🇺🇸You Can Only Know What You Can Prove🇺🇸

How to Prove What You Know?

by CEJames (researcher/author) & Akira Ichinose (editor/research assistant)

Held stone, open palm —

proof lives where the fingers grip;

doubt cannot reach there.

You watched the sun rise —

no argument needed, yet

words make others see.

CAVEAT: Keikoku (警告)

The content presented in this work is produced solely for educational, research, and creative purposes and does not constitute legal advice, a certified self-defense methodology, or the official position of any organization, institution, or government body.

All views and opinions expressed herein are those of the authors alone. Laws and statutes governing the use of force, personal protection, and related conduct vary by jurisdiction; readers and practitioners are strongly advised to consult a qualified attorney and to seek instruction from a certified self-defense professional before making any decisions regarding personal protection or the use of force.

Where this work contains fictional narrative, all names, characters, incidents, and dialogue are products of the authors' imagination and are not to be construed as factual, historical, or representative of any real person, living or dead, or any actual event. Any resemblance to real persons or events is entirely coincidental.

All content is protected under applicable copyright law. Unauthorized reproduction, distribution, or transmission of this material, in whole or in part, by any means — electronic, mechanical, photographic, or otherwise — is strictly prohibited without the express written permission of the authors.

Introduction: The Question Behind the Question

Let's be honest with each other from the start: this title is a little bit of a trap. 'You can only know what you can prove' sounds like a hard, clean rule — the kind of thing a tough-minded scientist or a skeptical philosopher might tape above their desk. But once you sit with it for more than a minute, it starts to squirm.

 

Because here's the thing. You know what a headache feels like. You know whether you slept well last night. You know you are reading these words right now. Can you prove any of that to me? Not really — not in any airtight, courtroom-ready sense. And yet you would be hard-pressed to say you don't know those things.

 

So the question immediately splits into two smaller, more useful ones: What do we mean by 'knowing'? And what do we mean by 'proving'? This essay is a walk through both — not as an academic exercise, but as a practical inquiry into how we build confidence in what we believe, how we communicate that confidence to others, and what happens when proof runs out.

 

Section I: What Does 'Knowing' Actually Mean?

The classical definition of knowledge, traced to Plato's dialogue Meno and refined in the Theaetetus, is 'justified true belief.' To know something, three conditions must hold: it must be true, you must believe it, and you must have a good reason — justification — for believing it (Plato, c. 385–360 BCE/1997).

 

That sounds tidy until Edmund Gettier blew it apart in 1963 with a three-page paper showing you can have justified true belief and still not have knowledge. Imagine you glance at a stopped clock at the one moment in the day when it happens to show the correct time. You have a true belief — the time is right — and you are justified in your belief — clocks normally tell time — but you don't actually know what time it is (Gettier, 1963).

 

Philosophy has never fully recovered from the Gettier problem. Dozens of alternative definitions have been proposed — reliabilist accounts, virtue epistemology, contextualist theories — and none has achieved consensus. This isn't a failure of philosophy; it is a sign that knowledge is genuinely complex (Steup & Neta, 2020).

 

For our purposes, let's work with a more practical framing: 

• knowing is a state of justified confidence in a proposition — confidence high enough to act on. 
• It lives on a spectrum. 
• At one end is the dead certainty of pure mathematics (2 + 2 = 4). 
• At the other end is the qualified confidence of everyday judgment (I'm fairly sure I left my keys on the counter). 

Most of what we know lives somewhere in between.

A Parable: The Farmer and the Rain

An old farmer in the valley told anyone who would listen that it was going to rain hard before the week was out. He hadn't checked a weather app. He'd read no forecast. But he'd watched those mountains for forty years. He felt the particular humidity in the morning air. He noticed the swallows flying low. He saw how his knees ached in a way they only did before a heavy storm.

 

It rained — hard — on Thursday.

 

Did the farmer know it was going to rain? In one sense, yes: he had genuine, reliable knowledge, rooted in decades of pattern recognition. In another sense, he couldn't prove it to the agricultural extension officer who came by Monday wanting data and weather maps. The farmer's knowledge was real. It was just non-transferable by conventional means.

 

This is what philosophers call tacit knowledge — the knowing-how that Ryle (1949) distinguished from knowing-that. The farmer's knowledge was embodied, experiential, and resistant to formal proof. Yet it reliably predicted the world. This distinction will matter throughout everything that follows.

Section II: What Does 'Proving' Actually Mean?

Proof is not a single thing. It shifts shape depending on the domain you are standing in.

 

In mathematics, proof means deductive certainty: a sequence of logical steps, each following necessarily from the last, that leads from axioms to a conclusion that cannot be denied without contradicting the axioms themselves. When Euclid proved that the square root of 2 is irrational, that was a proof. It doesn't need experimental confirmation. It doesn't expire.

 

In science, proof is something humbler and more honest: it means evidence sufficient to reject the null hypothesis with a defined level of confidence, or the repeated experimental confirmation of a predicted result. Scientific proof is probabilistic and revisable. Yesterday's proof is tomorrow's corrected understanding. That is not a flaw in science; it is how science works, and it is part of why science works so well (Popper, 1959).

 

In law, proof means convincing a judge or jury beyond a reasonable doubt in criminal matters, or on the preponderance of evidence in civil ones. Legal proof is not about truth in the cosmic sense; it is about what a reasonable person would conclude from admissible evidence, evaluated under procedural rules.

In everyday life, proof often just means: Can you show me enough that I would be foolish not to believe you? It is a social standard — context-dependent and sometimes quite forgiving.

 

The crucial insight here is that these standards are not ranked. 

• Mathematical proof doesn't make legal proof inferior; 
• they answer different questions. 
• Demanding mathematical certainty of a scientific claim is a category error. 
• Demanding scientific replication of a personal moral conviction is also a category error. 

Knowing which kind of proof is appropriate for which kind of claim is itself a form of wisdom.

A Response: The Courtroom

Think of it like a courtroom. The prosecutor doesn't have to prove, in the mathematical sense, that the defendant committed the crime. That standard would be impossible for almost any crime ever committed — you can always construct some chain of theoretical doubt. What the prosecutor must do is present evidence so convincing that no reasonable person would doubt it.

 

That is a social, practical standard. And notice: most of what we 'know' in daily life operates the same way. 

When you believe your colleague is acting in bad faith, 

when you judge that the stranger approaching you at night means harm, 

when you conclude that your doctor is competent — 

you are operating as a jury of one, evaluating evidence against a standard that is reasonable for the context. You probably can't prove any of these conclusions by mathematical or scientific standards. You don't need to. The standard for proof should match the weight of what's at stake.

Section III: The Bridge Between Knowing and Proving

Here is where the title's challenge gets most interesting. The phrase 'you can only know what you can prove' is really an epistemological position known as verificationism — the view, associated with the logical positivists of the Vienna Circle, that a statement only has cognitive meaning if it can in principle be verified (Ayer, 1952). Unverifiable claims, on this view, are not false — they are literally meaningless.

 

Verificationism had a heroic run in the early twentieth century. It gave philosophy a bracing clarity, and it did useful work clearing out speculative metaphysics that had accumulated over centuries. But it ran into serious trouble. For one thing, the verification principle itself cannot be verified by its own standard — making it, by its own definition, meaningless. This was not a minor objection.

 

Beyond that, verificationism couldn't make sense of large swaths of perfectly intelligible discourse: historical claims about unobservable past events, theoretical entities in physics (the electron, for instance, has never been directly observed), moral claims, aesthetic judgments, mathematical truths. The logical positivists eventually modified and softened the principle so many times that it collapsed under its own revisions (Hempel, 1950).

 

So: no, you cannot only know what you can prove — at least not if 'prove' is given a strict empirical or logical meaning. But the spirit of the demand is still worth honoring. The pressure to justify our beliefs, to ask ourselves what grounds we have for what we claim to know, is one of the most valuable habits of mind a person can cultivate. It doesn't always result in proof. But it makes us more careful, more honest, and less prone to the catastrophic confidence that attaches to unexamined conviction.

 

A Parable: The Scholar and the Student

A young student came to an old scholar and said, 'Teacher, I believe that honesty is the highest virtue. Prove it to me.'

 

The scholar was quiet a long moment. Then he said, 'Tell me what happened the last time someone lied to you.'

 

The student told him. And the time before that. By the time the student had described the third incident, he was angry.

 

'You see,' said the scholar, 'you already know the weight of dishonesty in your body. You feel it. You can give me a hundred reasons why honesty matters — philosophical, practical, relational. Not one of them is a proof in the geometric sense. But put together, they make the case. That is how we prove most of what matters. Not by demonstration, but by convergence.'

 

Convergence. Multiple lines of evidence, argument, and experience pointing toward the same conclusion. This is how philosophers speak of coherentism — the view that beliefs are justified not by individual foundational proofs but by how well they fit and support each other within a web of belief (BonJour, 1985).

 

Section IV: How to Prove What You Know

If proof is domain-dependent and most of our knowledge is probabilistic rather than certain, then 'proving what you know' is less about achieving certainty and more about constructing a persuasive, coherent, honest account of your grounds. Here are the key strategies.

 

First: triangulate. A single line of evidence is fragile. Three independent lines converging on the same conclusion are formidable. The detective who identifies the murderer doesn't usually have one piece of evidence; they have a pattern. If your personal experience, the testimony of credible witnesses, experimental data, and the logic of the situation all point the same direction, that is the strongest kind of proof available in practical epistemology.

 

Second: make your reasoning visible. Proving what you know requires not just the conclusion but the path. Show how you got there. Identify your assumptions. Acknowledge where the chain is weakest. A proof that conceals its assumptions is not really a proof — it is a performance. Intellectual honesty requires exposing the joints in your argument so others can evaluate them.

 

Third: engage the best opposing view. Confirmation bias is the single greatest threat to genuine knowledge. If you only seek evidence that confirms what you already believe, you will succeed in finding it — and you will not know. The test of real knowledge is whether it can survive a serious encounter with the best objection against it.

 

Fourth: understand the limits of your evidence. Knowing the kind of claim you are making, and the kind of evidence appropriate to it, keeps you from overclaiming. 'The available studies strongly suggest X' is honest. 'X is proven' often isn't.

 

Fifth: accept that some things are known without being provable. As Wittgenstein wrote in On Certainty, at some point one has to act from certainty — not from proof (Wittgenstein, 1969). Some things function as bedrock, not because they've been proven but because proof has to start somewhere. We cannot prove that we have hands. But if that is in doubt, everything else is in doubt too, and doubt loses its meaning.

 

A Response: The Mariner's Compass

Think of proving what you know as navigating by compass in open water. The compass doesn't tell you exactly where you are. It tells you the direction you're heading and lets you track your position relative to a known starting point. You don't need GPS precision — you need enough information to make the next decision well.

That's how proof works in practice. You gather enough evidence, triangulate enough sources, apply enough reasoning, and arrive at a position where doubt would require more effort and more implausibility than belief. You can't prove you will make port. But you have enough to steer by. That is what it means to prove what you know.

Section V: Counter-Argument — A Case for the Other Side

Let me pause here and do something important: take the opposing position seriously. Because there is a serious opposing position, and intellectual humility demands that I give it full weight rather than a polite summary before dismissal.

 

An empiricist in the tradition of David Hume might say: you have given a lot of ground to tacit knowledge, convergence, and coherentism, and in doing so you have made proof so flexible that almost anything can count. The farmer who 'knew' it would rain might have been right this time. But without systematic testing, without controlled conditions, without a method that distinguishes real knowledge from lucky guessing, you are validating intuition dressed up in philosophical language.

 

And there is something to this. History is littered with 'knowledge' that turned out to be confident ignorance — about race, about disease, about the structure of the cosmos, about the role of women in society. People who held those views were not obviously less intelligent or less sincere than those who later corrected them. They had what felt like convergent evidence, coherent worldviews, and cultural authority backing them up. And they were wrong. Catastrophically, sometimes murderously, wrong.

 

So the empiricist's demand — show me data, show me method, show me a way to test this that doesn't just confirm what you already believe — is not pedantry. It is a hard-won lesson from the history of human error. The philosopher of science Thomas Kuhn (1962) showed that entire scientific communities can be gripped by paradigms that systematically prevent them from seeing contrary evidence. If that can happen to disciplined scientists, it can happen to all of us.

 

Furthermore, the reliabilist philosopher Alvin Goldman (1979) argues that what makes a belief count as knowledge is whether the cognitive process that produced it is reliable — not whether it feels solid from the inside. Intuition, personal experience, and coherence can all feel very reliable while being systematically distorted. This is precisely why double-blind experimental design exists: because human beings are not reliably aware of their own biases.

 

I take this counter-argument seriously. I hold my own epistemological preferences with some humility, because the history of overconfident knowing is also the history of great harm. The appropriate response is not to abandon the practical standards of knowing and proving outlined above, but to hold them provisionally — to keep the door genuinely open to revision, and to take real objections more seriously than is comfortable. Epistemic humility is not uncertainty paralysis. It is the honest acknowledgment that our instruments of knowing are imperfect, and that the best thing we can do is use them carefully, transparently, and in community with other careful thinkers.

 

Conclusion: What Remains

After all of this, what can we say?

 

You can know things you cannot prove — in the full, technical sense. Personal experience, tacit knowledge, and the bedrock certainties that make all inquiry possible are real forms of knowing that resist formal proof.

 

You can prove things in degrees — mathematical proof is the strongest, scientific evidence is strong, legal proof is context-specific, and practical proof is the flexible currency of daily life.

 

The demand to prove what you know is a valuable discipline even when it cannot be fully satisfied. It checks overconfidence, surfaces hidden assumptions, and builds the kind of honest, examined knowledge that is most useful and most trustworthy.

 

Intellectual humility is not a sign of weakness. It is the appropriate epistemic posture for any being whose knowledge is finite, whose perceptions are limited, and whose history includes more error than any of us would like to admit.

 

And finally: in a world where the loudest voices often claim the most certainty, there is something quietly radical about saying, 'Here is what I know. Here is why I think I know it. Here is where I might be wrong.' That is not ignorance. That is the beginning of real knowledge.

 

Bibliography

Ayer, A. J. (1952). Language, truth and logic (2nd ed.). Dover Publications.

BonJour, L. (1985). The structure of empirical knowledge. Harvard University Press.

Gettier, E. L. (1963). Is justified true belief knowledge? Analysis, 23(6), 121–123. https://doi.org/10.1093/analys/23.6.121

Goldman, A. I. (1979). What is justified belief? In G. S. Pappas (Ed.), Justification and knowledge (pp. 1–23). D. Reidel.

Hempel, C. G. (1950). Problems and changes in the empiricist criterion of meaning. Revue Internationale de Philosophie, 4(11), 41–63.

Kuhn, T. S. (1962). The structure of scientific revolutions. University of Chicago Press.

Plato. (1997). Meno and Theaetetus. In J. M. Cooper (Ed.), Plato: Complete works (pp. 870–1011). Hackett Publishing. (Original works composed c. 385–360 BCE)

Popper, K. R. (1959). The logic of scientific discovery. Hutchinson.

Ryle, G. (1949). The concept of mind. Hutchinson.

Steup, M., & Neta, R. (2020). Epistemology. In E. N. Zalta (Ed.), The Stanford Encyclopedia of Philosophy (Fall 2020 ed.). Stanford University. https://plato.stanford.edu/archives/fall2020/entries/epistemology/

Wittgenstein, L. (1969). On certainty (G. E. M. Anscombe & G. H. von Wright, Eds.; G. E. M. Anscombe & D. Paul, Trans.). Blackwell.

Page of