Last updated: November 3, 2026

The thing in itself, in German Ding an sich, is Immanuel Kant’s name for reality as it is independently of how it appears to us. Kant argued that everything we can know comes to us through the forms of our own mind, space, time and concepts such as cause and effect, so we only ever know things as they appear, as “phenomena”. What they are in themselves lies beyond any possible experience. We can think about it, Kant said, but we cannot know it.

That idea, set out in the Critique of Pure Reason (1781, revised 1787), is one of the most influential and most argued about in the history of philosophy. This post explains what Kant meant, why he thought the thing in itself is unknowable, how it relates to the word “noumenon”, and why his successors found it so hard to live with.

No Kids, Only Things in Themselves

The meme above uses the classic two panel format. Kant, in an orange puffer jacket, turns away in disgust from “I have no kids” and points approvingly at “I’m only interested in the thing in itself.” It is a joke about a famously unworldly man. Kant never married, never had children and, according to the story his biographers loved to tell, lived by such a fixed routine in Königsberg that neighbours could set their clocks by his afternoon walk. The story is probably exaggerated, but he did spend almost his entire life in and around that one city on the Baltic.

The irony is that the thing in itself is the one thing Kant insisted he could not be interested in, at least not as an object of knowledge. Much of his greatest book is a careful explanation of why the question “what is the world really like, apart from us?” cannot be answered by theoretical reason. The meme has Kant pointing at the one thing he said nobody can point at.

The Problem Kant Was Trying to Solve

Kant was born in Königsberg, then in Prussia and now Kaliningrad in Russia, in 1724 and died there in 1804. He taught at the local university for decades, lecturing on everything from logic to geography, and published nothing major for about ten years while he worked on the problem that became the Critique of Pure Reason. He was 57 when it appeared. Three years later, in his essay “What Is Enlightenment?” (1784), he gave the age its motto, sapere aude, dare to use your own understanding, a phrase we also put on a shirt. The thing in itself is part of the same project: using reason boldly, and that includes finding out where reason’s limits are.

Kant started out as a fairly conventional philosopher in the rationalist tradition of Leibniz and Wolff, who believed that pure reason could discover the ultimate nature of reality. Then he read David Hume, who argued that we never actually observe causation, only one event following another, so that our belief in cause and effect is a habit rather than an insight. Kant later wrote that Hume had interrupted his “dogmatic slumber”, a moment C.M. Black describes in Waking Up: Kant’s Dogmatic Slumber and You.

Kant faced a dilemma. If Hume was right, then the sciences, which are full of necessary laws about causes, rest on nothing more than habit. If the rationalists were right, then reason could reach beyond experience, but metaphysics had been arguing for two thousand years without settling anything. Kant’s solution was what he called his Copernican revolution in philosophy. Copernicus had explained the movement of the heavens by moving the observer. Kant proposed that instead of assuming our knowledge must conform to objects, we should assume that objects, as we experience them, must conform to the structure of our knowledge.

The Glasses You Cannot Take Off

According to Kant, the mind is not a passive mirror. It actively organises what reaches it. Space and time are not features of things in themselves but the forms in which anything can appear to us at all, what he calls the forms of intuition. And our understanding brings concepts, which he calls categories, such as unity, substance and causality, that turn a flood of sensations into a world of objects. In one of his most quoted lines: “Thoughts without content are empty, intuitions without concepts are blind.”

The usual way to explain this is a pair of tinted glasses. If you were born wearing blue glasses that you could never take off, everything you saw would look blue, and you could never find out what colour things really are. The analogy is helpful and also misleading, which is worth noticing. Glasses can in principle be removed, and we know what they are made of. Kant’s forms cannot be removed even in thought, because without them there would be no experience at all, not even of a world without colours. The point is not that we see the world slightly distorted, but that “the world as we can know it” is always the world as structured by these forms.

This has a remarkable consequence that Kant thought was good news. Because space, time and causality come from us, we can know a priori, before any particular experience, that every possible experience will be in space and time and will obey causal laws. That is how mathematics and physics can be certain. Hume’s worry is answered. But the price is steep: this certainty only applies to appearances. About the thing in itself, our categories tell us nothing.

Phenomena and Noumena

Kant draws the line between two terms. Phenomena are things as they appear to us, structured by space, time and the categories. Noumena are, literally, “things thought”, objects that would be accessible to pure intellect without the senses. People often use “noumenon” and “thing in itself” as synonyms, and Kant sometimes does too, but there is a subtle difference that philosophers like to point out.

The subtle difference is this. Kant distinguishes a positive and a negative sense of noumenon. In the positive sense, a noumenon would be an object of an intellectual intuition, a kind of direct, non sensory knowing that perhaps God has but we do not. In the negative sense, a noumenon is simply something that is not an object of our sensory intuition, a limiting concept that marks where our knowledge ends. Kant allows only the negative sense. That is what people mean when they say the noumenon is “thinkable but unknowable”: we can form the idea of it, and we even need it, but we can never fill it with content.

An Apple, Taken Apart

It helps to run Kant’s distinction through a single ordinary object. You see a red apple on the table. You can tell that it is round, that it is about ten centimetres across, that it sits to the left of your cup, that it was there a minute ago and will be there in a minute, and that if you knock it off the table it will fall because of gravity. All of that is knowledge, and all of it is phenomenal: it concerns the apple as it appears in space and time and within the network of causes.

Now strip away, in thought, everything that depends on the forms of your experience. Remove its place in space, its duration in time, its causal connections, its colour, its shape. What is left? Kant’s answer is that you cannot say. Not because the apple is hiding something, but because every “something” you could name would already be framed by those same forms. The thing in itself is what remains when you subtract everything you can know, which is why it can be thought but never described. It is not a second, hidden apple. It is the apple, considered without the conditions under which anything can be an apple for us.

Science does not change this picture. A physicist can tell you that the apple consists of molecules, the molecules of atoms and the atoms of fields and particles. Each step is a deeper and more useful description of the apple as it appears to measurement. None of them is a description of the apple in itself, because instruments only extend our experience. They do not let us leave it.

Why We Still Need the Thing in Itself

If we can know nothing about the thing in itself, why not drop it? Kant’s answer is that it would be absurd to talk about appearances without something that appears. The very word “appearance” points to something that is not itself an appearance. In the preface to the second edition he insists that although we cannot know objects as things in themselves, we must be able at least to think them as such.

The thing in itself also does practical work. Because we cannot prove that the world in itself is a closed causal machine, there is room to think of ourselves as free, which morality requires. The same goes for God and the immortality of the soul. Kant summed it up in a famous sentence: he had to deny knowledge in order to make room for faith. That move connects the Critique of Pure Reason to his ethics, which C.M. Black explains in Kant’s Categorical Imperative, Categorically Explained.

Two Worlds or Two Aspects?

Here the professional arguments begin. On the traditional “two worlds” reading, there are two sets of objects: appearances in our minds and things in themselves somewhere behind them, causing them. That reading has an obvious problem. Causation is a category, and categories only apply to appearances, so how can things in themselves cause anything?

Many recent Kant scholars, following Henry Allison, prefer a “two aspects” reading. There are not two worlds but one, considered in two ways: as it appears to us and as it might be considered apart from the conditions of our experience. The tree you see and the tree in itself are the same tree. Other interpreters, such as Rae Langton in Kantian Humility (1998), read Kant as saying that we know only the relational properties of things, how they affect us and each other, and must remain ignorant of their intrinsic nature. No reading has won, and the question “what exactly is Kant’s thing in itself?” is still a live research topic. Anyone who tells you it is simple has probably not read the secondary literature.

The Objection Kant Never Escaped

The sharpest objection came almost immediately. In 1787 the philosopher Friedrich Heinrich Jacobi wrote that without the thing in itself he could not enter Kant’s system, and with it he could not stay inside. The reason is the problem just described: the thing in itself seems to be needed to explain why we have sensations at all, yet according to Kant’s own rules nothing can be said about it, least of all that it causes anything.

Kant’s successors took different exits. Fichte and then Hegel concluded that the thing in itself had to go, and that reality is fully knowable as the development of reason itself, the story we tell in our beginner’s guide to Hegel. Schopenhauer went the other way and claimed to know what the thing in itself is: a blind, restless will that drives the whole world, which gave his philosophy its famous pessimism, explored in Buddha vs Schopenhauer on Suffering. Neither move would have satisfied Kant.

What Kant Did Not Mean

A few misunderstandings are worth clearing up, because they come up again and again. Kant did not say that the world is an illusion or that it exists only in our heads. He called his position transcendental idealism, but he combined it with empirical realism: within experience, tables, trees and other people are perfectly real, and science tells us true things about them. The thing in itself is not a hidden “true world” that makes this one fake, the way Plato’s Forms make the shadows in the cave second rate, a contrast you can follow in the Allegory of the Cave.

Nor did Kant mean that the thing in itself is some mysterious substance we might one day reach with better instruments. No microscope or particle accelerator will get closer, because every instrument gives us more appearances, structured by the same forms. And he did not mean that we know nothing. We know a great deal. We just know it from a human point of view, and we cannot step outside that point of view to check it against reality as such, a limit Descartes had hoped to overcome with his method of doubt.

Why It Still Matters

The thing in itself sounds like the most abstract idea in philosophy, and yet versions of it keep turning up. When neuroscientists say that the brain constructs our experience of colour from wavelengths that have no colour in themselves, they are making a broadly Kantian point, though Kant would add that wavelengths are appearances too. When Thomas Nagel asks what it is like to be a bat and concludes that we may never know, he is pointing at another limit of perspective, discussed in Release the Bats!. And when Wittgenstein wrote that the limits of his language meant the limits of his world, he was redrawing Kant’s boundary in linguistic terms, as we explain in “The Limits of My Language Mean the Limits of My World”.

Kant’s lesson is not that reality is unknowable in some gloomy sense. It is a lesson about humility and confidence at once. We can be confident about the world as it appears to beings like us, which is the only world we will ever live in. We should be humble about claims to know how things are from nowhere. Some philosophers think Kant drew the boundary in the wrong place, and others that there is no boundary at all. But nearly everyone after him has had to say where they stand on it, which is why this one German phrase still fills seminar rooms. If you want the wider map of questions like this, start with What Is Metaphysics?. And if you want a shirt that sums up the whole problem, there is always Kant Touch This.