and more advanced artistic forms. Grosse's The Beginnings of Art is a good example of this interest and of a new scientific sociological and ethnological method. Not only is light thrown on one specific problem, that of the origins of art; the book is controlled throughout by an ideal of a scientific biology of culture and reveals the use of a genetic analysis of the facts of experience. One cannot imagine Grosse taking pleasure in the botanizing excursions, vague and ambitious, of Taine or having much confidence in the cultural mesh-bags of Schiller, Croce, or Faure, through which facts are constantly slipping. He is too scientific for that too intent on the check-up of experience. When he studies the dance he distinguishes its early mimetic and gymnastic types and sees them in relation to definite customs and beliefs and as expressive of impulses and feelings. Groos in The Plays of Man shows a like advance over the theories of Schiller and Spencer. It is one thing to say that man is wholly man only when he plays and to interpret art as Spieltrieb; it is quite another thing to trace in detail the playful activities of man and of the lower animals, show significant variants, as in the dramatic games of children, and to relate this playful activity to anticipatory instincts of use in serious living or to survivals of what was once of service to life. It is one thing to generalize with Spencer, quite another to demand and furnish verification in detail.

Art is to be read as part of a social text, which in turn is to be studied in the spirit of a painstaking scholar. The application of this method to aesthetics has resulted in large gain. Its successes are, however, most striking in the simpler and earlier forms of art. Thus may a war mask or totem pole be studied or a bit of pattern in its modifications and migrations, or a tribal dance, like the Snake Dance of the Hopi Indians. But when social influences are many, and art is advanced and personally colored, the method

either slips back to its older, vague form or vainly seeks help from biological guesses, as bold as they are questionable. In such a situation the only hope lies in using the cultural method only where it can be effectively used, and in showing a willingness to be as flexible in method as aesthetics is varied in content.

THE EXPERIMENTAL METHOD

The experimental method may be roughly defined as the commonsense method made scientific by greater care in selecting and interpreting, and by the use of methods and appliances which allow a more minute, accurate, and searching study of art material and aesthetic impressions. Both methods are empirical and unembarrassed; both operate by means of introspection, observation, and analysis. Certain glaring faults in the first method are avoided in the second. Suppose I say of a certain well-known picture, “I like it"; and then look for the sources of my enjoyment in the painting. I observe balanced grouping, graceful curves, color harmonies and contrasts; and I respond to the subject and its emotional associations. To what degree do all these contribute to my enjoyment? In order to answer this question I look within and try to discover the inwardness of my response. But if I do that and only that, I am caught within a single mood-and moods change. I neither observe exhaustively nor am able to escape the capricious influences of the moment. If I seek to standardize my enjoyment by comparing my responses at different times, I have gone but a little way, for they are, after all, my responses, and, individually colored as they are, may not be shared by others. If I interview these others and seek some sort of agreement, I am exposed to the "personal fallacy" and to an uncritical acceptance of the vague enthusiasms and biased judgments I meet. I am still moving on

the plane of the commonsense method and am paying the penalty. I cannot disentangle associations that carry me to the heart of a picture from such as hurry me away from it; I cannot discover the true value of a line, color, or compositional scheme.

If I am to force my way to a better method and a less subjective understanding of art, I must devise some type of experimental control. The first control that suggests itself is exhaustive analysis of the work of art itself. Observe color relations, measure distances, plot curves, trace ratios. Such a method, if judiciously used, clears away subjectivities; reveals the technique and often the aim of a picture or a piece of sculpture; and opens the way to a wider, comparative study of art. But it has its dangers. It may easily become too objective in the sense of overlooking visual illusions, personal impressions and preferences, and the part these play in the free, creative activities of art. Art, as it is created and enjoyed, is a psychical experience. A purely mathematical method may easily lead to a barren schematization as thoroughly right and as thoroughly wrong as a musical notation which awaits the interpretative stressing of a master.

The method of experimental psychology does not neglect this fact of preference. To Fechner belongs most of the credit of pioneer work in experimental aesthetics. In 1855 Zeising published Aesthetische Forschungen, a book whose title and table of contents seem to promise a scientific aesthetics. It turns against Hegel, contains a detailed analysis of aesthetic types, and offers studies in simple geometrical forms, in symmetry and proportion, in sound and rhythm. His aim is "to investigate beauty in the spirit of the student of the natural sciences, and to trace its causes in time and space relations, material and formal conditions; causes which produce the various effects of various aesthetic material." Pages 165-187, taken in conjunction with an

earlier essay of Zeising's, Neue Lehre von den Proportionen des Menschlichen Körpers (1854), gives an interesting and in many respects valuable analysis of the Golden Section Ratio. But all this promising material is too deeply embedded in philosophical terms and classifications.

It was Fechner who took the decisive step. In an early essay, Ueber die Frage des goldnen Schnittes (1865), he presents "certain empirical facts" against Zeising's overemphasis on the Golden Section Ratio; in another essay, Ueber das Assoziationsprinzip (1866), he makes a study of the associative factor in aesthetic experience; in Zur experimentalen Aesthetik (1872) he develops his method. All this, and much more of the same kind, is taken up into his Vorschule der Aesthetik (1876). Modern experimental aesthetics has rejected many of Fechner's conclusions and has advanced by many steps the technique of experimentation, but the point of view and the way of going about problems remain. It is therefore worthwhile to watch Fechner at his work.

Fechner cut out of white cardboard ten rectangles of the same area, ranging from the ratio 1:1, the square, to the ratio 2:5, a long narrow rectangle. One of these figures embodied the Golden Section Ratio, 21:34. They were submitted during a course of several years to a few hundred persons of both sexes, who were asked to express their preferences independently of use and other associations. The judgments were carefully tabulated, and it was discovered that the peak of the curve was in the neighborhood of the Golden Section Ratio, the preferences sloping sharply downwards at both extremes. Fechner was too careful an investigator to stop at this point. He studied various objects in common use, books, visiting cards, portfolios, stamps and objects commonly seen, as picture frames and discovered that the square was seldom used and that the ratios close to the Golden Section Ratio were frequent. He submitted

to children in nurseries the square and the Golden Section rectangle, watched their reaching out-changed the figures from right to left to guard against right and left-handedness and discovered no decisive preference.

What if the rectangle under consideration is a picture frame? Will not the pleasing ratio of height and breadth vary with the subject of the picture? Fechner tests this question experimentally. He puts pictures into classes: religious, mythological, genre, landscape and seascape, still life; compares examples of each class and tabulates results. Other illustrations of Fechner's method are: his study of vowel-color; his careful separation of the direct and associational factors in art; and his analysis of the direct values of sound, color, line, and of the effect of associations.

With this scientific control in terms of fact always in mind, Fechner is unwilling to reduce aesthetic experience to one principle or law. He formulates six: the principles of the aesthetic threshold, of summation, of unity in variety, of harmony and truth, of clearness, of association—and these are not given as an exhaustive list.

Since Fechner's time experimental aesthetics has developed rapidly. The technique of investigation has been improved, and the range of experimentation, widened. The improvement in technique can be traced to the invention of various laboratory devices. Revolving disks with color segments allow the study of contrast and fusion; eye-movements are studied by means of photography 2; reaction time

2 The following may serve as an illustration of how aesthetics benefits by certain technical improvements. Curves are supposed to be more pleasing than straight or irregular lines. That is an assertion on the part of commonsense which must be put to the test and, if valid, must be backed by some theory that accounts for the facts. One such theory offered was that the eye-ball, owing to the peculiar way it is set in and controlled by the muscles, works most easily in curves. But photography has shown that the eye moves not in curves but in very irregular broken lines. (cf. Valentine, An Introduction to the Experimental Psychology of Beauty, pp. 44-46).