I haven’t been happy with my ability to get a glowing quality of light in the limited palette portraits I have done so far, so I decided to take on the light in the wonderful Stone Bridge and see what I could learn. My copy is smaller than the original so lacks a lot of detail, and I didn’t try for exact shapes or drawing; my goal was mainly to try to get the feeling of light. I tried several changes in technique, and I think this was more successful than any of the others in that regard. The changes were mainly to follow some advice from Tad Spurgeon’s book, painting grays and colors in mostly distinct layers, and using chalk putty much more than titanium for whites since I lack lead white in my palette. I found that sticking to the more transparent putty white requires a mindset of preserving the light of the canvas much more than I have been doing. I don’t think that was his method, but it’s helping me avoid titanium chalkiness.
I started with a wash of yellow ochre, much lighter than the orange-brown washes I used previously. (Some of it can still be seen in the upper right linework.) The basic shapes and drawing were laid in with a reddish burnt umber and raw umber, then the foreground darks were done mostly with a back-and-forth of grays, the red umber, and a green made from black and lemon yellow (bismuth yellow is what I have). A layer of grays provided the atmospheric recession, but my black seemed to add a purple tinge that was wrong, so the green mix became important to balance that. (Overall, the key color in this painting turned out to be lemon yellow.) I never got the purplish touch out of the shadows in the big tree; Rembrandt’s are definitely warmer.
The sky and water were originally very problematic for the same reason. I set out to use only chalk putty as my white. This worked well over the light background, but when I added a little black hoping to get the blues, it added a purplish cast which was correct in some areas but was killing the warm aqua tones that give the sky a lot of glow. I tried tiny bits of blue and yellow added to the putty, and finally decided all I had that gave it the aqua tint was a decidedly non-Old Master’s phthalo blue mixed with the bismuth yellow. (I’ve read that their azurite could be a greenish blue, so perhaps that’s what he used.) A little black was added back in to move into the more purplish areas in the lower right and upper left.
The light horizon sky and cloud areas on the left were achieved mostly with white putty alone. Interesting that white over the yellow ochre ground gave a slightly rose tint, rather than the strictly cooling effect I expected. Everything in this type of painting is context! My sky is not nearly as finely graded as his; I had trouble here with the putty mix, which handled a lot like cake frosting, clumping up unless I heat set it frequently. I got close enough to his color that I was satisfied and let well enough alone. I’m pretty confident that I could make it work in oils.
I never figured out for sure if the drawn mass in the upper right is a foreground tree or a storm cloud, though my best guess is a tree. I didn’t come close to his technique here. I laid on a raw umber, then drew lines with ganged wires to pull back to the yellow. This was much too harsh; maybe those round bristles from a hairbrush would be better. It’s possible to me that he built a transparent putty texture over the ground and let it dry, then glazed dark and wiped the texture back. I just can’t tell on my reproduction, and it wasn’t my big concern, so I went with my wire version, aiming only for the general effect.
The grace notes on the trees and bridge are the heart of the painting to me; everything else builds to these. Here it seemed important that the ground color was light, with the ochre acting as a tiny support system for the lemon yellow/white highlights. It proved important that shapes have a grayed transition into shadow to help keep the lights luminous. There’s a wonderful loopy texture on his highlights that maybe comes from lead white, which I don’t have. It would be fun to do a full-size of just the tree to see if I could get it better.
As with the other copies, I am struck with the amazing ability of limited palette paintings to give you advice, that is to tell you especially when things look wrong. Even slightly discordant notes seem to jump off the page. And Rembrandt amazes me again with his complex support structure for values. So much of my first impressions are of virtuosic brushwork, sketchiness (very deceptive), and limpid grace notes. When you try to reproduce these, you find that all of it is carefully supported with finely graduated values and colors that allow shapes to read correctly and light areas to glow. I find some of this can be achieved with the “turbid medium effect” (love that name!) of glazing lighter color over dark, resulting in slight graying in addition to coloring. You can see the turbid medium effect in the distant trees on the left, where a little bit of the sky color is dragged over the tree tops to gray them. In other places a careful adjustment of grayed complementaries gives volume and makes colors more vivid.
I can still see things wrong and any number of adjustments I could make, but I’ve learned a lot and am ready to move on. Next up, perhaps I will try to apply some of what I’ve learned to another portrait of Terra.