Does God exist? Apparently not

"Is there a God?" asks Aquinas, and answers, "Apparently not". This method is, of course, the one he follows throughout the Summa, a method derived from the intellectual practice of the quaestio, of which we saw a little in the first chapter. In this, as in every case, St Thomas first presents objections, the strongest arguments he can find against the thesis enshrined in the formulation of the quaestio, or against the answer which he is going to adopt. Usually he presents us with three objections, but in this case he presents only two: two arguments which are still regarded as the most successful arguments against the existence of God. They can be labelled "the argument from evil" and "the argument from partial explanation" or "the argument from science".

Aquinas expounds the argument from evil as follows: "Apparently there is no God. For if one of a pair of contraries were to be infinite, the other would be completely destroyed. But in the word "God" is included the notion that God is an infinite good. If there were a God, then, there would be no evil to be found. But there is evil to be found in the world. Therefore there is no God."

It is worth noticing that, as St Thomas renders this argument, it is of a form that has already been discussed. It is an attempted demonstration of the non-existence of some supposed entity, on the basis of some features of the significatio which can be attributed to the name of that thing. We have seen St Thomas using this technique himself to demonstrate the non-existence of vacuum, of empty space, and discussing (and eventually rejecting) its use to demonstrate the non-existence of place. Equally, it will be remembered that Lavoisier allegedly used the same technique to demonstrate the non-existence of phlogiston. It is clearly a valid method of reasoning. It is also clear that for this kind of argument to work, just as for Aquinas’s own arguments in favour of the existence of God to work, there has to be a solid measure of agreement between the parties to the discussion on the signification of the word in question. In this case the one arguing against the existence of God has taken up a notion which is part of the signification of the word "God" in what is called "classical theism": that God is infinitely good. The defender of the existence of God, in a context in which classical theism, or a religious tradition which includes classical theism, is the norm, is in fact not likely to deny this.

There are many different versions of the argument from evil, but they all follow the pattern which is followed by the argument St Thomas gives. This is roughly as follows:

1. The existence of God and the existence of evil are incompatible.

2. Evil exists.

3. Therefore, God does not exist.

In so far as various forms of the argument from evil differ, it tends to be with regard to different kinds of evidence given for the first premiss. This is entirely appropriate. The argument is apparently of valid form; thus, if the premisses are true, the conclusion must be true. The defender of the existence of God has to attack the truth of one of the premisses.

The second premiss is all too obviously true. There are, indeed, religious and philosophical groups both in East and West who hold that the existence of evil is only an illusion. The claim appears to be confused. If they mean that if we could behold reality without confusion and without error, we would be free from evil, it is at least possible, though perhaps unlikely. If they hold that evil does not exist and there is only an erroneous (and therefore evil) appearance of evil, they stand self-contradicted.

When we claim that evil exists, of course, we are not claiming that evil is an entity or a reality. The existence we attribute to evil is existence in the sense of the true, esse ut verum, the kind of existence which we can attribute to blindness and other similar privations. All we need to maintain, to support the claim that evil exists, is that various people, things, actions or occurrences are bad: and this is obviously true. The second premiss is thus unchallengeable.

Thus discussion has to centre on the first premiss, that the existence of God and the existence of evil are incompatible. This can be variously supported. Aquinas suggests a metaphysical justification for it; a more common line nowadays would argue on more moral grounds. We would not claim that some man we knew was good if he had power to stop certain evils which he knew about, and did not do so. On the view of classical theism, God is supposed to be good, all-powerful, and all knowing. This appears inconsistent, since it appears that God does not stop evils which he knows about (being all-knowing), and which he has power to stop (being all-powerful). Therefore the existence of a good God such as classical theism supposes, and the existence of evil, are incompatible.

There are objections to this formulation of the argument from evil which query the notion of God’s almighty power or universal knowledge. But these defences do not seem of much value. Sometimes they go so far as to abandon traditional elements of classical theism, which seems a desperate move. Moreover, as Geach has pointed out, the argument from evil still stands even if we leave aside questions of God’s power and knowledge, and concentrate simply on the fact that God is the Creator of the world. Since the Five Ways profess to demonstrate the existence of a Creator of the world, such a form of the argument from evil would certainly overthrow the Five Ways. The form which St Thomas gives, in any case, does not depend on the notions of God’s power and God’s knowledge, but solely on that of God’s infinite goodness, a notion of which he himself makes use in the context of the Five Ways: at least in the Fourth Way, if not in the others. He cannot even defend himself, then, by claiming that God’s power and knowledge are further claims which he has not yet made, and should not be supposed in advance of his proof that there is a God.

It is sometimes suggested that the existence of evil and the existence of God are not incompatible, because evil proceeds not from God but from human free will. This "free-will defence", as it is called, appears ineffective as it stands. There seem to be innumerable kinds of evil which do not depend on human free will: an obvious example would be those sufferings of animals and of infants which are caused by natural phenomena. Moreover, the fact that an action proceeds from the free will of a creature does not mean that it does not also proceed from God. If human free actions did not proceed from God they would not be created. Creation, as we shall see later, is not so much something God once did, as something that God is continually doing: it is more like a performance than a production. There is nothing in the world, then, that is not God’s action. For some forms of the free-will defence, human actions would have to exist independently of God’s creation, and this is inconsistent with classical theism in general and with the thought of St Thomas in particular.

Nevertheless, there is an argument, related to the free-will defence, which may be consistent with the thought of St Thomas. What is objectionable in the free-will defence is the doubt it seems to throw on the notion that God is the cause of the world as a whole and of everything that is in it. However, we may derive from St Thomas’s account of the way in which per accidens existents are caused, an argument which may suggest that God is not the cause of everything in the world in exactly the same way. This argument does not throw doubt on the notion that God is the cause of the world, and of all that is in it, including the evil that exists: it only questions the mode in which we can attribute to God being the cause of evil.

We have seen that we should not attribute reality, real esse, to evil. But if St Thomas is right in his suggestion of what I have called the "derivativeness" of esse ut verum, there should be some reality on which the esse ut verum of evil is based. In general this is not difficult to see. Evil, like blindness or any other privation, needs to have some really existent subject. But the presence of some defect or privation in some really existent subject is not a per se existent but a per accidens existent. The real existent from which the existence of evil derives is thus a per accidens existent.

This point may be of value. For St Thomas, as we have already seen hinted, that which exists per accidens exists as the result of the coincidence of two different strands of explanation. The example he gives, following Aristotle, is that of a man digging in a field who finds a treasure. There is no doubt a reason why the man is digging at that point in the field: perhaps it seems to him the best place for a drainage ditch. No doubt there is also a reason why there is a treasure buried just there: perhaps the person who buried it thought that it would be easy for him to find there and difficult for anyone else to find. Once these two lines of explanation are given — a reason why our hero is digging at point X, and a reason why there is a treasure at point X — sufficient reason has already been given why this man should dig at point X and find a treasure. We do not need to seek a reason why this-man-should-dig-at-point-X-and-find-a-treasure, considered as a single phenomenon, as if it were a per se existent which requires a single explanation.

Similarly, though everything which happens happens by God’s will, it is not necessarily true that every description under which an occurrence falls is a description under which that occurrence is willed by God. It is sufficient that there should be some description, even a per accidens one, under which it is willed by God. Obviously, if God knows everything, God knows every description which is true of every occurrence which he wills. But so do I know many descriptions which are true of the occurrences which I will. As Kenny points out, when I walk across the field to the river I do a certain amount of damage to the grass and wreak a certain amount of havoc among various micro-organisms which live in that habitat. I am not so ignorant as to be unaware of this: but though I will to walk across the grass, and I know that I thus damage certain micro-habitats, is it therefore true that I will (in any interestingly strong sense) to damage the micro-habitats?

To this it might be objected that my will is not, as God’s is supposed to be, unlimited. I cannot be supposed to will everything which I do. The objection is unsound. God’s will is unlimited because God’s will extends to everything in the world, as mine does not. But this does not mean that God’s will extends to everything under every description which is true of it. It is a common teaching among traditional believers in both Christianity and Judaism that the text of Scripture is inspired. Every word, and therefore every letter, is, as it were, written by God and willed by God: willed by God, indeed, in a very special and direct way in which other ordinary occurrences in the world, though willed by God, are not. Nevertheless, most Christians and many Jews have regarded as misguided and even superstitious those who have sought to use the dispositions of the letters in the Bible, taken according to some numerical pattern, as vehicles of special messages from God. The first letter of Scripture, say some of the Rabbis, is aleph, the first letter of the Hebrew alphabet; the middle letter of Scripture is mem, the middle letter of the alphabet; the last letter is tau, the last letter of the alphabet. Surely this is no coincidence? And surely it is no coincidence that these three letters spell ’emeth, truth? Well, it seems to me that it is a coincidence, and it certainly forms no part of traditional belief, let alone of classical theism, that it is not.

Even if this argument against the truth of the first premiss — that the existence of God and the existence of evil are incompatible — is regarded by some as doubtful, there is at least one other argument. The first premiss relies on the accepted truth of the claim made by classical theists that God is supremely good. But it is also a claim of classical theism that God is transcendent and inscrutable. The doctrine of God’s transcendence is that God surpasses everything which we can say of him. All our language about God will contain at least some misleading elements. If we say that God is good we seem to imply that God’s goodness is something distinguishable from God, as Socrates’s goodness is something distinguishable from Socrates, and thus that God is in some way complex or composite. If, on the other hand, we say that God is goodness itself, we seem to imply that God is an abstraction, as the word "goodness" in our ordinary language means something abstract rather than something concrete.

It is therefore highly dangerous to suppose that when we say that God is good we must be attributing to God the sort of thing we would be attributing to a human being when we say she or he is good. A good human being is brave and resistant to the temptations of pleasure, when in the pursuit of some good thing which is difficult of achievement. But God is not brave, as there is nothing that constitutes a threat or danger for God. God is not resistant to the temptations of pleasure: God has no temptations. God has no projects for pursuing some good difficult of achievement: nothing is difficult for God. A good man is kind to animals: but Geach has argued that there is no reason to suppose that a good God must be kind to animals. This is just as well, as all the evidence seems to be that the world God has made is a system which is simply indifferent to the sufferings of animals.

An appeal to the traditional belief in God’s transcendence is thus of some value in taking away the force of the first premiss of the argument from evil. What reason do we have for supposing that God’s transcendent goodness is incompatible with the existence of evil? We may indeed have a strong feeling that it ought to be, but strong feelings about what ought to be the case are notoriously a poor guide to what actually is the case, even as regards this world; a fortiori, they will be a yet poorer guide to what actually is the case as regards God.

We can add to these considerations those drawn from God’s inscrutability. "Inscrutable" is one of those curious words which outside a technical field — here, that of natural theology — is used only in a cliché, in this case a rather offensive one. Westerners traditionally regard Orientals, particularly Chinese and Japanese, as "inscrutable". I take it that this is because it is hard for a Westerner to guess what an Oriental is thinking from what he or she does or says, presumably because of the differences between Oriental and Western physiognomy, and those between culturally differing modes of expression in gesture and speech.

If one is stupid and tactless enough one need not go to the ends of the earth, and spend one’s time being rude about their inhabitants, in order to reach a grasp of the meaning of "inscrutability" in human contexts. I am myself extremely stupid and tactless, and I find it practically impossible to make out what my own countrymen and women — perhaps especially women — are thinking, on the basis of what they do or say, unless it is painfully and embarrassingly plain and explicit. Thus most people are for me, in the relevant sense, inscrutable.

However, I am not in other ways grossly stupid: I have managed to lead most of my life so far with a fair degree of what might be considered success, according to various criteria. Thus the difference between myself and my fellows is not really all that great. Nevertheless, it appears that the differences between me and my fellows is sufficient to have given rise to a high degree of inscrutability. The moral is fairly clear: the differences between God and any of God’s creatures are infinitely greater than the differences between me and my fellows, and so we have to presume an infinite degree of inscrutability in God.

If God is inscrutable, you cannot tell what God is thinking, or what God means, or even what God is like, directly from a contemplation of what God does or says. If I have often been mistaken about the goodness or badness of my fellow-creatures on the basis of my judgement about what they do or say, a fortiori any creature may be yet more grossly mistaken about the goodness of God, on the basis of his or her judgement about what God does or says. This conclusion is highly agnostic, of course: but I do not need here to offer more than an agnostic conclusion. I am trying to overthrow the confidence with which people feel they can assert that the existence of God is incompatible with the existence of evil. I am not here trying to prove that God is good: all I am trying to show is that there is no sufficient reason to believe that the goodness of God is incompatible with the existence of evil.

Indeed, this is all that needs to be shown. As Davies points out, on the basis of the argument from evil, as expounded above, we can form another argument which runs exactly on all fours with it.

1. God exists.

2. Evil exists.

3. Therefore, the existence of God and the existence of evil are not incompatible.

This argument is a mere logical transposition of the argument from evil given above. If the argument from evil is of valid form — as it is — then this argument will be of valid form too: that is, if the premisses are true the conclusion will be true. The two arguments have the second premiss, "Evil exists", in common: thus the only difference between them lies in their first premisses, "The existence of evil and the existence of God are incompatible" and "God exists". Which of these two premisses is true?

Here there is no remedy but to investigate the two premisses, and the arguments for holding them. We have just taken a look at the first premiss of the argument from evil, and have seen that it depends on the truth of certain views of classical theism; and we have also seen that the resources of classical theism are also sufficient to cast doubt on it. There is, I have claimed, no good reason for holding that the existence of evil is incompatible with the existence of God. Meanwhile, there may well be good reasons for holding that God exists: we are about to examine them in the Five Ways.

Clearly, it is important that the arguments for the existence of God which we give should be independent arguments: that is, they should not be arguments which depend on the non-existence of evil or on the goodness of God, understood in some common, non-transcendent way. Someone who held that God exists because this is the best of all possible worlds would be vulnerable to the argument from evil, because the existence of evil would at the same time give him reason to question the truths of the premiss on which his argument for the existence of God is based. But the Five Ways are in this sense independent of the existence of evil.

It is worth pointing out, in any case, with Davies, that the argument from evil seems in practice to work less as an argument for not believing in God than as a reason for not bothering to consider what evidence there might be for believing in God. The possibility of turning the argument on its head, as Davies turns it, shows that this is intellectually dishonest. The arguments for the existence of God need at least to be considered, and compared with those for believing that the existence of God and the existence of evil are incompatible. I have suggested that the latter arguments are weak: it remains to be seen whether the arguments for the existence of God are any stronger.

Aquinas, in his reply to the first objection, does not need to enter into the complications we have discussed. He couches the first premiss of the objection in strongly metaphysical terms, rather than in the quasi-moral or quasi-aesthetic terms which would probably be used nowadays, and in terms of which we sketched out the reasons for believing the first premiss. As a result the objection Aquinas gives is not vulnerable to the objections we have made: believing in it does not depend on ignoring the subtleties of God’s mode of causation of the per accidens, or God’s transcendence or inscrutability. It is, however, vulnerable to the reply which he makes.

"To the first objection, then, we have to say, as Augustine does in his Enchiridion, ‘God is supremely good: so no evil would be allowed in God’s works were God not so good and almighty as to be able to make even evil good’. It is proper, then, to the infinite goodness of God to allow evils to exist and to draw good from them." This is clearly a denial of the first premiss, that the existence of God and the existence of evil are incompatible. The argument for this premiss, as sketched out in metaphysical form in Aquinas’s objection, is that the very existence of something infinitely good is incompatible with the existence of anything which is evil. Aquinas simply denies this: God’s bringing good out of evil is evidence of yet greater goodness than would be the non-existence of evil. There is no reason to believe in the metaphysical version of the first premiss of the argument from evil than there is to believe in any other version.

The next objection which St Thomas raises, the second argument for not believing in the existence of God, can be called the "argument from partial explanation". His version runs:

"Apparently there is no God. (...) The second argument is, that which can be achieved by a smaller number of originating principles is not brought about by a larger number. But, apparently, everything we see in the world can be achieved by some other originating principles even if there is no God. This is because natural things can be brought back to the originating principle of nature, while things which come about intentionally can be brought back to the originating principle of the human reason or will.

So there is no need to suppose that there is a God."

As a matter of historical anecdote, it is perhaps worth noticing here the use of the principle which is (obviously incorrectly) called "Ockham’s razor": in Aquinas’s formulation, that which can be brought about by a smaller number of principles is not brought about by a larger number. Aquinas’s formulation appears to be a metaphysical profession of faith in a principle of explanatory economy, while the principle usually attributed to Ockham is rather a normative principle of methodology. The practical results are the same, though Aquinas’s formulation appears in some ways more radically minimalist than Ockham’s, strange as it may seem to the, alas, large number of contemporary English-speaking philosophers who accept the myth of Aquinas as a scholastic multiplier of unnecessary entities, and of Ockham as the forerunner of the glorious empiricist reductionism of our day.

But leaving aside anecdote and rhetoric, the objection is a serious one. The claim it makes is that the world does not in fact need an explanation. Since the Five Ways are an attempt to prove that the world requires an explanation in terms of a relationship to something other than itself, namely God, the argument, if valid, undermines precisely the kind of argument which Aquinas favours. The claim that the world needs no explanation, though, is one that needs disambiguating. Davies tells of a famous debate on the wireless, about the existence of God, between Copleston and Russell. Russell, perhaps wisely, attempted a move similar to this objection: he tried to undermine Copleston’s position by denying that there was any need to look for an explanation for the world. Copleston asked whether this meant that Russell regarded the world as "gratuitous", like some existentialist philosophers; but Russell objected even to the word "gratuitous", as it seems to suggest that the world might have been otherwise. He preferred to say "The world is just there, that’s all".

"The world is just there, that’s all" is not the claim being made in the objection which Aquinas offers. Aquinas’s objection claims that the world has an explanation, or a number of explanations: nature and reason are those given. A similar kind of objection might claim that the world is such that it explains itself, or that in some other way needs no explanation; just as (on Aquinas’s account) God is such that he explains himself, or needs no explanation. Russell, by contrast, claims that the world is "just there". I cannot see how Russell’s claim differs from a stark refusal to bother to think whether the world needs an explanation or not. One thing is to look for an explanation and fail to find one; another is to show why the world requires no explanation; yet another is to show how the world explains itself; yet another, to show that it can be explained by e.g. nature or reason. Russell explicitly refuses to do any of these things, and I do not think that one would do him much of an injustice by characterising his attitude as "sulks". Copleston could equally have thrown a fit of the sulks and said "God is just there, that’s all", but I don’t think it would have made for a very good debate, or for very good wireless. Moral: not all the reasonableness is always on the side of those who are called rationalists.

As I say, the objection Aquinas brings is far more reasonable than Russell’s sulks. It has a place in a serious debate, of which St Thomas’s own arguments provide the other side. Like the argument from evil, it is common to this day — though, like the argument from evil, not exactly in the form which Aquinas gives us — and it is psychologically very forceful. The claim it makes amounts, roughly, to the claim that there is no need to look for God as the explanation of the world, because the world already has an explanation. The sum of the partial explanations which can be given for this or that part of the world — as Aquinas says, the sum of nature as an explanation for natural phenomena, and of human reason as an explanation for voluntary actions — provides a complete explanation of the world as a whole.

Aquinas’s own reply is perhaps, at this stage, not very enlightening. In it he refers us explicitly to his own Five Ways, as providing a reason for denying that the explanation offered for the world by the objection can be a complete one.

"To the second objection we have to say that since nature acts towards a determinate end in virtue of the direction of a superior agent, then we have to bring back the things that come about through nature to God, as well, as their first cause. In the same way, too, things that come about intentionally should be brought back to some higher cause: not the human reason and will, for these can change and fail, and everything that can change or fail must be brought back to some first unchanging originating principle which is necessary in its own right, as we have shown."

Since Aquinas himself refers us to his own developed arguments ("as we have shown"), we may perhaps omit for the present a careful examination of this reply. If Aquinas is right, going through the Five Ways will make it clear that the objection has no force. However, it will be useful to return to an examination of the apparent force of the original objection, perhaps couching it in terms which we would be more likely to use nowadays. This performance is of value not merely to help us to realise that the discussion St Thomas is involved in is not so alien to us as might at first appear, but also because some considerations which arise from an examination of the objection may provide a clue to the strategy Aquinas is following in the Five Ways.

Putting the matter crudely, the objection is as follows. If we have an explanation of the existence of each bit of the universe, can we not just lump all those partial explanations together? And if we do so, have we not then given an explanation of the whole universe? If such a complex of partial explanations existed, would not Aquinas’s search for an explanation of the universe as a whole be at least redundant? If we had such an explanation of the whole universe (because we had an explanation of each and every bit of it) is it not a bizarre idea to then start looking for an explanation of the universe as a whole? Just what justification could there be for a distinction between "the whole universe" and "the universe as a whole" which seems to be necessary if Aquinas’s project is not to be just ridiculous?

We can go further. It is true that we do not have an explanation of each and every bit of the universe, nor do we feel confident that we are likely to get one in the very near future. We are not Victorians, after all: we do not believe that science provides all the answers — yet. But we do have an idea of what sort of thing science has to do to explain each and every bit of the universe. Some bits are well explained, some less well, some scarcely at all, but we are on our way. Providing an explanation of the existence of each and every bit of the world is an intelligible and theoretically feasible project. We have not got such an explanation yet, and we may never reach it, but we know what it would be to have such an explanation. Victorian optimism may have been misplaced, but do we not have grounds for holding that it is at least possible that there should be a scientific explanation of each and every bit of the universe? If this is so, then if it is not obvious that our search for God is wholly redundant and ridiculous, it is only our current state of relative scientific ignorance which conceals the obviousness from us. This, in the eyes of any theist, is tantamount to a rejection of the existence of God. No classical theist can say, "There must be a God now, but as science develops the need for God will diminish until maybe God becomes redundant." (It is the fact that modern science is expected to come up with the explanations of this or that part of the world, incidentally, which has brought it about that this objection against the existence of God, which I have called "the argument from partial explanation", a label which fits Aquinas’s formulation well, is now often called "the argument from science.")

This argument is already quite concrete and understandable, but it has become a commonplace in the philosophy of religion to bring it down to earth yet more by what is now called the example of the five Inuit, or argument of the five Inuit. The argument goes as follows. If one were to meet a group of five Inuit standing on a street corner, the presence of this group is a surprising phenomenon, one which requires an explanation. But suppose you had an explanation of the presence of each one of the group: would it then still be sensible to demand an explanation of the presence of the group as a whole? Clearly not.

It is worth pointing out that this kind of argument is one which Aquinas himself would accept. We have seen above that for Aquinas there need be no reason for coincidences, for the per accidens as such. That which is per accidens is the result of the convergence of two or more lines of explanation or causality, and once each of those lines of explanation has been given, the per accidens existent is already explained. To seek for an explanation of the per accidens as such is to demand that it should be, in the modern jargon, "over-determined". If I seek an explanation of why I should have bumped into a friend in the street, there need be no other explanation than the explanation of why I am proceeding along street S from east to west at time t, plus the explanation of why my friend is proceeding along street S from west to east at time t. I can lump the explanations for each bit of the meeting together, and I have an explanation of the meeting as a whole. What goes for the meeting of me and my friend goes for the group of Inuit, too: if there is a reason why Inuk A is at point X at time t, and a reason why Inuk B is at point X at time t, and a reason why Inuk C is at point X at time t, and a reason why Inuk D is at point X at time t, and a reason why Inuk E is at point X at time t, then there is eo ipso a reason why there is a group of five Inuit at point X at time t.

More to the point, what goes for a group of Inuit goes for the world as a whole. In the example we have a complex and surprising phenomenon, the presence of a group of five Inuit; a phenomenon which excites our curiosity and makes us want to look for an explanation. The existence of the world, it is suggested, is in every way parallel: a complex and surprising phenomenon, a phenomenon which excites our curiosity and makes us want to look for an explanation. We ignore the siren call of Lord Russell to regard the world as "just there", as we would probably ignore an appeal from one of our more boring friends to regard the Inuit as just there. (My grandmother used to say "Don’t stare" and "mind your own business". That Lord Russell should have so much in common with my grandmother is another of those trivial but interesting sidelights on the history of philosophy.) In the example, we are supposed to find an explanation of the presence of each and every one of the Inuit who compose the group; and once we have that, we will not need to look for an explanation of the presence of the group as a whole. In the case of the world, we are on our way to having provided for us by science an explanation of the existence of each and every thing in the world. Equally, then, it is claimed, we do not need to look for an explanation of the existence of the world as a whole. The case against arguing for the existence of God, in so far as arguing for the existence of God means arguing for an explanation for the existence of the world as a whole, rests.

Clearly the time has come to tell a story. Once upon a time there were five Inuit standing on a street corner in Glasgow: Nanuk, Amoraq, Kadlu, Kotuko, and Angekok. The claim being made is that when we have an explanation of the presence of each and every one of the group we then needn’t ask for a reason for the presence of the group as a whole. Let us endeavour, then to find an explanation of the presence of each and every one of the group. Why is Nanuk there? Nanuk has a strong, if not necessarily a good reason for being there: he is deeply and madly in love with Amoraq, and where she goes, he goes. Why is Amoraq there? Amoraq is there because she is Kadlu’s wife (I told you that Nanuk’s reason for being there was strong but not necessarily good), and she is accompanying her husband. Why is Kadlu there? Kadlu is there because, not unnaturally, he does not wish to see his wife go off touring Europe without him, and particularly not in the company of Nanuk, of whom he has (justifiable) suspicions. Why is Kotuko there? Kotuko is there because he is the infant son of Amoraq and Kadlu, and is too young to be left at home on his own. Why is Angekok there? Angekok is there because (as, I understand, his name suggests) he is the village sorcerer or shaman, a person of some authority in the community, who is there to keep an eye on Kadlu and Nanuk, to make sure that they don’t start quarrelling or even fighting, thus lowering the high reputation the Inuit people have hitherto deservedly enjoyed in Scotland.

What is wrong with this story? As a novel, it’s pretty thin on both characterisation and plot, as a pulpit anecdote its moral is rather unclear. But as a story to explain why each and every one of the group of five Inuit is standing on the street corner in Glasgow, there is nothing wrong with it. Ask "Why is he there?" or "Why is she there?" of any one of the group and the story provides a perfectly valid explanation. If the thesis which this example was developed to support is correct, we can now lump all these partial explanations together and we will have an explanation of the presence of the group as a whole. To look for any further explanation of the presence of the group as a whole should be redundant and ridiculous.

One does feel, however, that not everything that needs to be said has been said. Yes, we might say, I can see a reason why each should be on a street corner in Glasgow, and these reasons, lumped together, do indeed give some sort of a reason why all should be there: what I cannot see is a reason why any should be there. Some explanation is surely still missing. Can I have found a sufficient reason for the presence of all when I have a crying need for a reason for the presence of any?

What is wrong with the story, what makes it unsatisfactory for us to lump together the explanations of the presence of each and call the resulting story an explanation of the presence of the group as a whole, is the following feature: the explanation of the presence of each member of the group is given in terms of his or her relation to some other or others of the group. Naturally I composed the story, such as it is, to display this feature. When the explanation of the presence of each member of the group is given in terms of his or her relation to some other or others of the group, then we cannot lump together the explanations of the presence of each member of the group and call it an explanation of the presence of the group as a whole. It is therefore not redundant or ridiculous, in such a case, to seek for a further explanation of the presence of the group as a whole.

This further explanation can take various forms. It could indeed be an explanation for the presence of each and every member of the group directly: in the imagined case, they might each be members of the North Manitoba Ethnic Dance Troupe which is about to perform at the Mayfest Arts Festival. Equally well, it could be an explanation primarily and directly for the presence of only one; maybe Kadlu is an exceptionally good footballer and is thinking of signing for Partick Thistle. In the latter case, one might want to say that Kadlu has a reason for being there, and that the others don’t. This would be a mistake. The others have a reason for being there: they are there, directly or indirectly, because Kadlu is, and Kadlu is there to sign for Thistle. Thus the reason for the presence of the group is Kadlu’s signing for Thistle. In this case Kadlu occupies an explanatorily privileged place relative to the rest of the group. If the story were told another way, another one, or several others, might occupy the explanatorily privileged place. Or, as has been said, it might be that none of them occupies an explanatorily privileged place; as in the dance troupe explanation, they might all have an equally strong and equally immediate explanation for their presence. The form of the explanation is at this point irrelevant: what is relevant is that the story as originally told cannot be the end of the matter, there must be an explanation of the presence of the group as a whole, over and above the various explanations of the presence of each member of the group

Equally unimportant, clearly, is the size of the group. What goes for five Inuit goes for ten, for twenty ... for as many as you have room for on the street corner. No matter how large the group, if the explanation of the presence of each is given in terms of that person’s relation to another of the group, we need a further explanation of the presence of the group as a whole. This remains true, we may point out, even if the group is infinitely large. We may grant that there is something odd about the picture of an infinitely large group of Inuit, or of anybody, standing on any finite corner of any finite pair of streets in Glasgow, but the point remains. Easier to imagine, perhaps, is a group of infinite, or at least indefinite, temporal duration. For as long as you care to mention there are five Inuit standing on the street corner, since when one goes another comes. If the presence of each Inuk is explained in terms of his or her relation to another of the group — e.g. if each new arrival explains his or her presence by saying "I’m here to take over from Whatsisname" — then even if the group is infinitely large, then we will still need an explanation of the presence of the group as a whole.

Let us apply the parallel. Science, let us admit, does adequately explain the existence of each and every thing in the world; or, if in fact it doesn’t, let us admit that it looks as if one day it might. If we ask for the explanation why this or that thing exists, science can give it to us, or at least can plausibly claim that one day it will be able to give it to us. In the same way our original story gave us a explanation of the presence of each and every Inuk. But the parallel continues: the explanation of each and every thing in the world is given in terms of that thing’s relationship to something else in the world. Clearly this is so: science very carefully avoids telling us about anything outside the world; science is rightly agnostic about anything outside the world. So it was with the Inuit: the explanation of the presence of each was given in terms of his or her relationship to some other member of the group. For this reason we required a further explanation of the presence of the group of Inuit as a whole, and for the very same reason we not only may, but must, require an explanation of the existence of the world as a whole.

The parallel continues yet further. This further explanation could relate each part of the world directly with some single explanation, as the dance-troupe explanation did to the Inuit; this, as we shall see, is the kind of explanation favoured by St Thomas in the Five Ways. Or the further explanation might relate in some more direct and privileged way to some individual part of the world, as the football explanation related directly to Kadlu and indirectly to the others; this would be similar to the way in which some people explain everything else in the world in terms of its relationship, direct or indirect, to the Big Bang, and then wonder about the explanation of the Big Bang. St Thomas, I think, would not object to this kind of explanation, but, as we shall see, he does not (in the Five Ways, at least) think he has sufficient evidence to establish the explanatory priority of any particular part of the universe.

Lastly, the parallel applies also to the question of the infinite extent of the whole. For St Thomas, the world is limited in physical extent. He also believes, though, that while in fact the world had a beginning in time, it might have existed for ever. He makes use in a number of the Five Ways of a step in the argument which claims "We cannot go on to infinity in this line". As we shall see, the sense in which he uses "we cannot go on to infinity" is a curious one, one which is compatible with the hypothetical everlasting existence of the world; he might have expressed himself more perspicuously by saying "If we go on to infinity in this line, there is no explanation". This is precisely the point we made when we imagined the presence of the group of Inuit indefinitely protracted in time.

This illustration — which the reader may feel has been over-laboured — has the additional justification that it can be used to illustrate what may be thought of as the overall pattern of the Five Ways. St Thomas’s claim, in each of the Five Ways, is that the world as a whole requires some kind of explanation; and the conclusion of each of the Five Ways is that the explanation of the world of the kind which that Way demands is something which we call God. In a sense, "God" is at this stage no more than a label for "the explanation of the world". St Thomas feels himself obliged to go on in later questions of the Summa to argue that, e.g., there can be no more than one God, that God does not enter into any kind of composition with the world, that God is good and almighty, and so forth, through all the traditional attributes which classical theism attributes to God. The conclusions of the Five Ways are very agnostic, in a sense, though not tentative: and it seems, for example, that the Third Way might be consistent with a multiplicity of physical or material Gods, such as the ancient Greeks traditionally believed in, and possibly that the Fourth Way is consistent with a multiplicity of formally distinct spiritual Gods, such as (perhaps) some of the Greek philosophers believed in.

Though the Five Ways can be seen to follow something of a common pattern, they differ essentially in detail. They differ essentially, in that though in each case what is being argued for is the need that the world as a whole has for an explanation which is distinct from it, in each of the Five Ways the world as a whole is being considered under a different aspect, under a different description, as a systematic whole, representing various different manners of systematisation. This is clear from the first sentence of each of the Five Ways. St Thomas begins by identifying a particular feature — call it feature X — a different feature for each of the Five Ways. This feature is displayed by this or that bit of the world: in each case St Thomas takes this claim as being obviously true, and does not argue for it. One of the distinguishing marks of feature X is that anything which displays feature X requires an explanation in terms of its relation to something else: the fact of displaying feature X means that whatever displays it requires an explanation, and does not explain itself. This is clearly a vulnerable point of each in the Five Ways: it is possible for the critic to argue that the feature which St Thomas identifies does not have this distinguishing mark. At various points we will see St Thomas defending his identification of a feature X against obvious criticisms of this kind: a good example is the defence he offers for the principle "omne movens ab alio movetur", everything which is in process of change has that change initiated in it by something else, in the First Way.

The next step in each of the Five Ways is likewise questionable. It is what Geach calls the "lumping-together" move. This consists in moving from the claim that this or that part of the world displays feature X to the claim that the world as a whole displays feature X. There is nothing particularly mysterious about this move: it can be made perfectly straightforwardly with regard to any number of features, and with regard to any number of systems which contain parts which display that feature. The pattern of argument is that any system which contains parts which display feature X is itself something which displays feature X. It is not disputable that this move works where feature X is, say, "complexity": any whole which contains complex parts is itself complex. Equally, it is not disputable that this move fails to work for "smallness": a system which contains small parts may itself be rather large, even, indeed, if all its parts are small. Here again there is room for the critic to raise a query: is the feature X which St Thomas has identified one which permits of this move? We can see the point being argued fairly explicitly in the Third Way. It is worth saying, however, that there is no obvious contradiction between this mark of feature X — that it permits us to make the lumping-together move — and the previously identified mark, that whatever displays feature X requires an explanation in terms of a relation to something different to it. Any alleged inconsistency will need looking at on its merits.

This brings us to the penultimate move in each of the Five Ways. It has been claimed that parts of the world display feature X, and that whatever displays feature X requires an explanation in terms of a relation to something else. If the lumping-together move is valid, we are obliged to admit that the world as a whole, which consists, at least in part, of things which display feature X, is itself something that displays feature X. On the supposition that an appropriate feature X has indeed been identified, we are forced to conclude that the world itself, as a whole, requires an explanation in terms of something else. It is worth pointing out that this move does not yet bring us directly to God. There might be something which explains the existence of the world as a system which displays feature X, which is in some way separate from or distinct from the rest of the world as a whole, which nevertheless itself displays feature X. If it does display feature X, then it too requires an explanation in terms of its relation to something else. We then must perform the lumping-together move over again, putting this new explanatory entity together with the rest of the world which displays feature X, and ask of this newly enlarged systematic whole, what is its explanation. We might, indeed, ask about that particular explanatory entity, itself displaying feature X, what is its explanation: this would be to attribute to it a position of explanatory privilege, such as we offered to Kadlu as footballing star in the story of the Inuit, such as some people nowadays would offer to the Big Bang, or such as St Thomas in the Summa contra gentes seems to want to attribute to the outermost sphere of the heavens. It is worth pointing out that in the Summa theologiae St Thomas abandons this strategy, and prefers that of performing the lumping-together move over again. Clearly this strategy is of greater generality, and does not depend so much on shifting physical theories which would identify what is explanatorily privileged in different ways at different times.

But now we come to the concluding moves of the Five Ways. No matter how many explanatorily privileged elements of the world we may identify at one time or another, we are entitled to perform the lumping-together move as often as we have reason to do so, and to recognise in the (now expanded) system of the world which we now behold, something which displays feature X. Now matter how big the world is, no matter how many extra elements, explanatorily privileged or not, we find ourselves obliged to build into it, the world does not cease to be something which displays feature X. And that which displays feature X requires an explanation in terms of a relation to something else. This remains true even if the world is of infinite extent: if the explanation of the presence of each member of the group of Inuit is his or her relationship to another member of the group, then the presence of the group requires an explanation, even if the group is of infinite size or of (at least) indefinitely long duration. The whole system — the group of Inuit, the world as a whole — will thus lack an explanation, until we posit the existence of something, in terms of a relationship to which the whole system which displays feature X can be explained. And this posited something, through the reasoning we have just expounded, cannot itself display feature X, or it would itself be just a part of the newly expanded system of things which display feature X. We therefore postulate the existence of some thing which does not display feature X, which therefore does not require an explanation in terms of its relationship to something else, in terms of a relationship to which the world as a whole is explained. And this, Aquinas tells us, we call God.

Two glosses are worth making at this point. The way we have explained the Five Ways makes it clear why it is absurd to ask "But who made God?" The God whose existence we have been forced to postulate, by the very force of the argument, does not display feature X, and therefore the question "Who made God?", or, more accurately "What is it in terms of a relationship to which the existence of God is explained?" does not arise. The second point to be made is that this last line of each of the Five Ways provides yet another point at which the critic may object. The critic may very well protest that he does not see why the explanatory entity thus postulated needs to be called "God": in more formal terms, that he cannot see why "Entity lacking feature X, in terms of a relationship to which the world as a whole, in so far as it is itself something which displays feature X, is explained" enters into the signification of the word "God". The pattern I have suggested for all of the Five Ways thus admits three crucial points for the objector, while in each of the Five Ways there may also be particular points for objections to be made, which relate to the different ways in which the accounts of the different kinds of feature X are developed. Displaying the arguments in this way should make it easier to understand the arguments of St Thomas and to recognise them as valid, if they are valid; but it will also make it easier to find points of questionable validity. The structure alleged is not a kind of trick to make the arguments more acceptable: it seems to me to be a useful piece of common ground on which supporters and opponents of the Five Ways can carry out their debate.

If this position, that there is a discernible single structure to be found in all of the Five Ways, our next move has to be to examine the detailed development of each of them.

Return to Contents

Go to Philosophy page