Because he is widely regarded as a strict and demanding moralist, and because his political philosophy is not as well-known as his ethics, Immanuel Kant is not as frequently mentioned as a classical liberal as are John Stuart Mill, Adam Smith, and Friedrich Hayek. In this essay, I will argue that he should be, based on a survey of his views of the moral roles of the individual and the state. First, I will explain that the portrayal of Kant’s ethics as cold and rigid is a caricature (much though not all of it due to Ayn Rand), when in fact they represent a commonsense morality, grounded in the dignity of the individual, with a surprising degree of flexibility. Then I will discuss Kant’s view of government, which represents the classical liberal ideal: a minimal state restricted to protecting individuals from wrongful interference from each other and the state, allowing them to pursue their ends with maximal liberty and freedom.

Kant’s Focus on the Individual

The starting point for Kant’s ethics is autonomy.[1] According to Kant, autonomy is the ability to make moral choices without undue regard or deference to either external authority or internal inclinations. Each of us should decide what to do in ethically fraught situations according to our own moral code, not what the government, church, or community tell us to do—unless we agree independently that it is the right thing to do. (For someone living under a dictatorship, as well a powerful church to which he belonged, Kant’s stance reflects well on his courage.) We should likewise not act merely according to our personal preferences or desires, which may pull us in directions opposed to our sense of morality. Both of these conditions ensure that our moral choices align with what we believe to be the right thing to do in our personal conception of what morality demands, in the sense of self-government that the term autonomy implies. To Kant, autonomy is not only an ability but also a responsibility, specifically to live up to our own moral standards without deferring to anybody else’s desires—or to our own.

Of course, no one exercises their autonomy perfectly all the time—none of us is perfectly moral, rational, or strong-willed. Nonetheless, based on our capacity for autonomous choice, every person is imbued with dignity, an unconditional, incalculable, and incomparable worth that not only distinguishes persons from things but also renders each person of equivalent moral value with every other. Again, given the rigidly hierarchical social structures at the time, the concept of the moral equality of all persons that Kant endorsed was very progressive, even if Kant himself did not live up to it in all of his personal opinions.

The autonomy and resulting dignity of each person is the basis for Kant’s categorical imperative, the bane of introductory philosophy students everywhere. There is no need for me to rehearse the details here, but I will mention several aspects of the procedure that generates moral duty. Perhaps the best-known version of the categorical imperative involves universalization: we should act on those plans of action or maxims that we could rationally will everyone follows. While this seems to be an empty exercise in logic—and a notoriously difficult one to put into practice—the reason why we universalize is inherently normative. Because all persons have dignity, we cannot allow ourselves moral license that we could not and would not extend to everybody else. Because no one is more special or worthy than anyone else, we must hold ourselves to the same standards to which we hold others, which is guaranteed by the thought exercise of universalizing our maxims.

The other well-known version of the categorical imperative puts dignity front and center: we must never treat other persons (or ourselves) merely as means to our ends, but always at the same time as valuable ends in themselves. This statement embodies Kant’s distinction between persons and things: things (which lack dignity) may be used simply as means to our ends, but persons (which possess dignity) cannot. This is not to say that we can never use other persons for own ends, but only that we must do so with respect. I use my local baristas to get my coffee fix in the morning, but I am polite (and, on a good day, even charming) and I readily exchange my money according to the terms laid out on the menu. I do not cheat, lie, or force my baristas to give me coffee, as coercion and deceit are the primary ways to treat a person merely as means. Of course, this goes the other way also: my baristas (or their employer) use me to earn wages (or profits), but they do so by offering to sell me a product I like (or I will go elsewhere for it).

Kant wrote that his categorical imperative was nothing but a formalization of the moral intuitions of the common person on the street: treat everyone the same and, if you need their cooperation, convince them to work with you, rather than lying or forcing them to. This is the social context of mutual and reciprocal respect that Kantian ethics requires, and it is very much in line with classical liberal thought.

Defying the Caricature of Kant as a Rigid Moralist

Much of the reason Kant is not widely acknowledged as a classical liberal may be that the duties he promotes seem rigid and demanding. Ayn Rand is well known for accusing Kant for demanding extreme altruism; not only is this is a misunderstanding of Kant’s views of beneficence, but of the nature of  moral duties in general in the Kantian paradigm.

In short, the categorical imperative generates two types of duties. Perfect duties, also called strict duties, allow for no exceptions in one’s own interests, and are normally negative in nature: do not lie, do not steal, do not kill, and so forth. Imperfect duties, or wide duties, do allow for exceptions or latitude because they do not demand certain actions (or abstention from certain actions), but only lay out attitudes that must be maintained and acted on when possible. (In this sense, they resemble how virtue ethicists conceive of virtues and virtuous action; this is no accident, and indeed imperfect duties are often called duties of virtue.[2]) Accordingly, these duties are normally positive in nature: help others, cultivate your talents, and so forth.

If any Kantian duties are rigid, perfect duties are. But note that these duties prevent us from taking actions that fail to respect the dignity of other persons, speaking both directly (in terms of personal impact) and indirectly (by carving out personal exceptions and thereby treating other person as being of lesser worth), so their rigidity is in service of basic classical liberal principles of noninterference. By contrast, imperfect duties are hardly rigid—quite the opposite, as they allow us to choose when and how to act on them. We need not help everyone we possibly can; we should take into account our other duties and responsibilities as well, including those to ourselves, as persons with dignity equivalent to all others.

Kant was very clear about the limited nature of duties of beneficence, saying for instance that we should never give so much that we become in need of help ourselves.[3] Neither are we required to take those actions that require the most sacrifice on our parts—Kant merely said that duties must be practiced for the sake of duty, not self-interest, to reflect a truly moral character. He recognized that we often have mixed motivations for our actions, including duty and self-interest, so we can never be sure we’re acting for the right reasons, although the actions may still be moral.

He also wrote that the duty of beneficence merely requires that we not be completely indifferent to the suffering of others—this is surely no extreme altruism that requites heroic selflessness on the part of agents (as Rand wrote). More generally, this reminds us that technically all Kantian duties are negative in that they derive from a maxim of action being rejected by the categorical imperative. The perfect duties “do not kill” and “do not lie” are clearly negative, but the positive duty of beneficence is actually a duty not to be indifferent to the needs of others, resulting from a rejection of a maxim of indifference by the categorical imperative. As long as we are not indifferent to others’ suffering, and do what we can when we can within our other life circumstances (for the sake of duty), we are moral agents.

When we recognize that all Kantian duties are negative in nature, we see that, when making recommendations for moral action, Kant was much more interested in telling us what we ought not do than what we ought to do. For instance, we must not lie, but we need not be completely forthright either: we are free to prevaricate, change the question, or be silent, as the circumstances dictate, provided we do not deceive outright. By the same token, we must not ignore completely the hardship of our fellow persons, but we are free to decide how and when to offer assistance, given our other commitments.

Abstaining from what you should not is fairly straightforward, but choosing what to do instead is more difficult, and Kant offers little guidance, other than to point to judgment, “a peculiar talent which can be practiced only, and cannot be taught.”[4] There are no rules or formulae to guide judgment, but only one’s intuitive sense of the moral law. Judgment is necessary not only in deciding how to act within our duties, but also to choose which of two or more conflicting obligations is one’s true duty. For instance, if I promised to help my best friend move today, and then my mother calls and needs my help, only my judgment can tell me which obligation to act on: my duty to keep my promise to my friend or my duty to help my mother. Kant wrote only that we must act on whichever has “the strongest ground of obligation,” and we must use judgment to decide which that is.[5]

Given the negative nature of duties and the ubiquity of conflicts among them, our judgment ends up being more useful than the duties themselves, which are just the building blocks of moral behavior. The need for judgment to decide how to implement and balance duties and obligations also has the effect of highlighting the surprising degree of flexibility in Kant’s ethics, where even perfect duties can be overridden by an imperfect duty if the latter has a stronger ground of obligation. This flexibility belies caricatures of Kant as a cold, rigid moralist, and reveal him to be a true supporter of freedom within the bounds of sociality.

Kant’s Minimal Government

One reason Kant’s political philosophy is often overlooked may be that it is nearly inseparable from his moral philosophy, both based on the inviolable dignity of the individual.[6]

Kant holds the state to the same moral standards as he does individuals: all must respect the dignity of persons and not use them merely as means while not at the same time as ends. His binding principle of governance, the Universal Principle of Justice, is almost identical to the universalization formula of the categorical imperative: the state can take only those actions that are consistent with the rightful freedom of all. To Kant, the state only exists to protect the autonomy of persons from wrongful interference from others; in other words, the only legitimate use of state coercion—and, to Kant, a necessary role of the state—is to prevent the coercion of one person by another. The enforcement of the individual’s right to noninterference, a central principle of classical liberalism, is the central role of the state in Kant’s political philosophy.

Furthermore, Kant did not support forced beneficence, either as a matter of virtue or justice. Because beneficence is an imperfect duty, there is no required or “due” amount of help that one person owes another, and therefore there can be no correlative right to assistance. Beneficence, to be truly virtuous, must be given freely, and cannot be coerced on the part of the needy by the state. This rules out any state welfare system, because any government assistance given to some citizens would represent forced beneficence on the part of others. As with many classical liberals, libertarians, and fiscal conservatives, Kant hoped that voluntary charity would be sufficient to help the poor and needy, but the minimal state he describes in his work, much like that of Robert Nozick that followed, does not demand it.

Kant’s minimal system of government, enforcing individuals’ rights of noninterference from each other as well as the state, should be sufficient to assure his reputation as a pre-eminent classical liberal. It only enhances that reputation to recognize that his moral philosophy embodies the same spirit of liberty and freedom, allowing maximal exercise of individuals’ autonomy to pursue their own ends, while helping others when we can, provided we do not interfere with the right of others to do the same. Together, the moral and political philosophy of Immanuel Kant seem to represent the ideal statement of classical liberalism, and I hope this essay contributes to his appreciation in that context.

Mark White thinks that Immanuel Kant has given the “the ideal statement of classical liberalism,” and that this statement is often overlooked both because Kant’s political philosophy is not well known, and because of the impression that his ethics are “cold” and “rigid.” White calls this impression a “caricature,” and he oddly attributes its prevalence to Ayn Rand’s influence. (I say “oddly,” because worries about Kant’s ethics being unduly rigid and cold have been pervasive since his own time, and they’re commonplace today in many circles where Rand has had no appreciable influence; in any case, Rand’s criticisms of Kant, which I’ll come to later, are distinct from the ones against which White defends him.[1]) To counter this impression, White interprets Kant as a “flexible,” “commonsensical” ethicist who valorized the freedom of the individual and, far from preaching radical selflessness, held that individuals have only modest obligations to one another, including most importantly the obligation to respect one another’s autonomy. This obligation stands at the heart of Kant’s political theory, and White describes how it led Kant to oppose the welfare state which, in effect, forces individuals to fulfill (what Kant takes to be) their duty to be charitable.

I am no expert on Kant or on early modern political thought, but my impression is he’s received comparatively little attention as a specifically political thinker because he did not write much specifically on politics, and what he did write is not widely thought to mark a great departure from or advance over the work of such earlier natural rights theorists such as John Locke, Hugo Grotius, and Samuel von Pufendorf. If these thinkers count as classical liberals, then at least insofar as his views are like theirs, Kant counts as a classical liberal. It would be interesting to consider whether the respects in which Kant’s politics differs from these earlier thinkers represent contributions to the classical liberal tradition or departures it, but I leave specifying these differences to those more knowledgeable than I.

White’s statement that Kant’s politics “is nearly inseparable from his moral philosophy” suggests that his primary contribution was not a new specifically political position, but a new ethical foundation for the idea of rights. This seems right to me, but doesn’t Kant already get the credit he deserves? In Anarchy, State, and Utopia, which is surely the most cited 20^th-century work on this subject within academia, Robert Nozick is explicit that the avowedly Kantian “principle that individuals are ends and not merely means” underlies his own defense of a strictly limited government.[2] And the Kantian character of Nozick’s position is often remarked on by sympathetic 21^st-century scholars.[3]

Of course, proponents of contrary political positions have also sought to ground their views in Kant’s moral philosophy. Nozick and John Rawls are often seen as the primary antagonists in late 20^th-century analytic political philosophy, and Rawls’ debt to Kant is acknowledged and unmistakable.[4] So, if the reason for thinking that Kant represents “the ideal statement of classical liberalism” is that his ethics provides the most compelling defense of classical liberal political principles, then, to assess this claim, we’ll need to weigh alternative views of what Kant’s ethics really implies. (I’ll only have a little to say about that below, but I hope to say more in later replies.)

More importantly, we’ll need to think about what we mean by “classical liberalism.” Is there really a single, determinate position here, such that we can ask whether it is implied by a certain moral theory? As I understand the term, it primarily denotes a tradition or movement that emphasizes freedom as a political value and that stretches from the 17^th-century natural rights philosophers and continues through 19^th-century political thinkers like John Stuart Mill, but doesn’t include the proponents of socialism or of the welfare state who came to be called “liberals” in 20^th-century America. Rather than being united by some clear-cut principle, these thinkers are united by shared interests; by relations of influence; by institutional affiliations (e.g. with political parties); and by related goals, allegiances, and adversaries that shifted gradually from one generation to the next, as classical liberalism faded into modern liberalism.

The same points apply to the sundry 20^th- and 21^st-century thinkers who (sometimes unwillingly) get labeled “classical liberals” or “libertarians.” This lot includes, inter alia, Rand and Nozick, who support a state with the sole function of protecting rights; Rothbard and other anarcho-capitalists who oppose the state altogether; and Hayek, Friedman, and the recent “bleeding heart libertarians” who embrace aspects of the welfare state. And it includes the four of us who have been invited to take part in this issue of Cato Unbound. Perhaps this will be a controversial claim, but I don’t think that there is any principle that unites all of us to whom these labels are applied, only a loose set of overlapping interests and influences. This sort of loose shared context often forms an excellent basis for conversations; and, especially, in the context of a culture where all our views are outside the intellectual mainstream, I am glad that there are organizations like Cato to facilitate such conversations. But I think it’s a mistake to conduct these conversations on the premise that we are fundamentally ideological allies, and I worry that this premise is presupposed in the question of whether Kant (or Locke or Mill or whomever) represents the best statement of classical liberalism.  I prefer to engage one another and the philosophers of the past with the goal of better understanding of the similarities and differences between our positions and the relative importance of these similarities and differences. This requires keeping in mind throughout that we may turn out to be adversaries with respect to the issues that turn out to matter most.

Towards that end, let me sketch briefly and without much argument my non-expert understanding on the classical liberal tradition, and how Kant’s ethics relates to it.[5] I see the liberalism of Locke and other 17^th-century thinkers as the political expression of a broader movement, the Enlightenment, that valorized reason and the individual’s pursuit of happiness on earth. Jonathan Israel quotes a 17^th century cleric who deplored this attempt to base politics on “life in this world.”[6] The Enlightenment flew in the face of the common sense of the time, according to which we are inherently sinful creatures who owe obedience to a God, in whom we must have faith, and to the clergy and kings who are His representatives on earth.

Kant coopted some of the Enlightenment’s language and used it to defend a purified form of the dogmas that had long been accepted as common sense but were newly under attack.  In particular, by defining morality in contradistinction to prudence, Kant gave a new prominence to the idea that morality requires sacrifice. The “freedom” (or “autonomy”) he extols is not the Enlightenment’s freedom to conceive and pursue ambitious, life-affirming goals, nor is it the freedom to follow one’s whims. Rather, it is the ability to obey a morality the entire content of which Kant derives from the notion that there must be something for the sake of which one must be always ready to sacrifice the whole of one’s happiness. Though a softer face is often put on it by present-day Kantians, this point is crucial to his derivation of the first formulation of the categorical imperative.[7]

It is this observation about the structure of Kant’s position, rather any concern about rigidity or heroic amounts of charity, that is the essence of Rand’s objection to Kantian ethics.[8] She recognized that Kantian ethics is flexible in many of the ways White describes, and she did not consider it a point in Kant’s favor.[9] (This is as good a spot as any to mention one interpretive claim White makes about Kant’s ethics that was new to me: he writes of “the surprising degree of flexibility in Kant’s ethics, where even perfect duties can be overridden by an imperfect duty if the latter has a stronger ground of obligation.” It would certainly be surprising if Kant thought this, especially given his insistence that motives of philanthropy give us no right to lie to murderers who inquire after the locations of their intended victims.[10] I’d be interested to hear more about which texts support this interpretation.)

This Kantian conception of morality as something essentially different from prudence was widely adopted by subsequent ethicists, who disagreed with Kant on specific content of morality. This includes Mill and other members of the classical liberal tradition, and it was one of a few related causes for this tradition’s gradually departing from its pre-Kantian roots. For Locke, rights defined the freedoms a rational and industrious person needs to pursue his happiness in the world. The Kantian conception of morality, which accords no value to this pursuit as such, leaves these rights with no foundation. Appeals to Kantian dignity are of no help, because a person has this dignity only qua being able to act dutifully, and acting dutifully has little to do with charting one’s course through the world or supporting oneself in it. How is one’s ability to be dutiful infringed upon by, for example, a political system that redistributes money? Perhaps the system eliminates some opportunities for dutiful action and creates others, but it needn’t treat anyone’s will (i.e., duty-following-faculty) as a means to anyone else’s (non-duty-based) ends. Kant himself may have regarded such a system as an affront to dignity, but he also thought this about almost all sex acts. What reason does a present-day Kantian have to follow Kant in the former case but not the latter?

By creating a gulf between morality and prudence, Kant undercut the moral basis for Lockean individual rights, and put in its place a moral framework that entails a different sort of society. Subsequent 19^th- and early 20^th-century liberals discovered much about the mechanisms of a market economy, and some of them did a great deal to extend freedom to women and racial minorities, but with regard to liberalism’s basic ideological orientation, I think the whole post-Kantian liberal tradition represents a series of steps away from a defense of genuine freedom. I expect that some readers and contributors will disagree, both because they understand the historical figures differently and, perhaps, because we have different views of what freedom is and what sort of society it requires.

Notes

The answer is complicated. Any reading of Kant’s massive corpus of writings initially gives one a mass of conflicting liberal and anti-liberal claims. 

He did argue for a robust freedom of speech, at least for the philosophical use of reason. He advocated property rights. He resisted some types of commodifications of human beings. And on numerous matters in foreign policy he called for more peace-oriented and humanistic policies.

Yet he also argued against allowing the smallpox vaccination: “So that states do not become overcrowded with people and thus stifled from the outset, two evils are placed in them as antidotes: smallpox and war.” More trivially, he was against allowing individuals to sell their hair to wigmakers (although buying wigs was allowable; Kant wore wigs). More seriously, he believed it was not a crime for an unmarried mother to kill her child, since bastards are illegitimately brought into the polity.[1] He complained that “a prolonged peace favours the predominance of a mere commercial spirit, and with it a debasing self-interest, cowardice, and effeminacy, and tends to degrade the character of the nation”[2] (note the “mere commercial spirit” barb, which has implications for Kantian economic policy). He held that the important teachings in educational institutions should be subject to the sanction of the state. He was ambivalent about the separation of church and state and allowed for the censorship of some church teachings. And he denied strenuously the right of individuals to rebel against evil rulers, no matter how tyrannical.[3]

Also, given Kant’s many negative remarks about Jews,[4] women,[5] and blacks and other races,[6] it is far from clear that he believes the majority of human beings are capable of agency at a level worthy of moral dignity. Of course, those negative remarks can be bracketed in Kant’s system as mere anthropology and discounted to lower status than his philosophical pronouncements. But the remarks do undercut his connection to the classically liberal ideal of universal freedoms or rights.

Still, we have a puzzle. Many subsequent liberal thinkers have been energized by Kant’s formulations of principles of respect for human dignity and treating individuals as ends in themselves. But they have also been shocked by his frequent anti-liberal principles and policies. Were the anti-liberal deviations merely the result of his being a man of his times, as he was born and raised in a strict Pietist Lutheran family in 1700s Prussia, after all? Was he simply inconsistent, since he wrote on thousands of topics across many decades? Is there some complicated philosopher’s way of making his whole corpus more or less consistent that deep scholarship can reveal?

I will argue that the anti-liberalism is much deeper in Kant’s philosophy than the liberalism. That means saying something about the ringlingly liberal-sounding principles that are indeed integral to Kant’s philosophy. That something is this: One must always interpret a comprehensive philosopher’s remarks on applied matters in the context of his philosophical system.

Let me give an exaggerated, fictional example to stress this point about method. Suppose one reads Chapter One of a philosopher’s book in which one finds affirmations that individuals should seek freedom and realize their true selves and that only in liberty can one find dignity and so on. In Chapter Two, one finds the philosopher arguing that one’s body is not one’s true self; rather one has an immaterial soul that is one’s real Self and that it comes to full actualization only upon separating from this physical realm. Chapter Three argues that when one’s true self is actualized, it does so by merging one’s self into an uber-Self of souls that are collectively real, and that one’s highest moral obligation is to achieve such a merger. As one works through Chapter Four, one learns that the philosopher believes that back in this material world the official state/church institutions are the temporal embodiment of one’s collectively true Self. Finally, in Chapter Five one reads the philosopher’s perfectly logical conclusion that individual freedom means obeying the state/church’s orders.

The example is fictional,[7] but it makes the point about interpretation: What some philosophers mean by self, individual, and freedom is the opposite of what other philosophers mean. So before settling upon an interpretation of a philosopher’s normative claims one always must investigate his or her more fundamental claims in metaphysics and epistemology. Many of us begin our study of Kant by reading his Groundwork as undergraduates, and we may interpret his formulations of the Categorical Imperative as being continuous with the liberal individualist tradition coming out of John Locke, Voltaire, and Adam Smith. But when we pursue Kant’s big books—most importantly the three Critiques—we find ourselves in an entirely different philosophical universe.  

And especially in the case of Kant, incorporating his fundamental distinction between phenomenal and noumenal realms is essential.

Kant argues that we experience only the phenomenal world with its features of time, space, cause and effect. This world is bounded, finite, and Isaac-Newton-physically deterministic. Yet beyond the phenomenal world is the noumenal, which is perhaps the realm of God, freedom, and immortality.

Kant also believes that if there is to be morality, it must come from a place of freedom. Consequently, that freedom must be outside of the phenomenal world, since the phenomenal is ruled by deterministic cause and effect in space and time. In other words, since morality depends upon freedom and the phenomenal realm is unfree, the origin of morality can only be noumenal.

The human self must then be a microcosm of the phenomenal/noumenal split. We are of course phenomenal beings, subject to natural law. But to be moral beings, an aspect of us must be free. We must have a noumenal self that is not subject to natural laws.[8]

So when Kant says that human selves are moral agents, that they have a capacity for dignity, and that their freedoms ought to be respected, he is speaking only about noumenal selves. He is not speaking about our phenomenal selves. Our phenomenal selves are not free and are consequently not in the realm of morality.

Further: Within Kant scholarship there is controversy over whether Kant meant for us to believe that noumenal selves really exist in a noumenal realm or whether he is merely positing them as regulative ideas. A regulative idea, in Kant’s philosophy, is one that we cannot experience or verify empirically but which our reason must posit if it is to make sense of some important philosophical project such as knowledge and morality.[9] The regulative-idea interpretation has traction, since we learn from Kant’s first Critique that any knowledge of the noumenal is totally closed off to us. So we must accept that a noumenal self may or may not actually exist—it may be merely a regulative idea that we posit philosophically in order to make sense of morality, if morality actually is possible.

In either case, freedom is a defining feature of the noumenal self, and bondage is a defining feature of the phenomenal self. And if only my noumenal self is free, and freedom is required for moral agency and responsibility, then my phenomenal self is not a morally responsible agent.

But it is precisely one’s phenomenal self that is one’s lived self—the self that makes a living through physical labor and commerce, that has sex and bears children, that consumes food and wears clothes, that listens to music and works out. That self, in the Kantian ontology, has no freedom at all. So it has no moral agency. And, Kant argues, it is therefore properly subject to a supposed metaphysical nugget—i.e., a noumenal self—in another dimension that issues commands to it.

Where does this leave the Kantian case for liberalism? To conclude this line of argument bluntly:

If our case for political liberty ultimately depends on following orders from a hypothetical voice from another realm, then we don’t have much of a case.

My recommendation: If we are looking for a principled philosophical justification of liberalism, then bypassing Kant is a good idea.  

Notes

Since it’s possible for thinkers to make mistakes in drawing practical implications from their own principles, the question “how good a classical liberal is Kant?” must be distinguished from the question “how good a foundation for classical liberalism does Kant’s philosophy provide?”

It might turn out, for example, that while Kant himself defends classical liberal policies, the more basic principles to which he appeals in doing so actually have far more illiberal implications. (This is essentially the Randian critique to which Mark White alludes.) Or it might be the other way around: that Kant’s principles actually support a far more liberal approach to politics than the one he ends up defending.

My own answers to those questions are complicated. I think Kant’s political writings have strong classical liberal, even libertarian strands, and that these stem in large part from his more basic principles. I also think his political writings contain deeply illiberal elements. In many cases, I take these to be the result of Kant’s misapplication of his own principles, and would expect a more consistent application of those principles to result in a still more thoroughgoingly libertarian set of policies. To that extent, I take the side of Kant’s defenders against critiques like Rand’s.[1] All the same, I also think that some of the illiberal aspects of Kant’s politics do have to be laid, at least in part, at the door of flaws in his basic philosophy; and here I find a kernel of truth in Rand’s criticisms.

The classical liberal components of Kant’s political philosophy are undeniable. His account of a system of rights as defining the “fully reciprocal use of coercion that is consistent with everyone’s freedom”[2] is an exemplary libertarian formulation, and his account of property rights[3] lies solidly within a Lockean tradition that Kant shares with such libertarian thinkers as Lysander Spooner,[4] Émile Levasseur,[5] Murray Rothbard,[6] and Robert Nozick[7] (though also, more complicatedly, Johann Gottlieb Fichte[8] and G.W.F. Hegel).[9] Indeed, I think Kant’s account of the justification of property rights is in some respects more subtle and profound than that found in many libertarian writers. He also applies his theory of property in some admirably libertarian ways – for example, in his defense of native land rights against colonialist expropriation.[10]

But however libertarian Kant’s theory of property may be in its basic outlines, it is far less so in application. And here I cannot agree with Dr. White’s statement that “Kant did not support forced beneficence,” or that Kant’s political theory “rules out any state welfare system.” As part of the social contract (a fictional social contract – more on that anon), Kantian citizens are understood to cede to the ruler a degree of ownership over their private holdings; hence Kant explicitly defends

the right of the supreme commander … as supreme proprietor (lord of the land), to tax private owners of land, that is, to require payment of taxes on land, excise taxes and import duties, or to require the performance of services (such as providing troops for military service) …. On this supreme proprietorship also rests the right to administer the state’s economy, finances, and police.[11]

Notice how the right to conscript citizens into military service is casually included here as part of the taxing function. The Kantian state’s claim on its citizens goes well beyond a mere claim to their external property.

In any case, the taxing power also includes “the right to impose taxes on the people … to support organizations providing for the poor, foundling homes, and church organizations,” which sounds like a state welfare system to me. Kant even stresses that such public support is to occur “not merely by voluntary contributions” but “by way of coercion,” explaining that taxpayers “have acquired an obligation to the commonwealth, since they owe their existence to an act of submitting to its protection and care, which they need in order to live.”[12] Kant goes on to maintain that taxation is morally superior both to lotteries (because they’re regressive) and to begging (the latter supposedly being “closely akin to robbery,” while taxation allegedly isn’t).

Kant’s (classically) illiberal streak goes much farther than mere support for state welfare. He holds that illegitimate children have “stolen into the commonwealth … like contraband merchandise” and so stand outside the protection of civil law;[13] he denies citizens a right of self-defense against the state by declaring the ruler to be above the law and not to be rebelled against;[14] he denies wage workers the right of self-defense against the wealthy classes by depriving them of the vote;[15] and he is so enamored of capital punishment as to insist that “if a civil society were to be dissolved by the consent of all its members … the last murderer remaining in prison would first have to be executed.”[16]

One of the most problematic portions of Kant’s theory of justice is his category of “rights to persons akin to rights to things.” These are rights to treat certain other people to some degree as one’s property, a right allegedly consistent with the categorical imperative not to treat others as mere means. Thus Kant justifies the coercive authority of husbands over their wives, since “this dominance is based only on the natural superiority of the husband to the wife”;[17] he likewise justifies a coercive authority of employers over employees, maintaining that a head of household is entitled to “fetch servants back and demand them from anyone in possession of them, as what is externally his.”[18]

How much more “Kantian” in spirit is Herbert Spencer, who rejects the husband’s authority over his wife as “utterly repugnant,”[19] and who describes the “master-and-workman type of industrial organization” as a “remnant of the régime of status” by which an employee is “temporarily in the position of a slave,” and so calls for the replacement of the wage system by a system of workers’ cooperatives.[20]

Now many of these cases can fairly be diagnosed, at least in part, as instances in which Kant has simply failed to apply his own principles correctly. In particular, he understands contract right as a right to specific performance – a “possession of another’s choice”[21] – whereas a more authentically “Kantian” approach, insistent on rejecting all forms of interpersonal subordination, would see personal services as inalienable, and thus would treat contractual obligation as rooted in the transfer of title to external resources only.[22] Likewise, using a greater degree of compulsion against a person than is necessary to restrain her from injuring others arguably treats that person as a mere means, a consideration that invalidates the institution of coercive punishment.[23]

But Kant’s theory of the social contract, which is the basis both for his claim that a political ruler is the supreme proprietor of citizens’ land, and for his rejection of the right of rebellion, is a bit more deeply rooted. Regrettably, Kant’s version of social-contract theory owes as much to Hobbes and Rousseau as to Locke.

To begin with, Kant’s theory that rights in the state of nature are simultaneously genuine and merely provisional is an unconvincing attempt to fuse a) the Lockean insight that rights exist prior to the state with the Hobbesian claims that b) we have no reason to respect others’ rights in the absence of law, and that c) law cannot exist without the state.[24]

Both of the Hobbesian claims, (b) and (c), seem incoherent. On the one hand, the view that in the state of nature one should withhold cooperation until other parties show themselves to be cooperative first is evidently not universalizable, which should be enough to condemn it by Kantian standards.[25] On the other, by confusing law with legislation,[26] Kant, in his very attempt to defend the rule of law, makes the rule of law impossible by exempting the ruler from its application.

If this Hobbesian muddle weren’t bad enough, Kant maintains that victims of state coercion are really free, since the state represents their true will via the social contract – so that, e.g., when the state taxes its citizens it’s really exercising their property rights on their behalf. Kant thereby adds Rousseau to the Locke/Hobbes mix – with the added twist that the contract need not even be genuine.

I’ve written elsewhere of the conflict between vicarious and immanent forms of liberalism.[27] Immanent liberalism is concerned to protect relations of mutual consent in people’s everyday lives, while vicarious liberalism displaces the chief locus of consent somewhere else – to the social framework as a whole, or to a merely hypothetical starting point, or both. Kant’s liberalism contains both immanent and vicarious strands; but the amount of emphasis placed on the vicarious strand threatens to poison the system, since actual curtailments of everyday freedom are all too easily justified as part of a system that one has tacitly consented to, or would hypothetically consent to.

And here I do think we find a set of problems that can be traced back to defects in Kant’s basic ethical principles. Kant’s theory of the categorical imperative draws a sharp distinction between duty on the one hand and happiness on the other, where happiness is seen primarily as a matter of following our subjective inclinations. Dr. White is right to point out that Kant is not nearly as hostile to individual happiness as he is often thought to be; nevertheless, Kant insists that there is no necessary connection between duty and happiness, so that they always can come apart even if they usually don’t.

On the one hand, the duty/inclination divide fuels Kant’s Hobbesian side; since inclinations are essentially nonmoral and always potentially antimoral, we can conclude a priori that people cannot be trusted without state compulsion. But on the other hand, the duty/inclination divide also fuels Kant’s Rousseauvian side, since the Kantian doctrine that we are autonomous only when choosing duty over inclination lends itself all too easily to vicarious liberalism and the privileging of indirect, tacit, or hypothetical consent over actual consent. This is what is right in the Randian critique of Kant.                                       

Where Rand goes wrong, I think, is in taking Kant’s theory of a duty/inclination divide as motivated by a pathological hatred of human happiness. The duty/inclination divide represents Kant’s attempt to solve a genuine philosophical problem; I think he gives the wrong solution to it, through grasping one half of the truth while discarding the other. And while Rand is right to identify Aristotle’s approach to ethics as offering the best way out,[28] I don’t think Rand quite manages to follow where Aristotle leads; instead, she ends up seizing the piece of the truth that Kant has discarded while discarding the piece he has seized. But that’s a point I’ll elaborate on in the rest of the discussion.

I thank my three colleagues for their comments; in this response I will try to address the themes common to all of them.

In general, the comments helped me see I should have been more circumspect in how I presented my thesis about Kant as the “ideal of classical liberalism.” Instead of promoting Kant as a broad figure, with all of his personal and philosophical complications, I should have limited my thesis to the essence of his ethical and political theory (as I summarized them), which is where I think the value lies, especially in terms of linking him to classical liberalism.

1.

When I think about Kant, I think in generalities, such as the basic ideas of autonomy, dignity, and duty, and the implications thereof—in particular, the judgment made necessary by the vagueness of duties and the inevitable conflicts between them, which leads to the flexibility I claim for Kant’s ethics. One can see this in the casuistical questions scatted throughout the Doctrine of Virtue, which represent a series of “but what if” questions that I can easily imagine him tossing out at his students to show how difficult moral decisionmaking is in the real world. I also focus on his essential comments about the role of government, particular the Formula of Justice that holds government to the same ethical standards to which we hold individuals, particularly to treat them with respect as persons with dignity.

I regard this general framework for ethical thinking as valuable in and of itself. As such, I don’t concern myself much about the phenomenal/noumenal distinction in Kant’s metaphysics, which Hicks emphasized in his commentary. To be sure, I completely agree with him that it’s contentious (at best). I don’t accept it, and I don’t think many modern Kantian ethicists do either, but we all believe nonetheless that Kant’s ethics are important and valuable, regardless of where freedom actually resides. As I said above, his basic positon starting with autonomy and dignity generates a system of ethics that governs the basic interactions between persons as well as between citizens and state, which can be applied to practical problems without being overly concerned about metaphysical complications such as the timelessness of the noumenal self.

2.

However, the internal inconsistencies in Kant’s moral philosophy that Hicks and Long mention are potentially very troubling in that they cast the integrity of the core of Kant’s ethics into doubt. Both commenters acknowledge that specific ethical statements must be interpreted in light of general ones, and I would add that we must also be willing to reject specific claims if they are not consistent with general ones. The fact that Kant expressed racist and sexist ideas that conflicted directly with his writings about universal dignity for all persons should not, in my opinion, diminish the value of those profound words any more than should his metaphysical complications. Of course, we must be mindful not to cherry-pick, endorsing as integral the things we happen to like and dismissing as irrelevant those we don’t. We must hold Kant’s specific claims to the standard he established through his core principles—as Kant recommends we do with our own ethical decisionmaking—and we must also be willing to hold some claims to be mistaken in light of those principles.[1]

In addition to the obvious examples of racism and sexism, I believe this applies to the notorious paragraph from the Doctrine of Right cited by Long (and an online commenter) that appears to endorse a system of coerced redistribution.[2] Its tone and content are in striking contrast to Kant’s more general statements regarding the narrowly circumscribed domain of state coercion (to limit coercion by others), yet some Kant scholars have tried to interpret it within his general political philosophy, generating a provisional defense of a limited welfare state.[3] Kant’s denial of a right of revolution also seems to be at cross-purposes with the rest of his political philosophy; it does reflect his strong belief that a state was necessary to protect rights (with which many classical liberals would agree), but cedes too much power in a tyrannical government by tying the hands of its citizens.[4] As Roger Sullivan writes, “Kant’s discussions of civil disobedience and of the right to revolution are obviously inadequate, both in terms of his own moral theory and in terms of the judgments of ordinary moral reasoning” (emphasis mine), concluding that the only explanation can be the despotic state Kant lived in at the time.[5]

I also believe that this identification of specific mistakes judged against general principles applies to the infamous case of the murderer-at-the-door that Salmieri mentions, in which Kant maintains (quite passionately) that you must tell the inquiring prospective murderer that his prey, your best friend, has hidden in your closet.[6] Much ink has been spilled and many trees felled by philosophers trying to square this circle, but I think the clearer path is to judge that he applied his basic principles incorrectly in his verve to promote honesty and effective communication as a value above most others. In general, perfect duties do not automatically take precedence over imperfect duties; rather, their relative grounds of obligation, the substance behind the duties, determine which is more important.[7] Because the specific ground of the duty of beneficence in this case is an innocent life, and the duty of not lying is owed to a person who wrongfully threatens that life, I think it is clear that, in weighing the two grounds, the former obligation prevails.

Consistency is an ideal goal, and certainly the initial presumption should be that all of a philosopher’s claims fit together. But this goal is rarely achieved, especially with systematic philosophers like Kant, and at some point we must be willing to consider that particular claims are mistaken, especially when we can defend such a judgment on the basis on the majority of his other work. While inconsistencies and contradictions are worthy subjects for philosophical inquiry and can sometimes be illuminating, there is also the risk that the bulk of a thinker’s writing that is consistent and focused can get lost (in the spirit of Oliver Wendell Holmes’ famous saying, “hard cases make bad law”). I maintain that once you separate the wheat from the chaff, the distillation of Kant’s ethical and political system is a fine statement of classical liberalism that sits well among other, more widely acknowledged advocates.

3.

But not all of the criticisms from my colleagues hinge on inconsistencies and complications; some questioned the core of Kantian ethics that I defend as an exemplar of classical liberalism. In particular, Long and Salmieri questioned Kant’s emphasis on duty and morality to the detriment of happiness and flourishing. This is one case (among several) in which Kant’s rhetoric works against him, for in fact Kant did prize human happiness, holding it up as the ultimate good (albeit proportionate to a person’s virtue).[8] On the level of specific acts, he stressed self-regarding duties such as to cultivate one’s talents as well as to practice beneficence to oneself; on the latter, he wrote that “lawgiving reason … permits you to be benevolent to yourself on the con­dition of your being benevolent to every other as well.”[9] Furthermore, this duty must be balanced with all others using our judgment, which may favor our own happiness, even to the exclusion of specific (and perfect) duties to others, depending on the specific choice circumstances.

Because persons can generally be trusted to pursue their own happiness, it was Kant’s intent to point out the limitations of that pursuit out of respect for the autonomy and dignity of other persons. In other words, Kant did not want us to sacrifice our happiness; he just wanted to make sure we didn’t gain it wrongfully at someone else’s expense. This corresponds well to ideas such as Mill’s harm principle and the principle of noninterference, providing further support of Kant’s core ethical ideas as consistent with classical liberalism.

I agree with Dr. White that the basic moral principles Kant endorses – specifically, the prohibition on treating persons as mere means – commit him to a broadly libertarian politics, and that Kant follows out the implications of that commitment to a considerable extent. But I’m less convinced by White’s attempts to downplay the centrality of some of Kant’s deviations from those implications.

Specifically, White acknowledges that the passage I cited from the Doctrine of Right “appears to endorse a system of coerced redistribution,” but he suggests that its “tone and content are in striking contrast to Kant’s more general statements regarding the narrowly circumscribed domain of state coercion (to limit coercion by others).” Here I disagree. It seems to me that Kant is thinking in Rousseauvian terms here: since all the individual property-holders are to be regarded as parties to the (fictional) social contract, the political disposition of their property is (as Kant sees it) in an important sense not coercive, since they’ve all agreed to have their individual holdings regulated by themselves in their collective capacity (as represented by the state).

The problem Rousseau faced was that of reconciling political authority with human freedom and equality. His solution was the idea that in the social contract, we all individually submit ourselves to all of us collectively, so that (so long as the collective rules in accordance with the general will) we are subordinated only to ourselves and so remain technically free. Through this “total alienation of each associate, together with all his rights, to the whole community,” says Rousseau, “each gives himself absolutely,” and since “the conditions are the same for all … no one has any interest in making them burdensome to others”; thus “each, while uniting himself with all, may still obey himself alone, and remain as free as before.”[1]

And here is Kant in high Rousseauvian mode:

The legislative authority can belong only to the united will of the people. For since all Right is to proceed from it, it cannot do anyone wrong by its law. Now, when someone makes arrangements about another, it is always possible for him to do the other wrong; but he can never do wrong in what he decides upon with regard to himself (for volenti non fit iniuria [= “to the willing, no injustice is done.”]).[2]

Thus on Kant’s view, when the state redistributes its citizens’ property, it does them no injustice, since the actions of the legislative authority are simply the people acting upon themselves.

This Rousseauvian approach pervades Kant’s political thought (albeit uneasily interwoven with Lockean and Hobbesian strands). For example, Kant’s justification of marriage is a Rousseauvian social contract writ small. For Kant, sexual intercourse is presumptively immoral, since it involves using one’s sexual partner as a means to one’s own pleasure. (He seems unaware that people sometimes try to give their sexual partners a good time as well. But then his personal experience in this area was not extensive.) The institution of marriage legitimizes the sex relation through a contract involving subordination that, in virtue of being reciprocal, in a certain sense ceases to be true subordination (even though the male spouse is the sole government of this mini-republic).

There is nothing surprising, then, in Kant’s taking redistributive taxation to be noncoercive, insofar as the state is to be viewed, not as violating its citizens’ property rights, but as exercising those rights on their behalf. This is the vicarious conception of consent that I referenced in my initial essay, and it is pervasive in Kant’s political thought (even if it has to share space with a more immanent conception congenial to libertarians). And in fact he is getting it straight from Rousseau, whose account of property in The Social Contract likewise sounds rather Lockean (with its talk of first occupants establishing property rights through labor) until Rousseau reminds us that proprietors, in order to render their holdings more secure, surrender those holdings entirely, along with themselves, to the collective sovereign, whereupon the united private holdings constitute a “public territory.”[3] In J. L. Austin’s words, “[t]here’s the bit where you say it and the bit where you take it back.”[4]

Moreover I do think that the sharp opposition, in Kant’s ethical thought, between our true will and our mere inclinations makes it easier for Kant to find his way to a vicarious, Rousseauvian approach to consent – just as his division between the noumenal and phenomenal realms makes it easier for him to embrace a fictional social contract that devalues real consent in favor of a merely posited consent. (After all, Kant’s view of free will is that we are pawns of causal determinism in the phenomenal realm but must think of ourselves as being free agents in the unknowable noumenal realm. If the move works for moral freedom, why not similarly for political freedom?)

The Lockean strand in Kant’s political thought is real and important. But it stands forever in the shadow of the Rousseauvian strand, which lies ready at any moment to devour and transmogrify it.

In this comment I’d like to focus on one element of Kant’s philosophy and its implications for liberal economic policy.

One kind of classical liberalism argues that liberty is good because individuals need freedom to be productive, trade, and enjoy the results. Property rights are then understood both as a necessary condition for successful economic activity—individuals need to control their productive assets, their trading goods, and their gains from trade—as well as being a recognition of the deservedness of one’s successful economic activity. There is thus a conception justice embodied in liberal economics: it’s just because individuals make and own the results of their choices and actions, for good or bad. Property rights in part embody this conception of economic justice: You get to keep what you earn.

Let us consider the common scenario in which a successful person will make a donation to a worthy cause but will also deflect any praise by saying “I’m only giving back.”

A standard rejoinder is to point out that “giving back” assumes that the giver has taken something from others in the first place—he or she has borrowed or stolen something, and in “giving back” is merely restoring it to its rightful owners. That zero-sum assumption is usually untrue in a market economy, since most donors have earned what they possess. So the claim that one is “giving back” contains within it an injustice: a false accusation of borrowing or theft.

Yet the phrase also denies the benevolence of the giver. If the giver is only giving back what is rightfully someone else’s, then he or she does not deserve any special praise for doing so. “Giving back” thus  implicitly contains a double injustice: it implies that the giver does not deserve what the giver has, and it denies the giver any credit for his or her benevolent act.

Now let us return to Kant and his Lectures on Ethics, specifically the lecture titled “Duties to Others.” 

Here Kant employs his usual distinction between inclinations and duties: actions done from inclination have no moral worth, while actions done from duty do. What then, Kant asks, does this mean for acts of charity? Charity done out of benevolence has no moral significance, since benevolence is a kind of inclination. By contrast, charity done out of duty has moral significance.

Kant then asks a follow-up question: On what is the duty to be charitable based? Why ought one be charitable, whether one is inclined to or not?

Kant’s answer is that to give charity to the poor is to make good on past injustices. Here is the key quotation. In giving to a person in need of charity, the giver

makes restitution for an injustice of which he is quite unconscious; though unconscious of it only because he does not properly examine his position. Although we may be entirely within our rights, according to the laws of the land and the rules of our social structure, we may nevertheless be participating in general injustice, and in giving to an unfortunate man we do not give him a gratuity but only help to return to his that of which the general injustice of our system has deprived him. For if none of us drew to himself a greater share of the world’s wealth than his neighbor, there would be no rich or poor. Even charity therefore is an act of duty imposed upon us by the rights of others and the debt we owe to them (p. 194). 

Thus in Kant we have the first part of the “giving back” claim made explicit: the zero-sum assumption and the consequent implication that one is merely returning something one has borrowed or stolen.

Kant here is indeed Rousseauvian, as per the second part of Rousseau’s Discourse on the Origin of Inequality, and his thought is in keeping with the connections Professor Long makes in his recent comment. Kant’s claim is also congruent with Proudhon’s later “Property is theft” line that is much beloved of the political left.

On the very next page of “Duties to Others,” Kant makes explicit the second assumption of the troubling uses of “giving back”:

A man ought not to be flattered for his acts of charity lest his heart swell with generosity and desire to make benevolence the sole rule of his conduct (p. 195).

Kant therefore denies that charity is properly a matter of benevolence or even of a duty to help the poor meet their needs, as previous thinkers had argued. To my knowledge, Kant is the first to argue that charity is a matter of justice — compensatory justice, to be precise.

This Kantian conception of charity-as-justice is in marked tension with the economic liberalism sketched at the beginning of this comment. If charity is a matter of justice, then there are implications for the role of the state, given that the state is an arbiter and enforcer of justice. In other words, Kant’s twist on the ethics of charity has consequences for those versions of modern political philosophy that see the welfare state as an instantiation of social justice. Kant’s argument in Lectures on Ethics thus undermines at least one classically liberal understanding of justice and property rights.

As Prof. White notes, much of the suspicion of Kant’s philosophy in libertarian circles is due to Ayn Rand’s severe condemnation. To be sure, Rand’s chief quarrels with Kant concern metaphysics, epistemology, and ethics, not politics; but Rand did think that Kant’s views in those more fundamental areas of philosophy had authoritarian political implications, even if Kant did not fully draw those implications himself. One of Rand’s students even devoted, with Rand’s approval, an entire book to blaming Kantian influence for the rise of Nazism in Germany.[1]

In a nutshell, the links Rand saw were as follows. In metaphysics and epistemology, Kant – according to Rand’s interpretation – denies that our minds can know reality; all we can ever know is a world of appearances constructed by our cognitive apparatus. In ethics, Kant demands that happiness be sacrificed to duty whenever the two conflict (and he does think they often conflict). The first view is an attack on human cognitive efficacy; the second is an attack on human happiness. (Deliberate attacks, in Rand’s view; when Kant goes wrong, it is through malice, not innocent error.) And those who have been taught that they cannot trust their own cognition and have no right to pursue their own happiness are intellectually disarmed against the claims of a totalitarian state.

I said in my initial essay that I thought Rand was partly right and partly wrong here. Let me elaborate.

Peter Strawson argues persuasively that there are two strands in Kant, one “austere” and one “transcendental.” The core of Kant’s “austere” project is to explore the “limits to what we can conceive of, or make intelligible to ourselves, as a possible general structure of experience”; this project Strawson considers legitimate and defensible. But in his “transcendental” moments, Kant mistakenly slides into thinking of those limits as a product of the “ways in which our experience is causally dependent” on “our own cognitive constitution.”[2]

In other words, Kant is on solid ground in seeking to investigate what picture of the world we can make sense of; his error lies in taking the results of that inquiry to be the product of our cognitive apparatus’ imposing certain structures on a reality that, but for that imposition, might have been structured entirely differently. This, as Strawson rightly notes, is incoherent; for if we cannot make sense of the world not having those structures, we cannot intelligibly entertain the possibility of the real world’s not having them, and so cannot intelligibly regard those structures as the mere products if our cognitive apparatus. This would be the mistake that Wittgenstein diagnoses as trying to think both sides of the limits of thought.[3]

The transcendental Kant is the Kant that Rand attacked.[4] And the transcendental Kant does indeed seem to undermine the reality and cognitive efficacy of the individual in a way that is politically troubling. But on Strawson’s reading, which I find persuasive, the transcendental Kant is not the most important Kant; rather the transcendental Kant represents Kant’s misunderstanding of his own project, which properly understood yields the austere Kant. (Now Rand wouldn’t like the austere Kant either; she’d find him too much of an a priori rationalist. But she wouldn’t regard him as an unmitigated disaster; after all, she had kind words for Spinoza.)

Thus the transcendental strand in Kant is a mistake, not a deliberate salvo against the human mind. Moreover, it’s not a stupid mistake or a mistake easily avoided; it’s all too easy to slide from identifying the limits of thought to seeing those limits as impositions on something that might otherwise be different.

With regard to ethics, while it’s true, as Dr. White points out, that on Kant’s view we have duties to ourselves, it’s still the case that Kant believes in the possibility (and the frequent, though not unvarying, actuality) of conflicts between duty and self-interest, and holds that in such cases we must choose duty, with self-interest being given no contrary weight at all. This is why Rand sees Kant as an extreme altruist.

Here, as in the case of his metaphysics and epistemology, I think Kant is sliding from a correct insight into a mistake.

The correct insight is that our motivation to do the right thing should not be conditional on how we happen to feel or what we happen to find strategically convenient. If the reason I respect other people’s rights is that doing so gives me a warm fuzzy feeling, then as soon as I become depressed and no longer get that fuzzy feeling, I can no longer be counted on to act rightly. If the only reason I behave cooperatively toward others is to incentivize them to behave cooperatively toward me, then as soon as I see a way of winning others’ cooperation without being cooperative in turn – say, by deceiving or defrauding them instead – I can no longer be trusted. A genuinely moral person’s commitments cannot be conditional in that way.

This much I think Kant is right about. Moreover, this is one of the parts of Kantian doctrine that is especially congenial to libertarianism of the natural-rights variety. Natural-rights libertarians often criticize consequentialist approaches to liberty for being unreliable in their commitments. As Murray Rothbard writes:

Another problem with the utilitarian is that he will rarely adopt a principle as an absolute and consistent yardstick to apply to the varied concrete situations of the real world. He will only use a principle, at best, as a vague guideline or aspiration, as a tendency which he may choose to override at any time. … To say that a utilitarian cannot be “trusted” to maintain libertarian principle in every specific application may sound harsh, but it puts the case fairly.[5]

But by placing duty at odds with self-interest, and demanding the sacrifice of the latter to the former in case of conflict, Kant arguably does downgrade the importance of human happiness, just as Rand charges. But there is a reason (other than malice) for Kant’s position. Just as, in metaphysics, Kant thinks that there are a priori constraints on what account of the physical world makes sense, so in ethics, Kant maintains that there are a priori constraints on what account of morality makes sense; the unconditional character of moral duty is one of those constraints.

So far so good. But Kant seems not to think that there’s much at all in the way of analogous a priori constraints on what happiness might be. Given that he holds to a fancy, highly constrained conception of morality and an un-fancy, rather unconstrained conception of happiness, it’s unsurprising that he thinks nothing can guarantee that the two will go in tandem:

If the doctrine of morals were merely the doctrine of happiness it would be absurd to seek a priori principles for it. … Only experience can teach what brings us joy. … But it is different with the teachings of morality. They command for everyone, without taking account of his inclinations …. He does not derive instruction in its laws from observing himself and his animal nature or from perceiving the ways of the world …. Instead, reason commands how men are to act even though no example of this could be found, and it takes no account of the advantages we can thereby gain, which only experience could teach us.[6]

This is where I wish to contrast Kantian ethics with the approach of the ancient Greek philosophers (in particular, Socrates, Plato, Aristotle, and the Stoics). To begin with, the Greeks draw a distinction between instrumental and constitutive means. As I’ve written elsewhere:

Suppose that I want to play the Moonlight Sonata; and so I save money to buy a piano, and to buy sheet music, and to take piano lessons and so forth, so that I’ll be able to play the Moonlight Sonata. These are all means to the end of playing the Moonlight Sonata; if you ask me why am I saving this money, why am I buying a piano, etc., I would say these are all means to my ultimate goal, which is to play the Moonlight Sonata.

But now suppose you come upon me in the middle of playing the Moonlight Sonata, and I’m hitting a particular note. And you ask me: “Why are you hitting that particular note? Is it just that you find that note valuable in and of itself?” And I would answer: “No, I’m playing that note because I want to play the Moonlight Sonata, and I can’t play the Moonlight Sonata without playing that note at that point.” Well, in a sense, then, playing that note is a means to playing the Moonlight Sonata; but it’s not a means in the other way. It’s not a means that’s external to the end; it’s a means that’s part of the end. …. So there are cases where a means can be a constitutive part of the end rather than being an external means to it.[7]

The Greek philosophers declare each individual’s happiness to be the supreme good and appropriate end for that individual. And morality they declare to be a means to that end. But they take morality to be a constitutive rather than an instrumental (external, strategic) means to happiness – for happiness, on their view, is an objective state of being with objective requirements, not a subjective feeling. (Here, again, I have in mind Socrates, Plato, Aristotle, and the Stoics. The Epicureans are a more complicated case.)[8] Moreover, the objective constituents of happiness are determined by conceptual rather than empirical considerations.

For example, when Aristotle tries to show that “honor” – by which he means being honored, i.e., being respected and admired, or having a good reputation – cannot be the central component of happiness, he doesn’t try to show that honor will leave us subjectively dissatisfied; instead, he argues that honor violates certain a priori constraints on the nature of happiness, such as being an end in itself, or not being excessively outside of our control.[9] Aristotle would thus firmly reject Kant’s dictum that “only experience can teach” what brings us happiness. The Greeks are in effect doing for happiness and morality together what Kant is doing for morality alone.

The advantage of the Greek approach to ethics over the Kantian approach, as I see it, is that it allows morality to have all the unconditional and a priori character with which Kant seeks to invest it, but without severing the tie between happiness and morality. Moreover, the way in which Greek ethics takes determining the contents of morality and happiness to be a matter of mutual adjustment offers libertarians a manner of reconciling rights-based and pragmatic arguments for liberty.[10]

Kant does not seem to understand the Greek view of ethics. For example, he tells us that “[t]he ancients saw perfectly well that happiness alone could not be the greatest good” – with the exception of Epicurus, who “took away the value from virtue, in that he made morality into a means to happiness.”[11] Here Kant is clearly using “happiness” to mean a subjective feeling – which he is entitled to do in describing his own views, but it’s not a position he should be attributing to Plato, Aristotle, or the Stoics, who quite explicitly did declare happiness, objectively understood, to be the greatest good.

Thus Rand is right in championing Aristotle against Kant, and in criticizing the latter for devaluing human happiness. But Kant’s position here was neither a fiendish plot nor a dim-witted blunder; rather, it was the product of a laudable attempt to protect the unconditionality of moral commitment, combined with a failure to see how happiness could be included in the deal. Nor can Rand claim much superiority over Kant on this point, since she herself is notoriously unclear as to whether the relationship between self-interest and morality is instrumental or constitutive, causal or conceptual.[12] Despite her professed admiration of the Greek approach to ethics, Rand is like Kant in not fully finding her way to it.

Professor Salmieri and Professor Long have drawn our attention to the tricky issue of the relationship between philosophical principles and practical application.

Long advocates a sensible moderate position, noting that thinkers can be more or less accurate about the actual connections between their principles and their practical policies, so on a case by case basis (especially in the case of Kant) we should be open to the possibility of separating a thinker’s philosophical and practical claims.

Salmieri suggests a tighter connection between philosophical principle and practical application, with the intriguing suggestion that major thinkers who apparently agree over practical matters do not in fact do so when one considers the philosophical justification of their practical positions. 

In his response, Professor White agrees that there are many non-liberal and many liberal elements in Kant’s practice. At the same time, in response to the troubling elements in Kant’s basic philosophy, he adopts a more radical severance of basic philosophy from normative application, suggesting that we can use Kant’s ethics “without being overly concerned about metaphysical complications.” 

I’d like to take up this point of method, as I disagree with White: When looking for a philosophical justification of liberalism, we must be overly concerned about metaphysical complications. Especially in the case of interpreting and using a philosopher such as Kant—a systematic philosopher if ever there were one—selecting what one likes from his ethics and omitting the metaphysical context runs the risk of cherry-picking on a grand philosophical scale.

Consider two simple examples to begin:

Two women explain why they cook for their families. The first says, Because I love cooking and I love my family. The second says, Because it’s my duty as the woman. As a matter of practice, both women consistently engage in the activity of cooking, so we can categorize them as cooks. But because their reasons for their actions are so different, it is questionable whether they belong in the same category as principled advocates of cooking.

In a second example, consider three people who believe that one should be truthful. When asked why, they say the following:

1: Because people will trust you, and want to be friends, and do business with you, and that will make you happy.

2: Because your parents and teachers told you to tell the truth, and you should obey them.

3: Because you will burn in Hell if you tell a lie.

As a matter of practical action, all three may be equally truthful. But when the action is bundled with the justification, their philosophies of truth-telling diverge significantly.

The philosophical principles become more important when we are categorizing “isms” (such as classical liberalism is meant to be), which bundle or integrate many principles, and when we are deciding which thinkers should be categorized in which ism.

To step up from the two simple examples to a more complicated one. Suppose that one were attracted to Christian ethics and its emphasis upon charity and forgiveness—but added that one doesn’t pay much or any attention to the talk about God, the revelations to the prophets, and the resurrection of Jesus in Christian theology. In this example, one separates the normative claims of Christianity from the metaphysical and epistemological claims. But by jettisoning the Christian metaphysics and epistemology, one is no longer defending Christian ethics. One might still advocate charity and forgiveness, but one’s advocacy is not uniquely Christian. Partly this is because many ethics in many religious and naturalist philosophies advocate charity and forgiveness, so a free-floating advocacy is no longer specifically Christian. Partly this is also because the Christian religion is intended by its creators to involve the entire Christian theological package.

In the same way, a “Kantian” ethicist is arguably not a Kantian ethicist to the extent that he or she is unaware, uninterested in, or actively jettisons the metaphysical and epistemological underpinnings of Kantian ethics.

Autonomy and dignity are not free-floating abstract qualities. One must specify the subject: Who or what is autonomous or dignified? And Kant is clear that he means the noumenal, not the phenomenal, self. Consequently, a principled stress upon autonomy and dignity alone is not sufficient to make one a Kantian, especially as other philosophers—Aristotle, Thomas Aquinas, and John Locke are prominent examples—have also stressed and defended the autonomy and dignity of human beings. A Kantian autonomous-and-dignified self is differentiated from the Aristotelian, Thomistic, or Lockean autonomous-and-dignified selves precisely by means of the differences in the underlying metaphysics. And that means we must always integrate a philosopher’s metaphysics with his or her normative philosophy.

White and I evidently have a very different understanding of Kant’s ethics. Some of the points on which we differ are well known areas where Kant scholars (of which I am not one) differ. But some of White’s interpretations seem to me to be idiosyncratic to him, and I see no basis for them in the texts.

Among these is White’s view that Kant thinks that duties often conflict and that, when they do, imperfect duties can sometimes trump perfect ones, because they can sometimes have stronger grounds of obligation. In my first post I expressed my surprise at this idea, and I asked what texts White thinks support this surprising interpretation. In White’s response, he pointed me to Metaphysics of Morals Ak 6:224. Here’s what I take to be the relevant portion of what Kant says there:

A conflict of duties would be a relation between them in which one of them would cancel the other (wholly or in part). But since duty and obligation are concepts that express the objective practical necessity of certain actions and two rules opposed to each other cannot be necessary at the same time, if it is a duty to act in accordance with one rule, to act in accordance with the opposite rule is not a duty, but even contrary to duty; so a collision of duties and obligations is inconceivable. However, a subject may have, in a rule he prescribes to himself, two grounds of obligation, one or the other of which is not sufficient to put him under obligation, so that one of them is not a duty. When two such grounds conflict with one another, practical philosophy says, not that the stronger obligation takes precedence, but that the stronger ground of obligation prevails.[1]

Far from saying what White would have him say, Kant says that genuine conflicts of duty are impossible (indeed inconceivable). If a person thinks she is in such a conflict, it is because one of more of the rules she takes to be a duty is not really a duty at all, because it is based on some “ground of obligation” that is insufficient to actually obligate her. One or both of the conflicting rules must me a mere maxim (viz. “a rule that the agent himself makes his principle on subjective grounds”).

I take it that the following would be an example of the sort of conflict Kant had in mind. Back when I was a teenage blues guitarist, I made a rule for myself that I would take advantage of every opportunity to play for an audience. Happily, a local bar converted itself into a blues venue for a time, and I established myself as a regular at their jam sessions. I didn’t drink there, no one ever asked for any ID, and it never occurred to me that minors might not be welcome. But then, one day, while I was chatting with the manager and about to go onstage, he said “You’re over 21 right?” in a tone that implied that I shouldn’t be there if I wasn’t. So my rule of taking advantage of every opportunity to perform had now come in conflict with the rule that prohibits lying. The basis of the former rule was my decision that taking every opportunity to play is a good way to develop my talents. According to Kant, I have an obligation to develop my talents. But I don’t have an obligation to specifically develop my talent at guitar, nor to develop it by playing at every single opportunity. The rule that I should develop that talent in that way is a mere maxim, and its grounds of obligation was a decision I made that did not actually obligate me. By contrast, the rule prohibiting lying is a duty that Kant thinks I can grasp a priori to be obligatory. Therefore, I had to be truthful, even if it meant violating my rule to seize every opportunity to perform.

If something is really a duty, whether perfect or imperfect, it has an authoritative ground of obligation—namely the fact that acting contrary to it would be acting on a maxim that I cannot will to be a universal law of nature. Kant thinks that it’s impossible for conflicting rules to have this ground of obligation, so he thinks duties can never conflict.

Perfect duties command one never to take certain prohibited actions (such as lying), whereas imperfect duties command one to take certain actions in at least some of the circumstances when taking those actions would not run afoul of one’s perfect duties. Thus the imperfect duty to develop one’s talents commands one to find some ways to develop some talents within the sphere of activity permitted to you by your perfect duties to refrain from lying, or euthanizing yourself, or having sex for pleasure, or doing any of the other things that Kant thinks we can each tell a priori that we are absolutely obligated never to do.[2] Any specific rules I make for myself about how to develop my talents (or about anything else) are applicable only within the sphere in which duty permits me to act. Laws instituted by a government that has legitimate authority over me are intermediate between duties and the rules I make for myself. The government has this authority in virtue of the duties to which I, my fellow citizens, and the lawmakers are all subject, and there are bounds on this authority—most notably, it doesn’t extend to laws compelling me to violate my duties. So if the government passes a law mandating that I lie, or prohibiting me from ever aiding the needy, I’m not obligated to follow the law. But if the government passes a law preventing me from hanging out in bars until I reach a certain age, I am obligated to obey, even if it conflicts with my self-made rules for developing my talent on guitar.

Now, let’s consider the infamous case of the murderer at the door inquiring after the location of his intended victim, who is hiding in my house. I have an imperfect duty to promote the wellbeing of others, and it commands me to do what I can, consistent with my other duties, to prevent innocents from being murdered. One rule I might adopt for myself (or even that might be legitimately legislated by government) would be to shelter intended victims in my house whenever possible. Perhaps it’s by following this rule that I got into my present predicament. But since I have a perfect duty not to lie, I can’t get out of the predicament by telling the murderer that the victim is elsewhere. I must find some other means. I can refuse to open the door, or to answer his question. Perhaps I can stall (without lying) for long enough to give the victim a chance to escape. But, if all these options fail, then I simply have to let the victim die.

White thinks Kant reached this conclusion only because of a misapplication of his own theory. He may be right. Many have argued that a perfect duty prohibiting lying does not really follow from Kant’s basic premises. However, this is not the main fault that White finds in Kant’s application. White thinks that Kant ought to recognize that the ground of my duty of beneficence, which White thinks is “an innocent life,” outweighs the ground of my duty not to lie in this case. But “an innocent life” isn’t the sort of thing that Kant thinks can ground an obligation at all. For him, all duties have precisely the same ground—namely, one’s inability to will that that a contrary maxim be a universal law of nature—or (to put what Kant thinks is the same point in a different ways) one’s treating humanity always as an ends and never as a mere means—or one’s respecting everyone’s autonomy. Other obligations, which are subordinate to these, derive from governments and agreements and the like that arise within the context of these duties. Issues like what happens to innocents—or to anyone else—don’t enter into the picture. To be sure, we have duties about how to treat innocents, but the consequences of our actions for their lives cannot as such be grounds of obligation, because consequences are just not the sorts of things that Kant thinks ground obligations. At least, that’s how I understand Kant, and I haven’t seen any evidence from White that supports his contrary interpretation.

Kant is not the “flexible,” “commonsensical” ethicist White wants him to be. If this is the sort of view he thinks is necessary to defend his political convictions, I suggest he look to other ethicists. W.D. Ross, in particular, comes to mind.[3] He argued that the various formulas through which Kant thinks one can determine what one’s duties are do not yield determinate answers, as Kant thought they did, to the question of what one should or should not do in various situations. Any action can be described at different levels of abstraction, such that it will be willable as a universal law under some descriptions and not under others.  Lying to the murderer at the door, for example, is universal when described as “a lie told to a murderous person,” but not when descried simply as a lie. With no simple algorithm for determining one’s duty, Ross instead generates a commonsensical list of “prima facie” duties, each of which has its source in some fact of the sort that (I gather) White thinks would count as a grounds of obligation. These prima facie duties aren’t actual duties, Ross argues, because they can come into conflict with one another. To determine one’s actual duty in a given circumstance, he thinks, one must exercise a nuanced particularistic form of judgment that is reminiscent of Aristotelian phronēsis. This seems very much like the theory of judgment that White finds in Kant. (I hope to say a bit about how I understand Kant’s view of judgment in a future post.)

For my part, I don’t see this sort of flexibility as a desideratum in ethics. On this one point I’m nearer to Kant than to Ross and White. A true moral principle, I think, is an absolute that applies without exceptions within the contexts in which it arises. If you find yourself in a situation where your moral principles seem to demand conflicting actions, you need to (as Rand put it) “check your premises.” You’ll find that there is a mistake somewhere in the chain of reasoning that led you to think that one of those principles is true or applicable to the situation at hand.

Moreover, I don’t think that the sort of Rossian flexibility White finds in Kant is of any help in grounding what I think of as the key insight of the classical liberal tradition: the idea that we each have an inalienable right to life, liberty, property, and the pursuit of happiness. Kant’s best claim to have made an important contribution here is that his moral theory is rigid enough to support such inalienable rights, whereas (for example) Mill’s is not. In the end, however, I think the critics are right that no specific content about our obligations or rights follows from Kant’s framework. And I think his basic conception of morality and its role in life represents an attack on the principles from which I think our rights to life, liberty, and the pursuit of happiness in fact arise. More on this in a later post.

I’ve defended the Greek philosophers as providing a better foundation for classical liberalism than Kant does. But I might fairly be asked: why then are these Greek thinkers generally so much less liberal than Kant in their actual political proposals?

The answer, I think, lies in one crucial insight that Kant, along with most other modern political thinkers, has that the Greek thinkers (meaning here, once again, mainly Socrates, Plato, Aristotle, and the Stoics) tended to lack.

The aforementioned Greek philosophers hold that the right reason to act justly is a love of virtue for its own sake, not a strategic concern to elicit cooperation from one’s neighbors.[1] Kant would agree (though he would quarrel with the Greeks’ understanding of what it means to love virtue “for its own sake”; they take it to mean loving virtue as an intrinsic component of happiness, whereas Kant would take it to mean loving virtue apart from any connection to happiness). For Kant, those who behave justly in order to win external rewards or avoid external punishments are acting merely in accordance with rather than from duty.[2]

But these Greeks go further, in thinking that no sociopolitical order can function properly unless most of its citizens are in fact motivated by virtue. Hence the Greeks see the inculcation of virtue as a crucial task of political institutions, and this accounts for most of the illiberal features of the ideal states put forward in Plato’s Republic and Laws, as well as (to a lesser degree) in Aristotle’s Politics. (Zeno the Stoic’s lost political treatise, also titled Republic, might seem an exception, since it reportedly dispensed with all mechanisms of state coercion; but Zeno’s ideal republic was also supposed to be a community of the wise only, and so did not contemplate the uncoerced cooperation of ordinary people.)

I’ve stressed that by the Greek philosophers I means specifically Socrates, Plato, Aristotle, and the Stoics, because there was another group of Greek philosophers – the Sophists – who very much looked to strategic thinking as the basis for social order.[3] (A partial exception is the Sophist Protagoras, who is portrayed in Plato’s dialogue of that name as arguing that we wouldn’t see as much social order as we do if most people were purely strategically motivated.)[4] But the Sophists also tended to hold that strategic thinking was the best reason to be just.[5]

Here Kant essentially splits the difference. He agrees with Socrates, Plato, Aristotle, and the Stoics that strategic motivations are the wrong reason to be just; but against them, and with the Sophists, he agrees on the importance of strategic motivations to the maintenance of social order. The distinctively modern insight that Kant shares with such thinkers as Thomas Hobbes, Bernard Mandeville, and Adam Smith is that the right incentive system can lead people to behave in a just manner.

“[T]he problem of setting up a state,” Kant tells us in Perpetual Peace, “can be solved even by a nation of devils … so long as they possess understanding.” The key to securing cooperation among the citizens is not to foster the “moral improvement of man”; rather, the “constitution must be so designed that, although the citizens are opposed to one another in their private attitudes, these opposing views may inhibit one another in such a way” as to incentivize cooperation.[6]

Here Kant arguably overstates his case, making the opposite mistake from that of Plato et al. Quite possibly no constitution could secure cooperation for a literal nation of devils; Protagoras is probably right in thinking that some degree of virtuous motivation in the populace is necessary in order for society to function. Thomas Paine surely strikes the right balance in tracing “that order which reigns among mankind” to a combination of “mutual dependence and reciprocal interest” on the one hand and “a system of social affections” on the other.[7]

But Kant, along with his fellow moderns, is right in seeing that, given the appropriate external incentives, the level of virtue needed for society to function is much lower than that needed to make the participants count as fully morally worthy.

The political authoritarianism of the ancient Greek philosophers has always been in tension with their ethical individualism.[8] But the perceived need for the state to mold its citizens’ characters arguably stems, at least in part, from the assumption that near-universal virtue is a prerequisite for proper social order. Kant’s retreat from that assumption allows him to defend a much less intrusive state than those we find in Plato or Aristotle.

Nor is Kant’s understanding of incentive systems limited to those deliberately constructed, like the formal constitution of a state. Unsurprisingly given his enthusiasm for Scottish Enlightenment thinkers like Adam Smith, Kant also makes room in his political philosophy for non-constructed “spontaneous order” or “invisible-hand” mechanisms. For example, in his Universal History Kant identifies the apparently destructive character of human beings’ competitive impulses, their “unsocial sociability,” as, paradoxically, one of the chief engines of social progress.[9]

To be sure, Kant’s understanding of spontaneous order has its limits. For one thing, as previously noted, he fails to see the possibility of law without the state. For another, he identifies legislative fiat as the sole possible origin of money: “In this way only (so it seems to me) could a certain merchandise have become a lawful means of exchange ….”[10] Kant fails to anticipate the arguments of Austrian economists like Carl Menger and Ludwig von Mises showing how a common medium of exchange could arise spontaneously out of a barter economy without state involvement.[11]

On the other hand, some Austrian economists have gone so far as to maintain that money can only arise in the bottom-up manner described by Menger and Mises, and could never be the product of state action. Murray Rothbard, for example, writes: “Money must develop out of a commodity with a previously existing purchasing power, such as gold and silver had. It cannot be created out of thin air by any sudden ‘social compact’ or edict of government.”[12] Here Kant deserves credit for seeing that this simply isn’t true; as Kant points out, a ruler can easily cause a given item to have universal exchange value simply by choosing to “levy taxes on his subjects in this material.”[13]

Roderick Long praises Kant and some of the Greeks (viz. Socrates, Plato, Aristotle, and the Stoics) for having a conception of morality that, unlike consequentialism, is “a priori,” “conceptual,” “fancy,” and “highly constrained.” He thinks the Greeks are superior to Kant because they marry this conception of morality to a similarly “a priori,” “conceptual,” and “fancy” conception of happiness. It is Kant’s “un-fancy” conception of happiness that led him to believe that it often comes into conflict with morality and to devalue it. Long thinks that Rand was right to prefer Aristotle to Kant and to criticize the latter for devaluing happiness, but he thinks her own view on the relationship between morality and happiness is little better, because it is beset by confusions that prevented her from adopting the Greek approach. I disagree with much about how Long conceptualizes and evaluates the relevant positions, but I think that there is an important element of truth in what Long says.

What separates the views Long calls “fancy” and “un-fancy” are the sorts of cognition and motivation that each view holds human beings are capable of. The un-fancy positions are essentially empiricist. Empiricism (in the sense I’m using the term here) holds that human cognition is sensory in character: we have sensory experiences, we remember experienced phenomena, we form associations among them, and we predict future phenomena on the basis of past experience. The corresponding view of human motivation holds that it too is sensory: we are moved only by desires for pleasure or other directly experienceable things. And we figure out how to satisfy these desires by taking actions that we remember having led to their satisfaction in the past. Some empiricists use the word “reason” to describe all or part of this manner of functioning, but Plato, Aristotle, and the Stoics all reserved that term for another sort of cognition that they thought human beings alone were capable of. Reason, as these philosophers understood it, enables us to understand the world in terms of universal concepts that enable us to know the natures of objects and to understand why they necessarily have certain attributes and act in certain ways. It also enables a rational or conceptual form of motivation.

Consequentialists usually have an empiricist view of happiness and of morality. They equate happiness with some sensory or quasi-sensory state such as pleasure or desire satisfaction, and they think that whether an action is moral or not is simply a matter of whether (and in whom) it tends to cause this state. The Greeks, by contrast, thought that happiness is something that can only be grasped and valued conceptually and that is (wholly or partially) constituted by having or acting on certain conceptual motives. Kant agreed that some such form of motivation was possible, and he thought that it was the essence of morality, but he did not think of it as a means to or component of happiness, because he understood happiness as desire-satisfaction—a state that he thought could (in principle at least) be achieved without reason.

Thus far I’ve said nothing about how reason works and how it relates to the sensory forms of knowledge and motivation. Different answers to this question yield fundamentally different philosophies. One answer is that the objects of reason are wholly different from those of sensory perception and that reason has access to them independent of the senses. This is the view that Plato takes in the Phaedo and other related dialogues. His descriptions of rational cognition are vague, as he himself acknowledges, but the one point on which he leaves no doubt is that rational cognition is radically opposed to and apart from the senses. The same goes for the Phaedo’s view of rational motivation. The only content that Plato gives there to virtue is that it involves a distain for sensory pleasure and a sort of purging of all things sensory. Plato’s valorizing of reason in this work amounts to little more than damning the senses and pleasure.[1]

Similarly, Kant’s valorization of duty amounts to little more than deriding inclination (i.e. the sensory-based desires the satisfaction of which he equates with happiness). It is true that Kant thinks that good people deserve happiness and that it is sometimes permissible to pursue one’s own happiness. But the central role happiness plays in his moral philosophy is that of a “a powerful counterweight to all the commands of duty which reason presents to him as so deserving of the highest respect.” Indeed, Kant thinks that the whole function of moral philosophy is to combat the influence of this counterweight which leads us to “rationalize against those strict laws of duty and to cast doubt on their validity, or at least upon their purity and strictness, and, where possible, to make them better suited to our wishes and inclinations, that is, to corrupt them at their basis and to destroy all their dignity” (Ak. 4: 406).[2] Moreover, as I indicated in my first post, Kant attempts to derive the content of duty wholly from its lack of any motivational basis in inclination.[3]

Plato and Kant may be right that morality involves motivation by reason rather than animalistic urges, but there is so little positive content to their understanding of rational motivation that I think it is more accurate to describe them as opponents of pleasure and sensory-based goals than as champions of conceptual motivation. (Similar points can be made about the Stoics, though the case is a bit more complicated, so I won’t take the time to develop this point here.)

Aristotle represents an alternative both to the Platonic view of cognition and to empiricism, for he holds that we develop rational cognition on the basis of sensory perception.[4] He says less than one might hope about the process by which it develops and about how we can ensure that we are performing the process correctly (so that the result will be knowledge rather than error), but his corpus contains numerous leads.

With regard to ethics, Aristotle held that moral motivation involves pleasure (including the sensuous pleasures that Plato damned in the Phaedo) and that it is largely based on lived experience. His view of and the other constituents of happiness is more contentful than Plato’s or Kant’s, and he says enough to give one a sense of how these constituents are meant to cohere into the distinctively human way of life that Aristotle thinks is the ultimate good.

But though I find much wisdom in Aristotle’s ethics and politics, I also find many positions that (I’m sure we’ll all agree) are deeply wrong: he advocated slavery and a politically subordinate role for women and people in certain professions; he recognized no virtue in productive work; he opposed money lending; and he was an opponent of Athens’ emerging middle class. No doubt many of these positions stem from prejudices of Aristotle’s time, and perhaps he cannot have been expected to have been able to see his way through them. But I don’t think it’s accurate to say (as Long does about Kant) that these flaws are mere misapplications on Aristotle’s part of his own methods or theses, for it’s not obvious to me that Aristotle had a method or thesis that would or could have prevented these flaws if it were more expertly applied. The problem with Aristotelian ethics, as I see it, is that it contains no means by which one can reliably distinguish things that are required by human nature for the human good from Aristotle’s (or some Aristotelian’s) personal preferences and prejudices. Or, to put the same point another way: the way the theory is not objective (though I think it does represent a stride towards objectivity).

The lack of objectivity is a concern with conceptual cognition generally. There is no shortage of sophisticated schemes for conceptualizing various domains, and from each there are profound conclusions that might be thought to follow necessarily. But the schemes and conclusions are not all compatible with one another, and it’s not obvious how to tell which, if any, is true. This muddle is what lends plausibility to empiricism: the idea that all such sophisticated schemes are hot air, and that knowledge cannot extend beyond a superficial acquaintance with phenomena and the patterns among them. I don’t think this empiricist position is tenable, but I don’t think many non-empiricist epistemologists have said much of value about how we can tell whether any given concept (or battery of concepts) will either enable genuine insights or mire us in error and confusion. There are attempts (such as Kant’s) to show that some small set of extremely abstract concepts is presupposed by all thought or experience; but, even if successful, these attempts would offer very limited guidance in reasoning. Consequently, when philosophers engage in “a priori” or “conceptual” reasoning, they are too often elaborating on or distilling their own prejudices and assumptions.

This elaborating and distilling of prejudices and assumptions is common in ethics. Kant’s “pure,” “a priori” moral philosophy is just certain elements of Christian (or more generally divine command) ethics purified of any element of personal value. In like manner I think the distinction between constitutive and instrumental values that Long admires is—in the form in which it appears in Aristotle, at least—a distillation of the prejudices of certain aristocrats who had contempt for work and were impressed with certain new forms of knowledge for which they could see no practical value.

The distinction we find repeatedly in Aristotle’s ethical and political treatises is between items that are valuable because they are useful in producing some further value, and useless items that are valuable intrinsically. Insofar as an item is in the first category, it is not valuable in itself, but only for its consequences, and these consequences are necessarily better than it in the sense that they must more valuable than it is and they must have this value independently of it. An item in the second category (insofar as it is in this second category) cannot be valued for the sake of any valuable consequences, but there is still a sense in which Aristotle thinks it can be valued for the sake of a further value. The item can be part of constellation of intrinsically valuable items which is more valuable than any of the members. If so, the item will be valuable both for its intrinsic value and for the contribution this intrinsic value makes to the larger whole of which the intrinsically valuable item is a part. Aristotle (in some moods, at least) seems to see happiness as such a constellation, made primarily of such intrinsically valuable activities as contemplating truths and taking virtuous actions. These activities are (what some later scholars have called) the constitutive means to happiness, and the items and activities by which we produce the values and conditions needed to remain alive and to engage in these activities are the instrumental means to happiness.[5] But it is unclear that Aristotle has any grounds for identifying all and only the activities he does as intrinsically valuable. Moreover, the idea that the value of productive activity is extrinsic to it figures in the justification of all the Aristotelian ethical theses that I described earlier as “deeply wrong.”

Let me close by indicating how Rand fits into the picture. Long sees her as someone who recognized that Kant was wrong to degrade happiness, but who was “notoriously unclear as to whether the relationship between self-interest and morality is instrumental or constitutive, causal or conceptual.” She is notorious for this only among readers who insist that the relation between a means and an end must be either instrumental and causal or else constitutive and conceptual. But this is a false dichotomy that Rand can show us the way out of. Here’s how I summarized her position in Ch. 6 of A Companion to Ayn Rand:

Atlas Shrugged and Rand’s subsequent essays present virtue neither as intrinsically good nor as an instrument by which one maximizes some such good as pleasure, desire-satisfaction, or longevity. Rand recognized a third alternative: the moral values and virtues are essential constituents of the ultimate end that is a person’s life, and they owe their status as constituents to the causal contribution they make to the sustenance of this life. On her view, there are no values apart from an individual’s ultimate value of his own life, and a person’s life is made up of the values and activities by which he sustains himself.[6]

Unlike the Greek views, Rand’s position makes clear what you would need to know to establish objectively that something is a moral value or virtue, rather than a subjective preference. Whether and how one can attain such principled causal knowledge brings us back to the questions I raised earlier about the nature and standards of conceptual cognition. But these questions (and Rand’s views on them) are a topic for another occasion.[7]

I’ve previously discussed some of the ways in which Kant deviates from Lockean liberalism in a Rousseauvian direction. I now want to discuss some of Kant’s deviations in a quite different direction: the Hobbesian one.

One of the crucial differences between Locke and Hobbes is that Hobbes sees cooperation, or at least anything approaching large-scale or long-term cooperation, as irrational, and consequently unstable, in the absence of government, since no one has any reason to cooperate unless they first have reason to expect the other party to cooperate as well, and this they cannot reasonably expect unless there is a state system in place to punish non-cooperators. Hence for Hobbes the state of nature is inevitably a state of war.

Locke, by contrast, thinks people have the capacity to recognize, and be motivated by, the moral law even when government is absent. Hence, while he has his own objections to the state of nature, Locke denies that it is necessarily a condition of violent conflict.

Given Kant’s insistence that human beings do have the capacity to be motivated by duty alone,[1] one might expect him to side with Locke. And sometimes he seems to, as when he says that “[w]hat is opposed to a state of nature is not a condition that is social,” for “in the state of nature, too, there can be societies compatible with rights ….”[2]

But while cooperation in the state of nature may be possible, in Kant’s view it is not rational. Given the pervasiveness of motives for predatory behavior (even though it is possible to resist those motives), we have no reason to cooperate until we first have evidence that the other party is cooperative, which without the state we can rarely have:

No one is bound to refrain from encroaching on what another possesses if the other gives him no equal assurance that he will observe the same restraint toward him. No one, therefore, need wait until he has learned by bitter experience of the other’s contrary disposition; for what should bind him to wait till he has suffered a loss before he becomes prudent, when he can quite well perceive within himself the inclination of men generally to lord it over others as their master …. And it is not necessary to wait for actual hostility; one is authorized to use coercion against someone who already, by his nature, threatens him with coercion.[3]

In other words, Kant is saying that in the state of nature it’s fine to initiate aggression, to make a preemptive strike against one’s neighbor, because one’s neighbor is a human being and so should be presumed guilty of aggressive intent until proven otherwise.

Notice that this Kantian position is essentially cribbed from Hobbes:

If a Covenant be made, wherein neither of the parties performe presently, but trust one another; in the condition of meer Nature, (which is a condition of Warre of every man against every man,) upon any reasonable suspition, it is Voyd: But if there be a common Power set over them both, with right and force sufficient to compell performance; it is not Voyd. For he that performeth first, has no assurance the other will performe after; because the bonds of words are too weak to bridle mens ambition, avarice, anger, and other Passions, without the feare of some coerceive Power; which in the condition of meer Nature, where all men are equall, and judges of the justnesse of their own fears, cannot possibly be supposed. And therfore he which performeth first, does but betray himselfe to his enemy; contrary to the Right (he can never abandon) of defending his life, and means of living. …[4]

The Lawes of Nature oblige in foro interno; that is to say, they bind to a desire they should take place: but in foro externo; that is, to the putting them in act, not alwayes. For he that should be modest, and tractable, and performe all he promises, in such time, and place, where no man els should do so, should but make himselfe a prey to others, and procure his own certain ruine, contrary to the ground of all Lawes of Nature, which tend to Natures preservation.[5]

In game-theoretic terms, Kant, like Hobbes, is advising us to keep defecting until the other party cooperates first. This is rather awkward for a theorist like Kant, who insists that we need to act only on maxims we can coherently universalize; after all, if everyone waits to cooperate until someone else makes the first move toward cooperation, then cooperation will never begin – a situation that would invalidate the Kantian imperative to leave the state of nature and set up a government, since setting up a government presumably requires cooperation. Robert Axelrod’s advice to cooperate first,[6] while based on instrumental considerations, seems more “Kantian” in spirit than Kant’s own position; and it’s certainly easier to universalize.

Just as Kant underestimates the extent to which cooperation can (indeed must) exist prior to the state, he also overstates the extent to which cooperation under the state must be the product of political coercion. After all, as Étienne de la Boétie pointed out in the 16^th century, and David Hume and William Godwin in the 18^th, the fact that the ruled generally outnumber their rulers shows that some motive other than sheer force must explain the extent of civil obedience.[7]

Even when Kant acknowledges the possibility of social cooperation in the state of nature, he denies that there can be law without civil government: without established government “there can be societies compatible with rights … but no law.”[8] And without law, rights in the state of nature are merely “provisional.”[9]

Here Kant, as previously noted, has made the mistake of confusing law with legislation.[10] The existence of nonstate, spontaneously evolved legal systems is nowadays a commonplace of libertarian social theory.[11] Of course Kant might refuse to regard these systems as genuine forms of “law” – but on what grounds? That they are not the product of a social contract? But neither are those of the state. Of course Kant instructs us to posit that the state’s laws originated in a social contract even though they did not; but why can’t we likewise posit that nonstate legal systems so originated?

Worse yet, Kant’s attempts to defend the rule of law actually undermine it, because in Kant’s eyes the ruler must be above the law: “The civil union … cannot itself be called a society, for between the commander … and the subject … there is no partnership.” ruler and ruled are “not fellow-members,” since one is “subordinated to, not coordinated with the other,” and only “those who are coordinate with one another” are “subject to common laws.”[12] If the ruler of a civil state cannot be subject to laws in common with the regular citizens, in what sense can such a state embody the rule of law?

In his writings on international law, Kant urges rival states to enter into systems of federation with one another: “Each nation, for the sake of its own security, can and ought to demand of the others that they should enter along with it into a constitution, similar to the civil one, within which the rights of each should be secured.”[13] But unlike the civil constitution, which sets up a state over individuals, the international constitution does not set up a super-state over states: “a federation of this sort would not be the same thing as an international state,” since “every state involves a relation between a superior (the legislator) and an inferior,” whereas an international federation is a relation among a “group of separate states which are not to be welded together as a unit,”[14] and is in fact a “voluntary coalition … that can be dissolved at any time.”[15]

In short, when discussing international relations Kant contemplates the possibility of a legal system that unites parties who remain free and equal and not subordinated to any overarching state. Why couldn’t such a relationship unite individuals rather than states? And why couldn’t the entire panoply of nonstate legal systems constitute such a federation?[16]

Indeed, if it’s true, as Kant says, that “every state involves a relation between a superior … and an inferior,” one might wonder how a moral philosophy as devoted to respect for personhood as is Kant’s could tolerate the existence of states. After all, as an “end in himself,” Kant tells us, each human being “possesses a dignity,” an “absolute inner worth,” that makes each of us “exalted above any price” and thus entitled to “moral self-esteem” which bids us to stand up for ourselves. “Be no man’s lackey,” Kant advises. “Do not let others tread with impunity on your rights.” There exists “a widespread propensity to servility in men,” but “one who makes himself a worm cannot complain afterwards if people step on him.”[17]

All this sounds good. Indeed, it sounds more than a little like Ayn Rand! But how can I respect you as a locus of priceless dignity if at the same time I regard you as my political subordinate, a legitimate object of my coercion? How can you respect yourself as a locus of priceless dignity if you regard me as your political superior, entitled to impose constraints on you that you have no corresponding right to impose on me? And how can a system of rights embody a “fully reciprocal use of coercion … consistent with everyone’s freedom”[18] if some members of society have rights of coercion that are denied to others – as is implied in the Weberian definition of the state as a “monopoly on the legitimated use of physical force”[19]?

Thus while some strands of Kant’s political thought lend themselves to authoritarianism, others point in the direction of free-market anarchism.

I find much both to agree and to disagree with in Greg Salmieri’s latest post (as he says he did in mine). As the present installment of Cato Unbound is winding down, time doesn’t permit me as full a response as I’d like – especially since what I do have to say doesn’t have much to do with Kant. (I hope Salmieri and I can return to these topics in some future venue.) In particular, to say as much as I’d like about how I see the role of concepts in knowledge would take us too far afield – though I can point in the relevant direction in a footnote.[1] So I’ll make just a few points about the ethical and political side of things.

I. Concerning the illiberal aspects of Aristotle’s philosophy, Salmieri says “it’s not obvious to me that Aristotle had a method or thesis that would or could have prevented these flaws if it were more expertly applied.” In response, let me quote myself from another context:

On an Aristotelian virtue-ethical account, right action is action that expresses the attitudes and dispositions appropriate to a flourishing human life, where the latter is conceived as a life that gives primacy to the exercise of distinctively human capacities. A life aiming primarily at sensual pleasure, or at mere survival, is rejected as subhuman, since it focuses on capacities that humans share with the lower animals, rather than being organized around the exercise of distinctively human traits. But superhuman lives are ruled out as well. Aristotle does urge us to strive for as godlike an existence as possible, but he makes clear that our human nature places constraints on this goal, and that actually becoming a god would not be a benefit for a human. Hence, the best life for a human being is one that navigates between the extremes of subhuman and superhuman ….

The Aristotelian virtues, too, can be seen as a mean between the subhuman vice of overvaluing, and the superhuman vice of undervaluing, our vulnerable embodiedness. To err on the side of the beasts is to be excessively concerned with our animal nature, our physical desires and physical security; this is the error of the common people, whom Aristotle regards as all too prone to take pleasure and material advantage as their primary goals, and to neglect the possibility of higher values that may require us to sacrifice comfort or even continued existence. To err on the side of the gods, by contrast, is to treat human beings as disembodied intellects for whom the animal nature is irrelevant; this is the error of philosophers like Socrates who see knowledge and virtue as sufficient for happiness, and dismiss external goods as unnecessary, aiming for a transcendent self-sufficiency that is not an option for embodied beings like us. …

Aristotle identifies the distinctively human capacity for reason and speech as the basis of our being naturally political animals, for it enables us to pursue our goals through discussion with one another. Moreover, Aristotle famously regards logos, reason or speech, as the essential trait around which a flourishing human life must be organized. This, it seems, is why Aristotle regards it as an essential component of a truly human life to deal with others politically, i.e., through reason and discourse – i.e., as conversation partners. But such an ideal creates a strong presumption against the use of force, and in favor of relying on persuasion as far as possible. Aristotle indeed affirms that it is unjust to rule by force rather than persuasion, insisting that statesmen should be as dependent on the consent of their subjects as doctors and pilots are on the consent of their patients and passengers respectively. I think, however, that Aristotle’s insight points in the direction of a more radical critique of force than he is likely to have recognized. To deal with others by force is to act in a subhuman manner, like a beast of prey; we live a more human life (and therefore, in Aristotelian terms, a better life) to the extent that our relations with other people embody reason and persuasion rather than coercion. Therefore, the need to avoid the bestial type of vice gives the virtuous agent reason to accept an obligation to respect other people as ends in themselves, rather than to treat them as mere means to her own ends. If this high-level human end places a constraint on the pursuit of lower-level, animal ends, so be it.[2]

In short, then, I do think some very foundational Aristotelean principles in fact provide the basis for an approach to politics much more libertarian than his own.[3]

II. Salmieri points to Aristotle’s view that “the value of productive activity is extrinsic to it” as the root of many of the illiberal features of his politics. I think that’s mostly right – though the contrary mistake, the Puritan overvaluation of labor, can lead to illiberal results too (a technophobic disapproval of labor-saving devices, for example). Surely productive work is sometimes merely instrumentally valuable and sometimes not.

In any case, I just want to point out that Aristotle is sometimes willing to entertain a more nuanced view of productive activity. He writes, for example:

Each person cherishes his own work more than he would be cherished by the work if it became ensouled. This happens most of all, like as not, in the case of poets; for they cherish their own poems excessively, holding them dear as offspring. … Now the cause of this is that existence is choiceworthy and lovable for all, and we exist in virtue of activity (for we exist in virtue of living and acting) and the work is, in a way, the producer in activity. He holds the work dear, then, because he holds existence dear as well. And this is natural, since what he is potentially, the work manifests in activity.[4]

Aristotle’s suggestion that we love the products of our labor because they are, “in a way, the producer [i.e.. ourselves] in activity” certainly comes close to the idea that activity can be valued for its own sake because it is productive of products beyond itself. 

III. This same idea – that a means can acquire constitutive value in virtue of its instrumental value – seems to be at the root of Salmieri’s suggestion that for Rand, “moral values and virtues are essential constituents of the ultimate end that is a person’s life, and they owe their status as constituents to the causal contribution they make to the sustenance of this life.” Salmieri offers this formulation as a way of avoiding the choice between treating virtues as instrumental and treating them as constitutive. But while I like the idea of basing virtues’ constitutive value on their instrumental value – also a common move in Stoicism and Epicureanism, as I read them – I don’t think it succeeds as a way of avoiding the choice I mentioned.

If virtues are constitutive elements of the ultimate end because of their causal contribution to that end, then their instrumental role must be logically prior to their constitutive role. In other words, the end to which virtues make a causal contribution cannot already be characterized as having those virtues as constituents. It would be circular to say that this is valuable because it contributes to that, if that is already conceptualized as including this. Hence the life to which virtues are constitutive means must be a more richly and thickly characterized version of life than the life to which virtues are initially identified as instrumental means – leaving Rand with the problem of deciding which of these two versions of life to identify with the ultimate end. 

I think, of course, that the richer version is the way to go; but if one takes that route, one then has to abandon the earlier, narrower conception as merely provisional – a passing “moment” in the conceptual construction of the ultimate end. Rand’s view, reaching forward to grasp the richer version while refusing to let go of the narrower version, strikes me as a case of wanting to eat one’s cake and have it too.[5]