Some issues are so important that it is worthwhile to go back and revisit them over and over again – Kant's understanding of ‘rational faith’ is one such issue. It is important for Kant scholarship ‘internally’, but it is equally important because Kant's thought is so often used either to buttress or to debunk various views on what the ‘rational’ person is required to believe concerning ‘God’. Indeed, one recent discussion announced that ‘moral faith’ – that complex of issues involving the highest good, the postulates of God and immortality, and the concept of ‘rational faith’ – is not only one of Kant's ‘central doctrines’, but also that ‘no aspect of Kant's philosophy is more controversial than moral faith’ (Beiser (Reference Beiser and Guyer2006), 588–589). Interestingly, Kant himself admits his awareness that in proposing the concept of ‘a pure practical rational belief’ he has proposed an ‘unusual’ concept which can be misinterpreted (Kant (Reference Kant and Gregor1997), 119 [5:44]).Footnote 1 In this essay, I propose to consider religious belief as Kant portrays it in the Critique of Practical Reason by examining an important vacillation in Kant's formulation of the ‘postulates’ of God and immortality – namely, a vacillation between the formulation that the postulates of pure practical reason postulate the ‘possibility’ of God and immortality and the formulation that the postulates of pure practical reason postulate the ‘existence’ of God and immortality. This vacillation recurs even in Kant's later work, Religion within the Limits of Reason Alone (Kant (Reference Kant, Green and Hudson1960) ).Footnote 2 Despite the numerous writings on Kant's view of the relation between morality and religion, many still entirely ignore this vacillation and of those who do note it, none has attempted to account for the vacillation. I propose that examining this vacillation can reveal important contributions Kant makes to the question of what the rational person is required to believe about God as opposed to permitted to believe. This examination will also address the ambiguity in common claims that the moral agent is ‘committed’ to the existence of God, or must ‘affirm’ God's existence; at the same time it will disagree with those who claim that for Kant the most we achieve is the idea of God as a useful fiction, a merely regulative ‘as if’.
The neglect of or ambivalence about the vacillation in Kant's formulation of the postulates is too prevalent to be documented fully, but here are a few examples. First, there continue to be accounts which simply ignore the vacillation between the postulate of God's ‘existence’ and the postulate of God's ‘possibility’ (which I will from now on refer to as the ‘existence postulate’ and the ‘possibility postulate’), concluding that Kant's ‘moral argument’ is clearly intended by Kant to end in the postulation of (required belief in) God's ‘existence’. Such accounts propose something like the following:
Since reason forces us to think these objects, we accept them as actual. Not to believe would be distrusting reason, it would be rational unbelief. The transcendental deduction of God as a postulate of pure practical reason shows that ‘There is a God’ is a meaningful proposition as well as that reason recommends it as true. It would therefore be irrational, for Kant, not to accept it as true. (Kuehn (Reference Kuehn1985), 167)Footnote 3
I take these conclusions to imply that Kant was trying to defend a rational requirement of belief in the existence of God.Footnote 4 My sense is that even if people no longer argue for (or against) the idea that Kant is offering a ‘proof’ of God's existence, most people still read Kant's intentions as requiring belief in the existence of God, and I will argue that this way stands in need of serious qualification.
Even in cases where commentators charge Kant with failing to ‘justify’ ‘positive belief’ in the existence of God, it is still assumed that Kant wanted to require it. James L. Muyskens, for example, wrote: ‘Kant wishes to argue for a stronger conclusion than the mere possibility of God's existence … the stronger conclusion, namely, that we ought to believe that God actually exists’ (Muyskens (Reference Muyskens1979), 61).
Moreover, there are other readings that seem to me ambiguous on this issue of what is required. For example, in an essay by Andrews Reath which serves as the introduction to the standard Cambridge version of the second Critique, Kant's deduction is presented in a way that leaves unclear whether Kant intended the postulate of the existence of God as a rational requirement. Reath writes:
The only way in which we can conceive of the highest good [a state of affairs it is our duty to promote] as a real possibility is by assuming the immortality of the soul and the existence of God as a moral author of the universe who has ordered the laws of nature to support the possibility of moral ends. Thus a ‘need of pure practical reason’ licenses us to postulate the existence of God and the immortality of the soul (Reath (Reference Reath and Gregor1997), xxviii, my emphasis)
Reath continues that these postulates are ‘rationally grounded beliefs’, concluding as follows: ‘somewhat oversimplifying, Kant thinks that we may assert that God exists as a moral author of the universe … [although this postulate is] only usable for the purpose of a conception of the world that supports the rationality of adopting the highest good as our end’ (ibid., xxx, my emphasis). Reath's claim that for Kant ‘the only way’ to conceive the possibility of the highest good is to postulate the existence of God seems to differ from his claim that ‘we may assert’ that God exists – the latter (unlike the former) suggests that the postulate of the existence of God is not a requirement, although it is rationally permitted. Thus, there is ambiguity about whether for Kant God's existence is actually, according to Reath, a rationally necessary hypothesis or merely a rationally permitted hypothesis.
In his recent summary of Kant's religious thought, Allen W. Wood suggests that, for Kant, ‘practical considerations can justify a belief, at least for the purposes of moral action, that there is a wise, benevolent, and just Providence ordering the world’ (Wood (Reference Wood2005), 180, my emphasis).Footnote 5 He alludes to ‘some subtle variations in Kant's statements of the conclusion of the moral argument, which relate to some serious questions about how strong a conclusion the argument establishes’: he suggests that Kant ‘sometimes seems to agree’ with the conclusion that we are only rationally committed ‘to the possibility that there is a God’ (ibid., 182).Footnote 6 Wood supports this more ‘modest’ position here by reference to the Religion, but not to the statements of possibility in the second Critique. Footnote 7 In any case, even when Wood acknowledges this more ‘modest’ position, he immediately adds that ‘it is equally evident that Kant thinks assent to the actual existence of God is better suited to a properly moral disposition than is this “minimum” ’ (ibid.). It is no longer clear what he takes Kant's ‘postulate’ to be.
In so far as Wood is claiming that there is a vacillation in Kant's thought, he echoes the recognition made earlier by Lewis White Beck. Beck was virtually alone at the time in even noting the presence of two contrasting strands, but still Beck concluded that ‘in spite of this diversity, Kant's meaning and intention in the Critique are clear: the three postulates, officially and deliberately so called, are the actual freedom of the will, the immortality of the soul, and the existence of God’ (Beck (Reference Beck1960), 259). Beck made no attempt to explain the vacillation in the text, nor has anyone else since him.
In what follows, I want to offer a way of accounting for the presence of both formulations of the postulates within the second Critique. I will preface the two main parts of my project by presenting the textual evidence for the contrasting strands on the postulation of God and immortality. Then, in the first part of my project, I will identify the role that the ‘possibility’ postulate plays. I will show how the ‘possibility’ of God is related both to Kant's requirement of the ‘possibility of the highest good’ and to his particular presentation of ‘virtue’ in the deduction of the postulate of immortality, as well as how this postulation of ‘possibility’ is consistent with Kant's understanding both of theoretical reason's limits and of practical reason's achievement. Since the discussion of Kant's notion of ‘pure practical rational faith’ continues to this day to lack any nuanced appreciation of the relevance of the ‘possibility’ postulate in the second Critique, I am considering the issue again (Ferreira (Reference Ferreira1983) ),Footnote 8 hoping to show how an appreciation of the ‘possibility strand’ reveals ambiguities and false dichotomies too often still found in contemporary discussions of ‘moral faith’ and ‘practical belief’, as well as how it is related to Kant's conception of virtue in the second Critique.
The second part of my project will focus on the role the ‘existence’ postulate plays in the second Critique, examining its relation to Kant's concept of moral ‘hope’. I will argue that the vacillating formulations are actually equivalent in so far as a ‘practical belief’ in the existence of God amounts to an active hope in God's existence, and this ‘practical belief’ or ‘hope’ entails a required theoretical belief in the real possibility of God. It might seem obvious that a ‘practical belief’ in God's existence is not a theoretical belief in God's existence, but Kant's claim that ‘it is morally necessary to assume the existence of God’ is too often taken to announce a moral requirement of theoretically ‘believing’ that God exists. That is, even when it is clear that Kant is denying that we can have knowledge of God, too often it is still claimed that Kant required for moral reasons a ‘belief’ in the existence of God; however, when theoretical belief is not clearly distinguished from that unusual creature, ‘practical belief’, such readings are misleading. Moreover, even commentators who recognize the importance of hope in Kant's account often assume false dichotomies in what they contrast with hope.
Preliminary textual evidence
First I shall look at the textual evidence of the contrasting formulations of the postulates of practical reason in the second Critique. If one were simply to begin reading the book from the beginning – i.e. the Preface – one would find in the opening two pages a succinct version of Kant's view of the postulates of God and immortality.Footnote 9 He writes there that the
concepts of God and immortality, which as mere ideas remain without support in the [whole structure of a system of pure reason], attach themselves to this concept [of freedom] and with it and by means of it get stability and objective reality [objektive Realität], that is, their possibility [Möglichkeit] is proved [by the fact that ‘freedom is real’]. (Kant (Reference Kant and Gregor1997), 3 [5: 4])
The emphasis here is Kant's own – ‘objective reality’ means that ‘their possibility is proved’.Footnote 10 He repeats this when he goes on to insist that ‘we cannot affirm that we cognize and have insight into – I do not merely say the reality but even the possibility of them’, but they are ‘the conditions of applying the morally determined will to its object given to it a priori (the highest good). Consequently, their possibility in this practical relation can and must be assumed’ (ibid., 4 [5: 4]).
Kant's emphasis on the ideas of the possibility of God and immortality is found again when he notes that a ‘postulate of pure practical reason’, as opposed to ‘postulates of mathematics’, ‘postulate[s] the possibility of an object itself (God and the immortality of the soul)’ and that the ‘certainty of the postulated possibility is not at all theoretical’, but nevertheless is a ‘necessary hypothesis’ (notwendige Hypothesis) with ‘subjective’, yet ‘unconditional rational necessity’ (ibid., 9 [5: 11n]). A ‘postulate’ is a required hypothesis, and just as what ‘can and must be assumed’ is the possibility of God, so too the ‘necessity’ qualifies only the hypothesis of the possibility of God.
Such claims about the ‘postulates’ of the ‘possibility of God and immortality’ are not confined to the Preface. They are reiterated later in Kant's explicit discussion in the Dialectic of how it is possible to think of ‘An Extension of Pure Reason for Practical Purposes without Thereby Also Extending Its Cognition as Speculative’ (section VII). There Kant writes that the practical needs of reason require ‘presupposing three theoretical concepts’, namely, ‘freedom, immortality, and God’ because for the ‘existence of the highest good possible in a world’ ‘the possibility of those objects of pure speculative reason, the objective reality which the latter could not assure them, is postulated’ (ibid., 112 [5:134], my emphasis). The practical needs of reason establish that ‘those concepts are real and really have their (possible) objects’ (ibid., 112 [5:134], my emphasis).
However, in Kant's explicit deduction in the sections on ‘The Immortality of the Soul as a Postulate of Pure Practical Reason’ (section IV) and ‘The Existence of God as a Postulate of Pure Practical Reason’ (section V), which, interestingly, fall between the above affirmations of the ‘possibility’ postulate, we find him apparently deducing more than the possibility of God from the need for belief in the possibility of the highest good. For example, he writes as follows: the moral law ‘must lead to the supposition of the existence of a cause adequate to this effect, that is, it must postulate the existence [Existenz] of God as belonging necessarily to the possibility of the highest good’ (ibid., 104 [5:124]); ‘the postulate of the possibility of the highest derived good (the best world) is likewise the postulate of the reality [Wirklichkeit] of a highest original good, namely, of the existence [Existenz] of God … the possibility of this highest good … is possible only under the condition of the existence [Dasein] of God … that is, it is morally necessary to assume the existence [Dasein] of God’ (ibid., 105 [5: 125]).
What should we make of these contrasting formulations? Nothing else seems to change – both formulations are found in discussions of the necessary conditions for rational pursuit of the necessary object of a moral will. In so far as a ‘postulate’ is a required hypothesis, the question is what is the sufficient morally required hypothesis: the hypothesis that God's existence is actual, or that God's existence is (in some as yet unspecified way) possible? And, more importantly, why?
The relevance of the possibility strand
The point of the second Critique is to establish a ‘transition to a quite different use of those concepts’ (ibid., 6 [5: 7]), like the concept of God, which were introduced in the first Critique. I suggest that a reconsideration of Kant's understanding of ‘possibility’ can show that once the postulation of ‘possibility’ of God and immortality is understood properly it constitutes a crucial transition achieved by practical reason that is also sufficient for Kant's purposes.
Logical vs. real possibility
First, the ‘possibility postulates’ in the second Critique must be understood in the light of Kant's nuanced understanding of ‘possibility’, which was already anticipated in the Critique of Pure Reason. Footnote 11 Kant distinguishes there between two kinds of ‘possibility’: ‘logical possibility’ and ‘real possibility’. He writes:
I can think whatever I please, provided only that I do not contradict myself, that is, provided my concept is a possible thought. This suffices for the possibility of the concept, even if I may not be able to answer for there being, in the sum of all possibilities, an object corresponding to it. But something more is required before I can ascribe to such a concept objective validity [objektive Gültigkeit], that is, real possibility [reale Möglichkeit]; the former possibility is merely logical [logische]. This something more need not, however, be sought in the theoretical sources of knowledge; it may lie in those that are practical. (Kant (Reference Kant and Smith1965), B xxvi, second emphasis mine)
Kant characterizes ‘merely logical’ possibility simply in terms conceptual non-contradiction, and he characterizes ‘real possibility’ as involving ‘something more’. In this passage he equates ‘real possibility’ with the ‘objective validity’ of a concept and with having a ‘possible object’, which is ‘something more’ than merely having a non-contradictory object. This ‘something more’ has a functional analogue in the ‘practical’ realm.
Kant also claims in the first Critique that it is a ‘necessary logical condition that a concept of the possible must not contain any contradiction; but this is not by any means sufficient to determine the objective reality of the concept, that is, the possibility of such an object as is thought through the concept’ (ibid., A220, my emphasis). He repeats: concepts have ‘objective reality’ when ‘they apply to possible things’ (ibid., A221). He is using interchangeably the categories of ‘real possibility’ and ‘objective reality’. In addition Kant there equates ‘logical possibility’ with the term ‘problematic’ and contrasts it with ‘objective’ possibility: ‘The problematic proposition is therefore that which expresses only logical (which is not objective) possibility – a free choice of admitting such a proposition, and a purely optional admission of it into the understanding’ (ibid., A75; B310).
I suggest that Kant is applying these distinctions he made in the first Critique when he sets about in the second Critique to show that it is not ‘inconsistent’ to ‘deny objective reality to the supersensible use of the categories in speculation and yet grant them this reality with respect to the objects of pure practical reason’ (Kant (Reference Kant and Gregor1997), 5 [5:5]). His goal is to show that ‘practical reason of itself, without any collusion with speculative reason, furnishes reality to a supersensible object of the category of causality, namely, to freedom’ (ibid., 5 [5:6]). Kant says that ‘objective reality is given to the ideas of God and immortality’, and this means for him that ‘their possibility, which was hitherto only a problem, here becomes an assertion’ (ibid., 4 [5:5]); he writes that in the ‘practical postulates’ an ‘object’ is given to a ‘merely problematical thought’ and thus the thought receives ‘objective reality’ (ibid., 112 [5:135]). He explains that the ‘increment’ to the ‘theoretical cognition of reason’ ‘consists only in this: that those concepts, otherwise problematic (merely thinkable) for it, are now declared assertorically to be concepts to which real objects belong … these concepts are real and really have their (possible) objects’, but ‘no synthetic proposition is possible by this reality granted them’ (ibid., 112 [5:134]). Here ‘problematic’ is defined as ‘merely thinkable’ – i.e. logically possible, without contradiction – and the transition from a concept which is ‘problematic (merely thinkable)’ to a concept ‘declared assertorically’ means that these ‘concepts are real and really have their (possible) objects’ (ibid., 12 [5:134]). Similarly, Kant's initial claim in the second Critique – namely, that when ‘objective reality is given to the ideas of God and immortality’, ‘all that is given is that their possibility, which was hitherto a problem, here becomes an assertion’ (ibid., 4 [5: 5]) – does not contrast possibility with assertion, but rather contrasts ‘problematical’ possibility with a real (asserted) possibility. A real or asserted possibility is one that is established as a required possibility, as opposed to an ‘optional’ one.
Kant re-emphasizes this in the section on the ‘extension of pure reason for practical purposes’, when he suggests that in certain cases the stronger possibility not only ‘need not’ but cannot be established by theoretical reason – namely, that the concept of God is given ‘no objective reality’ by theoretical reason, and so is ‘problematic (merely thinkable)’ (ibid., 112 [5: 134]). So, the ‘real possibility’ of God, the ‘objective reality’ of the concept of God, cannot be established by theoretical reason.Footnote 12 But as we saw earlier, such ‘real possibility’ or ‘objective reality’ of the concept of God can, Kant thinks, find grounding in the practical resources of reason.
Kant announces a substantive and distinctive ‘increment’ when he says that these ‘thoughts’ (of God, for example) ‘in which there is nothing impossible’ go on to ‘receive objective reality through an apodictic practical law’; at the same time he notes that this is a limited achievement, since ‘nothing further is accomplished in this by practical reason than that those concepts are real and really have their (possible) objects’ (ibid., 112 [5: 134–5], my emphasis). Kant's locution here is noteworthy – a ‘real’ concept (a concept with ‘objective reality’) is one which ‘really’ has a ‘(possible) object’.
Thus, although the phrase ‘real possibility’ is not found in the second Critique, Kant does use the term ‘objective reality’ and refers to a concept ‘really hav[ing] a possible object’ which he contrasts with what is ‘logically possible’ or ‘merely thinkable’. I take it that Kant's explicit distinction in the second Critique between a possibility that is ‘problematical’ and a possibility that is ‘assertorical’ also tracks the distinction in the first Critique between logical and ‘real’ possibility. Thus, it seems plausible that when Kant refers to the ‘possibility’ postulates in the second Critique, he is referring to ‘real possibility’.
This stronger (‘proven’, ‘established’, ‘assertoric’) kind of possibility that Kant insists on distinguishing from mere non-contradiction is crucial to Kant's attempt to show that practical reason receives an ‘increment’ that goes beyond what is theoretically achievable. But what is this ‘real possibility’ and how is it guaranteed? Kant is not alone in assuming such a distinction. Many would agree that there is intuitively a difference between a possibility guaranteed merely by the form of the words (logical non-contradiction) and a possibility that is further supported by the laws of nature, or the physical world as we know it (as opposed to the world as it could have been).Footnote 13 For Kant, ‘real possibility’ can be generated not only within the physical world as we know it, but also within the moral world as we know it.
Kant characterizes ‘real possibility’ or ‘objective reality’ in two ways – in terms of moral content and in terms of moral necessity. First, he affirms that even when there is no corresponding object of intuition we can say ‘that an object really exists’ so long as ‘a category as a mere form of thought is here not empty but has significance’ (ibid., 113 [5: 136]). In other words, the concept is ‘empty’ when it is merely non-contradictory, but it receives content or ‘significance’ when it gains an ‘object’ (objective content) by being determined to a certain extent by the needs of practical reason.Footnote 14 The needs of morality have, for Kant, a reality that cannot be denied. On his terms a concept gains ‘objective reality’ and an object gains ‘real possibility’ by its connection with the moral enterprise – one could say that the moral law is the stand-in for the sensible intuition that is lacking. The necessary connection to the moral law is as compelling as sensible intuition would be – it provides an otherwise unavailable content.Footnote 15 A logically possible object (the sum of all possible predicates, the first cause, God) changes status when it gains the specific content generated by the requirements of the moral law. Kant insists that some ‘theoretical positions are inseparably connected’ to the a priori principles of practical reason (i.e. the needs of the moral law) (ibid., 100 [5:120]) and this connection determines the concept (gives content to it) in a morally relevant way so that it is no longer merely non-contradictory or ‘empty’.
Second, when something which is logically possible is also integrally related to what is as ‘real’ as the moral law is for Kant, its possibility gains a kind of necessity – it is seen as part of the ‘real’ world in a way that something merely non-contradictory is not. The reality of the moral law and its demand on us confers a kind of reality on logically possible objects ‘inseparably connected’ to the moral law. For Kant, reason ‘must accept them’ since they ‘belong inseparably to the practical interest of pure reason’ (ibid., 101 [5: 121]). As he says, the moral law provides ‘something extra’ to certain logical possibilities, strengthening them by making them no longer optional, but required by a moral agent. This move from ‘empty’ concepts to ‘real’ concepts (concepts with ‘objective reality’) is the achievement of practical reason over theoretical, and it can be seen in terms of both added moral content and added moral necessity.
A ‘really possible’ object is, for Kant, as real as the moral law is real; however, this does not mean that the really possible object exists outside the understanding. Kant insists on the difference between a concept's object (its objective content or reality) and an object in the world when he says that the ‘reality of the concepts’ (ibid., 113 [5: 136]) which is generated by practical needs does not allow existential or synthetic propositions about the objects of those concepts:
it is absolutely impossible to cognize the existence of this being from mere concepts, because every existential proposition – that is, every proposition that says, of a being of which I frame a concept, that it exists – is a synthetic proposition, that is, one by which I go beyond that concept and say more about it than was thought in the concept, namely that to this concept in the understanding there corresponds an object outside the understanding, which it is absolutely impossible to elicit by any inference. (ibid., 115 [5: 139])Footnote 16
The ‘reality of these concepts’ of God and immortality is achieved when it is shown that they have ‘real possibility’; the extension possible to pure reason because of the practical needs of reason is not an extension which requires the affirmation of the theoretical synthetic proposition ‘There is a God.’ When Kant clearly contrasts the ‘concept in the understanding’ with the ‘object outside the understanding’, he is distinguishing between giving ‘reality’ to a concept and claiming that the object of the concept exists ‘outside the understanding’.
In sum, Kant has already implied that ‘something more’ than the mere logical possibility of God and immortality is needed by the practical interest of reason (I shall reconsider this in the following section on ‘Hope’). Without the category of ‘real possibility’ Kant could not argue for a ‘transition to a quite different use’ of the concept of God, a transition that exceeds what theoretical reason could achieve.Footnote 17
Since Kant insists that we get an extension ‘only so far as is directly necessary for pure practical purposes’ (ibid., 88 [5: 106]), he is looking for the minimal proposition sufficient for the practical interest of pure reason. What is at issue is the difference between what Kant thinks is rationally required of the morally committed person and what he thinks may be legitimate but not required. If one says that the theoretical proposition – the synthetic proposition – that God exists is the one Kant required to be believed, one is, in effect, arguing that Kant thought it is irrational not to believe in God if one engages in the moral enterprise. The presence of the ‘real-possibility postulate’ formulation suggests that Kant did not want to make such a strong claim; rather, it suggests that he recognized that the ‘room’ for faith which he sought to provide was a more limited achievement, though still a significant one. We now need to determine whether the belief that God's existence is ‘really possible’ is sufficient for the needs of the moral enterprise.
The postulate's sufficiency for moral purposes
To determine whether belief in the real possibility of God is sufficient for the needs of the moral enterprise, we have to reconsider what Kant is trying to guarantee. At least twenty seven times in the Dialectic he says that he is trying to guarantee the ‘possibility of the highest good’ – no less but no more.Footnote 18 He announces both in the Preface and in the Dialectic that the postulates of God and immortality are deduced from their direct connection with the ‘necessary object’ of a moral will (namely, ‘the highest good’) and he insists in the Dialectic that ‘it is the possibility of the highest good which reason points out to all rational beings as the goal of all their moral wishes’ (ibid., 96 [5: 115]). He justifies his claim as follows: since ‘promotion of the highest good … is an a priori necessary object of our will and inseparably bound up with the moral law, the impossibility of the first must also prove the falsity of the second’ (ibid., 95 [5:114]).Footnote 19 That is, if the highest good were not possible, commitment to the moral endeavour would be irrational. This need not put in jeopardy the validity of the moral law; rather it suggests that, although we would still be bound by the moral law, our commitment to it would be irrational in the sense that our actions would be without purpose (ibid., 119 [5: 143]).Footnote 20 The necessity of guaranteeing the ‘possibility’ of the highest good thus seems to remain a constant for Kant throughout the second Critique,Footnote 21 and I will assume that whenever Kant speaks of the ‘possibility of the highest good’, he means (because of its connection with the moral law) the real possibility of the highest good, rather than its mere logical possibility. (Kant already assumes that the logical possibility of such concepts can be achieved by theoretical reason.)
I should note that the order in which Kant presents his deduction of the postulates in the second Critique seems to separate the components of the highest good, aligning the postulate of immortality with the virtue component of the highest good, and the postulate of God with the happiness component of the highest good. However, they are more closely intertwined than Kant leads us to expect. This is both because already in the deduction of the postulate of immortality we find a reference to the need for God even before we get to the problem of apportioning happiness, and because (although Kant mentions it earlier,Footnote 22 he does not allude to it here) our duty to be virtuous includes promoting the happiness of others. However, I want to suggest that the relationship is even more complex, because the account of virtue assumed in Kant's deduction of the postulate of immortality has implications for the postulate of God, in addition to the one Kant explicitly notes.
Kant's formulation of the ‘highest good’ raises more questions than it answers, but the gist of it is as follows. For Kant, the highest good (in the second Critique, at leastFootnote 23), specifically ‘the highest good of a possible world’, is the ‘whole and complete good’ of ‘our pursuit of happiness’ under the conditions of morality such that the consequence is happiness in ‘exact proportion’ to virtue (ibid., 93, 108 [5:110, 129–30]). Although strictly speaking the ‘exact proportion’ of happiness to virtue could be fulfilled even if only a small amount of each were present, that would not be the ‘highest’ or the ‘whole and complete’ good, so we can assume that the ‘highest good’ is, for Kant, the achievement of the highest virtue and the greatest amount of happiness, causally connected.
What is required for rational commitment to the moral enterprise is that we affirm that the highest good is ‘really possible’. The question remains: what then are we rationally required to believe if we are to believe that such a state is really possible? Keep in mind that it seems plausible to expect that we would be required to believe something less to justify a belief in the real possibility of the highest good than to justify a belief in the actuality or existence or realization of the highest good.
I want to suggest that Kant's actual formulation of the deduction of the postulates in sections IV and V succeeds only in requiring a belief in the real possibility of God and immortality (despite the fact that this is precisely where Kant brings in the idiom of ‘existence’). This is, I will show, because on his own terms there is a necessary connection between God (and immortality) and the highest good, such that (a belief in) the existence of God (and immortality) would entail (belief in) the inevitability or actual existence of the highest good, and so would be more than is required to guarantee the real possibility of the highest good.
This necessary connection between God and immortality and the highest good is revealed in Kant's explicit assumptions about happiness and virtue. With respect to the second component of the highest good (happiness causally related to worthiness to be happy – virtue), Kant assumes that ‘to need happiness, to be also worthy of it, and yet not to participate in it cannot be consistent with the perfect volition of a rational being that would at the same time have all power’ (ibid., 92–3 [5:110]). Moreover, he believes that ‘all possible happiness in the judgment of a wise and all-powerful distributor of it has no restriction other than rational beings’ lack of conformity with their duty’ (ibid., 107 [5: 128]). Thus, it need not introduce heteronomy into morality to believe that if God exists, God necessarily will do what a rational omnipotent being would do to proportion happiness to virtue (ibid., 107 [5: 129]).Footnote 24
With respect to the first component of the highest good, Kant asserts that ‘the complete conformity of dispositions with the moral law is the supreme condition of the highest good’, even though ‘complete conformity’ or ‘holiness’ is ‘a perfection of which no rational being of the sensible world is capable at any moment of his existence’ (ibid., 102 [5: 122]). Rather than deducing that such holiness is not our duty, he concludes that we must assume ‘such a practical progress’ is possible, and so he deduces the ‘immortality of the soul’.Footnote 25 Kant acknowledges that immortality is not sufficient, however, when he adds that ‘the eternal being, to whom the temporal condition is nothing, sees in what is to us an endless series the whole of conformity with the moral law’ and the required ‘holiness’ ‘is to be found whole in a single intellectual intuition of the existence of rational beings’ (ibid., 103 [5: 123]). This is another role for God in relation to immortality (although it should be asked whether this already obviates the need for immortality). In any case, both conditions (i.e. immortality and God) are presented as necessary and sufficient for the possibility of the highest virtue.
I say they are presented as ‘sufficient’ (as well as ‘necessary’) because in the actual ‘deduction’ of the postulate of immortality found in section IV Kant raises no problem about moral ‘progress’ as such. It is remarkable that here (in the section explicitly devoted to the deduction of the postulates) Kant presents the task of fulfilling the virtue component of the highest good simply as the problem of achieving an ‘ideal’ of ‘holiness’ in a limited time. As long as we couple endless ‘duration’ (ibid., 110 [5: 133]) with a God who takes our virtuous efforts in at a glance, the highest virtue is inevitable. This optimistic account of unproblematical moral dedication is also assumed in Kant's earlier claim that ‘the utmost that finite practical reason can effect is to make sure of this unending progress of one's maxims toward this model [of holiness] and of their constancy in continual progress’ (ibid., 30 [5:33]).Footnote 26
Whereas Kant has required only that the highest good be conceived as really possible, on such an optimistic account of virtue any belief in the existence of God and immortality entails a belief that the highest good will necessarily be realized. That is, if immortality exists, there will be unending progress in virtue and if God exists, God necessarily takes things in at a glance (atemporally), so the highest virtue component is guaranteed; moreover, if God exists, God will appropriately apportion happiness to virtue. So, if we believe that God exists, we must also believe that the highest good as a whole is guaranteed (rather than ‘possible’).
It is noteworthy that Kant's account in section IV of the inevitability of the highest virtue component of the highest good seems at odds with his later account in the Religion Within the Limits of Reason (where he raises the problem of ‘radical evil’) – but there is also a tension within the second Critique itself in this respect. There are at least five references in the second Critique which while seemingly made in passing nevertheless reveal that the radical evil Kant explored in the Religion was not something new; it was something he had already thought about. He repeatedly mentions the idea of ‘the transgression’ of a moral law (ibid., 34, 70, 79, 122 [5: 37, 82, 93, 147]). In the course of his discussion of how choice has its basis in free causality, he writes:
these actions, on account of the uniformity of conduct, make knowable a natural connection that does not, however, make the vicious constitution [Bosheit] of the will necessary but is instead the consequence of the evil and unchangeable principles freely adopted, which make it only more culpable and deserving of punishment. (ibid., 84 [5: 100], my emphasis)
In line with this, he acknowledges ‘a continuing propensity to transgression or at least impurity, that is, the admixture of many spurious (not moral) motives to observe the law’ (ibid., 107 [5: 128]). Moreover, he intends to discuss the ‘grounds of that possibility’ of the highest good – ‘first, with respect to what is immediately within our power and then, secondly, in that which is not in our power but which reason presents to us, as the supplement to our inability, for the possibility of the highest good’ (ibid., 100 [5: 119], second emphasis mine).
Given Kant's insistence that ‘To satisfy the categorical command of morality is within everyone's power at all times’ (ibid., 33 [5: 36–37]), these comments in the second Critique are easily overlooked by the reader. More importantly, they seem not to have had any real effect on Kant's deduction of the postulates of God and immortality in the second Critique, since he does not bring their subject into the explicit deduction at all.
In sum, the explicit deduction of the postulates in sections IV and V expresses Kant's assumption of a necessary connection between God and immortality and the realization of the highest good – and this has the important consequence that it renders belief in the existence of God too strong a condition to support the possibility of the highest good. At the same time, it implies that the real possibility of God and immortality would be sufficient to support belief in the real possibility of the highest good. The real possibility of God and immortality are logically sufficient to guarantee the real possibility of the highest good because standard modal logic suggests that ‘what is necessarily implied by something possible is itself possible’.Footnote 27 This suits perfectly Kant's claim concerning the possibility of God: what is required for my commitment to promote the highest good, which is the necessary end of a morally determined will, is only that I believe that ‘it is not impossible that morality of disposition should have a connection, and indeed a necessary connection, as cause with happiness as effect in the sensible world, if not immediately, yet mediately (by means of an intelligent author of nature)’ (ibid., 96 [5:115], my emphasis).
The maximally legitimate conclusion of an argument that genuinely premises the ‘real possibility’ of the highest good thus coincides with the affirmation of the postulates of ‘real possibility’ of God and immortality actually found in the text. The ‘possibility postulates’ are actually the only legitimate conclusion of an argument which assumes such a necessary connection between God and immortality and the realization of the highest good. The necessary connection between the concepts of God and immortality and the highest good means that the minimal required belief would be belief in the real possibility of God and immortality.
The requirement of a belief in (or a theoretical assumption of) the ‘existence’ of God would be too strong a condition for Kant's purpose of guaranteeing the ‘real possibility of the highest good’, so Kant need not and should not argue for the postulate (the necessary hypothesis) of the existence of God. The requirement of a belief in (or a theoretical assumption of) the real possibility of God would be a strong enough condition to guarantee it.
The ‘possibility postulates’ may thus represent Kant's recognition that a moral argument which genuinely (strictly) premises the real possibility of the highest good rather than its actuality or inevitability will result only in the postulate of the real possibility of God's existence.Footnote 28 But if so, we still need to consider why the language of ‘existence’ is found in the section on the explicit deduction of the postulates of God and immortality.Footnote 29
The relevance of the existence formulation: the ‘hope’ that God Exists
Sections IV and V in the Dialectic provide Kant's explicit deduction of the two postulates: ‘The Immortality of the Soul as a Postulate of Pure Practical Reason’ and ‘The Existence of God as a Postulate of Pure Practical Reason’. It is precisely here that Kant shifts his language from the ‘postulate’ of ‘possibility’ of God to the ‘postulate’ of ‘existence’ of God; he refers to the ‘existence’ of God seventeen times (apparently using Existenz, Dasein, and Wirklichkeit interchangeably). The question is what are we to learn from this particular discussion of the existence of God. Kant is committed to explicating a ‘pure practical rational belief’ and we need to consider more precisely what a ‘practical belief’ is.
It has been noted that Kant usually uses phrases like ‘presuppose’ or ‘assume’ or ‘take for true’ when speaking of the existence of God in these two sections,Footnote 30 and it might be thought that whatever Kant means by assuming the existence of God is a practical activity that doesn't deserve the name ‘belief’. However, it is important that Kant does retain the term ‘postulate’ in the title of both sections (and elsewhere), and makes it a technical term, defining it as follows: a ‘postulate of pure practical reason’ is a ‘theoretical proposition, though one not demonstrable as such, in so far as it is attached inseparably to an a priori unconditionally valid practical law’ (ibid., 102 [5:122]). This suggests that the turn to the idiom of ‘existence’ marks a change from the earlier position concerning the postulate of the possibility of God, since in both cases he is talking about a ‘postulate’. What is the minimal theoretical proposition that must be held by the moral believer – what proposition would it be irrational of the believer not to hold? Has Kant changed his mind about which ‘postulate’ – which theoretical proposition – is required by the rational believer?
‘Practical belief’
Whatever Kant means by a rational ‘practical belief’ is clearly different from what he means by a rational ‘theoretical belief’. The relation between ‘belief’ and the theoretical employment of reason is complicated by several claims Kant makes. He insists that ‘postulates’ are not ‘theoretical dogmas’ but rather ‘presuppositions’ (ibid., 110 [5:132]) – however, he still holds that postulates are ‘theoretical propositions’. They are not ‘dogmas’ or claims to knowledge, but they are theoretical propositions. Kant alerts us, however, to an important change he is making when he speaks about God's existence: he qualifies his claim that ‘it is morally necessary to assume the existence [Dasein] of God’ by insisting immediately after that ‘there can be no duty to assume the existence of anything (since this concerns only the theoretical use of reason)’ (ibid., 105 [5: 125]). He explicitly claims that there is a contrast between a theoretical proposition used theoretically and a theoretical proposition which is not used theoretically (ibid., 105, 112 [5:125,134]). I suggest that once the difference between practical and theoretical belief is fully appreciated, the moral necessity ‘to assume the existence of God’ (ibid., 105 [5: 126]) need be no more than the ‘practical belief’ in the existence of God. However, this ‘practical belief’ has a theoretical dimension in so far as it involves a theoretical proposition. So when we ask whether the ‘postulate’ or ‘theoretical proposition’ that reason requires is the proposition that God exists (actually) or the proposition that God's existence is really possible, we need to remember this strange creature – a theoretical proposition that is assumed but not used theoretically.Footnote 31
So what does it mean that the theoretical proposition of God's existence is not used theoretically, but is held as ‘practical belief’? If ‘unbelief’ is the belief that God does not exist or that it is not possible that God exists, then technically ‘unbelief’ would be ruled out by the requirement of the belief that it is really possible that God exists, and this would mean that the ‘possibility postulates’ achieve Kant's aim of authorizing ‘pure practical rational belief’ so as to discredit the alternative of ‘unbelief’ (ibid., 121 [5: 146]). But is this a sufficient account of the ‘pure practical rational belief’ Kant endorses? Is there something else Kant wants us to learn in this discussion of God's existence?
Hope
I suggest that the idiom of the ‘practical belief’ in God's ‘existence’ also serves another purpose. I propose that the notion of a ‘practical belief’ can be clarified by noting that Kant refers to ‘hope’ repeatedly (at least thirteen times) in the process of explicitly deducing the postulates of God and immortality (ibid., 103–108 [5: 123–130]). Kant is not explicit about the logic or the psychology of ‘hope’, but we can tease out his assumptions.
For Kant, the practical employment of reason consists in ‘determin[ing] maxims of the will’ (ibid., 40 [5: 45]); it involves ‘effects possible through freedom’ or making things ‘real’ (ibid., 50 [5: 57]). The ‘practical’ is what is ‘made real through our will’ (ibid., 95 [5: 113]), so reason in its practical employment can only issue imperatives to behave in particular ways. In this case, this means acting in ways which promote the highest good, or equivalently, ways which amount to ‘acting as if’ God exists. It is important that, for Kant, striving to promote the highest good and ‘acting as if God exists’ are materially equivalent. Such action is ‘practical belief’ in God's existence.
Indeed, this seems to be what Kant alerts us to when he writes that
a need from an absolutely necessary point of view … justifies its presupposition not merely as a permitted hypothesis but as a postulate from a practical point of view; and granted that the pure moral law inflexibly binds everyone as a command (not as a rule of prudence), the upright man may well say: I will that there be a God, that my existence in this world be also an existence in a pure world of the understanding beyond natural connections, and finally that my duration be endless. (ibid., 119 [5:143])
Kant attests to the character of that willing when he immediately goes on: ‘I stand by this … and I will not let this belief be taken from me.’ These comments suggest a passionate and effective (as well as affective) attitude of hoping.
Thus, for Kant, a ‘practical belief’ is acting out a ‘hope’. A ‘hope’ that God exists is a hope that the highest good will be realized, and it is constituted by striving for the highest good. A ‘hope’ that God exists is not a ‘wish’ or mental entity – it is acting in certain ways, it is striving. A ‘hope’ that God exists seems to be for Kant a practical denial of God's impossibility constituted by striving. Hope and striving go together.
This latter point is extremely important for Kant and suggests two important qualifications. First, striving for the highest good involves positive action, often in the face of difficulty – it is not merely refraining from doing something. Striving for the highest good involves the effort to be virtuous and to promote the happiness of others. Sufficient motivation would seem to be lacking from a belief in the mere logical possibility of the highest good. Perhaps Kant's intermittent awareness of the problem of evil even in the second Critique, noted earlier, accounts for his sense that the neutrality of merely logical possibility of the highest good does not seem able to support (either logically or psychologically) the effort involved in striving to fulfil the moral law and promote the happiness of others.
Second, since hoping is a positive striving for the highest good, it requires a certain attitude of believing. One cannot act (hopefully strive) as if God exists while believing that it is false that God exists, or that it is impossible that God exists. In addition, it seems that the passivity of merely not believing in the impossibility of the highest good (and hence, God) would be insufficient for Kant. Not believing that God is impossible is a lack and it cannot sufficiently motivate. If acting as if God exists is, as Kant sees it, a matter of hopefully striving to promote the highest good, one cannot act as if God exists while having no opinion about God (as Wood (Reference Wood1970, 30–31) rightly notes).
Kant's view is in line with a number of philosophical accounts of ‘hope’. These take as criteria of hope the following kinds of things: that the object of hope must be desired by the agent, that the object of hope must be seen as a matter of importance by the agent, that the object of hope must be possible or probable, and that the object of hope be difficult to attain.Footnote 32 Kant's understanding of hope suggests that the kind of possibility at issue is determined by the criteria of positive desire, importance, and difficulty of attainment, which means that mere logical possibility of the object of hope would not be sufficient. Robert S. Downie's discussion of ‘hope’ supports the Kantian view that the ‘bare logical possibility’ or ‘mere logical possibility’ of what is desired does not adequately ground a hope.Footnote 33
Thus, for Kant, one's striving is effectively a ‘hoping’ for the highest good only if one believes that the highest good is ‘really possible’– and the theoretical belief that God really-possibly exists is required by the need to believe in the real possibility of the highest good. The possibility at issue in Kant's postulates is a possibility anchored in an unconditional moral interest which implies hopeful striving. ‘Hope’ correlates with ‘objective reality’ and ‘real possibility’ – not mere logical possibility. The ‘practical belief’ or ‘hope’ that God exists assumes a theoretical belief in the more-than-logical (i.e. the established or ‘proven’) possibility of the object of hope.
If a ‘hope’ (rather than a mere wish) that God exists (including its corollary required theoretical belief in the real possibility of God) is sufficient to account for our need to promote the highest good (for our rational commitment to the moral enterprise with its necessary object of the real possibility of the highest good), then that is all that is shown by Kant to be a requirement of the practical needs of reason.
The consistency of the two formulations
The two contrasting formulations of the practical postulates can thus be consistent with each other: with respect to God, the only required rational belief is either the theoretical belief in the ‘real possibility’ of God or, equivalently, the ‘practical belief’ in the ‘existence’ of God. The ‘practical belief’ that God exists involves the ‘theoretical belief’ that God is really possible. The talk about the ‘existence’ of God no longer concerns the theoretical belief we are required to have, but rather the ‘practical belief’ we are required to have.Footnote 34 If theoretical reason is forced to ‘accept’ a theoretical proposition for theoretical purposes, we have a case of a theoretical belief; if theoretical reason is forced to ‘accept’ a theoretical proposition but not for theoretical purposes, we have a case of a ‘practical belief’. The ‘pure practical rational belief’ that is required of the agent is the ‘ground of a maxim of assent for moral purposes’ (Kant (Reference Kant and Gregor1997), 121 [5:146]), not a theoretical assent to the proposition that God exists.
What Kant rationally requires of the moral person is the ‘assumption that God exists’ – namely, a hope constituted by ‘willing’ with all one's heart that God exists, by acting accordingly. A ‘practical’ belief in God's existence is, on my account of Kant, materially equivalent to acting in ways appropriate to the existence of God, with the crucial qualification that such action constitutes an active, striving, hope that God exists. The necessary condition of such hoping is the required theoretical belief that God is more-than-logically possible because God is integrally involved in a project we are necessarily committed to – but hoping does not require the theoretical belief that God exists. Moreover, this claim about the postulate of the real possibility of God's existence does not in any way deny that Kant's purpose is to propose an ‘objective’ requirement of reason (rather than a subjective ad hominem argument); the necessary hypothesis of the ‘real possibility’ of God would be sufficient to meet Kant's ‘objective’ goal: to render it ‘unavoidable for every rational being in the world to assume what is necessary for its [the highest good's] objective possibility’ (ibid., 119 note [5: 144]).
The language of ‘hope’ is appropriate to the proposition that God exists, not to the proposition that it is really possible that God exists. But, on his own explicit account of the deduction of the postulates, with his assumptions about both virtue and what an omnipotent rational agent would do (ibid., 93 [5:110]), the only postulate Kant reveals as necessary – the theoretical proposition that is needed because of its inseparable connection with our moral interest in striving for the highest good – is the postulate of the real possibility of God.
It can be plausibly argued that in the context of the Preface Kant emphasizes ‘proven possibility’ as the distinctive accomplishment of the practical needs of reason over and above what theoretical reason can accomplish, whereas in the Dialectic (having elaborated at length on the notion of the ‘highest good’) he begins a discussion of religious ‘hope’ in which the discussion of God's existence is appropriate. In other words, although Kant only needed the ‘possibility postulates’ to justify striving for the highest good, he recognized a need to make clear that ‘hope’ in God's existence is an affective and effective action, and so emphasized the affective and practical ramifications of ‘real’ possibility.
Relevance to the contemporary discussion
The foregoing analysis of Kant's account of ‘pure practical rational belief’ in God and immortality has made a case that the apparently opposed formulations of the practical postulates in the second Critique are actually consistent with each other. Kant introduces the category of ‘real possibility’ as absolutely central to his account of practical reason and defends it as anchored in his notion of the moral law (whose reality establishes the more-than-logical possibility of whatever is integrally related to it). Moreover, Kant presents the difficulty of striving for the highest good as requiring a motivation that cannot be sustained merely by belief in the logical possibility of the highest good, God, and immortality. On his own terms, then, once the categories of ‘real possibility’ and active hope are highlighted, the differences in formulations can be construed as being prompted by differences in context. We need to pay attention to whether the proposition is being used theoretically or not. If we assume that a ‘theoretical belief’ is a theoretical proposition used theoretically, then the only required theoretical belief is the belief in the real possibility of God's existence. The talk about the ‘existence’ of God no longer concerns the theoretical belief we are required to have, but rather the ‘practical belief’ we are required to have.
Kant adds to the distinction between ‘possible’ and ‘actual’ and between ‘possible’ and ‘impossible’ the further subdistinction between logical and ‘real’ possibility. His explicit attention to the distinctive category of ‘real (‘assertorical’ or ‘proven’) possibility’ reveals a variety of ambiguities and false dichotomies in many contemporary discussions. All those discussions which pose the problem in terms of the contrast between possibility and impossibility, or between believer and non-believer, are challenged by Kant to consider the category of ‘real possibility’. When commentators read the position of those who recognize ‘possibility’ as believing that ‘perhaps all we need for faith is to acknowledge a lack of flat contradiction with speculative reason’ (Insole (Reference Insole2008), 342), they assume a false dichotomy. Lack of contradiction is clearly not enough for Kant's notion of faith, but belief in the existence of God is not the only alternative left.
Moreover, in their consideration of what is ‘justified’ or ‘rational’ for the religious believer to hold, contemporary discussions often fail to distinguish between what is permitted as opposed to required by the religious believer. Kant makes the contrast between a ‘postulate’ and a ‘permitted hypothesis’ (Kant (Reference Kant and Gregor1997), 119 [5:143]) – a ‘postulate’ is a required hypothesis that cannot be demonstrated theoretically. Commentaries on Kant usually formulate the inquiry in terms of what is ‘justified’ by Kant (Muyskens (Reference Muyskens1979), 56) or what Kant considers ‘rational’, but it is important to recognize that both those terms are ambiguous since both can mean either ‘required’ or ‘permitted’. This distinction between what is rationally legitimate (because theoretical reason cannot preclude it) and what is rationally required (because it would be irrational not to believe it) is an absolutely crucial distinction for discussions about religious faith, and too many commentaries on Kant (and accounts inspired by Kant) fail to clarify the limits and extents of Kant's claims about what we have a duty to believe.
Finally, Kant offers a nuanced account of ‘hope’ which addresses some of the concerns of those who see hope as an alternative to belief as well as those who worry about religious belief being ‘practical’. Those who have rightly appreciated the importance of ‘hope’ in Kant's account of religion have sometimes (wrongly) proposed ‘a shift from belief to hope’ (Muyskens (Reference Muyskens1979), 56) or have assumed the false dichotomy of ‘hope’ versus ‘knowledge’ (Insole (Reference Insole2008), 334). For example, some have argued that when Kant's notion of ‘practical’ belief enshrines the ‘as if’ dimension, it ends up ‘gutting belief of any commitment to truth’ (O'Leary-Hawthorne & Howard-Snyder (Reference O'leary-Hawthorne and Howard-Snyder1996), 244). They have also given the following account of the ‘compromise’ which the radical (‘as if’) Kantian is said to try to forge between the voices of the speculative intellect and the practical intellect: ‘The speculative intellect says “God does not exist” or “It makes no sense to say ‘God exists’; the practical intellect says “There is a God”. The speculative intellect says “I don't believe in God; the practical intellect believes in God” ’ (ibid., 248). Such an account, however, misrepresents Kant in significant ways because it assumes a false dichotomy between a rational required theoretical belief in God's existence and a ‘practical belief’ with no theoretical or motivational implications at all. Such an account neglects Kant's nuanced attention to the theoretical and motivational implications of the notion of acting as hopeful striving, or alternatively, the theoretical and motivational implications of requiring a belief in the ‘real possibility’ of God's ‘existence’. Moreover, it ignores how the qualifier ‘is true’ is part of both the claim that God's existence is actual (is true) and the claim that God's existence is really possible (is true); claims of existence do not have a monopoly on claims to truth. Finally, I am arguing that Kant sees hope for God's existence and the realization of the highest good as required, as opposed to those who see Kant's view as supporting only a ‘right’ to hope (Day (Reference Day1969), 102).
An appreciation of the distinction between logical and real possibility also addresses Wood's suggestion that Kant's conclusion, because of its practical character, may not even count as a ‘belief’ (Wood (Reference Wood2005), 182; Wood (Reference Wood and Guyer1992), 404–405). The attempt to reject ‘theoretical proof’ as an option (as Wood and others rightly do) all too often carries in its train the assumption that there is no ‘theoretical’ accession to reason at all, but this assumption goes against Kant's intentions.Footnote 35 Kant insisted that theoretical reason received an ‘extension’ through the establishment (by the practical needs of reason) of the ‘objective reality’ of those theoretical concepts which could not gain ‘objective reality’ ‘by the theoretical path’ (Kant (Reference Kant and Gregor1997), 112 [5: 134]): ‘Theoretical cognition … of reason, in general, is extended by this [the ‘objective reality’ that practical needs establish] in so far as objects were given to those ideas by the practical postulates, a merely problematic thought having by this means first received objective reality’ (ibid., 112 [5: 135]). There is, he says, ‘an extension of theoretical reason and of its cognition with respect to the supersensible in general’ by the needs of practical reason, although it cannot determine these objects theoretically any further (ibid., 112 [5: 135]). This is compatible with saying that reason ‘has not in the least gained by this’ anything ‘for speculative purposes’ (ibid., 113 [5:136]). The proven or established possibility of these concepts is something that theoretical reason has to ‘accept’ and it is thereby extended: theoretical reason ‘must accept these [theoretical] propositions … as a foreign possession handed over to it’ (ibid., 101 [5: 120]); moreover, Kant says that ‘a subjective condition of reason enters into [the way we represent the achievement of happiness proportioned to virtue], the only way in which it is theoretically possible for it to think’ the possibility of the highest good in a manner conducive to morality (ibid., 120–21 [5: 145]). Wood's conclusion that ‘No philosopher who denies us proofs and evidence of God's existence, as Kant does, can honestly condone an actual belief in God’ (Wood (Reference Wood2005), 183) must, I think, be qualified: it neglects the relevance of the theoretical belief that God's existence is more-than-logically possible, which is a theoretical accession to reason, an actual theoretical belief. Kant's notion of ‘practical belief’, on the contrary, includes a required theoretical belief in God's real possibility and a required hopeful striving (a required practical belief in God's existence).Footnote 36
