Schopenhauer: Man can do what he wants, but man can't want what he wants.
|
|
Time to read 19 min
|
|
Time to read 19 min
Who said “Man can do what he wills, but he cannot will what he wills”?
What does Schopenhauer’s quote mean?
Did Schopenhauer believe in free will?
If we can’t choose what we want, are we still responsible for what we do?
Who said it before Schopenhauer?
Last updated: November 17, 2026
"Man can do what he wills, but he cannot will what he wills" sums up Arthur Schopenhauer's view of free will: you are free to act on your wants, but you never chose the wants themselves. The idea comes from his Prize Essay on the Freedom of the Will, which won the Royal Norwegian Society of Sciences' prize in 1839 and argues that at any given moment a person can want only one definite thing. The essay never puts it in that snappy form, though. The one-liner spread mainly through Albert Einstein, who cited Schopenhauer in his 1930 credo "What I Believe" and called the thought a lifelong consolation.
It sounds like a riddle on a mug. It is the compressed end point of a whole theory about what wanting is, where it comes from and why it feels free when, on Schopenhauer's account, it isn't. He explains that theory with an ordinary evening after work, and the evening does more persuading than most of his metaphysics.
The idea is Schopenhauer's, and the famous wording is a compression of it. In 1839 the Royal Norwegian Society of Sciences in Trondheim gave him its prize for an essay on a single question, which Christopher Janaway renders in his Cambridge edition as "Can the freedom of the human will be proved from self-consciousness?" Schopenhauer answered no. In 1841 he published the essay together with his essay on morality, which the Royal Danish Society had refused to award a prize, under the title The Two Fundamental Problems of Ethics.
The German text says it at greater length and with more precision. Early on, Schopenhauer puts the decisive question to the reader: "Kannst du auch wollen, was du willst?" ("Can you also will what you will?"). The second chapter ends with his answer. "Du kannst thun was du willst" ("You can do what you will"), but at any given moment of your life you can will only one definite thing and nothing else. The final chapter boils it down once more: "Der Mensch thut allezeit nur was er will, und thut es doch nothwendig" ("Man always does only what he wills, and yet does it necessarily"). All three passages can be checked in the public-domain edition of his collected works on archive.org.
The one-liner most English speakers know took a detour through physics. In "What I Believe", published in Forum and Century in October 1930, Einstein wrote that he did not believe in free will in the philosophical sense and cited Schopenhauer's saying that a man can do what he wills, but "he cannot determine what he wills". The German version in Mein Weltbild (1934) reads "Ein Mensch kann zwar tun, was er will, aber nicht wollen, was er will", and the familiar English line sounds much closer to that. Searches split between "wills" and "wants" because German wollen covers both. "Will" keeps the weight of Schopenhauer's technical term, the Will with a capital W on which his whole metaphysics rests. "Want" sounds more natural. The claim survives either translation: your actions follow from your will, and your will follows from what you are.
Schopenhauer did not present the necessity of willing as his own discovery. The fourth chapter of the essay, titled "Vorgänger" (predecessors), traces a line of thinkers who ran into the same problem before him, from Augustine and Luther through Hobbes, Spinoza, Hume and Joseph Priestley to Kant. He also quotes, in Latin, a famous image from a letter by Baruch Spinoza: a stone pushed into motion by an outside cause which, if it were conscious of its own striving, would think it was moving freely. As the IEP's entry on Spinoza summarises, the stone "would believe itself free to move, though it is determined by external forces." Replace the stone with a person who feels their desires but not their causes, and you are very close to Schopenhauer's point.
For Schopenhauer, every action follows from a motive acting on a character, as necessarily as an effect follows its cause. The feeling of freedom comes from surveying the options, and surveying them tells you nothing about what decides between them. His own illustration sits in the third chapter of the Prize Essay.
A man stands in the street and thinks: "Es ist 6 Uhr Abends, die Tagesarbeit ist beendigt" ("It is six in the evening, the day's work is done"). He could go for a walk, or to his club, or climb a tower to watch the sunset, or go to the theatre, or visit this friend or that one, or even walk out through the city gate into the wide world and never come back. All of it, he tells himself, is entirely up to him. Then he does none of it and goes home to his wife, "just as freely". Schopenhauer compares him to water telling itself that it could make high waves, rush downhill, plunge as a waterfall, shoot up as a fountain or even boil away, yet chooses to stay calm and clear in the pond. Water can do every one of those things, but each only when its conditions are present, and the same holds for the man. Both passages are in the original German and in an English rendering on Cambridge Core.
The man's monologue gets one thing right: nothing physically stops him. Schopenhauer grants that and calls it physical freedom, the absence of obstacles. He also grants intellectual freedom, a clear view of one's motives, which in the essay's appendix he says can be suspended by madness, intoxication or overwhelming emotion. What he denies is moral freedom, the idea that in exactly the same situation the man could just as well have wanted the club as his wife. Scholastic philosophers called this liberum arbitrium indifferentiae, the freedom of indifference. Self-consciousness, as Janaway's introduction explains, only ever reports "I can do what I will", a freedom of doing that says nothing about the willing itself.
What connects the evening to the decision is motivation. In On the Fourfold Root of the Principle of Sufficient Reason, his book on the rule that everything has a reason for being as it is, Schopenhauer calls motivation "die Kausalität von innen gesehn", causality seen from within. A motive is a cause that works through knowledge: the thought of dinner, the wife waiting, the tiredness after work. It only acts once you are aware of it, which on his account is why options you never think of can't tempt you. Once it is known, the Internet Encyclopedia of Philosophy summarises, an act of will follows from it as necessarily as "the motion of a billiard ball follows from its being struck."
The same evening sends one man home and another to the club because their characters differ, and on Schopenhauer's account each person's character is inborn and stays the same for life. Motives are the trigger. Character decides what the trigger sets off.
The vocabulary comes from Kant. Your empirical character is the pattern of behaviour that shows up over time, the one your friends could describe. Behind it sits what Kant called the intelligible character, which the IEP describes as "our character outside of space and time", "original, inalterable and inexplicable." Think of it as the core self that time doesn't touch, with the empirical character as the way that core plays out across a life. Janaway sums up Schopenhauer's claim that character is inborn and "constant and never changes." In §55 of The World as Will and Representation, Schopenhauer quotes Seneca, velle non discitur: willing cannot be taught. Teaching can change the means someone uses to get what they want, he writes, but it can never make them want something really different.
Our six o'clock man can do anything on his list, but which item pulls hardest was settled before he left the office, by a character he never picked in any moment he could remember and by motives he did not invent. He finds out what he wanted the same way his wife does, by watching where he goes. Schopenhauer says so in the same §55: the intellect learns the will's decision only after the fact, which produces what he calls "der Schein einer empirischen Freiheit", the illusion of empirical freedom.
The one thing he leaves open is easy to misread as a way out. He calls it the acquired character, and it consists of knowing your character, "möglichst vollkommene Kenntniß der eigenen Individualität", the fullest possible knowledge of your own individuality, gathered through experience. You can't become someone who loves the club. You can stop pretending you might, and stop making plans that only work for a person you aren't.
Schopenhauer's own biography supplies an uncomfortable test case. On 12 August 1821, in the Berlin lodging house where both of them lived, he got into a physical fight with a 47-year-old seamstress named Caroline Marguet, who had refused to leave the space outside his door. The Stanford Encyclopedia calls it "an angry shoving-match". She sued, the case dragged on for years, and he lost. According to the documents edited by the Schopenhauer scholar Arthur Hübscher, he ended up paying her five thalers a month in quarterly instalments, which he did for some twenty years. When she died he is reported to have noted, in Latin, obit anus, abit onus: the old woman dies, the burden goes. Read next to the Prize Essay, the episode is almost too neat. On his own theory, the temper on display that August was no passing mood. It was the character itself, which is exactly where he said the blame belongs.
For Schopenhauer, "will" means something older than any conscious decision: the blind striving at the core of everything that exists, which in us shows up as desire, hunger, fear, longing and aversion long before deliberation starts. In The World as Will and Representation (published in December 1818, dated 1819) he takes over Immanuel Kant's distinction between the world as it appears to us and the world as it is in itself, and claims to have found what Kant thought we could never know: the thing in itself is will. He later qualified this, and the Stanford Encyclopedia of Philosophy notes that in his later works it is only "to us" that the thing in itself appears as will.
He picks the word "will" because your own body is the one thing you know from two sides at once. From outside it is an object among objects, an arm moving in space. From inside the same movement is felt as wanting. Schopenhauer reads the rest of nature on that model, from gravity to the growth of plants, which is the boldest and most disputed part of his system and not something you need to accept to follow the quote. The narrower point is the one that matters here: looking inward, you don't find a neutral judge weighing options. In Janaway's phrase, the self we meet within is "fundamentally conative and affective", made of striving and feeling.
Three levels of will have to be kept apart, or the theory looks like a contradiction. At the top is the will as thing in itself, a single striving behind all appearances, which Schopenhauer considers free because causality, like the rest of the principle of sufficient reason, governs only appearances. In the middle is your individual intelligible character, which on his account also lies outside time. At the bottom are your individual acts of will, this craving now, that decision tonight. Those belong to the world of appearance, and like every event there they have causes. The famous quote is about the bottom level. How the middle level fits with the top one is a real weak spot, and Janaway's objections further down go straight at it.
The second volume of his main work, published in 1844, adds the piece that makes all this psychologically plausible. In the chapter "On the Primacy of the Will in Self-Consciousness" he reverses the usual picture of the mind. The will is "the substance" of a person and the intellect "the accident". Reason lights the way for desire instead of steering it, or in his own image, "the will is the heat, the intellect the light." That is why the six o'clock man learns his decision by watching his feet.
Schopenhauer keeps responsibility and moves it from individual actions to the character they come from. That is a stranger and more demanding position than plain fatalism, and it answers the obvious objection to the quote.
His starting point is a feeling he refuses to explain away. People have, in a phrase Janaway quotes from the essay, an "unshakeable certainty that we ourselves are the doers of our deeds." Schopenhauer draws the conclusion himself: "Where guilt lies, there must responsibility lie also." If each action follows necessarily from motive and character, the guilt has to attach to the character, to what the person is. The sentence quoted above, that man always does what he wills and yet does it necessarily, continues in the original: "Das liegt aber daran, daß er schon ist was er will", because he already is what he wills. Schopenhauer uses the scholastic formula operari sequitur esse, doing follows from being, and concludes in the final chapter that freedom, which cannot be found in the doing, "muß im Esse liegen", must lie in the being.
That also resolves the puzzle of how you can be blamed for a character you never chose in time. On his view your intelligible character is itself your own deed, which the SEP describes as "a timeless act of Will that the person essentially is." In §55 he puts the double perspective as a paradox: every thing as appearance is necessary through and through, while the will in itself is "völlig frei, für alle Ewigkeit", completely free for all eternity.
The same section draws a fine line between regret and a bad conscience. Regret, he writes, never comes from the will changing, only from knowledge changing. A bad conscience rests on "der Gewißheit, daß man den selben Willen noch immer hat", the certainty that you still have the same will. Change the evening slightly to see the difference. If the man goes to the club because he didn't know his wife was ill, he regrets it the moment he finds out. If he knew and went anyway, what he feels is worse, because no new information will change who he turned out to be.
This is where the quote leads into Schopenhauer's ethics. Seeing through the separation between yourself and others shows up in conduct as compassion, which he treats as the root of real morality: "alle wahre und reine Liebe ist Mitleid" (§67), all true and pure love is compassion. The only real break with the will comes one step further, in the denial of the will to live, an ascetic letting go of desire. Here knowledge acts as a "Quietiv", a sedative of all willing, instead of as one more motive, and he calls this the one place where the will's freedom enters appearance directly. Even then nobody decides their way into it. He describes it as something closer to grace.
Indian philosophy matters most at this point in his work. From it he took the "veil of Maya" as an image for the illusion that separates individuals, and the Upanishadic formula tat tvam asi, "that art thou", for the insight behind compassion. His source was remarkably indirect: the Oupnek'hat, a Latin translation published by Abraham Hyacinthe Anquetil-Duperron in 1801 and 1802 of a Persian translation of fifty Upanishads made in 1657 under the Mughal prince Dara Shikoh. According to Urs App's reconstruction from library records, Schopenhauer first borrowed it in Weimar in 1814, and in Parerga and Paralipomena (1851) he called it "der Trost meines Lebens", the consolation of his life. The IEP counts him as the first German philosopher to incorporate Eastern thought into his writings, and Douglas Berger's The Veil of Maya shows how much Eurocentric prejudice travelled with that pioneering interest.
Schopenhauer's split between doing what you want and wanting what you want still shapes how the free will debate is framed. His claim that character is fixed from birth draws most of the pushback, and modern psychology has added data he never had.
The diagnosis holds up best. Compatibilists, who think free will and determinism can coexist, have a classic definition going back to Thomas Hobbes that treats freedom as "nothing more than an agent's ability to do what she wishes". That is exactly the freedom Schopenhauer grants and finds insufficient. In 1971 Harry Frankfurt tried to secure freedom of the will one level up: you will freely, roughly, when the desire that moves you is the one you want to be moved by, a desire about your desires. His unwilling addict does what he wants when he takes the drug, so he acts freely in the ordinary sense, but he lacks freedom of the will because the craving is not the desire he endorses. That mirrors Schopenhauer's own distinction closely, though we found no source in which Frankfurt engages him directly. It also inherits a familiar problem, noted in the same SEP entry: people can feel alienated from their second-order desires too, so the question of where those come from moves up a floor.
The opposite camp accepts Schopenhauer's standard and rejects his verdict. Libertarians, in the philosophical sense, agree that freedom of doing is too little and, in the words of the SEP entry on incompatibilist theories, that "free will and determinism are mutually exclusive." They hold that we really could have willed otherwise, either because the causal chain leaves room at the moment of decision or because the agent herself counts as a cause. Their standing problem is the one Schopenhauer would press at once. If neither your character nor your motives settles the choice, it is hard to see why the result is yours rather than, as the same entry puts it, "a matter of luck." Randomness in physics doesn't obviously help, since an action that happens by chance belongs to you no more than one that happens by necessity.
The fixed character is where his own sympathetic translator hesitates. Janaway finds Schopenhauer's mostly anecdotal case for an unchanging character "less than fully convincing", and says that how a self outside space, time and causality is supposed to work "is not made clear." In a later paper he suggests a more consistent position would have been to deny freedom of will to the individual altogether. Nietzsche attacked from another side. In Beyond Good and Evil (§19) he accuses Schopenhauer of treating the will as the best-known thing in the world and adopting "a POPULAR PREJUDICE", when willing is "above all something COMPLICATED." In §21 he goes further: "The 'non-free will' is mythology", and in real life the question is only one of strong and weak wills.
Personality research puts pressure on the fixed character, though the details are kinder to Schopenhauer than you might expect. A meta-analysis of 92 longitudinal samples by Brent Roberts, Kate Walton and Wolfgang Viechtbauer (2006) found that people on average become more conscientious and emotionally stable over their lives, especially between 20 and 40. A 2017 review of 207 studies led by Roberts went further: during interventions, mostly psychotherapy, personality measures shifted noticeably within an average of 24 weeks, with emotional stability moving most. Roberts himself added a limit when the study came out: "You're not taking an introvert and making them into an extravert." The twist comes from a 2024 systematic review of deliberate personality change by Peter Haehner, Amanda Jo Wright and Wiebke Bleidorn. Simply having the goal to become a different person was only weakly related to actually changing. Structured interventions did produce modest change. That is close to velle non discitur with an asterisk: wanting to be someone else does little on its own, while a practice that keeps feeding you new motives and new self-knowledge can move the needle. Schopenhauer would file that under changing knowledge, not changing will. Whether that line can be drawn cleanly is precisely what the data leave open, and a defender who places the real character behind all measurable behaviour makes the theory very hard to test at all.
The water comparison from the six o'clock example has a similar weak spot. It is vivid, and it quietly assumes the point in dispute, that a person relates to motives the way water relates to wind and heat. Benjamin Libet's experiments on brain activity before conscious decisions often get pulled into this debate too, and Schopenhauer's claim that the intellect learns the will's decision afterwards can sound like an anticipation of them. The resemblance is ours, not a documented influence, and the SEP entry on free will notes that the standard reading of Libet's results has come in for severe criticism.
The worry that people who stop believing in free will start behaving badly rests on findings that have not replicated well. The effect Einstein described, that giving up free will makes you judge others less harshly, has held up better.
The worry has a famous source. In 2008 Kathleen Vohs and Jonathan Schooler had students read a passage arguing that free will is an illusion, and those students cheated more on a maths task than a control group. The study was widely cited and widely retold as a warning. A registered replication in the Many Labs 5 project, with 621 participants, found that the reading did not even change people's beliefs about free will, let alone their cheating. A separate set of four studies by Damien Crone and Neil Levy asked the question from the other end and found no sign that people who believe in free will are nicer.
The tolerance effect looks sturdier. In 2014 Azim Shariff, Joshua Greene and colleagues found that people with weaker free will beliefs favoured less retributive punishment of criminals, while their support for punishment that protects society was unchanged. Students who had just taken a neuroscience course recommended milder sentences than students who hadn't. In 2026 a survey of 8,917 people in 44 countries, led by Aaron Walton at the University of Melbourne, found that stronger belief in free will went with harsher reactions to everyday wrongdoing such as lying or acting selfishly. Correlations like these can't show that the belief causes the harshness, and the authors themselves are careful about comparisons between countries. Still, nearly a century after the essay, Einstein's claim that the conviction is "a perpetual breeder of tolerance" reads less like a nice sentiment and more like a testable prediction that has so far come out roughly right.
That may be the most usable thing in the whole package, whether or not you buy the metaphysics. In "What I Believe" Einstein added that the conviction stops us from taking ourselves or others too seriously and makes room for a sense of humour. If nobody picked their wants from a menu, a little less moral outrage at other people's wants seems reasonable. That includes the barber's, if the haircut goes wrong.
For more on where Schopenhauer's reading of India led him, see Schopenhauer and Hinduism: The Upanishads, Moksha and the Denial of the Will.
The idea sits in Schopenhauer's 1839 Prize Essay. The slogan version mostly travelled through Einstein.
Ask yourself "am I free?" and, according to Schopenhauer, all you get back is "I can do what I want". That answer covers acting and says nothing about wanting.
Motives are causes routed through the mind: once a reason reaches you, it moves you as surely as a cue moves a billiard ball.
Character is the hidden variable. Two people facing the same evening go different ways, and neither decided their own temperament.
Blame survives. It shifts from what you did to what you are, which is why a bad conscience cuts deeper than regret.
Spinoza's self-satisfied stone made the same point more than a century and a half earlier, and Schopenhauer said so himself.
Personality does change over a life and in therapy, but merely wanting to change does surprisingly little, which leaves Schopenhauer half right.
The scary studies about determinism and cheating mostly failed to replicate. The link between doubting free will and punishing less has held up, just as Einstein predicted.
You liked this blog post and don't want to miss any new articles? Receive a weekly update with the best philosophy memes on the internet for free and directly by email. On top of that, you will receive a 10% discount voucher for your first order.
Tax included and shipping calculated at checkout