I've been studying Locke recently and I'm having trouble understanding how his epistemic position differs from Kant's, and by implication, why did Kant see his epistemology as being so revolutionary given that it is not too different from Locke's. Both of them seem to assert the following:

  • Objects exist in the world independently of the perceiver.

  • We can never know the true nature of the objects, only the effect they have on ours senses.

  • Our mind plays an important role in the way our sense data is organized.

My questions

  1. What is it that I am missing here that makes Locke's realism and Kant's realism so different?

  2. Why did Kant see himself as having revolutionized epistemology, given Locke's results?

  3. My understanding was that Kant was mainly refuting Hume, why wasn't he concerned with Locke's results?


I think the difference between Locke and Kant is captured in the primary/secondary quality distinction:


Locke's view is just a step beyond naive realism. It is in some sense, our modern everyday view. Although he said we couldn't know the true nature of things, he thought position, motion, volume were primary qualities (qualities that existed in objects independent of the observer). I think this is close to the view most people take today... that there's an object out in space affecting our senses... our senses may not be accurate, but give some kind of analagous copy of the real object. ie: physics deals with position, velocity etc... and these quantities are taken as real and belonging to objects out there in correspondence with the objects of experience.

Berkeley took the step to say that all qualities are secondary. ie: position, velocity etc... are also experience dependent (if you take away all experiential aspects of objects, in what sense can you say it is at any spatio-temporal location).

Kant, imo, kept Berkeley's view that all properties are secondary, but then analyzed these secondary properties in greater detail (analyzed the structure of experience which is all that we can really do). All properties might be secondary, but space/time are preconditions of experience, and have a more fundamental role to play than smell or color. Same with categories of experience.

For Kant, the objects that exist independently of the observer (noumena) are not analogous to objects of experience. To him they were something completely alien we couldn't talk about with any intelligibility. Different from Locke's view.

Did some digging and found this term, 'indirect realism':


According to this, Descartes and Locke were indirect realists.

  • 1
    While you may be correct overall, I would be cautious with terms like "Kant kept Berkeley's view", because he actually only read the first available translations of Berkeley around 1783 (after the review of CPR being strongly influenced by it) and did not base any of his main ideas on him because of the historical fact that he couldn't read it in time. – Philip Klöcking Mar 29 '16 at 10:52

Kant brought into the epistemological discussion at least the following fundamental concepts, which are not present in Locke's epistemology:

1) Constructivist epistemology: We create or construct knowledge from the input of our senses by two capabilites of the human mind: intuition and categories. Knowledge is not an image of the outside world but the result of active construction thanks to our rational capabilities.

2) No essence: Things-in-themselves are necessary hypotheses to explain the input to our senses. But we cannot know any of their properties, notably not their essence.

3) A priori knowledge: When analyzing our rational capabilities we find a certain a priori knowledge: Appearance in space and time and the classificational concepts of the categories.


I haven't looked at any Locke in a long time, but let me grope around bit.

First, you are right that they are not as distinctly defined as we tend to think, partly because it was Kant who sorted his predecessors into the "rationalist-empiricist" slots to clear the scrub for his own work.

But I think the similarities are somewhat superficial. Locke, despite the blank slate analogy, did assume innate capacities, of course, but did not really show how they connected with sensible objects. He simply assumed that God would not equip us with reasoning capacities that did not somehow latch onto experience, enabling us to "trust our senses." Most people were more or less happy with this, though one would have to read Leibniz's criticisms. Hume showed how utterly vulnerable this stance was to skepticism, how little of certainty is left if we dispense with a priori necessity and if only the divine designer postulate is jettisoned.

Distressed by this, Kant undertook to demonstrate how experience itself reveals the underlying structure of reason and hence of a priori necessity, providing its own guarantee of knowledge. No flimsy assumptions of a mediating deity is needed to guarantee our senses "make sense." We can really "know" the object not because we are correctly receiving and interpreting the sense data from "out there", but because any object that can "be sensed" is already the necesary construction of a rational being.

Though Kant is prepared to strictly limit knowledge, he sought certainty within those limits, the synthetic a priori assurance that math and physics really did overcome the problems of induction and produce knowable "laws" of nature. Locke simply did not secure this, but I feel myself now fumbling to clarify further without consulting a book or two.


One additional note. Kant explored the a priori nature of interpreting the external world as more or less limited to the human understanding internally, whereas Locke didn't dispute that "things-in-themselves" exist in an established medium, IN SPACE and externally - regardless of whether a human being is conscious of them or not. Here's an excerpt from an article "LOCKE’S ‘THINGS THEMSELVES’ AND KANT’S ‘THINGS IN THEMSELVES’ by Yasuhiko TOMIDA. Kyoto University, JAPAN:

"The correspondence explained earlier is not the only correspondence between Kant’s and Locke’s frameworks. For Kant, sensible representations are produced according to the forms of our sensibility, and as he emphasizes in the Transcendental Aesthetic, in the case of our outer sense they appear with a spatial character. That is, though the representations appear in the mind, they are perceived as something in space. By contrast, Locke does not view space as an a priori form of our sensation, but as his argument in the chapter on perception in the Essay shows, the sensible ‘complex ideas of substances’ appear as something in space in spite of their general determination of ‘being in the mind’, and in our ordinary life we regard them as ‘things’ qua experiential objects. As is well known, in the chapter on perception, Locke analyzes the process of our spatial perception and discusses the close relationship between our visual and tactual sensations.[32] Thus, though Kant’s view is profoundly different from Locke’s, their framework shares important similarities."

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.