Hacker News Reader

Comment by ChuckMcM | Hacker News Reader

Solid parenting advice, #1 talk to your children even when they cannot talk, #2 read new books to your children every week. You can always re-read favorites but just keep the new books coming. And then combine #1 and #2 and talk to your children about the books you read.

There is an excellent Radiolab podcast on "Colors" [1] which has (for me) the fairly stunning revelation that you can't see a color you can't name. It starts by looking at the colors used in ancient texts and works its way forward. Some women have more cones than normal and are tetrachromatic so they can see more colors than the normal trichromatic human. But until you name the new colors apparently you can't see them (or recognize them).

People seem to need words to organize their thoughts and reason about things they see or read. The more words you give them, the better able they are to reason.

[1] http://www.radiolab.org/story/211119-colors/

View on HN

> There is an excellent Radiolab podcast on "Colors" [1] which has (for me) the fairly stunning revelation that you can't see a color you can't name.

As I understand, this has been tested and found wanting: you can ask people to take a group of things colored aoi / blue-green, when they have words for neither blue nor green, and ask them to come up with some additional categorization. Instead of categorizing at random or something, they form two distinct groups of... blue and green. And you can predict what colors a language will have based on the number of words it allocates to color (eg. if there's only two color words, they're always going to be black and white https://en.wikipedia.org/wiki/Linguistic_relativity_and_the_... )

> There is an excellent Radiolab podcast on "Colors" [1] which has (for me) the fairly stunning revelation that you can't see a color you can't name. [emphasis added]

Interesting idea, but untestable, therefore unscientific. Do you see why? A color we can't name isn't open to systematic investigation. If we try to get around this by asking a subject to compare the "unseen" color to a color patch, we're only given the "unseeable" color a name by proxy.

This kind of claim is very commonly made by people who don't understand science.

> People seem to need words to organize their thoughts and reason about things they see or read.

This argues that animals lacking language also lack the ability to think. Copious evidence disproves this idea. Video of a raven using tools to achieve a purpose:

http://www.youtube.com/watch?v=41Z6Mvjd9w0

> The more words you give them, the better able they are to reason.

The better able to reason with words. This is self-referential in the extreme.

I'd suggest that it's not so much any colour is "unseeable" as it is "indistinguishable" for some or most people compared to similar colours. And since our perception of colour is relative to different contexts, that makes the research even harder to account for. That all said, an "unseeable" colour to my mind really refers to how if you put two very, very similar colours together, only some people can tell them apart, and perhaps more often once they've learned to recognize the difference? No idea how close I am to the mark, as I haven't listened to any of this or read much source material. But I would compare it to colourblindness -- just because in a certain context someone can't tell red from green doesn't mean it can't be objectively, absolutely measured as such. Which is to say, just because you can't perceive it yourself doesn't mean it can't be studied: either the perception (false or true) or the absolute measurements of reality.

The point is that "color" does not physically exist[1], it is a construct of human perception and language. Thus you cannot scientifically study color at all--if you try, what you are actually studying is human perception and language.

This is not an unscientific concept, because human perception and language are naturally occurring, albeit incredibly complex, phenomena.

Animals do not see colors, they react to physical phenomena including photons. Humans use the concept of color to describe what animals do.

Certain animals who learn human language can communicate with us about color. That doesn't mean they can't think, it just means that color is a human construct so they need human language to reason about it.

[1] What physically exists are photons across a range of frequencies, but that's not the same thing as color.

> Animals do not see colors, they react to physical phenomena including photons.

Come on. This is a roundabout way to try to define animals in terms of what they don't have -- not color perception, which many animals have, but words to describe the colors.

Of course animals see colors, in the most basic biological sense. Many see colors more efficiently than we do, indeed see more colors, colors we can neither imagine nor name.

> ... it just means that color is a human construct so they need human language to reason about it.

Believe me when I tell you, animals don't need us to tell them how to reason about color. Why do you think the Peppered Moth changed color over a period of 150 years? It was the only way the species could survive their changing environment and the acute visual sense of their predators.

> What physically exists are photons across a range of frequencies, but that's not the same thing as color.

No, what we call "color" or "color perception" is the biological sensing of electromagnetic energy, which many animals have. So, since we and animals have color perception, and since some animals do it better than we do, the argument has no substance.

While we're on the topic, here are my favorite colors:

* Unforeseeable fuchsia, the color you never see coming.

* Statutory grape, the color of a dress worn by a 13-year-old girl.

You're basically doing what I said people do, which is to use color to describe how animals act. That does not mean that color, as a concept, has any meaning in the mind of an animal.

The electromagnetic spectrum is continuous; our concept of color is a completely arbitrary way of dividing it[1] up into sections. That was the point of the Radiolab episode (and a lot of academic research before it): different human cultures divide up the electromagnetic spectrum in different ways, each of which is as valid as another.

Animals might not have this concept at all; for example it would be just as physically valid to think of the electromagnetic spectrum the way we think about sound--as a continous spectrum from "quiet" to "loud". We know that divisions are arbitrary when it comes to sound--that's why the "this amp goes to 11" joke in Spinal Tap is funny.

[1] Edit to add: only a tiny portion of it

The frequency of the perceived light coresponds roughly to color. The frequency of the perceived sound corresponds roughly to pitch. So I think that "quiet" to "loud" is analogous to "pale" to "bright", when it comes to visible light.

>>different human cultures divide up the electromagnetic spectrum in different ways ... Yep, different human cultures divide up the sound spectrum in different ways too ... So what ?

Some people have absolute pitch perception, allowing them to reproduce a musical tone, that they had heard exactly, even without knowing its proper name, just like some people can distinguish between color hues, that other people lump together as generic "pink", "green", "blue" or "yellow".

Having a vocabulary of specific names may ease learning/training, but it seems to me, that no matter how many different words for colors you teach a child with daltonism, the child will still fail a color blindness test.

>> you can't see a color you can't name

> Interesting idea, but untestable, therefore unscientific.

Astonishingly, it has been tested and falsified. Some people have color synesthesia: they experience various patterns as colors. For example, in number-color synesthesia, the person experiences numbers as having colors regardless of their literal visual color. For example, the numeral 7 might appear as a particular shade of green.

Some of these color synesthetes are color blind due to genetic defects in their light detectors. Nonetheless they appear to experience the full range of synesthesic colors including ones they have never seen by direct perception. One person called them Martian colors.

> which has (for me) the fairly stunning revelation that you can't see a color you can't name.

Then how did they get named in the first place?

By other cultures. Example: show a set of color swatches all identical blue with one bluish green. Ask participants to indicate the exception. Speakers of languages with a blue-green distinction get it right essentially instantly (< 1 second); native speakers of languages without a blue-green distinction can take as much as 30 seconds and actually get it wrong (I.e., indicate one of the identical blue swatches)

How did the other cultures name them in the first place? Can't be turtles cultures all the way down the stack.

Reply on news.ycombinator.com