Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Re: Knowledge defined to overcome the Gettier problem

6 views
Skip to first unread message

olcott

unread,
Apr 26, 2021, 1:36:19 PM4/26/21
to
On 4/26/2021 12:19 PM, Richard Damon wrote:
> On 4/22/21 12:03 PM, olcott wrote:
>> Instead of defining knowledge as a justified true belief that leaves
>> gaps in the degree of connection between the justification of the belief
>> and its truth we could define knowledge as the complete comprehension
>> that an assertion is necessarily true and thus impossibly false.
>>
>> The assertion that semantic meanings are expressed using words proves
>> itself to be true on the basis that it is an example of semantic
>> meanings expressed using words.
>>
>> It is also irrefutable because every refutation will also be an example
>> of semantic meanings using words.
>>
>
> I will point out that some of your arguments actually show show that
> words do not always have precise semantic meanings, and thus it does not
> an irrefutable proof.
>
> The semantic meaning of a sentence are only properly expressed using
> words when the words have a proper agreed upon meanings in context.
>
> semantic meaning can also be not expressed by the words used, but the
> words express some other false meaning.
>

if we adapt the conventional definition of knowledge from:
[justified true belief]
https://plato.stanford.edu/entries/knowledge-analysis/#KnowJustTrueBeli
to become a [fully justified true belief] such that this justification
guarantees the truth of the belief, then the "Gettier problems" with
original definition cease to exist.

[Self-evidence] https://en.wikipedia.org/wiki/Self-evidence
In epistemology (theory of knowledge), a self-evident proposition is a
proposition that is known to be true by understanding its meaning
without proof...

"This sentence is comprised of words." is proved to be true entirely on
the basis of the meaning of the terms: {sentence}, {comprised}, and
{words} combined together to form the compositional meaning of the whole
sentence.






--
Copyright 2021 Pete Olcott

"Great spirits have always encountered violent opposition from mediocre
minds." Einstein

olcott

unread,
Apr 26, 2021, 2:05:37 PM4/26/21
to
On 4/26/2021 12:53 PM, Richard Damon wrote:
> But a string of words may have NO meaning or a FALSE meaning, so, by
> itself, does not express knowledge.
>

If the sentence is not guaranteed to be true on the basis of the meaning
of its words then [fully justified true belief] is not met.

olcott

unread,
Apr 26, 2021, 4:12:10 PM4/26/21
to
On 4/26/2021 3:04 PM, Richard Damon wrote:
> The question is how many sentences are ACTUALLY guaranteed to be true,
> just by the meaning of the words, under any possible interpretation of
> the words.
>
> And are those sentences really 'knowledge'.
>

Except for the Gettier problem [justified true belief] is accepted as
the definition of knowledge.

When we change that to [fully justified true belief] the Gettier problem
utterly ceases to exist.

That the following sentence cannot possibly be false:
"This sentence is comprised of words." proves that there is at least one
instance of [fully justified true belief].

> How do you actually KNOW that others agree with your definitions, and
> that you don't have something defined differently than others?

That is an entirely different can-of-worms.

olcott

unread,
Apr 26, 2021, 5:15:40 PM4/26/21
to
On 4/26/2021 3:58 PM, Richard Damon wrote:
> On 4/26/21 4:32 PM, olcott wrote:
>> On 4/26/2021 3:23 PM, Richard Damon wrote:
>>> I will admit not having time to fully study that document, but I don't
>>> think it makes the claim you think it does.
>>>
>>> The claim is that for SOME knowledge, those conditions are not enough.
>>
>> The Gettier problem is that there are some assertions that are both true
>> and justified yet there is no relationship between the truth and the
>> justification.
>
> Yes, SOME.

When you restrict this to [fully justified true belief] such that the
meaning of the words proves that the belief is true or current sensory
stimulus confirms that the words are true: "I am eating a sandwich now"
then the Gettier problem utterly ceases to exist with this revised
definition of knowledge.

>
>>
>>> Not for ALL knowledge. That there exist some things that may be true,
>>> you actually belive them to be true, and even have justification that
>>> they are true, but you don't really have knowledge because your
>>> justification and belief aren't based on other Truths.
>>>
>>> Showing examples where the cases are sufficient doesn't prove that they
>>> are always sufficient.
>>>
>>
>> It is an axiom. All the cases that are not fully justified do not count
>> as knowledge.

Matt Faunce

unread,
Apr 26, 2021, 8:27:41 PM4/26/21
to
Regarding, “This sentence is comprised of words,” each term within the
sentence has a meaning which is derived from outside of the sentence. Then,
the meaning of each outside source is derived from outside itself, etc.
Therefore the full justification of the truth of that sentence requires the
verification of each source in the chain of outside sources. Whether the
chain is infinite, or stops at the whole universe within a set time-frame,
or ends with God, is another question.


--
Matt

olcott

unread,
Apr 26, 2021, 8:31:55 PM4/26/21
to
Bullshit on that.

olcott

unread,
Apr 26, 2021, 8:35:58 PM4/26/21
to
On 4/26/2021 7:27 PM, Matt Faunce wrote:
The sentence is its semantic meanings encoded as symbols.

> Therefore the full justification of the truth of that sentence requires the
> verification of each source in the chain of outside sources. Whether the
> chain is infinite, or stops at the whole universe within a set time-frame,
> or ends with God, is another question.
>
>


--

olcott

unread,
Apr 27, 2021, 10:20:41 AM4/27/21
to
On 4/26/2021 7:27 PM, Matt Faunce wrote:
The meanings of words exist as a knowledge tree inhertance hierarchy of
relations between meanings. This is a mutually self-defining semantic
tautology. https://en.wikipedia.org/wiki/Ontology_(information_science)

> Therefore the full justification of the truth of that sentence requires the
> verification of each source in the chain of outside sources. Whether the
> chain is infinite, or stops at the whole universe within a set time-frame,
> or ends with God, is another question.
>
>


--

Matt Faunce

unread,
Apr 27, 2021, 9:32:02 PM4/27/21
to
That the sentence in question “is its semantic meanings encoded as
symbols”, and that it’s true, is not in question. What’s in question is
whether a person’s belief that it’s true can be fully justified by that
person.

(1) Does the believer have to also be the justifier? I assume so.

(2) Is the tree of term-meanings, that supports each term of a particular
sentence, able to be completely investigated by a human? Is it even finite?
I would argue that it’s infinite, but, that returns diminish as you reach
outward accounts for practical knowledge, i.e., highly justified true
belief, but never fully justified. However, you need to argue that it’s not
only finite but small enough to be practically possible for a man to fully
investigate.

(3) There are many vague terms, like beauty and justice, which elude a
universally accepted definition. These terms play a supporting role in a
full justification of the terms in your original sentence. I don’t think
anyone can complete a practical investigation of their meanings never mind
a full investigation.

(4) Everything in reality will change*. After you complete your survey of
meanings of supporting terms, if that’s possible, you’ll have to go back
through it again to make sure nothing from the beginning of the
investigation changed on you by the time you reached the end. And once you
complete that, you’ll have to do it again, and again... Only a God-like
mind can take a snap shot of the whole of reality.

* What was meant by “the heavens” changed, thanks to Ptolemy, Copernicus,
and Einstein for example. Aristarchus certainly had a different meaning
than his contemporaries.
The term “multiply” once had a meaning that proscribed the multiplication
by a fraction of one, e.g., 6×1/2=3 would have been expressed by 6÷2=3 or
6:2=3:1. The term’s meaning grew.

--
Matt

olcott

unread,
Apr 27, 2021, 10:01:05 PM4/27/21
to
I always thought that is has always been a horribly awfully terrible
mistake to associate belief with knowledge because all beliefs can
possibly be false and no knowledge can possibly be false.

I merely started with the original definition of knowledge that is
subject to the Gettier problem and correct the error to eliminate the
Gettier problem.

The scope of this investigation was to fix the error in the defitition
of knowledge to eliminate the Gettier problem. I did do that.

A [fully justified true belief] such that the justification makes the
belief impossibly false does do that.

Every expression of language that can be verified as necessarily true
entirely on the basis of it meaning defines the body of analytic
knowledge that meets the above criteria of [fully justified true belief].

Empirical knowledge is a whole other can or worms entirely. The best
that we can do with empirical knowledge is confirm that a set of what
appears to be sensations from the sense organs that are happening right
now correspond to their verbal description using language.

We cannot definitely confirm that physical realty, other minds or five
minutes ago actually exist.

Matt Faunce

unread,
Apr 27, 2021, 11:00:43 PM4/27/21
to
Well, it appears that you and I are on opposite sides of a debate that
before you posted here I thought was settled. I’m talking about the
analytic/synthetic distinction. Years ago I read Quine’s paper, Two
Dogmas... and agreed with his conclusion that there is no real distinction
between them, although I thought his presentation was bit tortured due to
more fundamental disagreements I have with him (—I’ve long been a
pragmatist who never liked logical positivism or the general mindset of
so-called analytical philosophers.) Before my first response to you I
looked at a couple of summaries of Quine’s paper to refresh my memory. In
searching I saw that the debate was never settled! That was a surprise to
me. I don’t know what to say at this point except I’m not willing to dig
deeper into the thinking of analytical philosophy in order to challenge
them or myself, depending on what I find. Not that my philosophy is
settled, it’s not. My unsettled area is the debate between Charles Peirce
and Joseph Margolis. Both are pragmatists, but they have important
differences.

--
Matt

olcott

unread,
Apr 27, 2021, 11:22:35 PM4/27/21
to
There is a very clear distinction between direct sense data from the
sense organs and abstract reasoning using language that requires no
basis what-so-ever in sense data from the sense organs. This proves that
the distiction between analytic and empirical is clear and unequivocal.

The first hand direct experience of the actual taste of strawberries and
the direct memory of these physical sensations is purely empirical.

The pure interconnected concepts of this sentence:
"This sentence has five words." Apart from its symbolic or phoenetic
physical represetnation is pure analytic.

There is a whole bunch of knowledge that spans both of these boundaries
and the analytic and empirical aspects can be easily divided.

Whatever aspect requires sense data from the sense organs or memory of
sense data form the sense organs is empirical. Whatever aspect can be
encoded using language is analytical.

> although I thought his presentation was bit tortured due to
> more fundamental disagreements I have with him (—I’ve long been a
> pragmatist who never liked logical positivism or the general mindset of
> so-called analytical philosophers.) Before my first response to you I
> looked at a couple of summaries of Quine’s paper to refresh my memory. In
> searching I saw that the debate was never settled! That was a surprise to
> me. I don’t know what to say at this point except I’m not willing to dig
> deeper into the thinking of analytical philosophy in order to challenge
> them or myself, depending on what I find. Not that my philosophy is
> settled, it’s not. My unsettled area is the debate between Charles Peirce
> and Joseph Margolis. Both are pragmatists, but they have important
> differences.
>

Settling this debate is merely a facet of mathematically formalizing the
notion of analytic truth. Tarski thought that he proved this impossible
yet seemed to have used the liar paradox as his basis.

http://www.liarparadox.org/Tarski_247_248.pdf
0 new messages