One may describe what a neural net does in propositional terms, post hoc. But there are no propositions present in neural nets. Neural networks do not function by making use of propositions — Banno

No, they function by instituting normative patterns. This they have in common with our propositional terms, because their organizational basis is the condition of possibility of propositional grammar. You would have to eliminate the ‘net’ aspect of neural nets, removing the ascription of patterned organization to one’s neural model, in order to sever the normative equivalence between neurological and propositional.

More ellipsis. — Banno

I just don't see the point of being gnomic when doing philosophy, but you do you.

For instance, as you note,

sometimes we use propositions. — Banno

and you yourself intend someday to tell a story that begins "Once upon a time there was an entity with a neural network,..." and ends "And they used propositions happily ever after."

If you don't know the middle bit yet, that's understandable. I suppose people waiting for the next installment would like some reassurance that there will be a middle bit. "Blah blah blah, the end" is not a good story.

Ok, so we agree that this interchange between ourselves is not going anywhere.

...their organizational basis is the condition of possibility of propositional grammar. — Joshs

What's that mean?

Neural networks are not von neumann machines. They do not manipulate symbols, they modify weightings.

We agree on that, at least?

There is some number n where n >= 0 such that “there are n coins in the jar” is true even if nobody has counted them. — Michael

You are just begging the question Michael. Sure it is true that someone could count the coins, and determine how many there are. But until someone does, there is no such thing as the number of coins in the jar. This requires that someone draws (judges) a relation between a particular number, and the quantity which the jar contains. Until then, the number of coins in the jar is undetermined.

All you are doing here is mentioning every possible number, and saying that one of them will be the number of coins in the jar, if counted. So you are assuming that it is possible that the coins can be counted and you say that one of the infinite possibilities will match. I would assume the same thing. But I think it's quite obvious that until they are counted, there is no such thing as the number of coins in the jar. No specific number has been assigned to that assumed quantity. Therefore there is no number to that quantity. And what you are saying above, is that out of all the possible numbers, you are quite sure that one will prove to be the correct number. So you apply some logic to justify your claim that the coins can be counted and one number will prove to be the correct number. However, this does not produce the conclusion that there already is a correct number, as you seem to think it does.

And how do you account for two people making contradictory judgements, much like you and I here? Is it just the case that we disagree or is it also the case that one of us is right and one of us is wrong? — Michael

This is a good question. If the two people honestly believe what they are saying, and are stating it to the best of their capacity, they are both making true statements, regardless of the fact that they disagree with each other. This is why knowledge requires justification as well as truth. We move to resolve these disagreements through justification. Right and wrong are judgements based in justification, whereas "true" is a judgement based in what one honestly believes. You can see a lot of overlap here, and that is "knowledge" as justified true belief.

Neural networks are not von neumann machines. They do not manipulate symbols, they modify weightings.

We agree on that, at least? — Banno

Ok, I’ll go with that , even though there are other ways of describing their functioning. Getting away from a computational approach to neural modelling is a good start. Next step would be dumping representationalism.

I believe there is a Kantian distinction between the "thing in itself" and noumena..... — Janus

Absolutely.

the former is a purely formal or logical requirement to the effect that if there is something as perceived there must be a corresponding thing as it is in itself...... — Janus

Ok. I would rather think the ding an sich as merely an ontological necessity; if there is an affect on us by a thing, the thing-in-itself is given immediately by it. The only difference between a thing and a thing-in-itself.....is us. So your notion of formal and logical requirement is too strong, methinks.

.....Noumena I take to signify the general hidden or invisible nature of what is affecting us pre-cognitively such as to manifest as perceptual phenomena. — Janus

It doesn’t hurt anything to think noumena as you say, but that wouldn’t the Kantian distinction. Simply put, phenomena arise legitimately according to rules. Noumena arise illegitimately by overstepping the rules. Noumena are possible iff what we consider as rules by which our intelligence works, are themselves unfounded, which is of course, quite impossible to prove. Which leaves them as entirely possible to another kind of intelligence altogether. Who knows....maybe that stupid lion thinks in terms of non-sensuous intuition, such that for his kind noumena are the standard. Too bad we can’t just ask him, huh?

Noumena are not complicated; assembling and comprehending the antecedents against them, are.

A flat-earther can claim to know that the world is flat. He nonetheless doesn't know that.
— Andrew M

That's what you say. He says he knows it, you say he does not know it. It's your word against his. We can move to analyze the justification, and show that your belief is better justified than his, but this still doesn't tell us whether one or the other is true. And if you argue that his is not knowledge, it's not because his belief is not true that it's not knowledge, it's because it's not justified. So we cannot establish the relationship between knowledge and true, in this way. — Metaphysician Undercover

That's kicking the can down the road. The flat earther will say he is justified in making his claim, you say he is not justified. It's your word against his.

If anything which may turn out to be false in the future cannot be correctly called knowledge, then there is no such thing as knowledge, because we cannot exclude the possibility of mistake. — Metaphysician Undercover

Infallibility isn't a condition of knowledge, as ordinarily defined and used.

It's more accurate to define "knowledge" as the principles that one holds and believes, which they apply in making decisions. That is a person's knowledge, regardless of the fact that it may later turn out to be wrong. This way, we don't have to decide at a later date that the knowledge we held before wasn't really knowledge. — Metaphysician Undercover

If it is later decided that your "knowledge" was wrong, then that just is to decide that you didn't have knowledge, as ordinarily understood. Thus we have a translation between ordinary usage and your way of speaking.

And the knowledge we hold now will later turn out to be not knowledge, onward and onward so that there is no such thing as something we can truly call "knowledge" because we can never exclude the possibility of mistake — Metaphysician Undercover

By that argument, there is also no such thing as something we can truly call a "kettle" because we can never exclude the possibility of mistake.

Ok. I would rather think the ding an sich as merely an ontological necessity; if there is an affect on us by a thing, the thing-in-itself is given immediately by it. The only difference between a thing and a thing-in-itself.....is us. So your notion of formal and logical requirement is too strong, methinks. — Mww

I seem to remember reading Kant where he says that if there are representations, then there must be something that is represented. I had interpreted this as being seen by Kant as a logical entailment. You seem to be saying it is an ontological entailment, so I'm wondering if there is a difference.

It doesn’t hurt anything to think noumena as you say, but that wouldn’t the Kantian distinction. Simply put, phenomena arise legitimately according to rules. Noumena arise illegitimately by overstepping the rules. Noumena are possible iff what we consider as rules by which our intelligence works, are themselves unfounded, which is of course, quite impossible to prove. Which leaves them as entirely possible to another kind of intelligence altogether. Who knows....maybe that stupid lion thinks in terms of non-sensuous intuition, such that for his kind noumena are the standard. Too bad we can’t just ask him, huh? — Mww

This is quite a novel way (for me at least) to think of noumena, If the ding an sich is an "unknowable X", unknowable in the sense that what it is in itself cannot be known, then I had thought of noumena as simply the general unknowable. This because the thing in itself is still thought as a thing, but a thing considered not as it is for us, but in itself, whereas I took noumena to signify what is unknowable, beyond even being thought of as thing or things. However, I am no Kant scholar, merely someone who has read some of his CPR and secondary sources about it; and it's also been a while. I'm more trying to tease out what are the implications of Kant's ideas, what we might think is implicit in them rather than explicit.

So, I am struggling to understand what you mean by this: "Noumena arise illegitimately by overstepping the rules. Noumena are possible iff what we consider as rules by which our intelligence works, are themselves unfounded, which is of course, quite impossible to prove."

It is too bad we can't ask him!

Infallibility isn't a condition of knowledge, as ordinarily defined and used. — Andrew M

f it is later decided that your "knowledge" was wrong, then that just is to decide that you didn't have knowledge, as ordinarily understood. Thus we have a translation between ordinary usage and your way of speaking. — Andrew M

There seems to be a contradiction here. The second quoted passage seems to be saying that if what we thought was knowledge turns out not to be true, then it was never knowledge in the first place. Doesn't it follow that knowledge (as distinct from what we might think is knowledge) cannot be false; and thus that it is infallible?

What do you think....is there a definition other than the nominal, that defines what truth is? — Mww

I don't think so. Suppose Alice says that it is raining outside. There is no general criterion that we can use to determine the truth of her statement (i.e., independent of its specific context). Instead, we need to look at what the statement is about, in this case, the weather outside.

Regarding your question, what do you think?

There seems to be a contradiction here. The second quoted passage seems to be saying that if what we thought was knowledge turns out not to be true, then it was never knowledge in the first place. Doesn't it follow that knowledge (as distinct from what we might think is knowledge) cannot be false; and thus that it is infallible? — Janus

Yes, knowledge cannot be false. But human beings, being fallible, are always capable of making mistakes or being wrong.

For example, Alice claims it's raining outside as a result of looking out the window. We can conceive of ways that her claim can be false (say, Bob is hosing the window), and thus not knowledge. But if it is raining outside, then she has knowledge.

So infallibility is not a condition of knowledge, whereas truth is. Another way of putting it is that Cartesian certainty isn't a condition of knowledge.

So infallibility is not a condition of knowledge, whereas truth is. Another way of putting it is that Cartesian certainty isn't a condition of knowledge. — Andrew M

But isn't truth infallible in the sense of its being incapable of being false? Your reference to Cartesian certainty suggests to me that we may be talking at cross proposes, so I'm not proposing that possessing knowledge means that one knows one is infallibly correct, but that the knowledge we possess, if it is to be knowledge, must be infallible.

I have wondered whether it ought to be said that we possess knowledge in cases where we cannot be certain, that is when we do not know that we know, but that is a whole other can of worms.

But isn't truth infallible in the sense of its being incapable of being false? Your reference to Cartesian certainty suggests to me that we may be talking at cross proposes, so I'm not proposing that possessing knowledge means that one knows one is infallibly correct, but that the knowledge we possess, if it is to be knowledge, must be infallible. — Janus

I was referring specifically to human fallibility. I prefer to say that a true statement cannot be false, just as it cannot be raining outside and not raining outside. But word choice aside, we agree.

I have wondered whether it ought to be said that we possess knowledge in cases where we cannot be certain, that is when we do not know that we know, but that is a whole other can of worms. — Janus

I think it sometimes can (@Srap Tasmaner gives some examples earlier in the thread), even if they often are found together. There's an interesting quote by philosopher Timothy Williamson on that subject here.

Infallibility isn't a condition of knowledge, as ordinarily defined and used. — Andrew M

I'm not proposing that possessing knowledge means that one knows one is infallibly correct, but that the knowledge we possess, if it is to be knowledge, must be infallible. — Janus

If S's knowledge that p is infallible, then S "cannot be wrong" that p. If that's just to say it is not possible that S knows that p and yet ~p, sure, that's impossible.

If S knows that p, and if we consider only possible worlds consistent with S's total knowledge, then p is true at all of those. p is, for S, epistemically necessary. But that's not to say that p is metaphysically necessary, which means there's a sort of odd gap. Any ~p-worlds that might exist are just epistemically inaccessible to S.

And that strikes me as curious. My knowledge that p creates in me an incapacity -- I become unable to know that ~p. Which is as it should be, but imagine reversing the analysis: suppose I do not know that p, and suppose further that I am, for whatever reason, utterly incapable of knowing that ~p. Then p-worlds are, ceteris paribus, consistent with my total knowledge, and only ~p-worlds at which I do not know that ~p. (At none of those will I know p either, because ~p.) This inability to be epistemically committed to ~p seems to greatly increase the likelihood of my landing at a p-world and knowing it. An inability to be wrong doesn't guarantee that you will be right -- you may never come even to hold a belief regarding p either way, much less know the truth -- but it surely helps.

(It's also curious that because we're interested in the complement, the weaker the commitment to ~p you are unable to make, the better for your chances of knowing that p: excluding worlds at which you only believe that ~p would be better; excluding worlds at which you take seriously ~p but are undecided, better still; excluding worlds at which you merely entertain the notion that ~p, better still.)

David Lewis has a paper that addresses infallibility. I've not read it yet.

++++

Dots I forgot to connect.

There's nothing particularly interesting about being right when you're right. Being right means really, really not being wrong.

But when Roman Catholics say that the pope is infallible with regard to certain, though not all, matters, what they mean is not only that whatever he has said is right, but that whatever he will say is right. He is unable to be wrong in these matters.

So the point I was making above is that when you're right, you pick up -- for free -- that inability to be wrong on this matter, and that feels like it's in the neighborhood of infallibility, though it's really just what being right is.

And that's why the reverse is interesting. An inability to speak ungrammatically doesn't mean you produce every grammatical sentence, but that every sentence you produce is grammatical. If you were unable to make faulty inferences, you wouldn't have every reasonable belief, but every belief you had would be reasonable, given your total evidence.

Maybe that's not much to get excited about though. Sounds a bit like playing not to lose, which is a notoriously bad strategy.

Next step would be dumping representationalism. — Joshs

Good. So we agree to moving away from a computational, representational approach to neural networking.

Then in what way does

description-dependence (go) all the way down. — Joshs

Apologies to everyone for the much shorter responses than your interesting questions deserve. I have a work commitment this week which will require more than the usual amount of my attention...

Yes, broadly it's (2). What I'm really saying is that language doesn't seem to me to be very much in the business of 'representing' anything at all so much as the business of manipulating hidden states. It obviously derives from a model of how those hidden states will respond (otherwise the action on them would be random), but I wouldn't, myself, expect to see very much by way of reflection in the language of those motivating models.

So to answer your second post (or just make my position even more confusing!) there's a difference between "there's a a kettle" and "the kettle is boiling" that is not found in the grammatical structure of predication. I see them as two different expressions for two different jobs, rather than see one as reflecting a hidden state and the second a predicating something of it (that same hidden state reflected by the former). Hidden states are whole and dynamic, linguistic entities are discrete and static. So linguistic entities can't really reflect hidden states, but I don't think that prevents them from being about hidden states, just that the 'aboutness' might be two-way (not just reflection but aspirational). "there's a kettle" might mean something like the intention that other's should use the word kettle for that which I model as such, whereas "the kettle is boiling" might be more intended to get people to stand away from the object and it wouldn't have really mattered if I'd said "the pot is boiling" instead.

As such, it's difficult to see any analysis of the truth of "the kettle is boiling" as being based on anything other than a post hoc assumption that the expression predicates something of the same "kettle" we have in mind when conducting this analysis.

So given that, from your description of "hidden states" -- I'd say these things are absolutely not connected. First we don't even have concepts with your neural model, that's sort of just "assumed" to ride along with the firing of neurons. And then with all the causal language being used "noumena" seems wholly innappropriate as a boundary condition for this discussion. I'd say this falls under "empirical psychology", so the transcendental conditions of knowledge won't effect what we have to say here even if we are Kantians. — Moliere

Great, thanks for the insight. I think my conversations with @Mww have moved along similar lines (the lack of overlap), but I can also see where there might be space for such a notion in our meta-theories. Hidden states themselves suffer from the same problem in that simply by positing them as causal, we have identified them (and so they're not really hidden). They can't really play a direct role in perception as such, but only in a meta-theory about perception. I can't look inside someone's brain and then look at my hidden-state-o-meter and see the connection, I can only put 'hidden states' as a place holder in my meta-model of how models are made.

I suppose the error theorist's task, then, is to lay out what discriminates a fantasy from a purposeful story -- "story" in the sense of our ability to parse the world into story form, ala "purposeful fiction". — Moliere

Yes. This kind of work (on social narratives and their function) is what I used to do my research on. It's a fascinating field - but then I would say that wouldn't I?

it is also evident, to me at least, that our language and how we conceive mentality does not match up, in any simple way, with this description. Now what? — Srap Tasmaner

Well... that's a massive question that deserves more time than I currently have for it. But... I think it leads us back to where I first interjected. If we're not conducting any kind of empirical investigation (nor constraining our models by the results of any such) then we're perhaps constructing an entity more like maths where axiomatic choices are made and consequences follow, but without any hooks in reality (as far as my limited understanding of maths goes). That's certainly as entertaining a pastime as any other, but it leaves us, much like maths, with judgements like 'elegance' or 'coherence' as our targets for a good model, rather than the more boring 'pragmatic utility' of the empirical investigation. All still fine so far, until... I aesthetically prefer my models to be pragmatic. My desire for a system to have pragmatic hooks into empirical sciences isn't dogmatic, it's aesthetic. I just like my theories that way. so any contribution I might make to the purely 'philosophical' constructions of how the world might be is going to end up that way whichever route we take to judging a theory's merit.

What I think we need to be careful about, is thinking the mismatch between a particular scientific model, on the one hand, and a philosophical one, on the other, indicates that one has not sufficiently slurped up the other yet, but it will. It's that "if all you have is a hammer" thing. — Srap Tasmaner

I agree. I think where we might differ is that I'd be more tempted to see the rogue philosophical position as a narrative, where you might have it more as an analytical truth?

"purposeful fictions" still contains the problematic "fiction"; I wonder if "narrative" would be better, leaping from non-symbolic to symbolic representation. Or perhaps "invention", we invent the kettle from the hidden state; but that loses something of the cooperative aspect. — Banno

I like 'narrative', but I've been told I use the word too often. I feel a renewed permission to revert to it now, though!

We must at some stage look for a bone of contention between us; It'll be something to do with the move from a neural net to a narrative. To my eye, building on Searle, at some stage there is a move from a hidden state to a narrative about a kettle, that has a logical form something like "This hidden state counts as a kettle"... — Banno

Even here though... I like '...counts as'. It covers a lot of what I was trying to get across to @fdrake in answering his questions above. The idea that speech is doing a job, in this case declarative - 'we'll treat this as a kettle'. It's declaring that any discrepancies we might have in resulting from whatever behaviours our neural networks are currently resulting in toward that hidden state, we should put them aside in favour of the more collaborative 'kettle'.

I believe there is a Kantian distinction between the "thing in itself" and noumena; the former is a purely formal or logical requirement to the effect that if there is something as perceived there must be a corresponding thing as it is in itself. .'Noumena' I take to signify the general hidden or invisible nature of what is affecting us pre-cognitively such as to manifest as perceptual phenomena. — Janus

Thanks. So 'noumena' might be closer to hidden states in that respect, but I'd be interested to hear what you think of what @Moliere says about the problem of causality. Hidden states are definitely considered causal.

But until someone does, there is no such thing as the number of coins in the jar. — Metaphysician Undercover

So are you saying that the number of coins in the jar is in some sort of superposition of all possible numbers until someone counts them?

Forget the word "true" for the moment: what kind of (meta)physics are you suggesting describes the nature of the world?

You are just begging the question Michael. — Metaphysician Undercover

I am asserting what our best understanding of the world entails. You brought up quantum mechanics earlier to support your argument, so you appear to accept the findings of scientific enquiry, and the findings of scientific enquiry are that the number of coins in the jar isn't in a superposition of all possible numbers until counted.

I would say that you are begging the question, saying that "there is no such thing as the number of coins in the jar [until counted]" without any evidence or reasoning.

The flat earther will say he is justified in making his claim, you say he is not justified. It's your word against his. — Andrew M

Right, but saying "I'm justified" is not acceptable justification. Nor is an appeal to authority, or to the norms of our society.

Infallibility isn't a condition of knowledge, as ordinarily defined and used. — Andrew M

Infallibility is a condition of "truth" as you use it, and "truth" is a condition of knowledge. So infallibility is a condition of knowledge, under those terms.

If it is later decided that your "knowledge" was wrong, then that just is to decide that you didn't have knowledge, as ordinarily understood. — Andrew M

No, I don't think that's right. People just change their minds about things, and sometimes if encouraged to, they will admit to having previously been wrong. But they are more inclined to say that their information (knowledge), was incorrect when the mistake was made, than that they didn't have knowledge at the time. This is because having incorrect knowledge allows them to pass blame elsewhere, toward the source of that incorrect knowledge.

If we look back, in retrospect, we see two possible principal causes of mistaken action, one being a lack of knowledge, the other being incorrect knowledge. The two are not the same, and we must maintain a distinction between them to be able to understand and prevent the causes of mistake in the future. You seem to be claiming that there is no difference between these two in ordinary usage of "knowledge", as if people don't distinguish between lack of knowledge and incorrect knowledge when assigning blame, and in other situations.

By that argument, there is also no such thing as something we can truly call a "kettle" because we can never exclude the possibility of mistake. — Andrew M

No, this is an incorrect conclusion as well. I define "true" with honesty. So if one honestly believes the item is "a kettle" then the person will truly call it a kettle, despite the fact that someone else might truly call it "une bouilloire". Excluding the possibility of mistake is not required for a human being to speak truthfully. That is supposed to be a feature of God, but not human beings.

So are you saying that the number of coins in the jar is in some sort of superposition of all possible numbers until someone counts them? — Michael

No, not really. I used the example of quantum mechanics to elucidate the type of problems which adhering to your principles brings about. We look at an electron as a particle, and we think, a particle has a determinate location, therefore the electron has a determinate location. The issue at hand is that "determinate" is not the same thing as "determinable".

Now, if you and I look at the jar of coins, you would say that there is a determinate number of coins in the jar, and I would say that there is a determinable number of coins in the jar. I differ from you, because I insist that an act of determination (measurement) is required to determine the number of coins, before we can truthfully say that the number is determinate. You seem to take this act of determination for granted, as if there is already a number assigned to the coins without an act of measurement. That, I say is a mistake. There is no number already assigned to the coins prior to being counted, just like there is no location already assigned to the electron prior to being determined.

Taking things like this for granted is a pragmatic principle which is extremely useful. If we want to know the quantity of coins, we assume that there is such a thing as the quantity of coins, therefore it is determinable, and we are motivated to count them. Further, we can use this assumption to state premises for logical procedures, like you did, which I said was begging the question. However, that there is a determinate quantity is just an assumption, which is not completely justified until after the count. So in the case of the electron, we might assume it is a particle, therefore it has a location which is determinate, and then we could proceed to determine it. In the end though, the original assumption, that the electron is a particle with a determinate location, is never justified. What is justified is that it has a determinable location

What this demonstrates, is that we must be very wary of these pragmatic principles, which we accept without proper justification, for the sake of facilitating our logical procedures. The principles are extremely useful, and even appearing to be infallible in the circumstances where they are heavily used, and so they appear to be universal. But then, when we expand the use of them, because of that appearance of universality, outside their range of applicability, this misleads us. Because the principle is so extremely useful, and infallible in its original application (there is a determinate number of coins in the jar), we are not inclined to use the tool of skepticism, to question that premise and see where it is faulty.

Forget the word "true" for the moment: what kind of (meta)physics are you suggesting describes the nature of the world? — Michael

Come on Michael, if I knew the answer to that, I'd have reality all figured out. And of course, so would everyone else because when one person gets it right everyone else jumps on board. I think metaphysics is an inquiry into the best way to describe the nature of the world. If one already knew the best way there would be no need for inquiry.

I am asserting what our best understanding of the world entails. You brought up quantum mechanics earlier to support your argument, so you appear to accept the findings of scientific enquiry, and the findings of scientific enquiry are that the number of coins in the jar isn't in a superposition of all possible numbers until counted.

I would say that you are begging the question, saying that "there is no such thing as the number of coins in the jar [until counted]" without any evidence or reasoning. — Michael

Look at it this way, it's very simple really. The truth of the phrase "the number of coins in the jar" implies that there is one specific number attached to, associated with, or related to, the quantity of coins in the jar. Can you agree with that? Now do you honestly believe that a particular number has already been singled out, and related to the quantity of coins in the jar, prior to them being counted? How is that possible?

I think the real problem here is that we've come from a tradition of religious beliefs. We have a religious history. So there are many old principles that we now take for granted, which are only properly supported by the concept of God. Isaac Newton for example, stated that his first law of motion (which we tend to take for granted), relies on the Will of God. This is similar to what I think about your belief, that there is a specific number already associated with the coins in the jar. It is a belief which was developed under the assumption that God has already counted them, and assigned that number to the quantity of coins in the jar.

Surely we know our collective fiction (which is the model, which is the world) in exactly the same way, and with the same level of surety, that we know Aragorn was king of Gondor. So, whence surprise?
— Luke

From hidden states. — Isaac

Hidden states are the world, right? (However, you also say that the model is the world?)

Who said anything about the world surprising us? — Isaac

I have been trying to.

Since it is possible that our model could be false in at least some respects, and that we could be surprised, it follows that there is more to truth than a mere "collective fiction".
— Luke

This is only true if the terms are interchangeable (that truth is about the model being surprising), otherwise your conclusion doesn't follow, hence you begged the question by assuming that relation in your argument for it. — Isaac

That's not at all what I've been attempting to say. I reject this reading that "truth is about the model being surprising".

What I have been saying is that if the model is the world which is a collective fiction, then there should be no surprises. This "collective fiction" view is your account of redunancy, yes? Do you agree that there are no surprises with the truth of the collective fiction that "Aragorn was king of Gondor"? If so, then I don't see why the same should not extend to all (other) truths. According to redundancy, therefore, there should be no surprises. What I mean by a "surprise" is that our expectations are not met. But if it is our collective fiction, then why would our expectations not be met? We should always expect that "Aragorn was king of Gondor" is true, right?

However, in opposition to this, you also state that "the hidden states the world is a collective model of may be modelled imperfectly". This imperfect modelling indicates that occasionally our expectations may not be met. And that indicates a problem for the "collective fiction" view of redundancy. If it is the view of redundancy that there are no surprises and our expectations are always met because of our collective fiction, then there should be no "imperfect modelling".

I take the position of redundancy to be that there are no matters outside of language, and that the model is equivalent to the world, whether that is your personal view or not.
— Luke

I really don't know where you're getting that idea from. — Isaac

I'm getting it from my understanding of redundancy. If "p is true" is no more than "p", then the model and the "hidden states" are one and the same. In other words, there are no hidden states, only the model; only the collective fiction. Hence my charge of relativism. There should be no place for a "better model" according to redundancy. I see this as being the reason @Srap Tasmaner asks what makes it a "better" model, instead of merely a free or random change to the existing/previous model.

A better model indicates that what makes "p" true (or "more" true) is having/getting our model of the world perfect, or closer to perfection. It means that "p is true" is not just whatever we call "p" at a given time; but is instead the best version of "p" of all time - the "p" that perfectly models the world.

Once the world and model come apart, then it is no longer redundancy/deflationism (at least, as I understand it).

There is no number already assigned to the coins prior to being counted, just like there is no location already assigned to the electron prior to being determined. — Metaphysician Undercover

But as I said, the findings of science are that the position of an electron isn't like the number of coins in the jar. The former is in a superposition, the latter is not. If you want to use science to support your position then you cannot pick and choose which bits you like.

Now do you honestly believe that a particular number has already been singled out, and related to the quantity of coins in the jar, prior to them being counted? — Metaphysician Undercover

I'm not even sure what you're asking. If you're asking if somebody has determined the number of coins before somebody has determined the number of coins, then of course not. If you're asking if there is some number of coins before somebody has determined the number of coins, then yes.

Your argument seems to commit a fallacy of equivocation.

he says that if there are representations, then there must be something that is represented....... — Janus

Yes. The text is rife with affirmations.
“....I cannot rest in the mere intuitions, but—if they are to become cognitions—must refer them, as representations, to something, as object, and must determine the latter by means of the former...”

.......I had interpreted this as being seen by Kant as a logical entailment. You seem to be saying it is an ontological entailment, so I'm wondering if there is a difference. — Janus

Again, no interpretive harm done, even if there is a great contextual and methodological difference. I would still hesitate to agree Kant sees it that way, for the very first paragraph of the text.....

“.....For how is it possible that the faculty of cognition should be awakened into exercise otherwise than by means of objects which affect our senses, and partly of themselves produce representations...”

.....indicates an ontological necessity, while the logical entailment, re: the possibility of awakening, resides in the system itself. But we see the awakening of the system constantly, and its negation is impossible, both entirely logical entailments, but irrelevant to its theoretical operation.
—————

I took noumena to signify what is unknowable, beyond even being thought of as thing or things. — Janus

Noumena do represent what is unknowable, but for different reason than the ding an sich is unknowable. The latter merely from lack of immediate access by us, its objective reality being given, the former from the impossibility of its objective reality being given, to which the access is then moot.

Regarding beyond even being thought of as things....

“....The understanding, when it terms an object in a certain relation phenomenon, at the same time forms out of this relation a representation or notion of an object in itself, and hence believes that it can form also conceptions of such objects....”

....it is clear noumena are thought as things, insofar as thought is the origin of conceptions of things. Briefly, understanding treats phenomena, which are mere representations of objects, as objects themselves, which it has no warrant to do, for the arrangement of the matter of objects into a form is the purview of intuition alone.

Another thing: representation formed out of relation, is not the same as representation formed out of sensation. In effect, when understanding thinks a representation as object in itself, the entire sensory apparatus is bypassed, which means the faculty of sensuous intuition, the kind we actually possess, is idle. For us, then, this has two consequences negating the possibility of experience itself. First, going forward in the methodology, if cognition of objects requires sensuous intuition, and it is idle, no empirical cognition is at all possible. Hence, noumena as un-sensible objects in themselves, are necessarily uncognizable, which is the same as unknowable. And second, when going backward in the methodology, if sensuous intuition is idle, and it is the case that the awakening of our cognitive system depends exclusively on the appearance in sensation of real objects, the objective reality of such non-sensuous objects in themselves can never be given, which is also the same as such objects being unknowable.

“....But I can think what I please, provided only I do not contradict myself...”, and we can see, thinking noumena is not contradictory, but to attribute substance, reality or even adjoin concepts to them, is a contradiction of the systemic methodology itself.

what are the implications of Kant's ideas, what we might think is implicit in them — Janus

Personally, my opinion is, first and foremost, that a priori conditions are not only possible, but necessary, and second, given from the first, that the human intellectual system has a natural, intrinsic, thus inescapable, duality.

I am no Kant scholar, merely someone who has read some of his CPR and secondary sources about it.... — Janus

Same here. Anything anybody says, even the relation of textual citations, with respect to Kantian metaphysics is no more than his own opinion.

What do you think....is there a definition other than the nominal, that defines what truth is?
— Mww

I don't think so. Regarding your question, what do you think? — Andrew M

I think I’m going to backtrack, unapologetically I might add. While you did get me to think above and beyond my cognitive prejudices, I found support for my original claim, truth is that in which a cognition conforms to its object (A58/B82), here.....

“...But although these rules of the understanding are not only à priori true, but the very source of all truth, that is, of the accordance of our cognition with objects...”

......found at A237/B296, quite obviously further along in the methodological thesis, so shouldn’t it be taken for granted he means an answer to “what is truth?”, which must be a definition of it, to be just that? To repeat what he doesn’t mean would be disastrous.

On the other hand, perhaps one could reject that “truth is.....”, is technically sufficient as a definition, but is rather merely an exposition of the conditions which make all truths possible. But the rejoinder to that would be that’s precisely what a definition does, serves as the criterion for the validity of any conception.

Personal choice, then?

The truth of the phrase "the number of coins in the jar" implies that there is one specific number attached to, associated with, or related to, the quantity of coins in the jar. Can you agree with that? Now do you honestly believe that a particular number has already been singled out, and related to the quantity of coins in the jar, prior to them being counted? How is that possible? — Metaphysician Undercover

A jar of coins either has no coins in it, or some coins in it. For the moment only, assume there is no other possible state for a jar. (We need neither claim nor stipulate that the number of coins in an empty jar is 0.) If a jar has no coins in it, we cannot remove a coin from it; if a jar has some coins in it, we can remove a coin from it, and If we were to remove one coin, then again the jar would have in it either no coins or some coins. This we know because a jar must have no coins in it or some coins in it. We count, from 1, as we remove coins from the jar, stopping when there are no coins in the jar; if the procedure does not terminate, then there is no number of coins in the jar. If the procedure terminates, then the number we have reached is the number of coins that were in the jar before we started counting.

The only difficulty we face is determining what it means for a coin to be in the jar. If the jar is quite full, so that some coins rest on other coins but above the lip of the jar -- that is, outside the space we think of as bounded by the jar -- shall we count those as in the jar or not? If a coin is partially within the space bounded by the jar and partially outside that space, shall we say the coin is in the jar or not? If our jar of coins is in such a problematic state, then our counting procedure is of no use until we agree which coins will be considered to be in the jar. If we cannot agree which coins to count, there is no point in counting them. Similar considerations apply to what is a coin.

But if we do agree what to count as a coin and which coins to count, we know there is a procedure available, and that we will be able to determine the number of coins currently in the jar, even if we have not yet made that determination.

Proof that such a procedure, if it yields an answer, must yield a unique answer, is left to the reader.

I can only put 'hidden states' as a place holder in my meta-model of how models are made. — Isaac

....and I can only put noumena as a placeholder in a meta-theory of how other intelligences function.

Risky business, indeed.

Good. So we agree to moving away from a computational, representational approach to neural networking.

Then in what way does
description-dependence (go) all the way down.
— Joshs — Banno

I am linking description (space of reasons, account, value system) , to scheme , scheme to pattern and pattern to reciprocal network of relations. Tying all of these together within an enactivist approach are a connected set of concepts characteristic of autonomous living systems: organizational and operation closure and sensory-motor structural coupling between organism and environment.

“ Organizational closure refers to the self-referential (circular and recursive) network of relations that defines the system as a unity, and operational closure to the reentrant and recurrent dynamics of such a system.

…autonomous systems do not operate on the basis of internal representations in the sub-jectivist/objectivist sense. Instead of internally representing an external world in some Cartesian sense, they enact an environment inseparable from their own structure and actions . In phenomenological language, they constitute (disclose) a world that bears the stamp of their own structure.”(Thompson , Mind in Life)

It is not as though any particular description or account is split off from the environment it interacts
with and organizes. This is a two-way street. A network of relations defining a space of reasons or the pattern of a neural net is in a relation of reciprocal
causality with the world of material processes.

@Joshs

Good. So we agree to moving away from a computational, representational approach to neural networking. — Joshs

Can you explain the difference?

I am linking description (space of reasons, account, value system) , to scheme , scheme to pattern and pattern to reciprocal network of relations. — Joshs

So what you were saying is that it is patterns all the way down, but miscalling them descriptions.

Sure, neural networks make use of patterns. They do not make use of names or propositions.

Even here though... I like '...counts as'. — Isaac

One supposes that this counting as is the result of neural processes yet need not be located in any particular process. There need be nothing in common, perhaps, in the neural patterns that enable one to make a cup of tea and the neural process that enables one to order quality Russian Caravan from an online supplier. Yet both are to do with tea.

Hence, anomalous monism.

And

In contrast, the latter thinks that Davidsonian "physical properties" and "the micro-structural level" are just theoretical suppositions that are meaningful only within a description or vocabulary. — Joshs

would be to claim that neural science is imaginary...

David Lewis has a paper that addresses infallibility. I've not read it yet. — Srap Tasmaner

Nor I. But I agree with Lewis that the standards of knowledge depend on the context:

Lewis argues that S knows that p is true iff S is in a position to rule out all possibilities in which p is false. But when we say S knows that p, we don’t mean to quantify over all possibilities there are, only over the salient possibilities.
...
The kind of position Lewis defends here, which came to be known as contextualism, has been a central focus of inquiry in epistemology for the last fifteen years. “Elusive Knowledge”, along with papers such as Cohen (1986) and DeRose (1995) founded this research program. — David Lewis - SEP

The flat earther will say he is justified in making his claim, you say he is not justified. It's your word against his.
— Andrew M

Right, but saying "I'm justified" is not acceptable justification. Nor is an appeal to authority, or to the norms of our society. — Metaphysician Undercover

The flat-earther is not claiming it is. He will point to what he regards as evidence for a flat earth. Is his claim thereby justified?

Infallibility is a condition of "truth" as you use it, and "truth" is a condition of knowledge. So infallibility is a condition of knowledge, under those terms. — Metaphysician Undercover

Let me put it differently: Cartesian certainty is not a condition of knowledge, as ordinarily defined and used.

I define "true" with honesty. So if one honestly believes the item is "a kettle" then the person will truly call it a kettle, despite the fact that someone else might truly call it "une bouilloire". — Metaphysician Undercover

The issue is not about what language one uses to refer to a kettle. It's that someone can conceivably, and honestly, mistake something for being a kettle that is not, or for not being a kettle when it is.

Excluding the possibility of mistake is not required for a human being to speak truthfully. That is supposed to be a feature of God, but not human beings. — Metaphysician Undercover

That's exactly the point. Someone might be mistaken about whether the object before them is a kettle. Similarly someone might be mistaken about whether they have knowledge. People can make honest mistakes. They thought it was a kettle when it wasn't. They thought they knew something when they didn't.