How to Paint an Oak Tree

A man who spent three years painting the same English tree repeatedly—in all weather, day and night—explains how exactly, and why.

Six years ago, Stephen Taylor came across a 250-year-old tree while on a walk in Essex, England, shortly after the deaths of his mother and close friend. The following excerpt comes from Taylor's new book, Oak: One Tree, Three Years, Fifty Paintings (Princeton Architectural Press), which documents his painting experiences. It is a profound treatise on beauty, change, and the enormous challenge of seeing the real world clearly.

Starting Points

A pochade (small sketch) box, developed in the nineteenth century to help painters work outdoors. The lid has slots for panels so that wet paintings can be transported without damage. There is a strap to permit the artist to paint standing up. Sitting down, the box rests on the lap. With a seat and an umbrella, it makes a small portable studio.

The natural world is visually very rich and constantly changing. From light reflecting on a mountain stream to the pattern of fallen leaves in a forest, nature is visually unpredictable and often very hard to see clearly. So I use painting to help me discover what is there. My methods are designed to clarify the complex colors and shapes I see. The approach is empirical, and I use science to help me observe. At the same time, a painting is more than a neutral record. My choice of colors is selective, and the picture expresses the associative and emotional concerns of the observer-painter. There are two sides to the way I work: one that takes place outdoors and another indoors. I trace the steps I took in creating Swallows at 9pm as a way to show my typical working process for a large painting. Seeing a picture develop from start to finish should help to get closer to the painting and to see how it can be both observational and expressive at the same time.


Outside, everything I do is to help my senses and my reactions adjust to the natural world. I want my work to come out of an intimate connection with a place. By the time I came to paint Swallows at 9pm, I had been working in the same field for six years, so I was familiar with its character. Each picture is unique and requires its own set of adjustments: for an evening picture like this, I arrive early to let my eyes adapt.

I strapped a pochade box around my neck so I could work standing up in the crop, which was chest high at this time of the year, and I mounted a digital camera on a tripod. I had a canvas seat to rest on, as well as hot coffee, snacks, fruit, and a tot of whisky. I take an extra set of old clothes—even in summer I can get cold if I’m standing still. I probably look like a tramp. I remember feeling utterly happy at this point, with my brushes, palettes, and rags, just looking...

I sometimes use photographs—taken years, days, or just minutes before—as inspiration. You can shoot, evaluate, adjust, and reshoot digital shots on the spot. I do not always start color studies with a photographic prompt, but I will take selective exposure shots as soon as I see the effect I’m after.

I had already taken photographs of the field of rape around six o’clock during a previous summer evening. (I had taken them because I felt that people don’t really look at rape as it responds to sunlight.) I wanted to use the bright yellow of the industrial crop to create an extreme contrast with the delicate tones of the trees, so I would paint in the early evening, when sunlight has a reddish color that makes the field look luminous.

I needed a visual source for swallows, so I went to a field of barley where I knew they hunted. I took some photos during the day that seemed to match what I’d seen at dusk in the rape field. I had to use bright light to get the speeds needed to capture such fast-moving objects.

I was set up under a clear sky, ready to go. I first put an approximate base key (a set of background colors that helps judge subsequent local colors more accurately than is possible on a white background) over the entire board using quick-drying acrylic. At six o’clock the crop started to catch the warm light of the setting sun, and I started to paint quickly in oil.

I make color mixes in clear, definite values, each one based on what I see and carefully compared with the other colors around it. This exacting comparison and mixing is fundamental to observational painting and is something students need to practice. Each color we perceive is effected by the colors surrounding it, and by analysing individual colors and their relationships to those around them, you can discover the unique character and lighting of a scene. Each study is a form of empirical discovery—and also a race, because the faster the light changes, the faster you have to work.

When I lay down the main colours, I start with big areas first. But in this case, they did not seem to work together. The light was fading very slowly, so I waited for something different to happen. I sat down and prepared another board in a darker key for the lower light that would come.

As the sun set behind me, the shadow of the earth rose on the horizon, reversing the usual gradient so that the sky above became brighter than the lower sky, which passed into a grey limbo. The crop was now in shadow but still luminous; detail in the trees on the horizon began to fall away. I noticed new, unimaginable red-blue greens within the rape.

I worked fast and in under an hour had laid down all I needed. At the same time, I took a series of selectively exposed photos. It was about nine o’clock when a small group of swallows flew straight at and passed me like bullets. The lead bird looked violet against the rape. I rapidly painted some colours to record this, then noticed the crimson light of a radio mast in the distance and included that, as well as the cold yellow of the headlight of a passing car.

I packed up and went home. 

Oil-on-board study with acrylic underpainting. The clearly defined color patches provide clear values, like individual notes on a piano, for the final painting.

Computer Work

The next day my study looked both unfamiliar and somehow complete. These were both good signs. I transferred the photos into Adobe Photoshop to analyse as layers to help me discover the distribution of colors, a method that I first used to study colors in a field of corn. Traditional oil painting also has layers: painters, like Titian, built images up from separate layers of color, wet onto dry. This is the way I work. I use the colours from the study made outdoors to determine colors for the wet-on-dry layers that I build up on my painting in the studio.

Sable points, synthetic brushes, hogs hair with filbert tips, round heads, square heads, fans, shaving brushes, housepainter’s brushes, soft Japanese water color brushes, brushes in useful states of destruction, sable riggers, some of my dad’s brushes, school painting brushes, handmade badger blenders from Germany, sign-writers’ brushes... I choose the brush that matches what I see. Because you can’t predict what you might see, you need to have a lot of options. The matching of tools to perceptions in this way is not new: two hundred years ago riggers (very long, fine sable brushes) were used to paint the rigging of ships. I sometimes use them to paint electric pylon cables.

The photos I took at the rape field were variously exposed for different parts of the scene. To find the distribution of a color in the painted study, I first chose the photo that was best exposed to show that colour. I then used Photoshop’s selection tool to pick out that color in the photograph. Photoshop can then indicate every point in the photo where that color appears. This breakdown reveals details not obvious to the naked eye and can clarify the scene in an unexpected way. I could now use this information to guide the distribution of paint of that color onto the canvas. Only three photos were needed for this painting. After some trial and error, I could use this method to see the color distributions in a way that would help me to read the scene more effectively.

Photograph showing the best exposure of the tonal range of sky that will match the range of the oil study.

Photograph exhibiting the best exposure of the distribution of yellows and greens.

Photograph showing an excellent range of tonal values in trees on the horizon.

The layers also revealed the color distribution as a texture gradient, a regular increase in the number of texture elements proportional to the distance from the viewer. In experimental psychology, a texture gradient is a recognised depth cue. Since each texture is so closely associated with a color, it seems more appropriate here to refer to them as color-textures.

Some experimental psychologists are currently studying these types of association as an important new channel in visual processing. For me, they are a fundamental means with which human vision can disentangle or parse nature.

This Photoshop selection layer picks out a range of yellows to show their distribution as a color texture. (Some yellows have been switched red to make them easier to see.) The selection also reveals a texture gradient, a strong depth cue. A printout is used in the studio. colors selected from the print are matched to corresponding colors of the oil study and are used to guide the distribution of the colors in the studio painting. The canvas has a grid to help transfer the distribution from print to painting.

This Photoshop selection shows the distribution of the darkest yellows and greens. The actual color mixes used in the painting will be the darkest yellows and greens from the oil study, but their distribution will be guided by a print of this image.

Photoshop selection layer showing the distribution of the darkest greens and blacks.

Studio Painting

I first had in mind a square painting, a night scene with the oak placed in the center (to pair with an identically composed picture, Swallows at 11am, set in the morning). This canvas had already been prepared with a dark acrylic underpainting. I decided to change the time of day of the scene but used this underpainting because the emotion I now wanted to express had a sense of floating on blackness about it.

A dark acrylic underpainting, originally for a night painting.

After the painting was finished and I had compared it to its finished companion, Swallows at 11am, I decided to call the new painting Swallows at 9pm. The reference to the birds and to time suggests temporal scales at play. At least a year has passed between the two images, as evidenced by the crop rotation, rape following barley. The birds would have migrated to Africa and back. Of course the same 250-year-old oak remains; it offers a very different way of measuring time. There are only a few days in the year when the crop looks like this, only a few minutes in the day when light is like this, only a few seconds to see the birds.

Early transparent oil layers still affected by the dark underpainting. As the paint layers accumulate, light onto dark, the lighter colors will become increasingly opaque, producing an effect of paler solid objects above transparent, darker ones below. The piece of cotton running across the canvas is a part of a grid used to help transfer the color distributions from the print out to the canvas.

In the ancient Roman world, bird watching was connected to augury. In the end there were four swallows, like little horsemen, and a tiny red light. I think this makes the beauty of the scene both uneasy and somehow more truthful.

Swallows at 9pm, detail. There are differences in finish across a painting that brings out the character of the elements in it. These differences also reveal the artifice involved in the act of representation. The patterning of distinct layers is visible, and some white grid dots remain.

Swallows at 9pm, detail.

Swallows at 9pm.