You know that Albert Einstein famously asked that very question. The exact quote is:

"We often discussed his notions on objective reality. I recall that during one walk Einstein suddenly stopped, turned to me and asked whether I really believed that the moon exists only when I look at it."

As recalled by his biographer Abraham Pais.

Why did Einstein, of all people, feel obliged to ask that question? — Wayfarer

Because he thought Pais was nuts. Anyway, physicist David Mermin has since resolved the question to everyone's satisfaction. :-)

The questions with which Einstein attacked the quantum theory do have answers; but they are not the answers Einstein expected them to have. We now know that the moon is demonstrably not there when nobody looks. — Boojums All the Way Through - N. David Mermin

The idea as human beings being subjects and the world being the object is what I'm referencing here. That we're thinking things in the sense Descartes meant -- consciously aware beings, and since Kant subjects with object as "phenomenon" and representation. Schopenhauer discusses this at length as well, as one of the most basic principles of all knowledge.

It's hard to see things any other way, I realize...hence why I'm wondering about others' opinions. — Xtrix

I normally think of them as:

Subject: A person or thing that is being discussed, described, or dealt with.

Object: A person or thing to which a specified action or feeling is directed.

In a scenario where Alice sees Bob, Alice is the seeing subject and Bob is the seen object.

↪Xtrix Can you expand the question more please? Don’t know what you’re asking. — I like sushi

It's like a "choose your own adventure" book. You fill in the details and then answer that. :-)

:up:

Chalk me up in the pro subject/object notion column, but I don’t see the world that way. — Mww

Performative contradiction? Or do you mean other people generally see the world that way, but not you?

It's worth noting the grammatical origin of those two terms which have often acquired different meanings in the history of philosophy (including with Descartes' subject-object dualism).

The subject is, according to a tradition that can be traced back to Aristotle (and that is associated with phrase structure grammars), one of the two main constituents of a clause, the other constituent being the predicate, whereby the predicate says something about the subject. — Subject (grammar)

For example, consider the sentence, "Alice sees Bob". Alice is the subject, sees Bob is the predicate and Bob is the object.

In the early 20th Century, self-awareness about the use of language in philosophy was marked by the linguistic turn and, specifically, ordinary language philosophy.

Ordinary language philosophy is a philosophical methodology that sees traditional philosophical problems as rooted in misunderstandings philosophers develop by distorting or forgetting what words actually mean in everyday use. — Ordinary language philosophy

How does she know it wasn't raining while Bob was hosing the window? Bob hosing the window isn't justification for it not to have rained earlier. — Harry Hindu

OK. Suppose Bob (who she believes to be honest and reliable) told her that it wasn't raining earlier and that he was hosing water on the window. Thus Alice would justifiably form the belief that it was not raining earlier.

This goes back to what I said about making objective observations. You seem to be saying that we check our knowledge when we get outside of the thing we are talking about. — Harry Hindu

No, there is no proof (or getting outside the thing we are talking about).

So, if we know that to really know whether or not is raining is to go outside and look, then looking out the window isn't proper justification for knowing it is raining. If this is the case, then no, Alice didn't know it was, or wasn't raining, because she didn't have proper justification. — Harry Hindu

Alice never has a "proper" justification if proof is required. That is model K1 (which we both reject). Alice's initial belief that it is raining is justifiable in both H1 and H2.

If all she needs is justification, then yes, Alice knew it was raining and now she knows something different - that it wasn't. — Harry Hindu

OK, that is model K3 (which you seem to accept and I reject). The consequence is that per hypothetical H2, Alice at the earlier time knew that it was raining when it was not.

If you want to bring up the possibility of Alice hallucinating while outside, how do you retroactively show she is hallucinating - by asking someone else? How do we know that they aren't hallucinating, or lying? — Harry Hindu

If you have reason to believe someone is hallucinating then you factor that into your judgment. But, as discussed, any reasons you have will fall short of proof, and there is always the possibility of being mistaken despite your belief being justifiably formed. That's the way it goes sometimes when proof is not obtainable.

If all you can have are justifications and truth is something elusive, then it stands that the only requirement for knowledge is justification. No one can ever know if Alice used the term, "know" correctly, if truth is a requirement for its correct usage, which is the same as saying it isn't a requirement at all. — Harry Hindu

Alice has no proof that she used her words correctly. It doesn't follow that she didn't, in fact, use her words correctly. Which she did in H1 on model K2.

You seem to think that rejecting model K3 means that one therefore requires proof of the truth. That's not the case. Proof is not a requirement for K2, only truth and justification is. Since Alice's belief is both true and justified in H1, it follows that Alice has knowledge in H1 per K2.

The central concept in computability is that a problem is solvable if there exists an effective procedure for deriving the correct answer. That is pretty much what the Church-Turing thesis says.
...
I wonder if computability and epistemology are ultimately not one and the same thing? — alcontali

I think so. The Church–Turing–Deutsch thesis takes this a step further and states that the universe itself is mathematically isomorphic to a quantum Turing machine.

... a modification of Turing's assumptions does bring practical computation within Turing's limits; as David Deutsch puts it:

"I can now state the physical version of the Church–Turing principle: 'Every finitely realizable physical system can be perfectly simulated by a universal model computing machine operating by finite means.' This formulation is both better defined and more physical than Turing's own way of expressing it." — The Church–Turing–Deutsch thesis

Are we talking about the state of one's knowledge, or the state of the weather? — Harry Hindu

Alice has formed a belief about the state of the weather (in H1 and H2). We are talking about the state of her knowledge (in terms of K1, K2 and K3).

When the window is being hosed, what is the distinction between, "I know it is raining" and "It is raining."? — Harry Hindu

When Alice forms the belief, "It is raining", her belief is about the state of the weather. Per K2, she knows that it is raining in H1 but not in H2.

If Alice further forms the belief, "I know it is raining", her belief is about the state of her knowledge. Per K2, this latter claim is justified in both H1 and H2 but only true in H1. Therefore she knows that she knows it is raining in H1 but not in H2.

The latter stems from the prior. You can't say, "It is raining." and say that it is about the weather without having some justification. It would be more like guessing. — Harry Hindu

That statement is still about the weather even if you are guessing. An unjustified claim can still be true, but it wouldn't be knowledge under models K1, K2 or K3.

I thought you said that we can't have proof, yet you are now saying that you can have proof retroactively? — Harry Hindu

No, a revised claim is subject to the same conditions for knowledge as the original claim. Which is to say, there is no proof for the revised claim either.

If Alice was using the term in the normative way, then she is using it as I have defined it. The veridical use would be your definition. But you have shown that the veridical use refers to something unattainable, while the normative use refers to having justification only. I think we are pretty much in agreement, it's just you haven't realized it yet. — Harry Hindu

The veridical use is attainable - see H1, K2 where Alice has a justified and true belief that it is raining, and therefore has knowledge on that model (K2).

I think we are pretty much in agreement, it's just you haven't realized it yet. — Harry Hindu

Here's a question for you. In H2, suppose that Alice later discovers that Bob was hosing the window with water. Thus she now knows that it wasn't raining earlier.

Did Alice know that it was raining at the earlier time?

So how does one get to know it is raining? — Harry Hindu

I've presented three models (K1, K2, and K3) that each provide a potential answer. Hypothetical's H1 and H2 demonstrate the consequences of each model for Alice's claim that it is raining.

On model K1, Alice never knows it is raining since proof is unobtainable.

On model K3, Alice knows it is raining whenever her belief that it is raining is justified. The problem is that in H2, she knows it is raining when it is not raining.

On model K2, Alice knows it is raining in H1 but not in H2.

Of those three models, K2 is the best explanation for ordinary usage. It's also the model philosophers generally accept (Gettier aside).

So imagine yourself in Alice's position where it appears to be raining outside. You can't easily distinguish H1 from H2 (nor would all the claim-defeating possibilities like water from a hose even come to mind). But per K2, if you are in H1 your belief meets the conditions for knowledge. If you are in H2, your belief does not.

So to answer your question, you get to know it is raining when you are in scenarios like H1, but not when you are in scenarios like H2. If you wish to, you can also do things like go outside to check. The benefit is that you can potentially rule out or discover some H2-like scenarios (though not all), but it also bears the practical cost associated with the checking. The standard for justification is a pragmatic middle-ground between not bothering to look at all (i.e., guessing or believing whatever one wants to believe) and endlessly checking (which will still fall short of proof).

How can you claim that your belief is true without proof? — Harry Hindu

Because it is justifiable to do so - that's the normative rule for when we can make knowledge claims. If we needed proof, no claim could ever be made.

Now that means that we can sometimes inadvertently make false claims. That is, we can use the word "rain" when there is no rain, which would be a misuse. Normatively, that is okay - we're not infallible and we don't hold anyone to that standard. But veridically, we would have made a mistake. If we have made a mistake then it's not knowledge. That's a logical consequence of the model K2 - it doesn't depend on anyone ever discovering the mistake.

If you don't need proof that you are using the term correctly, then it seems that thing you don't need proof of isn't a necessary component of "knowledge", or it's not important to know when you're using "know" correctly. — Harry Hindu

We ultimately want true claims, not false claims. That's why truth is a condition of knowledge. But we also recognize that the possibility of mistakes shouldn't preclude people from making claims at all. It just won't be knowledge if the claim is false.

If you can't enforce the rules for the use of the term. "knowledge", then that is to say that there aren't any rules when using the term. — Harry Hindu

The only way to enforce the rules upfront is either to require proof (per K1), which makes knowledge unattainable, or to allow false claims to count as knowledge (per K3).

What we do instead is to enforce the rules retroactively (per K2). That is, if we discover a misuse it is retroactively corrected. For example, suppose in H2 Alice later discovers that Bob was hosing water on the window. In that case, she recognizes that she didn't know it was raining at the time, she only thought she did.

That's strange. It was your claim that it is raining, not some state-of-affairs that it actually was raining. How do I know that you are right, when all you have to show is your justifications. If you can make a claim and assert that that is the state-of-affairs stipulated, then somehow you have gained true access to the world. How did you do that? — Harry Hindu

I didn't. There's a difference between making a claim that it is raining (which I did not do) and presenting a hypothetical within which Alice makes a claim (which I did do).

To ask how I know that Alice exists, for example, or how I know that it is raining is misplaced. We're discussing a hypothetical state-of-affairs that I've stipulated, not an actual state-of-affairs. The premises of the hypothetical are taken as a given or as a basis of agreement (unless there is some reason to take issue with them).

If there is no guarantee, then you can't know that you ever used the word correctly. — Harry Hindu

You're using the word "know" in the sense of K1 again. On K2, you know that you used a word correctly when you have a justified and true belief that you did. You don't need a guarantee, you just need those conditions to be met.

You can prove you have justification, — Harry Hindu

No, you can't prove you have a justification either. Nor did Alice have such a proof in hypotheticals H1 and H2. She hadn't ruled out the possibility, for example, that she was looking at a virtual (VR) window instead. Whether something counts as a justification assumes a particular model for justification, itself contestable. Taking it further, suffering a hallucination may preclude one from having any (or very few) justifiable beliefs at that time.

No claims are provable except within the context of a deductive proof.

Do you agree that word-use can be picked up and used without really knowing what they mean? — Harry Hindu

Yes.

Is there a difference between using words and understanding what words mean? — Harry Hindu

Yes.

It seems to me that we can use words that appears to be the correct use, because everyone else uses it that way, but there are cases where mass delusions exist and people use the same words like "I know the Earth is flat" without knowing the truth. — Harry Hindu

Yes. In those cases, only on K3 would they have knowledge (assuming their claim was justifiable, which is contestable). On K2, they thought they knew that the Earth was flat, but didn't actually know that.

How do you guarantee that your claim is true? How do you show that your claim is true for it to qualify as using the term "knowledge" correctly? — Harry Hindu

There is no guarantee. Alice can show that her claim is true in H1 by pointing out the window since that is what justifies her belief. She can't show that her claim is true in H2 since her claim is not true in that hypothetical.

But the only way to know it is true is to have proof. — Harry Hindu

You're asserting K1 here and the rest of your questions assume it. Knowledge on K2 and K3 does not require proof. They require justification, which Alice has since her belief was formed by looking out the window. Thus she knows it is true in H1 on K2 without proof. Interestingly, she knows it is true in both H1 and H2 on K3 (again without proof). That's because she can have false knowledge on K3.

In H2, K2, who is claiming that Alice does not know it is raining, and what proof do they have that their claim is true? — Harry Hindu

That it's not raining is the state of affairs stipulated in H2. That's prior to any claims.

My definition is the one that would be the "ordinary" definition, as it allows knowledge to be only about justifications, not truth - of which you need proof the claim is true to say that you are using the term "know" correctly. — Harry Hindu

OK, let's test our claims. I'll outline the three distinct models for knowledge that we've discussed here (K1, K2 and K3) and what they imply for two distinct hypotheticals (H1 and H2).

(K1) Knowledge requires proof and truth
(K2) Knowledge requires justification and truth
(K3) Knowledge requires justification

Note that the truth condition in K1 is redundant but I'll leave it there for clarity. That is, if there is a proof that it is raining, then it is true that it is raining.

Also, as we have discussed it, justification falls short of proof and so doesn't imply truth. That is, a justified belief can be false.

Finally, I haven't indicated belief (to reduce clutter), but it should be implied for each model.

--

(H1) It is raining. Alice looks out the window and forms the belief that it is raining (since it appears to her to be raining).

Per K1, Alice does not know that it is raining. That's because she lacks proof - she has not definitively ruled out all other possibilities such as water from a hose.

Per K2, Alice knows that it is raining. That's because her belief is justifiable and her belief is true.

Per K3, Alice knows that it is raining. That's because her belief is justifiable.

--

(H2) It is not raining. However Bob is hosing water on the window. Alice looks out the window and forms the belief that it is raining (since it appears to her to be raining).

Per K1, Alice does not know that it is raining. That's because she lacks proof - she has not definitively ruled out all other possibilities such as water from a hose.

Per K2, Alice does not know that it is raining. That's because her belief, while justifiable, is false.

Per K3, Alice knows that it is raining. That's because her belief is justifiable.

--

I'll leave it there for the moment. Do you agree with the above conclusions for each hypothetical?

"Truth" would be ever-elusive, and we would never know when to use the word, "truth" appropriately because we never have any guarantees (proofs) of what the truth is. — Harry Hindu

We know when to use the word "truth" when we have a justifiable belief for when to use it and our belief is true.

Note that you keep presuming that one needs a guarantee (proof) in order to know something. But that is an infallibilist definition of knowledge, not the ordinary definition.

So I agree that one can never prove that one has the truth. It doesn't follow that one can never know that one has the truth. That's because the standard for knowledge is an ordinary and pragmatic one, not an infallible and unattainable one.

Right. Which is to say that it has no definite identity. Which is another way of calling into question its actual existence. Which in turn has a lot to do with the whole Einstein-Bohr debate. — Wayfarer

As far as scientists can tell, all physical systems are quantum and so will produce the same quantum behavior as microscopic particles (one example is a visible piezoelectric tuning fork comprising about 10 trillion atoms). So the issue of identity and existence applies equally to large systems as small.

What if it's not 'a different notion of identity', but that it's not an identity at all. There are not two things which are identical. It is said that particles aren't even particles until they're measured, prior to that literally all that exists is a distribution of possibilities. It's not as if there's a particle whose whereabouts is unknown, it doesn't have an exact whereabouts, so also does not have an identity. In fact couldn't you say that the measurement is what confers identity on it, by giving it a position and making it 'this particle'? — Wayfarer

It's not so straightforward. While a position measurement makes the position definite (within the measured range), it also makes the momentum indefinite (i.e., the particle is now in a superposition of momenta). So the state of a particle always has at least some indefinite properties (due to Heisenberg uncertainty), regardless of whether the particle is measured or not.

A proof is a guarantee of the truth of one's claim.
— Andrew M

How do you guarantee the truth of your claim? — Harry Hindu

You don't. Knowledge doesn't require proof. It requires that one's belief is both justified (a standard that is lower than proof) and true.

In that case yes, but if we say, "It is raining" and it is raining, then we are talking about the actual state of affairs.
— Andrew M

What if we're brains in vats? How do you know your not a brain in a vat, or hallucinating when you say "It is raining."? — Harry Hindu

Ordinarily the model is applied to the natural world of everyday experience (where we can point to what we mean by rain), not a supposed brain-in-vat world. So that renders the first question inapplicable (at least for ordinary claims). For the second, the question of knowing that it is raining is just whether your belief is justified and true. You don't need proof that all the things that could go wrong didn't go wrong (though, per Gettier, it does need to be the case that things didn't go wrong in ways that undermine the claim - i.e., that one was hallucinating would preclude the claim from being knowledge).

As for sub-atomic particles - do they have any identity, if they don’t even have a position? I mean, electrons and photons notoriously manifest as waves in some contexts, and particles in others. So if they don’t have an identity prior to measurement then they're indistinguishable as a matter of principle - as you say - but perhaps not in the way you mean. — Wayfarer

Part of a quantum particle's identity is that it can have amplitude for two or more positions (i.e., a superposition).

This is just a different notion of identity to the classical idea where particles always have a definite position and momentum.

By 'identical' I mean similar in every detail; exactly alike. — believenothing

See identical particles which are indistinguishable in principle.

I don't like the wording here. It doesn't make any sense to say that some state-of-affairs is a truth-maker, as if some state-of-affairs makes some other state-of-affairs called the "truth". Which state-of-affairs are we talking about when using our knowledge - the state-of-affairs that made the truth, or the state-of-affairs that is the truth? Claims don't bear truth if they are wrong.

All you have done is state what makes truth and what bears the truth, but haven't explained what the truth is and how it is made by some state-of-affairs or carried in a claim. — Harry Hindu

In the Alice/rain hypothetical, the state of affairs is that the window is being hosed with water (i.e., it is not raining). When Alice looks out the window, she forms the belief that it is raining.

Her belief represents a state of affairs that has not obtained. Thus her belief is false. However if it were raining (i.e., if that state of affairs had obtained), then her belief would have been true.

That defines the ordinary meaning (or use) of our truth terms.

The standard model for knowledge is JTB. Since Alice's belief is false, she does not have knowledge. If it were raining, she would have had knowledge (since her belief would have been true and also justified by her looking out the window).

That defines the ordinary meaning (or use) of our knowledge terms.

Note that the logic of the use of those terms is demonstrated in that hypothetical. They are, in effect, formal models where claims are satisfied in particular scenarios (and not in others). Also, the models can themselves be contested, as with the Gettier problem (or with your claim that knowledge can be false).

In reality we're in the position of Alice in the hypothetical. Instead of stipulating a hypothetical world with full knowledge of what claims are true and false in that world, we are instead a part of the world we're seeking to represent. However the logic of the terms defined above apply in exactly the same way. That is, if we were in Alice's situation and formed the same belief, our belief would also be false and we wouldn't have knowledge.

If we can't have proof that one's knowledge is actually true, then it is illogical to say "truth" is a property of knowledge. — Harry Hindu

No. See above. If Alice's belief were true (i.e., if it were raining), then she would have had knowledge. Yet the justification for her belief (that she looked out the window) did not amount to a proof.

What is "proof"? — Harry Hindu

A proof is a guarantee of the truth of one's claim.

It seems to me that we can only ever talk about our knowledge, not the actual state-of-affairs. If we say, "I don't know if it is raining", then we're still talking about the state of our knowledge. — Harry Hindu

In that case yes, but if we say, "It is raining" and it is raining, then we are talking about the actual state of affairs.

Deterministic systems can behave probabilistically
Ignorance or rather the impossibility of knowing was the actual impetus for the development of probability theory — TheMadFool

:up:

I wonder how one differentiates the true random from pesudorandom? — TheMadFool

From an operational perspective, "true" randomness would come from outside the system of interest (such as from thermal noise or quantum phenomena that are used as inputs to the system), whereas pseudorandomness would be the result of computable processes within the system.

A determinist about the universe would regard them as ultimately the same thing. Randomness and chance would just be terms indicating one's ignorance of the relevant information for making correct predictions. Empirically, one could try to discover the underlying laws or causes that explain apparently random events.

Because...

we can introduce randomness or more accurately pseudo-randomness into a deterministic system. — TheMadFool

Yes, so does any puzzle remain for you?

BTW here's a quick-and-dirty example. Suppose my algorithm for 1000 dice throws is to take the first 1000 prime numbers, calculate their square roots, multiply their decimal expansion by 6 and round up to the nearest natural number. The algorithm is deterministic and the sequence is pseudorandom.

The first few throws are:
3, 5, 2, 5, 2, 4, 1, 3

Another path is saying, "I know that it is raining" is talking about your knowledge, not the rain. You are talking about the state-of-affairs that is your knowledge, not the weather. So, while you may know what rain is, you don't know much about your knowledge because it isn't raining outside. — Harry Hindu

Yes, that is my view. We can be mistaken about whether we know it is raining just as we can be mistaken about whether it is raining.

It seems to me that you don't have proper justification to claim to know anything until you make the proper observation from the proper perspective. — Harry Hindu

With the rain example, if we went outside we presumably shouldn't be mistaken about whether it is raining or not. But that still falls short of a guarantee or proof. People sometimes have hallucinations, holograms are possible, and there will be other possibilities I haven't thought of. (And that's before getting to the more skeptical hypotheses of brains-in-vats, Descartes' evil demon and the like.)

Consider the clock example of the OP. Normally we just glance at the clock to find out the time. In order for that to count as knowledge, should we verify that the clock is working first, as well as check it against other clocks? But even if we did that, we might still be mistaken (e.g., perhaps the clocks are wrong because of a daylight savings time change, or someone else just set all the clocks incorrectly).

These extra requirements would normally be overkill for forming a belief about the time or reporting it to others. Even if fulfilled, they still fall short of a guarantee or proof. And, in the end, they fail to match our ordinary use of knowledge terms.

That is, glancing at the clock or looking out the window for rain is normally considered a sufficient justification for our beliefs and claims. If someone asks how we know what time it is, they are normally satisfied when we say we looked at the clock just now.

So only in making the proper observation can we say that we possess knowledge and we obtain the proper observations when we are objective in our perspective. One observation isn't enough justification to make a knowledge claim. — Harry Hindu

However if only deduction provides a guarantee or proof, then the only mistake-proof claim we could make would be of our own existence, per Descartes. Which is a very different thing to ordinary knowledge of the time of day or whether it is raining (that doesn't require a guarantee or proof).

I'd like to give my own "solution" to the paradox:

A deterministic system can't be random and the die is behaving as if it is random. This implies that a random element was introduced into the system — TheMadFool

It doesn't imply that. A number-generating algorithm simulating 1000 throws can be completely deterministic. Given the algorithm, I could correctly predict every outcome before the simulation is run. For example, that throw 23 would produce a six. Yet all the outcomes taken together would follow a probability distribution.

Perhaps you just want to say that this is not true randomness, it only appears random. If so, the term you could use is pseudorandom.

However, if that's the case 2, and 4 should be false but they are true and indicate the die is behaving as if determinism is false. — TheMadFool

That 2 and 4 are true is completely consistent with determinism. A deterministic algorithm is able to produce pseudorandom numbers that follow a probability distribution. And the generated sequence is repeatable.

Because of the nature of number generating algorithms, so long as the original seed is ignored, the rest of the values that the algorithm generates will follow probability distribution in a pseudorandom manner. — Random seed

What is it that we are trying to accomplish when we say, "I know <something you "know">"? — Harry Hindu

We are trying to describe a particular state of affairs - the way the world is. If our claim is true then we have been successful in that endeavor.

Suppose Alice says "I know it is raining". Her intention is to describe a particular state of affairs - that she knows that it is raining. She achieves that if and only if that state of affairs has obtained.

So if she does know that it is raining then she has a justified and true belief that it is raining. That is, she has looked out the window (resulting in the formation of her belief) and it is, in fact, raining.

If any link in the chain is broken, the whole chain is broken. In which case the speaker has failed to accomplish what they intended.

Then why does it feel like you possess knowledge when you don't? — Harry Hindu

Because a mistaken belief doesn't feel mistaken when you have it.

To people several hundred years ago it looked as if the Sun went round the Earth. But what would it have looked like if it had looked as if the Earth turned on its axis?

As far as appearances go, both look the same. The difference is in the explanatory hypotheses.

(The example is from Wittgenstein.)

When you've already had the experience of claiming you have knowledge and then find out that you didn't, then that should cause some concern for any other knowledge you claim to possess AND cause concern about your very understanding of what "knowledge" is. When having knowledge and not having knowledge are indistinguishable at any given moment you make a claim, then how can you really know what you are talking about? — Harry Hindu

Consider a parallel example. Does discovering that it wasn't raining when you thought it was cause concern about your understanding of what "rain" is?

It shouldn't, unless there were some further reason to think there was a problem with your understanding (e.g., people consistently referring to what you call "rain" as "snow").

The notable difference is that rain is a concrete thing whereas knowledge is an abstraction. But being mistaken about what you think you know is usually just being mistaken about something more concrete, such as whether it is raining.

How can you ever say that you are claiming some truth at any given moment? — Harry Hindu

You can make a claim when it is justifiable to do so (such as looking out the window for the rain scenario). But there is never an infallible guarantee that a given claim is true, and knowledge, to obtain, does not require one.

Why do they claim to know things that they don't? If they claim to understand what knowledge and knowing is, then how can they misuse the terms? — Harry Hindu

Because people can make a mistake when they deploy those terms to describe a state of affairs.

Alice looks out the window and claims it is raining. It is not raining (someone was hosing water on the window). Therefore she made a mistake. Yet she understands perfectly well what the word "rain" means.

Truth is the actual state-of-affairs. — Harry Hindu

The standard use in philosophy is that a state of affairs is a truth-maker while a claim or a belief is a truth-bearer. So states of affairs obtain or fail to obtain (e.g., it is raining) while claims are true or false (e.g., Alice's claim that it is raining).

If it is a common understanding that what we claim we know can be faulty (which doesn't mean that it necessarily is all the time), then it should be obvious that when we claim we know something, doesn't mean that truth is necessarily involved. Truth has to be something separate. — Harry Hindu

Yes, and no-one disagrees with this. There is a language distinction between what we claim we know (which can be false) and what we know (which can't be false). When someone claims to know that it is raining when it is not, they have made a mistake - and they don't have knowledge.

Bingo! You finally got it! — Harry Hindu

I've been saying it from the beginning of our exchange, but I'm glad you've finally caught up! ;-)

So if this is common knowledge - that there is no getting around the fact that no-one has an infallible guarantee that any specific claim is true, then that means people use the term, "knowledge" in the way I have described it, not you. People understand that their knowledge is fallible and so don't use the term in a way that implies truth - only justification. — Harry Hindu

That people are fallible means that they sometimes misuse the term "knowledge" or "know". They sometimes claim to know things that they don't.

Here's the definition of use from Lexico: "Take, hold, or deploy (something) as a means of accomplishing or achieving something; employ." In the context of our discussion what are being deployed are words and sentences.

Now what has Alice accomplished or achieved when she claims to know it is raining when it isn't (as she may later discover)? She hasn't accomplished what she intended, because she has a mistaken belief.

As I mentioned earlier, this issue is not just confined to knowledge terms. That people are fallible means that they can mistakenly misuse any word (according to their own definition). They can point out the window and say "rain" when it's not rain (for example, it's water from a hose).

Not only does your version of "knowledge" not fit how people use it, it relegates truth into meaninglessness as well. If there is no infallible guarantee that any specific claim is true, what does it mean to be "true"? — Harry Hindu

It means that a claim describes a state of affairs as it has been defined. For example, we understand what the phrase "it is raining" means. So if Alice claims it is raining when it is raining then her claim is true. That is a correct use. Whereas if she claims it is raining when it is not raining, then her claim is false. That is an incorrect use (or misuse).

That's one way of putting it. Although I agree with the description as a commonly understood one, I'm very hesitant to employ the terminology as an explanation of his belief. I think we can do better, with less verbiage. — creativesoul

Fair enough. How would you explain it?

This isn't anything different than what you've already said. Your model is useless if you can never know when it's appropriate to use. Knowledge would be this state-of-affairs that we'd never know about because we can never know that knowledge is true because we only have justifications and justifications are not truths. They might be, but we'd never know. — Harry Hindu

There is no getting around the fact that no-one has an infallible guarantee that any specific claim is true. So there's no point criticizing the model on those grounds.

It happens to be the model that people ordinarily use, including yourself, except that you also allow justifiable, but false, claims to be knowledge. Your model is no more (or less) useful, it's simply a different choice of language to represent the same reality.

We seem to be in essential agreement.

My main point about prior beliefs is that Bob would presumably just glance at the clock and automatically form a belief about the time. He's not consciously reflecting on it, weighing up evidence, or making inferences from one belief to another. How he forms his belief happens "under the hood", so to speak, as part of the brain's internal processing.

However we can infer his beliefs and construct a chain of inferences as a representation of what's going on in Bob's subconscious.

We're working from different notions of what counts as a basic or rudimentary belief. Our exchange led us into the notion of whether or not a language less creature's belief could possible count as being well grounded. If it requires being based upon other beliefs, then we arrive at the notion of infinite regress... Somewhere along the line, some belief or other is not based upon prior belief.

Can those be identified and/or isolated, and can they count as being well grounded true ones? — creativesoul

I think so, at least in principle.

It also seems clear to me that there are a plethora of pre-existing belief underwriting the very ability to participate in time telling practices such as looking at clocks, so... To say that there is no implication that he needed another belief prior to forming that belief is most certainly wrong. — creativesoul

What I was referring to here is the idea that Bob first needed to think about whether the clock was working before he believed it was 3pm. And before that he needed to think about whether the clock was real or a hologram. And before that ...

He didn't do any of that. He just glanced at the clock, saw that it showed 3pm, and then got on with the rest of his day.

I otherwise agree that we can infer that he has other related beliefs (such as that clocks are generally reliable, that clocks are usually real and not holograms, and so on).

But... when we're offering an account of Bob's belief, they must be Bob's beliefs... right? — creativesoul

We're asking whether Bob's belief that it is 3pm counts as knowledge (and why or why not). It doesn't really matter if Bob simply believes that space aliens implanted his mind with the correct time. We only care that Bob believed the time that the clock showed and that the clock was working correctly. If he did and the clock was working correctly, then we say that Bob knew it was 3pm, regardless of his personal theories of things.

Fiveredapples has given an eloquent defense of his position, but my position is much simpler. I'm sticking to JTB as the definition of knowledge. I don't see any good reason to give it up. The clock was broke, therefore the person wasn't justified in their belief that the time was X. If they weren't justified, then they failed to meet the definition of knowledge under JTB. Gettier fails for similar reasons. — Sam26

I'm curious whether you think there can ever be justified but false beliefs. If not, then the T condition in JTB would be redundant since the J would already entail truth. Thus knowledge, on your view, would be justified belief (where justified entails the truth of the belief and its premises).

Right. So how do you know that you or someone else is using the term, "knowledge" correctly, so say things like, "I/You are using the term, "knowledge" correctly."? — Harry Hindu

I have a model for what knowledge is. For example, my model says that knowledge is always true. So if Alice says that she knows it is raining, but it's not raining, then Alice's claim doesn't satisfy that model. So she didn't use the term correctly (in the veridical sense - her claim may still have been justifiable).

Whereas her claim may be satisfied on your model (that doesn't include a truth condition for knowledge).

This seems to arrive at a problem regarding the origen and/or content of belief. It presupposes that all belief is premiss based. I've an issue with that as a result of the fact that premisses themselves are belief.

Seems to me that it would have to be the case that some rudimentary belief are not premiss based. If they need to be in order to qualify as being well grounded then such belief cannot count... by definition alone... for if the definition is good... they do not have what it takes. — creativesoul

Bob looks at the clock and forms the belief that it is 3pm. That's a basic or rudimentary belief with no implication that he needed another belief prior to forming that belief, which would just result in an infinite regress.

However we can nonetheless investigate the premises of Bob's belief. Those premises emerge as part of our analysis, not something we need to suppose were Bob's beliefs at that time.

Well yeah, because in using knowledge terms you are referring to your knowledge. What else could you be doing with those terms? — Harry Hindu

That's the intention, but there is no guarantee of success. Compare with using the word "raining". Alice might look out the window and say, "It's raining" when it's not raining (e.g., she instead sees water from a hose).

Her use is justifiable, but unsuccessful.

But you're not taking this to it's ultimate conclusion and that is how do we know that the aliens know the truth? How do you know that you have acquired the truth when you only have justifications to go on? Again, as you are defining it, you'd need to know that your knowledge is true, not only justified, in order to use the term "knowledge" correctly. — Harry Hindu

The justified claim only needs to be true in order to use the term "knowledge" correctly or successfully. It is the same in this respect to the use of the word "raining" above.

If you're not referring to your knowledge when using knowledge terms, then what do you mean when you use the terms? — Harry Hindu

The intention is to refer to knowledge. The reality may be different.

When I say "use" I mean making a particular sound or scribble to refer to the information one possesses about a particular state-of-affairs, like the steps one takes to tie their shoes, and the reasons why one should tie their shoes. What do you mean by the word "use"? — Harry Hindu

Employing words to accomplish things, such as communicating something about the world. Whether that use is successful or not depends on a lot of things coming together in the right way.

Would that exclude language less creatures' belief from being well grounded? — creativesoul

I don't think so. You, as a language user, could in principle identify the premises of any belief and check if they're true (and thus whether the belief holder could be said to have knowledge). But those premises are true (or not) independently of whether anyone does identify them.

So, in Russell's clock example, the belief is justified but not well grounded? — creativesoul

Yes, where well-grounded means that the belief as well as all the premises that the belief depends on are true.

Yes, exactly. Particular conventions have emerged because they are useful. They represent the pragmatic middle ground between infinite checking and no checking at all.