ISSN 2011-7477 n.° XV, julio-diciembre de 2011 Fecha de recepción: mayo 12 de 2011 Fecha de aceptación: junio 14 de 2011

ARTÍCULO DE INVESTIGACIÓN / RESEARCH ARTICLE

Allen W. Wood
Stanford University, Faculty of Humanities
wood2420@gmail.com

Los estrictos puntos de vista de Kant sobre la mentira han sido regularmente citados como una razón para pensar que hay algo fundamentalmente equivocado en la ética kantiana. Algunas de las afirmaciones de Kant parecen tan desmedidas que la mayoría de los kantianos que se han ocupado del tema han tratado de distanciarse de ellos, generalmente argumentando que no necesariamente siguen los principios de Kant. En este trabajo voy a hacer un poco de eso en parte, al cuestionar si el famoso ejemplo del "asesino en la puerta" "realmente se ajusta a los principios con los que Kant se refiere a éste. Discutiré si los argumentos de Kant sobre la veracidad son razonables o justificables al menos; si no lo es evidente por sí mismo. Esto es principalmente porque también creo que algunos de ellos, especialmente su posición en el tardío y, famoso ensayo sobre Un supuesto derecho a la mentira de la Filantropía (1797), ha sido mal interpretado. Mi principal objetivo aquí será el de corregir este malentendido.

Palabras clave
Kant, ética kantiana, derecho a mentir, filantropía, veracidad.

Kant's strict views on lying have been regularly cited as a reason for thinking there is something fundamentally wrong with Kantian ethics. Some of Kant's statements here seem so excessive that most Kantians who have dealt with the topic have tried to distance themselves from them, usually claiming that they do not (or need not) follow from Kant's own principles. In this chapter, I will do a little of that, partly by questioning whether the famous example of the "murderer at the door" really fits the principles Kant applies to it. By and large, however, I will argue Kant's views about veracity are reasonable or at least defensible, if not self-evident. This is mainly because I also think some of them -especially his position in the brief, late and famous (or notorious) essay On a Supposed Right to Lie from Philanthropy (1797)- have been badly misunderstood. My principal aim here will be to correct that misunderstanding.

Keywords
Kantian ethics, right to lie, philanthropy, Kant, truthfulness.

Let's begin with an elementary point of terminology. 'Lie' (Lüge, mendacium) is a technical term for Kant. It means: an intentionally untruthful statement that is contrary to duty, especially contrary to a duty of right. An intentional untruth, when it violates no duty of right, is called a falsiloquium, a term I will translate here, for the sake of convenience, as "falsification" (MS 6:238n, VE 27:447). In Kant's usage, therefore, it is an analytic proposition that a lie is contrary to duty, and hence analytic that lying is always wrong. (Actually, there is one possible exception to this last assertion, the case of the "necessary lie", which we will discuss later). However, it is by no means analytic that every falsification is contrary to duty (or wrong), and Kant does not believe that every falsification is contrary to duty. That it is an analytic truth that lying is contrary to duty tells us nothing about the conditions under which a falsification becomes a lie. It therefore implies neither strict nor lax moral views about veracity.

The next point to consider is not terminological. It is that Kant considers the prohibition on intentional untruthfulness in relation to two (and only two) kinds of duty. It is either a violation of a duty to right, or a violation of a perfect ethical duty to oneself (though of course Kant may regard many lies as violating both duties at once). Here we will first consider lying as a violation of a duty of right, and then turn to lying as a violation of a perfect ethical duty to oneself.

This brings to our attention Kant's basic distinction between right and ethics. The fundamental principle of morality derived in the Groundwork (in all its formulations) is the principle of ethical duties. But duties of right fall under a different principle: "Any action is right if it can coexist with everyone's freedom in accordance with a universal law, or if on its maxim the freedom of choice of each can coexist with everyone's freedom in accordance with a universal law" (MS 6:230). It is unclear (and controversial in the literature on Kant) whether the principle of right is based on the principle of morality or is independent of it (my own view, argued elsewhere, is that it is the latter)¹. Every duty of right, however, in Kant's view also generates an ethical duty, because respecting the innate right to freedom possessed by all persons is an ethical duty grounded on the right of humanity (according to FH) (MS 6:237).

Since the prohibition to lying, when lying is regarded as the violation of a duty of right, is a matter of right rather than of ethics, it is not natural for Kant to think about this duty by trying to derive it from the principle of morality (in any of its formulations), as a great deal of the existing literature on this subject tries to do - for instance, by considering this prohibition via the example of the lying promise discussed in the Groundwork (G 4:402-403, 422, 429-430). I think this error alone vitiates most of the arguments found in that literature, regarded either as interpretations of Kant's views in the right to lie essay or even as properly Kantian views on the subject of veracity.

Not every intentionally false statement is a lie, in the sense of a violation of a duty of right. Many such statements are merely falsifications. In order to understand how a falsification can become a "lie" (in the technical sense that it is a violation of a duty of right), we need to understand yet another crucial piece of technical terminology -the term 'declaration' (Aussage, Deklaration, Latin declaratio). All these terms, in Kant's vocabulary, refer to statements that occur in a context where others are warranted or authorized (befugt) in relying on the truthfulness of what is said, and makes the speaker liable by right, and thus typically subject to criminal penalties or civil damages, if what is said is knowingly false.

The fact that (in juridical contexts) Aussage and Deklaration are technical terms for Kant is usually missed by readers of the essay on the right to lie. But this is quite clear from his consistent use of the term throughout his writings, and especially in the Metaphysicsof Morals (KpV 5:44, MS 6:254, 258, 304 366). Sometimes Kant appends the adjective "solemn" feierlich) to "declaration," to emphasize the special significance of the term (R 6:159, MS 6:272, 304). One paradigm case of a declaration would be a statement made under oath in a court of law, where it is to be taken as probative (KpV 5:44, MVT 8:268, MS 6:272). Another clear case of a declaration would be a promise or warranty contained in the terms of a contract (MS 6:254, 272). However, because in Kantian ethics right is the larger rational system of morals (Sitten) that grounds mere positive legislation and the enforceable rights it secures, declarations are not limited only to statements with specific legal consequences. For example, Kant thinks that a person's solemn avowal of religious faith counts as a declaration (R 6:159, MVT 8:268).

Declarations must be truthful. Kant's main principle governing the prohibition on untruthfulness regarded as a violation of duties of right is this: An intentionally untruthful declaration is a lie, hence a violation of a duty of right. This applies chiefly to cases of untruthfulness that deprive someone of something that is rightfully theirs -such as a piece of property, or a choice it is their right to make. "The only kind of untruth we want to call a lie, in the sense bearing upon right (im rechtlichen Sinne) is one that directly infringes upon another's right. e.g. the false allegation that a contract has been concluded with someone, made in order to deprive him of what is his falsiloquium dolosum)" (MS 6:238n). Such a false declaration or "wrongful falsification", contrasts with a mere falsification, that is, an intentional falsehood that involves no infringement of right:

[One is] authorized to do to others anything that does not in itself diminish what is theirs, so long as they do not want to accept it -such things as merely communicating his thoughts to them, telling or promising them something, whether what he says is true and sincere or false and insincere (veriloquium aut falsiloquium); for it is up to them whether they want to believe him or not (MS 6:238).

The basis of all duties of right, according to the principle of right, is the protection of unhindered external freedom according to universal laws. What is rightfully mine includes property (MS 6:260-270), or various other things that can be made the objects of contracts, including the promised performances of others (MS 6:274-276), or also a choice that is mine to make unhindered by coercion or by the deception of others as to the consequences of my options. What belongs to me by right is regarded by Kant as falling under principles and duties of right because it bears on my external freedom under universal law. I am externally free only insofar as I can make use of what rightfully belongs to me, including the performances others have contracted and the unhindered choices that are rightfully mine to make.

In the context of right, a declaration is a statement made by another on whose truthfulness I am authorized to rely. If a declaration made to me is knowingly false, my freedom is wrongfully restricted. More generally, however, truthfulness in declarations in general is something on which all persons are authorized to rely, within a system of right (or external freedom of persons under universal laws). If someone lies in a court of law, for example, it is not only his adversary whose right is violated, but the entire system of right, which must presume the truthfulness of declarations made in legal processes. If someone is defrauded in a contract, it is not only this person whose right is violated, but the entire system of contract right, which is structured around the truthfulness of the declarations involved in contracts. (As we will see, this is what Kant regards as the crucial point of disagreement between himself and Benjamin Constant.)

Kant also puts this point in the following way: that when I make a lying declaration, "I bring it about, as far as I can, that declarations (Aussagen [Declarationen]) in general are not believed, and so too that all rights which are based on contracts come to nothing and lose their force" (VRL 8:426). The claim here is not that some particular lie might in fact shake people's confidence in trials or contracts (as if it by itself would cause them no longer to believe anyone). It is rather that the system of right is constituted by a set of laws that are universally valid - actions are right only if they can coexist with everyone's freedom under this system according to a universal law. A statement counts as a declaration whenever reliance on its truthfulness is required to secure people's rightful freedom under universal laws. Hence it is contrary to the very concept of right that it could be right to make an untruthful declaration when the truthfulness of that declaration is required by rational laws of right. By making such a declaration, I am in that sense acting in such a way as to deprive declarations made the system of right of their validity, whether or not that result is intended or actually occurs. Kant also puts it this way: "It cannot hold with universality of a law of nature that statements should be allowed as proof and yet be intentionally untrue" (KpV 5:44).

We could put this point in the terminology of John Rawls (Rawls, TCR) if we said that for Kant, right is a "practice" (the rational practice involving what is necessary to guarantee people rightful freedom under universal law). Truthfulness in making declarations is one of the rules of the practice. "Right" is, in effect, a rational framework for understanding, justifying and correcting not only state and legal institutions but also other kinds of understandings between people guaranteeing their freedom under universal laws. Kant attempts to justify the practice of right by showing the necessity of different aspects of it for protecting something regarded by Kantian ethics as of fundamental value -namely, the guarantee to persons of their external freedom according to universal law. The requirement that there be "declarations" at various points in the system of right is to be established by presenting contract law, judicial trials, and so on, as requiring them. Once the making of truthful declarations is established as part of the "practice" of right, the rule of right requiring that declarations be truthful goes with the practice of making them. The duty always to be truthful in declarations needs no further defense.

History of the famous example. In the famous late essay, untruthfulness is being considered (as the title of the essay, as well as its content, clearly indicates), solely as a violation of a duty of right. That essay is part of a controversy between Kant and the French writer Benjamin Constant. As I have said, the brevity of the essay, along with the common neglect of Kant's entire theory of right, often prevents readers from appreciating the precise nature of the question being addressed in it. They are so bedazzled by the famous example (chosen by Constant) that is discussed in it, and Kant's apparently unreasonable position on that example, that they never even notice certain unusual, artificial or even dubious features the example must take on if it is to be an illustration of the point Kant is trying to make.

The moral principle 'it is a duty to tell the truth' would if taken unconditionally and singly, make all society impossible. We have proof of this in the very direct consequences drawn from this principle by a German philosopher, who goes so far as to maintain that it would be a crime to lie to a murderer who asked us whether a friend of ours whom he is pursuing has taken refuge in our house (Constant, Des reactions politiques, quoted by Kant, VRL 8:425).

Constant is apparently responding to an example Kant had used in the Doctrine of Virtue - a servant lies to the police in saying that his master is not at home, and this lie enables the master to slip away and commit a crime:

For example, a householder has ordered a servant to say 'not at home' if a certain human being asks for him. The servant does this and, as a result, the master slips away and commits a serious crime, which would otherwise have been prevented by the guard sent to arrest him. Who (in accordance with ethical principles) is guilty in this case? Surely the servant too, who violated a duty to himself by his lie, the results of which his own conscience imputes to him (MS 6:431).

Constant radically modifies the example, however, transforming the servant of a would-be criminal into the friend of an innocent man who is trying to escape someone intending to murder him. What shocks people is that Kant's position about Constant's example is that the friend must not lie to the murderer:

Truthfulness in declarations (Aussagen) that one cannot avoid is a human being's duty to everyone, however great the disadvantage to him or to another that may result from it; and though I indeed do no wrong to him who unjustly compels me to make the declaration if I falsify it, I nevertheless do wrong in the most essential part of duty in general by such falsification, which can therefore be called a lie...; that is, I bring it about, as far as I can, that declarations (Aussagen [Declarationen]) in general are not believed, and so too that all rights which are based on contracts come to nothing and lose their force; and this is a wrong inflicted upon humanity generally (Kant, 8:426).

What the dispute is about. It is clear both in Constant's essay and in Kant's reply, that the real issue is the duty to speak truthfully in declarations in political contexts, and the alleged limits on this duty. This point is easy to miss because the example of the murderer at the door is not at all about the speech of politicians or statesmen. Constant's thesis is that moral principles can be applied to politics only by means of intermediate principles. Specifically, he claims the principle of truthfulness in declarations must meet the condition that those to whom one speaks have a right to the truth. Kant's counter-thesis is that the duty of right to be truthful in declarations is not limited by that condition.

Kant is usually interpreted as holding that while it may be permissible to refuse to answer the murderer's question, if you cannot avoid answering it (as Constant stipulates you cannot), then it is not permissible to lie to him, even if your truthfulness directly enables him to murder your friend (VRL 8:425-427). The natural reaction to Kant's position, so understood, is that it is a piece of rigoristic craziness. It is regularly used to call into question the moral sanity of any philosopher who could take such a position, and sometimes also to back up the crudely erroneous argument, discussed in Chapter 4, § 1, that the very notion of a categorical imperative commits Kantian ethics to a set of rigid moral rules. (That Kantian ethics should in principle admit possible exceptions to any moral rule was argued in in my book Kantian Ethics, Chapter 3, § 4.)

The usual interpretation of Kant's position gives no thought at all to the fact that he would see no violation of right whatever in a mere falsification uttered to the would-be murderer. Although the category of "declaration" includes more than assertions made under oath or in a contract, it is no part of Kant's theory to hold that just anyone who knocks on your door is automatically in a position to require from you a solemn declaration regarding the present whereabouts of some person. Perhaps a policeman, as in Kant's original example, might be in such a position. That is why Kant argues that the servant would be criminally liable as an accessory to his master's crime (MS 6:431). But someone merely appearing at your door with murderous intent normally would not. Of course if the murderer could not require a declaration from you, then telling him an intentional untruth would not count as a lie (mendacium). Kant explicitly allows that no lie, and no violation of right, occurs if we commit a falsification in order to prevent another from making wrongful use of the truth:

I can also commit a falsiloquium when my intent is to hide my intentions from the other, and he can also presume that I shall do so, since his own purpose is to make a wrongful use of the truth. If an enemy, for example, takes me by the throat and demands to know where my money is kept, I can hide the information here, since he means to misuse the truth. That is still no mendacium. (VE 27:447).

Sometimes Kant describes this situation, or one very much like it, in terms that make it permissible even to make a false declaration, and thus to tell a lie. This is the one possible exception to the proposition Kant seems otherwise to regard as analytic: namely, that lying is wrong, as being contrary to a duty of right. For there are passages in his lectures in which Kant invokes the traditional concept of a "right of necessity" (Notrecht, ius necessitatis), in which under compulsion in an extreme case of need or distress, a person is permitted to do something that would normally violate a rule of right. A "necessary lie" (Notlüge) occurs where someone forcibly compels you to make a declaration of which you know they will make wrongful use.

Yet since men are malicious, it is true that we often court danger by punctilious observance of the truth and hence has arisen the concept of the necessary lie, which is a very critical point for the moral philosopher. So far as I am constrained, by force used against me, to make an admission, and wrongful use is made of my statement, and I am unable to save myself by silence, the lie is a weapon of defense; the declaration that is extorted and then misused permits me to defend myself, for whether my admission or my money is extracted is all the same. Hence there is no case in which a necessary lie occurs except where the declaration is forced from me and I am also convinced the other means to make wrongful use of it (VE 27:448).

In Constant's example it is stipulated you have no alternative to making the declaration. Yet because the murderer has not forcibly extorted the declaration from you, a lie to him under these circumstances would not count as a necessary lie. Kant does not always seem happy with the idea of a "necessary lie". In some places he seems to question whether really there is such a thing, or he accepts the concept of a necessary lie only reluctantly (VP 9:940, VE 29:701). He roundly rejects the idea that we are permitted to deceive another simply because the other has deceived us. On the contrary, in Kant's view the right course of conduct will sometimes leave you open, at least to a limited extent, to the attacks of evil people; this policy of leaving yourself vulnerable is required of you by your respect for the right of humanity: "When one country has broken the peace, the other cannot do so in retaliation, for if that were allowable, no peace would be secure. And thus though [a lie to a deceptive or unjust person] may not infringe [his] particular right, it is still already a lie, and contrary to the right of humanity" (VE 27:447). But as a passage quoted earlier also clearly says, Kant thinks that when you are forced to make a false declaration in order to prevent the truth from being used unjustly, that would normally be considered a mere falsiloquium and not a mendacium, and it would violate no duty of right.

Can the murderer demand a declaration? It might be argued that in the example of the murderer at the door, there could be no question of making a statement having the rightful import of a declaration. Tamar Schapiro (2003) has argued that a Kantian has good grounds to make an exception to a duty when the duty is based on a practice between people, but the understanding on which the practice rests has been reduced to a sham, as through the systematic misconduct of some of the parties to it. To put it in the Rawlsian terms already mentioned, we might argue that the murderer at the door, through his wrongful intentions, has undermined the practice of right, and therefore that the concepts of "declaration" and "lie" that presuppose the rules of this practice, no longer apply to what I tell him.

I think Schapiro's theory of excuses or exceptions is entirely cogent within a Kantian theory of right, and so is its application to this case. Yet Kant himself clearly rejects that argument, at least as applied to this example. The reason is that he holds that even when someone intends to use a declaration unjustly, it might nevertheless be possible in principle for him to be entitled to a declaration. Thus Kant allows Constant to assume that this is true in the case of the murderer at the door. When someone unjustly requires a declaration of you, Kant holds, you do no wrong to him in falsifying your declaration, but you nevertheless do wrong to humanity generally by violating your unconditional duty to be truthful in all your declarations:

Truthfulness in declarations (Aussagen) that one cannot avoid is a human being's duty to everyone, however great the disadvantage to him or to another that may result from it; and though I indeed do no wrong to him who unjustly compels me to make the declaration if I falsify it, I nevertheless do wrong in the most essential part of duty in general by such falsification, which can therefore be called a lie.; that is, I bring it about, as far as I can, that declarations (Aussagen [ Declarationen]) in general are not believed, and so too that all rights which are based on contracts come to nothing and lose their force; and this is a wrong inflicted upon humanity generally (VRL 8:426).

This point is closely related to one of the main issues between Kant and Constant, Kant's rejection of Constant's claim that we owe truthfulness only to those who have a "right to the truth" (VRL 8:426). It might seem that Kant should agree with Constant here, since the Kantian distinction between a lie and a mere falsification might seem to amount to the distinction between saying something false to someone who has a right to the truth from you and saying something false to someone who has no such right. Yet Kant rejects any such account of the distinction, because he holds that the duty not to lie attaches to every declaration as such, and is not owed only to the person to whom it happens to be made.

On this issue, moreover, Kant seems clearly to be right. As we have already seen, the duty to make a truthful declaration under oath in court is not owed merely to the attorney who asks you the question, but involves you in a relation of right to the judge, the jury and to the entire process of justice. In relation to what ultimately interests Kant the most, the duty of politicians to be truthful in their public declarations is a duty whose performance must be relied upon by the public at large. In the essay on the right to lie, Kant sometimes distinguishes between a lie in the strict sense of the jurists, where the untruthful declaration violates the right of an assignable individual or individuals, and a lie in a broader sense, in which it violates the right of humanity (VRL 8:426; cf. VE 27:448). His main point in that essay is to insist on the validity of this broader conception, making an untruthful declaration wrongful (a lie, the violation of a duty of right) even where no assignable individual (with a "right to the truth") is wronged by it. This point seems especially pertinent if the real target lies in a political context, where statesmen or politicians make untruthful declarations to the public. For here it is the public at large, or humanity in general, and no assignable individual, whose right is infringed by the lie.

Kant's further view here, that you might be unjustly required to give a declaration, which you nevertheless have no right to falsify, seems more questionable. This is a first cousin of his also questionable view that you are required to obey even the unjust commands of a civil authority, as long as they do not require you to do something that is in itself wrong. But this questionable view clearly plays a role in his willingness to regard it as conceivable that the murderer at the door, even with his plainly unjust intent, might in principle be in a position to demand a declaration from you.

Once we appreciate all these points, we should begin to see how extreme, artificial (or even dubious) is the kind of case in which Kant's principles require him to say that it would be wrong to lie to the murderer at the door. If our statement to the would-be murderer is not a declaration, then we need not speak truthfully, because that would be a mere falsification, not a lie. If he extorts a declaration from us, intending to use it unjustly, then that would be a case of a "necessary lie" and would again be permissible. It is only where a declaration is unavoidable, yet not extorted, that lying to the murderer at the door would violate the right of humanity. Most people who read Kant's essay seem bedazzled by the thought that Kant is willing to say about any case of the murderer at the door that you may not rightfully lie to him. The glare prevents them from seeing anything else about the case, including any of the more specific principles involved.

What seems to me most implausible about Kant's claims about the murderer at the door is not that it would be wrong to make a lying declaration to him, given the conditions stipulated, but rather that the stipulated conditions could ever obtain in the case of a murderer at the door. That, however, is not the error with which Kant is usually charged. It would be an error merely about whether this example could really fit his principles, not an error infecting Kantian principles themselves.

In order to gain a better appreciation of the issue Kant means to address, we might do better to consider a different example, in which the necessary assumptions would be less artificial or implausible. I propose the following example, suggested by some of Kant's own examples in his lectures (VE 27:493, 506, 508), but even more by a remark by Sidgwick (Sidgwick, p. 97). "In speaking truth to a jury, I may possibly foresee that my words, operating along with other statements and indications, will unavoidably lead them to a wrong conclusion as to the guilt or innocence of the accused, as certainly as I foresee that they will produce a right impression as to the particular matter of fact to which I am testifying" (Sidgwick, p. 97). Sidgwick admits that the morality of common sense would call it truth-speaking to testify truthfully to the particular fact in question. But he appears to conclude -though without quite saying this explicitly- that there is a certain artificiality in the common sense notion of veracity here -as though a witness who is really interested in the truth might do better to lie about the particular fact in order to bring it about that the jury draws the right conclusion about the guilt or innocence of the accused. That would justify a conclusion directly contrary to the Kantian one about our example. However, it would also be a most remarkable conception of what the oath to tell the truth at trials binds us to do. It would imply that a witness might lie whenever they thought the jury would be misled in their conclusions about the case by truthful testimony and would be more likely to reach the right conclusions about the case by being fed intentionally false testimony. Imagine what would happen if a witness refused to take the oath without adding this qualification: "I will speak the truth, the whole truth and nothing but the truth, unless I am convinced that lying will result in a more just outcome, in which case I will feel free to lie." If Sidgwick were called as a witness at a trial, and I were in the position of the attorney cross-examining him with the aim of discrediting his testimony, I think all I would need to do is read to the jury this passage from the Methods of Ethics, and suggest to them the non-quite-stated conclusion that it looks like Sidgwick wants to draw. Unless Sidgwick were prepared to repudiate this interpretation of his meaning, and were capable of doing so quite convincingly, I submit that the jury would be quite justified in ignoring everything he says on the stand as totally lacking in credibility.

Or imagine a witness who insisted on modifying the oath to read: "I will tell the truth, the whole truth and nothing but the truth, unless I think this is a case sufficiently like the 'murderer at the door' example, and in that case I will feel free to lie". Such a witness's testimony would lack the credibility required for declarations to serve their function in a system of right. Such testimony would not be accepted, nor should it be. Suppose you are a witness under oath in a court of law. You are asked by the prosecutor a question the truthful answer to which will predictably result in the conviction of your friend (or in Kant's example, your brother), whom you know to be innocent, on a charge of murder. Here an unscrupulous prosecutor might play the role of the murderer at the door, the innocent defendant the role of his intended victim, and again you are faced with the choice between telling the truth and saving him.

No one should deny that this would be a deeply troubling predicament to be in, but my own considered view about it is the Kantian one: Unless I think the legal process is illegitimate, or a mere sham, I think I had better tell the truth and be prepared to live with the consequences. Otherwise (as Kant himself suggests) I am the one turning the process into a sham, by behaving according to a principle which, if generally followed, would bring all solemn testimony and all legitimate legal processes into discredit (VRL 8:426).

It is an important principle that an action may be wrong (a violation of the right, an injustice) even if the particular person against whom it is committed is in no position to claim that he is wronged by it. This is Kant's position about the murderer at the door. His murderous intent deprives him of any standing to claim that he is wronged by a lie that is told to him, yet it may still be wrong to lie to him. I think we can better appreciate the point Kant is making if we think of a different kind of wrong -for instance, torture. There are those who think that it may be permissible to torture certain people because, say, they are terrorists, or are themselves torturers, and therefore are flagrantly guilty of the very conduct that we are perpetrating on them-hence (the argument goes) they are in no position to claim that they are wronged by it. The Kantian position in this case might be that it could be true that they are in no position to say that they are wronged, but it still might be wrong of us to torture them -simply because of the kind of thing that torture itself is. Kant's term for it would be that torture is a wrong against humanity. A more contemporary way of making this Kantian point would be to say that when it is a question of the wrongness of torture, it does not matter whether the person tortured is a terrorist or a torturer. Torture is simply wrong. We must not do it, no matter who the victim is. Or as the slogan has it: "It's not about who they are, it's about who we are." It's not about whom we torture, it is about the wrongness of torturing. Kant's claim about lying to the murderer at the door (on the assumption that the falsehood is a lying declaration) is analogous to this position about torturing.

Political lying. As I have mentioned, the issue that appears to have really concerned both Kant and Constant is the duty of politicians and statesmen to be truthful in their official declarations. Here we surely need no "trolley problems;" there is no shortage of crying examples all around us in real life.

Stephen Holmes (1984), persuasively describes Constant's position in the dispute with Kant as the outcome of his experiences during the French Revolution, where the line separating police officials from murderers was not necessarily well-defined, and where declining to lie (even, we may suppose, in a declaration to a policeman or in solemn declarations in a political context) might easily result in you, or your friends, being sent to the guillotine. Under those circumstances, Constant's position is certainly understandable. Looking at the dispute from this angle, Kant might be faulted for failing to appreciate the extreme conditions that motivated it.

Kant's contrary view, however, belongs to his insistence in Perpetual Peace that for rulers and statesmen, political expediency must always be subordinated to principles of right, and that high office and political power -and the need to confront the kinds of decisions that go with the possession of such extraordinary power- earn no one an exemption from these principles. Maxims involving deception, moreover -denying the wrongs you have done, for example, or concealing your true aims and policies from the public- are prominent in that discussion (EF 8:375-376, 381-382). Kant's position on these issues seems to me clearly correct².

I find the dispute between Kant and Constant, considered in historical context, to be one in which each of the parties is making a valid point, but as I see it about quite different issues, though issues that can interact in real life. To the extent that this is the case, both positions are deserving of respect, and the disputants are to some extent talking past each other. To the extent that they are not, the dispute is a troubling one, about which a sensible person should experience a good deal of unresolved conflict.

More recent real life examples of lying declarations by political leaders and government officials leave me feeling far less ambivalent. Outrageously wrong political lying has played a decisive role in the political life of the U.S. certainly as far back as most of us can remember, and in the past five years it has become the chief determinant of governmental policy in virtually every area, from foreign policy to environmental policy. Lyndon Johnson obtained the Gulf of Tonkin Resolution from the Senate (with only two dissenting votes) by straightforwardly lying about what had happened there. The Pentagon Papers disclosed a systematic pattern of official lying to the public about the reasons for the Vietnam war, war policies and the facts on the ground in Vietnam. G. Gordon Liddy -convicted and imprisoned as a Watergate conspirator but now the popular host of a right-wing radio talk show- has repeatedly said that Richard Nixon and his associates were entirely justified in lying under oath to the U.S. Congress during the Watergate cover up. Oliver North -another popular figure in the far right media that now enjoy a near monopoly on the dissemination of public information in this country- has insisted it was right for him to lie to the Congress about covert sales of arms to Iran to finance (also covertly and illegally) the Contras in Nicaragua. Bill Clinton's lies about his personal misconduct, since they involved no wrongful exercise of governmental power in the public realm, were not impeachable offenses, but they were clearly wrong.

What should perhaps be uppermost in our minds is the outrageous political manipulation and falsification of intelligence leading up to the U.S.-British invasion of Iraq in 2003. This involved systematically untruthful declarations to the public by many officials of both governments, including the U.S. Secretary of State Colin Powell before the United Nations on February 5, 2003. Further, we should know how to judge these same officials when they offer the excuse that they were misinformed by their intelligence sources. In light of the fact that they not only picked and chose among those sources but even manipulated the gathering of intelligence with a view to rationalizing the policies they had already decided upon, we should say that this is nothing but a further lie compounding the wrongs they have committed. Reflection on recent history should make us more sympathetic with the position Kant takes in the right to lie essay.

Rules and exceptions in philosophy and real life. Philosophers are always looking for counterexamples to general theses, and this makes them look hard for exceptions to every rule of right or morality that might be proposed. As I have argued in Kantian Ethics, Chapter 3, § 4, Kantian ethics says they are right. For in moral philosophy it is an important truth in that due to the great complexities of human life, no moral rule simple enough to be practically useful can be framed so delicately as to be free of exceptions. But alongside this truth, philosophers should also appreciate another truth, which was always vividly before Kant's mind, and I think explains some of the things he says about lying as well as other subjects.

The following is a true empirical generalization about people's behavior in real life: People have a powerful tendency to use the fact that there are exceptions to moral rules in order to rationalize making exceptions when they should not. For this reason, the speech act of asserting truly that there are exceptions to rules is more often than not used to justify wrongdoing, while the speech act of asserting falsely that there are none is most often a rhetorical attempt (probably unsuccessful) to prevent wrongdoing. Sometimes, on the contrary, the opposition is between inflexible moral prejudice and an open-minded reasonableness that is trying to take circumstances into account. Philosophers prefer to imagine the latter situation, since it flatters them by making their subtle reasonings a force for good rather than for evil. But if we take human beings as they are, we must admit this is not the typical case.

If we take proper account of this true generalization, it tends to justify those moralists who rhetorically exaggerate the strictness of important moral rules, and to cast doubt on the wisdom, and even the moral integrity, of philosophers who derive conceptual titillation from devising counterexamples to them and treat such counterexamples as reasons for relaxing strictness of the rules. Kant shows himself to belong to the former class of moralists, for example, when he denies we should teach children that there can be "necessary lies," since (he says) "they would soon take the smallest excuse for a necessity, and often allow themselves to tell lies" (VP 9:490).

In this respect, people in power tend to be far worse than even the naughtiest of children. When they argue for exceptions to important rules restricting their conduct - using murderer-at-the-door arguments to justify lying, or ticking-bomb arguments to justify torture, or weapons of mass destruction in the wrong hands to justify preventive war - then you can be certain that they will lie to your face when there is no murderer at the door, use torture on prisoners when there is no ticking bomb, and start wars of aggression when there are no weapons of mass destruction.

Constant claimed that Kant's position would make political life impossible. The charge seems exaggerated, but the decisive Kantian rejoinder, which is surely no exaggeration, is that the policy of politicians to permit themselves lying declarations for supposedly worthy ends is precisely what does make possible much of what is utterly intolerable in our actual political life.

¹ See Kant's Ethical Thought (New York: Cambridge University Press, 1999), pp. 322-323; "The Final Form of Kant's Practical Philosophy," in Mark Timmons (ed.), Kant's Metaphysics of Morals: Interpretive Essays (Oxford: Oxford University Press, 2002), pp. 5-10. For a contrasting view, see Paul Guyer, "Kant's Deductions of the Principles of Right," in the same volume, pp. 23-64. However, I do not think the claims I am making here depend on which side of the dispute we take, as long as we are agreed on the undeniable fact that duties of right (Recht) and duties of ethics (Ethik) belong to two different spheres within Kant's entire theory of morals (Sitten).

² The issue Kant means to raise here is sometimes misleadingly stated by those arguing on the other side as an issue about whether the standards of "private morality" apply to the "public" realm of politics, statesmanship, war and the like. For example, Carl Schmitt holds that the political has its "own criteria" which are distinct from those that can be traced back to moral concepts of good and evil. See Carl Schmitt, The Concept of the Political, translated by George Schwab (Chicago: University of Chicago Press, 1996), p. 26. When Kant speaks of "morality" in relation to politicians, as he does in Perpetual Peace, the standards he is using are never those of private ethics but always of public right. If Schmitt's claim is that politicians or statesmen are bound not by the criteria of private morality, but by standards appropriate to the political realm, then Kant agrees. Kant thinks they are bound by the standards appropriate to their position as exercisers of public coercive force, which must be regulated by laws of right. These standards are looser than private ethical standards, since they relate to a system of laws that are in general coercively enforceable - and this is looser than the system of ethical laws through which each of us should inwardly regulate our private behavior. As I have already mentioned, however, not all standards of right are coercively enforceable, and Kant is famous (or infamous) for holding that subjects have rights against heads of state, and heads of state have duties of right, that no one is in a position to enforce coercively (as by violent revolution, which might be the only conceivable means for enforcing them). So Kant's position, while no doubt different from Schmitt's, is not as different from it (or as vulnerable to criticism) as people like Schmitt often think. The problem is that for many politicians (and the "realist" theorists who enjoy identifying with those who exercise great power over their fellow human beings), any constraint on the use of that power based on mere principle (rather than arising from external constraints or political self-interest) feels like an annoying incursion on their prerogatives according to "inappropriate" standards.