Whenever I try out some new technique I find a pattern in my habits confirmed – in writing and printing as well as in cooking or cycling, it seems to be a really universal principle. Recently it happened with oil painting. I tried out some baby steps around December or January, didn’t have a clue and screwed it all up. Threw all the material into a dark corner and forgot about it. This is the principle: I’d do some at best half-baked shit at first, then wait, typically for a few months, and then, in the second attempt, do something really cool. Apparently new ideas need to ferment first.
In May I took the oils up again. In the meantime I had acquired a few bits of knowledge waiting to be used practically – in particular about glazing. Instead of painting with opaque colours you take a monochrome underpainting and put transparent hues on top of it in order to separate shape and colour. This was very useful for me as a draughtsman with little to no ideas about colour. Of course almost everything I did in the process was more or less wrong at first, but I learnt a lot about how to make the next picture really fancy!
The ground for painting is a (roughly A4-sized) board of plywood rather than canvas. In almost every aspect wood is more problematic than canvas. It’s heavier, more difficult to transport, more cumbersome to prepare. It’s “working”, i. e., it’s expanding and shrinking, for a long time. Compared to canvas it’s absolutely inflexible and may even crack and break (for this reason museums are still very reluctant to lend panel paintings away, compared to the relative ease with which they send canvas abroad). Last but not least there’s the ridiculous problem of knotholes: If your panel has any, resin may seep into the colour later and produce dark spots years or decades after you considered your painting done.
Yet for largely glazing-related undertakings panel is recommended. Other than canvas it has a smooth surface. The thickness with which transparent paint is applied is directly tied to how dark it appears. Even light colours darken the picture overall, while thick layers of dark colours (e. g., picture ultramarine) have that wonderful and oxymoronic effect of seeming pitch black and maximally saturated at the same time.
In my case and similar renaissance-to-baroque paintings the layers tend to be extremely thin. The amount of paint used can be quite homeopathic. I’m literally talking about tiny beads of paint on the tip of a single brush hair! So the paint caught in the little space between canvas threads can make a fairly dramatic difference. Painters who used glazing extensively on canvas, e. g. Vermeer, would use very finely-woven linen, much finer than what you’d typically be sold by your local painters“ supplier.
First a very rough – a very very rough – charcoal drawing. Its purpose is to mark where to find which feature. This one here is actually far more detailed than necessary and at the same time far cruder than it should have been. Next time I’d rather do something with the look of a technical blueprint.
Four and a half hours of total working time. Keep in mind this is my first serious attempt at brushwork, today I’d be much faster not so much for technical ability but rather just because of having a clearer idea whereat to go.
The “monochrome” in this case means black-and-green. There are a few different classical underpainting techniques that mostly differ in the colour employed: Grisaille is grey, verdaccio green, bistre brown. The strength of verdaccio lies in that a greenish base takes out the saturation of the warm (brown, red, yellow) glazes on top that would normally yield too garish a skin tone.
Here I’m already working in oil: oil diluted with turpentine. When working in oil layers there is one rule that is to be followed all the time: Fat over lean. Pure linseed oil would be considered fat, paint coming straight from the tube is still relatively fat, and turpentine (which evaporates after a little while) counts as very lean. Oil paint dries not physically by evaporation but chemically by polymerising – extremely slowly – upon contact with oxygen. After some days to weeks it seems dry, but to fully react it takes many months and, like too fresh wood, works in the process. That’s not a problem unless slow-drying (fat, thick) layers are covered and sealed off with fast-drying ones. Paint may wrinkle, crumble, flake off. No, thanks.
Thus for the underpainting the leanest colours are most appropriate. In fact it’s possible to do without oil altogether: In ink or gouache, or in oil-containing but still extremely lean tempera. Blending is a bit more difficult (at least for me) in all of these, but still reasonably well possible.
One thing I regretted all the time in later stages of the picture is how I only took care of the skin in the underpainting. Definitely should I have included shirt and hair at this stage which in the end now look far flatter than they could have.
And generally the underpainting isn’t the strongest: The eyes are flat, proportions in the lower half of the face are fairly messed up and so is the hand – in short: everything really. For some reason this wasn’t so clear at that stage – self-betrayal, I suppose. Here’s one huge issue with this kind of layering: Corrections of previous stages are impossible (on the other hand, corrections of the currently wet uppermost layer are very easy because they don’t affect anything below them, unless being done very carelessly). Once the painting gets its first layer of colour the underpainting turns wholly immutable.
The first two attempts at a colour layer failed: I had used the wrong medium. Confused two different glass jars standing on my desk. They’re not even similar and it took me two days to notice why exactly the paint wouldn’t dry and stay on top.
This medium is a mix of linseed oil, turpentine and dammar varnish. The oil and turpentine adjust for fat-/leanness, the dammar – a resin also used as a final varnish – speeds up drying. Otherwise the same painting might take half a year. By this stage pictured on the left, the first coloured layer of burnt siena, three weeks have passed and nigh thirteen hours of painting, many of which were wasted on dead ends and correcting avoidable mistakes.
It’s possible to mix medium and paint right on the palette – there are some, though, who advocate covering the picture or selected parts of it in pure, colourless medium, then adding colour into that wet surface. The latter way has worked very well for me but it may not be optimal, I need to do more research on this. Just know that both are possible.
At one point, after four or five layers, I found the colours way too intense, even with the verdaccio below – so I figured I could add a complementary ultramarine layer. Blue + brown = grey, I thought. And it actually was a good idea that’d work that way – unless of course I’d take too much blue. This is what happened and gave me a severe case of cyanosis.
Glazing doesn’t just consist of colouring pre-set shapes like in a children’s book, although it comes close. Since, as mentioned, every layer darkens the picture a bit overall, it’s necessary to lighten up some areas manually. Opaque white is gently massaged into these places – and really the smallest amounts, lest it cover the layers below entirely. Compared to the glazes its impact feels much stronger. It’s possible to add a glaze and wonder whether anything has changed, while the tiniest drop of white left me afraid this was really way too much now – in fact, these seemingly tremendous amounts of white are necessary. As a child I used to wonder why white people were called “white” when in fact they’re rather pinkish. Painting now it feels as if (Centronorthern European) people are really scarcely any darker than sunlight. This summer I’ve already got substantially tanned, and yet still I am incredibly white. More than in the painting for sure: Let me blame the warm light bulb illuminating my desk, please.
Between the underpainting and the state to the left there have been about nine or ten layers. Finally I added some hair (and realised that the forehead ran out of skin, duh), and shirt. Apologies, by the way, for the terrible picture quality so far – it’s just that wet oil doesn’t go along so well with the scanner. This version had had enough time to dry, though, because it’s from when I returned from my three weeks in France. The next such painting will have a quality Praktica-based documentation, I promise.
Note the background. It’s glazed, too; dark green on a black ground. I had considered a red background in order to let the skin itself appear more muted – according to shady internet sources, Délacroix once said: “I can paint you the skin of Venus with mud, provided you let me surround it as I will”. But while it’s incredibly tempting to put some bright, undiluted tube paint somewhere, I came to find it too strong and went for a dark green instead of the kind you’d see in Cranach portraits, for example. Probably I could just have put the colour there as is, but I experimented with glazing a black background instead. Would it remain black, I wondered, or gain some saturation? The result was fantastic. In this state you can get an idea of the difference, because I had to correct something in the left half that led to losing transparency there. It required another transparent glaze, as did those areas of the shirt that lay in the shadows. Turned out that glazing uniformly over a pre-shaded underground looks much, much better than superimposing a shadow onto a coloured surface.
And yet – all done, now! By the end of July, with this one break of three weeks, I accumulated 21 hours and a half of drawing-and-painting time. The writing alone took one hour and a half, it’s awful to do with a brush when used to feathers – I was tempted to give it another layer in order to smoothen the differences in brightness between the different letters, but screw it, it looks more lively in this amateurish manner. Gone are the times, anyway, when painters who knew how to write could make some extra pennies.