Principles of Natural Theology / by George Hayward Joyce, S.J.

Chapter VII. Kant's Criticism and his Alternative Argument.

  1. Kant's Criticism and its Fruits.
  2. His Criticism of the Cosmological Argument.
  3. His Criticism of the Teleological Argument.
  4. His Alternative Proof from the Practical Reason.

1. Kant's criticism and its Fruits. Kant, it is well known, challenged the validity of the traditional proofs of God's existence. In his Critique of Pure Reason he examines the cosmological and the teleological arguments, and declares that they are both vitiated by a latent fallacy. In each case, he says, the reasoning of the ontological argument is surreptitiously introduced into the proof, so that the value of the conclusion stands or falls with that method of demonstration. That argument, however, he has already weighed in the balance and found wanting: it follows that these two famous proofs of God's existence are alike worthless. The mind, moreover, he contends, must recognize its incapacity to reach this conclusion by any demonstration of the speculative reason: for there neither are, nor can be, any other lines of proof save the three just mentioned. Either we argue from the peculiar constitution of the world of sense: or we argue from the nature of finite being as such, abstracting from any special laws of physical nature: or, lastly, we abstract from all experience, and base our reasoning on a priori conceptions alone. In the first case we have the teleological proof: in the second, the cosmological: while the third alternative gives us the ontological argument employed by St. Anselm and Descartes. Beyond these no other proofs by way of the speculative reason are possible. It is unnecessary to point out that this assertion that there is but one proof based on physical laws, and but one based on the nature of finite being, is a mere assumption, and altogether erroneous. Kant further undertakes to shew how it comes about that the human intellect in virtue of its very constitution is driven by a native tendency to the illusion of a personal first cause, and is thus disposed to buttress up its figment by fallacious arguments. What, however, he takes away with one hand, he seeks to restore with the other. He maintains that though the speculative reason cannot provide us with a proof of God's existence, the practical reason will supply the deficiency.

It may be of interest here to give a summary account of his views regarding the alleged tendency of the reason to posit a personal first cause irrespectively of any real grounds for belief in its existence. Some leading features of the Kantian system have been already explained (chap. ii., § 5). It will he remembered that he held that things as we know them -- phenomena -- are not objective realities, but are the work of our own mind operating on and organizing the internal subjective sensations which are its sole data. The realities to which those sensations are ultimately due -- the noumena -- we cannot know. Space and time, in which things are known to us, are forms of sensibility. Our conceptions of things, whether as substances or accidents, and all our judgments regarding them, are determined by the categories of the understanding. Reason likewise has its laws: and these laws determine for us the whole process of our ratiocination. In argument, the mind is not, as we fondly imagine, taking each step in obedience to the assured claims of truth and in conformity with a real world of which it is representative. It travels along a road prescribed by its internal constitution. Its characteristic operation is the arrangement of our cognitions in a systematic unity. In virtue of this natural tendency we are irresistibly led to view all phenomena in reference to the total of reality -- this total, however, being a mere ideal and having no objective validity. Further, we come to regard this ideal total as the common ground or cause of all things, and hence as containing in itself all perfection. In this manner we reach the idea of God. But "we have not," he says, "the slightest ground to admit the existence of an object corresponding to this idea . . . . The idea of this Being is essentially nothing more than a demand upon reason that it shall regulate the connection which it and its subordinate faculties introduce into the phenomena of the world by principles of systematic unity, and consequently that it shall regard all phenomena as originating from one all-embracing Being, as the supreme and all-sufficient cause."{1}

It is not to be denied that ever since Kant's time an impression has prevailed widely that the old proofs are no longer defensible. Possibly the mere fact that an eminent thinker had ventured to call in question such seemingly irrefutable arguments seemed by itself almost equivalent to a disproof. But another reason also, extrinsic it is true to the merits of the criticism, but none the less effective, operated in favour of this result. During the last century, rationalism, in the form either of naturalism or of idealism, had become strongly entrenched in the great centres of learning. It was only natural that thinkers who had discarded belief in a personal God should applaud Kant's conclusion, even if they might hesitate to affirm that his criticism of the proofs was in all respects sound. Thus it came about that those who admitted the value of the traditional arguments were regarded as out of date. Often the validity of Kant's objections is simply taken for granted, and the proofs of God's existence dismissed without more ado.{2} Even some of the apologists of revealed religion, eager not to be behind the fashion, discard them as untenable. In Lux Mundi (1889), a book written by men of real academical distinction in view of the intellectual difficulties against faith then commonly felt, and intended to provide a reasoned defence of Christianity, the author of the essay on "The Christian Doctrine of God," says that in his opinion "there can be no proofs in the strict sense of the word of the existence of God. . . . Neither conscience nor the speculative reason can demonstrate God's existence" (p. 103){3}: and proceeds to base his defence of the doctrine on the fact that belief in God is spontaneous and instinctive, and that it is confirmed alike by the experience of life and by the conclusions, so far as they will go -- which is not far -- of the speculative reason. Such an answer, it is plain, is no adequate reply to the rationalist attack. If philosophy can do no more for us than this on the most vital of all issues which fall within its scope, it is hard to avoid the conclusion that the popular idea which sees in metaphysics, once reckoned as the supreme science, nothing but profitless disputation leading to no practical result, is amply justified. Such, however, has never been the attitude of Scholastic thinkers. They have consistently denied that Kant detected a flaw in any one of the great traditional proofs of God's existence. On the contrary, they maintain that his analysis of these arguments is sophistical through and through, and destitute of all real value: and that the argument from the practical reason which he proposed as a sufficient substitute for those which he discarded, is itself quite incompetent to sustain the test of a critical examination. In the subsequent sections of the present chapter we shall endeavour to make good these contentions.

2. Criticism of the cosmological argument. Kant bases his attack on the cosmological proof chiefly, as we have already noted, on the ground that when analysed, it is found to reach its conclusion by the concealed employment of the ontological argument. It professes to argue from existing contingent beings to the existence of a necessary being, who is God. As a matter of fact, he maintains, it arrives at the existence of necessary being, not from the contingent existences which are its alleged ground, but purely from our ideal conception of the absolutely perfect being. He has other charges also against the proof. These, however, may for the moment be deferred. Our immediate task is to examine his first and principal gravamen.

The conclusion, Kant reminds us, is reached by establishing the two propositions: 'Contingent being involves the existence of necessary being': and, 'Necessary being is supreme perfection, i.e., God (Ens necessarium est Ens realissimum).' From these two premisses it is claimed that we have demonstrated God's existence. Yet so far, he assures us, is this from being the case that the second of the two propositions is worthless for our purpose, having no bearing on the existential world unless the validity of the ontological argument be presupposed. This he proceeds to shew as follows, The proposition, Necessary being is Supreme Perfection, is equivalent as logic teaches us to its converse. This converse would by the rules of logic take the form, Some supremely perfect beings are necessary beings. But as there can be but one supremely perfect being, we may state it as, Supreme Perfection is necessary being. This is neither more nor less than the ontological fallacy which has slipped in unobserved. The conclusion, which seemed at first sight to be based on experience, is really determined a priori by our conceptions, the notion of perfect being being conceived as necessarily involving its own existence in the real order.

The fallacy of this criticism is patent. In the cosmological argument the proposition, 'Necessary being is supreme perfection,' is concerned with an existing reality. We have already proved that a necessary being exists. We are now concerned to discover the nature of that being. Its necessity enables us to conclude to certain other of its attributes: and in this way we realize that this being of whose existence we are certain is in fact the living God, the source of all reality. In the ontological argument it is otherwise. The proposition, The perfect being is necessary, does not deal with an object whose existence is known. We are concerned solely with the concept of perfect being: and since we have only a vague and indeterminate notion of supreme perfection, we are not even aware whether such a concept is possible of realization in the objective order. We are not, in fact, justified in stating that supreme perfection is necessary, but only that supreme perfection, if it exists, is necessary.

It may well excite our wonder that Kant should have fallen into so manifest a sophism. But his discussion of the argument shews us how the error arose. He tells us that the a posteriori argument from experience only enables us to conclude to the existence of a necessary being of some kind or another, but is incompetent to shew us what particular thing possesses this attribute of necessity. "Experience," he says, "merely aids reason in making one step -- to the existence of a necessary being. What the properties of this being are cannot be learned from experience. . . . Experience is utterly insufficient to demonstrate the existence of this attribute [of necessity] in any determinate existence or thing."{4} He fails to see that there is no need for us to look to experience to shew us a determinate thing which is necessary: that just as reason is able to take the step which leads it from the existence of contingent being to the existence of necessary being, so it can go on to shew that necessary being cannot be limited, but must be being in the fullness of its perfection. Thus he comes to give a wholly erroneous account of the manner in which we reach the minor premiss, Necessary being is supreme perfection. "When we propose to ourselves an aim of this character," he says, "we must abandon the sphere of experience and rise to that of pure conceptions, which we examine with the purpose of discovering whether any one of them contains the conditions of the possibility of an absolutely necessary being." We find these conditions in our conception of Ens realissimum, Supreme Perfection. Moreover, not merely is Supreme Perfection conceived as having the attribute of necessity; but among all our conceptions it is the sole nature which has this attribute. Hence we are justified in identifying it with the Ens necessarium, whose existence we have proved, and saying, Necessary being is Supreme Perfection. But here, he contends, is the fallacy. We have no right to make this identification: for we have no ground for regarding our concept of Supreme Perfection as something which belongs to the real order. It is a mere concept: and when we treat it as something which belongs to the real order, and identify it with Ens necessarium, we are simply employing the illegitimate method of the ontological argument.

This account of the way in which the minor premiss is obtained is clearly quite inaccurate. Conceiving it thus, it is easy to understand how Kant came to regard it as invalid. But his criticism is throughout based on the strangest of misapprehensions.

Though this is Kant's principal accusation against the cosmological argument, it is not, as we have already said, his only one. Indeed, he goes so far as to say that it "contains a perfect nest of dialectical assumptions, which transcendental criticism does not find it difficult to expose and to dissipate": and he proceeds to enumerate four of the alleged assumptions. These we shall consider seriatim.

(1) In the first place, he urges that the category of causality belongs solely to the world of phenomena, and that we have no right to treat it as valid also for that world of noumenal realities to which God is held to belong. Causality is a mere 'form' imposed by the intellect. In virtue of our mental constitution we conceive every phenomenon as related to some previous one as its effect; but this does not justify us in believing that the relation holds good outside our minds, and in viewing God as the first cause of the world. The objection finds clear expression in the following passage:

"If the Supreme Being forms a link in the chain of empirical conditions, it must be a member of the empirical series, and like the lower members which it precedes, have its origin in some higher member of the series. If, on the other hand, we disengage it from the chain, and cogitate it as an intelligible being apart from the series of natural causes -- how shall reason bridge the abyss that separates the latter from the former? All laws respecting the regress from effects to causes, all synthetical additions to our knowledge relate solely to possible experience and the objects of the sensuous world, and apart from them are without significance" (op. cit. p. 382).

This objection rests entirely on Kant's theory regarding the data and forms of knowledge. That theory we reject as radically unsound. We have already argued at length (chap. ii.) that causality is no subjective contribution of our own without objective validity, but that it is a fact of the objective order: and that the principle of causality is a metaphysical truth of supreme certainty. The whole elaborate edifice of Kant's theory is, we contend, built on the gratuitous assumption made by Descartes, and accepted by Locke and Hume, that we are conscious of nothing but our own internal states.{5}

(2) As his next objection Kant urges that we have no justification for asserting the impossibility of an infinite series. The principles of reason, he holds, do not permit us to affirm the repugnance of such a series even as regards the world of experience. Much less, then, is it allowable as regards noumenal realities.

The objection is valid according to the principles of the Kantian philosophy. If there were a law of our mentality compelling us to regard every phenomenon as proceeding from a previous phenomenon, it would follow that the empirical series must be infinite a parte ante. We have already treated the question from the point of view of a metaphysic more conformable to reality. It was shewn that although there is room for difference of opinion regarding the possibility or impossibility of a series of causes whose sole function is to bring the subsequent cause into existence, this is not so as regards the causes whose actual operation is needed hic et nunc for the realization of the effect. An infinite series of these is intrinsically repugnant: and it is with these causes that the cosmological argument is concerned.

(3) In the third place we are told that the idea of necessity as applied to the First Cause is illegitimate and meaningless. The First Cause is, by hypothesis, unconditioned. But necessity is conceived as resulting from the fulfilment of requisite conditions. This, in fact, is the meaning of necessity, and apart from this no significance attaches to the notion.

To this we reply that Kant fails to distinguish between dependent entities and the independent, self-existent Being. Without doubt the necessity proper to dependent beings arises from the fulfilment of the conditions requisite for their realization: and we cannot conceive it save in this manner. But the self-existent First Cause is not a dependent being. It is not possible for us to form a positive concept of the necessity proper to such a Being, any more than of His infinity. Yet we gain a valid and significant idea of it when we conceive it as freedom from all conditions, just as we form a significant idea of infinity as being that which has no limits, and of eternity as that which has neither beginning nor end.

(4) Kant's fourth objection is that because we find no internal contradiction in the idea of a sum-total of reality, we conclude that this is capable of objective existence as an individual thing. We have, he maintains, no right to make this transition, unless we have first shewn the practicability of the objective synthesis. And yet, even did we do so, it would avail us little: for the synthesis thus vindicated would have reference solely to the world of phenomenal experience and not to reality.

The answer here is simple. We do not conclude the objective possibility of a Being, who sums up in Himself all reality, on the ground that we can detect no logical contradiction in the concept of the sum-total of perfection. We conclude to the possibility because we have shewn that such a Being exists. We demonstrate a posteriori that necessary being is, and by considering what attributes must pertain to a necessary being, we are able to conclude that this existing entity is the sum of all reality. Nor do we view this entity as the congeries of natural perfections, but as of an altogether higher order. In Him as in their sole source must be gathered up all that this world displays in such various forms and such diverse harmonies of goodness, beauty and truth. Yet created perfections, though of a lower order, nevertheless afford a sure basis for our argument. For they are objectively real: they are no phenomenal construction of our own minds.

It would seem, then, that "the nest of dialectical assumptions" was of Kant's own making. The traditional proof a contingentia is free from fallacy. But on the other hand, each one of his five objections is radically sophistical.

3. Criticism of the teleological argument. Of the teleological argument Kant speaks in terms of high regard. He says:

"It is the oldest [argument], the clearest, and that most in conformity with the common reason of humanity. . . . It would be utterly hopeless to attempt to rob this argument of the authority it has always enjoyed. The mind unceasingly elevated by these considerations which, although empirical, are so remarkably powerful, and continually adding to their force, will not suffer itself to be depressed by the doubts suggested by subtle speculation: it tears itself out of this state of uncertainty, the moment it casts a look upon the wondrous forms of nature, and the majesty of the universe, and rises from height to height, from condition to condition, till it has elevated itself to the supreme and unconditioned Author of all" (op. cit. p. 383).

On what ground, then, does he attempt to achieve the task which he declares to be a hopeless endeavour -- to rob this argument of its authority? The proof, he tells us, is based on the fact that there are in the world manifest signs of an arrangement full of purposes, and that "this arrangement is foreign to the things existing in the world -- it belongs to them as a contingent attribute." Things do not tend to these ends in virtue of their own nature: the arrangement must be imposed upon them from without by an intelligent cause. Now this reasoning carries us, he urges, no further than to an architect of the universe, whose efforts are limited by the materials in which he works. It will not take us to a creator: and consequently is no proof of the existence of God. Moreover, even as it is, we get no determinate idea of this architect of the world. We see that he must be 'very great' and 'very wise.' But such predicates as these merely indicate the relation which his greatness and wisdom bears to ours: they do not give us any definite knowledge regarding him. We have not any sufficient ground for judging that any attribute is absolutely his. Though wiser and greater than us, he may not be, absolutely speaking, wise or great. We cannot pronounce finally whether he is one or several, good or evil. Our knowledge is indeterminate. How then do we reach our conclusion? We fall back, Kant avers, on the cosmological proof. From this order and conformity to external aims, we infer the contingency of the world. The contingency of the world enables us to reason to necessary being: and then by the invalid ontological process we conclude that God exists. Only thus does our concept become determinate. For no other concept than that of the plenitude of reality, which contains all possible perfection, can he regarded as such.

Both these criticisms are unsound. It is, doubtless, true that the proof from final causes does not establish that the Cause who organized the world stands to it in the relation of Creator. It is unreasonable to demand that it should do so. To prove that God did not form the world out of preexisting matter, but created it, is a subsequent step to demonstrating His existence. The denial of a divine attribute may logically lead to the denial of God; but the questions involved are distinct: and a proof of God's existence is not bound to indude a proof of His attributes. It may, indeed, be urged that the teleological argument does not even shew that the architect of nature is self-existent: and that without this it cannot he said to shew that he is, in fact, God. Many theists admit that the objection as so stated is not without force. They grant that to be rigidly complete the teleological proof must be supplemented by the cosmological argument. That argument they hold, of course, to be absolutely free from any connection with the ontological proof. They point out, however, that practically, the teleological demonstration, even as considered separately, is decisive of the issue of theism. No one who admits the existence of an intelligent Author of Nature will have the smallest doubt that he has proved the existence of God. We have shewn reason in chap. iv. for thinking that even this concession is unnecessary. The teleological argument, as we pointed out, carries us not simply to an architect of nature, but to a First Source of order. Such a First Source must necessarily be self-existent.

The second objection, that the proof gives us no determinate knowledge regarding the Author of Nature is equally fallacious. Here, just as above, we must demand to keep the full treatment of the Divine attributes for later consideration. We are not bound, when establishing the existence of God, simultaneously to demonstrate the infinity of all His attributes. But we know that His wisdom and His power must be proportionate to the effects on which the teleological argument is based: that they are such as could devise and could realize the marvellous scheme of things which the created world offers to our contemplation -- surely a perfectly definite idea. The unity of the order displayed throughout the universe affords a weighty argument that He is One. But the formal proof that He is infinite in all His attributes, and the full discussion of His unity, are logically posterior to the question proposed to us.{6}

4. Proof of God's existence from the practical reason. While Kant denied the capacity of the speculative reason to establish God's existence, he believed that in the practical reason he had found a secure basis for holding this truth. Man's highest good (summum bonum), he tells us, consists, not in virtue alone, but in virtue united to happiness. For happiness is a good: and though virtue is a higher good than happiness, yet goodness wants something to its completeness if happiness be not joined to it. It is our duty to be ever aiming at the summum bonum, though the motive of our actions must be the moral law itself, and not the desire of the highest good. Here, however, we are faced by a difficulty. The causal series which constitute the world of experience do not operate in favour of virtue: they are altogether indifferent to it. Happiness does not result from a virtuous life, but from a knowledge of physical law combined with power to employ it for our advantage. The only manner in which it is possible to suppose a connection between virtue and happiness is to postulate the existence of a Supreme Cause distinct from nature and able to effect a union between the two. Such a cause we signify by the name, God. Hence we are justified in affirming the existence of God as the holy and omnipotent Author of the world.

Is this an act of intellectual cognition? Emphatically no, Kant replies. The speculative reason leads us to a very different issue, shewing us that our idea of God arises from a dialectical illusion. Our belief in God is a moral conviction based upon our practical needs. It is a 'postulate.' The act is not an act of knowledge, but of faith. This employment of the term 'faith' merits attention: for it involves a radical change in its significance. The word in its accepted meaning denotes an intellectual assent given to truths which we know to have been revealed by God -- an assent which is grounded on an assured certainty of God's existence, and of the fact of revelation. With Kant it is used to signify a blind adhesion, in default of knowledge, to the 'postulate' of God's existence. Strictly speaking, as a matter of speculation, the postulate is still an hypothesis; but in view of our practical needs it becomes a legitimate object for an act of 'faith.' Yet this act must not be employed as a basis for any theoretic conclusions as to the nature of God: our speculative knowledge is not thereby enlarged.{7}

Kant's reason for thus refusing to allow the fact of God's existence revealed by the practical reason to be employed as a basis for the extension of our speculative knowledge, was that he held 'synthetical' propositions to be possible only in regard of objects which are capable of presentation to the senses.{8} God, it is manifest, cannot be sensibly experienced.

The reasoning here is marred by a fatal flaw. There is in Kant's system no ground whatever for his contention that happiness must be connected with virtue. This is a legitimate conclusion, when we already know that God exists, and that He has imposed upon us the obligation of the moral law. In that case it follows that the Supreme Lawgiver will not have left His law without its due sanctions. But if we have no assurance of God's existence, and do not regard the moral law as a divinely imposed obligation, there is no reason whatever to suppose that virtue will secure happiness for those who practise it. Yet Kant assumes that it is so, and builds his conviction of God's existence on this assumption. He does not postulate God's existence as that of the law-giver who will avenge wrong-doing and reward the just. He consistently maintains that the moral law makes no appeal to any higher lawgiver than reason itself. He concludes to God's existence, only because he holds it evident that the summum bonum must include happiness. This, however, on his principles, is a gratuitous assumption. Happiness and unhappiness, as he conceives them, belong to the phenomenal order, being dependent on physical law. We have no ground for holding that they possess any noumenal counterpart. Moral excellence is noumenal. It may, so far as we know, be wholly unrelated to happiness. It would seem that, on Kantian principles, where moral obligation is concerned, the phenomenal facts of happiness and unhappiness should be wholly disregarded as being altogether outside the question.

But even were Kant's reasoning as valid as it is invalid, how lame and impotent is the conclusion to which he brings us! We desire to know not merely God's existence, but much else about Him. Does He concern Himself with us, or is He wholly occupied in His own essential bliss? Does He exercise an immediate and direct providence in regard of each several individual or not? And many other questions are there which it vitally concerns us to know. Kant replies that we must be content to be utterly ignorant of these things: that we can know nothing but the bare fact that He is: and that even this is not properly knowledge, but a conviction based on a practical need. Such a theory is, as has well been said, the union of practical dogmatism with complete speculative agnosticism.

{1} Critique of Pure Reason, Pt. II., Div. ii., Book 2, c. iii., § 7 (Meiklejohn's trans., p. 420).

{2} Cf. J. Ward, Realm of Ends (1911). "Can we then prove the existence of God? Attempts innumerable to prove this have been made -- as of course we know -- all of them reducible to one or other of the three forms called respectively the ontological, the cosmological and the teleological argument. The fatal defects of all these have, it is almost universally conceded, been clearly exposed once for all by Kant."

{3} Mr. Aubrey Moore. It would seem as if he had hardly troubled even to acquaint himself with the arguments which he so lightly sets aside. For he writes: "The so called proof a contingentia (which underlies H. Spencer's argument for the Unknowable) is an appeal to that very consciousness of dependence which some people consider a weakness and a thing to educate themselves out of." It is difficult to attach any reasonable meaning to such a statement.

{4} Op. cit., p. 373.

{5} The objection that it is impossible to argue from the chain of contingent causes and effects to a cause which, being necessary, is of a different order, is still urged by some writers. Thus Principal Caird says: "You cannot in a syllogistic demonstration put more into the conclusion than the premises contain. . . All that from a finite or contingent effect you can infer is a finite or contingent cause, or at most an endless series of such causes. But if, because the mind cannot rest in this false infinity, you try to stop the indefinite regress, and assert at any point of it a cause which is not an effect, which is its own cause, or which is unconditioned and infinite, the conclusion in this case is purely arbitrary." Introd. to Phil. of Religion, p. 129. It would seem as though he had failed to see that not the contingent substances themselves, but the analytic proposition "Contingent being involves the existence of necessary being" provided the premises of the syllogism. The same objection is found in Edw. Caird's Phil of Kant (1877), p. 646, and in Illingworth's Personality Human and Divine, p. 93.

{6} In the Critique of Judgment Kant calls the teleological argument in question on a new ground. He contends that the necessity which we feel of attributing finality to nature is purely subjective. We cannot study nature unless we view it as a unity. And we effect this end by conceiving it as a system such as a mind would have established; but it is quite conceivable that the order of nature may be due to purely mechanical principles without finality of any kind. This objection has its true origin in Kant's doctrine of the Categories. That teaching leaves, in fact, no room whatever for finality in nature: and Kant is thus logically compelled to suppose that the principle of finality is a sort of working hypothesis devoid of objective value.

{7} Critique of Practical Reason, Pt. I., Bk. 2, c. ii., § 5 - § 7 (Abbott's trans., pp. 223-234).

{8} Cf. the citation given on p. 225. On the significance of the term 'synthetical proposition' in the Kantian philosophy see pp. 48, 49.

<< Principles of Natural Theology >>