So you keep saying, but you've not given any account of why a dress cannot be both a red dress and a blue dress. — Isaac

I haven't claimed that it cannot. In fact I explicitly said above that it can. Maybe I should repeat myself?

A dress can be red and blue. Or it can just be red. Or it can just be blue. Or it can be some other colour or combination of colours.

I haven't claimed that it cannot. — Michael

...

A red dress isn't a blue dress. — Michael

I'm asking for support for the above assertion.

Saying that a red dress isn't a blue dress isn't saying that a dress can't be both red and blue.

Here's a red dress:



Here's a blue dress:



Here's a red and blue dress:

Yep. I'm not asking about dresses which are part red and part blue. I'm asking about dresses which are both all red and all blue.

I'm asking about dresses which are both all red and all blue — Isaac

If red and blue are different colours then it is a contradiction for it to be all red and all blue.

If red and blue are different colours then it is a contradiction for it to be all red and all blue. — Michael

Why? Why can a dress not be two different colours at the same time?

Why can a dress not be two different colours at the same time? — Isaac

Because that's a contradiction. You might as well ask why something can't be both a rabbit and a duck.

Because that's a contradiction. You might as well ask why something can't be both a rabbit and a duck. — Michael

Is it? I don't see that. A duck and a rabbit are two different sets of cells/organs. No one claims to see a duck where others see a rabbit (in real 3d objects) so we've no reason to assume anything can be both.

With colour, some people do claim to see blue where others see red, so that's default reason to believe that colour is the sort of property which can be of two kinds at the same time. We've no similar reason to believe species is such a property. If we did (half the world claimed some ambiguous object was a duck and half claimed it was a rabbit) then we'd have good reason to believe that species was the sort of property which can be of two kinds at the same time too.

The constellation Orion is in both the shape of a man with a bow, and the shape of a smiling cat. The shape of the constellation Orion is the sort of property which can be of two kinds at the same time.

With colour, some people do claim to see blue where others see red, so that's default reason to believe that colour is the sort of property which can be of two kinds at the same time. — Isaac

No, it’s a reason to believe that colour is in the head.

No, it’s a reason to believe that colour is in the head. — Michael

Why?

Because the colour one sees is determined by how one’s brain responds to signals from one’s eyes. The same external stimulation but different colour experience. Therefore colour isn’t a property of that external stimulation or of whatever is responsible for that external stimulation.

the colour one sees is determined by how one’s brain responds to signals from one’s eyes. The same external stimulation but different colour experience. Therefore colours isn’t a property of that external stimulation. — Michael

The conclusion just doesn't follow. A hidden state might have the property of causing one response in person A but a different response in person B. That would still be a single intrinsic property of the hidden state. There's nothing at all preventing us from calling that property its 'colour'.

You're invoking this notion of a 'colour experience' without any warrant. There's no activity in the occipital cortex corresponding to a colour experience. There's no evidence for it at all, and there's bags of evidence against it (the vast majority of hidden states cause exactly the same responses in almost all humans - excellent evidence that the colour is a property of the hidden state, not the person's mind )

Almost every human in the world agrees that the postbox in the village is the same colour as the bus.

We need an explanation for this extraordinary consistency.

We also have very, very rare cases where people disagree that two objects are the same colour.

The simplest explanation of the extraordinary consistency is that the colour is a property of some external state which we all interpret and the very, very rare cases are either mistakes or odd hidden states.

You're taking these extremely rare cases and saying that our entire explanation needs changing, an entirely fabricated notion of color experience needs to be proposed, without any physical evidence it even exists... Just to avoid the much simpler explanation that these rare cases are just that. Rare oddities.

We need an explanation for this extraordinary consistency — Isaac

The explanation is that they scatter light at a wavelength of 650nm and that when light of this wavelength stimulates the sense receptors in our eyes it triggers brain activity from which visual experience emerges and that colour is a quality of these visual experiences. Given that we tend to have the same eye and brain structure and given that physical processes are mostly deterministic the quality of our visual experiences are mostly the same. And when someone has tetrachromacy or brain damage or the like then they respond differently to the same stimulus and so the quality of the emergent visual experiences are different, i.e they see different colours.

Whereas your theory requires this “hidden state” invention and the requirement that the external stimulus has as many “hidden states” as there are ways of experiencing it. It’s overly complicated, there’s no evidence for it, and I would even say it’s incomprehensible. Something can’t be both all red and all blue. The fact that your theory requires this should show how problematic it is.

An even simpler example: fire causes most of us to feel pain. Pain isn’t some external “hidden state”; it’s a quality of our experience. Colour is of the same kind.

People seem so bewitched by the complexity of visual experiences that they think sight works differently to other senses. If a dog could talk it would probably make the same mistake about smell.

An even simpler example: fire causes most of us to feel pain. Pain isn’t a “hidden state” of some external thing; it’s a quality of our experIence. Colour is of the same kind.

People seem so bewitched by the complexity of visual experiences that they think sight works differently to other senses. If a dog could talk it would probably make the same mistake about smell. — Michael

:up:

Patterns emerge and are reinforced or altered in actual
contexts of interaction, rather than in rules or properties that supposedly exist before or outside of actual contexts
— Joshs

So why would those patterns emerge variable? What causes the variance? — Isaac

Variance cause variance. As Deleuze said, a change is a difference that makes a difference. Laws and patterned regularities are idealizations of continuous
qualitative change.

The atoms do collude together to form a teacup. That's why we can all see them as a teacup. That's why one of the available gestalts is that of a teacup. Because the atoms do indeed form the shape of a teacup. They also form the shape of dozens of other things which we ignore, choosing, instead, to focus on the teacup option. But it's wrong to say they're not in the form of a teacup just because they're also in the form of many other options. — Isaac

It’s not just a matter of shapes that atoms
form , but of the relationship between accounts and the varying senses of concepts like shape, size, space. Of these potentially infinite variety of accounts, are you giving priority to a certain empirical account from physics? Is this a ‘bedrock’ account, as Quine claimed, one which grounds all the others in an irreducibly real beginning?

I think Wayfarer might agree that the way to bedrock is to begin by asking what all possible accounts of any aspect of the world have in common, that is , what is the condition of possibility of empirical account-building?
I think an answer compatible with naturalism is possible, but it requires a naturalism utilizing recent models from biology, centering on niche construction. As the thinking goes ( I draw from Joseph Rouse here), linguistic conceptual accounts of the world are elaborations of practical perceptual interactions that are continuous with the role of niche building in non-linguistic animals under selective evolutionary pressure. I don’t think such models warrant taking an account from physics as normatively determinative. That’s a way of saying that physics is just beginning to take into account the temporal notions that Darwinism has contributed.

A Gestalt picture does not merely bind separate objects together, but creates an entirely new complex entity which did not exist before. It creates a new world of hierarchically structured new objects—a world which could not exist without Gestalt perception.

Our biologically-designed model of reality is thus superposed on the physical stuff of the world and structures it. It is with this reality that we interact.

It is useful to compare Pinter’s model of gestalt with that of Merleau-Ponty. For Pinter , a gestalt “is a complex of images that we may call the concept of dog. Visually, it includes a fluid composite image of what the most commonly seen dogs look like, viewed in various positions and from different angles. It includes general notions of the temperament and character of dogs, and the knowledge that dogs may be both loving and, in some circumstances, dangerous.”

From MP’s vantage , this definition of gestalt remains embedded in objective naturalism, which explains gestalts a s the result of a causal process of concept formation. For MP, as for the later Wittgenstein, a gestalt is never a move from the particular to the general. It neither involves composition nor decomposition.
A perceived object has its sense in relation to a background gestalt field as its irreducible basis. A gestalt is both subjective and intersubjective, producing perceptual and social fields of action.

Pinter’s contention that gestalts are ‘superimposed on the physical stuff of the world’ is problematic, and suggests that he, like Quine, take the results of physics to be normatively determinative. Only this orientation, I believe, can justify belief in the 'unreasonable effectiveness' of mathematics. We have to begin with a reified notion of the object which transcends the relativity of Pinter’s subjective gestalts in order to see mathematical logic as unreasonably effective. If you’re looking for a transcendental basis for scientific truth you’re better off with Putnam’s valuative realism than with an idealization of the objectivity of natural objects.

Interesting, but I wouldn’t judge Pinter’s book on the few excerpts I have provided alone. He doesn’t say anything about Wigner’s essay (although he does mention him a couple of times.) That is my own conjecture. I’m not looking for a transcendental basis for scientific truth so much as trying to understand what ‘transcendent’ and ‘transcendental’ mean. I certainly don’t want to posit physics as ‘normatively determinative’, and I don’t think Pinter does that, either.

Another snippet:

Sensations, beliefs, imaginings and feelings are often referred to as figments, that is, creations of the mind. A mental image is taken to be something less than real: For one thing, it has no material substance and is impossible to detect except in the mind of the perceiver. It is true that sensations are caused by electrochemical events in a brain, but when experienced by a living mind, sensations are decisively different in kind from electrons in motion. They are indeed “figments” because they exist nowhere except in awareness. As a matter of fact, they exist only as claims made by sentient beings, with no material evidence to back up those claims. Indeed, brain scans reveal electrical activity, but do not display sensations or inner experience. — Mind and the Cosmic Order

Why this appeals to me, is that I’ve always argued that reason is the relationship of ideas. Nothing physical comes into it, indeed, we can’t even form a notion of ‘physical’ without employing reason. So whereas physicalism wants to claim that reason is derived from or supervenes on or can be reduced to neurological data, what he is saying is that is real on a different plane altogether. And that lends support to one of the basic themes I’ve been working on since joined here: that existence and reality are not the same thing.

(I should add, Pinter also says that all the basic concepts of physics like velocity, mass, etc, are ultimately derived from the embodied experience of resistance, lifting, movement, and so on. In that sense they’re ultimately visceral in origin.)

1. Some object is red1 if it causes most humans to see red2, — Michael

Again, it pays to consider a wide range of examples. I think your argument here has complications caused by colour being a secondary quality. Try making the same point with a primary quality instead - does it still work?

SO the eggs might be rendered something like:

"Some object is an ovoid if it causes most humans to see an ovoid ..."

Is that something you wish to assert? Because it seems to me to be wrong.

@Isaac, arguments concerning colour tend to be futile.

Why do you say that? — Metaphysician Undercover

I say that because that's how neural nets work.

Knowledge of how neural nets work will not inform you as to whether or not they are representational. This would require knowledge of "why". Knowing how the human vocal chords produce the sounds which are words, for example, does not provide you with information as to whether or not the sounds are representational. Again, this is fundamental to "meaning".

Knowledge of how neural nets work will not inform you as to whether or not they are representational. — Metaphysician Undercover

Yeah, it does. If you set up a neural net to, say, add two numbers, nothing in the processing represents the numbers in the way that von Neumann architecture does. That's rather the point of connectionism.

Cheers.

Sorry Banno, but you're not making any sense at all. There is a vast multitude of different ways of representing. That a process is not representational in a specific way, does not mean that it is not representational in an absolute way.

Connectionist models are just representations, so they are built for a reason other than the reason for which the thing represented was built, as they are built to represent that thing.

And I'll tell you one more time, knowing how the thing works allows you to build a representative model of it. But this provides you with no information as to whether or not the thing represented is itself a representation. A child can copy a word without even knowing that it's a word. To determine whether the action is representational requires a knowledge of why the thing is doing what it does. And that is completely different from why the model of the thing does what it does, because they are produced for different reasons.

I certainly don’t want to posit physics as ‘normatively determinative ’, and I don’t think Pinter does that, either. — Wayfarer

Let me use Pinter’s own words to make the case that his is a classic neo-Kantian appearance-reality dualism.

To begin with, one of his chapter headings is ‘Do We See the World Realistically?’Does he mean by this that it makes no sense to distinguish between human or animal perceptions of the world, and the world as it supposedly is in some factual naturalistic sense? Or does he mean that there is a real world with intrinsic properties which sensate creatures do not represent accurately? It sounds like the latter to me:

“What an animal experiences seeing may be unlike a high-fidelity reproduction of reality, with all its complexity and inscrutability…. so long as all the experiences a creature has with objects are consistent with one another— with no discrepancies of any kind—the creature is far better off interacting in mind with usefully simplified and schematized replicas.”

“It is no different when you set out to solve a technical problem involving a real-world situation: You don’t want a photograph of the objective situation, but a diagram showing just the necessary information.”

Studies have show that “faithful representation is driven to extinction by non-veridical strategies based on utility rather than objective reality.”

“So long as its segmentation is self-consistent, the animal cannot ever become aware of a difference between its world-model and reality.”

So our accounts are a replica , a non-faithful representation of the ‘physical facts’ of an ‘objective situation’.

“Though our segmentation of reality is partly bound to physical facts, much of it is arbitrary.”

Pinter uses as an example of this arbitrary association between external reality and our conceptions of it the difference between color and wavelengths of light. The former are subjective representations and the latter are the physical facts.

“There is no logical connection between perceived color and the wavelength of light: It is an arbitrary association invented by nature.”

Neo-Kantianism courts skepticism because our representational filters prevent us from seeing the world as it is.

You might prefer to say the representation is distributed. In any case, it is clear that what goes on in a neural network looking at a cup is not at all like a cup, as you said. That's a basic misunderstanding. Th neural net is not making a model that you then see with your mind. It is your mind seeing.

The point has been made before, by myself and @Isaac.

First let me say I really appreciate the care you've taken to raise those points.

I don't think Pinter juxtaposes a real, physical world, with a world of appearances. It's not as if the real thing is hiding behind the sensory depiction of it. The first words in the book are:

Imagine that all life has vanished from the universe, but everything else is undisturbed. Matter is scattered about in space in the same way as it is now, there is sunlight, there are stars, planets and galaxies—but all of it is unseen. There is no human or animal eye to cast a glance at objects, hence nothing is discerned, recognized or even noticed. Objects in the unobserved universe have no shape, color or individual appearance, because shape and appearance are created by minds. Nor do they have features, because features correspond to categories of animal sensation. — Pinter, Charles. Mind and the Cosmic Order p1

He doesn't go on to say much about the world as it is in the absence of any observer, because (I think) in his view, there's nothing to be said about it.

Furthermore he says, as I previously put it, that the mind operates by different laws to physics, which are not derivable from it:

The mystery is that in an age when physics has carried us into such a fantastic and unimaginable reality, we still balk at the idea that there are mental phenomena which do not follow the rules of classical physics. Why is it so hard to accept that in a universe in which space-time bends and curves, where particles of matter weave in and out of existence, and space itself is particulate—why would it be strange to accept that the mind of living animals is something complex whose laws are not the same ones that have been familiar to us for centuries?

'the same ones' referring to the laws of physics. So he's not arguing for any normative role for science or that the understanding of the mind is derived from or dependent on physical laws. Where he appeals to science, is what cognitive neuroscience has discovered about the way the sentient mind organises cognitions into gestalts.

There's a section heading Materialism and Objectivity, which is where he explicitly mentions Kant and noumena, and the requirement of scientifically objective statements to be utterly devoid of any subjective sense, which, he says, results in:

The “zombie universe” of objective science [which] is exactly the mind-independent universe discussed in Chapter 2: It is the residue after all sensable qualities of objects have been taken away, leaving objects with no color, appearance, feel, weight or any other discernible features.

So he's not suggesting that the "zombie universe" is real, and its depiction in the sensory systems of animals is an illusion. Rather he says that what we instinctively take to be an external reality, is really the 'manufactory of the brain' (in Schopenhauer's exact words.)

our representational filters prevent us from seeing the world as it is. — Joshs

From a Buddhist (and even, possibly, even a neo-Kantian) perspective there is no 'world as it truly is', because 'the world' is itself a composite of conditioned factors continually arising and ceasing. Real being is something which it lacks. Seeing through the apparently solid reality of 'the world', is 'seeing things as they are' - but that can be a disillusioning realisation:

For Trungpa, a truly spiritual journey toward basic sanity has to begin with a sense of hopelessness — the recognition of the complete and utter hopelessness of our current situation. He assured his readers that they are required to undertake a major process of disillusionment in order to relinquish their belief in the existence of an external panacea that can eliminate their suffering and pain. We have to learn to live with our pain instead of hoping for something that will cause all of our hesitations, confusions, insanity, and pain to disappear. This theme is elaborated in [the book] Illusion’s Game:

Creating this kind of hope is one of the most prominent features of spiritual materialism… There are so many promises involved. So much hope is planted in your heart. This is playing on your weakness. It creates further confusion with regard to pain. You forget about the pain altogether and get involved in looking for something other than pain. And this itself is pain… That is what we will go through unless we understand that the basic requirement for treading the spiritual path is hopelessness (Illusion’s Game, pP. 61-62.) — Traleg Kyagbon Rinpoche

'Hopelessness' in the sense of abandoning hope of some ultimate gain or reward to be had.

Anyway, that's all a digression, although I've mentioned before I first encountered Kant through T R V Murti's book on Buddhist philosophy so that's obviously left an imprint.

The subjects which are *not* dealt with in Pinter's book are the nature of reason and the role of creativity, and many other such grand topics, but then part of the appeal is its simplicity and the fact it has specific focus. But he devotes quite a lot of the book to criticising what I think of as scientism (although he doesn't use that word either).

they scatter light at a wavelength of 650nm — Michael

An intrinsic property.

So I'd say "the postbox is red". You'd say " the wave particles you imagine are a postbox scatter light at a wavelength of 650nm"...

...and yours is the simpler explanation?

when someone has tetrachromacy or brain damage or the like then they respond differently to the same stimulus and so the quality of the emergent visual experiences are different, i.e they see different colours. — Michael

Nothing there contradicts what I've said.

your theory requires this “hidden state” invention — Michael

It's not my theory, it's called active inference, it's currently the leading theory of perception among cognitive scientists.

It’s overly complicated, there’s no evidence for it, and I would even say it’s incomprehensible. — Michael

It's the standard model now taught on most cognitive science courses, so it can't be that incomprehensible and with currently just over 160 papers in print on the subject I hardly think it lacks evidence. Perhaps if you're not an expert in cognitive science you might refrain from deciding arbitrarily what there is and isn't evidence for.

Something can’t be both all red and all blue. — Michael

So you keep claiming, yet still no argument to support it. Why can something not be both all red and all blue?

An even simpler example: fire causes most of us to feel pain. Pain isn’t some external “hidden state”; it’s a quality of our experience. Colour is of the same kind. — Michael

And yet we don't say the fire is in pain, we do say the fire is red.

People seem so bewitched by the complexity of visual experiences that they think sight works differently to other senses. — Michael

No people have spent decades studying the occipital neural circuits, modelling people's responses in different cases of brain damage, and directly experimenting on the occasional elective open brain surgery and have reached the conclusion that sight works thus.

And it's not that dissimilar to other senses in this respect.

We've concluded that we do not 'see' a model or a quale, or anything like that because there is absolutely zero evidence of any mechanism by which such observation could possibly happen in the brain.

Variance cause variance. — Joshs

Variance in what. There has to be medium for the variance to be a variance of.

Laws and patterned regularities are idealizations of continuous qualitative change. — Joshs

Change in what?

Of these potentially infinite variety of accounts... — Joshs

This is the assertion I take issue with. I see no grounds for believing it. That there exist multiple possible accounts is evident. I don't see any reason to conclude from that that the variety is near infinite, that's a huge and unwarranted leap (a near infinite one, in fact!). There are multiple accounts. That's all the data we have.

are you giving priority to a certain empirical account from physics? Is this a ‘bedrock’ account, as Quine claimed, one which grounds all the others in an irreducibly real beginning? — Joshs

I'm not personally giving priority to any particular account. I'm arguing that the range of possible accounts is constrained by the intrinsic properties of the hidden states of which they are an account. Physics need not even enter into it, these can be treated merely as hidden states in Markov nodal network system, no atoms or particles need be involved. Only information.

I think Wayfarer might agree that the way to bedrock is to begin by asking what all possible accounts of any aspect of the world have in common, that is , what is the condition of possibility of empirical account-building? — Joshs

Need they have anything in common?

linguistic conceptual accounts of the world are elaborations of practical perceptual interactions that are continuous with the role of niche building in non-linguistic animals under selective evolutionary pressure. I don’t think such models warrant taking an account from physics as normatively determinative. — Joshs

I agree.

Isaac, arguments concerning colour tend to be futile. — Banno

So I'm finding. And yet that's where these discussions tend to go (it seems safe ground from the anti-realist). I don't like squash either, but if it's the only game one's colleagues are willing to play, then one must either play it or play nothing. Discussion is a two person game and colour seems to be the topic of choice...

The neural net is not making a model that you then see with your mind. It is your mind seeing. — Banno

Spot on.