Kant's theistic solution


by Stephen Palmquist

1. The Problem of Transcendental Theology

Kant's transcendental philosophy begins with an attempt to solve the theoretical problem of the possibility of synthetic a priori judgments. In solving this epistemological problem Kant demonstrates how transcendental knowledge (i.e., knowledge of the synthetic a priori conditions for the possibility of experience) is possible only when its application is confined to the realm of empirical knowledge (i.e., to experience). He argues that space, time, and the twelve categories form the transcendental boundary line between what we can and cannot know. But this 'solution' itself calls attention to an even more significant problem: what is the status of that which lies outside the boundary of possible empirical knowledge? Kant reveals as early as CPR xxix-xxxi1 that this metaphysical problem of how to verify the fundamental human ideas of 'God, freedom, and immortality', upon which he believes all religion and morality depend, constitutes the deepest and most urgent form of the 'transcendental problem'. It should therefore come as no surprise when he devotes the entire Transcendental Dialectic, the largest section of the first Critique, to the task of solving this ubiquitous perplexity of human reason.

According to Kant our ideas of God, freedom, and immortality inevitably arise in the human mind as a result of our attempts to unify and systematize our empirical knowledge. In other words, reason naturally seeks for something beyond the limits of empirical knowledge which can supply unity and coherence to the diversity of facts which fall within that boundary. The problem is that the transcendental conditions which enable us to gain knowledge in the empirical world are unable to perform their function with respect to such ideas, because the ideas abstract from all sensible content, whereas the transcendental conditions (space, time and the categories) all require such content.

As is well known, Kant devoted considerable effort in the Transcendental Dialectic to the task of pointing out the implications of this transcendental problem for rational psychology (with its proofs of the immortality of the soul), rational cosmology (with its proofs of transcendental freedom), and rational theology (with its proofs of the existence of God). Interpreters often assume Kant sought to demonstrate the total uselessness of all such 'speculative' disciplines, especially when it comes to theology, where he offers his radical criticisms of the three traditional proofs for the existence of God. Since Kant's division and negative assessment of these proofs has become common knowledge among theologians and philosophers of religion, there is no need to rehearse his position here.

The attentive reader of Kant's writings will notice that he regarded the failure of the traditional proofs for God's existence not as closing the books on the issue, but as posing one of the most important problems for transcendental philosophy to resolve. Although some theologians fear that Kant's criticism of traditional rational theology could, in the long run, have a detrimental effect on the ordinary religious believer, Kant's disagreement with such a 'sophisticated' conjecture is explicit and to the point: In religion the knowledge of God is properly based on faith alone .... [So] it is not necessary for this belief [i.e., in God] to be susceptible of logical proof.... [For] sophistication is the error of refusing to accept any religion not based on a theology which can be apprehended by our reason.... Sophistication in religious matters is a dangerous thing; our reasoning powers are limited and reason can err and we cannot prove everything. A speculative basis is a very weak foundation for religion... [LE 86-7; cf. CJ 480-1] The problem, then is to discover the proper foundation which can be put in the place of speculation.

What is not so well known is that Kant saw his philosophical System not only as posing this problem, but as offering at least four distinct ways of solving it. So, even though Kant begins his theology on an essentially negative theological note, believing he has been able 'to discover the fallacy in any attempt [to prove God's existence theoretically], and so to nullify its claims' [CPR 667], nevertheless he devotes considerable effort to the task of showing how an honest recognition of the limitations of human reason leaves ample room for drawing affirmative theological conclusions in a theoretical discussion of God's existence and nature. In what follows I will examine these affirmations in turn, with a view towards assessing the common assumption that his theology defends an entirely negative position, suuch as deism, and thereby ascertaining his true attitude towards theology.

2. God as a Regulative Idea

Kant believes it is important for us to form some judgment on the question of God's existence despite the transcendental limitations imposed on human knowledge. He explains that there is a real need associated with reason itself [which] makes judging necessary even if ignorance with respect to the details required for judging limits us... If then it has been demonstrated that there can be here neither intuition of objects nor anything similar to such intuitions by which we could exhibit appropriate objects to our broadened concepts and thus make sure of their real possibility, nothing remains for us except first to test the concept with which we venture beyond all possible experience to see if it is free of contradictions, and then to bring at least the relation of this object to objects of experience under pure concepts of reason. By this we do not make the object sensuous. We only fit something supersensuous to thought in reason's empirical perspective, for without this precaution we could make no use of such a concept and would rave instead of think.2 Kant's criticism of the traditional proofs is actually designed to fulfil the first of these tasks, by demonstrating that belief in God cannot be logically contradictory, since God's existence, regarded as a constitutive part of the world, can never be proved or disproved, on the grounds that an intuition of God is, in principle, impossible. The second task is fulfilled in a lengthy Appendix to the Transcendental Dialectic [CPR 671-732], where Kant offers an alternative explanation of the epistemological status of the idea of God--an explanation which is often not treated very seriously by commentators.

Kant's first theological affirmation provides an explanation for a commonly experienced paradox, which Kant expresses in CPR 643 by asking: 'Why are we constrained to assume that some one among existing things is in itself necessary, and yet at the same time to shrink back from the existence of such a being as from an abyss?' Dialectical illusion results only if we try to subdue one of these natural tendencies. Those who try to prove God's existence theoretically are repressing the latter, while those who categorically deny God's existence are repressing the former. But if the truth which lies behind both tendencies is grasped, both errors can be avoided. The situation which gives rise to this paradox is that 'I can never complete the regress to the conditions of existence save by assuming a necessary being, and yet am never in a position to begin [such a regress] with such a being' [CPR 643-4].

The two sides of this paradox can be made compatible by recognizing the 'merely heuristic and regulative' character of the principles underlying each side:

The one [principle] prescribes that we are to philosophise about nature as if there were a necessary first ground for all that belongs to existence--solely, however, for the purpose of bringing systematic unity into our knowledge, by always pursuing such an idea, as an imagined ultimate ground. The other warns us not to regard any determination whatsoever of existing things as such an ultimate ground... [CPR 644-5e.a.] Whereas all theoretical arguments about the existence of God are bound to fail in their attempt to establish knowledge of God as an ideal object, these two principles suggest that the concept of God can have a valid use after all as long as it is regarded, less ambitiously, as an idea of reason. Since I have discussed the general character of this regulative employment of the ideas in some detail elsewhere [see FKK 452-5 and KE 190-6], I will proceed directly to a discussion of its implications for our theoretical understanding of the concept of God.

A theoretical discussion of God's existence and attributes, Kant argues, cannot be based 'upon the knowledge of such a being but upon its idea only' [CPR 729e.a.]. From the standpoint of theoretical reason our idea of God is postulated only problematically (since we cannot reach it through any of the concepts of the understanding) in order that we may view all connection of the things of the world of sense as if they had their ground in [it]... In thus proceeding, our sole purpose is to secure systematic unity... [709] In such usage God is 'an idea which reason is constrained to form as the regulative principle of its investigation of nature' [725]. As such, it is used as a principle for viewing empirical objects from a hypothetical, not an empirical, perspective. (The latter would be a constitutive use of the idea in reference to the world.)

Kant explains the proper use of an idea as follows: I think to myself merely the relation of a being, in itself completely unknown to me, to the greatest possible systematic unity of the universe, solely for the purpose of using [the unconditioned object] as a schema of the regulative principle of the greatest possible empirical employment of my reason. [707]

The purpose of Kant's whole treatment of the idea of God in CPR is to establish the right to use this theoretical concept from other, nontheoretical standpoints [see above, note 2]. His criticism of the traditional proofs does this by demonstrating that, although the concept cannot be instantiated in experience, it is at least not self-contradictory. The function of this concept as a regulative idea can therefore be put forward as a reasonable hypothesis (i.e., as plausible, though not provable), even from the standpoint of theoretical reason. Far from being an afterthought, this theory is at the core Kant's theological concern. As Despland points out in KHR 146: 'The unique strength of criticism is that "rational" is not restricted in meaning to cognitive. The Ideas of reason can be thought rationally without being objectified into possible objects of knowledge.'

'Hypotheses', Kant urges in CPR 805, are 'permissible only as weapons of war, and only for the purpose of defending a right, not in order to establish it.' They can be invaluable tools, when used 'in self-defence', in order to nullify 'the sophistical arguments by which our opponent professes to invalidate this assertion [of God's existence]' [804-5]. Yet they cannot be used dogmatically, since the sceptic can also produce opposing hypotheses. Since theoretical reason 'does not...favour either of the two parties', hypotheses can be used 'only in polemical fashion.' So a proper view of hypotheses limits dogmatists by refusing them knowledge, while yet limiting sceptics by upholding the right to believe. These warring parties, Kant explains, both 'lie in ourselves'; and the task of criticism is to remove 'the root of these disturbances' in order to 'establish a permanent peace' [805-6]. Once we recognize that hypotheses, 'although they are but leaden weapons', are required 'for our complete equipment' in fulfilling this purpose, we will see that there is 'nothing to fear in all this, but much to hope for; namely, that we may gain for ourselves a possession which can never again be contested.' By establishing peace in our system of theoretical knowledge, the regulative use of the idea of God directs our attention forward to the other Critical standpoints in anticipation of a more complete justification for belief in God.

This affirmation of the benefits of the regulative employment of our idea of God is frequently rejected prematurely by Kant's critics. One of the most common criticisms is that science (especially since Darwin) simply has no use for postulating 'the idea of God...as a heuristic device in the empirical study of nature' [KRT 145]. But this is based on a complete misunderstanding of the perspective from which Kant is speaking: he never intends the ideas to be used as regulative principles from an empirical perspective, such as that adopted by the natural scientist; for he insists that 'just because it is a mere idea, [the idea of God] is altogether incapable...of enlarging our knowledge in regard to what exists' [CPR 630-1]. Hence it cannot serve as the constitutive 'ground of the systematic order of the world' [709; cf. 724-5]. This function is fulfilled on the material side by the thing in itself and on the formal side by reason's architectonic forms [see e.g., 723-4].

Instead, the ideas are to function regulatively only in the context of reason's special hypothetical perspective. To think otherwise is to ignore the fact that metaphysics 'does not need the ideas for the purposes of natural science, but in order to pass beyond nature' [395n]. In other words, these regulative principles concern how 'to philosophise about nature' [CPR 644, q.a.], not how to investigate nature scientifically. Accordingly, Kant harshly condemns the latter approach:

To have recourse to God...in explaining the [physical] arrangements of nature and their changes is...a complete confession that one has come to the end of his philosophy, since he is compelled to assume something of which in itself he otherwise has no concept in order to conceive the possibility of something he sees before his very eyes.3

Just as the regulative use of an idea assumes it not to have 'creative power', but to 'have practical power..., and form the basis of the possible perfection of certain actions' [CPR 597], so also such regulative usage implies nothing about how we are to go about gathering empirical knowledge, but only about how we are to structure our beliefs about the source of the ultimate unity of that knowledge: much as a (e.g., religious) vision of the 'not yet' can act as a powerful force pulling us forward towards the realization of a hope, the idea motivates us to search for systematic unity in our philosophical explanations.4

Another frequent complaint against Kant's plea that we be satisfied with regarding God as a regulative idea is made by those theologians who are (as Kant says with respect to the moral philosophers of his day) 'dedicated to the omnipotence of theoretical reason' [Kt6:377]. He continues:

...the discomfort they feel at not being able to explain what lies entirely beyond the sphere of physiological explanation [e.g., the idea of God] provokes them to a general call to arms, as it were, to withstand that Idea, no matter how exalting this very prerogative of man--his capacity for such an Idea--may be.

That is to say, they reject the notion of God as an idea not because it is incoherent, but because it does not provide what they are looking for, viz. certain knowledge of God's existence and nature. Because Kant says, for example, that 'this Idea proceeds entirely from our own reason and we ourselves make it' [442], they disregard his many other claims to believe in a real, living God, as in traditional theism.5 Such a premature rejection of his position fails to recognize that, as in virtually every other aspect of his System, Kant often gives different answers to the same question when different perspectives are assumed. Hence, viewing 'God' from the theoretical standpoint as a man-made idea does not prevent us from adopting some other standpoint in order to affirm that a real, transcendent God actually exists.

3. Natural Theology

Kant's theory concerning the regulative idea of God is actually the least important of his three ways of affirming the rationality of theology; for 'the conception of a Deity...can never be evolved merely according to principles of reason's theoretical standpoint' [Kt7:400]. So in addition to such transcendental theology, he develops his own type of natural theology in the second and third Critiques. A thorough examination of his moral and physicotheological arguments for God's existence will help to reveal the systematic character of his general concept of God and to demonstrate the richness and depth of this 'guiding-thread' [cf. CJ 389] of his System.

Kant affirms the physicotheological proof in the third Critique, yet this does not nullify the limitations he places on it in CPR; for the standpoint from which it is discussed in CJ is judicial rather than theoretical. The same theoretical concept (God) is still under consideration; from the outset, however, Kant is now aiming to establish not theoretical knowledge, but only an empirical justification of a practical belief. Even in CPR Kant explicitly allows for such a usage: he argues that we are 'undoubtedly' permitted, if not required, 'to assume a wise and omnipotent Author of the world', as long as we realize that such an assumption does not in any way 'extend our knowledge beyond the field of experience' [725-6]. Elsewhere, he develops this idea a bit further:

Physicotheology...can enlighten and give intuitive appeal to our concepts of God. But it cannot have any determinate concept of God. For only reason can represent completeness and totality. In physicotheology I see God's power. But can I say determinately, this is omnipotence or the highest degree of power? [LPT 32-3]

The implicit answer, of course, is 'no'. For although it has a constructive role to play, physicotheology on its own is 'unable to...serve as the foundation of a theology which is itself in turn to form the basis of religion' [CPR 656]. Instead, Kant intends it to point the way outward from experience to moral activity, where theology has a more secure foundation.

Kant argues in CJ 389 that empirical reflection on 'the clearly manifest nexus of things according to final causes' requires us to conceive of 'a world-cause acting according to ends, that is, an intelligent cause-- however rash and undemonstrable a principle this might be for the determinant judgment.' He bases this conclusion on the specific phenomenon of finality in our experience of the world:

...the nature of our faculty of reason is such that without an Author and Governor of the world, who is also a moral Lawgiver [see below], we are wholly unable to render intelligible to ourselves the possibility of a finality, related to the moral law, and its Object, such as exists in this final end. [455]

In particular Kant emphasizes that 'the end for which nature itself exists' is man, and that 'it is upon the definite idea of this end that the definite conception of such a supreme intelligent World-Cause, and, consequently, the possibility of a theology, depend' [437]. Viewed from the judicial standpoint of CJ rather than the theoretical standpoint of CPR, this argument is, as Wood points out in KMR 174, directed not so much to the theoretical philosopher as to the ordinary man: 'the ordinary man "sees" nature as the work of God, and discerns in it--what no amount of empirical evidence could have demonstrated--the signs of a divine and morally purposive creation' [176]. Yet even from the standpoint of CJ physicotheology on its own is quite limited, for experience 'can never lift us above nature to the end of its real existence or thus raise us to a definite conception of such a higher Intelligence' [CJ 438; see also LPT 38]. Thus 'physical teleology urges us to go in quest of a theology. But it can never produce one' [CJ 440]; for 'physico-theology...is of no use to theology except as a preparation or propaedeutic and is only sufficient for this purpose when supplemented by a further principle on which it can rely' [442].

Rather than depending on the speculative proofs of transcendental theology, however, Kant's physicotheology depends on the proof provided by moral theology from the practical standpoint: 'underlying our procedure [in physicotheology] is an idea of a Supreme Being, which rests on an entirely different standpoint [than the judicial], namely the practical' [CJ 438]. Kant sums up the preparatory function of physicotheology in DV 481, where, in his example of 'a moral catechism' [479], the final comment of the pupil is:

For we see in the works of nature, which we can judge, a wisdom so widespread and profound that we can explain it to ourselves only by the ineffably great art of a creator of the world. And from this we have cause, when we turn to the moral order...to expect there a rule no less wise.

4. The Moral Argument as the Basis for Kant's Theism

Kant's moral argument for the existence of God is the only aspect of his solution to the problem of transcendental theology which has been duly recognized by his commentators. In its simplest form, his argument is fairly straightforward. After arguing that the highest good consists of the distribution of happiness to each person in proportion to his or her virtue, Kant points out that, given the nature of human virtue (viz., that it often requires us to deny our own happiness in order to obey the voice of duty), man on his own is unlikely to bring into being this ideal end of morality. Yet if the end or purpose of morality proves to be unattainable, moral action itself will be irrational. Hence, anyone who wishes to regard moral action as rational is constrained to postulate whatever is necessary to conceive of the possibility of the highest good. As is well known, Kant argues that the immortality of the soul and the existence of God are the two postulates which alone can save morality from the abyss of meaninglessness.

Although Kant's basic argument is familiar enough, its intended force is often misunderstood, especially by those who fail to take into consideration the different perspectives in Kant's System. In the first place, Kant's moral argument has little, if anything, to do with his theory of religion (a point often misunderstood by those who write on the latter subject). Instead, the postulate of God is intended to perform its function exclusively within the final stage of Kant's practical (moral) system, where it suggests that rational moral agents are, in fact, acting as if God exists whenever they act morally, whether or not they claim to believe in God. In other words, God's existence, though not theoretically provable, is nevertheless a necessary assumption for any moral agent who wishes to conceive of the highest good as being realizable (and therefore, of moral action as being ultimately rational).

What then are the specific implications of Kant's moral argument for the theologian's attempt to prove the existence of God? Kant's argument, as summarized in CJ 446, is that every moral agent needs a moral Intelligence; because he exists for an end, and this end demands a Being that has formed both him and the world [i.e., both freedom and nature] with that end in view.... Hence...there is in our moral habits of thought a foundation for...form[ing] a representation depicting a pure moral need for the real existence of a Being, whereby our morality gains in strength or even obtains --at least on the side of our representation-- an extension of area, that is to say, is given a new object for its exercise. The resulting concept of 'a moral Legislator' has no theoretical value; yet, Kant continues, the source of this disposition is unmistakable. It is the original bent of our nature, as a subjective principle, that will not let us be satisfied, in our review of the world, with the finality which it derives through natural causes, but leads us to introduce into it an underlying supreme Cause governing nature according to moral laws. --In addition...we feel ourselves urged by the moral law to strive after a universal highest end, while yet we feel ourselves, and all nature too, incapable of its attainment.... Thus we have a pure moral ground derived from practical reason for admitting this Cause (since we may do so without contradiction), if for no better reason, in order that we may not run the risk of regarding such striving as quite idle in its effects, and of allowing it to flag in consequence.


In CJ 482 he deliberates with equal clarity:

Kant's approach to theology is typically characterized as implying, in the words of Cupitt, 'a non-cognitive philosophy of religion which leaves the believer to be sustained in a harsh world by nothing but pure moral faith'.12 But in fact, Kant's theological and religious views are not so 'bleak and austere' as is often assumed. On the contrary, such an assumption, like most misinterpretations of Kant, rests on a failure to understand how the principle of perspective operates in his System. It is true that his practical postulates as such are not much help in facing the harsh realities of human existence, but they are not primarily intended to fulfil such an empirical role; for Kant offers us a good deal more in the way of equipping us with tools to cope with reality. The most significant of these, which concern Kant's view of God as participating in human morality and as relating on a personal basis with his creatures, are beyond the scope of our present inquiry.13

Nevertheless, the foregoing account of Kant's solution to the problem of transcendental theology has, I hope, made abundantly clear that Kant's theology is not that of a 'deist', as is so often assumed, but is the rational framework for a 'theism' which has rarely been adequately appreciated by his interpreters. This failure is due in part to the fact that theologians and philosophers of religion often group Kant on the side of those who argue 'that God is utterly unknowable', and that therefore 'theology is a useless effort'.14 The latter conclusion does seem to follow naturally from the deistic assumption that God is utterly unknowable, an assumption Kant apparently adopts in his denial of our ability to intuit God. But this interpretation reflects a rather narrow acquaintance with Kant's writings. For, even in the Preface to Religion within the Bounds of Bare Reason Kant says with no apparent irony that the philosopher and the theologian should see themselves not as rivals, out to destroy each other, but as co-workers, mutual friends and companions.

Kant defines theology as 'the system of our knowledge of the highest being'; it 'does not refer to the sum total of all possible knowledge of God, but only to what human reason meets with in God' [LPT 23; cf. CPR 659]. The 'knowledge of everything in God', which Kant calls 'theologia archetypa', is unattainable for man, while 'that part of God which lies in human nature', the knowledge of which he calls 'theologia ectypa', is attainable.15 Within the latter he distinguishes between deism and theism: 'Those who accept only a transcendental theology [i.e., knowledge of God based on the theoretical standpoint] are called deists; those who also admit a natural theology [i.e., knowledge of God based on the practical or judicial standpoints] are called theists' [CPR 659; see also 660-1; LPT 28-9]. Kant therefore believes the distinction between the theist and the deist concerns not only one's theoretical standpoint, but also one's particular (moral and empirical) experiences of the God whom such theories are intended to describe. Deists, then, are those who, after reflecting logically and/or transcendentally on the concept of God, come up with a positive answer to the question of His existence. Theists are open to these two perspectives, but regard them as only secondary to the more basic use of empirical and/or hypothetical perspectives in developing a theoretical affirmation of God. Only from the latter two perspectives can God be regarded not just as 'an original being or supreme cause' (as in deism), but also as 'a supreme being who through understanding and freedom is the Author of all things'. Thus, Kant asserts 'that the deist believes in a God, the theist in a living God' [CPR 660-1].

Kant demonstrates in numerous ways that he is, given the above definitions, a thoroughly theistic philosopher. Not the least of the reasons for regarding Kant as a theist is that, as we have seen, he replaces the deist's reliance on the theoretical standpoint with a theology firmly rooted in the practical standpoint. Thus he confesses in CPR 856: 'I inevitably believe in the existence of God..., and I am certain that nothing can shake this belief, since my moral principles would thereby be themselves overthrown, and I cannot disdain them without becoming abhorrent in my own eyes.' Ironically, the very criticisms of the traditional theoretical arguments for God's existence with which Kant begins his critical theology, though they were designed to pave the way for a practical theism, are (as we have noted) often the basis upon which Kant is misinterpreted as being himself a deist!16

Kant is indeed acutely aware of the problems posed to theological knowledge by human ignorance: 'Both in theology and in religion, but particularly in theology, we are handicapped by ignorance' [LE 85]. Sometimes even when we think we have knowledge, he tells us, we actually have 'no concept at all' of God [LPT 24]. But as Wood points out, this does not make him a deist [KMR 155,164], for he means by this that 'our concept of God is an idea of reason' [KRT 79], rather than a concept which rises out of abstraction from appearances. For Kant holds that 'we cannot intuit God, but can only believe in him' [LE 99]; yet 'in order to believe in God it is not necessary to know for certain that God exists' [81]. On the contrary, as Wood again conveys Kant's view, 'the "minimum" theology it is necessary to have is a belief that God is at least possible' [KMR 31]. Kant believes the ideas of 'God, freedom, and the immortality of the soul are the problems to whose solution, as their ultimate goal, all the laborious preparations of metaphysics are directed' [CJ 473]; and his System of Perspectives is intended to solve these problems once and for all by developing a theistic philosophy which rejects the false foundations offered by theoretical reason. Hence, in a choice between atheism, deism, anthropomorphism and theism, Kant would undoubtedly favour theism.17

Because Kant's theology guards against what might be called 'gnostic' errors (such as anthropomorphism), into which dogmatic theologians and philosophers of religion repeatedly fall, he is branded an agnostic. And because his theology likewise takes seriously the objections advanced by the atheist, he is branded a deist. Yet a perspectival interpretation reveals that his response to the problem of transcendental theology was that of neither a deist nor an agnostic, but a theist in a quite profound sense of the word. Ironically, those who label Kant as a deist or an agnostic are often those who would call themselves theists because of their affirmative response to the traditional arguments of speculative theology. Yet for Kant this is not good enough: no one can claim to be a theist on the strength merely of logical ingenuity, for theism depends on a belief in a God who manifests Himself as 'a living God' in our immediate experience, whereas the ontological and cosmological arguments portray God 'wholly separate from any experience' [LPT 30]. If anyone is a deist, then, it is not Kant, who believes in a God who purposely hides his true nature from us, but gives us enough evidence to make a reasoned step of faith, after which we are able to understand God's nature with sufficient clarity in terms of our finite human perspectives; rather it is those who put all their trust in the powers of theoretical reason and toil endlessly and in vain to attain knowledge which is not to be had by us men. The religious implications of Kant's theism are not always entirely consistent with orthodox Christianity; yet they are not as inconsistent as is often assumed. For, although it is couched in the difficult terminology of a highly complex philosophical System, Kant's theism is not significantly different (in its general intent, at least) from the theism expressed by the writer of 2 Cor. 4:7 when he proclaims that 'the transcendent power [he huperbol ts dunámeôs] belongs to God and not to us'.

S.R. Palmquist August 1991 Hong Kong



CKC: Lewis White Beck, A Commentary on Kant's Critique of Practical Reason (London: The University of Chicago Press, 1960).

KMP: Peter Byrne, 'Kant's Moral Proof of the Existence of God', in Scottish Journal of Theology 32 (1979), pp.333-43.

CPK: Edward Caird, The Critical Philosophy of Immanuel Kant,2 2 vols. (Glasgow: James Maclehouse and Sons, 1909).

EPR: James Collins, The Emergence of Philosophy of Religion (London & New Haven: Yale University Press, 1967).

KNT: Don Cupitt, 'Kant and the Negative Theology', in B. Hebblethwaite, and S. Sutherland (eds.), The Philosophical Frontiers of Christian Theology (London: Cambridge University Press, 1982), p.55-67.

KHR: Michel Despland, Kant on History and Religion (London and Montreal: McGill-Queen's University Press, 1973).

IK: Lucien Goldmann, Immanuel Kant, tr. Robert Black (London: NLB, 1971).

HCRS: Theodore M. Greene, 'The Historical Context and Religious Significance of Kant's Religion', in RLRA ix-lxxviii.

RPG: Heinrich Heine, Religion and Philosophy in Germany, tr. J. Snodgrass (Boston: Beacon Press, 1959).

GWB: Grace M. Jantzen, God's World God's Body (London: Darton, Longmann and Todd, 1984).

CPR: Immanuel Kant, Critique of Pure Reason (1781,1787), tr. N. Kemp Smith (London: Macmillan & Co. Ltd., 1929).

PFM: -----, Prolegomena to Any Future Metaphysics (1783), tr. L.W. Beck (New York: The Bobbs-Merrill Company, Inc., 1950).

WOT: -----, What is Orientation in Thinking? (1786), tr. L.W. Beck in CPrR 293-305.

CPrR:-----, Critique of Practical Reason (1788), tr. L.W. Beck (Indianapolis, Indiana: The Bobbs-Merrill Company, Inc., 1956).

CJ: -----, Critique of Judgement (1790), tr. J.C. Meredith (Oxford: The Clarendon Press, 1952).

DV: -----, The Doctrine of Virtue, tr. M.J. Gregor (Philadelphia: University of Pennsylvania Press, 1964).

RLRA:-----, Religion within the Limits of Reason Alone (1793), tr. T.M. Greene and H.H. Hudson (New York: Harper & Row, Publishers, 1960).

GTLA:-----, Of a Gentle Ton Lately Assumed in Philosophy (1796), in Essays and Treatises on Moral, Political, Religious and Various Philosophical Subjects, anonymous translator J. Richardson (London: William Richardson, 1798-9), vol.2, pp.159-87.

PM: -----, Progress in Metaphysics (1791; ed. F.T. Rink, 1804), tr. Ted B. Humphrey (New York: Abaris Books, 1983).

LE: -----, Lectures on Ethics (ed. P. Menzer, 1924), tr. L. Infield (London: Methuen & Co. Ltd., 1979), English pagination.

LPT: -----, Lectures on Philosophical Theology (ed. K. Beyer, 1937), tr. A.W. Wood and G.M. Clark (London: Cornell University Press, 1978), English pagination.

KPC: -----, Kant: Philosophical Correspondence 1759-99, tr. and ed. Arnulf Zweig (Chicago: The University of Chicago Press, 1967). AKF: J.C. Luik, 'The Ambiguity of Kantian Faith', Scottish Journal of Theology 36 (1983), p.339-46.

DMK: Donald MacKinnon, 'Kant's Philosophy of Religion', Philosophy L (1975), pp.131-44.

PC: Greville Norburn, 'Kant's Philosophy of Religion: A Preface to Christology?' in Scottish Journal of Theology 26 (1973), pp.431-48.

NPG Robert A. Oakes, 'Noumena, Phenomena, and God' in International Journal For Philosophy of Religion 4 (1973), pp.30-8.

FKK: Stephen Palmquist, 'Faith as Kant's Key to the Justification of Transcendental Reflection' in The Heythrop Journal XXV (October 1984), pp.442-55.

RUKT:-----, 'The Radical Unknowability of Kant's "Thing in Itself", Cogito III.2 (June 1985), pp.101-115.

SPO: -----, 'Six Perspectives on the Object in Kant's Theory of Knowledge' in Dialectica 40.2 (l986), pp.121-51.

KE: -----, 'Knowledge and Experience -- An Examination of the Four Reflective "Perspectives" in Kant's Critical Philosophy' in Kant-Studien 78 (1987), pp.170-200.

KCM: -----, 'Kant's Critique of Mysticism: (2) Critical Mysticism', Philosophy & Theology 4.1 (Fall 1989), pp.67-94.

DKR: -----, 'Does Kant Reduce Religion to Morality?', Kant-Studien, forthcoming.

RK:  Gabriele Rabel (ed.), Kant (Oxford: The Clarendon Press, 1963).

DKT: Herman-J. de Vleeschauwer, The Development of Kantian Thought, tr. A.R.C. Duncan (London: Thomas Nelson and Sons Ltd., 1962).

KPR: Clement C.J. Webb, Kant's Philosophy of Religion (Oxford: The Clarendon Press, 1926).

KNF: Don Wiebe, 'The Ambiguous Revolution: Kant on the Nature of Faith' in Scottish Journal of Theology 33 (1980), p.515-32.

KMR: Allen W. Wood, Kant's Moral Religion (London: Cornell University Press, 1970).

KRT: -----, Kant's Rational Theology (London: Cornell University Press, 1978).



1. See the Bibliography for the key to abbreviations. All references to CPR cite the page numbers of the second (1787) German edition. References to Kant's other writings (except where otherwise noted in the Bibliography) will cite the pagination of the Akademie edition of Kant's works. For translations which do not give the German pagination in the margins, the Akademie page number(s) will be followed by the English pagination in brackets.

2. WOT 136-7. Kant continues by explaining that 'the right of a need of reason enters as the right of a subjective ground to presuppose or assume something which it may not pretend to know on objective grounds' [137]. From the former, theoretical standpoint, this 'need of reason' to 'assume the existence of God' is 'conditional': the assumption only 'needs' to be made 'when we wish to judge concerning the first cause of all contingent things, particularly in the organization of ends actually present in the world' [139]. But from the latter, practical standpoint, 'the need is unconditional; here we are compelled to presuppose the existence of God not just if we wish to judge but because we must judge' [139].

3. CPrR 138; see also LPT 25-6. This seems at first to apply equally to Kant's own assumption of the thing in itself, which he does believe to be philosophically sound. However, he is speaking here from an empirical perspective, in the context of which the thing in itself, as positive noumenon, is indeed superfluous [see section 3 of SPO]; Kant's use of the thing in itself does not fall under this criticism because it assumes the transcendental perspective. Hence, when we read Kant warning us that 'a presumptuous readiness to appeal to supernatural explanations is a pillow for a lazy understanding' [On the Form and Principles of the Sensible and Intelligible World (Kant's Inaugural Dissertation), p.418, as translated in KPR 45], we must be careful not to interpret this too harshly, as does Webb when he says this claim means that 'the assumption of the supernatural is excluded on "critical" principles' [KPR 45]. For as we have seen, Kant actually encourages such an assumption in the appropriate circumstances, as long as it is put forward without a presumptuous attitude (i.e., as a theoretical hypothesis rather than a dogmatic knowledge-claim). If Kant's advice to us in such passages is that supernatural explanations are always inappropriate, then why does he himself make use of the God hypothesis throughout his writings? Rather, his message is that we must be careful to use them wisely--i.e., in such a way that they do not prevent us from relentlessly seeking natural explanations wherever possible.

4. Kant describes a 'hypothesis' in CPrR 126 as 'a ground of explanation'. As such, a proper understanding of his theory of the regulative use of the idea of God from the hypothetical perspective reveals it to be remarkably similar to modern attempts to defend God's existence as the best 'explanatory hypothesis' [see e.g. PC 441]. There are differences, of course, such as that the modern versions, while they perhaps benefit from their freedom from Kant's rather difficult and old-fashioned terminology, often suffer unnecessarily by mixing different perspectives uncritically (e.g., by assuming that rigorous logical argumentation is the primary, if not the only, tool available to defend or refute such hypotheses). But the two approaches are alike to the extent that they both attempt a theoretical defence of God's existence not on the basis that the God-hypothesis enables us to provide a better scientific explanation of the available data, but rather on the basis that the available data point beyond themselves to something which can best be explained philosophically in terms of the God-hypothesis. Thus in both cases the theoretical argument, when properly constructed, assumes a hypothetical, rather than an empirical, perspective.

5. The issue of Kant's theism will be discussed in more detail at the end of this paper.

6. KMR 161. Unfortunately, many interpreters make the very mistake in interpreting the underlying connotations of Kant's moral argument that Wood is warning against here. Webb, for example, claims that Kant's moral argument 'certainly is in no way calculated to express the religious man's conviction of the reality of the object of his worship' [KPR 66]. If 'the religious man's conviction' here refers to traditional, uncritical ways of believing in God, then of course Webb is correct, since the argument is directed to 'the moral man'. But the words 'in no way' are misleading, since (as I argue in DKR) Kant does intend his argument not only to be compatible with a religious standpoint, but also to provide a rational foundation for the fuller conception of the God of religion, as expounded in his own book, Religion within the Bounds of Bare Reason [i.e., RLRA].

7. See CPK 470. Of course, it almost goes without saying that Kant would totally reject the implications of the all-too-frequently repeated caricatures which cast doubt on the sincerity and/or validity of his moral proof. After mentioning the common complaint 'that while in his first Critique [Kant] has thrown God out the front door, in the Critique of Practical Reason he let Him in again by the back door', Rabel insists: 'There is not a shred of truth in this accusation' [RK vii]. Another myth goes back at least as far as Heinrich Heine, who claims Kant's revival of God in CPrR (after having put Him to death in CPR) is proposed by Kant 'half ironically', only because he recognized that 'Old Lampe must have a God' [RPG 119]. Heine's conjectures reach their height when he suggests that Kant may have developed his moral proof 'not merely for the sake of old Lampe, but through fear of the police' [276-7]!

We must admit, with Donald MacKinnon [DMK 133], that throughout Kant's treatment of God and religion, he often 'tries to do justice to what at a first reading he seems to dismiss out of hand.' But as long as we keep in mind Kant's reliance on the principle of perspective, the sincerity and reasonableness of his attempts to do this should be clear enough. Thus, rather than taking Heine's caricature too seriously, we can suggest a more appropriate version of Heine's story: perhaps Kant invented the moral argument in order to protect his faithful servant (and all others who humbly recognize, with Kant, the universal limits of 'common human understanding' [see e.g., CPR xxxii]) from the misuse he knew many philosophers would make of his negative criticisms of theoretical arguments for God in CPR. In other words, the moral proof explains not to Lampe (who has no need of a formal proof), but to Kant's fellow philosophers-- some of whom may well have joined Kant for lunch, and offered snide remarks attacking the servant's simple faith--why Lampe and all other human beings have nothing to fear from the limitations of theoretical reason. 8. As Webb puts it in KPR 68, 'Kant...definitely denies that the knowledge of God, the Object of religion, falls primarily or properly within the spheres of Physics [cf. the judicial system] or Metaphysics [cf. the theoretical system]. It is only...to be reached by starting...from the consciousness of duty or moral obligation [cf. the practical system].' Along these lines, Kant distinguishes between the moral argument as 'an argument kat' 'ánthropon valid for all men as rational [i.e., moral] beings in general', and 'the theoretical-dogmatic argument kat' alytheian' [PM 306].

9. Peter Byrne argues against Kant's moral proof in KMP 337:  'If 
knowledge of God is impossible then one cannot have grounds for 

believing or thinking that God exists.' He reaches this conclusion, however, only by presupposing an epistemology quite foreign to Kant, whereby knowledge is identified with justified belief [336; cf. section 3 of RUKT]. For Kant, unknowability by no means implies unthinkability. And he distinguishes between knowledge and belief by explaining that the evidence for a judgment one believes is true must be 'subjectively sufficient', but 'objectively insufficient', whereas the evidence for a judgment one knows is true must be 'sufficient both subjectively and objectively' [CPR 850]. For Kant, the relevant 'subjective' grounds are, of course, moral.

Oakes argues against the common assumption that anyone who believes knowledge of God is possible must reject Kant's doctrine of the unknowability of noumena [NPG 31]. He argues that Kant was wrong to construe 'all religious epistemology as necessarily a quest for noumenal knowing' [32], because our knowledge of God is, in fact, phenomenal [33]: 'any sensible experience of God...must be construed as providing knowledge which is partial or perspectival, i.e., knowledge solely from the vantage point of a finite knower.' Kant would, of course, agree that all knowledge is perspectival, but would argue that our 'sensible experience' is never a direct experience of God, in the way that our empirical knowledge is a direct experience (i.e., intuition and conception) of empirical objects. Rather, the religious person regards some experiences as coming from God by means of the God-hypothesis, which can never yield actual knowledge of a phenomenon called 'God'. Nevertheless, Kant's treatment of the experience of God is not far removed from that of Oakes, except that Kant never regards such experiences as capable of producing knowledge [see KCM 67-94].

10. Kant offers the theologian various tools to cope with the realities of human ignorance, in the form of analogical models for God's nature which represent a balanced and realistic view of some basic theological issues. These models, though rarely appreciated by his commentators, constitute an important aspect of Kant's systematic understanding of our theoretical conception of God's nature, though there will be no opportunity to discuss them in this paper.

11. 'Moral teleology' is the title Kant gives his moral proof in CJ to show its structural parallel to teleology proper (i.e., physical teleology). Each is teleological insofar as it concerns the purpose or final end which must be posited in order to explain a certain type of experience (viz. of either a moral or a physical end). Beck's criticism of the moral proof on this account [CKC 275] is therefore correct, but irrelevant, since this type of teleology is clearly distinguishable from that discussed elsewhere in CJ.

12. KNT 64. Likewise, Goldmann [IK 201] says Kant believes in: 'A transcendent superhuman God who has only practical and moral reality but who lacks independent moral existence...--a more unreal God could scarcely be imagined.' Such a comment is grossly unfair, however, since Kant never dogmatically proclaims that God has no such independent existence, but only warns that if He does, we could never grasp it as an item of our empirical knowledge.

13. I have discussed this issue in detail in KCM.

14. GWB 1; cf. 42-3. For instance, Goldmann says 'Kant rejected all positive religion' [IK 194]. Or, as J.C. Luik puts it in AKF 345, 'there is quite literally no Kantian theology, no religious knowledge for Kant.' Luik makes this assertion in the process of rejecting Wiebe's claim that for Kant '"knowledge" of God can be had', though only if it is 'inferred' from 'moral data' [KNF 531]. Although Luik's position would be correct as a description of Kant's theoretical standpoint, it ignores the fact that for Kant the practical and judicial standpoints are just as important; for they can each produce (at least in a symbolic sense) a kind of knowledge of their own. Thus Kant clearly states that 'all our knowledge of God is merely symbolic', whereas 'Deism...furnishes no knowledge [of God] whatsoever' [CJ 353].

15. LPT 23. In his 1796 essay, GTLA 391(164-5), Kant makes a similar distinction, between Plato's view of 'archetypes (ideas)' as intuitions which originate in 'the Divine understanding' but can be 'named directly' by man, and his own belief that 'our intuition of these divine ideas...is distributed to us but indirectly, as the copies (ectypa)...'

16. Zweig infers from a 1759 letter to Kant that Kant equates 'deism' with 'sanity' [KPC 35n]. Yet Hamann's letter actually portrays Kant as an arbitrator between Hamann the Christian and Berens the deist. Zweig's assumption that Kant was on Berens' side is not justified from the content of the letter, which seems instead to portray Kant in his usual, 'critical' position as a middle man.

Although Heine caricaturizes CPR as 'the sword that slew deism in Germany' [RPG 268], he believes CPrR was intended to revive it. Likewise, Greene regards RLRA as 'a deistic classic' [HCRS lxxvii; see also p.lxvi]. And Webb implies that Kant was a deist for most of his life when he says that in his Opus Postumum, Kant 'was prepared to repudiate ...the deism which had been so predominant in his youth--the deism which taught a merely transcendent God' [KPR 200-1]. Ironically, Vleeschauwer sees in this same work 'a public confession of deism' [DKT 177]!

There is, however, a growing rank of scholars who reject such interpretations. Despland, for example, argues that in his philosophy of religion 'Kant...moved beyond the classical deist position' [KHR 198; see also pp.199-201,228,262; and PC 431; KNF 515]. For as Collins puts it: 'Kant regards religious deism and the varieties of nature-based theism as incomplete, preliminary forms of religious life.' [EPR 117]. Indeed, as I have demonstrated elsewhere [see DKR and KCM], Kant moves beyond these to form a moral theism--one which is thoroughly compatible with his Critical principles. Indeed, Kant's theistic outlook is acknowledged so consistently throughout his writings that I would call into question even the assumption that Kant ever seriously defended a deistic position as such.

Kant's rejection of deism is, admittedly, usually expressed in very cautious terms--and understandably so, given the dominance of deism in the philosophical climate of his day. Nevertheless, some texts reveal his dissatisfaction with deism so clearly that all debate on this question ought to be a thing of the past. In a 1789 letter to Jacobi, for instance, Kant approves of his friend's refutation of 'the syncretism of Spinozism and the deism of Herder's God' [KPC 158]. And in PFM 356-7, Kant says that if theism and anthropomorphism are both abandoned, then 'nothing [would] remain but deism, of which nothing can come, which is of no value and which cannot serve as any foundation to religion or morals.'

17. It should be noted, however, that Kant reveals his dissatisfaction with the theoretical implications philosophers often impute to theism, by warning in CJ 395 that even theism 'is absolutely incapable of authorizing us to make any objective assertion.'