JTB: the myth of propositions and the Gettier problem

[Peter Holmes]

When Gettier cases demonstrate that some examples of justified true belief don't constitute knowledge, this exposes the mistake of describing knowledge as justified true belief - and particularly the mistaken truth condition: S knows that p iff p is true.

The main condition for our knowing something is the case is that it is the case - 'objective' knowledge that we Gettier-spectators uncontroversially possess. And the truth-value of a factual assertion has nothing to do with that. (Knowledge by description comes from acquaintance with a description whose truth-value is assumed and not at issue in a Gettier case.)

The JTB definition of knowledge demonstrates the myth of propositions at work: mistaking what's said about a state-of-affairs for the state-of-affairs, as though reality consists of true propositions - and not-reality of false ones. It's an ancient and powerful delusion.

[Terrapin Station]

His believe is NOT just that Jim is in the restaurant. His belief is that a specific x is in the restaurant and x is Jim. Again, it's an illegitimate substitution to say that Bill has a belief that "Jim is in the restaurant" where we're not talking about the person that Bill actually observed.

The person Bill actually observes is Jim's twin brother Tim; but Bill has no belief about Tim, since he doesn't even know Tim exists. He has a belief about the man he sees—the belief that the man he sees is Jim—, but he has no belief about Tim despite its being the case that the man Bill sees is in fact Tim and not Jim. There is a man in the restaurant (= Tim) of whom Bill believes that he is Jim, but there is no man in the restaurant of whom Bill believes that he is Tim, although the man in the restaurant of whom Bill believes that he is Jim is Tim.

This is an intensional context where Tim = the man Bill sees, but "Bill believes that Tim is Jim" is false and "Bill believes that the man he sees is Jim" is true. So the distinction between de dicto belief and de re belief seems relevant here:

https://plato.stanford.edu/entries/beli ... ctBeliAttr

This has nothing to do with what I was saying.

[Consul]

This has nothing to do with what I was saying.

What do you think has something to do with what you were saying!

[Terrapin Station]

This has nothing to do with what I was saying.

What do you think has something to do with what you were saying!

Do you understand why I'm saying that the Gettier example here (and in many, maybe most cases) rests on an "illegitimate substitution"?

[Consul]

When Gettier cases demonstrate that some examples of justified true belief don't constitute knowledge, this exposes the mistake of describing knowledge as justified true belief - and particularly the mistaken truth condition: S knows that p iff p is true.

The main condition for our knowing something is the case is that it is the case - 'objective' knowledge that we Gettier-spectators uncontroversially possess. And the truth-value of a factual assertion has nothing to do with that. (Knowledge by description comes from acquaintance with a description whose truth-value is assumed and not at issue in a Gettier case.)

The JTB definition of knowledge demonstrates the myth of propositions at work: mistaking what's said about a state-of-affairs for the state-of-affairs, as though reality consists of true propositions - and not-reality of false ones. It's an ancient and powerful delusion.

The JTB definition can be reformulated without any reference to (abstract) propositions:

(Let "s" stand for "state of affairs"!)

S knows that s iff
1. s is the case/s is a fact;
2. S believes that s;
3. S is justified in believing that s.

Then, a belief is true iff the believed state of affairs is the case/a fact. To believe truly is to believe what is the case/a fact. To believe falsely is to believe what is not the case/a nonfact.

(By "fact" I mean an actual, existing/obtaining state of affairs, and not a true proposition.)

[Consul]

Do you understand why I'm saying that the Gettier example here (and in many, maybe most cases) rests on an "illegitimate substitution"?

There is no "illegitimate substitution", because Bill's (true) belief that Jim is in the restaurant results directly from his (false) belief that the man he sees in the restaurant is Jim.

[Terrapin Station]

Do you understand why I'm saying that the Gettier example here (and in many, maybe most cases) rests on an "illegitimate substitution"?

There is no "illegitimate substitution", because Bill's (true) belief that Jim is in the restaurant results directly from his (false) belief that the man he sees in the restaurant is Jim.

I'm not asking if you agree with me. I'm asking if you understand why I'm saying what I'm saying.

[Consul]

Do you have anything to say on my response to the specific case presented by Gettier?

See: viewtopic.php?p=387799#p387799

[Consul]

If you aren't convinced by Gettier's original examples, maybe the following one will convince you of the insufficiency of the JTB definition of "knowledge", which I just thought up (and which is analogous to the sheep-in-the-meadow case):

Bill looks through the window of a restaurant and sees a man sitting at a table who looks exactly like his friend Jim; so he believes that Jim is in that restaurant. However, the man he sees isn't Jim but Jim's identical twin brother Tim, whose existence is unknown to Bill, because he never met Tim and (for some strange reason) Jim never told him that he has an identical twin brother. As it happens, Jim is in the restaurant too, sitting at another table that Bill cannot see through the window. So Bill's belief that Jim is in the restaurant is both justified by his seeing someone who looks exactly like Jim and true owing to Jim's presence in the restaurant. But does he really know that Jim is in the restaurant? No, he arguably doesn't; and if he doesn't, then justified true belief isn't the same as knowledge.

The JTB definition requires that p is true, Bill believes that p, and Bill is justified in believing that p. But what is p in this case? It is not that Jim is in that restaurant, but that the man BIll is seeing is Jim, which is not true, therefore even though Bill seems justified in believing that the man he is seeing is Jim, this does not constitute knowledge of Jim being the person he is seeing. In a second instance, Bill's false belief creates the conditions for having another belief: that Jim is in the restaurant, but evidently this means the person that Bill is confusing as Jim being seen in the restaurant, which is false. The fact that the real Jim (not being seen by Bill) is by mere chance in the restaurant, unknown to Bill, does not constitute for Bill knowledge of Jim being in the restaurant.

The relevant belief is Bill's belief that Jim is in the restaurant. I stipulated in my imaginary example that this is Bill's justified true belief which doesn't seem to equal knowledge. (Bill is an unreal person, so we cannot discover what he really believes, since unreal persons don't really believe anything.)

1. Bill believes justifiedly (but falsely) that the man he sees in the restaurant is Jim (because that man looks exactly like Jim).
2. So Bill believes justifiedly that Jim is in the restaurant.
3. Bill also believes truly that Jim is in the restaurant (because Jim is in the restaurant).
4. Therefore, Bill believes both justifiedly and truly that Jim is in the restaurant, and yet he doesn't know he is.

[Peter Holmes]

The JTB definition can be reformulated without any reference to (abstract) propositions:

(Let "s" stand for "state of affairs"!)

S knows that s iff
1. s is the case/s is a fact;
2. S believes that s;
3. S is justified in believing that s.

Then, a belief is true iff the believed state of affairs is the case/a fact. To believe truly is to believe what is the case/a fact. To believe falsely is to believe what is not the case/a nonfact.

(By "fact" I mean an actual, existing/obtaining state of affairs, and not a true proposition.)

I think your account of what we call knowledge - or knowing that something is the case - is interesting. But I'd like to run some suggestions by you.

1 Perhaps the opening should be 'S knows that s is or was the case iff...'
2 That would make 1: 's is or was the case'.
3 Then 2 could be: 'S believes that s is or was the case'.
4 And 3 could be: 'S is justified in believing that s is or was the case'.

But, on reflection, I'm wondering why belief and justification have to be conditions for knowing something is or was the case. (And, as you know, justification has itself been much debated in the JTB literature.)

I have the feeling that this need to provide so-called conditions for what we call knowledge comes from a metaphysical delusion. Words such as 'knowledge' and 'knowing' and 'know' can mean only what we use them to mean. So all we can do is explain the ways we use them.

More thought needed. Thanks again.

[Terrapin Station]

1. Bill believes justifiedly (but falsely) that the man he sees in the restaurant is Jim (because that man looks exactly like Jim).
2. So Bill believes justifiedly that Jim is in the restaurant.

If (2) isn't referring to Bill's belief that the man he sees is Jim, it's an illegitimate substitution, because that's not how belief actually works.

[Consul]

1. Bill believes justifiedly (but falsely) that the man he sees in the restaurant is Jim (because that man looks exactly like Jim).
2. So Bill believes justifiedly that Jim is in the restaurant.

If (2) isn't referring to Bill's belief that the man he sees is Jim, it's an illegitimate substitution, because that's not how belief actually works.

It's incoherent to say that Bill believes the man in the restaurant he sees is Jim, but he doesn't believe Jim is in the restaurant.

[Terrapin Station]

1. Bill believes justifiedly (but falsely) that the man he sees in the restaurant is Jim (because that man looks exactly like Jim).
2. So Bill believes justifiedly that Jim is in the restaurant.

If (2) isn't referring to Bill's belief that the man he sees is Jim, it's an illegitimate substitution, because that's not how belief actually works.

It's incoherent to say that Bill believes the man in the restaurant he sees is Jim, but he doesn't believe Jim is in the restaurant.

It's not saying that he doesn't believe that Jim is in the restaurant. It's that it's necessary to qualify "Jim is in the restaurant," because Bill doesn't believe that unqualified. He believes that "The man I see, who is Jim, is in the restaurant," whether it's stated that way or not. He doesn't believe something less than that, and he doesn't believe something more generalized than that.

[Consul]

It's not saying that he doesn't believe that Jim is in the restaurant. It's that it's necessary to qualify "Jim is in the restaurant," because Bill doesn't believe that unqualified. He believes that "The man I see, who is Jim, is in the restaurant," whether it's stated that way or not. He doesn't believe something less than that, and he doesn't believe something more generalized than that.

Again, there's nothing general about the belief that Jim is in restaurant, since it's a belief about one particular person and no one else.
Bill's belief "Jim is the man I see in the restaurant" or (equivalently) "The man I see in the restaurant is Jim" is false, but his belief "Jim is in the restaurant", which is derived from the former belief, happens to be true—and it stands as it is, no "qualification" is needed.

[Terrapin Station]

Again, there's nothing general about the belief that Jim is in restaurant, since it's a belief about one particular person and no one else.

There's something general about it in the sense I mean, in this context, if it's not referring to "The person I see, who is Jim."