What should I do?

Kant's second great question concerns morality. His answer starts neither with commandments, feelings nor consequences — but with the will of the person acting.

In 1785 Kant published the “Groundwork of the Metaphysics of Morals” — a short book with an enormous ambition: to identify and establish the supreme principle of morality, independent of worldviews, traditions and contingent experience. In three “transitions” it leads from everyday moral experience to philosophical justification. Its famous starting point is the good will: nothing in the world, or even beyond it, can be called good without qualification except a good will alone.

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.

Kant, Groundwork of the Metaphysics of Morals (1785), Ak. 4:394 — English translation

Talents such as wit and judgment, strengths of character such as courage, even happiness — all of these are only conditionally good for Kant: in the hands of a bad will they can do harm. The good will alone shines by its willing, not by its success. The course is set: morality is decided within, not by results.

Source: Wikipedia: Groundwork of the Metaphysics of Morals

Statements of “ought” are what Kant calls imperatives — and he distinguishes two fundamentally different kinds:

Hypothetical imperative

Binds only on condition of an end: “If you want to stay healthy, don't eat too much chocolate.” Such rules of skill and counsels of prudence oblige no one morally — give up the end and the rule lets go of you.

Categorical imperative

Commands unconditionally, independent of ends and inclinations. It obliges not through the content of an action but through its form — the universalisability of the maxim — and holds for every rational being.

The test

State your maxim → turn it, in thought, into a universal law → check for contradiction. If the maxim fails already in conception (if everyone made false promises, no promise would be believed — the very concept destroys itself), it violates a perfect duty; if it merely cannot be willed consistently, an imperfect one.

Not the golden rule

“Do as you would be done by” measures actions against one's own wishes. Kant dismissed it as trivial: it is merely derived from the categorical imperative and cannot be a universal law, since it grounds neither duties to oneself nor duties of love towards others (Ak. 4:430, footnote). The categorical imperative instead tests maxims against reason.

Sources: Wikipedia: Categorical imperative · Wikipedia: Groundwork of the Metaphysics of Morals

Kant insists there is only one categorical imperative. The famous formulas of the Groundwork light up the same principle from different sides:

1. 1
   Universal law formula

   “Act only according to that maxim whereby you can at the same time will that it should become a universal law.” (Ak. 4:421) The basic formula: whatever is to count for me must be thinkable — and willable — as a law for everyone.

2. 2
   Law of nature formula

   “Act as if the maxims of your action were to become through your will a universal law of nature.” (Ak. 4:421) A thought experiment: imagine a world in which your rule holds as exceptionlessly as gravity — does it still work?

3. 3
   Humanity formula (end in itself)

   “Act in such a way that you treat humanity, whether in your own person or in the person of any other, never merely as a means to an end, but always at the same time as an end.” (Ak. 4:429) People must never be merely used — the core of what we now call human dignity.

4. 4
   Autonomy and the kingdom of ends

   The will of every rational being is a universally legislating will: it obeys no alien law but its own — autonomy instead of heteronomy. And all self-legislators together form an ideal “kingdom of ends”, of which each agent should act as a lawgiving member.

Sources: Wikipedia: Categorical imperative (formulations); Kant quotations in English translation of the German original · Wikipedia: Groundwork of the Metaphysics of Morals

The Socratic mirror in the workshop is named after Socrates, who tested in conversation whether the convictions of the person he was talking to fitted together. The mirror does something similar with your own verdicts: if two of them pull in different directions, it puts them side by side and asks how the cases differ. It does not say which verdict is the right one.

Source: Wikipedia: Socratic method

The French-Swiss thinker Benjamin Constant objected: a principle by which one would have to tell the truth even to a murderer asking whether the friend he is pursuing is hiding in one's house would make any society impossible. Duties correspond to rights — and no one has a right to a truth that harms others. The benevolent lie, Constant concluded, must be permitted.

Kant replied in 1797 with the essay known in English as “On a Supposed Right to Tell Lies from Benevolent Motives” — and did not budge: truthfulness is an unconditional duty of right, a sacred command of reason that no convenience may restrict (Ak. 8:427). A lie always harms — “if not some human being, then it nevertheless does harm to humanity in general, inasmuch as it vitiates the very source of right” (Ak. 8:426, in translation): whoever lies, however good the intention, undermines trust in declarations as such — the very foundation on which all right rests. Admitting exceptions, Kant argues, would destroy the universality that makes principles principles (Ak. 8:430).

In his essay Kant himself works through what may follow from a lie at the door, and arrives at an argument about liability for unforeseen consequences.

Sources: Wikipedia: On a Supposed Right to Tell Lies from Benevolent Motives · Wikipedia (in German): Über ein vermeintes Recht aus Menschenliebe zu lügen (reception)