Now we’re in uncharted territory: this is the chapter of Making Color Sing where I stopped reading (and blogging) about this book in 2022. At the time, I thought, “This chapter is confusing. I must just be confused because I’m not advanced enough to get it.” Now I think “I was (and am) confused because this book is written in a confusing way.”
Let’s see if I can decipher it now.
We get off on the wrong foot immediately, with this first sentence:
Master artists do not describe a color simply as red, yellow, or blue. Instead, they qualify it as a warm blue or a cool red, a muted or an intense yellow.
Jeanne Dobie, Making Color Sing p. 30
Oh, Lord. “Master artists?” Give me a break.
This bit of snobbery really has nothing to do with the rest of the paragraph, which lays out of the main thesis of this chapter.
Insight #1: Create depth with color temperature
Pushing a color into a cooler shade causes it to recede, while pulling its neighbor into a warmer color makes the neighbor advance. Side by side, the two colors no longer appear to be on the same plane. You create space in your painting!
Jeanne Dobie, Making Colors Sing, p. 30
So, basically this chapter is about using color temperature to create a sense of depth. The crucial insight is the idea that things that are further away appear bluer.
It is my understanding that this is true in some circumstances because of atmospheric perspective. The further away something is, the more air there is between it and us. In that air, there are water molecules. Rayleigh scattering of the blue wavelengths in water molecules make that air appear subtly, transparently blue – the same reason the sky appears blue.
The actionable information for you, the a painter, is that since things that are further away appear bluer in real life, you can paint things bluer to make them seem further away in your painting.
In art books, I have generally seen this idea stated as “cool colors recede, warm colors advance.” I’m not sure if this originated with Dobie, but here she takes it a step further by making it an active action that you, the artist, can take: “pushing” elements into the background with the addition of more blue, or “pulling” them into the foreground with more warm red or yellow.
My Thoughts on Color Temperature as a Depth Cue
Color temperature (blueness) as a depth cue is, I think, a reasonable topic for a book on color. Unfortunately, it’s hard to teach in isolation outside of a larger lesson on depth cues, especially since it is one of the weakest. It is easily overridden by other depth cues such as:
- Value – lighter colors recede.
- Chroma – grayer colors recede.
- Compositional elements – e.g. overlapping, linear perspective.
Another problem is that there are many situations where the “cool colors recede” rule doesn’t apply:
- Sunrise/sunset – or any other atmospheric condition where the sky is a color other than blue/white/gray.
- Elements with a local color of bluish white (or blue for that matter) – these may appear less blue at a distance because they also become lower chroma.
- Small spaces – atmospheric perspective really only applies over long distances, such as in grand landscapes. You wouldn’t expect to note this effect indoors or between two adjacent buildings.
In truth, I think that color temperature has at best a supporting role in creating depth in a painting. When it is aligned with other depth cues, it can reinforce them. When it is in opposition, it tends to be overwritten by them.
Rather than “cool colors recede,” I prefer to say, “Bluer landscape elements may appear further from the viewer.” The softer language reflects how weakly I perceive the effect. Throughout this chapter, I tend to mentally replace the stronger language with weaker:
- recedes: may appear further from the viewer
- advances: may appear closer to the viewer
Insight #2: Color temperature is relative
Dobie explains that not only do entire hue categories have a temperature (blue is cool, red and yellow are warm), but there are also warmer or cooler specific hues within each category. You can have a “warm red” or a “cool red,” for example.
A lot of this chapter has to do with the relative color temperature of different hues, which I find tiresome since I think the warm and cool color binary makes no sense. These days, I tend to translate color temperature language into color bias language. Instead of talking about “warm red,” I will talk about “orange-biased red” or scarlet. Instead of talking about “cool red,” I might talk about “violet-leaning red”, or use a more specific term such as magenta, pink, or rose.
This next paragraph is the one that confused me into oblivion when I read this in 2022. Note that here Dobie is talking about a “neighboring color” as in the color that is adjoining in a painting or a scene, not on the color wheel.
The temperature of a color is always relative, because it depends on its neighboring color. Suppose you place French ultramarine blue next to cadmium red… Here there’s no trouble identifying the blue as a cool color and the red as a warm one. But now suppose you place the same blue next to lemon yellow. In this case, the yellow is the cool hue and the blue appears warmer than usual. In fact, the blue begins to look almost purple next to the cool yellow.
Jeanne Dobie, Making Color Sing, p. 30
I can now see a reasonable interpretation of this text – that the blue looks more purple when contrasted against the yellow compared to how blue it looks when contrasted against the red – but I initially interpreted this paragraph to be Dobie asserting that cool yellow is actually cooler than warm blue. This would be an extraordinary claim!
It is also contradicted by the accompanying illustration, which I’ve recreated below using Hansa Yellow Light (PY3) and Pyrrol Scarlet (PR255) in place of Lemon Yellow and Cadmium Red.
I was all ready to acknowledge that I had simply misinterpreted the text originally, but then, the caption of the image reads:
The yellow here seem to recede (cool) while the blue advances (warm).
What??? No it doesn’t!
One of the reasons it is obvious that yellow is in front is the layout of the diagram, which I copied faithfully from the book. The strip of yellow is longer on the top and bottom while the blue peeks out from both sides, making it strongly appear that yellow is on top or closer to the viewer.
Let’s make it more ambiguous with a tight crop.
Try to ignore my lack of neatness in allowing the colors to mingle a little at the border. Which color do you perceive as being closer to you? Is it yellow over blue, or blue over yellow? Personally, I think they are now so abstract that the question is meaningless – they appear to be on the same plane, which they are – but if pressed, I would say that yellow still appears closer than the blue.
Despite all evidence to the contrary, I simply cannot believe that Dobie is saying that cool yellow is cooler than warm blue. If cool yellow is cooler than warm blue, then it means that the “coolness” of cool yellow and the “warmness” or warm blue are absolute, not relative, as stated. If you translate it into color bias language, it’s the same as saying “green-leaning yellow is bluer than purple-leaning blue.” Clearly nonsense!
I can’t tell if she has misunderstood her own idea or if she has just chosen a spectacularly uninformative example and explained it in the most confusing possible way. I do agree with the sentence “The temperature of a color is always relative, because it depends on its neighboring color.” If I were to support it with an example, I would say something like:
“A ‘cool yellow’ like Lemon Yellow is cool compared to orange-yellow. But it is still warm compared to blue.”
It doesn’t have to be this hard.
Insight #3: Local color can fool you
Dobie urges the reader to “question your assumptions about local color.” I wish she defined “local color,” which I found to be an intimidating term upon first read. For those unfamiliar, it simply means the main color you know/expect an object to be under normal lighting conditions. For example, the “local color” of a tree trunk is brown.
But Dobie urges you not to just paint the tree trunk brown because you think it’s brown. Instead, notice how the colors of the object change due to light and shadow and time of day. Notice how your perception of the color may be different depending on the similarity or contrast with other colors around it.
Under an early evening light, for instance, a brown tree trunk may turn pink. Open your eyes! Local color can fool you. Is a road really dark and gray? After a rainstorm it may turn lavender and, under other conditions, may deepen to a rich red-purple… Does snow always have blue shadows? Not in Pennsylvania at four in the afternoon – then the snow turns an incredible pink from the low angle of the sun.
Jeanne Dobie, Making Colors Sing, p. 30
I think this is an extremely interesting insight and I want to know more! I wish it were an entire chapter with pictures, examples, and exercises. (I especially like that all the examples are things turning pink.) I’ll just have to explore it myself sometime. But for now, we soldier on!
Miscellaneous Insights About Outdoor Light
Dobie lists a bunch of facts about outdoor light. In this section she even includes a definition of atmospheric perspective, which makes me think that she thinks the the insight about “cool colors recede” is distinct from the idea of atmospheric perspective. Personally I think they are the same thing.
Other insights:
- The color of the sky is reflected in every plane that faces the sky, e.g. water, ground, rooftops. So on a day when the sky is blue, all of these things will appear bluer. If the sky is another color, expect that color to be present within the local color of these objects.
- The sky is more intense (higher-chroma) at the zenith and grayer at the horizon due to particles in the atmosphere.
- Bluish-white objects, such as clouds and snow, don’t obey the “bluer = further away” rule. Generally, white objects that are further away appear more creamy brown-gray. (I will tell you that this is also explained by atmospheric perspective, which causes items further away to appear bluer and lower chroma. In the case of a bright white object, the lower-chroma effect is more obvious.)
Exercises
Exercise #1: Stripes
There is an exercise buried in an image caption again! Again, I’ll provide my own painted version of the image.
The caption in the book reads:
Experiment with push-pull reactions: Merely painting a color light or darker does not give it a push-pull relationship. Change the same greens to warm and cool to create a push-pull reaction. Repeat this exercise with other colors, such as purple or blue.
Jeanne Dobie, Making ColorS Singe p. 31
What she appears to be saying is that the upper pair of color stripes, the lighter and darker values of the same hue of mid green, do not create a “push-pull” reaction, but the lower pair of color stripes do.
My problem with this is that I don’t see it. Both of these look pretty abstract to me. I don’t see one stripe as being closer to me than the other.
If I squint and force myself to see them as landscapes, then the lighter stripe appears to “recede” or be further away for both of them. Actually, there are actually several cues prompting me to see depth here:
- Value depth cues: Lighter values recede, which affects both examples.
- Chroma depth cues: Higher chroma values advance, which affects the bottom example (in addition to warmth)
- Composition depth cues: Relative position of the light over dark stripes prompts me to see this as a landscape and interpret the lighter color as the sky.
The big test of this would be a counterexample: if I do the same thing, but the bottom stripe in addition to being darker and higher-chroma than the top stripe is also cooler, will that “cooler” color recede?
I see the bottom stripe as being closer to the viewer in all three of these examples. Varying the color temperature does not affect it for me. Your mileage may vary.
This might be a first: I came in agreeing with Dobie’s general premise and, through the exercise, came to doubt it!
Exercise #2: Jewel in a Setting, Complementary Hue Edition
Let’s try to reel it back here.
I decided that I needed to create a truly ambiguous situation where it isn’t clear which element is closer to the viewer and the only difference is color temperature. So I chose the most unambiguously warm/cool hue pairing I could – orange and blue – painted some simple squares, as in the “setting of a jewel” example from Mouse Power, painting either the outer square or the center orange or blue. If “cooler colors recede,” then the blue should look like the background in both: an orange center or an orange frame should appear to be placed on top.
Urgh…. I don’t know anymore.
Exercise #3: Cool and Warm Yellow Background Comparison
Last week, I did a painting where the background is yellow – the warmer color is meant to be further back, while the cooler greens are meant to be in the foreground.
This chapter made me wonder if the use of cooler or warmer yellows for that background meadow might have made a difference in the sense of depth, so I did some thumbnails.
Following the theory, the yellow meadow in the top thumbnail, painted with cool Hansa Yellow Light (PY3), should be more successful at appearing to seem further away from the foreground, while the bottom meadows, painted with Hansa Yellow Medium (PY97) and Hansa Yellow Deep (PY65) should be more confusing or flatter.
I kind of see it, though my instinct is that all of them are fairly unsuccessful because every yellow is too bright, and that lighter value or more muted colors would probably have more of an effect.
Chapter Rating
I do not feel that this chapter adequately establishes its thesis idea. As it turns out, it is quite difficult to show examples where color temperature is the only depth cue, or the most important one, and Dobie doesn’t really try. Aside from the disastrous diagrams, the images included are almost entirely paintings that demonstrate exceptions – situations where the background isn’t bluer than the foreground – leaving the central premise, the idea that cool colors recede, largely undemonstrated. After engaging with the diagrams and exercise, I came away feeling less confident in the idea than I started!
But the most important problem with this chapter is the confusing way in which it is written, which almost seems intentionally designed to promote the wrong idea that color temperature is absolute, despite stating that it is relative. I was confused and misled by the chapter the first time I read it, and it actually took me a long time to unlearn the wrong ideas that I got. This goes beyond the type of gibberish that led me to close-read chapter 1, because it doesn’t just feel ambiguous and confusing – it feels like it is unambiguously saying the wrong thing.
I don’t want to give this chapter a low rating, because there are interesting ideas mixed in here, such as the thoughts about local color. But I wish these ideas had gotten their own chapter instead of being tossed miscellaneously in with unrelated ideas about atmospheric perspective and color temperature.
Overall rating: D.
Key Useful Ideas
- Warm colors pop forward, cool colors recede. Atmopsheric perspective causes faraway objects to appear bluer, lower-chroma, and lighter-valued. Use hue, chroma, and value to create or emphasize depth cues.
- Color temperature is relative. A cool yellow is cool compared to warm yellow, but not compared to blue.
- Local color can fool you. Colors appear different in different lighting conditions. Observe, don’t assume.