The question, "What is the leading characteristic of Japanese painting?" has often been put to me, and I have invariably answered, "Repetition with slight variation."

Of course, there are other qualities to consider, as the peculiar color-distribution, the calligraphic dexterity of brush-work, the willful neglect or exaggeration of detail, the grotesque division of space, and the economic manipulation of backgrounds which apparently look empty and yet enhance the pictorial aspect of the picture to a rare degree. But more important than any of these peculiarities of composition seems to me to be their laws of repetition with slight variation, because a composition of that order possesses the two principal elements of pictorial art: It is decorative and yet true to life. Its object is not to execute a perfect imitation of reality (only bad works of art do that) or a poetic resemblance of life (as our best painters produce), but merely a commentary on some pictorial vision which sets the mind to think and dream.

If the Japanese artist wants to depict a flight of cranes, he draws half a dozen or more which at the first glance look alike, but which on closer scrutiny are each endowed with an individuality of their own. He foregoes perspective and all other expedients, he simply represents them in clear outlines in a diagonal line or sweeping curve on an empty background, and relies for his effect upon the repetition of forms. A Western artist would have expanded this at least into a picture with a landscape or cloud effect as background; to the Japanese artist, working in the narrow bonds prescribed by custom and taste, any such attempt can not be expressed more forcibly than by simply depicting the objects with only a slight variation in their representation.

The first form introduces us to the subject, its appearance, and action; the second accentuates the same impression and heightens the feeling of reality by the slight variation in the appearance and action; and every following form resembling, at the first glance, a silhouette is simply a commentary upon the preceding one; and all together represent, so to say, a multiplication of the original idea.

And in the same manner as they respect lines and masses, they vary color-schemes, which often resemble each other, but are nevertheless endlessly varied in shade and line. The French illustrators and the German designers of the "Secessionist" School have adopted this method with considerable success. The painters, however, have been rather reluctant about following their example. They probably realize that their plastic style of painting would not harmonize with the idea of repetition, which is strictly decorative and specifically adapted to flat-surface work.

I know of only two men who have successfully adapted this law in their composition and created something like a new style. They are Puvis de Chavannes, and D. W. Tryon, the American landscape painter. Both, however, were wise enough to avoid repetition in a diagonal direction or in a curve arrangement. Chavannes, in his mural painting, is very fond of the parallelism of vertical lines. Not only the trees, but also his human figures are constructed in that fashion. His aim is to express dignity and repose, and nothing can accomplish it better than an architectonic arrangement of vertical lines, as, for instance, in his "L'Hiver."

Chavannes composes at largo while Tryon is satisfied with adagios and andantes. The latter was addicted for years to the parallelism of horizontal lines. Undoubtedly he went a step in the right direction, as the principal line-idea in all natural scenery is necessarily horizontal, and a painted landscape, where this parallelism is accentuated and elaborately worked out (balanced by vertical line work and oval shapes), will convey the idea of vastness and level expansion more readily than those in whose composition a horizontal monotony of lines has been neglected.

In artistic photography I have not yet encountered any attempt at repetitions with slight variation, and I would advise no one to take it up without devoting some profound study to it, and even then I believe it should only be utilized when life or nature spontaneously suggests it. I do not believe that it can be forced into photography without looking forced; but that the photographers have to decide for themselves. Whoever wants to make a study of it must learn to appreciate its various ways of application, and thereby get down to the very essence of its esthetic value, will find ample opportunity, not only in painting, but also in the other arts.

In musical composition it is very frequent. The pieces which treat variations of one theme are innumerable. In Western literature we find it in the refrains of ballads, in Poe's poems, and the work of the French symbolists, and above all else in the writings of Maurice Maeterlinck, this quaint combination of Greek, medieval, and Japanese art reminiscences. In architecture it has always been one of the leading elements, only with the difference that in Western architecture everything has to be subservient to symmetry, while the Eastern world also recognizes (at least in the ornaments) the right of unsymmetrical composition. In the Gothic style one can study the parallelism of diagonal line, and in the Baroque and Rococo the repetition of curves. In dancing, the arrangement of a ballet, nearly everything depends on repetition; many figures are nothing but repetition without variation. The performers themselves substitute the lack of too frequent changes in movement and action.

Even the variety stage affords at times good opportunities for study. I realized it when I saw the Barrison Sisters. They were an object lesson that should have interested any student of art. There were five pretty, gay ladies of fascinating leanness and awkwardness à la Chavannes, who could neither dance nor sing, but who, simply having been drilled by a manager to expound in coquettish movements and attitudes a French-Japanese code of frivolity, unconsciously expressed the Japanese law of repetition with slight variation. But no other American critic at the time dwelt upon their esthetic values, and I may, after all, have been mistaken in my judgment.

Nature and every-day life, of course, are in this instance past-masters. One only has to keep one's eyes open to discover the raw material which the artist utilizes.

But there is still another side to the question -- at least from the Eastern point of view. Not only the composition of Japanese artists is guided by the law of repetition, but also their inventive power. As inexhaustible as it seems, one will find that they have always treated a certain line of subjects. For instance, they have painted a crow sitting on a snow-covered fir-branch with the full moon behind a thousand times; but every painter who has handled the subject has tried to lend it a new individuality. Only the subject remains the same; treatment and conception are invariably charged with the personality of the artist.

We Occidentals do not seem to be capable of this; our aim is above all to be original; like Richard III, we roam through the fields of art and say: "An original idea! An original idea! My life for an original idea!" forgetting that originality does not consist of something that has never been done before, but rather in new ways of expression. And nothing tends more to the very opposite of the conventional and commonplace than to find a new variation of an old subject. Thousands of mother-and-child pictures have been painted by the old masters, but artists like de Forest Brush, Abbot Thayer, Tompkins, and Mary Cassatt have understood how to lend the time-worn subject a new note of interest.

The craze for originality is really the curse of our art, as it leads nearly always to conventionalism and mannerism. The artistic accomplishments of the Japanese are due largely to the fact of their never-tiring study of variation. They have realized that a beautiful idea always remains a beautiful idea, and that it takes as much creative power to lend a new charm to an old theme as to produce and execute and apparently new one, which, after all, may prove an old one.