A criticism of Knowledge Justified True Belief

This occured to me the other day when considering the generally acecpted Definition of Knowledge as Justified True Belief. (Ignoring the already known Gettier Example)

I know X

Therefore:

  1. I believe that X
  2. I am justified in believing that X
  3. X

Now as you can see part (1) and (2) are mental statements whereas (3) is an existentual one.

When considering how do we know that we know that X?
We can respond, I have a belief (1), It is justified (2), It is true (3)

But how do you know X is true?
Well I have a belief (1), It is justified (2), It is true (3)

But how do you know it is true?

I believe there is an infinite regress with subjective awareness of knowledge, therefore under this definition we can in fact never know anything.

Additionally

Ascribing knowledge objectively is again problematic

Bill knows that Y

Therfore:

Bill believes that Y
Bill is justified to believe that Y
Y

Unfortunately how do we know that Y?
This just reverts to the problem of infintie regress previously demonstrated for subjective knowledge.

As you point out, truth, justification, and belief, are the necessary conditions of knowing. Gettier has argued that although they are necessary conditions, they are not sufficient conditions. The jury seems to be still out on this.

Now, you object that putting the Gettier objection aside, there is something the matter with the truth condition because we don’t (or cannot) know that the truth condition is satisfied.

But now, things begin to get a little muddy. Suppose I answer that of course we can know the truth condition is satisfied if we have evidence that it is. Let’s just take an example: I (think) I know that Quito is the capital of Ecuador, since I believe it is; I have ample justification that it is; and, it is. Now you ask, “But how do you know it is the capital of Ecuador?” Well my reply is to turn to my justification. It says so in the latest Atlas; it says so in the World Almanac; and (although this is a lie) I have even phoned the Ecuadorean Embassy in Washington, D.C. and the person who answered the phone assured me that Quito is the capital. So, that is how I know it is true that Quito is the capital.

Now, maybe (although I don’t think so) your objection is just the usual skeptical objection that all this evidence is not justification enough since even given the evidence if might still be false that Quito is the capital. I won’t go into this objection, since I don’t think that is really your objection. If it is, let me know.

I think your objection is different. It is that even if all three of the conditions of knowing are satisfied we do not (or cannot) know that they have been satisfied. And, of course, in particular, that the truth condition has been satisfied. So your objection seems to be that even if the truth-condition (and the others) are, in fact, satisfied, we cannot know that they are satisfied, and if we cannot know they are satisfied, that means (and this is obscure) that they are “really” not satisfied.

But, now, by question for you is this: why do you think that unless we know that the conditions are satisfied, they are “really” not satisfied. After all, if they are satisfied, then they are satisfied (tautology) whether or not we know they are. To give an analogy, you know that in the game of chess, the conditions of checkmate are: the king is in check, and the king has no legal move. Now, suppose my king satisfies these conditions. Could I argue that I am not “really” in checkmate because I do not (or cannot) know those conditions are satisfied? Obviously not. The conditions, if satisfied, are satisfied, whether or not I know they are satisfied. Why, then, should the case of knowing be any different? If it is true that Quito is the capital, then the truth condition of “Quito is the capital of Ecuador” is satisfied, and knowing it is satisfied is irrelevant.

I think your “philosophical cramp” (as maybe Wittgenstein would have put it) is that you (perhaps in the back of your mind) accept the “KK principle” which is the principle that a necessary condition from knowing that P is knowing you know that P. But as Spinoza pointed out a long time ago, the KK principle implies a contradiction. For, as he argued, far from its being the case that in order to know, you have to know that you know, the truth is that in order to know that you know, you have first to know. So the KK principle implies that knowing you know is a necessary condition of knowing, and that is self-contradictory.

I don’t want to go on at length. But I will remark here that I think there are two motivations behind this view (which I think is not only false, but self-contradictory) that knowing entails knowing you know. The first is the old belief (I think bugaboo) that you cannot know unless you are certain, and that knowing you know confers certainty. (And this, itself, goes back to the skeptical objection I mentioned earlier). But the other motivation is, I think, even stronger. It is that unless you know that you know (and more particularly, unless you know that the truth condition is satisfied) it is illegitimate to claim or to say that you know. And to this, I have two replies, and then I will stop. 1. It is not true that it is illegitimate to claim you know unless you know you know. If “illegitimate” means anything here, it is clearly enough to believe you know to make it “legitimate” to claim that you know. My second reply is this: WE are now talking about the conditions not of knowing, but the conditions of claiming one knows. And that is a different issue.

I hope this helps in this tangled web your post brings up.

I covered this issue in “Fun with modalities.”

The problem you’re running into is that the term “know” is referentially opaque.

Just how does that work? If I know that some proposition. P, is true, and if P is intensionally equivalent to Q, then do I not know that Q is true? If I know that New York City is east of Chicago, then do I not also know that Chicago is west of New York City?

How do we know that Bill knows that Y? We have no way of ascertaining that Bill actually believes that Y, as his thoughts are private and inaccessible to us. Bill’s expression of belief does not have adequate weight, as he could be lying, or he could just falsely believe that he believes Y.

Troy:Would you please post a link to your discussion of modalities.

If we allow justification to be mental, then that removes the necessary connection to an objective, observable world. Therefore, justification must come from a publicly verifiable source, or from a publicly recognized authority.

Of course, we would get stuck in infinite regresses otherwise, we’d have to know we know we know etc

Well I dunno about you but I would like to be able to say “I know x” or “He knows x” without risking that it might be wrong by some ontological fluke.

Probably not actually

“I know that the morning star is about 700 million miles from the earth”
-/>
“I know that the planet venus is 700 million miles from the earth”

Knowledge (at least the belief part) is a genuinely intensional issue.

Yuxia,

“I would like to be able to say “I know x” or “He knows x””

‘To know’ is only to experience surety, and never anything more than that. All attempts to define it beyond that experience, actually take one further from what knowing is, and not closer. Because it is fundamentally an experience, it has no ontological foundation other than in the sense that ontology, that is ‘is-ness’, is that which is experienced.

Dunamis

That’s my Dunamis!!!

Oh boy, you guys are in for it now.

ah Dunamis, Descartes reincarnate.

So Dunamis are you willing to accept that we can know things that are false?

Cause to me that is quite absurd surely if I am to say “I know that the world is flat” you will respond “No you don’t because the world isn’t flat, you simply believe it is”

So I think the way in which we use the word “know” does imply ontological reference.

Yuxia,

I do not accept Cartesian Dualism.

“So Dunamis are you willing to accept that we can know things that are false?”

We can experience surety in something that will prove to be false, i.e. in something that from another perspective, thinking such to be so, will not produce the experience of surety. When something is experienced as to not produce the experience of surety, that becomes an experience of surety. As animals we are always orienting ourselves.

So I would accept that we can experience the surety of that things are false. But this is only an experience.

“So I think the way in which we use the word “know” does imply ontological reference.”

It certainly implies ontological reference, but implication does not prove nor guarantee. It really assumes ontological reference. This is the game of “knowing”. The body-as-mind assumes that the body-as-body reflects states of the world “beyond” it. The experience of this frequent correspondence is one of the fundamental experiences of surety that human beings have, but it still is an assumption.

Dunamis

The issue is not how, or even whether, we know that A knows that Y. The issue is under what conditions is it true that A knows that Y. Of course we cannot know for certain that A knows that Y no more than we can know for certain that Quito is the capital of Ecuador. We are fallible. In the same way, we cannot know for certain that checkmate has occurred. (We may be dreaming). But what has that to do with whether the truth-conditions for checkmate are" the king is in check, and the king has no legal moves. And what has the fact that we don’t know for certain whether A knows that Y, that the necessary conditions of knowing are true, justified, belief?

As I pointed out, no more than it is necessary to know that the conditions of checkmate are satisfied for the conditions to be satisfied, is it necessary to know that the conditions for knowing are satisfied for them to be satisfied.

Your objection is irrelevant.

I know with certainty that Dunamis is your name.
I know with certainty that Dunamis is not your name.

Yada,

You experience surety in each.

Dunamis

That implies that it is both true and not true that Dunamis is [his] name. That is a contradiction. So the conjunction: “I know with certainty that D is your name; and, I know with certainty that D is not your name” must be false.

Ken:

Please excuse my interuption, but I’d like to know more about this. I do not see the contradiction here.

Dunamis,

It is believed that there are billions of galaxies in the universe.

In the case of public knowledge, as it appears in a textbook, for me to know, it is enough if I can correctly recall ‘facts’, I do not have to subscribe to (believe in) that knowledge.

So it appears important to examine which type of knowledge we are talking about.

Dunamis, and other names you use, are all references to the same ontological being, you. The names are true references, as you expressly define them, and only as long as you maintain each reference.

Quito is Equador’s capitol by public definition of a reference. A batchelor is a single man by public dictionary definition. The Earth is empirically truly flat by anyone’s everyday observation.

How would you treat these?

Yada,

“How would you treat these?”

I treat all of these as experiences of surety. The subject is oriented within various discourses and experiences the surety of that orientation.

Dunamis

So, a three-year-old knows everything he repeats?

Yada,

I’m not sure I actually understand the question.
Yes, 3-year olds experience surety, as do dogs and dolphins. The enunciation of a “truth” by a 3 year old would only be “knowledge” to the degree that he or she experiences surety, (which may be only the surety of being conscripted into the adult world as a speaker). The degree to which a 3 year old can enter into discourses and find orientation within them would be to a large extent the degree of his or her “knowledge”, which again would be his or her experience of surety.

Dunamis