final

TheOntologicalArgument--handout.pdf

Home >Social Science homework help >Philosophy homework help >final

The Ontological Argument

The Ontological Argument was first thought of by St Anselm almost a thousand years ago.1 The essence of the argument can be stated very briskly.

1. God, by definition, is a perfect being. 2. It is better to exist than not to exist.

Therefore, God exists.

In an argument, one may define terms however one wishes, and premise 1 just reports one aspect—indeed, I have argued, the central one—of the theistic definition of God. So if anything goes wrong with the argument, then it must be in premise 2. But premise 2 looks pretty obviously right as well. Consider the question: which of these would be better for you: that you be vaporized now with a ray gun and thus that you cease to exist or that you continue to exist? However small an amount of benefit or enjoyment you’re receiving from

reading this, I doubt if you’ll really think you’d be better off if you didn’t exist. Of course, we can all imagine a situation where someone’s life was so bad that it would be better for them if they ceased to exist—maybe the Spartan boy I told you about in an earlier chapter was in such a situation. However, if the person in question was in all other ways well off, it would certainly be better for him or her if he or she existed rather than not; and God is obviously going to be maximally well off in all other respects, so it’s obviously going to be better for him (and indeed us) if he exists. The claim that it’s better to exist than not to exist seems then—minor and irrelevant quibbling aside—right. Both the premises of the Ontological Argument seem to be obviously true; taken

together they seem to lead in an obviously deductively valid way to the conclusion that God exists, which was something not so obviously true. If God’s by definition perfect, then of course—given that it’s better to exist than not to exist—he’ll have to exist. It’s impossible for the premises to both be true and yet the conclusion false and it’s obvious that both the premises are true. So it seems as if we’ve got a deductively sound argument for the existence of God the soundness of which is more obvious than is the existence of God. The Ontological Argument then seems to satisfy our criteria for being a good argument. It seems to, but does it?

♦ ♦ ♦

Co py ri gh t © 2 00 5. C la re nd on P re ss . Al l ri gh ts r es er ve d. M ay n ot b e re

pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or

ap pl ic ab le c op yr ig ht l aw .

EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 5/17/2017 4:18 PM via TARRANT COUNTY COLLEGE DISTRICT AN: 186624 ; Mawson, T. J..; Belief in God : An Introduction to the Philosophy of Religion Account: tarrant

It’s easier to spot that something has gone wrong with the Ontological Argument than it is to describe what it is that has gone wrong with it. One way of seeing that something has gone wrong with the Ontological Argument is to consider that if it worked, one could generate parallel arguments ad infinitum that proved the existence of any sort of entity one cared to mention. Allow me to introduce one such parallel argument.

The Senior Common Room butler at my college is pretty good at keeping us all in order. For example, there was one occasion, early on in my membership of the Senior Common Room, when I was dining at High Table. The main course arrived—to be served silver-service style by the butler. I was one of the first to be served, and scooted one slice of whatever it was off the salver, then prepared to take a second. The butler leaned forward slightly and sotto voce advised me, ‘One is usually considered sufficient, sir.’ I was a bit miffed at this, but left the second portion on the salver, which proceeded in his hands down the table towards the end, where the Senior Tutor was sitting. As it arrived there, I noticed that there had been exactly the right number of portions on the salver to mean that the Senior Tutor—the last to be served—took the last one. Had I taken a second portion earlier on, then the Senior Tutor would have been left without any main course; the assembled eyes of the Senior Common Room would have then worked their way back along the table, unerringly seeking out where this prob- lem had originated. Ultimately, they would have fallen on me, merrily munching my way through two portions. Were it not for the timely intervention of the butler, my career at my college might have been cut rather short.

So, as you can tell, my college’s butler is pretty good at keeping people out of scrapes. But even he isn’t the best possible butler; even he can’t quite be a Jeeves to my Wooster, which is—one might reasonably hypothesize—more of a sign of how much of a Wooster I am than how little a Jeeves he is. In any case, reflection on this incident prompts me to think that it would certainly be rather handy to have a butler at one’s side throughout one’s life, ready to assist one in making one’s way through the world with wise sotto voce advice. Let me define the term ‘Jeeves’ to mean the best possible butler for you. So, Jeeves will—by definition— always be on hand for you whenever you might need him.

1. Jeeves I define as the best possible butler for you. 2. If there is a better analysis of the Ontological Argument than this, it

would be better for Jeeves to be right by your side now, handing you this better analysis.

Therefore, Jeeves must be right by your side now, handing you a better analysis of the Ontological Argument than this one.

This argument seems as good as the Ontological Argument for the existence of God. The first premise simply reports my definition of ‘Jeeves’, so there’s nothing to be argued with there. The second premise reports the fact that if there

The Existence of God126

Co py ri gh t © 2 00 5. C la re nd on P re ss . Al l ri gh ts r es er ve d. M ay n ot b e re

pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or

ap pl ic ab le c op yr ig ht l aw .

were such a thing as a better analysis of the Ontological Argument than this, then, as it would be better if you were reading this better analysis, so a better butler would be one who was by your side with such an analysis. That seems to follow pretty obviously. Yet from these two premises it follows that Jeeves must be right by your side with a better analysis than this. Look for him. Is he there? No. The only way to escape the conclusion that he must be there seems to be to claim that this is the best possible analysis that one might attempt of the Ontological Argument. I might be happy to rest content with this conclusion, but I doubt that you are. The objection that if the Ontological Argument worked, then my Jeeves

argument and similar sorts of arguments would also work is sometimes called the ‘Overload Objection’ to the Ontological Argument; if the Ontological Argument worked, then we could overload the universe with all sorts of entities like Jeeves.

♦ ♦ ♦

So, something’s gone wrong with the Ontological Argument. What exactly? First, premise one is rather ambiguous. Is this premise using the term ‘God’ to

pick something out and then attributing a property, albeit an essential one, to it, just as you might say that this book here—waving it around—is by definition something that has pages? Well, if so, we could not know that the term ‘God’ had secured reference without already knowing the conclusion of this argument, that there is a God, so premise 1 would not be one that could be known to be true with more certainty than we knew the conclusion, that there is a God. This would be sufficient to undercut any claim that the Ontological Argument— however deductively sound—is good. However, if premise 1 is not using the term ‘God’ to pick out something and then attribute a property to that thing, then it must mean something like ‘If there is a God, then he is by definition perfect’, but if that’s what premise 1 really means, then although it can be known to be true without first needing to know that there is a God, it cannot support the conclusion that God exists but only—at best—the conclusion that if there is a God, then he exists. This conclusion is a rather unexciting one. We all knew that anyway. So premise 1, despite my initial enthusiasm for it, is in fact deeply questionable. Despite this being a sufficient reason to reject the Ontological Argument, for

the sake of completeness if nothing else we must look at the second premise. The second premise is also the one on which most philosophical criticism has focused. The second premise is ‘It is better to exist than not to exist.’ What can be said against this second premise?

♦ ♦ ♦

One can beat around the bush for quite a long time here, but eventually one gets to the point that was first made by Kant: existence is not a predicate. Let me explain what Kant meant.

The Ontological Argument 127

Co py ri gh t © 2 00 5. C la re nd on P re ss . Al l ri gh ts r es er ve d. M ay n ot b e re

pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or

ap pl ic ab le c op yr ig ht l aw .

I talked earlier about this book and a couple of the properties that it had: one of these properties was that it was, I guessed, then being held by you. I’m guessing it’s got that property again now. Kant’s point would be that while it is indeed a bona fide property of this book that it is being held by you, it’s not a bona fide property of it that it exists; saying that existence isn’t a predicate is a way of saying that existence isn’t a property that objects have. So, the following sentences as spoken by me are true (I’m supposing): ‘This book is being held by you’ and ‘This book exists,’ but—according to Kant—there’s a crucial difference between these two sentences. The first really does predicate something of the book. It picks out the book and asserts of it that it has a property, the property of being held by you. The second sentence, despite its grammatical similarity with the first, does not—according to Kant—do this. It doesn’t pick out the book and assert of it that it has a property, the property of existence. What does it do then? Answering that question is a bit more tricky. To do so I have to augment what Kant said with something said by a later philosopher, Gottlob Frege, and it’s going to take a moment or two for me to set out all the ideas we need if we are to understand what Frege said.

First then, I want to introduce a distinction between what I’m going to call concrete objects and what I’m going to call abstract objects. Examples of con- crete objects would be things like this book; the chair you’re currently sitting on; and your right hand. Examples of abstract objects would be things like the nature of education; the current government’s misconception of the nature of education; its consequent policies with respect to universities; and—to move away from my particular concerns—the number five. On what basis do we decide whether a given thing is a concrete or an abstract object; or indeed do we decide on any basis at all? (It may be that the distinction is a brute one—incapable of explica- tion in terms of anything else.) This question is not an easy one to answer, but fortunately for my present purposes we don’t need to answer it, assuming as I think it safe to assume that we all have a pretty good grasp on the distinction through the examples I’ve just given.

Armed then with the distinction between concrete and abstract objects, let’s consider the concrete objects that are the chairs in the room in which you sit. Let me suppose for the sake of argument that there are three of them. It’s natural for you to group the chairs in the room together in your mind for the purposes of discussion into one set, the set of chairs in the room. The set of chairs in the room is an abstract object the members of which are concrete objects. The abstract object that is the set of chairs in the room has properties that its concrete members do not have. It has the property of having a quarter of the number of members as the abstract object that is the set of chair legs in the room has (I’m assuming). The individual concrete objects that are the chairs in the room could not be said to have a quarter of the number of members as the set of chair legs in the room; that wouldn’t make sense. The individual concrete objects that are the chairs in the room have properties that the abstract object that is the set of chairs

The Existence of God128

Co py ri gh t © 2 00 5. C la re nd on P re ss . Al l ri gh ts r es er ve d. M ay n ot b e re

pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or

ap pl ic ab le c op yr ig ht l aw .

in the room does not have; they each have upholstered seats (I’m assuming). The abstract object that is the set of chairs in the room could not be said to have an upholstered seat; that would not make sense either. Now, consider the following two sentences as they might be spoken by you:

‘The chairs in this room have upholstered seats’ and ‘The chairs in this room are three in number.’ If one wasn’t thinking too carefully, one might say that each of these sentences had the same subject—the chairs in the room in which you sit—and predicated different things of that subject, having upholstered seats and being three in number. But with Frege’s help we can now see that the real subject of these two sentences is actually different, despite their similar grammar. The first sentence takes the concrete objects that are the chairs in the room as its subject; the second takes the abstract object that is the set of chairs in the room as its subject. The first is saying that the chairs in the room have a certain property, the property of having upholstered seats; the second is saying that the set of chairs in this room has a certain property, the property of having three members. With this in hand, we now have the tools to understand Frege’s interpretation of existence. Consider the sentence as uttered by you, ‘The chairs in this room exist.’ What

is the true subject of this sentence? Is it the concrete objects that are the chairs in the room? No. Is it the set of chairs in this room? Yes. Saying that the chairs in the room exist is saying that the set of chairs in the room does not have zero members. Saying ‘X exists’ is not then actually saying anything about X. It’s saying something of the abstract object that is the set of those things that’s picked out by the concept of X and it’s saying of it that it’s not the set with zero members. So, according to Kant and Frege, existence is not a property of con- crete objects; existence isn’t something that objects do, like breathing, only quieter. Rather, when one says that X exists one asserts something, not about X, but about the set of Xs and what one asserts is that the set of Xs is not the empty set, the empty set being the set with zero members. If there is a God, then the set of Gods is not the empty set, but the fact that the set of Gods is not then the empty set is not a fact about God; it isn’t a property of the concrete object that is God that the abstract object that is the set of Gods is not empty. Once we’ve shown with Kant and Frege’s help that existence is not a property

of God even if he does exist, premise 2 of the Ontological Argument can be seen to collapse. If existence is not a property of God even if he does exist, then it cannot be a property that it is better for him if he has. How then to explain our intuition that one’s ceasing to exist would be bad?

We have seen already that what would make one’s permanently ceasing to exist bad for one is that it would frustrate one’s flourishing and one’s desires. Never having existed would not have frustrated any flourishing or desires, so even though it is not good for one that one was brought into existence, death would be bad for one if it was permanently ceasing to exist, which is why—as we’ve seen—if there’s a God, he’ll ensure that our deaths are not our permanently

The Ontological Argument 129

Co py ri gh t © 2 00 5. C la re nd on P re ss . Al l ri gh ts r es er ve d. M ay n ot b e re

pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or

ap pl ic ab le c op yr ig ht l aw .

ceasing to exist. It’s not bad for the brother that I never had that he never existed; it would be bad for the sisters that I do have if they permanently ceased to exist.

So, to sum up my conclusions with regard to the Ontological Argument: the first premise is true—on both interpretations of it—if and only if theism is true; it is false on the interpretation of it that would be necessary for the argument to be a deductively valid one for theism if theism is false. This is a sufficient reason for us to conclude that the Ontological Argument is not a good argument for theism. The second premise is false if existence is not a predicate, which it is not. This too is sufficient for us to conclude that the Ontological Argument is not a good argument for theism. The Ontological Argument fails in two ways then as a deductive argument and—starting as it does from pure categories a priori—it cannot be turned into an inductive argument. I therefore conclude that the Ontological Argument does not provide any reasons for believing that ‘There is a God’ is true.

♦ ♦ ♦

This is the ‘classic’ version of the Ontological Argument and one so central to the tradition that other ‘versions’, in so far as they differ from it, run the risk of not counting as versions simply by virtue of their doing so. However, if we think of the essence of the Ontological Argument as simply its proceeding ‘from pure categories a priori’, then there are arguments that are recognizably different from the classic version and yet deserve to be considered versions of the Ontological Argument. I’ll close this chapter by discussing such arguments in terms of a rather generalized instance of them. It will be helpful, before I do this, to say a word or two about the notion of possible worlds as it is used in the presentation of arguments of this sort.

I began the book by using the word ‘world’ to refer to the physical universe as a whole, so that I might describe the perplexity that I claimed we have all felt at some moment in our lives when contemplating the world as a whole; the world as a whole raises in some sense a question to which we think God might be the answer. ‘World’ in this sense means ‘universe’; in this sense of ‘world’, God himself could not then be a resident of the world; if he exists, he exists outside the world; he has to in order to explain it. The important point to note now is that the notion of worlds in play when we talk of possible worlds in this context is different. Possible worlds in this context are to be understood as ways that everything might be or might have been. Thus, at least prima facie, it seems logically possible both that there’s a God of the sort we’ve been discussing in the first half of this book and also logically possible that there’s not. Neither involves a contradiction in terms. If that’s right, then—using this new notion of world— we might say that there’s a possible world in which the physical universe (the world in my original sense) is as it is and there’s a God on top and there’s a possible world in which the physical universe is as it is and there’s no God on top. God’s a resident of the first world and not of the second. Theists think that

The Existence of God130

pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or

ap pl ic ab le c op yr ig ht l aw .

the actual world is the first—the physical universe plus God (and perhaps various other supernatural beings)—and physicalists think that the actual world is the second—simply the physical universe as a whole.2

Consider then this argument:

1. It is possible that God exists, i.e. he does exist in some possible world. 2. If God exists in some possible world, then he exists in every possible world.

Therefore, God exists in every possible world, including the actual world, i.e. he actually exists.

The premises of this argument seem right. After all, we’ve just said that it is possible that God exists, which is the same as saying that he does exist in some possible world. And, even if it wasn’t one of God’s essential properties, given that it’s obviously going to be better to exist necessarily rather than merely con- tingently, then, if there is a God, he’s going to have that form of existence, he’s going to exist in every possible world. So the premises seem right and the conclusion seems to follow deductively from them. Again, perhaps the best way to see that something has gone wrong with this

argument is to see that if this argument worked, then a parallel argument would also work, one that we won’t want to say works. In this case, the reason we won’t want to say that the parallel argument works is not because its working would ‘overload’ the universe but because we can’t say that both arguments work as the second one working is incompatible with the first one working. This is the parallel argument:

1. It is possible that God does not exist, i.e. he does not exist in some possible world.

2. If God does not exist in some possible world, then he doesn’t exist in every possible world.

Therefore, God doesn’t exist in every possible world, including the actual world, i.e. he doesn’t actually exist.

Surely we have no less (non-question-begging) reason to believe premise 1 of the parallel argument than we do premise 1 of the original argument. The first five chapters have established that the concept of God is internally consistent; it describes an entity that it is logically possible exists and that it is logically possible does not exist. So something’s gone wrong with this version of the ontological argument. What? The answer lies in the ambiguity of the word ‘possible’ and the notion of

possible worlds. The first five chapters have established that the concept of God is consistent, so God’s existence is logically possible, which amounts to there being no inconsistency in saying that God exists. This, we may say, is indeed equivalent to God’s existing in some logically possible world, so let’s take premise 1 of the original argument in this way: it is logically possible that God

The Ontological Argument 131

pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or

ap pl ic ab le c op yr ig ht l aw .

exists, i.e. he does exist in some logically possible world. As such, we may agree with premise 1. But that God exists in some logically possible world (indeed in an infinite number of them) does not entail that he exists in all logically possible worlds, as suggested by premise 2. Rather, God’s non-existence is also logically possible; there is no inconsistency in saying that God does not exist. So God does not exist in some logically possible world, indeed he doesn’t exist in an infinite number of them. So we may reject premise 2 and thus the argument. If we take the word ‘possible’ to mean metaphysically possible, then we must accept the second premise. If God does exist in some metaphysically possible world, then he exists in every metaphysically possible world because if there is a God, he’s that on which everything else metaphysically depends. But then of course we have no (non-question-begging) reason to accept the first premise, that it’s metaphysically possible that God exists. Whether or not we think that this is true depends on whether or not we believe there’s a God. So this version of the Ontological Argument also fails as a good argument.

All versions of the Ontological Argument then fail to respect the categorical difference between manœuvring within a concept and discovering whether that concept, however understood, does or does not have an instantiation.3 If we’re going to find evidence of God’s existence—a reason to believe that there’s a God—we’ll have to consider more than the mere concept of God. Where shall we look? The only place we can: the world that he’s supposed to have created. We must see if he’s left any evidence of his existence there. It has seemed to many that it’s obvious that he has. Let us turn then to the Argument to Design.

The Existence of God132

pr od uc ed i n an y fo rm w it ho ut p er mi ss io n fr om t he p ub li sh er , ex ce pt f ai r us es p er mi tt ed u nd er U .S . or

ap pl ic ab le c op yr ig ht l aw .