consonance and discord

A glance at music may drive the argument home. If musical harmony were concerned only with the rules of what sounds well together, it would be limited to a kind of aesthetic etiquette. Instead of telling the musician by what means he can express what, it would teach him only how to behave. Actually this aspect of musical harmony has proved to be of no permanent value, because it is dependent on the taste of the period. Effects that were forbidden in the past are welcome today. This is precisely what has happened to certain norms of color harmony even with a few decades. For example, Ostwald, commenting in 1919 on the rule that saturated colors must be presented only in small bits, asserted that large-sized surfaces of pure vermilion, as found in Pompeii, are crude, "and all the blindly superstitious belief in the artistic superiority of the 'ancient' has been unable to keep attempts at the repetition of such atrocities alive." In reading this today, we may remember a painting of Matisse in which six thousand square inches of canvas are covered almost completely and quite satisfactorily with a strong red, and conclude that the proposed norm was nothing but the expression of a temporary fashion.

But—to come back to music—the rules of theory are hardly concerned with such matters. Arnold Schönberg says, in his Theory of Harmony: "the subject matter of the doctrine of musical composition is usually divided in three areas: harmony, counterpoint, and the theory of form. Harmony is the doctrine of the chords and their possible connections with regard to their tectonic, melodic, and rhythmic values and relative weight. Counterpoint is the doctrine of the movement of voices with regard to motivic combination . . . The theory of form deals with the disposition for the construction and development of musical thoughts." In other words, musical theory is not concerned with what sounds nicely together, but with the problem of how to give adequate shape to an intended content. The need for everything to add up to a unified whole is only one aspect of this problem, and it is not satisfied in music by drawing the composition from an assortment of elements that blend smoothly in any combination.

To state that all colors contained in a pictorial composition are part of a simple sequence derived from a color system would mean no more—even though perhaps no less—than to say that all the tones of a certain piece of music fit together because they belong to the same key. Even if the statement were correct, still next to nothing would have been said about the structure of the work. We should not know what parts it consists of and how these parts are related to each other. Nothing would be known about the particular arrangement of the elements in space and time; and yet it is true that one and the same assortment of tones will make a comprehensible melody in one sequence and a chaos of sounds when shuffled at random, just as one and the same group of colors will produce a senseless jumble in one arrangement and an organized whole in another. Also it goes without saying that composition requires separations just as much as connections, because when there are no segregated parts there is nothing to connect, and the result is an amorphous mash. It is useful to remember that the musical scale is suited to serve as the composer's "palette," precisely because its tones do not all fit together in easy consonance but provide discords of various degrees as well. The traditional theory of color harmony deals only with obtaining connections and avoiding separations, and is therefore at best incomplete.

-- Rudolf Arnheim Art and Visual Perception