Jump to content
Objectivism Online Forum

Gettier

Rate this topic


Recommended Posts

I hope a few people here are aware of Gettier's problems with Justified True Belief (JTB). We're looking at it in our Epistemology seminars, and I thought I had a response to it, but it turns out I may have made a mistake. I'll summarise what I said, and then post what I originally said.

Basically: something is true only once it has been identified as so. A 'fact' describes the relationship between our mind and reality - namely that our the a proposition in our mind reflects something in reality. This relationship is justification. In the "Justified True Belief", the word "true" is redundant, because it is implied by "justification".

But I read this thread where D.O. and Dan Edge discuss the Correspondence Theory, and D.O. makes a few marks about Gettier, and states an objection to Dan Edge that I don't quite understand, but which basically mean that truth does not come about through justification, and that 'false knowledge' is possible.

I was wondering if anyone can weigh in here on Gettier's problems, or if D.O or Dan Edge would care to say something, because I'm quite stuck on this issue.

And my statement on this issue in full (at least, the conclusion):

"Gettier's objections are perfectly sound given the premises of analytic logic, and it shows why it is just so insane, and why it really is just a modern day version of Platonism, where we treat true knowledge as some perfect form, which exists outside ourselves.

Gettier's claims are valid, so long as we accept that one can make some sort of "logically true", justified argument, without knowing the truth (without conforming to reality) - and so long as we accept that something can also be true without justification (that "objective truth" is some literal thing, out in reality, not requiring any kind of recognition on the part of the seeker for it to be "true")

However, I should note, that this does not mean that Justified True Belief is at all wrong, of course, except that the word 'true' is redundant. As O'ists, we hold that Truth and Justification are one and the same, and cannot exist separate of one another. All this Gettier problem shows is that abstracting "truth" from "justification" results in the erosion of any justifications, and results in the search for truth in something outside of what can be "purely reasoned" (thank you, Kant!)."

Link to comment
Share on other sites

Basically [for Gettier's problems with Justified True Belief (JTB)]: something is true only once it has been identified as so. A 'fact' describes the relationship between our mind and reality - namely that our the a proposition in our mind reflects something in reality. This relationship is justification. In the "Justified True Belief", the word "true" is redundant, because it is implied by "justification".

I'm not familiar with Gettier, but I would say his formulation as you state it is incorrect. In a primacy of existence approach, a fact is a fact whether a consciousness recognizes that fact or not. As an example, it is a fact that the Earth has an iron core -- and it was a fact before anyone discovered it. Facts exist apart from consciousness. In other words, existence exists even if no one is aware of it.

Frankly, I didn't find that thread helpful with regard to this particular issue, though it was an interesting discussion.

Link to comment
Share on other sites

David - reading that section, when she approaches the end, she says:

"What have I added to the term "reality" by saying "facts"? I have narrowed it. I have said: whichever aspects, events, or existents you happen to know, these are the facts of reality -- meaning: these are the things which actually exist." (Emphasis mine; for convenience)

Am I right to draw from this, that we can possess knowledge, of which we can be justifiably certain, but which turns out to be false? Does it turn out to be false in the face of facts? And, given that, are facts knowledge which can never be false?

I draw this last question from the statement by 'Prof B' which Rand affirms: "'Fact' designates existents, but it is used in a context in which it is relevant to distinguish knowledge from error". This statement implies that facts are absolutely true, but that we can have knowledge which is false (and a quick question - since you raised the "crow" issue in that other thread: what do we say about knowledge at the time when we are only aware of black crows? Can we saw we certainly and truthfully know that only black crows exist?).

Edited by Tenure
Link to comment
Share on other sites

David - reading that section, when she approaches the end, she says:

"What have I added to the term "reality" by saying "facts"? I have narrowed it. I have said: whichever aspects, events, or existents you happen to know, these are the facts of reality -- meaning: these are the things which actually exist." (Emphasis mine; for convenience)

Am I right to draw from this, that we can possess knowledge, of which we can be justifiably certain, but which turns out to be false? Does it turn out to be false in the face of facts? And, given that, are facts knowledge which can never be false?

I draw this last question from the statement by 'Prof B' which Rand affirms: "'Fact' designates existents, but it is used in a context in which it is relevant to distinguish knowledge from error". This statement implies that facts are absolutely true, but that we can have knowledge which is false (and a quick question - since you raised the "crow" issue in that other thread: what do we say about knowledge at the time when we are only aware of black crows? Can we saw we certainly and truthfully know that only black crows exist?).

The two problems of knowledge are fallibility and non-omniscience. The conceptual faculty is not infallible, the perceptual is. Conclusions that are the result of inductive reasoning are subject to revision due to expanding context, but not being omniscient is not a kind of error. Thus the issue of certainty does not apply to all knowledge; certainty applies only to some complex inductive knowledge.

Regarding the crows question, that is about induction. Dr. Peikoff's solution to the problem of induction is causality. You are certain that only black crows exists only when you know why all crows are black and any other color is an impossible contradiction. Lacking that explanation, you have no epistemological right to claim all crows are black, even if that is all you ever see.

(Usually swans are used for this problem, and crows are reserved for counting. How weird that this actually set me back a few seconds.)

Link to comment
Share on other sites

Ah! I only listened to the first few of those lectures in induction back last year. They're hazy in my memory, and I had no idea what the 'problem of induction' was, so it never really integrated properly. I knew he said *something* about the crow/swan thing, I just didn't know what it was. I thought it was something like the response to Descartes' doubt thing, "There's no reason to consider the arbitrary... unless you have reason; in which case, it wouldn't be arbitrary anyway". But, in fact, it seems what Peikoff was saying in this case was that this isn't about considering the arbitrary, but actually a matter of proper induction (and specifically, that enumerative induction is not good in and of itself).

Sorry if that seems I'm just saying back to you what you just said. I just find it helps cements a new bit of knowledge (and, if I make an error in understanding, it means someone can point it out to me).

So, regarding your first point about certainty: yes, indeed! And I was thinking this in class today. The professor was talking about Gettier's problem, and that it runs (in part) on this assumption: that it is impossible to have false knowledge.

And to clarify what this meant, he stated that a belief cannot both be justified (where justification could be the most strictest, most stringent kind of induction or deduction), and then turn out to be false. What this would lead to is a state of pure skepticism, where all *supposedly* justified claims are, potentially, false. At best, it means all things are probable, but not quite definite.

However, if I'm getting what you say here, this does not necessarily have to lead to skepticism, because, by the very nature of my discovering I am wrong: if I am capable of making justified claims which turn out to be false, I must have some infallible method by which to verify claims.

To sum up: I can have knowledge, of which I can be reasonably certain (which I can have no reason to doubt, but which is still fallible). I also have a method of verification, of which I am absolutely certain (which I have no reason to doubt because it is infallible).

As a consequence, knowledge is 'justified belief' which I have every reason to believe to be true -- and if it turns out that my justified belief is not true, it does not invalidate my method of knowing truth, it just means that I am not omniscient and not infallible.

Link to comment
Share on other sites

I draw this last question from the statement by 'Prof B' which Rand affirms: "'Fact' designates existents, but it is used in a context in which it is relevant to distinguish knowledge from error". This statement implies that facts are absolutely true, but that we can have knowledge which is false (and a quick question - since you raised the "crow" issue in that other thread: what do we say about knowledge at the time when we are only aware of black crows? Can we saw we certainly and truthfully know that only black crows exist?).

I'll have to re-read that section of ITOE, but I think you might be misunderstanding Miss Rand's point. It is not as if facts are always true, because it is not a true / false issue in this context. Facts are that which exists -- the raw data, so to speak -- and one can say those facts are there or they are not there, but it wouldn't be true or false. Also, one has to distinguish facts from mental processing, such as theories, as the starting point of knowledge.

For example, say a crime was committed and there is a certain body of evidence. That body of evidence are the facts -- how they are put together mentally in order to determine who is guilty of the crime is theoretical, and there can be many ways of putting the facts together mentally -- different ways of trying to integrate the facts to decide who committed the crime. The theory is not a fact, the evidence is the facts. Say one theorizes that Joe did the crime, based on the evidence (the facts), and the evidence does point that way, but you have no direct evidence, then you cannot say that Joe definitely committed the crime. For issues of rational certainly, new facts coming into the scene may be used to contradict the previous conclusion -- say he had an alibi, a fact, that makes it impossible for him to have committed the crime. But in all of this speculation as to who did it, the facts are the facts, and the processing to a conclusion are not a fact. One's mind corresponding to the facts of existence is what gives issues such as true or false, but the fact simple exist and are not true or false.

I don't know that one can have false knowledge, as I think that is a contradiction. If knowledge is rationally based upon the facts, I think you can only be wrong if you don't take the right facts into account -- say because they are unknown -- but what you rationally integrated from the facts that you do have could still be true knowledge for those known facts (like Newton versus Einstein). I think we would need some examples of what you mean by false knowledge so that it is not just a floating abstraction.

Link to comment
Share on other sites

In the "Justified True Belief", the word "true" is redundant, because it is implied by "justification".
There is a universe where I can accept this: it's one where "justification" requires following a logical method that cannot produce error. On the face of it, justification simply means "if you can come up with a story; an excuse". If you limit justification such that improperly generalized statements are not justified, then we're on the road to a theory where justified belief is necessarily true. (That is, the evidence directly proves a restricted statement A but not a broader A' -- then only A is justified, not A'). Additionally, for universal statements, the proof requires knowledge and proof of causation. Often, "justification" is not a full logical validation, is is simply the showing of a preponderance of evidence.
but which basically mean that truth does not come about through justification, and that 'false knowledge' is possible.
If I knew exactly what you were referring to, I could do better than just guess. So let me guess. The condition where [you grasp that a proposition describes a fact]1 is only possible if [you have logically validated the proposition]2. And clause 1 is "truth", clause 2 is "justification". Truth and justification are not the same thing: truth requires justification in order to come about. So that's the relationship between truth and justification, IMO. I don't think I've held (for over a half a decade) that "false knowledge" is possible, but false belief is possible. I'm not doctrinaire about this, but I think that "know" entails that the object of knowledge is a fact (though I grant that people talk as though you can have "false knowledge", an expression that occurs nowhere on the research CD). Thus the distinction is drawn between knowledge and error, crucially distinguished in terms of "fact".
since you raised the "crow" issue in that other thread: what do we say about knowledge at the time when we are only aware of black crows? Can we saw we certainly and truthfully know that only black crows exist?
This is in part a linguistic problem (which tells me I have to finish that dang paper on language and objective meaning). The linguistic aspect is the relationship between the word and the concept, where at a certain point in one's knowledge of society, you don't grasp what concept "crow" refers to. No theory of truth that depends on words will lead to anything but chaos -- we have to be talking about concepts. It is true that by their nature, all corvus caurinus (crow species) are black, which might lead one to say that all crows are black -- if one is unaware of the broader social fact that "crow" refers not just the the species that you are familiar with, but to multiple species -- it covers a genus. That's the linguistic problem.

The epistemological / scientific problem is that even if there were just one (black) species of crow, the conclusion is unjustified because you do not have evidence that something about being a crow causes the observed correlation. If you know rudimentary biology, you will know that albinism exists, which thwarts any "all X's that exist are black" claim. If you don't know that, you don't know what is necessary to get actual justification.

Link to comment
Share on other sites

I just finished typing up my notes on "The Art of Thinking" lecture on certainty available here.

I haven't read all of your notes, but enough of them to decide that I have to take that course; though I do wonder if it is appropriate to provide that thorough of a set of notes.

At any rate, did he give an examples of following the proper method with all the known facts, and yet still make a mistake? You mentioned the checkbook example, but I use a double entry method -- I do the addition and subtraction going down the list, and then I do the same thing going up the list. I do multiple entries before doing the mathematics. This method would seem to correct out mistakes -- i.e. if the two answers do not match up with the starting and ending figures, then I made a mistake, and I correct it. Though I suppose one type of error would be to write down the wrong amount for a single entry, or forget to carry the receipt with you for later entry. One aspect of making mistakes in a check book is with auto withdrawals that one overlooks while balancing one's checkbook.

Yes, I know, I should gear up to using a computer accounting program, but I don't have that many entires to worry about :D

So, can you give an example from the course of using the right method, having all the relevant facts, and yet still making a mistake?

Link to comment
Share on other sites

I haven't read all of your notes, but enough of them to decide that I have to take that course; though I do wonder if it is appropriate to provide that thorough of a set of notes.

Hell yes its appropriate. Ideas can't be copyrighted, and its not a transcript. It leaves a lot out, the notes are only thorough from the outline perspective of following all the key points. The presentation is sometimes quoted for short snippets but usually entirely omitted, paraphrased, or in my own words. By all means get the course, you won't regret it.

So, can you give an example from the course of using the right method, having all the relevant facts, and yet still making a mistake?

The checkbook example was a transcription error, so even the arithmetic was correct. The error goes undetected until the the bank statement arrives and then the contradiction is perceptual. An infallible method serves as a check on a fallible method. (Not that the bank is infallible, but that the differing balances are perceptual.)

Link to comment
Share on other sites

The checkbook example was a transcription error, so even the arithmetic was correct. The error goes undetected until the the bank statement arrives and then the contradiction is perceptual. An infallible method serves as a check on a fallible method. (Not that the bank is infallible, but that the differing balances are perceptual.)

Right. In other words, you perceive that the checkbook says you have a balance of say $523 and the bank statement says you have a balance of $554. The perceptual level evidence -- the facts -- are not questionable, since the sense are infallible, but since there is a contradiction, somewhere along the lines an entry was either overlooked or mistranscribed into the checkbook. One doesn't question the facts that one perceives (the two balance numbers), but due to the contradiction, one has to question the entries -- either of yourself or of the bank. You could question the arithmetic, but that kind of mistake is unlikely these days with computers.

By the way, banks do make mistakes. I once had a rent check deposited but something was wrong with the magnetic numbers at the bottom, and the bank put some sort of strip on it for identification with a different number than my check sequence number. I asked them about it and they made an investigation (since the check numbers didn't match from my records). In the end, the bank decided they couldn't prove it was my check, for some reason, and they reversed the charges, so I had free rent that month :) I was willing to assert that it was my check, but they followed their own procedure and it was in my favor.

Edited by Thomas M. Miovas Jr.
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...