Epistemic Humility: Nobody Owns the Truth
Epistemic humility is not relativism and not indecision. It is the recognition that our access to truth is limited, fallible, and incomplete, and that conviction should track evidence, not identity.
I keep returning to a sentence that sounds obvious until you sit with it:
Epistemic humility acknowledges that no single philosophy, individual, or group possesses a monopoly on truth.
The core idea is not that everyone is equally correct. It is that our access to truth is limited, fallible, and potentially incomplete.
That distinction is easy to miss, and missing it produces two opposite errors. One is dogmatism: treating your current model as reality itself. The other is collapse: treating the possibility of error as a reason to believe nothing. Both are failures of calibration. Epistemic humility is the attempt to get the calibration right.
1. What “epistemic” actually means
Epistemology is the study of knowledge. Not facts in isolation, but the conditions under which a mind is entitled to treat something as known:
- What can we know?
- How do we know it?
- What makes a belief justified?
- How certain should we be?
- How do we distinguish truth from what merely feels true?
So epistemic humility means having humility about your own knowledge. It is the recognition that:
I can be highly confident about something and still be wrong.
That is different from being indecisive. Indecision is the absence of a view. Humility is a disciplined relationship to the view you already hold.
Contemporary philosophers usually treat this as an intellectual virtue rather than a mood. Whitcomb, Battaly, Baehr, and Howard-Snyder define intellectual humility as owning one's cognitive limitations: noticing them, admitting them, and letting that recognition change how you inquire. Ian Church and Justin Barrett put it as a mean between two vices. Intellectual arrogance overrates the positive epistemic status of your beliefs. Intellectual diffidence underrates it. Humility is tracking that status accurately: not too high, not too low.
The psychological literature converges on a similar picture. Leary and colleagues describe intellectual humility as recognizing that a particular belief may be fallible, together with attention to the limits of the evidence and to your own limits in evaluating it. You can still believe you are right. You just refuse the additional claim that you couldn't be wrong.
That last clause is the whole game.
2. “No monopoly on truth” does not mean “there is no truth”
This is the most important distinction, and the one that gets lost the moment the sentence leaves a seminar and enters a group chat.
There are two very different positions:
Relativism:
Everyone has their own truth, so nobody can really be wrong.
Epistemic humility:
There may be an objective truth, but my understanding of it is incomplete and fallible.
The second position is much stronger. It preserves the existence of a target. Relativism dissolves the target and then congratulates itself for not missing.
Fallibilism, the closest technical cousin of epistemic humility, makes the same move. As the Internet Encyclopedia of Philosophy puts it: no belief can be justified in a fully conclusive way, and yet this need not imply skepticism. Most epistemologists are fallibilists and think knowledge is possible. The claim is not that we never know. The claim is that even our best knowledge remains, in principle, revisable.
Charles Peirce, who coined the term fallibilism, was explicit that this is neither skepticism nor relativism. He thought we may hold certain propositions as individually perfectly certain while still thinking it likely that some of them are false. Genuine doubt, like genuine belief, requires reasons. You do not get to doubt a claim merely because doubt is always logically available. (Gregory on Peirce)
A concrete case. Suppose three people disagree about why a company failed.
- Person A believes it was poor product-market fit.
- Person B believes it was bad distribution.
- Person C believes the founders ran out of capital.
Epistemic humility says:
One of these explanations might be substantially more correct. But before dismissing the others, examine the evidence and ask what each perspective captures that yours might miss.
It does not require you to conclude that all three explanations are equally valid. Product-market fit, distribution, and capital are not three “personal truths.” They are hypotheses about a causal structure. Some of them will explain more of the wreckage than others. Some will turn out to be downstream effects of a deeper cause. Humility is the refusal to freeze the first story you found emotionally satisfying.
This is also why “nobody owns the truth” is a dangerous slogan if you hear it lazily. It can be read as: every philosophy or group possesses some truth, so all frameworks deserve equal standing. That does not follow. A monopoly can be false without the market being efficient. A theory can be incomplete without its rivals being insightful. The humble move is not equal weighting. It is remaining open to the possibility that your current weighting is wrong.
3. It applies to individuals and to philosophies
The interesting part of the original sentence is that it does not limit humility to people.
You can become intellectually dogmatic at several levels:
Individual dogmatism
I understand this better than everyone else.
Group dogmatism
Our community has figured this out; outsiders don't understand.
Philosophical dogmatism
This philosophical framework explains reality; competing frameworks are fundamentally misguided.
Ideological dogmatism
If you disagree with our premises, you must be ignorant or corrupt.
Epistemic humility challenges all four. It says that a framework can be extraordinarily useful without being complete.
This is harder than it sounds, because frameworks do not just describe the world. They become part of identity. Once a model has paid rent, once it has helped you ship a product, win an argument, join a community, or feel coherent, abandoning it feels like self-betrayal rather than updating. Group and ideological dogmatism are especially sticky for this reason. The cost of revision is no longer intellectual. It is social.
Peirce saw this clearly. He listed ordinary error as a source of fallibility, but he also listed “subtle, often unconscious psychological and emotional motives including anxiety, self-interest, pride, group loyalty, and ideology.” Those are not bugs in other people. They are the default operating system. Humility is the attempt to build a second system that can notice when the first one is defending a map instead of reading the terrain.
The same pattern shows up in technical work. A founder who has a theory of the market, a researcher who has a theory of intelligence, a critic who has a theory of culture: each of these can be locally brilliant and globally incomplete. The failure mode is not having a theory. The failure mode is treating the theory as a closed system that has already exhausted what there is to see.
4. Why this matters: maps are not territories
Human beings have a tendency to confuse:
“This is the model through which I understand reality”
with
“This is reality.”
Those are not the same thing.
Alfred Korzybski put this in a sentence that has been quoted so often it now sounds like a poster: the map is not the territory. The original claim, from a 1931 paper later collected in Science and Sanity, is more precise than the slogan. A map is not the territory it represents, but if it is correct, it has a similar structure to the territory, which is why it is useful. Usefulness is structural resemblance, not identity.
That is the entire epistemology of models in one line. A good model compresses. Compression discards. What it discards may not matter inside the domain you care about, and may matter enormously the moment you leave that domain.
Science itself is the cleanest demonstration.
Newtonian mechanics was extraordinarily successful. Then relativity and quantum mechanics showed that classical mechanics does not describe everything. That does not make Newton stupid, and it does not make classical mechanics “false” in the everyday sense. It means the model had a domain of validity. Inside that domain it remains one of the most powerful maps ever drawn. Outside that domain, the neglected structure starts to leak.
Karl Popper turned this into a method. In Conjectures and Refutations, scientific theories are not digests of observations. They are inventions: conjectures boldly put forward, retained only tentatively if they survive our best attempts to break them. Even the theories we can no longer bring ourselves to doubt remain, in principle, hypothetical. Before a theory has been refuted, we cannot know in what way it may later have to be modified.
The same pattern appears everywhere:
A map can be extremely accurate without being the territory.
Epistemic humility is essentially remembering that your beliefs are maps. Product-market fit is a map. A political ideology is a map. A personality framework is a map. A training run’s loss curve is a map. Your autobiography is a map. Some of these maps are astonishing. None of them are the ground.
The practical implication is almost rude in its simplicity: when a model works, do not immediately promote it from local success to total theory of everything. Ask what it is silent about. Ask where it would fail. Ask what a competing map would have to get right in order to look as good as yours does from the inside.
5. Humility has a dangerous extreme
There is a trap on the opposite side:
Since I could be wrong, I shouldn't believe anything strongly.
That is not epistemic humility. That is epistemic paralysis.
You can simultaneously say:
I could be wrong.
and
The evidence currently gives me very strong reason to believe this.
You do not need absolute certainty that gravity exists to confidently predict that dropping your phone will cause it to fall. You do not need metaphysical finality to hire, ship, invest, or refuse a bad deal. Action does not require infallibility. It requires that the expected cost of acting on the current map be lower than the expected cost of waiting for a better one.
So epistemic humility is not:
certainty → zero
It is:
certainty → calibrated to evidence.
This is exactly Church and Barrett’s Aristotelian framing: a virtuous mean between arrogance and diffidence. If a belief enjoys only marginal support, humility requires you to hold it lightly. If a belief enjoys enormous support (simple arithmetic, well-replicated physical regularities, a pattern you have seen destroy ten companies in a row), humility requires you to value that support, not to perform uncertainty as a personality trait.
Peirce, again, is useful here. He rejected Descartes’ method of universal and arbitrary doubt. You cannot, and should not, doubt every belief just because a clever skeptic can construct a scenario in which it fails. Doubt has to be earned. Otherwise you do not get inquiry. You get a theatrical refusal to stand anywhere.
A useful mindset, popularized in forecasting by Paul Saffo since the mid-1980s, is:
Strong opinions, weakly held.
Saffo’s version is more operational than the LinkedIn version. You force yourself to a tentative conclusion with the information you have. Then you systematically try to tear it apart. The strong opinion is there to give inquiry a surface. The weak hold is there so the surface can be replaced. Without the first half, you never leave the waiting room. Without the second, you decorate the waiting room and call it a house.
Believe things strongly when the evidence warrants it. Remain willing to update when better evidence appears. That is not a contradiction. It is the only adult relationship to belief.
6. The deeper implication: disagreement becomes information
If you accept epistemic humility, then disagreement becomes potentially informative.
Normally, disagreement triggers:
How could they possibly believe that?
Epistemic humility allows:
What are they seeing that I'm not seeing?
That does not mean accepting their conclusion. It means investigating the information asymmetry.
Maybe:
- they have evidence you don't have,
- they are using a different model,
- they have different experiences,
- they noticed an assumption you overlooked,
- or they are simply wrong.
You do not know which one until you investigate.
John Stuart Mill made this the center of On Liberty. “He who knows only his own side of the case, knows little of that.” It is not enough to hear the opposing view as summarized by people who already agree with you. You have to hear it from people who actually believe it, in its most plausible form, until you feel the force of the difficulty your own view has to survive. Otherwise even a true conclusion is, for you, a dead dogma rather than a living truth.
This is psychologically expensive. Leary’s 2017 studies found that people higher in intellectual humility were more attuned to the actual strength of arguments, less likely to judge others solely by whether they shared a religious opinion, and less inclined to treat a changed mind as mere flip-flopping. Later work associates the trait with curiosity, more balanced evidence evaluation, weaker myside bias, and less derogation of people who disagree. That is a description of what it feels like when disagreement stops being an identity threat.
The founder version of this is familiar. A customer says the product is confusing. An engineer says the architecture cannot scale. An investor says the market is a feature, not a company. Each of those statements can be wrong. Each can also be the thing you needed to hear and were structurally unable to generate from inside your own map. Humility is not agreeing with them. It is refusing to treat the disagreement as noise by default.
This makes disagreement less of a threat to your identity and more of an opportunity to test your model. The test can still fail the other person. Models survive some collisions. They should not be protected from all of them.
7. The paradox: humility is not “I know nothing”
The most epistemically humble position is not:
I know nothing.
It is closer to:
I know some things reasonably well, many things imperfectly, and there are probably important things I don't even know that I don't know.
That last category is crucial.
There is:
Known knowns: things you know.
Known unknowns: things you know you don't know.
Unknown unknowns: things you aren't even aware you're missing.
The taxonomy became famous through a 2002 remark by Donald Rumsfeld, but philosophers of science have since treated it as a genuine epistemic distinction. Torsten Wilholt reframes known unknowns as transparent ignorance and unknown unknowns as opaque ignorance. Scientific progress, on this view, is often the conversion of opaque ignorance into transparent ignorance: making a limit explicit so that it can start to guide inquiry. Fields that lose sight of their genuine unknowns can keep accumulating data while drifting into epistemic aimlessness.
Socrates is the ancestral version of this. He did not, contrary to the bumper sticker, claim to know that he knew nothing. In the Apology, the sharper formulation is that he does not think he knows what he does not know. The politicians, poets, and craftsmen he questioned often had local competence. Their vice was extending that competence into a general claim to wisdom. Socrates’ small advantage, “because of this small detail,” as he puts it, was refusing that extension. (Macé on Socratic ignorance)
Joseph Ransdell argues that this is not a Zen koan. It is the refusal of a god’s-eye view. Human understanding is not the kind of understanding a god would have. Socratic wisdom is the recognition of that impossibility, not a performance of emptiness.
Epistemic humility is partly an awareness of the third category: the unknown unknowns. You can inventory your known unknowns. You cannot inventory the rest. The honest stance is therefore not “I have mapped my ignorance,” but “my map of ignorance is itself a map, and therefore incomplete.”
8. False humility is not humility
There is a counterfeit.
Allan Hazlett describes false intellectual humility as affected or pretended humility concealing intellectual arrogance. It often looks like insincere expressions of ignorance or uncertainty: the conspiracy theorist who insists they “just want to know what’s going on,” the operator who says “I’m no expert, but…” and then refuses every correction, the person who hedges every sentence while remaining completely unmoved by evidence.
This matters because humility has become a social signal. In some rooms, saying “I could be wrong” is how you get to keep being wrong in public. The phrase is doing interpersonal work, not epistemic work.
Real humility has a behavioral signature. It updates. It asks better questions. It can state a view clearly enough to be proven false. It can also, when the evidence is strong, stop hedging and take the hit of being definite. False humility does the opposite. It uses uncertainty as armor and uses “just asking questions” as a way to avoid the responsibilities of a claim.
The test is ugly and useful:
After contact with better evidence, did anything in your map actually move?
If the answer is never, you are not humble. You are either frozen or disguised.
9. What the research actually found
Philosophy gives the shape of the virtue. Psychology has started to measure its correlates, which is a different and more modest kind of knowledge, but still worth having.
People higher in intellectual humility tend to be more open, more curious, more tolerant of ambiguity, and less dogmatic. They distinguish strong arguments from weak ones more cleanly. They have a more accurate sense of when they might be wrong: less confidence on incorrect answers, without a corresponding collapse of confidence on correct ones. They are less likely to believe poorly evidenced conspiracy theories, and in at least one study they were more willing to take a COVID vaccine when the evidence supported it. (Leary, 2022)
That last point is important, because it breaks a folk theory: that humility means being equally open to every claim. It does not. If anything, calibrated humility should make you harder to move with weak evidence and easier to move with strong evidence. That is the opposite of both the crank and the weather vane.
There is no single agreed definition. Some accounts emphasize owning limitations. Some emphasize low concern for intellectual status. Some emphasize accurate tracking of epistemic status. Porter and colleagues, in a systematic review that found 18 definitions and 20 measures, treat this as an integrative problem rather than a settled one. That unsettledness is itself an instance of the thing being studied. We know something about the trait. We do not own a final theory of it.
The research also suggests why the virtue is rare. Overconfidence is the default. People overestimate their performance, overplace themselves relative to others, and hold unwarranted certainty in the accuracy of their beliefs. Humility has to fight a bias, not merely express a temperament. It is also domain-specific: you can be humble about astronomy and dogmatic about your industry. Trait-level humility is a tendency, not a halo.
10. A more precise sentence
The original sentence is good, but “no single philosophy, individual, or group possesses a monopoly on truth” can be heard as saying that every philosophy or group possesses some truth.
I would make it slightly more precise:
Epistemic humility is the recognition that our understanding of truth is necessarily fallible and incomplete. No individual, philosophy, or group should assume that it has exhausted the truth simply because its current model explains what it can see.
Or, more concise:
Epistemic humility is holding your beliefs with conviction proportional to the evidence, while remaining open to the possibility that your understanding is incomplete or wrong.
I think that gets closer to the real philosophical power of the idea.
Because ultimately, epistemic humility is not about believing less.
It is about becoming better at distinguishing what you know, what you believe, why you believe it, and how certain you actually have reason to be.
Those are four different questions. Most of our worst intellectual habits come from collapsing them into one. “I feel sure” is treated as “I know.” “I know” is treated as “I could not be wrong.” “I could not be wrong” is treated as “those who disagree are defective.” Each collapse is a small act of annexation. Enough of them and you start to believe you own the territory.
You don’t. Nobody does. The territory is larger than any map, including this one.
That is not a reason to throw the maps away. It is a reason to keep drawing them in pencil, press hard when the evidence is good, and stay willing to erase.