This is how I understood the categorical imperative: an action is moral if it, when universalized (the constraint), is good (utilitarian).

The need to universalize can be seen as a constraint: that is, one cannot apply different principles to different scenarios. If we are allowed to come up with sufficiently flexible principles, such as deciding whether telling a white lie is okay under all possible circumstances, then the categorical imperative idea will become useless. In this sense, I see the need to universalize as a constraint on the space of possible principles to consider.

I’m sure people will object to my view. Can you tell me if I am wrong, and if so, why?

  • What you suggest is essentially rule utilitarianism. It has little to do with Kant and categorical imperative, but the resulting set of rules can be close in practice to Kant's deontology. But for Kant following rules come first, we should do things because they are right in themselves, not because their consequences are "good", or "good on average" (whatever that means), as in utilitarianism. Not lying to a murderer at the door is a famous example of how those things come apart.
    – Conifold
    Aug 16, 2020 at 0:27
  • I understand, but in the Kantian world, how are the rules decided in the first place? As far as I can tell, the rules are defined using utilitarian principles (with constraints), no?
    – J Li
    Aug 16, 2020 at 20:17
  • 1
    No. Rules are decided based on intrinsic worth of an action, not utility of its consequences.
    – Conifold
    Aug 16, 2020 at 20:33
  • But what determines the intrinsic value of the action, other than “does this action lead to the greater good for society when universalized”?
    – J Li
    Aug 16, 2020 at 21:44
  • 1
    Nothing else, value is or is not inherent in it, and what it leads to be damned. "A good will is not good because of what it effects or accomplishes, because of its fitness to attain some proposed end, but only because of its volition, that is, it is good in itself and, regarded for itself, is to be valued incomparably higher than all that could merely be brought about by it in favor of some inclination and indeed, if you will, of the sum of all inclinations".
    – Conifold
    Aug 17, 2020 at 3:59

4 Answers 4


Both, rule-utilitarianism and the categorical imperative build on universalization. Yet, universalization works differently in these approaches.

Consider the following example of whether promises have to be kept at all cost or whether occasionally breaking them is in order.

According to rule-utilitarianism, one would have to compare the utility generated by the consequences of following the rule "keep promises at large" vs. the utility generated by the consequences of following the rule "keep promises at discretion". This may theoretically go either way, although the benefits of kept promises are probably overwhelming.

According to Kant's categorical imperative, one would have to evaluate whether breaking promises at will can be a universalizable maxim. The question is whether one can have the maxim of keeping promises at discretion and wanting this to be a universal law.

Why can't this maxim be universalized according to Kant?---The concept of a promise builds on the idea that it is kept. It is illogical (in Kant's terms) to at the same time grasp a maxim that builds on the idea of promise and have the universal version of the maxim nihilate this very idea.

So, these two "schools" work with totally different approaches/mindsets. Kant is about the possibility of the universalized maxim becoming a universal law (deontologic approach). Rule utilitarianism its about rule-action rendering consequences with a maximum of utility (consequentialism).

  • If I understand your answer, then you are effectively agreeing with me. I understand that Kant is concerned about rules, but the way to decide (once and for all) whether the rules are justified is to apply utilitarian principles with constraints.
    – J Li
    Aug 16, 2020 at 20:19
  • No. There is virtually nothing utilitarian in Kant. The way to decide whether an action is right is considering whether the maxim behind the action could be a universal law. But not on utilitariajn grounds. The maxim must be universalizable on grounds of reason alone. For instance: Not keeping promises is wrong because if promises wouldn't bind anybody the concept of promise which is part of the maxim would be nihilated.
    – Mr. White
    Aug 17, 2020 at 3:49
  • I see. Let’s take the following example then. Does the Kantian view lead to the conclusion that “you should not murder innocent people”? If so, why? Is there a reason behind such a conclusion that does not rely on something that is related to “what is good for people/society”?
    – J Li
    Aug 17, 2020 at 3:56
  • The judgement of killing people is a bit more complicated because the argument goes down to respecting others as free agents. Might be worth a question in itself. In a nutshell: The maxime of killing sb (innocent, as you say) is not universalizable because this would potentially result in everyone killing any other free agent at will. If you grasp the maxime of killing sb (innocent) you basically agree to being killed yourself. Thereby, you at the same time act as a free agent (by universalizing the maxime) and agree on this free agent being killed. A conflict of reason, according to Kant.
    – Mr. White
    Aug 17, 2020 at 5:14
  • Kant's approach has always been critisized as impractical. Even if this is true, his idea is unique and neat in way.
    – Mr. White
    Aug 17, 2020 at 5:17

Kant isn't easy, and I'm no expert. But let me try a slightly different turn, which may help clarify.

As noted, utilitarian or consequentialist ethics is concerned with the outcome of an action. The action is good if the consequence is good. But then, as you noted, you have to define what you mean by "good" and so on in an infinite regress of relative means and ends.

(As an aside, one reason Kant rejects this view is that it assumes we can actually predict outcomes, while in reality life is full of unintended consequences. And because of its relativism. You can always move the goal posts and redefine the good or appeal to simple majorities.)

Kant was deeply concerned with moral law in the emerging world of science and utility. His whole approach sought a path out of such relativistic dilemmas, developing a complex, utterly original philosophical set of critiques.

He assumes, in some sense, a human subject that is rational and "free" to make moral choices. Rather than point to "evidence" or appeal to axioms, he painstakingly demonstrates what "must already be the case" for such a being to exist. A moral being must have a capacity to both know what is "good" and yet be able to freely choose. What sorts of mental relations and categories must be universally the case for this to be so in the first place?

So, Kant is looking at the total logical makeup, relations, and categories that must exist "universally" for all such creatures, all "rational beings." He is not concerned with the "psychology" of this or that person or "sociology" of this or that society.

He uses the term "hypothetical imperative" to describe actions that are means to given ends, as in utilitarian "reasoning" towards some goal. If...then. But the "categorical imperative" is a rule that must be consistent with the very existence of "reasoning" itself. It must be "universal" for all reasoning beings and not contradict the categories of thought necessary to the very act of reasoning.

Morality is not so much about external ends, good or bad, but about the internal logic and coherence of a "being that can reason" and all that entails. Lying, for example, is always wrong even if it saves lives, because the act of lying "logically" contradicts the very basis of a lie, which is premised on the default assumption honesty that makes language possible to begin with. If everyone lies, that is, no one could lie.

Now admittedly, this is not a very useful guide to actual ethical choices. Nor is this a very satisfactory explanation. The only real and complete answer to your question pretty much entails all of Kant. But the way to start think about it is that the CI is justified more by internal coherence with "reasoning" than by the "reasons" given for some specific end. It is that ultimate "reason" for which reasoning itself is the end.

  • 1
    Thank you for the answer. I (think) I understand this point. My main complaint is that what Kant has laid out using “pure reason” often looks like an ex-post rationalization of what we would get from utilitarian reasoning. For instance, justifying the rule of “we shouldn’t murder innocent people” is easy from utilitarian grounds, but will be complicated under CI. Of course, I’m sure Kant will disagree with my view :)
    – J Li
    Aug 17, 2020 at 22:39
  • 1
    First, utilitarian views can run into a lot of problems, and it usually isn't so easy as it seems at first. Kant, of course, wasn't just working out ethics, but a whole system with a rational structure. So yes in a sense he was "rationalizing" intuitive ideas. He was very pleased that his CI produced, on a deeper, rational basis, almost the same intuitive idea as the golden rule, "do unto others..." which he thought gave it even greater credence. The CI does illuminate the essential, rational structure of the golden rule. Aug 17, 2020 at 23:05

As Rawls notes in AToJ, universality (and its sister generality) are common descriptors for moral claims, not just Kant, with Kant specifically emphasizing a concept of autonomy vs. these descriptors.

Now Kant says moral value is absolutely infinite, i.e. transfinally maximum for its order and not interchangeable or replaceable. So for Kant, you can't add the goodness of persons to get a larger amount of goodness in a group of people. Anachronistically, even adding the smallest infinity to itself as many times as itself, equals that infinity. Indeed, adding any size of infinity to itself as many times as itself, equals the basic addend. If you multiply an infinity by itself, some infinite number of times, then you can get a larger infinity, but in the physical world, it is hard to see where this arithmetic would appear (maybe summing over all persons in all possible worlds?) and again, in Kant, the infinity of moral value is greater than all transfinite numbers in particular, because it is not commensurate with them (whereas the alephs are all commensurate with each other).

So concepts whose logic tends towards utilitarianism when used a certain way, do not do so when used in other ways, despite appearances. (Consider that Moore spoke of ethical value as "what ought to exist for its own sake." If A ought to exist, and B ought to exist, though, it does not follow that the use of the phrase 'ought to exist' twice means that the group of A and B ought to exist twice as much as either A or B alone, does it?)

  • I appreciate your answer and don’t fully understand it (due to my own lack of knowledge in philosophy). My whole point is that the way Kant judges an action to be moral or not sounds rather consequentialist to me. For instance, “lying is bad because if we universalize it, it is bad (for the society)”. Isn’t that a consequentialist judgment?
    – J Li
    Aug 16, 2020 at 21:54
  • Consequentialism is perhaps an unfortunate name for the idea. A moral principle isn't consequentialist because it relates to conditional/hypothetical factors but if it is aimed towards maximizing the amount of good, as if goodness is a finite quantity like gold or uranium. So maybe "productivism" might've been better? As it is, the name it got stuck to it as a contrast to people, like Kant, who said things like "let justice be done, though the heavens would fall." Aug 17, 2020 at 0:30

TL;DR: They are different on a more fundamental level.

Moral theories seek to explain why something is wrong. In that sense, Kantian deontology and utilitarianism are completely different, even if you add additional constraints to each such that they arrive at the same conclusions (the possibility of which is itself doubtful, see below). Kant argues issues of morality must be a priori, that is, not contingent on empirical things like the specific natures of the people involved or how the action in question actually plays out. On the other hand, utilitarian is consequentialist, that is, the consequences determine whether something is good or bad. More specifically, classical utilitarianism argues something is good if it maximizes the pleasure for the most number of people--this is certainly an empirical question.

Going back to the point about the possibility of introducing additional constraints to make the two approaches converge, I am highly doubtful this is possible because as long as the two theories are putting forth conceptions of the good that are not intensionally equivalent, then one will always be able to come up with a hypothetical counterexample that one theory endorses with and the other rejects.

Also important to note, as Conifold and Clyde Frog pointed out, your understanding of Kant's Categorical Imperative is wrong--if construed the way you do, Kant would also be consequentialist, at which point it would be possible to employ various constraints to make it equivalent to (some formulation of) utilitarianism. But Kant is most certainly not consequentialist--as Clyde explains, he's more concerned with whether the universalization of a maxim will result in a logical, teleological, or practical contradiction (see Christine Korsgaard's paper Kant's Formula of Universal Law on this if you're interested; the PDF is easily accessible if you Google it).

EDIT: Here are some excerpts directly from Kant (taken from https://www.earlymoderntexts.com/):

So an action’s moral value doesn’t lie in the effect that is expected from it, or in any principle of action that motivates it because of this expected effect. All the expected effects—something agreeable for me, or even happiness for others—could be brought about through other causes and don’t need the will of a rational being (Grounding, 401)

Obviously the false promise isn’t made prudent by its merely extricating me from my present difficulties; I have to think about whether it will in the long run cause more trouble than it saves in the present. Even with all my supposed cunning, the consequences can’t be so easily foreseen. People’s loss of trust in me might be far more disadvantageous than the trouble I am now trying to avoid, and it is hard to tell whether it mightn’t be more prudent to act according to a universal maxim not ever to make a promise that I don’t intend to keep. But I quickly come to see that such a maxim is based only on fear of consequences. Being truthful from duty is an entirely different thing from being truthful out of fear of bad consequences; for in the former case a law is included in the concept of the action itself; whereas in the latter I must first look outward to see what results my action may have. (Grounding, 402)

How can I know whether a deceitful promise is consistent with duty? The shortest way to go about finding out is also the surest. It is to ask myself: Would I be content for my maxim (of getting out of a difficulty through a false promise) to hold as a universal law, for myself as well as for others? Immediately I realize that I could will he lie but not a universal law to lie; for such a law would result in there being no promises at all, because it would be futile to offer stories about my future conduct to people who wouldn’t believe me; or if they carelessly did believe me and were taken in, would pay me back in my own coin. Thus my maxim would necessarily destroy itself as soon as it was made a universal law. (Grounding, 403)

  • I understand your point. My question is about the criterion that determines whether an action is justified in the Kantian view. I might be misusing the term, but the Kantian criterion strikes me as consequentialist. For instance, “lying is bad” because “if we universally allow lying, this is bad for the society”. Therefore, isn’t the judgment consequentialist and deriving from whether some action (again, when universalized) is bad for society?
    – J Li
    Aug 16, 2020 at 21:49
  • @JLi Read the first and last paragraphs of my answer again: "Kant argues issues of morality must be a priori" and "he's more concerned with whether the universalization of a maxim will result in a logical, teleological, or practical contradiction." Regarding lying, Kant does NOT say that it is bad for society if everyone lied (why would he if he's trying to interrogate what it means for something to be bad?); in fact, he opposes that approach explicitly. Instead, he argues lying completely undermines communication if universalized, and without communication, there's no lying.
    – Zachary
    Aug 17, 2020 at 6:18
  • I understand your point. However, I want to point out two things. First, it seems that the “a priori judgments” that Kant makes are often quite similar to what a consequentialist view will generate (when applying the constraint that the results have to be universal rules), suggesting that perhaps such “a priori rules” are not so a priori after all. Second, if we are really forced to give up any notion of consequences, I really don’t see any reason why “let’s murder people at will” cannot be a universally applicable rule. What is intrinsically wrong with it?
    – J Li
    Aug 17, 2020 at 6:37
  • @JLi Have you read Kant (and I don't mean this aggressively)? It is one thing to argue that the Categorical Imperative reduces to utilitarianism (as many have), and a completely different thing to say that Kant's arguments are consequentialist. In the former, you only need to explicitly state your formulation of the CI, which might not require a close reading of Kant (although of course people could argue against your formulation), whereas in the latter, you have to quote Kant. I myself, for example, have argued that Rawls is utilitarian, and that required extensively reading Rawls.
    – Zachary
    Aug 17, 2020 at 6:49
  • I have not read Kant. I have only read people interpreting him. I’m aware that my views, at best, standing on shaky grounds. To be clear with my point: I do not claim that Kant’s reasoning is consequentialist. I meant CI can be reduced to utilitarianism with constraints — the constraint being that rules have to be universal and cannot be “very state contingent”. I am also insinuating that I think a lot of what Kant claims to “derive from pure reason” are really motivated by consequentialist thinking (such as the murder example), and the “reason” provided are more “rationalizations”.
    – J Li
    Aug 17, 2020 at 6:54

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .