What is the difference between reason and faith? The question is framed in what I would call ‘the treacherous singular’. The structure of the question implies a particular form of answer and makes other assumptions about notions that occur in the same region of our network of thoughts and understandings. If I were happy to play this game I might reply in kind by offering a simple formula purporting to sum up my own answers to the cluster of questions that are implicit in the simple form given above. The form makes the question and the answer appear more straightforward than they are. Perhaps I might answer in the questioner's style by stating the conclusions of my paper in these words:

A simple idea is at the heart of my paper, but one that is hard to absorb and to live by in one' s wider thoughts about reason and faith. The simple idea is that faith is itself a mode of reason. This means that there cannot be a sharp distinction between reason and faith, whether or not one tries to express such a distinction in the treacherous singular, or perhaps in the impoverished plural, a mode of expression in which a few separate but simple factors are identified as what faith is or what reason is.

The categories of reason and faith are often contrasted. When reason gives out, we are told that we have to rely on faith. Such exhortations are made particularly in the context of religion. When for instance, we face some personal tragedy which may well seem inexplicable, we are told that faith can help us through it. Very often faith is referred to in a vacuum. Presumably faith in God is usually meant, but all too often God drops out of the picture, and it seems that all we need is faith, not faith in anything or anyone, but just faith. We are thus encouraged to add what seems to be a magic ingredient to our lives, which can transform everything. Perhaps at the back of such thinking lies some Calvinist notion of the corrupt character of human reason. As a result it may seem that we cannot rely on our judgment, which is the product of the fallen and sinful nature of humanity. Instead we must depend on ‘faith’ which may, or may not, be given us by the grace of God.

This title represents, I suppose, a kind of challenge; for there seems at first sight some incompatibility between the practice of logic and mysticism, a contrast between the rational and the intuitive, the tough minded and the tender-minded. In taking up this challenge, I propose to argue with the help of two thinkers commonly admired for their attention to logic and its rights. I shall refer for the most part to St Thomas Aquinas but with occasional reference to Wittgenstein. Whatever may be said of the latter, it seems to me quite clear that St Thomas was a mystical thinker in that he was centrally concerned with the unknown and, in one sense, ineffable mystery of God and that he devoted a great deal of thought and writing to the problems associated with speaking of what is, in this sense, ineffable. I want to argue that in what is sometimes misunderstood as his dryly rational approach, even in his arguments for the existence of God, he is in fact engaged in, and inviting the reader to be engaged in, a mystical exploration, which is not at all the same thing as a mystical experience. Here the key notion is that of what he refers to as esse.

To talk of a logic of mysticism may sound distinctly odd. If anything, mysticism is alogical; it would be uncharitable if not false, on mature consideration, to call it illogical—though many, without due deliberation, might be tempted to use that term. Wittgenstein comes close to calling it illogical. In his lecture on ethics he draws attention to the logical oddity of statements of absolute value (Wittgenstein 1965). But he does not accuse the mystics or prophets or religious teachers of contradicting themselves or of invalid reasoning. What he accuses them of may be something worse, namely, talking nonsense, of not giving sense to the words they use or the expressions they utter. Russell (1921) and Ayer (1936) come to much the same conclusion but by a different route.

In this paper I shall be examining the following claims:

that ‘God’ has no meaning, in the sense of ‘sense’; that it is a proper name analogous to a Russellian proper name in that it has reference only. More crudely, that we cannot describe God in any way but only name Him and refer to Him by name;

that ‘God’ has no meaning, in the sense of ‘sense’ in that the nature of God is fundamentally inexpressible. That God is, as some have said, ‘wholly other’ or even ‘the wholly other’.

Holding God to be transcendent does not mean having to regard the grammar of the word ‘God’ as isolated or unique or inscrutable: and in speaking of grammar I use this word in its familiar sense, not in some ill-explained neo-Wittgensteinian sense. I want to make a methodological suggestion. When a sentence containing the word ‘God’ is puzzling, it may help to look at a grammatically possible replacement for the word. For example, if we wish to understand the statement that God is man's last end, let us consider what it implies to say that gold, or sex, or military glory, is a given man's last end.

‘There is a kind of eloquence’, maintained St Augustine,

which is manifestly inspired by God. Biblical writers have spoken with this kind of eloquence. … ‘On the other hand’ they have uttered some passages with a beneficial and salutary obscurity (salubris obscuritas), to exercise and, in a sense, to polish the minds of their readers, to break down aversions and spur on the zeal of those who are anxious to learn, as well as to conceal the meaning from the minds of the impious. (Augustine 1947: IV.6.9; IV.8.22, with one amendment to the translation)

Martin Warner's subtle and far-reaching synthesis of philosophical theology and philosophy of language belongs in a cluster of papers he has written on related topics (Warner 1985, 1990a, Introduction to 1990b) so it would be helpful to begin by setting out this wider context. His concerns overall cover three interlocking subjects: biblical interpretation, biblical translation, and reform of the liturgy. All pose a central conundrum, which in its briefest formulation is just this: what kind of meaning is involved in each case? Warner's particular focus is on the role of distinctions like content and style, truth and connotation, literal and figurative, ultimately semantics and pragmatics.

It was, I believe, Thomas Arnold who wrote: ‘Educate men without religion and all you make of them is clever devils’. Thus the Headmaster of one famous school summarized pithily the view of the relationship between religion and ethics which informed educational theory and practice in this country for at least a further century. There is a confusion of two different assumptions usually to be found in this context. The first is that religious belief can provide an intellectual foundation (logical, or epistemological, or sometimes both) for moral belief; the second is that the effect of religious teaching is to improve behaviour according to the norms of some particular set of moral beliefs.

Professor Sutherland has argued that ‘God wills the good’ should be regarded as an analytic truth, with the consequence that any account of what is God's will in which it does not appear to be good is either a mistake about God's will or a mistake about what is good.