# Trying to visualize Ternary Form with this piece

I am using this piece: "Ob ich manchmal Dein gedenke" (Josephine Lang, 1872) to try and visualize ternary form. I know the ternary form is ABA' but I am a little confused. Does it go A, then A repeats, and then B, then B repeats, and then A again slightly different? When I have tried to label the sections, I seem to run out.

I tried to have A section: m.1-20 and then B section: 21-28(?). But then what follows doesn't sound like A prime. How would you label the sections in this piece?

The score may be found on IMSLP.

• I took a look at that PDF and honestly, that A' section is so different from the A section that I'd rather say this song is in rounded binary form than ternary form. IMO, you picked a poor piece to practice visualizing ternary form with. Oct 7, 2022 at 3:30
• @Dekkadeci I could make a diagram that would be a solid ternary form with both an internal and external extension for the A' section, but I agree there is wiggle room on formal analysis and binary and ternary can be very similar. One thing about rounded binary is that the BA (second) section usually repeats as a unit. I don't think I see that in this piece, but I'd be happy to be corrected. One thing that suggests binary in this case is the restatement of A is only partial. Oct 7, 2022 at 3:50

It's fair to call this ternary form, but as an exemplar it's not straightforward — particularly because the final A section is significantly different from the initial one.

A: 1–8
B: 9–16
A': 17–32
Coda: 33–36

The accompaniment is clearest in terms of delineating the A section(s), because of the repetition of the dotted eighth—sixteenth Eb major chord followed by four eighth-note chords.

There is a clear change of harmony / mood that occurs in m. 9, the fully diminished seventh chord (especially following an obvious cadence in m. 8), which is the signal for a new section.

The return of the A section is clear, since the first four bars (mm. 17–20) repeat the piece's opening four bars. However, the variation/extension that beings in m. 21, but really gets moving in m. 22, obscures/confuses any sense that we've returned to the A section.

The main structure of the piece ends with the PAC in the home key of Eb major in m. 32, thus the remainder of the piece is a coda to affirm Eb major and allow for the emotion to taper gradually.

• Agreed on all counts, except possibly exactly where the internal expansion occurs in A'. Oct 7, 2022 at 3:53
• @ToddWilcox My explanation of the expansion was sloppy. Better now? Oct 7, 2022 at 3:56
• I was thinking maybe the extension starts in 23, but according to my forms professor, it's not always possible to objectively, definitively, precisely delineate an expansion, especially internal ones. The one thing the prof insists on being precise is the length of the expansion, which I think we would both agree is eight measures. Oct 7, 2022 at 4:00
• @ToddWilcox Ah, I better see where you're coming from. I placed the expansion at 22, because that's where the harmony changes significantly. I can understand, though, how a good case can be made for m. 23. And I agree that we agree that the A' section is eight bars longer than the A section. Oct 7, 2022 at 4:33

First, I strongly suggest you start your study of form from the smallest elements and work your way to larger ones (if you haven't already done this). Normally when analyzing form, we want to start from the macro structure, but when we learn about form, it helps to learn the smaller components (motives, cadences, phrases, sentences, periods, etc.) first. You'll definitely want to understand expansions of phrases and other auxiliary material, otherwise you're more likely to get confused by expansions that will make a section seem unusually long or short compared to the other sections.

For a ternary form, one thing you'll want to look for is a return of content from the A section. Without giving away everything, here are some things I notice about this piece that should help you understand the form:

• There are no repeat signs, so no, the sections are not repeated in the classical or baroque sense. Baroque and classical ternary forms often have repeats for each section, especially for dance music, but not always. As this was written in the 19th century, it comes from a time when "repeats" were not usually literal and therefore were written out. Note that forms are often a bit different for vocal music, such as this piece.
• Right at the beginning there is a motive starting on G, followed by different material starting in measure 3. The first motive returns on G in measure 5, and then again much later in measure 17.
• The piece is in Eb major, but there is a PAC in the key of Bb (the V of Eb) in measure 8.
• The last four measures are an auxiliary suffix like a codetta or post-cadential extension following the PAC in Eb (the I of the piece) in measure 32.
• It looks to me like there are internal expansions within the A' section. If I'm right, that means that the first two sections will be shorter than the third section.

My advice for understanding the form of a piece:

• Try to hear and see the macro structure first, but if that doesn't happen, start looking at the micro structure.
• When in doubt, find cadences and make sure to analyze what key the cadences occur in. This is crucial to distinguish half cadences to the V chord of the home key versus authentic cadences in the key of the fifth degree after a modulation.
• Sections are often composed of structured phrases and/or phrase groups, so be on the lookout for sentences, periods, and phrase groups that are similar to sentences and periods.
• Also consider the harmonic and thematic stability of phrases and sections. When you hear or see less stable material, that's a clue that the B section may have begun, but of course many B sections are stable, so this is just a clue, not a determinant.

When things come back, it's ternary form.

m1-7 is A
m8-16 is B
Then, at m17, A comes back.

It goes off the rails at m22, but that's OK. He's just building and resolving tension until the end (we can talk more specifically about how that happens, but it's not relevant to the question).

From a bird's-eye view, it's ABA.

• "Things come back" is not always a sign of ternary form. Rounded binary (A-B-A') and even rondos (A-B-A-C-A(-...)), strophic form (A-A'-A"(-...)), and sonata-allegros (specific version of (I-)A-B-C-A-B'(-Z)) all have "things come back". Oct 8, 2022 at 3:07
• @Dekkadeci Leon Stein, in Structure & Style, writes "the distinctive feature of the ternary pattern is the element of restatement or return." He lists "incipient three-part song form" (which seems to be the same as "rounded binary"), "first/second/third rondo forms", and "sonata-allegro" all as ternary forms. He talks about strophic songs in another chapter, and says "strophic, however, is not form-defining...strophic no more refers to a specific form than does prelude or nocturne." Oct 10, 2022 at 1:05
• The sonata-allegro (without introduction) is pretty much a fancy rounded binary form, not a ternary form. The B' section is too grossly different from the B section (wrong key, for starters) for the sonata-allegro to be a ternary form. Also, how is strophic not form-defining when there is such widespread agreement that it involves the same music repeatedly (note how long en.wikipedia.org/wiki/Strophic_form has existed, for example) and there is no such agreement for preludes or nocturnes? Oct 10, 2022 at 4:22
• These definitions are clearly controversial. I like the book I referenced, but everyone's free to make up their own mind. Oct 10, 2022 at 20:28

I think you just need to watch a few details of the terminology - binary, rounded binary, ternary, small ternary, large ternary - and understand rounded binary and small ternary are synonymous. "Ternary" is a bit too generic for your question.

Superficially `A B A` applies to both rounded binary and large ternary forms.

Rounded binary uses those letters but in this specific kind of structure: `||: A :||: B A' :||`.

Large binary uses those letters but the idea is the `A` and `B` sections are distinct and self contained sections.

The difference is mostly about the stuff labelled `B`. In rounded binary `B` is continuation and contrast material, but in large binary is has a clear expository function.

An important concept related to all this is standing on the dominant which is characteristic of the `B` of rounded binary. Standing on the dominant is not expository. It's harmonic contrast and continuation. It prepares for a return to the opening tonic.

In this song the stuff to label `B` is mostly about standing on the dominant, it's not expository. It doesn't present a new key, it's not a new theme.

This song would be called either rounded binary or small ternary form.

William Caplin's Classical Form discusses all these concepts in great detail.

A form that demonstrates both types of `A B A` in one composition is the menuet and trio. An example is Haydn's Haydn Sonata in C Major, Hob XVI 1. The menuet is the `A` part, the trio is the `B` part, and the second `A` part is the da capo repeat of the menuet. That is a larger ternary form. But, the menuet and trio are both examples of rounded binary (or small ternary) form `||: A :||: B A' :||`.

19th century pieces like piano dances or preludes also provide good examples of large ternary form. Look at examples like Chopin's Mazurkas - Mazurka Op. 24 No. 2 and Op. 17 No. 4 - are nice examples with double barlines and key/mode changes marking out the `B` sections clearly.

• This is a very good discussion of forms. I hope you'll consider opening a self-answer question like "What's the difference between rounded binary and ternary forms?". A quick search suggests we don't have such a question, and I think the value of your answer is greater than what it will achieve here. It should be seen; I think it's likely to get lost as the fourth answer to a question about a specific piece. In short, I would delete this answer and re-place it as the answer to a question that highlights it. Oct 8, 2022 at 19:42