Kant’s Transcendental Idealism: the ‘Copernican Revolution’ of Philosophy

Königsberg, 18th-century Germany: the philosopher Immanuel Kant, back from one of his regular evening walks around the town, is deeply troubled.

That morning, he’d finished reading David Hume’s 1748 work, An Enquiry Concerning Human Understanding — and his intellectual world had come crumbling down.

Hume, philosophy’s great skeptic, had surveyed the field before him and concluded that our lack of progress in metaphysics, epistemology, and philosophy generally shouldn’t really surprise us, for we are fated to never truly understand not just the status of external reality — but virtually anything about life.

Indeed, the approaches of rationalism (à la Descartes and Spinoza) and empiricism (à la Locke and Berkeley) fail not only in providing certainty about the external world, Hume argues, but also in establishing truth around causation, inductive logic, and even our own existence as continuous selves.

The arguments Hume develops to support these rather gloomy conclusions about what we can know are some of the most robust in philosophy.

He essentially demonstrates that everything we take for granted is not known, and never can be, as strict proof plays no part in human life outside mathematics.

Hume shows that most of reason’s claims are invalid, and that we know almost nothing: our thoughts are connected not by logic but by association of ideas, and our behavior is guided not by genuine understanding of reality but by habit and custom.

Reading Hume, Kant was dumbfounded. Here were some very rigorous arguments that seemed to undermine all confidence in our ability to establish any certain knowledge about the human condition, with no easy way to respond.

Was Hume right? Were all human theories ultimately doomed to groundlessness and mere evidence-based ‘belief’?

Kant, thoroughly roused from a self-described “dogmatic slumber”, thus decided that answering the questions Hume posed about the human limits of knowledge was the most important intellectual task of all.

But where to start? Kant was persuaded by Hume’s critique of both rationalism and empiricism as models for knowledge — so what else was there?

Here, after much deliberation, Kant introduces a new model: what he called transcendental idealism.

Kant’s transcendental idealism: a new model for knowledge

While rationalism is focused on deducing truths from starting principles using reason, and empiricism focuses on extracting knowledge from our experience of the world, transcendental idealism essentially tries to establish knowledge by analyzing what makes our experience of the world even possible: by investigating the preconditions that enable us to experience anything at all.

In other words, Kant is urging us to take a step back here. Let’s not just take our experience of the world for granted, he says. Instead, let’s ask: what makes our experiences possible? Why do we have experiences? What kind of experiences could we have? What sort couldn’t we have?

By conducting an investigation into the metaphysics of experience in this way, Kant thinks, perhaps we’ll get closer to securing the foundations of knowledge.

For instance, the English philosopher John Locke famously stated that at birth the mind is a tabula rasa — a blank slate — and that as we grow our experience of the world furnishes the mind with knowledge.

But for Kant, this cannot be so: the mind cannot be blank. For experience to be possible at all, the mind must be equipped with certain potentialities for understanding.

Our experience of the world, Kant explains in his 1781 epic Critique of Pure Reason, is determined wholly by the nature of our sensory apparatus and nervous systems: if our eyes or brains were different — like those of a bat or a dolphin, for instance — the world we experience would be different, too. He writes:

it remains completely unknown to us what objects may be by themselves and apart from the receptivity of our senses. We know nothing but our manner of perceiving them; that manner being peculiar to us, and not necessarily shared by every being.

The importance of this profound insight should not be understated: it implies that the world around us is a reflection not of reality, but of our specific, limited apparatus for interpreting that reality.

As I outline in my article on Locke’s tabula rasa, Locke actually began this work himself with his analysis of color: he showed that color is a characteristic not of reality itself but of our interaction with reality.

Kant brings this work to its startling conclusion by arguing that not just color or other sensory experiences, but time and space themselves, are features not of reality, but of our experience of reality.

Kant’s Copernican Revolution

Kant described this new way of thinking as the “Copernican Revolution” for our theory of knowledge. Just as the 16th-century astronomer Nicolaus Copernicus turned astronomy inside-out by hypothesizing that the earth moved around the sun, rather than the other way round — elegantly solving lots of problems by doing so — Kant turns our theory of knowledge inside-out by theorizing that our minds conform not to reality, but that reality conforms to our minds.

In other words, we do not start with reality: we have no choice but to start from where we are, with our brains’ potential for possible experience. Our cognitive systems then structure whatever is ‘out there’ in ways that make it possible for us to experience it.

Kant thus places us as knowers at the center of his metaphysical picture: we are looking out not at ‘ultimate reality’, but at a world constructed by our sensory and cognitive apparatus.

All of our experiences, Kant notes, take place in time and space, and are governed by natural laws like causation.

Our mistake, Kant thinks, is to assume that such structures are thus independent, built-in features of ultimate reality.

Really, all we can say is that, in our experience, such basic features are always present. We cannot extend such features to reality away from our experience without making a logical leap.

If we take seriously Kant’s Copernican Revolution, the placing of the experiencer at the center of empirical word, then we have to take seriously the idea that basic features of the world like time, space, and causation do not exist independently of our experience.

Rather, they represent ways in which our cognition is structured: they are built-in features not of reality, but of experience.

The upshot of this is two separate realms: the day-to-day empirical world we all experience, which Kant calls the ‘phenomenal’ world; and reality in-itself, which Kant calls the ‘noumenal’ world.

While we can explore and analyze the world of ‘phenomena’ — i.e. the day-to-day empirical world that the natural sciences investigate, our world of time, space, matter, and causation — we can never directly access noumenal reality in-itself, because we are forever destined to experience things as determined by our sensory and cognitive apparatus.

Together, Kant says, we live permanently in the phenomenal world, the world of our collective possible experiences — and our collective possible experiences are not reality.

The difference between ‘phenomenal’ experience and ‘noumenal’ reality

We might think: that’s all well and good, but surely the world of our possible experiences must correspond to reality in-itself in important ways. This apple I’m holding might be my experience of the apple, but surely it must be a pretty good representation of how it actually exists when removed from my experience, in-itself.

Not so, says Kant. Reality in-itself is, necessarily, distinctly different from the world we experience.

This difference is illuminated brilliantly by the late philosopher Bryan Magee, who in his commentary on Kant in his 1999 book Confessions of a Philosopher writes:

What is inside my head is not the Empire State Building but a set of visual data which I interpret as being the Empire State Building. The same is true of tactile data. The process of feeling a physical object — or, even more so, bumping into it — has something especially thingy about it, and can easily tempt us into believing that here is the brute reality of an independently existing world; but of course such feelings of bumpings are as inside the head as visual data are… All our experience of the external world is of this character. The direct and immediate awareness is in our brains, and must take a brain-dependent form to be awareness-material at all…

Indeed, the world we experience is determined by the possible forms our experiences and perceptions take: they all take place in time and space, are governed by causation, have associated colors, tastes, smells, sounds, and so on.

Thanks to our apparatus — our senses, nervous systems, and brains — we cannot conceptualize things as appearing in any way other than in such forms. Our cognitive systems impose structures that must be conformed to for experience to be possible at all.

Reality in-itself, by contrast, has no such restrictions; it exists apart from all these things. It’s what remains when all of these ‘necessary structures for possible experience’ are stripped away.

As Magee continues:

If there is a reality that exists independently of brains (and nearly all of us believe that there is) then it cannot exist in forms that are brain-dependent — despite the fact that these are the only ways in which we can apprehend it or form any notion of it… So as far as knowledge of objects goes there must be at least two realms, one of things as mediated to us by our brain apparatus and the other of whatever actually is, independent of all such brain and body operations…

The point is this: the world that appears to us, the world of phenomena, necessarily depends on the nature of our experiencing apparatus — it exists in terms of certain mind-dependent forms and categories.

Noumenal reality, by contrast, does not (and cannot) depend on our experiencing apparatus for its existence. Rather, noumenal reality exists independently, away from our minds — and we can never pierce through the veil of our mind-dependent experiences to determine the true intrinsic nature of mind-independent reality.

To adapt a further example from Magee: just as a photograph of a waterfall is not a waterfall, it’s a photograph; and just as a written description of a waterfall is not a waterfall, it’s a written description; and just as a sound recording of a waterfall is not a waterfall, it’s a sound recording: so our experience of reality is not reality, it’s experience.

And experiences cannot be intrinsically ‘like’ reality any more than a photograph is intrinsically the same as the thing photographed, or a sound recording the same as the thing recorded: it mediates it through a certain apparatus to capture something that is determined by the apparatus.

Put another way: noumenal reality is nothing like the world we experience, because it is not structured by the forms our cognitive systems otherwise impose.

We cannot possibly experience (or know anything about) unstructured noumenal reality, for it exists outside our conception of time, outside our conception of space, colorless, odorless, tasteless, shapeless, formless, non-conceptual.

Whatever the apple is in-itself, therefore, as a nonspatial, nontemporal thing, it does not — and necessarily cannot — resemble the apple that we observe and discuss.

The limits of human knowledge

So, says Kant, we shouldn't ever mistake our experience of reality for reality itself, because it’s not: it’s experience. We can make all the scientific progress we like in dissecting the world of our possible experiences, the world of time and space and causality. But we should never mistake this world for independent reality: it is reality as experienced by humans, nothing more.

Magee brilliantly summarizes Kant’s vision for how we experience the world thus:

If we think in terms of the metaphor of catching things in the network of experience, these are the meshes of our nets. Only what can be caught in these meshes is available to us. Anything that passes through them untouched will not be picked up by us, and nor will whatever falls outside our nets altogether. Only what these nets catch will be ours, and only what they can catch can be ours. What they do catch is a contingent matter, depending on what there is to be caught, but what they can catch is determined by the nature of the nets themselves, and we live permanently within their capacities and their limitations.

Who knows what the universe is really like? With our incredibly limited apparatus for detecting and understanding reality, there might be all sorts of things surrounding us that we don’t even know are there.

All we can comment on, Kant says, is the ‘phenomenal’ world of possible human experience: we can never directly access ‘noumenal’ reality itself.

A world without subjects is a world without objects

A slightly different way to frame Kant’s arguments is the following: there cannot be an ‘object’ without a ‘subject’ — the two are co-dependent.

Take the subject away, and it becomes incoherent to even discuss distinct objects, for a subject is required to make such distinctions in the first place.

It follows, Kant says, that objects by themselves are unknowable. Time, space, causation — these are features of the subject/object interaction. We cannot describe them as features of ‘ultimate reality’, because we can never strip the ‘subject’ (us) from the world.

Kant’s ultimate noumenal ‘world-in-itself’ — a world without subjects — is forever out of reach.

Kant wasn’t trying to deny that we could master or learn more about our environments, the day-to-day empirical world of our possible experience; he’s just saying that this world is forever entangled with concepts and structures that we, as experiencing subjects, bring to the table. We cannot get behind possible experience and its necessary features to access things in themselves.

The upshot? We shouldn’t equate the world of our experience with ‘ultimate reality’, for ultimate reality requires the stripping of all subjects.

We should thus limit our scope of inquiry accordingly: science investigates the empirical world, the world of our collective possible experience; it does not, and can never, investigate ‘ultimate reality’.

What are the advantages of Kant’s transcendental idealism?

Kant doesn’t intend to block us from ultimate reality simply for the sake of it; if we recall, he’s searching for a way to improve philosophy, to answer Hume’s skeptical concerns, and ultimately to provide a model for knowledge that reconciles modern science with religion and morality.

He thinks his Copernican Revolution — placing us as experiencing beings at the center — is the best way for us to move forward, and grants us a number of advantages.

Among these advantages is the grounding of human knowledge on secure, a priori foundations. As Kant writes:

If intuition has to conform to the constitution of the objects, then I do not see how we can know anything of them a priori; but if the object (as an object of the senses) conforms to the constitution of our faculty of intuition, then I can very well represent this possibility to myself.

To illuminate what Kant really means here, and why he believes his solution is so powerful, let’s consider one of the problems Hume identifies with our everyday conception of the world: that many of the ‘natural laws’ that we take for granted have no basis in logic.

Take causation. How do we establish causation between, say, event A and event B?

Well, Hume says, typically we observe that event B always follows event A, and then conclude that A thus causes B.

But this is a leap. All we can say, Hume points out, is that, based on our observations so far, B has always followed A. This is no reason, logically, to believe that B will follow A in future, nor to assert that A causes B.

According to Hume’s analysis, it turns out our entire concept of ‘causation’ is thus grounded not in logical necessity, but in habit and custom — and he applies this kind of skeptical analysis not just to causation but to many other sacred cows of metaphysics.

For Hume, a lot of our ‘knowledge’ simply has zero basis in logic; we cannot really be said to ‘know’ anything at all.

How, then, does Kant’s system combat Hume on issues like causation? Rather ingeniously, as it turns out.

While Hume is trying (and failing) to impose the law of causation onto the world, Kant flips the model on its head: for us to possibly experience it, he says, the world must conform to the law of causation.

Indeed, the world we experience must conform to certain fundamental laws and structures — causation among them — because the human mind constructs them according to those laws.

In other words: for us to possibly experience it, the world we experience has to be a certain way. Causation is thus a necessary law, Kant says, because experience would not be possible without it: it would be impossible for us to experience a world in which a given event didn’t have a cause.

So, causation has no logical basis in Hume’s system; but it does in Kant’s.

For Kant is not just looking to explain why the empirical world appears to us as it does; as we’ve seen, he wants to establish a new metaphysics for experience, and does so by establishing the necessary preconditions required to make experience of the world possible in the first place.

Such necessary preconditions include time, space, causation, and so on, Kants says: their existence within the empirical world, therefore, cannot be doubted.

This is the advantage of Kant’s transcendental idealism: by establishing the boundaries of possible experience, Kant in turn establishes some novel a priori metaphysical truths that had eluded the philosophers who came before him.

Of course, that we cannot then extend the existence of time, space, or causation to ‘ultimate reality’ is regrettable — but it’s an unavoidable consequence of grounding knowledge on the a priori foundations of possible experience, Kant says:

[F]rom this deduction of our faculty of cognizing a priori […] there emerges a very strange result […], namely that with this faculty we can never get beyond the boundaries of possible experience, […and] that such cognition reaches appearances only, leaving the thing in itself as something actual for itself but uncognized by us.

Kant’s system means that having knowledge of anything beyond the boundaries of possible human experience is impossible.

Physics, chemistry, and biology thus inform us only about the world it is possible for humans to experience; they have nothing to say about reality ‘in-itself’.

Limiting the scope of science in this way, however, is not all bad, Kant thinks, for it creates room for religious faith.

Indeed, while we can’t know about the noumenal world, about reality-in-itself, we can still believe things about it. Kant’s unknowable noumenal reality was conveniently plugged by his faith in God. He writes:

I had to deny knowledge in order to make room for faith.

Kant thought his transcendental idealism thus reconciled modern science and religion, and provided a metaphysical picture that offered solutions for many of the skeptical problems raised by Hume.

Objections to Kant’s transcendental idealism

Alas, while Kant thought he was advancing philosophy, many saw his transcendental idealism as a huge step backwards in our quest for securing the foundations of knowledge. Ultimate reality is forever out of reach and we can never know anything about it? Surely we cannot be happy with such an outcome…

Finding holes in Kant’s arguments has thus been a favorite past-time for many a metaphysician and thinker since.

While offering a full survey of all the possible objections to Kant is beyond this article’s scope, let’s briefly explore a few common approaches.

Firstly, we can point out that Kant seems to be rather contradictory about exactly what we can know about noumenal reality in-itself. On the one hand he rules it out as totally unknowable; on the other he says we know it exists, that it is non-spatial, non-temporal, and that it affects our senses.

On that note, if noumenal reality has no time, no space, no causation, and so on, then exactly how does it affect our senses? How can it ‘cause’ certain phenomena to appear to our senses, if causation has no existence outside of human experience?

Another common objection to Kant is to say that it traps us inside our own minds, and leads to an inescapable, unacceptable form of skepticism about the external world.

Some argue that Kant’s transcendental idealism, on close inspection, actually just collapses down to the subjective idealism of George Berkeley, who claimed that minds (and their contents) are the only things that exist.

Others question the coherence of Kant’s noumenal reality in-itself. If we can never know anything about it, if it’s forever out of reach, then what’s the point of it? Nietzsche took this position: who cares?

If we can never strip ‘subject’ from ‘object’, Nietzsche thinks, then the conclusion to draw isn’t that a mysterious noumenal realm of objects ‘in themselves’ is forever lurking out of reach; rather, the conclusion to draw is that such a realm is incoherent. The subject/object interaction is the only reality we need to concern ourselves with…

However, despite the fierce criticism Kant’s philosophy has faced since the publication of Critique of Pure Reason, his impact on metaphysics, epistemology, and philosophy generally is absolutely undeniable.

Whether you agree with his arguments or not, Kant’s transcendental idealism is an ingenious attempt to make sense of some of philosophy’s most difficult questions. Can we have a priori knowledge about the world? What is the scope of scientific inquiry? Do our experiences reflect reality?

One will not go far in philosophy without bumping into Kant along the way.

What do you make of Kant’s analysis?

• What do you make of Kant’s arguments that our cognitive systems are forever bound up in our perceptions of the world?
• Do you think we can get behind experience to access reality in-itself?
• Do you agree with Kant’s view that some aspects of the world will be forever hidden to us?
• Or do you think knowledge about the world can transcend our human limitations?⁣

Learn more about Kant, epistemology, and metaphysics

If you’re interested in learning more about Kant, epistemology, and metaphysics, I cover all of these topics in more detail in my Life’s Big Questions guide, which distills the great philosophers’ best answers to life’s big questions.

You might also like my list of the best books on Kant and his philosophy, as well as my epistemology reading list and metaphysics reading list.

You might also enjoy the following related reads:

• Is the World Around Us ‘Real’?
• Metaphysics: What Is It? Why Is It Important Today?
• George Berkeley’s Subjective Idealism: The World Is In Our Minds
• John Locke’s Empiricism: Why We Are All Tabula Rasas (Blank Slates)
• I Think Therefore I Am: Descartes’ Cogito Ergo Sum Explained

Finally, if you enjoyed this article, consider subscribing to my free Sunday breakdown. I distill one famous idea from philosophy each week; you get the summary delivered straight to your email inbox, and are invited to share your view. Consider joining 16,000+ subscribers and signing up below: