Smith-Kettlewell Eye Research Institute
 Christopher W. Tyler
« Previous    Next »        
  Introduction   |   One point   |   Construction lines   |   Construction method   |   Globes   |   Canaletto   |   Summary

What Perspective Constructions Reveals About Optical Aids

Hockney and Falco claim that artists use of optics as explains the 'optical look' that appeared in paintings after about 1430. As supporting evidence, they show that many paintings reveal different vanishing points in different regions, compatible with the use of optics with a small clear zone, moved from region to region to construct each part of the image.

I show several lines of contrary evidence, implying that Renaissance artists constructed their compositions purely through artistic intuition, without optical aids (or accurate geometric methods).

  1. Most Renaissance artists show perspective discrepancies within local regions, refuting an accurate use of optics.
  2. Construction lines reveal the use of geometric perspective.
  3. Even if the optics were moved, all vanishing points should still lie in the horizon, a requirement violated among the discrepant vanishing points observed.
  4. Optical aids such as mirrors and lenses are not photographs; they cannot explain the ability to capture elements in motion such as rearing horses.
  5. The Hockney/Falco/Graves demonstrations reveal that optical projection has a very narrow depth of field. The 'optical look' therefore should include many regions being painted out of focus. No Renaissance paintings exhibit the literal optical look of such optics.
  6. Falco's "Rosetta Stone" of Lotto's 'Santa Lucia' does not have a unified geometry of its central pattern element, even though apparent blurring implies a single optical projection zone. The perspective is locally haphazard, and is compatible only with intuitive construction.

Perspective Analysis

If an artist copied from a projected image, the local perspective should be undistorted. Reconstruction of the perspective geometry in each region of the image should reveal any systematic imperfections, implying both an incomplete knowledge of perspective rules and lack of use of optical aids.

Jan Van Eyck and the Northern Optical Movement of the early 1400s



In proportion, this Madonna would have been about 30 feet tall; therefore this image could not have been a single optical projection. This leaves the possibility that the cathedral could have been a background projection, with the Madonna later superimposed upon it. This view may be examined by inspecting the detailed perspective of the architecture.

This painting is only about 12" high, so could have been projected optically in separate projections for the upper and lower halves. The incredible architectural detail gives the opportunity to examine the perspective construction.

However, separate (but accurate) vanishing points for different local patches of the apse make it clear that van Eyck was following a variety of construction rules, not copying from an optical projection of any kind.

Point at the name "Campin" in the diagram at right to see Robert Campin's Annunciation (at the Prado in Madrid). The detailed architecture in this early Annunciation (1426) provides the opportunity to evaluate Campin's painting methods.

There are many different vanishing points even within local regions of the design, so Campin can have used neither optics nor geometry.

< Previous       Next >