I did another guest-blogger thing recently, this time on my friends Kim Van Der Hoek and Kelley Sanford's In the Artist Studio, a very informative art blog where they have this feature called Ask the Expert. You can find some excellent tips on this blog so I encourage you to check it out.
The following is my contribution to In the Artist Studio, reposted here on Studio Notes. Hope you find it interesting!
I’d like to do a little analysis on some conscious decisions that I made along the way on a recent landscape painting that I did. The ideas I’m sharing may not be new to you, but hopefully they will serve as a reminder to be mindful of the “basics” no matter whether you’re a beginner or a much more experienced painter.
Back on the Road, 18 x 36 inches, oil on linen
The painting, Back on the Road, is a studio piece, and it is entirely invented. I have done similar scenes on location so I had a very clear idea of the kind of mood I wanted and what I needed to do to achieve it.
The design is not particularly innovative or unique. I have the horizon splitting the canvas near the middle, a road taking the viewer into the picture, the big tree just off center. These are all common design elements and would be pretty boring if we didn’t do something to alleviate the predictability.
You’ve all heard this good advice; “Don’t put the horizon right in the middle.” Why not? because it’s boring to have the canvas split in two equal halves. I agree. so why not just make them unequal by giving a lot more visual weight to one half. Sure the horizon may go right across the middle, but all the dark masses, textures, colors, brush activity, perspective, manmade objects are on one half, and the sky as the other half serves as a big passive area, with very more subdued range of values, colors, etc. That makes the two halves not equal. Far from it. If you make sure that the visual impact of the two halves are unequal, having the horizon right across the middle is a non-issue, in my book.
The road leading the viewer into the picture is a common device, too. In order to make it more interesting, I made road curve and also go up and down small hills. Every time the road turns or the incline changes, I had to plot a new vanishing point. So the road alone has at least five–may be six–vanishing points. Tedious? Yes. Basic? Yes. Worth the effort? Yes!
The big tree is near the center, but just a little off to the right. That is a very basic design decision. However, I don’t think there’s anything inherently wrong with having your focal point in the center, as long as it looks absolutely intentional and not because you forgot to think about it.
Again, how you place the rest of the visual elements to create an interesting arrangement becomes critical.
But don’t forget; just because the focal point can be in the center doesn’t necessarily mean it should be. If moving the focal point off-center allows you to improve the design, and allows you to make your statement more clearly or efficiently, by all means, don’t leave it in the middle!
The lightest part of the sky is behind the big tree to create a greater contrast.
OK, so the big tree is my focal element. I spent a lot of time shaping that thing so that it had a good, strong silhouette. But a strong shape doesn’t by itself make it a focal point. It needs contrast! I always think in terms of value contrasts first, so I made sure I had unmistakable value contrast at my focal point. You can see the sky behind the dark foliage is lighter than the areas flanking it. In this way, I consciously used maximum value contrast, strong silhouette, (relatively) sharp edges, and selective color (more intense yellows and oranges in this area, only), and detail (the silhouette of the foliage has much smaller shapes and calligraphic interest than any other tree) all to support my main focal point.
There are several distinct, albeiet close, value steps in the clouds.
If you take a closer look at the sky, you can see the variations in color and value. From the lightest apricot color behind the big tree, to the shadow colors (5) and (6), there are several steps from lightest to darkest.
Also consciously modulated are the intensity of these variations. I put the highest chroma at (1), and made sure the others didn’t outshine that color.
I pushed the color toward red at (3) and (4). To do this, I added Cadmium Red and Alizarin to the mix. Which makes the color redder, but since these redsvare so intense, it was necessary to also knock the saturation down a notch by adding a tiny bit of blue as well.
Using softer edges for form shadows, and sharper edges for cast shadows and lit edges.
Now let’s look at forms in the sky. Because these cloud masses are dense, the light hitting them reveals forms, much like solid objects. By imagining where the clumps are forming, we can decide where light and shadow patterns go. Since the light is coming from the right (see the cast shadows of the telephone poles?) the right side of any “clump” would be lit, and the left side and the bottom of the same cloud mass would be in shadow.
I tried to break up the forms a little bit, to make it more fluffy and organic, but the form principle is intact. As for edges, where you would expect to see a form shadow edge–that is, where the form turns away from the light into the shadow, I used softer edges because there is a transition from light to shadow. This includes forms turning under.
See, it’s not so different from painting simple spheres.
Where the lit edge shows up against a darker cloud, the edges are sharper. However, because we are talking about cloud masses, even the sharper edges aren’t razor sharp; they look sharper just in comparison to the softer turning edges. It’s all relative.
Notice the sharpest edge in the clouds are used near the focal area. Another device to bring the eye there.
So as you can see, there is a pretty good range of colors and values in the sky, and each of these variations are used purposefully, whether to highlight the focal area, or to show that the form has a shadow side, or to provide a transition between light and shadow.
Violets in the clouds taken out of context.
It does looks like many hues are represented; yellow, yellow orange, orange… all the way to a blueish gray in some of the shadow areas of the clouds. But as you know, colors are relative. What appears violet in this context may surprise you when taken out of context.
The two shadow colors may look violet in the picture, but if I had just mixed red and blue on the palette and stuck it on, it would be completely out of harmony, sticking out like a tuba in a string quartet.
So how do you get these subtle colors? Just as I did with the reddish variations, I mixed a violet (probably made from ultramarine, ivory black, cad red, alizarin and white) into the apricot pile, a tiny amount at a time. I kept checking the value and chroma and fine tuned it until I had what I wanted. Mixing violet into a puddle of peachy color of course gave me a muddy gray, but that’s just what I needed.
The distant hills are painted in the color of the atmosphere. (They don’t have violet gray grass growing on them)
The distant hills are more or less a darker version of the violet, with a slight increase in the chroma. I’m playing up the atmosphere by completely ignoring the local color here. The rule is, the more atmospheric effect you have, the less relevant the local color becomes.
Things like hills way in the distance become just variations of the atmosphere because essentially, what we are seeing is the veil created by the particulate matter in the air between us and the hills, lit by the sun and/or the ambient light of the sky. We’re essentially painting the color of the veil, not the hills behind it.
So the atmospheric perspective used effectively will create the illusion of depth. This is very useful in a pure landscape painting where there are no man-made objects to give us linear perspective. But often if you look, you can find elements in a landscape which you can exploit to bring in some linear perspective.
In my painting, I have the road, which is an obvious thing since it’s man-made and we understand it as parallel lines going towards a vanishing point. I also used things like edges of fields, how telephone poles and trees diminish in size systematically. As this was an invented landscape, things like the edges of fields are made up elements specifically to show linear perspective. The view makes perfect sense without them. But including them help to create a sense of vast distances.
Some obvious and not-so-obvious devices to show linear perspective in a landscape.
Perhaps the least obvious, but just as important, are somewhat random-looking strokes on the ground plane that conform to the linear perspective by pointing to a vanishing point. Especially when you have a foreground that doesn’t have much in it, it can be difficult to make it look like it’s level ground. (Or inclined, if that’s what you’re trying to depict) In some cases, the up and down strokes used to describe grass in the foreground end up making the entire foreground vertical, like a face of a cliff. Strokes that suggest a vanishing point will not only help the ground lie flat, but it will contribute to the sense of depth.
Losing edges between tree masses.
We’ve been talking about some of the things that go into creating a more complex, believable visual environment. But not all the tools are about adding complexity. Some are about editing out the unnecessary elements. Simplifying the design strengthens the impact, and one of the most useful tools to move in that direction, is to combine shapes by losing edges between them.
In my painting, many of the dark tree masses connect, creating fewer shapes rather than a whole bunch of little tree shapes. Our eye doesn’t have a problem perceiving the trees as we intend them to be, even if the combined mass look more like blobs and strips. The context informs the viewer what these abstract shapes are, so we don’t need to give them unnecessary information by painting each individual tree, branch and leaf.
The shapes being connected don’t necessarily need to be same types of objects, like tree and another tree; you can connect tree and grass, grass and barn, shadow on the side of the barn to the shadow cast on the ground… any two shapes with similar value can be connected.
Sometimes, the shapes need to be separated, even if they’re similar in value. It all depends on whether losing the edge there strengthens or confuses the image. You want your statement to be clear, but if connecting certain shapes creates a silhouette of a poodle (or whatever) that completely misleads the viewer, then may be you want to avoid that.
OK, that’s about all I wanted to say about this painting. If you are a beginner and found this information overwhelming, let me tell you that I’ve been there, I know. And so has every great painter. They’ve all had to learn the basics, one canvas at a time. It just takes lots of practice. But remember, mindlessly going through the motions doesn’t count towards your canvas mileage. You have to be aware of what it is you’re trying to achieve in each painting. If you are, you’re more likely to spot mistakes, or elements that don’t contribute to your aim. And if you can spot them, you can fix them.