We all agree that you count as an atheist if you say – whether out loud or in the secrecy of your heart – that there is no God. You count as the equivalent of an atheist, perhaps, if you say things – whether out loud or in the secrecy of your heart – that logically imply there is no God. Howard Robinson believes, and would have the rest of us believe, that Gareth Moore was just such an atheist-equivalent. That is, Robinson believes, and would have the rest of us believe, that Gareth was the Mass-saying, Confession-hearing, Dominican equivalent of an atheist. To which I say, once again: there is not a single good reason to believe Gareth was any such thing.

Let's begin by reminding ourselves of our duty to regard everyone as innocent until proven guilty. This lays on us an obligation, as we ponder Robinson's case, to think of Gareth as an orthodox Christian, Mass-saying, Confession-hearing Dominican. To be sure, the presumption of Gareth's innocence is rebuttable. It would only be rebutted if someone demonstrated, beyond all reasonable doubt, that Gareth had said things that showed him to have denied the existence of God. Well, did he say such things?

What is beyond reasonable doubt is that Gareth believed this:

1. There is no such thing as an invisible person named ‘God’

Does it follow that Gareth was an atheist? That is to say: does what Gareth undoubtedly believed entail that there is no God? A given proposition, p, logically entails another, q, just in case it's impossible for p to be true and q false. Clearly, on its own, the thing that Gareth undoubtedly believed does not entail atheism. What we need is a little modus ponens:

1. (1) If there is no such thing as an invisible person named ‘God’, then there is no God.

2. (2) There is no such thing as an invisible person named ‘God’.

Therefore, (3) There is no God.

We reach a crux: while Gareth undoubtedly believed (2), he says over and over again that he does not believe there is no God. Now, to be sure, people notoriously fail to follow through on all of the possible inferences that might be drawn from their own beliefs. And sometimes, when invited to do so, they come to see that their own beliefs were not, after all, fully consistent. Showing how all of this works is what put Socrates in business. But Gareth was a perfectly intelligent chap. Did he really fail to draw the right conclusion from the conjunction of (1) and (2)? On the contrary, I suggest that what Gareth did was to reject (1). So for him, there was no inconsistency whatsoever in affirming (2) whilst denying (3).

That leaves (1). To bring out the disagreement as clearly as possible, let's recast it in the form of a categorically ontological claim. The thing I say Gareth would have denied is this:

1. (4) What it would take for God to exist is for there to be an invisible person named ‘God’.

In his original article, Robinson found Gareth guilty as charged because he took rejecting (4) to be tantamount to embracing atheism. That makes (4) well worth looking into.

For starters, there is ample reason to think (4) false. This is a logical or semantic point: ‘God’ is not a name. Names are bits of natural language, coming into natural languages by means of acts of naming (‘I name this ship “H. M. S. Pinafore”. May God bless her and all who sail in her.’) But any such act of naming presupposes the possibility of demonstratively identifying the bearer of the name. The speaker needs to be able to indicate that it is this ship over here that she is naming, not that one over there. If (4) is true, then whoever first named God needed to have been absolutely sure they had succeeded in naming the right invisible person, the one they intended to name using the word ‘God’. But how could any human speaker have been guaranteed of success in such an endeavour? For all we know, the ranks of invisible persons in whom our original God-namer believed may have included Thor and Wodin. Or it could have run to all the angels and archangels, the cherubim and seraphim. Hence, it's conceivable that the invisible person who was first called ‘God’ turns out to have been an angel. That such a thing is logically possible is the premise of Philip Pullman's His Dark Materials. (Indeed, it reveals how poorly Pullman understood the logic of ‘God’ that he seems to have thought this possibility made a half-decent case for atheism).

Moreover, when names migrate from one natural language to another, they standardly do so by transliteration. ‘Beijing’ is a rude English approximation to a string of sounds in a certain dialect of Mandarin. But ‘God’ has entirely Germanic roots, having of absolutely no Semitic heritage. So ‘God’ cannot be in any way a transliteration of the Hebrew term it is used to translate.

Someone might object: ‘But God is God, the father almighty. Surely God could have named himself “God”’. To which I say: that's not how the story goes. According to the telling in Exodus, when Moses encounters the mysterious presence who addresses him from the burning bush, he asks for the speaker's name. The reply he gets takes the form of a Hebrew expression usually translated as ‘I am who I am’. That would be like asking someone their name at a garden party and being told ‘I am who the host thinks I am’. One might well feel fobbed off, and simply move along. So Gareth appears well within his rights to say that ‘God’ is not a name.

In his reply to me, Robinson says what appear to be two conflicting things about whether he thinks Gareth believed that ‘God’ is a name. In one place, he writes that Gareth ‘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’ (“Reply”, p. 6). The upshot, it seems to me, is that, as Robinson reads him, Gareth held that ‘God’ names a certain set of ‘phenomena in our experience’. But later on Robinson says, ‘according to Gareth – and to Phillips – “God” is not a name or a referring expression, so it does not and cannot refer to anything’ (“Reply”, p. 9). Being unclear on the matter of Robinson on namesFootnote ¹, I propose we simply set aside (4), and replace it with

1. (5) What it would take for God to exist is for a certain person or thing to exist.

I think it's safe to assume Gareth would have denied (5), while Robinson obviously affirms it. Well now, does the rejection of 5 commit you to the equivalent of atheism? I suggest it does not. To see why, we need to understand that Gareth had two perfectly good, entirely orthodox Christian reasons for denying (5).

The first emerges from any thoughtful reflection on the doctrine of creation ex nihilo. The opening words of the Nicene Creed as they appeared in their earliest translation into English (in the Prayer Book of 1549) say

It's worth noting that the English translators felt the need to insert the count-noun ‘thinges’ where the original languages have none. (The Latin reads ‘factorem … visibilium omnium et invisibilium’). Now it is surely analytic that whatever this credal sentence says is something that orthodox Christians are bound, on pain of heresy, to believe. But the sentence, and the doctrine of creation ex nihilo it enforces, seem to have a pretty straightforward logical or, if you like, ontological consequence. Since God made all things – that is, since God is the creator of absolutely everything, of all that is, both seen and unseen – it seems to follow that God cannot logically or ontologically be a thing of any kind. Ergo, God cannot be a person or a thing, if the words ‘person’ and ‘thing’ are taken to be kind-terms whose extensions fall in the class of beings or entities named by the venerable expression ‘all thinges’. So if according to orthodox Christianity, God the creator is wholly increate, it seems to follow that God cannot to be numbered among the sum of all things. One cannot be both the uncreated creator of ‘all thinges’ whilst being counted among them.

I said Gareth had a second compelling orthodox Christian reason for denying that God is an invisible person or thing. This has to do with Christianity's proscription of idolatry. If God were to be any kind of thing, if God were discoverable among the sum of absolutely everything, then our worship would perforce be directed towards a creature. And it is central to orthodox Christianity that to worship any created thing is to fall into idolatry; it would be to worship a false God.Footnote ²

I hasten to add that Gareth was far from being alone amongst orthodox Christians in rejecting (5). At the time he was working on Believing in God, a certain position in philosophical theology was very much in the air at Blackfriars. Denys Turner, who ranks among its ablest contemporary defenders, calls it ‘Herbertical Thomism’, naming it for its redoubtable founder, Herbert McCabe. Herbertical Thomists take their lead from two of St Thomas's cardinal teachings. First, Thomas taught that we cannot know what God is, only what God is not. And second Thomas said that God cannot be placed in any kind whatsoever: there no such thing as the sort of thing God is. For these reasons, Herbert frequently and cheerfully attacked the view articulated in (5). A fine example is what he said in an article published in New Blackfriars in 1980, one that Gareth is almost certain to have read:

Allow me to make two points. One is that Robinson affirms, while Gareth and the Herbertical Thomists deny, (5). Second, and more importantly, Robinson appears unwilling to countenance the possibility that any orthodox Christian could deny (5). Even worse, Robinson appears unwilling to countenance the possibility that any orthodox Christian might have good reasons for denying (5). But since actuality entails possibility, Robinson is shown to be wrong on both counts.

At this juncture, a worry arises. Isn't it just plain common sense that if anyone says, as Herbert and Gareth manifestly do say, that

1. (6) God is not a thing.

It must follow that they are thereby committed to

1. (3) There is no God?

Perhaps both Gareth and Herbert, all fired up by thoughts such as (6), lapsed inadvertently into atheism. Here some help from Quine would not go amiss. Quine said that to be is to be the value of a bound variable. (A bound variable is one preceded by a quantifier in a well-formed sentence) The Quinean dictum is admirably reticent on what it would take for anything to be anything, other than for it to licence truthful statements of the form

1. (7) There is an x such that x is F.

And I propose that what both Gareth and Herbert (together with all orthodox Christians) want to say is captured by

1. (8) There is an x such that x is God.

I see no reason at all to think that affirming (8) is anything but fully consistent with affirming (6). Doing so would, after all, be on all fours with affirming both

1. (9) The square root of -1 is not a thing,

and

1. (10) There is an x such that x is the square root of -1.

So (6) does not, after all, and contra Robinson, commit you to atheism.

I shall now revisit what I called in my article Robinson's ‘indirect evidence’ that Gareth was the equivalent of an atheist. A better term might be ‘circumstantial evidence’, for it comes down to this: Gareth, deeply influenced by Wittgenstein, worked in the philosophy of religion, so he must have been a ‘radical Wittgensteinian’. And ‘radical Wittgensteinians’ are all of them atheists. Obviously, this notion of being a ‘radical Wittgensteinian’ deserves closer scrutiny. As I read Robinson, he uses the term to mean a Christian philosopher of religion who embraces two fundamental doctrines.

The first doctrine we might think of as a form of eliminativism. ‘Eliminativism’ originally named a position in the philosophy of mind taken by hard-line materialists such as the Churchlands. Most of us would likely include on our list of what there is, such things as our minds and thoughts. After all, we change the former and we think the latter. In doing so, we would be treating minds and thoughts as things that exist. But, say the Churchlands and their materialist friends, there are in fact no such things, there being only brains and neurons, and the various kinds of physical states that brains and neurons can get themselves into. Philosophers who say this are unerringly frank about their aim, which is to eliminate certain items from our ontology all together. This makes theirs an essentially ontological project. So, to be an eliminativist in the philosophy of mind is to advance the claim, ‘There are no minds, there are only brains and neurons and so on’. Generalizing, we can say that a philosopher counts as an eliminativist about an entity F, or a given kind of entities, the Fs, just in case she advances a claim of the form, ‘There are no Fs, there are only Gs’.

We can now say that Robinson's ‘radical Wittgensteinians’, being God-eliminativists (or should it be ‘eliminate-God-ivists’?) affirm the claim

1. (11) There is no God; there is only the observable world of people, medium sized dry goods, and other material objects.

The second doctrine that Robinson ascribes to his ‘radical Wittgensteinians’ is semantic. It arises from an obvious problem confronting any God-eliminativist who wished nevertheless to continue using Christian language and leading a Christian life.Footnote ⁴ How could a God-eliminativist coherently do such things? The answer, says Robinson, lies in the ‘radical Wittgensteinian’ insistence that religious language is entirely expressive. In particular, according to Robinson, it's a matter of holding that the subject-terms used in Christian discourse (such as ‘God’, ‘grace,’ ‘the Holy Spirit’, and so on) are entirely non-referential. And the ‘radical Wittgensteinians’ apparently go on to add that Christian sentences that purport to make true claims such as, ‘God loves us each and every one’, are nothing but expressions of spiritual attitudes and injunctions to practice religion. As a result, the proper logical form of such a sentence would presumably be given by something like, ‘Attention each and every one of us! Engage in religious practices which make us feel infinitely loved!’ In consequence, although many sentences containing the word ‘God’ appear to be in the running for being true or false, they are, by ‘radical Wittgensteinian’ lights, entirely free of cognitive content.Footnote ⁵

Now we need to ask: Is there any reason to think that Gareth embraced this pair of doctrines? Let's start with the first. Robinson accuses Gareth of what boils down to God-eliminativism on this basis of Gareth affirming

1. (6) God is not a thing.

I have said enough, I hope, to show that Gareth's affirming (6) in no way committed him to the belief that

1. (11) There is no God; there is only the observable world of people, medium sized dry goods, and other material objects.

But I would add that your typical eliminativist makes no secret of his ontological predilections. The Churchlands positively trumpeted theirs from the rooftops. Or take another kind of eliminativist, the good bishop Berkeley. He, of card-carrying idealist fame, was an eliminativist about material objects. But Berkeley never shied away from telling us in the plainest of plain English that there is no such thing as matter, there are only sensible objects whose esse is percipi. If Gareth really was an eliminativist, why would he have played his cards so close to his chest? Why did he not come right out and assert (11)? Surely it is more charitable to think – not to mention just plain true to say – that it never crossed Gareth's mind to assert such a thing as (11)?

Among Robinson's other reasons for attributing God-eliminativism to Gareth is the fact that Gareth drew a parallel between Wittgenstein's treatment of pain and his own treatment of God. Robinson apparently takes Wittgenstein to have been an eliminativist about the mental. Ergo Gareth must have been an eliminativist about God. In saying this, Robinson seems to attribute to Wittgenstein a series of claims such as

1. (12) There is no such thing as pain, there is only pain-behaviour in the right kind of circumstances.Footnote ⁶

It really would be absurd for anyone to assert (12). Confronted with such a proposal, one immediately wants to ask: ‘But surely, among the circumstances that make it appropriate to say that someone is in pain is that they are actually in pain? How could there be no such thing?’ Robinson thinks that Wittgenstein was a bit of a fraud. Perhaps so. But only a buffoon could say a thing such as (12).

Furthermore, the suggestion that Wittgenstein was an eliminativist about the mental in general, or about pain in particular, flies in the face of what he actually said. To take one instance, in §304 of the Investigations, Wittgenstein imagines somebody saying to him, ‘But surely you will admit that there is a difference between pain-behaviour accompanied by pain and pain-behaviour without any pain?’ To which he replies, ‘Admit it? What greater difference could there be?’Footnote ⁷ In point of fact, far from embracing eliminativism about pain, Wittgenstein meant to be rejecting a particular philosophical account of it. This was the conception of pain as a logically private object about whose existence only the subject could be certain. Hence, to the extent that Gareth was modelling his treatment of God on Wittgenstein's treatment of pain, Gareth could not have been an eliminativist about God.

That leaves the question of Gareth's alleged strong expressivism, the idea that religious sentences are never in the running for being true or false. Notice that if I am right about the eliminativism, then Robinson is batting on a rather sticky wicket in insisting that Gareth was a strong expressivist. This is because the only respectable reason anyone might have for being a strong expressivist about a given region of our thought and talk would have to be their own prior commitment to a relevant form of eliminativism. Ayer was an eliminativist about moral entities and moral properties. Given his radical empiricism, he denied there were any. This in turn motivated his commitment to emotivism. He had to account for the fact that we say things such as ‘lying is wrong’, thereby apparently predicating a moral property of a moral kind. But if Gareth would have eschewed (11), what philosophical motive could he have had to say that religious sentences are never in the running for being true or false?

Be that as it may, in his reply to me, Robinson supports the charge that Gareth was a strong expressivist with two lines of argument. The first is that, according to him, Gareth denied that ‘God’ was a referring expression. It seems that Robinson got the impression that when I said that Gareth denied that ‘God’ was a name, I meant that Gareth was denying that ‘God’ was referential.

In response, I want to say three things. First, of course I agree that the game is up if Gareth said or implied that ‘God’ has no reference. Only, second, Gareth never said it. Nor did he say anything that remotely implied it. In my original article, I was genuinely puzzled by Robinson's claim that, according to Gareth, ‘Both “poltergeist” and “God” are essentially names for absences, not for real agencies’ (p. 356). How, I wondered, might you go about naming something that was necessarily (and not just contingently) absent (in the way that Gareth said that God – being invisible, intangible and so on – was non-contingently absent from sense experience)? In a spirit of interpretative charity, I thought: Well, Robinson must mean that Gareth believed that ‘God’ is shorthand for one or more definite descriptions. That was the point I was making in speaking of how one might refer to a hole. I then went on to say what I shall say again: Robinson is dead wrong about the names bit. Gareth insisted that ‘God’ is not the name of anyone or anything, not even the name of an absence or a set of phenomena. From this Robinson inferred that I took Gareth to be saying that ‘God’ does not refer. I did no such thing.Footnote ⁸ And in point of fact, there is decisive textual evidence against Robinson on this score. It can be found in Gareth's discussion of the parallels between God and the equator, similarities that Gareth found very illuminating. In each case, he says, we come to believe in something (the equator, God) not on the basis of empirical evidence (by inferring their existence from other things we see) but simply by being taught how to use the words (‘the equator’, ‘God’). But in thereby coming to believe in the existence of the equator or in the existence of God, we arrive at a belief in the existence of something invisible, intangible and so on.

Enter Gareth's man, Otto. Inclined towards scepticism about the equator, he might say: ‘Surely there is no such thing as the equator. There is no evidence for its existence, and no one can ever point to anything and say: “See, that's the equator over there”’. And obviously Otto is here acting as a foil for an atheist such as the younger Flew. For that's exactly what Flew implied in his parable of the gardener: ‘Surely there is no such thing as God. There is no empirical evidence for God's existence, and no one could ever point to anything and say: “See, that's God over there”’. In the book, in replying to Otto's concerns, Gareth said, speaking in his own voice, ‘And yet of course the equator does exist, and it is invisible. And it is not an accident that it is invisible; it is part of our concept of the equator, belongs to the use of the word “equator”’.Footnote ⁹

The implication is that one good way to reply to the Flews of the world would be to say, ‘Of course God exists, and yet God is invisible, intangible and so on. The whole point is that it's no accident that God is invisible, intangible and so on – it's part of the very concept of God’.

But notice that in order for Gareth to make this reply to a sceptic about the equator, Gareth must have been using the expression ‘the equator’ to refer to nothing other than the equator, the very thing whose existence he is insisting on in this passage. He never comes close to saying, in response to Otto, ‘Ah, your problem is that you have misconstrued the logic of the expression “the equator”. It is not a referring expression at all’. Similarly, in order for Gareth to respond to Flew in the way he does, he has to be using ‘God’ to refer to God, the very God in whose existence Gareth takes himself to believe. He never comes close to saying to atheists like Flew, ‘Ah, your problem is that you have misconstrued the logic of the expression “God”. It does not refer’.Footnote ¹⁰

A final remark on truth. I argued that Gareth should be construed as a cognitivist about appropriately formed Christian sentences. My point was that Gareth took utterances such as ‘God loves us each and every one’ to be in the running for being true or false, that is, to be strings of words that are more than mere expressions of (non-cognitive) attitudes. In his reply, Robinson said that Gareth was actually following Strawson in cleaving to expressivism about truth itself, thereby construing uses of the word ‘true’ as ways of simply expressing one's (non-cognitive) agreement with some relevantly asserted sentence. I find this suggestion of Robinson's highly unlikely simply because of Wittgenstein's influence on Gareth. When it came to the question of truth, the Wittgenstein of the Investigations pretty clearly followed the Frege-Ramsey line. This is sometimes called the disquotational theory of truth, being one in which sentence-schemas of the form

lie at the heart of the story.Footnote ¹¹ Not only does the Frege-Ramsey view of truth sit comfortably with what I called moderate expressivism, in point of fact, I attributed the latter to Gareth on the assumption that he must have endorsed the former.Footnote ¹²

At the outset I said that we owed it to Gareth to assume his innocence, and that, unless it could be shown beyond reasonable doubt that he held one or another view that was the equivalent of atheism, Gareth deserved to be acquitted of the charge. In a court of law, if the defence can provide an innocent explanation for all the evidence produced by the prosecution, that explanation in itself constitutes reasonable grounds for finding the accused to be innocent. I have just provided such an explanation for all of the evidence Robinson adduces to show that Gareth's views were ‘equivalent to atheism’.

With that, I hereby move to acquit.