What is color?

[m74z00219]

Hi all,

I don't really expect an answer to this question, because I suppose it depends on a pretty complete understanding of consciousness....

Still, what in the world is color? Sure, we can map different colors to different frequencies of light, but knowing that is totally satisfying. The color green might correspond to an electromagnetic frequency of 550 THz (thereabouts), but how does the developing body go about "assigning" green to that frequency??

Of course, there's the other interesting philosophical question (or has this been addressed experimentally?) of whether or not different humans perceive the same colors (those without injury or abnormality)....

Anyway, looking forward to some interesting commentary.

Thanks all,
M74

[Nick Theodorakis]

...
Of course, there's the other interesting philosophical question (or has this been addressed experimentally?) of whether or not different humans perceive the same colors (those without injury or abnormality)....
...

That is experimentally accessible. There is a color matching test in which subjects look at a monochromatic source, and attempt to match its color by combining three primary colors.

It turns out that humans have two different variants of the red-sensitive photoreceptor molecule slightly different absorption maxima. Thus the two groups perceive red differently and this can also be demonstrated on that color test.

Nick

[Substantia Innominata]

Still, what in the world is color? Sure, we can map different colors to different frequencies of light, but knowing that is totally satisfying. The color green might correspond to an electromagnetic frequency of 550 THz (thereabouts), but how does the developing body go about "assigning" green to that frequency??

Consider how it goes about in assigning the term 'chair' to a seating furniture with (generally) four legs. I mean 'green' as such is evidently not a color, but is what you read (or think about) and what you have here: green. That's just a word. A fancy term, we invented in order to describe and label phenomena we experience. We might easily switch the semantics, rename, and say that what has been (described as) 'green' will now be (described as) 'red'. So the cucumber 'becomes' red. Provided the word spreads and we regain consensus, there'd be no problem. At least I don't realize any.

I wonder whether the brain might not in fact do something very similar to that we do (consciously) in language. That is, invention. It senses differing optical input and hence simply must come up with some means of indicating these differences, a means to differentiate and identify these different wavelengths (which quite often hint at the nature or class of objects 'in sight', or to the fact that the same object is in a different condition, like, for example, an apple that was red but has become rotten and thus changed its reflecting wavelength, in one word, its color.) Out of this basic necessity it invented the qualitative aspect(s) of color.. that we experience when perceiving a certain wavelength. It could have done a job far more unpleasant?!

Of course, there's the other interesting philosophical question (or has this been addressed experimentally?) of whether or not different humans perceive the same colors (those without injury or abnormality)....

In a fairly strict sense, I'd say, we do not perceive color at all. What we perceive is electromagnetic radiation, but ('in itself') that has no color. Only when we process our optical perception of this radiation in the brain will we 'make up' a color for it, consequently corresponding, actually sticking (it) to it. And this (then) given color will change, obviously with the frequency of the accountable radiation, so that we'll be able to differentiate between different frequencies, accordingly. At least the mode in which our eyes, our optical nerves, and those brain portions responsible handle "optical input" seems to be similar between individuals (of the same species in comparable condition). From there of course it doesn't follow that we see or experience the same color(s) identically. But it follows we're able to reach consensus, i.e. consensus as to the point when and where colors differ; as well as consensus pertaining to the question "which color is which" and "what color has x?" The (ripe) banana would be yellow, the (ripe) tomato rather red, and so on. This alone is covered (and only can be covered) by experiments of the kind described by Nick Theodorakis. Yet beyond that, we cannot know. Since there begins the domain of subjectivity and that is territory closed to all, apart, for sure, from the respective ('owning') subject.

[Ken G]

Yes, the key point here, I would say, is that "color" is a word, and "red" is a word, and therefore it is whatever we choose to make it. There's no such "thing" as color, it is a choice we make about organizing our experience. But this is nothing new, all language is labeling experience. We have no way to know the experience of "red" is the same between two people, and indeed we'd be silly to think it would be the same (or even the same for one person when they are young versus old), but we can point to it and say "that's red" and someone else can say "OK, got it", and of course they don't "got it" because there's not actually some particular thing there to "get", but they have learned how to use that label in a way that others will generally agree with.

On another thread, on consciousness, the concept of "qualia" came up, where qualia are the internal experiences of something that transcends the physical reality of the external phenomenon itself. Color would be a classic example of a quale. Now, some reject the concept altogether, saying there is no experience of red, but I think they are actually objecting to something else, to the implication (that should never have been there in the first place) that the "experience of red" is not something purely personal, that it must have some absolute character to count as a quale. I see no reason to give qualia that arbitrary and impossible criterion-- we still need a word for the things we experience that cannot be registered (even if they can be tracked) on an instrument of some kind, and "qualia" seems fine to me, even if it is not the same experience from person to person or even day to day. All the same, we have an "experience of red", and it is that experience that allows us to recognize red-- and nothing else.

[Trebuchet]

To some conspiracy minded folks, "true color" is an absolute, so that every time they see a NASA photo labeled "false color" or "approximately true color", it's evidence of a massive gubmint coverup of aliens/UFO's/the Moon Hoax/faces on Mars, etc.

[TrAI]

That is experimentally accessible. There is a color matching test in which subjects look at a monochromatic source, and attempt to match its color by combining three primary colors.

It turns out that humans have two different variants of the red-sensitive photoreceptor molecule slightly different absorption maxima. Thus the two groups perceive red differently and this can also be demonstrated on that color test.

Nick

Seems to me that this test would show differences in the eyes response to the frequencies, like mutations that cause different sensitivities in the proteins used as photo detectors, however, at this stage it is still about frequency, not color. Color is the way the brain represent the different values of the photo receptors, not a characteristic of the light itself, it may very well be that, just like how you can map a set of invisible frequencies to any arbitrary set of colors to create a false color image, the assignment of the colors to specific combinations of stimuli is arbitrary to some extent, and forms during early infancy. So it may be different from person to person, we would not really know, and it would be impossible to test, even if we had some sort of neural interlinking technology, as there would be no way to know if any difference were due to an actual difference in perception of color, or just that the link hardware was not properly trained/adapted to other structural differences in the brains of the people using it.

[novaderrik]

i've always (well, not "always", but for about 20 years or so) had a theory that people see colors differently- what my brain "sees" as blue, someone else's brain might "see" as the color i would consider brown.
this is my totally unscientific theory about why some people like brighter colors and why some people like earth tones and why some people like color combinations that i would consider to be ugly .

[Ken G]

i've always (well, not "always", but for about 20 years or so) had a theory that people see colors differently- what my brain "sees" as blue, someone else's brain might "see" as the color i would consider brown.
this is my totally unscientific theory about why some people like brighter colors and why some people like earth tones and why some people like color combinations that i would consider to be ugly .

That is very much an issue that comes up when people think about "qualia", but the problem is, how could you ever tell the difference between two people who find different color combinations ugly because they see them differently, or because they have a different idea about what is ugly? Those two notions do not seem distinguishable, the experience of "color" and the experience of "ugly" are so orthogonal to each other that no a priori connection seems to exist, the linking of the two seems arbitrary (OK, it doesn't seem arbitrary to you, to you those colors "really are" ugly, but how can you say it is impossible that someone else might find them pretty without seeing them the same way you see colors you consider pretty?). In other words, the experience of a color isn't something different from a color, that's what a color is. But it is a relative concept, not an absolute concept, meaning that a color is however a particular person experiences it. There's no "redness" experience, that you could say someone else is swapping for the "blueness" experience-- there's just the experience you have when you see red, and the experience I have when I see red, and that's what "red" means.

Some have taken that to dispute the whole idea of "qualia" like redness, but I say they are merely imposing an artificial requirement onto the qualia and then rejecting them when they don't satisfy that artificial requirement. We encounter the alternative approach to this in physics all the time-- say with "proper time." In the twin paradox of relativity, two twins can experience different proper times between the same two events, and no one says this disputes the existence of a concept of proper time, it just means that it depends on its subject, like redness does.

[Substantia Innominata]

Yes, the key point here, I would say, is that "color" is a word, and "red" is a word, and therefore it is whatever we choose to make it. There's no such "thing" as color, it is a choice we make about organizing our experience. But this is nothing new, all language is labeling experience. We have no way to know the experience of "red" is the same between two people, and indeed we'd be silly to think it would be the same (or even the same for one person when they are young versus old), but we can point to it and say "that's red" and someone else can say "OK, got it", and of course they don't "got it" because there's not actually some particular thing there to "get", but they have learned how to use that label in a way that others will generally agree with.

Exactly.

Color would be a classic example of a quale. Now, some reject the concept altogether, saying there is no experience of red, but I think they are actually objecting to something else, to the implication (that should never have been there in the first place) that the "experience of red" is not something purely personal, that it must have some absolute character to count as a quale. I see no reason to give qualia that arbitrary and impossible criterion-- we still need a word for the things we experience that cannot be registered (even if they can be tracked) on an instrument of some kind, and "qualia" seems fine to me, even if it is not the same experience from person to person or even day to day. All the same, we have an "experience of red", and it is that experience that allows us to recognize red-- and nothing else.

I'd say that, at least in the respective literature and discourse, color is the classic example of a quale, even though the choice as such appears rather arbitrary; since to my mind there's nothing less "classic" about smell, sound (in the sense of being 'heard'), or even light, which, of course, is also a color, or, respectively, an indicator of how strong colors are, or can be discerned: no light = no colors, anymore. Maybe it's become traditional, or 'color' just sticks easily and in this case you don't have to make clear everytime what exactly you're referring to; like in the case of 'sound', which can be a quale (when heard), but also something 'objective', while still in air.. as simple pressure waves. As an aside, (for example) in German we actually tend to draw this distinction, in a lexical and semantic way that is. So we would distinguish in terms of the word used when adressing the 'sound in air' (or sound in a physical sense), like in the 'speed of sound', or the 'sound barrier' (that'd be Schall; Schallgeschwindigkeit; Schallmauer, respectively) or the 'sound (when) being heard', which would rather be termed Geräusch, which in turn could be best translated into English as 'noise'. In other words, it is 'noise' when you've (or anyone/-thing) heard it, but 'sound' otherwise and in general. Funnily that's also why this proliferating thought experiment concerning the 'tree falling in the woods' does work 'best' (in some sense only) in languages like English, that do not normally tend to draw this distinction. As this, really, is all it boils down to. (~There'll definitely be no noise as long as not heard, but there'll still be sound waves, well.. at best. ) Still, by no means everyone distinguishes so sharply in everyday speech, in whatever tongue, despite the fact that it would make life easier, sometimes. But, yes, this was intended to be about color, not about noise, or the falling trees once more.

I'm fairly pleased whenever the qualia-thing and the language-thing pop up together in these discussions, because that's just what I hinted at, above, when wondering whether the brain might not do something similar to what we do, only consciously, in language. The qualitative aspect of experience.. seen as a sort of language, or even metalanguage, of the brain. Its very own solution, to come up with a depiction of, and to indicate, to represent, somehow, what is going on around (as well as inside) and what it receives (from.. who knows where).

Color is the way the brain represent the different values of the photo receptors, not a characteristic of the light itself

..here again, like I stated above, the thing with the 'light'. These discussions unfortunately always come with the danger of finally devolving into pure word bickering.. what do we mean when we say what? Yet, anyway, when writing 'light', what (I guess!) you really meant was nothing but electromagnetic radiation.. and that's then the harder part, namely, to imaginge and to intuit, that light as well, of course, is not a characteristic of the aforementioned radiation itself. It has no light. Comes without any light. Just like a nice bundle of ultra-hard x-rays will not let me read in the dark. The 'light' inside the radiation bundle is, at this point, exactly the same as its 'color', i.e. in this qualitative sense (only known to us, consciousness) not present. It is there only.. well, numerically, given as a potentiality, contained as information. Just like the color, which will 'result' from certain characteristics, like wavelength, frequency, etc. The light is, already, like 'noise' in the sense I was speaking of; the corresponding 'sound' of the radiation bundle then could be taken to be.. its relative intensity in just that spectrum we humans call the visible spectrum. Because if this is pronounced sufficiently, there'll be light.. and, certainly, nothing else is light. Similar to noise: It is nothing 'out there'.

[grant hutchison]

Another interesting issue with regard to colour and language relates to how we divide up colour space. It's reasonably well-known that different languages divide colour space differently (some languages don't differentiate "blue" and "green" with different names, for instance, while others subdivide that area more than English does). If asked to differentiate between colours that don't have separate colour names in our "native" colour system, we are a little slower to tell them apart than if we're asked to differentiate across a colour boundary. Maximally distinguishable colours are therefore distributed differently across colour space according to the language one uses. It's a good reminder that colour is a neurological event, and to some extent a learned neurological event.

Grant Hutchison

[Ken G]

I'm fairly pleased whenever the qualia-thing and the language-thing pop up together in these discussions, because that's just what I hinted at, above, when wondering whether the brain might not do something similar to what we do, only consciously, in language. The qualitative aspect of experience.. seen as a sort of language, or even metalanguage, of the brain. Its very own solution, to come up with a depiction of, and to indicate, to represent, somehow, what is going on around (as well as inside) and what it receives (from.. who knows where).

The way I look at it, we have the experience, and we have the intersubjective connecting of the experience between two experiencers. The latter is what language is for, but you are raising a second way to think about language, which is a kind of internal language that a brain uses to "talk to itself." This would the represent something more than communication, but actual processing, like when you take an output and feed it back in as an input. Whether or not language is essential for thought is an oft-debated issue in linguistics.

The 'light' inside the radiation bundle is, at this point, exactly the same as its 'color', i.e. in this qualitative sense (only known to us, consciousness) not present. It is there only.. well, numerically, given as a potentiality, contained as information.

This is very important, because physics is all about information, to the point that some people begin to imagine that the information is "out there", when in fact it is "in here", in our minds. Information is how we think about the world, like you are saying about "noise", it is not "in the light." However, the light does affect the information we get from it, just like the sound affects the noise.

[HenrikOlsen]

Consider how it goes about in assigning the term 'chair' to a seating furniture with (generally) four legs.

Chair is whatever we defined "chair" to mean.

[Strange]

Another interesting issue with regard to colour and language relates to how we divide up colour space. It's reasonably well-known that different languages divide colour space differently (some languages don't differentiate "blue" and "green" with different names, for instance, while others subdivide that area more than English does). If asked to differentiate between colours that don't have separate colour names in our "native" colour system, we are a little slower to tell them apart than if we're asked to differentiate across a colour boundary. Maximally distinguishable colours are therefore distributed differently across colour space according to the language one uses. It's a good reminder that colour is a neurological event, and to some extent a learned neurological event.

Grant Hutchison

Don't languages with more colour words tend to add them in (roughly) the same order? Which implies some sort of underlying neurological structure supporting the learned events (confirming your "to some extent").

[grant hutchison]

Don't languages with more colour words tend to add them in (roughly) the same order? Which implies some sort of underlying neurological structure supporting the learned events (confirming your "to some extent").

The observation that languages seem to acquire basic colour words in distinct order is usually credited to Berlin & Kay in 1969, although they seem only to have rediscovered that fact. But it's true for only a small number of colours. There are languages that distinguish only dark and light. There are languages that distinguish dark, light and red. There are languages that distinguish only dark, light, red and either yellow or green. There are languages that distinguish dark, light, red, yellow and green. And there are languages that add blue to that mixture. Beyond that, Berlin & Kay's ideas about the subsequent addition of other colour names (brown, pink, orange, purple, grey) don't seem to be borne out by observation: languages seem to be rather random about how they added other colour distinctions beyond the core group. But we do seem to have some sort of hard-wired salience hierarchy for the basic shades.

More on the connection between colour naming and perception:
Russian distinguishes between two kinds of blue (siniy and goluboy) in the same way English distinguishes two kinds of red (red and pink). If you ask Russian-speakers which of two blue sample shades matches a reference shade, and repeat the experiment with various sample pairs, then the speed of response varies. The farther apart in colour space the samples lie, the faster the matching response. But if the samples straddle the siniy/goluboy border, the response is also faster. The same level of "colour separation" is somehow more salient if it crosses the linguistic border. English speakers don't show this effect at the siniy/goluboy border, which has no demonstrable salience for them. And Russian speakers don't show this effect if you give them a distracting verbal task to perform while they perform the colour-matching task, or if you get them to perform the task using only the visual field that connects to their non-linguistic cerebral hemisphere (usually left visual field, right hemisphere).
That said, there's no evidence that people who lack specific colour terms are unable to distinguish those colours: colour-matching tests show that everyone performs equally well when given time to reflect. But our speed of response seems to be tied at some deep level to our ability to access verbal labels for colour space.

Grant Hutchison

[Strange]

That said, there's no evidence that people who lack specific colour terms are unable to distinguish those colours: colour-matching tests show that everyone performs equally well when given time to reflect. But our speed of response seems to be tied at some deep level to our ability to access verbal labels for colour space.

Intersting. I had thought that my wife had soemthing like Perfect colour" (by analogy with perfect pitch) but it maybe it is really because she uses a very wide range of colour words (half a dozen for different shades of red, for example) which makes it easier for her to match colours.

[grant hutchison]

Intersting. I had thought that my wife had soemthing like Perfect colour" (by analogy with perfect pitch) but it maybe it is really because she uses a very wide range of colour words (half a dozen for different shades of red, for example) which makes it easier for her to match colours.

It's difficult to know the resolution of this effect. The Russian-language research I mentioned is only a couple of years old, and I don't know if anyone has looked at finer discrimination within the broad, categorical colour spaces. Also I wonder if, at this finer level, people populate colour space with names because they are good at recognizing and remembering fine colour distinctions, or if they are good at recognizing and remembering fine colour distinctions because they've spent the necessary time to learn the labels. (I recall Robert Redford saying in an interview that he wasn't colour-blind, he was just colour-ignorant ... at which my wife declared that I was colour-stupid.)
But verbal labels do seem to be generally important: there's recent research showing that if you are given a name for an object or concept, you are subsequently able to retrieve and compare information about that named entity better and faster than if it has no associated label. Richard Feynman's famous little parable about how his father taught him about birds (in which Feynman derided the learning of names as valueless) turns out to be wrong: knowing the names of birds lets you recognize them faster, as well as letting you retrieve associated information more easily. A corollary to that is that learning definitions is important: if a word triggers only blurry and poorly-defined associations in your brain, you're not letting it perform its useful discriminatory task.

Grant Hutchison

[Ken G]

Richard Feynman's famous little parable about how his father taught him about birds (in which Feynman derided the learning of names as valueless) turns out to be wrong: knowing the names of birds lets you recognize them faster, as well as letting you retrieve associated information more easily. A corollary to that is that learning definitions is important: if a word triggers only blurry and poorly-defined associations in your brain, you're not letting it perform its useful discriminatory task.

And although that is an excellent point, there may be a tradeoff there. I might imagine that Feynman could counter that knowing the names of things makes it easier and faster to do certain kinds of thinking about those things, along the lines of sorting in boxes. But we might imagine that such labeling could also inhibit other kinds of thinking, the kind we normally call "outside the box." Labels act to herd our thinking, making it faster if we respect the boundaries we are being herded into, but Feynman might view a deeper level of understanding as coming from being able to reason across those boundaries-- rather his own specialty. I think that was the point about the birds, though he would no doubt also have been interested to hear the opposing information that you raise (and thanks for those extremely interesting and important scientific results).

One might even imagine ways to test this hypothesis. I don't know if this would be a good way to do it, but just as an example, imagine we show people color mixtures involving reds and blues giving purples, and ask them to identify the two shades that went into the mix. This would be a challenge for most people, so we could limit to painters or some such thing. Then we could use English and Russian speakers, and see if the Russians, unhindered by the artificial boundary between red and pink, do a better job of finding the right shades of red that go into that mix, but tend to mess up the blue tints, looking for them to be either siniy or goluboy. English speakers might do better in the opposite vein. If that turned out to be true, it would suggest what the "politically correct" movement has claimed all along-- language can shepherd thought, and that is not always a good thing.

[HenrikOlsen]

Richard Feynman's famous little parable about how his father taught him about birds (in which Feynman derided the learning of names as valueless) turns out to be wrong: knowing the names of birds lets you recognize them faster, as well as letting you retrieve associated information more easily. A corollary to that is that learning definitions is important: if a word triggers only blurry and poorly-defined associations in your brain, you're not letting it perform its useful discriminatory task.

Grant Hutchison

The fable is about how knowing only the name of the species doesn't tell you anything about the bird, not about how names are used for categorizing and retrieving knowledge.

And I have to say it's a bit strange to read about this research for me, since I have no trouble remembering lots of things about people, but I have trouble remembering their names.

[grant hutchison]

The fable is about how knowing only the name of the species doesn't tell you anything about the bird, not about how names are used for categorizing and retrieving knowledge.

In Feynman's story (at least, the version told in 'What Do You Care What Other People Think?'), the child Feynman can tell that his father doesn't know the names of the birds, and is simply making up ad hoc names in order to illustrate that they don't tell us anything about the bird. Setting aside for the moment the fact that many bird names do tell us something about the bird, I'm simply pointing out that this paternal policy of "not learning the name" (while it makes a nice point) turns out to be misguided.

And I have to say it's a bit strange to read about this research for me, since I have no trouble remembering lots of things about people, but I have trouble remembering their names.

Ah, but we also have a bit of brain devoted to storing faces, through which we can do the same memory-access trick.

Grant Hutchison

[HenrikOlsen]

So in reality, what's useful to retrieve information for something is to learn a label for it, whether they are a face (which we apparently have specialized circuitry to recognize faster than most other patterns so they are quite useful as labels) or a name.

Incidentally, I just realized that me not remembering names is irrelevant for this, as that's treating the name as part of the information to be retrieved rather than as a label to retrieve it by.
When someone refers to the person by name I still know who they are talking about.

I've also realized that I'm best at remembering names when I've seen the face at the same(within short term memory time) time as I've seen their name in writing. For some reason a name I've seen written down is a lot easier to remember than one I heard.

[grant hutchison]

If that turned out to be true, it would suggest what the "politically correct" movement has claimed all along-- language can shepherd thought, and that is not always a good thing.

I think the old Sapir-Whorf hypothesis, in which our thoughts are strongly constrained by our language, to the extent that we cannot entertain certain thoughts, has been long disproven by experiment and example. (In fact, in some quarters it was politically correct to utterly deny the validity of Sapir-Whorf, since any other stance might open a back door to racism.) What we seem to be left with is that there are weak but detectable effects from the structure of language. For instance, if you are a speaker of a language that uses grammatical gender, you have an experimentally detectable tendency to associate gender-stereotypical attributes (strength, nurturing, etc) with nouns of the appropriate grammatical gender. If given a list of word pairs, consisting of nouns and people's names, you'll remember pairs better in which the grammatical gender of the noun and the gender of the name match each other. And we have evidence that if your language makes you pay particular attention to some aspect of the world, then you'll probably be good at dealing with that aspect of the world. There are Australian Aboriginal languages, for instance, which routinely use global rather than personal coordinates: you don't raise your left hand, you raise your west hand (if facing north) or your south hand (when facing west). Native speakers of these languages have a pretty good sense of direction, since they have to routinely track their orientation. However, they have not the slightest difficulty communicating in terms of "right" and "left" if that is required.
So the constraints seem to be rather mild. I'd submit that if you want to think outside the box there's nothing stopping you doing so. Defining your new concept clearly and giving it a name early will work to your advantage.

Grant Hutchison

[grant hutchison]

When someone refers to the person by name I still know who they are talking about.

Ah, I thought you were like me. My wife spends a lot of time bellowing names at me incredulously: "Steve Perkins. Steve Perkins. STEVE PERKINS! How can you not remember who Steve Perkins is? We worked with him for five years. You shared an office for six months. He saved you from drowning." (OK, I'm exaggerating. But you get the idea.)
Whereas at parties she sidles up to me and says: "Who's the guy with the beard over there? I'm sure I've seen him before." And I usually know who it is. But, since I can't remember his name, I have to keep coming up with biographical details until she remembers. (At which point she usually shouts the name aloud triumphantly, and the guy comes over to talk to us, because he thinks he's being hailed as a long-lost friend.)
It passes the time.

Grant Hutchison

[PetersCreek]

Interesting discussion. Based on my experiences in art and photography, I've considered myself as having a fairly refined perception of color. When I was painting, I often referred to colors as their nearest equivalent oil paint color, such as Cerulean, Cobalt, or Phthalo blues. These days I still try to place them on the color wheel but I also catch myself mentally playing with RGB sliders. It makes me wonder if the egg came first or the chicken? Does color perception lead to a refinement of language used to identify them or does the refinement of language lead to greater selectivity among colors? Or both?

[Ken G]

I think the old Sapir-Whorf hypothesis, in which our thoughts are strongly constrained by our language, to the extent that we cannot entertain certain thoughts, has been long disproven by experiment and example.

But it is not the claim of sensible political correctness that language controls our thoughts, merely that it influences them.

(In fact, in some quarters it was politically correct to utterly deny the validity of Sapir-Whorf, since any other stance might open a back door to racism.)

Political correctness means many things. When it is "done right", it just means that people are made aware of how language can prejudice their thinking. When "done wrong", it means that truth itself is malleable by what we would wish to be true in a perfect world, as if asserting something enough times can make it true, or not asserting it can make it false. It is a shame that the latter form is so notorious that most people are so completely unaware of how much more prevalent the former form is, in the trenches of people fighting things like racism and sexism. I think the problem is more how that focus "bleeds over" into other areas where it is far less appropriate, until what should be an objective truth becomes conditional on ideology, leading to the "shooting the messenger" problem! Blaming that on political correctness is misplaced-- the blame falls on human foibles, and reactionary modes of thinking are just as susceptible as politically correct modes.

What we seem to be left with is that there are weak but detectable effects from the structure of language.

It is likely that the strength of the effect is context-dependent, so quantifying the strength in one clinical setting is hardly generalizable to every influence. Thus, the interesting results you report only establish that the principle itself is well-founded-- language can affect thought. How much, and in what way, is going to depend on every different situation, possibly sensitively in some (which is more or less the basis of the concept of language-as-propaganda). No doubt it gets exaggerated by those who wish to make a big deal of it, and swept under the rug by those who consider it inconvenient.

So the constraints seem to be rather mild. I'd submit that if you want to think outside the box there's nothing stopping you doing so.

Is the issue whether or not something is stopping me, or is the issue whether or not I am receiving enough encouragement, by language or education, to do so? Frankly, I see quite a few posts on this very forum that just strike me as coming from people who have not received enough encouragement from their language or education to step outside those comfy mind-limiting boxes (present company excepted).

Defining your new concept clearly and giving it a name early will work to your advantage.

That is certainly true-- if you can get others to accept the idea of thinking about such a new definition. Unfortunately that can be a "big if". I'm not saying it's not a good idea to have clear definitions, I'm saying it is not good to be the slave of language rather than its master, and those who depend too heavily on language to do their thinking for them (and as you point out it can be quite good at doing that, which is often a good thing but not always) nevertheless can fall into the latter trap. I believe that was Feynman's point about the bird names, even though you're right that ideally we know both the names and the organizing principles.

[HenrikOlsen]

That is certainly true-- if you can get others to accept the idea of thinking about such a new definition. Unfortunately that can be a "big if". I'm not saying it's not a good idea to have clear definitions, I'm saying it not good to be the slave of language rather than its master.

The point here is that if you define and name your concepts early, your thinking about them will be clearer.
Getting others to accept the definition is part of the different problem of communicating your conclusions once you're done with the thinking.

[Ken G]

The point here is that if you define and name your concepts early, your thinking about them will be clearer.
Getting others to accept the definition is part of the different problem of communicating your conclusions once you're done with the thinking.

Certainly, so we must all try to be careful about definitions. My added point to that is, along with Feynman, I see a danger in paying too much attention to the language, and not enough attention to the ideas behind the language. When defining models appropriate to the real world, like birds or electrons or measurement outcomes or consciousness, the purpose of the definition is not to carry all the salient features of the phenomenon so defined, but merely to conjure the appropriate object. It still leaves a great deal of thinking after the object has been so conjured, and I think it was the overlooking of that element that Feynman was bemoaning. He makes the same point about the term "friction"-- he hated that if someone asks "what causes heat when two rough objects rub together", the answer is given "friction", and then everyone goes home, as if the question were really answered instead of just giving a label to the answer. In education, this gets to the problem of the appearance of knowledge, and a nice A grade, when there is in fact no knowledge at all, just appropriate manipulation of labels.

[astromark]

Excellent descriptive text... leeds to refinement of meanings as required and thus a precision of description is as a learned method.

A complex understanding of the language being used is a great starting point... Red is often some other shade of so many relatives...

I would say that from just this thread a greater understanding can be achieved.. Just touched on I have some more to offer...

'What is colour?' The hue reflected. might be a good way of saying it. Learned perception for being able to recognise the shade you perceive..

How does nature do this ? The Mantis does not need to know the frequency of green..

Just that having a green hue seems to work for hiding amongst the other green stuff...

Colour is the reflective value or hue of that reflected light.

Do we all see the same colour ? No. but most see a similar shade and hue... we have learned to name it as such...

The Harrier Hawk has a info red ability. It see's the colour of its victims body heat and urine...

and ultraviolet is also used to help some insects find prey...

The visible spectrum is just a small part of the electromagnetic spectrum... learning and naming. Is a learned response.

[HenrikOlsen]

Just that having a green hue seems to work for hiding amongst the other green stuff...

For the vast majority of prey animals just standing still and making no noise works for hiding, even while wearing a bright orange safety vest.

[Grashtel]

For the vast majority of prey animals just standing still and making no noise works for hiding, even while wearing a bright orange safety vest.

IIRC most animals (or at least mammals) can't tell the difference between orange and green (explaining why ginger cats can be successful predators) so the bright orange vest is actually not bad as camouflage.

[Ivan Viehoff]

Sure, we can map different colors to different frequencies of light, but knowing that is totally satisfying. The color green might correspond to an electromagnetic frequency of 550 THz (thereabouts), but how does the developing body go about "assigning" green to that frequency??

What frequency are you mapping brown to? Surfaces have complex properties which can't be summarised in a single frequency.

I think actually we have to map colours to relative response of the various receptors in our eyes. Many birds have an additional receptor to us. Our 3 colour receptors give us 8 basic colours (if you include black), but for birds the 4 colour receptors mean that they have 16 basic colours. Birds must therefore live in a much more colourful world than us. It is not just that they detect an ultraviolet colour, it is all the interactions between UV and the other colours give them a whole additional world of colour on top of ours.