Groundwork of the Metaphysics of Morals, I-II, decoded


The only good thing is a good will. Intelligence, wit, courage and wealth can be misused. A clever, calculating villain is worse than a stupid one.

A good will is just plain good. It is not good because it leads to good things; it is good in itself, and it cannot be abused. Even if a good person has bad luck and accomplishes nothing, her good will shines, like a jewel, by its own light.

It is possible to do the same action with different motivations. If you own a store and do not cheat your customers, you might be honest, or you might be acting honestly because cheating them is bad business in the long run. In this case, you are not acting on the principle of honesty; you are acting out of long-term selfishness.

Similarly, many people take joy in being kind to others. However nice these people are, they are not good people. If people act to make themselves happy, they are being selfish; that they are helping an old woman across the street should not make their actions more noble. A person who brings happiness to others because it makes her happy is a selfish jerk just like everyone else.

An action’s moral worth does not come from its consequences. An action is good if it is done according the right ‘maxim’, if it is done for the right reason, if it is done with a good will. The will is between the idea of what should be done and the actual accomplishment of that idea. A person who wants (and tries) to bring the idea into reality is doing the right thing.

Suppose my wife is pregnant, and I am racing to get her to the hospital. On the way, I run over my neighbour’s dog. Did I do something bad? No. I was trying to do the right thing—to bring life safely into the world—and it didn’t work out according to plan. But I wanted it to, and my will is what matters. I did not intend to do anything wrong, and so I should not be blamed.

Duty is acting out of respect for the laws of morality. As I showed above, consequences cannot be predicted. They are the effects of the will, but not the energy of the will. Wanting the right thing is what really matters. But what are the right things to want?

I’ve already removed any consequences that might be motivations. I should not want, for example, to keep the peace, make people happy, help good people, or punish bad ones. All of these are the results of actions. The only thing that remains is this: I should want to be a good person. My will should be to act out of respect for the laws of morality.

What, then, are the laws? First, a real law must determine my duty and tell me what to will. Second, a real law obviously cannot depend on who I am. Laws apply to everyone; that is what makes them laws. Finally, a good law cannot depend on the effects of my actions.

Since I have removed all the the good consequences that would come out of respecting the law, I can only want one thing: to conform to the law, out of pure respect for lawfulness. This principle can guide me and delineate all my duties: I am never to act otherwise than so that I could also will that my maxim should become a universal law.

This means, in short, that there is one law that makes all the others. The one law is this: Act according to the law. In other words, do unto others as you would have them do unto everyone else.

Imagine that a man wanted to break a promise. His maxim would be “False promises are acceptable”. If we put this to the test, we imagine applying this maxim to everyone. Can everyone make false promises? No. If everyone made false promises, promises would be worthless, so a promise would not be a promise. The maxim, as soon as it is made universal, destroys itself in a contradiction.

No matter how inexperienced I am in the ways of the world, no matter how dumb I am, I can always find out the answer to any ethical quandary. I only need to ask myself this: Could I make my decision a rule for everyone to follow? If not, then it must be rejected.


Although many actions look ethical, far fewer are ethical. Many actions are done from selfishness. In fact, it is impossible to say that any action has ever been done purely out of duty and not at all out of selfishness. We simply cannot tell.

All moral ideas must come only from within. They must, in other words, be completely a priori. This is how they get their power. If they were not a priori, they would come from abstraction from circumstances in the world, and when those circumstances changed, so too would the laws.

The process of going from a law to an ‘ought’ is making ‘imperatives’. Imperatives are commands of reason. Because they come from universal laws, imperatives are objective. They do not depend on who you are, and everyone will agree to them. Sometimes our wills are weak, and sometimes we choose how to act out of selfishness instead of doing what is rational. Nonetheless, it is always clear what should have been done.

There are two kinds of imperatives: categorical and hypothetical. Hypothetical imperatives are of this kind: if x then y. X is the goal. Y is the step to take. If you want a good job, go to school is an example of a hypothetical imperative.

If we were like God and could know in advance with perfect certainly how everything would work out, the obligations of hypothetical and categorical imperatives would be the same. The things that make me happy would be the things that I am required to do, and the actions required by duty would make me happy. Obviously, though, this is not the case. People think that money will make them happy, but they forget about the envy and anxiety that money creates. People want to live forever, but who would want to live forever in misery? Empirical ethics is, at best, only a rough guide: be frugal, be kind, work hard, that sort of thing. Therefore we must look for an a priori categorial imperative.

Categorical imperatives are hard to explain. Categorical imperatives are ‘good in themselves’. They are the commands of the will according to reason, which in turn is in accordance with the universal and objective laws. “Be good to your mother” is a categorical imperative. There’s something disgusting about saying “If you want a cookie, be good to your mother”. Everyone should always be good to their mothers. That is categorical.

A categorical imperative applies to everyone. It cannot be contradicted—otherwise, of course, it would not be categorical and it would not be imperative. By stripping away all the “if… thens” of hypothetical imperatives, I can see what the categorical imperative is. What is the one rule that all people must follow? It is this: Everyone must follow the rules.

That, in short, is the categorical imperative. I can be more precise, though: “Act only on that maxim whereby thou canst at the same time will that it should become a universal law.” I can also be a bit less precise but more clever: “Do unto others as you would have them do unto everyone else.”

All duties can be derived from the categorical imperative. Here are a few examples.

A husband would like to sleep around on his wife. The maxim of his action will be: “Everyone should commit adultery.” But this cannot be a universal rule. If all people cheated on their spouses, cheating would no longer be cheating. Adultery requires that a spouse not know. If everyone slept around, every spouse would know, and adultery would become not adultery. This would be a contradiction.

A woman would like to borrow money and not repay it. Her maxim will be: “Promises must be broken.” This cannot be a universal law. If everyone were allowed to break promises, a promise would not be a promise. That too is a contradiction.

A woman would like to capriciously hit her son. The maxim is “people may hit one another to make themselves feel better.” This cannot work: if everyone hit one another, we would all feel worse. The action we had undertaken to feel better would lead to agony, a contradiction.

A poor man would like to buy a new Ferrari, impoverishing his family. His maxim “Everyone should go into debt to make themselves happy.” The contradiction is obvious: not everyone can go into debt; some people must be creditors for others to be debtors. He cannot ethically go into debt.

Sometimes we want an exception from the universal law. We want to get away with something we know is wrong. Then there is a contradiction in our own wills: we want there to be a law, but do not want it for ourselves.

All rational things are ends-in-themselves, and none should be used as a means or as a stepping stone to some other goal. Irrational beings have value only as means. That is why they are called ‘things’. Rational beings have absolute worth. That is why they are not things, but are human beings. Things have prices; ends do not have prices.

Every person thinks of herself as an end and worthy of respect, and every person knows that all other people deserve our full respect—even if we are sometimes reluctant to admit it. Because this is true for all people, this principle is objective, just like the categorical imperative. Accordingly, the practical imperative will be: Treat everyone (including yourself) as an end, never as a means.

If we look back on other ethical philosophies, it is easy to see why they failed. They all held that people were bound by laws. They did not see that the only laws that people are bound by are those they make for themselves. A person can be compelled to act in a certain way, but she cannot be compelled to will a certain way. She might be made do things that lead to good consequences, but she cannot be made to want the right things to happen. Therefore, nobody can be made to truly follow the law. Every person must choose to follow the laws.

Every person must be her own lawmaker. This may sound like everyone will choose to make and follow laws that lead to her own advantage. But, as I said, the laws are derived from the categorical imperative, and so they will be objective. In other words, everyone must, due to logic, agree on the laws. All rational beings must agree that they must treat themselves and everyone else as ends-in-themselves. This results in a systematic union of common, objective laws. It results in what I call “a kingdom of ends”. Of course it is only an ideal.

Still, when a person is a member of the kingdom of ends, she gives laws, and she is subject to them at the same time. She is a ruler who chooses to be a subject … subject to the rules. This is what freedom of will allows. Freedom of will lets us be lawmakers and law-followers at the same time.

We can now end where we started, with the unconditionally good will. An unconditionally good will would act in such a way that the maxim of its actions could be a universal law and never contradict itself. The only way a will can never contradict itself is by following the supreme law, the categorical imperative, “Act always on that maxim that you can will to be a universal law”. This, then, is the formula for an absolutely good will.