(2 of 2)
If the paintings are about the upper extreme of recognizability — the point at which a high frequency of visual information almost drowns one's reading of a face as a face — Close's preliminary drawings are about the lower extreme.
How little information do you need to recognize a face? How generalized can it become before the specific relation ship of features falls apart? And which features are the first to go? Close's meth od is to grid off a rectangle into squares — up to 600 of them — and to use each square as a part of the portrait. The coarser the grid, the more detail is lost.
Close does not take this very far: the expressions of the mouth and the eyes go first, taking the nuances with them; the head becomes a masklike assemblage of volumes.
Computer analysis of photographs might press the inquiry further, but Close is restrained by his desire to make drawings rather than diagrams. The ink-drawn squares, each with its precise ration of diagonal shading, give one a visual effect that belongs to the same family — though not the same order of majestic intensity — as Seurat's chalk drawings; the spots of pastel in the stud ies for Linda are distributed with a dogged aesthetic zeal that recalls Signac.
Like the full-scale paintings, these notes are the products of a bracing intelligence whose workings go far beyond the simple counting of facts.