I'm a physicalist simply because it seems a default for me, and I need a good reason to discard it. — Isaac

Simple: the colors, sounds, smells, tastes and feels aren't properties of the physical environment you interact with. Or at least not when it comes to our physical models.

Or to say it a better way, nobody has succeeded in explaining how they are.

Simple: the colors, sounds, smells, tastes and feels aren't properties of the physical environment you interact with. Or at least not when it comes to our physical models. — Marchesk

But they're properties of my brain. I mean, when brains are interfered with those things respond differently, so I don't see that as a reason to discard physicalism.

But they're properties of my brain. I mean, when brains are interfered with those things respond differently, so I don't see that as a reason to discard physicalism. — Isaac

They may be in reality, but brain models of neurons and neurotransmitters don't include sensation. That's just a correlation or outcome that we know exists from having brains.

The problem according to the esteemed Nagel is that subjective experiences (what it is to be a conscious being) are unique - "single point of view" in his words. Thus, since objectivity, a necessity for any physicalist theory, would be forced to ignore subjective experiences this will, in effect, make such theories incapable of explaining consciousness - the phenomenon they were built to explain.

What I don't understand is how and why "single points of view" (subjective experiences) precludes objectivity?

Is Nagel relying on the definitions of "objectivity" and "subjectivity" when he says any objective theory would be forced to exclude subjective experiences? This seems wrong since we may objectively study subjectivity or subjectively study objectivity without running into problems.

Is Nagel saying something more powerful in terms of relevance to his argument that it's impossible to view subjectivity under an objective lens? How did he come to have this belief?

Frankly I'm puzzled.

Could this not be the case the with the true coral perimeter? — Isaac

I mean it might be? To my mind there are two types of comparisons. Model output comparisons, like predictions or parameter estimates to other predictions or parameter estimates (how similar are these two models?), and comparison of model output to data (how similar are the predictions or parameter estimates to those in the data?). We only know "the true values" of the parameters in the data if we've simulated the data ourselves.

Regardless, I think that, like in Friston's approach, the model errors don't behave like predictions or parameter estimates, they're modelled as coming from some distribution, and when we take input data or perturb an external state as the result of an outputted prediction, there is some error which comes from a comparison of an observable whose value is not an output of some model. That's $T$ or $e$ "when exactly specified" in the coral measurement error model, the 'input data' in the coral model would be $L$.

Edit: a good rule of thumb in my book is that error terms crop up in models whenever they have a dependence on something external to it. If any model, even the human body-brain system has an 'error', there's an external source (a process which does not output model internal predictions or parameter estimates, but instead outputs observables/data that relate to the modelling process) associated with it.

I believe Nagel was saying that science uses an objective view from nowhere (perspective-less or lacking subjectivity) to create explanations. But the subjective doesn't fit into these explanations. And yet subjectivity is part of the world.

If the universe went bang, and stars formed fusing heavier elements leading to life evolving, then somehow subjectivity emerged.

pieces can be removed from an item, but ‘moments. cannot be removed — I like sushi

I go with matter can be removed but form cannot.
————————

but I do find it interesting even though I don’t grasp what he meant/means exactly. — I like sushi

There doesn’t seem to be any general consensus in the literature for either Brentano’s or Husserl’s intended meaning for noesis and noema either, so you’re not alone. Seems to me they’re trying to give the mind some character, instead of treating it as an abstract apex placeholder.
————————

There is certainly more than a hint at the distinction between ‘act’ and ‘object’ yet they are more like different sides of the same ‘object of intentionality’. — I like sushi

Again, for me this reduces to the distinction between intuition and appearance. As Brentano claims, “...Every mental phenomenon is characterized by (...) the intentional (or mental) inexistence of an object, and what we might call (...) reference to a content, direction towards an object (which is not to be understood here as meaning a thing), or immanent objectivity...”.

Everybody says the same thing, just in different ways.

Good luck with the fresh perspective.

They may be in reality, but brain models of neurons and neurotransmitters don't include sensation. That's just a correlation or outcome that we know exists from having brains. — Marchesk

This is another thing that keeps cropping up in discussions involving neuroscience that baffles me. What isn't just a correlation? If I throw a ball in the air on the moon, compared to the earth its relative falling speed is 'just a correlation' with the gravitational mass of the respective planets. Switching my light on is 'just a correlation' with illuminating my room. What is it that's marking out the correlation between neural activity and mental phenomena that singles it out for such unique inductive doubt?

Perhaps you’re drawing off the fact science can monitor the stimulus/reaction complex, and thereby affirming the antecedent, by re-naming the stimuli as disposition to act, hence, belief. (?)
— Mww

In a sense, yes, but any problem with doing so would only arise from a position that some previous definition existed whose only flaw was its inability to be thus monitored, and I'm not at all convinced that such a definition existed. — Isaac

Here, I suppose such a problem would arise, because if belief is held to be a subjective institution, re: judgement, and thereby defined with a priori predicates alone, it certainly cannot lend itself to empirical monitoring.

there is some error which comes from a comparison of an observable whose value is not an output of some model. — fdrake

OK, I'm quite taken by the factor that errors are modelled as coming from some distribution, that's definitely an important thing for model dependent realism to account for, but... (you knew there'd be a but, right?)

there is some error which comes from a comparison of an observable whose value is not an output of some model. — fdrake

...how? How do we get access to an observable that's not an output of some model? We can't take in any sensory data without it being merely confirmatory of some model.

We could filter data back out again, possibly?

We don't accept raw data, any raw data, so all raw data is both selected and filtered through some model. If we knew what the model did, could we recreate the raw data by extrapolating backwards? Is that what you're getting at by pointing out the equivalent position of e to L in forming T?

I don't know of any neuroscientific support for the idea, I mean we could theoretically 'see' such recal happening because it'd need forward driving connections from the sensory processing cortices, but it sounds possible.

Here, I suppose such a problem would arise, because if belief is held to be a subjective institution, re: judgement, and thereby defined with a priori predicates alone, it certainly cannot lend itself to empirical monitoring. — Mww

Yes, in a sense that's the reason I think it better to 'black box' the whole thing and look at the behaviour as indicative of what the whatever-it-is does to the inputs. I just can't see the need for descriptions of static mental states, they're not something I experience.

...how? How do we get access to an observable that's not an output of some model? We can't take in any sensory data without it being merely confirmatory of some model. — Isaac

It's like $y = mx + c + e$, the independent variable $x$ doesn't have a model. It's playing the part of data in the model. Like when an error is observed in Friston's hierarchy and passed down. A prediction is compared to some input data, the error propagates down the hierarchy of models, producing adjustments.

I believe Nagel was saying that science uses an objective view from nowhere (perspective-less or lacking subjectivity) to create explanations. But the subjective doesn't fit into these explanations. And yet subjectivity is part of the world.

If the universe went bang, and stars formed fusing heavier elements leading to life evolving, then somehow subjectivity emerged. — Marchesk

Aah! Thanks for the explanation.

I wonder if that's entirely true. Scientific objectivity doesn't mean you ignore essential and defining aspects, here subjective experiences, of the object of study. Rather scientific objectivity is specifically designed to eliminate observer bias and in no way does it/should it overlook, in this case, subjective experiences.

I may be completely wrong here but the claim here seems to be, if you take the argument to its logical conclusion, nothing that is ever subjective can be scientifically studied due to objectivity being necessary in science. While this may somehow shield consciousness from attack it also makes consciousness a woo-woo pseudoscience. I don't know which is more preferable here - following Nagel into pseudoscience or give up one of my favorite beliefs that there is something about consciousness that defies the physical.

It’s given that everything human, happens because of the brain. I reject out of hand that what it means to be human, can be discovered on an o’scope. And I reject it because I just don’t like it; it’s anathema to my ego. Pretty piss-poor reason for rejecting a form of relative proof, I know, but tell you what.....even if you get your

look at the behaviour as indicative of what the whatever-it-is does to the inputs — Isaac

, you haven’t really learned what you set out to discover, insofar as you’ve got the “how” but not necessarily the “how come”, because your subject himself may not even know.

I was under the impression methodological naturalism was created to circumvent the likelihood of error... — Mww

If you understood me as implying anything to the contrary, we ought chalk it up to poor writing on my part.

I wonder if that's entirely true. Scientific objectivity doesn't mean you ignore essential and defining aspects, here subjective experiences, of the object of study. Rather scientific objectivity is specifically designed to eliminate observer bias and in no way does it/should it overlook, in this case, subjective experiences. — TheMadFool

Well, it's because certain properties of experience are perceiver-dependent and not in the objects themselves. The air feels cold, but that doesn't mean it is cold in an objective sense (the feeling of cold varies between individuals and time). It just feels cold to you now.

The categories don’t determine errors, and we don’t choose them. Errors arise from irrational or illogical associations the subject thinks, and categories are merely the theoretical conditions which make cognition possible, so they exist in their entirely long before we give them names. Obviously, because we always cognize first, speak later, and never the reverse, about any one thing.

The idea of categories solves a problem, If you think it just causes another one, that would be on you, wouldn’t it? — Mww

This bit began with my pointing out that identifying the premisses is key. You agreed, then remarked that it is often the case that there is an unrecognized categorical error at work, such that the conclusions/model does not necessarily follow from those premisses.

I'm wondering about the grounds for charging another's position/argument/reasoning with such an error. If it is a categorical error, then I presume that is one kind of error. Specifically an error in categorization.

Is that much right?

Do you believe that we can be mistaken about that which exists in it's entirety prior to common language use?
— creativesoul

Depends what you mean by mistaken. — Isaac

Being mistaken about X is forming, having, and/or holding false belief regarding X.

Having pain is the experience. I have direct access to having pain of my own, and I have indirect access to another's. There are two kinds of accessibility here, yet you've claimed we have none.

Having pain is the experience. I have direct access to having pain of my own, and I have indirect access to another's. There are two kinds of accessibility here, yet you've claimed we have none. — creativesoul

Yeah, I should have said indirect. But it's also the case that we don't always have that indirect knowledge. It might exist in really subtle physiological cues, but we can't read brain activity that accurately, and we haven't put chips in everyone's heads yet. But sure, often enough we have some evidence to what people are experiencing.

...categories are merely the theoretical conditions which make cognition possible, so they exist in their entirely long before we give them names. — Mww

So, we must surely abandon Kantian language here. For anything that exists in it's entirety prior to our naming and descriptive practices qualifies for that which exists in and of itself(Noumena). Furthermore, we must have knowledge of both our premisses and that which exists in it's entirety prior to our premisses in order to perform a comparative assessment between the two. That comparison is required in order to know that one has indeed committed the error called "categorical".

Looking back, I see I could have registered the statement without including a mistake you wouldn’t have made.

You claimed that you do not believe other people exist. You're now speaking about believing a range of things in different contexts, and not believing a single thing...

Are you familiar with the notion of performative contradictions?
— creativesoul

Yes, but I don't see how it applies here, you'd have to flesh the argument out. — Isaac

Simply put, you've claimed to think X but not believe X. In addition, you've claimed to not think that others exist, and yet here you are...

So, we agree that fleshing out the premisses supporting one's conclusions is key. Based upon the latest exchanges, I'm curious about what you're referring to when you wrote "unrecognized categorical error such that the conclusions do not necessarily follow from the premisses"...

Example?

Pretty much, with the caveat that “categorization” might not carry the proper inflection. One shouldn’t confuse speculative categories such as Aristotle’s or Kant’s, with Ryle’s semantic categorical mistakes. Rush’s song “Time Stands Still” is a categorical mistake of semantics; space is a property of objects is a categorical error of reason.

Pretty much, with the caveat that “categorization” might not carry the proper inflection. One shouldn’t confuse speculative categories such as Aristotle’s or Kant’s, with Ryle’s semantic categorical mistakes. Rush’s song “Time Stands Still” is a categorical mistake of semantics; space is a property of objects is a categorical error of reason. — Mww

We have before us now, a listing of categorical errors...

Which ones will be used to render judgment upon whether or not a premiss of our choosing qualifies as being guilty of categorical error as compared/contrasted to other kinds of errors?

Like when an error is observed in Friston's hierarchy and passed down. A prediction is compared to some input data, the error propagates down the hierarchy of models, producing adjustments. — fdrake

Yes, but in Friston's model the sensory input is from perception, not the world, and that's vitally important for the free-energy principle to work. There only any variance minimising incentive between the perception and the cortices which model the causes of that perception. As Friston himself says "There's no reason at all why the Bayes optimal model would be true, truth is not important here". The error that's propagating is one between the perception (low level hierarchy model) and the beliefs (higher level hierarchy model). There's nothing to cause the lowest level hierarchy model to minimise variance, not are there any mechanisms by which it could.

I reject it because I just don’t like it; it’s anathema to my ego. Pretty piss-poor reason for rejecting a form of relative proof, I know — Mww

On the contrary, I'd go as far as to say it's the only justifiable reason to reject anything that isn't overwhelmingly contradicted by the evidence to the contrary. Pick your model and defend it until it's indefensible. I don't think we can handle any other way of approaching uncertainty.

, you haven’t really learned what you set out to discover, insofar as you’ve got the “how” but not necessarily the “how come”, because your subject himself may not even know. — Mww

I don't think the 'how come' is a measurable thing by any metric, so it's not a relevant investigation. Pick your reason.

Being mistaken about X is forming, having, and/or holding false belief regarding X. — creativesoul

That's just tautotlogy. What makes a belief false?

Simply put, you've claimed to think X but not believe X. In addition, you've claimed to not think that others exist, and yet here you are... — creativesoul

Did you not read what I wrote, or not understand it, or not agree with it. If you're not going to actually respond in any way to what I write there's little point in continuing is there?

"There's no reason at all why the Bayes optimal model would be true, truth is not important here" — Isaac

This comment is discussing box 1 and 2 in the linked paper.

Aye, I agree with this! It even looks like something of a category error in terms of the internal state; when watching a bike move, is it represented as having wheels as a unity moving in tandem or two wheels linked through the bike? But there's still questions of accuracy and adaptation which are relevant. If you have a goal of opening a door, your hand position needs to adapt to where the handle is and how it works; there's an accuracy constraint involved with the door's location, functionality and so on; irrelevant of how it's split up into perceptual features.

There's an interface of perceptual features that are associated with external states. If I've read it right there's some function $f$ that "specifies the dynamics of external causes" dependent upon specific external causes $\gamma$, our processual model of them $f$, our proposed actions $\alpha$ and with some error $w$. Our actions promote certain external causes; which actions are chosen depends on previous internal states/sensations/actions/external causes, our sensations/perceptions process external causes; how they are processed depends on previous internal states/sensations/actions/external causes.

Focus for a moment on the $\gamma$, if I've read things right these are "hidden causal states" which are later associated with environmental parameters $\theta$ rather than available sufficient statistics $x$, again if I've read it right. They are hidden, but inferred upon by the whole active-modelling process under some representation ("recognition dynamics"?).

Our ability to act well in an environment depends upon having a good model of it; to update our model through some error minimising in response to our current goals and current environment; the model doesn't just take input from previous modelling steps, it takes input from external states with their own dynamics under some processual representation*. It would be unable to guide action if it did not have a satisfactory (sufficiently accurate) representation* of the external states as they are relevant to our goals and bodily constraints.

I think you're emphasising that the starred representation* ($f$, I think in the paper) is another flavour of model; which it is; but it's also an observation process of relevant structures for us in the environment (we sample in accordance with it). What would be the point of all this modelling without its ability to promote accurate, relevant action - pulling causal levers whose structures we partially represent in the world? Like a codification of the world into what's relevant and available to us, and accuracy-prone (evaluable) perception and action within that codification.