Isaac - would you actually agree that "you burn your hand if you touch the kettle" would be the same as ""you burn your hand if you touch the kettle" is true" though? — fdrake

Well, yes, to a point. Although the topic here is 'truth' I think the reason we've diverged so much is that the simple redundancy that there's nothing more to "P is true" than "P" tends to lead to an assumption that that's all there is to say on the matter of truth. Whereas, to quote Ramsey "there are interesting problems in the vicinity".

The issue of the degree to which (and implications thereof) external states constrain our use of language that tries to refer to them is just such a 'problem in the vicinity', in my opinion.

I see language as a tool, part of our suite of 'active states', which themselves from part of the Markov boundary, but (by necessity) active states are influenced only by internal states (if they weren't they wouldn't be active states, they'd be sensory states), so we have a chain of external states>sensory states>internal states>active states(language use).

If we accept that model, then the extent to which language mirrors external states is, it seems, not entirely dependant on sensory states, but rather on the intent of active states. To use an analogy with perception, I see language more like saccades than V1 modelling, part of the active state response, not the passive state reception.

What do you see as being a significant difference between the "hidden states" that give rise to our models or collective representations, and the noumena that are represented as phenomena? Or perhaps Isaac, if he agrees with you, can answer that question in a more informed way than you can. — Janus

I don't feel qualified to comment on the potential differences because I wouldn't claim to know very much about Kant's noumena. From a complete layman perspective though, Kant's noumena are often referred to as the thing-in-itself, yes? Taking that literally (perhaps erroneously, though) I think the difference would be in that hidden states do not posit any 'thing' at all, they are an informational construct, about data, not material composition. As such they can be an implication of a data model, whereas any thing-in-itself would be ontological? But as I say, I'm not sure as I don't have a deep understanding of noumena.

If we accept that model, then the extent to which language mirrors external states is, it seems, not entirely dependant on sensory states, but rather on the intent of active states. To use an analogy with perception, I see language more like saccades than V1 modelling, part of the active state response, not the passive state reception. — Isaac

It's quite difficult for me to tell precisely how this limits your agreement with ""P" is true iff P". I can think of two possibilities, but I doubt they are exhaustive.

( 1 ) The first is that because our engagement with the world is simultaneously pragmatic and representational; passive states feed to active states feed to environmental interventions feed to passive states...; representational accuracy is not the only criterion by which the environment is parsed, so the environmental objects and processes which are referred to in "the kettle is boiling" aren't really environmental since what is referred to is part of the internal states of someone's body/brain/mind.

Your difficulty then comes with the interpretation of equating the kettle boiling with a state of the environment. The equation itself could have two subcases:
( 1 a ) Strict identity; this is undermined by the fact that the process is representational rather than presenting a numerical identity between brain-states/body-states. In other words, there is no one-one correspondence between "brain-stuff" (activation patterns, mind states) and "environmental stuff" (water molecules heating up etc) because modelling provides an inferential summary rather than an unfiltered presentation of environmental stimuli.

(1 b) Representational equivalence; by this I term I mean that the aggregate of brain and body states in the inferential summary in ( 1 a ) counts as an environmental object sortally, even though one is a brain-body state and one is an environmental state. You could "count" the kettle and the class of brain-states regarding it as "the same thing", just one is represented (environmental object) and one is the product of representation (internal state). This is undermined because the inferential summary is pragmatic.

( 2 ) It doesn't matter that the modelling relationship is inferential and pragmatic rather than presentational, what matters is that the states identified as perceptual ones are internal (of the brain, body) rather than external (of the environment).

In both of these cases, the content of language is treated as constrained by the how the content of perception is generated, since the semantic content of any expression is a historically informed inferential summary of internal states which is also an internal state. The distinction representation (models) and symbolisation (expression) doesn't matter for your analysis of truth, since they're both either not the appropriate sort of equivalence as in ( 1 ) or merely internal as in ( 2 ).

The distinction between ( 1 ) and ( 2 ) is that ( 1 ) cares about the properties of the internal state (which block statement truth somehow), and ( 2 ) just cares about the property that they are internal (which blocks statement truth by itself).

That's how we determine the truth of a proposition, through judgement. How could the truth of a proposition be determined, except by a judgement? — Metaphysician Undercover

A proposition being true and a proposition being determined to be true are two different things. There is a correct answer to "how many coins are in the jar?" before we actually count them.

Actually, what you've just stated, that one must be right and the other wrong, is just a judgement itself, made by you, as Mww has already pointed out. — Metaphysician Undercover

It's not just a judgement. See above.

A proposition being true and a proposition being determined to be true are two different things. — Michael

That's exactly what I don't agree with obviously.

There is a correct answer to "how many coins are in the jar?" before we actually count them. — Michael

If you think so, explain to me who has counted the coins in the jar and stated the answer. An "answer" is something stated as a reply to a question. If no one has counted the coins, and it was not determined at the time of placing the coins in the jar, and the jar has been watched, then no one knows how many there are, and no one has stated the "correct answer". The "correct answer" will be determined when the coins are counted. Therefore, there is not a correct answer to that question before the coins are counted. The number of coins is undetermined, no "correct answer".

This, I propose to you, is where we cross the line between honesty and dishonesty in our philosophy. We honestly know that unless the number has been determined, there is no correct answer. The correct answer is undetermined, it does not exist. However, we assume that since the coins could be counted, there is potentially a correct answer, and we allow that this potential answer has actual existence, and we say as you do, "there is a correct answer". This, I tell you, is a dishonesty, because we know very well that there is a difference between what actually exists and what potentially exists, yet we allow this division to be nullified, because it simplifies our use of mathematical language. We do not have to account for the process of counting, (See the difference between actually and potentially infinite for example). The abundant consequences of this sort of dishonesty are very evident in the issues of quantum mechanics.

It's not just a judgement. See above. — Michael

That's right, it's not just a judgement, it's a special type of judgement, a dishonest judgement made for the sake of facilitating our language use, especially mathematical languages (See above). When we are well convinced that "the truth" could be determined, we jump to the dishonest conclusion that the truth already is determined, for the sake of avoiding philosophical discussion about the required process of determination. The fact that this is a mistake is fully exposed in quantum mechanics. The particle's location really is not determined before the process of determination, and it is obviously mistaken to think that it is. Therefore it is only the process of determination (the act of measurement) which can determine "the correct answer". And the character of that assumption, that there is a correct measurement, prior to the measurement being made, is fully exposed for the lazy, and dishonest, attitude that it truly is.

An "answer" is something stated as a reply to a question. If no one has counted the coins, and it was not determined at the time of placing the coins in the jar, and the jar has been watched, then no one knows how many there are, and no one has stated the "correct answer" — Metaphysician Undercover

I can say "there are 66 coins in the jar" and that claim can be true even if I haven't counted the coins in the jar and even if nobody knows how many coins are in the jar.

It's not the case that my claim retroactively becomes either true or false after someone has counted them. And it's not the case that if two people count the coins in the jar and come to a different conclusion that both of them are right.

The fact that this is a mistake is fully exposed in quantum mechanics. The particle's location really is not determined before the process of determination, and it is obviously mistaken to think that it is. Therefore it is only the process of determination (the act of measurement) which can determine "the correct answer". — Metaphysician Undercover

We're not talking about quantum states though. It's not the case that the number of coins in the jar is in a superposition of all possible numbers until they're counted.

Your account of truth appears inconsistent with the (meta)physics of the world.

I can say "there are 66 coins in the jar" and that claim can be true even if I haven't counted the coins in the jar and even if nobody knows how many coins are in the jar. — Michael

So you say, but as I explained, I think you are being dishonest in your statement. You are using "can be true" which honestly implies possibility, to make it appear as if "66" actually is true. It is not.

To "be true" is very clearly a judgement. And if no one has counted the coins, who do you propose has made that judgement, God? Obviously, you are not even proposing that the judgement has been made, you only say "can be true", meaning it is possibly true. Yes, just like 65, 64, etc., are all potentially true answers. But that does not justify the claim that there is a true answer.

It's not the case that my claim retroactively becomes either true or false after someone has counted them. And it's not the case that if two people count the coins in the jar and come to a different conclusion that both of them are right. — Michael

This does not address the issue. Prior to being counted your answer, 66, "can be true", meaning it has the potential to be true, just like other numbers. When the coins are counted, there is a correct answer. There is no retroactivity involved. Prior to being counted, all the answers had the potential to be true, and after counting, the judgement is made.

Retroactivity is the mistaken route which the others proposed. After determining that what was accepted as "knowledge" is determined to be incorrect, they propose that we retroactively declare that it was really not knowledge. But then everything which we commonly call "knowledge", may at any moment, be shown to be not knowledge. Retroactive judgements is a mistaken venture.

This issue was very well discussed by Aristotle thousands of years ago. His solution was an exception to the law of excluded middle, to account for the reality of potential. So things which require a judgement, like the famous sea battle example, are neither true nor false, prior to the judgement being made. "There will be a sea battle tomorrow" is neither true nor false. And, after tomorrow passes, we do not retroactively say that there was a truth or falsity to that statement yesterday. We simply must face the fact that there neither is a truth nor a falsity to these statements which require a judgement, prior to the judgement being made.

We're not talking about quantum states though. — Michael

I used quantum mechanics as an example of how that type of dishonest thinking, which you display, causes problems in application. The particle "can" (potentially) have a location, but that does not justify the claim that it does actually have a location.

I take nominal to mean that the definition can't be employed to establish which statements are true — Andrew M

Ahhhh....now I see how you related classical to nominal. My go-to reference doesn’t use nominal to qualify the definition, so thanks for that.

Actually, we end up with...the nominal definition, the one used by classical logicians including Aristotle, ad recounted in Kant, re: truth is that cognition which conforms to its object, can’t be employed to establish which statements are true. But “what is truth?” isn’t asking about statements, it is asking after the conditio sine qua non with respect to the truth of our empirical cognitions in general, and insofar as the nominal definition involves a circularity, according to Kant and as stated in last part of the pg 45 comment...there isn’t any by means of the use of it.

Hence, the implication of what definitions are supposed to do, re: “...the criterion of the possibility of a conception is the definition of it...”, and because truth is a valid conception, hence its possibility is given, it must meet the criterion of being defined. Which is what the “...assumed as granted and thereby presupposed” is meant to indicate. What do you think....is there a definition other than the nominal, that defines what truth is?

I don't feel qualified to comment on the potential differences because I wouldn't claim to know very much about Kant's noumena. From a complete layman perspective though, Kant's noumena are often referred to as the thing-in-itself, yes? Taking that literally (perhaps erroneously, though) I think the difference would be in that hidden states do not posit any 'thing' at all, they are an informational construct, about data, not material composition. As such they can be an implication of a data model, whereas any thing-in-itself would be ontological? But as I say, I'm not sure as I don't have a deep understanding of noumena. — Isaac

Perhaps, with our powers combined, we could come up with something that works for us. Obviously to make these comparisons one has to have an interpretation of Kant, so there's going to be some controversy with respect to which interpretation we're favoring. But if we don't mind stirring that pot and wanting to have some kind of rough idea, I'd claim I have some knowledge of the noumena. (EDIT: heh, well... as a philosophical concept, at least! :D I'd be contradicting myself the other way...)


In my understanding of the distinction we have to step back and look at the philosophical landscape of the time to see what sorts of debates were being taken seriously by philosophers: Is space relative, or absolute? Are we free, or are we determined by the laws of physics? Does God exist? Is the soul immortal?

From the particular examples that Kant works through we can see that his target is metaphysical theories. Further, these metaphysical theories are demonstrated to be undecidable since the only way we settle whether some statement is true is by referring to what we collectively experience, and these particular theories and judgments attempt to get "outside" of our experience and assert the truth of things we have no connection to.

By "no connection", I always harp on the fact that one of the categories is "causation", and the noumena is outside of the categories, and so no we cannot make sense of the noumena by applying the category of causation to it -- it does not cause phenomena. With respect to our scientific knowledge, at least, it's a purely negative category (with respect to the other two powers of the mind, practical reason and aesthetic/teleological judgment, the noumena plays a different role -- but with respect to scientific knowledge, it's purely negative)

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.

is that they're purposeful fictions. — Isaac

I like this notion of purposeful fictions.

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". That might go some way to making this notion more appealing.

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.

Your account of truth depends on a (meta)physics that isn’t the case. The number of coins in the jar isn’t in a superposition of all possibilities until someone has made a judgement.

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?

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

Thanks for the input. From what I understand of the discussion with @Isaac and @Janus though, a state (or collection of states) being modelled is a precondition of being able to count as knowledge, experience, representation, an object etc. It can be construed to be inappropriate to even say that something is a kettle because, allegedly (if I've read it right), what is related to is not an environmental object but inferential summary of the model's current state. It's something the modelling process is doing, not a direct relationship between mind and world. And we can't 'get behind' the modelling process to "sneak up to the noumenon" so to speak.

I agree with you that it's actually an inappropriate comparison with the noumenon, but my claim is that @Isaac and @Janus seem to be operating under the assumption that such an inferential summary has broadly Kantian import. By that I mean the process of modelling plays the role of the categories+schematism, and the internal+receptive states play the roles of appearance, phenomenon and the modelling process itself plays the role of the faculty of sensible intuition. In total this renders us unable to reach "beyond" experience into the object since experience is equivalent to the modelled state, just as an object is always and only given through/into sensible intuition.

I agree with you that you don't have to interpret the hidden state/internal state/external state/sensory state account with the internal/external split detailed above, but if you do treat it like that internal and external look a lot like phenomenon/noumenon (even when it doesn't have to be interpreted that way). I believe this misalignment is fundamental in keeping this discussion unresolved.

@Srap Tasmaner - might be interested in post above here.

I can say "there are 66 coins in the jar" and that claim can be true even if I haven't counted the coins in the jar and even if nobody knows how many coins are in the jar. — Michael

You seem to be missing the point that @Metaphysician Undercover is an anti-realist, and his account of truth is some version of verificationism. (For meaning, it is sufficient that the coins can be counted, but for truth, it must actually have been done.)

Thus, as a self-organising system, we must, by definition, have internal states, and boundary states (and there must exist external states). Without these three states we cannot say that there is a system at all, we cannot define it from 'not-system' without defining a boundary and (as far as data is concerned) that boundary must be Markov boundary if the internal network is any more complex than a single ring of nodes. — Isaac

I can't imagine disagreeing with any of this.

But 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?

One option is to say, well, we've moved on. Our languages are the ossification of a folk psychology that we know better than now. Of course we'll have trouble expressing this new view of things in the terms of the old paradigm. You can even soften the pitch a little by claiming only that the new view is different, rather than less wrong, but the hope is still that it is a more fruitful paradigm for inquiry. There must be reason to switch, and problems with the old paradigm provide plenty of motivation there.

But I think it's not that simple. There is an almost irresistible temptation to identify mentality with the internal states of such a system -- a sort of "what else could it be?" But much of the last fifty years of philosophy in the English-speaking world has been devoted to showing that this identification is mistaken. This cluster of issues became important precisely because of the promise of early work in artificial intelligence, generative linguistics, and brain science -- everything that would become cognitive science -- and the realization of some philosophers that we might be able to say we had finally found the mind, and it is the brain. Endless debate ensued, some of which continues, and some of which seems very old-fashioned now. But in the meantime, other philosophers noticed that our mental concepts and all of our language, in fact, seem not to respect this apparently natural identification of the mental with the internal. I find those arguments pretty persuasive.

There's no problem for science here. If you tell a scientist that what he's investigating turns out not to be X, he can shrug and go on investigating whatever it is he is investigating. Whether it's X is not really his concern. He may have been considering writing a popular piece for Scientific American explaining how his work changes our understanding of X, and now he can just not do that and spend his time in the lab instead. (Occasionally a scientist will decide that showing up philosophers is part of the job.)

But there's still plenty for philosophy to worry about. Science, so far, may have failed to straight up solve issues of mentality for us, but it has perhaps sharpened (at least indirectly) what those issues are.

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.

All of which is to say that knowledge, for instance, is not a relation that holds or fails to hold between the internal and inferred external data nodes of a self-organizing system. Apples and oranges.

might be interested — fdrake

Will be reading back through this latest run of posts and maybe commenting. Threw in something else in the meantime.

Do you want to keep truth functions?

The idea here (for me) is that despite having to posit hidden states as part of our informational meta-thoery (see my post to Srap above), these states can still be proper objects of reference. "the kettle" doesn't refer to my model of a kettle, it refers to (in the informational model) the hidden state itself. It's like us all speculating what's in the room next door. the subject of our speculations isn't our speculations, the subject of our speculations is what's actually in the room next door. As such, the best way I can find of 'translating' an active inference model to talk of "kettles" is to say that "kettle" refers (when it refers at all, that is - not all uses are referenential, of course) to the hidden state we're modelling, the contents of the room we're speculating about. — Isaac

I'm even more confused by this! Can the hidden state denoted by "the kettle" boil? Even though boiling is a collective fiction? The first predicates boiling of the hidden state, the second predicates boiling of the collective or an agent or a modelling system (the kettle boils vs the collective fiction has the kettle boiling).

"the kettle" doesn't refer to my model of a kettle, it refers to (in the informational model) the hidden state itself. — Isaac

One can see why @Joshs mistakes this for the thing-in-itself, or some such.

"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.

I recall a month or so back a conversation in which it was said (possibly Joshs, again) that the mind creates reality, and we asked the obvious question, if mind creates reality, what does it create it from? Here you are answering that question, showing how the kettle is created by a neural net that interacts with stuff outside it.

Thanks for your reply. 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"...

"the kettle" doesn't refer to my model of a kettle, it refers to (in the informational model) the hidden state itself.
— Isaac

One can see why Joshs mistakes this for the thing-in-itself, or some such. — Banno

If one interprets this in Davidsonian terms, then the ‘thing in itself’ equates to real physical properties. I agree with Rorty and Putnam that description-dependence goes all the way down.

“The difference between a Davidsonian non reductive physicalist and a Rortyan naturalistic pragmatist is that the former does not deny that there really are physical properties at the micro-structural level, because the efficiency of a physical vocabulary is a sufficient reason to extend its claims to ontology. 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. They think that it is sufficient for a denial of the existence of physical properties at the level of ontology, precisely because they are still description-dependent.”(ELIMINATIVE MATERIALISM ELIMINATED:
RORTY AND DAVIDSON ON THE MIND-WORLD RELATION, Istvan Danka)

I recall a month or so back a conversation in which it was said (possibly Joshs, again) that the mind creates reality, and we asked the obvious question, if mind creates reality, what does it create it from? Here you are answering that question, showing how the kettle is created by a neural net that interacts with stuff outside it. — Banno

And I would add that if the net is the description, then the outside stuff the neural net interacts with cannot be assumed to harbor any features, properties, substance in themselves that are description-independent.

Fair enough; you've so little else to work with. — Banno

Yes, you have given so little to work with when it comes to just what you are wanting to say, beyond bare assertion and aspersion, Remember I've claimed no expertise regarding what the idea of "hidden states" entails. That's what I'm trying to find out in order to compare it with what I know the idea of noumena entails; which is basically that we are pre-cognitively affected, and that we have no conceptual purchase on what constitutes that pre-cognitive affect. The "hidden" in 'hidden states' seems to be suggesting a similar idea.

As such they can be an implication of a data model, whereas any thing-in-itself would be ontological? But as I say, I'm not sure as I don't have a deep understanding of noumena. — Isaac

@Mww

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.

I can't speak for Isaac, but I think you have my position pretty much right, except that I would say that we cannot help in ordinary discourse having the hidden state count as a, for example, kettle. Whether we call it a kettle, a hidden state or a noumenon, though, is a matter of what "language game"; we happen to be playing and is a matter of stipulation, not of fact.

What does seem to be a fact is that we are pre-cognitively affected and that we have, and can have, no conceptual grasp of that process. This is where I lose patience with Banno; as he seems to be simply asserting, ad nauseum, that it is a matter of fact that it is a kettle. But then I'm not sure what is behind his apparent position, since he offers no detailed reasoning.

...description-dependence goes all the way down. — Joshs

Neural networks do not use propositions. Hence, some explanation will be needed if they are "description-dependent".

Neural networks do not use propositions. Hence, some explanation will be needed if they are "description-dependent". — Banno

Now this is in agreement with the idea of noumena, which are understood to be affecting us, but not in any way dependent on descriptions (conceptualization), nor in any way amenable to being described.

Neural networks do not use propositions. Hence, some explanation will be needed if they are "description-dependent". — Banno

Neural networks instantiate patterns of normatively oriented practical engagement with a world. One can also think of these patterns in terms of forms of description, accounts, schemes , values. Propositions are one peculiar, culturally contingent linguistic product of such normative patterns.

Now this is in agreement with the idea of noumena — Janus

No, it isn't.

No, it isn't. — Banno

Yes it is. See I can play that stupid game too.

I can play that stupid game too. — Janus

Hey, you win.

No, you do. Or at least I don't feel I've won anything.

One can also think of these patterns in terms of forms of description, accounts, schemes... — Joshs

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.

But there are no propositions present in neural nets. — Banno

There is somewhere that propositions are present?

More ellipsis.

Yes, sometimes we use propositions.