I heard about Kant's reasoning that lying that you return money or about the leads to contradiction in conception.

But how could he even prove that lying under any circumstances leads to contradiction in conception?

My example:

Let us assume some kind of soft extortionists. They threaten people to give them money but only if people have them. But they do not check if it is true since they are afraid of consequences (it can even result in murder and/or imprisonment).

Now let's assume that all people lie to them that they do not have money. Since they all lie, extortionists do no get money. Even if they are in doubt of people's words. After many unfortunate attempts they give up.

The maxim in this case would be following:

1) In the case one is being approached and asked if he has money by a stranger 2) in order to reduce extortining incidence 3) the one should tell to the stranger that he doesn't have money.

Isn't this a counterexample for "do not lie" duty? If so, how can Kantianism even prove that perfect duties even exist? And moreover the perfection of any particular duty.

  • 2
    Possible answer can be found here. You do not seem to understand what contradiction in conception means for Kant. The case you mentioned is explained under the last header in the case a) (inner contradiction). – Philip Klöcking Apr 4 '18 at 9:57
  • @PhilipKlöcking, the thing is that there is no contradiction in my example which justifies the lie. Thus being a counterexample for Kant's "do not lie" imperative. This puts lying to contradiction in the will, making "do not lie" imperfect duty. – rus9384 Apr 4 '18 at 10:29
  • 2
    You fail to formulate a particular maxim, you just describe a situation where consistent lying would "solve" a particular situation, describing it from the outside. This does not constitute a counterexample to the Kantian universalization at all (as explained under the link above). Again, I beg to try to understand what is happening in Kantian terms first. If you did, you would understand that you'd need to find a maxim including lying that when universalized through the CI does not end up in logical incoherences. – Philip Klöcking Apr 4 '18 at 10:41
  • @PhilipKlöcking, but then does an example with murderer trying find kids valid? Since after universalization it is not possible for such a human to exist, it seems it is not valid thus not being a concern of Kantianism? – rus9384 Apr 4 '18 at 10:46
  • 1
    That seems to be a pretty bogus maxim. Clearly when one does the lying for extortion, it's for the hope of extorting. – virmaior Apr 5 '18 at 2:50

The duty not to lie is not in virtue of the consequences that would result from some particular act, but because if we universalze the maxim of lying (generally speaking) there would no longer be any upside to lying. If lying is universal law and I tell you, “give me $50 or I’ll kill you”, there would be no reason for you to fear for your life because the universality of the maxim of lying implies that my threat is empty.

Kant’s “proof” of a contradiction is an a priori argument, and it seems plausibe if taken at face value. The biggest concern is whether the intuitive result of universalizing some action is sufficient to establish its moral value - intuitively, probably not.

  • Wait, wait... if he does not lie, that means he really will kill me if I don't give $50. Why is there no reason to fear? About appliability of Kantian ethics... we do not live in perfect society, thus Kantian ethics probably fail. – rus9384 Apr 5 '18 at 9:14
  • 3
    Sorry to be off-topic, but this kind of logic always annoys me. If someone tells you "give me $50 or I'll kill you" there is already no reason to believe them. If they willing to kill they are much more likely to be willing to lie about it. And if they are willing to kill for $50 they are much more likely to be willing to kill for no reason at all. Once you give them the $50, you have witnessed their evil and they should kill you to prevent others from finding out how evil they are. The idea of counting on the honor of person who threatens your life is kind of bizarre... – user9166 Apr 6 '18 at 22:57
  • 2
    The argument is not that lying universally takes away the advantage of lying. It is that lying universally takes away the ability to pass on the abstract notion of truth to anyone else. Truth is hard enough to fathom when most folks are trying to maintain it. In a world lacking that value, it would be impossible to even explain. – user9166 Apr 6 '18 at 23:02

Try looking at is from an ethical management perspective. Kantianism is a denotological theory concerned with outcomes and maxims.

A maxim is a rule such as

To always respect a persons autonomy Or

To never respect a persons autonomy.

To prove one is ethical and that the duty is, as you say, perfect in that it is indisputable as the only correct choice regarding the party that the duty is owed to, the categorical imperative is used.

Kantianism in a way views ethical choices regarding outcomes where the outcomes are like absolutes, either one way or another, no in between which creates the perfect duty.

Categorical imperative steps:

1 universality Is the maxim applicable to any situation or is it self defeating. For instance, if everyone was to always respect the autonomy of others, no one would ever be coerced or lied to and society would be able to function and decisions could appropriately be made, self fulfilling. On the other hand if everyone was to lie and deceive and coerce everyone, no one would be able to function, therefore never respect autonomy is not the ethical action.

2 respect Does the maxim respect others as ends in themselves rather than means to an end? Meaning is the intention of the duty holder selfish or selfless? Selfless is required to pass step 2

3 Respecting autonomy in itself Does the maxim respect the full decision making ability of others, and not coerce, lie or deceive them so that they can come to their own conclusions.


The attack is directly on the notion of lying itself (not necessarily having anything to do with money).

The point is that it controverts the use of language, which is something necessary to us. If everyone lies as much as they want, and does not feel bad about it at all, then we will be confused all the time as to what they do or do not actually believe. No one would harbor the expectation that statements were trustworthy. And therefore any given lie would no longer deceive. So the idea of lying would no longer mean anything. That is a contradiction in concept -- too much lying removes the possibility of the effectiveness of lying. And in so doing it removes the ability to communicate directly at all, which would be a major loss.

There is no disputing the logic there. The question is not whether the things he has defined exist. The question is whether their existence actually matters. That, he 'proves' by taking up 'autonomy' as his principal value, and applying a very simple version of the Golden Rule in a ruthlessly rigorous fashion. If you were the extortionist, you would not want everyone to lie to you. We need to more directly address the morality of extortion, rather than working around it by thwarting its effects.

Otherwise, we are not addressing morality as a community, we are using the criminal to justify our own misbehavior. Once you do that with one category of misbehavior, it applies to a wide range of things you might forgive, including extortion of your own. That is using a human being as a mere means to our own psychological comfort.

So he has foreseen your argument, and he addressed it explicitly, even down to the idea that you should not lie to prevent a murder.

That does not prove he is right, but it does mean that you need to make that more global decision for some other reason.

Edit ---

OK. A perfect duty has to come from a contradiction in concept. And, at least for Kant, a perfect duty, if it can be determined always overrides a contigent duty -- one where the maxim allows for conditions or measures.

The duty against murder is perfect -- if everyone killed someone, there would be no one left to kill.

The duty against lying is perfect -- if everyone lied all the time, no one could lie effectively.

The duties you put forward, like preventing criminal activity or managing the population size, are duties, but they are contingent, because autonomy allows different notions of criminality or deprivation, in a way that death or the ability to communicate do not involve different versions and degrees.

You can punish criminals, or not, and still be able to punish criminals, or not. There is not a level of letting crime be profitable where it is no longer possible for crime to be profitable. (Quite the opposite.)

You can breed, or not, and still be able to breed, or not. Having another baby in a world at peak capacity remains possible -- that baby would just be incredibly unfortunate.

In an asymptotic sort of way, badly handled contingent duties allow things to go badly wrong, but there is not a point where neglecting the duty makes continued neglect of the duty absolutely logically meaningless.

Badly handled perfect duties make for a conceptually contradictory world.

For Kant, the perfect duty always supercedes the contingent duty, because of the way the metaphysics that underlies the ethics works. You don't have to agree with that metaphysics. But there is not a conflict here that Kant has overlooked.

There is a rule in Kantianism for exactly what you are complaining about in this case. You can ignore the distinction that matters over and over again, but that will not change the fact that this is addressed in Kant's writings.

Klocking already pointed you at the sources in comments, so I am not going to redundantly do so again.

  • I think my line of thought is that "lying"is a very broad class of things. We can also think that if everyone has children soon the planet will run out of resources and it is not sustainable. But it only is true if everyone has more than 2 children. The same with lying here, in some cases it can lead to the language not being usable, and sometimes it can't. And if we accept this, then the next problem is how can we know that all actions in the given class (e.g. "cheating": all kinds of cheating) are not universalizable? – rus9384 Apr 24 '20 at 19:03
  • He has an answer for that, too. A perfect duty has given markers, which he discusses, and if you can converge on a perfect duty, you do not instead pursue a contingent duty (one that recognizes more criteria.). The point is, he has already foreseen your argument. And what I have given is his response. I do not need to convince you he is right, in order to answer the question as posed. – hide_in_plain_sight Apr 24 '20 at 19:15
  • Yes, but the question is, how general the action should be to be considered a perfect duty. How does one decide it? As I said, having children is an action, and a very general one. While having only 2 or fewer children would be contigent. – rus9384 Apr 25 '20 at 8:49
  • OK, I have answered that question, which others have already answered in comments, or by giving sources, very explicitly and directly in the edit above. There is a definition of what contradiction in concept means, and if you have a contradiction in concept, it is stronger than any other duty established in a less thorough way. I think I am the fourth person to say that here. It remains true. And it remains your answer. – hide_in_plain_sight Apr 25 '20 at 23:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.