Improve your Color Harmony

"Winter Pines", Oil on Canvas, 20x12 inches
Jerry Lebo, 2007

I apologize for not having posted to my blog in a while. As some of you may know, I was traveling during January--which partly explains my silence. But I also managed to catch a case of “Dengue Fever” while visiting one of several tropical locations on my itinerary. There were only about 100-200 cases of Dengue Fever in the US last year—so most doctors don't even know how to treat it. I fortunately had access to a good doctor and have slowly recovered. But it has set me back in the studio and blog a bit—as I have been not been able to do much over the last three weeks.

Anyway, I am back in the studio and thinking about art--and I am very excited to be back blogging. I plan to use my blog over the next few week to talk about color, which is a topic my students are asking a lot of questions about these days. So, I plan to post some of the ideas and lessons I am sharing with them--and I hope you will be able to take something away that you will be able to use in your own art.

The first issue I want to talk about is "color harmony". In my experience, beginning and intermediate artist are always looking for a solution to color problems. Their colors look dingy, or their paintings don't have any spark--and look drab or dull (especially next to other artist's work). Or, their color is way over the top and looks gaudy and/or disorganized. They discover that painting like Monet or Matisse is not as easy as it looks.

After enough frustration, an artist will typically seek out new ideas (usually go to the bookstore or sign up for a class) and soon discover "color triads" and the famous "color wheel"--and various other color systems--and soon think that there is a formula or book they can read and learn about color. They buy books called "color harmony" or "color for artists" and read them carefully, then they go back to the studio and try out various ideas--and perhaps their paintings get a bit better--but generally they end up unhappy. The core reason is simple, there is no way for another person to give you advice about what color you should use—or what will make your paintings better. You have to discover it yourself. Color is a personal experience--and what one person likes about color is not necessarily what you will like. Also, learning how to use color requires an active engagement—not a passive one. You cannot learn about color by reading a book--you have to look and learn visually.

This leads me to my main hypothesis, the main reason I think artists struggle with color is simple: They don't spend enough time working with it in an active way. I remember when I was studying at the Washington Studio School—no one talked about color. Everyone talked about painting and drawing—but color was assumed to be something you knew about. We talked about values—but never really talked about how and if color became important in painting. Not surprising I have spent a lot of years struggling with color in my own painting. Anyway, I am going to try to talk about some of what I have learned over the years.

The first issue to understand is that the human eye is very sensitive to color and can see millions of colors--while it can only see a much narrower band of grayscale value (some think about a hundred values). At the same time, the eye is easily fooled by color and finds it difficult to see the difference between colors. I know this sounds contradictory—but it is true. They eye can see a lot of colors, but at the same time has difficultly seeing differences between two colors—and (as I have mentioned before in previous posts) the eye has a hard time seeing value and color at the same time.

I was doing an exercise with one of my students the other day where she had a series of red color chips taken from the Munsell Student Color Set. These are basically value and chroma combinations in one color range, in this case red. If you look at these chips all in a pile, some of them look very close in value and color—and are vary hard to tell apart. I took two of the chips that appeared very close in color/value and put them on top of a larger swatch of red color in a reproduction of a painting. One of the chips was very close to the color in the book, but when the other chip was put on the color—it was clear that it was remarkably different. So how can two chips that look so similar sitting together on the table—look so different when placed on large swatch of a color. The answer is—without a reference point—your eye finds it very hard to decide the difference between two colors. But, once you have a reference color—it has the effect of accentuating their difference (I talked about this effect in a previous post). In fact the difference can become so glaring that you will feel stupid you couldn’t see it before. Try it sometime—mix two reds that are close in value and chroma—perhaps one-half value step apart. Then paint a piece of paper with one of the colors. Put the other color on top of it. The difference that was hard to see when comparing two piles of paint—will become obvious when one color is on top of the other.

Okay, so what does this have to do with color harmony? The point I want to make is that when an artist goes to mix a “red”--there are thousands and thousands (if not millions) of variations in red to chose from--and believe me (I have tried it), a small change in color can make a big difference in what happens in your paintings. And, where is the reference for the right color? In your mind? If you get it wrong by just a bit--you may take your painting in a whole different direction.

Don’t believe me? Next time you are at a museum buy a reproduction of a painting you like—and then mix a color that is just slightly off of one of the colors in the painting (it doesn’t even have to be that much off). Paint over the reproduction with the “slightly off” color and then stand back and look at the painting from a distance. I think you will see very quickly that even a slight change in color can have a big effect in the overall look and feeling of a painting. Color is strange in this way—you have to mix the right one out of the possibly millions of choices or you won’t get the effect you are after. I think this is one of the biggest hurdles to improving color in paintings—taking the time to find the right color—not the “close enough” color. Most artists are in a rush to paint. Thus, rather than finding the right solutions—they stop at something “close”.

I will be giving you some exercises to help with color over the next few weeks. But, let me first say that learning about how to use color is a lifelong pursuit. If you have been following my blog, I have been talking a lot about values and working with grayscales. If you have been doing the exercises that I recommend, you have seen how hard it is to learn to see grayscale values. Well, mastering color is much harder than grayscale values. In fact, the problem with color is that there is no scale, formula, or system that you can use to learn about it. As Albers says, it is purely a psychological experience--and thus a moving target. Even your own sense and taste for color will change over time--so there is no way to learn a system and then walk away. It will be an on-going challenge that can keep you busy for a lifetime.

Okay, before I close let me give you a simple exercise to get started working on improving your sense of color. This is one I often use myself. First, find a reference—a colorful photo or some colored paper—or even a painting that has been reproduced. Look for a reference that has clear areas of color to work from—that is, areas of color large enough so you can make direct comparisons. Pick a color in your reference and try to mix that color on your palette. Do this exercise with the photo sitting a foot or so away from your palette—so you can see it, but not make a direct comparison while mixing. When you think you have it matched, compare the mixed pile directly to the photo reference. How close did you get? Don’t settle for "close enough"—try re-mixing until you have an exact match. If you have never done this exercise before, I would be surprised if you can even get it exact on the first try. If you are struggling, move the photo reference onto your palette so you can make a direct comparison while you are mixing—I bet it will be easier to get close. But even with the reference sitting there on your palette--you may struggle to get an exact match. Okay, here is the final twist. Do the exercise with a limited palette (say, blue, yellow, red, white). See how close you can get with only the primaries--that is my favorite way to do the exercise.

I recommend you do this exercise until you have mixed 5-6 different reference colors. If you feel it is too easy—time yourself. See if you can make a perfect match in less than two minutes or less than one minute. Or move the reference further away from you. I know that may sounds strange, but I believe that just taking the time to mix colors will improve your painting. If you do this as a warm-up every day before you start painting—you may be surprised with the result. You should soon not only be able to mix the color you want more closely and more quickly—but you will soon realize that you are able to see the subtle differences between various colors much more clearly as well. You will also be better able to remember colors that you have mixed in your minds eye.

Okay, that is it for now. There is much more to be learned about color, but this is a start. I will give you some more exercises in coming posts. Now, go to your studio and paint.

All the best, sixtyminuteartist.

Painting from Photographs

"Evening Sky Study", Oil on Panel, 8 x 4.5 inches

Jerry Lebo, 2007

Two posts back I showed you a simple exercise to help see grayscale values more clearly by painting directly on a printed photo. If you have tried this exercise, you have probably noticed that it may not lead to a very interesting result--at least not something that you would want to hang on the wall. There are several reasons. First, as Josef Albers points out in his book, "Interaction of Color", photos lose the finer nuances and delicate relationships that one would see in nature. Put another way, there is much less information to paint from when painting from photos in comparison to standing outside and looking at the real thing.

Another problem with painting from photos is that photos distort value relationships. If you have every compared a photo of something to the actual subject, you will have seen that colors change, and values tend to merge together. For example, the shadows in photographs tend to darken and simplify—and the lightest lights tend to get higher in value and merge together. If you want to test this—try this experiment. On a cloudy day, look up at the sky and take a few minutes to notice the all the shapes and value changes you can see. Then, take a picture. You will see very quickly (assuming that you did not apply a variety of photo filters and exposure techniques) that the subtle variations will be lost in the photo. In some cases, the sky may simply become a single value. As Alber's puts it in his book:

"…photography registers all lights lighter and all darks darker than the more adjustable eye perceives them. The eye also distinguishes better the so-called middle grays, which in photography after are flattened if not lost."

There are other distortions taking place in photos that are less obvious. For example, take a picture of the Grand Canyon and print it out as a 4x6 print—do you think that the photo will give you the same sensation as standing at the Grand Canyon--obviously not? A photo is much smaller in scale than the real thing, so the physical sensation and peripheral sensations of standing there are lost. This is in addition to the value and color distortions I mentioned above. It is no wonder people's eyes glaze over when you show them vacation photos—these are a poor substitute for the real thing.

This brings me to the point of this posting. That is, that a good painting has to convey something to the viewer--and a direct reproduction of photo is unlikely to provide the necessary information to convey that sensation. Another way to think about this is to consider paintings that you like--and compare them to photos you have seen of the same thing. For instance, would you prefer to see a Manet painting of a bowl of peaches—or a photo? What would it take to make a photo of peaches as interesting, or better, than a painting? As Edgar Payne says, "A pictorial representation (painting or photo) is always a translation." The art is thus in the translation. So, if you are going to paint from photos—don’t rely entirely on the photo—think about the translation.

There are of course many ways to translate visual ideas in paint, and I am not going to be able to say much in a few paragraphs. Instead, let me focus on some ideas to help you if you are going to paint from photos in your studio.

1. Don’t lose your focal area. A good painting needs an area of emphasis, or focal area. One of the problems I see in photos, since they tend to distort values, is that they may lose or even move the artist’s intended focal point. For example, in landscape photos the far shadows darken and tend to draw undue attention. In other cases, the subtle changes in values that naturally occur in photos will reduce the emphasis/complexity out your intended focal area. Or, the opposite could occur—where more than one point of emphasis emerges (for example, due to a over-darkening of a secondary shadow). Think back to what drew your attention when you took the photo—and work to ensure that that is the focal point in your painting. Don’t rely on the photo too much.

2. Add variation in shadows. I can almost always tell a painting that has been painted from a photograph—mainly due to the darkness and simplification of the shadows. The most common mistake I see is either to make the distant shadows too dark—or the near shadows too simplified (and dark?). Put an object on the still life stand and take a picture. If you compare the photo to the actual subject—you will see there are subtle value and color changes in the shadows that will lost in the photo. If you are going to paint from the photo, you will need to put these back. If you hold your photo directly under a bright light, you may still be able to see some of the subtle changes that were actually there. If not, I recommend mixing a least three shadow values (two of which will be a half value step or so above the darkest value you plan to use). You should use these higher two values for the majority of your shadow area(s). The darkest value can then be used as an accent as needed—noting that the darkest values tend to occur where two objects meet. A cast shadow will always have some reflected light in it, so will never be the darkest dark—despite the fact they may appear that way in a photograph.

3. Reserve your highest highs. Just like shadows, the highest highs in photos tend to brighten and merge. Thus, do the same as you would for shadows. That is, don’t use white or near white values right off the bat. Start by taking down the highest values in your photograph by least 1 to 1.5 value steps. For example, the sky in most photos will appear to be nearly white—but if you look outside—the sky is usually 1-2 value steps lower (hint: hold a white cloth out at arms length and look at the sky--which is darker?). Also, the local color of the highest values in your photo may be lost or distorted—that is, they will appear washed out. You may thus need to add a bit of color back into these areas (the color of the sky on a cloudy day is not white). If you don’t’ know what color to add—start with the local color of the object and the color of your light source. The highlight is usually some variation of these.

Let me close with a quote from JF Carlson from his book on landscape painting where he says, "We must not train our eyes to copy tone for tone, but think of the bearing of such colors and harmonies up the main idea of our picture.” In other words, the problem with painting from photographs is not only that they are not an accurate record of values and/or colors—but a good painting is more than a direct copy. It is a translation of reality. So, when painting from a photograph, always keep in mind the sensation and idea that you are trying to communicate—and adjust your value and colors based on what is needed to covey that to the viewer. Don't be a slave to your reference photos. Even if you managed to get it perfectly correct, it is not likely to result in a compelling painting, so go ahead and loosen up.

Hope that is helpful.

All the best, sixtyminuteartist.