Learning to See


Learning to See

This post has been moved to Brendan Learns to See.

On Thursday, the first of November 2012, I learnt how to see.  I have done art courses on and off for many years now. I’ve avoided drawing “hard stuff” like heads and hands, although I’m pretty comfortable with feet.  Having a headless drawing is a bit… lost.  So, I resolved to learn how to draw heads and hands.  The weekend before last, I thought about learning other things.  It dawned on me that learning other things had involved a lot of practice, but that I had never really practised drawing. So I figured I would make the effort practising,  I have drawn from photos on the computer screen.  Last Thursday, something switched in my brain.

“Before” sketches (apologies for the white balance – took a long while to get it right):

29 October 2012
  30 October, model 1
     30 October Model 2
30 October Model 3
   31 October

I think you can see from these that I was clearly struggling with how to represent the face, the elements within it, and its relationship to the rest of the skull/head.  I probably made a little bit of progress on 31 October.   The drawings also show some use of grids for placement of elements, especially in the separate head studies.

But – here is where something flipped in my head.  These are before and after drawings of the same model, albeit in different poses:

   
30 October  1 November

It is hard to believe I drew these two days apart.  This is where I was at last night:

It is not just a case of practising heads, therefore I’m getting the heads better, nor is it that I have suddenly had an enormous spurt of hand eye coordination.  Rather, it’s case of seeing differently.  It is not obvious from these photos, but the accuracy of all of my proportions throughout the recent drawings has improved markedly (I would guess they are now accurate to maybe 5%, down from, say, 20%).  I have also managed to draw hands more or less properly in this drawing – probably the first time I have ever drawn hands passably well.   Here are the hands:

Also of surprise to me is that I am now able to draw the models’ hair.  My rendering of hair improved dramatically in the 1 November sample, and has improved again in the 4 November drawing.

The key thing was loading the photos into Inkscape, drawing them freehand from the monitor, then going back and using the vector drawing elements to overlay grids to show sizing relationships.  Comparing these relationships as shown by Inkscape to those measured (with a ruler) on my drawings somehow snapped my brain into a different mode of seeing (presumably R-Mode to use Betty Edward’s nomenclature – I own, but have not worked from her book), one where I was simply absorbed in the drawing.  I seemed unable to accurately judge variance between reality and  the drawings just from sight alone unaided.

Moreover, now I am actually interested in seeing.  Looking back now I can tell before that, a lot of my problem was  – and this is hard to find the right words to express it – I couldn’t be bothered actually looking at what I was supposed to be drawing.  Now I’ve found I’ve spent the whole weekend looking closely at everyone’s faces, seeing how they curve, how they fit with the rest of the skull etc.

Also, having used the overlays for half a dozen or so drawings I now no longer feel I need them.   The following weeks will be telling in this regard.

I keep looking at the drawings and can’t believe I drew them.  So, here’s the thing – I think if you want to draw, you probably can do it.  It’s not about drawing straight lines or anything it really is just a matter of seeing differently.

2 Responses to “Learning to See”


  1. 1 Will 5 November 2012 at 2:23 pm

    This is soo true.

    I took art class in high school when many of my academic peers were taking languages. This lesson, that you don’t see what you think you see, was probably the best scientific training I got in high school – more important than any single idea I leant in ‘science’.

    The easy way to give someone else the experience is to get them to copy a photograph. First have them copy the photograph right way up, then have them copy it upside down. The second approach breaks their normal perceptual mechanisms so they tend to ‘copy the lines’ rather than interpret the image and then try to draw their interpretation. The upside down approach is almost always a much better image.

    Drawing like this should be a staple of science classes. (And historically was.)


  1. 1 Learning to See « Brendan Learns To See Trackback on 6 November 2012 at 3:09 pm

Leave a comment




Blog Stats

  • 279,029 hits

OSWALD Newsletter

If you would like to receive OSWALD, a weekly open source news digest please send an email to oswald (with the subject "subscribe") at opensourcelaw.biz