1 Introduction

The aim of this text is to consider Wittgenstein’s approach to religious faith, and to explain how he understood the difference between religious belief and non-belief.Footnote 1 In particular, I would like to focus on what it is, according to him, that philosophy can say about religious faith. As is well known, his conception of philosophy differs radically from traditional ones, and this difference also shows up in his treatment of the questions with which I shall be dealing here. For Wittgenstein, philosophy’s task is not to justify or question religious beliefs, but to understand them (cf. Phillips 1999: 163) – or point out that there are limits to understanding them. I will therefore try to show in the present text how Wittgenstein conceived of the sort of understanding which can be achieved by philosophy.

In popular discussions about religious belief and non-belief, an assumption is frequently made to the effect that it is quite clear what the difference between these consists in: that understanding the difference between them does not itself constitute a serious problem. Such an approach is also often accompanied by a conviction that the serious philosophical question is just that of who is right – the believer or the non-believer.

On the reading put forward here, Wittgenstein’s approach to religious faith represents a distinct alternative to the above. As I have already suggested, he holds that seeking to justify or question the various practices in which we are engaged or which we encounter in our lives is not a valid aim for philosophy – rather, its aim should be to understand them. Of course, an attempt to understand a certain practice may result in an elevation of status or diminution of significance as far as that same practice is concerned, but this would not mean that philosophy had somehow justified it, or proved it to be illegitimate. For example, some investigations within the realm of the philosophy of mathematics might serve to demonstrate that so-called “meta-mathematics” belongs to mathematics, and that it is itself, in fact, a set of related mathematical calculi – in which case it could not be a basis for mathematics as such. Such considerations could result in a downgrading of meta-mathematics, but one could not say of them that they had proved it wrong. My point will be that the connections obtaining between insights arrived at in the philosophy of religion on the one hand, and religious practices themselves on the other, can also to some degree be construed like this.

While aiming to present Wittgenstein’s conception of what is involved in understanding religion, I must nevertheless straightaway concede that this is not something that can be fully achieved in a single article. Therefore, I shall focus exclusively on those of his remarks concerning beliefs and statements of a religious kind that belong to his later writings and lectures.Footnote 2 Although I do not think there is any total break between his so-called “early” philosophy and his so-called “later” one, I do consider the latter far more mature, in that his later writings try to take into account the overall diversity and richness of our language games and forms of life. Thus, his later philosophy can be said to be more realistic, at least in the sense in which this term was used by Cora Diamond in her book The Realistic Spirit (Diamond 1991a).

The first question I consider is this: is there any such thing as Wittgenstein’s philosophy of religion? I will try to show that there is a sense in which his remarks on religion can be treated as an expression of a certain philosophy of religion. Then, I will focus on the four questions pertaining to the nature of religious beliefs and utterances that I take to be discussed in his writings and lectures: (1) Are there beliefs held by a non-believer that contradict certain of the religious beliefs held by a believer? (2) What are the differences and similarities between controversies relating to religious questions and those pertaining to scientific or everyday empirical ones? (3) How should the relationship between one’s acceptance of certain religious beliefs and one’s participation in a certain religious form of life be descriptively characterized?Footnote 3 (4) What role do pictures play in religion? Afterwards, my next step will be to show that Wittgenstein’s answers to these questions need not be interpreted as theses constitutive of some philosophical theory of religion. My discussion of this point, which I take to be crucial where giving a proper and adequate interpretation of Wittgenstein’s remarks on religion is concerned, includes an examination of the following objection to his philosophy of religion: namely, that his remarks on religious belief and religious assertions are incompatible with what believers themselves say about them. I shall try to show that this objection is either unjustified or does not represent a serious challenge to his philosophy of religion. I end my considerations by concluding that the aim of Wittgenstein’s philosophy of religion is, in a sense, a very modest one. His remarks can be understood as reminders which only try to present, and not to explain, those features of religious beliefs and utterances that are right before our – i.e. both believers’ and non-believers’ – eyes.

2 Is there any Such Thing as Wittgenstein’s Philosophy of Religion?

Someone might say that my attempt to present Wittgenstein’s philosophical approach to religion is unjustified, not for the reason that it is inadequate, but rather on account of the fact that Wittgenstein did not formulate any mature and elaborated philosophical view regarding the nature of religious beliefs and practices. The arguments that may be invoked in support of such a view are fourfold. Firstly, there are relatively few remarks of Wittgenstein concerning religion coming from the later period of his philosophical activity (i.e. from 1935 on), and moreover these ought not to be treated as forming a coherent whole. Secondly, the major texts presenting Wittgenstein’s approach to religion are, in fact, records of his lectures kept by his students, and as such these might be suspected of misrepresenting to a certain degree Wittgenstein’s actual thinking (cf. Graham 2014). Further to this, it has been pointed out that there is no conclusive evidence to show that the material collected as Lectures on Religious Belief (for which I will henceforth use the abbreviation LRB) was intended to make up or appear in lectures belonging to a single series (Wittgenstein 2017, editors' note: 86–87). Thirdly, Wittgenstein’s remarks on religion were not elaborated by him in the sort of way that, say, his remarks on mathematics or psychology were: i.e. they were not reformulated and corrected many times over (Pichler 2016: 60). Fourthly, many of Wittgenstein’s remarks on religion express his personal views and existential dilemmas, and so cannot be treated as an expression of his philosophy of religion.

Even so, the arguments presented above do not suffice to prove either that Wittgenstein’s remarks on religion form only an incoherent set of thoughts, or that they lack philosophical significance. Neither conclusion can be drawn, even though almost all the statements contained in these four arguments are true. In my view, these four points only license one to accept the following claims: (1) that Wittgenstein’s remarks on religion are not so elaborated as his remarks on mathematics or psychology (which in itself is a quite trivial conclusion), and (2) that it would appear to be much harder to give an adequate reading of Wittgenstein’s remarks on religion than to present such an interpretation of his philosophical remarks concerning other questions. However, these two claims are by no means incompatible with the belief that his remarks on religion form an important and valuable contribution to philosophical discussions of religion. Of course, such a position can only be regarded as justified if the remarks in question are actually found to be insightful and helpful when attempting to elucidate certain questions in the philosophy of religion.

3 Four Questions Concerning the Difference between Religious Belief and Non-belief

Wittgenstein, in discussing the question of the difference between religious belief and non-belief, pays attention to several aspects of this issue, each of which I will briefly discuss below. (1) Are there beliefs held by a non-believer that contradict certain of the religious beliefs held by a believer? (2) What are the differences and similarities between controversies relating to religious questions and those pertaining to scientific or everyday empirical ones? (3) How should the relationship between one’s acceptance of certain religious beliefs and one’s participation in a certain religious form of life be descriptively characterized? (4) What role do pictures play in religion? It should be emphasized, however, that the aim of Wittgenstein’s discussions of these questions is not to formulate some theory of religious belief; rather, his goal is only to remind us of what one would be inclined to say in certain circumstances and what consequences one would be inclined to draw from what is said. That is why I would not wish to present Wittgenstein’s considerations concerning religious belief and language in anything like the following terms: “Wittgenstein endorses theses X, Y and Z, and supports these using arguments A, B and C”. Of course, this does not mean that I am denying that Wittgenstein makes use of arguments in his philosophy: I only would like to say that they play a different role in his philosophy from that which they play in traditional philosophy, in that they serve to remove instances of confusion, and not to support theses.

3.1 Does a Non-believer Contradict a Believer?

In LRB (Wittgenstein 1967a), Wittgenstein discusses the question of whether someone who does not believe in the Last Judgment contradicts a person who believes in it. He seems to be denying that it could be the case that the non-believer holds a certain belief contradictory to a religious belief entertained by a believerFootnote 4:

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.

And then I give an explanation: “I don’t believe in…”, but then the religious person never believes what I describe.

I can’t say. I can’t contradict that person. (Wittgenstein 1967a, LRB: 55)

Moreover, as Wittgenstein notes, if such a thing is not possible, then it would seem to be the case that the believer and non-believer do not and cannot understand each other. Yet Wittgenstein does not accept such a description of the controversy between the believer and the non-believer:

That again is Greek to me. My normal technique of language leaves me. I don’t know whether to say they understand one another or not. (Wittgenstein 1967a, LRB: 55)

Wittgenstein’s remarks on controversies concerning religious beliefs have been interpreted in different ways. For example, according to Nielsen, they can be perceived as the expression of a certain kind of fideism – i.e. the viewpoint that holds that religious statements cannot be criticized by non-believers because believers ascribe a different meaning to them than non-believers do (Nielsen 1967, 2001).Footnote 5 So on this specific version of fideism there cannot be any contradiction between a statement by a believer that “God will judge all people at the Last Judgment” and a statement by a non-believer that “It is not true that God will judge all people at the Last Judgment”, because these individuals are not ascribing the same meaning to these sentences. Barrett likewise holds that, according to Wittgenstein, describing disagreements between believers and non-believers in terms of contradiction is out of the question: “Thus, disagreements between believers and unbelievers can only be pseudo-controversies and pseudo-contradictions, or apparent controversies and contradictions. They do not operate on a ground common enough to generate controversies” (Barrett 1991: 207).

Meanwhile, Diamond presents a different interpretation of Wittgenstein’s remarks. On her view, Wittgenstein does not rule out the possibility of the non-believer denying what the believer has to say about religious questions – for example, the possibility that the non-believer’s statement about the Last Judgment contradicts the believer’s statement about it. According to her, Wittgenstein’s words should be treated as only expressing his personal response to the statement that there will be a Last Judgment. She claims that one should not assume that an atheist responding negatively to the question of whether there will be a Last Judgment or not asserts nothing contradictory to the proposition asserted by the believer. In her opinion, Wittgenstein does not exclude the possibility of the relation between these propositions being appropriately recognized as one of contradiction (Diamond 2005: 133).Footnote 6

Even so, neither of the readings presented above seems entirely adequate and convincing – though Diamond’s interpretation is, in my opinion, much closer to the spirit of Wittgenstein’s remarks than those of Barrett and Nielsen. (It is worth noting that Nielsen presents a more nuanced interpretation of both Wittgenstein’s and Wittgensteinian philosophy of religion in his later texts, where he admits, for instance, that Wittgensteinians do not claim that religious beliefs cannot be understood by non-believers (cf. 2005a, b)). First of all, Wittgenstein does not claim that the believer and non-believer ascribe completely different senses to the sentence “There will be a Last Judgment”, and that for this reason they do not understand each other and the statement put forward by the former cannot be contradictory to that offered by the latter. In fact, Wittgenstein says:

I don’t know whether to say they understand one another or not. (Wittgenstein 1967a, LRB: 55)

According to him, then, it is not helpful in this case to appeal to ordinary or normal criteria for determining whether the statements are contradictory, or whether the believer and the non-believer are ascribing the same meaning to this sentence, because the controversy under consideration here differs from ordinary or normal ones.Footnote 7

Secondly, Diamond’s remark that Wittgenstein, in LRB (Wittgenstein 1967a), presents only his personal response to the statement that there will be a Last Judgment is also questionable. According to Diamond, the remarks appearing on page 70 of the published text confirm such an interpretation, yet it is not evident that these should be construed as vindicating her invoking of the distinction between what he personally would say in such a situation and what one would say in such a situation. Indeed, the remark quoted above (“I don’t know whether to say they understand one another or not” [my italics]) indicates that Wittgenstein is speaking not just about his personal response to someone’s belief in the Last Judgment, but also about how he would interpret the disagreement between the believer and the non-believer.

Thirdly, it seems that Wittgenstein neither advocates the view that there cannot be any contradiction between believer’s and non-believer’s statements concerning, for example, the Last Judgment or the Resurrection, nor the view that the disagreement between them ought to be described in terms of their endorsement of contradictory statements. On my reading, he would not only say that he does not consider his non-belief tantamount to an acceptance of a proposition contradictory to one accepted by the believer, but also that we have no solid grounds for assuming that the non-believer’s statement that there will not be a Last Judgment genuinely contradicts the believer’s statement that there will be a Last Judgment, even though the sentence used to issue the former statement is – from a syntactic point of view – the negation of that employed when making the latter one. Of course, this does not mean that such grounds cannot be found, but it does mean – among other things – that we do not know how we would go about showing that the overt form of these statements reflects their logical (grammatical) form and, on this basis, justifies treating them as contradictory.

So why is it that there are no grounds for describing such a disagreement in terms of a relation of contradiction between statements? Wittgenstein’s answer is, I believe, contained in the following remark:

And then I give an explanation: “I don’t believe in . . .”, but then the religious person never believes what I describe.

I can’t say. I can’t contradict that person.

In one sense, I understand all he says – the English words “God”, “separate”, etc. I understand. I could say: “I don’t believe in this,” and this would be true, meaning I haven’t got these thoughts or anything that hangs together with them. But not that I could contradict the thing. (Wittgenstein 1967a, LRB: 55)

We have no grounds for doing so because we do not have grounds for ascribing a common content to the statements of the believer and the non-believer.Footnote 8 However, according to Wittgenstein, this does not justify a conclusion to the effect that they cannot understand each other.

To be sure, one could criticize the view presented above, claiming that such statements can, after all, be treated as contradictory because, as I have already pointed out, the one is – from a syntactic point of view – the negation of the other and, moreover, the believer and the non-believer can ascribe the same meanings to the words occurring in the sentences used to make these statements. (This last point seems also to be accepted by Wittgenstein: “The difference might not show up at all in any explanation of the meaning.” (Wittgenstein 1967a, LRB: 53)). Wittgenstein’s response to this objection might then take the form of saying that if one wishes to use the word “contradiction” in that way, one is free to do so, but one must remember that where controversies are concerned, cases in which both sides agree on what the content of the disputed statements amounts to differ from ones where they do not. Hence, one must remember that there is a significant difference in these two kinds of use of the word “contradiction”. This response, however, might not seem satisfying. One might well ask whether it can really be claimed that the proposition (i.e. the meaningful sentence) “There will be a Last Judgment” asserted by the believer has a different content from the seemingly identical proposition that is a constituent of the negative proposition (“There will not be a Last Judgment”) asserted by the non-believer, in circumstances where it is assumed that both persons are ascribing the same meaning to the words that occur in the sentence “There will be a Last Judgment”. It could seem that such a thing is not possible, as in this case any such difference as to the content could not be a result of some occasion-specific character that the sentence takes on (in that differences pertaining to when and where the sentence is actually uttered do not change the content). Provided that the meanings of the words occurring in the sentence are explained in each instance in the same way, the content of the proposition should also be the same.

What, then, for Wittgenstein, is the reason for thinking that the content could be perceived as different in this case? Here one should note two points. Firstly, belief in the Last Judgment plays a central role in the life of the believer. For the latter, it means that they must choose a certain way of living their life – a certain way of acting, thinking and feeling (cf. Putnam 1992: 154), and this, in a sense, is tantamount to a transformation of the very concept of the Last Judgment.Footnote 9 By contrast, the non-believer may well see no essential connection whatsoever between endorsing a belief in the Last Judgment and choosing a certain way of life (Wittgenstein 1967a, cf. LRB: 56). Secondly, when a religious person affirms a belief in the Last Judgement, they are accepting a certain picture – one to which they constantly have recourse in the context of their life. Meanwhile, for the non-believer, the proposition under consideration need not be connected in an essential way with any such picture:

Here believing obviously plays much more this role: suppose we said that a certain picture might play the role of constantly admonishing me, or I always think of it. Here, an enormous difference would be between those people for whom the picture is constantly in the foreground, and the others who just didn’t use it at all. (Wittgenstein 1967a, LRB: 56)

I will discuss these two points in what follows below. Before doing so, though, let me sum up what has been said so far about the question of contradiction. The reader should note that Wittgenstein is not claiming that controversies arising between believers and non-believers cannot be described in terms of contradiction. Rather, his approach to this question consists in – among other things – a juxtaposition, for purposes of comparison, between, on the one hand, contradictions that appear in scientific or everyday forms of discourse, and, on the other, certain pairings of statements pertaining to religious questions that look like contradictions. He points out that ordinary contradictions differ significantly in their character from the sorts of divergence evinced in connection with the beliefs held by the religious believer on the one hand, and by the religious non-believer on the other. That is his reason for not being inclined to describe such disagreements in terms of contradiction. For example, disagreement over the Last Judgment could take the following form: the life of believers is guided by the picture of God who is going to be the judge of all people, whereas non-believers have no thoughts about that and appeal to no such picture, treating accounts of the Last Judgment as merely fictional. If the disagreement takes this form, then it is clear that the non-believer is not engaged in treating the Last Judgement as some possible event which, as a matter of actual fact, will not happen. To use Wittgenstein’s metaphor, one can say that in this case, where the non-believer is concerned, “statements” about the Last Judgement do not belong to the logical space of genuine assertions. That is the reason why a description of the disagreement as involving contradictory stances is questionable here. Of course, the controversy may take another form, in that the non-believer may claim that the description of the Last Judgement should be treated as an empirical description of a certain possible event in the future which will never happen. However, in this case one can also doubt whether the controversy should be presented as a matter of contradictory beliefs or statements being entertained or asserted, as the religious person’s belief in the Last Judgment does not seem to be an ordinary empirical belief, and in the present context this also furnishes a reason for thinking that one should not say of the religious person that he or she believes in something that the non-believer is engaged in denying.

3.2 Ordinary Controversies versus Controversies between Religious Believers and Non-believers

In order to better understand what I said in the last paragraph, and to explain why Wittgenstein is not inclined to describe disagreements between believers and non-believers in terms of relations of contradiction, we should consider the following question: what is the difference between controversies concerning religious beliefs – for example, the dispute over whether there will be a Last Judgment or not – and controversies pertaining to ordinary empirical facts or scientific explanations of these facts? I would like to start my discussion of this by pointing out that if not believing in the Last Judgment were to be treated as a case of believing in the proposition to the effect that a certain empirically possible event will not happen in the future, then it would be quite natural to treat believing in the Last Judgment as a case of belief similar to that which scientific predictions involve. However, as Wittgenstein rightly emphasized, such a belief understood in this particular way would not actually constitute a religious belief:

Suppose, for instance, we knew people who foresaw the future; make forecasts for years and years ahead; and they described some sort of a Judgement Day. Queerly enough, even if there were such a thing, and even if it were more convincing than I have described but, belief in this happening wouldn’t be at all a religious belief. (LBR: 56).

In my view, a brief examination of the difference between foretelling the future and religious prophecy will enable us to understand Wittgenstein’s point more effectively than otherwise. Firstly, foretelling the future can rest on knowledge of past and present events and the laws of nature (in which case it is identical with prediction), while religious prophecy is not based on such knowledge. Secondly, one can say that in the case of foretelling the future statements are verified just by what subsequently happens, in that they are descriptions of future events, whereas in the case of prophecy the words uttered by a prophet are treated by religious persons as an expression of the will of God. So, from the religious point of view, in one sense there are no false prophecies, and in the other false prophecies are in fact just such false predictions as happen to be taken to be prophecies. Thus, prophecies are in a sense much more similar to promises and warnings than to predictions. (Of course, one could say that in this case the difference between what is promised and what is predicted has a quite unique character, in that knowledge of God’s promises itself suffices to predict the future.) However, it is worth noting that, even for believers, it is only immediately obvious that a given prophecy has been fulfilled in a very limited set of cases (cf. Anscombe 2008). This feature distinguishes prophecies from ordinary promises. (Nevertheless, the foregoing approach to the relationship between prophecy and foretelling the future differs from, say, Anscombe’s approach, in that she seems to employ the phrases “uttering prophecies” and “foretelling the future” in roughly the same sense – as when she writes that “The mere fact that someone is a wonder-worker, or utters prophecies that are fulfilled, certainly does not show he is divinely attested” (2008: 37)).

According to Wittgenstein, a belief justified in a scientific way – i.e. on the basis of empirical evidence and scientific theories – is a case of scientific rather than religious belief. Such a statement may, of course, prompt one to ask whether any empirical evidence can serve to justify religious beliefs, and I would like to consider this question in connection with Holland’s claim that there are certain events which are empirically certain yet conceptually impossible (cf. Holland 1965). (Holland’s approach to this question has been criticized by Winch (1987c) and Phillips (1993d), and the assumptions entertained by each side in the debate are discussed in Palmer (1995)). Firstly, it is worth noting that the expression “conceptually impossible” used by Holland to refer to such events as the turning of water into wine or someone’s rising slowly and steadily three feet into the air is somewhat incongruous. It would be much more natural to say that such events are, or at least seem to be, physically impossible from the point of view of contemporary science. Secondly, if certain perceptions, i.e. certain empirical evidence, did justify the claim that certain events that are physically impossible from the point of view of contemporary science had undoubtedly occurred, this would not already justify ipso facto certain religious beliefs, because the occurrence of certain events that cannot be scientifically explained is not by its very nature necessarily a religious phenomenon. These two points suggest that if certain perceptions are to constitute reasons for entertaining certain religious beliefs, they must be understood not merely as perceptions of events that are physically impossible from the point of view of contemporary science. However, this does not mean that such perceptions are conceived by believers in terms of a two-factor model – namely, as ordinary perceptions supplemented with either some religious attitude towards the objects or events perceived or some religious interpretation of these. (It is worth noting that two-factor models of moral and religious beliefs have been criticized by authors inspired by Wittgenstein’s philosophy (cf. McDowell 1998a; Schönbaumsfeld 2014)). That is to say, it seems that perceptions treated by believers as reasons for holding religious beliefs do not, when seen from a religious point of view, divide up into perceptions of natural events (be they ordinary or – as with the turning of water into wine, for example – extraordinary) on the one hand, and religious interpretations of these natural events (for instance, the construal of the turning of water into wine as a revelation of Christ’s divine power) on the other. It seems that from a religious point of view, such perceptions should rather be described in terms of seeing as – meaning that the turning of water into wine is seen by a believer as a special intervention of God which reveals his goodness and power. (It should be noted that according to commentators such as Graham, the concept of seeing as plays an important role in understanding religion (cf. Graham 2014: 112–113)). Of course, there is a fundamental difference between seeing a certain figure in a picture as a duck and seeing a certain event as an act of God. In the latter case there is no agreement between believers and non-believers to the effect that certain events are indeed to be seen as God’s actions: some non-believers will say that believers are under the illusion that they see certain events as God’s actions.

The difference between religious and scientific beliefs consists, among other things, in the fact that reasons for the former “look entirely different from normal reasons” (Wittgenstein 1967a, LRB: 56). Here it is worth adding, in passing, that he was aware of the possibility of entertaining beliefs that would be such that one might not be sure whether to characterize them as religious or scientific (Wittgenstein 1967a, LRB: 56). Nevertheless, that does not furnish a reason for not acknowledging the distinction in question. (He discusses an analogous question in On Certainty (Wittgenstein 1972: §§ 97–98).) Indeed, we find him presenting his conception of the specific character of religious reasons in the following fragment from Culture and Value (henceforth CV):

A proof of God ought really to be something by means of which you can convince yourself of God’s existence. But I think that believers who offered such proofs wanted to analyse & make a case for their ‘belief’ with their intellect, although they themselves would never have arrived at belief by way of such proofs. “Convincing someone of God’s existence” is something you might do by means of a certain upbringing, shaping his life in such & such a way.

Life can educate you to “believing in God”. And experiences too are what do this but not visions, or other sense experiences, which show us the “existence of this being”, but e.g. sufferings of various sorts. And they do not show us God as a sense experience does an object, nor do they give rise to conjectures about him. Experiences, thoughts, – life can force this concept on us.

So perhaps it is similar to the concept ‘object’. (Wittgenstein 1998, CV: 97)

This quotation contains several suggestions. Firstly, arguments intended to justify belief in the existence of God do not engender such a belief, but may still serve to reinforce the latter and enhance the believer’s understanding of what it involves.Footnote 10 For instance, they may enable one to see what the difference between having this belief and entertaining a belief in the existence of any given object consists in. In contrast to arguments of this type, scientific arguments intended as furnishing justifications for belief in the existence of a certain object – e.g. a certain planet – do very often engender belief in the existence of that object. Secondly, a certain way of living may make one believe in the existence of God. For example, some particular type of upbringing, or certain situations and problems that one encounters in one’s life, may lead to an embracing of some sort of faith (cf. Schönbaumsfeld 2014; Cottingham 2009). Thirdly, to believe in God is, in some sense, to be forced by life to apply the concept of God in one’s life. What this means is that someone who starts to believe in God does not do so because they have discovered that this concept can be applied to a certain being (i.e. that the word “God” refers to a certain object or entity) (cf. Phillips 2005a: 81). It seems that for Wittgenstein, a person who is beginning to believe in God is rather to be seen – when looking at it from the religious point of view – as someone to whom the real content of the concept of God is starting to be revealed. However, it should be emphasized that this statement does not imply that non-believers can never understand propositions in which the word “God” occurs. Moreover, it is worth noting that even amongst those non-believers who find no grounds for admitting that the word “God” expresses a coherent concept, there will be some who are not prepared to dismiss the term itself as entirely meaningless.

If such an interpretation of the character of disagreements between believers and non-believers is right, then in fact these are not controversies in the normal sense of that term. In the case of ordinary disagreements, either each side of the controversy recognizes what is said by the opposing one as meaningful, or one of them regards statements made by the other as actually meaningless – as happens when they question whether the concepts used by their opponents are not in fact simply confused and incoherent. According to Wittgenstein, disagreements between religious believers and non-believers need not possess the latter characteristic: non-believers who acknowledge no grounds for using such terms as “God” or “the Last Judgment” can reject claims to the effect that the utterances in which such expressions occur are simply meaningless. But in that case, how does their taking issue with this claim manifest itself? In my opinion, it primarily shows itself in the fact that these non-believers continue to engage in the dispute with religious persons.

The dispute between religious persons and non-believers may basically take two forms. The first of these consists in this: that non-believers join the dispute with the intention of attaching to religious terms the same meanings that believers invest in them.Footnote 11 One of the aims of non-believers of this sort will be to learn from believers what the meaning of such terms actually amounts to. However, the dispute can also take another form: non-believers can treat the religious discourse as a quasi-scientific one (there being, of course, many different ways in which religious utterances can be interpreted as constituting quasi-scientific statements). In this case, religious utterances are criticized as amounting to statements that fail to be confirmed by any scientific evidence. Wittgenstein points out that a person who adopts a viewpoint according to which scientific evidence can confirm or falsify religious utterances is operating on entirely different plane from a religious person – there is, as it were, a gulf separating them:

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 1967a, LRB: 53)

Why can one say that there is a gulf between these two people? The main reason is the fact that the believer does not treat religious questions as questions to which scientific evidence or scientific doubts are relevant. By contrast, a non-believer of the sort invoked by Wittgenstein in this fragment is, it seems, one who thinks it impossible to ascribe any other kind of sense to utterances such as “There will be a Last Judgment” or “Jesus Christ was resurrected” than that ascribable to statements of empirical fact such as are, or at least can be, based on scientific evidence. The other reason why there is a gulf between them is that while, for the believer, the acceptance of such statements entails – or at least should entail – a complete change in their way of living, for a non-believer of the sort prepared to treat religious utterances as scientific statements of an unjustified kind, our acceptance of these does not differ fundamentally from our acceptance or rejection of other scientific statements.

3.3 The Role of Faith in the Lives of Believers

According to Wittgenstein, then, in order to comprehend the difference between religious and scientific beliefs it is necessary to determine the role played by religious beliefs and utterances in the lives of believers themselves. He characterizes faith – especially Christian faith – in the following terms:

Christianity is not based on a historical truth, but presents us with a (historical) narrative & says: now believe! But not believe this report with the belief that is appropriate to a historical report, – but rather: believe, through thick & thin & you can do this only as the outcome of a life. Here you have a message! – don’t treat it as you would another historical message! Make a quite different place for it in your life. – There is no paradox about that! (Wittgenstein 1998, CV: 37)

Amongst other things Christianity says, I believe, that sound doctrines are all useless. That you have to change your life. (Or the direction of your life.) (Wittgenstein 1998, CV: 61)

It appears to me as though a religious belief could only be (something like) passionately committing oneself to a system of coordinates. Hence although it’s belief, it is really a way of living, or a way of judging life. Passionately taking up this interpretation. And so instructing in a religious belief would have to be portraying, describing that system of reference & at the same time appealing to the conscience. And these together would have to result finally in the one under instruction himself, of his own accord, passionately taking up that system of reference. (Wittgenstein 1998, CV: 73)

These quotations show that on Wittgenstein’s view, religious faith should not be regarded as merely a set of beliefs that are endorsed by believers. (We should note, though, that the same material does not support non-cognitivist readings of Wittgenstein: as Schroeder and Cottingham rightly point out, the assertion that “it’s belief” makes up an integral part of the above quotation (Schroeder 2008; Cottingham 2009)). Faith is something that changes the life of the believer completely (cf. Burley 2012: 79) – or, more precisely, someone who has religious faith lives their life in a wholly different way from someone who does not. For example, a belief in the Last Judgment could manifest itself in the fact that the believer, whilst making various decisions, always asks whether God will approve their choice, and a belief in Christ’s resurrection could be disclosed through courageous acts that spring from a hope that, eventually, all evil and all of the suffering we encounter in our lives will in some sense be overcome. One may then ask whether the above explanations show that there is, for Wittgenstein, a fundamental difference between religious and scientific beliefs, and to this question one could answer “no”, pointing out that in Wittgenstein’s opinion, the acceptance of scientific beliefs also carries important practical consequences. Even so, while that would of course be correct, it does not justify the claim that there is no important difference between scientific and religious beliefs. At most, it shows that this difference cannot be explicated in terms of the thought that whereas religious utterances are expressions of pseudo-beliefs that actually amount to commitments to certain ways of acting, scientific statements express beliefs that serve only to represent reality – ones that, moreover, correspond to the kind of statements that can be separated, as it were, from the other spheres of our lives (cf. Schönbaumsfeld 2007: 189–190).

As I have already pointed out, the difference consists firstly in the fact that our reasons for endorsing scientific beliefs differ completely from our reasons for endorsing religious ones, and secondly in the fact that what is entailed by an acceptance of religious beliefs is something with a quite different character from the consequences that follow from embracing scientific beliefs. One’s endorsement of scientific beliefs is not necessarily connected with a readiness to make a fundamental change in one’s life. What this means is that although the endorsement of a certain set of scientific beliefs may cause a fundamental change in one’s life, it does not have to – and, moreover, from the point of view of science a person who accepts a certain scientific belief (e.g. the belief in global warming) and does not fundamentally change his or her way of life is not doing something wrong – i.e. they can be criticized from a moral, but not from a scientific, point of view. On the other hand, when it comes to endorsing religious beliefs, things look different (Wittgenstein 1998, cf. CV: 37): one of the criteria for this is a readiness to make a fundamental change in one’s life. So, the person who accepts certain religious beliefs and does not fundamentally change his or her way of life is doing something wrong from the religious point of view. Moreover, a fundamental change in one’s life connected with an endorsement of religious beliefs is often understood – from the perspective of those who make such a change – in a radically different way than a fundamental change in one’s life connected with the endorsement of certain scientific beliefs is understood by those who change their lives as a result of the latter. The former can be understood in terms of God’s grace, or a gift: that is, in terms of something independent of the believer’s will. The latter, however, is commonly understood in terms of a subject’s decisions, whether conscious or unconscious, rational or irrational.

3.4 The Role of Pictures in Religion

Wittgenstein, when discussing the specific character of religious beliefs, emphasizes one further aspect of them, which is that they are often expressed by means of pictures. But how should one understand the word “picture” in this context? As is well known, he uses this word in various ways both in his early and, especially, his later writings. In his later writings, unlike in the Tractatus, he does not regard all meaningful sentences as pictures that serve to represent reality (Wittgenstein 2009: § 23). It is obvious that when he speaks about the role of pictures in religious language he is not concerned only with paintings, although he thinks that paintings can also play an important role in explaining the meaning of religious language (Wittgenstein 1967a, cf. LRB: 63). The later Wittgenstein treats, for example, linguistic expressions not used in their literal senses as pictures. So, both allegories and metaphors can be characterized as pictures. And this, in turn, leaves room for one to distinguish between those pictures that can be replaced by expressions having a literal meaning without any loss of sense, and those that cannot be so replaced without undergoing a significant loss of sense.

The expression “I feel pangs of conscience because I lied” seems like a good example of a picture of the first kind, as it can be substituted with the sentence “I feel morally guilty because I lied” without any important loss of sense. By contrast, the sentence “God’s eye sees everything” (Wittgenstein 1967a, LRB: 71) is an example of a picture that, according to Wittgenstein, cannot be replaced by any other expression without a significant loss of sense. Any substitution of this picture with another expression would change the meaning of the utterance: it is an essential feature of the sense of this utterance that it attributes having an eye to God. That is the reason why he makes the following remark: “The whole weight may be in the picture” (Wittgenstein 1967a, LRB: 72). The weight of the picture expressed by the words “God’s eye sees everything” consists among other things in the fact that in likening God in some important respect to a human being the picture makes God into something very close to a human being. According to Wittgenstein, religious discourse is an example of a discourse in which pictures of this kind are often employed, and, moreover, one in which they play an important – not to say essential – role. This aspect of Wittgenstein’s approach to religious language is emphasized by Winch and Diamond (Winch 1987a, Diamond 2005, cf. also Schönbaumsfeld 2007: 179–182).Footnote 12 The impossibility of eliminating such pictures without an important loss of sense does not mean that there is no way to explain their meaning: one can partially elucidate the sense of the sentence “God’s eye sees everything” by pointing out that it suggests that God is omniscient – i.e. that God knows everything.

Nevertheless, the statement that God is omniscient does not capture the whole sense of this sentence. Moreover, it should be added that one can doubt whether what is conveyed by the sentence “God is omniscient (God knows everything)” is such that the word “knows” carries the same literal and unambiguous meaning there as it does in, for instance, the assertion “Kant knew that the Sun is bigger than the Earth”. It seems that when we apply the word “knowledge” to God we are using it in a different way to that involved when applying it to people, or to other finite beings to which we may also be inclined to apply it.Footnote 13 If one accepts the classical definition of knowledge – according to which it is a set of true and justified beliefs (and assuming for the sake of simplicity that one need not take into account the implications of Gettier problems for this) – then one can doubt whether God knows this or that: i.e. one can doubt whether he has justified and true beliefs about this or that in just the same literal and unambiguous sense in which people are said to have such beliefs. Firstly, one can wonder whether beliefs can be ascribed to God. Generally, it is assumed that it is not the case that for any proposition p, belief in p is necessarily equivalent to the truth of p. Moreover, it seems that the fact that this equivalence does not hold partially determines the meaning of the word “belief”. However, in the case of God, if one were to be inclined to apply the word “belief” to God, one would have to say that all of God’s beliefs are necessarily true and all truths are necessarily believed by God. This shows that when we apply the term “belief” to God, we are using it in a significantly different way than when we apply it to people or animals. Secondly, the claim that God is in a position to justify certain propositions, or even that he needs any such justifications, also seems questionable. To sum up, there is no reason to accept the view that when one asserts the sentence “God is omniscient” one is attributing a certain property to God in some unambiguous and literal sense. (The question of whether “words directed at God” are used univocally, equivocally or in some other way is interestingly discussed by Mulhall (2015).)

As regards pictures of the sort that cannot be substituted for by any other expressions without a significant loss of sense, it is worth noting firstly that such pictures are employed in both religious discourse and poetry, and secondly that religious faith is very often expressed in the language of poetry, which itself makes use of such pictures. For example, there are many texts in the Bible exhibiting this characteristic (e.g., Psalms, Song of Songs, 1 Corinthians 13).

On the other hand, the claim that there are important similarities between many religious utterances and poetry may be criticized in the following terms: those religious utterances that are metaphors (i.e. those pictures used in religious discourse that are not replaceable by certain expressions possessing a literal meaning without a significant loss of sense) are devoid of any cognitive significance, and if they are an indispensable element of the religious discourse in question, then a certain important part of the latter will likewise be devoid of any cognitive meaning. Of course, this objection can be interpreted as leading either to the conclusion that one should recognize certain important religious utterances as being devoid of cognitive meaning, or to the conclusion that the Wittgensteinian view concerning religious discourse is itself mistaken – i.e. the objection can be construed along the lines of modus tollens. In my opinion, the objection hinges on an unreflective acceptance of a premise to the effect that the language of poetry is essentially devoid of cognitive meaning. (Such a view seems to have been endorsed, for example, by Carnap (1959).) However, it is by no means obvious that such a premise holds true unrestrictedly: while it is obvious that, in principle, poems present neither theories explaining the nature of the world and mankind, nor scientific data which could verify (i.e. confirm or falsify) such theories, the thesis that they cannot say anything true about the world or mankind seems fairly questionable. Philosophers as different as Gadamer and Goodman claim that poetry can, in some sense, convey such truths (cf. Gadamer 2004; Goodman 1978).

Of course, these remarks do not of themselves prove that both poetry and religious discourse have a cognitive character. My own view is that the question of whether pictures of the sort that cannot be eliminated without entailing an essential loss of sense have a cognitive meaning or not cannot be decided in a completely general way. The best approach would be to study particular cases. Moreover, it should be admitted that questions concerning the cognitive status of particular pictures are of a much more subjective nature than scientific questions. It is not uncommon for metaphors perceived by some as offering the deepest insights into human existence to be regarded by others as some sort of grotesque description of human fate or an expression of unconscious obsessions.Footnote 14

4 Does Wittgenstein Embrace a Philosophical View Concerning Religion?

We may wonder why it is that Wittgenstein seeks to draw attention to the questions just discussed that pertain to religious beliefs and utterances. Does he do so in order to present a certain view regarding their nature? In one sense, it is entirely misleading to say that Wittgenstein’s remarks concerning religious beliefs and utterances are the expression of a certain view about them – at least if this view is conceived as tantamount to his entertaining a theory or an opinion with respect to them. In another sense, though, it is quite uncontroversial to assert that he is presenting a certain view where religious beliefs and utterances are concerned: i.e. a certain way of perceiving them. So why is it quite misleading to treat his remarks on religious beliefs and utterances as the expression of a certain theory or set of opinions about them? Here we can point out that ever since the time of writing the Tractatus he had stood opposed to any conception of philosophy according to which its task is to formulate philosophical theories, opinions, views and theses. According to him, philosophy’s aim is to remove confusions engendered by misunderstandings of the grammar (logic) of our language through a proper elucidation of that grammar (logic), where the latter process consists in our furnishing a perspicuous presentation of how, in fact, we use the particular fragment of our language that is misunderstood. Wittgenstein presents his approach to philosophical views and opinions in the following way:

On all questions we discuss I have no opinion; and if I had, and it disagreed with one of your opinions, I would at once give it up for the sake of argument because it would be of no importance for our discussion. We constantly move in a realm where we all have the same opinions. All I can give you is a method. I cannot teach you any new truths. (Wittgenstein 2001: 97)

Some influential commentators have sought to question whether Wittgenstein is adequately characterizing his own way of doing philosophy here. In their opinion, one can discern certain philosophical views within Wittgenstein’s remarks on language, mind and mathematics (cf. Dummett 1978a, b; Kripke 1982; Blackburn 1990). This way of interpreting Wittgenstein’s remarks does not seem to be right (it is overtly criticized by, amongst others, McDowell (1998b) and Diamond (1991b); moreover, even philosophers who do not themselves share Wittgenstein’s view about the nature of philosophy concede that it is open to question (Wright 2001)). However, it should be emphasized that it is wrong not because one cannot perceive Wittgenstein’s remarks as the expression of certain philosophical (metaphysical) views – of course, they can be perceived in such a way and, as a matter of fact, are so. In order to show that Wittgenstein’s remarks on mathematics, language, mind and religion are not inconsistent with his remarks on the nature of philosophy itself, it is sufficient to show that the former can also be understood in a different way: that is, not as an expression of certain philosophical (metaphysical) views, but simply as reminders of how, in fact, we use our language (cf. Wittgenstein 2009: §116, §127).

Before showing that Wittgenstein’s remarks on religion can be understood as being the sort of utterances that serve to remind us of how religious language-games function, I would like to briefly discuss the approach to interpreting his philosophy that perceives these remarks as amounting to the expression of a certain philosophical theory or standpoint. In my opinion, this way of reading Wittgenstein’s remarks is motivated firstly by an assumption to the effect that the main aim of philosophy is to formulate philosophical theses and theories, and secondly by an understanding of these remarks that is itself one-sided (cf. Wittgenstein 2009: § 593). So interpretations of this kind can rest on two kinds of ground. The first is the conviction that Wittgenstein must have had a certain view about the essence of religious beliefs and utterances, where this is itself taken to mean that he must have thought that there is a certain set of properties that will explain why certain beliefs and utterances are classified as religious ones. (I suppose that Wittgenstein would say that to equate the former formulation with the latter is to perform a “sleight of hand” or “conjuring trick” of sorts, as while the former is ambiguous and can be interpreted in a quite innocent way, the latter constitutes an interpretative assumption that turns out to be inadmissible). At the same time, as I mentioned above, the interpretations that ascribe philosophical theses and theories to Wittgenstein can also rest on a one-sided understanding of his remarks. Such one-sided readings of his remarks on the subject of “mind” result in attempts to classify his philosophy of psychology as a form of behaviourism, while one-sided readings of his remarks on mathematics produce interpretations of his philosophy (as it relates to this domain) as endorsing a kind of strict finitism or full-blooded conventionalism. In the case of his remarks on religion, this one-sidedness very often consists in treating his remarks as a sketch for or introduction to a non-cognitivist theory of religion.

According to a theory of this last kind, religious beliefs and utterances are only expressions of certain attitudes toward the world: for instance, of acceptance or gratitude – or moral attitudes.Footnote 15 (If one also embraces moral non-cognitivism, then moral attitudes are to be explained in terms of attitudes or feelings that are themselves completely describable in non-cognitive terms). For example, the whole content of the belief that God will judge all our deeds can be reduced to this: that while making any decision we should consider whether it is morally right or wrong. The whole content of the belief that everything is in God’s hands can thus be reduced to attitudes of hopefulness and acceptance of reality, and the feeling that we cannot change our fate. There can be no doubt, however, that these readings ascribing a stance of religious non-cognitivism to Wittgenstein are in fact mistaken. (We may note that this is a point on which commentators as diverse as Burley (2012: 15–16), Mulhall (2011), Phillips (1993a), Schroeder (2008) and Schönbaumsfeld (2014) all agree.Footnote 16) As we can see, he rejects any interpretation of religious beliefs and utterances according to which their sole role in human life is to express attitudes:

Lewy: “In this case, you might only mean that he expressed a certain attitude.”

I would say “No, it isn’t the same as saying ‘I’m very fond of you’” – and it may not be the same as saying anything else. It says what it says. Why should you be able to substitute anything else? (Wittgenstein 1967a, LRB: 71)

When Wittgenstein makes comments about the nature of his own remarks concerning religion, he notes several things: firstly that he does not wish to say anything with which religious people would disagree, secondly that he only wants to determine what consequences believers draw from their use of certain given words, and thirdly that if he “wished to say anything more [he] was merely being philosophically arrogant” (Wittgenstein 1967a, LRB: 72). Wittgenstein’s aim, then, is not to give an explanation of the sort that would seek to spell out what it is that the religious beliefs and utterances can in essence be reduced to. Why not? One reason is the fact that such an explanation can turn out to be incompatible with what believers themselves think about religious matters. Of course, this need not mean that believers’ interpretations of the nature of the latter must be right, but it does mean that if certain interpretations of religious beliefs and utterances are not compatible with utterances belonging to a given religious practice, then those interpretations count as mistaken or confused.Footnote 17

I would like now to develop the suggestion I made above, to the effect that, in a sense, it is quite uncontroversially the case that Wittgenstein presents a certain view about religious beliefs and utterances. This claim can be considered uncontroversial if all it is saying is that Wittgenstein’s remarks present us with the grammar of some religious language for elucidatory purposes. Such an interpretation is justified, because according to him “essence is expressed in grammar” (Wittgenstein 2009: § 371). However, it is worth emphasizing that when Wittgenstein speaks about essence he does not mean something hidden: it is not, as it were, a source from which overt properties of the phenomena under consideration – e.g. overt properties of religious beliefs and utterances – spring. The essence of these phenomena can be recognized by examining the religious forms of life to which certain given beliefs and utterances belong, together with the role played by these beliefs and utterances in the lives of believers and non-believers. Of course, such a formulation is sketchy, to say the least, and cannot serve as a satisfying elucidation of religious beliefs and utterances in the absence of a more detailed descriptive account of religious forms of life themselves. For this reason, even though it is not possible here to give a detailed description of the latter, a few remarks about them should be made. First of all, when one sets out to describe a religious form of life, one should take into account moral decisions, prayers, rituals and ceremonies, attitudes toward the world and other people, and so on. It is worth noting that there are important connections between these elements (Winch 1987b), and, moreover, that they are also connected with other elements of believers’ lives (Burley 2018; Phillips 1993b, c). However, to think that there must be one single property which all elements of a religious form of life have to possess in order to count as elements of such a form of life is confused.

A description of the grammar of a religious language-game will therefore contain descriptions of various aspects and elements of a religious form of life, and a description of the role played by religious beliefs and utterances in the lives of believers and non-believers. And if one treats Wittgenstein’s remarks on religion as descriptions of this kind, it turns out to be the case that they need not be construed as expressing a certain philosophical standpoint regarding the nature of religious beliefs and the meaning of religious utterances. The philosophical character of these remarks consists neither in the fact that they present a certain philosophical thesis concerning the rightness of religious utterances (this point, indeed, seems to be accepted by the vast majority of commentators on Wittgenstein), nor in the fact that they present some definite philosophical thesis concerning the meaning of such statements. Their philosophical character consists in the fact that they are issued in order to clarify the particular circumstances in which religious expressions are used and religious statements are made, and the particular consequences believers and non-believers are inclined to draw from the acceptance or rejection of such beliefs and utterances. To call descriptions of such circumstances and consequences philosophical theses or elements of a philosophical standpoint is entirely confused. For instance, it seems obvious that the following quotations are not expressions of certain philosophical theses about religious language:

The word ‘God’ is amongst the earliest learnt – pictures and catechisms, etc. But not the same consequences as with pictures of aunts. I wasn’t shown [that which the picture pictured]. (Wittgenstein 1967a, LRB: 59)

“You can’t hear God speak to someone else, you can hear him only if you are being addressed.” That is a grammatical remark. (Wittgenstein 1967b: § 717)

The existence of such material shows that Wittgenstein’s remarks on religion can be understood as being of the sort that serve to remind us of how we use the expressions that occur in religious utterances, rather than as expressions of a certain philosophical standpoint regarding the essence of religious beliefs and utterances. Nevertheless, such an interpretation might seem far-fetched. Let us therefore consider the following quotation:

Queer as it sounds: the historical accounts of the Gospels might, in the historical sense, be demonstrably false, & yet belief would lose nothing through this: but not because it has to do with ‘universal truths of reason’! rather, because historical proof (the historical proof-game) is irrelevant to belief. This message (the Gospels) is seized on by a human being believingly (i.e. lovingly): That is the certainty of this “taking-for-true”, nothing else.

The believer’s relation to these messages is neither a relation to historical truth (probability) nor yet that to a doctrine consisting of ‘truths of reason’. There is such a thing. – (We have quite different attitudes even to different species of what we call fiction!) (Wittgenstein 1998, CV: 37–38)

It seems that Wittgenstein, in the above, is formulating a certain philosophical thesis about religious beliefs and, moreover, that this thesis is incompatible with what actual believers think and say about such matters. The thesis seems to be that from the point of view of Christian faith, it does not matter whether the events described in the Gospels happened or not. However, such an interpretation of the fragment just quoted represents a misreading. Wittgenstein merely shows – in an emphatic way – that there is a radical difference between accepting a certain proposition on the basis of scientific evidence and doing so out of religious faith. He simply makes a grammatical observation to the effect that from the religious point of view, scientific evidence (proof) is not only inessential, but also actually irrelevant. More specifically, to say that Christians’ belief in Jesus Christ’s resurrection is not based on scientific evidence is to make a grammatical remark with the aim of reminding us about how the word “belief” and its cognates are used in the context of Christians’ faith in resurrection.

The assertion that scientific evidence is not relevant to religious beliefs is, of course, a very general one. Hence, it must be explicated in more detail if one is to show that the objection to my interpretation that says that “this remark is not a grammatical remark but a philosophical thesis” is invalid. The main reason for construing it as a philosophical thesis rather than an uncontroversial grammatical remark is the fact that some believers themselves tell us that scientific evidence can support certain religious beliefs. So, how should one understand the remark to the effect that scientific evidence is not relevant to religious beliefs?

Firstly, this remark does not mean that no religious beliefs are justified by believers’ perceptions. Secondly, it does not mean that one should never seek to justify religious beliefs on the basis of scientific evidence. Wittgenstein is not saying that one should not do this. The purport of the remark is rather the following: that trying to explain in a detailed way what actually justifying religious beliefs on the basis of scientific evidence involves leads, in the end, to the realization that we do not have any clear idea about what it would amount to. In order to better explain what I mean here, I will focus on one example. Let us consider the statement asserting that the belief in Jesus Christ’s resurrection can be verified or falsified by scientific evidence. Of course, the content of this statement depends on the content of the belief in Christ’s resurrection. If the content of the latter is, roughly, that Jesus Christ died (i.e. brain-death occurred) and then, after three days, his dead body started to function again as a living human organism (exhibiting the appropriate vital signs) and functioned thus for a certain period of time, then the content of the statement is, in principle, clear. It is basically clear what scientific evidence would confirm or disconfirm the belief. Such evidence could consist in observations pertaining to physiological processes taking place in Christ’s body.

However, the fundamental question that should be raised as regards the content of the belief in Jesus Christ’s resurrection when it is explicated in this way is this: is this belief really a religious belief, or is it rather a purely scientific one? In order to better understand the problem, let us imagine that someone believes that Ludwig Wittgenstein died, and that after three days his dead body then started to function as a living human organism again. Is his or her belief a religious belief? It seems obvious that this belief is a scientific one, and there is no reason to treat the belief about Jesus Christ discussed above differently: both beliefs are only concerned with biological and historical facts. To be sure, one might object that being a scientific fact does not exclude being a religious fact – for example, one could claim that for Christians, Christ’s resurrection is both a religious and a scientific fact. However, this objection is, in my opinion, not valid. First, it is not clear in what sense the fact that certain biological processes took place in Christ’s body constitutes a religious fact. Second, even if one were to argue that a certain fact is both religious and scientific, in that the religious fact is reducible to a scientific one, this argument would in fact just show that speaking of a religious fact at all in that instance was redundant. (The word “religious”, in this context, would be like Wittgenstein’s “wheel that can be turned though nothing else moves” (Wittgenstein 2009: § 271).

The above discussion shows that a proper construal of the remark to the effect that scientific evidence is not relevant to religious beliefs requires that we do not treat religious beliefs as scientific beliefs. In short, to attempt to explicate the content of religious beliefs in terms of description, explanation and prediction of scientifically verifiable states of affairs is to fall into confusion.

I would now like to return to the objection that some people treat scientific evidence as relevant to religious beliefs – for example, the religious belief in Christ’s resurrection. If this objection reduces to just the claim that becoming acquainted with certain scientific evidence could be the cause of one’s endorsement or rejection of a belief in Christ’s resurrection, then I would say that Wittgenstein would not have any reasons to disagree with the idea that scientific evidence could, in this sense, be relevant to religious beliefs. However, if the objection is intended to mean that a religious belief – such as belief in Christ’s resurrection – is justified by scientific evidence, then in response one should pose a number of questions. Firstly, do scientific or religious criteria decide what kind of observations can count as scientific evidence for Christ’s resurrection? Secondly, what does it mean to say that scientific evidence justifies a religious belief in Christ’s resurrection? Thirdly, should it be expected that such evidence will convince a non-believer that Christ was resurrected (in the religious sense of the word “resurrect” – i.e. that, for example, the Son of God overcame death once and for all)?

With regard to the first question, we may note that if, on the one hand, scientific criteria were to decide it, then in the end the belief in Christ’s resurrection would be – to the extent that it was considered verifiable by scientific evidence – a purely scientific belief. On the other hand, though, if religious criteria were to decide it, then certain observations that would count as scientific evidence for the belief in Christ’s resurrection would be treated as having a religious significance. That is to say, certain observations that play the role of scientific evidence for a certain scientific belief concerning Christ’s body would also be treated as furnishing religious grounds for holding a certain religious belief: the observations in question would not merely justify accepting that, for instance, certain biological processes occurred in Christ’s body, but also that, say, the Son of God overcame death. So these observations, in the context of religious discourse, would not be treated as scientific evidence.

Turning to the second question, the claim that scientific evidence justifies religious beliefs would seem to have no determinate sense, because certain logical (grammatical) connections that hold within scientific and religious discourses do not hold between propositions of the natural sciences and religious statements or judgments. One can say that certain conventions regarding the use of religious and scientific expressions are not adopted in our language. Of course, one can adopt certain conventions that will establish justificatory connections between scientific propositions and religious statements. For instance, some believers may adopt the convention that certain biological processes that took place in Christ’s body have deep religious significance, and prove that the Son of God was resurrected. However, the possibility of making such moves only goes to show that we can ascribe a new and trivial sense to talk of “justification of religious beliefs by scientific evidence”.

In relation to the third question, it seems that it is not irrational to deny that Christ was resurrected in a certain religious sense of the word “resurrect” (corresponding, say, to the thought that the Son of God overcame death), while accepting as reliable scientific evidence observations confirming that Christ died and that, after three days, his dead body then started to function as a living human organism again. Of course, non-believers may adopt various strategies for dealing with such observations. They may claim that scientific theories of the future will one day be able to explain the observed facts, or that human abilities are so limited that no such theories will ever be able to explain such extraordinary, yet completely natural, facts. It is worth adding, also, that it could happen that a certain non-believer, after becoming acquainted with such observations, starts to believe that the Son of God was resurrected. However, such a situation is not properly described by stating that they concluded on the basis of scientific evidence that the Son of God underwent resurrection.

To sum up, the remark that scientific evidence is not relevant to religious beliefs boils down to the assertion that there is no clear and non-trivial sense in which religious beliefs are confirmed or rejected thanks to scientific evidence. Hence, the remark should not be read as a protest against doing something, but as indicating that, in fact, what it is for that to be done has not been determined.

5 Do Believers Agree with Wittgenstein’s Elucidations of Religious Language Games?

As I have already mentioned, Wittgenstein’s approach to religious beliefs and utterances is sometimes criticized as being incompatible with what believers themselves are wont to say about their own faith (Trigg 1973; Bailey 2001; Hyman 2001; Law 2017).Footnote 18 However, there are at least two reasons to question this.Footnote 19 Firstly, such an objection to his remarks on religion may be motivated by the confused conviction that he is formulating some sort of philosophical theory concerning religious beliefs, such as a non-cognitivist position, and by the fact that such a theory is incompatible with how believers interpret their own religious beliefs. Such a motivation for this line of criticism can be questioned because, as I have already pointed out, Wittgenstein’s remarks on religion need not be read as an expression of a certain philosophical standpoint or theory, and so are not incompatible with how believers comprehend their own religious beliefs. The second reason for questioning such an objection is the fact that believers’ explicit interpretations of their own religious beliefs and utterances are not, as a rule, merely uncontroversial explanations of the content of those beliefs or the meaning of those utterances: rather, they themselves are sometimes recognizable as nascent philosophical theories.Footnote 20 (Arrington rightly points out that if this is the case, then Bailey’s criticism of Wittgenstein’s remarks on religious beliefs, according to which the Austrian philosopher offers a mistaken explanatory account of the nature of such beliefs, is unjustified (Arrington 2001).) Wittgenstein rightly seems to reject any approach to understanding the character of one’s utterances or beliefs that would say that the description of the nature of these given by the person uttering these words or holding these beliefs cannot be confused and must necessarily be correct.Footnote 21

For example, Wittgenstein suggests that what mathematicians are inclined to say about the nature of mathematics should not be treated in principle as furnishing an adequate elucidation of its nature. Mathematicians’ statements about the nature of mathematics are rather something that calls for a “philosophical treatment” (cf. Wittgenstein 2009: § 254). It seems that he also embraces a similar approach where statements concerning religious beliefs issued by believers are concerned. According to him, certain interpretations of the nature of religious beliefs and utterances can be confused or inappropriate (cf. Burley 2012: 62):

What seems to me ludicrous about O’Hara is his making it appear to be reasonable. (Wittgenstein 1967a, LRB: 58)

Theology that insists on certain words & phrases & prohibits others makes nothing clearer. (Karl Barth)

It gesticulates with words, as it were, because it wants to say something & does not know how to express it. (Wittgenstein 1998, CV: 96)

Even so, this does not mean that Wittgenstein thinks that if an interpretation of religious beliefs presented by a believer happens to be confused or inappropriate, one should then accept an interpretation of a kind incompatible with that presented by the believer. Indeed, grasping this point is vital for understanding his approach to religious beliefs, and I would say that it is even essential to a proper comprehension of the character of his philosophical investigations more generally. Failure to appreciate this point could lead one to think there is a fundamental incoherence afflicting Wittgenstein’s philosophy – one consisting in the fact that Wittgenstein would then seem to be on the one hand claiming that one should not question participants’ own understanding of a given practice, while on the other hand admitting that those same participants could themselves understand it wrongly. The fact is, though, that there is no such incoherence in his philosophy.

In order to better understand this, one ought to draw a certain distinction, which consists in recognizing that one can conceive understanding a practice in two ways. On the one hand, it may consist in the possession of some competences that are such as to be irreducible to the ability to give just a verbal explanation of the nature of a given practice – making it the type of understanding one might call “implicit”. These competences will include the skills which any participant of the practice has to possess, the capacity to recognize what is important for participants in the practice, and the ability to determine what role its participants are ascribing to certain acts and activities. For example, a mathematician can be said to understand the practice of mathematics because he or she is in a position to prove certain mathematical theorems, understand many mathematical proofs, see connections between various mathematical theorems, and use these to solve mathematical problems.

Where religion is concerned, one can also talk about a form of understanding that is implicit. Of course, that need not mean that there is an exact analogy between an implicit understanding of mathematics and an implicit understanding of religion. Someone who “understands” – in this implicit manner – a certain religion is able to indicate what counts from the point of view of that religion as being important in life, and to identify the roles that believers ascribe to certain rituals and types of action. They will possess these competences either because they are (or used to be) a member of a given religious community, or because they participate (or used to participate) in practices somehow connected with this religion and hold certain attitudes toward the world that are to some degree similar to those that believers have that count as religious ones. The second element of this disjunction is, of course, very vague and indefinite, in that almost any practice can be connected somehow with any other practice. However, its indefiniteness and vagueness should not be treated as a flaw, because it seems that such an implicit understanding is gradable: it is reasonable to assume that if one participates in practices closely related to a given religion, then this makes it more plausible that one will understand this religion better than someone who participates in practices that are only loosely related to it. Moreover, an analogous observation can, I think, also be made with regard to attitudes. As examples of such practices and attitudes one can cite the practice of burying people, the practice of swearing an oath of eternal love, the attitude of gratitude, the sense of the insignificance of our existence, the feeling that the existence of the world is a mystery, and the attitude that takes everything happening to us as possessing a deep sense and not being accidental.

On the other hand, one can conceive understanding a practice in a different way. It may consist merely in the ability to give a verbal explanation of the nature of a given practice – what one might call an “explicit understanding”. It should be emphasized, however, that it is neither necessary to understand a given practice in this way in order to participate in it, nor sufficient in principle to do so.

It is not a necessary condition because, for example, a competent mathematician may not accept any verbal explanations of the nature of mathematics, and a religious person may neither be in a position to formulate, nor be acquainted with, any theories of religion. At the same time, it is not a sufficient condition because, for instance, a person who formulates even a fully adequate explanation of the nature of the game of football may be unable to actually play it. The above remarks show that there is a fundamental difference between an explicit and an implicit understanding of a practice, and that it is even possible that various verbal explanations of the nature of a given practice may not convey any implicit understanding of it at all. For example, in case of a religion, such verbal explanations may not enable us to determine what role believers ascribe to those rituals and ways of acting recognized as being religious in character or important from the religious point of view. But why, we may ask, is it generally the case that verbal explanations of the nature of a given practice tend to not be conducive to an implicit understanding of it? I think that one of the reasons is the fact that such explanations can turn out to be manifestations of a conceptual confusion, and that this can be so no matter whether they are proffered by a believer or a non-believer. The same point can be made, mutatis mutandis, for mathematics and for other practices: the verbal explanations of the nature of mathematics given by mathematicians may be entirely confused. (One cannot exclude a priori the possibility that platonism, intuitionism and formalism in the philosophy of mathematics are completely misguided.) So, a participant in a given practice may not be able to give an adequate verbal explanation of the nature of the practice, and in this sense may not understand the practice, and that is the reason why, for Wittgenstein, what participants in a given practice say about it cannot be taken to be authoritative and may even be seriously confused.

In Wittgenstein’s view, then, we are not obliged to endorse statements expressing an attempt to put forward an explicit understanding of a given practice. At the same time, he does not seek to question those statements that count as moves within a given practice – where the ability to issue these partially constitutes an implicit understanding of the practice. According to him, it would be a philosophical arrogance to question the latter. In short, the aim of philosophy is not to change our language-games, but to describe them:

Philosophy must not interfere in any way with the actual use of language, so it can in the end only describe it. (Wittgenstein 2009: § 124)Footnote 22

It is worth noting, however, that Wittgenstein’s remark should not be read as an expression of a form of linguistic conservatism. He does not mean it as a prescription asserting that philosophy should not interfere “with the actual use of language” because only the actual use is the correct one. Wittgenstein’s point is rather that it would be wrong to call the activity that aims at reforming our language games rather than just describing them as “philosophy”. (One could make this point by invoking Phillips (1999), who says that Wittgenstein’s conception of philosophy is a contemplative one). To be sure, one could criticize this interpretation by pointing out that it also presents Wittgenstein as formulating a specific prescription concerning the word “philosophy”. However, such an objection would not be justified, as Wittgenstein’s remarks on philosophy are meant only to show that there is a radical difference between philosophy and such practices as science, literature, politics and religion. The latter can all be developed because, among other things, people who participate in these practices invent a new vocabulary or start using the old vocabulary in new ways, whereas it is completely unclear what is going on when philosophers start to use the language of science, politics, or religion in a new way. Are they doing philosophy, or are they trying to “do” science, politics or religion? On the one hand, it is obvious that if certain people start to use the language belonging to science, politics, or religion in a new way, they will be treated as attempting to practise science, politics or religion, not as trying to understand the nature of these practices. On the other hand, such philosophers insist that they are doing philosophy, and trying to understand the nature of these practices. According to Wittgenstein, they do not see clearly what they are in fact doing, and cannot really decide what they want to do. Their problem is of the form “I don’t know my way about” (Wittgenstein 2009: § 123).

It is worth emphasizing that the fact that philosophical inquiries can convince us that we should accept certain distinctions – for instance, that between an ‘implicit’ and an ‘explicit’ understanding of a practice – does not entail that someone who distinguishes an ‘implicit’ from an ‘explicit’ understanding is actually propounding a philosophical view or thesis. Moreover, utterances which explain how one should understand a given distinction are not philosophical theses either: they are merely grammatical remarks. So, it is neither the case that someone who claims that the proposition “2 + 3 = 5” is a priori and the proposition “Biden is older than Trump” is a posteriori is propounding philosophical theses, nor the case that someone who explains that the former proposition is a priori because it is not justified by any empirical evidence and the latter is a posteriori because it is so justified is doing so. In short, if philosophy is understood as “marshalling recollections for a particular purpose” (Wittgenstein 2009: § 127), then it turns out that what we say while philosophizing does not serve to express philosophical views, but to remind us of what we already know in respect of how we use our language.

To sum up, if one is prepared to distinguish an implicit understanding of a practice from its explicit understanding, then there is no reason to think that Wittgenstein’s philosophy is incoherent in the way suggested. He is not questioning participants’ understanding of a given practice insofar as this constitutes an implicit understanding of it. Moreover, he thinks that philosophy aims at furnishing us with an expression of this kind of understanding of a practice for the sake of thereby dissolving certain philosophical problems. Such problems, in turn, stem from mistaken interpretations and explanations of the practice being considered – ones that correspond to various ways of understanding the nature of that practice explicitly. So all Wittgenstein is questioning is the idea that, as a rule, various verbal interpretations of the nature of a given practice must be correct.

One can argue that some verbal interpretations of the nature of a religious practice have no impact on the practice in question, and for that reason cannot be treated as belonging to the latter. The claim that they do not have any impact on the practice seems to be justified by the fact that believers whose religious way of life is similar (in the sense that they make similar decisions in similar circumstances, have shared religious beliefs, and participate in a similar fashion in the same religious rituals) can accept explanations of the nature of the practice that vary significantly one from another. So it seems that one can say that in such cases verbal explanations of the nature of a religious practice do not do any real work: such explanations, to use Wittgenstein’s phrase, are just idling (cf. Wittgenstein 2009: § 132).Footnote 23 To be sure, one could protest by pointing to the fact that various theological disputes have exerted a significant impact on many religions, and such a remark would be true, but one should emphasize here that there is a distinction to be made between those explanations and interpretations which are, as it were, ‘internal’, and those which can be said to be ‘external’. On Wittgenstein’s account, the latter amount to philosophical theories or conceptions, whereas the former can be neither endorsed and justified nor rejected by philosophy. Philosophy can only describe internal explanations and interpretations: for example, it can describe them in the context of an examination of the various elements of a religious way of life. Thus, one can still acknowledge them, but only as elements of a given religion: i.e. elements of a certain practice.Footnote 24

As an example of an internal religious explanation that aims, among other things, at showing the connection between the belief in Jesus Christ’s resurrection and the belief in the future resurrection of all people, one can take the following words of St. Paul:

Now if Christ is proclaimed as raised from the dead, how can some of you say there is no resurrection of the dead? If there is no resurrection of the dead, then Christ has not been raised; and if Christ has not been raised, then our proclamation has been in vain and your faith has been in vain. We are even found to be misrepresenting God, because we testified of God that he raised Christ—whom he did not raise if it is true that the dead are not raised. (1 Cor. 15, 12-15)

But in fact Christ has been raised from the dead, the first fruits of those who have died. For since death came through a human being, the resurrection of the dead has also come through a human being; for as all die in Adam, so all will be made alive in Christ. But each in his own order: Christ the first fruits, then at his coming those who belong to Christ. (1 Cor. 15, 20-23)

St. Paul sets out to show that there is a strict connection between Christ’s resurrection and the resurrection of all people. However, his argumentation is not, for instance, of the form that would seek to suggest that if one could not justify in a scientific way the claim that Jesus died and then came back to life, one would have no rational grounds for believing that there will be a moment in the future when all people come back to life. St Paul’s argumentation should rather be read as asserting that the only firm ground for Christian belief in the future resurrection of all people is belief in Jesus Christ’s resurrection, and that belief in Jesus Christ’s resurrection is something unquestionable.

In order to better understand the distinction between internal and external interpretations of religion here, it may also be worthwhile to contrast St. Paul’s words about the belief in Jesus Christ’s resurrection with what philosophy could say about this belief. On the one hand, philosophers’ external explanations and interpretations could aim to make the grammar of the belief in Christ’s resurrection clear: they can, for example, point out that this belief is – to use Wittgenstein’s phrase from On Certainty – the central element of the Christian world-picture. On the other hand, they could try to present certain “theories” of resurrection: namely, that resurrection is possible in virtue of the fact that the body of a resurrected person is subject to an as yet unknown biological process, or that resurrection consists in the fact that the body of a resurrected person is subject to a certain metaphysically comprehended process that is empirically unverifiable.

The conclusion that I would wish to draw from this discussion of the distinction between internal and external interpretations and explanations of a given practice is essentially this: that one can, after all, distinguish philosophical explanations of a given practice from moves made within that practice, because only internal interpretations and explanations actually count as elements of the practice itself. Thus, my argument to the effect that Wittgenstein’s approach to religious beliefs and utterances is not incoherent is not in fact based on some sort of questionable presupposition that is simply assumed. Wittgenstein the philosopher does not seek to question any internal interpretations of a religion, even if Wittgenstein the person may not accept them and may even reject them. And so this possibility of rejecting them does not have a philosophical, but rather a religious or existential character. According to Wittgenstein, philosophers can only question philosophical interpretations of religious beliefs and utterances. Moreover, this very remark applies also to those philosophical explanations of religious beliefs and utterances that are formulated by believers themselvesFootnote 25 – just as, by the same token, a philosopher can only question philosophical interpretations of mathematical theorems formulated by mathematicians, and not the theorems themselves.

6 A Brief Comparison with the Readings of Phillips, Schönbaumsfeld and Mulhall

My interpretation of Wittgenstein’s remarks on religion is largely convergent with certain interpretations already advanced by several other authors. However, I would like to focus here on the features of my reading of the remarks that distinguish it from the readings presented by Phillips, Schönbaumsfeld and Mulhall. At the same time, it is worth noting that what primarily distinguishes my interpretation from others’ is a specific juxtaposition of ideas that has already appeared – albeit in a more or less developed form – in those readings.

Firstly, on my interpretation, Wittgenstein’s remarks do not express a certain view regarding religious belief – if this is understood as something furnishing an alternative opinion to those held by philosophers. So, they do not express an alternative view to, for example, Plantinga’s or Swinburne’s positions. This seems to be a point of difference between Schönbaumsfeld’s reading and my own, in that there is a suggestion in her book that Wittgenstein presents a vision of religious belief that amounts to an alternative to Swinburne’s view (Schönbaumsfeld 2007: 157). Of course, this difference may be only a question of terminology – the word “vision” can be understood in various ways.

Secondly, my interpretation seems to be, in a sense, both more minimalist and more radical than Phillips’ and Schönbaumsfeld’s readings. In order to explicate what I mean by “minimalist” and “radical”, it will be helpful, however, to first point out that I largely concur with Phillips and Schönbaumsfeld as regards the following points: that Wittgenstein’s philosophical remarks on religion are grammatical remarks (Phillips 1993e: 243, Schönbaumsfeld 2007: 165), and that they describe, amongst other things, the circumstances in which religious language is used or religious beliefs are accepted or rejected, together with the consequences drawn by believers and non-believers from the endorsement or rejection of religious statements or beliefs.

My reading is minimalistic in the sense that it emphasizes that philosophy contributes to an understanding of religion and religious belief only through giving descriptions of this kind. It should be noted that such descriptions are not, contrary to Phillips and Schönbaumsfeld, grounds for concluding that certain views or beliefs are mistaken – for instance, ones about immortality or the existence of an omnipotent being. (As regards the former example, the tension between Phillips’ conception of philosophy and his approach to particular philosophical questions, especially that of the immortality of the soul (cf. Phillips 2005b), is interestingly discussed by Burley, who writes that “[t]he danger in Phillips’ approach is that by dismissing philosophical reconstructions of the religious beliefs in question he can easily appear to be dismissing the religious beliefs themselves as inherently confused” (Burley 2012: 112). In relation to the latter example, see Schönbaumsfeld (2007: 160).) According to my reading of Wittgenstein’s remarks, if such views or beliefs exist at all, philosophy cannot show that they are mistaken, because its task is neither to draw any conclusions nor to defend any opinions that would be controversial (Wittgenstein 2009: § 128; 2001: 97).

My reading is radical in the sense that it takes Wittgenstein’s remarks to be an expression of a philosophical criticism of certain statements purporting to be about the nature of religion and religious belief, but only insofar as those remarks are to be understood as showing that such statements are either nonsensical or not about religion or religious belief. In other words, the Wittgensteinian critique shows that the alleged explanations of the nature of religion and religious belief are in fact certain strings of signs that are either devoid of meaning or have a meaning, but one not acknowledged by the authors of these explanations as the intended one. To sum up, and unless I am mistaken, the interpretation presented above stresses in a more emphatic and radical way than either Phillips or Schönbaumsfeld that Wittgenstein is not criticizing any views about religion, but merely helping us to see that certain sentences that seem to express such views do not in fact express any coherent position regarding religion, and instead only engender confusion of one kind or another.

Now, let us briefly compare Mulhall’s reading with my interpretation. According to him, Wittgenstein recognizes religious utterances as nonsensical: “Wittgenstein’s view of ethico-religious utterances (early and late) is that they are sheerly nonsensical (as opposed to being instances of illuminating nonsense, inevitably misfiring attempts to articulate what is ineffably so), and that that is precisely why we have a use for them” (Mulhall 2015: 21–22; see also 2018). In my opinion, though, it is a mistake to say that the later Wittgenstein treats religious utterances as sheerly nonsensical. (Moreover, contrary to what Mulhall suggests, Diamond does not seem to share his conviction that, for the later Wittgenstein, religious utterances are sheerly nonsensical. Of course, Diamond points out that certain religious utterances have a unique character or status, since “to be a great riddle is to ‘allude’ to language whose full transparency to us is ruled out” (Diamond 1991c: 282), but she does not describe them as nonsense: she talks of them in terms of what she calls “promissory meaning”.) The later Wittgenstein rather points out that if one wants to find out what sense, if any, is ascribed to religious utterances, one has to attend to their use. Of course, this answer may seem unsatisfactory, because it does not settle the question of whether one should treat religious utterances as nonsensical if, for example, one finds out that they generate paradoxes. Yet on my view, even if Wittgenstein is in a sense inclined to acknowledge that linguistic constructions that generate paradoxes can be called nonsensical, he also seems inclined to acknowledge that there is a vital difference between such a construction as “this sentence is false” and descriptions of God that apparently generate paradoxes. The former, for him, is a useless construction that does not play any important role in our life: “it is of no use” (Wittgenstein 1976: 207). The latter, meanwhile, fulfil a central role in the lives of many believers. One might say that the difference I am pointing to here between Mulhall’s interpretation and my own is merely terminological, with Mulhall insisting on applying the word “nonsensical” to certain religious utterances and I myself refraining from doing so. And the difference may perhaps seem inessential. Yet I do think that my approach is more faithful to the spirit of Wittgenstein’s philosophy. Indeed, were we to take the liberty of substituting “Philosophy” for “Theology” in the following remark by Wittgenstein, it might then be adduced in support of my reading:

Theology [read “Philosophy”] that insists on certain words & phrases & prohibits others makes nothing clearer. ...

It gesticulates with words, as it were, because it wants to say something & does not know how to express it. Practice gives the words their sense.” (Wittgenstein 1998, CV: 96)

7 Concluding Summary

In what follows below, I would like to briefly recapitulate the conclusions that, in my view, can be drawn from my discussion of Wittgenstein’s remarks on religion.

There are no convincing reasons for claiming that these remarks cannot be treated as the expression of a certain philosophy of religion – though this is one that is undoubtedly less developed than his philosophies of mathematics and psychology.

Wittgenstein’s philosophy of religion, as well as his whole philosophy, exhibits a quite specific character: it aims to make clear what we say and what we think. It pursues, among other things, a clarification of the role played by religious beliefs and utterances in the lives of believers and non-believers. According to Wittgenstein, it is not the task of philosophy to validate or undermine such beliefs and utterances.

How is it possible to achieve this clarity? Wittgenstein thinks that it is possible through giving a description of various aspects of religious beliefs and utterances. Such a description consists, inter alia, in determining the circumstances in which one makes religious statements and the consequences one is inclined to draw from accepting these statements.

However, one should note that some philosophers may misunderstand the role such descriptions play in clarifying religious language-games. For instance, Wittgenstein does not mean that if, as a consequence of accepting the statement “God will judge all our deeds at the Last Judgment”, we treat even the smallest decisions as extremely significant, then the sense of this statement can be explained in terms of the idea that it expresses a certain attitude or prescription requiring us to treat all our decisions as extremely significant. He warns us against such an interpretation of the role that descriptions can play in the context of philosophical investigations, discouraging us from any such “one-sided diet”, in that it leads us to confusions. I would add that one such philosophical confusion is a kind of reductionism in the philosophy of religion which has been mistakenly ascribed by some commentators to Wittgenstein himself.

The aim of Wittgenstein’s philosophy of religion is, in a sense, very modest. It is simply to remind us about what we – believers and non-believers – know about religion. Wittgenstein reminds us of what believers, qua believers, do and say. According to him, philosophy cannot give us anything more, and this, of course, does not mean that there is something else to be given by philosophy, but which philosophy is unable to give it to us; rather, it simply means that if a certain activity or certain investigations offer us something more, then this activity is not philosophy and these investigations are not philosophical ones. Moreover, such reminders are all that is needed to achieve philosophical clarity – including, in our case, with respect to religious beliefs and utterances. One reason why philosophy does not furnish us with anything more is that all attempts to either validate or undermine religious faith belong, in fact, to some kind of either religious or anti-religious discourse, and not to philosophy itself: just as meta-mathematics is not philosophy, so an apology for or critique of religion is not philosophy. Of course, these discourses can be objects of philosophical reflection, but the solutions to the problems considered within these discourses are not solutions to philosophical problems.

A second reason why philosophy does not yield anything more is that it does not aim to provide us with causal or functional explanations – explanations, that is, of the phenomenon of religion of the sort that various scientific disciplines, such as psychology and sociology, are able to propose. Causal and functional explanations of religious belief do not count as philosophically elucidatory clarifications. Moreover, as Wittgenstein emphasizes, in order to achieve clarity with regard to the essence of religious beliefs, one need only give a description of the role played by these beliefs in our lives. Such a description will be a description of the grammar of religious language. When, for example, Wittgenstein investigates the essence of the belief in Christ’s resurrection, his idea is that such investigations are concerned with what we would call “belief in Christ’s resurrection”, and in order to pin this down what one needs to do is furnish criteria for the use of this expression, so that one must point out, inter alia, the grounds cited by believers for affirming the statement “Christ is resurrected”, and the consequences they draw from their acceptance of it. Of course, it is by no means an easy task to identify criteria for the use of expressions belonging to religious language-games; however, the difficulty here does not lie in this demanding the discovery of some new facts, but in its requiring us not to ignore or forget any of those aspects of religious practices right before our eyes.