Sean Daily

A friend texts you late at night: “If my boss calls, tell her I was with you all evening. Please.” It is a small lie. Nobody gets hurt, probably. Your friend keeps a job, and you keep a friend. So why does something in you hesitate? Kant’s categorical imperative is one of the most famous attempts to explain that hesitation. Immanuel Kant argued that the question is not “What happens if I lie this once?” but “What would I be willing for everyone to do?” Below you will find what the categorical imperative is, how its main formulas work step by step, why Kant thought you may not lie even to save a life, and where his critics think he went wrong. By the end, you can decide what you would text back.

〰

Kant’s Categorical Imperative: What It Is and Where It Comes From

Immanuel Kant was born on April 22, 1724, in Königsberg, Prussia (today Kaliningrad, Russia), and died there on February 12, 1804, according to Encyclopaedia Britannica. He set out his moral theory in several books, but the short Groundwork of the Metaphysics of Morals (1785) is where most readers meet the categorical imperative. Britannica calls it more readily comprehensible than his later Critique of Practical Reason, despite its forbidding title.

The book opens with a bold claim. In T. K. Abbott’s public-domain translation: “Nothing can possibly be conceived in the world, or even out of it, which can be called good, without qualification, except a good will.” Intelligence, courage and even happiness can all be used badly. A good will, Kant says, is good in itself, not because of what it achieves.

To see what a good will looks like, Kant offers a shopkeeper. As Abbott renders it, “it is always a matter of duty that a dealer should not over charge an inexperienced purchaser.” A prudent shopkeeper charges a child the fair price, but maybe only because cheating would hurt business. His action matches duty without being done from duty. The Stanford Encyclopedia of Philosophy (SEP) summarizes Kant’s view this way: actions done from self-interest or even sympathy can conform to duty, but only actions done from duty have moral worth.

〰

Two Kinds of “Ought”: Hypothetical vs. Categorical

Most of the “oughts” in your day are conditional. You ought to leave by 8:00 if you want to catch the train. You ought to stretch if you want to avoid a sore back. Kant calls these hypothetical imperatives. In Abbott’s wording, they “represent the practical necessity of a possible action as means to something else that is willed.” Drop the goal and the command disappears.

A categorical imperative is different: it “represented an action as necessary of itself.” It does not say “if you want X.” The SEP explains that it applies to you simply because you have a rational will, not because of any end you happen to hold.

	Hypothetical imperative	Categorical imperative
Form	“If you want X, do Y”	“Do Y” (no condition)
Who it binds	Only people who have goal X	Every rational being
Example	“If you want to be well-liked, do not lie” (Britannica’s example)	Do not make a lying promise
What happens if you drop the goal	The command no longer applies	The command still applies

〰

The Formulas, Step by Step

Kant said there is only one categorical imperative, but he stated it in several ways. The SEP and the New World Encyclopedia both discuss four: Universal Law, Humanity, Autonomy and the Kingdom of Ends. Commentators disagree on how many are truly distinct and on Kant’s claim that they are equivalent. The first two do most of the everyday work.

1. Universal Law. “Act only on that maxim whereby thou canst at the same time will that it should become a universal law.” A maxim is simply the personal rule behind your action, such as “When I need money, I will borrow it and promise to repay, knowing I won’t.” The SEP lays out the test as a procedure. First, state your maxim. Then imagine a world where everyone acts on it. Then ask two questions:

• Can that world even be conceived? If not, you have a “contradiction in conception,” and you are violating a perfect duty (one with no exceptions).
• Could you rationally will that world? If not, you have a “contradiction in the will,” and you are violating an imperfect duty (one that leaves room for judgment about when and how).

The lying promise fails the first question. If everyone made false promises, Kant writes, it “would be in vain to allege my intention in regard to my future actions to those who would not believe this allegation.” Promising itself would collapse. “Hence my maxim, as soon as it should be made a universal law, would necessarily destroy itself.” A maxim of never helping anyone in need, by contrast, describes a world you can picture. The New World Encyclopedia explains why it still fails: you cannot rationally will it, because you yourself will someday need help.

2. Humanity. “So act as to treat humanity, whether in thine own person or in that of any other, in every case as an end withal, never as means only.” Notice the word “only.” You use the barista as a means to coffee every morning, and that is fine, because they agreed to the exchange. What Kant forbids is using someone in a way they could not consent to, such as deceiving them so they serve your plan.

3 and 4. Autonomy and the Kingdom of Ends. These ask you to act as if you were helping write the moral law for a community of free, rational people. “By a kingdom I understand the union of different rational beings in a system by common laws,” Kant writes, and “In the kingdom of ends everything has either value or dignity.” Things have a price and can be swapped. Persons have dignity and cannot. The New World Encyclopedia notes that this idea later influenced John Rawls.

Kant’s example	Type of duty	Which test it fails
Suicide to escape suffering	Perfect duty to oneself	Contradiction in conception
Making a lying promise for money	Perfect duty to others	Contradiction in conception
Neglecting one’s talents	Imperfect duty to oneself	Contradiction in the will
Refusing to help others in need	Imperfect duty to others	Contradiction in the will

A short detour, because readers often ask: isn’t this just the Golden Rule? Kant did not think so. As the Wikipedia overview of the categorical imperative reports, he saw the Golden Rule as dependent on what each person happens to want. In a footnote to the Groundwork he points out that a convicted prisoner could use it to ask the judge for release, since the judge would not want to be locked up either. The universal law test asks a different question: not “What would I want done to me?” but “Could this rule hold for everyone?”

〰

The Strongest Objections

1. The murderer at the door. The French writer Benjamin Constant argued that if truth-telling is an absolute duty, you would have to tell a murderer where his intended victim is hiding. Kant answered in a 1797 essay, “On a Supposed Right to Tell Lies from Benevolent Motives,” and accepted the conclusion: you may not lie, even then. He held that a lie always harms someone, if not a particular person then humanity in general. The scholar Allen W. Wood calls the essay “famous (or infamous),” and in 2025 Jens Timmermann published a whole book on it. One detail is often forgotten: both Kant and Constant accepted that simply refusing to answer would be consistent with the categorical imperative. The argument assumes, for the sake of argument, that silence is not an option.

2. Is acting from duty really better than acting from love? If you visit a sick friend because you care, rather than because duty commands it, did you do something with less moral worth? Many readers find that conclusion cold. The SEP notes that Kant’s defenders reply that he means duty must be decisive in your deliberation, not that warm feelings are worthless.

3. The tests don’t always work cleanly. Even sympathetic readers admit that some of Kant’s own derivations are shaky. The SEP calls his argument about developing one’s talents unconvincing as stated. Arthur Schopenhauer went further, arguing (as Wikipedia summarizes) that the categorical imperative is secretly hypothetical and egotistical, since we reject certain maxims because we fear their results for ourselves. Wikipedia’s article judges this criticism weak, but the worry it raises survives in a softer form.

4. Is Kant a consequentialist in disguise? John Stuart Mill and R. M. Hare, as the SEP reports, read Kant’s tests as really being about outcomes. David Cummiskey has argued that respecting the value of every person leads to producing the best overall results. Others, such as Barbara Herman, Paul Guyer and Allen Wood, defend readings in which a value (rational agency, autonomy or humanity) grounds the rules. Many scholars treat this as one of the liveliest open debates in Kant studies.

〰

The Categorical Imperative in Everyday Life

You can run Kant’s test in about ten seconds. Before you pad a résumé, fake a sick day or write a glowing review for a friend’s restaurant you never visited, state your maxim honestly and universalize it. If everyone padded résumés, résumés would stop meaning anything, and your padding would stop working. That is the same self-destruction Kant saw in the lying promise.

The Humanity formula reaches into newer territory too. The SEP’s discussion of artificial intelligence argues that, on Kant’s view, responsibility stays with the people who design, fund and use these systems, and that dilemmas like the choices of self-driving cars do not get obvious answers from Kantian principles. What the principles offer instead is a question to keep asking: is anyone being treated merely as a tool?

Now go back to your friend’s text. Kant’s answer seems clear: don’t lie to the boss. You could refuse, or stay silent, but you may not say what is false. Yet the price is real. Your friend may lose a job, and you may lose a friend, while the boss loses nothing she would ever notice. A consequentialist would weigh those outcomes and might tell you to lie. Kant would say that the moment you start lying when the stakes feel right, you have made your honesty conditional, a hypothetical imperative after all. Which of those costs are you prepared to pay?

〰

Questions to Think About

• Think of a rule you broke recently. What maxim were you acting on, and could you will that everyone act on it?
• Is there any situation where you would lie to the murderer at the door? If so, what does that tell you about whether your duty not to lie is categorical?
• Does a kind act done out of love deserve more praise than the same act done out of duty, or less?

〰

FAQ

Q: What is the categorical imperative in simple terms?

It is Kant’s basic moral rule: act only on principles you could will everyone to follow, and never treat a person merely as a tool. It applies to everyone regardless of their personal goals.

Q: What is the difference between a hypothetical and a categorical imperative?

A hypothetical imperative depends on a goal (“If you want X, do Y”). A categorical imperative commands unconditionally, whether or not you want anything in particular.

Q: How many formulations of the categorical imperative are there?

Four are usually discussed: Universal Law, Humanity, Autonomy and the Kingdom of Ends. Scholars disagree about how many are truly distinct, and Kant claimed they are equivalent ways of stating one principle.

Q: Would Kant really refuse to lie to a murderer?

In his 1797 reply to Benjamin Constant, yes: he held that lying is wrong even then. He and Constant both accepted, though, that refusing to answer would not violate the categorical imperative.

〰

Related Reading

• Trolley Problem Explained: Switch, Footbridge, and Hard Choices
• Social Contract Theory Explained: Hobbes, Locke, and Rousseau
• Chinese Room Argument Explained: Can a Machine Understand?

This article is an introduction, not a substitute for reading the original texts.