Music Theory on the Radio: Excavating Hans Keller’s Functional Analysis

Society for Music Theory 2017 • Arlington, Virginia • November 5, 2017

William O’Hara • williamevanohara@gmail.com

On the evening of September 7th, 1957, listeners to the BBC Radio’s Third Programme were treated
to a unique broadcast of Mozart’s String Quartet in D Minor, K. 421. The performance by the Aeolian
String Quartet was punctuated between movements by several minutes of additional music, composed
and arranged by music critic Hans Keller. The broadcast was the public’s first exposure to what Keller
called “wordless functional analysis” (abbreviated “FA”). Or as he more vividly described it: “the
musical analysis of music.”1
Keller’s broadcast began with the following brief announcement: “The Unity of Contrasting
Themes, an experiment in functional analysis, by Hans Keller. The Aeolian Quartet play Mozart’s
String Quartet in D Minor (K. 421) and analytic interludes between the movements designed to show
how the contrasting themes and movements hang together. The analysis is entirely wordless. It
consists of a continuous score, except for a three-minute silence, for the recreation of the listener,
after the unity between the slow movement and the minuet has been shown. Mr. Keller calls his
method “functional analysis” because instead of descriptively dissecting a piece of music, it is intended
to isolate the unifying functions of the organism that is a living work of art. The programme begins
with the complete first movement and ends with the complete last.”2
After the announcer’s brief introduction, the form of the broadcast went something like Table 1
on your handout. The first two movements were played in their entirety, each followed by an analytical
interlude. The third movement closed the first half of the broadcast, and was followed by a three-
minute intermission, during which the audience was asked to reflect on what they had just heard. After
the brief intermission came the most ambitious of the analytical interludes, which seeks to tie the third
movement—the Minuet—to the already-heard first movement, and the upcoming fourth.
The broadcast seems to have been a moderate success. It received enthusiastic responses from
listeners, who said they were intrigued by Keller’s premise, although many admitted that they didn’t
fully understand what he was trying to do. Reasons cited by listeners for their confusion included their
own lack of prior musical knowledge, the absence of verbal or sonic signposts to differentiate Keller’s
analytical contributions from Mozart’s original,3 and even sleepiness due to the late hour of the
broadcast: 10 o’clock in the evening.4 Keller himself made more grandiose claims: “Reactions to my
first wordless FA,” he wrote in a letter to The Musical Review on December 9, 1957,
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 2

Show that proportionately, my appreciative audience is pretty evenly distributed among

composers, teachers, practical musicians, musicologists, critics, amateur musicians, and music
lovers. So far, the genuine success of the method … has in fact proved immeasurably wider than
I hoped, but I should be the last to deny that it is too early to point to a victory of musical over
unmusical analysis. We shall see.5
The response, however, was not universally positive. Eric Blom, music critic for The Observer, hated
the repetitiveness of the analysis, and complained: “If anyone succeeds in making me hate Mozart’s
music, it will be Hans Keller’s boast to have done so.”6 And a letter to the editor of The Music Review
wonders who actually benefits from Functional Analysis, and questions Keller’s claim that FA is accessible
to non-experts.
Keller’s method of Functional Analysis is one of the most famous aspects of his musical output.
Yet while Keller often speaks of functional analysis as a coherent project—writing at one point that
FA constitutes “a body of knowledge which is gradually building up as a result of the method’s musical
fact-finding”—there is little indication of what that knowledge actually is, nor of the precise methods
by which FA exposes it. This is probably due to the fact that Keller’s “musical analyses of music” have
been mostly inaccessible to scholars—Nicholas Cook, in fact, names this specifically as the reason
that he omits any details of Keller’s method from his book.7 Until 2001, only two of the FAs were
available in print, with the rest held in manuscript form by the library at Cambridge University. In
2001, Gerold Gruber published a full edition based on Keller’s manuscripts and notes.8 Scholars now
have the opportunity to study Keller’s scores in detail, and I believe that they have much to offer to
our understanding of twentieth-century tonal theory, and much to teach us about Keller and his
contemporaries. So today in this talk, we will begin the work of analyzing Keller’s functional analyses
in order to test his sometimes grandiose claims and lay out clearly the contours of his theory.

I. Functional Analysis and Unity: Deciphering Keller’s Theory

Keller’s descriptions of his own project are scattered throughout the popular and academic press
over the late 1950s and early 60s. While his comments on FA are short and frequently cryptic, when
read together they begin to circle around a few themes. Keller’s project is animated by two central
ideas, which are laid out within the first two pages of his earliest major essay on musical analysis: a
chapter on Mozart’s string quartets in a collection called The Mozart Companion, published in 1956. The
first of these ideas is the distinction between the description of music (for which Keller has nothing but
contempt) and the analysis of it; this distinction leads to his quest for a purely musical, non-textual
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 3

form of criticism. Keller lays out his opinions on analysis and criticism in no uncertain terms: “What
usually goes by the name of analysis,” he writes,
is nothing of the sort. Most critics have never grasped the essential distinction between analysis
and description. Description gives a verbal account of what you hear, and is essentially
unnecessary. Can anyone seriously suggest that a music-lover has to be told that a contrasting
theme is a contrasting theme? Verbal or symbolic analysis shows, on the other hand, the elements
of what you hear.9
By “description,” it seems that Keller means the kind of narratively driven analysis most closely
identified with critics like Donald Tovey. Keller believes that descriptions of music are redundant, and
he yearns for a method that will expose the way in which tonal masterworks operate; hence, the title
“Functional Analysis.”
The second central idea is an aesthetic one: the notion that “a great piece [of music] grows from
an all-embracing idea.”10 This latter notion is familiar from many of the twentieth century’s most
famous theories of tonal music, such as those of Schenker and Schoenberg. Keller’s “all-embracing
idea” bears at least a passing resemblance to Heinrich Schenker’s Ursatz or “fundamental structure,”
and is also closely related to Arnold Schoenberg’s Grundgestalt or “basic idea,” and Rudolph Reti’s
“motif.” Keller himself names these theorists as three of the four primarily influences on his work:
Reti, Keller writes, “exaggerates the melodic aspect” of music analysis; Schenker, he says, goes the
opposite direction by emphasizing harmony over melodic and rhythmic features. Arnold Schoenberg
is his Goldilocks, and gets it just right: Keller believes that his music and writings are indispensable
for any music analyst. Finally, Keller’s fourth, non-theoretical influence is his former violin teacher
Oskar Adler. Keller writes that Adler’s “uniquely organic and motif-conscious way of playing taught
[him] more about the essentials of chamber-musical forms and textures than any analytical teacher
could possibly have done.”11
Keller also has a significant non-musical influence, whom he mentions only implicitly: Sigmund
Freud. Along with his musical background, Keller was an enthusiastic student of Freudian
psychoanalysis. His first professional publication, in 1946, was an essay entitled “Male Psychology,”
which argued against the typical Freudian emphasis on masculine normality, and criticized the clinical
tendency to pathologize women’s experiences.12 Psychodynamic theories influenced some of his early
musical writings as well, such as “A Slip of Mozart’s: Its Analytical Significance,” which reads a missing
accidental in the manuscript of the Overture to Le nozze di Figaro as evidence of Mozart’s
preoccupation with the dominant-key secondary theme.13 His distinction between the manifest diversity
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 4

of the musical surface and the latent unity that underlies it is taken precisely from Freud’s identification
of “manifesten und latenten Trauminhalt [dream content]” in The Interpretation of Dreams. Keller uses the
very clear English cognates found in A.A. Brill’s 1913 translation, but it’s safe to say that he would
have known the original German, since he grew up in Vienna.14 In order to explain how even painful
or frightening dreams can still engage in wish fulfillment (as his previous chapter argued), Freud argues
that the images, emotions, and narrative elements of a dream—the manifest content—may be
“distortions” or transformations of hidden—or latent—content. Interpreting the meaning of a dream
thus becomes a process of decoding the meanings that are hidden beneath the dream’s disjunct or
impressionistic surface. By borrowing these terms, Keller casts his theory of musical unity as a musical
version of Freud’s therapeutic method: presented with the musical surface, the analyst traces common
elements throughout the work, revealing hidden meanings that elude the average listener, and are
perhaps even inaccessible to the composer.15 This hidden knowledge, for Keller, underlies the
connections between music that would become central to his first Wordless Functional Analysis: both
motivic/thematic similarities between movements, and also the compositional tendency for motives or
phrases in disconnected movements to complement and complete one another.

II. Keller’s FA1: Mozart’s String Quartet in D Minor, K.421

So, we have two concerns to keep in mind as we approach Keller’s method of functional analysis:
his interest in thematic unity across movements of a single work, and his disdain for verbal descriptions
of music. In the rest of this talk, I’d like to explore how these interests play out in his most emblematic
Functional Analysis. FA number 1, the inaugural broadcast, was written to be filled in between the
movements of Mozart’s D Minor String Quartet, K. 421.
Keller’s first analytical score attempts to demonstrate two things: how the different themes within
a movement—the first, third, and fourth movements in particular—are compositionally related to one
another; and how the same motivic ideas echo across the string quartet’s four different movements. To
accomplish this, Keller frequently works at two different levels of analysis. First, he illustrates how the
piece should be heard in a linear fashion, as a series of transformations from one theme to the next.
Second, with his Freudian analyst’s cap firmly on his head, he reminds listeners to keep the whole
quartet in mind by periodically surfacing themes from different movements, whether they have already
been heard, or are yet to arrive.
First, we’ll deal with the disposition of themes within the first movement. One of the principal
challenges facing the composer of a minor mode sonata is how to treat the second theme when it
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 5

comes back in the recapitulation: leave it in major, as it originally appeared, or re-write it into a minor
key? In the first movement of K. 421, Mozart chose to re-write the F major Secondary theme into D
minor for the recap. In fact, he altered it even more than necessary, testing the limits by which it can
even be called the same theme. In Keller’s hands, this recomposition of the exposition’s secondary
theme—which I’ve labeled SMAJ—into minor becomes an analytical tool: Keller casts this “third
theme” (which I will label SMIN, for “Secondary Theme, minor version”) as an equal player alongside
the other two. As we see here, he uses it as the linchpin in a transformational process by which the
Primary’s rhythmic, melodic, and harmonic contours are connected to the exposition’s SMAJ theme, by
way of the recap’s SMIN.

In other words, the piece develops “out of order.” Here, we see reduced versions of all three first
movement themes. The thick black arrow across the top of the figure indicates the linear progression
through the piece. But, as we will see, Keller’s Functional Analysis argues that they should be
considered in a different order: Primary Theme first, moving into the Secondary Minor, and finally
the Secondary Major.
Let’s turn to Keller’s analytical score, in your handout. Section A1, which is played after the first
movement, begins with two statements of the quartet’s opening phrase, each time with the fourth
measure omitted. Next, he instructs the first violinist to “remove violin and bow.” She begins to clap
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 6

on the downbeat in m. 11, replacing the cello’s bass line as it goes silent, and helping to emphasize the
off-beat rhythm being played by the second violin and viola.
Here, in the absence of anything but a rhythmic accompaniment, we can hear Keller’s first
transformation: the inner strings switch to sixteenth notes in m. 13, audibly illustrating the similarity
between the guitar-like figure that accompanies the first theme [SING/PLAY], and the more active
sixteenth notes that underpin the second theme [SING/PLAY]. In Keller’s version, however, we
remain in D Minor: we are about to hear the version of the second theme found in the recapitulation,
not the exposition. In m. 14, the claps end, giving way again to the cello. The first violinist must quickly
take up her instrument to play the minor version of the second theme. After four measures, Keller
again leaves us hanging on the dominant in measure 18, with an incomplete cadential 6/4.
The sparse texture that follows—in which only one instrument plays—is a hallmark of Keller’s
analytical style, and appears again and again throughout his fifteen FAs. In m. 19, he introduces his
most characteristic analytical technique: the repetition and gradual transformation of short melodic
fragments. Here, Keller breaks down what we have just heard. The violinist plays an obvious fragment
of SMAJ, twice. Then, in measures 21-23, we hear two altered versions of this fragment, emphasizing
the interval from D to F and the interplay of D and C#. The latter fragment then expands the D-F
leap from a third to a tenth, repeating again for emphasis. Measure 24 starts to put these pieces
together by bringing us back to the first theme. This version of the theme is embellished by a
portamento on the D-string: Keller wants us to hear the physical connection between the two registers,
as the violinist literally drags D upwards by a minor tenth.
After yet again playing the piece’s opening three measures, Keller brings about the final analytical
point of this segment: in measures 28-29, he audibly compares the incipit of SMIN (which we have just
heard several times) with the closing figure of the Primary theme. This leads us to the transition, the
end of which is then compared back to SMIN. These motivic comparisons end in m. 43, in which a
descent through scales degrees 3 and 2 sets up F Major. The SMAJ theme follows: here, we are meant
finally to realize that this major version of the theme is the end result of a motivic process that runs
through the whole movement.
Keeping all that in mind, let’s listen to excerpt A1. [PLAY]
From this example, we can see how Keller thinks of complete movements. Through functional
analysis, he wants to illuminate the composer’s background for us, the listener. Mozart’s
“background,” in Keller’s psychological sense, is his pre-compositional activity such as planning or
sketching, or it’s his synoptic view of all his materials at once as he’s writing the piece. And Keller
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 7

wants to make that synoptic view of the piece into our own background, so that we can use it to
interpret the foreground that we hear in a standard performance. By re-arranging and stringing
together the themes as he does, Keller makes the argument that Mozart conceives of the first
movement’s themes as a cohesive unity, which is progressively elaborated as he writes the piece. He
attempts to make this nonlinear temporality of composition into our linear temporality of listening, in
order to convince us of his thematic analysis without using any words.
So, despite unfolding in the completely linear manner of a radio broadcast, Keller’s analysis
actually proposes a different kind of temporality than we usually find in motivic or thematic analysis.
He tracks the development of musical ideas, but those ideas don’t necessarily unfold in the order in
which we would normally hear the piece. Keller’s analytical recomposition in A1 helps us to jump
around the first movement in order to hear its conceptual contours. In the other big example that
we’ll consider in a few minutes, his music drags us around all four movements, forging thematic
connections among each.

III. Motivic Trees, Chains, and Networks

In other words, Keller presents a somewhat different view of motives than Arnold Schoenberg’s
famous concept of “developing variation.” Take, as a contrast, this figure from Scheoenberg’s 1947
essay “Brahms the Progressive,” one of the most famous formulations of the idea of “developing
variation.”16 The figure analyzes the second theme from Brahms’ String Sextet No. 1 in Bb, Op. 18
(1860). There are two motives here, a and b. Each iteration is marked by triangular brackets, and an
ever-rising number (a, a1, a2, and so forth) marking each motive’s successive variation. With this
notation system, Schoenberg not only demonstrates which of the two initial ideas informs any given
fragment, but also tracks each time we hear a new version of that idea.
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 8

We see something similar in an earlier, lesser known treatise: Fritz Cassirer’s Beethoven und die
Gestalt (1925) anticipates Schoenberg’s analysis of Brahms. (Although, as a sidebar, we know that
Schoenberg owned a copy of Cassirer’s book, and began discussing developing variation in 1917).17
In Cassirer’s analysis of Beethoven’s Appassionata Sonata, we again see a musical surface saturated by
a pair of contrasting motives: the descending minor triad of a, and the elaborated upper-neighbor
motion of b.

Cassirer, like Keller, is interested in showing motivic continuity across movements. He is as

economical as possible with his motivic labels. For example, he identifies this moment from the
opening of the development [mm. 65–69] as a variation on the opening measures: the a1, a2, a2
sequence marked “Thema” is precisely the same, it’s just in E major instead of F minor.

Still, by the time we reach the Appassionata’s second movement, we are already in the double digits
when it comes to motivic variations, and the analysis is in danger of becoming phenomenologically
opaque: the simple upper neighbor motion that opens the movement is easy enough to hear, but how
much is our listening really informed by the knowledge that this is the 14th, or 15th, or 16th new
variation on neighboring motion that we’ve heard?
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 9

Keller’s Functional Analysis, by necessity, avoids this aspect of other motivic analyses. Working
in the form of analytical scores, Keller uses no brackets and no labels to make his points. This is fine—
the radio audience can’t hear a bracket anyway! The arguments are delivered in a purely auditory form,
even though I would argue they require the kind of decoding that I am undertaking in this talk. As Keller
shuttles back and forth and plays different melodic fragments next to each other in various
combinations, we have the sense that the fragments he isolates are all connected to one another, but
not necessarily in a direct way. In other words, Keller’s conception of thematic process—of the latent
Freudian unity buried under the “manifest” diversity of the musical surface, ends up unfolding not
through reference to singular, originary forms:

nor through a successive chain of developing variations:

Either of those descriptions might apply to the motivic analyses of Schoenberg and Cassirer. Instead,
Keller conceives of a piece of music as a network of thematic resemblances:
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 10

The Primary theme of the first movement appears first, of course, but it doesn’t seem to have a sense
of primacy in his analysis. As I said, there are no labels to declare that one fragment or another is
motive form a, or b, or a1, and so forth. And even despite his temporal ordering, Keller leaps back
and forth in order to get us to hear a whole series of transformations between and among the motive’s
various appearances throughout the D minor quartet.

In the days before the conference, I decided to cut my second example for time but I have still included it in the
handout, for reference.

IV. Listening to Abstraction

We can draw another distinction between Keller’s method and the primary tradition of motivic
analysis. Consider the notation of Rudolph Reti, shown here. These images are from Reti’s Thematic
Process in Music (1951). They represent his analysis of the motivic resemblance between the first
movement of Beethoven’s Ninth and the second, the scherzo, and the Adagio third movement. (He
goes on through various segments of the fourth, but we’ll leave those out – it’s not a very famous part
of the work anyway…)
First movement: Second movement:
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 11

Third movement:

Reti’s idea of thematic process highlights simple motivic shapes that form a common thread
through multiple themes of a single movement, or among movements of a single piece—a familiar
theme by now. In his analyses, Reti performs visual abstractions on standard notation, using one of
two techniques. He either uses a second staff to show the notes that form his basic pitch cells (as in
the first two movements of the Ninth Symphony), or he uses full-sized notes to represent his basic
motivic shapes while printing the other notes in a smaller font (as he does with the third movement).
The latter aspect of his practice is what has drawn criticism from those skeptical of Reti’s tendency to
search for patterns. In fact, the analysis of the third movement is pretty bad from a theoretical
standpoint: it just cherry-picks notes that aren’t harmonically or metrically significant, in order to
create a shape that’s consistent with the first two movements.
But it is the visuality of Reti’s diagrams that I wish to highlight here, in contrast with Keller. While
the styles of motivic analysis we have seen from Cassirer and Schoenberg highlight surface features,
Reti’s work uses visual signifiers to point to deeper structures below the musical surface. I believe that
his argument here is not so different from Keller’s. Indeed, Reti’s views on musical unity anticipate
Keller’s, and the latter acknowledges this influence. “[The composer],” writes Reti, “strives toward
homogeneity in the inner essence, but at the same time toward variety in the outer appearance.
Therefore he changes the surface but maintains the substance of his shapes.”18 Reti highlights these
“inner essences,” or what Keller might term the background, by means of notation. In a sense, he
does all of the necessary abstraction for us. Keller, on the other hand, shows no notation to his
audience: they only hear his analyses. The audience has to do a lot of the work of active listening. This
kind of engagement was considered entirely appropriate for the BBC’s Third Programme in the late
1950s.
As Jenny Doctor writes,
Functional Analysis was particularly suited to the philosophy and identity of the Third
Programme. … A mixed cultural arts [channel], the Third nightly presented music, drama,
experiments in arts broadcasting, and talks, with no hampering of ‘fixed points’ (such as news or
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 12

time signals). It was intended for culturally aware audiences who would devote their full attention
to listening.19
The audible nature of Keller’s medium thus shapes the way that his analyses work. Even with an
assumed audience of culturally literate and musically engaged listeners, may not be enough merely to
play two themes back to back in order to suggest a connection. Instead, Keller performs minute
transformations that are immediately audible – such as the transformation of the Trio theme from
major to minor. He also repeats and alternates short fragments a number of times, to make sure his
listeners can follow. So, while he intends for his functional analyses to represent and draw out the
“background” of a piece (in much the same way as Reti and Schoenberg expose deeply buried
patterns), Keller does so almost entirely by means of the foreground or surface of the music.
Keller’s functional analyses thus take the form of teaching a listener about the piece, or perhaps
even the form of a conversation among friends. “Hey, listen to this,” he seems to say (albeit without
words!). “Doesn’t it sound like this other thing? Isn’t it interesting how they both outline D – F, then
walk down through C#...” and so forth. “Now listen to this other thing.”
Along with a deeper understanding of thematic processes and an alternate model for motivic
analysis, Hans Keller’s method of functional analysis has one final insight to offer music theorists in
the twenty-first century: his work is a rare example of public music theory. Public musicology is a
growing field, as scholars attempt to reach outside of the academy in new ways, but examples of public
music theory are relatively rare. Keller’s example of a purely musical form of discourse that avoids
engaging in disciplinary debates and eschews technical terminology can inspire us to invent new ways
of doing analysis that might be more accessible to audiences not already well-versed in music theory.
His ideas might also be useful for the analysis of non-notated music, such as pop music or folk
traditions. As Timothy Warner has recently written, for example, Hans Keller’s functional analyses
operate on principles not unlike those used by DJs when they identify songs that can be smoothly
bridged by musical transitions, or by mashup artists who identify common features of two different
recordings.20 Thinking about ways in which we can follow Keller’s lead, and transform the kinds of
music-theoretical insights that are often expressed through complex and specialized visual diagrams,
into insights than can clearly be heard by someone who has never seen an Ursatz or a transformational
network, will sharpen our theorizing of all repertoires, and help music theory continue to develop and
thrive in the 21st century.
In making these two realizations, we return to Keller’s earlier assertion that Functional Analysis
is a method. It is not merely a method carried out by him (for it would seem to have no adherents but
 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 13

him); rather, it is a method for teaching someone how to listen. So, if Functional Analysis is to be a
“body of knowledge,” as Keller calls it—that is, if we are meant to learn anything by listening to his
work—then surely it is how to listen and how to identify structural elements and similarities. Although he
offers us a model of developing variation that is different from Schoenberg’s, it is a model that shows
us directly how developing variation works, and exemplifies the kinds of details that make it audible,
and the connections within and among movements that must be forged in order for it to be
meaningful.

1
See Hans Keller, “The Musical Analysis of Music,” in Hans Keller: Essays on Music, ed.
Christopher Wintle with Bayan Northcott and Irene Samuel (Cambridge and New York: Cambridge
University Press, 1994), 126–128. Originally published in The Listener 58/1497 (December 5, 1957).
2
Hans Keller Archives, Cambridge University Library, Ex 4/1: “Introductory Text to the
Broadcast of Keller’s First Functional Analysis, 7 Sep 1957).” I am grateful to Susi Woodhouse for
transcribing this document for me.
3
Keller originally proposed that a pianist should play his analytical interludes in between the
movements of a string quartet recording, but later jumped at the chance to have the Aeolian Quartet
perform the entire broadcast.
4
BBC Written Archive Centre document “R9/6/69 Audience Research” (reproduced in
Garnham, Hans Keller and the BBC, 38) reports: “Several [listeners] wished they had armed themselves
with a score, or that the programme had been timed earlier, as they found themselves ‘unable to make
the necessary effort for sustained concentration’ so late in the evening.”
5
Hans Keller, “To the Editor of The Musical Review: Functional Analysis,” The Music Review
(Dec 9, 1957), 83–84.
6
Eric Blom, “The New Approach,” The Observer (April 8, 1956), 10.
7
Cook, A Guide to Musical Analysis, 91–92.
8
FA1 was published in issue 22 of the British music magazine The Score (1958), while FA 14
appeared in Music Analysis 1-2 (1985),
9
Keller, “The Chamber Music,” 90–91.
10
Hans Keller, “The Chamber Music,” in The Mozart Companion, ed. H.C. Robbins Landon and
Donald Mitchell (London: Rockcliff, 1956), 90–91.
11
Keller, “The Chamber Music,” 93.
12
Hans Keller, “Male Psychology,” British Journal of Medical Psychology 20 (1946), 384–388.
13
Hans Keller, “A Slip of Mozart’s: Its Analytical Significance,” Tempo 42 (Winter 1956-1957),
12–15
14
Keller, hailing from Vienna, would have been familiar with the original German terms. See
Sigmund Freud, Die Traumdeutung (Leipzig and Vienna: Franz Deuticke, 1906), 104; or its English
version, The Interpretation of Dreams, trans A.A. Brill (New York: The MacMillan Company, 1913), 114.
15
Keller brags, for example, that his functional analysis of Benjamin Britten’s Second String
Quartet successfully taught the composer—a close friend of Keller’s—something about his own
compositional process. “My method,” Keller writes, “confined itself to the composer’s own pre-
compositional thought, partly conscious, partly unconscious. [Britten] had thus learnt a lot about

 O’Hara, “Music Theory on the Radio” • SMT 2017 • Nov 5, 2017 14

himself from my FA of his Second Quartet.” See Keller, “Functional Analysis of Mozart’s G Minor
Quartet,” 73.
16
See Arnold Schoenberg, “Brahms the Progressive,” in Style and Idea, ed. Leo Stein (New
York: The Philosophical Library, 1950), 74.
17
See Arnold Schoenberg, Coherence, Counterpoint, Instrumentation, Instruction in Form, ed. Severine
Neff, trans. Charlotte M. Cross and Severine Neff (Lincoln: University of Nebraska Press, 1993).
18
Rudolph Reti, The Thematic Process in Music (New York: The MacMillan Company, 1951), 13.
19
Jenny Doctor, “’Vital to the Very Survival of Music as a Living Art’: Hans Keller’s Radio
World.” Music & Letters 85/4 (November 2004), 618.
20
Timothy Warner, “Approaches to Analysing Recordings of Popular Music,” Ashgate Research
Companion to Popular Musicology, ed. Derek B. Scott (Farnham, UK: Ashgate, 2009), 138
 Music Theory on the Radio: Excavating Hans Keller’s Functional Analyses
Society for Music Theory • Arlington, Virginia • November 5, 2017
William O’Hara • Gettysburg College • wohara@gettysburg.edu

Table 1: Formal Outline of Mozart, K. 421 with Keller’s Analytical Interludes

Functional Analysis #1: Mozart, String Quartet in D Minor, K. 421 (1957)

As written Added by Keller
I. Allegro
A1: incorporates primary theme and secondary theme
(in both exposition and recapitulation forms) from mvt. I
II. Andante
A2: incorporates themes from mvt. II and mvt. III (minuet),
along with new triple-meter material
III. Menuetto

Interval (3 minutes)
A3: extended interlude, incorporating Mvt. III’s minuet and trio themes,
along with primary themes from mvt. I and mvt. IV.
IV. Allegro ma non troppo [to m. 32]
A4: very brief, based on mvt. IV, first variation [mm. 25 – 32]
IV. Allegro ma non troppo [mm. 25 – 48]
A5: incorporates mvt. IV, first variation, and mvt. I, primary theme
IV. Allegro ma non troppo [mm. 49 to end]

Table 2: Hans Keller’s Functional Analyses1
FA # Composer Piece analyzed Broadcast(s)/Performances Performer(s) Notes
BBC3: Sept. 7, 1957; Dec. 11,
String Quartet in 1957 NDR: Feb. 5, 1958
FA 1 Mozart Aeolian String Quartet Published in The Score 22 (Feb. 1958)
D Minor, K. 421 ISM: ISM Conference, Stratford-
Upon-Avon, Dec. 31, 1957.
String Quartet in BBC3: March 2, 1958 Pro Musica (BBC3) &
FA 2 Beethoven
F Minor, Op. 95 NDR: May 5, 1958 Hamann (NDR) Quartets
Piano Concerto Denis Matthews, piano;
FA3 Mozart in C Major, K. BBC3: Dec. 7, 1958 Chas. Mackerras & The Described in Music Review essays (1956)
503 Goldsbrough Orchestra
DSS: August 1958 (repeated
String Quartet in Dartington (DSS, BBC3)
August 1959);
FA 4 Haydn D Major, Op. & Benthien (NDR) Commisioned by BBC
NDR: Autumn 1959
64, No. 5 Quartets
BBC3: March 11, 1960
String Quartet in NDR: recorded Jan 17, 1959;
FA 5 Haydn F Major, Op. 50, broadcast with FA 4 in Autumn Benthien Quartet Commissioned by NDR
No. 5 1959
String Quartet in
NDR: recorded Jan 16, 1959;
FA 6 Haydn E✠ Major, Op. Hamann Quartet Commissioned by NDR
broadcast date unknown
20, No. 1
String Quartet in
NDR: broadcast on Jan. 21,
FA 7 Haydn D Minor, Op. Hamann Quartet Commissioned by NDR
1959; recording date unknown
76, No. 2

1
Compiled from Keller, “Wordless Functional Analysis—The Second Year and Beyond—I”; Keller, Functional Analysis: The Unity of Contrasting Themes (Frankfurt am
Main: Peter Lang, 2001); Renee Atcherson, “Bibliography,” Hans Keller: A Memorial Symposium, Music Analysis 5/2-3 (1986), 407–440; and archival sources
(Cambridge University Libraries)
 Piano Concerto BBC3: broadcast May 6, 1959; Clifford Curzon, piano;
FA 8 Beethoven in G Major, Op. manuscript dated March 29, Stanley Pope & London Dedicated to Clifford Curzon
58 1959 Symphony Orchestra
FA Piano Sonata in [unknown; performed in concert, not Susan Bradshaw and Keller (1960a, 73) notes that 9b will be
Mozart
9a A Minor, K. 310 radio] Susan McGaw, pianists the piece for piano quintet; never written.
Clarinet Quintet completed March 9, 1961;
FA Thea King, clarinet;
Mozart in A Major, K. Hampton Music Club, March
10 English String Quartet
581 24, 1961; BBC: March 3, 1962
FA String Quartet in 1961 Aldeburgh Festival [ran
Mozart Dartington Quartet Commissioned by Benjamin Britten
11 F Major, K. 590 June 28 – July 9, 1961]
Radcliffe Festival of British
Music, October 23, 1962
String Quartet
FA [earlier broadcast possible; Keller
Britten No. 2 in C
12 (1984) claims that Britten
Major, Op. 36
commissioned FA11 after hearing
FA 12 on the radio.]
Brandenburg
Completed May 5, 1963;
FA Concerto No. 3
J.S. Bach premiered at 1963 Tilford Bach
13 in G Major,
Festival [May-June, in Surrey]
BWV 1048
FA String Quartet in
Mozart BBC: November 4, 1978 Published in Music Analysis 4/1-2 (1985)
14 G Minor, K. 516
FA Piano Quartet in David Fanning, piano, & May 22, 1985 letter from HK to Mischa
Mozart Manchester Univ., May 10, 1985
15 G Minor, K. 478 the Lindsay Quartet Donat
Abbreviations: BBC3 = BBC Radio Third Programme; NDR = Norddeutscher Rundfunk (Hamburg);
DSS = Dartington Summer School (Devonshire, England); ISM = Incorporated Society of Musicians

 Keller’s FA1 (excerpts)

A1: To follow first movement

P

Allegro moderato
° b˙ œ œ œ œ œ#œ œ j
Violin I & j J œ‰ ∑
œ™ œ œ™ # œ œ œ œ™

Violin II &b ‰ ‰ ‰ ‰ œœ
. . œ
‰
œ. # œ. œ. ‰ # œ ∑
œ. œ. œ. œ. œ. œ. œ. œ. œ. . . œ. œ.
Viola Bb ‰ œ œ œ ‰ œ œ. œ. ‰ œ. œ. œ. ‰ œ. œ. œ. ‰ œ. œ. œ. ‰ œ ∑
. . . . . œ. œ.
?
¢ b˙
Violoncello ˙ ˙ ˙ ˙ ∑
˙

P
5 (remove violin & bow)
° b˙ œ œ œ œ œ#œ œ j
& j J œ‰ ∑ ∑
œ™ œ œ™# œ œ œ œ™

&b ‰ ‰
.
‰
‰ œœ ‰
. œ œ œ ‰ ∑ ‰ ‰
œ. œ. œ. œ. œ. œ. œ œ œ
. . . . #
. .œ . # œ. œ. œ. œ. œ. œ. œ. œ. œ.
B b ‰ œ œ œ ‰ œ œ œ ‰ œ œ œ ‰ œ œ œ ‰ œ. œ. œ. ‰ ∑ ‰ œœœ‰ œœœ
. . . . . . . . . . . . œ. œ. œ. . . . . . .
?
¢ b˙ ˙ ˙ ˙ ˙ ∑ ˙ ˙
˙

10 (clap) (take violin)

° b
& ∑ ¿ Œ ¿ Œ ¿ Œ ¿ Œ ∑

&b ‰ ‰œœ ‰ ‰ ‰ ‰ ≈ ≈
œ. œ. œ. . . œ. œ. œ. œ. œ. œ. œ. œ. œ. œ. œ. œ. œ. œ
œ ºœ œ œ œ œ œ œ œº œ œ œ œ
Bb ‰ œ œ œ ‰ œ œ œ ‰ œ œ œ. ‰ œ. œ. œ. ‰ œ. œ. œ. ‰ œ. œ. œ. ≈œœœœœœœ≈ œœœœœœ
. . . . . . . . œ º
? ∑ ∑ ∑
¢ b˙ ˙
2 SMIN
14
° b ∑ Œ œ #œ œ œ œ #œ œ Œ œ œ œœ
& œ J

&b ≈ œ œ œ œ œ œ ≈ œ œ œ œ œ œ ≈ œ œ œ œ œ œ ≈ œ œ œ œ œ œ ≈ œ œ œ œ œ œ ≈ œ œ œ œ œ œ
œ º œ º œ º œ º œ º œ º
B b ≈ œ œ œ œ œ œ œ ≈ œ œ œ œ œ œ ≈ œ œ œ œ œ œ œ ≈ œ œ œ œ œ œ ≈ œ œ œ œ œ œ ≈ œ œ œº œ œ œ œ
œ º œ º œ
? Œ Œ Œ Œ Œ Œ
¢ bœ œ œ œ œ œ
f mf p

SMIN SMIN
17 œ œ œ œ
° bœ œ Œ œ œ bœ œ J J #œ Ó œ #œ œ œnœ #œ œ Œ œ œ œ œ œ
& J J J J

& b ≈ œ œœœœœœ ≈ œ œœœœœœ

º
≈ œ œœœœ
º œœ ≈ œ œœœœ œœ ∑ ∑
º º
Ω œœ
œœœ Ω Ω
B b ≈ œ œœœœœœ
º ≈ œ ≈ ≈ œbœ œœœ œœ ≈#œ œœœœnœ œ ∑ ∑

? œ œ œ
Œ Œ Œ ∑ ∑
¢ bœ œ

Intermediate
Step (IS) IS P P
21
° b #œ œ Œ œ œ œ œ œ œ œ Œ œŒ œŒ œ ˙
& j
œ œ# œ œ œ œ œ œ™# œ œ œ œ œ™ œ
‰ ‰
&b ∑ ∑ ∑ ∑
ÓœœœÓœœœ
Bb ∑ ∑ ∑ ∑ ‰ ‰
œ œ œ
Ó Óœœœ
‰‰‰‰‰‰‰‰
? ∑ ∑ ∑ ∑ ˙ ˙
¢ b
 SMIN P 3
26
° b œ œ œ œ œ#œ œ œ #œ œ œ œ #œ œ Œ œ œ œ œ œ œ œ Œ Œ ‰ œJ
& œ™# œ œ œ œ™ J œœ Ó

&b ‰ ‰ ‰ ‰ ∑ ∑ ∑
œœœ œ# œ œ # œ œ œ
Ó œ œ œ Ó Ó Ó
Bb ‰ œ œ œ ‰ œ œ œ ‰œœ œ ‰ ∑ ∑ ∑
Ó
Ó Ó Ó œœœ
? b ‰˙ ‰ ‰ ‰ ‰ ‰ ‰ ‰ ‰‰ ‰ ‰ ‰‰‰‰ ∑ ∑ ∑
¢ ˙ ˙ ˙

31 ˙ œ
° bœ œ œ œ#œ œ œ™ œ œ ™ #œ œ œ œ ™ J
& œ œ œ œ œ #œ œ Œ J
j œœ
& b ‰ nœ # œ nœ ‰ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ ‰ œ œ œ ‰ œj œ œ ‰ J œ

B b ‰ nœ œ œ ‰ Œ ‰ œ œ #œ œ œ œ œ ‰ #œJ œ œ ‰ œJ œ œ ‰ #œJ œ œ
œœœ œ
n œ œ œ œ œ œ nœ œ œ bœ œ œ
?b Œ Ó ‰ J ‰ J ‰ J ‰ J
¢ ˙ ˙ œ

TR

35 œ œ œ œ#œ œ
° b œ œ œœœœœœ œœ œ œ™ œ œ œ. œ. œ œ œ
& #œ Œ J #œ Œ
Ϫ
œ œœ œœ œ œ #œ œ œ œ œ œ#œ ‰ Œ
&b œ œ œ œ œœœœ œœœ Œ J J
Ϫ
œ
B b #œ ‰ œj nœ#œ œ nœ œ œ œ Œ œ œ™ œ œ œ œ œ œ ‰ Ó
J Ϊ J
œ œ œ œ œ œ œ œ œ# œ œ #œ œ
? ∑ Ó œ
‰J œ œ J‰Œ
¢ b
 TR IS IS
4
39
° b Œ ‰ #œ œ œ œ œ œ Œ œ #œ œ œ œ œ œ Œ œ œ œ œ œ œ œ Œ Ó
& J

&b ∑ ∑ ∑ ≈≈‰ Œ ≈ œœœœœœ

œ
Bb ∑ ∑ ∑ ≈≈‰ Œ ≈œœœœœœœ

? ∑ ∑ ∑ Ó Œ
¢ b œ

SMAJ
43
° b œ nœ œœœ œ œ œ œœœœ
& œ œœœœœ œ Œ #œ œ œ œ œ œ œ œ œ Œ

&b ≈ œ œ œ œ œ œ œ ≈ œ œ œ œ œ œ œ ≈ œ œ œ œ œ œ ≈ œ œ œ œ œ œ ≈ œ œ œ œ œ œ œ ≈ œ œ œ œ œ œ œ
œ œ
Bb ≈œœœœœœœ≈œœœœœœœ ≈œœœœœœœ≈œœœœœœœ ≈œœœœœœ œ ≈œœœœœœœ

? Œ Œ œ Œ œ Œ Œ Œ
¢ bœ œ œ œ

II. Andante
46
° b œ œ œœ Ó j j j
& œœ œ œ œ œ œœ œ Œ œ œœ œ œœ œ Œ 68 œ ‰ œ œ œ œ œ œ œœœ ‰ ‰

6 j‰ j j œj ‰ ‰
& b ≈ œœœœœœ Œ Œ ∑ ∑ 8 œ œ œ™ œ
œ œ
B b ≈ œ œœœœœœ Œ Œ ∑ ∑ 6 œ ‰ œ œ™
8J J œ œJ œJ ‰ ‰

? Œ Ó ∑ ∑ 6 œ‰œ ™ j‰ ‰
¢ bœ 8J Jœ œ™ œ
Play second movement as written.
 A3: To follow third movement
MIN
103
° b4 œ ™ œ 3 œ œ œ œ œ#œ œ œ œ œ œ œ œ œ™ œ œ œ œ
Violin I & 4 4 œ
4 3 œ
Violin II & b4 Œ 4Ó œ™ œ œ œ œ œ Œ œ™ œ œ #œ œ Œ œ™ œ

Viola B b 44 Œ
3
4Ó
œ™ œ œ œ œ œ Œ œ™ œ œ œ Œ œ Œ œ™ œ
œ™ œ ˙™ ˙™ œ œ™ œ # ˙ ™ nœ œ™ œ
Violoncello
? 4 3 Œ Œ
¢ b4 4
TRIO

109
° b œ œ œ™ œ œ œ œ œ œ œ#œ œ œ ## > œ ™ >œ œ ™
Vln. I & œ #œ ˙ œ> #œ ™ œ œ™ œ
>
#
Vln. II & b œ nœ œ œ
Œ #œ ˙™ nœ œ œ ˙ Œ #
œ
Œ Œ
œ #œ œ#œ œ œ œ bœ œ ##
Vla. Bb œ Œ Œ œ œ ˙ Œ œ Œ Œ
? n˙ ™ bœ
Œ
œ™ œ œ œ œ œ œ ## œ
Vc.
¢ b œ ˙ Œ Œ Œ
TRIO - Simplified

115 > Ϫ >

œ ™ > œ ™ >œ œ™ >œ œ >œ >œ
> œ œ œ >˙ ,
° ## œ œ ™ œ œ™ ˙ ,
œ œ œ
b
Vln. I & œ œ œ œ
#
Vln. II &#œ Œ Œ œ œ œ œ Œ Œ ∑ ∑ ∑ ∑ b
œ œ œ œ
œ
Vla. B ## Œ Œ œœ œœ œœ œœ Œ Œ ∑ ∑ ∑ ∑ b
œ œ œ
? ## Œ Œ
Vc.
¢ œ œ Œ Œ ∑ ∑ ∑ ∑ b
 Simplified TRIO - Minor MIN - Simplified
2
122
œ œ #œ œ œ ˙
œ œ œ œ œ œ œ œ œ œ œ ˙
Vln. I &b œ œ œ œ

Simplified TRIO - Minor MIN

130
> œ œ #>œ œ œ >˙
° b œ œ™ œ œ œ œ œ œ œ#œ œ œ œ œ œ
Vln. I & œ œ œ
>

Vln. II &b ∑ ∑ ∑ ∑ Ó œ™ œ œ œ œ œ Œ œ™ œ

Bb ∑ ∑ ∑ ∑ Ó œ™ œ œ œ œ œ Œ œ™ œ
Vla.

œ™ œ ˙™ n˙ ™ œ œ™ œ
Vc.
? ∑ ∑ ∑ Ó Œ
¢ b

137
° b œ œ œ™ œ œ œ œ œ œ œ™ œ œ œ œ œ œ œ#œ œ œ œ™ œ
Vln. I & œ #œ ˙

œ nœ œ
Vln. II & b œ #œ œ
Œ
œ™ œ œ œ
Œ #œ ˙™ nœ œ œ ˙ Œ

œ
œ Œ œ™ œ œ œ Œ
#œ Œ œ
Vla. Bb œ Œ œ#œ œ œ œ bœ œ œ Œ
˙
™ œ ™ œ n˙ ™
? b#˙
nœ bœ œ™ œ œ œ œ œ œ
Vc. Œ Œ œ ˙ Œ
¢

MIN / P MIN / P MIN / P

144
œ œ œ œ#œ œ ˙ œ œ œ œ#œ œ œ œ œ œ#œ Œ 4w
Vln. I &b œ œ ˙ 4
 3
MIN / P P
151
° b œ ˙ œ œ œ#œ ˙ œ
Vln. I & ˙™ ∑ j J
œ™ œ œ™ # œ œ œ œ™

Vln. II &b ∑ ∑ ∑ ‰ ‰ ‰ ‰ œœ
œ. œ. œ. œ. œ. œ. œ. œ. œ. . . œ.
Bb ∑ ∑ ∑ ‰ œœœ‰ œœœ ‰ œ
Vla.
. . . . . . . œ. œ. ‰ œ. œ. œ.

Vc.
? ∑ ∑ ∑ ˙ ˙
¢ b ˙ ˙

156
° bœ œ œ œ#œ œ j ˙
Vln. I & œ œ œ œ œ #œ œ ‰ Œ œ™ œ
j

Vln. II &b ‰ œ #œ œ. ‰
# œ. œ. œ. œ œ œ œ œ œ œ œ
‰ ‰
. . œ. œ. œ. œ. œ. œ.
Vla. B b ‰ œ. œ. œ. ‰ Œ ‰ œ œ #œ ‰ ‰
œ. œ. œ. œ œ. œ. œ. œ. œ. œ.

Vc.
? Œ Ó ˙ ˙
¢ b˙ ˙
œ

159
° b œ œ œ œ œ#œ œ j
Vln. I & œ™ # œ œ œ œ™ J œ œ œ œ œ #œ œ ‰ Œ

Vln. II &b ‰ ‰ œ œ ‰ ‰
. . œ.
œ. œ. œ. œ. # œ. œ. # œ. œ. œ. œ œ œ œ œ œ œ œ
Vla. B b ‰ ‰ œ œ #œ
œ. œ. œ. ‰ œ. œ. œ. ‰ œ. œ. œ. ‰ Œ
œ. œ. œ. œ

Vc.
? Œ Ó
¢ b˙ ˙ ˙ ˙
œ
 P MIN
162 œ™ œ # ˙ ™
3
˙ nœ œ ™ œ n˙ ™
Vc.
?b ˙ ˙ 4 Œ
˙ ˙ ˙ ˙ w
P
170
? b bœ Œ œ™ œ >œ œ œ >œ >œ
Vc. œ> ˙ ™ ˙™ ˙™ ˙™ ˙™ ˙™ ˙™

MIN
180
° bÓ œ™ œ œ œ œ œ œ œ#œ œ œ œ œ œ œ œ œ™ œ œ œ œ œ œ œ™ œ
Vln. I &

&b ∑ Ó œ™ œ œ œ œ #œ œ œ
Vln. II œ Œ œ™ œ œ œ Œ œ ™ œ œ nœ

Vla. Bb ∑ Ó œ™ œ œ œ œ œ Œ œ™ œ œ œ Œ œ Œ œ™ œ œ œ Œ
œ™ œ ˙™ n˙ ™ œ œ™ œ # ˙ ™ nœ œ ™ œ n˙ ™
?
Vc.
¢ bœŒ Œ Œ

187
° bœ œ œ œ œ œ#œ œ œ #œ ˙ œ ™ œ œ œ Œ œ œ Œ 68 œ œ œ œ œ œj
Vln. I & œ

6
Vln. II & b œ Œ #œ ˙™ nœ œ œ ˙ Œ ∑ ∑ 8 ∑

Vla. B b #œ Œ œ œ#œ œ œ œ bœ œ œ
˙ Œ ∑ ∑ 6
8 ∑

? b bœ Œ œ™ œ œ œ œ œ œ 6
Vc.
¢ œ ˙ Œ ∑ ∑ 8 ∑
2
FINALE
194
° b œ ™#œ œ œ œ œ #œ ™ œ œ j ‰ ‰ Œ ™ ‰ ‰ j œ ™#œ œ œ ™ œ œ #œ ™ œ œ j ‰ œ ™ œ
Vln. I & œ œ œ

j j j j j j j
Vln. II &b œ œœ œ œ œœ ‰ ∑ œ œœ œ œ œœ œ

Vla. B b œ™ œ œ œ™ œ œ œ #œ œ ‰ ∑ œ™ œ œ œ™ œ œ œ #œ œ œ
J J J
œ œœ œ œ œœ œ œœ œ œ œœ œ
Vc.
?b J J J ‰ ∑ J J J J
¢

MIN FIN MIN

199
œ œ œ œ#œ œ œ ™ j œ™ œ œ œ œ#œ œ œ
Vln. I &b œ œ Œ ‰ ‰ œ œ ™#œ œ J ≈ ‰ Œ ™ ‰ ‰ œ ™ œ œ œ

FIN MIN
204
œ ™ œ j œ ™ œ œ ™ œœ . œœ œœœœœ 3
Vln. I &b J ‰ œ #œ #œ ™ œ™ œ™ ‰ ‰ œ ™œ œ œ œ œ œ œ#œ œ œ ‰8

FIN
210
° b3 ‰ ‰ j 6 œ ™#œ œ œ ™ œ œ #œ. ™ œ œ j ‰ œ œ œJ ‰ œ œ œJ ‰ œ œ œ ™ œ ‰ œ™ œ
Vln. I &8 œ 8 œ J

3 6œ j j j j j
Vln. II & b8 ∑ 8 œœ œ œ œœ œ #œ ™ œ œ nœ ™ œ œ œ œœ ‰

Vla. B b83 ∑ 6 nœ ™ œ œ œ™ œ œ œ #œ œ œ œ #œ œ™ œ œ #œ #œ#œ ‰

8 J J J J
œ œœ œ œ œœ œ œ
? b83 ∑ 6 J J J J œ œ œ œ ‰
Vc.
¢ 8 J J œ J œ

MIN
215
œ œ œ œ œ#œ œ œ œ œ œ œ œ ‰ œ œ œ œ œ ‰ œ œ œ œ œ ‰ 9 Œ ™ Œ ™ Œ œ œ 6
Vln. I b
& œ 8 8
 FIN
3
220 œœ œ œœœ œœ
° b 6 œJ ‰ œ œ œJ ‰ œ œ œ ™ œ ‰ j œ™ œ œ œ™ œ œ œ ™ œ œ œ ‰ J‰ J‰
Vln. I & 8 J œ J
6 j
Vln. II & b8 ∑ ∑ œ œœ œ œ œ œ œJ nœ bœ œ j
J J J #œ
Vla. B b 68 ∑ ∑ #œ ™ œ œ œ ™ œ#œ œ œ œ œJ œ œ œ j
J
J œ
? b 68 ∑ ∑ œ œœ œ œ œ œ #œ œ œ œ j
Vc. J J J J J œ
¢

MIN
225 Ϫ
Vln. I
° b
& œ ‰ œj œ ™ #œ œ œ ™ œ œ #œ ™ œ œ œj ‰ œj œ ™ #œ œ
J œ™ # œ œ

Vln. II & b œ™ œ œ œ ‰ ∑ ∑ ∑

Vla. Bb œ™ œ œ j ‰ ‰ ∑ ∑ ∑
œ
Vc.
? œ ‰ ∑ ∑ ∑
¢ bœ J œ

229 etc.
° b 4 œ œ œ œ œ #œ
Vln. I & œ™ # œ œ œ™ œ œ 4 œ™ # œ œ œ œ™ J

4‰
Vln. II &b ∑ 4 ‰ œ œ
. . œ.
‰
œ. # œ. œ.
œ. œ. œ.
Bb ∑ 4‰
Vla. 4 œ œ œ ‰ œ. œ. œ. ‰ œ. œ. œ.
. . .
? ™ 4
Vc.
¢ bœ œ™ 4˙ ˙ ˙