half. The second half, commencing at (b), can easily be perceived to have a freer harmonic basis than either of the first sections, and so leads the mind away from the Tonic and Dominant centres in order that they may come in fresh again for the conclusion; and having carried the figure on to an apparently disproportionate length (which serves the excellent purpose of breaking the monotony of constant pairs of bars), finally, at (c), resumes the little tail-piece of the first half and thereby clenches the whole into completeness. The manner in which this answers the requirements of artistic construction is very remarkable, and it will be found hereafter that it does so throughout on a precisely similar scheme, in miniature, to that of a 19th century Symphony movement. It would be natural to suppose that this was pure accident if there were not other ancient examples of the same form coming from the most opposite sources. The above Branle is a French dance tune; if we turn from it and take the most famous German Chorale 'Ein' feste Burg' (1529), the principles of its construction will be found to be identical. It is so well known that it is needless to quote it.[1] It will be sufficient to point out that the first half of the tune ends at the conclusion of the second line; and of this half the first line ends on the Dominant and the second on the Tonic, precisely as in the Branle; and it is then repeated for the third and fourth lines. The music to the fifth, sixth, seventh, and eighth lines answers to the passage between (b) and (c) in the Branle, and like it presents a variety of harmonic bases; and to clench it all together the music of the second line is quoted to conclude with, precisely as in the little tailpiece of the first half in the Branle. It is impossible not to feel the force of this as a point of musical form when it is once realised; it has the effect of completeness for a short tune which is unrivalled. If we turn to far other sources we shall find an early English specimen in the well-known 'Since first I saw your face' (1607), in which the second and last line will again be found to be identical, and the other points of the scheme to conform in like manner. Even in Italy, where the value of form does not seem to have been so readily appreciated as by Teutons, we find a little Sinfonia for flutes in Giacomo Peri's 'Euridice' (1600)—the first musical drama performed in modern Europe—which at least has the one important feature of repeating a little characteristic figure of the cadence of the first half to conclude the whole. It must not be supposed that this form was by any means universal so early as the middle of the 16th century—a time when notions of harmony proper, as apart from polyphony, were but dawning, and the musical scales and keys as we now know them were quite vague and unsettled. It is wonderful enough that there should be any examples of Form at all in such a state of musical language; for Form as now recognised depends greatly upon those two very elements of harmonic bases and relation of keys; so that what was then done in those departments must have been done by instinct. But by the middle of the 17th century musical knowledge in these respects was much more nearly complete, and the scope of composers proportionately widened. Accordingly we find a greater freedom in the treatment of forms; but the outline of the same form on a larger scale is found to predominate in the instrumental works of the tune, especially such as pass under the names of dances; though it is probable that those sets of them which were called 'Suites,' or 'Sonatas,' or 'Ordres,' were rather purely Musical than Terpsichorean. In the ecclesiastical Sonatas (Sonate di Chiesa) the style still continues fugal and polyphonic.
It would be impossible to give even a faint idea of the number of examples of this form which are to be found in these dance-tune suites, but it will be well to take some typical specimens and indicate the points in which they show development. In Corelli's Chamber Sonatas there are many clear instances. Thus, in the Giga of Sonata IV of the 'Opera Quarta,' there is the usual division into two halves. Of these the first is again divided into two phrases, the first phrase all in the Tonic key, D; the second then modulating to the key of the Dominant and closing in it. The second half begins with a sort of development of the figures of the first part, then modulates to nearly related keys, and after passing back to the original key concludes with a quotation of the last few bars of the first half. In this scheme there are two points of advance on the previous examples; the first part concludes in what we will henceforward call the complementary key, or key of the Dominant, instead of merely passing to it and back and closing in the principal key—by that means establishing more clearly the balance between it and the principal key; and secondly, the first part of the second half of the movement presents some attempt at a development of the features of the subjects of the first part, and real free modulation. The Corrente and Giga of the 7th Sonata of the 'Opera Seconda' are also remarkably clear specimens of repetition of the end of the first part as a conclusion to the whole, since full six bars in each are repeated. Both examples are however inferior to the above-quoted Giga in respect of the conclusion of the first part being in the principal key—like the older examples first quoted as typical—though like that Giga they are superior to the older examples in the free modulations and reference to the conspicuous figures of the subjects in the first section of the second half of the movements.
Domenico Scarlatti (1683–1757) was a contemporary of Handel and Bach, being but two years older than the former [App. p.637 "they"]; nevertheless he must be considered as historically prior to them, inasmuch as the very power of their genius would make them rather the prophets of what was to come than representatives of prevalent contemporary ideas. Domenico Scarlatti left many examples of Studies or Sonatas which are
- ↑ It is given on p.484
