What Can You Really Know? Carneades and the Art of Doubt

Imagine you’re looking at identical twins. They’re standing side by side, dressed the same, making the same face. You’re absolutely sure you’re looking at your friend Alex—until you learn that Alex is out of town, and this is actually his brother Sam, who you’ve never met before.

Everything about the experience felt exactly like seeing Alex. The impression was clear, vivid, convincing. And it was wrong.

Now imagine a philosopher tells you: “Every time you think you know something for certain, you could be making exactly this kind of mistake. Not just about faces, but about everything.” Would you believe them? Would you still be able to live your life?

These are the kinds of questions that drove a philosopher named Carneades (car-nee-uh-deez) to develop one of the most challenging and influential arguments about knowledge ever devised. He lived about 2,200 years ago, in Athens, and he never wrote a single book. Everything we know about him comes from other people writing about his arguments—which is fitting, because his whole philosophy was about what happens when you can’t be sure of anything.

The Argument Against Certainty

To understand Carneades, you first have to understand who he was arguing against. The main opponents were called the Stoics. (Not the same as today’s “stoic” meaning someone who doesn’t show emotion—these were philosophers with a detailed theory of knowledge.)

The Stoics believed that human beings could achieve something remarkable: perfect certainty. They thought that under the right conditions, your mind could grasp reality so clearly and accurately that there was simply no room for error. They called this a “cognitive impression”—a mental experience that, by its very nature, could only come from something real. It was stamped into your mind like a wax seal, bearing the exact shape of whatever caused it.

How would you recognize a cognitive impression? According to the Stoics, you just would. It would feel different from ordinary impressions—clearer, more vivid, somehow obviously true. If you only ever said “yes” to impressions like that, you’d never be wrong. You’d have knowledge, not mere opinion.

Carneades thought this was nonsense.

His basic argument was simple and devastating. He pointed to all the ways human beings can be fooled: dreams, hallucinations, optical illusions, mistaken identities, and even gods or demons sending false messages (ancient philosophers took this possibility seriously). The problem is that these false impressions can feel just as clear and convincing as true ones. A vivid dream feels real while you’re having it. A mirage looks exactly like water. A case of mistaken identity feels like recognition.

If false impressions can feel just as certain as true ones, then you can never tell the difference just by how an impression feels. And if you can’t tell the difference, then no impression is guaranteed to be true just because it seems obvious. The Stoics’ perfect “cognitive impression” doesn’t exist—or if it does, you can’t know when you’ve got one.

Carneades pushed this further with another argument. What if two things are exactly alike? The Stoics denied this could happen—they thought everything in the universe was unique. But Carneades pointed to examples like twins, eggs, or manufactured goods. If two things could be identical, then even if you had a perfectly accurate impression of one, you couldn’t tell it apart from the other. You’d always be vulnerable to confusing them.

The conclusion: if you want to be certain you’re never wrong, you’d have to stop believing anything at all. You’d have to suspend judgment on everything.

But How Do You Live Like That?

This is where things get interesting. The Stoics had a powerful objection: if you can’t be certain of anything, how can you act? How can you decide what to eat, where to walk, whom to trust? If you truly suspend judgment about everything, wouldn’t you just freeze?

Carneades had two different answers to this question—and here’s a strange fact: he seems to have defended both of them on different occasions. (Philosophy tip: when a philosopher argues for two opposite positions, they might not believe either one. They might be trying to show that both are better than the position they’re attacking.)

The first answer was: you don’t need to believe anything to act. Instead, you can just follow impressions that seem probable, without fully committing to them. It’s like the difference between jumping into a pool because you’re sure there’s water (belief) and jumping into a pool because it looks like there’s water and you want to swim (following a probable impression without certainty). You can live your life based on what’s likely true, even while admitting you might be wrong.

The second answer went further: you can form beliefs and opinions, as long as you hold them tentatively. You can say “I think this is true, but I might be wrong” and still act on it. This is what philosophers call “fallibilism”—the idea that you can have justified beliefs without certainty.

Both answers say the same thing to the Stoics: certainty isn’t necessary. We can manage with probability, with evidence, with checking our impressions against each other. If you see something that seems like water, you can test it—look at it from different angles, dip your toe in, see if other people confirm it. The more checks it passes, the more confident you can be. But you’ll never reach absolute certainty, and that’s okay.

What This Means for Right and Wrong

Carneades didn’t just argue about knowledge. He also applied his skeptical method to ethics—questions about how to live and what’s good.

His most famous performance was in Rome, where he gave two speeches on consecutive days. On the first day, he argued that justice was good and necessary. On the second day, he argued that justice was foolish and self-defeating. (The Roman politicians were apparently horrified that a philosopher would argue against justice. They sent him home.)

What was he doing? He was showing that you could make a strong case for either side of almost any ethical question. The arguments for treating everyone fairly are powerful—but so are the arguments for looking out for yourself first. The Stoics claimed their ethical system was obviously true and shared by all reasonable people. Carneades wanted to show that it wasn’t obvious at all.

He also developed a framework for thinking about what people are actually pursuing in life. He identified three basic things that all humans naturally want: pleasure, freedom from pain, and “natural advantages” (things like health, strength, and having your basic needs met). Then he showed how different ethical theories just pick different combinations of these. The Stoics said virtue was the only good, and virtue meant acting to obtain natural advantages whether or not you succeeded. Another view said the good was simply enjoying those advantages. A third said it was a combination of virtue and pleasure.

Carneades himself argued for different positions at different times—probably not because he kept changing his mind, but because he wanted to show that many different ethical systems could seem reasonable. If you can’t be certain about knowledge, you certainly can’t be certain about how to live.

Does the Future Already Exist?

One of Carneades’ most subtle arguments was about fate and free will. The Stoics believed everything was fated—every event caused by previous events in an unbreakable chain. The Epicureans (a rival school) thought this destroyed free will and moral responsibility, so they invented “the swerve”—random, unpredictable movements of atoms that broke the causal chain.

Both sides agreed on something that seems obvious: if every statement about the future is either true or false right now (which is a basic principle of logic called “bivalence”), then the future must already be fixed. A statement like “you will read to the end of this article” is either true or false right now, and whatever it is, that’s what will happen. This seemed to support the idea that everything is determined.

Carneades argued that this was a mistake. The fact that a statement about the future is either true or false doesn’t mean anything about fate. It just means we understand what the words mean. You can accept that every statement has a truth value without believing that the future is fixed by causes.

So what does cause our actions? Carneades suggested that our minds have a kind of autonomy. When you make a decision, the cause isn’t some external fate pushing you around. The cause is you—your mind, your character, your nature. This isn’t random (like the Epicurean swerve), and it isn’t predetermined (like the Stoic fate). It’s just… you deciding.

(Sound convincing? Carneades probably didn’t expect you to accept this fully. He was showing his opponents that there were alternatives to their views, not necessarily claiming to have the final answer.)

Did Carneades Believe Anything?

This is the hardest question about Carneades, and nobody really knows the answer. We know he argued both sides of many questions. We know he never wrote anything down. We know his students disagreed about what he actually thought.

Maybe he was a pure skeptic who doubted everything and never committed to any position himself. Maybe he had his own positive views but only revealed them through arguments against others. Maybe he was honestly searching for truth and used debate as a tool to get closer to it, without ever reaching it.

His student Clitomachus thought Carneades defended the position that wise people should never assent to anything—never say “yes, this is true”—but could still act by following probable impressions. Another student, Philo of Larissa, thought Carneades was a “probabilist” who accepted that some beliefs are justified, even if never certain.

Both sides have evidence. Both sides have arguments. And that’s kind of the point. In a world without certainty, even your interpretation of your teacher’s philosophy can’t be certain.

Why This Still Matters

You might be thinking: “Okay, but I know some things. I know that I’m sitting here reading this. I know my own name.”

Carneades would respond: do you? Could you be dreaming that you’re reading this? Could you be a brain in a vat (or whatever version of that thought experiment ancient philosophers had)? Could you be experiencing a hallucination that feels exactly like reading?

The point isn’t to make you doubt your own name. The point is to notice that certainty—the kind that can’t possibly be wrong—is much harder to achieve than we think. And maybe that’s okay. Maybe we don’t need certainty to live good, rational lives.

The most important legacy of Carneades is the idea that you can be a skeptic without being paralyzed. You can admit that you might be wrong, test your beliefs against each other, hold them tentatively, and still act decisively. This is a middle path between the dogmatist who claims certainty for everything and the pessimist who thinks nothing can ever be known.

It’s also an invitation to keep asking questions. If Carneades could challenge the most respected philosophers of his time, on questions they thought were settled, you can challenge your own assumptions too. The goal isn’t to end up believing nothing. It’s to end up believing things for better reasons.

Key Terms

Term	What it does in this debate
Cognitive impression	A Stoic idea about a mental experience so clear and accurate it could only come from something real—which Carneades argued didn’t exist
Suspension of judgment	The practice of refusing to say “yes, this is true” when you can’t be certain; Carneades suggested this could be a way to avoid error
Probable impression	Carneades’ alternative to cognitive impressions; an impression that seems convincing and holds up to checking, even without certainty
Fallibilism	The idea that you can have justified beliefs that might be wrong—you don’t need certainty to have good reasons
Bivalence	The logical principle that every statement is either true or false; Carneades argued this doesn’t mean the future is fixed
Swerve	The Epicurean idea that atoms sometimes move randomly, breaking the chain of cause and effect; Carneades thought this was unnecessary

Key People

Carneades (c. 214–129 BCE): A philosopher from North Africa who became head of Plato’s Academy in Athens. He never wrote anything, argued both sides of every question, and became famous for challenging the idea that certain knowledge is possible.
The Stoics: A school of philosophers who believed the universe was rational and orderly, that everything happens by fate, and that human beings could achieve perfect certainty through “cognitive impressions.” They were Carneades’ main targets.
Clitomachus: Carneades’ student and successor. He interpreted Carneades as a strict skeptic who thought wise people should never form beliefs, only follow probable impressions.
Philo of Larissa: A later head of the Academy who interpreted Carneades differently—as someone who accepted that we can have justified (though never certain) beliefs.

Things to Think About

Carneades argued that you can’t tell the difference between a true impression and a false one just by how they feel. But doesn’t experience teach us to recognize when we’re being fooled? Or does the very fact that we sometimes can’t tell prove his point?
The Stoics wanted to live without any opinions—only certain knowledge. Carneades offered two alternatives: live without opinions but follow probabilities, or have opinions while admitting you might be wrong. Which seems more appealing to you? Are there downsides to each?
If you could never be certain about anything, would you still make the same decisions about friends, school, and the future? Or does the possibility of being wrong actually matter to how you live?
Carneades argued for justice one day and against it the next. Was he being dishonest? Or is there value in understanding both sides of an argument, even if you disagree with one of them?

Where This Shows Up

Science: Scientists don’t claim to have absolute certainty. They propose theories, test them, and accept them tentatively—exactly the fallibilist approach Carneades described.
Courtrooms: Juries are asked to make decisions “beyond a reasonable doubt,” not with absolute certainty. This is Carneades’ idea of probability in action.
Everyday arguments: When you and a friend disagree about what really happened, you’re facing the same problem Carneades raised: how do you know your memory is accurate?
Artificial intelligence: Current debates about whether AI can “know” anything or just process information echo Carneades’ questions about what makes an impression trustworthy.
Modern skepticism: The idea that we should question our assumptions and hold beliefs lightly is central to science, democracy, and critical thinking—all descendants of the skeptical tradition Carneades helped shape.