Kant’s 1797 essay “On a Supposed Right to Lie from Love of Humanity” has done more than any of his other works to scare students off his moral theory. Interpreters have little time for it. They call it “grotesque”, “shocking” or “morally perverse”. This is not surprising. The central thesis of Kant’s short piece is that, when only a bold and confident lie might save a friend, one must, if asked, answer truthfully and thus betray his hiding place to the person who wants to kill him. Asserting what one does not believe to be true – ‘lying’ for short – is always wrong, regardless of the consequences of telling the truth. We do not have to read the essay, let alone grapple with its arguments, to find Kant’s position unappealing. A lecturer’s three-sentence summary in an introductory ethics class will do.
And yet, we should not dismiss Kant’s essay out of hand, if only because we would be missing out on many historical, exegetical and philosophical goodies. It is, for instance, far from clear that in real life a lie will shield the friend from harm or that a truthful reply will help the assailant find his victim. Lying may not protect the victim if the person put on the spot is mistaken about his whereabouts. For Kant, a lie is opposed to (subjective) truthfulness, not to (objective) truth. A householder may honestly believe that the victim is hiding in his house. If he is actually hiding elsewhere, a lie may facilitate the victim’s discovery. This is the story of Jean Paul Sartre’s The Wall.
Moreover, why assume that the assailant will take the householder’s answer at face value? “Tell the truth. He will think it a lie!” says the Göttingen theologian and orientalist Johann David Michaelis, who argued for a robust duty of truthfulness a little earlier than Kant. So, we must differentiate lying from deceiving. This distinction, which is sometimes ignored in the Kant literature, is neatly illustrated by Talleyrand’s quip about Prince Metternich and Cardinal Mazarin: The cardinal deceived but he never told a lie, whereas the prince told lies all the time but he failed to deceive anyone. A lie that is disbelieved need not be deceptive. What is more, Kant appears to think that we can lie without even intending to deceive, as he does not include the traditional ‘intention to deceive’ in his central definitions of lying. This move has recently gained some momentum among colleagues working in the philosophy of language. Kant may well have been an ‘anti-deceptionist’ and, at least in that regard, our contemporary.
We have already seen that the case of lying to a would-be murderer is not, as is often alleged, Kant’s own. It has its roots in the Hebrew Bible and in Xenophon’s Memorabilia; it was formalised by Augustine in his De mendacio; and by 1797, it had been long been a stock example of academic disputations. Michaelis and Fichte, amongst others, discuss it independently of Kant. We have ‘trolley cases’; they had ‘lying to murderers’. Here is the example as told in a story book for children published in 1792. A captain is looking for his servant, who he suspects has stolen his dog:
Hat on head and sword in hand, the captain furiously stormed through the door: Is my man with you? he asked, and he looked around the room with sparkling eyes.
Spoiler alert: The servant, who did not steal the dog, is hiding in a locked wardrobe nearby; the householder succeeds in changing the captain’s mind; captain and servant are reunited; no person or animal is harmed. And here is the case as told by an anonymous Kantian before Kant himself weighed in on the debate:
Someone flees to my room, obviously terrified, and asks me to hide him. A stranger, his sword bared, comes running after him; snorting with anger and vengefulness, he asks me whether his enemy is with me. Should I save his life by lying or sacrifice him by telling the truth? I am of the view that it is not just that one may, but that one must deny that the stranger is in the room. Other Kantians maintain, we know, the opposite view.
Kant would not have discussed the example had he not been confronted with the charge of absolutism about truthfulness by Benjamin Constant and his translator, Karl Friedrich Cramer, in 1797. Entirely without warrant, Constant and Cramer attribute to Kant the thesis that
it would be a crime to lie to a murderer who asked us whether a friend of ours, whom he was pursuing, had perhaps taken refuge in our house.
We know that Kant had not previously defended an absolute duty to be truthful in his publications. But there can be no doubt that he does so now.
Now, some of those who seek to save Kant from his own moral theory argue that trying to deceive the assailant by a lie should be permitted because the latter is trying to deceive the householder about his intentions. But that is questionable even if we ignore the fact that lying and deceiving are distinct phenomena. Do we really want to make a licence to lie conditional upon the would-be murderer’s duplicity? As we now know, the murderer traditionally pursues his victim openly, brandishing a weapon. And if we are allowed to lie to a duplicitous murderer, should we not be allowed to lie to an honest one? Moreover, the details implicit in Constant’s allegation explain why the householder cannot remain silent, an option that would be innocuous as well as effective: The assailant forces the householder to say either yes or no by threatening to kill him too. That is why Kant does not entertain reticence as a possible way out.
Another tactic popular with interpreters who wish to soften the rigorism of the essay on the ‘supposed right’ is to point to Kant’s early lectures on ethics. At that time, Kant believed that untruthful declarations were admissible in certain extreme cases. This strategy is equally questionable, if for different reasons. Kant’s theory of truth and truthfulness changes significantly over the years. In the 1770s, lying is considered wrong because it violates a tacit social contract established to enable the exchange of information. Kant speaks out against lying. But within a contractualist framework, there is room for an untruthful assertion (Lat. falsiloquium) that do not deserve the morally charged epithet of a lie (mendacium). By the 1790s, the core duty of veracity has become an ethical duty speakers owe to themselves, not to the addressee; and now it is absolute. Untruthfulness is vicious because it is deeply corruptive. The story of the 1797 essay is different still because it is an exercise not in ethics but in the philosophy of law. Kant argues for truthfulness on the grounds that lying, or a right to lie, threatens to undermine the possibility of contracts, a grave wrong inflicted on humanity generally. This argument is woefully inadequate philosophically. And whilst it is not new, it had not been in evidence for years when Kant decided to revive it in 1797. To make things worse, it is not even Kant’s own. Michaelis had presented it in his inaugural pamphlet at Göttingen University in 1750. Kant just readapted it to have something to say to Constant nearly half a century later.
Finally, the case is commonly called that of the ‘murderer at the door’. This is a mistake. Constant does not mention a door. Nor does Kant. Murderer and householder must meet somewhere near the victim’s place of hiding for the former to pose an immediate threat. But, see above, the encounter may well take place inside the householder’s lodgings. In the late eighteenth century, people often left their doors unlocked; and the householder would not have had enough time to lock up when the friend barged in. In other versions of the case, the two parties meet in the street. In any case, there is no reason whatsoever to assume that they meet as the householder answers the door. And why would he?
As we bid farewell to the ‘murderer at the door’, I bid farewell to a project that has been with me for nearly 25 years. The book is still far from perfect. I say little about Kant’s ethical absolutism about lying since, as one must, I discuss the essay as a contribution to the Doctrine of Right. The Doctrine of Virtue remains a topic for another occasion, or perhaps for a colleague. Moreover, I will no doubt change my mind again as discussions continue. New historical evidence may be brought to light. I tore up two manuscripts because I was unhappy with what I had written. So, I am under no illusions. But there is, I very much hope, enough material in this book to revive and enrich the debate about Kant’s theory of truth and truthfulness.
Kant and the Supposed Right
to Lie by Jens Timmermann
Latest Comments
Have your say!