Notes on Form for MUTH 251

John Paul Ito

Sections:

1 Introduction 1
2 Binary Form 1
3 Simple and Rounded Binary Forms 2
4 Sectional and Continuous Binary Forms 3
5 Charting Form 3
6 Using Subscript Numbers and Primes 4
7 Ternary Form 5
8 Compound Ternary Form 7
9 Transitions, Codas, and Codettas 8

1) Introduction

During term 3 we learn about units larger than the period or double period. These forms
may be entire movements or they may comprise clearly defined sections within
movements. The individual names will be introduced as we go; collectively they are
usually referred to as small forms or song forms. This latter term has a long
pedagogical history, but it can be misleading; few of the pieces of music we will look at
could be described as songs, and while some songs are in these forms, many songs are
organized in other ways.

2) Binary Form

Any movement or well-defined section of a movement that has a clear two-part form has
what is called a binary form. We have already seen that grouping units must divide into
either two or three smaller units; binary form names division into two parts at the level of
the movement or of the significant section of a movement. These two parts may be
defined in a variety of ways (e.g. contrast of key or of instrumentation), but the most
common way is by repeating each part.
Notes on Form for MUTH 251 – John Paul Ito © 2007 2

Getting to the end of the first section and then resuming from the beginning defines the
boundaries of the first section: the beginning is already a beginning by merit of starting
the section, and the end is defined as an end by the return to the beginning – recognizing
a new beginning, what preceded the new beginning is retroactively understood as a very
significant ending. Passing the end of the first section the second time, the listener is then
prepared to hear what comes next as another new beginning; this defines the beginning of
the second section. Similarly, at the end of the second section, the return to the beginning
of the second section tells the listener that they just heard an ending; the return to the
beginning of the second section also confirms that beginning as a beginning. In this way
the two parts are defined for the listener. In many cases these repeats are indicated by
repeat signs; in some cases, in order to introduce variations the second time through, the
repeat may be written out.

We will designate the parts of binary forms using capital letters, A and B.

Example 1 shows a binary form, the theme from the final movement (a theme and
variations) of Beethoven’s Piano Sonata in E, Op. 109.

3) Simple and Rounded Binary Forms

Binary form comes in two varieties: simple binary form and rounded binary form. Both
have the characteristic two sections, A and B; the difference between the two has to do
with the structure of the B section.

The beginnings of B sections of binary forms are often somewhat harmonically unstable,
in one way or another avoiding tonic harmony. Binary forms end with cadences in the
original key, so at some point there will be a return to tonic harmony. In some cases this
return is synchronized with a recollection (sometimes fleeting, sometimes extensive) of
the melodic material that began the A section. In such cases we hear the outlines of a
three part form: a beginning, a contrast, and a return to the beginning, both melodically
and harmonically. If it is a binary form, though, this three-part structure is subsumed
within the more dominant two-part organization defined by the repeats. We call this
rounded binary form, and to describe it we will need two hierarchical levels. At the
highest level we have A B, but then we specify A=a, B=ba. Binary forms that lack this
rounding are called simple binary forms.

Because the main melodic material from the A section in Example 1 never returns, this
passage has a simple binary form. In contrast, look at Example 2, the second Bourrée
from Bach’s Fourth Cello Suite. Here, after four measures of contrasting material, the
material from the A section returns to close out the B section. This is a rounded binary
form.
Notes on Form for MUTH 251 – John Paul Ito © 2007 3

Distinguishing between simple binary and rounded binary forms:

The following chart lays out the difference between the two kinds of binary form.

simple binary rounded binary

Return of opening melody: maybe yes

Return of opening melody

synchronized w/ return of tonic key: no yes

4) Sectional and Continuous Binary Forms

In addition to simple and rounded, there is another pair of descriptors that may be applied
to binary forms: sectional and continuous.

When a grouping unit ends with an authentic cadence in the same key in which it began,
we describe that unit as harmonically closed. This is because it could stand alone as a
complete harmonic unit. When a grouping unit ends with a half cadence or with an
authentic cadence in a different key, it is called harmonically open; this is because it is
not harmonically complete by itself.

If the first section of a binary form is harmonically closed, the binary form is described as
sectional, because each of the main sections could stand on its own. When the first
section is harmonically open, the binary form is described as continuous, because the
first section needs the second to provide harmonic closure – the whole cannot be divided
into two self-standing parts.

5) Charting Form

As with the smaller forms learned in earlier terms, we will represent these larger forms
graphically. Here some differences may exist between sections of the theory core, and
the procedures developed here may not apply exactly to your section. You should
consult your instructor about this.

As before, we will represent form on a timeline using slurs for grouping units. But now,
in order to make it easier to see the larger aspects of the form, we will omit subphrases
(this aspect in particular is likely to vary from section to section).
Notes on Form for MUTH 251 – John Paul Ito © 2007 4

In general we will name sections of a form using letters; because we will eventually have
a significant number of nested levels, we will use several kinds of letter. At the highest
level (not needed yet) we will use bold face capital letters: A, B, etc. Below this simple
capitals: A, B. Then lower case: a, b, c, etc. For the smallest level (when needed) we
will use Greek letters: α, β, γ, δ, ε, etc.

Example 3 is a chart of the Beethoven passage given in Example 1, and Example 4 charts
the Bach cello suite movement from Example 2. Note that the music is charted as
written, not as heard; the A and B sections each appear once, not twice. This will be our
usual practice with notated repeats; written-out repeats will be charted separately, as the
variant has different measure numbers from the original.

There are just a few features that deserve special attention. In Example 3, only the large
sections of the binary form are named. The individual phrases are not named because
there are no returns that could be made apparent by doing this; the phrases would be
named a, b, c, and d. In Example 4, because of the rounding of the binary form, lower-
case letters are needed to describe the three-part organization, aba, that is found inside of
the larger binary form, AB. Also, the slur over the first four measures has two labels, A
and a. This occurs because those measures function as a single unit at two different
hierarchical levels.

In general, we will use as many levels of letter names as are needed to show thematic
returns. Example 5 shows the Minore (an unusually named Trio) from the Minuet and
Trio from Beethoven’s Piano Sonata Op. 22, a rounded binary form in which precise
charting of thematic relationships requires another level of letter names. As shown in
Example 6, which charts this movement, the two phrases of the A/a1 section have been
labeled as α1 and α2. This is in order to show the precise relationship with the a2 section.
a1 has two phrases where a2 has one, with a2 having a closer relationship to the second of
a1’s phrases, α2. This is because both organize the sixteenth notes in the left hand as
descending thirds (look at the first of each four-note group) and because both end with
PAC’s. The grouping unit labeled a2 is therefore also labeled as α2′. Greek letters were
used for the chart of rounded binary form shown in Example 6 but not for the chart of
rounded binary form shown in Example 4. This is because the additional level of letters
allows a relationship to be more precisely described in Example 6, while in Example 4
they would have provided no additional information. Take note of this as a general
principle: while details are always relevant to the larger musical picture, in making these
charts they should be included if and only if they allow the precise differentiation of
similar sections.

6) Using subscript numbers and primes

The previous example uses two notations to describe similarity and difference. There is
some similarity among units called α, but α1 and α2 are different. Similarly, there is
similarity between units called α2, but α2 and α2′ are different. In general, subscript
numbers will be used to mark larger differences, primes smaller differences. The exact
Notes on Form for MUTH 251 – John Paul Ito © 2007 5

differences marked will vary from piece to piece. In this case, α1 and α2 are similar as
the two phrases of a parallel period, but differentiated in their melodic continuation (the
second moving more uniformly in a chain of descending thirds) and by the keys and
types of their cadences. α2′ is associated with α2 because both share the descending chain
of thirds and both end in PAC’s – two shared points of contrast with α1. α2′ receives the
prime because it stays in the tonic where α2 modulates to the dominant.

7) Ternary Form

Some pieces or major sections have a similar structure to rounded binary forms, but they
lack the repeats – there is some initial music, this is followed by some contrasting music,
and at the end the initial music returns, possibly with some alteration. In a rounded
binary form, the repeat would result in a motion from the end of the second a section to
the beginning of the b section, the motion that fuses b and a together as the larger unit B.
Lacking the repeat, such pieces are almost always heard as three-part forms, or ternary
forms, designated ABA. (To avoid confusion with compound ternary forms, introduced
below, ternary forms that are not compound will sometimes be described as simple
ternary forms.)

Deciding whether a piece or section is in rounded binary form or ternary form is

straightforward: if b and a are repeated as a unit it is a rounded binary form, and if not it
is a ternary form. (Bear in mind, of course, that repeats may be written out, probably
with some variation.)

There are also features that are characteristic of rounded binary or ternary forms, but that
are not necessary, defining features.

The most typical ternary form is extremely stable, even static. All of the major sections
will generally be harmonically closed, ending with authentic cadences in the same keys in
which they began. Each section can stand on its own, with its own melodic interest, and
when the A section returns, it returns in its entirety.

In contrast, the most typical rounded binary form (especially for Haydn, Mozart, and
Beethoven – Bach’s practice has somewhat different norms) is a much more dynamic
form. The first section will usually modulate to some contrasting key; in major this will
likely be the dominant, in minor either the mediant or else the minor dominant.
Harmonically open, it will not be stable and self-sufficient like the typical A section of a
ternary form; rather it will demand continuation. (This amounts to a claim that
continuous rounded binary forms are more typical than sectional ones.)

The b section will be even less stable, whether compared with the A section or with the B
section of a ternary form. It will often lack melodic material of its own, perhaps using
some motive from the first section to create short, fragmented segments of music, and it
will almost certainly be open harmonically, as it must lead smoothly into the big event of
Notes on Form for MUTH 251 – John Paul Ito © 2007 6

the form, the return of the material from the first section in the original tonic key. The b
section usually ends with a half cadence or with a cadence that overlaps the return of a.

Finally, the concluding a section will often (again, especially for Haydn, Mozart, and
Beethoven) be shorter than the original. Its main tasks are to state the opening melodic
material and then to bring cadential closure in the original tonic key, and these tasks do
not require the kind of full-breathed exposition that an initial section will usually offer:
the first section will often have two phrases, the return of the a section only one. (There
may well be some post-cadential extension to allow the reverberations of the conclusion
of the drama to resound; for this reason the contrast in length is more consistent when
counting phrases than when counting measures.)

The following chart summarizes this discussion:

Deciding between rounded binary and ternary:

rounded binary ternary

Defining criterion:

Repeats (ba): yes: aa(ba)(ba) no: ABA

Typical features:

A closed more often not usually

B closed virtually never frequently

B new melodic material more often not usually

A:A′ 2:1 (esp HMB) 1:1

These contrasts are exemplified by the Minore from Beethoven’s Op. 22 that was given
as Example 5 and by Chopin’s Mazurka Op. 33, No. 3, shown in Example 7 and charted
in Example 8. In the Beethoven, the A section is harmonically open, modulating to the
dominant; the B section is also harmonically open, starting in C minor and ending in G
minor; the B section is melodically unstable and based on the main motive from A; and a1
is twice as long as a2, counted either by measures or by cadences. In contrast to this, the
Chopin has: major sections that are all harmonically closed; new material for the B
section that has its own stable phrase structure; and opening and closing A sections that
have the same length.
Notes on Form for MUTH 251 – John Paul Ito © 2007 7

As already emphasized, deciding between ternary and rounded binary form is always
based on the presence or absence of repeats in the approach used here (check with your
instructor as this may vary). A rounded binary form will sometimes have almost all of
the other features typical of ternary forms, but it will be considered a rounded binary
form if it has the repeats. (It is common, for example, for rounded binary forms to have
closed A sections; if this were not the case, the term “sectional rounded binary form”
would be of little use.) And there are a good number of ternary forms that have all of the
features typical of rounded binary forms (open A and B sections, fragmented, subsidiary
B material, and a second A section half the length of the original) but that lack the
repeats. Because of this one decisive feature, they will be categorized here as ternary
forms. You may decide that you really hear such pieces as rounded binary forms; if so
there are a good number of musicians who would agree with you. If you encounter this
or some other disagreement with these notes, consult your instructor about how best to
convey both the “correct answer” and the way you actually hear the music.

8) Compound Ternary Form

When at least one of the main sections of a ternary form is itself a small form (binary,
rounded binary, or ternary), the larger ternary form is called a compound ternary form.
In the Baroque period, the most common compound ternary forms were paired dances,
such as Gavotte 1 and 2 or Minuet 1 and 2. It was generally either stated or understood
that the performer should go back and play the first dance again after finishing the
second. (The instruction for this “da capo,” means “to the head” in Italian.) The
sequence of performance would be (for example) Gavotte 1 – Gavotte 2 – Gavotte 1,
creating a ternary form; because each of the two dances would almost certainly have a
binary form, the whole would have a compound ternary form. (Remember, though, that
it is only necessary for one of the main sections to be a small form in order for the whole
to be a compound ternary.) This practice survived in Classical and Romantic periods in
the form of Minuet and Trio and Scherzo and Trio movements, and Classical and
Romantic composers also found new opportunities for compound ternary forms,
especially in slow movements.

We have already seen the second Bourrée from Bach’s Fourth Cello Suite as Example 2;
Bourrée 1 is given as Example 9, and together they constitute a compound ternary form.
Compound ternary forms become large enough that a single phrase diagram with all of
the information will probably be hard to read. The notes will therefore chart different
levels separately. The whole form will be charted as in Example 10, showing only the
largest level of the form, using bold-face capitals to name the largest sections, and with
written-out measure numbers replacing the railroad tracks (or, as here, using movement
names if the second of the pair is independently numbered). Each of the major sections
will then be charted separately. Example 11 charts the first Bourrée, and Example 12
charts the second. In the context of the paired movements as a compound ternary form,
Example 12 is needed as a replacement for Example 4 because of the letters used.
Capital letters A and B had already been used in Example 11 for the two sections of the
Notes on Form for MUTH 251 – John Paul Ito © 2007 8

binary form, and reusing them would have implied a similarity of material. Because
Example 11 did not use lower-case letters, a and b could be used for the subsections of
the rounded binary form in Example 12. Be careful about when letters need to be used
consistently and when they don’t. There is no need for letters at different levels of
hierarchy to match, even when they are applied to the same section of the music. For this
reason the subsections of the second Bourrée can be called a and b even though the larger
sections are called C and D. It is important, though, that at each level of the hierarchy
(bold capitals, regular capitals, etc.) the letters be used consistently across the analysis;
this is why we replaced Example 4 with Example 12.

Note that the large chart creates an exception to a rule, as repeated music is charted twice.
This will be our usual practice with da capo (or da segno) repeats as opposed to simple
repeats; because the repeated music is heard after an interval and not immediately
following the first hearing, the form will not be visually evident unless the repeated
music is charted twice. The identifying information under the slurs will make the da capo
clear.

9) Transitions, Codas, and Codettas

Paired dance movements make for particularly straightforward examples of compound

ternary form because the closure of each major section is so clear, both to the listener and
to anyone looking at the score. When a compound ternary form is written out as a single
movement without da capo, as often happens in slow movements from the Classical
period and in 19th-century character pieces, it is not as easy to determine the form from a
quick glance at the music. But while this notational device makes the form less obvious
to the eye, the main sections may remain just as obvious to the ear. When composers
want to create greater continuity in a compound ternary form, they sometimes add music
at the ends of the major sections to smooth over the divisions in the form.

Example 13 shows a compound-ternary-form movement in which this happens, the slow

movement from Mozart’s Piano Sonata in C major, K. 330. The first A section, mm. 1-
20, is a simple binary form in F major, and the repeats in the music that follows seem to
define mm. 21-36 as another simple binary form in the parallel minor. But what are we
to make of mm. 37-40? Their key and motivic content make them group strongly with
the preceding music, as does the start of the second A section in m. 41. They sound like
(and function as) a post-cadential extension, but the fact that we hear this extension only
after the second playing of the cadence introduces extra complexity into their analysis.
The repeat back to m. 29 after the cadence in m. 36 defines that cadence as the end of the
body of the B section. The simple binary form is complete when we hear the cadence in
m. 36 the second time; the PAC back in the section’s key of F minor provides all the
closure that is needed. Because the post-cadential extension that follows is heard only
once, because it is not included in the repeated music, it is not a part of the body of the B
section; it is best understood as a tack-on following the end of the main action. A section
of this sort will be called a codetta, using an Italian word that reflects this status as an
optional addition.
Notes on Form for MUTH 251 – John Paul Ito © 2007 9

A codetta is a small coda, and a coda is a tail, in Italian. A tail is often not necessary for
an organism’s survival; cutting off the tail will rarely lead directly to death. Similarly, a
movement or a large section of a movement may come to a point of clearly sufficient
closure, but then have some extra music that follows this point of closure. The extra
music will be called a coda or a codetta. Extra music at the end of a movement is called a
coda, and extra music at the end of a large section is called a codetta. This status as
“extra” should be understood in terms of function within a form; it is not meant to imply
that this music is somehow less important or less satisfying. Though still viable, many
animals will be much less happy without their tails (for example because of difficulties
dealing with flies), and similarly a coda or a codetta will often be an essential part of the
expressive quality of a piece of music.

Returning to the example at hand, mm. 41-60 return to the material of the A section, this
time without the repeats. This section is still heard as a simple binary form, despite the
lack of repeats; this is because we already understand that formal structure from having
heard the repeats when the material was first presented.

Measures 61-64 are another post-cadential extension, and again we hear them as lying
outside the body of the second A section. This is because the end of the A section was
clearly defined by the cadence in m. 20, and by the start of new material following it;
when that cadence returns in m. 60, we again understand it as concluding the body of the
section.

At this point we must decide whether mm. 61-64 are a coda or a codetta. This may seem
odd, as above I stated that “extra music at the end of a movement is called a coda, and
extra music at the end of a large section is called a codetta.” The key point to grasp here
is that mm. 61-64 are found both at the end of the movement and at the end of the second
A section; this means that they could be heard either as a coda or as a codetta, depending
on whether they are understood to extend the entire movement or only the second A
section. The two options are charted as Examples 14a and b.

The criteria used in making this decision are not precise, and a judgment call will often
be necessary. Most often it has to do with a sense of proportion; a very short extension is
more easily heard as a codetta modifying just the final section, while a much longer
passage of music could be strange as a codetta, as the final extension could start to rival
the length of the music being extended.

In this case the content of the extension makes the decision relatively straightforward.
Measures 61-64 are extremely similar to mm. 37-40, and both belong to the world of the
B section in all respects except the major mode of the ending. The melodic material
clearly comes from the start of the B section, as do the repeated 16th’s in the left hand.
Measures 61-64 would therefore make a rather odd end to the second A section; if mm.
41-64 were performed as an independent movement, the ending would seem somewhat
out of place. The final four measures are a coda, not a codetta; by returning to the
material of the B section they round out the entire piece and help to unify the whole.
Notes on Form for MUTH 251 – John Paul Ito © 2007 10

Because the final measures are heard as a coda, Example 14a presents the larger-scale
form of the movement, but we will chart is as in Example 14c. Note that Example 14c
appears to violate one of the most basic rules of grouping structure: it has five groups at
the highest level. This is because Example 14c is understood to represent the analysis
shown in Example 14a. Because the final measures are labeled as a coda, we understand
that they follow and confirm the ending of the body of the piece as a whole. So as shown
in Example 14a, at the highest level there are two groups. And because mm. 37-40 are
labeled as a codetta, they are understood to group together with the B section; therefore
the body of the piece is divided into three subgroups, not four, in Example 14a. The
labels “codetta” and “coda” in Example 14c give all of the information needed to
understand the grouping structure shown in Example 14a; the extra slurs in Example 14a
are therefore unnecessary.

The final kind of musical connective tissue is called a transition. In a compound ternary
form, transitions occur between the major sections. In contrast to codettas, which
confirm the closure of sections, transitions lead on to the next section harmonically and
also often in terms of motivic content and texture.