Geometry and composition

Last post I said I’d write next about why I use geometry in my paintings. But before I do that, I thought it might be useful to give an example of how I use it to compose and structure my work.

So I’ll describe how I composed the painting Hotel. As it’s fairly recent, that should make it easier to remember exactly what I did, which could be more problematic if I tried to describe composing an older work. Although I approach all of my painting with a fairly consistent method, I don’t use a formula. Each painting is its own thing from the outset, and evolves according to its own internal logic. This can make it hard to remember, or reconstruct, exactly how the result was arrived at. Heck, sometimes I don’t even use geometry at all.


Hotel started out, like most of my work, as a small thumbnail sketch, the sort of thing you doodle in the margins while you’re thinking about something else. Unusually, however, I didn’t really develop the sketch. I didn’t try out various alternative shapes, work out patterns of darks and lights, or any of the other things I normally do before I commit the time to turn the sketch into a painting. This time I was fairly confident I had enough to start with, even though it was pretty rudimentary.

One of the first things I noticed about the sketch was that the most important horizontal division (indicated by a red line overlaid on the drawing) cut the frame at approximately the golden section, or what you’d get if you multiplied the vertical dimension by 0.618. The ratio can be explained by a simple diagram, where the length of the whole line A:B divided by the length of the largest part A:C is equal to the length of the largest part A:C divided by the length of the smallest part C:B …


This gave me a good starting point for structuring the rest of it. I don’t worship the golden section, but I do find it useful for reasons I’ll explain in my next post.

I also noticed a problematic vertical right in the center of the design that I would need to deal with. Cutting the rectangle into two equal halves either vertically or horizontally is generally considered bad form.

I went straight from this tiny drawing to roughing out the full size work. Here’s the underpainting with the compositional geometry superimposed:

image with geometry superimposed

Here the main divisions are indicated in red. These are the vertical and horizontal lines that result from applying the golden ratio to the whole frame. The blue lines indicate secondary divisions, which result from applying the golden ratio to the main divisions. Finally, the green lines are diagonals derived from those divisions.

At this point, the composition was mostly set. In terms of the structural lines, the biggest change I made was to shift the problematic central vertical a little to the left, congruent with the centremost blue secondary division. You can also see I’m starting to work out a pattern of darks and lights, although I’m not done yet …

Here’s the finished painting again:


So what’s the point of all this fiddling about with geometry and structure? And if I’m so into that, what’s with the parts of the painting that don’t conform to any ratio, and why do some of the lines kind of follow the structure and kind of not? Why don’t I just paint the diagram, like any sensible reductionist would do?

All will be explained in my next post.

Comments are closed.