whole can only be taken as an illustration of a poem in sonata form. But this nevertheless has some importance, as showing the acceptance of the aptitude of sonata-form for such purposes by a composer who was by no means in full sympathy with the lengths to which Schumann was prepared to carry the romantic theories.
Among other living composers who treat sonata-form in a poetic fashion, we may name Raff and Rubinstein. The works of the former are always admirable in the treatment of the instruments, and both composers frequently present subjects of considerable fascination; but neither have that weight or concentration in structural development which would demand detailed consideration. Poetic treatment is commonly supposed to absolve the composer from the necessity of attending to the structural elements; but this is clearly a misconception. Genuine beauty in subjects may go far to atone for deficiency and irrelevancy in the development, but at best it is only a partial atonement, and those only are genuine masterpieces in which the form, be it ever so original, is just as clear and convincing in the end as the ideas of which it is the outcome.
The whole process of the development of the Sonata as an art-form, from its crudest beginnings to its highest culmination, took nearly two hundred years; and the progress was almost throughout steady, continuous, and uniform in direction. The earlier history is chiefly occupied by its gradual differentiation from the Suite-form, with which for a time it was occasionally confounded. But there always was a perceptible difference in the general tendency of the two. The Suite gravitated towards dance-forms, and movements which similarly had one principal idea or form of motion pervading them, so that the balance of contrasts lay between one movement and another, and not conspicuously between parts of the same movement. The Sonata gravitated towards more complicated conditions and away from pure dance-forms. Diversity of character between subjects and figures was admitted early into single movements, and contrasts of key were much more strongly emphasised; and while in the Suite, except in extremely rare cases, all the movements were in one key, amongst the very earliest Sonatas there are examples of a central movement being cast in a different key from the rest.
In a yet more important manner the capacity of the Sonata was made deeper and broader by the quality and style of its music. In the Suite, as we have said, the contrasts between one movement and another were between forms of the same order and character—that is, between dance-forms and their analogues; but in the Sonata the different movements very soon came to represent different origins and types of music. Thus in the early violin sonatas the slow introductory first movement generally shows traces of ecclesiastical influence; the second, which is the solid kind of allegro corresponding to the first movement of modern sonatas, was clearly derived from the secular vocal madrigals, or part music for voices, through the instrumental canzonas which were their closest relations. The third, which was the characteristic slow movement, frequently showed traces of its descent from solo vocal music of various kinds, as found in operas, cantatas, or other similar situations; and the last movement earliest and latest showed traces of dance elements pure and simple. A further point of much importance was the early tendency towards systematic and distinct structure, which appears most frequently in the last movement. The reason for the apparent anomaly is not hard to find. The only movement in the group on a scale corresponding to the last was the second, and this was most frequently of a fugal disposition. The fugue was a form which was comparatively well understood when the modern harmonic forms were still in embryo; and not only did it suffice for the construction of movements of almost any length, but it did not in itself suggest advance in the direction of the sonata kinds of form, though it was shown to be capable of amalgamation with them when they in their turn had been definitely brought to perfection. In the dance movements on the other hand, when the fugal forms were not used, all that was supplied as basis to work upon was the type of motion or rhythm, and the outlines of structure had to be found. As long as the movements were on a small scale the structure which obtained oftenest was the equal balance of repeated halves without contrasting subjects, of which the finest examples are to be found in Bach's Suites. The last movement was in fact so long a pure suite movement. But when it began to take larger dimensions, emphasis began to be laid upon that part of the first half of the movement which was in the dominant key; then the process of characterising it by distinct figures or subjects became prominent: and by degrees it developed into the definite second section. Meanwhile the opening bars of the movement gradually assumed more distinct and salient features, making the passage stand out more clearly from its immediate context; and in this form it was repeated at the beginning of the second half of the movement, the second section being reserved to make a complete balance by concluding the whole in a manner analogous to the conclusion of the first half. So far the change from the suite type of movement rests chiefly on the clearer definition of parts, and more positive exactness in the recapitulation of the subjects; but this is quite sufficient to mark the character as distinct, for in the movements of the Suite (excluding the prelude) balance of subject and key were never systematically recognised. The further development of binary form, in which the recapitulation of the distinct subjects was reserved for the conclusion, took some time to arrive at, but even at this early stage the essential qualities of sonata-form are clearly recognisable. The Violin Sonata was naturally the kind which first attained to perfection, since that instrument had so great an advantage in
