Back on the Road, 18 x 36 inches, oil on linen
I guess the summer is here–still mid-June, but we have already had a couple of 100F+ days. I hope it's not going to be like this all summer long!
Not so much nowadays, but when I was younger, summers meant road trips. You know, on the open road, radio blasting, diner food, destination unknown, (or at least lodging uncertain), one hundred bottles of beer on the wall… ah, the good old days….
The open road is one of my recurring motifs, and today I thought I'd share the process of one such recent painting. This is a larger studio painting, 18 x 36. I started by doing some thumbnails in my sketchbook. Nothing too tight or detailed, just a general idea.
I knew that I was going to have a pinkish sky, a large central eucalyptus tree, the open road taking us into the picture and into the distance. I didn't define the smaller tree groupings because I knew I would be moving those around later. I thought this was an area where I could have a lot of flexibility. Still, I had a fairly good idea of what it was going to look like.
I started by toning the canvas. I used a mixture of ultramarine and Transparent Oxide Red, with Gamsol. Just brushed it on like watercolor, and wiped most of it off. I tone the canvas to kill the white (easier to judge color/ value on a non-extreme context)
I then used the same mixture to draw out my design. Keeping things fairly loose from the get go. I try to take advantage of the fact that drawing trees and clouds are more forgiving than drawing buildings or figures. I avoid tightly defining shapes at this point.
Here I've indicated the telephone poles. I've also used a lighter wash to delineate the distant hills. If I can get a good sense of where I'm going with the design at this early stage, I feel fairly confident about pulling this off. Sometimes it's a struggle and I end up abandoning it at this stage. You can usually spot major design issues before getting too far along. The problems are almost never small detail-ly things. They're usually big issues, like balancing visual weights, symmetry vs. asymmetry, focal point(s) or lack there of, and whether I have a single, clear statement or not. If I can't make that statement in three values, it's probably too complex.
This is what it looks like close up. As you can see, the mixture of T.O.R and Ultramarine and Gamsol makes a nice neutral, clean dark wash. It's transparent, and can be pushed towards warm or cool by adjusting the amount of one or the other color.
I usually aim for a slightly warmer tone here. Also, I try to keep the paint very thin. Dark, but not thicker than absolutely necessary to achieve that dark value. As this is a wash using solvent, it'll dry fairly quickly and much lighter, too. Which is not a problem as I will cover it all up anyway. It's more important to me to keep it thin so that subsequent layer stays clean as I apply strokes on top of this dark brown color.
Here I have started painting with colors. From here on, 99% of the application is opaque. I'll maintain transparency only where it's very dark, like the interior of the tree peeking through the green top layers. Other than that, I like to paint everything opaque. The rule is, if you can see color or detail, it's because it's illuminated (if not directly by the sun, then by reflected or ambient light) and if it's illuminated, it's opaque. A handy rule, though not carved in stone because opacity and value can be manipulated independent of each other. It's just one of those rules of thumb that works well most of the time.
Except for the finished image, all the progress shots are taken with my iPhone so I apologize for the lower quality. But I think the points come across OK.
At this point, I'm trying to figure out the relationship between the foreground and background. The greens in the foreground are much higher in chroma. Richer. As we go back in space, I'm dropping the chroma (making it grayer) and moving it toward the blue-violet grey of the distant hills.
I do this by dividing the distance–from the closest to the farthest–into three or four sections, and systematically reducing the saturation by using less yellows. In actuality, it's not as simplistic as dropping yellows - the subtleties require mixing all three primaries plus white to get a nice muted color, but using less yellow in the distance is a big part of the color mixing process here.
Determining the general color of the sky. I know I want it to be pinkish. But not screaming pink. I want a dusty pink. Red and yellow plus white makes a nice peachy orange color, and I can make it more muted by adding blue to the mix.
There are many variations of the dusty, rosy color in the sky and some parts are more orange than pink, but a lot of it leans toward red violet. Violet is one of those colors for which making warm and cool variations are fairly easy. Red violet is warmer, and blue violet is cooler. If I use the same puddle to make these two close variations, I have the foundation for the light and shadow colors of the violet sky.
I tried to make them closely related, so as to not have too jumpy a contrast in the sky. I want the sky to be active and interesting, but not so much so that it takes away from, or competes with, the foreground elements.
It's difficult to tell whether the sky color works before the whole thing is filled in. Often a test patch of color may look good, but turns out to be wrong when it becomes contextual. I think the only ways to deal with this problem is experience, and doing small color studies to test it out. When in doubt, try it!
Now I'm starting to add value variations in the sky, in an attempt to form some volumous cloud masses. Using a cooler, darker color variation of the general dusty pink, I try to imagine which areas fall away from the light source.
The light is coming in from the right, so the left side of the imagined puff balls gets the shadow colors. So too, do the bottoms of the puff balls.
I go back and forth between the light and the shadow to design the sky. This may take a few hours or a few days. I like the organic nature of the process where I'm just pushing and pulling until it starts to make sense. I don't typically do a tight design of clouds or foliage masses beforehand- I like to make them up as I go, allowing myself to change my mind as many times as necessary.
Getting close to finish. I've covered the canvas, added thicker strokes over almost everything, and refined the relationships between shapes. The tree masses had light and shadow colors on them already, but my lights were too dark so they were hard to see. I clarified the light-shadow pattern by pushing the value range apart; lighter lights.
I moved the right edge of the road inward. Not that it was wrong before, but it made an odd shape near the bottom of the picture. By bringing in that edge, I think we can step into the picture a little more easily, and not dwell on the shape there. Yes, it makes the road narrower, but that's OK. This road doesn't have to be a highway.
The slight overall color change between this shot and the earlier ones, is due to the fact that the earlier ones were taken with my phone, and the last one was done with a decent set up.
I'm waiting for my frame to arrive now. I think it'll look nice with a heavy, dark frame. Can't wait to see it framed!
If this kind of post interests you, and you'd like a more in-depth, hands-on instruction, you may be interested in an upcoming workshop I'm doing in Lowell, Michigan. It's a 3-day plein air landscape painting workshop (weather permitting–in adverse weather we'll take it inside) and I'll be covering all the important points I talked about on this post, and then some. Find out more on the Franciscan Life Process Center's website!