I

Rationality in matters of belief is, as we know, a difficult concept to characterise. We manage to sort things out fairly well in restricted and mundane contexts without much by way of a deeper understanding. Outside these contexts, conceptual understanding falters. It falters just when it is needed. Anyone prepared to offer an account of the concept, it seems fair to suppose, will be offering something inadequate, at least if the account is not uselessly schematic: ‘a proper respect for truth’ let us say. Can we say, at a minimum, that rationality presupposes consistency? No, not even this, for the paradox of the Preface shows that a rational individual's beliefs ought to form an inconsistent set.Footnote ¹ That our understanding of rationality on belief is pretty limited is a thought which might occur to anyone with a background in philosophy, and it is only reinforced when one encounters disagreements among writers on justified belief which seem to become more entangled year on year.Footnote ² Perhaps we should not be expecting something too formulaic. Rationality is at least in part a matter of having good critical standards, and this is so often, in practice, an untidy business. We might think of particular safeguards: of being able to appreciate the fallibility of eye-witness reports, for example, or of being sensitive to the shiftiness of modal expressions.

An essential element in a rational strategy for belief must be to judge well which among an indefinite number of possible suggestions are to be taken seriously – as distinct from being dismissed out of hand. ‘Open-mindedness’ can only appear virtuous because of tacit restrictions on its scope. Only a mind that is closed in many ways is in a position to make enquiries. The question whether this involves a priori knowledge can only invite a yes-and-no answer: the acquiring of knowledge by empirical means presupposes knowledge already possessed. The pump must be primed. And this is so even if this acquisition of knowledge by empirical means is accidental. Pasteur's causal remark in his inaugural lecture, that in Matters of observation chance can only favour the prepared mind, is more deeply the case than perhaps be realized.Footnote ³

The filtering is mostly quite unpremeditated – so perhaps it was misleading of me to call it a strategy. So many possibilities never occur to us though they would instantly be excluded (as impossibilities) if they did. It is not that one shuts one's eyes in the face of doubt. They are already shut. Other possibilities will be borderline cases, more or less beyond the pale, and this is where judgment is called for. It will be a shifting borderline, and changes will be the cause anxiety and meet with accusations of prejudice. The distinction has a bearing on what we do. When for example one cries out for help in an emergency in case one might, just might, be heard, the range of what is to be taken seriously comes into play. And the same is true when one is invited to take precautions. Pascal's wager cannot sensibly be discussed if absolutely anything is up for consideration.

These judgments as to the borderline will inevitably be made from within the perspective of a certain group educated in a certain way, a fact of life which disconcertingly introduces a parochial quality which can (in practice, and quite possibly in principle too) only be worked around up to a point. ‘Think what happens when the missionaries convert the natives’ Wittgenstein once remarked, and not at all in the oddly censorious spirit in which such a remark would nowadays be made. ‘At the end of reasons comes persuasion’.Footnote ⁴ Even this filtering aspect of rationality in belief is not easy to describe well. But at least we may say this, that each of us must rely to a considerable extent in this matter of filtering on what is thought, and of course what has been thought, by others, by those who ought to know what is true in regard to this or that, insofar as anyone does. If the matter comes up at all, we think of their sensitivity, breadth of learning, character – especially intellectual character, robust good sense, resistance to local fashions, etc. – while making proper (that is to say considerable) allowance for the peculiar waywardness of the learned.

II

What then can be taken seriously? It is in connection with this question that the work of a philosopher like Elizabeth Anscombe has a peculiar interest. For she was not only that rare thing, a philosopher of the first rank, and someone with a wide competence in the field; she was also a philosopher with an abrasive and unsettling world-view. Her arrival in the community of analytic philosophers – I mean philosophers who are cautious, piece-meal, plain-speaking, logic-informed, science-respecting, and, one hopes, historically-aware – destroyed a certain consensus among them as to what could more or less be excluded from consideration. I am thinking of a time, perhaps a rather exciting time intellectually, when Cambridge, Vienna and Oxford were deservedly at the centre of the philosophical world. We should not attribute the initiation of this change entirely to Elizabeth Anscombe. I might have referred to the whole household of philosophers who were in and out of 27 St John's Street, c. 1950. But I suspect her influence was central. She was a walking sed contra. Her scepticism, every bit as lively as that of other philosophers, was directed at different and perhaps more appropriate targets.

Hers was a dissent which disturbed the peace not only intellectually but up to a point socially. She did not fit in. Philippa Foot, who came to know her so well, reports how even she had at first been alarmed by her. Anscombe, she said, was much feared by her colleagues and often avoided: ‘the last thing they wanted was Elizabeth around’. People thought they were going to be challenged.Footnote ⁵ Isaiah Berlin, not someone who would easily have been intimidated, once tried to avoid having to meet her, claiming to find her unsettling and prone to ask awkward questions.Footnote ⁶ Anscombe's dissent could not simply be dismissed as a personal quirk, of a piece with the superficial eccentricities of her clothing (of not caring too much what she wore) and of her smoking habits. In fact she was dangerously sane. Nor could her dissent be hygienically detached from her philosophical thinking. It was bound up with it. In this respect it was quite unlike, let us say, the somewhat surprising devotion to the occult which we find in Yeats. Above all, no one could dismiss Anscombe as superficial. ‘Oxford philosophy tended to be clever,’ said Bernard Williams, ‘but Elizabeth conveyed a strong sense of the seriousness of the subject, and how the subject was difficult in ways that simply being clever wasn't going to get round’.Footnote ⁷ All this certainly made philosophy more interesting – especially at a time when many people were inclined to suppose that philosophy could have in the end nothing of interest to say, interpreting or misinterpreting Wittgenstein's remark that it ‘leaves everything as it is’.

A consideration of Elizabeth Anscombe's view of the world and its connection with her work in philosophy has now been made easier by the publication of Faith in a Hard Ground, a posthumous collection of her articles, all admirably ordered here by the editors.Footnote ⁸ One or two of them are well-known from her Collected Philosophical Papers,Footnote ⁹ but many have been pretty inaccessible and some are published here for the first time. Though the book contains many (suitably upsetting) articles in moral philosophy my attention here is confined to the author's dissenting faith which is singled out in the title, and to her concept of faith. For some the book will outrage and intrigue in equal measure. I imagine that an angry reader might tear up his copy. But he might also immediately leave the house to buy another – Anscombe's writing so often calls for a second look. Her thought, writes Jonathan Dancy, is ‘so idiosyncratic and so personal, she does things in quite her own way, she does not give the reader any help at all and she doesn't try to put things into any kind of order’.Footnote ¹⁰ I feel diffident in presuming to expound her opinions, seeking to introduce what order I can. Michael Dummett, remembering the St John's Street Anscombe from his undergraduate days, tells us that ‘she was always especially critical when I wrote things with which I expected her to agree.’Footnote ¹¹ I can only get away with it because she is no longer here to read the result.

She was of course a Catholic Christian. (The doubled-up description is hers: 11, 57, 66, 201, 206, see also xii.) This stance certainly marked her off not only from the orthodox atheistic mainstream, that is to say, from people who suppose somewhat rashly that Christianity could not be a serious option,Footnote ¹² but also from those who would willingly reach for religious-sounding language if this could be done without the risk of saying anything too definite with it. Her Christian beliefs, acquired in her teensFootnote ¹³ when she was already intellectually pretty sharp, no doubt enabled her to avoid being deprived of independence of mind by her close association with Wittgenstein. It also helped her to escape the religiosity of so many of those who came under his influence, insofar as help was needed.

It will readily be suggested that an analytic thinker as capable as Elizabeth Anscombe could only hang onto her worldview by ‘compartmentalising’ (see xiii). Ayer had similar thoughts as to the Christian convictions of Michael Dummett, his successor as Wyckham Professor at Oxford, and a philosopher he much admired. Surely he must somehow divorce his religion from his philosophy?Footnote ¹⁴ But we might wonder what this ‘compartmentalising’ or ‘divorcing’ is thought to amount to. Dummett himself writes: A man's philosophy ought not to be controlled by his religious beliefs; philosophy ought always to be free thought. If the results of someone's philosophising appear to be coming into conflict with what he otherwise firmly believes, he ought to conclude that they cannot be correct, though he is unable to see at present where or how they have gone wrong.Footnote ¹⁵ Would that count? Would the drawing of such a conclusion represent a failure in rationality, or might it be rationally required? We will touch on this question below.

If Anscombe's dissenting world-view proved disturbing to others, the content of it was to her the cause of permanent astonishment (155). Her astonishment was not I suspect at the mere thought of the existence of a creator of the world – a possibility taken very seriously by Mill in his essay Theism. I do not say that the matter is trivial. But people might take the thought of a First Cause, considered simply on its own, rather in their stride – almost in the way a sensible person will regard the initially surprising but ultimately boring statistics of the size or age of the universe, statistics so memorably put in their place by F. P. Ramsey.Footnote ¹⁶ What's so special about being a first rather than a second cause? ‘It is possible in some sense to believe in deity’, Anscombe wrote, ‘and not to believe that deity knows or cares what we do’ (147). What was truly amazing to her was the whole Christian story, involving such divine knowing and caring – I mean of course the remarkable and real possibility that this story might be true. Our sensitivity to surprise in this regard is surely blunted by centuries of everyday acceptance by people around us: Anscombe herself felt the need to heighten her awareness of the wonder of it all (155). It is a story of many historical contingencies. One need only think of what happened one sunny morning in Palestine not so very many centuries ago, when a young woman living in obscurity happened to answer with a Yes. (It could so easily have been a No, as Anscombe remarks, 155–6.) It is the Christianity of this or that particular incident, in the setting of other contingencies both forwards and backwards in history, which should really make us catch our breath. That, and the way in which such a story must in part be known: the possibility, which Anscombe called ‘astounding,’ that there could be such a thing as taking God at his word (2).

The astonishment of the academic bystander that a philosopher of the first rankFootnote ¹⁷ can have had such convictions might be put down to two prejudices concerning history. These prejudices serve to limit the scope of what suggestions we are prepared to take seriously. The first involves a patronising attitude to what comes earlier in human thought. ‘Poor dears, they could hardly have known any better’. ‘Surely we should have grown out of that by now?’. One thinks perhaps of Sir James Frazer and his story of the long ascent of man from Magic to Science. It is as if Moses or Methuselah could hardly have had anything of importance to say to us as they were so woefully ignorant of quarks and statistics. A philosopher ought to be somewhat protected from this particular snare – more especially if closely involved from time to time with ancient philosophy in a way that is not simply antiquarian. The second prejudice is this, that history can only be ‘one damn thing after another’, with nothing of the deepest significance for mankind happening at any particular moment – such as the contingency of that young woman's answer. I am not of course thinking here of mere technical fixes, like the invention of the wheel.

III

One thing stands out in this collection. Anscombe was never much concerned to make her worldview attractive – if by that we mean reassuring rather than impressive.Footnote ¹⁸ Reassurance in a teaching, I expect she thought, might be the very thing to make one suspect a cheat. Most disconcerting of all would be the suggestion that God might actually go so far as to punish us (118).Footnote ¹⁹ The distaste among contemporary Christians for the thought that we live under threat is indeed hard to miss. People will call such a God ‘punitive’ and hence impossible. That label in place, they will feel entitled to think no more about it. Peace of mind is what it is all about. A gospel-reading on the separation of the sheep from the goats will nowadays be followed up by a homily on the importance of preserving the diversity of species.Footnote ²⁰

Those who are hostile to religion like to say that religious belief is wishful thinking.Footnote ²¹ They hardly pause to discriminate one such belief from another. Religions evidently differ widely, but in this respect, it is supposed, they are at one. An accusation of wishful thinking assumes that the belief is both agreeable and irrational (many agreeable beliefs are of course rational on any account of rationality). Such an accusation in regard to religion runs into difficulties in regard to the first condition, at least in regard to an un-modern, un-prettified Christianity. The demands made are demands, and are often quite specific and severe. Christianity is not just ‘something which helps one get through the day’; those who have died for their belief never get through the day at all. On a more everyday level one only has to think of what might be taught as to the indissolubility of a valid marriage, and how one could be caught out, and how devastating that might be. Thoughts of wishfulness are particularly implausible where someone changes religious allegiance from something less to something more demanding, as quite often happens. Or are we going to say that people only become Christians because they cannot bear not to be living under threat? Demandingness aside, it is not in the least surprising that a thoughtful man can very much want there not to be a God, Thomas Nagel affording a well-known example.Footnote ²² Others might wish not to know, preferring the ever extended uncertainty of the quest. But why need there be any special motive behind our beliefs, whether religious or other, whether threatening or agreeable? So often we are simply told that p and we believe that p – as children in particular, but also in later life.

In these essays Anscombe continually draws our attention to what we might call the bad news of the Gospel. I mean: bad by contrast with what increasingly appears to be the general expectation. Anyone who claims to be disturbed by ‘negativity’ will, I am glad to say, find plenty of it here. ‘To read only, say, the Gospel according to St Mark and to pay close attention to all that our Lord is recorded as saying, is frightening’ (152). Rather generally her Christian outlook is a forbidding one, and in more than one sense. At least it is likely to seem forbidding to fallen man (61), to members of ‘a damaged race with darkened minds’ (155).Footnote ²³ Anscombe writes with particular scorn of ‘religious confectionary’ (63) ‘whose promotion is proper to a salesman’ (64). This indeed is something of a recurrent theme. Two distinguished Jesuits, Arthur Vermeersch and Bruno Schuller, are taken to task, the first for adopting a ‘world-pleasing role’, the second for attempting ‘to anesthetize his readers against the effect of knowing the ten commandments’ (166). Both are accused of using their talents ‘to go along with the world, to reassure it and flatter it’ (169). Needless to say, there is not the least hint of world-pleasing in Anscombe's essays. Julian Barnes began his recent meditation on death, Nothing to Be Frightened Of, by saying: ‘I do not believe in God, but I miss him’. On being invited to comment, the philosopher Jonathan Barnes (the author's brother), dismissed the remark as ‘soppy’. Soppy is what Anscombe ain't.

We tend to forget when we hear consoling prayers recited at gravesides that Christianity has always taught that we live dangerously, indeed in quite remarkable danger. There is indeed something to be frightened of. ‘All shall be well, all manner of things will be well,’ said Julian of Norwich, her words being everywhere repeated. This, it would seem, is what everyone longs to be told. It is the hopium of the people. But well for whom? Not perhaps for all. Not perhaps for many. Julian's remark reflects Isaiah, 3.10, but with a crucial difference: ‘Say ye to the righteous that it shall be well with him’. That might suggest a miniscule number. And even so, the righteous might reach salvation only with difficulty, as we are told in 1 Peter, 4.18: the thought being expressed in passing as if too obvious to require further comment. Christians learn to say, almost by rote ‘Thy will be done’. But as people rightly say, we have to be careful what we ask for. Bernard Williams once wrote of ‘this terrible thing, religion’. There is a sense in which the description is apt, at least in regard to the religion he was probably thinking of.Footnote ²⁴ Wittgenstein by contrast took the thought of being Absolutely Safe as a paradigm of religious intimation (59). Surely this was not one of his deeper insights. One is reminded of the blissful epicure: ‘Fate cannot harm me, I have dined today’. In any case it goes quite contrary to what Wittgenstein was prepared to say elsewhere, disagreeing with the eschatological cheerfulness we find in Origen.Footnote ²⁵

At this point, however, someone will surely be reaching for a way out: ‘Isn't God supposed to love us, to love us greatly, and love us in a way we have done nothing to earn?’ Well of course. The account of our redemption, supposing it to be true, shows us that. It is a story beyond wonder. Moreover this love is universal, that is to say it is a love for every human being. Yet from here on, the confectionary salesmen have tended to take over. They will talk of course of unconditional love. But universal love need not mean unconditional love, and the Christian story shows it not to be unconditional. Anscombe in particular thought that God's continued patience with mankind was not to be presumed upon (Papers, Vol. 3. 71). The idea that we are loved unconditionally is one we naturally like to hear, but it rather calls for the Jonathan Barnes treatment.Footnote ²⁶ And if God's love is not unconditional it is not impartial either. The Bible after all tells us continually of a chosen people. And there is no suggestion that they were chosen on the basis of merit or especial need. But isn't impartiality required by God's justice? Not at least in this way. Choice which is arbitrary in relation to merit or need is so often compatible with justice, even for us humans. Those who expect impartial benevolence are confusing God with Godwin.

What we find rather plainly expressed on the topic of God's love in Aquinas may usefully serve to shake our assumptions. And the same indeed might be said for Aquinas on God's goodness, a related topic. I am not of course arguing that Aquinas must be right, but only that he may be so, and that for modern readers he opens up the possibilities. Aquinas denies for example that God loves human beings more than anything else he has created.Footnote ²⁷ This promises to overturn a common conceit. When we think of the scope of the divine love we humans keep quite lowly company. God's love, Aquinas claims, is so universal that it extends to items like a pebble, since a good is continually done to a pebble by the mere fact of being sustained in existence.Footnote ²⁸ We should also think of what has to be compatible with God's love. God's world is a world in which everyday a tower of Siloam falls, causing random death, injury, and misery. This is part of the story. Indeed, Anscombe says, and quite without fuss, that God is the cause of countless evils. ‘If perishing is an evil for what perishes’ – which is argued for here – ‘then the creator of the world cannot but be a creator of constantly repeated evils, or lacks of the excellence of lasting’. And the same is true of defects and illnesses. Even a canker on a tree is an evil for the tree, and there are so many such cankers. And there is no suggestion here that a canker on a tree is only there because without it some greater good to something else could not have come about. All this has to be faced, and Anscombe was not going to allow Aquinas to wriggle out of it (140–1). She must however have known that people in the present day would not be in the least willing to regard the mere perishing of perishable things as evils – death of a mayfly, or even the collapse of a soap bubble, both of which conspicuously lack the excellence of lasting. This is perhaps because an evil is thought of as a bad thing to happen simpliciter, rather than a bad thing to happen to something or other. And perhaps because ‘evil’ has come to have creepy connotations, and hence something safely unconnected with the lives of ordinary folk. She must have known too, and enjoyed the thought, that apoplexy would be caused by her failure to mention pain. Isn't that what the problem of evil is all about? Fawns dying in forest fires, and all that? I think she would reply that the occurrence of a pain is not an evil. It need not represent a ‘failure of an organism to work properly’ (her phrase, 141) – in normal circumstances it would be an instance of proper working, and hence no evil at all. An evil in this regard would be a pathological inability to feel pain – in one's finger ends brought about by leprosy let us say (as Paul Brand so memorably discovered). People will talk of a compassionate God. But it is not at all evident that a being incapable of passion would be capable of compassion. All in all, for all the talk of God's love, we are left with a bleak picture. It is so very far from anything envisioned by Leibniz.

Let us return to the thought that we walk this world in danger, for risk is itself one of these evils. This danger tends to be concealed from us by a false teaching, popular even (or should I say particularly?) in Catholic circles: that it is sufficient to act in accordance with one's conscience. It goes along with the view (or pretence) that everyone's opinions are to be respected. I even came across the resonant phrase ‘respect for conscience and the dignity of each person's convictions’ – indicating that this was something we have at long last come to appreciate.Footnote ²⁹ That this thought should have been taken up is not be wondered at. It is religiously important. Respecting all religions is a prophylactic against the possibility of having to respect just one.

This teaching as to the sufficiency of conscientiousness Elizabeth Anscombe was continually concerned to undermine. ‘Even among Catholic Christians there is often a bad state of confidence that one is all right – because one's conscience is clear’ (153).Footnote ³⁰ In Anscombe's view it was all too easy to act inexcusably in doing what one's conscience permits. She quotes 1st Corinthians, 4.4: ‘True, my conscience does not reproach me at all, but that does not show that I am acquitted’ (116). In the Summa Theologiae there are two adjacent Articles about erroneous conscience; they form a short clear non-technical bit of the work, of obvious importance to any reader interested in the effect of error on acting well (I-IIae, Q. 19, A. 5 and 6). Curiously however, as Anscombe points out, the second of these Articles is ‘less well known’ (107). It is worth our asking how this could be, for they are obviously meant to be read together. One difference is plain. The message of the first Article is popular, of the second unpopular; for the second tells us that an erroneous conscience does not, or does not always, excuse – and excuse is what is everywhere wanted. The story is regularly put about that we cannot act really badly unwittingly, so that if we think we are in the clear we are in the clear, peccadillo apart. However, we can be blamed for unwittingness, and the fault of not having bothered to find out is not always venial (115). This helps to explain why so few of us are prepared to believe that we might be in a very bad state (153). We tend wrongly to assume that when we act badly we must be burdened by pricks of anxiety (155). Not that such anxiety would save us, for the ‘endless bellyaching’ of the conscientious does not guarantee the requisite seriousness (97). We forget that even actions done out of a sense of duty can be profoundly and culpably evil (230). All this might seem evident, almost commonplace, but it seems insufficiently appreciated. Thus it is often stated in obituaries that the deceased had been ‘deeply committed to his sense of moral values’ (or some such). This could only very contingently count as a compliment. Again, it will be said, in an irenic spirit, that there is room for honest disagreement in regard to some controversy in ethics. An innocent enough remark one might think, but I suspect it will sometimes represent a covert appeal to the falsely indulgent doctrine of conscience under discussion.

The status of good faith is not as easily won as we tend to suppose, at least if we are using ‘good faith’ to mean something which exonerates. There is certainly no refuge in being (what is called) normal. To be normal is to be defective. People recognise this when they talk, with Kant, of the crooked timber of humanity. We need to have doubts about ourselves: ‘rather generally, it must be good for anyone to think “Perhaps in some way I can't see, I may be on a bad path, perhaps I am hopelessly wrong in some essential way”’(Papers, Vol. 3, 37). We can be blamed for not taking the trouble to see, or for getting ourselves into a position of not being able to see. We certainly cannot consider ourselves in good faith merely on account of being ‘caring’ (116). It would not be safe to live a life admired and respected by one's community, as a well-regarded doctor say (153). Nor I suspect would it be sufficient to be a paid-up ‘member of the human rights community.’ Bernard Williams writing about ethics feels free to talk in rather general terms about people who count as ‘horrible’.Footnote ³¹ He says nothing however about how they are to be identified. No doubt we are expected to think ‘Well, we all know who they are.’ They tend to live in far away places, or so we suppose. Anscombe's reflections on good faith rather suggests that our knowledge in this respect is not far reaching.

The teaching we need to live by will quite often be contrary to the spirit of the age (191). Being a Christian, Anscombe argues, means separation. This of course has nothing to do with a stand-offish attitude to one's fellow men, still less with pride. The Jews were instructed to welcome the stranger ‘for you were once strangers in the land of Egypt’. Christians were commissioned to teach people of every nation – the task could not be more inclusive. And all are made in the image of God.

Separation in the present context means separation ‘from what the unchristian world in the West approves and professes’(171, also 116, 213). And also indeed from much of what the apparently Christian world professes (61–2). The necessity – surely ever more obvious with each decade – of separation is charted across many pages in this book (see e.g. 206–7, 238). Standing apart from what the unchristian world approves and professes would not be a sufficient indicator of truth in these matters as anyone can see; it would merely be something needed. A Christian is not to be passively co-opted by what comes to be taken for desency and moderation. He is called to be a misfit; to disturb the consensus and to be ridiculed and hated for it. This is even more evident now than when Anscombe wrote. A liberal society is one in which people can be prosecuted for calmly expressing quite ordinary Christian teachings.

This emphasis on separation distances her from Wittgenstein. One thinks particularly of Wittgenstein's oddly undiscriminating remark to Drury that all religions were wonderfulFootnote ³² – though it is possible that Wittgenstein, irritated by being classed as a positivist, was here simply cocking a snook at the positivist assumptions of his time. Wittgenstein also said, rather comprehensively, that if someone came to him with religious doubts he would very likely find himself raising doubts about the doubts (58). For Anscombe, at least, the question would be: doubts in relation to which religion? So often, the appropriate response would be the encouragement of doubt, religions being so very various. She saw in some of Wittgenstein's followers a certain inclusiveness characteristic of paganism: the view that all peoples have their gods about whom it is contemptible to be scornful – though to be sure one might naturally favour some over others, in the way one might prefer to follow the local traditions one has been brought up with or support one's local football team. The modern notion of ‘culture’ (as in the coinage ‘multiculturalism’) should strike us as a distinctly pagan notion in this sense. Religion is treated as a matter of local colour. Which of these many gods one followed could hardly count for much, since what really mattered was something else: a certain presence in a man's heart, ‘something rich and significant in the depths of the self’. Anscombe would simply have asked us how such a thing could matter (60). What could it profit a man to gaze into his soul and lose the whole world?

People are apt to be impressed by an ersatz which they call spirituality. Anscombe herself talked about spirituality; but for her, this was not an attitude of mind we could adopt, or a path we could follow. It was a capacity, a remarkable capacity – and one we all possess, and not therefore like musical ability, uncannily manifested by some and not by others.Footnote ³³ The spirituality of man was a capacity to become engaged with what is ‘eternal’, as something to be achieved perhaps, or as something to be called upon, or again feared. Aristotle for example invites us, in Nicomachean Ethics 1177b, to ‘think with the immortals’ insofar as we mortals can (149). It would not be necessary to have a belief in God in order to exercise this capacity; a belief in Nirvana might be sufficient (74). In The Virtues (Cambridge University Press, 1977, 72–3) Peter Geach relates how an early encounter with the philosophy of McTaggart served to keep alight ‘a longing for the infinite and the eternal that was not to be quenched by the noisy winds of the world in the storms of youth’. McTaggart was of course an atheist.

Man, Elizabeth Anscombe said, is both flesh and spirit.Footnote ³⁴ We might interpret this to mean it belongs to man – uncartesianly undivided – both to weigh a certain amount and to weigh things up. Yet that does not quite capture the way things are, for spirituality in Anscombe's view was something even narrower than rational capacity: it went beyond the capacity to use language, not being guaranteed by this last. She seems to have thought of it as a latent power of insight (75). The obliqueness of her language here is perhaps inevitable; sometimes one must not begrudge a pinch of salt. I do not expect, however, that she would have seen nothing but good in all the very various ways such a capacity might be exercised. For Anscombe, the spiritual nature of human beings, who are yet animals, is made plain when we realise that we can share a defect with angels (64). Bad or inadequate exercise would indeed be the rational expectation. Our spiritual capacity is there for a purpose, rarely achieved: it is to enable us to live in God's presence – as Brother Lawrence tried so constantly to do as he worked year on year in his monastery kitchen (159).Footnote ³⁵

Christian understanding of the importance of separation, she argued, is something which is owed to the precious and quite indispensible Jewish inheritance. The Jews were unpopular among pagans, and Anscombe quotes extensively and amusingly from the disdainful things Tacitus had to say about them. They were held to be credulous, and not because they had religious beliefs – pretty well everyone had something of the kind – but because they were serious about what they believed (56). They were unpopular because they stood out: religiously they would not play along and be more or less like everyone else, adopting an easy-going respect for the local colour of difference. ‘To the elder Pliny, they were known as a people distinguished by their insults to divinities.’ And this was no misrepresentation. Such divinities were ‘abominations.’ The Prophet Ezekiel boldly referred to the gods of the Gentiles as shit or turds – though the word, Anscombe says, is ‘usually translated more prettily as “filth”’(53). These days, all this would be thought to represent a failure in charity. But on the contrary, it is a manifestation of charity. It is not a licence to be pointlessly hurtful.

To be sure, a Jew or a Christian with this inheritance would not have to suppose that a being worshipped outside the Jewish or Christian tradition could not be the God of this tradition. Anscombe saw the true God in the God of Xenophanes, Anaxagoras and Aristotle. ‘The thing that is remarkable about these philosophers is not to have had a false picture of deity’ (149).Footnote ³⁶ It is striking with what respect, affection almost, she writes about Anaxagoras in particular, mentioning the annual holiday for the children of Lampsacus which was set up in his memory in accordance with his dying request (56). Pascal's much-adopted contrast between the God of the philosophers and the God of Abraham was not her contrast. ‘Xenophanes, Anaxagoras and Aristotle believed in the existence of that deity whose name is “I am”’ (147). There is thus no suggestion here that people outside this Jewish or Christian tradition walk in total darkness and can have nothing to contribute. She tells us in particular how in Aristotle's view ‘the goods of fortune are worth nothing except so far as they help us to worship and contemplate God’: surely her own view of the matter too, and something of central importance.Footnote ³⁷

Christians who choose to live ‘in the world’ – who hope to marry and have property, let us say – are described here as a pretty suspect sort of Christian. It was doubtful in Anscombe's view whether they (she or they) could be sufficiently detached from the things of the world (64), friendship with which is enmity to God (191). It would not be enough to respond to the ordinary demands of self-discipline that any reasonable human being can see the necessity of (65). Even those somewhat removed from ‘the world’ could hardly count themselves as safe. Anscombe supposed that bishops in particular must live in exceptional insecurity, being tempted to betray those in their charge, offering them perhaps ‘a degenerate form of religiosity’ (208). However, none of this presupposes that God is a particularly demanding God who for inscrutable reasons of his own would set peculiarly high standards for humanity to follow. The standards, I imagine, would be suitably low and would for the most part be set (if that is the right word) by the circumstances of human life. At one place Anscombe says with emphasis that all anyone has to do is to keep the commandments (‘in faith and in hope’ 63).Footnote ³⁸ This would represent a tough rather than a fancy standard. It is just that one would be most unlikely to keep to this minimum without a degree of this detachment she talked about.

All in all then, discomfort is in order. A failure to recognise this fact shows up a basic unsoundness in belief: ‘false worshippers of God are easily known by the following sign: they condemn the fear of God’ (61–2). This invites a striking contrast. Bertrand Russell was prepared to argue that people cling to their religious beliefs out of fear, the fear of ceasing to be. ‘Fear is the basis of religious dogma’ he said. ‘Immortality removes the terror from death’.Footnote ³⁹ This diagnosis must of course have been fairly popular among philosophers of the time. A propensity to religion would have been seen as a weakness, an unworthy refuge.Footnote ⁴⁰ One thinks here too of Philip Larkin's suggestion in his many of poem Aubade that religion was ‘created to pretend we never die’. Pleasing to bleak late his readers no doubt, but not I think to his thoughtful readers. Atheists might do better to say that people only become Christians out of boredom. For certainly, the world through Christian eyes is so very much more interesting. Every minute, even for the plainest person on the greyest of days, is then momentous.Footnote ⁴¹

The idea that anxiety about death, put to rest by teachings about immortality, is somehow the basis of religious belief cannot be based on any serious research or reflection. In the history of the Jews thoughts about an afterlife are a late development,Footnote ⁴² and their beliefs represent almost a paradigm of religious belief. And of course the religious beliefs that Russell will have had in mind are an inheritance from it. Russell recognised of course that religion does not entirely succeed in removing the anxiety. What it cannot remove it dignifies. Religion he complained ‘has dignified certain kinds of fear, and made people think them not disgraceful. In this it has done mankind a great disservice …’. Russell spoke of this fear with melodramatic disdain. ‘I believe that when I die I shall rot … but I should scorn to shiver with terror at the thought of annihilation’.Footnote ⁴³ But why terror rather than simple regret? There is certainly reason for this last; an afterlife might provide our only opportunity to show the gratitude we owe to those we have lost touch with, and for expressing our sorrow to those we have so grievously hurt.

But let us keep with terror. It is important to remember that there is a fate worse than annihilation. Atheist advocates of mercy killing remind us of this over and over. And what would be the worst fate of all? It would be the perpetual loss, perpetually perceived, of everything which truly matters to us, from love and joy and security to the most basic of creature comforts. Christianity of the un-prettified kind need not be joyless and gloomy, short of laughter and of good red wine. Nevertheless it teaches that we live at the risk of such a loss indeed, the suggestion seems to be that more will suffer than avoid it.Footnote ⁴⁴ Christianity teaches that there are four ‘last things’ we need to remember, to meditate upon. Counting as far as four seems beyond the powers of the contemporary religiose. In the light of all this, we must correct Russell's remark: immortality adds terror to death. To fear disaster would not be a sign of weakness. It would be the most elementary mark of the very rationality which Russell so rightly claimed to prize. Fear in Christianity is not what is called fear of the unknown. It is fear of the known.Footnote ⁴⁵

IV

If a certain religious world view is not attractive in the sense of reassuring (and why should we expect it to be?) we must still ask the more important question: whether it is attractive to reason – that is to say, attractive to someone who has good critical standards and a determination to know what it is both possible and necessary to know. The intelligence, learning and character of a philosopher like Elizabeth Anscombe may expand the range of what is up for serious consideration; but then this consideration must take place. This of course represents the point of view of an outside observer. An insider will talk instead of what is defensible. And here we need to note how many difficulties Anscombe was apparently prepared to make for those who might wish to defend her own religious outlook. It might be thought, to give just one example, that Christianity requires a notion of immaterial substance if only to characterise God and the human soul. Anscombe rather surprisingly regarded such a notion as ‘delusive’. She argued that our category of substance is one which is at home in the material world so that we do not know where we are with it when we try to apply it outside the realm of chalk and cheese, to angels let us say – though in so saying we must give up the assumption, which comes so naturally to us, that what things count as substances exhaustively constitute what is actual (71).Footnote ⁴⁶

We should not be surprised that Anscombe makes things difficult. Philosophy is not apologetics. And our discipline as we know is rather good at making things difficult. Indeed there just are difficulties for any view of the world. I allowed myself to talk above about the ‘orthodox atheistic mainstream’ in modern philosophy, as if what is called scientific naturalism – seen here as distinct from that respect for scientific carefulness and achievement which we all must share – represented a default position, a robust refuge against paradox. But hardly so. Goodman's problem of projection, threatening to undermine our expectations about emeralds, has hardly been answered. It would be no good positing the uniformity of nature, even if one was prepared to ‘posit’. For the question is in what way uniform?Footnote ⁴⁷ And this is not alt. But Anscombe, in a sympathetic discussion of Quine, argues that if we adopt his naturalistic outlook no sense can be made of our use of intentional verbs, when we recognize for example that a bit of language ‘says that’ something or other. Such verbs which are basic to human life and thought. She rather suggests that if we were to follow Quine we would be left a-wandering in a desert (262–4). In recent years ‘eliminative materialists’, with whom Quine is sometimes associated, have detailed the remarkable costs of adhering to eliminative materialism. No one it seems is prepared to pay up.

Christianity naturally has its own problems. Anscombe in effect and often in intention brings them to our notice. The article ‘Faith,’ reprinted here, might seem more or less totally to undermine the argumentative underpinnings of her Christian world view. For many years Christians, Catholics in particular, had been taught that their religion was backed by a structure of argumentative considerations, available at least in outline to all, and which safely delivered the correct outcome. It was a programme which went back at least as far as the eighteenth century; Lessing for example was clearly concerned to attack its historical aspects. Anscombe was remarkably hostile to this whole approach. She was pleased to be able to say that it was now ‘out of fashion’, although, as she pointedly added, this was not because better thoughts about faith were now commonly to be found. ‘There is a vacuum where these ideas were once prominent’. She regarded this whole structure of apologetic argument as something ‘fanciful’, based on prejudices, ‘prejudices … about what it is to be reasonable in holding a belief’ (13).

To be sure, her unorthodoxy was more philosophical than doctrinal. In regard to the content of Christianity she had high praise for ‘that excellent thing’, the so-called Penny Catechism, as once taught to children by ‘un-reconstructed’ Catholic teachers (114).Footnote ⁴⁸ Anscombe, we might say, would have believed as a matter of course what Catholic 8 year olds were being taught. But as to what the 16 year olds were being taught, well, that would be another story; philosophy of an unsatisfactory kind would by then have put in an appearance. She had no time, for example, for the thought, common in sophisticated Catholic circles, that there was something called a natural law argument (that is to say a distinctive bit of theory) which could be deployed to show that an activity – it might be contraceptive intercourse, say – was wrong.Footnote ⁴⁹ And in her article on transubstantiation she presented not what many people might have expected from a philosopher, that is to say yet more theory, but rather what might be called a child's eye view of the matter, what the smallest child would learn from the simplest of teachings. ‘What is grasped by the simple’, she said, ‘is not better understood by the learned’ (91). It might take a certain sophistication to see that this os. I think of a remark attributed to Pasteur that the more he knew the more his faith resembled that of a Breton peasant. Anscombe was pretty cool about the traditional philosophical attempt to explain what was going on: the comical suggestion that one parcel of Lockian inscrutable we-know-not-what, called ‘substance,’ was being replaced on the instant by another sort of the inscrutable same, a change we would hardly expect to notice (86–7).Footnote ⁵⁰

The traditional apologetic structure of argument started with what were called ‘preambles to faith’, the groundwork established by (what was thought of as) reason, something foundational and pretty generally available. Anscombe wanted to talk here of ‘pre-suppositions’ rather than ‘preambles’. This difference would come out as follows. It is a commonplace fact that we can believe someone. It might be one's doctor say, or the writer of a letter which has just arrived in the post. It is a propositional matter: one believes one's informant that such and such is the case. Now if one believes someone, when reading a letter say, there are various things which one presupposes – that the ostensible writer exists for example, and that the letter is from him (13). However something presupposed is not something established as a first step, as the traditional apologetic supposed. It might not be thought of at all. Indeed it need not even be a temporally prior belief (17), and might even be occasioned by the arrival of the letter.

It is evident that although Anscombe found the need to be destructive in this article, her main task was to describe what faith is, rather than to explain how it could often be rational. Someone who has faith, in this usage, is not someone who harbours a groundless belief – whether inappropriately, or, as is quite often the case, appropriately. To have faith is to ‘believe someone’ in the way just mentioned, and it would be misleading to call such a belief as lacking a ground, unless one had reason to regard the testimony in question as worthless. But it might be natural to talk of a lack of an independent ground. Taken in this sense, there would be no interesting contrast between faith and reason, for it is often, very often, entirely reasonable to believe what one is told, indeed without a second thought. It is an essential aspect of human life.

Someone would be said to have divine faith who believes God that such and such. Believing God is, as Anscombe naturally acknowledges, a notion which is far more problematic, more problematic in sense, than the everyday notion of believing a human informant. Locke, defining faith, talks of (something extraordinary) ‘some extraordinary way of communication’ (Essay, Bk. 4, Ch. XVII, 2), and that in a way must be right. In some way the message from God is thought to get through: it might indeed be via human teachers (often quite unlearned). But – as in the case of St. Augustine – it might even be a consequence of overhearing a remark, addressed to someone else about something else, and one's hearing it in a way to suit one's condition (18). I once heard a toddler, rather struggling with the idiom, exclaim to her father who was no doubt larking about: ‘What are you doing on earth!’. That could have been just such a moment.

By extraordinary I do not mean rare. Indeed, being in a position to believe God must be universal among human beings with the use of reason, at least at some time in their lives, if Peter Geach is right to say both that divine faith on their part is conceptually necessary (The Virtues, 20–22), and that a chance of salvation turning on the rightful choice is offered to everyone of ordinary human capacities who has passed what is called the age of reason. This last is constantly emphasized in his writings.Footnote ⁵¹

It is true of course that ‘faith’ in religious contexts is nowadays often used in a different way. Anscombe took strong exception to the practice of using the word simply to mean a religion, as in the modern coinage: ‘faith schools’. For in her view this usage had had a very bad effect (‘disgusting’ was her word!) on thought about these matters. We have tended, she argued, to lose sight of the particular and important thing, found within certain religions only (2), this notion of believing God. ‘A’ faith would not be any old religion but a religion that involved such believings, whether reasonably or not. We might also lose sight of something else: of the pervasiveness and necessity of mundane human faith, the faith which rests on merely human informants, and which has nothing in particular to do with religion. There is an important sense in which every school, good or bad, religious or atheistic, is a faith school, in which the teachers are passing on what they themselves have been taught in faith schools. Though we must distinguish here the passing on of what is known groundlessly,and coming to know by a chain of testimonial inheritance beginning with someone who just knows. The chain cannot begin with faith.

A sceptic can of course reject out of hand the thought that there are such divine tellings, but he would not thereby be making a mistake about what faith is. Elizabeth Anscombe was rather willing to go along with apparently hostile descriptions of faith. She quotes Ambrose Bierce's definition from The Devil's Dictionary: ‘The attitude of mind of one who believes without evidence one who tells without knowledge things without parallel’. She concluded, after a short discussion, that this is more or less correct – since people will characteristically believe extraordinary things as a result of picking up what they are told by quite unlearned informants who do not offer evidence (14).Footnote ⁵² Philosophers talk of ‘fideism’. But these reflections might suggest to us that this made-up word is one we have no use for. Fideism seems either too universal or too absurd to deserve a name: too universal and every-day if it is simply a matter of often believing N that p, too patently absurd if it is a matter of believing N that N exists.

As we have seen, a man who believes N that p thereby presupposes certain things. What then are we to say when one of the presuppositions turns out to be false? In particular, can he still be said to believe N that p, even if N does not exist? It is natural to say yes, Anscombe suggests, adding that it would be very inconvenient not to be able to talk in this non-committal way (16). Believing N, just like admiring or fearing N (but unlike kicking N), does not presuppose the existence of its object. We must, she adds, say something similar about the case where N does exist but has not conveyed that p. However, a Christian will mean by faith: believing God where what is presupposed is true (19). In this narrower sense, if there is no God, or if God has never revealed anything, there is no divine faith either.

That we are in fact recipients of a divine message cannot, Anscombe insisted, be something we learn on the authority of a sage, a philosopher or a theologian (18). It is easy to see why she might want to have said this, but it seems to me to introduce a remarkable uncertainty. I am reminded of those who claim that so and so ‘has a vocation,’ or has, on a certain occasion, had an answer to his prayer, and of our natural doubts as to how anyone could know such things. We are very ready to believe that at least some of such claims are rash even when they are confidently made. How can we tell that they are not all rash?Footnote ⁵³

It is not, says Anscombe, that the learned have nothing important to contribute. But ‘the only possible use of a learned clever man,’ Anscombe says, ‘is as a causa removens prohibens’(18). I imagine this task of obstacle removal to be considerable, partly because of the activities of other learned clever men. Anscombe herself in a brief article reprinted here lists twenty opinions common among modern Anglo-American philosophers which are inimical to Christianity, each of which, as she claims, represent philosophical errors recognisable as such without Christian presuppositions (66–8). That in itself represents a good deal of ground clearing. And no doubt there are many further obstacles not introduced by philosophers but current in the general teachings of our time.

There is one respect in which we can appreciate the difficulty which leads to the abandonment of the traditional expectation of apologetic underpinning. Anscombe asks us to consider ‘the faith of the simple’, and there is indeed a problem here if it is assumed, curiously but understandably, that this faith or certitude cannot be less available to them (12, 91). But how can this assumption hold? It might be hard to see how the simple could be entitled to opinions at all where the learned disagree, let alone to certitude. And it must further be acknowledged that we are all fairly ‘simple’ in relation to this topic or that. This becomes more and more evident as science, mathematics and history become ever more specialised. McTaggart argued that in view of the prevalence of controversy in religious matters, people could not reasonably take their opinions on trust, and could only accept religious doctrines on the basis of their own metaphysical enquiries. And most people, he added, would be quite unable to undertake such studies, lacking learning, ability or time. We were thus driven to the conclusion ‘that, whether any religion is true or not, most people have no right to accept any religion as true’.Footnote ⁵⁴

What McTaggart said could be even more threatening to simple faith than might at first appear. For this limitation upon the unlearned would still obtain even if the rational acceptance of a religion turned merely upon the removal of obstacles, at least insofar as the learned disagreed over whether the obstacles were removable. But perhaps rational acceptance merely demands the belief, which need not be rash, that the obstacles are removable, removable by someone now or at some future time? Think of the way people believe that there are solutions to various paradoxes which arise in philosophy, even though no one can yet provide them. This whole topic, ‘scholarly disputes and simple minds,’ calls for further exploration. Philosophers of religion will tend at this point to have recourse to dodgy thoughts of ‘religious experience’, conveniently undeniable from within and not subject to impertinent scrutiny from without. But I suspect that our problems here are not at all confined to questions of religious belief.

V

If one merely reads the article ‘Faith’ in this collection an uncommitted observer might begin to wonder why anyone could sensibly acquire or retain Christian beliefs. The next article in this book, a substantial piece about Lessing on history, prophecy and miracles, might seem further to support this verdict. Anscombe not only criticised the old apologetic argumentation for its faulty understanding of rationality in belief. She saw it as buoyed up by an inappropriate self-confidence. In this article in particular she repudiated the idea put about by Catholics and others that an examination of the New Testament record would demonstrate to pretty well any fair minded person that various alleged miracles, and in particular the resurrection of Christ, probably occurred (25–6). Not true, she said. Such a person if challenged could reasonably reply that he had simply no idea how such extraordinary stories got into circulation and how people came to believe them. Indeed we are often in this kind of position in regard to exotic claims. The all-too-familiar disjunctive demand ‘either to accept the claim or to explain the error’ would itself be unreasonable. ‘You have to know a man, and to know him extremely well, to judge the truthfulness of that sort of story: a point which I think makes ridiculous the argument that we have to believe what the earliest Christians said they had observed, or said they had [learned?] from people who had observed them’. After all, she said, people just love wonderful stories (32). If an illustration were called for, we might remember here the accounts of Daniel Dunglas Home, and his levitations out through one window and in through another, seventy feet over the street below, and of the unimpeachable witnesses to these events.Footnote ⁵⁵ On the topic of miracles Anscombe rejected both the traditional Christian apologetic and the Humean anti-apologetic. This last is disposed of rather briefly in a later essay in this volume, with appreciative references to the critical work already put in by two distinctly nonchristian commentators, C. D. Broad and Antony Flew (40–48).

However Anscombe showed here that she had not abandoned the apologetic task – perhaps remembering that Christians are ‘to give a reason for the hope that is in them’. There is no post-Wittgenstein suggestion here that just because Christian practice involves various ‘language games’ (obeying, asking, thanking, greeting, praying, are all listed as language games in Philosophical Investigations, Sec. 23) a call for epistemic justification is thereby rendered inappropriate. For Anscombe, bad apologetic arguments as to miracles etc. are not bad qua inappropriate; they are bad qua inadequate. Moreover the bad tends to undermine the standing of the good (26). Her attitude to natural theology was thus quite different from the total hostility we appear to find in Karl Barth, for whom faith meant ‘no crutches’, and who said that ‘If you had historical evidence [for the resurrection] it would not be the resurrection!’.Footnote ⁵⁶ We surely have some such evidence – though we should be asking with some caution what evidence for the resurrection would have to be like, for as Anscombe points out – making difficulties as usual – ‘the death of Christ would be refuted, in normal circumstances, just by his reappearance alive’ (28).

Anscombe's attitude to the traditional apologetics is more of a piece with her objections, published very early in her career, to C. S. Lewis's way of dismissing ‘naturalism’ (reprinted in Papers, Vol. 2, 224–232).Footnote ⁵⁷ She was however prepared to argue that there was a rational way into Christianity, persuasive though not demonstrative, available to an impartial and indifferent observer. It would not cast aside any thought of a proof of God's existence. She made repeated efforts to construct such a proof as a teenager, rejecting one after another all five of her versions (Papers, Vol. 2, Introduction). And as an undergraduate she had been much disturbed by the thought that there could not be a proof of God's existence, at least on the basis of what we know about the world, since truths about the world would all be contingent whereas God's existence could not possibly be so (xvi).Footnote ⁵⁸ Late in life she claimed to have discovered in Anselm what might well be a proof of God's existence, something other than the would-be proof which has always been attributed to him; she could not make up her mind whether it was valid.Footnote ⁵⁹ It is rather striking that people will think it worth insisting that there cannot possibly be such a proof when we find them on the very next day insisting, with some show of philosophical sophistication, that nothing whatever (outside mathematics let us say) can be proved.

Reason would dictate, Anscombe suggested, that God, if he exists, is to be worshipped. Aristotle saw this – and not on the basis that God had commanded it, for Aristotle, as far as we know, did not believe in the existence of divine commandments (120–1). The point presumably is this: that worship, where it is called for, involves the recognition of a special kind of worth and of the appropriateness of a certain response. But of course, Christianity is more than a mere belief in the existence of God and the thought that he is to be worshipped. And here, in Anscombe's view, a solid and reasonable argument for Christianity would be available. It would however follow a path unlikely to be taken by anyone today. It would start from an appreciation, on the part of someone who already believes in the existence of God, of the impressiveness of the moral law as laid out in the Old Testament, a person who only later comes across the New Testament and finds in it not only something continuous with the older books, but a solution of puzzles and paradoxes in them. It has been all too easy for us in the West to miss this impressiveness, she argued, for the Hebrew ethic has for so long represented just ‘morality’ to us that it has been difficult for us to see and admire it (36), having had nothing to contrast it with. An appreciation of this background is necessary. ‘Only if a man is impressed by the Old Testament, to the extent of being inclined to take it as his teacher, has the argument from prophecies and miracles any weight’ (34). ‘When Augustine said that the fulfilment of the prophesies in Christ was the greatest proof of his divinity, what he said was true; but the proof requires a very special position on the part of someone who is to consider it. … That is why the kind of apologetic … which did not assume that position, was so vulnerable and stupid’. Mary Geach relates how she had once asked her mother why she accepted her religion, and Anscombe had replied: because she believed the teaching of Christ (xviii). And once again we must surely say that a certain background is presupposed. One has to be considering the truth of ‘the whole revelation’ and seeing that things fit together. The miracle stories are consonant with all this. But they are not independently capable of convincing the indifferent historian so as to provide a foundation (37).Footnote ⁶⁰

Christianity seen in this way is not something which can be cut off from its Jewish root. ‘To begin as a Jew and see in Christ the promised Messiah’ was, Anscombe said, ‘the “royal road” to the Catholic faith’ – as her daughter Mary Geach reports in her Introduction. Mary Geach relates how she once took a Jewish friend to the Easter Vigil. This friend said afterwards to Elizabeth Anscombe that she now understood that ‘Christianity was making available to the Gentiles the riches which belonged to Israel’ and Anscombe had agreed with her (xxv–xxvi).Footnote ⁶¹

Let us suppose that somewhere in accordance with this sketch what might be called the case for Christianity is to be found. It seems important to stress, however, that elements of this line of argument would not need to be appealed to, or even be known about, by countless people who, quite reasonably, profess the Christian faith. In regard to faith, as we saw, a thinker is no more privileged than a tinker. Simple men are taught what to think by teachers who may themselves be quite simple. ‘According to faith, this simple man and his teacher have a belief in no way inferior to that of a very learned and clever person who has faith’ (14). It is perhaps not unusual for the simple to be on a par with the learned in this way, and in matters which have nothing to do with religion.

VI

Here I will return, as promised, to Anscombe's remarks about the prejudices so many of us harbour about what it is to be reasonable in holding a belief (13). Such prejudices are surely to be expected–both in regard to the requirement for evidence and for understanding. And by ‘prejudice’ we should not think of ill will. We extrapolate from local cases, court procedures, say, or the routines of normal. Science, without thinking of the framework within which these everyday things occur. The topic becomes interesting when we come to realise against initial thoughts, that there can be national belief without evidence, and indeed without a deep understanding of what is believed. One would have liked to ask Anscombe at this point: What then are we to say? What is it to be reasonable in holding a belief? To be sure, as our remarks at the outset have indicated, the notion is very difficult to characterise, and she might not have had a comprehensive answer to give. All the same, we do not need to know how to bake a cake before being able to list several ways of not going about it. There are in fact a few hints or clues to what she might have said.

Anscombe would have wanted to point out how we have tended to misconceive the place of testimony. ‘The greater part of our knowledge of reality rests upon the belief that we repose in things we have been taught and told.’ This is a matter ‘of huge importance for the theory of knowledge.’ ‘Nor is what testimony gives us 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’ (all 3). She was writing thirty years ago, but although there has since been some attention to testimony as a philosophical topic I think the lesson is still not well learned. It is still perhaps not appreciated how deeply our possession of language and our ability to think and to appreciate reasons is bound up with the verbal transmission of information. Reliance on testimony is not just a shortcut that given more time we could do without.Footnote ⁶² And Anscombe's remark about undetachableness seems apt. Evidence spoken in a courtroom is not for the most part what we call the evidence of our senses even though it is something we hear and see, certain noises, certain patches of colour. Sometimes we are shown items, exhibits. But we can only interpret their significance by relying on what we have been taught – very often without any memory of having been taught it, when we initially acquired language for example. We might also think here about proof, either mathematical or forensic. Proof sounds like something quite independent of testimony. However people who believe that there is a proof of this or that nearly always believe it because they have been told so. Or they will believe the premises because they have been taught that they are true, or have picked up such a conviction from their cultural surroundings. And even a student who is taken through a proof of this or that in the classroom will trust his teacher that there are no subtle objections to its standing as a proof. None of this need be rash.

‘The belief that we repose in things we have been taught and told’ extends more widely than we might expect. It would not just be a matter of trusting eye-witness reports, even at several times removed. Indeed we sometimes judge that there were witnesses because we know the event happened rather than judge that the event happened because we know there were witnesses. For we know that a dramatic event taking place in public would have been seen and reported. Neurath's image of our system of belief as a boat which has to be repaired while at sea is familiar and useful. But Anscombe noticed a way in which this analogy fails, at least if we are supposed to conclude that we can pick any plank of the ship and test it for rottenness. There are beliefs we cannot test, for the following reason: a doubt about them brings with it a doubt about anything we might test them by.Footnote ⁶³

We can often see the influence of On Certainty in Anscombe's remarks. The possibility of investigation presupposes mere acceptance. That is to say, the quest for grounds emerges from a groundlessness (perhaps not always quite the same for everyone) which cannot be dispensed with. So Anselm's remark ‘I believe in order that I may understand’ need have no particular reference to religious belief. And acceptance is not for the most part a policy – it is more of an inevitability. In the beginning was the word. And here ‘a word’ means something propositional, as in ‘a word in your ear’ (see also Anscombe's own usage, 19).Footnote ⁶⁴ Investigation comes later. Investigation, like ostensive definition, is only possible when a place has been prepared for it.

We should note a distinction Anscombe claimed to see between kinds of knowledge. Some of our knowledge she called ‘connatural’, to be distinguished from the knowledge which accumulates in libraries for example (200).Footnote ⁶⁵ Connatural knowledge is bound up with the virtues. Prudence gives one a nose for what is foolish, justice for what is unjust, etc. – a perceptiveness which takes over when the guidance of the rules run out, as they must eventually do (100).Footnote ⁶⁶ Reasonableness in belief is no doubt an aspect of prudence, shrewdness or good sense, and it will be a matter of connatural knowledge to have a nose for foolish belief. Wittgenstein naturally rejected witchcraft beliefs, but thought it naïve to suppose that science had somehow demonstrated that such beliefs were mistaken, or at least unreasonable (on the ground that there was “insufficient evidence” perhaps). He told Elizabeth Anscombe that he would want to stop someone becoming a witch doctor even though he was not well able to say why (58, and Papers, Vol. I, 125, see also Wittgenstein's On Certainty, sec. 609–612). Anscombe would have thought his instinctive attitude in the matter perfectly reasonable. She suggested that (said she was inclined to think that) it was not science or common sense or Western enlightenment but (a certain kind of) religious outlook which enabled us to argue against superstitious beliefs in spirits (79–80).

These various remarks about the concept of rationality are not I think partisan. Efforts have been made in recent years to construct what might be called ‘labelled epistemologies’ which one naturally suspects to be designed to give factional results, rather as a suitably Biblical Mathematics might be constructed to ensure that π comes out to exactly 3, in order to accommodate the measurements of Solomon's architect (I Kings, 7. 23). A partisan fix in regard to reasonableness would be recognised as self-defeating.Footnote ⁶⁷ That said, it is not too clear when a fix counts as partisan. A fix which simply excludes what is not to be taken seriously would be fine. And Alvin Goldman rejects an epistemology of warrant partly for the reason that it would be too hard on atheists, and that might be fine too (Op. cit., Sec. 5). In any case the attempt to set up a fix in regard to our present topic would seem unnecessary. Christian and atheist philosophers, a Geach and a Quine let us say, can largely agree upon these fundamental matters.Footnote ⁶⁸

VII

When it comes to the relation between her religious belief and the reasons in support of it, Anscombe brings up two particular difficulties and I shall end by briefly considering them. One concerns certainty, and the other concerns relevance. She rather leaves it to the reader to sort things out. I shall discuss these questions in terms of the web and the will.

The first difficulty could be put in this way. How can we base certainty upon uncertainty? The ‘central objection against faith,’ Anscombe says, is this: ‘it is [supposed to be] certain, its grounds – if it has grounds – at best probable: what excuse can one have for it, then?’ (23). Must not this transition be illegitimate? This is more or less the objection we also find upheld in the last chapter of Anthony Kenny's Faith and Reason. Can one rationally believe that p for the reason that q with a commitment stronger than one's commitment to q? ‘Unless the relevant stories [from history] can be as certain as the commitment which faith demands of the believer, the commitment is, so far forth as it is faith, irrational.’Footnote ⁶⁹

The second difficulty is this. Someone comes to adopt Christianity, and he supposes that he does so solely by way of what he considers probable arguments. He then finds out that the arguments are entirely worthless. But, Anscombe says, ‘he is supposed to hold on to faith just the same’ (23). How is this to be squared with the claims of rationality?Footnote ⁷⁰ Anscombe does not answer this latter question, or at least not directly Yet we can perhaps make some progress with it. We might point out that such ‘holding on’ is of a type accepted in other contexts, even by those hostile to Christianity. This is very evident in ethics. A Christian might believe that it is bad to steal simply on the ground that stealing is contrary to God's law. But if he ceases to believe in God and his law he will probably continue to believe that stealing is bad, and will be praised on all sides for continuing to do so. Is this response only thought to be reasonable on the assumption that adequate grounds can be provided by others – that such grounds are here-and-now available, though perhaps only to be pieced together from volumes buried in dusty library stacks? I do not think we are entitled to rely on such an assumption. Philosophers, if they are any good, will often find it difficult to explain the way in which bad conduct is bad and they might well give conflicting accounts. An ironical paragraph by Bernard Williams and Michael Tanner, replying to Anscombe's ‘Contraception and Chastity’, is relevant. Responding to Anscombe's remarks on how to read an encyclical, they comment: ‘It is a curious feature of Prof. Anscombe's view of these things that both Church and Pope turn out to be much better at conclusions than reasons’.Footnote ⁷¹ Williams and Tanner rather cut themselves with their own knife. It is part of the human predicament rather generally that we are often better at conclusions than reasons. This is especially evident in moral philosophy. And in metaphysics too, if we suppose, with Bradley, that this is matter of the finding of bad reasons for what we believe upon instinct.

It is helpful – in regard to both of these difficulties – to picture rationality as involving something like a web, a web of belief – to draw on the familiar analogy we associate with Quine and to some extent with Wittgenstein. What stands fast is held fast by what stands around it (On Certainty, 144). To be sure this picture, as with the picture of a building resting on foundations,Footnote ⁷² can be expected to mislead in certain ways. A spider's web has a boundary. Does the web of belief? We might want to say that certain propositions, propositions ‘at the boundary,’ are properly regarded as basic, but the criteria for inclusion will be very vague and the boundary a shifting one, so much so that the very word ‘basic’, or the image of a web with a boundary, will seem not quite right. We seem to need the category of ‘fairly basic,’ but a web does not have semi-anchorages.

Insofar as we can talk of a boundary it will not simply be bounded by descriptions of what the individual has seen or heard. The supposition that belief-webs are so bounded, perhaps referred to as empiricism, would itself be an item in this or that philosopher's own belief-web, an item which he might, after taking thought, be reluctant to retain. Yet it could hardly have been an item in Elizabeth Anscombe's own belief web – for the idea that we can have knowledge without observation is so prominent in her work.Footnote ⁷³ The a-priori thought that empiricism, thus understood, must be right, is perhaps connected with the idea that all we ‘really’ hear is sound, rather than a teaching – an idea which simply invites philosophical disentanglement. Hearing a sound does not have to be comparatively foundational; rather as we saw earlier, we often infer that we must have heard certain noises because we now know this or that – these noises being the presumed causal origin of what we know. This kind of empiricism will tend to think of testimony as itself a kind of experience, vicarious experience, ‘seeing with another's eyes’. But the cases where this figure is apt – a blind person's being informed by a sighted companion of the approach of a bus, say – seem quite exceptional.Footnote ⁷⁴ These remarks represent a kind of counter-empiricism. They maybe represent settled progress. Anscombe talks about ‘the eclipse in recent philosophy of the notion of “the given”’ – comprised of and confined to the empirically given basic truths. ‘Asked what was given, a present-day English speaking philosopher would be very likely to say “the lot”’ (Papers, Vol. 1, 84, ‘present-day’ meaning 1965). The given, that is to say, is the whole of the structure of Neurath's boat.

We are apt to draw a false contrast between knowing by argument and knowing by experience. Thomas Nagel writes: ‘Most people are believers or nonbelievers in the existence of God not as a result of argument, but in a much more basic way. They either see or experience God's presence in the world and in their lives or they don't’.Footnote ⁷⁵ There is something artificial about this. The claim both about argument and about experience are misleading, whether about belief in God or about other matters. Within a web formed largely by testimony the phrase ‘determined by argument’ will often have an indeterminate sense. As to the claim about experience inappropriately suggests the duck-rabbit. And what is here called experience would be better called interpretation, interpretation in the light of other items in the web.

Sensory experience would indeed provide some anchorage, but so too would those matters of which in the law of evidence judicial notice is taken, such as what are termed ‘notorious facts’ – which occasionally of course will not be facts at all, as was once the case with one of the ‘certainties’ listed by Moore while delivering his Howison Lecture. In cases of conflict the individual's own sense experience would not be the final arbiter. This will be so even in cases of what we call empirical enquiry. The possibility of empirical enquiry in any case can become closed in regard to the past, in that traces of events can quite disappear. Anscombe drew our attention to the great class of historical assertions which must in general be true, though the time for disproving any one of them is past (28), none of which could be held certain. Not that certainty in history is never attainable, or could only be second-class certainty. Anscombe objected to the way Lessing lumps together everything historical as of inferior certainty to what one knows as a result of one's own experience (27). And in particular she denied the claim in Lessing that nothing in our knowledge of history can be completely certain. And she meant it. She showed herself willing to bet her salvation on the real existence of Alexander the Great (26).

When it comes to the question ‘what is to be tested by what’ we will often have a certain choice, as was already evident in Quine's original account. This shows, that what is called ‘rationalisation’ is not always a sign of contemptible weakness as superficial rationalism regularly suggests. I say we will often have a choice. But in Quine's ‘Two Dogmas’ account we always have a choice. Empirical evidence, on this account, never settles anything – let one's respect for experience and for logic be ever so strong. Given the evidence any statement can be held true, we just have to make suitable adjustments elsewhere in the system. And I do not think he meant to point out that any statement can be held true by someone sufficiently unreasonable or insufficiently concerned with truth, for that would rather go without saying.Footnote ⁷⁶ This Quineian evidential indeterminacy would surely be impossible to live with. Not for nothing do people talk about the evidence in relation to this or that. We want to know for example whether someone actually committed the crime for which he is to be punished. He would surely have a complaint if our verdict turned on Quine's ‘vague scheme of priorities’: an unwillingness to put up with complexity or inelegance in our system of beliefs, or on our laziness when it came to making changes, which is the sort of thing he had in mind. In some way Quine's ‘always’ has to be modified. About this I am not sure what to say. I will merely repeat what I said earlier: Only a mind that is closed in many ways is in a position to make enquiries.Footnote ⁷⁷

At any rate, there is sometimes a choice. The remark from Wittgenstein cited above – that what stands fast is held fast by what stands around it – suggests a misleading passivity. What look like grounds for believing this and that can become adopted in the light of those beliefs already held. We saw earlier that presuppositions were not necessarily foundations. Hence Anscombe can say, rather contrary to expectation: ‘It is probably common among us [Catholics] to believe in fulfilled prophesies and miracles because we believe the Catholic religion and they are part of our doctrines’ (25). The same is doubtless true, we might suppose, about Christian belief that prayers are sometimes answered, something which can only be weakly ‘evidence based’ at best. Surely no one becomes a Christian because they prayed for rain and it then rained, even if this happened on several occasions.

If our beliefs form a web which routinely involves the reasonable individual in a choice as what is to be held true, it is very misleading to think that rationality, perhaps called respect for evidence, marches one ineluctably up to various points, after which something called faith takes over and one is invited to make a glorious, or reckless, ‘leap’ – which might (to choose a non religious example) take the form of insisting, in the face of Russell's teasing scepticism, that the world is more than say 6,000 years old. This is certainly not the notion of faith under consideration in this book.Footnote ⁷⁸ The analogy of the web, in Wittgenstein or in Quine, undermines the thought that to every suggestion there is a reasonably well defined package specific to it, called ‘the evidence’, something we might rashly (or gloriously) go ‘beyond’. And if the talk is of the ‘available’ evidence, thoughts of a well defined package are even more obviously inappropriate. A complaint raised in everyday life that someone is rashly ‘going beyond the evidence’ seems to presuppose a limited context that does not exist when we are discussing reasonableness in belief more generally.

Thoughts of a leap into the blue yonder give a false picture of the place of epistemological choice. We might make a comparison here with utilitarianism. Utilitarians have typically assumed that there is such a thing in every circumstance as ‘the’ right action, after due allowance for joint winners. Perhaps the common view of rationality in belief is analogous. That is to say, there is supposed always to be such a thing as ‘the’ right, or ‘the’ rational, belief, the rationally obligatory belief, relative to our circumstances and perhaps abilities. And in a limited context this might seem correct. But perhaps there is a place for the rationally permissible, that is to say for mere permissibility. This might even allow considerations of loyalty to determine belief within the limited range of the permissible, for saying something of the form, ‘As an X-ist, I believe in such and such’. This idea seems to have a pedigree. Quine for example in ‘Two Dogmas’ tells us what he thinks about this or that as an empiricist. 40 years later, in The Pursuit of Truth, he appears as a staunch empiricist.Footnote ⁷⁹ People will talk of ‘faith’ here, perhaps unfortunately as it has nothing in particular to do with testimony. Galen Strawson for example writes: ‘My faith, like that of many other materialists, consists in a bundle of connected and unverifiable beliefs’.Footnote ⁸⁰

Bas Van Fraassen talks about the way we philosophers adopt a stance. It might be part of such a stance I suppose that, like Mach, one refuses to believe in atoms. And now I see, on re-reading what I said towards the outset of this article, that I myself found it natural to use this Fraassian word in describing Anscombe as a Catholic Christian. And what is a stance? At least I would want to say this: that a stance is not a matter of feeling. It is more a matter of will than of wish, of resolution than of mere reaction. It would not follow of course that a stance is beyond criticism.

The possession of a stance might help one to hang onto a belief in the face of epistemological mood-swings. Michael Dummett, on being asked about his religious belief, described how such mood-swings had affected him: how one might begin to wonder whether one's thinking as a whole might not have gone awry, and whether one could rest anything on the very general reflections that seem to be involved.

This, it seems to me, is both admirable and troubling in equal measure. Is not this a form of wishful thinking? We might at least insist that the belief in question, to which one is moved by a loyalty of this kind, must be within the range (no doubt vaguely defined) of the rationally permissible. But that said, it would seem rash to suppose that a philosopher of such manifest intellectual power and seriousness could be making an elementary mistake about what reasonableness in belief, or a proper love of truth, requires of us.

The reflections in this paper taken more generally seem themselves to leave us with a choice. Do we understand enough about what it is to be rational to be able firmly to say that divine faith, that is to say, believing God, taking him at his word, cannot rationally be possible in this modern world? Or, if we are surprised to find the persistence of Christian belief even among those who appear to have remarkably acute critical standards, does our reaction merely show that we do not understand enough about what it is to be rational? I am strongly inclined to think the latter. The persistence of crude thoughts about reason is very striking. Rationalism, in the tub-thumping sense, is the failure to appreciate our conceptual difficulties. But what better thoughts are available to us? Anthony Kenny began his lectures on faith and reason by saying that he had found it much more difficult to define reason than to define faith, indicating that this was something of a surprise.Footnote ⁸² That there is this difference in our understanding of these concepts is well brought out in Elizabeth Anscombe's discussion of these questions.