There's an age old problem (though I'm not sure of it's age exactly) regarding epistemic justification: how can I be justified in anything that I know to be true, even a principle as basic as modus ponens? After all, when I say

Given: If P then Q
Given: P

Therefore: Q

How have I really proven 'Q'? What if someone were to deny that I can conclude Q from the given premises: how could I convince him? How can I prove that modus ponens actually works? This problem is often referred to in a humorous manner, such as in Lewis Carroll's "What did the Tortoise say to Achilles". In his 'story', the only way to convince someone else of a syllogism would be to construct a similarly structured syllogism - but, of course, there's no way to validate the second syllogism, and so we require a third, etc.

While I've always been happy to leave this as it, I have a friend who studied under Peter Klein who's convinced that the fact that there needs to be an infinite amount of steps for justification is in itself a justification. This seems to me to be absurd: at no point is there ever a jump to say 'ah ha, this one works!' any more than the first case. While I've read Klein's papers, I remain unconvinced.

Can someone please explain how an infinite regress of justifications can provide actual epistemic justification?

  • As you can see from the Wiki's entry you are referring to, the issue is still debated between philosophers (Quine, Maddy); so, the best way to do is to read them. But basically, the issue is: it is better the "standard explanation" : we need to put somewhere a nail to hang our arguments (i.e.we cannot justify everything in a not-circular way) or we prefer to accept as justification an infinite regressive argument ? – Mauro ALLEGRANZA Jul 15 '14 at 7:56
  • @MauroALLEGRANZA my friend who was in Klein's lectures insisted that I'd agree if someone explained it to me properly, which he was incapable of doing (by his own admission). I thought maybe there would be a more eloquent/expressive Klein-follower here to explain. When you say "an infinite regressive argument", what do you mean? What's the argument? (If you can respond please do so as an answer) – This lad Jul 15 '14 at 8:03
  • I'm only quoting from Wiki's entry : "On this view, to be justified in believing P is to possess a reason R1 to believe P, and a reason R2 to believe R1, and a reason R3.....and so on, ad infinitum. Justification is, so to speak, 'turtles all the way down'". Do you prefer a justification which stops saying : "we must start from something which is nor more justifiable" of a justification which goes on ad infinitum ? – Mauro ALLEGRANZA Jul 15 '14 at 8:04
  • @MauroALLEGRANZA I don't understand how that can meaningfully be called a justification – This lad Jul 15 '14 at 8:09
  • 2
    If someone disbelieves Modus Ponens, it basically means in effect that they don't know what "if" means, and/or don't acknowledge that it is part of useful apparatus for reasoning. If they refuse to understand it, then the only way to convince them (if that is possible) is by demonstrating the utility of reasoning using conditionals. If they will not be convinced, their viewpoint is simply incompatible, i.e. there is a respect in which they are not your peer. – Niel de Beaudrap Jul 15 '14 at 8:50

This is my position, which could conceivably be wrong. I am not aware of any unanswered criticisms of it.

There is no infinite regress because justification is impossible, unnecessary and undesirable. If you assess ideas using argument then the arguments have premises and rules of inference and the result of the argument may not be true (or probably true) if the premises and rules of inference are false. You might try to solve this by coming up with a new argument that proves the premises and rules of inference but then you have the same problem with those premises and rules of inference. You might say that some stuff is indubitably true (or probably true), and you can use that as a foundation. But that just means you have cut off a possible avenue of intellectual progress since the foundation can't be explained in terms of anything deeper. And in any case there is nothing that can fill that role. Sense experience won't work since you can misinterpret information from your sense organs, e.g. - optical illusions. Sense organs also fail to record lots of stuff that does exist, e.g. - neutrinos. Scientific instruments aren't infallible either since you can make mistakes in setting them up, in interpreting information from them and so on.

What about Klein's specific argument? This is given here:


He assumes that justification is possible, necessary and desirable. He then argues that other accounts of justification don't work and that the best objections against infinitism don't work. The objections he addresses are (1) You can't do an infinite number of steps. Klein claims that what matters is that there is no proposition that couldn't be justified, even if it isn't actually justified. (2) If there is such an infinite stack of propositions then at some point they will be so complex that no finite mind can grasp them. Klein claims that the propositions need not increase in complexity in this way.

In reality, Klein's position is not tenable and this is not primarily because you can't make an infinite number of justifications, but, rather, because even the first justification doesn't work.

We don't create knowledge (useful or explanatory information) by showing stuff is true or probably true for reasons so how do we create knowledge? We can only create knowledge by finding mistakes in our current ideas and correcting them piecemeal. You notice a problem with your current ideas, propose solutions, criticise the solutions until only one is left and then find a new problem. We shouldn't say that a theory is false because it hasn't been proven because this applies to all theories. Rather, we should look at what problems it aims to solve and ask whether it solves them. We should look at whether it is compatible with other current knowledge and if not try to figure out the best solution. Should the new idea be discarded or the old idea or can some variant of both solve the problem?

See See "Realism and the Aim of Science" by Karl Popper, especially chapter I and "The Retreat to Commitment" by W. W. Bartley III.

  • 1
    I know I'm probably wasting my time, but I'm going to make a small suggestion. Can you change the first sentence to state that as your belief (or if you prefer you can it's Poppers or you following Popper) rather than an absolute certainty that has been decided? Or just restructure it to present the options of foundationalism and non-fundationalism concerning justification? and then say why you prefer the latter. – virmaior Jul 15 '14 at 11:17
  • 2
    Also, this answer does not answer the question and is as such off topic. A good answer would explain Kleins argument to OP (or another argument why an infinite regress suffices for justification). – Lukas Jul 15 '14 at 11:19
  • I have amended the post. Also, @Lukas you say my post should state why "why an infinite regress suffices for justification" but that's not my position. However, I have amended my answer to say what is wrong with Klein's position. – alanf Jul 15 '14 at 11:31
  • Rather than adding a disclaimer at the top (which is problematic in its own right), you could just tone down the universal: "There is no infinite regress because justification is impossible, unnecessary and undesirable." --> "I think justification is impossible and unnecessary. Thus, I avoid the infinite regress." – virmaior Jul 15 '14 at 12:26
  • 2
    @alanf: Well, frankly, noone asked for your position, and this board is no discussion-board, but a Q&A Site. Since OP asks for an explanation of Kleins Argument, relevant answers are those that explain Kleins Argument. Everything else is irrelevant. – Lukas Jul 15 '14 at 17:49

There is not actually an infinite regress. There is a long, subtle, and biologically ancient regress that you have adapted to unconsciously and simply called it you. But the interesting point, is that what you call "you", is shared at some point, at what your partner calls him/herself.

That is, that long chain is connected to your listener's chain. That is what the purpose of dialog is: to continue working each others way down each's respective chains of causal thought until one reaches and creates through linguistic feedback the point in common in which you agree. Call it a shared "cultural meme", but consider that since your perception and cognition systems evolved along similar paths, it actually gets down to shared genetic ancestry.

It's very much like the correspondence of physics and math. There the universe co-evolved with consciousness, so that there is a point in common that makes the universe have sense and not be simply and blandly arbitrary -- which would otherwise be its nature.


With the 1st statement (Given: If P then Q), you are making a "truthful" assertion. Since the validity of the 3rd statement is conditioned on the validity of your assertion, your listener may or may not accept the validity of the 3rd statement, depending on how credible you have been in the past, your reputation, your credentials, etc., which would be the source of your "justification."


Infinite regress seems like a flagrant violation of Ockham's razor. The razor is not provable, but it makes discussion possible. Otherwise people just live in their own heads and struggle to find common ground.

Infinite regress would also seem to hit the cosmic wall around the time of the big bang.

I suspect infinite regress is also a way to avoid accusations of special pleading, which seem to come with the territory in epistemology. Consider: is the uncaused cause special pleading? Because without it you get infinite regress.

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.