What to do when you don't know what your values are
"I don't actually know what I value" is one of the most common things people say right before they try to write a personal principle, and it is almost never true in the way it sounds. It is not that the values are missing. It is that "what do you value" is a bad question to ask directly, because the honest answer requires noticing your own patterns, and patterns are invisible from the inside when you go looking for them on purpose. You do not need a clearer soul. You need a different angle of approach.
Why the direct question fails
Ask yourself "what do I value" and the mind does not search your history — it searches for socially acceptable nouns. Family, growth, honesty, freedom, health. These are not wrong, exactly. They are just too abstract to have come from you specifically; almost anyone on earth would nod at the same list. A value you arrived at by brainstorming answers to a direct question is indistinguishable from a value you arrived at by remembering what values are supposed to sound like. That is why the exercise leaves you with a tidy list and the nagging sense that you still don't know anything you didn't know an hour ago.
The fix is not to think harder about the same question. It is to stop asking it directly and instead look at four places where a value shows up as a side effect of something else you were already doing.
Where to look instead
1. What you react to, not what you admire
Admiration is aspirational and therefore unreliable — you can admire patience in others for years without it telling you anything about your own rules. Irritation is not aspirational; it is involuntary, and that makes it a better instrument. What kind of behaviour in other people makes you disproportionately annoyed, for a reason that seems small when you say it out loud? People who cancel plans at the last minute. People who take credit quietly. People who ask a question they could have answered themselves. That flash of irritation is usually the underside of a rule you hold for yourself and notice being broken by someone else. You rarely get angry about the absence of a value you don't actually have.
2. What you complain about more than once
A single complaint is noise. The same complaint, made about different people or situations over months, is a value with the serial numbers filed off. If you keep telling the same story about a boss who didn't listen, a friend who only calls when they need something, a team that ships sloppy work — the throughline is not really about any one of those people. It is what you keep measuring everyone against, whether or not you have ever written the standard down.
3. Where your money and time actually went
Stated preference is what you say you care about. Revealed preference is what a bank statement and a calendar say, and the two disagree more often than anyone likes to admit. Look at the last twelve months, not for a category you're proud of, but for the thing you would defend to a skeptical friend without embarrassment — the course, the trip, the ordinary Tuesday habit, the person you always make time for even when it costs you something. What you would defend is closer to a real value than what you would list first.
4. What you would actually quit over
Most people can name a long list of things they claim to care about and a very short list of things they have ever walked away from. The short list is more honest. Think of the jobs, relationships, or arrangements you have left, or seriously considered leaving — what was the actual line that got crossed? That line is a floor, and a floor is one of the more useful kinds of value to have written down, because it does not require inspiration to consult; it only requires noticing when you are near it again.
Write it as a hypothesis, not a conclusion
Once one of these methods surfaces something — a repeated irritation, a recurring complaint, a defensible expense, a line you have already crossed once and would not cross again — write it down as a guess, not a verdict. "I think I actually care more about being trusted with hard truths than about being liked" is allowed to be wrong. Treat the first version as a draft you are going to test against your own life for a couple of weeks, not a statement you now have to defend to yourself. The people who get stuck are usually the ones who felt they had to be certain before they were allowed to write anything down.
Start with one, not a list
If the honest answer is still "I don't know," do not try to produce eight values before you have found one. Pick the single strongest signal from the four places above — the irritation that comes up the most, or the line you would not cross twice — and write one plain sentence about what it asks of you. Live with it for two weeks. A single tested sentence tells you more than a brainstormed list of ten, because it generates actual evidence: moments where it held, moments where it didn't, situations that made it easy and ones that made it expensive. That evidence is the material the next value gets built from.
Sometimes "I don't know" is actually a conflict
Occasionally the blankness is not an absence of values but two of them arguing quietly enough that neither wins clearly — wanting security and wanting freedom, wanting closeness and wanting independence. If every attempt to write a value produces an immediate, specific counterexample from your own life, you may not be missing a value. You may be sitting on an unresolved trade-off, and naming both sides honestly is more useful right now than forcing a single tidy answer. "I have not decided how much stability I am willing to trade for how much freedom" is not a failure to find your values — it is a more accurate one than most people manage to write.
This is close to where CreedOS starts for people in exactly this position: instead of a blank text field, its templates span six areas — efficiency, social, emotion, money, health, and cognition — so the first move can be reacting to an existing option rather than inventing one from nothing, and its AI creed generation takes a rough, half-formed thought and turns it into a title, a category, and behavior standards you keep editing from there. It is free, with no in-app purchases.