Wednesday, October 26, 2005

Prove to me that God exists and I'll believe it!

This is slightly revised version of talk given to Rice University Undergraduate Philosophy Society, 11/4/91.

I take the title I have been given for this talk to express the attitude of certain people who do not altogether rule out the possibility that there is a God but see nothing likely to convince them that there is. That is the typical attitude of the agnostic. And so it is to the agnostic that I shall address myself.

Not, however, to just any agnostic; for there are some who cannot be fruitfully addressed. Some, for example, cannot say what, if anything, would convince them that there is a God. I have found that most such agnostics so limit what might count as evidence or proof as to preclude evidence or proof that there is a God. That kind of narrow-mindedness is usually impervious to argument. But most of you are probably not guilty of it, for it is quite rare—except among academics who pride themselves on their broad-mindedness.

More often, agnostics have an mistaken idea of what would count as God. Consider the painful fact that we live in a world in which the wicked often prosper and the innocent often suffer. Many agnostics think that even if there is some entity responsible for the way the world in general is, that entity is unworthy of worship and obedience—and hence cannot be God. That is one way of posing the so-called "problem of evil," a problem that theists must and do confront. But it ought at least to occur to such agnostics that if there is an all-powerful, all-knowing, perfectly good God, he might see fit to arrange the universe quite differently from how they would in his place and do so in a way that is unclear to them. Until they take that possibility seriously, their idea of God remains so anthropomorphic that it would be pointless to try to persuade them with arguments that something called "God" exists.

Admittedly, some agnostics are free of such errors. Their idea of what would count as God is pretty close to that of the theistic mainstream, and they have a more or less defensible notion of what would count as a compelling case that there is such a God. It's just that they have yet to find any case compelling. Such open-minded, clear-headed agnostics are the only sort to whom a theist may usefully propound an argument that there is a God. Some of you, I hope, count as such agnostics.

Now the question what would succeed in persuading anyone in particular that God exists is a highly personal one. So, all I can do as a theist in a setting such as this is to try to present an argument that is cogent in an objective sense: one that really does establish its conclusion (or at least render it probable), and thus should persuade those who understand it even if, for whatever reason, some people remain unpersuaded. Such an argument would be a proof that God exists. It might fail to prove to you—the open-minded, clear-headed agnostic—that God exists; but if my argument is objectively cogent, then the failure is as likely to be yours as mine.

Very well then: what criteria must an objectively cogent argument satisfy? It is easy to answer that in broad terms. The first criterion is that of validity: an argument is valid just in case its conclusion follows necessarily from its premises if it is a deductive argument, or its conclusion is rendered highly probable by its premises if it is an inductive argument. The second is that of soundness: an argument is sound just in case it is valid and all its premises are true. The third criterion is that of independent decidability. If an argument is to be objectively cogent, then the question whether it is sound must be decidable by some reliable, publicly accessible method that does not require prior knowledge that the conclusion is true. That essential since an argument that we can't certify as sound without prior knowledge of the conclusion is not useful as an argument.

Now we can decide the validity of many sorts of arguments by the reliable, publicly accessible methods of deductive and inductive logic; and it is in fact quite easy to produce a valid argument for nearly any meaningful claim. And so my main task is to show how the premises of some valid argument that God exists can be verified by some reliable, publicly available method that does not require prior knowledge that God exists. The main difficulties may be raised in two questions: by what method might we look to verify such premises, and by whose judgment would such premises thereby be verified? I cannot answer those questions tonight; I can only point you in the right direction by constructing the right sort of argument, i.e., one that I think satisfies the criteria I have cited and allows for good answers to the questions I have posed.

Before proceeding further, though, I must dispose of a major red herring: believers as well as unbelievers often reject what I want to do as impossible, irrelevant, or both. Indeed most people these days are given to saying that belief in God is a matter of faith rather than of reason. And though I find that most who say so are unclear about what they mean either by 'faith' or by 'reason', they are clearly negative about those instances of reasoning that consist in philosophical arguments for the existence of God. That is only to be expected from unbelievers; but ironically enough, believers will often go further.

Some Christians, for instance, accept as revealed truth the doctrine of total depravity, thus holding that human reason unaided by revelation is too corrupt to attain any knowledge that it would facilitate one's salvation to have. From that point of view, arguing as I shall do appears almost as a kind of impiety. And even many believers who would admit that my aim is not impious are inclined to consider it irrelevant. How many people, after all, come to believe in God primarily as a result of philosophical arguments? You may be sure they are few. To most people who pay the matter any notice at all, natural theology—i.e., that branch of metaphysics which aims to achieve knowledge of God without taking anything from divine revelation as a premise—seems like a quaint if occasionally irritating sideshow to the real business of ordering their minds, hearts, and lives.

Well, natural theology is not for everyone. But it is for many of you, else you wouldn't be here. You want to know just how much there really is in the whole business. I think there is a lot in it. That is what I want to begin showing you.

As many of you probably know, there are traditional clusters of arguments for the existence of God. Each cluster comprises many arguments, but the clusters themselves are surprisingly few. In my view, arguments from each cluster contribute to what the literature terms a cumulative-case argument for theism, and it is just such an argument that I shall present in briefest outline at the end. In aid of formulating such an argument, I shall consider only the four clusters most widely discussed by philosophers: the modal, the cosmological, the teleological, and the moral. I want to frame arguments from each of these four as briefly as possible, so as to show as briefly as possible why each is valuable but none are objectively cogent by themselves.

Modal arguments are sometimes classed together under the heading 'the ontological argument'; but for reasons I cannot discuss here, I shall leave aside ontological arguments that are properly so-called. Modal arguments are called such because they employ the logical modalities of possibility and necessity, and may be generically framed in the following simple argument:

(1) If God possibly exists, then God necessarily exists
(2) God possibly exists
(3) Therefore, God necessarily exists.

The first premise may seem odd, but it has to be true. It means that if anything could be God, then there is a God that exists eternally, uncausably, and unpreventably. That is because God cannot be a contingent being: God is not the sort of thing that happens to exist but might not have, or that happens not to exist but might have. God cannot come to be or cease to be: God cannot be caused to exist, or just pop into existence, or die. However extraordinary, no such being could be God, but would be only one more item of our world. The same goes for any alleged God who undergoes change; such a being would not be God, but another constituent of our world, though perhaps a pretty special one. Whatever would count as God would have to be eternal, uncausable, and unpreventable, so that it always exists, never really changes, and could neither come into existence nor pass out of existence. Hence, if it could exist at all, it must exist.

The simple modal argument I have presented is obviously valid. But since Hume, many philosophers have been inclined to reject its conclusion, and hence its second premise, as false on logical grounds alone. Their argument goes like this: no existential statement (i.e., no statement to the effect that such-and-such exists) could be a necessary truth; for a necessary truth is one whose denial entails a contradiction, and no statement of the form 'x exists' entails a contradiction; hence no statement of the form 'x necessarily exists' is true; hence it cannot be true that God necessarily exists. And given (1) as well, we may also conclude that God does not possibly exist—i.e., that (2) is false.
But that argument is itself invalid. For from the fact—if it is a fact—that no existential statement is such that its denial entails a contradiction, it does not follow that any statement of the form 'x necessarily exists' is false. In particular, to say that God necessarily exists is not to imply that 'God does not exist' entails a contradiction; it is rather to imply that, given the sort of thing God would be, God exists eternally, uncausably, and unpreventably if at all. The word 'necessarily' in this context is not about the modal status of the statement 'God exists', but is rather a shorthand description of how God exists if God exists at all. So if (3) is false, that is not because it is saying something that logic alone can teach us is false.

The real problem with the modal argument is that there seems to be no reliable, publicly available method for verifying premise (2) that does not require prior knowledge that God exists. In general, we find out that a thing possibly exists (i.e., could exist) in one of two ways. The first is to extrapolate from what one knows to exist. That, e.g., is what astronomers do when they hold that there could be life on some planet orbiting some other star; given what we know about the physical universe, that is a reasonable view, even if it turns out to be false. The second way to find out whether something could exist is simply to find out that the thing does exist, from which it follows trivially that the thing could exist. Obviously, to verify (2) by finding out that God exists would render the modal argument useless as an argument. But from what may we extrapolate in order to make it reasonable to think that God could exist? God is not the sort of thing, like life on a planet, that might not have existed if indeed it does exist, or that exists by first coming to be. In view of this, some theists would argue that God, though not physically possible, is logically possible. Thus, given a complete description of the concept of God, we would find that God is, in a certain sense, like planets but unlike things that are both black-all-over and not-black-all-over. Like the former and unlike the latter, the claim that there is a God would not entail a contradiction; hence God could exist (is possible). But of course, there is no agreement on just what such a description would entail or on whether we could fully understand it even if we could agree on all that it would entail. Short of knowing that there is a God, there seems to be no reliable way to find out just what God is; if so, then one cannot know that God's existence is possible without knowing that God exists. It seems we are at a loss here. By itself, the modal argument does not get us very far—though it does highlight something about what God is that must always be kept in mind.

The next two clusters of arguments—the cosmological and the teleological—are the best-known, and involve essentially extrapolatory moves. Their aim is not to establish directly that God could exist but that God does exist. The cosmological argument may be generically framed as follows:

(1) Whatever exists, and is contingent, has been caused to exist by something other than and not part of itself
(2) The world exists and is contingent
(3) Therefore, something not of the world has caused the world to exist.

Even if sound, of course, such an argument does not establish that the cause of the world is God. But for the moment I want to leave that issue aside and assess the argument for what it purports to do, not criticize it for failing to do what it does not purport to do.

The cosmological argument as I have presented it is clearly valid. But are both its premises true? Consider (2) first. To say that the world is contingent entails that the world is not the sort of thing that exists eternally, uncausably, and unpreventably. Obviously, whether that's true or not depends on what the world is. Suppose, for example, that the world consists only of matter-energy and the changes matter-energy undergoes, and that matter-energy is neither created nor destroyed. Then one might be inclined to affirm that the world is necessary not contingent. But even if the world consists only of matter-energy and the changes it undergoes, and even if the first law of thermodynamics is true, that law does not entail that the world can neither be created nor destroyed; it only entails that such changes as occur in the world do not involve the production or destruction, but merely transformations, of matter-energy. And so the sort of crude physicalism I have hypothesized for argument's sake is no basis for saying that the world is necessary.

There are, of course, other more philosophical arguments that the world's existence is necessary—the best known being Spinoza's. Each of those arguments, if you look at them in their respective contexts, stands or falls with the philosophies to which they severally belong. Of those philosophies, all I can say here is their chief attraction has been their explanatory power: if any are true, then everything is explicable in such a way that nothing could have been otherwise. Such philosophies often attract people, like philosophers and scientists, who crave maximum intelligibility of the sort we find in, say, geometry or elementary symbolic logic, where every theorem is strictly deducible from a few axioms using a few obvious rules of inference. Such philosophies entail determinism: the thesis that whatever is the case is either necessarily so in itself or is rendered inevitable by the past and the laws of nature. But determinism does not seem true to most of us—who find it incompatible with, e.g., our sense of moral responsibility—and in any case determinism finds no more support in contemporary physics than in ordinary experience. Many people, then, are right to think it a marvelous and puzzling fact that the world exists, and continues to exist, at all. At any rate, there is no good reason to hold that the world is the sort of thing that exists uncausably, unpreventably, and eternally. And so we may safely say that the existence of the world is contingent.

Premise (1) is harder to justify. True, ordinary experience only presents objects that are caused to exist, and so one might think that we can infer (1) by induction. But the contingent objects that experience presents are the sorts that come to be through causally efficacious processes which follow laws: the laws of nature. Whatever the world as a whole is, it cannot be an object of that sort; it is rather that totality which includes all such objects, whatever else it may include. So if the world has indeed come to be, it cannot have done so through a causal process that obeys the laws of nature. And some have found in this grounds for denying that the world's existence is causally explicable at all. Such explanations as we have of the causes of particular things involve appeal to generalizations that get refined and expressed as scientific laws; a causal explanation involving no such appeal seems by contrast to hypothesize a unique, essentially arbitrary event that we can neither observe nor replicate. In face of this, many philosophers think that the contingency of the world might consist just in its happening to exist, even though it might not have existed; as Bertrand Russell put it, "the world is just there, and that's all." The world might be eternal and uncausable, for all we know, but still be preventable in some way; for we may admit that the world is the sort of thing that might not have existed; but in fact it exists—end of story.

Some proponents of the cosmological argument have tried to counter that attitude by premising the so-called principle of sufficient reason as an a priori truth. In other words, they claim that whatever is the case, there is sufficient reason why it is the case, and that such a generalization is either self-evident or otherwise immune to counterexamples. If they are right, then there is reason enough why the world exists, and thus a complete explanation of the world's existing.
But such a tack is unpromising. First, the principle of sufficient reason as generally formulated does nothing to rule out that sufficient reason for the world's existence may be found in that of its constituents. Thus, to the question "Why does the world exist" one might simply answer "Because each of its constituents do," without any clear violation of the principle. One might, of course, so formulate the principle as to rule out such a trivial answer. But that would bring us square up against the historical fact that the usefulness of the principle of sufficient reason varies inversely with the precision with which it is formulated.

Vague formulations which command broad assent do not suffice for the purposes their proponents usually have in mind; more precise formulations that would suffice if true either are untrue or appear to be true only if some version of the cosmological argument is independently cogent. Following Aristotle, for example, Thomas Aquinas argued that there cannot be an "infinite regress" of causes of being, even though there could in principle be an infinite regress of causes of becoming. Thus for all philosophy can tell us, the world might always have existed, and so there might have been infinitely long causal chains by which one contingent individual after another comes to be; but an account of why any such individual can continue in being at all must terminate somewhere-or-other. That, I think, is undeniably true. But Aquinas also thought that what is identified at the terminus of any such explanation must be a necessary existent that causes whatever is caused: namely, God. I believe that; and if Aquinas is right, then some version of the principle of sufficient reason is also true. But Aquinas' conclusion does not follow just from the fact that there cannot be an infinite regress of causes of being. If you examine Aquinas' arguments carefully, you find that some of them get the conclusion to follow by relying on an Aristotelian cosmology that is now outdated; others get it to follow only by bringing in metaphysical theses that few philosophers, including theistic ones, have even found intelligible, much less acceptable. More work needs to be done, especially by way of clarifying the logic of causal explanation.

The cosmological argument does call attention to the contingency of the world, as well as to the fact that the existence of the world so conceived cannot be explained in the sort of way in which the known constituents of the world can be. But by itself, the cosmological argument as typically framed does not show that the existence of the world is explicable by the activity of what would count as God. Partly for that reason, some people prefer the teleological argument.

Generically framed, the teleological argument (or: the argument from design) goes like this:

(1) The world as a whole is well-ordered
(2) Good order obtains only by the design of something intelligent
(3) Therefore, there is an intelligent designer of the world as a whole.

Like the modal and cosmological arguments I have framed, that one is logically valid. And if both its premises are true, we have both a reason why the world as a whole is explicable and a fairly specific sense in which the world is explicable. But there are two difficulties with the argument.
First, it does not establish that the designer of the world must also be the creator of the world, and hence is not by itself an argument for the existence of God. However, in a few minutes I shall overcome that difficulty by combining the teleological with the cosmological argument in one mega-argument. The second difficulty is more serious: there seems to be no apposite method for verifying either of its premises.

Consider premise (2). One might argue by analogy that the pervasive and intricate order we find in the world needs to be accounted for. After all, if our lives are to be well-ordered, we have to think and plan and take appropriate action; things don't just happen to fall into place for us (unless we are toddlers, in which case others will usually order our lives for us, in face of the chaos we spread). And so one might think that, in general, chaos is the natural state of things-unless some intelligent super-agent orders them for a certain end or purpose. But that doesn't follow. In fact, we have no experience of a chaotic universe being reduced to order; the natural state of things, by all appearances, is fundamentally orderly. Although instances of local order may indeed require explanation, it is not at all obvious that the general order of things itself requires any explanation other than the lawlike interrelations among things that would be exhibited by a good inventory of the sorts of things there are. There seems to be no reliable, publicly accessible method for showing that the general order of things calls for explanation in the first place.

The problem with premise (1) is similar. By what reliable, publicly available method could we determine that the world as a whole is well-ordered in the first place? One might say the answer is obvious: the method of modern physics; and the world, by all appearances, is well-ordered. Well, the universe studied by astrophysicists may well be; but if it is, how do we know that that is all there is to the world? For all science can tell us, there might be other universes, or countless non-physical beings; the questions whether or how such things might be related to the physical universe we know are dark ones indeed, which it is not obvious that science or any other human method can answer. Until we know the answers, it seems, we are in no position to verify premise (1).

Despite such difficulties, the teleological argument does focus attention on a very important fact: if there is a genuine explanation why the world works as it does—and by 'genuine' I mean one that consists in more than just an account of the world's content and structure—that explanation cannot be in terms of scientific laws but only in terms of the intentions of some intelligent personal agent. At any rate, I have not encountered any alternative sort of explanation that would do the job.

In that respect, the appeal of the teleological argument dovetails with that of the moral argument. The best-known moral argument for the existence of God is Kant's. I have no time to describe and evaluate it; but fortunately there are others too, and I think the gist of them all can be captured in the following, generic version:

(1) Moral laws tell us how, in general, we really should behave, not merely how we or others want to regulate our behavior
(2) Such laws can only be thought of as promulgated by a supreme and benevolent legislator
(3) Therefore, there is a supreme and benevolent legislator.

Like the others I have framed, that argument too is logically valid.

Now the mere presence of premise (2) might seem to undermine the argument from the outset; for you may well ask why anybody should believe (2) unless they already believe in God. But premise (1), I submit, is true just in case premise (2) is. If morality is not just descriptive, but is also normative—so that it tells us not just how people do choose to regulate their lives, but how they should—then its ultimate source must be beyond us in a supremely powerful and benevolent will to which our obedience is due. The converse also holds. Why?

Well, if your personal morality is just a list of rules that you set for yourself, then it does not necessarily tell you how you should behave, but merely how you want to regulate your life. Of course one might well reply that one's personal morality should conform to some larger, collective morality. But even so, if the latter is ultimately just a way of codifying social prejudices, preferences, and goals, then it only serves to clarify how most people of a given human community want its members to behave—not necessarily how they ought to behave. In other words, the question whether your personal rules are the ones you should be following, and/or the question whether some socially codified set of rules are the ones people ought to impose on each other, still remain open. In face of those questions, there are only two possible responses.

You can insist that, in the final analysis, moral norms simply tell us how certain individuals or social groups want to regulate their lives. In that case, saying that one "ought" to behave in such-and-such ways reflects nothing more than some person's or group's preference. The force of ought here is merely psychological: ultimately, the voice of conscience is just the internalized voice of society. This result would be what is called "moral relativism." Or you can admit that there is some fundamental moral law—perhaps the so-called "natural law"—in light of which we can and must judge whether some norms by which people want to regulate their lives are ones by which they should regulate their lives. The first response entails rejecting premise (1), and with it premise (2). The second raises another question: what gives this law its binding force?

Some philosophers argue that this law binds simply because it describes what general forms of behavior are really good for us, as opposed to what may only seem good to some people at some times. From that point of view, it is supposed to be obvious that we ought to do what's really good for us—just because that's what's really good for us. But one difficulty here is that it is not clear who this "us" is supposed to be: people taken individually? Collectively? The people now alive, or posterity too? But even supposing for argument's sake that what's really good for the one is ultimately good for the many, and that what's thus good for the living is always good for the yet-unborn, why should we always do what's good for us? How can we rule out the possibility that there are some goods incompatible with our individual and/or collective flourishing qua human beings—goods for the sake of which such happiness might be worth forgoing? Perhaps humanity is just a doomed evolutionary experiment, whose chief value consists in teaching other races, by our example, how not to live. Or maybe what's really good for us is not good for the universe as a whole: the big stage of the universe might be more important than the actors who cross it, so that at some point we actors would do well to sacrifice our careers for the good of the theater. In that case, our highest duties would be murder and suicide. The Albigensian heresy of medieval France entailed just that.

In order to set aside such possibilities as morally irrelevant, one must show that there is no good we can intelligibly pursue that is incompatible with and preferable to what would promote our flourishing as human beings. From that it would follow that, by doing only what's really good for us (whatever that is), we do as well as we can; and doing as well as we can is, of course, all and only what we should do. But we thereby do as well as we can only if the general scheme of things embodies such values as we could (a) recognize as compatible with what's really good for us, and (b) pursue under that description. And why think that?

The only reason I can find to think so would be that the general scheme of things is intelligently designed by some benevolent agent who thereby intends, among other things, what's really good for us. That person would be God. By designing us and the general scheme of things in a certain way, God would be in effect a supreme and benevolent legislator. For if God has set things up so that doing what's really good for us is doing as well as we can, then doing what's really good for us is all and only what we should do. In that case, our obligations coincide with what would make us truly happy; and an agent responsible for such a state of affairs is surely benevolent.

It is of course psychologically possible to treat certain moral precepts as laws without being aware of any such supreme legislator. People manage it all the time—partly by noticing that cultivating the cardinal virtues is essential for a modicum of happiness in this life, and in particular by noticing that certain good reasons for others not to harm them are equally good reasons for them not to harm others. But the explanation why the relevant precepts are laws, and thus binding on all of us, must posit a supreme lawgiver.

The only way to avoid this result, it seems to me, is by embracing moral relativism. But then you will have abandoned the idea that we are ultimately bound by moral obligations that we do not simply choose to impose on ourselves. That in turn means abandoning the idea that what we do can be evaluated in any terms other than its fitness for attaining such ends as we happen to have. Ultimately, you will have rejected the common belief that moral laws tell us not merely how we or others want to regulate our lives, but how we should. And by thus rejecting premise (2), you will have rejected premise (1). They stand or fall together.

The moral argument I have presented would be no good if there were no reliable, publicly available method for establishing that there is an objectively binding moral law in the first place. I think there is such a method, but I cannot pursue that issue here. For my present purpose, the key thing about the moral argument is how it shows that a belief which is very important to most people—namely, that certain moral norms bind us absolutely, whether some of us care to acknowledge that or not—is best supported in theistic terms. In that respect, the moral argument resembles the teleological argument by exhibiting that the only way to explain certain things (assuming those things are so) is in terms of a supreme, intelligent, and benevolent agent, namely God.

Where, then, does all this leave us? Let me propound to you an argument that I think combines the virtues of the four generic ones I have reviewed, while avoiding at least some of their difficulties.

(1) The world is the totality comprising every entity that really changes, and is thus contingent
(2) The existence of the world so defined either is not explicable at all, or is explicable as the creation of a necessary, intelligent, and benevolent being, which would be God
(3) There is better reason to think that the world's existence is explicable than that it is not
(4) Therefore, there is better reason to think that the existence of the world is explicable as the creation of God than not.

In defense of that logically valid argument, I have time to make only a few brief remarks. First, I have defined the world in the way I have because it is about just such a world that the question "Why does the world exist?" can intelligibly arise. Since it seems clear that that question does intelligibly arise—whether or not we have the resources for answering it—I must here rest my case for (1). Premise (2) can be verified in light of my considerations on the four major argument-clusters I have reviewed. It is premise (3), I think, that many of you will find most objectionable. In its defense, I will say only this: the only justification for rejecting it would be to show that the world, as I have defined it, is not the sort of thing whose existence can be explained by anyone at all, no matter how intelligent or well-informed they may be. But that, I am certain, would be an instance of the narrow-mindedness that I have said is impervious to argument. The most important question about my argument is: is there a reliable, publicly available method for verifying each of the premises? I think there is, but I must now leave that to you to verify for yourselves.

© Michael Liccione 1999. Permission is granted only for private, non-commercial use of this article. Reproduction or use in any other form without the express written consent of the author is prohibited.