Thursday, July 15, 2010

More about the Shiskin lesson

Here is the painting for which the sketch I showed you last night must have been the study. This was provided by thehiddenplace in the comments. Thanks, behind the scenes a lot of e-mail arrives asking questions and providing me with interesting information. Perhaps someday this blog will be self writing.

Several people have asked me when I would do a workshop in California. I might be there in the fall, if you are interested in that e-mail me and let me know. My e-mail address is over on the right under ask Stape. I would love too, by the way, but there needs to be some interest.
I received this e-mail. I have highlighted the e-mail message in aqua velva and then inserted my commentary in red text.

Dear Stape,

Your breakdown of Shishkin’s piece was thought-provoking- great bullets. I am realizing more and more that my color sense is downright horrendous, so I am trying to improve my decision-making:Here exactly is how my Color thought process would work in copying this piece:

a) The trees look like cad yellow and cobalt blue- that’s my light, but there are some burnt sienna notes, probably for the local tree color.

I don't know what color the piece actually is, there is a lot of variation in monitors and digital captures, but for the sake of this exercise lets assume it is true. I don't use cobalt much except in skies, but that might work. I would have to hit it once and see. My default choice would probably be ultramarine.

b) That means the shadows should be the complement, which is an orangey-violet. I’ll use ultramarine blue (because I want to use a different blue than in the light) and cad red (because that’s browner and warmer than aliz. crimson) and I’ll mute it with raw umber( that will help me keep it dark enough, plus warm). Or maybe just ultramarine blue and burnt sienna (that’s orangey!)- that way I’m incorporating the burnt sienna I’ve already used (color harmony!). But wait, isn’t this a case of cool light, warm shadow? Maybe I should use lemon yellow, not cad yellow, for the lights!

I would probably reach for the cobalt violet, I might modify it with a little ocher. I have no idea if Shiskin actually used it, but it is about the right hue. I would also have ultramarine and burnt sienna handy. Again I would try that, if it looked good. OK, if not, I would try another ploy. It is hard to know without actually trying to hit the note and then comparing them.I have no lemon yellow only cad yellow vulgaris. I can cool a yellow if need be so cad. lemon seems extraneous. I know a lot of landscape painters do like it though. I have only one cadmium, cadmium yellow, sometimes I have cad red light but I am just as happy with the proprietary reds like Rembrandt's or Sennelier's. I like those which are closest to vermilion.

c) For the dark or medium midtones I should mix the light color with this shadow color, because we should see some related color of the tree in light in that mid-shadow.

I would try to mix up the color I perceive rather than trying to arrive at it mathematically. I would mix a color that "went" with the family of hues in my lights. I probably still have a pool of color on my palette left over from painting the lights that I could doctor up and use too.

d) The grass is getting more sunlight, so it’s the cobalt blue with a greater part of cad yellow(or maybe lemon yellow, because it’s supposed to be a cool light), and some aliz. crimson to mute it a bit; the shadow grasses would be ultramarine blue and a duller yellow, like yellow ochre, muted with some cad red (because that’s warmer than aliz. crimson).

That sounds good but I would have some viridian rather than the cobalt blue on the bottom of this mixture.I have cad red but almost always try to use something else, it doesn't seem to be a very clean mixer, but in this instance it would probably work. I would again be peddling cobalt violet there. I would expect that Shiskin used vermilion, that was the common warm red in those days. It is a great color and I learned to paint using it, Now it is replaced by cadmium red because it is mercuric sulphide and nasty poisonous.Which brings us to an interesting dilemma, should a copy be done with original period materials, or is it OK to use contemporary pigments. I would argue the latter as long as you can get close to the original look, because the real point is to learn, and you won't have period materials when you are doing your own painting.. Also you might not want to use some of the colors that those guys did, they have been replaced because of their toxicity and the availability of better, more permanent pigments.

e) The sky I want the coolest, so it’s ultramarine blue and aliz crimson.

f) White of course wherever I need a lighter value.

If you wanted to be like Shiskin, that would have to be flake (lead) white of course. Use Titanium instead.

Does my logic sound correct, am I close, or am I insane?

The first thing that happens when battle is joined is that all planning goes out the window.You really have to try a mixture and then check to see if it is right. If fit is not, the question becomes, what can I add to it to make it right, does it need more x? Sometimes the root color is wrong and you need to start a new mixture in hopes that will be correct.

Plus this is just to copy a successful work- it’s many times worse when I am deciphering what is live in front of me. Any advice would be helpful, but if it is just that I am insane then I apologize for taking your time. In case of this, I did not put this as a blog comment.

I have an almost instinctive system of grabbing the colors to make a note before me. Some of this is personal preference, and some is experience. They have worked for me outside. But there are other ways as well, maybe better. It is awfully hard to make anything more than a guess without actually mixing up the note and seeing if it will work. I can hit any color I see except for some weird colors in man made materials like Sears ponchos.

Signed, My Kids Have To Dress Me Or Everything Clashes


Robert J. Simone said...

The difference between finding some set of color recipes and mixing from observation is that when mixing from observation there is more than one way to skin a cat. The options can cause confusion. "Which way do I go?" A general thought process seems more useful than a recipe book. I like to start a puddle and then ask questions rooted in the attributes of color and value: Does the puddle need to be darker or lighter? Warmer or cooler? More intense of more neutral? Making adjustments according to how I answer the questions gets me where I need to go. Also useful in the process is filling in the blank, "This puddle is too ____." Like learning anything it becomes second nature and we don't have to think as much as we can just do.

Works for me, anyway. And helps my students.

Lucy said...

This artist should not feel crazy. It's the process we all go through. Greens are so hard! I've thrown away a lot of green paintings! I think it's the thinking (and looking)that counts, and getting where you need to be with value, hue & temperature.
Also, shishkin may have used entirely different pigments. He may not have ued cadmiums at all. Maybe like Corot, he liked Prussian, yellow ochre, red ochre, Flake white, black, and burnt sienna.
Stapleton, your explanations really make sense. Thanks.

R Yvonne Colclasure said...

Sounds like "Clash" is over-thinking this thing. The best thing to do is just try it and see what you get. It is only paint and time and it can be painted again with different mixtures. I doubt the paint police are lurking at the door to pounce if it doesn't turn out just right the first time. But then what do I know? I am only a rank amateur.

willek said...

I think the two Shishkin Versions represent the clash that often occurs between what we see and what we know. In this case, he chose to get rid of the filtered green light that pervaded the whole scene. I'm sure it was there and probably was what attracted him to the scene to begin with, to go with those more conventional/acceptable, redder lights. You think?

billspaintingmn said...

This is getting exciting! The mystery is in the paint!

jade said...

I find the study more interesting than the final. it's looser, less detailed. the final is pretty of course but it has so much detail, I find it a bit boring. *ducks and runs*

JonInFrance said...

Yeah , I prefer, by far, the study. R. Schmid starts his last video with a mixture of yellow ochre pale + viridian - which gives that sort of pale green

Mary Bullock said...

He not only changed the colors in the final painting, he also changed the composition. The trees are much closer and more to the right.
Which composition is better Stape?

Stapleton Kearns said...

Yes that is how it works out in the real world. I often mix colors almost automatically.If someone asks me in a note I will sometimes not know. But I can make it again if need be.

Stapleton Kearns said...

I would suspect that Shiskin more likely had chromium colors and vermilion, both of which are out of use today.

Stapleton Kearns said...

I thought that too, but thinking is a good way to arrive at the ideas. I overthink EVERYTHING (I suppose, but maybe not).

Stapleton Kearns said...

Perhaps but there may just be some skewing of the color in one of the photos. Its hard to say.....

Stapleton Kearns said...

Remember, its not in the paint!

Stapleton Kearns said...

I think that is more a contemporary preference for looser work, that would not have occurred to a viewer in Shishkins time.

Stapleton Kearns said...

Join in France
I love Schmidt, but I haven't sen the video.

Stapleton Kearns said...

I don't know if the study was "cropped" .I guess I like the big, finished picture best. I am a big Shiskin fan and that's how I want them to look.So I come to this with expectations.