Friday, July 16, 2004

The "Armstrong Ontological Argument" for God's Existence: "A concept greater than which first meets the eye"

by Dave Armstrong (10 March 2003)

The ontological argument, originally formulated by the 11th-century Christian philosopher St. Anselm, is fascinating and ingenious, has a long and illustrious history, and involves more than might be apparent at first sight. It's very subtle and requires one to think in ways which are not the usual, everyday modes of thinking and analyzing. But it is not impossible to grasp.

I give it the (tentative) old college try in my own formulation of it.

The atheist asks why anyone should accept the possibility of the maximally great being, and even that only in a possible world, not the actual one. This doesn't strike me as an extraordinary or unreasonable concession or admission at all. In other words, it is simply admitting, "God might exist in another possible world."

We imagine that any number of things may exist (especially in other worlds) that we don't believe exist in our world. The premise, one must always keep in mind, is only about a possibility. We posit the existence, of, e.g., extraterrestrial life, even though there has been no proof of it whatsoever. We reason that there are (demonstrably, I believe) thousands, millions of galaxies, and that it is rational to believe that life may have developed in another one besides ours. Many educated people believe this (and -- to note in passing -- it is not incompatible with Christianity, according to so prominent a Christian apologist as C.S. Lewis), but there is no hard evidence for it.

To accept the mere possibility of a God in some other possible world does not, it seems to me, involve much more "faith" than belief in extraterrestrial life. Every book of science fiction imagines another world that doesn't exist in fact. A book which was sheer nonsense and had no plausibility as another world whatever, would not succeed. Thus, it is arguable that most readers grant some possibility of this imagined world, even for the book to succeed as a piece of entertainment.

One could think of several reasons why it is reasonable to accept the premise of a possible God-Being: anthropological study, showing that human beings are overwhelmingly religious, and that they usually believe in one or more deities, or the fact that the existence of God has been a prominent aspect of philosophy from the beginning, and that many of the greatest thinkers throughout history have been theists or Christians (thus it would seem unreasonable to rule out the very barest theoretical possibility), or similarities in moral codes and individual consciences across various cultures (not to mention minds and consciousness themselves), which lead many to believe that there is a unifying Mind and Benevolent "Force" which lies behind all that.

To deny the premise, on the other hand, involves one in considerable difficulty, I think, because it is far too "certain" to remain plausible: it claims far too much for its own knowledge. Let's examine such a scenario a bit:
(anti-A) There is no possible world in which maximal greatness is instantiated.
How can a person know this? By what criterion is anti-A more plausible or worthy of belief than A? For to believe this would be (it seems to me) to believe the proposition:
(anti-A2) No such thing as God exists, and no such thing can possibly exist in any possible, imaginable, conceivable universe.
Now, if that is true, then why is the topic of God and theism so prominent in philosophy? If indeed theism were as silly and foolish as belief in fairy tales, leprechauns, unicorns, mermaids, centaurs, or other fanciful, absurd mythologies, why does the question continue to occupy great minds (both in favor of theism, and opposed to it?). One doesn't devote any time to sheer nonsense: Alice-in-Wonderland worlds or linguistic gibberish.

No one (with three brain cells) seriously considers as any possibility that the earth is flat, or that the moon is made of green cheese. If the notion of God is in that kind of immediately dismissible category, then it is quite strange that rational, thoughtful, intelligent people devote so much time and energy to it. Therefore, the rational person must (given all these considerations) grant the bare possibility of God in another possible world, and this is all that premise A of the argument requires.
I don't see that it is all that big of a deal to simply admit the possibility of such a God. It would seem to follow from a healthy intellectual humility or self-questioning. 

We are not infallible beings. We are fallible; therefore we can't place so much confidence in our own beliefs and mental processes that we can start ruling out possibilities of scenarios in possible worlds. If there is no possibility at all, why do Internet lists (and philosophy clubs and associations) devoted to the question of God's existence, exist, and draw very sharp people to them, willing to spend time on the question? How many lists and clubs and associations devote themselves to flat earths and moons made of green cheese? There may be such lists, but who are the members?!

The atheist might reply that they are trying to persuade the theist of the error of his ways, but does any round-earther spend time trying to dissuade flat-earthers? Do the greatest minds spend time trying to reason with the worst cases in a mental hospital? No, of course not. The very existence of the discussion proves that reasonable minds worth being reached and interacted with, believe in theism; therefore its bare possibility (and only in other possible worlds at this point) must be granted, and in effect, is granted, by the evidence of the very way that atheists (at least the more respectable and thoughtful ones) act and think, vis-a-vis theists and Christian thinkers. Actions speak louder than words (or thoughts). So let's revisit again what denial of A involves, and develop it a little further:
(anti-A2a) No such thing as God exists, and no such thing can possibly exist in any possible, imaginable, conceivable universe.
(anti-A2b) Such an inconceivable, unimaginable, impossible thing cannot have any conceivable, imaginable, possible rationales or defenses in its favor.
(anti-A2c) Things which have no conceivable, imaginable, possible rationales or defenses are not worth talking about; indeed, cannot be rationally and meaningfully discussed at all.
Conclusion: Internet lists and clubs and philosophical associations devoted to the question of God's existence are worthless, meaningless enterprises, if we accept premises anti-A and anti-A2 (a,b,c). That being the case, we should shut such endeavors down immediately and talk only about agreed-upon concrete realities. Or atheists can admit [prominent Christian philosopher] Alvin Plantinga's Premise A, in which case his argument can be allowed to proceed with (by his own appraisal), the largest hurdle removed.

The atheist might reply that a being could still be maximally great within its own world but not in all other worlds. But truly "maximal" greatness is greatness in all possible worlds. It is a matter of simple definition. "Maximal greatness" is not confined to one world alone. It transcends that limitation, because it is the greatest greatness imaginable.

Or the atheist might argue that it is not possible for moral perfection to exist in a morally imperfect world. But this would be a smuggling in of the notion of morally imperfect world, too early in the game. At this point, we are discussing all possible worlds, and remain in the realm of the hypothetical (but following unarguable logical principles). 

Secondly, such a criticism fails to distinguish between such a hypothetical being and the world in which it is found. The argument is far from creation, which then introduces difficulties like the problem of evil, and so forth. Moral perfection is simply one of the aspects of being maximally excellent. 

Thirdly, even if creatures of said Being were imperfect, that doesn't necessarily reflect badly upon the Being-Creator (which gets back to the thorny issue of free will of created creatures to contravene the will of their creator). These are all reasonable and important considerations, but far ahead of the actual argument as stated.

The characteristic of maximal greatness is not confined to one world. Even the laws of logic and mathematics are not confined to the actual world but (quite arguably) apply to all possible worlds. How could, for example, possible world #47 exist and not exist simultaneously, or have the property of relativity ubiquitously and also not have it, or be expanding and contracting simultaneously?

One atheist I interacted with readily admitted that "the conclusion does actually follow logically from (A)." He went on to deny (A), of course. But by his own words, if one can establish the rational credibility and plausibility of A (itself a mere hypothetical), then the argument succeeds -- if the goal is to show that theism is at least as rational as atheism (as Plantinga himself states). That's all I ever claim even for the cosmological and teleological arguments, which I consider the best ones in favor of theism.

Given what Plantinga already granted, of course the theist can also grant atheism as a logical possibility (even one which could be rationally established by modifying this very argument). The theist has no problem conceptualizing a possible world without God (the opposite of Plantinga's Premise A in one sense). That is a humdrum, unremarkable admission. What I find extraordinary is the denial of the very possibility of A: that God could possibly exist in a possible world. If the denial of A (or, anti-A) is easier to believe than A, then the atheist would have a point, but how does one argue that it is an easier thing to believe? Much of the point of the argument is to show that theism is every bit as plausible as atheism. Until I am shown why anti-A is more plausible than A, then I will continue to assert that this version of the ontological argument succeeds in its stated purpose.
It's hardly possible to either prove or disprove A':
(A') There is a possible world in which maximal greatness is not instantiated.
or another notion opposed to A; what I called "anti-A-2":
(anti-A2) No such thing as God exists, and no such thing can possibly exist in any possible, imaginable, conceivable universe.
But then it is also impossible to disprove A or hold that it is impossible in all possible worlds. So that places (it seems to me) theism and atheism on, at very least, equal rational ground, insofar as this argument is concerned.

One possible hypothetical doesn't preclude another possible hypothetical. Possibility is not actuality. Only the instantiation of one or the other precludes the contrary state of affairs. But Premise A (as I keep reiterating) was only about possibility. The theist says, "sure, it is conceivable that the universe might have existed, might have been eternal, without a God to create or oversee it." Anything is conceivable, if it isn't inherently nonsensical or self-defeating.

Why, then, does the atheist find it so hard to conceive of a possible world (and an actual world) with God? Most atheists I know claim that they would be willing to believe in God, given what they deem as a proper, compelling amount of evidence. How can one even be willing to theoretically believe in something they claim is inconceivable? They must already have the (sensible / non-nonsensical) concept in their head to even conceive that they might conceivably believe it, given enough verification and proof and philosophical plausibility.

The way I view the argument (which is apparently Plantinga's opinion, also) is that it shows both atheism and theism equally plausible and rational options, before anything else is considered. Of course, for me, the theist, the "anything else" is a massive amount of cumulative evidences which enter into the acceptance of Premise A. Neither side should be too confident about what philosophy or logic alone can accomplish. One also has to take into account plausibility and the mysterious process by which one arrives at axioms and premises in the first place. It is true of virtually any argument, that the premises (almost always) have to be supported elsewhere, and can be disputed. So this is no problem unique to the ontological argument. It's a worthy goal to show the equal rationality of theism with atheism, as this is routinely denied (and often assumed without argument).

Of course the theist believes that theism is more rational and plausible, and that atheism is ultimately irrational and implausible, but in terms of individual arguments, a rational equivalence is a good outcome. For an atheist to even admit such a thing is already a huge victory, because so much of atheist predisposition is based on the notion that Christianity is inherently intellectually inferior, "primitive" or "antiquated," based on mythology and wish-fulfillment, etc. ad nauseam.

I acknowledge that a world without God is conceivable, if one (theoretically only) starts from scratch, with no prior axioms of God's existence (on any grounds whatsoever). In other words, it is the provisional stance of a (non-theist, non-anything) skeptic, for the sake of argument only. Once one gets into the intricacies of the logic involved in the ontological argument and theism generally, there is a sense in which, indeed, a theist cannot admit this, because it would involve self-contradiction.

For my part, I was (above, in certain places) thinking in terms of conceptualizing possible, conceivable worlds which are other than what theists believe them to be in reality. If it is impossible for us to envision any possible realities other than the one we accept, then it follows that our view is well-nigh unfalsifiable, and amounts to an irrational fideism. As I have always opposed that, I must accept that I could be wrong and that other worlds are conceptually and actually possible. I do not believe this myself; only that things might have been other than what they are. There is a sense in which anything not logically impossible, is possible.

If we ask the atheist to do such a thing in reverse (i.e., conceive of a possible world with God, which is the requirement of Plantinga's first premise), it seems to me that we must do the same ourselves, in the opposite direction, strictly for the sake of argument, and an acknowledgement of what "possible worlds" means, in the broadest possible sense. I think my statements were and are permissible, in the sense just explained.

When I state such things, I am momentarily stepping outside an espousal of the ontological argument or any other argument for theism, even theism itself, for the sake of argument and pure conceivability of other worlds. In fact, the ontological argument itself allows this (as I understand it) because it seems to offer the choice of God as necessary or impossible. If that is the choice that the logic of the ontological argument (viewed as a reductio ad absurdum) entails, then certainly I can conceive of the atheist state of affairs, while not accepting it myself.

Arguing for the basis or non-basis of any premise of any argument is a distinct endeavor from the argument proper itself. Therefore, in arguing for premise A of the ontological argument (however presented), one is not necessarily bound to the logic of the ontological argument itself. One is not yet "in" the conceptual and logical framework of the ontological argument (ontology and modal logic).
What one can conceive as possible is not the equivalent of one's own belief. One can certainly conceive of a world in which a necessary being did not, in fact, exist. I can conceive of a world in which the Godhead subsisted in four persons rather than three, or one where Jesus died by drinking hemlock, like Socrates, rather than by being crucified. Therefore (if one grants this), one can also conceive other worlds and argue within those theoretical frameworks, in order to look for inconsistencies in opposing arguments.

I hasten to add that this is thought within an exclusively philosophical framework. The Christian, however, believes in attainable knowledge beyond mere philosophy (revelation, experience, legal-historical knowledge, etc.). Theologically, as an orthodox Christian, I don't believe that God's existence is contingent or optional at all (nor does St. Thomas Aquinas or any Thomist). The Christian believes that God is the self-existent being and could not not exist. That is (orthodox Christian) framework in which Anselm begins, because he believes in the impossibility of absolute separation of faith and reason (according to the usual medieval synthesis of faith and reason). Faith and reason exist in harmony and do not contradict each other at all.

I wholeheartedly agree, but methodologically, I think it is possible to temporarily separate the two true forms of knowledge, in making particular arguments. This doesn't entail a suspension of one's beliefs; it is only a methodological matter. One can play the game of philosophy as if atheism or skepticism were true, in order to examine arguments. In response to the argument against the espousal of the proposition, "God might conceivably exist in other possible worlds," I reply that His non-existence was conceivable in some possible world.

It is not yet arguing in either the paradigm of the ontological argument or the larger Christian one, to discuss with an atheist the plausibility or non-plausibility of a premise of the ontological argument, which he himself is considering adopting or rejecting. For premises are necessarily (it seems to me) accepted on grounds other than those established by the arguments of which they are but the beginning-point. Otherwise, circularity would obtain.

I think we must distinguish between the following two sets of propositions:
1a. Necessary beings must exist. God is such a being (by definition --the very meaning of the word); therefore He exists.
1b. There is such a thing (in possible worlds) as a necessary being, but whether such a being exists or not is a separate issue to be determined. Even if God is defined as such a being (even uniquely so), this does not yet prove (by reason or logic alone) that this God exists. This may be a world such that a given necessary being is not necessarily existent.
Philosopher Graham Oppy, in his paper, On "The Ontological Argument": A Response To Makin (1991); originally published as "Makin On The Ontological Argument", Philosophy, 66, 255, January 1991, pp. 106-114 (, makes the same point as my 1b above:
. . . the discussion of ontological arguments needs to be carried out in the context of a modal logic which allows that accessibility relations between worlds are not--e.g.--symmetric (so that one can say that it is possible for a state of affairs to be necessary and yet for it not to be the case that that state of affairs actually occurs).
2a. God is a contingent being Who happens to exist (just as my four children do, but wouldn't have if I had never met and married my wife).
2b. The God Who is a contingent being does not, in fact, exist (in the same way that unicorns might, but don't exist).
The ontological argument excludes 2a and 2b as matters of definition (God by definition cannot be a contingent being). Christianity does the same. Atheists, of course, may define even the theoretical God Whom they disbelieve in such terms. If they claim to be arguing against the God of Christianity, they cannot, of course, do so , because that would entail a fundamental confusion as to the nature of the God with Whom they claim to be dealing in their argument.

I think some of the confusion regarding the ontological argument lies in the distinction between 1a and 1b. Christians believe both that God exists, and that He cannot not exist. He is pure Being or Existence (unlike ourselves, who are His creatures, and entirely contingent upon His decision to create us). Thus, all Christians accept 1a as a matter of course. But Christianity is not philosophy. It may be consistent with true philosophy, and not irrational or incoherent at all (I certainly believe so), but it is something different from philosophy per se. Philosophy simply does not constitute the sum of all knowledge.

Thus, in a philosophical world, apart from the prior beliefs of Christians (which presuppose 1a and therefore exclude in actuality 1b -- including St. Anselm, based on Christian theology and belief), one might (wrongly) deny that the being Who is by definition maximally great and self-existent, exists (on various other grounds). In other words, the atheist is not bound by Christian or theistic assumptions (that may be arrived at either philosophically or non-philosophically).

God is a perfect, necessary being, by definition, but atheists need only deny this definition of God or deny that such a God exists at all in order to escape your statement: "God necessarily exists because contingent existence . . . cannot be a property of a perfect being." The atheist will try to deny that a maximally great being Who possesses these characteristics in the first place, exists in the first place.
I think they are dead wrong, of course, and that they cannot establish this rationally or conclusively, but they are within their "logical rights" to do so, because conceivability does not exclude even a necessary being from existing in the first place. Even the ontological argument (as stated by its advocates) establishes either that God is a necessary being or that His existence is impossible.

It is precisely for this reason that the premise of Plantinga's ontological argument is so controversial, and why even he admits that the argument does not prove God's existence, but only that theism is rational. The atheist's difficulty, on the other hand, is to prove that such a being is inconceivable in actuality or in any possible world (Plantinga's first premise). I don't think this can be done, and -- that being the case -- Plantinga's argument succeeds in demonstrating the rationality of theism, even though it is not a flat-out proof of theism or disproof of atheism.

I firmly believe that God exists, but on many other grounds. I would agree with Plantinga that belief in God is "properly basic," and epistemologically equivalent to belief in other minds. On that basis it is entirely warranted and not opposed to reason at all. And that is how every person (of whatever intellectual capacity) can believe in God, without having to master elaborate reasoning like the ontological argument or even the more-easily understood teleological and cosmological arguments.
One can distinguish belief, logic, and conceivability. I believe in one sort of world; I can conceive of many possible worlds, including one without God (I could also conceive of many and accept none; taking an agnostic position). 

I don't see how anyone can deny this. It is presupposed every time one sits and watches a science fiction or fantasy movie or reads a novel along those lines. It is assumed (consciously or not) every time one truly understands another point of view that they are critiquing. For how could someone argue against theism if they don't have the slightest clue what it is they are arguing against (since they can't comprehend or "conceive" it)? That makes no sense. But one can obviously conceive situations that would be contradictory if they both existed, as long as one doesn't assert that they exist simultaneously.

I'm not yet convinced that merely conceptualizing something and calling it by a certain defined word reaches to the level of logical necessity and existence. That's where my evidentialist and empiricist nature stumbles. But that's what makes consideration of the ontological argument fun, too, because it is so different from the way I (and many other Christian apologists and philosophers) usually think and analyze things.

Philosophy and faith/religion are two different things. Philosophy can only go so far. The atheist has no supernatural faith, so he is confined to the world of strict philosophical speculation (and scientific knowledge, which is also a branch philosophy, so it all reduces to philosophy). In faith, all Christians think God's nonexistence is unthinkable, of course. The very concept of God demands this. But I don't think that can be established by philosophy apart from faith. 

I believe that St. Anselm would agree with me because his thought on the ontological argument is strictly within the context of religious faith. He is very explicit about that in his writings. St. Thomas Aquinas, on the other hand, is relatively more concerned about writing to folks who don't share the same faith (especially in his Summa contra Gentiles). Thus, he tries to find a different epistemological starting-ground. Empiricism makes more sense in that situation, since it is something "other" which can be verified apart from a precommitment of some particular worldview or faith.


Joseph Saffioti said...

This is all very interesting. I've had a preoccupation with Anselm's ontological argument for quite some time and go back and forth. I wonder though, if it is more of a prayer rather than a proof?

I have a sense that St. Anselm might find all this confusion and discourse concerning his Proslogion somewhat amusing!

The challenge I think in this argument is in the definition of God. Anselm's starting point is theological faith. He knows by faith that God is "He who is" and therefore nothing greater can be conceived. This theological principle illuminates his entire argument, and leads him to valid conclusions. The argument is not false given its premises.

The problem is that the "definition" of God presumed by Anselm cannot be known apart from theological faith. In fact, while our usage of the word "God" can be defined, God himself cannot be defined by the human mind strictly speaking, since He is not an object, but a subject- specifically three persons.

This raises the question of motivating someone to accept our definition of God and of admitting at least the possibility of an all-knowing, all-benevelent, and all powerful God, supremely one and personal. In the end though, we are not dealing with reality as such, but with definitions, and what follows from definitions.

Not that Anselm committed this fallacy. I would argue that Anselm's starting point was not a definition of God, but rather Divine Revelation which cannot deceive. I think its possible that later interpreters took his "proof" the wrong way, supposing that Anselm was arguing from definition to reality, as opposed to understanding that Anselm was taking a theological principle as his starting point and sketching out its necessary conclusions ("faith seeking understanding").

That being said, St. Bonaventure refined Anselm's proof even further: 1) God is necessary existence; 2) existence must exist; 3) God exists.

Again, we can accuse Bonaventure of shoddy argument, just like Anselm. But in fact the argument is valid once one accepts the definition of God. Yet, the definition of God is derived from Revelation, not from human reasoning. One wonders whether St. Bonaventure or Anselm are simply shunning the vanity of human reasoning when setting forth their respective proofs?

What I struggle with is the assertion that Anselm even meant this to be a demonstrative proof? And even if we examine the ontological proof, what we have is a proof derived from definition. However, that doesn't really seem to prove anything- it only proves that "it follows from this understanding of God that He exists..." In other words, on a strictly rational plane, what we have is a valid argument. Yet valid arguments are not necessarily "true" arguments.

For that reason, I still keep coming back to the traditional cosmological proofs, along with some of the more recent arguments from moral and subjective experience. Perhaps the mere personal and subjective experience of raising the question about God's existence contains some basic clues as to the existence of a personal God?

Just some thoughts..~JS

Joseph Saffioti said...

I guess I would also add that this seems to be St. Thomas' objection to the use of Anselm's Proslogion as a demonstration of God's existence. Aquinas doesn't seem to object to the truth of what Anselm is saying, but rather to employing his words as a demonstrative, a priori proof. One can draw valid conclusions from definitions; but one cannot prove the existence of things from mere definitions.

One can define a circle, but the definition does not prove that a perfect circle exists in reality. This is all Plato and Aristotle. The neo-Platonists of his day confused these distinctions and misinterpreted Anselm so as to suppose that God, whose essence is not knowable by the human mind, could be known in a self-evident way once the definition is granted.