There is a rather curious tension in Kantian scholarship: Kant has recently been hailed as a radical precursor to modern feminism, yet modern feminists can easily find a deep-seated misogyny in what Kant actually wrote about women. Some interpreters take Kant’s theory of rationality and his moral theory, and the central themes of equality, autonomy, and respect for persons, as providing the basis of "a robust and liberating feminism" (Mosser 1999; see also Schott 1998 and many of the papers in Schott, 1997). Indeed, his moral theory entails some (very radically, for his time) egalitarian conclusions, regarding the rights and duties of every human being, qua human being. The generality and universality of his moral theory—particularly its application to all rational beings and thus to all human beings irregardless of gender—underscores the power and generality of Kant’s moral philosophy. Marvin Fox (1949) highlights this in his introduction to The Fundamental Principles for the Metaphysics of Morals (Abbott translation):

Several writers have criticized Kant for failing to avoid this "all too common pitfall", however. Many of Kant’s comments about the body, sexuality, and especially about women and women’s place in the family and in society, certainly show their author to be very much a—rather typically misogynist—man of his times. Most of Kant’s moral pronouncements on such matters reflect the particular views of bourgeois 18th century Germany. Because of Kant’s misogyny, his "disdain for the body" and his conclusions about marriage, sex and women, Barbara Herman (1993), describes Kant as "the modern moral philosopher feminists find most objectionable" (p. 50). In light of this, and the fact that Kant argues that such conclusions are deductions from the Moral Law and the Laws of Nature, it is understandable that many feminist authors feel justified in minimizing Kant’s relevance to feminism. The recent emergence of concern with feminist interpretations of Kant would seem to indicate, however, that this dismissal is an unfortunate consequence of particular pronouncements Kant made, rather than a problem with his entire system.

Furthermore, on a closer inspection, many of Kant’s particular conclusions about women seem to straightforwardly contradict conclusions he draws about the rights and duties of all human beings. Trying to make sense of these internal tensions is what initially drew me into this topic: How is it that Kant can argue that all human beings are naturally equal and free, and that nobody can give away this natural freedom and equality, nobody can enter into a contract that makes them cease to be their own master, yet also argue that a marriage can make the husband the master of his wife ("he the party to direct, she to obey")? How can Kant further argue that this kind of master/servant relationship between a husband and wife "cannot be regarded as conflicting with the natural equality of the couple" (The Metaphysics of Morals (MofM) p. 98)? How can Kant oppose people having special rights and privileges by virtue of the status into which they are born, and argue that all human beings must have the ability to become active citizens (be able to vote), while automatically excluding all women from active citizenship, arguing that no woman—no matter how astute, financially and politically independent, rational or capable she is—can do what (at least in principle) the poorest and most dull-witted of male serfs and apprentices can do?

One of my aims here is to—as much as is possible—make sense of these tensions. In particular, I will show that these conclusions are not in fact derived from a Kantian account of the Moral Law. In addition, I will explore the extent to which, in spite of, as Fox puts it, his "emphasis on reason alone as the foundation for the correct moral judgement," Kant makes the simple mistake of taking "the particular moral ideals of [his] own society" and arguing that they "are, or ought to be, universally recognized". In doing so I aim to treat Kant in accord with what he expresses as one’s duty to respect an author in the use of their reason:

Thus in addition to making the—by now rather trivial and pedestrian—accusation that Kant elevated particular morals and laws of his society to the status of universal truths, I’ll also give a—more demanding and interesting—explanation of why he thought that these judgements were universally applicable. I’m going to explain how Kant could have thought that particular features of the local, contingent laws and mores of middle-class 18th century Germany could be universal truths, deducible through reason alone.

Several others have attempted to explain these conclusions. Susan Mendus (1987, p. 36) argues that Kant was just unable to distinguish between what is contingent in his society and what is a command of reason. Howard Williams (1983) contends that Kant’s recognition of the ideal state in which all were free, autonomous and equal had to be mitigated by the realities of putting these ideals into practice. According to Williams, Kant had to temper these ideals with pragmatic concerns about the nature of human relationships and of women’s and men’s particular natures, and the best way to get as close as possible to an ideal society. Morris Cohen (1962) suggests that, to Kant, women being subservient to men and being passive rather than active citizens, are direct consequences of the laws of nature and the moral law, and thus are necessarily so. While there is certainly some textual support for these interpretations, I will offer an alternative explanation for why Kant did not draw the egalitarian conclusions many think that one should draw from his moral theory; one that tries to address concerns and views to which Kant explicitly refers, about the connection between the Laws of Nature and the Moral Law.

I will reconstruct Kant’s reasoning, or in some places I should say that I will construct that reasoning, since Kant often seems to arrive at conclusions without showing exactly how he got there; such as claiming that this can be derived from that, without showing exactly how it can be derived and what extra premises are needed. At the same time, I intend to uncover "the subjective ground that determined his judgements that, by an oversight, he took to be objective"; to identify the flaws in the arguments. And there are such flaws. The arguments Kant uses repeatedly incorporate empirical observations and pragmatic judgements local to 18th century Germany as (often unarticulated) premises; premises which are taken to be universally true, yet which in modern Western societies at least, are generally agreed to be false.

The flaws in Kant's arguments are logical as well as factual. We might forgive someone living in the eighteenth century for not knowing better than to advance arguments that are variations on tried-and-true traditional theme: that women are "inferior" to men in some important way; that, because of this weakness or inferiority, women need (for their own good, or for the good of the marriage, or for the good of society) to be constrained in some way; that, also because of this weakness or inferiority, women lack a requisite for active participation in the political life of the society, and so on. Forgiving Kant, however, is more difficult to do. There is much in his moral and political writings that straightforwardly contradicts or undermines some of the premises in his arguments. Kant’s argument that the husband’s right to be in charge "cannot be regarded as conflicting with the natural equality of the couple" (MofM p. 98) highlights logical moves Kant makes which seem rather indefensible for someone as unquestionably astute as Kant. I’m going to concentrate on his arguments that a woman must of necessity obey her husband, and that women cannot be active citizens and represent themselves in civil matters, laying out some of the tensions with other conclusions Kant makes about human beings in general. I’ll also show some of the—to my mind rather sneaky—tactics Kant employs to make some of the conclusions more palatable.

One purpose of this paper is to point out where some of these errors lie and to attempt to make sense of Kant’s position on these matters. The principal reason for trying to explain away these internal tensions, is to attempt to rehabilitate some of the (admittedly debatable, from a feminist perspective) moral insights of Kant’s system, by freeing them from the weight of the disagreeable conclusions Kant drew about the moral, marital and legal status of women. I aim to show that these pronouncements can be seen, certainly not as defects in Kantianism, nor as flaws in Kant’s ability to reason morally, but as a result of an admirable, but questionable, metaphysical picture of the place of humankind in the natural world.

Unfortunately, I do not think that Kant can be entirely excused on these grounds. When some of the rather devious logical moves and circular reasoning he employs are laid out, he can also be seen in a more malicious light. Accordingly, it is natural to ask a further question: is Kant simply endorsing unthinkingly the views of his time, not being critical enough to see that some of the mores and laws of his society might justly and morally be otherwise? Is he simply uncritically supposing that the way he found women’s "nature" to be—a "nature" constrained and shaped by the laws and mores of his time—is a universal and necessary aspect of the "nature" of women? Alternatively—interpreting him more maliciously—is Kant reacting against (perhaps potential) criticism, and defending his culture’s views about women? Is he "force-fitting" the mores and laws of his time, by attempting to show how these laws and mores are in accord with the moral law, rather than deriving the conclusions that ought to follow "naturally" from the moral law? The fact that someone as logically adept as Kant supports his more disagreeable conclusions with the devious rhetorical moves and fallacious argument forms I will show him to be using, I argue, points towards the latter interpretation. I think that seeing Kant’s own particular political objectives in this light can likewise free some of the rather useful aspects of the Kantian approach in general from the rather disagreeable particular pronouncements Kant himself made. We can safely ignore these particular pronouncements, while not necessarily rejecting their (alleged) Kantian foundation, since these pronouncements are not in fact natural conclusions from that foundation.

In general, equality, reciprocity and symmetry feature prominently in Kant’s descriptions of the relations between people:

When Kant discusses marriage this seems especially so:

Such passages focussing on the equality and mutuality of possession in marriage prompt Zweig (1993, p. 300) to applaud Kant’s "enlightened departure from the one-sided prejudice that in marriage the husband owns the wife and her property."

Kant also endorses the idea that all people are born equal (he is opposed to inherited titles of nobility; see MofM, p. 139). He maintains that no-one can lose this natural equality, apart from having it legally revoked through being found guilty of committing a crime:

Nevertheless, Kant supports and attempts to justify laws which "say of the husband’s relation to the wife, he is to be your master (he is the party to direct, she to obey)" (MofM, p. 98). It seems surprising that Kant could endorse laws which seem obviously in conflict with his belief that people are naturally free and equal, and that this cannot change—one cannot cease to be one’s own master—merely through any contractual agreement. In examining the way Kant supports such laws in detail, I attempt to make sense of this tension and to draw out exactly where the errors in Kant’s argument lie.

One way to interpret this tension, is as an ideal theory, and the adjustments that must be made due to pragmatic considerations and to human nature in order to apply the theory in practice.² Williams (1983, pp. 179-81) interprets the tensions in Kant’s views on women in this way: "The woman preserves her natural equality in principle, but forgoes it in practice for the sake of the common interest she shares with her husband." On this view, although all human beings are equal in the ideal—in theory—it happens that practically, in order for marriage to achieve its ends, one partner must be superior to the other:

So for there to be a harmony, or unity, of will between the partners in a marriage, argues Kant, it’s practically necessary that one partner is the master; one person must command, the other must relinquish the natural equality they would possess in the ideal situation, and obey. Otherwise there will be "nothing but wrangling" and progress towards the purpose of marriage (whatever that purpose is) will be slow, perhaps stalled.

Two ironies that arise from this passage deserve highlighting, before we consider whether this theory/practice distinction is an appropriate explanation for the above tension. The first is that Kant seems to keep the idea of reciprocity alive here. The husband is superior in one respect, and the wife is superior in another.

I say it "seems" so because the wife’s superiority is rather hollow, and in the end doesn’t carry much weight. Zweig (1993, p. 291) remarks about this passage that "the ‘superiority’ of the woman to which Kant alludes should not mislead the reader into thinking that Kant saw women as entitled to power."

The second irony is that Kant does speak in other places of a "unity of will" arising in a marriage and arising in a friendship, neither of which seem to involve dominance and superiority. In marriage a "unity of will" arises through mutuality, in reciprocity of possession of each other:

Thus by reciprocity, in giving of one’s person to one’s spouse, and by the spouse doing the same, a couple become a unity of will. It is by this reciprocal arrangement, argues Kant, that we can "apprehend by reason how a commercium sexuale is possible without degrading humanity and breaking the moral laws" (LE p. 167). This arrangement is quite contrary to his remarks that a harmonious union can only be achieved if one partner is subject to the other.

Additionally, Kant argues (MofM, pp. 261-63) that "a harmony of the will of one with another" is possible in a quite different way; this time in a friendship. For friends this is possible without the necessity for one friend to be superior to the other. This is achieved through the same kind of reciprocity of concern that he describes at the end of the above passage. Good friends can harmonize their wills through "sharing sympathetically in the other’s well-being" (MofM, p. 261), by each reciprocally taking on the other person’s ends as their own.

Kant’s formula for achieving a unity of will in any relationship (especially friendship) between rational beings is one that embraces two factors: love and respect.

Admittedly, however, Kant is speaking here of the ideal friendship, the sort that would not be found in the real world.

Kant thinks of it as impossible, because it’s impossible to know if the friends’ love and respect for each other is in fact equal. This ideal, however, is a practically necessary one, in that Kant says that we have a duty at least to strive towards friendships like this (MofM, p. 261).

It wouldn’t be unreasonable, it seems, to paint marriage in similar terms: equal mutual love and respect as an ideal towards which the partners in marriage should strive. This is especially so when we consider that marriages, like friendships, are "moral relations of rational beings, which involve a principle of harmony of the will of one with that of another". This would also fit the "ideal of theory tempered by practical necessity" interpretation of Kant’s reasoning.

However, perhaps because of the absence of respect and love in his conception of marriage, Kant doesn’t consider that a "harmony of will" based on mutual love and respect could, even in the ideal case, exist between a husband and wife in the same way that it could, at least in the ideal case, exist between friends. ³ Friends and spouses are obviously very different sorts of relationship to Kant.

So we could, with Williams (1983, pp. 179-81), interpret the tension between Kant’s emphasis on the natural equality of all people, and his endorsement of laws which make the husband the wife’s master as being due to the ideal of the theory being tempered by practical reality. Thus Kant could be seen as allowing people to forego the natural equality which he praises as an ideal in The Metaphysics of Morals, because of practical necessity given the nature of men and of women and the purpose of marriage. It could be that, to Kant, one of the partners must relinquish their ideal equality, in order to further the practical aim of promoting the common interest of the couple.

This sort of move could be seen to be being made in the passage from Theory and Practice cited above, but Kant’s claims about the natural equality of all people directly counter this interpretation. He does not speak of equality as only an ideal, which can be relinquished for practical reasons. In fact, he says that no-one can forego their equality, especially through entering into a contract (T&P, pp. 75 and 76). It’s a necessary aspect of humanity, which is in some sense inalienable.

Another reason for rejecting the theory vs. practice interpretation of this tension is that his defence of laws that make the husband the wife’s master is given in the Metaphysics of Morals, which is supposed to express the non-empirical, a priori part of ethics. So Kant doesn’t see the wife as having to relinquish her equality for reasons to do with empirical, pragmatic considerations. Or at least he shouldn’t be expressing them here if this is the case. (As I’ll explain soon, however, Kant doesn’t seem to think that either partner has to relinquish their natural equality.)

But however we excuse Kant’s mistake, it is a mistake. This first premise in Kant’s argument for his conclusion that the wife should be subservient to the husband—the premise that one person being in charge is the only way the unity of will required in order to serve the purpose of marriage could be maintained in marriage—is one which Kant takes to be necessarily true, yet it’s one which we can see to be contingent. As I have already mentioned, there are other ways a unity of will can exist between two people; some of which (e.g. friendship based on mutual love and respect) Kant himself comments on. Nonetheless, to have a unity of will in marriage, Kant mistakenly maintains that it’s necessary that one of the partners must be the master.

His second premise is about which partner should be in charge. He says that the person in charge should be the person better able to further the common interests of the household (MofM, p. 98). Conceding, for the sake of argument, the premise that someone must be in charge in order to achieve a unity of will, this second premise seems felicitous. Granting that furthering the common interests of the household is one of the central purposes of marriage, the person in charge should be the person best able to do this job.

The next premise is one driven by empirical observations (but it’s one found, surprisingly, in the MofM (p. 98); as is most of this argument). Practically, the person best able to further the interests of the household is the husband, says Kant. The husband’s dominance, he says, should be "based only on the natural superiority of the husband to the wife in his capacity to promote the common interests of the household" (MofM p. 98). Presumably this "natural superiority" is due to the man’s "naturally superior" strength and courage (Anthropology, p. 219).

This empirical claim that the husband is naturally superior to the wife in the capacity to promote the common interests of the household, is one which contemporary readers would see as mistaken. It is mistaken for two reasons. It’s mistaken because having superior strength and courage doesn’t necessarily make one the person best able to further the interests of the household; having cunning, wisdom, intelligence, sensitivity to the needs of others, and financial acumen might be seen to be among the more important characteristics. It’s also mistaken because there is little reason to suppose that the male is naturally superior in all these characteristics. Kant (as far as I’m aware) does not attribute wisdom, intelligence, or acumen to either gender in higher proportions.⁴

Of course, Kant could see it to be easier for the male to take the commanding role rather than the female, because of respectability and fear. It was probably easier for Kant to see the wife taking orders from the husband, and obeying, than for the man to be obeying a woman. The woman would obey, because the man is often physically stronger. So for pragmatic reasons, men being in charge makes more sense to Kant than the woman being the master.

Kant’s argument so far, is that in order to attain the unity of will necessary in marriage, someone has to be in charge. This should be the person best able to carry out the task of furthering the interests of the household. Kant takes the male to be most able in this regard. And although Kant appears to view each of the premises in this argument as true a priori each premise can be seen, from our contemporary viewpoint, to be empirical and contingent—except perhaps the premise that if someone is in charge, then that person should be the person best able to further the interests of the household.

Now comes the especially disputable and disagreeable part of Kant’s argument. The husband’s legally supported right to be in charge, based on his superior ability to further the interests of the household, Kant puzzlingly argues, "cannot be regarded as conflicting with the natural equality of the couple" (MofM, p. 98). This seems to undermine Williams’ (p. 119) interpretation of the reason for the inequality, that the wife preserves her natural equality in principle, but voluntarily renounces it in practice for the sake of the common purpose she shares with her husband. There’s no need for her to do this, because there is, to Kant, no conflict between the husband being the wife’s master, and the natural equality of the couple. How Kant manages to perform this piece of logical prestidigitation is worth examining closely.

There is no conflict between the husband’s dominance and the wife’s right to equality, argues Kant, because the husband’s right to dominate is derivable from the duty to be unified and equal with respect to the purpose of marriage (MofM, p. 98). Kant does not show how this derivation might work, however. But what he does say is worth paying close attention to, because there is a glaring flaw in this argument. The flaw is the kind of flaw of which it is hard to imagine someone as logically adept as Kant being innocently unaware.

I have already given much of the argument deriving the husband’s right to be in charge from the couple’s duty to have a unity of will and of activities. As I have already explained, Kant holds that in order to produce this unity, one partner must be in charge, directing the actions of the other. This "master" should be the husband, since to Kant the male is naturally the best person for the job. The important extra premise Kant adds here is that the unity of will that ought to be present in marriage is derived from "the duty to unity and equality" (MofM, p. 98). Thus, for Kant, since the conclusion that the husband ought to be the master is derivable from the duty to unity and equality, a law based on these reasons that gives the husband the right to dominate his wife is not in conflict with the couple’s duty to equality.

But of course it is in conflict! It’s in conflict not only with our early twenty-first century sensibilities, but it’s also in conflict with conclusions Kant himself draws from the moral law. To highlight this conflict, and to show why it seems that Kant has little excuse for arguing for the absence of any conflict, it will help to pay attention to a distinction between what we might call natural equality, equality among people as human beings, and legal equality, equality among people as subjects of the state. This distinction initially appears to be important in distinguishing the legally sanctioned inequality within the household from the more important morally required natural equality of all human beings, qua human beings. However, although the two types of equality seem to be different, when it comes down to it they turn out to be practically identical, since legal equality must always accord with natural equality. An explanation of the distinction also affords a clear view of what Kant means by "the duty to equality" in this context.

There is only one innate right, says Kant (MofM, p. 63), "belonging to every man by virtue of his humanity": the right to "Freedom (independence from being constrained by another’s choice), insofar as it can exist with the freedom of every other in accordance with a universal law." This innate right to freedom, he continues, involves and is not distinct from an innate right to equality. This is a right that "belongs to everyone by nature." This natural equality involves "independence from being bound by others to more than one can in turn bind them." Natural rights like this rest purely on a priori principles to Kant, and thus do not need to be legislated for.

We can contrast this natural equality with what we might call the principle of legal equality. This legal equality involves the equality of each with all the others as subjects before the law. This must be in accord with the natural equality of all people, since, as Kant describes it, the state (and thus presumably the state’s laws) must be "in accordance with the pure rational principles of external human right," (T&P, p. 74) one of which is the right to natural equality. So the rights people have through legislation must not conflict with people’s natural rights.

We can also contrast both this legal equality and this natural equality with what might be called the natural inequalities that exist between people, by virtue of their condition. Kant claims that this sort of inequality is perfectly consistent with the legal equality of all as subjects before the law. Kant –perhaps rightly– identifies the inequality between rich and poor as an inequality arising between people by virtue of their possessions, an inequality which does not conflict with people’s right to be treated equally as subjects under the law (T&P, p. 75). But next he identifies the wife’s obedience to her husband as an example of an "inequality of... particular rights... with respect to others" (T&P p. 75); another inequality which he claims is consistent with all people’s legal equality.

I see this claim as incorrect. The wife’s subservience to her husband is directly in conflict with what Kant expresses as her right to natural equality. Her natural equality in virtue of her humanity means that she has the right to "independence from being bound by others to more that one can in turn bind them." And since subjects’ legal equality must accord with this right to natural equality, the wife’s obedience to her husband is also in conflict with her legal equality.

So when Kant claims, in his defense of laws which make the husband the wife’s master, that the husband’s right to be the wife’s master is derivable from a duty to unity and equality, whether he refers to legal equality or to natural equality makes little difference. From either we can derive that no person should be bound by someone to more than they can in turn bind them. We can also agree with what Kant himself concludes from people’s natural equality: that no person can relinquish their natural equality and make someone else their master. To do so would be to relinquish your humanity, which is inalienable.

Now I want to look closely at how Kant defends these laws that give the husband the right to be his wife’s master (MofM p. 98). He does so by arguing that it does not conflict with the natural equality of the couple. This lack of conflict, he argues, is because the husband’s right to direct is derivable from the duty to unity and equality with regard to the purpose of marriage. There is a conflict, as I’ve just shown, and Kant employs a rather insidious piece of logical sleight-of-hand in what appears to be an attempt to disguise that conflict.

The conjurer’s trick is to conjoin the duty to unity with the duty to equality. The husband’s right to dominate is apparently derived from the conjunctive duty to unity and equality. It is not. In fact, it is derived from the duty to unity alone (in the way Kant sees that duty, as I explained earlier). The duty to equality is just "along for the ride," so to speak; it plays no part at all here. But by claiming that this conclusion is derived from the conjunction of these two duties, Kant has made it appear that the husband’s right to dominate is derived from the duty to equality, and thus is not in conflict with the natural equality of the couple. Kant has made what should be a glaring contradiction look perfectly innocent, by conjoining the duty to equality together with the duty to unity with regard to the purpose of marriage, and claiming that the husband’s right to be the master of his wife is derived from this conjunction, when it’s only derived from one of the conjuncts.

There is an easy way to highlight the contradiction here: an almost textbook argument by reductio ad absurdum. Kant holds the duty to equality as fundamental, inviolable, where the duty to unity merely facilitates efficiency in following one’s duty to promote the purpose of marriage. So let’s start with the more fundamental duty to equality. From the duty to equality, and from Kant’s stipulation that no-one can voluntarily renounce their natural equality and cease to be their own master, not even through a contract (T&P, p. 75-76), we can conclude in one step that neither partner in marriage should, or could, become the master of the other through the consummation of the marriage contract. If we now bring in a duty to unity, we get a direct contradiction. From this duty to unity with respect to the purpose of marriage (as Kant conceives of the marriage relationship; where friendship, love and respect are conspicuously absent), Kant concludes that one of the partners in marriage must be the master of the other. So, if we already affirm the duty to equality, then introducing the duty to unity gives us a contradiction. The duty to equality gives us that neither partner can be the master of the other. And the duty to unity with respect to the purpose of marriage gives us that one partner must be the master of the other. So if we start with a duty to equality, then we can conclude by reductio ad absurdum that it is not the case that there is a duty to unity. These two duties are incompatible.

A second, more benign, conclusion is that a duty to unity as Kant conceives of it here, is incompatible with the duty to equality. That is, the duty to unity should be compatible with a duty to equality, and so it must manifest itself in some other way than the subservience of one partner to the other. And given that Kant does see that a unity of will between two rational beings being possible in ways other than one person being in command, such as in a relationship based on mutual and equal love and respect, a reformulation of the consequences of the duty to unity and equality would be a more felicitous conclusion.

The duty to equality must take precedence. This is because, to Kant, the duty to natural equality is a pure rational principle, which rests only on a priori principles (MofM, p. 63). The duty to unity with respect to the end of marriage, in contrast, is merely a pragmatic recommendation designed to make the marriage work more efficiently towards its end, by eliminating the inefficiencies caused by "wrangling" between the partners. Thus a law that says of the wife "he is to be your master" is an unjust–even immoral–law, because it directly contradicts the innate right to equality. And the custom that the wife obey the husband is just as immoral.

So if there is both a duty to equality and a duty to unity with respect to the purpose of marriage, these must manifest themselves in a different sort of marriage than the master/subject relationship envisaged by Kant. These duties dictate a marriage where a unity of will is achieved, in something like the way Kant sees it achieved in friendship, by a duty to strive towards a relationship of equal mutual love and respect, with a proper balance between the closeness generated by love and the distance required by respect (MofM, p. 261). Thus each partner takes on the ends of the other as their own ends too, so that each is concerned for the welfare and happiness of the other, while also respecting the other. (Although Kant’s conception of respect, as being a limitation of intimacy so that the two do not make themselves too familiar to each other (MofM p. 261) might be disputed.)

Kant’s view of marriage as a master/servant relationship is a significant factor in the second area I want to look at in detail: his view of women as citizens and their right (or rather, their lack of right) to play an active part in the political life of the community. Not being one’s own master, being under the direction of another, is one of the criteria that disqualifies people from playing a role as an active citizen in the political life of the community. Additionally, the reason why women can’t play an active role in politics is also a reason for Kant’s view that women should be obedient to their husbands; a deeper reason than the premise that it will further the purpose of marriage.

Kant separates the citizens of a commonwealth into two sorts: active citizens who have the right to play a part in the political process—that is, to vote—and who thereby play a part in the creation of laws, and passive citizens who do not have the right to vote. To have the right to vote, one must have what Kant calls civil independence. Someone with such civil independence is someone "owing his existence and preservation to his own rights and powers, not to the choice of another among the people" (MofM p. 125). Whoever has a say in the creation of the commonwealth’s laws, by voting, must have this civil independence. This is because legislating for a commonwealth,

requires freedom, equality and unity of the will of all the members. And the prerequisite for unity, since it necessitates a general vote (if freedom and equality are both present), is independence (T&P, p. 77). 5

To be an active citizen then, one must be one’s "own master" (T&P, p. 78); one must be an independent member of the commonwealth capable of "acting from his own choice in community with others" (MofM, p. 126). So if you are under the direction or protection of another, and thus do not act purely from your own choice, you can’t be an active citizen; that is, you are not fit to be allowed to vote. This, presumably, is because whoever directs you or protects you (and thus to whom you owe a debt of gratitude) could influence the way you vote.

In spite of this stratification, however, social mobility is an inalienable right to Kant. This right arises from the natural, and thus the legal, equality of all citizens:

This social mobility is also in accord with Kant’s in principle opposition to any person having special rights or privileges purely by virtue of the way they are born:

It seems from these comments that if a woman, through her talent, industry and good fortune, managed to get to a position where she was her own master, and able to support herself, so that she could make choices of her own free will, independently of the influence of any other person, then that woman should have the status of active citizen, and thus the right to vote. A law that legislates otherwise would be an unjust and immoral law.

Yet, surprisingly (or perhaps not surprisingly, given his ideas about women in other areas), Kant categorically denies women this right. As the passages I cited earlier attest, Kant believes that to participate in the political life of the commonwealth, a person must have civil independence. Any person who has civil independence can be an active citizen. And any person can ideally work up to active citizenship, by becoming civilly independent; that is, by gaining the ability to ensure his own preservation and existence independent of the influence of others (MofM, p. 126). But in Kant’s system a woman cannot, even in principle, work her way up from passive to active status. Women are automatically excluded from active citizenship. When discussing who should qualify as an active citizen, the first qualification Kant makes is that an active citizen must be an adult male (T&P p. 78; MofM p. 126), in spite of his claim that every person should be able to work up from passive to active status (T&P p. 78; MofM p. 126). So it seems that social mobility, based on your natural equality as a human being, is an inalienable right; except if you’re a woman. This, among other comments, leads Zweig (1993, p. 297) to comment that:

Given that women are rational agents capable of recognizing and endorsing moral laws—and Kant did see women as rational beings (Anthropology, p. 216)—it seems that Kant should reject as unjust any law which treats women differently than men. This, unfortunately, he does not do.

It seems that Kant believes that women must necessarily be dependent on men; a woman will always owe her existence and preservation to someone else, presumably a father and then a husband. Thus women can never be active citizens. Although Kant does not explicitly state why this is so, it is possible to lay out what appear to be the premises in his argument for the conclusion that women are necessarily dependent on, and subservient to, men. (Again, Kant doesn’t actually articulate this argument, but merely implies it, and bases his arguments on its conclusion.) In what follows, I offer an interpretation of Kant’s position, which illustrates how he could have coherently believed such premises to be objectively true.

As an initial attempt on a construction of the argument, recall that Kant describes the qualification for active citizenship as being capable of acting on one’s own choice in community with others, being someone neither under the protection nor the direction of anyone else. From a person’s civil independence—being able to ensure their own existence and preservation—Kant argues, "follows his civil personality, the attribute of not needing to be represented by another where rights are concerned" (MofM, p. 126). Thus if you don’t have the ability to stand up for your own rights and to look after your own property and earn your own income from that property, then you lack civil independence, and thus you need to be represented by another where your rights are concerned.

Women have "weaknesses", says Kant (Anthropology, p. 105), which necessitate their being in a state of legal tutelage (he states neither what these weaknesses are, nor whence this necessity arises). This "tutelage" means that the person’s "inability to use his own understanding in civil matters" (T&P, p. 105) is based on legal arrangements, not on any deficiency in that person’s abilities (which makes the claim about "weaknesses" seem even more hollow). Thus the law requires women to rely upon the protection of someone else when it comes to such matters; somebody else must speak for them. The legal system in Kant’s time also denied women the right to administer their own property:

Thus since a woman is not "able" to represent herself, to stand up for her rights in court, but must be always under the protection of another, and because she must employ a representative to administer her property, women therefore lack civil independence, and thus can never be active citizens.

Yet Kant seems to reverse the flow of reasoning here. His original premise was that if someone lacks civil independence, then that person needs a custodian to represent them in civil matters. In his argument about women’s legal position, he is committing the fallacy of affirming the consequent, arguing that since women require someone to represent them in civil matters, they therefore have "weaknesses" which entail a lack of civil independence.

But the reason for women’s lack of ability to represent themselves isn’t a consequence of their lacking the requisite abilities for civil independence. We must be attentive to the sense of "able" used here. On the very same page in the Anthropology, Kant admits that women—by their nature—are quite capable of self-reliance:

Thus it’s not that a woman is incapable of looking after herself and of standing up for her own rights. It turns out that here the "weaknesses" Kant speaks of are not natural weaknesses, but legally enforced weaknesses. A woman isn’t able look to after herself and stand up for her own rights because the legal system prevents her from exercising whatever ability she has. Kant continues:

The law requires women to employ a representative, even though—as Kant admits—many women would be quite capable of representing themselves. But in spite of the fact that at least some women have these abilities, the law excludes all women from having the right to exercise them. So women’s "inability" to represent themselves and to ensure their own existence and preservation, isn’t an inherent inability which would fail them (one that would also fail any man, on his view) in qualifying for civil independence. Rather this "weakness" is due to the legal system not allowing women the right to exercise the abilities they often do in fact have; abilities which would entail that at least some women would qualify for active citizenship, were they unfettered by the law. So at least some women—by their nature—do have the capability to represent themselves and to ensure their own existence and preservation (as do some men). Thus, based on abilities Kant himself acknowledges, women should have the right to earn active citizenship—a right every male has.

It’s unsatisfyingly circular to justify a legal system which denies women active citizenship on the basis of the claim that women don’t have the ability to represent themselves in matters of their rights and their property, when they don’t have this "ability" simply because they are legally prevented from representing themselves, even though they do have the capability to do so.

If Kant relied only on the above argument, however, he’d be in a fairly shaky position. But the above argument, which I’ve reconstructed from passages in the Anthropology, is merely a shallow cut. Kant’s reason for women’s lack of civil independence has deeper roots than mere recourse to the laws of the time. Kant believes that women necessarily lack civil independence; a view which comes out of Kant’s view on women and morality in general. He believes that reason dictates that women must be under the direction of men, in the interests of morality.

The argument, if Kant had ever expressed it, would run something like this: To Kant, the moral course of action is that action which would be dictated by reason (following the Categorical Imperative), that action which a purely rational agent would perform of necessity. But humans are not purely rational agents, we are "animals with reason", we have emotions and inclinations as well as reason. For me to be acting morally then, I must be able to ignore my inclinations and be motivated by my reason alone. Situations in which I perform the moral action, but for reasons other than rationally grasping that this is the moral action, such as being instructed to do it or acting out of an emotional desire for a certain outcome, Kant describes merely as moral behaviour (Anthropology, p. 220). So if people follow their inclinations instead of acting from rational motives, then their actions won’t be moral; they might not even approach moral behaviour if, from emotional motives, they perform actions other than the action which reason dictates. If they did behave morally, this would be by accident.

Although Kant does not explicitly say this, he intimates in several places that women are less capable than men of following the moral law, or performing the action dictated by reason. Contrary to some interpretations, this is not because Kant denies (implicitly) that women’s nature has a connection with reason, or that, as Susan Mendus (1987, p. 36) interprets him, "in the kingdom of rational beings there are only adult males." Nor is it as Zweig (1993, p. 292) claims, that women are "less fully rational than men (though of course even men fall short of total rationality)." Kant quite explicitly states that all human beings, both women and men, are rational (Anthropology, p. 216). Women are just as connected with reason as men are, but women, are more emotional than men, Kant claims in several places (e.g. Anthro, p. 218, p. 224). Women are "too weak to control their emotions," and emotions overpower reason: "To be subject to emotions and passions is probably always an illness of the mind because both emotion and passion, exclude the sovereignty of reason" (Anthropology, p. 155). Thus, to Kant, women are less capable than men of following the moral law, not because they are less rational, but because men are much better at controlling their emotions and are therefore more consistently motivated by their reason alone. In contrast, women are more emotional, so they are more likely to allow their inclinations to motivate their actions, overpowering the rational motive for the moral action.

Thus on Kant’s view men, being more consistently motivated rationally rather than emotionally, are the better moral agents. And since women should at least behave morally (perform the morally required action, even if not motivated by rationally grasping that this action is the morally required action), women must be restrained from following their inclinations, by being told what to do by men. Otherwise they will often behave immorally. So a woman must obey her husband, so that she "can be brought if not to morality itself, then at least to that which cloaks it, moral behaviour, which is the preparation and introduction to morality" (Anthropology, p. 219-220).

This, then, is why women can never be active citizens: to Kant a woman must of moral necessity not act under her own direction. So that she does not act immorally, she must be restrained from following her inclinations, and be controlled by a man—her father or husband, presumably—a man who can know through reason the right thing for her to do, and because of his ability to suppress his emotions and inclinations to tell her otherwise, will instruct her to do what is in fact the right thing for her to do. Thus women must of moral necessity be subject to a man’s choices, and thus women can never aspire to gain civil independence, and thus to active citizenship. This is also part of Kant’s reason for believing that women’s subservience to their husbands is necessarily so, and not contrary to the moral law.

Thus women have a natural equality (as do all human beings), but they also have a natural dependence, because of the moral law and the very nature of women, as Kant saw it. This is one of the places where empirical observations have not been entirely cleared from the Metaphysics of Morals, as Kant expressed was his intention in the Groundwork (p. 4). Kant has given a contingent empirical observation about the nature of women—that they appear to be more emotional than men and thus are less capable of acting according to the dictates of their reason—the status of necessary fact; one which could not be otherwise. Let’s look at several interpretations that attempt to shed light on exactly why Kant could have seen this observation about the nature of women to be necessarily true.

Hannelore Schröder (1997) sees Kant’s reasoning as almost completely politically motivated. She argues that Kant defends patriarchal marriage "Because Kant particularly seeks to protect the ‘one-sided arbitrariness’ that is the legal privilege of men to subject women to their rule" (p. 294). She sees his aim as that of securing the "Patriarchal Order" and the privileges that follow from it for Hausherren (patriarchs). Thus it appears that Schröder sees Kant as actively defending patriarchal advantage, for purely political ends, having nothing to do with morality. In fact, Schröder concludes that in defining a woman as a means to the ends of men (p. 295), Kant is "taking his categorical imperative ad absurdum, canceling it himself" (p. 296). I’m not convinced that this is so. (And I would rather see these conclusions cancelled out by the categorical imperative, rather than allowing this fundamental moral principle to be undermined by a few contradictory conclusions, albeit by the author of that principle.) My reason for not resting here is partly because, as Kant himself argues in the selection I quoted at the beginning of this paper, it seems necessary to uncover the "subjective ground that determined his judgement, that, by an oversight, he took for objective." How did Kant justify to himself (rationalize?) these conclusions so that they seemed to him to be necessarily true?

On a first attempt, this might sit well with Susan Mendus’ (1987) (fairly uncharitable ⁶) interpretation of Kant, that he "fails to distinguish between what is merely contingent and accepted in his society and what is a command of reason." She claims that Kant was just plain unable to distinguish the contingent and circumstantial from the a priori and necessary (Mendus, p. 36). I’d like to give Kant more credit than this; it was certainly not merely that Kant was not clever or insightful enough to realize that what he took for a priori truths were actually contingent facts that could have been—and indeed probably were—otherwise. The reasons for Kant’s belief that these observations were necessarily true runs much deeper than this shallow cut.

Another interpretation, almost as unfitting, is Williams’ (1983) interpretation of Kant’s position, as an ideal of the moral law, which must be tempered by pragmatic concerns. A state in which all members are free, independent and equal, is an ideal of reason; one which we can’t expect to be realized. On Williams’ interpretation, Kant moderates this ideal because of pragmatic concerns and empirical observations of the human condition: "The formal equality of each person within civil society is contradicted by his actual economic and social dependence on other persons" (Williams, 1983, p. 181). And it just happens that while practically, Kant could see that apprentices could work their way up to active status by learning their master’s trade and setting up their own businesses, Kant "does not seem to envisage any circumstances under which women could play an active part in politics" (p. 181). Thus although Kant, on this interpretation, sees the ideal position as one where all people are free, equal and independent, it happens that the nature of women and pragmatic considerations about the best way to achieve the end of marriage, determine that this ideal cannot be achieved in practice.

But it seems that if this were so, then Kant despairs of even attempting to work towards this ideal, while, as I mentioned earlier, insisting we should attempt to work towards the ideal of friendship. Thus Kant, curiously, ends up with the conclusion that the present state of affairs is the best we can hope for; we will not get closer to the ideal than this, and thus the present state of affairs could not (should not, perhaps) be otherwise. Anything else would be further from the ideal. Moreover, given the contradictions between this practice and what Kant paints as inviolable rights to equality, this seems an unlikely interpretation.

A little closer to the mark is Cohen’s (1962) interpretation (which it appears Mendus misinterprets). According to Cohen, Kant believes that women being necessarily dependent on men is a dictate of reason, and Kant thinks this is so because the laws of nature and the moral law make it so. Kant saw women’s emotional nature to be a consequence of the laws of nature, which in Kant’s time were seen to contain a priori propositions (Cohen 1962, p. 296). And the moral law is also "revealed clearly and unmistakably in the conscience of all mankind" (Cohen 1962, p. 297). Thus the conclusion that women are by nature emotional, and thus necessarily dependent on men for moral guidance is derived directly from the laws of nature and the moral law. And since both of these laws are knowable a priori, this conclusion that women are dependent on men for moral guidance and protection is true a priori to Kant, and thus necessarily true.

However, on a fourth interpretation—my own—we can see the nature of women arising, not by being nomologically produced, but by being teleologically directed. The Anthropology From a Pragmatic Point Of View is key here. It is from a pragmatic point of view deliberately. As Holly Wilson (1997, p. 381) argues,

Wilson argues that in the introduction to the Critique of Judgement Kant "Proposes reflective teleological judgement of purposiveness as a theoretical view of nature that allows us to view nature in such a way that our free action is thus enabled" (p. 381). Thus, as a precondition for the possibility of morality, human beings must be seen as free agents, and when we view the natural world as one that allows for this possibility, we are required to view nature teleologically, as itself purposeful. Kant’s understanding of "Nature’s purposes" in the design of human beings are the key to understanding why he sees women’s nature as he does.

To accurately characterize the feminine sex, says Kant, "we must use the principle which served as Nature’s end in the creation of femininity." And this end, he continues, is able to reveal its underlying principle, "which does not depend on our own choice, but on the higher design for the human race" (Anthropology, p. 219). From this underlying principle of the ends towards which women were designed—(1) the preservation of the species, and (2) the improvement of society and its refinement by women—Kant derives two consequences for the nature of women (Anthropology, p. 219-220):

Thus women are of necessity to be both morally guided and physically protected by men. Nature has given women strong inclinations because of the need to protect Nature’s "most precious pledge, namely the species, in the shape of the embryo by which the species was to propagate and preserve itself" (p. 219). These strong inclinations –especially "fear of injury and timidity towards similar dangers"– Kant reasons, could lead a woman to act immorally, even in spite of her recognition that another action is morally required of her. A woman should thus be guided by a man, who does not have that fear and timidity. Women also demand that men give such guidance gently and politely, although Kant appears to believe that for a woman to be brought to moral behaviour and act contrary to her other inclinations, a certain amount of fear and timidity towards her husband would be an effective motivator.

The importance of the union of marriage, and the obligation to preserve the species are also the reasons for the man to have greater strength. Nature has given men greater strength than the female: "...in order to bring both, who are also rational beings, together in intimate physical union for the most innate purpose, the preservation of the species" (Anthropology, p. 216). Furthermore, this strength and this obligation are the basis of the man’s right to be in charge. This can be seen clearly in Kant’s remark that the man "...builds on the right of the stronger to give the orders at home because he has the obligation to protect his home against outside enemies" (Anthropology, p. 217). Thus the woman ought to see her husband, quite literally, as her master (one she fears as well as loves). These, Kant claims, are consequences of Nature’s design; a design whose underlying principles are revealed to humans. And since this is directed by Nature’s design, it is necessarily so. It could not be otherwise.

Whichever of the preceding interpretations—or any other—which might be offered to account for Kant's reasoning in denying women active citizenship, it seems that we can quite justifiably say that Kant was mistaken. He was mistaken about the need for a (gender-based) stratification of citizenship, he was mistaken about women’s nature being essentially more emotional and, because of this last point and most importantly for present purposes, he was mistaken about women being morally dependent on men’s direction. He was certainly mistaken about these being necessarily the case.

The fact that women were dependent on men in Kant’s time is almost entirely a consequence of social and legal conventions, which are by no means necessarily true. Kant saw women as being legally, economically, socially and morally dependent on the guidance, tutelage, and protection of men; a view which has been well falsified empirically since then. And, one suspects, this view was falsified in more than a few instances in Kant’s time, though the falsifying instances were dismissed, often patronizingly and rudely by Kant and his contemporaries (for example, see Zweig 1993, p. 291, n4)

In their book Women’s Choices, Mary Midgley and Judith Hughes argue that we might be able to find excuses for the views of women expressed by more historical philosophers, because they perhaps failed to see that things could be different (although even Aristotle had Plato’s arguments). They continue:

The view of women as essentially emotional, weaker than men, requiring moral guidance from men, and as necessarily subservient to men was certainly being challenged in Kant’s time. For instance, Mary Wollstonecraft’s (1792) A Vindication of the Rights of Woman, challenged many of the claims about the nature and purpose of women at the time. She was especially critical of comments made by Rousseau, who strongly influenced Kant’s picture of the nature of women (see Van De Pitte 1978; Mendus 1987, p. 22). Rousseau, especially in his Emile (1762/1972), was not just blindly and uncritically accepting the generally accepted view of women, but quite blatantly defending this perspective against criticism and pushes towards change. For example, Rousseau says:

We may legitimately wonder which camp Kant falls into. Was Kant was merely unthinkingly and uncritically accepting the laws and mores of his time regarding women? Was Kant suffering from a general human condition which he himself decries as looking to experience instead of to reason:

That is, perhaps Kant simply thinks that in general people were happy with the legal system as it was in eighteenth century Prussia, and that there could not practically be a better system. Perhaps Kant simply judges people’s happiness and rights in terms of the status quo, and prefers this state "to the dangerous task of looking for a better one". But perhaps, like Rousseau, Kant was actively defending those practices against challenges that they are unjust, and reacting against moves towards change. I conclude with some remarks on this last question.

The way Kant arrives at conclusions, such as his claim that the husband being the wife’s master does not conflict with the natural equality of the couple, certainly gives one pause. Recall that Kant did this by conjoining the duty to unity and the duty to equality as an initial premise. He then claimed that since the husband’s right to be the master is derivable from the duty to unity and equality, it is not in conflict with the duty to equality. But Kant’s conclusion that the wife should be subservient to her husband, that he is her master, is derived solely from the duty to unity, and it is in conflict with the duty to equality, a conflict which I showed by pointing out the consequences of that duty given by Kant himself: that nobody should be bound by others more than they can in turn bind them, and that nobody can cease to be their own master. His argument that women must necessarily be passive citizens also gives us reason to suspect that Kant is actively defending this practice. He defends this practice by saying that women are not able to defend themselves and must therefore be in a state of tutelage. However, this lack of ability is defended by appeal to the fact that women are legally prevented from exercising abilities he himself admits they have. Their lack of ability is not a natural incapacity, but a consequence of the very law Kant is defending.

With guileful reasoning like this, by ignoring contradictory conclusions that he himself draws elsewhere (but around the same time), and by ignoring the Categorical Imperative (or refusing to apply it universally, to all human beings), it seems less likely that Kant is simply not being critical enough, that he is just unthinkingly accepting the laws and mores of his culture, and looking to experience instead of to reason. Seen in the light of the purported attempt to exercise his reason, and the kind of reasoning that results, it seems more likely that Kant is reacting to criticism of, and is actively defending, the status quo.

Further supporting this conclusion is a work arguing for an egalitarian society, published in 1792 by Kant’s close friend, Theodor Gottleib von Hippel. Here von Hippel reproaches the French Constitution (formed after the 1789 revolution) for allowing "a whole half the Nation to be forgotten." He continues:

Here we find a close friend of Kant’s demanding justice for women, protesting against the very principles for which Kant argues: excluding women from active citizenship and ruling that the husband is his wife’s master. ⁷ Allen Wood (1999, p.340) further points out that von Hippel’s book, published anonymously, was thought by some to have been authored or co-authored by Kant; a claim which Kant explicitly denied. From this denial and the close relationship with von Hippel, we can presume somewhat safely that Kant probably read the book. He was at least aware of the controversy surrounding it and similar challenges to the way women were treated. For instance, many of these challenges (e.g. Olympe De Gouges (1791) "The Declaration of Human and Civil Rights for Women") appear to have been engendered by the explicit exclusion of women from the French constitution. According to Schröder (1997, p. 276) this caused a something of a political scandal, a scandal that was partly the target of von Hippel’s criticism. However, some of Kant’s more pernicious defenses of women’s subservience and exclusion (in the (1785) Metaphysics of Morals) predate this controversy. Thus defending against challenges such as von Hippel’s cannot be the sole reason for Kant’s defense of such laws.

Because of this historical context, and because of the contradictions we can find in Kant’s writings, it seems that instead of following the natural logical consequences of his moral theory, Kant is defending the practices of his own culture, by trying (unsuccessfully, as I have argued) to show that they are natural consequences of the moral law. Rather than thinking deeply and critically, and letting the conclusions which follow naturally from the moral law emerge from the process, Kant seems to have taken a custom or law of his society, and then attempted to derive this conclusion from the moral law. That is, Kant seems to be "force-fitting" the mores and laws of middle-class 18th century Germany so that they appear to be supported by the moral law.

This seems especially so where those customs are embedded in the laws of his time. Perhaps this defence is a consequence of his conservatism combined with Rousseau’s influence and Kant’s belief that we have a duty to respect the law, and to obey it. On the subject of criticizing existing laws, however, Kant is surprisingly ambiguous. Although citizens should be permitted to voice objections to unjust laws, he counsels that in passing general and public judgements the citizens "must not transcend the bounds of respect and devotion towards the existing constitution" and that it would never be permissible "to offer any verbal or active resistance" (T&P, p. 84-85; my italics). If Kant seriously believed in this duty to not offer even verbal resistance to the laws, then he might see it as his place, given his influential position, not to even "pass general and public judgements" against the laws of his time. But even then, there seems little reason to defend those laws; remaining silent would have been just as much in accord with this duty. And Kant certainly did not do that on this subject.

Thus, since the rather disagreeable remarks Kant makes about women (their necessarily being under the direction of husbands and fathers, and their necessarily being only passive citizens) are not in fact derived from his moral theory, I suggest that feminists’ reluctance to engage Kant’s moral and political theory because of these pronouncements is understandable, but unwarranted. Admittedly, there is much for feminist scholars to disagree with in Kantian theory: feminists can and do justifiably criticize the focus in Kantian ethics on universality, impartiality and rationality, and the focus away from particularity, from partiality and from care. However, objections to Kant himself and the particular pronouncements he made about women should not impede an honest engagement with Kantian ethical theory. The admirable recent attempts to offer feminist interpretations of Kant’s approach, I hope, are simply the beginning of a thorough engagement with this rich and potentially fruitful body of work.

1. Kant's view of sexual relations as using a person as an object, as a means to gratification, and his view of marriage as a way of showing how such "mutual use" of one another can conform to the moral law is an interesting topic in itself. I'm going to bypass this aspect of marriage. For a thorough treatment, see Herman (1993).

2. This is a common theme in Kant: it's found explicitly in his "On the Common Saying 'This May be True in Theory, but it does not Apply in Practice'," and also in the relationship between his Anthropology from a Pragmatic Point of View and The Metaphysics of Morals. The views expressed in The Metaphysics of Morals, are supposedly "completely freed from everything which may be only empirical" (Kant 1785/1995, Preface, p. 5), they express the part of ethics that Kant believes can be known a priori. The Anthropology applies this a priori part of ethics to the phenomenal world; see Fred Van de Pitte's (1978, p. xxi) introduction to the Anthropology.

3. The reason Kant doesn't describe marriage in terms of love and respect could also be interpreted as a consequence of Kant's view that relationships between rational beings are based on love and respect, where he doesn't see marriage as being a relation between such beings. Many writers (Mendus 1987, for example) have interpreted Kant as holding that women are not strictly rational. This, however, is not the case. Kant explicitly says that all human beings, women and men, are rational (Anthropology, p. 216). (I'll have more to say about this later on.)

4. Kant does, however, mock "scholarly" women: "As for scholarly women, they use their books somewhat like a watch, that is, they wear the watch so it can be noticed that they have one, although it is usually broken or does not show the correct time" (Anthropology, p. 220).

5. It seems ironic that Kant argues that a just public law requires the unity of will of all the members of the commonwealth, yet he maintains that only some of the members of the commonwealth are entitled to vote on such legislation. The italicization of "all" in the passage cited is even Kant's (or rather the translator's, following "the style and substance" of Kant's original German). This irony is explained, however. A commonwealth whose constitution is based on such a unity of will of all the citizens, Kant concedes, is an ideal, which "we need by no means assume... exists for a fact, for it cannot possibly be so." (T&P, p. 79) It's not clear that the reason for this is that not all the members of the commonwealth have the right to vote on such legislation, however. The reason is rather that a state, with its attendant constitution and legislation, could not have actually been created by such a unity of will, because such a unity would require that the people were already united under a constitution. But this ideal of reason does have some measure of practical reality, as a benchmark for the legislator to judge the justness of his laws. The legislator is obliged "to frame his laws in such a way that they could have been produced by the united will of a whole nation" (T&P, p. 79). An unjust law is one with which it is not even possible for a whole people to agree.

6. It almost appears that Mendus intentionally takes the least charitable interpretation of Kant, by (among other things) very selectively reading and citing Cohen's (1962) article. Mendus cites the passage where Cohen paints Kant's comments as apparently absurd, and then she halts (Mendus 1987, p. 23). But what Cohen was actually doing was painting Kant's comments as apparently absurd, and then, in the sentence immediately following the piece that Mendus cites, explaining some factors that make Kant's apparently absurd position more intelligible (Cohen, 1962, p. 296).

7. I owe thanks to Elizabeth Brake (personal communication) for pointing this out to me. Schröder (1997) begins with a good summary of von Hippel's arguments.