F. H. Jacobi (Wikipedia) argued in the late 18th century against Kant that any foundation of knowledge will inevitably lead to an infinite regression of justifications (thus any foundation of knowledge will eventually lead to nihilism).

I know that many Neo-Kantians have tried (sometimes successfully) to "fix" this issue in Kant's philosophy (Kant himself tried in Opus Postumum). But I haven't heard of that in the 20th (and 21th) philosophy of metaphysics. So, was it just taken for granted after the Neo-Kantians' solutions? Was it ignored? Or was it discussed and I just haven't seen this discussion (and hopefully will get references if this is the correct option)?

  • 1
    There's no need to look to Neo-Kantians or anyone else for that matter, because Kant himself addressed this question. The foundation of knowledge would lead to an infinite regress if it weren't grounded in God as the original source of reason. Kant asserted, "For the greatest systematic and purposive unity, which your reason demands as a regulative principle to ground all investigation of nature, was precisely what justified you in making the idea of a highest intelligence." (A699/B727) So without God, there would be no possibility for knowledge.
    – user3017
    Feb 25 '18 at 20:14
  • 1
    Furthermore, if Jacobi had understood Kant's argument concerning the antinomies of reason, he would have realized that his infinite-regress argument was a moot point which Kant had already carefully worked out in the Critique of Pure Reason. Therefore, there never was any issue that would have needed fixing in Opus Postumum, nor is there anything for modern philosophers to defend which Kant hasn't adequately defended himself.
    – user3017
    Feb 26 '18 at 1:40
  • @PédeLeão and what if we don't want to base knowledge in God? Feb 26 '18 at 10:50
  • 1
    You're free to base your philosophy on whatever makes you happy. However, John Maynard Keynes expressed the fundamental principle that "...in any formal inference the conclusion is implicitly contained in the premises, and affirms no absolutely new fact." Any theory of learning must observe this, so denying God, the only possible source of a priori concepts, would require a new form of logic whose conclusions are based on nothing at all. Atheists try to avoid this, hoping that complexity and obscurity can hide the fact that their epistemological assertions defy the laws of logic.
    – user3017
    Feb 26 '18 at 14:07
  • 1
    And nor likely one on the thought of Yechiam Weiss soon ;). Or mine come to that. Feb 27 '18 at 14:18

Infinite regress problems are a result of a misunderstanding of epistemology. Epistemology is about how knowledge is created, how you can distinguish what ideas you should adopt and act on, and similar problems. Philosophers have commonly tried to solve this problem by saying there is a process called 'justification': a process that can make conclusions true or good or something like that - justificationism. An argument's conclusions are true if its premises and the rules applied applied to those premises are correct. So to show the conclusion of an argument is true, you must show the premises and rules are correct. To do that, you either assert by fiat that they are true, which proves nothing, or you make an another argument. If you make another argument, then you have the same problem as the original argument and you are no better off.

The notion that god somehow solves this problem doesn't make any sense. If you want to use god as a foundation then you either assert stuff about god by fiat, or you argue, which gives rise to the same problem.

Nevertheless, many philosophers ignore or deny or obfuscate this problem and spend their time arguing about induction or god or whatever. They argue endlessly about this issue and you can read vast piles of books about epistemology. People outside academic philosophy may occasionally think something along the lines that this is a difficult problem but progress is being made or something like that and point to some book or other that they haven't read or understood as being the state of the art.

One philosopher who didn't take this way out was Karl Popper. Popper bit the bullet and said that justification is impossible. Knowledge is created by noticing problems, making guesses about the solution, criticising the guesses and using the guesses that survive this process. The truth of those guesses is not guaranteed, nor is it probable or anything like that. But you can select ideas based solely on whether they survive criticism - that is, whether they leave relevant problems unsolved. For a guide to Popper's epistemological work, see


  • 1
    Isn't this a very old idea? I mean Socrates asked questions of the people in the polis and thereby got them to refine their ideas. Popper happened to be interested in Greek philosopy, and not just Socrates - Parmenides too. Yet by your reasoning it seems that one ought not to be interested in this and be merely interested in Popper, and just a small fraction of him; I mean his fallibilism; it strikes me as a very narrow outlook. Feb 26 '18 at 15:29
  • 2
    Making guesses and criticizing them presupposes concepts as well as some standard of evaluation, so Popper's theory is circular in nature. And nobody asserts God by fiat; rather, God has revealed His own existence in various ways. Kant demonstrated how God's role in our epistemology is logically consistent with reason.
    – user3017
    Feb 26 '18 at 15:36
  • @MoziburUllah I must say, to add to your comment (although not merely as old as the Greeks, but at least older then Popper) - I actually found the falsification idea in Schelling too. Feb 26 '18 at 17:32
  • @PédeLeão In the context of any specific argument some standard of evaluation and some ideas are used. But they are conjectures and can be refuted by criticism in subsequent arguments. There's no circularity because there's no attempt at justification. Your remarks about God don't address the problem I pointed out. Either Kant made an argument with rules and premises that could be wrong, or he assert God by fiat at some point. There is no third option.
    – alanf
    Feb 26 '18 at 20:35
  • @MoziburUllah Popper's fallibilist epistemology is directly relevant to this question. His writings about the pre-Socratics are not as relevant. Also, Popper's falliblism is more fully developed and explained than most fallibilist ideas before or since the time he wrote.
    – alanf
    Feb 26 '18 at 20:55

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.