The work of Immanuel Kant follows an arc from his theoretical philosophy to his ethics and finally to his philosophy of religion. That arc is indicated by Kant’s three famous questions for philosophy: “What can I know?” What ought I to do?” and “What may I hope?” (Kant 1787/1929, 635 [A805; B833]).
Kant’s sequence of inquiry suggests that each phase of his thinking was built on the one that preceded it. Thus, the theoretical philosophy sought to understand the relationship between sense experience and human reasoning, and culminated in the discovery of the a priori principles of human cognition. Working from the assumption that human conduct, too, requires an a priori organization, Kant sought the principles governing practical reason and found one in the categorical imperative.
When we turn to Kant’s philosophy of religion, however, it is not immediately clear how his thinking in this area arises from or builds on either the theoretical or practical philosophy. Kant’s deals with religion primarily in the second half or “Dialectic” of the Critique of Practical Reason and in the Religion within the Boundaries of Mere Reason. In these works, Kant argues that practical reason leads us to postulate the existence of a just and compassionate God and the reality of life beyond death. Both concepts, Kant argues, spring from our rational need to unite happiness with virtue. But since Kant had banished happiness from any place in moral reasoning, his philosophy of religion seems not merely discontinuous with his ethics but radically opposed to it.
In what follows, I intend to argue that this is not at all true. Kant’s philosophy of religion is based firmly on his ethical reasoning. Kant’s penetrating understanding of moral reason led him to perceive a serious problem at the heart of rational moral justification. This problem, Kant concluded, could only be resolved by introducing transcendent religious concepts, for which his theoretical philosophy by challenging both religious certainty and empirical dogmatism, had prepared the way. As Kant famously put it in the Critique of Pure Reason, “I had to deny knowledge in order to make room for faith” (Kant 1787/1929, 29 [Bxxx]).
To better understand the logic of Kant’s arguments, we must take a journey into contemporary moral philosophy. I am a student of John Rawls, the great twentieth century American moral philosopher. Although Rawls focused primarily on political philosophy, his most important work, A Theory of Justice, offers a method of moral reasoning that, as Rawls himself acknowledges, closely approximates and even illuminates Kant’s ethics.
To create a just society, according to Rawls, we must imagine the basic principles of that society as being agreed to—literally voted on—by all participants in a purely hypothetical choice situation which he calls “the original position of equality.” In the original position, says Rawls, all individual are permitted to pursue the satisfaction of their desires—their happiness. But, to render the resulting principles fair and not influenced by the accidents of good or bad fortune, all participants in the “original position” must choose those principles from behind “a veil of ignorance” that deprives them of all knowledge of the particular circumstances of their lives: their sex, race, ethnicity, economic position, family background, educational attainments and so on.
Choice under these circumstances leads to a largely egalitarian social order in which basic equal rights and liberties are held by everyone, and in which economic differences are permitted only if, in the longer term, they benefit the least off members of society.
I won't further develop Rawls’s political views and their implications for moral reasoning. For our purposes what is important are the ways in which Rawls’s thinking illuminates Kant’s ethics, a theme about which Rawls has written (Rawls 1971, section 40).1 In brief, both Rawls’s “original position” and his concept of the “veil ignorance” are echoed in Kant’s categorical imperative. That principle states: “Act only in accordance with that maxim through which you can at the same time will that it become a universal law” (Kant 1785/1996, 73 [4:421]). As the very wording indicates, implied here is an imagined legislative conception very similar to that of Rawls. In choosing to act, each of us must assess our implicit policy (our maxim) in terms of its likelihood of acceptance by all other rational agents as a law governing everyone’s conduct.2
It is true that the kind of rational impartiality and objectivity built into Rawls’s original position by the veil of ignorance is not immediately indicated here. It is suggested by many of Kant’s other ideas, including the formulation of the categorical imperative that requires us to respect human and rational nature as an end in itself. But above all it is contained in Kant’s insistence that our particular needs and desires (our personal conceptions of happiness) must not be the governing considerations in our thinking. Kant’s criticism of making happiness the “determining ground” (Bestimmungsgrund) of our willing is not directed at happiness itself. Indeed, Kant states at the very beginning of the second Critique that “to be happy is necessarily the demand of every rational but finite being and therefore an unavoidable determining ground of its faculty of desire” (Kant 1788/1996, 159 [5:25]). But Kant’s great insight, no less than Rawls, is that everyone’s enduring well-being and happiness is predicated on our willingness to put selfish pursuits aside. Within a moral community, no one can make their own personal happiness the supreme consideration governing their willing. That is a prescription for anarchy. For the common good, everyone must subordinate their pursuit of personal happiness to those rules they could advocate under conditions of impartiality.
At this point, however, a profound question and problem of moral reasoning arises. That question is “Why should I be moral?” Why in formulating the rules that govern my life, should I put aside all my self-knowledge, including the knowledge of my strengths and assets, and reduce myself to a shared human condition? Even before Rawls made clear the impartiality requirement in moral reasoning, this “Why Should I be moral?” question has drawn the interest of contemporary rationalist moral philosophers—and has elicited various efforts to answer it.
One entire line of thinking holds that the question itself makes no sense because our very understanding of morality answers it. It is the very nature of morality to trump self-interest. This position is well stated by the philosopher Kurt Baier:
Moralities are systems of principles whose acceptance by everyone as overruling the dictates of self-interest is in the interest of everyone alike . . . .The answer to our question “Why should I be moral?” is therefore as follows. We should be moral because being moral is following rules designed to overrule self-interest whenever it is in the interest of everyone alike that everyone should set aside his interest. (Baier 1958, 314)
Baier here offers what amounts to a general rational justification of the institution of morality and its subordination of self-interest to the interests of everyone. And he is right that it is in everyone’s best interest, taken as a whole, for there to be the institution of morality and for everyone to do their part in sustaining it. But we are not now looking at this matter from the standpoint of “everyone” taken as whole. We are looking at it from our own particular standpoint as someone whose interests are threatened by moral obedience.3
To put this another way, Baier’s answers—and the answers of all who point here to the general usefulness of morality—involves a form of circular reasoning. It is clear that if we look at matters impartially, we would all advocate subordinating self-interest to our collective interests. But when I am in a specific situation where my self-interest is at risk, and I ask, “Why should be moral?” it will not do to tell me that I should look at the matter impartially because it is that very impartial standpoint that I am calling into question.
The question “Why should I be moral?” thus finds us in an unusual and unique situation of rational indetermination. Ordinarily, when I am pulled in different directions by two pressing desires or goals, it is rational for me to try to put aside the immediate attraction of each one, and evaluate the two desires or goals impartially and objectively. This is the essence of reasoning. But in the acute conflict situation when moral reasoning vies with rational self-interest our very means of ordering the dispute breaks down. For as Baier’s inadequate answer shows, resort to impartiality here always favors obeying morality, and it is that very obedience that is under question.
There have been many other attempts to answer the question “Why should I be moral?” A leading one today, bolstered by the authority of science, is that morality is built into our DNA as a result of our evolution in collaborative groups. Simply put, those of us who did not feel and heed the tug of duty were eventually excluded from group protections and failed to survive. However, while such answers may offer an explanation of why we experience feelings of moral obligation, they will not answer the justificatory question of why it is rational for me to privilege such obligation. After all, there are many built-in evolutionary impulses such as my attraction to high calorie foods that I rationally decide to resist when it is not in my best interests to act on them. Why should this not apply as well to the impulse to morality?
Thus we can say that contemporary moral theory has exposed a profound problem for moral reasoning. While it has made the logic of morality clearer than ever before, it has also heightened the challenge of justifying an individual’s moral obedience.
This brings us back to Kant, who I believe, perceived this problem at the heart of moral reasoning more clearly than any philosopher before him. As soon as he had completed his analysis of the logic of practical reason in the first book or “Analytic” of Critique of Practical Reason,Kant turned in the second book or “Dialectic” to the concept of the highest good. He defines this as involving a state of the world in which virtue and happiness are possessed by each person and in which “happiness [is] distributed in exact proportion to morality (as the worth of a person and his worthiness to be happy)” (Kant 1788/1996, 229 [5:110]. Where the highest good is realized, each person holds moral obedience as his or her top priority, but is also able to experience happiness in direct proportion to that commitment.
We may ask why Kant, after having put happiness aside as an object of moral reason not only reintroduces it here, but would have it in exact proportion to virtue, with there not being a single instance of virtue going without its reward.
The answer, I believe, has to do with the problem of individual moral justification. We have seen that no indisputable rational answer can be given to the individual who, when self-interest is at risk, asks, “Why should I be moral.” This puts each moral agent in a quandary. He or she may be deeply inclined to doing what moral reason bids, but whichever way he or she proceeds, reason in the form of morality or prudential reason issues a commanding “No.” If the individual chooses to act immorally, one’s rational conscience will offer its unavoidable condemnation. But even in acting morally, self-interest in the form of rational prudence will raise its objection: “Aren’t you being foolish to sacrifice yourself and those you love in the name of impartial morality?” This quandary can lead to moral paralysis, or worse, choice in a direction that results in inevitable self-blame and remorse.
None of this would be true, of course, if morality and personal happiness could never diverge, if every moral act, however seemingly self-sacrificial, led to a proportionate measure of personal well being. Note, too, that the correspondence must be exact. It is not enough here to argue, as a Baier might do, that morality is generally conducive to everyone’s well being, because only the absolute confidence that one’s own well being will not be sacrificed can prevent a conflict between moral and prudential reason. But this exact correspondence between virtue and happiness is what the highest good means, and Kant has introduced it, I believe, precisely to address the question and problem we have identified.
Now, however, the response to the question of moral justification is raised to a still higher level. If the state of affairs signaled by the concept of the highest good is attainable, our problem is resolved. But is the highest good possible? Is it possible for moral conduct in the world to reach a state where there is not a single instance in which moral agents will see a discrepancy between their moral choices and the happiness they seek and deserve? Significantly, Kant’s argument now turns to this question.
Before doing so, however, Kant is compelled to address an answer to the question of the possibility of the highest good, which he judges to be misleading and inadequate. This is the answer offered by the Greek schools of philosophy, notably the Stoic and Epicurean schools. Despite their differences, both schools saw an intrinsic linkage between virtue and happiness. In Kant’s terms, both saw the virtue and happiness as analytically related, with the concept of virtue necessarily implying proportionate happiness. They did so by focusing on the “self-contentment” (Selbstzufriedenheit) or “peace of mind” (Seelenruhe) that they believed necessarily accompanied all moral willing. When a person resists impulsive and selfish desires and does the right thing, both schools taught, he or she experiences a state of calm well being—contentment—which is virtue’s own reward. Thus, morality may sometimes appear to require the sacrifice of one’s happiness in the relinquishment of immediate satisfactions, but it more than pays back these sacrifices with a form of well-being that is higher, purer, and more enduring.
But Kant will have none of this. In a telling paragraph at this point in the second Critique he characterizes this whole argument as involving a kind of optical illusion, which confuses happiness, which is the object of our natural impulses and desires, with the pleasure we obtain from the exercise of reason whenever it disciplines those same impulses and desires. But it is happiness at which our reasoned willing aims, and it cannot be replaced by a state of mind merely associated with its pursuit.
In addition to this, in a passage at this juncture in the Critique Kant points out the circularity involved in taking the contentment associated with moral action as morality’s reward:
If a human being is virtuous he will certainly not enjoy life without being conscious of his uprightness in every action . . . but in order to make him virtuous in the first place, and so before he he esteems the moral worth of his existence so highly, can one commend to him the peace of mind [Seelenruhe] that will arise from the consciousness of an uprighteness for which he as yet has no sense? (Kant 1788/1996, 233 [5:116])
Here Kant is criticizing essentially the same kind of circular reasoning that we find in many contemporary efforts to answer the question “Why should I be moral?” In all these cases the individual is urged to answer this question by taking an impartial standpoint—here it is the standpoint of our approving moral conscience—the adoption of which is the very thing being questioned.
The failure of such analytic answers to our question leads Kant to assert that the union of virtue and happiness is not definitional in this way, but synthetic: it involves a real causal relation. In our possible experience, virtue must actually be rewarded with proportionate happiness.
But how is this possible? This returns us to the question of the possibility of the highest good, and it brings us directly to the center of Kant’s philosophy of religion. At this point in the second Critique, Kant introduces two “postulates” of practical reason. These are beliefs whose possibility must be affirmed if we wish to provide coherence to the whole structure of practical reason. The first of these is the postulate of the immortality of the soul, which Kant believes is needed to ensure eventual attainment of the degree of virtue associated with the highest good. The second is the postulate of the existence of God, understood as the supreme cause of nature who is able to proportion happiness to individuals in keeping with the worth of their moral disposition.
I will not go into these postulates at length. My focus is on the moral challenges that underlie Kant’s philosophy of religion rather than the outlines of his religious beliefs. The first postulate, in any case, appears to have had a rather short lived presence in Kant’s thinking since it is replaced by a focus on divine grace in the Religion within the Boundaries of Mere Reason after Kant had discovered the problem of radical evil. The second postulate is the familiar idea of a righteous, omnipotent, and omniscient God who can bend nature, which seems radically indifferent to morality, to God’s moral will.
Kant’s introduction of the second postulate as an aspect of practical reason is sometimes called his “moral proof of the existence of God,” but nothing could be further from the truth. Not only is this not a proof, because Kant’s epistemology rules out either theoretical or experiential demonstrations of truths beyond our world of sense experience, but also, and more directly, because the beliefs associated with these postulates in themselves are in no way rationally required. Near the end of the Critique of Practical Reason, Kant describes the highest good and the transcendent beliefs associated with it as “a voluntary determination of our judgment” arising from our moral disposition. He adds, “It can therefore often waver even in the well-disposed but can never fall into unbelief” (Kant 1788/1996, 257 [5:146].
What Kant is telling us, I think—and what the whole argument in the “Dialectic” of the second Critique is trying to help us understand—is that human practical reason is essentially and unavoidably in conflict with itself. The commands of prudential reason—the form of reason that aims at securing our happiness—and the commands of moral reason—the reason that governs everyone’s pursuit of happiness—appear inevitably to clash. If we wish, we can choose to live with this. We can freely heed just one side of reason, prudence or morality, and live with condemnation from the other side. But, we have another choice. Although we can never opt for prudence without incurring moral self-blame, we can choose to heed the voice of conscience and adopt those transcendent religious beliefs whose possible reality silences all complaints from the side of prudential reason.
We can see therefore that Kant is not proving anything to anyone who rejects belief in God or a commitment to morality, nor is he saying that we are rationally required to be moral and accept the religious postulates. Rather, he is developing the underlying rational assumptions of those who have already chosen to commit to morality and who seek to rationally understand and justify that commitment to themselves. The religious beliefs identified by Kant are best thought of as the conceptual underpinnings of a free commitment to both morality and rationality.
Kant is keenly aware that belief in the truth of the postulates takes us beyond the accustomed sphere of our cognition, which relies on phenomenal experience to gain knowledge of the world. Indeed, in the first Critique he had demolished transcendental religious proofs that involved flights of thought into realms beyond our possible experience. Thus, in the closing pages of the second Critique he addresses a final major question by asking whether it is rationally allowable to entertain beliefs beyond the reach of theoretical knowledge. Can we rationally accept the beliefs associated with a moral faith even when these beliefs receive no support from our experiential knowledge?
Answering this question, Kant points out that every function of reason has an “interest.” The interest of theoretical reason—or as he calls it here, “speculative reason”— consists in restricting “speculative mischief” and “rejecting as empty subtle reasoning everything that cannot accredit its objective reality by manifest examples to be shown in experience” (Kant 1788/1996, 237 [5:120]). But, says Kant, practical reason, too, has an interest. This involves “the determination of the will with respect to the final and complete end” (Kant 1788/1996, 237 [5:120]). This final and complete end, we know, is the concept of the highest good, which entails holding to the possible truth of the transcendent religious postulates.
Can speculative reason then prohibit the holding of beliefs that go beyond its warrant? “No,” Kant firmly replies. If what was involved here were mere private wishes and beliefs, speculative flights would not be allowed. But in cases where the very viability of practical reason is at stake, practical reason must take priority, since, in Kant’s words, “all interest is ultimately practical and even that of speculative reason is only conditional and is complete in practical use alone” (Kant 1788/1996, 238 [5:121]).
What Kant is correctly telling us here, I think, is that our very lives as member of communities of rational beings rest on our respect for the moral law. While it is important for our existence that we preserve the ordered pursuit of knowledge, including sound scientific and philosophical inquiry, even these activities depend on respect for morality. Indeed, a philosophy or a science unfettered by moral restraint would be worse than no such knowledge at all. So if certain religious beliefs going beyond the reach of scientific or empirical proof or disproof (but not contradicting certain knowledge) are needed to support the moral life, then these beliefs are rationally allowable, and speculative reason has the duty of trying to knit them up with everything else it knows.
This concludes my exposition of the main outlines of Kant’s religious arguments in the second Critique. I could stop here since I believe I have explained how these arguments rest on a keen understanding of the challenges posed by commitment to reasoned morality, insights that Kant came to only after developing a full understanding of morality’s rational basis. But I want to go a bit further now and show how these insights, once again in conjunction with a penetrating understanding of the nature of moral reasoning, led Kant to adopt several other beliefs from our biblically-derived religious traditions, notably a belief in the radical imperfection of human moral willing and our need for completing the moral project with the support of divine grace. The issues here are so complex, that I can only sketch some of Kant’s chief arguments.
I believe that when Kant undertook the project of his critical philosophy late in life, he did not see all the conclusions to which his work would lead. This is especially true where the practical philosophy is concerned. Thus, I suspect that when he wrote the Foundations of the Metaphysics of Morals, where almost no mention is made of the religious underpinnings of morality, he did not yet fully understand the logic of the religious positions he only sensed lay ahead. Even more so, when finishing the Critique of Practical Reason, I believe he did not yet see how his arguments there would undermine our ability to achieve the moral worthiness required by the concept of the highest good, and would lead to his rediscovery of the problem of human sin in the Religion within the Boundaries of Mere Reason.
This latter problem has its start in the insight that human beings are rationally free either to accept or reject obedience to the moral law. I say “rationally free” because, as the argument I have outlined reveals, our freedom to be immoral extends to the exercise of reason itself. It is not just that instinct and desire sometimes overpower reason, as the Greek moral philosophers believed. Rather, reason itself succumbs to selfishness because reason cannot unequivocally justify its moral commands. As we have seen, in instances of difficult moral choice reason’s two employments, prudence and morality, can conflict and there is no resolution of the dispute by appeal to a third or higher exercise of reason that does not involve circularity. We can be moral and rational, but as Kant’s arguments show, doing so requires resort to religious beliefs that we are not rationally required to hold. Thus, reason itself is implicated in wrongdoing. We are free—radically free as rational beings—to accept or reject morality, to heed the voice of conscience or to ignore it and selfishly pursue our personal happiness.
There is no hint of these insights in the “Dialectic” of the second Critique. There Kant sees the challenge before the moral individual as one of choosing morality and the religious beliefs needed to support it. Any moral failures along the way are made up for in the unending opportunity for renewed virtue afforded by the postulate of immortality. Sin never enters into the picture. But five years later, with the 1793 publication of the Religion, Kant’s insights have deepened. Now he sees that the degree of human moral freedom that he had discovered in the second Critique opens the way to perpetual human moral transgressions.
Several acute moral insights drive Kant to this conclusion. One is the recognition, just mentioned, that when caught between obeying morality or selfishly pursuing our vital personal interests, we are free—and not just free in the sense that we can impulsively act irrationally, but that we are rationally free to move in either direction. The second insight is the observation that moral reason cannot permit even a single instance of defection from obedience to the moral law. In other words, the categorical imperative is universal not just in its extension in space—we must always take into account the interests of all other moral agents—but in time as well. If we are to regard ourselves as morally worthy, every act of our willing—in our past, present, and future—must evidence our giving priority to duty over self-interest.
This seems odd. Why must I be absolutely good, and why must I commit myself to the unerring choice of duty over self-interest? Isn’t there an acceptable middle ground somewhere between outright selfishness and total commitment to duty, a partial or conditional acceptance of duty? For example, can’t I will to obey the categorical imperative in most cases, except when the most urgent personal needs intrude? On these rare occasions can’t I give myself license to defect from duty? And in doing so, am I not a morally better person than one who never or rarely takes duty into account?
Kant’s answer to these questions, developed in the opening pages of the first book of Religion, is a firm “no.” There is no mid-position between absolute obedience to duty and the outright rejection of duty. Kant explains this in the following words:
The moral law in the judgment of reason, is itself an incentive . . . and whoever makes it his maxim is morally good. Now, if the law fails nevertheless to determine somebody’s free power of choice with respect to an action relating to it, an incentive opposed to it must have influence on the power of choice of the human being in question; and since, by hypothesis, this can only happen because this human being incorporates the incentive (and consequently also the deviation from the moral law) into his maxim (in which case he is an evil human being), it follows that his disposition as regards the moral law is never indifferent (never neither good nor bad). (Kant 1793/1996, 73 [6:24])
What Kant is saying here, I believe, is that any conditional or qualified commitment to the moral law is really no commitment at all, because its ultimate determining ground, the tie-breaker in all cases of conflict, is self-interest. It is true that people who make a conditional commitment to morality may vary in terms of the threshold they set for the point at which self-interest takes priority over duty. But all fail to absolutely prioritize duty above self-interest, and in this respect, all are equally unworthy.
This derivation of the unyielding priority of the moral law—Kant’s so-called moral rigorism— leads him directly to the discovery of universal human sinfulness. For even a single free past immoral choice—and who can say that they have never once freely chosen wrongly—betrays a will not firmly oriented to moral obedience. Looking forward, who can say that their moral dedication is so firm that they will never fail to prioritize morality? From these insights, Kant is led to agree with the biblical conclusion that no morally honest human being can attest to his or her own self-worth. All have fallen short and all must confess the possibility that at its root their will may be morally deficient. Here we see the discovery of the doctrine of radical evil that marks the whole first book of the Religion.
With this discovery, Kant’s philosophy of religion rooted in the concept of the highest good faces a new challenge. In the second Critique, the highest good was imperiled by the possible disconnection of virtue from worldly happiness. But now, it is the integrity of virtue itself that is in question. For if all human beings fall fatally short of virtue how can the human project—or each human individual—achieve the goal of the highest good, and be anything but a moral failure? Above all, how can we rationally commit and recommit to moral striving in the face of the despair that accompanies such moral self-condemnation?
Kant’s answer to these questions involves a concept of divine grace. He introduces this in the Religion and returns to it five years later in his 1798 treatise The Conflict of the Faculties. This concept had never before been suggested in Kant’s writings, and for good reason, for as Kant perceives, the idea of a morally supportive divine grace seems to challenge the very autonomy—free and willed choice—on which morality depends. If my willing is renewed or sustained by God, how can I take credit for my moral accomplishments? How can grace ease my negative self-estimate if it is not me who wills but God?
I am not going to closely examine Kant’s answers to these questions. These answers are so insightful and have generated so much additional commentary in the secondary literature that doing so would require me to deliver another address. Suffice it to say here that in resolving the moral problem of grace, Kant reapplies the basic approach he had developed in the first Critique. There he had shown that the apparent contradiction between our certainty of causal determinism and our experience of human moral freedom can be resolved by an admission that we are incapable of understanding ultimate reality—things in themselves—and that these limits to our cognition also forbid us from dogmatically denying that moral freedom is possible. Similarly, now with regard to grace, he affirms that the limits of our knowledge prevent us from understanding how our moral freedom, in the form of a renewed striving for moral goodness in the wake our own failures, can be compatible with divine assistance, but that the same limits also prevent us from denying that grace and moral freedom can cohere. What is important, Kant concludes, is that we renew and continue our moral striving. A rationally permissible practical faith in God’s gracious goodness makes this rationally possible.
On this note, I will conclude. I have elucidated a dizzying maze of concepts, and I clearly cannot defend every move that Kant makes or my interpretations of them. What I have tried to convey is that the whole edifice of Kant’s philosophy of religion rests on the base of his ethics and the startling new problems and challenges which that ethics revealed. More than two hundred years later, Kant’s understanding of ethics and its challenges continues to merit our keen attention.