Sheer Poison? Wittgenstein’s Philosophy of Religion

Abstract

Anscombe once said to Anthony Kenny that “On the topic of religion, Wittgenstein is sheer poison”. This paper offers an assessment of that view. I take it that Anscombe meant that Wittgenstein was a bad influence rather than that his views were necessarily false, although she seems to have been uncertain about what exactly his views were. In “Paganism, Superstition and Philosophy”, she identifies five ideas that make up “a certain current in philosophy which has a strong historical connection with Wittgenstein”. I identify some of the sources of these ideas, in Wittgenstein’s writing and in work by some of his followers, and consider what Anscombe’s objections to them might have been. I also look at whether we should think of these ideas as belonging either to Wittgenstein’s philosophy of religion or to his personal beliefs. This will involve some consideration of how far we can, or should try to, separate the personal from the philosophical. So far as he held objectionable views about religion, I argue that these ought to be considered personal rather than philosophical.

1. Introduction

Elizabeth Anscombe once said to Anthony Kenny (2019) that “On the topic of religion, Wittgenstein is sheer poison.” (Kenny 2019, pp. 12–22, p. 20). This paper offers an assessment of that view. I will also try to understand Anscombe’s perspective on the philosophy of religion, from which Wittgenstein might have appeared to be poisonous. This will involve going into some detail on her religious beliefs and philosophy of religion. Readers who care only about whether Wittgenstein was guilty as charged, and not Anscombe’s reasons for bringing the charges, might want to skip Section 4. Anscombe probably meant that Wittgenstein was a bad influence rather than that his views were necessarily false, although she seems to have been uncertain about what exactly his views were. In “Paganism, Superstition and Philosophy” (Anscombe 2008a), she identifies five ideas that make up “a certain current in philosophy which has a strong historical connection with Wittgenstein.” (Anscombe 2008a, pp. 49–60, p. 58). I will identify some of the sources of these ideas, in Wittgenstein’s writing and in work by some of his followers, and consider what Anscombe’s objections to them might have been. In Section 3, I will also look at whether we should think of these ideas as belonging either to Wittgenstein’s philosophy of religion or to his personal beliefs. This will involve some consideration of how far we can, or should try to, separate the personal from the philosophical. In Section 4, I will return to Anscombe’s views and what they suggest about faith and the kind of Wittgensteinian philosophy of religion that she rejects. Given Wittgenstein’s explicit comments on how he thinks philosophy ought to be conducted, we should not treat his personal opinions as philosophical theses.

2. Wittgensteinian Philosophy of Religion

Anscombe’s respect for Wittgenstein is well known, so it is surprising to find her calling him “sheer poison.”1 It is unlikely that she meant he had false beliefs, since she was not sure exactly what he believed and was reluctant to attribute false beliefs to him, even if she suspected he may have held them. Evidence for this comes from the following anecdote. She once asked Wittgenstein what he would do if he had a friend from a culture that practiced witchcraft who intended to become a witch doctor. Would Wittgenstein try to dissuade this friend from doing so? “We walked in silence for a space and then he said: ‘I would, but I don’t know why’.” (Anscombe 2008a, p. 58). She says of this that she is inclined “to think that a vestige of the true religion spoke in him then.” (Anscombe 2008a, p. 58). She does not appear to like the thought of Wittgenstein’s having false beliefs on such an important matter. This generous (or, perhaps some might think, patronizing) approach to interpreting Wittgenstein’s remarks can also be seen in her saying that ‘Truth’ is “one of the names of God” in connection with her memory of Wittgenstein’s having said that he loved the truth, although the idea of loving God was one with which he had difficulty (see Anscombe 2011, p. 71).

However, as mentioned already, she also criticizes what she describes as “a certain current in philosophy which has a strong historical connection with Wittgenstein.” (Anscombe 2008a, p. 58). She seems to be choosing her words carefully here, so as to bring out a connection with Wittgenstein without actually saying that the ideas are his. The current in question has the five following component parts:

• Religions, being radically unlike collections of scientific claims, are not true or false: “There is no such thing as a religion’s being true”.
• Religious belief is more like being in love than “believing anything true or false”.
• Complete honesty is required for a prayer to count, and this is discovered through personal experience: “by being confronted with a personal problem”, a person “who has been taught to pray finds out that you have to be completely honest for your utterance to count as prayer”.
• Within religion, “So far as realisation and understanding are concerned, what is in question is an attitude which one man may have and another lack”.
• Sometimes in religious matters (i.e., religion and adjacent attitudes, feelings, etc.), one misuses language in a way that expresses something important but cannot be false. For instance, I might think that nothing anyone does can harm me “because in some inexpressible way nothing that might happen to me at anyone’s hands would count as harm. Why? There isn’t a reason.” (see Anscombe 2008a, pp. 58–59).2

Some of these ideas, or at least ideas like them, can be traced back to Wittgenstein’s writings and can be found in Culture and Value (in the case of points 1, 2, and 4) (Wittgenstein 1998) and the Lecture on Ethics (in the case of point 5) (Wittgenstein 1993).3 We might have to look elsewhere for the origin of point 3. I will say little about points 4 and 5, partly because both are connected with the idea that religious belief is not truth-apt, which is already covered by points 1 and 2, and partly because I take Anscombe’s objection to point 5 to be simply that it presents parts of religion as nonsense. Wittgenstein is explicit about this in the Lecture on Ethics, and it is not hard to imagine why a religious believer would object, especially one who believes that religion is concerned with truth. Before I say more about points 1, 2, and 3, however, it will help to look more closely at what Anscombe says about Wittgenstein’s attitude toward religion.

She mentions his having raised in a notebook the question of whether “it would matter for the Christian religion whether Christ did any of the things recorded of him, or indeed existed at all”. (Anscombe 2008a, p. 58). She also says that Wittgenstein once “reproved a Jain friend for speaking somewhat contemptuously of the Jain beliefs about the journeyings of the soul after death” and that he said that, if a friend came to him with religious doubts, “he would probably raise doubts about the doubts” (Anscombe 2008a, p. 58)4. Of all this, she then comments:

Now I can’t speak for Wittgenstein, I only report one or two things here from his tongue or pen. But there is an attitude or fairly characteristic strain of thought in some of his followers which seems like a partial reflection of some of what I have quoted, though as I will formulate it I don’t mostly claim that anyone has definitely spoken so.

(Anscombe 2008a, p. 58)

It seems likely that the remark mentioned here from Wittgenstein’s notebooks is the one that appears in translation in Culture and Value, as follows:

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. The 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.

(Wittgenstein 1998, pp. 37e–38e)5

This seems relevant to points 1, 2, and 4 from Anscombe’s list—religious beliefs are neither true nor false (1), being religious is more like being in love than ascribing to certain propositions (2), and religion is a matter of attitude (4).

It is possible that she might also have D. Z. Phillips in mind for at least some of the current of ideas that she identifies, especially regarding point 3.6 In The Concept of Prayer, Phillips (2014) makes a number of claims with which Anscombe might have disagreed. When he first raises the question of what is to count as prayer (see Phillips 2014, p. 130), he notes there that he has already distinguished between prayer and pseudo-prayer. This could be relevant to Anscombe’s third point about what it allegedly takes for prayer to count. In his chapter on “Superstition and Petitionary Prayer”, he distinguishes prayer from incantations and spells, writing the following:

The importance of prayer, to a large extent, depends on the role it plays in the life of the person who offers it. If the prayer is an isolated occurrence, having little or no relation to the life of the individual prior to or after the act of ‘praying’, one tends to doubt whether what has been said can illustrate what prayer is. One might put forward a general thesis that the more tenuous the relation between the prayer and the rest of the person’s life, the more suspect the prayer becomes; the likelihood of superstition increases.

(Phillips 2014, p. 115)

Phillips does not explicitly mention honesty here, as Anscombe does in her point 3, but if a person never otherwise prays and their prayer has only a tenuous relation to the rest of their life, then one might be inclined to doubt the sincerity of their belief in God. Phillips is also sceptical about much petitionary prayer, as follows:

When deep religious believers pray for something, they are not so much asking God to bring this about, but in a way telling Him of the strength of their desires. They realize that things may not go as they wish, but they are asking to be able to go on living whatever happens. In prayers of confession and in prayers of petition, the believer is trying to find a meaning and a hope that will deliver him from the elements in his life which threaten to destroy it: in the first case, his guilt, and in the second case, his desires.

(Phillips 2014, p. 121)

Anscombe prayed for at least one of her children when they were sick or injured. Her famous habit of smoking cigars came about because she promised God to give up cigarettes when her son Charles was seriously ill (see Richter 2023, p. 38). Was she then not so much asking God to bring about her child’s recovery, but asking “to be able to go on living whatever happens”? Would she recognize this as an accurate description of what she was doing? I doubt it.

Part 3 of the current that Anscombe opposes is the claim that a person who has been taught how to pray finds out, by being confronted with a personal problem, that their prayer must be completely honest in order to count. Anscombe was, in this case, confronted by a grave personal problem, about which she prayed. According to Phillips, if she were a deep believer, she would not have been asking for her child’s life to be saved, but, rather, to be able to go on living if he died. She would have been, in his view, telling God the strength of her desire that her son should live, and if the prayer had been an isolated occurrence, then it would more likely have been mere superstition. Anscombe, I suspect, believed that she really was praying for her son’s life and that anyone else in similar circumstances might well be doing the same thing, even if they rarely prayed otherwise. There is a saying that there are no atheists in a foxhole. What Phillips says suggests that anyone who ceases to be an atheist in that kind of danger is merely superstitious. Anscombe’s view might be that some people only realize the truth of God’s existence in extreme danger.

What Phillips says about petitionary prayer does not appear to match Anscombe’s point 3 perfectly. He does emphasize the importance of honesty and raise the question of what really counts as genuine prayer, but he does not say that one discovers through facing a personal problem that complete honesty is required for prayer to count as prayer. Then again, as we have seen, Anscombe does not “mostly claim that anyone has definitely” spoken in just the ways that she identifies.

An even likelier source than Phillips, especially for Anscombe’s point 3, is Peter Winch.7 In his paper “Can a Good Man be Harmed?”, Winch quotes Mark Twain’s Huckleberry Finn, where Huck says “You can’t pray a lie—I found that out” (Winch 1966, p. 56). Winch says of this that what Huck has “become aware of is a necessary truth” (Winch 1966, p. 57). In the novel, Huck finds out that you cannot pray a lie by trying and failing to say in prayer that he would “do the right thing and the clean thing, and go and” inform an escaped slave’s owner where the slave now was (quoted in Winch 1966, p. 56). The lie, presumably, is that informing in this way is right and clean. Or perhaps it is simply the claim that this is what Huck was going to do, since he does not bring himself to do it. Either way, by trying to make a false claim in a prayer, he comes to realize that prayer requires honesty in order to count.

What he realizes, Winch says, is “an analytic truth and perhaps one that he had been taught before in being shown what it is to pray” (Winch 1966, pp. 57–58). In this sense, it is something that he already knew. The problems that he now faces, however, the problems involved in helping an escaped slave when one has been raised to believe that this is something one ought not to do, enable him to appreciate the point and force of this truth for the first time. Perhaps Anscombe also had Phillips in the back of her mind, but Winch seems like the likeliest source of the third part of the current that Anscombe describes.

What about parts 1 and 2 of this current? Although part 1 insists that religion is not like science, this does not mean that the two can have no similarities. For instance, if they are alike in being different language games or spheres of discourse, then this would support the claim that truth and falsity in the normal or scientific sense do not belong to religion. Phillips talks about religion as being, like science, a kind of domain of its own, and quotes Winch and Norman Malcolm in support of his view (see Phillips 2014, pp. 23–25). Winch writes the following:

criteria of logic are not a direct gift of God, but arise out of, and are only intelligible in the context of, ways of living or modes of social life. It follows that one cannot apply criteria of logic to modes of social life as such. For instance, science is one such mode and religion is another; and each has criteria of intelligibility peculiar to itself. So within science or religion actions can be logical or illogical: in science, for example, it would be illogical to refuse to be bound by the results of a properly carried out experiment; in religion it would be illogical to suppose that one could pit one’s strength against God’s; and so on. But we cannot sensibly say that either the practice of science itself or that of religion is either illogical or logical; both are non-logical…

(See Phillips 2014, p. 24)8

Such modes of social life, Winch and Phillips acknowledge, can overlap, and so are not all sealed off from one another, as Phillips notes in a footnote appended to this quotation. This matters because Kai Nielsen has famously argued that there is a “cluster of dark sayings” (Nielsen 1967, p. 192), more or less Wittgensteinian, that have a tendency to generate what he calls Wittgensteinian fideism. Such a fideist believes that “Philosophy cannot relevantly criticise religion; it can only display for us the workings, the style of functioning, of religious discourse” (Nielsen 1967, p. 193). Winch can seem to be saying that religion, as its own mode of social life with its own criteria of logic, is protectively sealed off from rational criticism. Whether Wittgenstein, Winch, or Phillips is rightly read as being such a fideist has been disputed, but there is ammunition in what I have just quoted from Winch for a Wittgensteinian fideist to use in order to try to protect religion from rational attack or defense. Anscombe would surely not want religion to be defended or protected in this kind of way. She distinguishes between true and false religion, and false religion does not get off the hook by being non-logical. True religion is defended by being shown to be true, or, at least, by combating attacks on it.

If we wanted to take Anscombe’s side against Winch, one way we might question his view would be to doubt his use of the word ‘illogical.’ It might be bad science to ignore the results of an experiment, but would it be illogical to do so? He presumably has in mind the idea that science has a kind of logic of its own. If this is better referred to as something like a set of ground rules, then his point might still stand. But if one experiment, however properly carried out, produces anomalous results, might it not be all right for scientists to set these results aside while other work goes ahead, or until the anomaly is explained or resolved? This seems to be a question that might have to be looked at on a case-by-case basis. The same might go for talk about pitting one’s strength against God’s. Trying to fight an omnipotent being certainly sounds not merely foolish but absurd, and yet we find Jacob wrestling with God in Genesis 32:24–30. There is mystery here, but not what I would call a violation of the logic of religion or religious language.

Winch and Phillips might have shared what Anscombe calls Wittgenstein’s dislike of rationality, or would-be rationality, in religion. He would describe this with a characteristic simile: there is something all jagged and irregular, and some people have a desire to encase it in a smooth ball: looking within you see the jagged edges and spikes, but a smooth surface has been constructed. He preferred it left jagged. I don’t know how to distribute this between philosophical observation on the one hand and personal reaction on the other.

(Anscombe 1981b, p. 122)

One question to consider in connection with this is what exactly Wittgenstein’s simile means (and whether he is correct). Another is what Anscombe means by the distinction between philosophical observation and personal reaction (and whether she is correct). She goes on to say that “Wittgenstein’s attitude to the whole of religion in a way assimilated it to the mysteries: thus he detested natural theology. But again, what part of this was philosophical (and therefore something which, if right, others ought to see) and what part personal, it is difficult to say.” (Anscombe 1981b, p. 123). This clears up what she means about this distinction. The philosophical is what others ought to accept if it is right, while the personal is more a matter of preference or attitude, about which reasonable people might disagree.

The simile of the irregular mass covered over with a smooth ball is reminiscent of Wittgenstein’s image of a sharp copy of a blurred original in Philosophical Investigations §77 (Wittgenstein 1958). It might help us to understand what he thinks about religion if we look at what he says about this other simile. Here, he writes the following:

[I]magine having to sketch a sharply defined picture ‘corresponding’ to a blurred one. In the latter there is a blurred red rectangle: for it you put down a sharply defined one. Of course—several such sharply defined rectangles can be drawn to correspond to the indefinite one.—But if the colours in the original merge without a hint of any outline won’t it become a hopeless task to draw a sharp picture corresponding to the blurred one? Won’t you then have to say: “Here I might just as well draw a circle or heart as a rectangle, for all the colours merge. Anything—and nothing—is right.”—And this is the position you are in if you look for definitions corresponding to our concepts in aesthetics or ethics.

There are possibly significant differences between the contrasts here. In one, the jagged and irregular is contrasted with the smooth and the spherical, while in the other, the blurry and merging is contrasted with the sharp and distinct. With regard to ethics and aesthetics, Wittgenstein implies that important concepts cannot be defined. With regard to religion, however, he is not necessarily talking about defining concepts. In each case, however, something messy is tidied up, and Wittgenstein seems to be (in the case of ethics and aesthetics) or simply is (in the case of religion) against such attempts to make the crooked straight and the rough places plain. We should, he seems to believe, look and see how things are and then describe them accurately, not make them out to be neater than they are or more like some ideal that we believe they should resemble.

An important test case for his view is Roman Catholicism, because the Catholic Church is philosophically sophisticated and has views that can seem at odds with some of Wittgenstein’s. It does not take itself to have a mere mess of beliefs and practices, as might be suggested by the words “something all jagged and irregular” (Anscombe 1981b, p. 122). Hence, perhaps, why Anscombe wrote the following:

Of his later work Wittgenstein said ‘Its advantage is that if you believe, say, Spinoza or Kant this interferes with what you can believe in religion but if you believe me nothing of the sort.’ I do not know whether he was right about this.

(Anscombe 1954; quoted in Haldane 2020, p. 57)

The Roman Catholic Church teaches that God’s existence can be proved and seems to be committed to natural theology. As Anscombe says, “In natural theology there is attempted reasoning from the objects of the world to something outside the world. Wittgenstein certainly worked and thought in a tradition for which this was impossible” (Anscombe 1981b, p. 123). Indeed, the Catholic Church has decreed that, “If anyone says that the one, true God, our creator and lord, cannot be known with certainty from the things that have been made, by the natural light of human reason: let him be anathema.” (First Vatican Council 1868. Emphasis in the original).

If Wittgenstein is anathema, then it is interesting to consider whether this is because of a personal view of his or because of something to which he is philosophically committed. It might be very important, to Catholic Wittgensteinians, for instance, to work this out. If it is not the result of a philosophical commitment, is this because we can interpret his philosophy in a non-obvious way or because we can interpret the Church’s decree in a non-obvious way (as an expression of a certain kind of passion, for instance, rather than as a kind of philosophical thesis)? Charitable readings of both Wittgenstein and the Church suggest that he personally falls foul of this decree, but that his philosophy does not imply that his followers must do so. I will return to this question in Section 3.

The second part of the current that Anscombe criticizes, and the last that I shall discuss here, is the idea that religious belief is more like being in love than believing a proposition. An obvious place to consider as the possible origin of this idea is the remarks collected in Culture and Value. For instance, we have the passage quoted above from 1937 about the Gospels being seized on lovingly. Also, in 1946, Wittgenstein wrote that, “Amongst other things Christianity says, I believe, that sound doctrines are all useless. That you have to change your life. […] For a sound doctrine need not seize you; you can follow it, like a doctor’s prescription.—But here you have to be seized & turned around by something.—(I.e. this is how I understand it.)” (Wittgenstein 1998, p. 61e).

The parenthetical remark seems to make it clear that Wittgenstein is speaking personally here, or at least tentatively. The rest of what he says here does not do this, but given that its source is a notebook rather than anything prepared for publication in a philosophical work, I think it is wise to treat it as no more than a personal remark.

3. The Personal and the Philosophical

Wittgenstein’s personal views on religion, which might be called a philosophy of religion, are of a particular type. William James, Tolstoy, and Kierkegaard are relevant to this kind of understanding of religion. It is practical, non-rational, and passionate. James writes about religion as a psychologist and defines it as “the feelings, acts, and experiences of individual men in their solitude, so far as they apprehend themselves to stand in relation to whatever they may consider the divine” (James 2009, p. 27).9 He is explicitly not concerned with religious institutions or theology (see James 2009, p. 26). Tolstoy, similarly, regards “Christianity neither as an inclusive divine revelation nor as an historical phenomenon, but as a teaching which gives us the meaning of life” (Tolstoy 1892). Kierkegaard’s pseudonym Johannes Climacus implies that passionate idolatry is preferable to Christian worship in the wrong spirit (Kierkegaard 1974, p. 179f). Wittgenstein’s apparent adoption of a generally Jamesian, Tolstoian, and Kierkegaardian view of religion is perhaps not strictly incompatible with Roman Catholicism, but it certainly has a different flavor. The Catholic Church shows more concern for what is true than for what works in a self-help kind of way, and where others might regard faith as taking priority over reason, the Church tends to see the two as going harmoniously together. Pope John Paul II’s encyclical Fides et Ratio, for instance, begins by saying that “Faith and reason are like two wings on which the human spirit rises to the contemplation of truth” (John Paul II 1998).

Wittgenstein’s philosophy, in what we might call the official sense, in contrast with his idiosyncratic personal views, consists of a method or methods (see Wittgenstein 1958, §133, which says both that “we now demonstrate a method” and that “There is not a philosophical method, though there are indeed methods”) and does not argue for any theses (see Wittgenstein 1958, §128). So, it is doubtful whether it really could be incompatible with Catholic Christianity, and, one might think, it cannot conflict with any religion. How could a method for dealing with philosophical problems be incompatible with a set of religious beliefs? Bringing “words back from their metaphysical to their everyday use” (Wittgenstein 1958, §115) does not mean, for instance, rejecting belief in everything supernatural. Wittgensteinian philosophy does not reject Catholicism because, on the contrary, it “leaves everything as it is” (Wittgenstein 1958, §124). This is why one can agree with Wittgenstein’s later philosophy whatever one’s religious beliefs might be, while the same is not the case if one agrees with Spinoza or Kant, for instance.

What, though, we might wonder, about Wittgenstein’s assertion that philosophy “cannot give” the actual use of language “any foundation”? (Wittgenstein 1958, §124). Does this not imply that attempts to provide God-talk with a foundation are misguided? Wittgenstein talks in Philosophical Investigations §122 and §124 of what philosophy is “for us”, and when he practices philosophy, he does not go in for constructive metaphysical projects. So, philosophical proofs of the existence of God are not what he has in mind when he talks about philosophy as he conceives it and recommends that it ought to be practiced. But this does not mean that he is rejecting such proofs on philosophical grounds. That is, Wittgenstein recommends going about philosophy in a certain kind of way, and this way does not include trying to prove anything. It does not follow that he denies the existence of God, nor even that he denies that it is possible to prove the existence of God, or so we might think based on what he says about how philosophy ought to be practiced and about what philosophy, as he (re-)conceives it, is. What, though, if we attend not only to what he says, but to what he does? Does his introduction of a new conception of philosophy not imply a rejection of most of the work of philosophers such as, for instance, Aquinas? Why would we need Wittgenstein’s new method or methods if the old ways were all right? Is this not at least in tension with the Church’s view that God can be known by the light of human reason?

Joseph Cardinal Ratzinger, the year before he became Pope Benedict XVI in 2005, drew on the first two parts of the current that Anscombe identifies to connect Wittgensteinian ideas about religion with those of Rudolf Bultmann and of many unnamed Catholic preachers (see Ratzinger 2004, pp. 215–16).10 He cites the page of Culture and Value from which I quoted above and a page from On Certainty in which Wittgenstein mentions the thought-guiding or -supporting idea that “Every human being has parents” and asserts “that the idea of ‘agreement with reality’ does not have any clear application” in every case. (Wittgenstein 1975, p. 29e)11 It is not clear exactly which passages Ratzinger finds objectionable in these pages of Wittgenstein’s work, but it is clear that he sees Wittgenstein as one of many people who treat religion as a kind of “play”, “as a kind of being-in-love”, and as a “beautiful fiction” (Ratzinger 2004, p. 216). I disagree with this reading of Wittgenstein, but I think that Ratzinger is right to see a tension between Wittgenstein’s later philosophy and the kind of religion that he and Anscombe believed in.

There is a tension, I think, but not a contradiction. Attempts to show that we can have certain knowledge of the one, true God are not going to count as what Wittgenstein calls philosophy, but it is also no part of philosophy so conceived that it defends any such thesis as “It is impossible to prove the existence of God” or “Talk of ‘proving the existence of God’ is nonsense.” Paul O’Grady addresses this point in a recent paper (O’Grady 2024). According to him, Wittgenstein’s

unsystematic, therapeutic approach to philosophy has a huge substantive claim implicit in its practice, namely, that no positive constructive approach to philosophy is viable.

(O’Grady 2024, p. 331)

One solution to the alleged dilemma of wanting to avoid making positive philosophical claims about what philosophy can or cannot do while also using a method that allegedly presupposes that it cannot do anything constructive is to use Wittgensteinian methods for destructive, and perhaps also testing, purposes, but not for constructive ones. This goes against the (alleged) implicit claim, but it still allows for piecemeal engagement with any positive claim that might be made in support of, say, the existence of God or free will, or whatever it might be. Thus, if someone produces alleged proof of the existence of God or the immortality of the soul, say, then we can check to see if it really makes sense, try to obtain a perspicuous overview of the relevant areas of language, and so on. A committed Wittgensteinian might be convinced that there must be some mistake, some nonsense, somewhere in the purported proof. But a Wittgensteinian who is true to Wittgenstein’s word will not dismiss something as nonsense unless and until the anticipated nonsense has actually been unearthed, and there ought not to be any presumption that it will be.

This kind of approach might be especially attractive to a Catholic philosopher such as Anscombe. Another potential advantage of this kind of approach to philosophy is integrity. One does not turn what is, at most, a claim implicit in a practice into a dogma, and one thinks one’s thoughts through, perhaps eventually coming to see that they are nonsense, but perhaps not, rather than prematurely writing them off as metaphysical-and-therefore-nonsense. Wittgenstein is not a logical positivist. Philosophy practiced as he recommends is not supposed to encourage intellectual laziness or fear of rogue conclusions. Houses of cards are to be destroyed (see Wittgenstein 1958, §118), but a properly Wittgensteinian philosopher will avoid dogmatism and any “preconceived idea to which reality must correspond” (Wittgenstein 1958, §131). Problems are to be dealt with one at a time and, as with the method described in Tractatus 6.53, in an ad hoc way. There is no a priori showing that all remarks of a certain kind are nonsensical or instances of language doing no actual work. Perhaps someone with a lot of experience in practicing philosophy this way would develop a nose for nonsense, but suspecting nonsense, with however much confidence, is not the same as either demonstrating it or curing those who want to speak it. If anyone is to be cured of such a thing, it might be beneficial to encourage them to say what they really want to say. This might provide sufficient rope for their own hanging. But if it does not, then perhaps what they are saying is not nonsense after all. The non-dogmatic and, hence, truly Wittgensteinian philosopher should allow for this possibility.

Anscombe says that she does not know how to distribute Wittgenstein’s preference for keeping the jagged jagged between philosophical observation and personal preference. Perhaps she is right not to know. Perhaps, that is, the personal cannot neatly be separated from the philosophical without violence. James C. Klagge has written of, and implicitly endorsed, the “belief that there is an intimate connection between Wittgenstein’s life and his work; this connection is crucial to understanding both” (Klagge 2001, p. x).

If this is true, then the personal cannot always be neatly separated from the philosophical. Wittgenstein’s personal beliefs may well have influenced what he said about religion. However, if he was true to his own pronouncements about how philosophy ought to be practiced, then he should not have made any controversial claims about religion in his philosophical work. Any that he did make, inadvertently, should have been withdrawn as soon as an objection was made.12 We should not take him, therefore, to be committed to any debatable views about religion qua philosopher. As an individual, of course, he may have had all sorts of peculiar or objectionable beliefs. Anscombe’s claim that he was sheer poison, then, should probably be taken to refer to some combination of his personal beliefs and the philosophical pronouncements of some of his followers. We might share these views, especially if we are not Catholic, but so far as they are debatable, they are not examples of philosophy undertaken as the later Wittgenstein intended.

4. Anscombe on Wittgenstein and the Philosophy of Religion

In an apparently unpublished book review, Anscombe says that Wittgenstein’s later thinking “has not been sufficiently digested for us to see its consequences” (Anscombe n.d., p. 3).13 One problem is that “Philosophy is pretty difficult, and so Wittgenstein’s is too” (Anscombe n.d., p. 3). A second is that he changed his mind, or was at least capable of doing so. At one point, she says, he thought that it was meaningless to say that someone believed something if this thing was itself meaningless, and he once said that he did not know what the doctrine of transubstantiation meant. So, if he was implying that the doctrine of transubstantiation is meaningless, there may have been a time when he would have said that it was meaningless to say of someone that they believed in transubstantiation. But in On Certainty, he writes that Catholics believe this doctrine. Perhaps, therefore, he had changed his mind about what it makes sense to say about others’ beliefs by that time (see Anscombe n.d., pp. 3–4). Even if Wittgenstein did not mean to reject transubstantiation as meaningless, and if Anscombe did not know exactly what to make of his remarks on the subject, it is still obviously true that he could have changed his thinking on any subject.14 It is also true, as noted in the first problem, that what he meant is not always clear.

A third problem is the lack of a written record of Wittgenstein’s beliefs on religion when Anscombe was writing.15 The lecture notes we have are from students, and Anscombe points out that each lecture would have lasted about two hours, but all we have from three of them is a total of around twenty pages (see Anscombe n.d., p. 1). Much has been left out.

There are some doctrines associated with the Swansea school (D. Z. Phillips and company) that she thinks might be “Wittgenstein’s fault”, but it is “difficult to judge”. (Anscombe n.d., p. 2)16 These include the idea that religious believers do not pray for actual things, such as rain or victory in a race, to happen, and the idea that they equally do not believe in the Last Judgment as an event that will really take place in the future. She, clearly, does believe it is such an event, and takes joy in the teaching she received from Fr. Richard Kehoe that it is all right to pray for anything that it is all right to want (see Anscombe n.d., p. 2; and Geach and Gormally 2008, p. xvi). This is further evidence that she had a different view of petitionary prayer than Phillips had.

What exactly her view was is not easy to say, because she did not write extensively or in general about the philosophy of religion. In the introduction to her Collected Papers Volume III, Anscombe says that her interest in the philosophy of religion has been in particular questions, such as the nature of faith or transubstantiation, rather than in more meta questions.17 She relates this to her claim there that her “interest in moral philosophy has been more in particular moral questions than in what is now called ‘meta-ethics’.” (Anscombe 1981a, p. viii). She then adds parenthetically that “The analogous thing is unrestrictedly true about philosophy of religion”. She gives two examples to show this point, one an essay on transubstantiation and the other “Faith” (in Anscombe 1981a, pp. 113–20).

This is an important essay for understanding her view of religion. In this paper, she describes a teaching that she says used to be heard about the rationality of the Catholic Christian faith. She first gives “the more extravagant form of this teaching” (Anscombe 1981a, p. 113). According to it, God’s existence could be proved, then the divinity of Jesus Christ, and then the establishment by him of a Church with a Pope at its head, whose job included teaching. Thus, one could prove that the teachings of the Church are true. One problem with this “extravagant” belief is that it leaves it unclear how grace could then be essential to faith in the teachings of the Church (Ephesians 2:8 says “by grace are ye saved through faith; and that not of yourselves: it is the gift of God.”).

A “more sober form” of the idea “wouldn’t identify the church by its having the Pope” and would, thus, commit the believer “to the Christian faith, rather than suggesting that he had as it were signed a blank cheque to be filled out by the Pope in no matter what sum” (Anscombe 1981a, p. 113). Then, an even more sober form would point out that the New Testament is dependent on the Old Testament (Anscombe 1981a, pp. 113–14). These sober variants would be more honest and truthful than the extravagant version, she says, but would have the disadvantage “that no one could suppose it quite easy for anyone to see that what Jesus established was matched by the Catholic Church that we know” (Anscombe 1981a, p. 114). There was also “a graver problem” regarding “the faith of the simple” (Anscombe 1981a, p. 114). Their faith was supposed to be just as good as that of the learned, and yet they could not do, or follow, the proofs that intellectuals produced and studied.

(Compare the faithful simple mentioned here with the “man, who does not follow the rather elaborate reasoning of the philosophers”, who “simply says ‘I know it is in any case a disgraceful thing to say that one had better commit this unjust action’.” (Anscombe 1981a, p. 42). These simple people and the things they simply say are very much the kind that Wittgenstein seems to have preferred.18 Anscombe need not have had Wittgenstein in mind when writing about them, however, and she surely did not have him exclusively in mind.)

Faith, Anscombe believes, is believing God (see Anscombe 1981a, p. 115). Not, that is, believing in God or that God exists, but believing what God has said. This, of course, involves believing that God is not a fiction and has spoken, but it is not the same thing as that. As she says, “the presuppositions of faith are not themselves part of the content of what in a narrow sense is believed by faith” (Anscombe 1981a, p. 117).

The belief of the “simple man and his teacher”, which I mentioned above, is “in no way inferior to that of a very learned and clever person who has faith”, she says (Anscombe 1981a, p. 116). No such clever person is needed to explain or show that “the evidence is that God has spoken” (Anscombe 1981a, p. 118). Rather, “The only possible use of a learned clever man is as a causa removens prohibens” (Anscombe 1981a, p. 118). That is, clever arguments might be needed to combat false thinking that gets in the way of faith. Perhaps Wittgensteinian destruction of houses of cards would be helpful, or something of the sort. Nothing constructive is, apparently, needed in Anscombe’s view.

Faith, then, is believing God. What this means is addressed in her paper “What Is it to Believe Someone?” (Anscombe 2008b). Here, she clarifies that believing God used to be called divine faith, to distinguish it from human faith, which is believing a human being (see Anscombe 2008b, p. 2). To speak of faith would usually be to speak of divine faith, understood in this way. Now, however, ‘faith’ has come to mean the same thing as religion or religious belief. This might sound like a simple change in the meaning of a word, which it is, but Anscombe sees it as being highly significant, as she explains thus:

This is a great pity. It has had a disgusting effect on thought about religion. The astounding idea that there should be such a thing as believing God has been lost sight of.

(Anscombe 2008b, p. 2. Emphasis in the original)

Immediately after these words, she quotes Genesis 15:5 in a translation that I do not recognize and might be her own.19 She has it as “Abraham believed God, and that counted as his justification”. The King James Version, for what it is worth, has “And he believed in the LORD; and he counted it to him for righteousness”. Omitting the word ‘in’ is important to her point about what this passage means, and she is not obviously wrong to omit it. It is absent in the New International Version, the English Standard Version, the Douay–Rheims Version, the Catholic Public Domain Version, and others. She credits Kierkegaard’s pseudonym Johannes de Silentio with having kept the story of Abraham well known “even to ignorant intellectuals”, but also complains that he does not confront “the first point of the story, that Abraham believed God” (Anscombe 2008b, p. 2). The ignorant intellectuals who follow Kierkegaard “like sheep” consequently deluge us “with rubbish about ‘believing in’ as opposed to ‘believing that’. Like the chorus of animals in Orwell, there is a claque chanting ‘believing in goo-ood, believing that ba-ad’.” (Anscombe 2008b, p. 2).

Her anger here is evident, and what it is directed at is close to the first three, especially the second, parts of the current that she identifies as possibly inspired by Wittgenstein. If we reject belief that in favor of belief in, then religion will not really be true or false and will be more like love or some similar attitude. One can believe in God, as one can love God, without, it might seem, believing that anything in particular (except, perhaps, that God exists).

However, Anscombe does care about attitudes as well as beliefs that. Faith in the sense of believing another is of crucial importance, according to her, “The greater part of our knowledge of reality rests upon the belief that we repose in things we have been taught and told” (Anscombe 2008b, p. 3). So, it is vital that we believe those who teach us, such as parents, teachers, and priests. Not that we ought to believe all such people, or each particular thing they tell us, but if children did not, in general, have faith in the people who raised them, then they would have a hard time developing any knowledge of reality. Furthermore, Anscombe goes on to say that what we get from testimony is not “entirely a detachable part, like the thick fringe of fat on a chunk of steak. It is more like the flecks and streaks of fat that are distributed through good meat; though there are lumps of pure fat as well” (Anscombe 2008b, p. 3). If we take this last observation to be not just about beef, but about knowledge from testimony, then Anscombe is implying that some of what we learn in that way could be detached from the rest of what we know, but most of it cannot be, or not at all easily.

This connects, I think, with what Wittgenstein (1975) says about Catholics in On Certainty, which Anscombe edited (along with G. H. von Wright) and translated (with Denis Paul). In §239, Wittgenstein contrasts the Moorean certainties that “every human being has two human parents” and that physical substances such as bread and wine do not change into other things, at least not without a change in appearance of some kind, with what Catholics believe, namely that Jesus had no human father and that “in certain circumstances a wafer completely changes its nature”, despite the fact that “all [empirical] evidence proves the contrary.” To be a Catholic is to have a significantly different view of reality from other people. This need not come from being raised as a Catholic, but, in most cases, it will be, and it will always involve relying on another person or book as an authority (see Anscombe 2008b, pp. 4, 6). This is an ethical matter, since “It is an insult and it may be an injury not to be believed” (Anscombe 2008b, p. 9). For a child not to believe their parent is, at least usually, disrespectful. The same goes for pupils not believing their teachers. Sometimes, of course, parents and teachers deserve disrespect, but not always, and surely God never does (either because God is so good or because God does not exist).

5. Conclusions

There is, or so at least Anscombe appears to believe, a kind of non-cognitivist Wittgensteinian or post-Wittgensteinian philosophy of religion, according to which religion is something like a loving attitude or a (hard to articulate) belief in (something, but not some thing, called) God. She does not simply reject this part of religion. She too loves and believes in God. But she emphasizes that she also believes God and the things God has said. She believes that these things are true. It might be very difficult to prove that such belief is rational, but it does not follow that it is simply a mistake to try to do so.

Her religious beliefs are clearly different from the beliefs of those such as Peter Winch and D. Z. Phillips of the Swansea school. Her beliefs also seem to be quite different from Wittgenstein’s, whether or not those were the same as Phillips’. However, if we take seriously Wittgenstein’s claims about what he is up to in his philosophical work, particularly in the Investigations, then we will not take his religious beliefs to be part of his philosophy. Reading Wittgenstein in this way should allow us to remove any poison there might be from his philosophy.20