Friday, December 26, 2008

Defining into Existence

The ontological argument:
1. We can think of a greatest conceivable being – call it 'God'.
2. It is greater (qua being) to exist than not.
3. So for any being X to qualify as the greatest conceivable, X must exist.
4. So God [the greatest conceivable being] exists

The refutation:
Distinguish the properties things are conceived to have vs. those they really do have. (Santa is conceived to be a material person, made of flesh and blood, but in fact he's a mere figment of our imaginations.) When we think of the 'greatest conceivable being' in #1, is it actually great? Or merely conceived to be great? If the latter, then the premises are insufficient to establish the conclusion: it may be possible for us to conceive of some X as supremely great (existing) without it actually being so.

In other words: insofar as we can imagine the qualities of a supreme being, one of the things we imagine may indeed be that it exists ("according to the fiction", so to speak). This act of imagination is compatible with the being in question not actually existing -- and hence not actually qualifying as 'supreme'. It's one thing to be imagined as the greatest conceivable, and quite another to really be that way in fact. So, for the ontological argument to be valid, we would have to replace the first premise with:
1*: We can conceive of some being, 'God', and this being is in fact the greatest conceivable being.

But of course if we accept premise 2 then there's no reason at all to think that 1* is really true. There's no reason to think that any being actually possesses the qualities required to qualify as the greatest conceivable (which, after all, include necessary existence).

The ontological argument raises a lot of other interesting issues: whether existence is a predicate, whether it is a perfection, whether there is a unique set of qualities that would render a being insurpassably great, etc. But I think the above counterargument -- based on the distinction between the actual properties of a represented thing vs. the properties it is represented to have -- cuts to the core of the matter.

It's also worth noting that all I've done here is bring out the core insight that finds inchoate expression in the slogan, "You can't define God into existence." I recall there was a bit of a brouhaha a while ago as theists complained about Dawkins' so swiftly "dismissing" the ontological argument on these grounds. But it seems to me that Dawkins was exactly right. It's like Zeno's paradox. The argument is very obviously unsound -- sophistry, even -- and you don't have to be a philosopher to see this. The difficulty simply lies in pinpointing the error, and explaining it in a philosophically sophisticated way. Any old fool can recognize that there's an error there to be found (they even have a rough sense of what the error is).

Granted, the ontological argument is philosophically very rich, and rewards further discussion and exploration in the philosophy classroom. But that's not because there's any real question as to its soundness (as would be relevant to a popular book on the question whether God exists). It's simply because the mistakes - and the further questions it raises - are so subtle and interesting. It is, in short, of purely academic interest. (Not that there's anything wrong with that.)

28 comments:

  1. Distinguish the properties things are conceived to have vs. those they really do have. (Santa is conceived to be a material person, made of flesh and blood, but in fact he's a mere figment of our imaginations.)

    Let's make that distinction in the case of Santa. There are properties that Santa really does have in contrast to those he's conceived to have? I don't think that can be right. Otherwise, you'll find yourself quantifying over imaginary beings. If Santa really does have property P, then since Ps is true, then (Ex)Px.

    ReplyDelete
  2. This doesn't really affect my point. If we accept the Quinean dogma that quantification involves ontological commitment, then we will deny that any properties can be ascribed (de re) to merely imaginary beings. So this just makes it all the more obvious that conceiving of Santa as material does not entail that Santa really has this property (because the name 'Santa' doesn't refer). But - to be as ecumenical as possible - even if we allow talk of non-existent entities, and allow that existence is a predicate, and all that, still to conceive of God as existing does not imply anything about whether God actually exists.

    ReplyDelete
  3. Throughout the post, though, it sure sounded like you were presupposing the existence of Meinongian objects or something like that: e.g., "the actual properties of a represented thing vs. the properties it is represented to have" makes it sound like there really is a represented thing (even in cases of downright nonexistence).

    ReplyDelete
  4. How does this matter? As noted above, the Meinongian approach is the most charitable to the ontological argument, and thus offers the strongest possible refutation. But I think everything I said could be translated into a strictly Quinean framework -- it would just be clunkier and more long-winded. (In any case, I don't think there's really anything substantive at stake. Like most ontological debates, the question of Meinongianism strikes me as merely terminological. But let's keep this thread on topic.)

    ReplyDelete
  5. The following might constitute a problem.

    So there's this represented thing or intentional object or Meinongian object or whatever. And though we may represent it as having the property of being the greatest conceivable being, in fact it doesn't have that property.

    But without that property, why think that bringing this object before our mind counts as bringing the greatest conceivable being before our mind? Shouldn't we be bringing an object that actually has that property before our mind? We were supposed to agree, after all, that the greatest conceivable being exists in the understanding at least.

    Or we can ask: is there anything about this object that qualifies it for the job (e.g., any properties that it does have)? or would any other object serve equally well?

    ReplyDelete
  6. "But without that property, why think that bringing this object before our mind counts as bringing the greatest conceivable being before our mind?"

    It doesn't. One way to re-state my point is to note that we can't bring the greatest conceivable being (de re) before our minds. (The thing we have in mind cannot actually qualify as supreme, because it does not actually exist.)

    All we can do is make a 'de dicto' representation as of "the greatest conceivable being". The de dicto qualities of our representations are all that we have introspective access to, so this is all we could be asserting with the claim that the GCB 'exists in the understanding'. The phrase "in the understanding" functions here analogously to "in the fiction".

    As explained in the main post [cf.(1*)], there isn't the slightest reason to think that we can have in mind ('think about', or 'represent') a being that actually is the greatest conceivable. One is only tempted to think otherwise insofar as one forgets that 'the mind' is used here as a representational space, not a real location.

    It can be convenient to use the language of intentional objects, but we should take care not to take the reification too literally. Questions about "which" non-existent object we have in mind (or whether some "other" could take its place) are, I think, strictly nonsensical. I tend to think that any sensible question can be asked in purely qualitative terms, i.e. without appeal to de re identities. But that goes double when we're talking about mental content!

    (If you insist that there are substantive "identity" facts, then you shouldn't go anywhere near 'intentional object' talk, since there certainly aren't substantive identity facts about them.)

    ReplyDelete
  7. So are you just explaining in more words how this argument is begging the question? Again, this economics undergraduate humbly apologizes for his philosophical ignorance, but it seems to me that just because we can conceive of a greatest being that it doesn't mean it actually exists, and this may just be a short and unenlightened way of saying what you are.

    I am also interested in what it means to exist. It's like saying that infinity does exist, but in a strictly material sense, it does not. Likewise, God may exist as the sum of all information, and thus be a being that is omniscient, but He doesn't really exist materially.

    Perhaps this is completely irrelevant, but the infinity analogy could continue to hold throughout your post, particularly when we talk about defining into existence.

    1^ : We can conceive of some being, 'infinity', and this being is in fact the greatest conceivable being

    There is no reason to think that there can be a greatest conceivable set of numbers, humor me, that contains all possible numbers that does in fact contain all possible numbers, but by definition, that is infinity.

    ReplyDelete
  8. Richard,

    I'm not sure what you mean by 'conceiving X to be Y' here; it looks like you mean that it means "imagining X to have the property of being Y". But this is not conception as usually understood in ontological arguments (at least of the limited post-Cartesian sort that you have in mind here), because such arguments are usually found in systems where we can conceive things we can't imagine. Moreover, usually in such contexts 'conceiving X to be Y', taking conceiving in that sense, is not a coherent claim: we don't conceive X to be Y, we conceive X, and then judge that it is Y. But divergence between judgment and reality is trivial -- the argument itself presupposes it, since it presupposes that one might falsely judge that the GCB does not exist, and purports to show why such judgments are necessarily false.

    ReplyDelete
  9. Brandon, I'm fine with using 'conceive' in a broader sense than 'imagine'. (It needn't involve any kind of sensory "imagery", for example.) But I don't see how that changes anything. The essential point is that it is still a form of representation, and our judgments about what is true "in" the representation should not be confused with judgments about what is true in reality.

    ReplyDelete
  10. This doesn't really affect my point. If we accept the Quinean dogma that quantification involves ontological commitment, then we will deny that any properties can be ascribed (de re) to merely imaginary beings.

    Hard to follow. The suggestion that we compare the "real" properties (whatever 'real' is supposed to mean here) with the conceived properties is not something that can be done. So whatever that was supposed to show, it is simply not something that can be done with imaginary objects.

    But you add,
    . . .still to conceive of God as existing does not imply anything about whether God actually exists.

    Again, hard to follow. If God is genuinely conceivable. as seems to be conceded here (compare Chalmers on ideal conceivability), then such a being is broadly logically possible. But if God is broadly logically possible, then there is a world in which he has the property of necessarily existing. If that's true, then (assuming S5 as all do) He actually exists.
    So everything turns on God instantiating properties that are compossible. It's sometimes urged that the conceivability of God is no guarantee of God's possiblity, since there are metaphysical impossiblities that are conceivable. But in the context of God's existence, this sort of objection is confused. The metaphysical impossibilities that are conceivable are all negations of necessary a posteriori propositions. It is obvious that 'possibly, it is not the case that God exists' is not a necessary a posteriori propostion. So it's negation--viz., necessarily, God exists--is not the negation of an a posteriori necessary propostion. So no worries arise for Anselm from metaphysical impossibilities that are conceivable. The upshot is that you can get from the conceivability of God to God's existence, provided that you can genuinely conceive of God. This is why so much time is spent on arguments against the view that Anselm's God is genuinely concievable.

    ReplyDelete
  11. Mike - the point of my response was that we don't need to "compare" real vs. conceived properties. It is enough to make the distinction: imagining or conceiving of a being as existing is not thereby to conceive of a being that actually exists. That shouldn't be hard to follow. I'm not sure how to say it any more simply than that.

    As for the S5 stuff, I certainly don't think the existence of a necessary being is 'ideally conceivable' in Chalmers' sense. But that's not the sense I was talking about here, so it's kind of irrelevant. (The ontological argument I'm discussing should not be confused with the S5 modal argument.)

    ReplyDelete
  12. This comment has been removed by the author.

    ReplyDelete
  13. The linked argument strikes me as transparently unsound -- much more obviously so than the argument I've been discussing. But it is similar in that the error rests on a basic failure to understand the representational nature of mind. (There's no reason a finite being can't create a representation - or "idea" - of the infinite.)

    ReplyDelete
  14. This comment has been removed by the author.

    ReplyDelete
  15. But I don't see how that changes anything. The essential point is that it is still a form of representation, and our judgments about what is true "in" the representation should not be confused with judgments about what is true in reality.

    I'm still a little puzzled here; the ontological argument, as you present it, doesn't seem to make quite this confusion -- rather, it depends on the distinction between the two, since it's (quite explicitly) an argument from a judgment about the content of the representation to a judgment about what is true in reality, where the warrant for this inference is explicitly given: if the reality-judgment did not match the concept-judgment we'd be committed to the contradiction (i.e., the conceivable GCB that is not GC). You can't be arguing that inference from concept-judgments to reality-judgments are never warranted -- thought-experiments and 2D semantics both make such inferences -- nor that the (alleged) warrant is a bad one -- 'We have to accept X because ~X is a contradiction' is a pretty good warrant for inferring X. The argument you lay out, after all, is not an inference directly from 'God can be conceived to exist' to 'God exists'; it's an argument that we can conceive the concept 'God' [premise 1] but this concept would be inconceivable (because incoherent, [premises 2+3]) if God did not exist. Or am I reading the argument laid out in a different way than you intend?

    [I take it that this is fairly closely related to Mike's concern.]

    ReplyDelete
  16. Brandon - my response is that the alleged "contradiction" relies on the conflation. Once we avoid this confusion, we find that there's nothing incoherent in conceiving of a greatest conceivable being (de dicto, or 'as represented') that is not the greatest conceivable (de re, or 'in reality') due to actually lacking the requisite property of existence.

    ReplyDelete
  17. Hi, Richard,

    This cleaner way of stating it clarifies things quite a bit; if I recall correctly, this objection to the ontological argument is much like that of Prior ("On Some Proofs of the Existence of God"), and was specifically put in the terms you use here by Parsons (in _Nonexistent Objects_). Peter Millican has an interesting paper in which he argues a similar, but slightly different, sort of claim (although such claims are harder to press against Anselm, who has a rather sophisticated and complex theory of what you are calling conception 'de dicto' and conception 'de re', than post-Cartesian arguments such as the one you're looking at, which have a more naive notion of conception):

    The One Fatal Flaw in Anselm's Argument (PDF)

    But I'm worried that this might beg the question against the ontological argument. How do we know "that there's nothing incoherent in conceiving of a greatest conceivable being (de dicto, or 'as represented') that is not the greatest conceivable (de re, or 'in reality')"? Certainly the two come apart in other cases, but it seems that precisely what is at issue is whether they can do so without contradiction in the particular case of the GCB. So the question is: what actually prevents there being something so great that to conceive of it de dicto you have to conceive of it de re?

    Adriano,

    You might be interested to know that a (much more extensively developed) version of your "ontological argument, akin to the cartesian one" was in fact proposed by one of the Cartesians (Malebranche), and was a significant part in the big debate between Malebranche and Arnauld, with Arnauld taking Richard's line above and Malebranche arguing that Arnauld could provide no account of how a finite being could have an idea of an infinite being that did not presuppose that it already did.

    ReplyDelete
  18. What prevents there being something so great that to conceive of it de dicto you have to do a cartwheel? They're just different things. I don't even see any prima facie grounds for the proposed connection here.

    ReplyDelete
  19. I don't even see any prima facie grounds for the proposed connection here.

    Of course you do: lots of people have claimed that there is such a proposed connection on the basis of arguments like the ontological argument (there are other arguments besides the OA that make similar moves; e.g., Spinoza-like arguments on substance or Sommers-like arguments for the necessity of "Something exists", so it's not even as if the proposal of the connection is ad hoc), which could be taken to show that there is at least one case, perhaps more than one, where de dicto conception commits you to something de re, on pain of contradiction. Precisely what is in question on this point is whether the OA, or any arguments like it, can be regarded as successful in such a way; and it doesn't seem that you've done more than just deny that it can.

    We always have to distinguish between diagnoses of arguments and refutations of them. Your account is a reasonably plausible diagnosis of the type of OA you give; there are at least a few reasons to think it or something like it might be true, and plenty of reasons at least to take it seriously as a promising account of the argument, if certain things are granted. You are still far short of 'pinpointing' the error, to use the term you use in the post; all you seem to have done is given one plausible candidate (of several) for where it might possibly be pinpointed, assuming there is in fact an error, as most people suspect. This is quite worthwhile; but as a refutation it begs the question.

    ReplyDelete
  20. This comment has been removed by the author.

    ReplyDelete
  21. Brandon - no, really, I'm truly baffled. To recap: I sketched out a standard version of the OA, and showed it to be invalid (and that if we strengthen the first premise so as to restore validity, the new premise is question-begging). Your response, so far as I can make sense of it, is to say that even though I've shown the OA to rest on a generally invalid inference, maybe in this case it's really valid after all. I haven't seen any reason given for thinking it is so. So all I'm seeing here is special pleading: maybe my refutation isn't really a refutation. Maybe. But I'm not seeing any reason to take this skeptical proposal seriously.

    Anyway, I'm not interested in playing burden-of-proof tennis here. For my purposes, I'm satisfied that (i) my 'diagnosis' undermines the apparent validity of the OA (that's what makes it a refutation, by my lights). It now strikes me as very obviously invalid. And (ii) although one can always raise the skeptical possibility that an apparently invalid argument is "really" valid for reasons unknown, I'm not seeing any such reasons in this case. So unless you can spell out some positive reasons for thinking that the inference in this case is valid after all, I consider the case closed.

    ReplyDelete
  22. To recap: I sketched out a standard version of the OA, and showed it to be invalid (and that if we strengthen the first premise so as to restore validity, the new premise is question-begging).

    No, you asserted without any argument that it was invalid, or at least it looks like this may be what you've done. Your whole argument consists of making the distinction de dicto and de re and saying, "it may be possible for us to conceive of some X as supremely great (existing) without it actually being so" and then saying the same thing in different words: "This act of imagination is compatible with the being in question not actually existing". But one can argue that this is Seriously, go back and re-read what you actually wrote; it isn't obvious that you've done anything more here than simply deny that it is valid. Merely making a distinction and waving your hands doesn't magically get you a refutation. You have to show that the distinction actually does what you claim it does in this particular case. You have not shown that the form of the argument is generally formally invalid, since all you've noted is that there are times when the distinction makes the argument invalid. You have not shown that the argument is not materially invalid even if not formally invalid, and thus not shown that your 'proof' of invalidity is any stronger than accusing an argument of (say) making the fallacy of composition or division, where arguments are formally invalid except in certain types of domains. You have not shown what makes the OA different from thought experiments like, say, the zombie argument, which certainly looks like it begins with conceiving the possibility of zombies de dicto and concluding the possibility of zombies de re (and is regularly opposed with objections that are the same or analogous to those that are used against the OA). All you've done is suggest (over and over again) that your distinction would make the argument invalid, in terms that aren't obviously more than simply a denial of the possibility of ontological arguments.

    There is no burden-of-proof tennis; my point was simply that you have not done what you originally claim to have done, or, if you have done it, you have done it so obscurely and in such a roundabout way that it's not at all clear that you have done it.

    ReplyDelete
  23. Whoops; the incomplete sentence before "Seriously" should read, "But one can argue that this is just a denial of the possibility of ontological arguments, if there is no proof that one must accept that the distinction disallows inferences from one to the other."

    ReplyDelete
  24. Brandon, I suspect we're talking past each other again.

    Here's the important point: the version of the OA I presented was initially plausible-seeming. But once one understands the distinction I'm drawing, the argument no longer seems plausible, since the distinction dissolves the apparent contradiction in conceiving of a GCB without it existing. Hence the argument will lose its grip on (most of) those who understand the distinction I'm drawing here.

    You complain that my argument merely consists of drawing the distinction and saying, "it may be possible for us to conceive of some X as supremely great (existing) without it actually being so". Sure, and that suffices. Most people who had previously felt the pull of the OA will now, through appreciating this point, be able to recognize it as a sophism. All the plausibility of the original argument (to my mind, at least) depended on neglecting this very possibility.

    Note that if we flesh out the inference from 3 to 4, it goes something like this:

    3. For any being X to qualify as the greatest conceivable, X must exist.

    3b. So for us to be thinking of the greatest conceivable being in P1, that being must exist.

    So 4. God exists.

    But 3b is only plausible as long as we ignore my distinction (effectively conflating P1 and P1*). Instead, the most we can obviously conclude here is:

    (3a) in conceiving a GCB (de dicto), we must therefore be conceiving it as existing.

    Now, you can always speculate that maybe there's some (non-obvious) way to get from 3a to 4. I haven't seen any such argument, but you can always speculate that some other ontological argument could be offered to this effect. I don't mean to rule that out. I'm just saying that this argument has been defused.

    To repeat: I must emphasize that I'm not interested in "proving" that "the distinction disallows inferences from one to the other." If someone is inclined, even after understanding what I've said here, to still find the OA plausible, then I've nothing further to say to them (at least until they spell out the extra steps in the argument so as to make it seem remotely plausible to me again). I've made what I take to be an obvious point about one particular argument, a point that (I trust) most of my readers will immediately grasp. I just don't care about the rest. If the refutation isn't obvious to you (once understood), then you're not in my target audience.

    ReplyDelete
  25. (I should add that my objection here isn't relevant to the zombie argument. If you think otherwise, you haven't understood it. I might write another post clarifying the disanalogy at some point, since it seems to be a common mistake.)

    ReplyDelete
  26. Hi Richard,

    I posted earlier and got inundated with posts afterward that made me question my grasp of your post (all of which used really fancy language about philosophy). But after reading, I have realized that the posts don't add too much to the discussion, particularly because your claim, I think, is far less aggressive that people are making it out to be. Certainly this is no proof that God doesn't exist.

    I'm sorry that my earlier post does not address your post directly, rather other aspects of this ontological argument, namely what it means to exist.

    But, correct me if I'm wrong, the way I understand it is that the original argument is invalid because 4 does not follow, rather only 4* (or i guess your 3a from above) follows:

    4* So God [the greatest conceivable being] must be conceived to have the property of existing

    And thus, when we make the change in 1* so 4 will follow, but then 1* is dubious because there is no reason to believe that there is a greatest conceivable being out there, and even if there was a God that existed (which I believe), it still begs the question because you are assuming that it exists even though you are using that assumption to prove it's existence.

    Ya? I mean, Santa must have the property of existing when we conceive of him because only existing beings can man a sleigh and deposit presents under a tree. It just wouldn't make sense to conceive of a Santa that didn't exist (de dicto), likewise it doesn't make sense to conceive of a God that doesn't exist. That doesn't mean he exists though.

    ReplyDelete
  27. Hi Roscoe, yup, that's exactly it. (The ontological argument provides us with an additional reason why it "doesn't make sense" to conceive of a God that doesn't exist -- i.e., he would lack a 'perfection' and so wouldn't qualify as 'God' at all -- but this point by itself doesn't seem to have any implications for what beings exist in reality.)

    ReplyDelete
  28. Oh, and as for your argument strategy, I think it would help if you admitted that it may be possible for us to conceive of some X as supremely great (existing) and with it actually being the case that it exists, but only that this argument doesn't prove it logically. This would help weed out all the people that would attack your argument by saying that we aren't actually conceiving of the GCB if it doesn't exist de re (I mean, they totally don't understand what you are going for, but this could help them make the connection).

    I mean, I could conceive of a sweet car, and it must obviously exist for it to be sweet, and you know what? the car is right in front of me and I own it, so it does, in fact exist. But if I were to go up to some random person and tell them about this sweet car, it doesn't mean they have to believe I actually own it. This, again, might help an audience that may not feel comfortable with denying the existence of God (and thus may close their mind to your arguments if you present it the way you have).

    Just trying to help =)

    Roscoe

    ReplyDelete

Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)