A few weeks ago, I went to the Cleveland Museum of Art with my sketchbook, and some pencils. I sat down on a bench looking at a bust, and spent five hours trying to draw it. This was an incredibly valuable exercise – it’s something I definitely want to try again. It highlighted two really important things to me:

1. I need to spend more time measuring things. I spent a little bit of time measuring, but I was self conscious holding up my pencil like that while people were walking around, so I didn’t take as many measurements as I should have. The end result was pretty dramatically out of proportion.
2. I need to simplify more efficiently. After measuring, I immediately started drawing the shadows, as if it were a Bargue plate study. But I should have tried to understand the planes and structure of the sculpture better before I started drawing shadows.

The next day, I and played with the contrast on the photo I took until I got this:

Then, I traced the shadows in Illustrator to study the shapes of the shadows a bit more.

I was really surprised to see that even ignoring EVERYTHING else, if I render the deepest shadows, and nothing else, the result is still a very recognizable representation of the bust.

This sort of epiphany told me a lot about how to get an accurate representation of a thing. To make it look like the thing – the proportions are really all that matter. Here, there isn’t a “value structure” so to speak, there are just shapes. And the proportions of those shapes, and the proportion of the negative space between those shapes is what makes this recognizably the bust in the photograph. All the little tricks and techniques that go into making the painting “come to life” will never fix a bad drawing.

With that in mind, I decide to trace the photograph onto watercolor paper. I did this by placing the photo under the paper, and taping them onto a window. The light coming in through the window was enough to see through the paper so that I could trace the photograph.

Then, I painted the drawing. I started with just dirty water. Instead of getting paint from the wells in the palette, I just used water to clean the big center areas. I then dipped a brush into that dirty water, and used that to paint the first shapes. Then, I used a little less water on the brush and dipped into those same dirty puddles and painted the next darkest shapes. I kept doing this until I got to the point where I needed to paint the deepest darkest shadows. At that point, I pulled a bit of phthalo blue and raw umber onto the palette, and painted the darkest shadows with that.

This taught me that I usually go too dark, too fast. By relying on the dirty water on the palette first, I was “sketching” the shapes with paint first. Because I went that route, those early values aren’t bossy, and the darkest darks really do punch through. The way I usually paint, I use so much pigment early on that I never am able to get really dark values.

This was the epiphany for me here. Start with less pigment. Build up your values slowly.. don’t rush to render those big chunky shadows that lock the drawing into place and really refine everything. It can be tempting to render those deep dark shadows right away, because they do such a good job of selling the rendering. It’s not until those go in that the painting really looks like something. But, if I put more trust in the process, and I’m more patient with my values, I will end up with something I’m more proud of.

So, Stan Miller was right. Get the drawing right, and get the values right. Everything else is candy.

When I finished the face, I painted the background. I just used the leftover paint on the palette to render the background, and covered it in salt and mists of water to get a lot of granulation and texture. I really like how this works, and it’s really easy. Unfortunately, I accidentally painted into the cheek a bit. I tried to blot it out with a paper towel but by the time I noticed it was too late – so there’s a big ugly stain there. Oh well, lesson learned.