Almost everywhere in the architectonic of the system of Kant’s transcendental idealism one finds the transcendent. Kant’s critical philosophy establishes key relationships with the unrepresentable: in epistemology (the noumena), in ontology (the thing in itself), in ethics (noumenal freedom) and in aesthetics (the mathematically conceived sublime). Moreover, this inaccessible space of the unrepresentable allows for the theoretical promotion of a legitimate primacy of the ethical – even in the least obvious instance, aesthetics, the experience of the sublime is said to lead to, or to be conducive to, moral behaviour – and the promise, connected to it, of the preservation of religion. In the Critique of Pure Reason, Kant sets up the conditions for what seemed to be this, his great spiritual wish: not a tolerated – because disguised – atheism but rather a truly rational faith in God consistent with the tenets of his transcendental system.
Examining Kant’s philosophy of religion here will both illuminate its own intrinsic merits, tensions and problems and also bring to light the manner in which it then allowed Schopenhauer to argue for his specifically post-Kantian variety of atheism, an ontological atheism which in turn greatly influenced the philosophy of Nietzsche. The present chapter will be structured as follows: first, a little scenesetting by means of a brief and therefore selective look at the general argument of Kant’s major work, the Critique of Pure Reason, followed by a study of Kant’s own construal of God as existing wholly outside of space and time and his subsequent falsificationist approach to the study of religious scripture. The final section of this chapter turns to examine Kant’s elaborate ‘moral argument’ for the existence of God. This proof is without any doubt the locus of Kant’s attempt to construct a positive philosophy of religion within the constraints of his ‘critical’ system but it will not prove to be ultimately convincing and two possible ways in which it might be seen to fail will be developed. Given the ultimate inadequacy of Kant’s positive attempt to rationally justify the positing of God within the critical system, it will be suggested that it would be sensible for us to suspect Kant’s deism and accept Kant’s own occasional admissions that his philosophy allowed rather than compelled a theistic commitment.
Kant’s project can be roughly characterised as a close attention to our absolutely (as it is supposed) invariant perceptual and conceptual faculties and their implications for the study of metaphysics. Kant thus might be said to position himself much in accordance with the main thrust of specifically modern philosophy from René Descartes onwards by beginning his inquiry with and from the individual epistemological subject (and its knowledge). Yet one problem that notoriously arises even at this early point in Kant’s attempt to establish philosophy on such a subjective footing is that, although Kant does not wholly overlook the philosophic problem of other minds, he nonetheless – unlike his predecessors, Descartes and Berkeley – has disturbingly little to say on this particular aspect of alterity (a charge familiar from Jean-Paul Sartre’s L’É tre et le Néant). In truth, to establish our knowledge of other subjects, pretty much all Kant has to say is that
If I wish to represent to myself a thinking being, I must put myself in his place, and thus substitute, as it were, my own subject for the object I am seeking to consider (which does not occur in any other kind of investigation) [CPR A 353–4]
which, far from solving the sceptical problem of other minds, serves only to highlight it. It is worth emphasising this problem at the outset because far from being a non-issue it might in fact be thought to impinge upon a consequential theological problematic within the Kantian philosophy.
What is meant by this somewhat grandiose claim is simply that to such typically modern philosophical endeavours as Kant’s, which are anchored so centrally to epistemological subjectivity, there belongs, at least in principle, a suspicion (which haunts modern phenomenology and is arguably confirmed in Hegel’s Phenomenology of Spirit) that the subject himself, the beginning of all philosophical inquiry, might in fact be a rather special kind of spiritual being. After all, although an essentially privileged one, God is nevertheless presumably also a subject (even Kierkegaard insists on this, despite his repeated claims concerning an absolute difference obtaining between God and man). In other words, broadly Cartesian philosophies, such as Kant’s, that start off from the subject, if they do not successfully defuse scepticisms concerning an intersubjective world, might be left with the extravagant idea that the subject posits everything, including himself. And, in this connection, it is noteworthy that the thought that the self apparently indicated by self-awareness or the cogito might itself be God is, in fact, briefly entertained by Descartes in the Meditations on First Philosophy.
The first of Descartes’ six Meditations famously inaugurated the modern concern with epistemology by calling into doubt all our beliefs. Specifically, it relied on an argument from illusion, the suspicion that we might be continually dreaming – later laid to rest because of dreaming’s unconnectibility with past experience – and the even more extreme idea that an evil demon might be deceiving us to thereby render all our beliefs collectively suspicious. In the second Meditation, Descartes then discovers a single and now well-known truth – or perhaps he rediscovers an Augustinian truth – on the basis of which he will rebuild his knowledge from robust foundations: namely, that his own existence is indubitable: I think therefore I am: cogito ergo sum. In the third Meditation, Descartes then investigates other of his ideas and finds that one in particular, the idea of God, could not have been generated by himself, since it is the idea of infinity and he is but a finite creature. This argument, coupled with the rather more basic a priori assumption that ideas must have adequate causes, leads Descartes to suppose that the idea of God has to be innate and implanted in us by the creator Himself,1 thereby saving the Cartesian project from an apotheosis of the self. To the modern reader, however, Descartes’ argument might seem to be flawed in either of two ways: either in its assumption that our idea of infinity is not just the concept of our own powers with their limitations and imperfections removed or by its basic assumption about causality being jeopardised by later worries raised by David Hume. Yet it was neither proto-Humean worries concerning causality nor equally empiricist concerns over the idea of infinity that most troubled Descartes. Rather, he was most immediately concerned with the unlikely possibility that the idea of an infinite God might be generated by his own self if that self is God. This potential problem, however, is recognised and raised by Descartes only to be then quickly dispatched with what must surely be still considered to be a knock-down argument: by pointing out that the subjectivity found by the cogito argument has not always known itself to be God and so it therefore demonstrably does not possess omniscience as one of its attributes and so it cannot, after all, be God.2 Descartes writes:
Perhaps also I am something more than I imagine myself to be and all the perfections I attribute to the nature of a God are in some way potentially in me … Still, all these excellences do not belong to or approach in any way the idea I have of a Divinity, in whom nothing is to be found only potentially but all actually existent. And is it not even an infallible argument of the existence of imperfection in my knowledge that it grows little by little and increases by degrees?3
At this point, it might be thought to be still open to the particularly obstinate sceptic to desperately suggest that Descartes may be wrong in thinking doubt an imperfection. But this last objection is evidently invalid for if Descartes is wrong in thinking doubt to be an imperfection then he is still, qua maker of mistakes, an imperfect being.4
Kant too, although he sophisticates Cartesian self-knowledge by distinguishing between the phenomenal self as an object of inner sense and the existence indicated by apperception also entertains the thought that the self revealed by self-awareness might be God in his Lectures on Philosophical Theology:
When I think I am conscious that my ego thinks in me, and not in some other thing … I exist for myself and am not the predicate of any other thing … Either I must be God himself or God is a substance different from me. [LPT 75]
Kant’s answer to this potential puzzle – an explicit answer in the Lectures but only implicit in the first Critique – is essentially the Cartesian one that God does not think in the manner that we do but it is coupled with the important Kantian qualification that the knowledge that pertains to the divine mind is qualitatively as well as quantitatively different: ‘All his knowledge must be intuition, and not thought, which always involves limitations’ (CPR B 71). When Kant – to be knowingly followed by Kierkegaard5 – suggests that God does not think what he means is that God’s knowledge is not conceptual but is ‘intellectual intuition’ which creates rather than perceives objects. As this point may also be put, since human knowledge is always partly discursive, the recognition of God’s extradiscursive omniscience allows Kant to remain as untroubled as Descartes by the solipsistic argument to the effect that our subjectivity must be identified with God (the Refutation of Idealism section and the first Analogy of Experience of the first Critique arguably disprove the idealism of objects – although not of subjects – in the external world, further suggesting we are not responsible for everything around us). But the idea of extra-discursive omniscience, though, might be thought to lead to a troubling problem of its own.
The problem is this: the fact that, according to Kant, divine ‘intellectual intuition’ creates rather than discerns objects seems to have a serious implication for another of Kant’s own doctrines, that of the purported spontaneity of our own human understanding, as it seems to render the reconciliation of our understanding’s spontaneity with God’s productive omniscience problematic. The source for these concerns here is a series of remarks by H.E. Allison in his Kant’s Transcendental Idealism, a text otherwise not centrally concerned with Kantian religious thought. Allison maintains that:
The difficulty stems from the productive, archetypal nature of intellectual intuition. In conceiving of myself as known by such a mind, I would be constrained to regard the spontaneity of my own thought as the product of something else. This is … a contradiction.6
Allison does not draw any specific conclusion from this contradiction but I suspect he means to suggest that we might do well to regard the place for a supreme Being in Kantian thought with a measure of scepticism, if not cynicism. Yet there is a viable alternative: we might do just as well to concede Allison’s point about the conflict between, on the one hand, omniscience construed as intellectual intuition and, on the other hand, our understanding’s spontaneity, but then rise above that very conflict by construing divine omniscience as something other than intellectual intuition. Furthermore, we should not feel compelled to state what exactly this ‘other omniscience’ need be. Admitting our ignorance of what God is seems to be neither untrue to the measured scepticism of Kant’s own writing nor untrue to the biblical criticism of idolatry. It also has the secondary advantage of granting God the power to represent to himself objects not present without actually making them present.
Returning to the main line of argument in the first Critique, Kant thus embarks upon the process of circumscribing the structure and function of the human intellectual and sensible capacities. In the Transcendental Aesthetic, which opens the main text of the Critique of Pure Reason, Kant argues that all our experience is not simply passively received from the external world as on the empiricist model of perception but is rather subjectively organised, partly by what he calls two a priori forms of sensible intuition, which are space and time. Space and time, according to Kant, are not derived from experience and do not in fact exist outside our actual or possible experience but rather are the sensible ways in which we experience our world. Kant had two arguments to demonstrate that space and time are not derived from the external world and a further two aiming to show that they are intuitions rather than concepts. It is worth mentioning the arguments in summary fashion (concentrating on the arguments concerned with space: the corresponding arguments for time are parallel formulations), as a grasp of their structure would seem to be a prerequisite for an understanding of the nature, and the implications, of Kant’s concept of God.
In the first place, Kant argues that space and time are given to us prior to sensory experience because any spatial relations we perceive presuppose space as a whole. We cannot derive the notion of a unified spatial field from noticing space relations between particulars or from noticing spatial properties of particulars because any spatial characteristics that we might derive from an object already presuppose such a unified spatial field, such a field also being presupposed by the distinction between objects and the distinction between objects and our own embodied self in the first place. Our knowledge of space must therefore be a priori: absolutely independent of experience. This, however, was not Kant’s only argument for the a-priority of space and time. A second argument which Kant used to this effect – Schopenhauer considered this particular argument to be the knock-down one (W II 33, PP II 44) – is to emphasise the fact that we cannot represent to ourselves the absence of space, although we can think what Locke called ‘pure space’, that is, space empty of objects (A 23–24=B 38–9). The fact that we can represent to ourselves space empty of objects but not objects devoid of space is taken by Kant to suggest that space is an ineliminable part of perception in a way that objects are not. Nevertheless, despite Schopenhauer’s appreciation, various criticisms of Kant’s arguments could be brought to bear on the discussion at this point. In opposition to the first argument, for example, one could claim that whilst it does show that the outer world cannot be represented except as spatial and so space cannot be derived from the experience of an external world, it nevertheless leaves open at least the possibility that spatiality and the outer world are intuited contemporaneously.7 And in response to the second argument, P. Guyer has pointed out that even if space was an empirical representation it could conceivably become so entrenched that it could not be imagined away, even if any particular object could.8 Yet here we are not going to labour any such criticism of the structure of transcendental idealism itself. Rather, it seems better to provisionally accept the framework of transcendental idealism – in both its Kantian and subsequently its Schopenhauerian form – so as to examine the implications this idealism may have for arguments concerned with theism and atheism.
Returning to the argument of the Transcendental Aesthetic, Kant further believed not only that space and time are a priori but also that they are intuitions and not concepts. His first argument to show that space is an intuition (Anschauung; broadly meaning: ‘something looked at’) and not a concept is that it is a unified individual thing: all particular spaces are just parts of space as a whole (in the sense that if we wanted to draw a certain figure, we must also already have the space in which to draw it). So all talk of diverse spaces really refers to parts of the same space. Particular spaces are therefore not instances of a distinct concept but rather parts of a unitary whole which must be something immediately sensed (the hidden premise here being that we perceive individual things but conceive universals). The second argument to show that space and time are intuitions is by common consent more difficult to discern but, at least according to Allison’s reconstruction of the argument, suggests that space is divided by introducing limitations or boundaries and can have an infinite amount of parts, which is how an intuition is divided; whereas concepts are divided intensionally into other concepts within it as component parts, which are not infinitely divisible.9 Space, Kant again concludes, therefore must be thought to be an intuition – something picturable – rather than a concept. It may also be pointed out that Kant also has an argument from geometry and an argument from ‘incongruent counterparts’ to show that space is an a priori intuition but examining even the outlines of these extra arguments would take us much too far afield at this point (the latter does not even appear in the Critique of Pure Reason and, in any case, can prove nothing with regard to the status of time, to which it lacks any appreciable application).
The primary conclusion of these four compressed arguments – and it will be with this conclusion and the implications that follow from it that we shall often be concerned – is groundbreaking: space and time are taken by Kant to exist only in a subjective or even anthropological sense: ‘We deny to time all claim to absolute reality; that is to say we deny that it belongs to things … independently of any reference to the form of our sensible intuition’ (CPR A 35–36=B 52; the same claim is made for space at A 46=B 63). Yet it should be pointed out that commentators have often made the objection that even if Kant has succeeded in establishing space and time as forms of our intuition this does not exclude them from a simultaneous objectivity (sometimes known as the ‘neglected’ or ‘missing alternative’ argument). S. Körner maintains that ‘It is always logically possible that what we see under the form of space and time is so ordered independently of our perception.’10 Adherents of this view would presumably want Kant to recant his strict ontological claims about the non-spatio-temporality of the noumenal world, which they would see as an unjustified move, and perhaps submit instead to a kind of Husserlian transcendental epoché, refraining from comment on the spatiotemporal character of the world as it exists in itself. Kant himself, however, allowed space for what Paul Guyer has called ‘a theological argument against the ultimate reality of space and time.’11 This argument consists of the charge that the conception of space and time as objective forms of objects is incompatible with natural theology in as much as God Himself would therefore have to be spatiotemporal in this account, which is obviously absurd (CPR B 71–2]. However, as Guyer rightly points out, the objective view of space and time is not incompatible with natural theology so long as we suppose only that space and time are genuine properties of some but not all things in themselves. In the light of this admittedly elementary distinction, Kant’s theological argument against objective spatiality and temporality and therefore against Körner’s objection must be seen to fail (it might also be seen to be disabled, as S. Gardner points out, by the contentiousness of its implicit premise that the concept of God is coherent12). On the other hand, to Körner’s objection it may be more successfully retorted that whilst it may be true without specific reference to the Kantian philosophy, Körner’s point cannot be considered to be by itself decisive against Kant himself since, in his Antinomies of Pure Reason, Kant had attempted to show that irreconcilable contradictions result from taking space and time to be objective aspects of reality and so, taken as a whole, the Critique of Pure Reason does exclude space and time from being transcendentally real, at least if one considers the antinomies section to be successful.
His case for the transcendental ideality of space and time being made, Kant goes on to argue – in the Transcendental Analytic – that human experience is further structured by twelve a priori logical ‘categories’. The categories seem to have been found in the so-called ‘Metaphysical Deduction’ by an analysis of all the kinds of judgements that there are according to Aristotelian logic. The categories are then collectively justified in the ‘Transcendental Deduction’ by an – even by Kant’s standards – uncommonly tortuous and obscure argument, the spirit of which is that the use of these categories is intrinsic to the temporally extended nature of experience as such.13 This seems to be because when we perceive an object Kant thinks we actually also judge it to be an object, judging being a conceptual operation that is temporally extended. Grasping objects as objects is taken to be an active affair of the mind in this way because any given array of sensations has to be apprehended as – that is, has to be judged to be – an object in a way that experience itself is impotent to carry out. And grasping objects is taken to be conceptual because judgement is nothing other than the employment of concepts (judging some shape in one’s field of vision to be an object is to bring it under the concept of an object). And given that conceptually judging objects takes place over time, it requires an abiding self to synthesise (it could not, in other words, be different selves who brought elements a + b together into one complex representation, it must be one unitary self-consciousness, which Kant, unfamiliarly, terms the ‘transcendental unity of apperception’). Thus it is our perception of the synthesised manifold itself that allows us to be sure of an abiding self. As this point may also be put, objectivity (a synthesised manifold that we perceive) and subjectivity (the abiding self that synthesises in perception) cannot be presented without each other. The last thing to note here is that this unity is experienced as active: Kant repeatedly insists the mind is aware of the spontaneity of its acts of synthesis. As previously noted, however, this seems inconsistent with the notion of God qua intellectual intuitor and abandoning the latter notion was suggested (there remains the option of abandoning, or qualifying, the former).
This necessity of the structure of both perceiving and thinking that is built into human experience from the very outset means that we possess the advantage of being able to perform what Kant calls synthetic a priori judgements, by which he means judgements that are about the world we experience rather than merely about the meanings of the concepts involved (synthetic judgements) but that are nevertheless possible independently of experience (a priori judgements). A paradigm case of such judgements is the metaphysical claim: ‘every event has a cause.’ Such a judgement of universal causality cannot in fact be rationally derived from or justified by experience – as has been evident at least since Hume’s Treatise on Human Nature – but Kant claims that we can nevertheless know it holds true of all possible experience because causality is one of the twelve categories found in the ‘Metaphysical Deduction’ (in the Second Analogy a deeper and more specific justification of causality as supporting our notion of an objective time-series is offered, thereby arguably making good any suspected argumentative deficit in the ‘Metaphysical Deduction’). Our synthetic a priori judgements were therefore only possible independently of experience because they told us about the way our mind regulated nature – through its forms of intuition and its categories – rather than about nature as it might be thought to be in itself. Human thought could consequently not represent reality as a whole. Kant nevertheless believed that we should not subscribe to empirical idealism of the Berkeleian sort and supported this by sophisticating Locke’s hesitant theoretical position regarding material substance by the addition of a rather complicated argument (which surfaces both in the First Analogy and in the Refutation of Idealism and which I will not rehearse here) concerning the awareness of ourselves as extended in time requiring the existence of enduring entities in the external world. These enduring entities were themselves, when considered outside of space and time, unknowable and Kant referred to this unknowable realm beyond representational experience as the intelligible world and these unknowable entities (or entity) as the Noumena (or Noumenon). Kant also thought that our reason, in spite of being forever divorced from this intelligible world, could not but attempt to reach that unknowable reality, and the totalising aberrations or illusions of reason, by means of which the mind dogmatically posits God (and a soul and a world thought of as a completed whole) were to be regarded as natural and unavoidable. For Kant, the idea of God is therefore neither innate, as it was in Descartes’ third Meditation; nor is it empirically formed, as empiricists such as Locke and Hume thought, by enlarging the ideas of our own nature with the idea of infinity and removing our imperfections. Typically, Kant creates an ingenious compromise between rationalism and empiricism here: the Kantian idea of God is an idea that though not innate is nonetheless inevitably created. This tendency toward forming an idea of God, given that it is natural to all humans, means that monotheism is a trend, Kant notes in his first Critique, to be found transculturally and transhistorically located:
In all peoples, there shine amongst the most benighted polytheism some gleams of monotheism to which they have been led, not by reflection and profound speculation but simply by the natural bent of the common understanding. [CPR A 590=B 618; see also LPT 73, CPrR 168]14
Furthermore, much like the psychopathological complexes outlined by Freudian psychoanalysis, the three sophistications of reason that Kant postulated and explored do not cease to function even when they are detected and have their invalidity clearly revealed to the subject of the aberrant thought process (CPR A 339=B 397; A 297=B 353).
Kant included in the first Critique a Transcendental Dialectic to counteract such natural ‘illusions’. We will mention but will not spend much time assessing either the general merits of the Dialectic or the particular merits of Kant’s refutation of speculative theology, partly because it has been discussed elsewhere by others but also because there is substantially more to Kant’s philosophy of religion than this aspect of transcendental idealism’s criticism of rationalism. Yet it is worth providing an overview of Kant’s attacks as they are presented in the Transcendental Dialectic (Book II, chapter III, sections three to six).
Kant considers all proofs of God to be instances of one of three types (CPR A 590=B 618). The first type is the ontological proof, which argues a priori that the concept of God analytically entails his existence (after the manner of one of the arguments of Descartes in the Meditations). Oversimplifying, we might say the point being made here is that ‘God’ in its normal meaning means, amongst other things, an all-powerful, all-knowing and existent creator. Thus the claim that God exists is guaranteed by the fact that the meaning of the term God includes existence in its definition. This is taken to fail by Kant because it assumes that existence is a characteristic which could function as a genuine predicate of a concept, whilst Kant – following Gassendi but contra ordinary language – famously disputes this (though whether he is right to do so is not an issue that can be studied in detail here). The second argument is the cosmological, which argues that contingent things must have been caused to exist by something else which, if also contingent, must in its turn have been caused, and so on until we reach a necessary being. This second proof failed according to Kant because it extended the concept of cause outside of the world of our possible experience and further failed to identify the concept of cause with an all-powerful and all-good God (at least without surreptitiously reintroducing the ontological argument to supply the absent predicates). The third argument is the physico-theological proof: in essence, the argument from design. Put crudely, it argues that this world shows order in an analogous way to a watch and since a watch has a purposive creator we may presume the same to hold for the world. This proof from apparent purposivness in nature, Kant argues, is at the most only licensed to posit an architect and not a creator of the world (and he might have added, as Hume’s biting Dialogues on Natural Religion did, that for all we know that architect might now have expired). To postulate a creator ex nihilo it would have to fall back on the cosmological proof, which itself relied upon the ontological (unlike Hume’s, Kant’s counterargument is therefore powerless against someone who wants only to prove a superhuman architect and not an all-powerful creator – but Kant quite implausibly thinks that no one would be interested in such relatively reduced aims15). All possible proofs of God thus eventually collapse into the ontological proof, which itself is – as we have already seen – fallacious, according to Kant.
So much for the general structure of Kant’s overall reasoning in the Critique of Pure Reason. It is time to narrow down our focus and examine some of the main implications for theistic religion of this intriguing account: Kant’s construal of God as existing outside of space and time and then his falsificationist scriptural hermeneutics.
Kant unconditionally ruled out the possibility of human contact with a divine being. The first step that allows Kant to do this is his premise that we can perceive nothing and therefore know nothing that is not in space and time. (We have already noted the thinking behind that premise.) His second step is to construe the supreme being as just such a non-temporal and non-spatial existence, as we have also already acknowledged in considering Kant’s ‘theological argument against the ultimate reality of space and time’. This construal of God is also illustrated in other of Kant’s important works: in Religion within the Limits of Reason Alone, we read that ‘The existence of God in space involves a contradiction’ (Rel 130 n); whilst in the second Critique, Kant writes of ‘The infinite Being, to whom the condition of time is nothing’ (CPrR 149). The condition of time is said to be nothing to God because ‘if God were in time he would have to be limited. But he is a realissimus, and consequently he is not in time’ (LPT 71). Kant thinks that ‘God is wholly distinct from the world and has no connection at all with space and time’ (LPT 104) because otherwise spatial and temporal boundaries would limit God, and a restricted God is, by definition, a contradiction. It should probably be stressed here that we cannot picture or represent such an atemporal and aspatial God, since on Kantian premises we cannot represent to ourselves anything lying outside what are essentially our forms of representation, the ways that we picture things at all. As Kant puts it in the Critique of Judgement:
We think of the eternity of God as presence in all time, because we can form no other concept … or we think of the divine omnipresence as presence in all places, in order to make comprehensible to ourselves His immediate presence in things. [CJ 337]
But however established this conception of God as outside of temporal and spatial determinations might be – itself a controversial issue, as there is still debate in philosophical theology as to whether God should rather be construed as eternally existing through time – it nonetheless means that, on Kantian premises, we are unable, even in principle, to encounter God sensibly at all. Since we necessarily see the world through space and time but God as conceived of by Kant exists outside such qualifications, then, as Kant himself puts it, the ‘feeling of the immediate presence of the supreme being would constitute a receptivity for which there is no sensory provision in man’s nature’ (Rel 163). Yet this is quite obviously not an atheistic position, since by the same token knowledge of the non-existence of God is similarly ruled out in principle.
From a traditionally Judaeo-Christian-Islamic religious perspective, the spaceless and timeless God of Kant’s philosophy appears to expressly contradict the revelations of God which we find in scripture (it could be argued that this is also a problem for certain other, more orthodox, Christian theologians who are committed both to Platonism and the Christian scriptures, but that is another, more expansive story and the wider point will not be argued here). We are told by the Bible that God made the heavens and the earth in six days and that he has intervened in our physical world in various visible capacities. For a Christian, the problem of the Kantian aspatial and atemporal God contradicting the biblical account might appear to be very stark indeed, since according to the Nicene creed – published by the council of Nicaea in 325 to combat the heresy of Arianism – Jesus Christ was a wholly divine figure who nevertheless entered into human history and experience. So according to Christian tradition, God entered space and time but according to Kant: ‘God is wholly distinct from the world and has no connection at all with space and time’ (LPT 104). Consequently, from what we know of Kant’s – seemingly partly evasive – Christology, it seems that Kant, instead of sacrificing human reason itself to this paradox about an eternal God becoming finite and accordingly seeing Christianity as being rationally indefensible, as Christians such as Kierkegaard (himself possibly forced into such a position by his prior acceptance of a broadly Kantian epistemology16) were to appear to do, was instead scarcely inclined to treat Christ as divine.17
But might we not suspect this position to be irreligious, substituting for the Christian God a ‘God of the philosophers’? This is why Kant could be charged with being a ‘deist’ in a way that Kierkegaard could not (terms such as ‘deists’ and ‘deism’ are meant to refer to those who reject the evidence of historical revelation of God but believe the existence of God to be nonetheless assured by reason). Has Kant irreligiously spurned divine revelation? Investigating Kant’s examination of biblical theology in Religion within the Limits of Reason Alone is the only procedure that will allow us to see if Kant can answer such a charge.
In the first few pages of Religion, Kant acknowledges that philosophical theology is not the only kind of religious thought: a religion such as Christianity which has been historically revealed, at least in part, through specific, extant, canonical texts must include hermeneutic reflection as part of its discursive apparatus. Yet, to rehearse the point made above, the Kantian God and the God revealed through the Bible seem to oppose each other. The Kantian God remains outside of the human world of space and time and history whilst the God of scripture, especially (but not only) Christian scripture is a providential God who fully enters into human historical affairs. How does Kant attempt to resolve this conflict between reason and revelation? In a late work, Kant set to work solving it by analysing the relationship between biblical theology and his own epistemology. Put bluntly, the analysis of biblical theology in Religion within the Limits of Reason Alone effectively discredits the Bible as an unquestionable source of divine revelation by the apparent detour of entering into theological debate about the relationship of priority between God and the moral good.
Readers acquainted with the history of philosophy will doubtless recall that this debate is in fact of truly ancient lineage, finding its locus classicus in Plato’s Euthyphro: ‘Is the holy approved by the Gods because it is holy, or is it holy because it is approved?’18 In that dialogue, the character of Euthyphro himself espouses the second clause (7a), a position which is duly problematised by Socrates, who argues that strife within the Greek pantheon on moral questions vitiates any recourse to the Gods as final ethical arbiters. This theoretical difficulty within Greek polytheism is ignored in Socrates’ own admission in Plato’s Apology that a so-called divine voice had advised him not to take certain courses of action and is in any case logically absent from the corresponding position in any monotheistic religion, as seems to be demonstrated in many of the writings of the Christian author, Søren Kierkegaard. By means of illustration, it is worth now pointing to the logic of one of these texts in particular: Fear and Trembling. It should, however, first be mentioned that Kierkegaard was an author who was intensely concerned with ironising his prolific output, in part by ascribing it to various narrators. According to more recent authors, the (internally consistent) views of these narrators are therefore not directly attributable to Kierkegaard himself. Many now therefore attribute the views of, say, Fear and Trembling to its pseudonymous author, Johannes de Silentio. Such a procedure has not, however, been followed in this book. This is chiefly because Kierkegaard’s work is used here merely as a expository apparatus in the articulation of Kant’s views.
Fear and Trembling explicates the akedah: the incident of Abraham’s near sacrifice of his son in Genesis 22, a narrative common to all three of the Semitic monotheistic faiths (though Islam is at odds with both Judaism and Christianity in claiming that the son in question is Ishmael and not Isaac). Crudely, Abraham receives a summons from God commanding him to sacrifice his son. Abraham neither doubts nor challenges the vox dei and, knife in hand, he takes his son to the designated spot only to be then granted a last-minute reprieve. One of Kierkegaard’s chief arguments in this text is simply that given such sanguinary stories in the Bible one cannot reconstruct God as wholly moral without doing some serious violence to the authority of scripture. It seems that on such a view, which in fact accords to some measure with that of St Augustine, we are to obey the revelations of God even if they seem madness to our moral standards – even if, as Kierkegaard elaborately states, our moral standards then themselves become the temptation that would prohibit us from doing God’s will.19
Yet Christianity is essentially moral for Kant in a way that it is not for Kierkegaard. Kant’s view is that God is essentially good, a conclusion that follows on naturally from the Kantian ‘moral proof of God’, which is itself based on the necessity and utterly obligatory nature of morality as seen by Kant. Kant argues that we are thereby given reliable criteria for recognising as either spurious or (potentially) authentic revelations or commands that might or might not be believed to have come from God instead of having to trust them by a criterionless faith, after the manner of Kierkegaard. In other words, since it has been philosophically decided by Kant that God is to be wholly good, anything in the scriptures that suggests otherwise must be reinterpreted by us to fall in with the philosophical (Kantian) view of God. Thus is the revelatory power of religious scripture (and by implication the freedom of the God of religious scripture and the Christian tradition) subordinated to certain of the tenets of Kantian philosophy.20 Our moral reason therefore supplies a criterion for our decision on the authenticity of outer revelation from religious scripture and so, as Allen W. Wood has rightly put it, ‘our moral conception of God provides us with a means of determining the moral purity – and consequently the possible authenticity – of the alleged revelation of such a God’.21 There is no rational way of knowing whether a seeming revelation of God is absolutely true but there is a rational way of knowing whether it is false: if it does not meet the test of our internal moral reason. It is Kant’s considered opinion that claims to divine revelation can never be verified – but that they can be conclusively falsified.
If some claim to revelation ‘flatly contradicts morality it cannot, despite all appearances be of God (for example were a father ordered to kill his son who is, as far as he knows, innocent)’ (Rel 82). This remark clearly alludes to the narrative of Abraham and Isaac, which is also explicitly mentioned later in Religion when Kant notes that since an ostensibly divine injunction is always interpreted by men ‘Even did it appear to have come from God himself (like the command delivered to Abraham to slaughter his own son like a sheep) it is at least possible that a mistake has prevailed’ (Rel 175). Since Kant’s moral proof of God is taken to furnish us with an indication of God’s existence and also with actual (moral) information about God, it furthermore provides a falsificationist guiding thread for biblical exegesis: scripture is not seen by Kant to be a higher court of appeal than those conclusions supplied to us by our faculty of reason; our moral reason limits what scripture can reveal.
Although the Kantian and biblical Gods do seem to contradict each other, Kant attempts to explain this by pointing out that a wholly moral God is the philosophically rigorous one whereas the God of the Bible is, in a sense, not to be wholly trusted since the Bible itself is not a vehicle of autonomous revelation. The answer to the Pascalian objection that Kant is not dealing with ‘the God of Abraham, Isaac and Jacob’ as revealed to us through holy scripture is to concede that Kant has indeed replaced that personal and historical revealed God who is inscrutable to our moral sense and reason. The God whom Kant has put in His place is aspatial, atemporal, unconditionally moral and non-interventionist. Yet this transparently conflicts with the biblical God who enters history and who sometimes seems incommensurable with normal human moral standards, but Kant’s implied answer to this is unequivocal: so much the worse for scripture. Given, in other words, that the Judaeo-Christian God is one wholly active in the course of history and that Kant seems not to hold this, Kant must be regarded as a philosophical rather than a religious monotheist: we must deem Kant to be a deist. And it should be noted in this connection that one Kantian commentator has suggested that Kant was the ‘last great exponent’ of deism, whilst another author has admitted that Kant’s concept of God is ‘little removed from that of deism’.22
Kant conspires to present the God of his philosophy as being identical with the God of our religious tradition, but construing Him in Kantian fashion as absolutely excluded from space and time appears to challenge the content of Christian scripture; the scriptures of the Judaeo-Christian religion are further challenged by being presented not as definitively revealing God but as providing falsificationist support for an autonomous morality. Yet it remains to be mentioned that Kant does attempt to philosophically justify the positing of his moral God, principally by means of the so-called ‘moral argument for the existence of God’. Kant’s moral proof of God, as we have already remarked, is taken to furnish us with an indication of God’s existence and also with actual (moral) information about God, which obviously backs up Kant’s approach to reading scripture.
Kant’s critical philosophy both withdraws God from the world of experience and from direct revelation through scripture. To be sure, it still does suggest that we are nonetheless compelled to form an idea of God but even conceding – against a mass of sociological and historical evidence to the contrary, as Locke was already aware in the Essay – that the human mind is led to form an idea of one God, we still have no reason to believe this idea to be anything other than a unavoidable fantasy. But whatever the demerits of the account of illusions (the details of which have largely been set to one side here), Kant’s central indication of God, found in all three Critiques, is based squarely on the importance of our moral lives. However, there are flaws to be found within the structure of Kant’s argument. There is a sense in which this may not come as a startling disclosure: there are, after all, few explicit believers in Kant’s specific version of moral deism today. Nevertheless it might yet be argued that a generic type of moral deism still haunts the thoughts of more theologically minded reflective people on this subject, though perhaps less as a temptation than as a last resort, a residual feeling that in order for the world and our abrupt lives to be morally justified, God must exist. If this be accepted, then there is even more reason to examine a major philosopher’s attempt to construct a proof of God based on ethics. Now, the moral proof’s final formulation is to be found in a sequence of passages in the Critique of Judgement under the section heading ‘Of the Moral Proof of the Being of God’. It is this portion of text that will form the spine of our explication and examination of the moral proof, although it will also be supplemented by the other Kantian accounts, where this will prove helpful.23
Kantian humanity, we must bear in mind, rigorously separates itself from nature: ‘Rational nature separates itself out from all other things by the fact that it sets itself an end’ (Gr 99). Moreover, the ends it sets itself include moral ones. However, we have also to remember that although Kant seems to claim that the formal moral law we set ourselves requires us to act regardless of consequences, our human sensible nature still requires us, if we are not to despair, to have some end in mind when we act: happiness: ‘There is, however, one end that can be presumed as actual in all rational beings (so far as they are dependent beings to whom imperatives apply) … by a natural necessity – the purpose, namely, of happiness’ (Gr 79).
Happiness has such a purchase on us ‘by a natural necessity’ because we not purely noumenal beings: our real, sensible existence involves certain needs, the satisfactions of which are captured in the distinctively human concept of happiness: ‘Happiness is the satisfaction of our desires’ (CPR A 806=B 834); it is ‘the maximum of well-being in my present, and in every future state’ (Gr 81). Dieter Henrich nicely states that happiness
is the self-assuredness of the person who experiences the satisfaction of all his wishes and who sees all his struggles end. Only the person who sees his entire life as a complete success is happy. The concept of happiness implies, therefore, that it is possible to unite all wishes and desires in such an aggregate. Kant disputed this possibility.24
So, happiness is a desire that only rational beings can have – but only those rational beings who are subjected to needs, wishes and desires. The moral proof is empirical – ‘concerns us as beings of the world’ (CJ 298) – only therefore to the extent that we need to know that we have interests of sensibility, although from this empirical premise we can then infer that as rational as well as sensual creatures we would also desire happiness. Our desire for happiness can consequently be said to be distinctively human because, although the non-human animals can of course be attributed desires, they cannot readily be said to possess the second order desire for the fulfilment of their first order desires. Our desire for happiness can also be said to be distinctively human in yet another sense: it would also be alien to God, who is obviously not subject to needs. Given, then, that we have a sensible side which desires happiness to our human nature as well as our intelligible side, the highest good for beings such as we are – empirico-transcendental beings – is a happy moral perfection: a summum bonum or ‘highest good’ that cares for our actual sensual needs (happiness) as well as our moral requirements (CJ 300).
As partly intelligible and as partly sensible beings we have a dual aim (moral integrity and happiness). Nevertheless, this dual aim need not fissure us irreconcilably, as it can be united: ‘Virtue and happiness together constitute the possession of the summum bonum in a person’ (CPrR 135). More specifically, happiness and virtue are not (eudaemonistically) identified here; rather: our highest good is a place where virtue is rewarded with the happiness we all desire. That is to say, in order not to be torn in two different directions, we need to aim at a summum bonum that answers the claims of both parts of our nature. But how is such a happy and moral perfection to be achieved? Happiness and moral worth are only contingently (if at all) related in this sensible world. Nature, therefore, affords little hope for the systematic reward of good. Likewise, man as a species is extremely limited as regards his ability to control the consequences of his actions in the natural world and so man, no more than nature, can be expected to harmonise virtue with deserved happiness. Indeed, there is no a priori guarantee that the moral law will not drastically conflict with our sensible pursuit of happiness. In such a disharmonious case, our practical reason would be antinomically torn between the sensible claims of happiness and the intelligible claims of virtue, claims neither of which we can eschew. Why not?
On the one hand, eschewing the claims of morality is not an option because it is Kant’s own view that the moral good is a ‘non-hypothetical’, ‘apodeictic’ or ‘categorical imperative’: Kant probably more than any other major philosopher emphasised the obligatory role of duty and thought that what had moral worth was our intentions. Bringing these two features together, we can say that to act in a morally worthy manner is to act from the motive of duty alone, which commands us irrevocably (what our duty actually consists in is to act according to maxims that are not contradictory, though this topic will not be further examined here). And on the other hand, eschewing the claims of human sensibility in the same way that we might eschew the claims of those of our baser desires that drag us down to the level of feral nature is no easy option for Kant because the desire for happiness is not a part of non-human nature: our desire for happiness is distinctively human (it is a second order desire for the fulfilment of first order desires). Eschewing the claims of happiness would thus be like writing off the call of our own human nature. Furthermore, it would, as at least some of Kant’s remarks very strongly suggest, lead us to a despair in which we gave up acting ethically altogether.
Since neither man nor nature can ever be hoped to systematically harmonise that which we want as sensible humans and that which we desire as intelligible beings we must therefore – to do justice to both of the rational claims on us and to avoid a despair in which neither claim could be answered – assume an effective harmonising force to exist outside the sensible world. That is to say, we are obliged to be moral and so if we can only be so by postulating a force that rewards virtue with happiness then we are also obliged to practically postulate that force. The only theoretical framework within which such a proportionate causal relation obtaining between virtue and happiness can be posited, however, is a theological one:
We must assume a moral world cause (author of the world) in order to set before ourselves a final purpose consistent with the moral law, and so far as the latter is necessary, so far … the former must also be necessarily assumed, i.e. we must admit there is a God. [CJ 301]
We can only imagine a realm where people are rewarded for their goodness (which is what we want to aim at if we are to be true to both our sensible and our intelligible nature and not to despair) as being under the command of an omniscient and omnipotent God who will take upon himself the task of organising, in G. Michalson‘s censorious characterisation, ‘a mysterious proportioning process occurring after my death in an unimaginably remote noumenal zone.‘25 Yet the supposition of a wise author and ruler is, it should be stressed, conditional upon us accepting the claims of morality and of our sensible nature. If we believe in morality and we also need happiness to aim at, our ultimate goal can only be a rewarding afterlife administered by an essentially benevolent personal God.
It should probably be stressed that Kant‘s moral proof is a proof that results from practical reason, not theoretical reason, which means that it results not in an objective finding (‘God exists‘) but in a kind of necessary existential commitment: we have to assume and hope for God‘s existence to marry our two separate goals and stop us from despairing. ‘I believe in the existence of God’ must be our conclusion, not the simpler ‘God exists’. Yet even on this existential basis certain objections can be made to the moral proof, although I should say that the question of the acceptability of Kant‘s account of the ‘categorical imperative‘, or absolutely obliging nature of morality, that underlies the moral proof is not an issue which we can discuss adequately here, so that important aspect of Kant‘s practical philosophy will have to be largely set to one side. There are, however, at least two other objections that can now be raised.
First, it is arguable whether Kant‘s psychological account of us ‘as beings of the world’ really rings as true as Kant himself supposed. Let us consider the position in what is probably its most comprehensive statement, articulated around the conceit of a righteous but faithless man, where it at least seems to be suggested that without the hope for happiness we would cease to act morally:
His effort is bounded; and from nature … Deceit, violence and envy will always surround him, although he himself be honest, peaceable, and kindly; and all the righteous men he meets will, notwithstanding all their worthiness of happiness, be yet subjected by nature, which regards not this, to all the evils of want, disease and untimely death, just like the beasts of the earth. So it will be until one wide grave engulfs them all (honest or not, it makes no difference) and throws them back – who were able to believe themselves the final purpose of creation – into the abyss of the purposeless chaos of matter from which they were drawn. The purpose, then, which this well intentioned person had and ought to have before him in his pursuit of moral laws, he must certainly give up as impossible. [CJ 303]
This is a nightmarishly well-made point yet it is hardly an uncontroversial one and it might be possible to assail Kant‘s moral proof of God here at its root by simply questioning whether we actually need the expectation of individual human happiness as an end for human action in the absence of which we would despair. The assumption that we do strive for such a happiness (which only God can systematically provide) to so impel us is seemingly central to Kant‘s moral proof but is it really possible that acting morally without the belief in adequate reward in terms of individual happiness would lead us to despair? Can anything else can be thought to function just as well as happiness in terms of motivating us to act without incurring dehabilitating melancholy?
One might think that an attachment to a personal cultivation of the virtues, particularly those tied up with a more terrestrial and more attainable perfectibility, might be enough to keep us acting without sensing any futility. Could there be a theory of the virtues which would move men in accordance with morality? Amongst theories of virtue and perfectibility, those of Aristotle and Alisdair MacIntyre are probably most likely to be mentioned. But since the virtue ethics of Aristotle has a problematic residue of teleological arcaneness, this leaves us with the writings of MacIntyre, specifically, After Virtue.
After Virtue is an explicitly anti-Kantian attempt to reconstruct a justification of moral action, seeing itself as radically departing from Aristotle‘s biology and from the moral grounding of modernity. MacIntyre thinks that the notion of virtue is linked to that of a social practice and it will now be argued that his attempt to ground the motive of human moral activity on such attainable grounds is at least as plausible as Kant‘s attempt to connect motivation with happiness.
MacIntyre introduces the notion of a kind of constant we can aim for that is as cross-cultural and as trans-historical as is the expectation of happiness or as would be a purported human metaphysical telos: a ‘practice’. So although there is, for MacIntyre, no given biological telos of a human life as such, there are nevertheless social practices, such as sailing or playing in a string quartet, that are found in some form or other across all human cultures and that clearly constitute goals for human desire. His technical definition of a practice runs as follows:
By a practice I am going to mean any coherent and complex form of socially established cooperative human activity through which goods internal to that form of activity are realised in the course of trying to achieve those standards of excellence which are appropriate to, and partially definitive of, that form of activity, with the result that human powers to achieve excellence, and human conceptions of the ends and goods involved, are systematically extended.26
This is a compact definition and so it might be worth explaining in a little more detail here what ‘goods internal to a practice’ actually are. Such goods, we might say, are those goods which can only be achieved through participation in that specific practice and such goods must moreover have historically evolved standards of excellence internal to them. Human lives are thus intertwined with social practices and are, in a sense, therefore given certain goods (many of the technicalities and qualifications of MacIntyre‘s account which are unnecessary for my purposes have been omitted here). Virtue then becomes the name for those human capabilities that allow us to pursue practices and therefore aim for the goods internal to those practices. Resilience allows us to pursue the good internal to the practice of sailing a coracle; assiduity allows us to pursue the different good internal to the practice of playing the triple harp and honesty allows us to pursue the good internal to, say, playing such a game as chess. And all these practices, because they have historically developed standards of excellence, call for the virtue of accepting the judgement of a legitimate authority on our part: as novices or beginners, we have to accept the judgement of a past master as to what the good of a particular kind of musicianship such as playing the triple harp consists in. MacIntyre‘s concept of a virtue thus requires the background of a practice and the corresponding notion of recognised internal goods and therefore his main disagreement with Kant lies in his suggesting that in order to be reasonably motivated to act (without moral weakness) we need not aim for a highest good partly constituted by a – otherwise than theologically, unattainable – happiness but rather that we can aim at fostering virtues which support goods internal to socially given practices, goods that do matter to us, arguably as much as does happiness. Where Kant juxtaposed moral integrity and happiness, MacIntyre connects moral integrity and practices. Put differently, both Kant (at least in an important part of the moral proof, at any rate) and MacIntyre seem to agree that some pre-existent desires are somehow involved in practical reason but MacIntyre points out that one kind of moral reasoning appeals to desires for goals that do not need God to help us achieve them.27 It is not, then, that we find our happiness in practices but rather that practical excellence replaces happiness qua goal. Nor have we here identified the goal of our sensible nature (internal goods) and that of our intelligible (moral worth). The point being made here is rather that replacing happiness with excellence as the goal of our sensible nature allows us to fulfil that goal ourselves (that is, omitting God) without compromising our morality. Our different natures now have compatible goals and not incompatible goals that only a God could reconcile.
It is also worth remarking here that, even if, as in any case seems unlikely, the moral man‘s resolve did break in the way that Kant thought it would if he did not have faith in God and a desire for happiness, perhaps despair might not be so unconducive to ethical rectitude as Kant seems to suppose. Since Kant‘s time, we have seen repeated instances of people coping with personal tragedies of religious faith that are wholly unaccompanied by moral weakness or failure. The alternative Kantian idea that such despair necessarily leads to immorally self-serving acts or even acts of malicious evil is perhaps today only the commonplace of a certain kind of modern European narrative centrally concerned with sociologically disconnected ‘loners‘: for real-life individuals suffering from, or working through, the kind of post-Heideggerean anxiety that we find in the existentialist novels of atheists such as Sartre and Camus generally cannot be said to fall into the kind of highly immoral behaviour which we often associate with the ‘heroes’ of such fiction.
The second objection to Kant‘s moral proof I should like to raise is that even if the conclusion of this moral proof is that we must admit, albeit in an existential manner, that personally we must believe that there is a God, we must also admit that the God in which we must believe is a rather impoverished variety of Deity (when compared with the God of the Christian tradition). The particular problem here might be said to be that the moral proof can only establish some but by no means all of what Kant claims for it. For it seems that only the simple existence of a moral rewarder of virtue must be assumed to motivate us to act ethically in Kant‘s system (although some form of resurrection seems to be implied for us humans). What this then further suggests is that although we must assume an effective harmonising force to exist outside the sensible world in order to act morally, there is still quite a leap from positing that harmonising force to believing in the traditional Christian God of infinite power, mercy and wisdom. As Y. Yovel has pointed out, the initial introduction of God into Kant‘s argument actually depends upon certain of our subjective limitations – that is to say, our inability to imagine an ‘immanent principle of justice’. Yovel writes of the Kantian moral proof of God as follows:
This procedure of postulation consists of two distinct stages. At the initial stage, which alone has logical necessity, all that we postulate is a vague and indefinite principle … Of this something we know nothing except that it is there and it fulfils the function described … but here our subjective limitations come into play, forcing us to imagine this factor with the aid of metaphoric, anthropomorphic imagery … and regard that ‘something’ as a supreme personal being, endowed with understanding and will, who is the ‘moral author of the world‘, that is, God.28
Implicit in Yovel‘s characterisation here is the truth that, as Michalson has it: ‘The God of the moral argument is chiefly an instrument in the realisation of a rational goal and little more … certainly Kant‘s argument does not account for the full roster of divine predicates.‘29 Nor does it suggest the idea of the Messiah (the coming or second coming of the Messiah being important to all the Semitic monotheistic faiths). So without the addition of our anthropomorphic imagery, all that logically follows from the moral proof of God is in fact some kind of instrumental principle of justice and it is highly unobvious that we should identify this bare principle either with the revealed Christian God of history, mercy and redemption or even with the God of the rationalist philosophers, who was the most perfect being, an uncaused cause and who held a providential design for the world: ‘The traits or attributes of the deity who is at issue in the first Critique are considerably more numerous than those of the God produced by the moral proof.‘30 Some of the kind of problems that we might associate with Kant‘s own aggressive tactics toward theological argumentation in the Transcendental Dialectic might therefore be thought to come home to roost here: there is, for instance, little reason to be found in this particular moral argument, which logically proves only a principle of justice, why we must consider this ‘God’ to be the creator – or sustainer – of the world. This instrumental principle of justice admittedly has to have the power to make the summum bonum achievable but still, a less than omnipotent and in any case arguably impersonal demiurge could accomplish such a task. It is in such a connection that D.M. Mackinnon has aptly noted: ‘For Kant, God is less the creator than the ultimate judge.‘31 I would like to add that there seems to be no clear conceptual connection between the notion of a rewarder of just acts and the idea of a personal creator God that would bridge the clear gap in Kant‘s argument – a gap as vast as that between what the cosmological and the physico-theological proofs attempt to prove and what they do in fact prove without the backing of the ontological argument.
The moral proof, even considered to be the result of practical and not theoretical reason, therefore proves only a ‘vague and indefinite principle’. This ‘principle’ is not the God of the Christian faith. Yet there are two obvious ways of refurbishing the sparse Kantian concept of a principle of justice that results from the moral proof with a more substantial inventory of divine predicates. The first would be to return to one of the traditional demonstrations of an all-powerful God (such as that to be found in the ontological proof of Descartes). The second would be to concede, alongside thinkers such as Pascal and Kierkegaard, that the real core of the Christian faith in a personal God is to be found in sacred revelation through scripture after all. Yet neither of these argumentative routes are live options for Kant because he has already shut them both off in a decisive fashion in the Transcendental Dialectic of his first Critique and in Religion within the Limits of Reason Alone, respectively.32
In rejecting the possibility of human sense experience of God, and in abandoning the prospect of the disclosure of God through scripture, Kant arguably changed the face of the interpretation of religious experience – and the lack of feeling now encountered in the presence of the religious is perhaps the weight by which we appropriately measure our loss.
We know that Kant himself thought that his removal of ontological questions of God‘s existence was actually open to a fideistic reading in the context of the critical system as a whole, albeit a fideistic reading associated with rationality in a way that Kierkegaard‘s fideism was not: ‘I have therefore found it necessary to deny knowledge in order to make room for faith’ (CPR Bxxx). And later: ‘For although we have to surrender the language of science we still have sufficient ground to employ, in the presence of the most exacting reason, the quite legitimate language of a firm faith’ (CPR A 745=B 773).
Nevertheless, as we shall see presently, one immediately post-Kantian philosopher will vigorously deny this claim; whilst his one-time disciple will then go on, not only to raise the idea that Kant‘s ‘Konigsbergian nihilism’ is ultimately lifedenying but also to argue that many possible responses to it – Schopenhauer‘s included – suffer precisely the same debased fate.
1This is, then, a rather traditional proof of God from a consideration of his creatures, except that Descartes begins his proof not from the visible world but from his own mind and ideas – as of course he must do, the visible world not being assured of its reality until a beneficent God‘s existence has itself been proved, see A. Kenny, Descartes (Bristol: Thoemmes, 1993), p. 127. Berkeley‘s proof of God also begins from his own mind and ideas – but in his case for idealist rather than sceptical reasons.
2I said that this haunted modern phenomenology: Maurice Merleau-Ponty employs the Cartesian exit thus: ‘I can never recognise myself as God without necessarily denying what I am trying in fact to assert‘, Phenomenology of Perception (London: Routledge, 1994), p. 359.
3Descartes, Discourse on Method and The Meditations, pp. 125–6.
4B. Williams, Descartes: The Project of Pure Enquiry (Harmondsworth: Pelican, 1985), p. 145.
5S. Kierkegaard, Concluding Unscientific Postscript to Philosophical Fragments (Princeton: Princeton University Press, 1992), p. 332: ‘God does not think, he creates.‘
6H. E. Allison, Kant‘s Transcendental Idealism, pp. 289–90.
7S. Gardner, Kant and the Critique of Pure Reason (London: Routledge: 1999), p. 83.
8P. Guyer, Kant and the Claims of Knowledge (Cambridge: Cambridge University Press, 1992), p. 347.
9Allison, Kant‘s Transcendental Idealism, p. 93.
10S. Körner, Kant (Harmondsworth: Penguin, 1984), pp. 37–8. Relatedly, Körner has repeatedly pointed out that all transcendental arguments fail to be uniqueness proofs, that is, that they leave open the possibility that another set of conditions could allow the experience in question to occur.
11Guyer, Kant and the Claims of Knowledge, p. 352.
12Gardner, Kant and the Critique of Pure Reason, p. 103.
13Kant himself further distinguished between the ‘objective’ and ‘subjective’ deduction, the latter referring to the first edition‘s analysis of the faculties (CPR Axvii). We may, following D. Henrich, regard the objective deduction as proof that the intuitions are subject to the categories, while regarding the subjective deduction as a proof of how they are so subject; see Henrich‘s meticulously argued ‘The Proof-Structure of Kant‘s Transcendental Deduction’ in Kant on Pure Reason, ed. R.C.S. Walker (Oxford: Oxford University Press, 1982), pp. 66–81, esp. p. 75.
14This natural dialectic is seen by Kant to take on one of two forms. A good account describing these illusions in their specificity (the idea of a first cause and that of an Ens Realissimum) can found in A.W. Wood‘s Kant‘s Rational Theology (Ithaca: Cornell University Press, 1978), pp. 25–79.
15J. Bennett, Kant‘s Dialectic (Cambridge: Cambridge University Press, 1974), p. 256.
16See R.M. Green, Kierkegaard and Kant: The Hidden Debt (Albany: State University of New York Press, 1992), p. 136: ‘Kierkegaard‘s very understanding of why Christ is the “absurd” presupposes Kant‘s epistemology.’
17On this, see G.E. Michalson, Kant and The Problem of God (Oxford: Blackwell, 1999), p. 102.
18Plato, Euthyphro, 10a.
19S. Kierkegaard, Fear and Trembling (Harmondsworth: Penguin, 1992), p. 88; see also Kierkegaard, Concluding Unscientific Postscript to Philosophical Fragments, p. 267: ‘Abraham was not heterogeneous with the ethical. He was well able to fulfil it but was prevented from it by something higher, which by absolutely accentuating itself transformed the voice of duty into a temptation.’ Also St Augustine, City of God (Harmondsworth: Penguin, 1996), Book XIV, Ch. 15, p. 575; Book XVI, Ch. 32, p. 694: ‘Abraham‘s obedience is renowned in story as a great thing, and rightly so, because he was ordered to do an act of enormous difficulty, namely to kill his own son‘; ‘Abraham, we can be sure, could never have believed that God delights in human victims; and yet the thunder of a divine command must be obeyed without argument.‘
20Given the many apparent inconsistencies in scripture it might be argued that any consistent interpretation depends, at least tacitly, upon certain prior philosophical commitments. It has been suggested here only that Kant‘s commitments are not religiously acceptable. I have eschewed trying to answer any objections to religious tradition itself here.
21A.W. Wood, Kant‘s Moral Religion (Ithaca: Cornell University Press, 1970), pp. 205–206.
22C.J. Webb, Kant‘s Philosophy of Religion (Oxford: Clarendon Press, 1926), p. 12; M.G. Rearden, Kant as Philosophical Theologian (London: Macmillan, 1988), p. 172. That the consensus is not absolute is evidenced by S. Palmquist in Kant‘s Critical Religion (Aldershot: Ashgate, 2000), pp. 83–4.
23Kant formulates the moral proof of God in many places, for example CPrR 150–58, CPR B 425–6, A 811=B 839-A 815=B 843, A 828=B 856. Fortunately, the argument does not differ significantly from the second Critique to the third (although some of the formulations in the first Critique are perhaps a little too uncomplicated to rely on as being definitive statements).
24D. Henrich, The Unity of Reason: Essays on Kant‘s Philosophy (Cambridge: Harvard University Press, 1994), p. 77.
25Michalson, Kant and The Problem of God, p. 116.
26A. MacIntyre, After Virtue (London: Duckworth, 2000), p. 187.
27It could further be argued at this point that as happiness is enjoyed by man qua phenomenal being, then it follows that Kant cannot postulate any noumenal happiness anyway.
28Y. Yovel, Kant and the Philosophy of History (Princeton: Princeton University Press, 1980), pp. 89–90.
29Michalson, Kant and The Problem of God, p. 21.
30Ibid., p. 34. Michalson sees the moral proof as also being undermined by the fact that in Kant‘s later works on history the theological summum bonum is apparently replaced by that of a moral commonwealth, located in space and time, with the result that God, who played a central role in the summum bonum, is marginalised (Kant and The Problem of God, p. 25). Yet according to M. Despland in Kant on History and Religion (Montreal and London: Queens University Press, 1973), this apparent antinomy need not really concern us because whilst Kant‘s philosophy of history is concerned with the political reform of the state and the league of nations, his philosophical theology takes the prolonged view and centres upon the Kingdom of God as the ultimate religious goal.
31D.M. Mackinnon, ‘Kant‘s Philosophy of Religion’ in Philosophy, April 1975, vol. 50, no. 192, pp. 131–44, at p. 138.
32It might be thought that there is yet another Kantian proof of God: the first Critique tells us that even for the man apparently devoid of right sentiment ‘enough remains to make him fear the existence of a God and a future life … this may therefore serve as negative belief … a powerful check on the outbreak of evil sentiments’ (CPR A 830=B 858). Although the negative inflection of the concept (of belief) here is a recognisably Kantian procedure – one is reminded of Kant‘s concept of negative pleasure in the third Critique‘s discussion of the sublime (CJ 83) and of negative moral perfection in the second (CPrR 189) – it is nevertheless misleading. For negative belief is not actually any kind of belief but is rather an agent acting as if he believed. This issue is further explored, twelve years later, in Religion Within the Limits of Reason Alone. In this work, Kant notes that the recidivist who entertains no hope of moral improvement glimpses an ‘incalculable misery‘; a ‘cursed eternity’ (Rel 63): representations psychologically powerful enough, despite their potential untruth, to serve as an incentive if not to goodness then at least to restraint and God-fearing ‘without our having to presume to lay down dogmatically the objective doctrine that man‘s destiny is an eternity of good or evil’ (Rel 63). It will be noted that neither the account in Religion Within the Limits of Reason Alone nor that which we found earlier in the Critique of Pure Reason declare that fear depends on the malefactor being committed to a belief in God and in the existence of human subjectivity in perpetuity: ‘Nothing more is required than he at least cannot pretend that there is any certainty that there is no such being and no such life’ (CPR A 830=B 858). This ‘proof from fear’ then, on closer inspection, is not a proof after the manner of Pascal at all but rather a form of policing behaviour with the end of getting people to act as if they believed in God. Et Exspecto Resurrectionem Mortuorum Moreover, given Kant‘s rigorism or strict doctrine of the categorical imperative, any deeds done from other motives, like fear, than those of duty can never be classed as moral (see the Groundwork‘s often-made contrast between conformity with duty and from the motive of duty: only the latter is ascribed ‘moral content‘). So it is a peculiarity of this supposed ‘proof’ that it neither commits men to a belief in God nor truly evinces the moral good.