The definition of knowledge under critical rationalism

Pfhorrest

The traditional philosophical definition of knowledge, dating back at least to Plato, is that knowledge is justified true belief. That is to say that it is not enough merely to believe something to be the case, and it is not even enough for that belief to turn out to be true, but for someone to know something they must also have a justification for their belief, a reason to believe it, because it would not constitute knowledge to simply guess at an answer to a question (or otherwise come to believe it for insufficient reason) and just by luck turn out to be right.

Edmund Gettier has since proposed that even justified true belief is not enough to constitute knowledge, to the extent that reasons to believe something can sometimes be imperfect, can suggest beliefs that nevertheless turn out to be false, yet we nevertheless want to say that someone can still be justified in believing something for such reasons. Because if justification can be imperfect, someone could be justified in believing something that, despite that justification, might nevertheless turn out to actually be false, and in such cases we would not want to say that it counts as knowledge to be misled by imperfect justifications to believe something that could nevertheless have still been false but, by an unrelated coincidence, does happen to also be true, just not for the reasons justifying the belief.

This problem could trivially be remedied by insisting that only perfect justification, the kind that guarantees the truth of something, is good enough to turn true belief into knowledge; but that would imply that knowledge of almost any substantial topic, where such certainty cannot be obtained, is thereby impossible.

My response to this problem is similar to that of Robert Nozick: I say that knowledge is believing something because it is true, such that not only does one believe it, and it is true, but if it weren't true one wouldn't believe it.

This last condition can, I think, be considered a different sense of "justification" from the usual one, and so salvage the traditional definition of knowledge, albeit only by turning the concept of justification on its head, which I argue needs to be done anyway to have a workably rational method of deciding what to believe.

Namely, rather than the usual justificationist sense of rationalism, whereby no belief is justified until it can be supported from the ground up somehow, instead any belief is justified (including contrary ones) until there is support to the contrary, i.e. reason to rule that belief out -- an epistemological position called critical rationalism, supported by philosophers like Kant and Popper.

IOW, under a critical rationalist conception of knowledge, there is no Gettier problem at all, because justification in such a paradigm doesn't mean what Gettier assumes it does.

Isaac

any belief is justified (including contrary ones) until there is support to the contrary, i.e. reason to rule that belief out — Pfhorrest

How is the existence of a reason to rule it out not also a belief? All you have here is competing beliefs - the belief in A and the belief in a reason to rule out A.

Pfhorrest

It is easy to ground a disproof of something: just show a contradiction. E.g. if A = B and C, and you can show that B and C are contrary to each other, you can rule out A.

(It could still be the case that B, or C, you just know that you have to reject at least one of them, so it can't be the case that A. In general, ruling things out never narrows down to one specific remaining possibility, only a narrower range of possibilities).

Trying to ground a positive proof of one specific thing, on the other hand, inevitably leads down an infinite regress.

Isaac

if A = B and C, and you can show that B and C are contrary to each other, you can rule out A. — Pfhorrest

All of which requires a belief in the truth-preserving nature of logical functions, a belief that 'A = B and C', and a belief that 'B and C are contrary to each other'.

If you also had a belief that A, how does your system tell you anything about your belief that A? It remains the case that you are either mistaken about A or mistaken about the relationship between B and C, or mistaken about the relationship A = B and C (or you're mistaken about logic itself).

As you have no means of determining which of those mistakes are the case, you are not discarding a belief because it can be ruled out. You are discarding a belief because you wish to replace it with another.

TheMadFool

↪Pfhorrest

I was always bothered by one thing about Gettier Problems. It seems to be tautologous in the Wittgensteinian sense i.e. it doesn't add to what we already know.

There are two kinds of arguments viz. 1. deductive and 2. inductive. We needn't worry about deductive arguments as they're foolproof justifications in that if the argument is sound it's impossible that the conclusion is false. There is no room for error with deductive arguments is what I mean.

Coming to the other strain of arguments viz. inductive arguments, it is already known, in fact it's contained in the definition of such arguments, that the inference is probabilistic in the sense that there's a gap between inductive justifications and their conclusions which requires, as some would describe it, a leap of faith.

Gettier problems, in my humble opinion, are essentially about this justificatory gap in inductive arguments. In other words, we were already aware of the problem even before Gettier formulated his now famous argument regardind justified true belief. :chin:

KerimF

↪Pfhorrest

Let us imagine there is a certain proposed belief/idea.
Who, in your opinion, has the last word to approve it is true or not and/or it is useful or not?
Thank you.

Isaac

We needn't worry about deductive arguments as they're foolproof justifications in that if the argument is sound it's impossible that the conclusion is false. There is no room for error with deductive arguments is what I mean. — TheMadFool

As I said in my first comment, this is of no use with knowledge claims because we have no means of distinguishing premises from conclusions. We cannot say that our belief in A is justified by the deductive argument 'If B then A, B therefore A' because our belief that B might be what is at fault, or our belief that 'if B then A'.

The whole approach rests on the flawed assumption that we build up our beliefs one block at a time from some first principle like an inverted pyramid. There's scant evidence that we actually do this and abundant evidence that we don't.

magritte

I am missing something.
Are we talking about my own personal belief or something that is believed by everyone because all humans by necessity believe or perhaps because science believes ? Inductive belief can be either one and personal justification is quite different from what is innate or a law of physics.

TheMadFool

As I said in my first comment, this is of no use with knowledge claims because we have no means of distinguishing premises from conclusions. We cannot say that our belief in A is justified by the deductive argument 'If B then A, B therefore A' because our belief that B might be what is at fault, or our belief that 'if B then A'.

The whole approach rests on the flawed assumption that we build up our beliefs one block at a time from some first principle like an inverted pyramid. There's scant evidence that we actually do this and abundant evidence that we don't — Isaac

"Because"??!!

Philosophim

My response to this problem is similar to that of Robert Nozick: I say that knowledge is believing something because it is true, such that not only does one believe it, and it is true, but if it weren't true one wouldn't believe it. — Pfhorrest

Lets go over why this doesn't work.

1. People believe in things that aren't true all the time, but believe they are true.
2. People believe in things that aren't true, and ALSO do not have any contradictions to their conclusion that is true, based on a lack of evidence or critical thought.
3. The only way to know that what they believe isn't true, is to have an "outside observer" who knows the actual truth.
4. But if 1 and 2 are the case, then how can we trust the outside observer? Couldn't they also have the issue of 2? How do we know they aren't the same as 2? Meaning we haven't really discovered knowledge, because our current proposal has a giant hole it cannot fix.

The issue is trying to state that knowledge is a 100% grasp of the truth. Knowledge is a tool of rationality. It is our best approach to claiming the truth, much better than mere inductive beliefs. Basically, knowledge is a rational conclusion from the limited evidence and thought capability that we have. But it must always have the caveat that it is a conclusion that might lack the whole picture. While the only thing we can rationally conclude within the picture is likely the best course of action, we must always be open to the fact that we do not have the entire picture.

Kenosha Kid

My response to this problem is similar to that of Robert Nozick: I say that knowledge is believing something because it is true, such that not only does one believe it, and it is true, but if it weren't true one wouldn't believe it. — Pfhorrest

This requires knowledge of whether or not the thing is true, the knowledge that is under question. If one already knows it is true, belief is irrelevant. In JTBs, the truth is verified after the belief. I believe the ball is red. I look at the ball. It is red. The above has knowledge of the truth as a prerequisite for the belief which is in turn a prerequisite for the knowledge.

Edmund Gettier has since proposed that even justified true belief is not enough to constitute knowledge, to the extent that reasons to believe something can sometimes be imperfect, can suggest beliefs that nevertheless turn out to be false, yet we nevertheless want to say that someone can still be justified in believing something for such reasons. — Pfhorrest

I've seen this, but never seen a sensible example wherein the belief is both false and justified, or true and not justified either. Examples include 'X told Y that Z was true', as if believing X is a given, which bypasses the justification aspect.

Isaac

↪TheMadFool

I'm not sure his better to explain it than I just did. In order to support the notion that a belief that A somehow changes property as a result of some logical deduction, becomes knowledge that A, the facts which prime the deduction have to have some property different from the belief that A, but in no case do they. The 'fact' that B is just another belief.

The deduction hasn't had any impact at all on the belief that A, all it's shown it that it is logically consistent (or not) with a belief that B. It remains a belief that A.

Pfhorrest

It remains the case that you are either mistaken about A or mistaken about the relationship between B and C, or mistaken about the relationship A = B and C (or you're mistaken about logic itself). — Isaac

Sure, but you’re still mistaken about at least one of those things, so you know it can’t be the case that all of them are true at once, and the range of possibilities is thus narrowed.

The whole approach rests on the flawed assumption that we build up our beliefs one block at a time from some first principle like an inverted pyramid. — Isaac

You get that opposing that assumption (or rather, the assumption that that is the correct way to form beliefs) is what critical rationalism (as opposed to traditional justificationist rationalism) is all about, no? I’m not making that assumption, I’m explicitly opposing it.

Who, in your opinion, has the last word to approve it is true or not and/or it is useful or not? — KerimF

Nobody, but that’s more a question about institutional knowledge and epistemic authority than the topic of this thread which is individual knowledge.

Are we talking about my own personal belief or something that is believed by everyone because all humans by necessity believe or perhaps because science believes ? — magritte

See just above, although I do think that the methods of individual knowledge formation form the basis of the institutions of knowledge as well.

we must always be open to the fact that we do not have the entire picture. — Philosophim

I agree completely and didn’t mean to say anything to the contrary. It’s rather quite the whole point of critical rationalism that you never narrow down to any single certainty, only rule out some possibilities leaving fewer (but always multiple) remaining options.

It sounds like maybe you’re misreading the same thing Kenosha is below...

This requires knowledge of whether or not the thing is true — Kenosha Kid

No more so than traditional JTB. Really the whole “truth” component of both traditional and my modified JTB is a historical vestige that’s rather redundant. Knowledge is merely justified belief, where justification itself implies a reason to think it is true; we only bother saying “justified TRUE belief” because before the justification criterion was added, the standard was simply “true belief”. It would have been better if the “true” had simply been replaced by “justified”.

Philosophim

Knowledge is merely justified belief, where justification itself implies a reason to think it is true; — Pfhorrest

any belief is justified (including contrary ones) until there is support to the contrary — Pfhorrest

Ok, I just wanted to make sure this was what you really believed, as on its face, it seemed contradictory.

I'm surprised you've never visited my thread here https://thephilosophyforum.com/discussion/9015/a-methodology-of-knowledge/p1

I essentially start with the same Popper premise, but flesh out what justification entails. You might like it. And if you don't read it, that's fine to, just an invitation.

Isaac

Sure, but you’re still mistaken about at least one of those things, so you know it can’t be the case that all of them are true at once, and the range of possibilities is thus narrowed. — Pfhorrest

No. You always were mistaken about one if these things, they merely exhaust the set. If you've narrowed it, what was the possibility you've eliminated?

You get that opposing that assumption (or rather, the assumption that that is the correct way to form beliefs) is what critical rationalism (as opposed to traditional justificationist rationalism) is all about, no? — Pfhorrest

No, I don't get it. It doesn't sound like that at all. You suggested a belief that A can be held as knowledge (ie changes status by some significant property) as long as there is no...

reason to rule that belief out — Pfhorrest

If you agree that there being a lack of "reason to rule that belief out" is just another belief (let's call it a belief that B), then I'm struggling to see how its mere existence changes the functional status of the belief that A.

It doesn't make A more likely (there's no reason at all to think my belief that B any more likely to be the case than my belief that A so believing both cannot make each more likely to be the case).

It doesn't make A more robust (if anything it makes it more intractable)

So what does a belief that B (the belief that there's a lack of reason to rule A out) do to the belief that A to change its status at all?

Isaac

↪Pfhorrest

Perhaps an example might help rather than all these As and Bs.

Say I believe that there are unicorns in my back garden. How does my additional belief that there's no reason not to believe there are unicorns in my back garden change its status in any way. It seems to me to just remain a belief. Its just now accompanied by a related belief.

KerimF

Nobody, but that’s more a question about institutional knowledge and epistemic authority than the topic of this thread which is individual knowledge. — Pfhorrest

Thank you for clarifying.
So, it has nothing to do with my life :)
Wish you the best.

Pfhorrest

Ok, I just wanted to make sure this was what you really believed, as on its face, it seemed contradictory. — Philosophim

I don’t see the apparent contradiction. Can you elaborate?

I’m basically applying the same standard of justification to belief as we usually do to action, at least in the modern free world: any action is by default justified, until it can be shown somehow wrong. We’re not obligated to do nothing at all except those things that we can prove from the ground up that we must do. We would normally consider than an absurd standard for justifying our actions, but it’s all too common to apply that standard to justifying our beliefs.

No. You always were mistaken about one if these things, they merely exhaust the set. If you've narrowed it, what was the possibility you've eliminated? — Isaac

You’ve narrowed the possibilities you’re aware of being possible by realizing that certain combinations of things are not possible. They were always not possible, sure, but we’re talking about your awareness of the possibilities.

No, I don't get it. — Isaac

See the above analogy to justification of actions, I think that will clear it up.

Isaac

You’ve narrowed the possibilities you’re aware of being possible by realizing that certain combinations of things are not possible. They were always not possible, sure, but we’re talking about your awareness of the possibilities. — Pfhorrest

But you haven't 'realised' certain combinations of things are not possible, you just believe it to be the case. I'm not seeing how that makes any difference to the status of any one of those beliefs.

See the above analogy to justification of actions, I think that will clear it up. — Pfhorrest

That basically opposes the idea that beliefs need to be justified from the ground up. Fine, i'm on board with that. That's only half the claim you've made. The other half is that a mere belief becomes something more than that when we also believe there is a lack of "reason to rule that belief out". It's this second half I don't get. Why does this second belief make any difference to the status of the first.

Pfhorrest

mere belief becomes something more than that — Isaac

I think maybe this is the point of confusion. I’m not talking about transforming beliefs into anything else, but just when a belief is or isn’t justified, or warranted.

Isaac

I think maybe this is the point of confusion. I’m not talking about transforming beliefs into anything else, but just when a belief is or isn’t justified, or warranted. — Pfhorrest

That is something else, though. It acquires this status 'justified' simply by virtue of there being another belief which references it. How does there being another belief about our first one change the functional status of the first belief in any way at all?

Take my example of believing there are unicorns in my back garden. That's just a belief. Then I also believe there's no good reason not to believe there's unicorns in my back garden. My first belief is now a 'justified' belief. What's changed about it to warrant the new name?

Pfhorrest

It acquires this status 'justified' simply by virtue of there being another belief which references it. — Isaac

Not on my account. On my account it's justified by default. Justification is the initial status of a belief, and it can only get worse from there, not better. That's the whole point of critical rationalism. The way you're interpreting it is equivalent to the usual justificationism, and the problems you're pointing out are exactly the problems I have with justificationism.

Take my example of believing there are unicorns in my back garden. That's just a belief. Then I also believe there's no good reason not to believe there's unicorns in my back garden. My — Isaac

It's not a matter of whether you believe there are no good reasons not to believe there's unicorns in your back garden, it's a matter of whether or not there are good reasons, and whether your belief would be changed by the presence of them.

This does raise a sort of meta question, of whether we can ever know whether we actually know anything. But even if we can't ever be sure that we do know things, it's still nevertheless possible that we do know things. This thread is about my account of in which circumstances someone does actually know something, whether or not they or we can know for sure that they know it. If they believe it, and they would disbelieve it were there actually reasons to do so (whether or not we or they can be sure there are not such reasons), then they know it.

Isaac

↪Pfhorrest

But once you've got away from actually knowing and into ways the world actually is you've removed the need for justification altogether. Why use (or even introduce) the actual state of the world {there being no reason not to believe there are unicorns in my back garden}. Why not just the state if the world {there being unicorns in my back garden}?

"It is justified to believe there are unicorns in my back garden if the state of the world is such that there's no reason in it not to believe that" seems absolutely no improvement on saying "It is justified to believe there are unicorns in my back garden if the state of the world is such that there are unicorns in my back garden".

I don't get why you'd choose the former over the latter. If you're prepared to talk about actual states of the world, why not just talk about the state of the world which the belief is about, rather than the state of the world wherein it is absent of reasons not to believe the belief in question?

So...

This thread is about my account of in which circumstances someone does actually know something, whether or not they or we can know for sure that they know it. — Pfhorrest

...they know A if A is in fact the state of the world, they don't know it if it isn't. What's wrong with that?

Pfhorrest

That would be saying that knowledge is simply true belief, which then leaves open the problem of people who form beliefs in a way not responsive to the state of the world but their beliefs just happen to coincide with reality.

(Which is really the problem at the root of Gettier problems, too).

Phrasing it in terms of the counterfactual clarifies that the belief has to be responsive to reality to remain justified. If you believe unicorns are in your back yard and would continue believing that despite evidence to the contrary, but as it so happens there are unicorns in your back yard, you didn’t really know that. If you would be responsive to evidence to the contrary, and there just isn’t evidence to the contrary because your belief is correct, then you know something.

Isaac

If you believe unicorns are in your back yard and would continue believing that despite evidence to the contrary, but as it so happens there are unicorns in your back yard, you didn’t really know that. — Pfhorrest

1. Why not? And 2. Who on earth believes something despite also believing there's evidence to the contrary (sufficient to counter that belief)? A few forms of severe mental illness might do this but I can't think of any real cases off the top of my head.

If you would be responsive to evidence to the contrary, and there just isn’t evidence to the contrary because your belief is correct, then you know something. — Pfhorrest

This just describes everyone. You're not distinguishing one group of beliefs from another here. People do not just spawn random beliefs out of thin air, they believe them because they have some reason to. If they persist despite evidence to the contrary its because they find that evidence insufficiently compelling on the basis of some other belief.

What it sounds like you're describing is people who persist in their beliefs despite evidence which you find sufficiently compelling, and it sounds a lot like you just want to find ground on which to diminish those beliefs in the light of your judgement of the evidence.

Kenosha Kid

No more so than traditional JTB. Really the whole “truth” component of both traditional and my modified JTB is a historical vestige that’s rather redundant. Knowledge is merely justified belief, where justification itself implies a reason to think it is true; we only bother saying “justified TRUE belief” because before the justification criterion was added, the standard was simply “true belief”. It would have been better if the “true” had simply been replaced by “justified”. — Pfhorrest

It was more that a justified belief is knowledge if the belief is determined to be true e.g. by observation, deduction, etc. The part I quoted made knowledge of truth a prerequisite of a justified belief instead. It seems circular to define knowledge in terms of a belief in something because it is true. If I already know it is true, belief is irrelevant.

Pfhorrest

Why not? — Isaac

Because guessing (or otherwise non-responsive-to-reality belief formation) and just happening to be right doesn’t rise to the standards that we hold knowledge to. Knowledge is supposed to be some kind of important relationship between belief and reality, not just coincidence. This is, again, the root of the whole Gettier problem.

Who on earth believes something despite also believing there's evidence to the contrary (sufficient to counter that belief)? — Isaac

Once again we need to distinguish between believing there is good evidence to the contrary and there actually being good evidence to the contrary. If we don’t, then we have to concede that every belief anyone ever has is equally justified, i.e. there is no such thing as epistemic error, because as you say, everyone THINKS they have good reason to believe as they do, but often they don’t.

determined to be true e.g. by observation, deduction, etc. — Kenosha Kid

That just is justification.

Anyway, see above with Isaac wrt the difference between thinking something is true (or reasons are good) and it actually being so.

TheMadFool

↪Isaac

I'm not sure what you mean here. Deduction is, to my reckoning, the relationship between premises and conclusion such that the latter follows from/is a consequence of the former. It differs from consistency in that two propositions maybe consistent but don't constitute a deductive argument.

Isaac

Once again we need to distinguish between believing there is good evidence to the contrary and there actually being good evidence to the contrary. If we don’t, then we have to concede that every belief anyone ever has is equally justified, i.e. there is no such thing as epistemic error, because as you say, everyone THINKS they have good reason to believe as they do, but often they don’t. — Pfhorrest

I wasn't asking about the motivation. I'm sure you're desperate to distinguish these things (I'm less bothered myself). But motives are beside the point. If you can tell what the state of the world actually is (that it actually contains or doesn't contain evidence for a belief), then justifying the belief by this method becomes completely unnecessary. Why would we be at all concerned whether our beliefs were justified or not, we can just directly check whether they're actually the case or not. Justification ceases to be of any importance.

The whole point of justified beliefs is that they're more likely to be true than less justified ones. The use of justification is premised on the fact that we don't have direct access to the way the world is and so have do deal with more or less likely beliefs about its state. You can't then propose a method of justification which relies on us knowing directly the way the world actually is (whether it does or doesn't contain evidence for our belief). We need a method of justification which concedes the same premise as motivated us to want one in the first place.

Echarmion

I say that knowledge is believing something because it is true, such that not only does one believe it, and it is true, but if it weren't true one wouldn't believe it. — Pfhorrest

Eliezer Yudkowski has said something similar, defining knowledge as "the ability to be more confused by fiction than by reality". If you can equally explain every outcome, you know nothing. This links knowledge with the concepts of information and entropy.

It also, however, limits knowledge to the physical. Knowledge can then only be gained about things that are falsifiable, i.e. subject to a prediction. You couldn't know anything about morality, for example.

Namely, rather than the usual justificationist sense of rationalism, whereby no belief is justified until it can be supported from the ground up somehow, instead any belief is justified (including contrary ones) until there is support to the contrary, i.e. reason to rule that belief out -- an epistemological position called critical rationalism, supported by philosophers like Kant and Popper. — Pfhorrest

It would then presumably follow that only those beliefs can be considered knowledge that have no justified contrary beliefs, i.e. all contrary beliefs are rules out. But, if we insist on some objective notion of truth, a belief can be true before we are able to find significant arguments to rule out contrary beliefs. We'd then have to conclude we have knowledge of something even though we are similarly believing contrary things about it. That doesn't sound very useful.

Isaac

I'm not sure what you mean here. Deduction is, to my reckoning, the relationship between premises and conclusion such that the latter follows from/is a consequence of the former. It differs from consistency in that two propositions maybe consistent but don't constitute a deductive argument. — TheMadFool

Yeah, that's my understanding of it too. I wasn't questioning your understanding of deduction I was questioning it's application to knowledge about the state of the world. This bit...

We needn't worry about deductive arguments as they're foolproof justifications — TheMadFool

They're not foolproof justifications, they don't help at all. Let's take an unknown state A and an inference about another state B, which I'll call 'b. If we have the logical relationship 'if B then A' about the state of the world, we can deductively conclude A because B. But by definition we do not know B either, we only know 'b. If we could directly know B we wouldn't be trying to deduce A we'd just directly experience it.

Now imagine I have a belief ~A, I also have a belief 'if B then A, and a belief B. I carry out the deduction 'if B then A, B, therefore A'. But this clashes with my belief ~A. So I carry out the deduction 'if B then A, ~A, therefore ~B. But this clashes with my belief B so I carry out the deduction... and so on. We've no way of knowing whether my belief ~A counts as a theoretical conclusion (and so should be discarded by the first deduction), or as a premise (and so B should be discarded by the second deduction).

The definition of knowledge under critical rationalism

Welcome to The Philosophy Forum!

Categories

More Discussions