlled by entirely different “grammars,” is rooted, I think, in a failure to acknowledge that our lives in language are not made up of an affirmation of individual propositions that add up, somehow, to a “grammar.” Our beliefs, our practices, the ways we experience the world and our neighbors, are all part of the same fabric of life, and they do not attain their significance one by one.

The person who argues that religious beliefs in, say, the afterlife, or virgin birth, or magical rituals such as rain dances and animal sacrifices, are mistaken because they go against the facts proven by natural science, holds a specific belief about what it means to hold a belief. This kind of reasoning relies on the idea that there has to be one, and one only (so-called) “propositional content” of these expressions that they all, in some way or other, relate (or fail to relate) to. That is, “resurrection” means one thing, and one thing only; and the beliefs we have all relate to that, quite regardless of whether we are in a religious or a scientific context. (The number of theories that are some kind of variation on this thought is too large to go through here.) There’s also a predominant view of what a concept is that is in play here, suggesting that there the meaning of a sentence is its propositional content, and that that content remains the same from context to context (a form of invariantism)—and there are too numerous theories about these things too to go through in the limited space of this article as well.

One way to spell out this “enormous gulf between us” that was mentioned earlier is to say that the context of the religious belief is different from the context of the scientific belief, and this difference is what creates the gulf. This is, again, partly true and partly misleading. It is obviously true that the contexts of religious or magical or ritualistic practices are very different from what goes on in laboratories, offices, and seminar rooms. It is obviously true that “the grammar of religious beliefs” and the “grammar of scientific beliefs” are different in these two contexts, and that one cannot “employ” one grammar that belongs to one context in the other, or vice versa. But these are observations that ought to be treated with caution, for they are easily inflated. The risk is that one downplays the fact that one man can easily move between these contexts, indeed can live and be fully alive in these movements. The religious scientist is not a contradiction.

One of the problems here is that the idea of contexts being islands, and that one cannot move from one to the other. Such an understanding of contexts works on the assumption that the meaning (say, the propositional content) of the concepts in question is locked to their context/island, as if the meaning of our words would be lost if one found a sturdy enough boat and travelled the gulf between us. This is one of the reasons why I think it is dangerous to put too much money on the hinge metaphor (cf. Bagheri Noaparast 2025; Hoyt 2025; Coliva 2025) Hinge-epistemology has emerged as a central theme in contemporary Wittgenstein scholarship. The importance of hinge-theory is rooted in the fact that once one acknowledges the importance of context, it is easy to assume that each context has its own organizing principles, and this also makes communication between spheres difficult, if not impossible. I will suggest that the hinge-perspective “hinges” on assumptions about propositional meaning that their emphasis on context actually calls into question, and contribute to the misleading idea of fortified “spheres.”

Even though it makes sense to say that the propositions that stand fast for Rudolph Carnap are not the same as the propositions that somebody like Simone Weil’s thinking “hinges” on—it does not follow from that that these individuals are locked, trapped by the doors that hang on these supposed hinges, and unable to move towards each other. Indeed, as I will argue, it is misleading to say that propositions “stand fast” one by one. We can perhaps describe them individually, but they don’t come to stand fast individually. That there are some connections and lines to be drawn between various inflections of concepts and ways of seeing things is not a threat to rationality; rationality does not require one-ness.

It is important to see that—quite regardless of various common philosophical ideals according to which everything would be easier if every concept had one correct specified definition—it is quite obvious that we do live with different concepts (pertaining to the same word), and we are also not only OK with moving between these varying concepts (pertaining to the same word) in daily life, but we actually need those differences (sometimes nuanced and fine-grained, sometimes provocatively different) to achieve precision in our expressions, and sensitivity in our understanding. If you insist that “love” has to mean one thing, then you’ll lose the ability to understand the differences between love for one’s partner, one’s children, one’s parents, one’s God, one’s tractor. If you insist that these are all completely different, you lose the ability to see how they relate to each other. If we cling on too tightly to the idea that the scientistic soul and the religious mind anchor their lives on different hinge propositions, and that they, because of that, will never be able to meet, we are still working with an understanding of contextuality that is too narrow, and we are still trying too hard to tie meaning to a singular unit (like “the proposition”) and its criteria for correct application.

4. Lovers

I want to return now to Wittgenstein’s thought that explanations are “too uncertain” (Wittgenstein 2020a, p. 34); to uncertain, that is, for being relevant for efforts to explicate the nature of religious beliefs, their firmness. So far, we have only noticed that explanations are too hypothetical in nature, but I don’t think this is the only reason why they become too uncertain.

One aspect of Wittgenstein’s writing on these matters that deserves some attention here, though it may appear fairly insignificant, is the point at which Wittgenstein seems to resign, both in terms of what he, himself, finds himself able to say about his own views, as well as from the effort of explicating these differences “fully” about others. As we are confronted with what looks like a clash between a deeply religious person and one that is not, Wittgenstein denies the possibility that this could be settled, at least not from an external point of view.

One can only resort to description here, and say: such is human life.

Compared to the impression that what is so described to us, explanation is too uncertain.

Every explanation is a hypothesis. But someone who, for example, is unsettled by love will be ill-assisted by a hypothetical explanation. It won’t calm him or her.

(Wittgenstein 2020a, p. 34)

The first remark about how one can only resort to description is important. Some of the reasons why we experience a clash between religious and non-religious forms of understanding are due to failures at the descriptive level. But this is not a dead end. Rather, this is where the descriptive work needs to commence. I assume many philosophers will discern a touch of resignation in the formulation “such is human life,” as if that merely means that we have now gone as far as we can, and all we can do now is shrug our shoulders and acknowledge, accept, or merely tolerate the other’s otherness. I don’t think this should be seen as a resignation from the responsibility to seek understanding. But we may have to work on our expectations.

The person whose idea of understanding entails that we need to come to the point where one of two alternatives is proven right, may not get what her or she wants. But that is not the real endpoint here. As I have tried to show, there are a number of philosophically troublesome assumptions in play here. One is the assumption that the meaning of expressions can be attained by looking at concepts as mere “representational devices” (The term “representational device” is borrowed from Cappelen 2018, p. 3; cf. Forsberg 2022b). Another is the intimately related effort to think that the sense of our expressions always can be adequately cashed out through and analysis of the sentence’s “propositional content,” which is something very different from an expression is a significant context if use (cf. Baz 2003; Forsberg 2022a, Ch. 4). A third is the related idea that “meaning” is something that belongs to the words themselves, and our “meaning it,” as well as the other’s effort to hear me in the right way, the Austinian “uptake” (Austin 1975, p. 116f.), are not really relevant. These three “views” are all at work in forming the further assumption that it is clear that the religious and the non-religious person “refer” to the same thing with a formation of words such as “Last Judgment.” So far, one may say that there are difficulties belonging to our understanding of language that block communication and add to the feeling of resignation here.

Further problems concern the concept of belief; what we think it is made of, as it were; but also how we “rank” our beliefs in terms of “firmness.” Here, philosophical views of how beliefs are grounded and need support from “evidence” may also block our imagination. It is easy to assume that if I can give reasons, or refer to natural facts or scientific data, my belief is stronger than the beliefs I have for which no such data are available, or accessible, or relevant. So, on the model Wittgenstein questions, “having a belief” means something like “being in a state of mind, of which one is aware, which has a clear content, and this content is grounded on evidence, and the more evidence, and the stronger the evidence, the firmer the belief.” Wittgenstein’s take on this issue is almost the opposite to the more traditional view, call it “analytic” if you want; for his view is that we will completely misconstrue the “firmness” of religious beliefs is we approach them through the lens of such assumptions (about language, the human mind, and what it means that a belief is “justified”). It is only in a limited sense that the religious and the non-religious person talk about “the same belief” when they talk about, say, a “Last Judgment.”

How should we compare beliefs with each other? What would it mean to compare them?

You might say: “We compare states of mind.”

How do we compare states of mind? This obviously won’t do for all occasions. First, what you say won’t be taken as the measure for the firmness of a belief? But, for instance, what risks would you take?

The strength of a belief is not comparable with the intensity of a pain.

An entirely different way of comparing beliefs is seeing what sorts of grounds he will give.

A belief is not like a momentary state of mind.

(Wittgenstein 2007, p. 54)

We are here pushed to question the idea that (all) beliefs are something we hold—as if having a belief is always a matter of holding a thought in one’s mind, evaluating it (rationally) and then, finally, endorsing it (if the conditions are right). I think there may be moments when it is plausible to think about beliefs in this way, but it is not right about all beliefs and not all the time. And the moment at which one may say to think about a belief in accordance with such a model is, strikingly, more in the mode of reflecting upon the belief than it is about having it, being alive in it. Sure, we can do that sometimes—“Do you believe …?”; pause for reflection; “Yes I do, and my reasons are…”—but it would be a mistake to take that outside-stance to one’s own beliefs as a picture of what beliefs always are like, and what they must “live up to,” as it were, in order to qualify as a true belief.

Wittgenstein also talked about religious beliefs as being “unshakable,” and they are so in a man’s life “not by reasons or by appeal to ordinary grounds for belief, but rather by regulating for all in his life” (Wittgenstein 2007, p. 54).

Now, the “Lectures on Religious Belief” were delivered in 1938, and as textual evidence they are not unproblematic, since they are not literally written by Wittgenstein but based on his student’s lecture notes. So, in terms of exegesis, caution is required. But it seems to me important that what we see Wittgenstein doing here is developing a more structural kind of thinking that comes to be prominent in the very late writings of his on certainty and G. E. Moore (Wittgenstein 1969; cf. Forsberg 2024). Wittgenstein’s “Remarks on Frazer’s Golden Bough” is also from the same period, one part presumably dictated in 1931; a second set of notes was written in 1936; and he returned to make remarks about Frazer as late as 1948 (Wittgenstein 2020b, p. 1f.).

Wittgenstein’s reflections on religious beliefs can thus be seen as one of the themes that led him to rethink the way that beliefs are held, and what it means to be certain about one’s beliefs, overall. What is suggested in all three of these texts is that “one belief” is rarely a singular entity that we evaluate and pass judgment about. They all belong to a movable and changing fabric. It is only in a limited sense, and seen from a rather peculiar perspective, that one can think that I have one belief. We can single beliefs out, and study them perhaps, but only at the expense of removing them from the fabric from which it attains its meaning and carries its genuine weight. A belief is, in that respect, an abstraction. Hence, a “hinge proposition” has to be so too.

It is for these reasons that Wittgenstein claims that “We must plow over language in its entirety” (Wittgenstein 2020a, p. 44). This means that the descriptive task mentioned above, which is called for, for example, when we come to see that we are out not in tune with one another, is no small task. The ability to say “such is human life” may require that we plow over human life in its entirety, or one human’s life in its entirety, in order to attain a genuine understanding of one’s, or the others’, beliefs—be they religious or not! (And, of course, this is a claim that needs some salt, since the idea that there’s a point at which “language in its entirety” is plowed through, or that there’s a point at which I can say “now I understand this other fully,” is, taken too literally, obviously empty.) The idea that religious beliefs have the character of “regulating” all in life (Wittgenstein 2007, p. 54), does not mean that there is one singular proposition that functions as a hub around which everything else revolves. It means, rather, that the religious beliefs (in plural, and in plural forms) are present in all areas of life, though not always as heightened.

This takes us back to the beginning. It can now be said that Frazer’s tone-deafness and the lack of imagination of the scientistic philosopher (that leads him to think that a proper understanding of our religious expressions can only be attained if we approach them in the same way as we approach beliefs about, say, the features and properties of medium-sized physical objects) are similar. They both err at the descriptive level. This is why the descriptive task is the crucial issue, and it connects with a certain tone-deafness —as if they are analyzing a song, note by note, not realizing that what gives each individual note whatever point, power, or push it may have comes from a visceral experience of the whole. So, the point is not the singular note, or belief, or a question of whether or not it “fits.” The descriptive task is the effort to learn to sing along.

What somebody like Frazer, or the scientistically oriented philosopher of religion, thus misses at the descriptive level, is not a matter of not trying (to see or understand the religious expression). What is missing is a form of self-reflection. The idea that the religious beliefs can and should be approached from those particular angles is not, itself, a singular belief that exists completely meaningful outside their form of life. So, it is by trying to hum their own tune, a tune that is in total harmony with their own wholes, their own background beliefs, that the other’s songs, or efforts to join in, simply ring false.

I have argued that it is problematic to lift a sentence from the messy muddle that is our lives and study it in isolation. But since we can do that (although hardly ever without losing the real meaning it carries), it is easy to go wrong in the required descriptive work. This means that how we approach a topic in our efforts to give rich enough descriptions that enlarge our understanding is absolutely crucial.

The passage that I used to highlight the necessity of the descriptive work about how our religious beliefs take shape in our lives ends with a striking little observation that highlights how we look as we seek to describe matter:

But someone who, for example, is unsettled by love will be ill-assisted by a hypothetical explanation. It won’t calm him or her.

(Wittgenstein 2020a, p. 34)

We can look at unhappy love from a strictly medical, physiological perspective. It is possible to describe movements in the brain, hormonal levels, and speak about how they may affect bowel movements, and so on. This is a way of “looking at” unhappy love. You can also seek to understand the importance of the other, the weight that the other person’s absent soul places on the unhappy lover’s shoulders. You can embrace the person suffering, talk, try to see what he or she sees, and listen. This is “looking along” with the other. And I think this offers us a much better way of approaching the descriptive task ahead of us, that the simplistic distinction between “the scientific” and “the religious.”

This distinction between “looking at” “looking along” is borrowed from Lewis’ short essay “Meditation in a Toolshed” (Lewis 2014a) from 1945. Lewis gives a number of examples to illustrate this distinction.

A young man meets a girl. The whole world looks different when he sees her. Her voice reminds him of something he has been trying to remember all his life, and ten minutes casual chat with her is more precious than all the favours that all other women in the world could grant. He is, as they say, ‘in love’.” Now comes a scientist and describes this young man’s experience from the outside. For him it is all an affair of the young man’s genes and a recognised biological stimulus. That is the difference between looking along the sexual impulse and looking at it.

(Lewis 2014a, p. 230f.)

The mathematician sits thinking, and to him it seems that he is contemplating timeless and spaceless truths about quantity. But the cerebral physiologist, if he would look inside the mathematicians head, would find nothing timeless and spaceless there—only tiny movements in the grey matter.

(Lewis 2014a, p. 231)

The savage dances in ecstasy at midnight before Nyonga and feels with every muscle that his dance is helping to bring the new green crops and the spring rain and the babies. The anthropologist observing that savage, records the that he is performing a fertility ritual of the type so-and-so.

(Lewis 2014a, p. 231)

To this list, we can add the strict musician, or musical theorist, who claims that it is a measurement of “good music” that it does not break the rules of music, in contrast with the curious young who discovers her own soul in the screams of a local punk band. These are all descriptions of the distinction between “looking along” and “looking at.” I think this distinction captures something that Wittgenstein is after quite well, and it does so without relying on simplified uses of words such as “scientistic” or “religious.” And it does so without suggesting that we have to choose between the “scientific” or the “religious” perspective. And it does so, without speculative notions of “hinge propositions” that organize two distinct “grammars.” Lewis’ examples also make it very clear that it would be strange to think that one of the sides has rationality, reason, and truth on its side, and the other one doesn’t, since looking along and looking at can be the perspectives of two different scientific endeavors. If we stick to a simplistic science vs. religion distinction, this will not be as obvious. It also helps us see that the fact that they can’t contradict each other does not mean that there’s no room for understanding. The idea of “different grammars” or “incompatible hinge propositions” leaves no room for conversation. But we can learn to “sing” and look along. I think this brings us closer to what Wittgenstein was after.

We may say that Lewis is also, so clearly, thinking along with Wittgenstein for example when he suggests—in a form of cultural criticism that Wittgenstein most certainly would have agreed with—that “It has been assumed that if you want the true account of religion you must go, not to religious people, but to anthropologists, if you want the true account of sexual love you must go, not to lovers, but to psychologists” (Lewis 2014a, p. 231). Of course, learning to sing, play, and look along may not be easy. Prejudices are one thing that plays a crucial role here. And this is precisely why a too simplistic division between “languages,” or “grammars,” or “hinge propositions” can lead us astray. As long as our opposition is between “scientific” and “religious” (or, for that matter, “superstition” or “magic), it is more or less impossible for the contemporary consciousness not to lean towards, if not side, with one over the other. What Wittgenstein is trying to make us see is precisely that those kinds of simplifications are leading us astray. I think that Wittgenstein’s thinking aligns perfectly with Lewis’s claim that “a great deal of contemporary thought is, strictly speaking, thought about nothing—all the apparatus of thought busily working in a vacuum” (Lewis 2014a, p. 233). And that is because they have only looked “at,” and never “along.” So, whereas it may be impossible to fully account for a religious life from a strictly naturalistic and scientific perspective: it is not impossible to learn to look along, and to acknowledge, with Lewis (and, I think, Wittgenstein) that “We must, on the pain of idiocy, deny from the very outset the idea that looking at is intrinsically truer or better than looking along” (Lewis 2014a, p. 233). In a different terminology, one could say that Wittgenstein and Lewis help us see that a pure “outside” perspective is an abstraction. But this does not mean that there’s a strict distinction between the outside and the inside perspective. It is (merely) the idea that the outside perspective is the true perspective to be prioritized that is criticized. It does not follow from that that we cannot take different stances, or that the two stances taken intermingle.

In sum: just because we can see that religious beliefs, as well as expressions of what Frazer studied under the name of “magic,” do not work in the same way as scientific statements do (in terms of representation and vindication); and just because religious beliefs require a rich contextual and anthropological perspective to be seen as meaningful, and that the descriptive work required to do that well is difficult and complicated; it does not follow that the “beliefs” that “we” have and rely on from within a Western scientistic worldview do not also belong in a larger fabric of practices and meaningful words and phrases and ways of looking at things; as if the nature of all beliefs is one that fits perfectly with our practices of formulating singular-sense-propositions that are possible to vindicate or refute by means of a very limited of number of possible practices—like measuring, weighing, calculating—of vindicating or refuting them.

There’s a mythology that we need to “plow through in its entirety” in magical and religious languages. There’s a mythology that we need to “plow through in its entirety” in the science-inspired Britishness of Frazer’s being.

To summarize: I began, a bit unusually, by a metaphoric built around the notion of hearing and listening, aiming to show that when there’s a breach in our expectations, we are also called to an exercise in self-critique, since the breach tends to reveal what we take for granted, what we expected to hear, and that understanding the other is thus often a moral matter. Section 2, “Savages,” showed that this kind of self-criticism was lacking in Frazer, and this is partly what led him to treat religious and magical expressions as faulty science. Part of the problem here is the assumption that there is a non-contextual propositional meaning that remains the same from context to context (e.g., religious and scientific). The problems with that assumption were discussed in Section 3, where I critically examined the idea that the world can be divided into two spheres and questioned the idea that each “context” has its own organizing principles (hinges). Central to my argument here is Wittgenstein’s thought that sentences, utterances, or propositions are not attained individually, and not meaningful individually, one by one—a thought that makes us see that understanding may require that we “plough through” language, or a life, in its entirety. This is also the guiding thought of this concluding section (4. “Lovers”). A result of this discussion is that if we tie the notion of rationality to tightly to the kinds of beliefs that we have adopted for a number of reasons, or because we have accepted the arguments speaking in favor of them, we end up with a too narrow understanding of rationality, and we fail to see that most of our most central beliefs do not fit this model. We must learn to hear the differences out in order to reach understanding and avoid projecting one model onto all cases.

In other words, the idea that religious life and scientific life are incompatible comes together with the belief that we are, almost in principle, unable to learn to listen differently; as if you could never learn to listen attentively in a religious register if your ears are scientifically trained. I think this is a dangerous simplification. While it is true that it may be difficult for a person with a Mendelssohnian mindset, who thinks that only music where everything is in its right place is good music to learn to appreciate, say, Arnold Schönberg’s atonal compositions, or the hard dissonant and monotonous growling of a post-metal band like Amenra, there’s nothing that makes the effort of learning to “listen along” or “looking” along with these impossible. These things are perhaps unlikely, but, logically speaking, they are not fundamentally different from the person who has studied love as a purely biological phenomenon, and convinced him or herself that love is merely enhanced activities of this or that area of the brain that has this or that effect on the nervous system as a whole, to fall in love. To truly fall and learn to “look along” with the other.