God, arguments for the existence of

1. Cosmological arguments
2. Ontological arguments
3. Ontological arguments (cont.)
4. Teleological arguments
5. Teleological arguments (cont.)
6. Other theistic arguments

Cosmological arguments start from some obvious and general but a posteriori fact about the universe: that there are contingent beings, for example, or that things move or change. We find first steps towards such an argument in Plato (Laws 10); Aristotle (§16) (Metaphysics 12; Physics 7, 8) gives it a fuller statement; the medieval Arabic (especially al-Ghazali) and Jewish philosophers (especially Maimonides) gave elaborate statements of the argument; but its locus classicus (for Westerners, anyway) is the first three of the famous 'five ways' of Aquinas' Summa theologiae. Following Aquinas, Duns Scotus (§§7-11) presented a subtle and powerful version of the argument, and in modern times the most influential versions of the argument are to be found in the works of Leibniz (§3) and Samuel Clarke (§1). (The most influential criticisms of the argument are given by Hume (§6) and Kant (§8).)

Following William Craig (1980), we may distinguish substantially three versions of the cosmological argument. First, the so-called kalam (Arabic, 'speculative theology') argument, developed by Arabic thinkers (for example, al-Kindi (§2) and al-Ghazali). Put most schematically, this argument goes as follows:

(1) Whatever begins to exist is caused to exist by something else.
(2) The universe began to exist.
(3) Therefore the universe was caused to exist; and the cause of its existence is God.

The second premise was supported by arguments for the conclusion that an 'actual infinite' is not possible: it is not possible, for example, that there have been infinitely many temporally non-overlapping beings each existing for at least a second; alternatively, it is not possible that an infinite number of seconds have elapsed. These arguments proceed by pointing out some of the paradoxes or peculiarities that an actual infinite involves (see Infinity). For example, suppose there were a hotel with infinitely many rooms ('Hilbert's Hotel'). The hotel is full; a new guest arrives; despite the fact that each room is already occupied, the proprietor accommodates the guest by putting them in room 1, moving the occupant of room 1 to room 2, of room 2 to room 3, and in general the occupant of n to . No problem! Indeed, when a large bus containing infinitely many new guests pulls up, they too can all be accommodated: for any odd-numbered room n, move its occupant into room 2n (moving the occupant of that room into , and so on), thus freeing up the infinitely many odd-numbered rooms. In fact, if it is a busy weekend and an infinite fleet of buses pulls up, each with infinitely many new guests, they too can all be easily accommodated. And the question is: is it really possible, in the broadly logical sense, that such a hotel could actually exist? The friend of the kalam argument thinks not, and adds that no other actual infinite is possible either. If so, then the universe has not existed for an infinite stretch of time, but had a beginning. Contemporary cosmological theory in physics has seemed to some to provide scientific, empirical support for the claim that the universe had a beginning; according to 'Big Bang' cosmology, the universe came into being something like 15 billion years ago, give or take a few billion (see Cosmology §3).

Given that the universe has a beginning, the next step is to argue (by way of the first premise) that it must therefore have had a cause; it could not have popped into existence uncaused. And the final step is to argue that the cause of the universe would have to have certain important properties - properties of God.

The second kind of cosmological argument is the kind to be found in the first three of Aquinas' five ways. His second way, for example, goes like this:

(1) Many things in nature are caused.
(2) Nothing is a cause of itself.
(3) An infinite regress of essentially ordered efficient causes is impossible.
(4) Therefore, there is a first uncaused cause - 'to which', says Aquinas, 'everyone gives the name of God.'

There are two points of particular interest about this argument. First, Aquinas disagrees with a premise of the kalam argument, according to which it is impossible that there be an actual infinite. He argues that it cannot be proved that the universe had a beginning; he thinks it possible (though false) that the universe has existed for an infinite stretch of time. How then are we to understand premise (3)? Aquinas is here speaking of a certain kind of series, an 'essentially ordered' series, a series of causes in which any cause of an effect must be operating throughout the whole duration of the effect's operation. It is only such series, he says, that cannot proceed to infinity. (Aquinas gives the example of a stick moving a stone, a hand moving the stick, and so on.) So the upshot of the argument, if it is successful, is that there exists at least one thing which causes other things to exist, but is not itself caused to exist by anything else.

But could there not be many such things? And would each of them be God? This brings us to the second point of interest. Aquinas argues that there must be a first unmoved mover, a first uncaused cause, a necessary being, and the like; but his theistic argument is not finished there. In the next eight questions he argues that anything that was a first efficient cause would have to be immaterial, unchanging, eternal, simple and the possessor of all the perfections to be found in those things dependent upon it - in a word, God. It is therefore incorrect to follow the usual custom of criticizing Aquinas for hastily concluding that a first cause or unmoved mover or necessary being would have to be God.

The third sort of cosmological argument is associated especially with Leibniz and Samuel Clarke; according to this version of the argument, there must be a sufficient reason for the actuality of any contingent state of affairs. Therefore there must be a sufficient reason for the existence of any contingent being - but also, says Leibniz, for the whole series of contingent beings. This sufficient reason must be the activity of God.

Anselm's ontological argument has excited enormous controversy (see Anselm of Canterbury §§3-4). Aquinas rejected it, Duns Scotus 'coloured' it a bit and then accepted it; Descartes (§6) and Malebranche endorsed it; like Duns Scotus, Leibniz thought it needed just a bit of work to be successful; Kant rejected it and delivered what many thought to be the final quietus (though others have found Kant's criticisms both intrinsically obscure and of doubtful relevance to the argument); and Schopenhauer thought it a 'charming joke'. Although in the twentieth century it was defended by (among others) Charles Hartshorne, Norman Malcolm and Alvin Plantinga, probably most contemporary philosophers reject the argument, thinking it a joke, but not particularly charming.

Anselm's version goes as follows:

Hence, even the fool is convinced that something exists in the understanding, at least, than which nothing greater can be conceived.... And assuredly that, than which nothing greater can be conceived, cannot exist in the understanding alone; for suppose it exists in the understanding alone; then it can be conceived to exist in reality; which is greater.

Therefore, if that, than which nothing greater can be conceived, exists in the understanding alone, the very being, than which nothing greater can be conceived is one, than which a greater can be conceived. But obviously this is impossible. Hence, there is no doubt that there exists a being, than which nothing greater can be conceived, and it exists both in the understanding and in reality.

(Proslogion, ch. 2)

This argument is a reductio ad absurdum: postulate the nonexistence of God, and show that this leads to an absurdity. Perhaps we can outline the argument as follows:

(1) A maximally great being (one than which nothing greater can be conceived) exists in the understanding (that is, is such that we can conceive of it).
(2) It is greater to exist in reality than to exist merely in the understanding.
(3) Therefore, if the maximally great being existed only in the understanding, it would be less than maximally great.

But it is impossible that the maximally great being be less than maximally great; hence this being exists in reality as well as in the understanding - that is, it exists. And clearly this maximally great being is God.

The earliest objection to this argument was proposed by Anselm's contemporary and fellow monk Gaunilo in his On Behalf of the Fool (Psalm 14: 'The fool has said in his heart "There is no God"'). According to Gaunilo, the argument must be defective, because we can use an argument of the very same form to demonstrate the existence of such absurdities as an island (or chocolate sundae, or hamster, for that matter) than which none greater can be conceived. (Says Gaunilo: 'I know not which I ought to regard as the greater fool: myself, supposing that I should allow this proof; or him, if he should suppose that he had established with any certainty the existence of this island.') But Anselm has a reply: the notion of a maximally great island, like that of a largest integer, does not make sense, cannot be exemplified. The reason is that the properties that make for greatness in an island - size, number of palm trees, quality of coconuts - do not have intrinsic maxima; for any island, no matter how large and no matter how many palm trees, it is possible that there be one even larger and with more palm trees. But the properties that make for greatness in a being - knowledge, power and goodness, for example - do have intrinsic maxima: omniscience, omnipotence and being perfectly good.

The most celebrated criticism of the ontological argument comes from Immanuel Kant, who apparently argues in his Critique of Pure Reason (Transcendental Dialectic, bk II, ch. III, section 4) that if this argument were sound, the proposition 'there is a being than which none greater can be conceived' would have to be logically necessary; but there cannot be an existential proposition that is logically necessary. Sadly, his reason for making this declaration is itself maximally obscure. He adds that 'existence is not a real predicate', which is widely quoted as the principal objection to the argument. Unfortunately this dictum is of dubious relevance to Anselm's argument and a dark saying in its own right. What might it mean to say that existence is not a real property or predicate? And if it is not, how is that relevant to the argument? Why should Anselm care whether it is or not?

Perhaps we can understand Kant as follows. The argument as stated begins with the assertion that a maximally great being exists in the understanding; the idea is that this much is obvious, whether or not this being also exists in reality (that is, actually exists). Anselm then goes on to reason about this being, arguing that a being with the properties this one has - of being maximally great - cannot exist only in the understanding, but must exist in reality as well. So the argument depends upon the assumption that there is a maximally great being, and now the question is: does this being actually exist? Use the term 'actualism' for the view that there are not (and could not be) things that do not exist; the things that exist are all the things there are. Note that if this is true, then existence is a very special property: it is redundant, in that it is implied by every other property; anything that has any property (including the property of being maximally great) also has existence. But if actualism is true, the ontological argument as formulated above cannot work. For if it is not possible that there be things that do not exist, then in saying initially that there is a maximally great being, one that at any rate exists in the understanding, we are already saying that there exists a maximally great being, thus begging the question. If no maximally great being exists, then there simply is no such thing as a maximally great being, in which case we cannot (following Anselm) suppose initially that the maximally great being does not exist in reality and then argue that this being would be greater if it did exist in reality. If actualism is true, existence is a redundant property; but then to say that there is a maximally great being that exists in the understanding is already to say that there really exists a maximally great being. So perhaps Kant's puzzling dictum should be seen as an early endorsement of actualism.

Of course Anselm might reply that the fault lies not with his argument, but with actualism; in any event, there are other versions that do not conflict with actualism. Charles Hartshorne (1941) claimed to detect two quite different versions of the argument in Anselm's work; the second version is consistent with actualism and thus sidesteps Kant's criticism. This version proceeds from the thought that a really great being would be one that would have been great even if things had been different; its greatness is stable across possible worlds, to put it in a misleading if picturesque way. So say that a being has maximal excellence in a given possible world W if and only if it is omnipotent, omniscient and wholly good in W; and say that a being has maximal greatness if it has maximal excellence in every possible world. Then the premise of the argument (thus restated) is simply:

Maximal greatness is possibly exemplified.

That is, it is possible that there be a being that has maximal greatness. But (given the widely accepted view that if a proposition is possibly necessary in the broadly logical sense, then it is necessary), it follows by ordinary modal logic that maximal greatness is not just possibly exemplified, but exemplified in fact. For maximal greatness is exemplified if and only if there is a being B such that the proposition

B is omnipotent and omniscient and wholly good (has maximal excellence)

is necessary. If maximal greatness is possibly exemplified, therefore, then some proposition of that sort is possibly necessary. By the above principle, whatever is possibly necessary is necessary; accordingly, that proposition is necessarily true and hence true.

So stated, the ontological argument breaches no laws of logic, commits no confusions and is entirely immune to Kant's criticism. The only remaining question of interest is whether its premise, that maximal greatness is possibly exemplified, is indeed true. That certainly seems to be a rational claim; but it is not one that cannot rationally be denied. A remaining problem with the argument, perhaps, is that it might be thought that the epistemic distance between premise and conclusion is insufficiently great. Once you see how the argument works, you may think that asserting or believing the premise is tantamount to asserting or believing the conclusion; the canny atheist will say that he does not believe it is possible that there be a maximally great being. But would not a similar criticism hold of any valid argument? Take any valid argument: once you see how it works, you may think that asserting or believing the premise is tantamount to asserting or believing the conclusion. The ontological argument remains as intriguing as ever.

Teleological arguments start from contingent premises that involve more specific features of the universe, features which in one way or another suggest that the universe has been designed by a conscious and intelligent being. These arguments have often been developed in close connection with modern science; they have been endorsed by many of the giants of modern science, including Isaac Newton. Here is a classic statement of the argument by William Paley (§2):

In crossing a heath, suppose I pitched my foot against a stone, and were asked how the stone came to be there, I might possibly answer that for any thing I know to the contrary, it had lain there for ever: nor would it perhaps be very easy to show the absurdity of this answer. But suppose I had found a watch upon the ground and it should be inquired how the watch happened to be in that place, I should hardly think of the answer which I had before given, that, for any thing I know the watch might have always been there. Yet why should not this answer serve for the watch, as well as for the stone? For this reason and for no other: viz., that, when we come to inspect the watch, we perceive (what we could not discover in the stone) that its several parts are framed and put together for a purpose.

([1802] 1804: 1-2)

Paley then points out that the universe and some of its parts - for example, living things and their organs - resemble a watch, in that they give the appearance of having been designed in order to accomplish certain purposes. An eye, for example, looks like an extremely subtle and sophisticated mechanism designed to enable its owner to see. But the only serious candidate for the post of designer of the universe is God.

Kant, who had little but contempt for the cosmological and ontological arguments, was much less dismissive of this one. He still rejected the argument, however, pointing out that at most it shows that it is likely that there is a designer or architect of the universe; and it is a long way from a designer to the God of the theistic religions, an almighty, omniscient, wholly good creator of the world, by whose power the universe sprang into being. Of course, a cosmic architect - a being who has designed our entire universe, with its elements ranging across many orders of magnitude from gigantic galaxies to the minutest things we know - is no mean conclusion, and it seems churlish to dismiss it with an airy wave in order to point out that there is something even stronger that the teleological argument does not show.

Many people, however, have rejected the teleological argument even taken as an argument for a designer. The eighteenth-century proponents of the argument invariably mentioned the apparent teleology in the biological world; but (so say the critics) Darwin changed all that. We now know that the apparent design in the world of living things is merely apparent. The enormous variety of flora and fauna, those enormously elaborate and articulate mechanisms, and finely detailed systems and organs such as the mammalian eye and the human brain give a powerful impression of design; but in fact they are the product of such blind mechanisms as random genetic mutation and natural selection. The idea is that there is a source of genetic variation which produces mutation in the structure and function of existing organisms. Most of these mutations are deleterious; a few are adaptive and their lucky owners will have an adaptive edge, eventually coming to predominate in a population. Given enough time, so the story goes, this process can produce all the splendid complexity and detail that characterize the contemporary living world (see Evolution, theory of).

Of course there is little real evidence that these processes can in fact achieve this much: naturally enough, we have not been able to follow their operation in such a way as to observe them produce, say, birds or mammals from reptiles, or even human beings from simian precursors. And even if we did observe the course of animate history (even if we had a detailed record on film), this would by no means show that blind mechanisms are in fact sufficient for this effect; for of course there would be nothing in the film record to show that those random genetic mutations were not in fact guided and orchestrated by God.

Still, the critic of the teleological argument claims not that in fact evolution has been accomplished just by these blind mechanisms, but that it could have been; if so, there is a real alternative to design. That these mechanisms really could have produced effects of this magnitude is far from clear; we have little real reason to suppose that there is a path through the space of possible animal design plans, a path leading from bacteria to human beings, and such that each new step is both adaptive and reachable from the previous step by mechanisms we understand. Still, the suggestion does perhaps damage the teleological argument by suggesting a naturalistic candidate for the post of producer of apparent design.

But organic evolution addresses only one of the areas of apparent design. There is also the origin of life; even the simplest unicellular creatures (prokaryotes such as bacteria and certain algae, for example) are enormously complex and upon close inspection look for all the world as if they have been designed; it is fair to say that no one, so far, has a decent idea as to how these creatures might have come into being just by way of the operation of the regularities of physics and chemistry. There are also the various considerations connected with the so-called 'fine tuning' of the universe. First, there is the 'flatness' problem. The mass density of the universe is at present very close to the density corresponding to the borderline between an open universe (one that goes on expanding for ever) and a closed universe (one that expands to a certain size and then collapses). The ratio between the forces making for expansion and those making for contraction is close to one. But then shortly after the Big Bang this value would have to have been inside a very narrow band indeed. Thus Stephen Hawking (1974): 'reduction of the rate of expansion by one part in 10¹² at the time when the temperature of the Universe was 10¹⁰ K would have resulted in the Universe's starting to recollapse when its radius was only 1/3000 of the present value and the temperature was still 10,000 K' - much too warm for the development of life. On the other hand, if the rate of expansion had been even minutely greater, the universe would have expanded much too fast for the formation of stars and galaxies, required for the formation of the heavy elements necessary for the development of life.

Another kind of fine tuning was also necessary: of the fundamental physical constants. If any of the four fundamental forces (weak and strong nuclear forces, electromagnetic force, electron charge) had been even minutely different, the universe would not have supported life; they too must have been fine-tuned to an almost unbelievable accuracy. And the suggestion, again, is that given the infinite range of possible values for the fundamental constants, design is suggested by the fact that the actual values fall in that extremely narrow range of values that permits the development of intelligent life.

But there is a naturalistic riposte. Since the 1970s, several different sorts of 'inflationary' scenario have shown up. These postulate the formation (at very early times) of many different universes or subuniverses, with different rates of expansion, and different values for the fundamental constants. These inflationary models are motivated, in part, by a desire to avoid singularities and the accompanying appearance of design. If all possible values for the fundamental constants and the rate of expansion are actually exemplified in different subuniverses, then the fact that there is a subuniverse with the values ours displays no longer requires explanation or suggests a Designer. Many of these scenarios are wildly speculative and unencumbered by empirical evidence, but (if physically acceptable) they do tend to blunt the force of a design argument from fine tuning. (Of course, someone who already believed in God and saw no need to eliminate suggestions of design might be inclined to reject these suggestions as metaphysically extravagant.) But there are also counterarguments here; the discussion goes on. It is hard to see a verdict, at present, on the prospects of this form of the argument. The teleological argument seems to have enormous vitality; its epitaph is often read, but the argument regularly reappears in new forms. As for a final evaluation, the best perhaps comes from Kant, who said that this argument 'always deserves to be mentioned with respect. It is the oldest, the clearest and the most accordant with the common reason of mankind' (1787: 520).

We have examined the Big Three among theistic arguments, but there are many more. First, there are moral arguments of at least two sorts. These are arguments that the very nature of morality - the unconditioned character of the moral law - requires a divine lawgiver. You might find yourself utterly convinced that:

Morality is objective, not dependent upon what human beings know or think or do.

You may also be convinced that:

The objective character of morality cannot be explained in terms of any 'natural' facts about human beings (or other things), so there could not be such a thing as objective moral law unless there were a being like God who legislates it.

Then you will have a theistic argument from the nature of morality. This argument can go in either of two directions: some people think we can simply see that moral obligation is impossible apart from a divine will and lawgiver, while others think that the dependence of moral obligation upon the will of God is the best explanation for its objectivity and special deontological force.

A second main type of moral argument is due to Kant (1788), who argues first that virtue deserves to be proportionally rewarded with happiness: the more virtuous you are, the more happiness you deserve. But nature by itself does not seem able to guarantee anything like this sort of coincidence. If morality is to make sense, however, it must be supposed that there is such a coincidence; practical reason, therefore, is entitled to postulate a supernatural being with enough knowledge, power and goodness to ensure that we receive the happiness we deserve as a reward for our virtue. So taken, the argument is for the rationality of making the assumption that there is a being of this sort; it is not really an argument for the actual existence of such a being. This argument receives criticism from several sides: some hold that we do not have to assume that there is proportionality between virtue and happiness in order to carry out the moral life; others (for example, many Christians) argue that both happiness and the ability to live a moral life are gifts of grace and that if we really got what we deserve, we should all be thoroughly miserable.

There are many other theistic arguments - arguments from the nature of proper function, from the nature of propositions, numbers and sets, from intentionality, from counterfactuals, from the confluence of epistemic reliability with epistemic justification, from reference, simplicity, intuition, love, colours and flavours, miracles, play and enjoyment, morality, beauty, the meaning of life, and even from the existence of evil. There is no space even to outline all these arguments, so we will look at just three.

First, the argument from intentionality (or aboutness). Consider propositions - the things that are true or false, that are capable of being believed, and that stand in logical relations to one another. Propositions have another property: aboutness or intentionality. They represent reality or some part of it as being thus and so, and it is by virtue of this property that propositions (as opposed, for example, to sets) are true or false. Most who have thought about the matter have found it incredible that propositions should exist apart from the activity of minds. How could they just be there, if never thought of? Further, representing things as being thus and so - being about something or other - seems to be a property or activity of minds or perhaps thoughts. It is therefore extremely plausible to think of propositions as ontologically dependent upon mental or intellectual activity in such a way that either they just are thoughts, or else at any rate could not exist if not thought of. But propositions cannot be human thoughts; there are far too many of them for that. (For each real number r, for example, there is the proposition that r is distinct from the Taj Mahal.) Hence the only viable possibility is that they are divine thoughts, God's thoughts (so that when we think, we literally think God's thoughts after him).

Second, there is the argument from sets or collections. Many think of sets as displaying the following characteristics: (1) no set is a member of itself; (2) sets (unlike properties) have their extensions essentially - hence many sets are contingent beings and no set could have existed if one of its members had not; (3) sets form an iterated structure - at the first level, there are sets whose members are nonsets, at the second, sets whose members are nonsets or first level sets, ..., at the nth level, sets whose members are nonsets or sets of index less than n, ..., and so on.

It is also natural to follow Georg Cantor, the father of modern set theory, in thinking of sets as collections - that is, as things whose existence depends upon a certain sort of intellectual activity, a collecting or 'thinking together' as Cantor put it. If sets were collections, that would explain their having the first three features. But of course there are far too many sets for them to be a product of human thinking together; there are many sets such that no human being has ever thought their members together and many such that no human being could think them together. That requires an infinite mind - one such as God's.

For a third example, consider the argument from appalling evil. Many philosophers offer antitheistic arguments from evil, and perhaps they have some force. But there is also a theistic argument from evil. The premise is that there is real and objectively horrifying evil in the world. Examples would be certain sorts of appalling evil characteristic of Nazi concentration camps: guards found pleasure in devising tortures, making mothers decide which of their children would go to the gas chamber and which be spared; small children were hanged, dying (because of their light weight) a slow and agonizing death; victims were taunted with the claim that no one would ever know of their fate and how they were treated. Of course, Nazi concentration camps have no monopoly on this sort of evil: there are also Stalin, Pol Pot and a thousand lesser villains. These states of affairs, one thinks, are objectively horrifying, in the sense that they would constitute enormous evil even if we and everyone else came perversely to approve of them.

Naturalism does not have the resources to accommodate or explain this fact about these states of affairs. From a naturalistic point of view, about all one can say is that we do indeed hate them; but this is far short of seeing them as intrinsically horrifying. How can we understand this intrinsically horrifying character? After all, as much misery and suffering can occur in a death from cancer as in a death caused by someone else's wickedness. What is the difference? The difference lies in the perpetrators and their intentions. Those who engage in this sort of evil are purposely and intentionally setting themselves to do these wicked things. But why is that objectively horrifying? A good answer (and one for which it is hard to think of an alternative) is that this evil consists in defying God, the source of all that is good and just, and the first being of the universe. What is horrifying here is not merely going contrary to God's will, but consciously choosing to invert the true scale of values, explicitly aiming at what is abhorrent to God. This is an offence and affront to God; it is defiance of God himself, and so is objectively horrifying. Appalling evil thus has a sort of cosmic significance. But of course there could be no evil of this sort if there were no such being as God.

See also: Agnosticism; Atheism; Deism; God, concepts of; Natural theology; Religion and epistemology; Religious experience