John H. Hick, Philosophy of Religion (3d edition), 1983.


Grounds for Belief in God

The conception of God outlined in the preceding chapter is the outcome of many centuries of development during which some of the most acute minds of the. western world have sought, in a cooperative venture, to discover the fuller meaning and deeper implications of the stream of religious experience recorded in the Bible. Throughout the remainder of this book, as we raise philosophical questions about the Judaic-Christian concept of God, our central and controlling question must be concerned with the reasons behind the belief that there is any such Being. We must examine the grounds on which religious persons have claimed to know that God exists.

In this chapter we shall examine the most important of the philosophical arguments offered to justify belief in the reality of God. These traditional "theistic proofs" are of great philosophical interest and have been receiving more rather than less attention from both secular and religious writers in recent years.


The ontological argument for the existence of God was first developed by Anselm, one of the Christian Church's most original thinkers and the greatest theologian ever to have been archbishop of Canterbury.1

Anselm begins by concentrating the Christian concept of God into the formula: "a being than which nothing greater can be conceived." It is clear that by "greater" Anselm means more perfect, rather than spatially bigger.2 It is important to notice that the idea of the most perfect conceivable being is different from the idea of the_most perfect being that there is. The ontological argument could not be founded upon this latter notion, for although it is true by definition that the most perfect being that there is exists, there is no guarantee that this being is what Anselm means by God. Consequently, instead of describing God as the most perfect being that there is, Anselm describes God as the being who is so perfect that no more perfect can even be conceived.

First Form of the Argument

In the next and crucial stage of his argument Anselm distinguishes between something, x, existing in the mind only and its existing in reality as well. If the most perfect conceivable being existed only in the mind, we should then have the contradiction that it is possible to conceive of a yet more perfect being, namely, the same being existing in reality as well as in the mind. Therefore, the most perfect conceivable being must exist in reality as well as in the mind. Anselm's own formulation of this classic piece of philosophical reasoning is found in the second chapter of the Proslogion.

If then that-than-which-a-greater-cannot-be-thought exists in the mind alone, this same that-than-which-a-greater-cannnot-be-thought is that-than-which-a-greater-can-be-thought. But this is obviously impossible. Therefore there is absolutely no doubt that something-than-which-a-greater-cannot-be-thought exists both in the mind and in reality.

Second Form of the Argument

In his third chapter Anselm states the argument again, directing it now not merely to God's existence but to His uniquely necessary existence. God is defined in such a way that it is impossible to conceive of His not existing. The core of this notion of necessary being is self-existence (aseity)3 Since God as infinitely perfect is not limited in or by time, the twin possibilities of God's having ever come to exist or ever ceasing to exist are alike excluded, and God's nonexistence is rendered impossible. The argument now runs as follows:

For something can be thought to exist that cannot be thought not to exist. Hence, if that-than-which-a-greater-cannot-be-thought can be thought not to exist, then that-than-which-a-greater-cannot-be-thought is not the same as that-than-which-a-greater-cannot-be-thought, which is absurd. Something-than-which-a-greater-cannot-be-thought exists so truly then, that it cannot be even thought not to exist.

Criticisms of the Argument

In introducing the ontological argument, Anselm refers to the psalmist's "fool" who says in his heart, "There is no God."4 Even such a person, he says, possesses the idea of God as the greatest conceivable being: and when we unpack the implications of this idea we see that such a being must actually exist. The first important critic of the argument, Gaunilon, a monk at Marmoutiers in France and a contemporary of Anselm's, accordingly entitled his reply In Behalf of the Fool. He claims that Anselm's reasoning would lead to absurd conclusions if applied in other fields, and he sets up a supposedly parallel ontological argument for the most perfect island. Gaunilon spoke of the most perfect of islands rather than (as he should have done) of the most perfect conceivable island; but his argument could be rephrased in terms of the latter idea. Given the idea of such an island, by using Anselm's principle we can argue that unless it exists in reality it cannot be the most perfect conceivable island!

Anselm's reply, emphasizing the uniqueness of the idea of God to show that the ontological reasoning applies only to it, is based upon his second form of the argument. The element in the idea of God which is lacking in the notion of the most perfect island is necessary existence. An island (or any other material object) is by definition a part of the contingent world. The most perfect island, so long as it is genuinely an island -- "a piece of land surrounded by water" and thus part of the physical globe -- is by definition a dependent reality, which can without contradiction be thought not to exist; and therefore Anselm's principle does not apply to it. It applies only to the most perfect conceivable being, which is defined as having eternal and independent (i.e., necessary) existence. Thus far, then, it would seem that his argument is able to withstand criticism.

Can Anselm's argument in its first form, however, be defended against Gaunilon's criticism? This depends upon whether the idea of the most perfect conceivable island is a coherent and consistent idea. Is it possible, even in theory, to specify the characteristics of the most perfect conceivable island? This is a question for the reader to consider.

A second phase of the debate was opened when Rene Descartes (1596-1650), often called the father of modern philosophy, reformulated the argument and thereby attracted widespread attention to it.5 Descartes brought to the fore the point upon which most of the modern discussions of the ontological argument have centered, namely, the assumption that existence is a property or predicate. He explicitly treats existence as a characteristic, the possession or lack of which by a given x is properly open to inquiry. The essence or defining; nature of each kind of thing includes certain predicates, and Descartes's ontological argument claims that existenc must be included among the defining predicates of God. Just as the fact that its internal angles are equal to two right angles is a necessary characteristic of a triangle, so existence is a necessary characteristic of a supremely perfect being. A triangle without its defining properties would not be a triangle, and God without existence would not be God. The all- important difference is that in the case of the triangle we cannot infer that any triangles exist, since existence is not of the essence of triangularity. However, in the case of a supremely perfect being we can infer existence, for existence is an essential attribute without which no being would be unlimitedly perfect.

This Cartesian version of the ontological argument was later challenged at two levels by the great German philosopher, Immanuel Kant (1724-1804).6

At one level he accepted Descartes's claim that the idea of existence belongs analytically to the concept of God, as the idea of having three angles belongs analytically to the concept of a three-sided plane figure. In each case the predicate is necessarily linked with the subject. But, Kant replied, it does not follow from this that the subject, with its predicates, actually exists. What is analytically true is that if there is a triangle, it must have three angles, and if there is an infinitely perfect being, that being must have existence. As Kant says, "To posit a triangle, and yet to reject its three angles, is self-contradictory; but there is no self-contradiction in rejecting the triangle together with its three angles. The same holds true of the concept of an absolutely necessary being."

At a deeper level, however, Kant rejected the basic assumption upon which Descartes's argument rested, the assumption that existence, like triangularity, is a predicate that something can either have or lack, and that may in some cases be analytically connected with a subject. He points out (as indeed David Hume had already pointed out in a different context)7 that the idea of existence does not add anything to the concept of a particular thing or kind of thing. An imaginary hundred dollars, for example, consists of the same number of dollars as a real hundred dollars. When we affirm that the dollars are real, or exist. we are applying the concept of the dollars to the world. Thus to say of x that it exists is not to say that in addition to its various other attributes it has the attribute of existing, but is to say that there is an x in the real world.

Essentially the same point has more recently been made by Bertrand Russell in his analysis of the word "exists."8 He has shown that although "exists" is grammatically a predicate, logically it performs a different function, which can be brought out by the following translation: "Cows exist" means "There are x's such that 'x is a cow' is true." This translation makes it clear that to say that cows exist is not to attribute a certain quality (namely existence) to cows, but is to assert that there are objects in the world to which the description summarized in the word "cow" applies. Similarly "Unicorns do not exist" is the equivalent of "There are no x's such that 'x is a unicorn' is true." This way of construing negative existential statements -- statements that deny that some particular kind of thing exists -- avoids the ancient puzzle about the status of the "something" of which we assert that it does not exist. Since we can talk about unicorns, for example, it is easy to think that unicorns must in some sense be or subsist or, perhaps, that they inhabit a paradoxical realm of noonbeing or potential being. Russell's analysis, however, makes it clear that "unicorns do not exist" is not a statement about unicorns but about the concept or description "unicorn" and is the assertion that this concept has no instances.

The bearing of this upon the ontological argument is as follows. If existence is, as Anselm and Descartes assumed, an attribute or predicate that can be included in a definition and that, as a desirable attribute, must be included in the definition of God, then the ontological argument is valid. For it would be self-contradictory to say that the most perfect conceivable ^ being lacks the attribute of existence. But, if existence, although it appears grammatically in the role of a predicate, has the quite different logical function of asserting that a description applies to something in reality, then the ontological argument, considered as a proof of God's existence, fails. For if existence is not a predicate, it cannot be a defining predicate of God, and the question whether anything in reality corresponds to the concept of the most perfect conceivable being remains open to inquiry. A definition of God describes one's concept of God but cannot prove the actual existence of any such being.

It should be added that some theologians, most notably Karl Barth, have seen Anselm's argument not as an attempted proof of God's existence, but as an unfolding of the significance of God's self-revelation as One whom the believer is prohibited from thinking as less than the highest conceivable reality. On this view, Anselm's argument does not seek to convert the atheist but rather to lead an already formed Christian faith into a deeper understanding of its object.9

The ontological argument has perennially fascinated the philosophical mind, and in recent years there have been a number of new discussions of it, some of the most important of which are listed in footnote 10.

1 The ontological argument is to be found in Chaps. 2-4 of Anselm's Proslogion. Among the best English translations are those by M. J. Charlesworth in St. Anselm's Proslogion (Oxford: Clarendon Press, 1965 and University of Notre Dame Press)—from which the quotations here are taken—and Arthur C McGill in The Many-Faced Argument, eds J. H. Hick and A. C. McGill (New York. The Macmillan Company, 1967, and London Macmillan Sc Company Ltd , 1968)

2 On occasions (for example, Proslogion, Chaps 14 and 18) Anselm uses "better" (mehus) in place of "greater "

3 See p. 8

4 Psalms 14:1 and 53:1.

5 Meditations, V. It is not entirely clear whether Descartes received the basic principle of his ontological argument from Anselm. When questioned by Mersenne about the relation of his own argument to Anselm's, he was content to reply, "I will look at St. Anselm at the first opportunity." (N. Kemp Smith, New Studies in the Philosophy of Descartes, London: Macmillan & Company Ltd., 1952, p. 304.) Descartes also makes another and different attempt to prove God's existence: Discourse on Method, IV and Meditations, III.

6 Immanuel Kant, Critique of Pure Reason, tr. N. Kemp Smith (London: Macmillan & Company Ltd., 1933 and New York: St. Martin's Press, 1969). "Transcendental Dialectic," Book II, Chap. 3, Sec. 4.

7 David Hume, A Treatise of Human Nature, Book I, Part III, Sec. vii.

8 This aspect of the theory of descriptions is summarized by Russell in his History of Western Philosophy (London: George Allen & Unwin Ltd., 1946 and New York: Simon & Schuster), pp. 859-60. For a more technical discussion, see his Introduction to Mathematical Philosophy (1919), Chap. 16.

9 See Karl Barth, Anselm: Fides Quaerens Intellectum, 1931 (London: Student Christian Movement Press Ltd. and Richmond, Va.: John Knox Press, 1960). Barth's interpretation is criticized by Etienne Gilson in "Sens et nature de 1'argument de saint Anselme," Archives d'histoire doctrinale et litteraire du moyen age, 1934, pp. 23f.

10 Charles Hartshorne, The Logic of Perfection (LaSalle, Ill.: Open Court Publishing Co., 1962), Chap. 2, and Anselm's Discovery (LaSalle, Ill.: Open Court Publishing Co., 1965). Jonathan Barnes, The Ontological Argument (London: Macmillan and New York: St. Martin's Press, 1972). Alvin Plantinga, The Nature of Necessity (Oxford: Clarendon Press, 1974 and New York: Oxford U. Press, 1979), Chap. 10, and God, Freedom, and Evil (London: George Allen & Unwin, Ltd., 1974 and Grand Rapids: Wm. B. Eerdmans Publishing Co., 1978), Pt.II


The next important attempt to demonstrate the reality of God was that of Thomas Aquinas (1224/5-1274), who offers five ways of proving divine existence.11 Unlike the ontological argument, which focuses attention upon the idea of God and proceeds to unfold its inner implications, Aquinas's proofs start from some general feature of the world around us and argue that there could not be a world with this particular characteristic unless there were also the ultimate reality which we call God. The first Way argues from the fact of motion to a Prime Mover; the second from causation to a First Cause; the third from contingent beings to a Necessary Being; the fourth from degrees of value to Absolute Value; and the fifth from evidences of purposiveness in nature to a Divine Designer.

We may concentrate upon Aquinas's second and third proofs. His second proof, known as the First-Cause argument, is presented as follows: everything that happens has a cause, and this cause in turn has a cause, and so on in a series that must either be infinite or have its starting point in a first cause. Aquinas excludes the possibility of an infinite regress of causes and so concludes there must be a First Cause, which we call God. (His first proof, which infers a First Mover from the fact of motion, is basically similar.)

The weakness of the argument as Aquinas states it lies in the difficulty (which he himself elsewhere acknowledges)12 of excluding as impossible an endless regress of events, requiring no beginning.

However, some contemporary Thomists (i.e., thinkers who in general follow Thomas Aquinas) have reinterpreted the argument in order to avoid this difficulty.13 They interpret the endless series that it excludes, not as a regress of events back in time, but as an endless and therefore eternally inconclusive regress ot explanations. If fact A is made intelligible by its relation to facts B, C, and D (which may be antecedent to or contemporary with A), and if each of these is in turn rendered intelligible by other facts, at the back of the complex there must be a reality which is self-explanatory, whose existence constitutes the ultimate explanation of the whole. If no such reality exists, the universe is a mere unintelligible brute fact.

However, this reinterpretation still leaves the argument open to two major difficulties. First, how do we know that the universe is not "a mere unintelligible brute fact"? Apart from the emotional Lcoloring suggested by the phrase, this is precisely what the skeptic believes it to be; and to exclude this possibility at the outset is merely to beg the question at issue. The argument in effect presents the dilemma: either there is a First Cause or the universe is ultimately unintelligible; but it does not compel us to accept one horn of the dilemma rather than the other.

Second (although there is only space to suggest this difficulty, leaving the reader to develop it), the argument still depends upon a view of causality that can be, and has been, questioned. The assumption of the reformulated argument is that to indicate the causal conditions of an event is thereby to render that event intelligible. Although this assumption is true on the basis of some theories of the nature of causality, it is not true on the basis of others. If, for example, as much contemporary science assumes, causal laws state statistical probabilities,14 or if (as Hume argued) causal connections represent mere observed sequences,15 or are (as Kant suggested) projections of the structure of the human mind,16 the Thomist argument fails.

Aquinas's third Way, known as the argument from the contingency of the world, and often monopolizing the name the cosmoloqical argument, runs as follows. Everything in the world about us is contingent—that is, it is true of each item that it might not have existed at all or might have existed differently. The proof of this is that there was a time when it did not exist. The existence of this printed page is contingent upon the prior activities of lumberjacks, transport workers, paper manufacturers, publishers, printers, author, and others, as well as upon the contemporary operation of a great number of chemical and physical laws; and each of these in turn depends upon other factors. Everything points beyond itself to other things. Saint Thomas argues that if everything were contingent, there would have been a time when nothing existed. In this case, nothing could ever have come to exist, for there would have been no causal agency. Since there are things in existence, there must be something that is not contingent, and this we call God.

Aquinas's reference to a hypothetical time when nothing existed seems to weaken rather than strengthen his argument, for there might be an infinite series of finite contingent events overlapping in the time sequence so that no moment occurs that is not occupied by any of them. However, modern Thomists generally omit this phase of the argument (as indeed Aquinas himself does in another book).17 If we remove the reference to time, we have an argument based upon the logical connection between a contingent world (even if this should consist of an infinite series of events) and its noncontingent ground. One writer points as an analogy to the workings of a watch. The movement of each separate wheel and cog is accounted for by the way in which it meshes with an adjacent wheel. Nevertheless, the operation of the whole system remains inexplicable until we refer to something else outside it, namely, the spring. In order for there to be a set of interlocking wheels in movement, there must be a spring; and in order for there to be a world of contingent realities, there must be a noncontinggnt ground of their existence. Only a self-existent reality, containing in itself the source of its own being, can constitute an ultimate ground of the existence of anything else. Therefore, if there is an ultimate ground of anything, there must be a "necessary being," and this being we call God.

The most typical philosophical objection raised against this reasoning in recent years is that the idea of a "necessary being" is unintelligible. It is said that only propositions, not things, can be logically necessary.and that it is a misuse of language to speak of a logically necessary being.18 This particular objection to the cosmological argument is based upon a misapprehension, for the argument does not make use of the notion of a logically necessary being. The concept of a necessary being, used in the main theological tradition (exemplified by both Anselm and Aquinas)19, is not concerned with logical necessity but rather with a kind of factual necessity which, in the case of God, is virtually equivalent to aseity or self-existence. For this reason, the idea of God's necessary being should not be equated with the view that "God exists" is a logically necessary truth.

There remains, however, an important objection to the cosmological argument, parallel to one of those applying to the First-Cause argument. The force of the cosmological form of reasoning resides in the dilemma: either there is a necessary being or the universe is ultimately unintelligible. Clearly such an argument is cogent only if the second altemative has been ruled out. Far from being ruled out, however, this second alternative represents the skeptic's position. This inability to exclude the possibility of an unintelligible universe prevents the cosmological argument from operating for the skeptic as a proof of God's existence—and the skeptic is, after all, the only person who needs such a proof.

Today there is an important neo-Thomist group of thinkers who hold that there are valid forms of the cosmological argument; some of the most important writings from this point of view are listed in footnote 20.

11 Thomas Aquinas, Summa Theologica, Part I, Question 2, Art. 3. For an important recent philosophical study of Aquinas's arguments, see Anthony Kenny, Five Ways: St. Thomas Aquinas's Proofs of God's Existence (London: Routledge & Kegan Paul Ltd., 1969 and Notre Dame, Ind.: University of Notre Dame Press, 1980).

12 Aquinas, Summa Theologica, Part I, Question 46, Art. 2. See also Summa Contra Gentiles, Book II, Chap. 38.

13 For example, E. L. Mascall, He Who Is (London: Longmans, Green & Co., 1943), Chap. 5.

14 Cf. Hans Reichenbach, The Rise of Scientific Philosophy (Berkeley: University of California Press, 1951), Chap. 10.

15 David Hume, An Enquiry Concerning Human Understanding, Sec. 7.

16 Kant, "Transcendental Analytic," in Critique of Pure Reason.

l7 Aquinas, Summa Contra Gentiles, Book II, Chap. 15, Sec 6.

18 See, for example, J. J. C. Smart, "The Existence of God" and J. N. Findlay, "Can God's Existence Be Disproved?" in New Essays in Philosophical Theology, eds. Antony Flew and Alasdair Maclntyre (New York: The Macmillan Company and London: Student Christian Movement Press Ltd., 1955).

19 See p. 8.

20 Mascall, E. L., He Who Is. Austin Farrer, Finite and Infinite, 2nd ed. (London: Dacre Press, 1960). For an interesting recent presentation of the First Cause argument, appealing to current scientific cosmology, see William Lane Craig, The Kalam Cosmological Argument (London: Macmillan and Company Ltd., and New York: Barnes & Noble, 1979). For general treatments of cosmological arguments, see William Rowe, The Cosmological Argument (Princeton, N.J.: Princeton University Press, 1975) and William Lane Craig, The Cosmological Argument from Plato to Leibniz (London: Macmillan and New York: Barnes & Noble, "1980).


This has always been the most popular of the theistic arguments, tending to evoke spontaneous assent in simple and sophisticated alike. The argument occurs in philosophical literature from Plato's Timaeus onward. (It appears again as the last of Saint Thomas's five Ways.) In modern times one of the most famous expositions of the argument from, or to, design is that of William Paley (1743-1805) in his Natural Theology: or Evidences of the Existence and Attributes of the Deity Collected from the Appearances of Nature (1802).21 The argument is still in active commission, especially in more conservative theological circles.22

Paley's analogy of the watch conveys the essence of the argument. Suppose that while walking in a desert place I see a rock lying on the ground and ask myself how this object came to exist. I can properly attribute its presence to chance, meaning in this case the operation of such natural forces as wind, rain, heat, frost, and volcanic action. However, if I see a watch lying on the ground, I cannot reasonably account for it in a similar way. A watch consists of a complex arrangement of wheels, cogs, axles, springs, and balances, all operating accurately together to provide a regular measurement of the lapse of time. It would be utterly implausible to attribute the formation and assembling of these metal parts into a functioning machine to the chance operation of such factors as wind and rain. We are obliged to postulate an intelligent mind which is responsible for the phenomenon.

Paley adds certain comments that are important for his analogy between the watch and the world. First, it would not weaken our inference if we had never seen a watch before (as we have never seen a world other than this one) and therefore did not know from direct observation that watches are products of human intelligence. Second, it would not invalidate our inference from the watch to the watchmaker if we found that the mechanism did not always work perfectly (as may sometimes appear to be the case with the mechanism of the world). We would still be obliged to postulate a watchmaker. Third, our inference would not be undermined if there were parts of the machine (as there are of nature) whose function we are not able to discover.

Paley argues that the natural world is as complex a mechanism, and as I manifestly designed, as any watch. The rotation of the planets in the solar system and, on earth, the regular procession of the seasons and the cornplex structure and mutual adaptation of the parts of a living organism, all suggest design. In a human brain, for example, thousands of millions of cells function together in a coordinated system. The eye is a superb movie camera, with self-adjusting lenses, a high degree of accuracy, color sensitivity, and the capacity to operate continuously for many hours at a time. Can such complex and efficient mechanisms have corne about by chance, as a stone might be formed by the random operation of natural forces?

Paley (in this respect typical of a great deal of religious apologetics in the eighteenth century) develops a long cumulative argument drawing upon virtually all the sciences of his day. As examples of divine arrangement he points to the characteristics and instincts of animals, which enable them to survive (for example, the suitability of a bird's wings to the air and of a fish's fins to the water). He is impressed by the way the alternation of day and night conveniently enables animals to sleep after a period of activity. We may conclude with an example offered by a more recent writer, who refers to the ozone layer in the atmosphere, which filters out enough of the burning ultraviolet rays of the sun to make life as we know it possible on the earth's surface. He writes:

The Ozone gas layer is a mighty proof of the Creator's forethought. Could anyone possibly attribute this device to a chance evolutionary process? A wall which prevents death to every living thing, just the right thickness, and exactly the correct defense, gives every evidence of plan.23

The classic critique of the design argument occurs in David Hume's Dialogues Concerning Natural Religion. Hume's book was published in 1779, twenty-three years earlier than Paley's; but Paley took no apparent account of Hume's criticisms—by no means the only example of lack of communication between theologians and their philosophical critics! Three of Hume's main criticisms are as follows.

1. He points out that any universe is bound to have the appearance of being designed.24 For there could not be a universe at all in which the parts were not adapted to one another to a considerable degree. There could not, for example, be birds that grew wings but, like fish, were unable to live in the air. The persistence of any kind of life in a relatively fixed environment presupposes order and adaptation, and this can always be thought of as a deliberate product of design. The question is, however, whether this order could have come about otherwise than by conscious planning. As an alternative, Hume suggests the Epicurean hypothesis. The universe consists of a finite number of particles in random motion. In unlimited time these go through every combination that is possible to them. If one of these combinations constitutes a stable order (whether temporary or permanent), this order will in due course be realized and may be the orderly cosmos in which we now find ourselves.

This hypothesis provides a simple model for a naturalistic explanation of the orderly character of the world. The model can be revised and extended in the light of the special sciences. The Darwinian theory of natural selection, for example, presents a more concrete account of the apparently designed character of animal bodies. According to Darwin's theory, there are in every generation small random variations between individuals, and species are relatively well adapted to their environment for the simple reason that the less well-adapted individuals have perished in the continual competition to survive and so have not perpetuated their kind. The "struggle for survival," operating as a constant pressure toward more perfect adaptation, lies behind the evolution of life into increasingly complex forms, culminating in homo sapiens. To refer back to the ozone layer, the reason animal life on earth is so marvelously sheltered by this filtering arrangement is not that God first created the animals and then put the ozone layer in place to protect them, but rather that the ozone layer was there first, and only those forms of life capable of existing in the precise level of ultraviolet radiation that penetrates this layer have developed on earth.

2. The analogy between the world and a human artifact, such as a watch or a house, is rather weak.25 The universe is not particularly like a vast machine. One could equally plausibly liken it to a great inert animal such as a crustacean, or to a vegetable. In this case the design argument fails, for whether crustaceans and vegetables are or are not consciously designed is precisely the question at issue. Only if the world is shown to be rather strikingly analogous to a human artifact, which we know to be designed, is there any proper basis for the inference to an intelligent Designer.

3. Even if we could validly infer a divine Designer of the world, we would still not be entitled to postulate the infinitely wise, good, and powerful God of the Judaic-Christian tradition.26 From a given effect we can only infer a cause sufficient to produce that effect; therefore, from a finite world we can never infer an infinite creator. To use an illustration of Hume's, if I can see one side of a pair of scales and can observe that ten ounces is outweighed by something on the other side, I have good evidence that the unseen object weighs more than ten ounces; however, I cannot infer from this that it weighs a hundred ounces, still less that it is infinitely heavy. On the same principle, the appearances of nature do not entitle us to affirm the existence of one God rather than many, since the world is full of diversity; nor of a wholly good God, smce there is evil as well as good in the world; nor, for the same reason, of a perfectly wise God or an unlimitedly powerful one.

It has, therefore, seemed to most philosophers that the design argument, considered as a proof of the existence of God, is fatally weakened by Hume's criticisms.

21 Paley's book has become available in an abridged version, ed. Frederick Ferre, in the Library of Liberal Arts, 1962.

22 For example, Robert E. D. Clark, The Universe—Plan or Accident? (Philadelphia: Muhlenburg Press, 1961).

23 Arthur I. Brown, Footprints of God (Findlay, Ohio: Fundamental Truth Publishers, 1943), p. 102.

24 Hume, Dialogues Concerning Natural Religion, Part VIII

25 Dialogues, Parts VI, VII.

26 Dialogues, Part V. Cf. An Enquiry Concerning Human Understanding, Sec. XI, para. 105.


Since Hume's time a broader form of design argument has been offered by F. R. Tennant27 and others, claiming that when we take account of a sufficiently comprehensive range of data—not only the teleological character of biological evolution but also man's religious, moral, aesthetic, and cognitive experience28—it becomes cumulatively more probable that there is a God than that there is not. Theism is presented as the most probable world-view or metaphysical system.

These thinkers claim that a theistic interpretation of the world is superior to its alternatives because it alone takes adequate account of man's moral .and religious experience, as well as giving due place to the material aspects of the universe. Needless to say, this claim is disputed by nontheistic thinkers, who point in particular to the existence of evil as something that fits better into a naturalistic than into a religious philosophy. The problem of evil will be discussed in Chapter 4; the question to be considered at the moment is whether the notion of probability can properly be applied to the rival hypotheses of the existence and nonexistence of God.

Two main theories of probability, the "frequency" theory and the ".reasonableness of belief" theory, are found in contemporary writings on the subject, developing what are sometimes called the statistical and inductive senses of probability. According to the first, probability is a statistical concept, of use only where there is a plurality of cases.29 (For example, since a die has six faces, each of which is equally likely to fall uppermost, the probability of throwing any one particular number at a given throw is one in six.) As David Hume points out, the fact that there is only one universe precludes our making probable judgments of this kind about it. If—impossibly—we knew that there were a number of universes (for example, ten) and if in addition we knew that, say, half of them were God-produced and half not, then we could deduce that the probability of our own universe's being God-produced would be one in two. However, since by "the universe" we mean the totality of all that is (other than any creator of the universe), clearly no reasoning based upon the frequency theory of probability is possible concerning its character.

According to the other type of probability theory, to say that statement p is more probablejhan statement q is to say that when they are both considered in relation to a common body of prior (evidence-stating) propositions, it is more "reasonable" to believe p than q, or p is more worthy of belief than q.30 The definition of reasonableness of course presents problems; but there is another special difficulty that hinders the use of this concept to assess the "theous" or "nontheous" character of the universe. In the unique case of the universe as a whole there is no body of prior evidence-stating propositions to which we can appeal, since all our propositions must be about either the whole or a part of the universe itself. In other words, there is nothing outside the universe that might count as evidence concerning its nature. There is only one universe, and this one and only universe is capable of being interpreted both theistically and nontheistically.

It has been suggested that we may speak of "alogical" probabilities and may claim that in a sense that operates in everyday common-sense judgments, although this is not capable of being mathematically formulated, it is more likely or probable that there is than that there is not a God.31 According to this view, the considerations that support the. God hypothesis are entitled to greater weight than those that suggest the contrary hypothesis. This, however, is clearly a question-begging procedure, for there are no common scales on which to weigh, for example, the human sense of moral obligalian against the reality of evil, or humanity's religious experience against the fact of human iniquity. Nor is there any valid sense in which it can be said that a religious interpretation of life is antecedently more probable than a naturalistic interpretation, or vice versa. Since we are dealing with a unique phenomenon, the category of probability has no proper application to it.

27 F. R. Tennant, Philosophical Theology, II (Cambridge: Cambridge University Press, 1930), Chap. 4.

28 Richard Taylor in Metaphysics (another volume in the Foundations of Philosophy Series), Chap. 7, makes striking use of man's cognitive experience in a reformulated design argument.

29 See, for example, Morris R. Cohen, A Preface to Logic (London: Routledge & Kegan Paul Ltd., 1946 and New York: Dover Publications, Inc., 1944), Chap. 6.

30 See, for example, Roderick M. Chisholm, Perceiving (Ithaca, N.Y.: Cornell University Press, 1957), Chap. 2.


The moral argument, in its various forms, claims that ethical experience, and particularly one's sense of an inalienable obligation to fellow human beings, presupposes the reality of God as in some way the source and ground of this obligation.

First Form

In one form the argument is presented as a logical inference from objective moral laws to a divine Law Giver; or from the objectivity of moral values or of values in general to a transcendent Ground of Values; or again, from the fact of conscience to a God whose "voice" conscience is—as in the following passage by Cardinal Newman:

If, as is the case, we feel responsibility, are ashamed, are frightened, at transgressing the voice of conscience, this implies that there is One to whom we are responsible, before whom we are ashamed, whose claims upon us we fear. . . . If the can of these emotions does not belong to this visible world, the Object to which [the conscientious person's] perception is directed must be Supernatural and Divine.32

The basic assumption of all arguments of this kind is that moral values are not capable of naturalistic explanation in terms of human needs and desires, self-interest, the structure of human nature or human society, or in any other way that does not involve appeal to the Supernatural. But to make such an assumption is to beg the question. Thus, an essential premise of the inference from axiology to God is in dispute, and from the point of view of the naturalistic skeptic nothing has been established.

Second Form

The second kind of moral argument is not open to the same objection, for it is not strictly a proof at all. It consists of the claim that anyone seriously committed to respect moral values as exercising a sovereign claim upon his or her life must thereby implicitly believe in the reality of a transhuman source and basis for these values, which religion calls God. Thus, Immanuel Kant argues that both immortality and the existence of God are "postulates" of the moral life, i.e., beliefs which can legitimately be affirmed as presuppositions by one who recognizes duty as rightfully laying upon one an unconditional claim.33 Again, a more recent theological writer asks:

Is it too paradoxical in the modern world to say that faith in God is a very part of our moral consciousness, without which the latter becomes meaningless? . . . Either our moral values tell us something about the nature and purpose of reality (i.e., give us the germ of religious belief) or they are subjective and therefore meaningless.34

It seems to the present writer that so long as this contention is not overstated it has a certain limited validity. To recognize moral claims as taking precedence over all other interests is, in effect, to believe in a reality, other than the natural world, that is superior to oneself and entitled to one's obedience. This is at least a move in the direction of belief in God, who is known in the Judaic-Christian tradition as the supreme moral reality. But it cannot be presented as a proof of God's existence, for the sovereign authority of moral obligation can be questioned; and even if moral values are acknowledged as pointing toward a transcendent ground, they cannot be said to point all the way and with unerring aim to the infinite, omnipotent, self-existent, personal creator who is the object of biblical faith.

31 See, for example, Tennant, Philosophical Theology, I, chap. 11.

32 J. H. Cardinal Newman, A Grammar of Assent, 1870, ed. C. F. Harrold (New York: Davidl McKay Co., Inc., 1947), pp. 83-84.

33 Critique of Practical Reason, Book II, Chap. 2, Sees. 4 and 5.

34 D. M. Baillie, Faith in God and Its Christian Consummation (Edinburgh: T. & T. Clark, 1927), pp. 172-73.


It has also been claimed that various special happenings of a publicly observable kind, such as miracles and answers to prayer, establish the reality of God. It is doubtless true as a matter of psychological fact that a sufficiently impressive series of such happenings, if personally witnessed, would move almost anyone, however skeptical, to believe in God. But no general proof of divine existence, valid for those who have not experienced such events, can be based upon this fact. For they can always either disbelieve the reports, for reasons classically stated by David Hume in his essay on Miracles,35 or accept them but give them a naturalistic interpretation. The relatively new but potentially highly significant science of parapsychology has already greatly enlarged the range of naturalistic explanations of the "supernatural" for those who are willing to extend the sphere of the natural to include such phenomena as extrasensory perception (telepathy), awareness of future events (precognition), and even the alleged power of the mind to influence directly the movements of matter beyond the boundaries of one's own body (psychokinesis).

More private but still dramatic manifestations of God in vision and dream, by inner voice, numinous feeling, or mystical or ecstatic experience have also convinced many of the reality of God. Once again, though, it is not possible to found upon these experiences a general proof of divine existence. As the skeptical Thomas Hobbes remarked, when a man tells me that God has spoken tohim in a dream, this ". . . is no more than to say he dreamed that God spake to him."36 This point has been made again in recent philosophical critiques of the claim to meet the unseen God in a spiritual "I-Thou" encounter. Although the believer has no doubt had the experience that he or she reports and regards as an I-Thou awareness of God, the having of this experience does not guarantee the truth of the experiencer's interpretation of it. It may be that he or she had the experience described but that the correct explanation of it can be given by psychology rather than, by theology.37

In short, any special event or experience that can be construed as manifesting the divine can also be construed in other ways, and accordingly cannot carry the weight of a proof of God's existence.

From this discussion, it is evident that the writer's own conclusion concerning the theistic proofs is negative. None of the arguments which we have examined seems qualified to compel belief in God in the mind of one who lacks that belief. However, it should be said in conclusion that many religious thinkers would disagree with this assessment and hold that one or another of the traditional arguments, or several of them in combination, are rationally persuasive.

35 An Enquiry Concerning Human Understanding, Sec. X.

36 Leviathan, Chap. 32.

37 See, for example, C. B. Martin, Religious Belief (Ithaca, N.Y.: Cornell University Press, 1959), Chap. 5; and Ronald W. Hepburn, Christianity and Paradox: Critical Studies in Twentieth- Century Theology (London: C. A. Watts & Company Ltd., 1958 and Indianapolis: The Bobbs- Merrill Co., Inc., 1968), Chaps. 3 and 4.

38 For interesting contemporary positive treatments see James F. Ross, Philosophical Theology (Indianapolis: The Bobbs-Merrill Co., Inc., 1969), and Richard Swinburne, The Existence of God (New York: Oxford University Press, 1979).