Can we know anything with certainty?
I've always been the kind of person who needs to know why.
Why does this decision feel right? Why do I believe the things I believe? It's a trait that's served me well, and occasionally driven the people around me insane.
This semester, I’ve been studying epistemology for a philosophy course, and I ran into a question that stopped me completely:
Can we ever know anything with certainty?
Well, my first instinct was - obviously yes. I know I exist. I know the sun will rise tomorrow. I know the people I love. These feel like givens, the kind of claims it would be embarrassing to doubt out loud.
But throughout the past months, the longer I sat with this question, the less solid that instinct felt. And I want to walk you through why (and I promise you it’ll be a fun one, or at least that was me). Because I think this question, if you actually take it seriously, changes the way you relate to everything you think you know.
What does it even mean to know something?
When you say "I know X," you mean something stronger than "I think X." You're claiming your belief actually maps onto reality, not just that it feels true, but that it is true.
But here’s the first problem, our access to reality is always filtered. Through our senses, our memory, our language, our assumptions. We hold maps, never the territory itself. And that gap between the map and the territory? It never fully closes.
The famous French philosopher René Descartes wanted to close it completely. He was writing in the early 1600s, a world of contradictory claims, where the Church, Aristotle, and Galileo's new science were all saying different things about how reality worked. His question was the same one that unsettled me: “how do you sort good knowledge from bad, when the very beliefs you'd use to judge might themselves be wrong?”
His answer: strip everything down. Doubt everything you possibly can, and see what survives.
And now, let’s have some fun, sit back, and allow me to walk you through how he’s done it.
Mental Experiment 1 - Your senses
Think about the last time your eyes played tricks on you. Or you misheard something clearly. Or you were absolutely sure you'd left your keys somewhere and found them somewhere else entirely.
Our senses aren't perfectly reliable. They make mistakes. So here's Descartes' first question:
If your senses have deceived you before, how do you know they aren't deceiving you right now?
You might brush this off. "Sure, minor errors, but broadly my senses work." Fair. But we aren’t done yet.
Mental Experiment 2 - Maybe you’re dreaming
Have you ever woken from a dream so vivid it took a second to realise it wasn't real? In that dream, you weren't suspicious. You had no internal alarm saying this isn't real. You were just... in it.
So how do you know, right now, that you're not dreaming?
What signal would you even look for? The dream version of you had no signal either.
Well, this is a tough one, and harder to dismiss. It goes beyond senses, it challenges whether the entire world you think you're experiencing is actually there.
Mental Experiment 3 - The evil genius
This is where it sounds ridiculous at first, but stay with it.
Imagine a being - powerful, deceptive, and endlessly intelligent. Whose entire purpose is to make you believe things that aren't true. Not just tricking your senses. But warping your ability to reason at the deepest level. Making you believe 2 + 2 = 4 when it doesn't. Making your logic feel airtight when it's been engineered to mislead you.
Now, and this is the real question - how would you know?
If your reasoning itself could be corrupted, then you can't trust any conclusion you reach, including the logical ones that felt bulletproof before. This layer attacks the system that generates all beliefs.
The one thing that survives
After all of that, Descartes finds exactly one thing he cannot doubt - and yes, the famous line:
“I think, therefore I am.”
If you try to doubt that you exist, you catch yourself. The act of doubting is itself proof that something is thinking. You can be deceived about the external world, your senses, your logic, but you cannot be deceived into non-existence, because the deception itself requires a mind to deceive.
Descartes called this bedrock. The one indubitable foundation from which all knowledge could be rebuilt.
But here's where I think it quietly falls apart.
After that statement, Descartes needs to reconstruct the rest of the world. He does it by arguing that a non-deceiving God guarantees our faculties are trustworthy. But strip away that argument, and you're left with one single truth: I exist. Everything else we actually care about knowing like science, the people around us, the world, remains in doubt.
Three reasons absolute certainty is out of reach
Let's set Descartes aside and ask the honest question: could some better approach get us there?
I don't think so, and here are my three reasons that come from completely different directions.
Infinite regress of justification: Any belief rests on reasons. But those reasons are themselves claims, which need their own reasons. At some point, every chain either becomes circular or rests on an assumption you've simply decided to accept. There is no bedrock that justifies itself. Following this logic, where every connection of knowledge depends on others, and no strand is absolutely secure on its own. Absolute certainty about any single claim would require certainty about everything it connects to.
The limits of Empirical knowledge: This is David Hume's unsettling observation. Most things we learn about the world comes through experience. When you drop your phone, you expect it to fall. Why? Because it has always fallen before. And we are reasoning inductively. The assumption that future resemble the past. But what actually justifies that? You cannot prove it through experience, because appealing to the fact that nature has been uniform is itself an inductive argument. The result is that the foundation of all empirical knowledge is an assumption we rely on but can’t rationally ground.
Our tools of thought language are inherently imprecise: We think and communicate in language, and language is inherently fuzzy. Every term is defined using other terms. For example, suppose this statement: There is a cup on this table. Seems straight forward. But what counts as a "cup"? Is it the shape? The function? What if it has a hole in the bottom? If the building blocks of our language are approximate, then every claim we build from them inherits that approximation. And when our tools of reasoning have a ceiling. How can we be absolute certain of things?
Taken together, absolute certainty isn't something we haven't worked hard enough to find. It's structurally off the table for beings like us with limited perception. We are, inescapably, reasoning about models of reality rather than reality itself.
So does this mean we know nothing for certain?
No. And this is the part I keep coming back to.
The mistake is assuming knowledge requires absolute certainty in the first place. Kant offers a better frame: instead of asking whether a belief is perfectly indubitable, ask whether it's objectively valid, whether any rational person, given the same evidence, would reach the same conclusion. Grounds that are sharable, assessable, and rationally compelling. Not a feeling of certainty. Something more honest than that.
Under this standard, mathematics and logic achieves the strongest form of what of reasonable certainty, the most robust conclusions we can reach.
Other knowledge like science, lived experience - achieves something real but revisable. Compelling given current evidence. Open to correction as better evidence emerges. This is actually what allows knowledge to improve. The whole history of science is better explanations replacing good ones, and that only works because knowledge stays open.
Okay, so what’s the takeaway?
I went into this question looking for certainty. I came out with something I think is more useful.
Knowledge isn't a building that needs perfect foundations before it's safe to live in. It's more like an interconnected web that is fallible, always capable of being revised. And its value is in the honest, never-ending pursuit of better understanding of the universe.
And maybe there's something genuinely freeing in that. Not knowing for certain doesn't mean you can't act, trust, build, or commit. It means you hold your beliefs with honesty, firmly enough to move on them, lightly enough to update when the evidence changes.
I think that's what knowing actually looks like.
David So
