rapidly passing through G and A minor, and so to the dominant
of E, in various phases of tender humour and cheerful climax.
The keys are remote but not unrelated, the modulations are
smooth, and the style is that of a witty improvization. Beethoven’s
second subject is intensely agitated; its modulation begins
like Haydn’s as regards key, but its harmonies are startling
and its pace tremendous. Its regular rising bass carries it in
two steps to a totally unrelated key, through which it is urged
by the same relentless process with increasing speed, and when
it is at last driven to the threshold of the key which it seeks as
its home there is a moment of suspense before it plunges joyfully
into its cadence. Such resources as this enable Beethoven to
give rational dramatic force to every point in his scheme, and so
they soon oust those almost symbolical formulas of transition
and cadence which are a natural feature in Mozart’s music and
a lifeless convention in imitations of it. The growth of Beethoven’s
forms is externally most evident in his new freedom of
choice for the complementary key. Hitherto the only possible
key for the second subject was in major movements the dominant,
and in minor movements the relative major or dominant minor.
A sonata which begins by treating all directly related keys as
mere incidents in establishing the tonic, will very probably
choose some remoter key as its main contrast; and it is worth
while trying the opening of the Waldstein sonata (Op. 53)
with the simple alteration of C sharp and A natural for C
natural and A sharp in the bass of the twenty-first bar, so
as to bring the whole transition to the second subject on
to the orthodox dominant of G, in order to see, on the one
hand, how utterly inadequate that key is as a contrast to the
opening, and, on the other hand, how unnecessarily long the
transition seems when that is the key which it is intended to
establish.
3. The Sonata as a whole.—The history of the Waldstein sonata marks the irrevocable transition from Mozart to Beethoven (see iv. 88); and in his rejection of the well-known Andante in F (which was originally intended for its slow movement) Beethoven draws attention to the problem of the sonata as a whole, and the grouping of its movements. The normal sonata, in its complete (or symphonic) form, consists of four movements: firstly, a quick movement in that sonata form par excellence to which our discussion has been hitherto confined; then two middle movements, interchangeable in position, the one a slow movement in some lighter form, and the other a dance movement (the minuet, or scherzo) which in earlier examples is of hardly wider range than a suite movement. The finale is a quick movement, which may be in sonata form, but generally tends to become influenced by the lighter and more sectional rondo form, if indeed it is not a set of variations, or even, in the opposite extreme, a fugue. Aesthetically, if not historically, this general scheme is related to that of the suite, in so far as it places the most elaborate and highly organized movement first, corresponding to the allemande and courante; while the slow movement, with its more lyric character and melodious expression, corresponds to the sarabande; the minuet or scherzo to the lighter dance tunes or “Galanterien” (such as the gavotte and bourree) , and the lively finale to the gigue. But just as the whole language of the sonata is more dramatic, so are the contrasts between its movements at once sharper and more essential to its unity. Hence, the diversity of outward forms within the limits of these four movements is incalculable.
The first movement is almost always in the sonata form par excellence, because that admits of higher organization and more concentrated dramatic interest than any other. Often after such a movement a slow piece in the form conveniently known as A B A, or simple “ternary” form (i.e. a broad melody in one key, followed by a contrasted melody in another, and concluded by a recapitulation of the first) is found to be a welcome relief, and of great breadth of effect. Of course in all true classics the very simplicity of such movements will be inspired by that sense of rhythmic freedom and possibility of development that permanently raises sonata forms from the level of a mere decorative design ; nor, on the other hand, is there any limit to the complexity of form possible to a slow movement, except that imposed by the inevitable length of every step in its slow progress. Still, the tendency of slow movements, even more than of finales, is to prefer a loose and sectional organization. Sonata form is frequently used in them by Haydn and Mozart with the success attainable only by the greatest masters of rhythmic flow; but even in their works the development is apt to be episodic in character, and is very often omitted.
The minuet, in Haydn’s and Mozart’s hands, shows a surprising amount of rhythmic variety and freedom within the limits of a dance tune; but Haydn, as is well known, sighed for its development into something larger; and, though Beethoven had long emerged from his “first period” before he could surpass the splendid minuet in Haydn’s quartet in G major, Op. 77, No. 1, he achieved in the scherzo of his Eroica symphony the first of a long line of movements which establish the scherzo (q.v.) as an essentially new art-form.
The only condition that affects the forms of finales is that a sonata involves a considerable stretch of time, and therefore its end must be so designed as to relieve the strain on the attention. In a drama or a story the deeper artistic necessity for this is masked by the logic of cause and effect, which automatically produces the form of an intrigue ending in a dénouement. In music the necessity appears in its purest form. There is no need for finales to be less serious than first movements; or even, in certain ways, less complex; but the attention which could be' aroused at the outset by problems must be maintained at the end by something like a solution. Hence the use of the lighter rondo forms, which, by dividing the work into shorter and more distinct sections, make the development easier without unduly limiting its range. Hence, also, the influence of rondo style upon such finales as are cast in true sonata form; and hence, lastly, the paradox that the fugue has occasionally been found a possible means of expression for the finale of a dramatic sonata. For the complexity of the fugue, though incessant, is purely a complexity of texture, and the mind in following that texture instinctively abandons any effort to follow the form at all, finding repose in the change of its interests.
Now, just as within the typical scheme of first and second subject development and recapitulation in the first movement, there is room for genius in the contrasting of different rhythms and proportions, so, within the limits of the simple four-movement scheme of the whole sonata is there room for genius in the contrast of various types and degrees of organization. The complete four-movement scheme seldom appears in works for less than three instruments. Beethoven was the first to adopt it for solo sonatas, and he soon thought fit to make omissions. In Haydn’s work for less than four instruments it was not even necessary that the “sonata” form itself should be represented at all. Its essential spirit could be realized in the melodic and rhythmic freedom of a group or couple of more sectional movements, nor did Beethoven (in Op. 26 and Op. 27, No. 1) consider such works unworthy of the name of sonata, or (in Op. 54) incapable of expressing some of his most original ideas. No design is known to pure instrumental music that is not possible as a movement of a sonata, if it has the characteristic freedom of rhythm and is not much over a quarter of an hour in length. There is no form that has not been so applied; and, indeed, the only instrumental form that has maintained a larger development outside than inside the scheme of the sonata is that of variations (q.v.).
As the scope and complexity of the sonata style grew, so did the interdependence of its movements become more evident. With Mozart and Haydn it is already vital, as we have seen in the crucial case of Mozart’s G minor quintet ; but the differences between one scheme and another are not remarkable until we study them closely; and, except in key-relationship, it would be difficult to trace anything more concrete than principles of contrast as interacting between one movement and another. But Beethoven’s dramatic power finds as free expression in the contrasts between whole movements as it finds within the movements themselves. In his later works, the increase in harmonic
