"Blue is black with green, like the sea."

July 18, 2011 1:00 PM

"We certainly cannot follow the example of Odysseus and, going down to Hades, tempt with a bowl of blood a representative sample of native speakers to label particular areas of the standard Munsell color continuum ..."

David Wharton's Latin Color Bibliography collects quotations from ancient literature and modern research on how languages classify colors, and tries to work out the meanings of color words in classical Latin.

The title quote comes from Maurus Servius Honoratus in Isidore's Etymologies, alluded to in comparison with Tzetzal Maya descriptions of blue and purple. The pull quote comes from John Lyons' chapter in The Language of Color in the Mediterranean, quoted here.

Bonus: Aulus Gellius talks about the differences between Greek and Latin colors, green and grey horses, and marmor flavum, the "yellow marble" of sea foam. (Scroll down for translation.)

posted by nangar (14 comments total)

"We certainly cannot follow the example of Odysseus and, going down to Hades, tempt with a bowl of blood a representative sample of native speakers to label particular areas of the standard Munsell color continuum ..."

They're obviously not trying hard enough.
posted by Faint of Butt at 1:03 PM on July 18, 2011 [3 favorites]

Holy shit, this is why I come to Metafilter. Benigne scribis.
posted by everichon at 1:07 PM on July 18, 2011 [2 favorites]

One reason we can't label particular areas of the standard Munsell color continuum, is that it's not continuous. The color-wheel slices are reasonably so, but the enveloping manifold has all kinds of spikes and dents. Yellow can in practice seem as bright or brighter than white. Don't have the site handy, but the word on the street is that it might take four color primaries to construct a perceptually smooth manifold.
posted by StickyCarpet at 1:20 PM on July 18, 2011

"We certainly cannot follow the example of Odysseus and, going down to Hades, tempt with a bowl of blood a representative sample of native speakers to label particular areas of the standard Munsell color continuum ..."

They should try it with a Pantone book.
posted by Mcable at 2:00 PM on July 18, 2011 [1 favorite]

One reason we can't label particular areas of the standard Munsell color continuum, is that it's not continuous.

Turning vaguely serious for once, theoretically it IS continuous. That's the whole concept of it. The 3d shape of it isn't symmetrical, but that's a different issue. And the Munsell system identifies 10 primaries.
posted by Mcable at 2:17 PM on July 18, 2011

They should try it with a Pantone book.

I don't think the Pantone book would tempt them either unless you went specifically to the graphic designer's hell. You'll know it by the slightly off colors and bad kerning on the signs.
posted by doctor_negative at 2:22 PM on July 18, 2011 [2 favorites]

You'll know it by the slightly off colors and bad kerning on the signs.

And the signs are all set in Arial and Comic Sans.
posted by Mcable at 2:33 PM on July 18, 2011

Anyone interested in the intricate relationship between colour words and colour perception should read Through the Language Glass by Guy Deutscher. Among other things, it covers Homer’s unusual (to modern English speakers) use of colour vocabulary, including the surprisingly accurate analysis in Gladstone’s 1876 magnum opus on Homer. (And, yes, it’s that Gladstone — the nineteenth-century British Prime Minister.)
posted by Tetch at 4:24 PM on July 18, 2011 [2 favorites]

This general background may be appropriate. IANAA (anthropologist).

A world not only full of gods, but strangely colored to modern eyes.

It also suggests the depth of bullshit evinced by modern racialist thought about classical antiquity, which this book shows in detail.
posted by bad grammar at 5:58 PM on July 18, 2011

And the Munsell system identifies 10 primaries.
posted by Mcable

Don't believe the hype. This book, COLOR SPACE, Kuehni, identifies some spikes and cavities in the Munsell space. If that's not the right reference, I'll try to post another.

Theoretically, giant spikes can be differentiated to continuity at some tiny level. The right word might be smooth, but I think we could say reasonably perceptually continuous. There are kooky spikes in the blue-green darks, among other places.

Those 10 Munsell primaries are landmarks in a three-space, and that three-space has cliffs and chasms.
posted by StickyCarpet at 7:02 PM on July 18, 2011 [1 favorite]

This reminds me of a section of Cytowic's The Man Who Tasted Shapes.

My real name is in my profile, so I try extra hard to be civil. Cytowic is a dick. Check out Lawernce Marks's Unity Of The Senses, if you can find it.
posted by StickyCarpet at 7:05 PM on July 18, 2011

everichon, benigne.
posted by nangar at 1:57 AM on July 19, 2011

> A world ... strangely colored to modern eyes.

Well, not really. Take the "yellow" foam thing from the Gellius quotation. In Gellius' description, if I'm following it, flavum is shade of viride (green), or at least often names things we consider shades of green, usually "a mixture of green and red and white" (ie. "yellow") but also pale green or even pale blue-green.

So, given that "marble" can be a verb in English, the verse fragment marmore flavo caeruleum spumat mare means something like "the blue sea is marbled with pale foam." (And "sky-colored" is probably a better translation than "blue" for caeruleum. It can also mean "sky-grey.") There's nothing really weird about it, though the usage of flavum apparently varied in ancient times as well. Favorinus thought this description was odd too.

Gellius' description of flavum makes sense if you think in terms of the way RGB hexadecimal colors work (ie. a description purely in terms of combining different colors of light). Red is FF0000. bright green is 00FF00. FFFF00 (red + green) is yellow. 888800 (the same combination at half intensity) is olive green. So flavum, according to Gellius, seems to cover any pale color with a major a major green component from about FFFF00 (yellow) to 88FF88 (pale green) to 88FFFF (pale cyan).

There's nothing optically weird about this at all. Latin color terms seem to name perfectly coherent categories when you have somebody like Gellius to explain them. But they don't match up with the ranges designated by English color names very well.
posted by nangar at 4:43 AM on July 19, 2011

Getting involved in the Munsell derail a bit.

The Munsell color system describes colors in terms of five basic colors, corresponding approximately to German color names. They're not "primaries" in the sense of light primaries or print primaries. (I'm not sure if the correspondence to German was intentional, but most research in optics in Munsell's time would have been published in German. It seems likely he just adopted the classification used in German literature.)

I'm not sure what's up with the "spikes." I think this is due to the differences between RGB color space and visible color space. Computer monitors can't mix highly saturated blue-greens for instance, though you can certainly see them, and they're certainly printable with the right pigments. (Cuprous phthalocyanine works nicely.)

Also, the human eye is less sensitive to colors in the blue range than it is to other parts of spectrum, so visible color space is not symmetrical. We simply are less sensitive to some colors than others. RGB space describes what computer monitors can display. Munsell's system attempts to describe what you can see. They're not describing exactly the same thing.
posted by nangar at 6:44 AM on July 19, 2011 [1 favorite]