We must therefore now ask, is the question 'Why anything?' legitimate? Is it, as some say, less a legitimate question than a question-begging question, a philosophically disguised version of the 'wife-beating' question? For it would seem that those who wish to deny the legitimacy of the question do so because they too assume, as I do, that if you may legitimately ask it, then it has to have an answer. If it has an answer, then the name of the answer would have to be 'God', for the answer would bear the name of the 'Creator' of all things, visible and invisible, 'out of nothing'. Of course, I should say that if 'God' is the name of the answer, then, though the question is intelligible to us, the answer could not be - but the atheistic opponent would say that it is just because the answer could not be intelligible to us that the question lacks sense. To which I would respond: if the question makes sense then the sense it makes requires that the answer must lie beyond our comprehension. But that does not settle the matter, for the atheist will still demand to know why it is a question which
I am compelled to ask, and so am constrained thus to answer. Even more, why should I be required to concede that the question makes sense at all?
And, of course, among those for whom the question does not make sense is Bertrand Russell, who maintained on a famous occasion in discussion with Frederick Copleston on the BBC Third Programme that all you can say about the world, however it has come about, is that it is 'just there, that's all'.1 There cannot be a question 'How come there is anything there?' because you could not give any account of the answer, the business of accounting for things belonging within the world; there is no question which can have a purchase on anything which might count as the cause of it. Clearly, that is something to be argued about as theist does with atheist. For Thomas it is an intelligible question, one the answer to which would bear the name 'God'.
But Russell's resistance to this argument-strategy seems to this extent justified. All these questions about items in the universe either have their
1 In The Existence of God, ed. John Hick, London: Macmillan, 1964, p. 175.
answers in terms of other parts of the universe, or else explain why we have the sort of universe we have rather than all the other possible universes:2 that is what it means to say that they have to do with the explanation of what there is, and so answers to the different question, 'Why is there this universe?' One bit explains another and we could grasp all the explanations required if we had enough information about all the parts of it. But Russell says that we cannot explain there being anything at all rather than nothing at all in that sort of way; at any rate we cannot explain the absolutely contingent fact of there being anything at all in terms of how so far we have used the word 'cause', because it is not possible to get the notion of 'cause', of which we know the meaning only from our experience in the world, to do any work, to carry any meaning when used of the universe itself. The statement 'x causes there to be something rather than nothing' collapses, Russell says, into nonsense: for neither the variable 'x', nor the verb 'caused there to be' can possibly bear any meaningful substitution when used of everything there is. Hence, it would seem that the question 'Why anything?' is question-begging in the wife-beating manner. For to suppose that there is such a question to be asked about everything is already to suppose that it would make sense to speak of a 'cause' of everything; and it is Russell's view that 'cause' cannot bear any such sense. In this conclusion, of course, Russell shares common ground with Kant, for whom 'so employed, the principle of causality, which is only valid within the field of experience, and outside this field has no application, nay, is completely meaningless, would be altogether diverted from its proper use'.3
It may seem surprising to some theists that Thomas was rather more in sympathy with Russell's view of this than with the imprudently optimistic Christian apologist. At any rate he would have agreed with Russell to this extent, that the question 'What accounts for this cat?' is of a wholly different kind from 'What accounts for there being anything?' And he would have agreed with Russell's reasons: for if our ordinary causal questions about particular bits of the universe are answerable in terms of other bits of it, the trouble with the question 'Why is there anything at all?' is that it is in the nature of the question that you have run out of bits of the universe in terms of which to give an answer. And that is just another way of saying what I said earlier, namely that if nothing existed nothing would be missing, for there would be nothing for it to be missing from. Hence, in the same sense in which we can understand questions in the
2 Of course, one possible answer to that last question might very well be that all the possible universes actually exist. But even if that is true, there could still be an explanation of why that is so.
3 Kant, Critique of Pure Reason, B664, p. 528.
form 'Why is there this rather than that?', we cannot understand what would count as an answer to the question 'Why is there anything rather than nothing?'
Moreover, it is clear that for Thomas the question at the very least lies at the limits of logical oddity, and just how odd can be best understood from its eccentric syntax, from the curious logic of the 'rather than' or - and here we return for the last time to the question - of 'difference'. As we have seen, we can get such relational expressions going when we can supply symmetrical values for the variables p and q in the expression 'p rather than q': for example, 'red rather than green'. The 'rather than' has the force of an intelligible contrast because red and green are both colours, and so we know what they differ as. But what is to be made of the 'p rather than q' if 'red' is substituted for p and 'Thursday' for q? - for it would seem odd to consider what 'red' and 'Thursday' differ as. All the same, as I suggested in chapter ten, no created difference between two things, properties or descriptions, however diverse, is beyond all possible containment within some context which could make sense of how they differ, and since I happen to think of days of the week as having colours, in that context it makes perfectly good sense to contemplate the disjunction 'red rather than Thursday', though admittedly it is the rather special one in which Thursdays are blue. But eccentric as the 'rather than' has become in this case, an ultimate oddity is inflicted upon the 'rather than' if you substitute 'anything whatever' as a value for p and 'nothing' as a value for q. Has the 'rather than' any meaning left? Is it still intelligible? In a way, yes, it is intelligible, it has the force of a very radical sort of 'might have been', of an existential contingency pushed to the very limit. A thing which is red, like a letter-box in the UK, might have been green, as letter-boxes are in Ireland, but there are no doubt good reasons why they are red in the UK and green in Ireland, some prior states of affairs -which account for the colours they are - providing a kind of 'causal narrative'. But if we could imagine that rather than there being anything at all there might have been nothing at all, we have, indeed, some force of contrast going for this 'might have been' but not one to be accounted for in terms of antecedent states of affairs, no possible background context to make sense of it, no explanatory causal narrative, for a fortiori there is nothing left to account for the fact that there is something rather than nothing, no bit of the world there functioning to explain the existence of things, but only 'nothing'. And 'nothing', as we have seen Thomas to say, is not a peculiar sort of causally explanatory 'something'; it is not an antecedent condition; it is certainly not the cosmologist's 'random fluctuations in a vacuum'; neither, alternatively, is there some specialised theological sense which might give force to that sort of 'out of' which is 'out of nothing'; the 'ex' of 'ex nihilo' means, Thomas says, just the contrary: the negation negates the ' out of itself, as if to say, 'We have a making here, but no "out of",4 no antecedent conditions, so no process, no event; an "after", but no "before".' It is just for this reason that the notion of a ' cause of everything' strains at the lines of continuity with our ordinary, intra-mundane, explanatory employments of cause with a force such that, for Russell, it has there broken free of its moorings altogether. And if Thomas would resist taking the conclusion to that extreme, at least his sympathies, one would guess, are nearer to Russell than to the imprudently optimistic Christian apologist.
That said, for Thomas, Russell's conclusion is one step ahead of the case which supports it. For Russell's saying that 'that the world is is just a fact' is itself questionable as to its epistemological standing. When by way of answer to the question ' How is it that anything at all exists?' you say, as Russell does, ' It's just a fact', your answer is not itself an empirically ' factual' answer to a question concerning what the facts are, because you are not asking any sort of empirical, factual question. As Wittgenstein makes clear in the Tractatus, there is no possible sense of fact' in which ' that there is anything at all' can be a fact, Russellian 'brute' or otherwise. It is clear enough, even, or perhaps especially, on Russell's account, that ' that the world is' cannot be a ' fact'. For it is on his account that a ' fact' is, and can only be, what some true statement states concerning something in the world, for ' facts' need contexts to be facts in, and ' nothing' is not a context. Moreover, if not a ' fact', neither is ' It just is' an explanation, as if, accepting the reasonableness of the question 'What accounts for there being anything?', it offers the answer 'Nothing' by way of rival to the theistic answer ' God'. On the contrary, as we have seen, it is a refusal of legitimacy to the question itself; it is to say that there is nothing to the question, there is no question to answer. The question does not arise.
But the question does arise, at any rate in the sense that it is a question which gets asked; for human beings ask questions, they 'wonder at causes', as Thomas says,5 and ought not be stopped doing so prematurely. They ask Why anything?' with seeming intelligibility, so that if you are to rule the question out, that is to say, proscribe it, it would seem reasonable to ask on what grounds it should be refused legitimacy: you cannot without some grounds for doing so declare an end to discussion by ruling it out just by fiat, for that is to reduce the refusal to no more than saying, ' It's just a fact, and that it's just a fact is just a fact', which is to lapse into the merely assertoric. And for the matter of that, quod gratis asseritur gratis negatur. For which reason, if the world's existence is not just
4 ST 1a q45 a1 ad3. 5 Aquinas, In IMetaphysicorum, I, lect.1, 35.
a fact - because not a 'fact' at all - then neither is the question whether its existence does or does not need an explanation a matter of free evaluation, as if anything at all which is not a question of'fact' is a question of how you choose to view things. The question arises as a causally explanatory question - it has grammatically the same shape that demands for an explanation of events in the world have - and to that extent the question retains its lines of continuity with all the causally explanatory questions which lead to it. But its logical oddity lies in its self-cancelling character: for we know this much at least about what must count as its answer, that the bringing about of anything 'out of nothing' cannot be any kind of causal process such that any kind of causal law governs it, for it is not in any sense a 'process'. Hence, if, being a causal question, the answer to it must have the character of a cause, we have, in thus answering the question, lost control over the understanding of the causality involved.
It is, moreover, important to understand correctly what it is that we could not understand about the nature of the divine causality: our 'loss of control' is such as to make it irrelevant what kind of causal explanation is in question. It may very well be true that, as Swinburne says,6 the divine causality is best understood on the model of human, intentional, 'agent' causality, rather than on the model of efficient, natural, causality; and it may very well be true that, as Kerr says,7 Thomas's model of efficient causality is in any case nearer to our contemporary conceptions of'agent causality' than to that of a post-Humean efficient causality. But either way we would have to enter the same apophatic reserve in ascribing causality to God - and here Mackie appears to agree with Thomas. As he says, it is only 'by ignoring such key features [of human intentional activities as their embodiment, as their being fulfilled by way of bodily changes and movements which are causally related to the intended result, and so as having a causal history] that we get an analogue of the supposed divine action'.8 By the time you have performed the necessary apophatic surgery on this 'agent causation' as predicated of God, there is no more left to it than in the case of any other causality in need of surgical reduction as predicated of God. Indeed, the same is left, whatever one's causal model, namely whatever it is that answers to the causal question, 'Why anything?' Whatever our model of causality, we know that we do not know in what way God is a 'cause'. We know this not because we do understand what kind of cause God is, and so know that God is not a cause in any ordinary sense. On the contrary, it is because we know only what kinds
6 Richard Swinburne, The Existence of God, Oxford: Oxford University Press, 1979, pp. 130ff.
7 Kerr, After Aquinas, pp. 46-8. 8 Mackie, Miracle of Theism, p. 100.
of cause there are in creation that we have to concede the mind's defeat in respect of the divine causality. Whatever 'grip' theological language has on anything at all it has on the world, and on a question which arises about its existence, a question which is the expression of a kind of ultimate astonishment: it might not have been at all, and that it is has been brought about. That is not, and could not be, just a 'fact' in the world; nor is it just a fact about it. But if, as I have said, an atheist is to offer anything more to the debate than to brush the question 'Why anything?' aside with a merely rhetorical gesture of refusal, we should need to hear of a reason for denying its legitimacy.
And the main reason for doing so is, as we have seen, that the question is said to involve the circularity of the wife-beating question: that it presupposes an answer to what it purports to ask about. The question, I have said, is 'causal' in shape. And it is true that if you can sensibly ask a causal question about the world then you have presupposed that a causal answer concerning the world makes sense, and you have got God in one move; or at least you have got to a point where, given a number of subsequent moves, what is recognisably the God of Christian belief can be shown to exist. But, the argument goes, at least since Kant we have known that there is reason to doubt - if no more than that - that a causal answer concerning the world could make sense. You cannot, therefore, without circularity press the case for saying that a causal answer about the world does make sense on the strength of an assumption that the question 'Why anything?' is legitimate. In argument what is merely assumed has no strength at all.
The case for denying the question's legitimacy would therefore seem to rest on a general epistemological ground, namely that causal language does not, because it could not, have any application to the world's existence as such. And such a ground is clearly contestable in principle. After all, if it involves a circularity to rest a case for saying that causal language does have application to existence as such on the grounds that the question 'Why anything?' is legitimate, it is but to traverse the same vicious circle in the opposite direction to rest the case for causal language's not having application on the grounds that the question is not legitimate. Furthermore, if any presumption is to be made on either side, it is that the case against the question's legitimacy has to be made out in such general and essentially contestable terms as Kant's critical rationalism, given the prima facie reason for supposing its legitimacy. For prima facie - that is to say, other things being equal - there is nothing to be said for ruling out a question, unless there is an overwhelming reason in principle for doing so, if, as in this case, that question lies so obviously in continuity with the sort of intra-mundane causal questions human beings naturally persist in asking about the world. And this antecedent probability that the question is legitimate survives, notwithstanding the fact that the answer is known in advance to break the links of continuity with those less radical causal questions, placing the answer beyond the reach of our powers of comprehension. What is at stake here - in one way or another it is the central issue of this essay - is the conflict between two forms of theological 'agnosticism', that of Kant and that of Thomas Aquinas.
Was this article helpful?