Sight Size Self Portrait Drawing – November 2009

In the last series of posts I covered a sight size copy of a portrait drawing ofLady Sarah Spencer by Sargent.

I was particularly interested in trying to move beyond a mark-for-mark copy and to develop myability to conceptualise form. I wanted to see what I could learn from Sargent’s approach to simplifying, strengthening andconceptualising form, and then apply that to a drawing from life.

In the first post I started by establishing the main anchor points. I then tried to build upa conception of the form in line, and finally added light and shadow to finish the drawing.

Throughout that drawing, I was developing a process that I thought I would be able to apply equally well to drawing from life. Thisself portrait sketch is the result of applying the process to a live model.

In the comments on the last post, David quite rightly pointed out that getting models to sit for you can be a bit of problemwhen you’re learning. I’d love to have done this drawing from a model since it would have been easier to use the sight sizetechnique. But since I too didn’t have access to any models, I used my own lumpy old head as the model. I’ve heard it saidthat it’s impossible to do a sight size self portrait. Well, I’m here to tell you that that’s nonsense, you absolutely can. Itjust takes a bit of thinking about and setting up.

I first had the idea of a sight size self portrait when readingArt and Illusionby E.H. Gombrich (Amazon UK link.)This book delves into the psychology of perception as it relates to art and is a fascinating read in it’s ownright. I’ve read it twice and plan to read it again. Somewhere in the book Gombrich mentions in passing that your reflection in amirror is exactly half the size of your actual head, no matter how far away you are from the mirror.

I should explain this a bit further. Gombrich’s own example is the best one to use since it’s so practical and you can test it for yourself.The next time you’re in the bathroom and the mirror is steamed up, trace around the outline of the reflection of your head in the mirror.Now move back. You’ll see that your reflection as it is projected onto the surface of the mirror always stays the same size, exactlyhalf the actual size of your head. That means that if you have a drawing board side by side with a mirror, and stand the usual eight to tenfeet back for sight size, you can copy your reflection much the same as you would a Bargue drawing, or any sight size subject, andthe resulting drawing will be exactly half the size of your head. Cool.

I had to try it. Hopefully this photo shows clearly enough how I had this portrait set up, if you ignore the presence of thecamera. I’m standing about eight feet back from the mirror, roughly lining myself up between the mirror and the drawing boardI have stuck to the wall next to it. The lighting was slightly complicated to set up and needed two independent lights on stands, oneto light the drawing and one to light me. Unfortunately I haven’t got a good photo of the lighting set up but I’m going to do anotherof these sight size self portraits soon so I’ll describe that in more depth then.

To establish the centre measuring line, I’ve stuck a length of black cotton vertically to the surface of the mirror andreplicated it with a central line drawn down the paper. This gives me a set up not unlike theone I use for copying Bargue plates. I can now proceed to measure out the drawing in much the same way as I’veoutlined in the above description of my copy of Bargue plate 5.

I should point out here though there is an added complication. When you have the cotton thread in front of your eyes the reflectionof your hands in the mirror necessarily obscures some of the subject – your reflection. That is a bit awkward. Because of that, Irestricted the amount of measuring I did to establishing the main points of reference, and then proceeded largely by eye from there.

Here’s a close up of the drawing at the first stage. At this point, I’m at about stage three of theprocess I outlined in the series of Sargent copy posts.

I’ve established the overall size and shape of the head by sight size measuring with a thread, and I’ve developed up a threedimensional wire frame of what I think are the main planes and forms.

Although I’ve been pretty careful, I did find that I had to do a fair amount of adjustment as I went on with this drawing, much lessthan I would ordinarily expect to do when copying a flat drawing. I assume that can be attributed to both working from a threedimensional, moving subject (me) and also being able to do less exact measuring due to the aforementioned reflected hands.

Nonetheless, the main advantages of the sight size method are preserved with this set up: direct, like-for-like comparison of the drawingwith the subject, and distance from the subject allowing a concentration on overall effect rather than detail.

In fact, this second advantage is effectively doubled in a sight size self portrait like this since although I’m copying the projectionof my reflection on the mirror’s surface (which makes the surface of the mirror the picture plane) in perceptual terms I’m double thedistance away from myself. Actually, it’s just struck me asI write this that that means I can work closer to the easel than I did here and still be a good distance away from my reflected self. I’llhave to try that.

At this stage I’ve moved on to the beginnings of the stage of the process I outlined in the fourth post inthe Sargent portrait copy series.

All I’m really doing here is establishing the main shadow shape and covering it with a reasonably even tone. I’m not attempting to doany modelling within the shadow, and in fact I did very little of that in the finished sketch. I’ve kept any detail of the forms to thehalf tones and the lights in the final drawing.

I was looking through the Bargue book the other day andnoticed that whilst Bargue does put some detail into the shadows, it’s very minimal. All of the fine modelling goes into the half tonesin the Bargue plates, there isn’t even any in the highlights. But that doesn’t seem to hinder the palpable feeling form that all theBargue plates have.

It may make sense that most of the detail is in the half tones. Generally speaking, the half tone will be at the most oblique angle tothe light, so what you have in half tones is basically raking light. That would tend to accentuate shadows and show up even gentleundulations of the form more clearly, perhaps.

Certainly it’s pretty clear in this detail of Bargue plate 56 that there is very little detail in either the lights or thedarks. All the subtle drawing is in the half tones. The darks here are almost completely flat. In the original plate, there’s a bit ofreflected light on his bum-cheeks but apart from that the main shadow is made up of a single, flat tone.

It’s only just struck me that although the paper used for the original Bargue drawings is toned, Bargue doesn’t introduceany detail into the light areas in the form of highlights. In his copies of other masters’ drawings in part two of the book there areno highlights either, just a general light area with no detail. He seems to use the natural tone of the paper for this area, andall values relate to that. Very interesting.

But look at how much subtlety of value handling has gone into the shoulder on the left here, and the lower back. Contrast that with the almostcomplete lack of modelling in the main shadow area and the main light area. Looking at the difference between how the two shoulders havebeen handled is particularly informative. It’s like a window into Bargue’s thinking about handling form. It’s the half tones in thisBargue plate that create the feeling of three dimensional form more than anything else I think. Hmm, have to bear that in mind inthe future.

Anyway, back to my considerably more amateur effort.

I do appear to have learned at least one thing from my previous attempt at the Sargent copy: at least I’ve gone in with my darkestdark in the background here. I had a black cloth hung up behind me when I was drawing this, so the shadow side of my head appearedslightly lighter against it. I’ve preserved that value relationship in the drawing, although I’m not so sure it was a good idea todo so.

I’m wondering now how much nicer this drawing might have come out if I’d allowed the left side of the head to disappear completelyinto shadow and let the viewer create the rest in their minds. There’s something special going on in the mind when that happens,Gombrich calls it ‘the beholder’s share.’

Some of my favourite painters leave a lot up to the beholder to finish in their own minds.Velazquez springs to mind immediately, and by extension Sargent. Zorn, and later Rembrandt and Titian too, and Veronese. Why is thatfeature so beguiling? Well, our perceptual systems work byactively creating a version of reality for us based on limited information picked up by our senses. In this way, our version of realityis learned by experience, and much what we think we perceive directly is in fact is our own inference. That’s how many optical illusions work.Allowing the viewer more leeway for interpretation may in fact create a more convincing image than giving too many cues and too muchinformation, since with fewer cues they’ll have the opportunity to fill in the gaps in a more meaningful way for themselves. Unfortunately adiscussion of that is beyond the scope of this post right now, but it’s something I think about a lot and will doubtless return to at some pointin the future.

In the stage above, I’ve also deepened the value of the shadow side and started to introduce some of the half tones between the lightand dark areas. Throughout this process, I’m trying to relate everything back to the forms. Some of the half tones I’m introducing hereare more local tone and not created by the form. I think it’s important to keep a conceptual distinction between those in your mind asyou work. Knowing what creates a given value, whether it’s from light, cast shadow, edge shadow or local value can help you to decide howto handle it.

At this stage I’ve really just worked on what I already had. The background has been filled in at the darkest value my charcoal cango to. I think that’s making the face read a little light on the shadow side though, everything else needs bringing down a notch or two tocompensate.

There’s also a lot more work gone into the half tones. The larger the form, the less sharp the turn away from the light, producing largerareas of half tones. My forehead has quite a wide area of half tone because it’s one of the bigger forms. The nose has much less halftone because it’s a much sharper curve and turns away from the light more quickly.

I’ve tried to differentiate the cast shadow of the nose across my left cheek by keeping the edges fairly sharp. It’s not an edge shadowso there are no half tones to speak of. Looking at the drawing at this stage though, I do wish I’d worked into the half tones on my leftcheek where the edge shadow meets the light though. That transition looks a little too sharp to me and makes the cheek appear to turn awayvery abruptly.

Here’s the final version again.

There’s a lot wrong with this drawing, a lot I could have done better. But I do think it’s proved two things: that you can do selfportraits sight size and that the approach I used for the Sargent copy can be used equally well on a live model.

This drawing was actuallydone at the back end of last year. I’m thinking about doing another one of these very soon though, perhaps trying to introduce something ofBargue’s approach, leaving the extremes of dark and light largely featureless and concentrating on half tones. I think there might besomething to be learned from that.