I. DERIVATION OF THE RONDO.

Our investigations into the formal element of early dance music have thus far revealed two plans, "binary" and "ternary," i. e., two-part and three-part. For such short pieces as the inventions of Bach, and for many of the separate dances in suites, the two-part form was adequate, but when instrumental music began to develop on broader lines, so that each of the halves was extended to a considerable length, the advantage of the three-part form with its "restatement after contrast" was readily perceived, and it came to be frequently used.

Among the early experiments in form we find a kind of extension of ternary form by the repetition of its separate parts. Such pieces sometimes consist of but one stable phrase (A) with alternating phrases of an indefinite character, while others alternate two set phrases, as: A, B, A, B, A, B, A, etc. The one fixed principle in these pieces seems to be that they should end with the phrase with which they began. In primitive songs this fixed part constituted the chorus, with which the solo melody alternated, having, of course, different words for each verse.

[PNG] [[audio/mpeg]]

FIGURE XXII. NOUEL DE LAS FLOUS. (CAROL OF THE FLOWERS)

Chorus.

Come with us, sweet flowers, and worship Christ the Lord,

Let your perfumes hover round the Babe adored.

1st & 2nd Sopranos.

Chorus.

Come with us, &c. (D.C.)

1. Modest violet, hiding in the grassy shade,

Thou canst say how humble He for us is made.

Come with us, &c.

2. Lily fair, low bending in the sun's warm light,

Thou dost tell that He is pure as thou art white.

Come with us, &c.

3. As thou, Pansy, shinest forth in bright array,

So doth He His majesty to man display.

Come with us, &c.

4. As thou, Rose, wide-opening dost thy scent impart,

So His love expanding, draws each sinful heart.

Come with us, &c.

"The question is sometimes raised whether in the primitive carol the chorus began, or whether, as in many of our own popular songs, it waited until the end of the first solo verse. Probably the former is of the greater antiquity; in any case, it is from it that the rondo[15] is derived."—Hadow, "Sonata Form."

An example of this primitive type of carol will be found in Figure XXII.

This is an ancient carol from the old province of Bas-Quercy (now Lot-et-Garonne) in the southwestern part of France.

The obvious weakness of this form, when applied to instrumental music, is its monotony. One would soon weary of a bald repetition over and over again of two phrases or two melodies to which no variety was imparted, such as the change of words supplies in the foregoing carol. In order to avoid this disadvantage the natural step to take would be to impart, by some means or other, variety to the music; and this was soon perceived by composers. The idea of a fixed part remained, i. e., the chief musical idea was retained in its original form, but the secondary melodies were varied. Once this change had taken place the rondo became a frequent medium of musical expression. Specimens of the early rondo may be found in Purcell's song, "I Attempt from Love's Sickness to Fly," and in Bach's, "Passepied en Rondeau," from the fifth English Suite. The formula for these two pieces is A, B, A, C, A.

Another interesting point is the plan of the harmony of the contrasting sections in the rondo. The first of these (B) would naturally follow the prevailing custom for "sections of contrast," and be in the dominant, or, if the piece were in minor, in the relative major (see Chapter II.); but the second (C) offered a further means of variety, and the instinct of composers led them to treat it in a free manner and not confine it to any one key. Each of the examples of rondo form referred to above adopts this method of procedure.

While this early form of the rondo possessed a certain charm, it was somewhat rigid in effect, since the various sections were separated from each other by a full close or complete pause. They were like little blocks that fitted together into a definite, if somewhat stiff pattern.