COLOUR

[ruby sparks]

"People universally believe that objects look colored because they are colored, just as we experience them. The sky looks blue because it is blue, grass looks green because it is green, and blood looks red because it is red. As surprising as it may seem, these beliefs are fundamentally mistaken. Neither objects nor lights are actually “colored” in anything like the way we experience them. Rather, color is a psychological property of our visual experiences when we look at objects and lights, not a physical property of those objects or lights. The colors we see are based on physical properties of objects and lights that cause us to see them as colored, to be sure, but these physical properties are different in important ways from the colors we perceive".

[1999, attributed to Dr Stephen Palmer, Professor of Psychology (speciality: Cognition), Visual Perception Laboratory, University of California at Berkeley].

https://plato.stanford.edu/entries/color/

-----------------------------------------------------------------

Claim 1: objects are not themselves coloured, they do not have colour.

Claim 2A: Colour is a psychologically-experienced 'mental' phenomenon only. Colour does not really exist other than in this way.

Claim 2B: Colour is a psychologically-experienced 'mental' phenomenon of consciousness only. Colour does not really exist other than in this way.

I think claim 1 is the easier and more recognised to be the case. I might hold that one quite strongly.

Claim 2A is, I think, not something that can be shown to be the case by any reasonable standard and is therefore (I would separately claim) an unresolved issue, but it is my inclination to go along with it and so I will start off defending the statement quoted above (which is apparently in blue).

Claim 2B is slightly more onerous, and may be even more up for debate, imo.

Does anyone have any views on the topic?

 

[Politesse]

Palmer's statement - that color is a psychological property of visual experience - is a more accurate statement than either of these, which I would modify to "perceptive property of visual experience". Any organism with light receiving organism needs a mechanism needs a corresponding perceptive system to help it sort through the input that is has received and form responses. But this perception doesn't always or even usually involve colors, nor do all color-perceiving species see the same colors. Indeed, even as individuals our perception of color often differs for physiological or cultural reasons.

 

[ruby sparks]

Palmer's statement - that color is a psychological property of visual experience - is a more accurate statement than either of these, which I would modify to "perceptive property of visual experience". Any organism with light receiving organism needs a mechanism needs a corresponding perceptive system to help it sort through the input that is has received and form responses. But this perception doesn't always or even usually involve colors, nor do all color-perceiving species see the same colors. Indeed, even as individuals our perception of color often differs for physiological or cultural reasons.

Thanks.

As I said, it seems mistaken to say, for example, that the redness of a red chair is because the chair itself is red. I even remember this from secondary school physics lessons on optics. And yet it's SO counterintuitive. Turn off all the lights in the sitting room before you go to bed at night and it seems intuitively true that the red chair is still there in the dark.

It seems to be accepted among the relevant scientific experts that the chair isn't red, but the light reflected from it may be, although this is what I tend to question. I don't even know how waves of electro-magnetic radiation (for example) could have a red colour in them. I would prefer to think that they have other non-colour properties that stimulate the parts the non-coloured brain that then produce the weird conscious experience of redness.

And here is something that Erwin Schrodinger (of Schrodinger's Cat fame) said in 1944:

"We may further ask: Is radiation in the neighbourhood of wave-length 590 nanometres the only one to produce the sensation of yellow? The answer is: Not at all. If waves of 760 nanometres, which by themselves produce the sensation of red, are mixed in a definite proportion with waves of 535 nanometres, which by themselves produce the sensation of green, this mixture produces a yellow that is indistinguishable from the one produced by 590 nanometres. Two adjacent fields illuminated, one by the mixture, the other by the single spectral light, look exactly alike, you cannot tell which is which. Could this be foretold from the wave-lengths - is there a numerical connection with these physical, objective characteristics of the waves? No."

 

[fromderinside]

Palmer's statement - that color is a psychological property of visual experience - is a more accurate statement than either of these, which I would modify to "perceptive property of visual experience". Any organism with light receiving organism needs a mechanism needs a corresponding perceptive system to help it sort through the input that is has received and form responses. But this perception doesn't always or even usually involve colors, nor do all color-perceiving species see the same colors. Indeed, even as individuals our perception of color often differs for physiological or cultural reasons.

Nuh-uh The Biological Basis of the Experience of Constant Colour Categories https://www.biorxiv.org/node/145849.full

[FONT=&quot]In summary, there was a trivial variability in assigning colours to different categories by subjects of different ethnic and cultural origins. This is a pointer to an important principle of the organization of the sensory brain, at least in terms of colour vision, namely that there is a very significant similarity in the inherited computational mechanisms for generating colour categories in all humans.[/FONT]

 

[Politesse]

Palmer's statement - that color is a psychological property of visual experience - is a more accurate statement than either of these, which I would modify to "perceptive property of visual experience". Any organism with light receiving organism needs a mechanism needs a corresponding perceptive system to help it sort through the input that is has received and form responses. But this perception doesn't always or even usually involve colors, nor do all color-perceiving species see the same colors. Indeed, even as individuals our perception of color often differs for physiological or cultural reasons.

Nuh-uh The Biological Basis of the Experience of Constant Colour Categories https://www.biorxiv.org/node/145849.full

[FONT="]In summary, there was a trivial variability in assigning colours to different categories by subjects of different ethnic and cultural origins. This is a pointer to an important principle of the organization of the sensory brain, at least in terms of colour vision, namely that there is a very significant similarity in the inherited computational mechanisms for generating colour categories in all humans.[/FONT]

I'm well aware of the Berlin/Kay study the students in your link were trying to replicate.

Variance isn't the same thing as absolute separation, however. Just because there is a biological basis for color perception (something my post pointedly does not deny) doesn't mean that perception of colors isn't influenced by culture. The visible color spectrum is absolute. Decisions about where to place the boundaries of color categories is not, and our ideas of color category strongly impact perception. For example, Ruby's post above asserts that he perceives a chair as being in a "red" category despite currently registering to his eyes as "black"; that isn't because of his physiology, it's because the learned category of "redness" has been assigned to the chair, and that idea is more persistent than the incoming data. One could easily imagine a culture in which it was not common to assign permanent color categories to objects, nor to identify something as having a "colored" quality. So we should see this as an area of potential cultural variation. Your linked study actually demonstrates this, as it reports that there were differences of color categorization, and with some regions of the color spectrum more commonly than others. This doesn't prove that color perception is autonomous from physiology, but it does establish that there are variations in perception based on differential cultural categorization.

Berlin and Kay would not be surprised, for instance, that "purplish" category is one of the object of most contention, since most languages by volume lack a basic color term for "purple" or its equivalents, while a tiny minority have more basic terms for "purplish" colors than Indo-European languages do. The study you link has way too few participants to make the case either way, since most of the cultures included in the study had only one representative subject and no primary speakers on non-Indo-European languages were included, but B & K themselves have previously made a strong case for these categories impacting self-reported color perception strongly. We're all looking at the same "colors", but we do not divide them up the same way, and therefore perception also varies. It makes no sense to categorize something as being a "green chair" if your culture lacks a category for "green things" semantically separate from "blue things". But since color percpetion is linked to a physiological reality, you will predictably call it a "blue/green chair", not a "red/orange" one. Most issues of perception and culture are like this: a material reality, but shaped and circumscribed by the experience of enculturation to an incomplete degree.

As for physiology, color-blindness color-impairment, and fourth-cone phenomena are all well known and well documented situations.

 

[Politesse]

Palmer's statement - that color is a psychological property of visual experience - is a more accurate statement than either of these, which I would modify to "perceptive property of visual experience". Any organism with light receiving organism needs a mechanism needs a corresponding perceptive system to help it sort through the input that is has received and form responses. But this perception doesn't always or even usually involve colors, nor do all color-perceiving species see the same colors. Indeed, even as individuals our perception of color often differs for physiological or cultural reasons.

Thanks.

As I said, it seems mistaken to say, for example, that the redness of a red chair is because the chair itself is red. I even remember this from secondary school physics lessons on optics. And yet it's SO counterintuitive. Turn off all the lights in the sitting room before you go to bed at night and it seems intuitively true that the red chair is still there in the dark.

It seems to be accepted among the relevant scientific experts that the chair isn't red, but the light reflected from it may be, although this is what I tend to question. I don't even know how waves of electro-magnetic radiation (for example) could have a red colour in them. I would prefer to think that they have other non-colour properties that stimulate the parts the non-coloured brain that then produce the weird conscious experience of redness.

And here is something that Erwin Schrodinger (of Schrodinger's Cat fame) said in 1944:

"We may further ask: Is radiation in the neighbourhood of wave-length 590 nanometres the only one to produce the sensation of yellow? The answer is: Not at all. If waves of 760 nanometres, which by themselves produce the sensation of red, are mixed in a definite proportion with waves of 535 nanometres, which by themselves produce the sensation of green, this mixture produces a yellow that is indistinguishable from the one produced by 590 nanometres. Two adjacent fields illuminated, one by the mixture, the other by the single spectral light, look exactly alike, you cannot tell which is which. Could this be foretold from the wave-lengths - is there a numerical connection with these physical, objective characteristics of the waves? No."

There isn't a "red color" in the wavelength; the light has an absolute wavelength (ie a physically measurable distance between parallel wave peaks), which your brain then sorts into a learned category. Input and response. It would be more accurate to state it as a acquired rule rather than a natural category: eg. "when a wavelength interval between 700–635 nm or so hits my cornea, I will perceive it as 'red' and refer to it that way from that point onward." You could change the rule - say, expand the category to 750-610 nm - and end up with entirely different categories (and perceptions) without changing any real physical property of the objects you are looking at.

You could say the same of almost any sort of object sorting. Say I were to give you a tray of index cards with a bunch of words written on them, and asked you to sort them into nouns and adjectives. There isn't a "noun in the card", and indeed neither nouns nor words themselves exist independently of social consensus. But you wouldn't struggle to identify which cards refer to nouns and which ones refer to adjectives, because your brain isn't looking for the "noun in the card", but rather trying to recognize the word itself, which has already been sorted into that category by previous instruction.

BTW I think most color specialists would heartily disagree with Schrödinger there, we know a lot more about how wavelengths interact with each other now than we did in 1944.

 

[ruby sparks]

There isn't a "red color" in the wavelength; the light has an absolute wavelength, which your brain then sorts into a learned category. Input and response. It would be more accurate to state it as a rule rather than a category: eg. "when a wavelength interval between 700–635 nm or so hits my cornea, I will perceive it as 'red' and refer to it that way from that point onward." You could change the rule - say, expand the category to 750-610 nm - and end up with entirely different categories (and perceptions) without changing any real physical property of the objects you are looking at.

You could say the same of almost any sort of object sorting. Say I were to give you a tray of index cards with a bunch of words written on them, and asked you to sort them into nouns and adjectives. There isn't a "noun in the card", and indeed neither nouns nor words themselves exist independently of social consensus. But you wouldn't struggle to identify which cards refer to nouns and which ones refer to adjectives, because your brain isn't looking for the "noun in the card", but rather trying to recognize the word itself, which has already been sorted into that category by previous instruction.

Ok. But maybe there's a difference between nouns and colours. For example, surely a newborn baby with normal baby-vision (not very good I believe) still has varied colour experiences, even if they have not learned to name them.

So although what you say is interesting, which it is, it isn't, er, shedding light, on the issues of whether colours are 'merely' experiences of consciousness (possibly involving what are called qualia) or whether light, for example, has colour properties of itself. I think your opinion on that seems similar to mine, but neither of us can claim to be sure.

As to different people perceiving colours differently, I believe that this is the case. There are even said to be gender differences (or at least I read something about that). But again, that could be down to different brain structure at the very fine level, or something like that. Or, as you say, the terms used to describe the colours could be learned, and there would, as you say, be merely broad agreement about, say, what 'red' is.

I suppose the thing I am most interested in is 'where or what is the colour' regardless of whether people agree what colour it is, or even see the same one, interesting though that is. If that makes sense.

That said, the variations and discrepancies we are talking about do not exactly seem to support the idea that there are objective colour properties 'out there in the world'. But they don't eliminate the possibility either.

 

[Politesse]

There isn't a "red color" in the wavelength; the light has an absolute wavelength, which your brain then sorts into a learned category. Input and response. It would be more accurate to state it as a rule rather than a category: eg. "when a wavelength interval between 700–635 nm or so hits my cornea, I will perceive it as 'red' and refer to it that way from that point onward." You could change the rule - say, expand the category to 750-610 nm - and end up with entirely different categories (and perceptions) without changing any real physical property of the objects you are looking at.

You could say the same of almost any sort of object sorting. Say I were to give you a tray of index cards with a bunch of words written on them, and asked you to sort them into nouns and adjectives. There isn't a "noun in the card", and indeed neither nouns nor words themselves exist independently of social consensus. But you wouldn't struggle to identify which cards refer to nouns and which ones refer to adjectives, because your brain isn't looking for the "noun in the card", but rather trying to recognize the word itself, which has already been sorted into that category by previous instruction.

Ok. But maybe there's a difference between nouns and colours. For example, surely a newborn baby with normal baby-vision (not very good I believe) still has varied colour experiences, even if they have not learned to name them.

So although what you say is interesting, which it is, it isn't, er, shedding light, on the issues of whether colours are 'merely' experiences of consciousness (possibly involving what are called qualia) or whether light, for example, has colour properties of itself. I think your opinion on that seems similar to mine, but neither of us can claim to be sure.

As to different people perceiving colours differently, I believe that this is the case. There are even said to be gender differences (or at least I read something about that). But again, that could be down to different brain structure at the very fine level, or something like that. Or, as you say, the terms used to describe the colours could be learned, and there would, as you say, be merely broad agreement about, say, what 'red' is.

I suppose the thing I am most interested in is 'where or what is the colour' regardless of whether people agree what colour it is, interesting though that is. If that makes sense.

There is no functional difference between "where the color is" and "what the color is called"; color is the category. Wavelength is the natural property, which those categories are attempting to organize. Since wavelength occurs naturally on a spectrum rather than in absolute categories (even the numbers applied to wavelengths are the result of an arbitrary but standardized categorization of space) it will always be inaccurate to refer to a wavelength as "being a color" even if everyone uniformly agreed on what the best color terms are. Your perception is taking in information from the naturally perceived wavelength, but your prefrontal cortex is going to kick in once you try to sort those wavelengths into conscious categories, and it will borrow on experience and custom to do so.

 

[ruby sparks]

There is no functional difference between "where the color is" and "what the color is called"; color is the category. Wavelength is the property, which those categories are attempting to organize. Since it occurs naturally on a spectrum rather than in absolute categories, it will always be inaccruate to refer to a wavelength as "being a color" even if everyone uniformly agreed on what the best color terms are.

Are we on the same...wavelength?

When I said 'where' I meant is the colour in the object, or in the light coming from it, or only in our brains, and if that last one, then is it a purely a 'sensation' (like pain for example). If you are unlucky enough to have someone prod the soles of your feet with a cattle prod, it will be painful, but there is no 'pain' property in electricity, it seems. Just for good measure the sensation of pain isn't actually in your feet but that's another issue in some ways (though fascinating of itself).

There are similar issues with all (conscious) sensations, including hearing (does a tree falling in a forest make a noise if there is nothing there to hear it, etc) and indeed taste and smell.

Of course even scientists use the terms 'sound waves' and so on, I believe, but in a way that's for convenience. The air disturbances do not necessarily have in them or of them the property of 'sound'.

 

[Politesse]

There is no functional difference between "where the color is" and "what the color is called"; color is the category. Wavelength is the property, which those categories are attempting to organize. Since it occurs naturally on a spectrum rather than in absolute categories, it will always be inaccruate to refer to a wavelength as "being a color" even if everyone uniformly agreed on what the best color terms are.

Are we on the same...wavelength?

When I said 'where' I meant is the colour in the object, or in the light coming from it, or only in our brains, and if that last one, then is it a purely a 'sensation' (like pain for example). If you are unlucky enough to have someone prod the soles of your feet with a cattle prod, it will be painful, but there is no 'pain' property in the electric current. Just for good measure the sensation of pain isn't actually in your feet but that's another issue in some ways (though fascinating of itself).

There are similar issues with all (conscious) sensations, including hearing (does a tree falling in a forest make a noise if there is nothing there to hear it, etc) and indeed taste and smell.

My answer is unchanged. "Color" is a mental category, produced in response to a physical reality (the wavelength of the light bouncing off the object). This is easily demonstrated: change the quality of the light bouncing off the object, as in your example, and the initial perception of its color will change even if your mental idea of its color stubbornly refuses to.

I'm not clear what you mean by "purely a sensation", as sensations are almost by definition an interaction between a sense organ and the external world, unless the perceptive mechanism has been entirely hijacked and is simulating those experiences rather than reporting them (as in a hallucination).

 

[ruby sparks]

.... it will always be inaccruate to refer to a wavelength as "being a color"...

Inaccurate as to which colour, yes, point about categories taken, but inaccurate to say that light has/is a colour property (of some sort)?


ps this thread is a spin-off of a thread on moral realism, so as such my initial interest was in the philosophical issue of colour realism.

 

[ruby sparks]

There is no functional difference between "where the color is" and "what the color is called"; color is the category. Wavelength is the property, which those categories are attempting to organize. Since it occurs naturally on a spectrum rather than in absolute categories, it will always be inaccruate to refer to a wavelength as "being a color" even if everyone uniformly agreed on what the best color terms are.

Are we on the same...wavelength?

When I said 'where' I meant is the colour in the object, or in the light coming from it, or only in our brains, and if that last one, then is it a purely a 'sensation' (like pain for example). If you are unlucky enough to have someone prod the soles of your feet with a cattle prod, it will be painful, but there is no 'pain' property in the electric current. Just for good measure the sensation of pain isn't actually in your feet but that's another issue in some ways (though fascinating of itself).

There are similar issues with all (conscious) sensations, including hearing (does a tree falling in a forest make a noise if there is nothing there to hear it, etc) and indeed taste and smell.

My answer is unchanged. "Color" is a mental category, produced in response to a physical reality (the wavelength of the light bouncing off the object). This is easily demonstrated: change the quality of the light bouncing off the object, as in your example, and the initial perception of its color will change even if your mental idea of its color stubbornly refuses to.

I'm not clear what you mean by "purely a sensation", as sensations are almost by definition an interaction between a sense organ and the external world, unless the perceptive mechanism has been entirely hijacked and is simulating those experiences rather than reporting them (as in a hallucination).

I think you and I would broadly agree about what colour is. It's more or less just a term to describe 'what something feels like' And when I say purely a sensation I mean that it isn't anything else. For example, I might have the sensation that there's a red object in my sitting room, but the existence of the object is not purely a sensation (there's still really an object in the room after I turn the lights out). I'm suggesting the redness is nothing more than the sensation.

By contrast there are colour realists, who would say that colours are physical properties of objects and/or light, just as spatial dimensions, forms and shapes are.

 

[Politesse]

.... it will always be inaccruate to refer to a wavelength as "being a color"...

Inaccurate as to which colour, yes, point about categories taken, but inaccurate to say that light has/is a colour property (of some sort)?


ps this thread is a spin-off of a thread on moral realism, so as such my initial interest was in the philosophical issue of colour realism.

Yes, because color only exists as a mental category. The light itself cares not what category it is sorted into by an observer. A wave with a wavelength of 556 nm will still be a wave with a wavelength of 556 whether some observer calls it (and perceives it) as "red" or just as generally"dark-colored" as in a bicolorate language/culture.

 

[Politesse]

My answer is unchanged. "Color" is a mental category, produced in response to a physical reality (the wavelength of the light bouncing off the object). This is easily demonstrated: change the quality of the light bouncing off the object, as in your example, and the initial perception of its color will change even if your mental idea of its color stubbornly refuses to.

I'm not clear what you mean by "purely a sensation", as sensations are almost by definition an interaction between a sense organ and the external world, unless the perceptive mechanism has been entirely hijacked and is simulating those experiences rather than reporting them (as in a hallucination).

I think you and I would broadly agree about what colour is. It's more or less just a term to describe 'what something feels like' And when I say purely a sensation I mean that it isn't anything else. For example, I might have the sensation that there's a red object in my sitting room, but the existence of the object is not purely a sensation (there's really an object in the room). I'm suggesting the redness is.

By contrast there are colour realists, who would say that colours are properties of objects and/or light.

Light definitely has measurable properties, from wavelength to amplitude to frequency. But none of those properties are "colors". Colors are mental categories circumscribed by socially normalized color terms.

 

[ruby sparks]

Light definitely has measurable properties, from wavelength to amplitude to frequency. But none of those properties are "colors". Colors are mental categories circumscribed by socially normalized color terms.

Yes, because color only exists as a mental category. The light itself cares not what category it is sorted into by an observer. A wave with a wavelength of 556 nm will still be a wave with a wavelength of 556 whether some observer calls it (and perceives it) as "red" or just as generally"dark-colored" as in a bicolorate language/culture.

I think we agree.

The thing is, if someone came along now and said, no, colour is also an independent property of stuff in the world just like, for example, shape or size (which we also perceive) are, we'd be unable to demonstrate that we were correct and they weren't. I think.

 

[steve_bank]

Color is defined by objective measured wavelengths. We learn to associate wavelengths with a name like red, green, and blue.

Eyes detect color and ttasmit to the brain which discriminates between colors. Physioilgy.

Saying color only exists mentaly is like saying a rock exists onlty mentaly.


I have been listening to NPR. Scientists making profound revelations that are no really all that profound. They have to find something to write abut. Publish or perish.
The old past metaphysical debates pre science are obsolete. These days arguing whether perceptions are real or not is silly.

 

[ruby sparks]

Color is defined by objective measured wavelengths. We learn to associate wavelengths with a name like red, green, and blue.

Ok, and I'm not saying it's conclusive (because it isn't) but...

(a) A colour experience resulting from a single light source at 590 nanometres wavelength (normally associated with yellow) can look exactly the same as two light sources of 760 nanometres (usually associated with red) and 535 nanometres (usually associated with green) arriving from two different directions.

Also,

(b) There is the reverse situation in a colour cube illusion such as this....



...where the wavelengths of the light reaching our brain from the middle square on the 'top' face and of the face 'nearest' us are the same, but the colours are experienced very differently.

and

(c) It is also possible for anyone to have colour experiences in the absence of any light.

Do the above effects not suggest that the brain can essentially 'make up' colours (to either be the same as or different from each other when they aren't, in terms of wavelengths, or experience colours when there are no wavelengths of light at all)?

In other words, colour experience seems as if it possibly could be at least somewhat decoupled from light input; does not depend only on the objectively-measured properties of the light I mean.

Which, if correct, would seem to place at least some colour experiences in the mental realm. And then the door would be ajar.

These days arguing whether perceptions are real or not is silly.

The perceptions could be real of course. The question is whether certain properties of them are in the world outside as well as inside brains.

As previously mentioned, electricity can cause the (private, subjective, conscious, inner) experience we call pain without electricity having the property of pain. So there is a precedent, it seems.

 

[steve_bank]

It is still your eyes and yoor brain.

An artist I knew in the 70s handed me a color strip and asked me how many shades I could see. I saw maybe a half dozen, she said she saw a lot more.

Eyes and brains vary/ I heard it said that maybe the painting Starry Night was not imagined, it is how Van Gough saw reality.

I watched a show that talked about a small number of people who's eye and audio nerves are crossed connected. Audio can cause colors to change.

We learn to associate something called red with a color. When you see red and I see red how do we know what I perceive is the same? We can't.

 

[ruby sparks]

I watched a show that talked about a small number of people who's eye and audio nerves are crossed connected. Audio can cause colors to change.

Yes, synesthesics can apparently, I believe, hear a word in the dark and experience a colour, yet there is no light, or as you say experience a colour change after only a change in audio. How come, if it's the wavelengths of light that transmit their colour property into the brain?

It's inconclusive, obviously, but, in conjunction with my previous examples, it does suggest that the brain can at least sometimes create its own colour experiences, which I think is at least a foot in the door, or some supporting evidence for, colours as mental experiences.

If there's one thing we can say about the human brain it's that it generates some weird stuff. It generates its own consciousness just for starters. And then there's what we call thoughts. And self. Do not underestimate the brain's ability to produce weird things.

Evolution, of course, doesn't care about (and our survival does not depend on) whether there are colours 'out there' or not. All that matters (for survival etc) is that it's useful to tell the difference between stuff (one radiated wavelength or amplitude from another for instance). We don't yet know why we have any conscious experiences, why anything feels like anything at all. But it does seem fairly clear that some phenomena are very likely purely mental (private, subjective, conscious, inner).

 

[steve_bank]

I watched a show that talked about a small number of people who's eye and audio nerves are crossed connected. Audio can cause colors to change.

Yes, synesthesics can apparently, I believe, hear a word in the dark and experience a colour, yet there is no light, or as you say experience a colour change after only a change in audio. How come, if it's the wavelengths of light that transmit their colour property into the brain?

It's inconclusive, obviously, but, in conjunction with my previous examples, it does suggest that the brain can at least sometimes create its own colour experiences, which I think is at least a foot in the door, or some supporting evidence for, colours as mental experiences.

If there's one thing we can say about the human brain it's that it generates some weird stuff. It generates its own consciousness just for starters. And then there's what we call thoughts. And self. Do not underestimate the brain's ability to produce weird things.

Evolution, of course, doesn't care about (and our survival does not depend on) whether there are colours 'out there' or not. All that matters (for survival etc) is that it's useful to tell the difference between stuff (one radiated wavelength or amplitude from another for instance). We don't yet know why we have any conscious experiences, why anything feels like anything at all. But it does seem fairly clear that some phenomena are very likely purely mental (private, subjective, conscious, inner).

For me the rational scientist type, or so I imagine myself. it is the way we evolved. Np more no less. There is n0 answer to why, there is only the way it is.
.