(UPDATE 17 October: The commenters have convinced me that the argument in this post is no good. But I'll let the post stand because some of what I say is right, and because I don't want to deprive my critics of their target. My argument has this most excellent anti-Continental virtue: it is clear enough to be refuted.)
The atheist’s denial of God must proceed via his concept of God: since his whole point is that there is no God, his claim cannot be about God, but only about his concept of God. Otherwise, he presupposes the very thing he is concerned to deny. What he is saying about his concept of God is that it is not instantiated: nothing answers to it in reality. In general, denials of existence, if true, cannot be about the things whose existence is being denied; they must be about some surrogate item (a concept, description, property, propositional function . . .) and must say of it that nothing answers to it. In this respect, God denial is not different from Cerberus denial. Given that Cerberus does not exist, one cannot say of him that he does not exist; one can only say of one's concept of him that it is not instantiated.
But suppose the atheist’s concept of God is mistaken. Then its failing to apply to anything would not show that God does not exist. God might still exist even if every extant concept of God is uninstantiated. For every extant concept of God might be mistaken or inadequate.
Consider a more mundane example. Suppose you hold the Expert title in chess, but I mistakenly think that you hold the Master title. Then my concept of you, which includes the subconcept chess Master, fails to apply to you. It fails to apply to you because one of its subconcepts fails to apply to you. Indeed, it may be that nothing is in joint possession of the properties included in my concept of you. But you still exist! You would still exist even if no one had an adequate concept of you.
Second example. Suppose your concept of the Holocaust includes the subconcept only genocide to have occurred in modern times. Clearly, this concept of the Holocaust is not instantiated since other genocides (see here for a possible example) have occurred in modern times. Does this give any aid and comfort to Holocaust deniers? Should we conclude that the Holocaust did not occur? No. All we can validly conclude is that the Holocaust so conceptualized did not occur. But all that that means is that the concept is faulty.
Of course, the case of God is much more difficult: we are not dealing with a physical object or empirically documentable event. Nevertheless, the point remains that the noninstantiation of the various extant God-concepts does not show the nonexistence of God. For it may be that each of these concepts is defective.
Our finitude insures that our concepts of even the most mundane things and persons will be inadequate. And if my concepts cannot capture the haecceity of my friend Harry, then a fortiori they will not be able to capture the haecceity of a being whose omni-attributes and transcendence remove him from the realm of the finite. Saying this, I am not presupposing the existence of God; I am saying something about the nature of God.
Only if the atheist’s concept of God is adequate, is the noninstantiation of this concept equivalent to the nonexistence of God. Only if the atheist's concept of God completely and with total accurancy captures the essence of God would the noninstantiation of this concept be equivalent to the nonexistence of God. It follows that the atheist must have adequate knowledge of what God is in order to deny that God exists. To have such knowledge the atheist would have to be quite the theologian! Clearly, the atheist lacks such knowledge: all he has is his concept of God, a concept that he cannot be sure is anything more than a concoction of his intellect. The atheist lacks such knowledge because we all do. The theist is in no better epistemic position: he operates with a concept of God that he cannot be sure is adequate to the divine essence.
A sophisticated theist like Aquinas does not claim to know what God is, but only that God — whom he refuses to pin down conceptually — exists. The asymmetry, then, is this. To deny God — God himself and not a mere figment of one’s imagination — one must know God’s essence. But that is what the divine transcendence makes it impossible for us to know. To affirm God, however, does not require knowledge of the divine essence, but only knowledge that there exists (in plenary mode) Something on which everything else depends for its existence (in derivative mode).
But then do I know what I am talking about when I say that God exists? Not exactly. My reasonings are gestures in the direction of the Transcendent, empty reasonings that one hopes will one day received intuitive fulfillment. The divine reality cannot be captured in theological formulas. Religion remains more quest than conclusions.
Great post. I have one question though. Can't the atheist disprove the existence of God by showing the impossibility of a certain attribute it is thought that He would require (such as omniscience) or by showing the incompatibility of two or more of the attributues it is thought that He would require? Coincidentally, Alan Rhoda has an interesting series on the second case (that of incompatibile properties) going on now. (See also his "Is Theism Internally Contradictory?" post for an interesting discussion.)
I'm afraid I don't see a meaningful asymmetry here. If the theist can assert that some vague concept of God is instantiated, why can't the atheist assert that the same vagueness is not?
You argue that the theist needs not understand the transcendent nature of God to assert that God exists, because all he has to do is claim that "There exists Something on which everything else depends for its existence (in derivative mode)."
I should think, though, that an atheist, with equal confidence, might assert that "It is not the case that there exists Something on which everything else depends for its existence (in derivative mode)."
Neither party is likely to adduce any compelling evidence; both claims are declarations of faith.
I can't see how denying the existence of God in any way depends on one's concept of God. Why think that 'God' refers to God only if I happen to have the right concept in mind? Why think 'God' manages to refer to God only if someone has the right concept of God in mind? I manage to refer to many of the presocratics by name, though I have almost no concept of them, and certainly no accurate concept. I manage to refer to King George IIII, though I might well be mistaken in everything I believe about him. These of course are just Kripke observations. So having the right concept is unlikely to be a necessary condition of successful reference. Same thing goes for denying that a term refers. When the atheist denies that the Judeo-Christian God exists he is claiming that no one in that tradition is refering to any being with the term 'God', quite apart from the concepts those in the tradition happen to have.
But what Bill means to be claiming in these remarks is not that the atheist cannot deny the existence of God without having a perfect concept of him. It is not a question of succeeding in refering to God. Rather, it is a question of being justified in denying the existence of God. Bill's claim, properly understood, is that the atheist can't be justified in denying the existence of God unless he has the right concept of God. Only if his concept of God is exactly right can he say that the non-existence of instances of that concept entails the non-existence of God.
Thanks for the comment and for referring me to Alan's interesting post. You employ the phrase "it is thought that He would require." This amounts to saying that our concept of God -- which I distinguish both from God and from the essence of God -- includes certain subconcepts, say, justice and mercy. But all-justice and all-mercy seem incompatible, as Drange argues. Well, suppose they are. It then follows that that particular concept of God cannot be instantiated. But that does not show that God does not exist; it shows that God conceived in precisely that way does not exist.
Your second comment is indeed very troubling. Every God-concept will include as a subconcept immaterial being. If this subconcept fails of instantiation, then every God-concept fails of instantiation. So I was wrong to say the following:
The atheist needn't have adequate knowledge of the divine essence, all he needs is knowledge that a subconcept like the one mentioned above is not instantiated. So he needn't be much of a theologian. But he has to be one outstanding metaphysician since he has to know that the whole of being is such that there cannot be an immaterial being. He needn't have an adequate concept of God to deny the existence of God, but he must have an adequate concept of being -- of what it is to be or exist -- to deny the existence of God.
So although my original argument fails, perhaps a more complex one will succeed. What I said in the early paragraphs of the post seems correct: showing that a particular concept of God is not instantiated does not show that God does not exist. But to show that NO concept of God is instantiated it seems that one has to know that reality is such as to exclude the very possibility of God. Apart from being question-begging, this knowledge is as difficult to obtain as the knowledge that God exists.
This was my point; that the theist's and the atheist's assertions are on symmetrical footing - they both stand on faith.
Well, one would have to have some concept of God, wouldn't one? Say, the concept of an immaterial creator of the physical universe.
Suppose Kripke is right and reference is not routed through sense (or concepts or definite descriptions, etc), but is direct. How then does 'God' refer to God? How can a causal chain lead from a use of 'God' to a being ouside space-time?
How does a Kripkean analyze negative existentials? You suggest that 'God does not exist' be analyzed as '"God" does not refer to anything.' But the nonexistence of God cannot depend on the existence of names, can it?
I think my argument is lousy. I haven't made a case that the positions are not symmetrical.
But do you suppose that most atheists insist that God cannot exist? Certainly that was never what I claimed in my atheistic days. I think that most atheists are content to believe that God simply does not exist.
How refreshingly gentlemanly, and how rare in blogospheric discourse. My respect for you deepens once again.
If you say that God does not exist, then that allows two interpretations: (a) God does not exist as a matter of contingent fact; (b) God does not exist as a matter of logical (metaphysical) necessity, which is to say: God is impossible.
Now if one means by 'God' a being that has all perfections, a being supreme in every respect, worthy of worship, free of all defect, "that than which no greater can be conceived," (Anselm) then it would seem to be built into the very concept of God that God exist necessarily if he exists at all. In other words, mere contingent existence is a defective mode of existence. On the other hand, if God does not exist, then he is impossible.
For creatures, the alternative is contingent existence/contingent nonexistence. For God, necessary existence/necessary nonexistence (= impossibility).
Thanks. Intellectual honesty is sine qua non for any serious discussion. One must admit when one is wrong when one has been shown to be wrong. But one shouldn't give in too quickly. It's little like chess. In a dead lost position one should resign and not hope for a heart attack on the part of the opponent, but as long as it is unclear one should fight on.
I wouldn't say, though, that the theist's and atheist's positions both rely on faith since each can marshall argments -- it is just that the arguments seem to balance each other out. Only one of the positions can be true, but I think both are reasonable.
Suppose Kripke is right and reference is not routed through sense (or concepts or definite descriptions, etc), but is direct. How then does 'God' refer to God? How can a causal chain lead from a use of 'God' to a being ouside space-time?
Yes, good. One clear way is through Abraham who was spoken to directly by God. There is a causal connection, supposing He did so talk. Another way is through the prophets and saints. These are our causal connection to God, if there is one at all. The causal chain is then transferred downward from Abraham to us. That is how we refer to God using 'God'. And that is indeed the best theory of reference on offer. But that is precisely what the atheist is denying. He is denying that by any name or none the word 'God' refers to God.
This is not a question of having a good argument or justifiation that God does not exist or any reasonably articulated conception at all. The atheist need not have any such thing. He can simply and reasonably find the whole story incredible. And indeed, it is an incredible tale. This is not to deny that many of us nonetheless "cred" it.
I thought you might bring up Anselm's argument, although of course the God that possesses "every possible perfection", and so must exist necessarily, is not the only imaginable God. The atheist can simply say: "Oh, that's not the God I was talking about; I just meant one that created the Universe, sees all, answers prayers, and so forth. Such a God could exist, I suppose, but I believe He doesn't. The one you are talking about, of course, doesn't exist either - but necessarily so, if you like."
This, of course, is exactly the issue you discuss in your post, but is a completely consistent position for an atheist to take.
I am reminded of:
This sounds right. So the idea is that the first use of 'God' or whatever is first fixed by the use of a definite description, but the reference of the proper name is nonetheless direct and nor routed through the sense of a description?
Malcolm,
I would argue that no God worthy of the name could be a contingent being. This is not yet the ontol arg.
Fair enough; there are still many places for the atheist to stand. And of course, we can disagree about whether a God worthy of the name can be a contingent being; for some atheists I'm sure that a contingent God is a perfectly sufficient candidate for disbelief, and without question there are atheists who are quite content to deny the existence of God without appreciating the arguments in favor of His non-contingency.
What could that mean? I have no idea what a term being causally connected its first use would be.
He may well have established the reference of the term via a definite description. He may have said, "Let 'God' refer to the being than whom no greater can be conceived,"
Kripke states that such definite descriptions might be used to "fix the referent" of the term. I'm not sure if that is what you mean by "establish the reference" of a term. And his examples are always ones when I am in proximity of the referent. Bill's worry was how we might deploy a causal theory of reference when the referent is not causally related to us. I do not get a causal connection by simply saying "let 'God' denote the greatest possible being" any more than I get a causal connection by saying "let 'Joe' refer to the tallest being in the closest possible world to ours in which I am 8' tall". We are not causally connected to possible worlds and we are not (obviously) causally connected to the supernatural world. So I intended to establish such a connection through Abraham, etc.
Not sure where to place this. I’m reading Ebon Musing’s essay “A Ghost in the Machine” (http://www.ebonmusings.org/atheism/ghost.html) and wondering if you either have or could comment sometime on such an argument for atheism.
Thanks,
Tom
That is right. And I agree with what you said above: there is still much that is good and right in your post.
Mike,
You write:
I would have problems with that too. That's why I said: "What is necessary is rather that our use of that term is causally connected with the first use of that term." I don't see any special difficulty with that. Suppose I say "Bob is a nice guy." My use of the name 'Bob' here is causally connected with some earlier use of the name, perhaps when I heard someone use it, and that person's use of the name is causally connected with some earlier use, and so on until we get back to the first use of the term, which would have been the use that fixed its reference. (Yes, I take establishing the reference to be the same as fixing the reference.) As long as you don't distort what I said, there seems to be no real problem here.
Kripke's examples of reference fixing may have all involved referents that were in proximity to the reference-fixer, but as long as the reference can be fixed through definite descriptions, as you concede, there is no reason why the referent need be present. A theory of reference of the sort sketched by Kripke does not require a causal connection with the referent, only a causal connection with the fixing of that reference. So the worry you attribute to Bill is unjustified, and what you say to allay it unnecessary.
(1) Bill's worry is expressed by Bill. See [10.17.2006 3:47pm].
(2) I say that reference can be fixed using a definite description, but I don't see how it follows that, as long as that is true, "there is no reason why the referent need be present." In the initial dubbing ceremony I am causally related to lots of objects; the definite description permits me to discriminate among those. In the famous Gareth Evans case, I have to dsicriminate between a set of twins in naming them. I can't stand in the next room and dub one of them 'Bob' by saying "I dub the one nearest the exit 'Bob'". For, if I am given the other child, the name 'Bob' will refer to it, not the other. Further, in the dubbing ceremony I can get the description wrong and still refer to the intended object. So the description is not doing the semantic work.
(3) Even in cases where the reference is fixed by a defnite description (Kripke I think talks about the reference to Neptune being fixed by a definite description that refers to the perceptible effects of Neptune) there is either a perception of the object or a perception of the causal consequences of the object (as in the Neptune case)
(4) When it comes to sophisticated causal theorist (Devitt, say) there have to be lots of perceptions of the object and lots of dubbings. I have no idea on a causal theory how a singular term is supposed to get connected to the right object if that object is not causally related to the initial referer via some causal relation (perception, for instance, or indirect perception). I have less of an idea how it might happen for objects to which we could not be causally related.
rel="nofollow" href="http://plato.stanford.edu/entries/reference/#CauThe"
In particular note section 2.3 at which Marga Reimer discusses the dual advantage of Evans' hybrid theory.
The advantages of Evans' theory appear to be considerable. Evans himself claims that his theory effectively combines the virtues of the descriptive theory with those of the causal theory, while avoiding their respective vices. Like description accounts, it accounts for cognitive significance (of the sort evidenced by sentences like (2) through (5)) as well as reference; like causal accounts, it preserves the intuition that one cannot refer to something with which one has no causal link whatsoever. Moreover, Evans' theory avoids the problem of ignorance and error. For it denies that reference is determined by ‘fit’ or ‘satisfaction’ of any sort of descriptive content. (emphasis added)
So, as I've noted, for reference to succeed on causal theories the initial dubbing ceremony must ensure a causal connection to the object dubbed. Uttering a definite description about God obviously does not establish any causal connection.
I agree that what you say about causal theories of reference is true of most causal theories of reference. But it is not true of the theory sketched by Kripke in Naming and Necessity. His is a causal theory in the sense that there must be a causal connection between the first (reference-fixing) use of the word and later uses of the word, not between the person fixing the reference and the referent itself.
I'll have to review the literature, but if there is one case of reference in which the reference is achieved not by causation but by the satisfaction of a description, then it would seem that the causal theory presupposes the description theory. Kripke seems to say the opposite of this, however, in Naming and Necessity, p. 97, n. 44.
Thanks for the pointer to that note. I don't follow your claim about the one theory presupposing the other. If there is one case of reference in which the reference is achieved not by causation but by the satisfaction of a description, then it follows not that the ancestral chain view of reference outlined by Kripke presupposes the description theory, but only that it is consistent with a description theory of reference fixing. (Cf. p. 96, n. 42: "the primary applicability of the description theory is to cases of initial baptism.")
Mike,
Your (2) and (3) above are in tension. You grant that
(i) Reference can be fixed using a definite description.
But do not see how it follows from this that
(ii) There is no reason why the referent need be present.
Well, (ii) fails to follow from (i) only if fixing reference via a description requires that the referent be present. Otherwise the referent need not be present. But then in your (3) you grant that on Kripke's view reference can be fixed in the absence of the referent, so long as there is some acquaintance with its causal consequences, as in the Neptune case. So does the referent have to be present, or is it sufficient that it have causal consequences with which the reference-fixer is acquainted? (Surely Neptune was not in any reasonable sense present to Leverrier.)
If your position is that on Kripke's view there must be at least a perception of the causal consequences of the referent in order for a description to succeed in fixing the reference of the name, then we still do not need an Abraham in order to succeed in refering to God. It would be sufficient that the person who fixes the reference of 'God' be acquainted with some causal consequences of God (e.g., the physical world) in the way that Leverrier was acquainted with the causal consequences of Neptune.
But to reiterate, there is no evidence--or at least none has yet been presented--that on Kripke's view reference-fixing requires "either a perception of the object or a perception of the causal consequences of the object." I would be happy to see such evidence if it's out there, but I don't think it is.
Interesting. I said "one clear way" reference to God can be established is through Abraham. There might be other ways and your suggestion is an interesting one. But it raises an interesting worry. When I try to name Neptune indirectly as the cause of some of observed perturbations, I might wind up naming God. As with all other contingent events/objects the cause of the perturbations is finally God. But there is proximate cause talk that would help here, I suppose.
2. Disallowing comments from a particular person, or deleting an offensive, off-topic, or otherwise substandard comment, has nothing to do with censorship. People who think otherwise confuse censorship with lack of sponsorship. I am under an obligation not to interfere with anyone's exercise of legitimate free speech rights. But I am not under any obligation to aid and abet anyone's exercise of free speech rights, legitimate or illegitimate.
3. The Comments area is not an open forum for anyone to say anything about any topic. As the name implies, it is primarily for commenting on the author(s)' posts. But to comment on them, one must have read them. And if I have spent three hours on a post, a reader will not understand it in thirty seconds. Secondarily, the Comments area is to facilitate civil discussion between and among commenters as long as the discussion remains on-topic.
4. Some undesirables: The skimmers, those who cannot read but only read-in. The sophists who, abusing argument, argue for the sake of argument. The ideologues, those who are out for power, not truth. The uncivil. The illogical. The politically correct. Worst of all, perhaps, are those who exemplify the anti-Socratic property: those who think they know what they don't know. If Socrates was famous for his learned ignorance, these types are marked by their ignorant unlearnededness.