Friday 20 April 2012

The Ontological Argument for God's existence


Nobody's claiming the argument proves THIS...

Ontological arguments seek to prove the existence of God simply by armchair reasoning - by thinking through concepts logically rather than venturing out into the world to find material evidence. Ontological arguments usually start from a definition, from which the existence of the entity logically follows. Here is one example:
  1. By definition, God is a supremely great existent being
  2. Therefore, God exists.
The argument is valid (see previous post). However, is it convincing? No. This is because only somebody who is independently prepared to admit the conclusion (2) will accept the premise (1). Therefore, we would say that the argument is ‘circular’, since the acceptability of the premise depends on whether we already believe the conclusion. Or rather, the argument assumes its own conclusion. Though valid, the argument doesn’t prove anything (though this guy might disagree).
Consider a more sophisticated version, attributed to St. Anselm of Canterbury (1033 - 1109):
  1. God is the greatest conceivable being
  2. If something is the greatest conceivable being, it possesses all great-making properties
  3. A great-making property is a property that, if had, makes a being greater
  4. Existence is a great-making property
  5. Therefore, the greatest conceivable being possesses the property of existence
  6. Therefore, God possesses the property of existence
  7. Therefore, God exists
This new argument is valid. But is it sound, and therefore convincing to any degree? Though it is thought fashionable these days (mostly to people who haven’t studied any philosophy) to simply declare this argument absurd and walk away, this isn’t a rationally acceptable thing to do - reasons must be given for why an argument is unsound before we reject it.
The problem is that this argument cannot so obviously be declared false as the first one, since it is not so obviously circular. It’s all very well to assert ‘you can never define something into existence’, but this isn’t really pointing out a specific flaw in the argument, which is a harder task than many think. Whether this version of the Ontological Argument is good or not, it poses an interesting logical problem nonetheless, and has premises to which even the atheist can agree. Unfortunately, accepting the conclusion renders atheism incoherent.

I love ontological arguments - if you give them a chance (which you should), they are endlessly fascinating. More to come on this topic.

22 comments:

  1. I would reject 4. existence is not a property. A object that exists has properties, not the property of existence specifically.

    Do you allow for non-existent things to have properties?

    ReplyDelete
  2. Consider Sherlock Holmes. He has properties ascribed to him - the property of having a long nose, the property of having a sharp mind etc. There doesn't seem to be much wrong with this.

    Consider the Taj Mahal. It's not incoherent to suggest that if the Taj Mahal lacked the property of existence, it would be less great. It would retain its identity as the Taj Mahal, but as a non-existent concept.

    With these considerations, the idea that existence is a property doesn't seem very problematic.

    ReplyDelete
  3. It states in proposition 1 that God is the greatest conceivable building. It seems to me that, from that premise, it is our conception of God that must include the property of existence.

    ReplyDelete
  4. Sorry, greatest conceivable being*

    ReplyDelete
  5. Terms are ill-defined. "Existence" here refers to a physical existence or physical reality, not a conceptual one. Thus Sherlock Holmes exists as a character in the mind of its readers, but lacks a physical reality. I do not think terms such as "existence" should be used without clarification as to their meaning, as it can lead to confusion. If you cannot define a concept beyond some nebulous reference to intuition, do not use it.

    ReplyDelete
  6. On the contrary, all argument must be supported by intuition at some point.

    ReplyDelete
  7. More relevantly, invoking the intuition argument isn't even applicable here; the fact that Anonymous was confused by the term existence is proof enough that you cannot just really on the fact that your readers have the same intuition as you.

    ReplyDelete
  8. In fact, Anonymous' objection is THE classic objection to the ontological argument, which is due to Kant.

    ReplyDelete
  9. Are you being deliberately dense?

    ReplyDelete
  10. No. What I said is true. Kant's objection is that existence is not a predicate. It's on wikipedia if you're interested. Anonymous wasn't confused by the term 'existence' - in fact, you've just arbitrarily restricted it to physically existing things, which is misleading in the context of the argument, since the argument can only run if existence includes non-physical entities. (That conceptual entities exist in some sense isn't by any means controversial - numbers, for example, properties, relations, etc).

    ReplyDelete
  11. I am obviously aware that existence can refer to a conceptual existence. But you say the following:

    "Consider the Taj Mahal. It's not incoherent to suggest that if the Taj Mahal lacked the property of existence, it would be less great. It would retain its identity as the Taj Mahal, but as a non-existent concept."

    What does this mean? The only sensible interpretation if this is: if the Taj Mahal lacked the property of *physical* existence, it would retain its *conceptual* existence. Yet you now claim that my restriction of existence to physicality is inappropriate in this context.

    So let's consider the alternative - are you suggesting that if the Taj Mahal lacked existence *of any kind*, it could somehow maintain its *identity*? When you consider the Taj Mahal as lacking the property of existence, what do you mean?

    This is what I mean by terms not being well defined. We're all aware that existence doesn't just mean physical existence. But qualifiers are necessary, as the absence of a qualifier implies that by "existence" you mean every conceivable type of existence, whether that be physical or conceptual. But why don't you clarify - given you've stated that you don't mean physical existence, can you explain your Taj Mahal point.

    ReplyDelete
  12. Moreover, it is INSUFFICIENT to REFUSE to define terms by referring to INTUITION as if INTUITION is the same for everyone. INTUITION IS NOT THE SAME. As an educated person born and raised in England, my INTUITION tells me it is wrong to suppress the freedom of the female in my society - YET IN THE MIDDLE EAST WOMEN can BEATEN AND KILLED for talking to Western soldiers - and this is viewed as INTUITIVELY MORAL.

    This is an extreme example. But certainly evidence of THE PROBLEM OF INTUITION. If physicists relied on intuition as much as you do, PHYSICS WOULD NOT HAVE PROGRESSED FOR THE LAST 100 YEARS. It is INTUITIVELY RIDICULOUS to suggest that TIME IS NOT ABSOLUTE, that things that go VERY FAST do in fact SHRINK. It is INTUITIVELY RIDICULOUS to suggest that on a sub-atomic scale, particles can TUNNEL THROUGH OTHER PARTICLES.

    INTUITION IS INSUFFICIENT.

    ReplyDelete
  13. Fair enough. Let's restrict 'existence' to 'physical existence' for these purposes. I am saying that The Taj Mahal, conceived as not to exist, would be less great than it is in reality. We can perfectly well conceive of things which don't exist - Sherlock Holmes, for example. These concepts can have properties, such as long nose, sharp mind etc. But they don't contain physical existence as one of their properties.

    Such consideration is supposed to reply to Kan't point that 'existence is not a proper predicate.' Kant thinks that in order for something to have properties, it must first exist, and this entails that existence can itself be a property. However, reflection on Taj Mahal, Sherlock Holmes etc shows this to be intuitively false. We can meaningfully talk about concepts such as Sherlock Holmes, and then imagine that if such concepts possessed physical existence, they would be greater than if not.

    ReplyDelete
  14. I understand the point you are making - my complaint was that it is important to distinguish between types of existence rather than vaguely referring to existence, and then citing intuition as an excuse for this lack of precision.

    ReplyDelete
  15. The appeal to intuition wasn't to obscure any lack of precision - it was to ground the plausibility of Taj Mahal and Sherlock Holmes cases.

    ReplyDelete
  16. Tom is definitely right in this argument. Face it Hank, you're just a not as good version of me. Why don't you fuck off and stop ruining everyone's lives around you.

    ReplyDelete
  17. Although I believe Kant's famous critique shows the essential error in this argument, I do as well think there is an implicit assumption here which could be questioned, i.e., assuming existence to be a real predicate, why would we deem it to be a 'great-making' predicate in the first place? Why would existence be 'greater' than non-existence?

    What criterion(s) ought we to refer to in order to determine what is to count as 'great'? Can we even claim such objective criterions to exist at all?

    ReplyDelete
  18. The existence of 'great-making' properties does depend on objective greatness. But it seems plausible to say that 'omnipotence' is 'great-making', in the sense that a being that has it is greater than a being that doesn't have it. I realize I'm being hazy here, but this sounds like something many might want to say, especially if 'greatness' is defined as 'worthy-of-worship-ness.' A being of omnipotence is more worthy of worship than a being without omnipotence.

    ReplyDelete
  19. Why existence is not a first level predicate (=express a property of objects)? Because objects must be instantiated (=exist) to have properties at all. So what is the meaning of "existence"? The meaning of existence is "to be instantiated" (=to exist, to be a second level predicate that is satisfied by first level properties and not by objects).

    Compare:

    a) Red-bearded Italians abound.
    b) Piero is a red-bearded Italian.
    c) Therefore, Piero abounds.

    a') Perfect beings necessarily exist.
    b') God is a perfect being.
    c') God necessarily exists.

    Each of these arguments are formally identical but they are logically odd. They are not valid because, in both of them, second level predicates are being attached to objects (that is, expressions of the logical category of objects or singular terms) in both conclusions (c and c'). "Exists", like "abounds" make sense only when attached to first level properties like in premises a and a', but they don't make any sense when attached to objects/singular terms. There is a confusion, then, between logical categories (first/second level predicates) which ontological arguments inherit without noticing.

    Existence cannot, logically --that is, in an intelligible sense--, be a property alongside being green, or being good, or having two legs ("like breathing, only quieter", says CJF Williams). Neither are truth nor identity. It is a second level predicate (a kind of numeral) like "...is true for all x", or "... are a lot", "... abound", "... are six" (like in "the persons working at that shop are six" --I'm not sure if this is good English, sorry).

    Kant was right about existence.

    Strongly suggested reading: What is existence?, and Being, Identity and Truth, both by Christopher F. J. Williams.

    Thanks for this great blog.

    ReplyDelete
  20. Really interesting. A related criticism might be the following. To attribute properties to an object, we use the following formula:

    ∃x(Px) - 'there exists an x such that x has P.'

    Where 'Px' is to attribute a property P to an object x. A case can be made for 'existence' being excluded as a property P, since it is already built into the existential quantifier ∃.

    However, in the above examples, it is argued (against your definition of existence) that an object need not be instantiated to have properties. That is, properties can be instantiated by a concept. Thus, the scope of '∃' can be widened to include conceptual existence, and thus 'physical existence' can then be a candidate for being property P. This is the move Plantinga makes, in (I think) 'The Nature of Necessity.'

    ReplyDelete
  21. (though I admit, extending the concept of 'exists' like this does appear a bit dubious...)

    ReplyDelete

Blog Directory
BLOG DIRECTORY, Submit blog free, Promote Blog, Best directory