Two-Reprise and Binary are not Equivalents

Dimitar Ninov

Dear Colleagues,

Flipping through the pages of several newly published theory books, I was astonished to realize that, in some cases, their authors seem to assess musical form more visually than aurally. As a result of that kind of mechanical analysis, clear-cut small ternary forms are pronounced as "rounded binary". Regarding these "new observations" I want to state:

A two-reprise design is not an immediate equivalent of a binary form! Repetitions signs neither make or alter a true formal design!

If the form is ABA, and sections BA are repeated together, this does not mean that it is a rounded binary form! The latter has a general AB design, where section B contains an underdeveloped recapitulation of section A. As a compromise, this could be visually shown as (A) (Ba) – two main parts.

These sections may be repeated as in A://Ba:// and that would be a two-reprise rounded binary form.

Therefore A: // BA: // is not a binary form but a ternary form. Whether A is absolutely the same or varied (A') – it makes no difference. In some instances, where only sections BA are repeated together but the initial A is not, some musicians call this "a five-part ternary form" ABABA, which could be written also as

A // BA://

I was advised that the "new" concept (about the two-reprise form being an equivalent of binary) was typical for American music theory, but this statement proved false. As soon as you open the textbook of Kostka/Payne, you will realize that they explain clearly the rounded binary as a binary form with an underdeveloped recapitulation, and label it as A B 1/2A. Curiously enough, I was taught the same concept in my student years in Europe, only that my teachers labeled it as AB and explained to us that a small recapitulation is attached part B but it does not make a stand-alone musical thought. This international coincidence in teaching methods means that there is some widely-spread practice of teaching this form with the clear understanding that repetitions signs do not alter the true formal design. Notice the term "a two-reprise ternary form" is used a few times in the Kostka/Payne book. It reveals the fact that parts 2 and 3 of a ternary form are repeated together.

Moreover, the thin line between small ternary and small rounded binary is the size of the recapitulation per se. If it is a phrase (even an extended one) - the form is rounded binary, for that phrase is attached to section B and makes a portion of it. If it is an equivalent of a one-part form: a self-standing period or phrase group or sentence of Schoenberg which has at least two phrases – the form is ternary, because part three stands alone as a complete musical thought. But that is not all - even if section A were twice as big (such as a double period or a repeated period, for example) – versus a recapitulation that is only a simple period – the form is still ternary.

Thank you for your attention.

Best regards,

Dimitar Ninov

Texas State University

San Marcos, Texas 

 

 

 

Jason Yust

Dimitar,

This is your opinion but you should recognize that some of the issues you raise have been debated for hundreds of years--specifically the binary vs. ternary issue. Some people feel very strongly one way or another. In my opinion the whole dispute is a canard. What do you mean by ternary? The distinction has to be defined. If ternary means "in three parts" then group two of the parts and you have a binary form.

The issue of reprises is not really a matter of dispute. If the reprises are there, they mark a significant formal division (could it be otherwise?). However, in no case I know of does anyone require the notated repeats to define a particular formal type.

Dimitar Ninov

Dear Jason,

Thank you for your input. Please read as follows.

1. "This is your opinion..." No, Jason - it is not only mine. As I mentioned in my letter, that is the opinion of a number of American and European scholars who did not necessarily know one another but came to the same conclusion. Reference: Tonal Harmony, 7th ed., pp 321-326.

2. "What do you mean by ternary?"...If ternary means "in three parts" then group two of the parts and you have a binary form." Following your logic, you may group the refrain of a rondo with a subsequent episode and call that a single part. Is this how you analyze music? I do not believe so. We are talking about a small form here (small binary) which has two distinguishable basic formal parts that are fully fledged and balance or unbalance each other.

3. "If the reprises are there, they mark a significant formal division (could it be otherwise?)"
No, Jason - not always do the repetition signs mark a formal division. Do not confuse organic formal division with mechanical sectional division. Sometimes they coincide, but sometimes - they do not.

Reference - a quote from Kostka/Pane, 7th ed., p. 322, concerning the first part of a Haydn Minuet: "The form is two-reprise sectional TERNARY - A://BA://. Yes, the small ternary form is a part of the compound ternary form of the Minuet. The visual sections are two, the organic division is in three parts, with the recapitulation (part A) being a stand-alone, fully fledged completion of the piece. It is very unfortunate that, as soon as some authors see the above scheme (commented in Kostka/Payne as two-reprise ternary), they announce it as "rounded binary". This is why I referred to "visual versus aural assessment". Let me repeat one more time something which is not only my opinion: sometimes there is a difference between organic formal division and mechanical sectional division. The term "two-reprise" is mechanical an it does not reveal the form per se. This is why my teachers did not emphasize it at all. At best, they might mention something like "a five-part ternary", implying that parts B and A are repeated, while the exposition is not, for example.

It does not matter if the repeats are notated physically or shown as double bar lines with dots; - when you do not have a score you do not know how the excerpt was notated.

4. A simple ternary form is one which has three basic parts, the outer ones typically being equivalents of a one-part form (period, phrase group, sentence, etc) and represent stand-alone complete musical thoughts. By exception, sometimes we may observe a simple ternary form made by phrases alone, the outer ones being independent phrases ending with a PAC (Mozart, the famous theme from the French song on which he composed 12 variations, or the song O Tanenbaum).

Thank you,

Dimitar

John McKay

Dimitar--

I certainly agree there can be advantages to the kinds of distinctions you are drawing.  However, while there are many theorists who agree with you, Jason is also correct that this is a matter of convention and definition, not some sort of musical "law."

And Jason is absolutely correct that historically these debates have raged with learned people on both sides.  The idea you claim to be "new" is actually a VERY old one, arguably with a stronger historical pedigree in terminology.

Perhaps most famously, Reicha's conception of sonata form (generally acknowledged as one of the first detailed descriptions of the form) was labeled the "Grande Coupe Binaire," emphasizing the binary nature of the structure, even as he delineated the second section into two primary parts.  And that was not a small "rounded binary" with an underdeveloped "recapitulation" as you put it -- he was describing full-blown sonata form.

Whether or not you agree that that description is still a useful one within theoretical systems today, it was a fundamental element of how such structures were perceived historically by many theorists and composers.  I would also add that your description of an "underdeveloped recapitulation" is a bit anachronistic in terms of the development of form, since sonata form (with a "full" recap) was partly developed from a tradition of smaller binary (or "two-reprise," if you prefer) structures.

Lastly, do note that while you credit Jason with placing perhaps undue emphasis in formal division on repeat signs, you are inserting a particular emphasis in formal division on the "recapitulatory moment" (often not marked in score at all) which you view as the beginning of the second "A" (or "a" or "1/2A" or whatever).  But that recapitulatory moment was often quite varied in early to mid 18th century structures, sometimes with a clearly demarcated return to "A" in theme, other times with only hints of it, sometimes with a strong formal division at that point (with phrase demarcation, prior cadence, etc.), sometimes not.  Insisting that ALL such two-reprise structures must fit EITHER "AB" or "ABa" (with an implicitly well-defined "recapitulatory moment") can also overlook significant complexities in how these structures were actually composed.  But creating a strict division between "binary" and "ternary" as you suggest can force analysts into false dilemmas when dealing with actual literature.

Again, this isn't an argument against your distinctions as being a useful convention in modern theory -- it's just noting that your argument is not the only way to classify things.

Dimitar Ninov

Dear John,

I appreciate your information and I agree with some things. I will begin with your last sentence.

"Again, this isn't an argument against your distinctions as being a useful convention in modern theory -- it's just noting that your argument is not the only way to classify things."

 – Thank you for the complement, John! I agree with this (:), and my posting may be thought of as a strongly expressed belief that "two reprise is not always a binary form" – even if there are arguments about where the line of distinction lies in modern theory or whether there is a line at all...

"Whether or not you agree that that description is still a useful one within theoretical systems today, it was a fundamental element of how such structures were perceived historically by many theorists and composers.  I would also add that your description of an "underdeveloped recapitulation" is a bit anachronistic in terms of the development of form, since sonata form (with a "full" recap) was partly developed from a tradition of smaller binary (or "two-reprise," if you prefer) structures."

– On the contrary, a possible contemporary interpretation of a typical classical sonata form as "Grande Coupe Binaire," is anachronistic; today the prevailing opinion is that the sonata form is a compound ternary form of a special type, and I share this opinion.

Yes, sonata-allegro may have arisen from a binary structure in some early forms wherein development and recapitulation fused intrinsically into one section because of a lack of clear thematic recapitulation, distinct caesuras, dominant pedal preparations, etc. In this sense it may have stemmed from the so-called Baroque or Old binary form (such as the ones in many Bach keyboard works where the second section is twice as big).

But the above forms are not "rounded binary". The latter term is relatively new and it hints to a clearly distinguishable thematic recapitulation. Since this recapitulation is not self-sufficient, it represents a natural part of section B. In other words, rounded binary is typically more symmetrical and smaller than the old Baroque binary form (from which the old sonata form may have arisen), and it may look like the following scheme (with more possible variations):

A	B
ab	ca' or cb'

"But that recapitulatory moment was often quite varied in early to mid 18th century structures, sometimes with a clearly demarcated return to "A" in theme, other times with only hints of it, sometimes with a strong formal division at that point (with phrase demarcation, prior cadence, etc.), sometimes not."

– Exactly, as I mentioned in my previous comment. If you asked me for my practical opinion, some of the old binary forms are physical ternary forms of an unusual type ABC (meaning that there is no thematic recapitulation but there is a third part), where B and C are separated with a PAC. However, tradition in terminology here is so strong, that I had to relinquish my attempt to fight against wind mills...

The bottom line is that "binary" in principle (including Baroque binary) is one thing, while "rounded binary" is a result of perception which developed later, in comparison with a small ternary form which has a clearly distinguishable thematic recapitulation. When this thematic recapitulation is not fully developed, it feels like being a part of section B, hence the term "rounded binary" (the form is rounded with a partial reprise of the initial A (typically stating its first phrase, but sometimes its second).

Thus, in Kostka/Payne they teach students in a way which I support and share (on many other topics I criticise this book but I think that the topics of cadence and form are decently presented). Paradoxically, all the newer books I have reviewed mechanically declare any two-reprise form as a rounded binary. I have other colleagues at Texas State who are astonished with this concept. Fortunately, many schools still use the K/P book and I do not have to excuse myself for supporting some portions of it.

I recognize there is a theoretical debate which is still going on. And it is all good that we can express our positions. This also comes to the problem of teaching. While I am willing to present different opinions before my students and gently guide them through those labyrinths, stating my opinion as well and supporting it with that of some authors, there are narrow-minded teachers all over the place who are so strict that they are ready to penalize a student for not agreeing fully with a "new trend" or "an exact definition" or a chart. I am sure that defeats one of the main purpose of higher education: to teach students to think critically; to provoke them creatively to ask questions, to discuss, to argue, and to formulate their own definitions. 

Thanks again.

Best regards,

Dimitar