This discussion often misses the point of absorption and emission spectra. the 'actual' color depends on the lighting conditions. Sometimes we know it should be one color in daylight conditions, so it appears different to our mind than it would otherwise be in other lighting, which explains one fascination with art. Moreover, our eyes use a different mechanism to see in the dark, and our minds combine the experience of different objects with known colors in the dark as they are perceived with visual purple. Finally, and perhaps most significantly. very low-level edge-detection mechanisms distort the actual perceived colors at boundaries, so one color really does look different to us depending on the color next to it.

Right. So we can stop trusting telescopes that there really are more stars in the sky than we can see with the naked eye. — Benkei

I don't see what instrumentation has to do with this. The fact is that the vast majority of colours which we see, talk about, refer to, and describe, are combinations of different wavelengths. The various reds which we see are no different from this. Have you ever seen someone mixing red paint? So your restricted definition of red, to a particular range of wavelength, while it might be useful for some scientific purposes, is not a true representation of what the average person refers to as "red".

You have restricted your definition of "red" to a particular type of red, some sort of pure red. When do you ever see pure red? There are many other instances of red, different combinations of wavelengths, which do not qualify as "red" under your definition, and it is not wrong to call these particular red objects, instances of red. But it is wrong to say that we shouldn't call any instance of colour "red" unless it conforms to your restricted definition. It's like you've determined the "ideal" red, and you do not think that anyone should call anything "red" unless it fits this ideal.

Thats exactly what I am saying. Red paint contains pigments which appear red in daylight. But in other lighting conditions, the red paint can appear to be orange, or purple, or brown, or gray. the strange thing is, two pigments can appear the same color in daylight, but different colors in different lighting conditions. That is what the great artists understood.

For a common object, such as a strawberry, we do not think of it being orange when it is lit such that it appears orange. We still think of it as red from our experience.

That is why color is more than just a combination of wavelengths.

I should add, it is amazing how much that is misunderstood even now.

Take Rothko, for example, who was an artist interested only in large swathes of color. He painted by candlelight. Art galleries now pay millions of dollars for his canvases and display them in bright incandescent light. I actually had an argument with the MOMA in NYC about it. It said basically, showing his canvases in candlelight was too weird for Americans to accept, and there would only be a demand to see it 'properly' in bright light instead. Which, given the other weird things MOMA displays, made it rather pointless continuing the debate, so I gave up. So now we cant even see the art Rothko actually painted at all. The name Rothko is now more important than the colors he wanted to show us. Just a thought why it is so misunderstood here now.

For basic scientific statements about various experiences and illusion related to color perception, see:for example:

http://www.archimedes-lab.org/color_optical_illusions.html

http://mentalfloss.com/article/54448/5-color-illusions-and-why-they-work

https://en.wikipedia.org/wiki/Color_constancy

http://www.stareclips.com/

https://en.wikipedia.org/wiki/Ideasthesia

it is amazing how much that is misunderstood even now. — ernestm

Indeed, especially scientific accounts based on arguments from illusion, according to which you'd never see a real colour, only your own hallucination of something unseen. :-}

what we *mean* by *real color* is how a large uniformly, flat colored object appears in daylight. That is just a standardized reference by convention. It has very little to do with how we actually perceive color.

I don't see what instrumentation has to do with this. — Metaphysician Undercover

It's instrumentation that can tell you there's no red in the image. The rest of your post is an argument to ignore progress and return to the Middle Ages.

Actually, your method for determining colour proceeds from a faulty premise. You assume that a particular colour has a corresponding range of wavelengths. That this is a false premise is evident from the fact that most of the different colours which we talk about have no such corresponding range of wavelengths.

Furthermore, it appears like you want to reduce "colour" to a specific set of primary colours, each having a corresponding wavelength. But any such reduction is known to be purely arbitrary, and abstract. So all you've done is produced an arbitrary, ideal "red", and you are claiming that if an encountered instance of red does not match this ideal red, it does not qualify as "red".

So you have a false premise, that any particular colour, such as red, can be defined by a particular wavelength, and from this you produce a false "ideal red", which is define as a range of wavelengths. Then you proceed to argue, from this false conclusion, that if a particular instance of colour does not match your ideal, it does not qualify as being the referred to colour.

And your mode of justifying your false premise appears to be pure assertion. I'm right and your wrong, because science supports my position. But if you looked closely, you would see that science does not support your false premise. Science has produced a specialized definition of "red", which is suitable for use within the instrumentation which you refer to, but is not suitable for use when referring to human perception.

We only have a limited number of "color"-receptive cones in our eyes, and all of them are triggered by an assortment of wavelengths of varying degrees. It is a rare circumstance when only one of our cones is triggered, or none as in the case of no light being in the environment. You have to account for the fact that we can't see at all when light is completely absent from the environment.

Our brains then assemble a "picture" based on these varying degrees that each cones is triggered by. In other words, the brain mixes the strength of these signals from each "color"-sensitive cone to create the colors we experience.

It doesn't really matter what one person claims as being "the ideal red". After all it is possible that we all experience different colors when interacting with the same light. All that matters is that the symbol, or representation, of that light is consistent - that the effect is always the same per the cause for that particular person. This way we can still communicate about what we see without anyone being the wiser choice of what is the "ideal color".

All that matters is that the symbol, or representation, of that light is consistent - that the effect is always the same per the cause for that particular person. — Harry Hindu

Don't we need some consistency between individuals as well? This consistency gives us what some people call inter-subjectivity, which in some cases it is argued, qualifies as objectivity. It's interpretation. If we all agree as to the meaning of a particular word, then that word has "objective" meaning (in the sense of inter-subjective), though it might not have an ideal objective meaning in the sense of an independent Platonic Form. The independent Platonic Form could allow us to theoretically judge the inter-subjective meaning, if we had access to that Form.

Isn't this the same with colours? Seeing is a mode of interpretation as well. There is an inter-subjective meaning of "red" which provides us with the common meaning of the term. Benkei appears to be claiming that there is a truly objective "red", an ideal definition of red, and even to know this Platonic Form, through science. It is suggested that we should judge our inter-subjective interpretation against this Form. But I think Benkei derives this ideal in a faulty way. There may be such a Platonic Form of red, but Benkei has not described it.

Seeing is a mode of interpretation as well. — Metaphysician Undercover

What is interpreted? We might interpret the presence of a silver oval in our visual field as a round coin. But we don't get to interpret its presence, nor the coloured shape. In this sense seeing precedes interpretation.

But we don't get to interpret its presence, nor the coloured shape. — jkop

Why not? I would say that all these things are interpretations. That there is a presence, that there is a colour, that there is a shape, all of these are interpretations. If you consider that the coin consists of atoms and molecules, then ask yourself why do you see it as the presence of a single, coloured, shape, instead of individual molecules, or atoms. Interpretation is inherent within seeing.

One thing that makes it difficult is that those are weird colors. It's kind of an array of grayish or "dirty-looking" teal.

It looks all blue and green to me. But what do I know?

.

.

If you consider that the coin consists of atoms and molecules, then ask yourself why do you see it as the presence of a single, coloured, shape, instead of individual molecules, or atoms. Interpretation is inherent within seeing. — Metaphysician Undercover

So what would seeing its atoms and molecules be an interpretation of? :-}

I don't think you see the atoms and molecules; it is not your interpretation of some undifferentiated swarm of individual particles which sets the visual features of the object that you see but the parts of the object which are present in your visual field and the optics and biology of seeing. The presence of a single, coloured shape is set by the objective facts of seeing. Hence seeing precedes interpretation.

Human eyes can see millions of different shades of colour. This is not because there are millions of different wavelengths between 400 and 740. — Metaphysician Undercover

There's a difference between colour and chromaticity. Sensitivity to colour change for humans is in the range from 1nm to 10 nm.

Benkei appears to be claiming that there is a truly objective "red", an ideal definition of red, and even to know this Platonic Form, through science. It is suggested that we should judge our inter-subjective interpretation against this Form. But I think Benkei derives this ideal in a faulty way. There may be such a Platonic Form of red, but Benkei has not described it. — Metaphysician Undercover

There isn't a Platonic Form, there's consensus on the wavelengths associated with colours when an object absorbs light and reflects light back. And I mean this in the sense of spectrum. The science is pretty clear on this. If we then "think" we see red, when in reality there are no colours emitting with a wavelenght between 622nm to 780nm, then we've been fooled by our very fallable perception.

A colour blind person (deuteranopia) sees purple as blue. He's wrong and so are you. Why are we comfortable with saying he's wrong? Because we understand his perception is "broken". Now what we have with the strawberries is a "broken" perception that is shared by most everyone due to the white-balancing we automatically apply due to high levels of teal in the picture. Very useful in everyday life, when moving from indoors and outdoors - for this particular picture it's a nuisance.

I'm perfectly fine trusting a spectrometer on this matter.

we've been fooled by our very fallable perception. — Benkei

Our belief of what it is that we perceive is fallable, not perception, because belief is representational.

There isn't a Platonic Form, there's consensus on the wavelengths associated with colours when an object absorbs light and reflects light back. And I mean this in the sense of spectrum. The science is pretty clear on this. If we then "think" we see red, when in reality there are no colours emitting with a wavelenght between 622nm to 780nm, then we've been fooled by our very fallable perception.

A colour blind person (deuteranopia) sees purple as blue. He's wrong and so are you. Why are we comfortable with saying he's wrong? Because we understand his perception is "broken". Now what we have with the strawberries is a "broken" perception that is shared by most everyone due to the white-balancing we automatically apply due to high levels of teal in the picture. Very useful in everyday life, when moving from indoors and outdoors - for this particular picture it's a nuisance. — Benkei

I think it's incorrect to say that because someone responds differently to the same input than most people that their perception is wrong.

It seems me that when you say "wrong" you just mean "uncommon". But as we're trying to take a more philosophical approach to colour perception, you need to use more exact language and not just play along with the everyday narrative that conflates "right" with "ordinary".

And what of someone with tetrachromacy? Is their percpetion "broken" if they see as blue what most other people (i.e. people without tetrachromacy) see as purple? Or if they see it as some colour that the rest of us don't even know?

there's consensus on the wavelengths associated with colours

This is the important point. There's an agreed consensus about what wavelengths are associated with what colours, but those wavelengths are not themselves those colours. The colour red is the sensory quality that is usually elicited by the stimulation of light with a wavelength of ~620–740nm, but that's it.

And there's no reason for it to be considered wrong (as opposed to just uncommon) for an organism to have such a sensory quality elicited by light of a different wavelength – or by stimulation of something other than light, e.g. in the case of synaesthesia.

This is the important point. There's an agreed consensus about what wavelengths are associated with what colours, but those wavelengths are not themselves those colours. The colour red is the sensory quality that is usually elicited by the stimulation of light with a wavelength of ~620–740nm, but that's it.

And there's no reason for it to be considered wrong (as opposed to just uncommon) for an organism to have such a sensory quality elicited by light of a different wavelength – or by stimulation of something other than light, e.g. in the case of synaesthesia. — Michael

If we define red as light with a wavelength between 620-740nm, then calling any light with a different wavelength "red" is wrong. By definition. And that's my point condenced I suppse. We are capable of verifying our own perceptions with instruments. To then lift the subjective experience up as the definitive answer to whether an object is red or not is a step back. All the way back to before Copernicus.

The strawberries therefore appear red and we can even explain why but they aren't red.

(Also the answer is a bit more subtle because the picture does emit light in that wavelength range of red as part of the white/gray areas but something is red or reddish due to a concentration of a particular spectrum of wavelength in a particular area but that is more about chromacity).

If we define red as light with a wavelength between 620-740nm, then calling any light with a different wavelength "red" is wrong. By definition. — Benkei

Sure. But that goes back to the question I asked before; does the "red" in "I see red strawberries" refer to that kind of light? I don't think so. When I say that I see red strawberries I'm not (either explicitly or implicitly) saying that I see that kind of light. So there's equivocation here. On the one hand we might use "red" to refer to that kind of light – and that's an ad hoc scientific stipulation – and on the other hand we might use "red" to refer to the qualitative aspect of seeing, which is what I think we're doing when we say "I see red strawberries".

The strawberries therefore appear red and we can even explain why but they aren't red.

But that's just wrong, if by "red" you mean "light with a wavelength between 620-740nm", because those strawberries don't appear to reflect light with a wavelength between 620-740nm. That would be equivocation, based on the fact that most people ordinarily see things that reflect light with a wavelength between 620-740nm as being red.

So as I've said before, it seems to me that the two instances of "red" in "it appears red but isn't red" mean/refer to different things, which means that the claim that there's some sort of conflict or error in the appearance is wrong. The "red" in "it appears red" refers to a certain sensory quality and the "red" in "it isn't red" refers to a certain type of light. And there's no prima facie reason that it's wrong (as opposed to just uncommon) for this type of light to elicit this type of sensory quality.

MU,

Don't we need some consistency between individuals as well? This consistency gives us what some people call inter-subjectivity, which in some cases it is argued, qualifies as objectivity. It's interpretation. If we all agree as to the meaning of a particular word, then that word has "objective" meaning (in the sense of inter-subjective), though it might not have an ideal objective meaning in the sense of an independent Platonic Form. The independent Platonic Form could allow us to theoretically judge the inter-subjective meaning, if we had access to that Form.

Isn't this the same with colours? Seeing is a mode of interpretation as well. There is an inter-subjective meaning of "red" which provides us with the common meaning of the term. Benkei appears to be claiming that there is a truly objective "red", an ideal definition of red, and even to know this Platonic Form, through science. It is suggested that we should judge our inter-subjective interpretation against this Form. But I think Benkei derives this ideal in a faulty way. There may be such a Platonic Form of red, but Benkei has not described it. — Metaphysician Undercover

After all, it is possible, being that we are genetically similar, that we do experience the same colors with the same wavelength or assortment of wavelengths. But even if we didn't, the consistency comes from being in a shared world. The same wavelength of light may create different colors in the mind, but the same wavelength ALWAYS triggers the same color in the mind for each person.

When you learned you colors, you learned to associate the word, "red" with the color you see when that particular wavelength interacts with the cones and rods in your eyes. I may experience a different color. But we both experience our color consistently as a result of the same wavelength of light. This is why we can agree on the wavelength, without experiencing the same color.

Sure. But that goes back to the question I asked before; does the "red" in "I see red strawberries" refer to that kind of light? I don't think so. When I say that I see red strawberries I'm not (either explicitly or implicitly) saying that I see that kind of light. So there's equivocation here. On the one hand we might use "red" to refer to that kind of light – and that's an ad hoc scientific stipulation – and on the other hand we might use "red" to refer to the qualitative aspect of seeing. — Michael

I agree we can use the word "red" in different contexts but when we disagree on the redness of something, how are we going to arbitrate this question? I have no problem with you referring to the strawberries as appearing red but if we want to definitely answer the question whether they are red, we have to conclude they aren't irrespective of our subjective experiences of seeing.

But that's just wrong, if by "red" you mean "light with a wavelength between 620-740nm", because those strawberries don't appear to reflect light with a wavelength between 620-740nm. That would be equivocation, based on the fact that most people ordinarily see things that reflect light with a wavelength between 620-740nm as being red. — Michael

I don't follow. I don't think I'm saying much else than the following analogy "the desert appeared to have an oasis and we even know why but there wasn't an oasis."

There isn't a Platonic Form, there's consensus on the wavelengths associated with colours when an object absorbs light and reflects light back. — Benkei

What is the consensus on the wavelength associated with grey, white, or burgundy?

Human eyes can see millions of different shades of colour. This is not because there are millions of different wavelengths between 400 and 740. — Metaphysician Undercover

True, but there are millions of different combinations of colors triggered by millions of different combinations of the strengths of the signals coming from the rods and cones in our eyes. It's no different from creating millions of different colors from just varying degrees of the three primary colors

I don't follow. I don't think I'm saying much else than the following analogy "the desert appeared to have an oasis and we even know why but there wasn't an oasis." — Benkei

You're saying that a thing can appear red even if it isn't. And you've said that "red" refers to light with a wavelength between 620-740nm. So you're saying that a thing can appear to reflect light with a wavelength between 620-740nm even if it doesn't. I reject the notion that those strawberries appear (with the naked eye) to reflect light with a wavelength between 620-740nm. The wavelength of the light that is emitted by those pixels isn't part of the appearance at all - it's just the cause. So as I explained in the last paragraph (edited in before your response), there's two different meanings/referents of "red", and you seem to be equivocating them.

I agree we can use the word "red" in different contexts but when we disagree on the redness of something, how are we going to arbitrate this question? I have no problem with you referring to the strawberries as appearing red but if we want to definitely answer the question whether they are red, we have to conclude they aren't irrespective of our subjective experiences of seeing. — Benkei

When we disagree on the redness of something, are we disagreeing on the wavelength of the light that is emitted/reflected by it or are we disagreeing on its appearance? If the former then, yes, there's an independent fact-of-the-matter. If the latter then, no, there isn't.

And I'd argue that we're disagreeing about the latter. We can both measure and agree upon the wavelength of the light but nonetheless disagree on its colour. Because although stimulation by the former is what elicits the experience of the latter, they're not the same thing.

When we disagree on the redness of something, are we disagreeing on the wavelength of the light that is emitted/reflected by it or are we disagreeing on its appearance? If the former then, yes, there's an independent fact-of-the-matter. If the latter then, no, there isn't. — Michael

As I said: I'm willing to have a spectrometer to have a last say on this. Are you?

And I'd argue that we're disagreeing about the latter. We can both measure and agree upon the wavelength of the light but nonetheless disagree on its colour. Because although stimulation by the former is what elicits the experience of the latter, they're not the same thing. — Michael

I agree they aren't the same thing. However, I don't believe we can arrive at two different answers. So when we search for an explanation as to why we disagree, we discover "red" is associated with a certain wavelength. We then discover there's no red in the picture and we even discover how that comes about.

The difference then is that I'm willing to say: "oops, I was wrong to interpret that as red" and you continue to maintain it's red.

By that token the sun still revolves around the earth and the illusionary oasis is really there. The latter with deadly consequences.

So what would seeing its atoms and molecules be an interpretation of? — jkop

You don't see them.

The presence of a single, coloured shape is set by the objective facts of seeing. Hence seeing precedes interpretation. — jkop

isn't it objective fact that seeing is interpretation?

There isn't a Platonic Form, there's consensus on the wavelengths associated with colours when an object absorbs light and reflects light back. And I mean this in the sense of spectrum. The science is pretty clear on this. If we then "think" we see red, when in reality there are no colours emitting with a wavelenght between 622nm to 780nm, then we've been fooled by our very fallable perception. — Benkei

As I said, what we see as colour is a mixture of wavelengths, not a pure wavelength. I read some of the discussion on the strawberry picture, and apparently there are grey pixels which are composed partially of red wavelengths. So despite the fact that there is red wavelength within the composition of the pixel, the pixel itself is grey, and therefore is not called red. But grey is not a particular range of wavelength itself, it is a mixture.

(Also the answer is a bit more subtle because the picture does emit light in that wavelength range of red as part of the white/gray areas but something is red or reddish due to a concentration of a particular spectrum of wavelength in a particular area but that is more about chromacity). — Benkei

This only indicates that you are wrong in defining colour based on particular wavelengths. There is no particular wavelength for the various greys, they are combinations of wavelengths, And, since the grey pixels do emit red wavelengths, it is contradictory for you to say that there is no red in the picture, according to your definition of red. You refer to the other definition of colour to say that the pixels are grey, then as Michael indicates, it is only by equivocation that you can say there is no red in the picture.

I agree we can use the word "red" in different contexts but when we disagree on the redness of something, how are we going to arbitrate this question? I have no problem with you referring to the strawberries as appearing red but if we want to definitely answer the question whether they are red, we have to conclude they aren't irrespective of our subjective experiences of seeing. — Benkei

It all has to do with the mixtures of wavelengths, and how our eyes detect them. If red mixed with other colours, within a tiny pixel makes a colour called grey, then we are justified in saying "colour", in general, is a mixture of wavelengths. If mixing these grey pixels with certain blue pixels in a larger scale makes the colour red, then we are justified in saying that the strawberries are red.

It all has to do with the mixtures of wavelengths, and how our eyes detect them. If red mixed with other colours, within a tiny pixel makes a colour called grey, then we are justified in saying "colour", in general, is a mixture of wavelengths. If mixing these grey pixels with certain blue pixels in a larger scale makes the colour red, then we are justified in saying that the strawberries are red. — Metaphysician Undercover

Sure. So how about you get the image into paint and zoom in on that apparent red pixel until that pixel fills your screen. Then get back to me with which colour it has.

EDIT: I'd like to point out that the colour of that faux-red pixel doesn't change because of the surrounding blue, we are not "mixing" them in our perception. We're quite capable of seeing individual pixels at these resolution (1080p, just stick your nose in the screen).

The blue merely influences how we perceive it due to false signal it gives us that we ought to white balance the image for outdoor circumstances. Then our brains filter out blue from the grey, which makes the grey appear red. Taking a closer look (literally!) shows the error.

As I said: I'm willing to have a spectrometer to have a last say on this. Are you? — Benkei

No, because that only tells us the wavelength of the light that is emitted/reflected by the object; it doesn't tell us anything about the object's appearance, and when I talk about the strawberries being red, I'm talking about their appearance.

I agree they aren't the same thing. However, I don't believe we can arrive at two different answers. So when we search for an explanation as to why we disagree, we discover "red" is associated with a certain wavelength. — Benkei

That's the problematic part. What do you mean by saying that red is "associated" with a certain wavelength? If only that most people would ordinarily see objects that emit/reflect such light as red then I wouldn't disagree. But you seem to be making the stronger claim that if someone doesn't see such objects as red (or if they see objects that emit/reflect light with different wavelengths as red) then there's something wrong with what they see; that the appearance is an error. And that would be an unjustified assertion that doesn't follow from the fact that there is a common relationship between light of a certain wavelength and certain colour appearances.

The difference then is that I'm willing to say: "oops, I was wrong to interpret that as red" and you continue to maintain it's red.

By that token the sun still revolves around the earth and the illusionary oasis is really there. The latter with deadly consequences. — Benkei

The difference is that when I say "I see red strawberries" the word "red" is referring to a type of appearance (rather than to a particular type of light), and so its truth just is in the appearance, whereas when I talk about the Sun revolving around the Earth, or about there being an oasis, the sentences intend to refer to some trans-appearance fact-of-the-matter.

(Of course, you might say that the word "red" refers to a type of appearance but also intends to refer to some trans-appearance fact-of-the-matter, suggesting a naive view of perception, but that would be a category error).

So as I said before, given that the "red" in "it appears red" means/refers to something different to the "red" in "it isn't red", that something appears red₁ but isn't red₂ gives no a priori reason to suggest that the appearance is an error. Such a conclusion can only stem from equviocation.