I am pleased to see Timothy Hinton come to Gareth Moore's defence. I felt uncomfortable having to be (as I saw it) so negative and sceptical in what was a memorial volume for an old friend.

Hinton accuses me of at least three more or less distinct mistakes. First, I quote too selectively, focusing on two passages and not seeing the book as a whole. Second, even in those passages, I miss out bits that do not suit my case. Third I operate by ‘guilt by association’, taking the affinity Gareth's method has with D. Z. Phillips's as showing that he must share Phillips's denial of survival and resurrection and general opposition to the kind of metaphysics that seem to be integral to religious belief. He then provides what he believes to be an alternative interpretation to mine of what Gareth is doing. It is on the last that I will concentrate, because the availability of Hinton's alternative interpretation is crucial to whether his criticisms of my reading are justified.

Hinton says, ‘Believing in God is an astonishingly clever and elaborate philosophical response to Antony Flew's paper “Theology and Falsification”’.

As Hinton explains, Flew's paper concerns Wisdom's parable about the discovery of a seemingly cultivated clearing in a jungle:

The point of the story is that religious beliefs, in the form they are represented as being held, are unfalsifiable and so empirically vacuous, if not strictly meaningless. The various tests given in the parable – electrified fence, bloodhounds etc. – are not meant to be the kinds of tests that would detect God. To think so would put Flew in the same category as Krushchev, who said that when the first astronaut, Gargarin, failed to see God, that showed atheism was true. Rather Flew can be taken to be thinking of the kinds of arguments based on the supposed nature of the universe, such as first cause and design, which he believed, at that time, science to have undermined, step by step.

According to Hinton, Flew imputes, in using this parable, two vital philosophical assumptions to both sceptic and believer:

Let me deal with the epistemological assumption first. I do not think that Hinton is fair either to Flew or to traditional theology in imputing to either the belief that ‘to come to believe in God is to make an inference from observed phenomena…’. The question is whether one can appeal to such phenomena in defending religious belief, and the idea that one can is fundamental to any natural theology. There is no suggestion that this is, normally, at least, the way we come to religious belief, or that God is fundamentally postulated to explain such things. If Gareth were just denying the theoretical-hypothetical status of religious belief in God, he would be attacking a straw man, as far as serious theology and philosophy of religion are concerned. As Anselm says at the beginning of Proslogion, ‘I believe so that I may understand’, and such arguments are part of the understanding. That Flew himself did not make this confusion is shown by his own future trajectory. As is well known, towards the end of his life Flew became convinced by the ‘fine tuning’ version of the argument from design, and so became convinced that a ‘gardener’ was needed, but he did not attach any religious significance to this - it was a bare deism. He did not think one could follow on from the initial conclusion ‘and this we call God’ to any reasons for imputing the traditional properties to the designer. He knew perfectly well that there was more to religious belief than the minimal conclusions of natural theology, but held that a rational defence of religious belief appealed to such arguments.

One's first reaction to the metaphysical assumption is that the characterization of God here looks entirely normal, with the possible objection that God, as opposed to the persons of the Trinity, is not to be regarded as a person. The imputation of invisibility and bodilessness is both standard and, as we shall see, something that Gareth seems to endorse. Hinton seems to me to be rather slippery on the divine nature. He says that Gareth's point is that God is not something that is at a particular place, such that you could point him out, as you could ‘Charlie’, and that there is a ‘logical or grammatical difference between the presence of God and of middle sized dry goods’:

These remarks might give rise to various thoughts. First, one might be irritated by the Wittgensteinian use of the notion of ‘grammar’. In so far as it is a grammatical fact that chairs, etc. are present at particular places, this is because it is a fact about chairs and ordinary material objects in general that they have spatio-temporal location: grammar is not what determines this. If God exists as more than a linguistic phenomenon that is essential to certain social practices and ‘forms of life’, His nature, whatever it is, is not determined by grammar either. But more of this when we come on to poltergeists.

Second, it might strike one that the metaphysical remarks are entirely uncontroversial – no theologian has ever claimed that God is located in the same way as a chair is. Nor is there any reason to think that Flew thought that ‘God’ must be so located. Third, one might wonder how God could be ‘present, everywhere, at all times’, as Gareth says he is, unless he were ‘invisible and bodiless’, as Flew is supposed falsely to have thought him to be – unless, perhaps, he is identified with the universe, in Spinozistic fashion; or unless, of course, the sense of ‘present at all times’ is not to be taken in what most of us would think of being a literal sense. And that brings us to a serious problem.

Hinton is very keen that Gareth affirms truths such as that God is not absent, His presence is ubiquitous, etc. Unfortunately, all that this shows is that, on the Phillipsian picture, everything that is traditionally affirmed of God is still affirmed, it is just given a different – non-referential – sense. This is why (as I said in the original article) Gareth was upset at being classified with Don Cupitt's approach. Cupitt is in the same tradition as Braithwaite's ‘Christian empiricism’, which denies the factual or metaphysical claims of religion, but wants to preserve the moral and spiritual dimension. The radical Wittgensteinians, by contrast, simply deny that what appear to be the metaphysical claims are to be understood referentially, but are wholly immersed in the ‘form of life’ side of religion. And, indeed, Hinton seems to agree that talk of God is not referential, as we shall see.

The fact that Phillipsians are keen to say all the traditional things about God makes it very difficult to be sure how to take it when they affirm traditional formulae. That is why I chose the passages I did to explain Gareth's views – they are the points at which the non-referentiality shows up. The examples I chose were Gareth's discussion of poltergeist phenomena, and his remark that religious truths are not discovered, but are created.

Hinton's response to the ‘poltergeist’ case is not convincing. He claims that Gareth was merely denying the epistemological point that ‘religious beliefs are reached by making inferences from observable phenomena’. But this does not fit what Gareth says, for he seems clearly to say that poltergeists are not something inferred, they are just names for the phenomena in our experience, and the implication is that God is like that. It is not just that we do not rely on such phenomena to ground our beliefs, but that there is not a divine presence ‘behind’ the phenomena: the phenomena, and the practices associated with them, are all there is. This, at least, seems to be the force of how he uses the parallel with poltergeist phenomena.

This leads back to the point about ‘grammar’, chairs, and God. Phillips, in his article in the memorial volume about Gareth says the following:

‘Looking at our practices’ is a hopelessly vague and underdefined notion, unless it is backed up by some theory about the nature of the context in which the practice takes place. When I look at our practices with religious language, I am strongly inclined to think that they show that we think of God as an immaterial spirit who actually created the world, sustains it in existence, etc. It is only against a background of the assumption that ‘practices’ has a restricted sense, implying that there is nothing more to them than what one might loosely call their ‘social role’, that a traditional understanding is ruled out. This parallels how poltergeists are treated by Gareth.

The justification of saying that the treatment of poltergeists is positivistic is shown by how similar it is to the maxim that Bertrand Russell gives in his ‘logical atomism’: ‘Whenever possible, substitute constructions out of known entities for inferences to unknown entities’.Footnote ² This is exactly the model for Gareth's treatment of poltergeists. It is also how Wittgenstein's anti-private- language argument works. Sensation reports do not refer to some phenomenon that lies behind the criteria that prompts our use of the terms, but captures the syndrome of causes and effects that ground our use.

A distinction must be made here between what one might call ‘strict construction’ and ‘informal construction’. Early positivists, with a strict verificationist theory of meaning, adopt the former because they claim that the terms introduced on the basis of the evidence mean the same as – are just a kind of shorthand for – the syndrome or evidence on which they are based. The later Wittgenstein does not follow this ‘translation’ form of construction; the meaning of the introduced term is not the same as that of the evidence reports on which it is founded, because its use or function is different; it serves a different purpose. But, though it is a genuine conceptual expansion, it is not an ontological one, for it does not introduce any further entities.

It seems I did not express myself clearly, on how I take Gareth to be using the word ‘God’. Hinton says

But that is my main point: according to Gareth – and to Phillips - ‘God’ is not a name or a referring expression, so it does not and cannot refer to anything; only in that sense is it ‘about’ an absence. This is entirely different from saying that God is a quite different thing from a chair or a human being. The term ‘thing’ is so general – too general even to be ‘generic’ in the Aristotelian sense - that to say that a noun does not refer to any thing is to deny that there is any such thing. It makes it analogous to the word ‘sake’ in ‘I did it for her sake’. It works in a context, but not by referring to anything: ‘sakes’ do not exist. Similarly, for Wittgenstein, ‘pain’ does not refer to something – ‘I am in pain’ has a different, contextual, non-referring use. Phillips is, and Gareth clearly seems to be, assimilating God-talk to cases of this kind.

If ‘God’ is not a referring expression, it does not refer to anything, and ‘name’ in this kind of logico-grammatical context just means ‘referring expression’, so that if ‘God’ is not a name, it is not a referring expression. In this case, if one wants to continue saying ‘God exists’, or ‘there is a God’ one must be using the expression in some way quite different from the way expressions of this form are normally used. And not just in the sense that God is not like a middle-sized dry good, for we can refer to all kinds of other things, such as properties, scientific theories, the number seven, a disease and many other things. If God exists at all, you should be able to refer to Him. If religious concepts such as ‘God’ are not referential, their use can only be expressive.

Hinton agrees that Gareth was some kind of expressivist, but distinguishes between strong and moderate expressivism. In the former case,

He then quotes passages from Gareth where Gareth affirms that the predicates ‘true’ and ‘false’ apply to religious statements. Unfortunately, this does not show that the statements are cognitive. Strawson's classic 1949 paper ‘Truth’ sets down, following F. P. Ramsay, the line on truth that was followed by most philosophers who wished to preserve ‘truth’ for forms of discourse they did not treat in a cognitive fashion:

That Gareth was well aware of this option - which would also certainly have been Phillips's opinion - cannot be doubted, so his affirmation of the truth of religious statements proves little or nothing.

One of the passages that Hinton cites against me strongly indicates that Gareth took this kind of attitude to the use of ‘true’:

The last sentence makes it pretty clear that what is being claimed is that the use of ‘true’ and ‘false’ has a role in this discourse. This understanding of it is exactly what Strawson is recommending; it signifies agreement in a practice, it does not imply that it is used to affirm some correspondence to any further state of affairs, as a genuine cognitive use would require.

If religious terms such as ‘God’ were moderately expressive because statements involving them can be true or false, what does the truth of ‘God exists’ consist in if ‘God’ does not refer to anything? Any realist sense of ‘true’ with sentences of the form ‘X exists’ or ‘there is an X’ must involve ‘X’ being a referring expression and having a reference if it is true, but, according to Gareth, and Hinton when expounding him, when ‘X’ is ‘God’ it does not refer. A ‘moderate expressivism’ could only take the form of saying that religious language, such as sentences involving ‘God’, as well as having reference, also carry, for example, strong normative force. This latter provision may be correct, and indeed it was clearly Flew's view, because his late acceptance of design did not lead him to religion; but without the former there is no ‘moderation’, and the referential role that the moderate version needs is denied by Gareth and Hinton.

The accusation that I employ ‘guilt by association’ is based on my quoting Phillips on survival of death, including bodily resurrection, Phillips dismisses such ideas:

Hinton says, ‘Robinson interprets Phillips to be saying that it would be utter nonsense to speak of surviving death’. The implication of calling this an ‘interpretation’ is that it is not uncontroversial to claim that this is what Phillips is saying. Apart from the minor point that I do not use the word ‘utter’, I think it is clear that saying ‘we do not know what to make of these words’ implies that they don't mean anything, which is equivalent to their use being nonsense. The only other thing such words could mean is that we are our failing in our understanding of something that does, or might, make sense, and it is plain that Phillips is not meaning to leave this option open; it is plain from the ‘grammar’ of ‘death’ that one cannot survive it.Footnote ⁴

But this is not Hinton's main point. He says that I accuse Gareth, as a ‘radical Wittgensteinian’, of being a ‘card-carrying materialist’, and that what Gareth says about heaven shows that he cannot be a materialist who denied the possibility of survival. Notice, however, that Phillips says that neither survival nor resurrection is ‘necessary presuppositions of the immortality of the soul’. Anyone who followed this line could – and probably would – equally say that talk of heaven did not depend on any such belief. This, of course, does not itself show that Gareth did not have these beliefs in a literal sense, but I believe that we have seen that he did not leave himself with the metaphysical or semantic resources to interpret them literally. My accusation of ‘materialism’ does not, of course, concern the ethical sense of that term, but exactly means that other forms of discourse than those referring to physical things are not referential, and so do not, in any acceptable sense, have ontological implications.