Binary is a piece that divides into two parts and ternary one that divides into three. It all seems very simple; and it would be if composers would write according to the book. Although there are many pieces that are binary or ternary there are also many that do not divide so defiantly. In these cases the books on form disagree, some saying binary and some ternary; and, to make things more difficult, some writers have invented new methods of description. However, the words binary and ternary are still in common use, and must be explained.


At first this was used for short movements. It will be found in the seventeenth and eighteenth century suites, and is suited to the short dance movements which were in vogue then. Since about the middle of the eighteenth century composers have made little use of it, except for small piano pieces. Examples occur in Schumann’s Album of the Young and Mendelssohn’s Song without Words. Many simple songs and hymn tunes are in binary form.



The four lines of the words are shown by the commas in the music, the main break coming at bar 4. The two parts are equal in length, and the note next in importance to the tonic (the dominant) ends the first section. In musical content, the two parts are different, the first showing a rise of a sixth in bars 1 and 3 and the second using quaver scale-passages, descending in bars4-5 and ascending in bars 6-7. The drop of a fifth in bars 5-6, extending through a sixth (b’-e’-d’), helps to preserve the unity of the song.

In many pieces of this type the first cadence is on the dominant, without modulation, but the effect is often made stronger by modulating as in the below example.


There is a definite modulation to the dominant at bar 4. The two sections are similar in rhythm and in their general type of movement. The rise of the fourth which appears three times in the first part is used once in the second.

In the eighteenth century suite the first part of a binary movement normally modulated to the dominant; or, if the movement were in the minor, to the relative major, or the dominant minor; or merely ended on the dominant chord, without modulating. There are exceptions. In the eight suites of Purcell there are two movements in which the tonic chord is used for the end of both parts. The same happens in Menuet II of the French Suite in D minor by Bach. Occasionally the first sections ends on a dominant chord and the second section begins in the tonic or reaches it in a bar or two. In general, however, it can be said that the progress of the first part is towards the dominant.


The above example shows a species of binary form in which each section is of the same length. There are eight bars in each part with the first ending on the chord of the dominant (there is no modulation). The dotted crotchet figure is the main feature of the first part; the second has heightened interest because of the dotted quavers in bars 12 and 14. Each part is repeated, making thirty-two bars in all.


He more common type of binary movement, eight bars answered by sixteen. The first part modulates to the dominant, and the second begins in the dominant and works back to the tonic. The return journey goes to C sharp minor at bars 12/13. At bar 17 there is a passing reference to the opening theme. This is not quite so marked when heard because the bass part is different from that used at the opening. The opening of the second section in B major (bars 9 and 10) is an exact transposition of the opening of the first section. In this movement the second section is longer than the first – eight bars are answered by sixteen. This became customary. Modulations to the dominant is easy, and quickly done, but in coming back to the tonic other keys are passed through, to give variety and this takes more time.

In binary form there may be very little repetition (example 4) or none at all (example 3). It nevertheless gives an impression of coherence. This is partly because it represents as its best a development of one or two opening figures and partly because of its key scheme which is almost always symmetrical, the first part moving from tonic to dominant (or the relative key), and the second part returning.

Simple Ternary Form

In the three sections that make ternary form, the first and the third are similar. The third section may be shorter or longer than the first, it may be varied in detail or be an ornamental version, but the essential requirement is that the opening section shall be repeated in some recognizable way.

Example 5 is an illustration of a symmetrical piece in ternary form. The first section, a sad melody in D minor, has eight bars. The second section, a quick dance in D major, also has eight bars (with repeat). The third section uses the same melody a the first (except the grace-note) but the melody is in a lower register for the first four bars ad in octaves for the last four.

There are several other examples of ternary form in Schumann’s Album for the Young, but the middle section is not always such a contrast. In a short piece, too much contrast destroys unity. Often the key remains the same and the theme of the second section is only slightly different from the first.


This simple ternary, A B A, is very rare in folk songs or national melodies for the simple reason that a three, or six-line stanza is a rarity.

Most example of ternary form in folk songs or national song divide the normal four lines of the verse as follows:

Line 1 Tune A
Line 2 Tune B
Line 3 Tune C
Line 4 Tune D

Now if we write AABA as (repeat start) A (repeat end) BA (double bar line) we have two equal divisions which some would say should logically be called binary. Which is more important, equality of length or repetition of theme? If we hold to the first, the above structure is binary; if to the second, it is ternary. Repetition always strikes the listener forcibly; he is much less likely to be conscious of the relative lengths of sections; and for this reason it is generally accepted that ternary is the correct label for this type of form.

Another deciding factor is that of key. The dependence of the second part on the first is marked in binary form because the first part usually ends in the dominant. In ternary form the first part usually ends in the tonic and can therefore stand by itself.

In the larger structures the difference between binary and ternary is clear; but in small pieces, which usually have little contrast of key or thematic material, the lone of demarcation is not always clear. (Example 6)

In example 6 the Minuet and Trio from Haydn’s Quartet Op.3, No.6, both sections have a return to the main theme. In the Minuet, the return is at bar 17 and in the Trio at bar 33. The Minuet modulates to the dominant at the double bar, but the Trio does not. It could be argued therefore the Minuet is in binary form and the Trio is ternary. It seems absurd to describe procedures so similar with different names. Binary and ternary yare convenient enough labels for extended pieces, but are not always suitable for the smaller ones.