But now we must entertain another sort of objection to the proposition that the argument-strategy of Thomas's proofs is shaped by the force of
12 In short, there is a genuine argument to be had, for example, with those who reject theism on the grounds of its incompatibility with the existence and quantity of evil in the world.
the 'Why anything?' question. The objection here is less the systematic objection to the legitimacy of the question in principle, than exegetical, concerning its relevance in the context of Thomas's arguments. It will be contested that the question 'Why anything?' is Leibnizian in origin13 and that it does not occur in situ where Thomas's arguments for the existence of God are formulated. And so it will be asked on what account of Thomas's actual text can my case for the strategically determining character of that question be made good. In answer to which objection it is necessary to take up two issues concerning the general character of Thomas's argument-strategy for a proof of God. The first issue is that if the question is not raised by Thomas in that form, on what grounds may one read his arguments for the existence of God in terms of it? And the second issue follows from the first: if you do read Thomas's argument-strategy in terms of the Leibnizian question, how is Thomas's procedure for proof of God to be distinguished from Leibniz's, superficially similar in general structure as they are? For both appear to argue for the existence of a 'necessary being' by inference from the contrasting contingency of the world - its existence rather than nothing - a contingency which appears to be what forces upon our minds the question, 'Why is there anything at all?' in the first place.
It is convenient to take up the second issue first, and Mackie's discussion of Leibniz's argument-strategy helps to clarify the distinction between his and Thomas's: the distinction would appear to lie in the distinct conceptions of contingency which underlie them. For Leibniz, Mackie says, a thing's existence is contingent if it depends upon how things are, such that, had things been otherwise, that thing would not have existed. Hence, by contrast, an existence is necessary if it is not dependent upon how things are, for it would exist whatever states of affairs obtained. Now Mackie argues that Leibniz's argument for the existence of God relies upon the premise that the world as a whole is contingent in this sense, but that this premise is not available: though we have some grounds for thinking that each part, or each finite temporal stretch, of the world is contingent in this sense upon something else, we have ... no ground for thinking that the world as a whole would not have existed if something else had been otherwise; inference from the contingency of every part to the contingency of the whole is invalid.14
Be that conclusion as it may, the second account of contingency maintains that a thing's existence is contingent if and only if it might not have existed,
13 See 'On the Ultimate Origination of Things', in G.W. Leibniz: Philosophical Writings,
14 Mackie, Miracle of Theism, p. 84.
and a thing's existence is necessary if and only if it is not the case that it might not have existed, and this account of contingency and necessity appears to be Thomas's - as employed for example in his 'third way'. But, as Mackie says, to argue on this account of the contingency of the world as a whole to the existence of a necessary being requires showing that there has to be a necessary being, a proposition with which Thomas undoubtedly agrees. Of course, of such a being, if it exists, it has to be the case that its existence is necessary. But, Mackie argues - and here again Thomas agrees with him - it does not follow from the meaning of 'necessary being' that the proposition 'A necessary being exists' is a necessary truth. A thing's character of being 'necessary' is not, and cannot be, any part of an argument that it does exist, and, as Thomas says, Anselm's so-called 'ontological' argument, which appears to rely on this fallacious inference, must for that reason be invalid.15
It was Kant's contention that all forms of cosmological argument for the existence of God, including Leibniz's, must fail in virtue of their reliance on this fallacious inference underlying 'ontological' arguments.16 But Kant's argument that all cosmological proofs must presuppose the invalid inference from a thing's character of being a necessary existent to the necessity of its existence would need to be investigated further, in view of Thomas's explicit disavowal of Anselm's fallacy in the article of the Summa which immediately precedes that in which the 'five ways' are expounded.17 At any rate it is clear that for Thomas 'God exists' is a contingent truth, for its denial, though false, is not a contradiction; for, as we have seen, a thing's actuality (esse) is never derivable from any characterisation of its kind.18 Indeed, it is precisely for this reason that God's existence cannot be demonstrated from any account of what God is and that any demonstration of God's existence would have to argue 'from the world' and not by the 'ontological' strategy. Such a demonstration will show that 'God exists' is true; and, he thinks, God's necessary existence can itself then be further demonstrated. Likewise, that there can be only one God can, he thinks, be demonstrated, even if (as we have seen) it would not be contradictory to think, as Hume surmised,19 that there
15 ST 1a q2 a2 corp.
16 See Kant, Critique of Pure Reason, B 635-6. 17 ST 1a q2 a2 corp.
18 Incidentally, this suggests a further difference between Thomas and Kant, for Kant rejected the ontological argument on the grounds that 'existence is not a predicate'. As we have seen, Thomas maintains that existence in the sense of 'actuality' is a predicate, and, in that sense of actuality, is not reducible to that sense of 'exist' as the existential quantifier ('there is an x') which, of course, Kant and Thomas agree is not a predicate (as also Mackie, see Miracle of Theism, p. 83).
19 David Hume, Dialogues Concerning Natural Religion V, ed. and introd. Norman Kemp Smith, Oxford: Clarendon Press, 1935, pp. 207ff.
might be many gods answering to a causal question about the world's existence - for, as we have seen Thomas to say, logically 'God' is not a proper name, but a nomen naturae, and so possesses a logically proper plural.20 Thomas, for sure, thinks it false to say there are many gods; and he thinks that it further follows (easily enough) from a proof of God's existence that there could only be one God, since neither any one of a team of gods taken severally, nor any team of gods taken collectively, could stand as a good answer to the question, 'Why is there anything at all?' In any case, it is clear that, in so far as Kant is right that cosmological arguments such as Leibniz's depend upon the invalid inference of the ontological argument, Thomas's proofs are to be differentiated from that of the eighteenth-century rationalist philosopher. Thomas's, at least, do not so depend. Thomas's 'Why anything?' question arises from a sense of the world's contingency quite different from Leibniz's.
Which brings us to the second issue, now long postponed, concerning how to place Thomas's arguments for the existence of God within the structure of the exposition of the Summa Theologiae, such that there would appear to be good grounds for reading the five ways in terms of the 'why anything?' question. Milbank suggested that one of the reasons for doubting that the 'five ways' presented in Summa Theologiae 1a q2 a3 could have been intended to be full-blown apodeictic proofs is their manifestly cursory character. Moreover, if it mattered to Thomas as much as I claim it did that rational proof is possible, it is said21 again that he could reasonably have been expected to provide us with more than the compressed, elliptical and, on his own standards of argument, clearly insufficiently articulated, schemata for argument that we find at the outset of his Summa. And it is true that, as set out in that work, the 'five ways' look more like an aide-memoire for proofs than the adequately complex and full-blooded exposition which would be required to be convincing as 'stand-alone' arguments. Nonetheless, there may well be a practical reason for this which has to do with the pedagogical purposes of the Summa Theologiae and one not bearing the weight of theological significance which Milbank attaches to Thomas's brevity. This vast Summa is often represented as a beginner's manual, intended for students setting out on the process of theological learning, and so it is - as Thomas says in the general Prologue to the work, he intends it ad eruditionem incipientium. But if this is its general purpose it is perhaps better to see it as a manual for the teachers of beginners than as a textbook for the use of beginners themselves, for
21 For example, by my colleague in the Cambridge Faculty of Divinity, Dr Anna Williams, in private correspondence.
that general Prologue makes clear that the revisionary purpose his work is meant to serve is principally curricular in character,22 that of setting out the ordering of the questions in a manner which is more coherently teachable than is done in the standard teaching texts available at the time; presumably Thomas has in mind the unhelpfully haphazard structure of Peter Lombard's Sentences. But if that is its purpose, then it is perfectly understandable that Thomas should be more concerned with argument structures than with the detailed exposition of the arguments themselves, for he could expect teachers to fill in those details which would be beyond the knowledge and experience of beginners. Hence, even if, taken on their own, the expositions of the 'five ways' are unconvincingly abbreviated, it is perhaps fairer to read them more as amounting to heads of argument, outlines for a five-fold argument-strategy, than as arguments proper in which would be set out in full exposure the connective tissue of inference, suppressed as it is in the text as it stands.
Secondly, it might further be conceded that in any case Thomas does not see the validity of the particular proofs he sets out in Summa Theologiae 1a q2 a3 as having any crucial role to play within the theological and methodological architectonic which is the Summa as a whole. This is not to say, however, that the arguments thus elliptically set out are not valid proofs, or at least proof-structures, or that Thomas did not consider them so; only that the question whether they are valid or not does not matter, as it were, from the point of view of the Summa's 'architecture'. But that, in turn, raises the question what the 'architectonic' considerations are within which the 'five ways', and their relatively cursory character, fit.
We derive one kind of answer to this question if we suppose, as I have supposed throughout this essay, that what matters to Thomas from the point of view of faith, and of its theological articulation, is not that any particular proofs of the existence of God are known to be valid, but that the possibility of such proof is not denied on grounds of faith. If that is so, then it might very well be thought sufficient to the purpose for a theologian merely to sketch the outlines of the sort of argument which
22 'We have come to the conclusion that beginners in this learning are in multiple ways handicapped ... in part because of the multiplication of pointless questions, articles and arguments; in part also because what such [students] need for the purpose of acquiring knowledge are not transmitted in accordance with the discipline's [own proper] ordering, but as required by [the ordering of] books commented on, or as the opportunities for disputation dictate; and partly because of the frequent repetition, the tedium and confusion which [those books] generate in the minds of their hearers.' - 'Considerav-imus namque huius doctrinae novitios . . . plurimum impediri: partim quidem propter multiplicationem inutilium quaestionum, articulorum et argumentorum; partim etiam quia ea quae sunt necessaria talibus ad sciendum, non traduntur secundum ordinem disciplinae, sed secundum quod requirebat librorum expositio, vel secundum quod se praebebat occasio disputandi; partim quidem quia eorundem frequens repetitio et fas-tidium et confusionem generabat in animis auditorum.' STPrologus.
could be considered to do the theological job required, and what would not: Anselm's 'proof will not do,23 and no good theological purpose is served by supposing that it will. But the arguments presented as the 'five ways' illustrate how and where human reason unaided by grace serves a theological purpose, exercising its powers, as reason characteristically does, in formal demonstration of God from creatures. But there is another kind of answer, more positively supportive of the reading of the 'five ways' as intended to be formally valid demonstrations, and that answer involves conceding that their probative value was never intended to be read off from the abbreviated text of Summa Theologiae 1a q2 a3 taken on its own.
For there is much to be said for the view that the article containing the 'five ways' is intended merely as the point of entry into a circle of highly complex argumentation which is completed only with the discussion of God as Creator, some forty-three questions further on in the Prima Pars, and that the reader is not expected to read the 'five ways' as having any validity as proofs except within that complex circle of argument taken as a whole. On one score at least this is clear. As I said in chapter 1, the demotic optimism of the five times repeated refrain, et hoc omnes dicunt Deum, cannot be conceded with any degree of plausibility at least until it has been shown, to any reader at all, Christian or otherwise, that what the 'five ways' have shown to exist is also extensionally equivalent to what 'all people' believe God to be, namely the Creator: and we get no argument to this effect until question 44. Moreover, even that argument will be insufficient to convince the Christian believer until it is shown that, whether as 'first cause' or even as 'Creator', the God thus demonstrated to exist is extensionally equivalent to the trinitarian God of Christian faith; and the demonstration of that is not begun until question 27. In short, if Thomas does think, as I believe he does, that a principal methodological purpose of the 'five ways' is to articulate a primitive meaning for the ratio Dei, constituting, as I argued in chapter 2, the formal object of sacra doctrina itself, then, from that point of view alone, the arguments presented in question 2 article 3 could not be fairly conceived as intended to 'stand on their own' for any good theological purpose whatever. They serve the purpose that rational proof may serve only within that wider philosophical and theological context.
Was this article helpful?