The best way to know if something belongs in the painting is to leave it out.
Ok, I confess I don't always make a preliminary study before I "try out" the image. But I almost always wish I had. If you're impatient, like me, please read this post anyway. I'll have more to say about how to follow up on what happens when you just dive right in.
When you are drawn to a particular subject, it may seem unnecessary to make a thumbnail sketch or a value study. If a juxtaposition of interesting shapes presents itself, you may feel as if you need to get totally involved right away, before the "flow" gets interrupted.
I get that, believe me. There is an ideal process in my imagination that goes like this: Inspiration leads to action. Action leads to expression. Expression = Success!
Realistically, though, the experience is usually more like: "Wait. let me try that again", which leads to the same mistakes, which leads to another attempt. And so on. So much for the flow.
I want to make a case for giving a thoughtful approach a try. The sacred "flow" will not be halted because you allowed your intelligence to play a part in the process. Think and experiment all you want. As soon as you make the first brushstroke, the current will begin to carry you again.
Before you launch into what you hope will be a proper painting of a new subject, take a minute to assess how ready you really are. Go down the list: Value, Composition, Edge quality, Color. Do you feel confident in each category? If so, off you go. But if not, see if you can give a name to your uncertainty.
When I consider this photo of Hart Pass, below, I know right away that I have a good sense of what I want to do with color. I also know that I like the bold contrast of light and dark. I may want to turn down the value of that big triangular shadow in the center of the page. It could look like a big hole in the scene if I don't lighten it a bit. I believe I can see that well enough in my mind's eye that I don't need to make a study to test the idea. But what about edge quality? I'm not sure the illusion of space would survive having so many hard edges, but which ones should I soften? Aha! The unanswered question.
Now I know what sort of study I need to make. I'll make a very quick sketch that has no hard edges at all. Then I can use that to tell me where I wish I had one or two. Maybe three. I can pause and ask where a hard edge would be essential and make a note of that information, adding locations one at a time, stopping when there are just enough.
If checking in at the gut level doesn't reveal how confident I am, there are a few simple questions, based on where my paintings most often go wrong that I can ask for each of the four main variables:
Color: How many colors will be enough?
Value: What should the value range be in the foreground? Middle ground? Background?
Is the pattern of values appealing?
Edge Quality: Where do I really need hard edges?
Composition: Are the major shapes where I want them?
Are there too many shapes?
Remember, you are asking these questions to inform an interpretation of the subject, not a duplication of the photo. The two approaches will give you very different answers.
Often just focusing your awareness by asking questions reveals the answers, and no study is required. Sometimes assumptions are revealed that are self-imposed obstacles to expressiveness. Stay receptive to ideas when you are in the planning stages of a painting. It is a creative process that helps you get out of your own way.
Whatever form of study I decide to make, I like to start by seeing shapes. I often look for the shapes that need to be separated from each other to understand where they are in space. How will you use color to separate the shapes? How about value? Edges? Composition?
In the photo below, there is just enough overlap to tell us where the buildings are in relation to one another. This would be the right time to make sure there are no unfortunate convergences of profile that cause the shapes to appear to be in the same plane. It's easier to move a shape now, when it's not yet made of paint.
Sometimes. though, you may not care about creating a convincing illusion of space. Notice how the buildings make one big shape when you compare them with the sky. Their geometric patterning contrasts dramatically with the simplicity of the plain blue wash. I can see how deliberately flattening the space could express something of the experience of being surrounded by big buildings. A couple of pencil studies, just outlines of the shapes in slightly different positions, should reveal how such a treatment might be done.
It helps to be clear about what is important to you in the subject you intend to paint. For the downtown scene, for example, the assumption that a feeling of space was essential turned out not to be true. That changes how the variables need to be employed. If you no longer need to separate the individual buildings you probably want to emphasize the similarities rather than the differences.
For homework, please find an image or a live scene you'd like to paint. Assess your readiness and devise a study that will provide answers to one or two unanswered questions. If you keep it down to just a couple of questions the odds are better that you'll see the answers.
Make your study quickly. It's not a painting, and it will do its job just as well if it looks wrong. We're just after information, not fame.
When you're ready, paint the picture. Keep the study where you can see it while you paint, and please bring everything; studies, notes and paintings, to our critique.