Monday, June 4, 2012

Famous Photo of the week

The Steerage, Stieglitz.
As a composition, this is divided in two top to bottom. The upper mass of people is neatly framed by various objects and lines, centering your attention on the white hat. The man's face and hand, if not the actual hat, is placed at about 1/3 of the way down the frame (rule of thirds!) while the various linear objects in-frame give more or less strong diagonals, especially the gangplank. The implied line of sight from the man in the white hat could be considered to be interrupted by the gangplank, but I think one can argue convincingly that it passes behind the gangplank , down into the second mass of people (the ones in Steerage).
The relative tidiness of the framing above contrasts with the relative untidiness below, arguably echoing in the composition the contrast between the two classes of people involved, although this might be a bit of a stretch.
As a composition, I think this one is fair rule-following, as far as the simple rules of composition we run in to over and over. It is a particularly fortuitous arrangement of shapes and lines, a little random to be formal as such, but while it's loose I think it's pretty obedient to convention.
Of course the point of this image is, as is so often with the really good ones, essentially political.


  1. How is this a composition of thirds? I see a horizontal line across the middle of two dark regions. The "hat at thirds" is really stretching things, don't you think?

    I think this is what bothers me about the "rule of thirds" - it's too subject to post facto confirmation.

    Here's the obvious rebuttal: everything that has something in the center, is ALSO broken into "thirds" because now you have the thing in the center and the stuff on either side. Consider the Karsh portrait of Churchill as an example of what I am talking about.

    1. All rules of composition are woefully subject to post facto confirmation, I suspect. On the other hand, that is to an extent how we figure out why things work, isn't it? We build models, and then try to fit them to the data we have. There is definitely something pleasing about this image, and I think it's worthwhile trying to figure out what!

      The hat is definitely not at a 1/3, but I think you can make the case that the man's hand/face/hat, the "man himself" in a sense, is pretty close. Wherever it is, the hat and the face are clearly fortuitously placed in the frame. Maybe the point is that it balances with the dark area under the gangplank or something?