Schoenberg's Serial Odyssey

Author(s): Silvina Milstein

Source: Music & Letters , Feb., 1992, Vol. 73, No. 1 (Feb., 1992), pp. 62-74
Published by: Oxford University Press

Stable URL: https://www.jstor.org/stable/736147

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/736147?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access to
Music & Letters

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
 SCHOENBERG'S SERIAL ODYSSEY

BY SILVINA MILSTEIN

A RECENT BOOK by Ethan Haimo' explores the vicissitudes of Schoenberg's journey of 'com-
positional discovery' from incipient dodecaphonic thought around the time of the outbreak
of the First World War to the consolidation of many features of twelve-note technique which
characterize the music written after 1928; and it offers a comprehensive summary and revi-
sion of much that has been written on the subject in the last four decades. Although the
discussion focuses on the later atonal and early twelve-note works, Haimo's approach is
teleological, a constant search for those features which characterize Schoenberg's composi-
tions of the late 1930s. By combining biographical information, commentary on the sketch
material, detailed chronological studies, analysis of the music, and theoretical speculation,
he attempts an ambitious portrayal of the composer at work.
The protagonist of the 'serial odyssey' emerges as a 'secretive' man, who calculates his
manoeuvres in order to prevent his music from falling into oblivion: '[He covers] his tracks,
emphasizing in his public statements the historical foundations of his new approach while
minimizing the undeniable systematization of his compositional technique' (p.2).2 Schoen-
berg's statements on this subject are consistent and unequivocal. In his essay 'Schoenberg's
Tone Rows', he writes:

At the very beginning, when I used for the first time rows of twelve tones in the fall of 1921, I foresaw the confu-
sion which would arise in case I were to make publicly known this method. Consequently I was silent for nearly
two years. And when I gathered about twenty of my pupils together to explain to them the new method in 1923,
I did it because I was afraid to be taken as an imitator of Hauer, who, at this time, published his Vom Melos
zur Pauke.3 I could show that I was on the way to this method for more than ten years and could prove so by ex-
amples of works written during this time. But, at the same time, already I did not call it a 'system' but a
'method', and considered it as a tool of composition but not as a theory. And therefore I concluded my explana-
tion with the sentence: 'You use the row and compose as you had done it previously'. That means: 'Use the same
kind of form or expression, the same themes, melodies, sounds, rhythms as you used before'.4

Even if Schoenberg's claims that the new method constituted only a means to fortify tradi-
tional musical logic in the absence of tonality were only a smoke-screen, and even if we
accept that the increasing degree of consistency and pervasiveness of twelve-note thought
eventually caused a rupture which forced the music beyond the boundaries of traditional
musical language, we could still not escape being shaken by the disruptive tension between
the old and the new. Ultimately my criticism of Haimo's book centres not on its standpoint
in relation to such matters but mainly on the fact that it understates the striking synthesis
between traditional and innovative modes of thought in this music.
Haimo often sees a line of causality between not necessarily related matters such as the
reception of Schoenberg's music and his silence on questions of compositional systematiza-
tion:

One can hardly blame Schoenberg for engaging in such self-protective tactics . . . To have gone into details, to
have revealed his compositional secrets would have opened himself to even more attacks and could have
deflected attention permanently away from his music. (p. 2)

S Schoenberg's Serial Odyssey: the Evolution of his Twelve-Tone Method, 1914-1928. By Ethan Haimo. pp. xi
+ 192. (Clarendon Press, Oxford, 1990, ?30. ISBN 0-19-315260-6.)
2 Henceforth, page references in parenthesis are to Haimo, op. cit.
3 'Schoenberg most likely means Vom Wesem des Musikalischen, which was published in 1923, rather than
Vom Melos zur Pauke, which appeared in 1925-and was dedicated to Schoenberg', editor's note, Style and Idea:
Selected Writings of Arnold Schoenberg, ed. Leonard Stein, with translations by Leo Black, London, 1975, p. 523.
4 'Schoenberg's Tone Rows', Style and Idea, p. 213.

62

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
The tendency to undermine Schoenberg's explanations is combined with an unquestioning
trust in current analytical methods:

Over the past thirty years, real progress has been made in the understanding of this music. .. Given the power-
ful theoretical background provided by Babbitt and the possibilities opened by the availability of Schoenberg's
sketches and manuscripts, we are now able to provide answers to substantive musical questions, both stylistic
and analytic. (p. 3)

Despite the powerful analytical tools available to us, our time seems to me poorly equipped
for coming to terms with the intellectual motivations of Schoenberg's music, for the Second
World War has effaced points of references and vital lines of historical continuity, which we
can only tentatively attempt to reconstruct.
Although Haimo is critical of some of the work of the second generation of American
analysts, his views are closer to them than to those of Babbitt, Lewin and Perle, the main
differences lying in their motivation and scope. Babbitt's writings have provided theoretical
foundations for dealing with the systematic aspects in Schoenberg's music, but perhaps more
fundamentally they constitute the basis for a theory of the twelve-note system which
American composers have been exploring 'since Schoenberg'. In 'Set Structure as a Com-
positional Determinant', the foundation-stone of theoretical studies on Schoenberg's twelve-
note music, Babbitt stated his position unambiguously:

Much of the discussion will be motivated by and centred about the initial measures . .. of the third movement
of the Schoenberg Fourth Quartet, the object of the observations mentioned above, not only-or even
primarily -for the purpose of analysing this excerpt, but in order to infer similar and further extensions of the
properties and methods it exhibits.5

Neither Babbitt nor Lewin nor any of the other American scholars working in the 1950s and
'60s attempted to put forward a comprehensive explanation of Schoenberg's music but,
rather, pointed towards the constructional potential of certain of its systematic aspects and
proposed new extensions to them.
In recent years the character of the exegetical enterprise has undergone profound
changes.6 While occasionally displaying genuine historico-analytical concerns, this work
generally exhibits a marked preoccupation with matters of mainly theoretical interest.
Schoenberg's music has proved an inexhaustible well of compositional procedures, which
have been generalized, formalized, and turned into the corpus of twelve-note theory. Com-
mon to the new generation of American scholars of twelve-note music is a preoccupation
with demonstrating that behind the immediate fact that 'Schoenberg's twelve-tone composi-
tions display surface features that strikingly invoke large-scale tonal forms' lies a much
deeper truth: 'despite surface similarities to tonal idioms, Schoenberg's twelve-tone music
represents a distinctly different form of musical life'.7
Schoenberg's Serial Odyssey characteristically emphasizes the high degree of refinement
and the pervasiveness of the systematic mode of thought. Haimo singles out DieJakobsleiter
and other incomplete compositions of the 1914-18 period as the initial steps in the voyage
towards ever-increasing systematization, which culminated in the Variations for Orchestra.
For Haimo,

the techniques developed in the period 1914-28 would remain the basis of Schoenberg's twelve-tone method
through all of the compositions after Op. 31. This is so because these were not isolated features, but an in-
tegrated compositional idea, one in which phrasing, form, rhythm, metre, developing variation, and harmony
were related to one another by the structure of the referential set. (p. 181)

In contrast, Babbitt never suggests that Schoenberg either achieved or even aimed at total
integration in a systematic sense, restricting his comments to the 'compositional manifesta-
tions of the work's set structure'. 8 For instance, when discussing an example of rhythm in the

5 'Set Structure as a Compositional Determinant', Perspectives on Contemporary Music Theory, ed. Benjamin
Boretz & Edward T. Cone, New York, 1972, p. 129.
6 See the bibliography in Andrew W. Mead, 'The State of Research in Twelve-Tone Music', Music Theory
Spectrum, xi (1989), 40-44, which provides a comprehensive survey of recent American writings.
7 Idem, '"Tonal" Forms in Arnold Schoenberg's Twelve-Tone Music', Music Theory Spectrum, ix (1987), 67.
8 'Set Structure as a Compositional Determinant', p. 139.

63

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
Fourth String Quartet, far from claiming that pitch structure determines rhythm he
presents it as 'a counterexample to the assertion that Schoenberg's rhythm is "unrelated" to
the twelve-tone structure of his pitches'.9
In the course of the last decade, scholarly work on Schoenberg has turned increasingly
positivistic, adopting certain not unproblematical ideas-some prompted by comments by
Schoenberg himself- as icons of a new orthodoxy. The following section examines some of
the limitations and problems of these ideas as they manifest themselves in Haimo's book,
and proposes alternative ways of looking at this music.

Each characteristic was introduced as part of a developing critical process that sought to make the structure of
the set the compositional determinant for all dimensions of the musical fabric. (p. 41)

It is a truism that in his music Schoenberg integrated traditionally derived principles of

musical discourse with the constructional potential of the method, but there are conflicting
views as to the nature and extent of such integration. Boulez and Perle maintain in different
ways that archaic tonal idioms and twelve-note procedures stand in a capricious relation-
ship. Schoenberg's practice, however, has been justified from two radically different angles.
On the one hand, Rufer and Leibowitz consider the twelve-note method itself as a vehicle of
historical continuity. Since their analyses do not explore the relationship between detail and
the whole, they fail to expose the constructional scope of the twelve-note method in Schoen-
berg's conception. On the other hand, Hyde claims that the generative force of detail in
Schoenberg relies exclusively on the twelve-note method itself. Since Hyde regards thematic
and formal elements exclusively as vehicles for twelve-note relations, her work constitutes
more an account of coherence between music and a generating set than an explanation of
musical thought.'"
Although Haimo admits that principles of developing variation operate to some extent in
these compositions, this is not investigated. Nor is he interested in exploring tonal and/or
linear motion in this music. Like Hyde, he views this music mainly in twelve-note terms.
Accordingly, when dealing with the use of a popular tune in the third movement of the Suite
Op. 29, he states:

Fitting a tonal theme into a twelve-tone composition may seem somewhat anomalous, and so it is. But
Schoenberg is not serious: this is but one of many playful, or tongue-in-check references in this composition. It
should not prevent us from recognizing the serious accomplishments of this movement. (p. 131)

Should we not instead recognize here a step towards a synthesis?

Some of the limitations of the systematic approach are evident in Haimo's discussion of
the preliminary drafts for the main theme of the third movement of the Wind Quintet. This
melody, which uses each of the twelve pitch-classes once, displays many interesting con-
structional features which required considerable drafting. Haimo discusses the sequence of
sketches in great detail and concludes enthusiastically:

Here, before our eyes, we see Schoenberg wrestling with a problem, and in solving it finding the key to enter a
new world of compositional possibilities. In Schoenberg's long odyssey, the compositional discoveries of the
third movement of the Quintet mark one of the more decisive stages. (p. 123)

Haimo's account of the journey stops, surprisingly, with the final draft of the theme, symp-
tomatically ignoring a subtly important change Schoenberg made in the full score: the addi-
tion of the bassoon's Eb in bar 7.
As Haimo explains, the manner in which Schoenberg associates sets and derives themes
here is related in an intricate way to the symmetries of the set, indicating a concern for devis-
ing systematic procedures, both to assign the elements of the set to the different polyphonic
constituents and to control continuity. For this reason, it is striking from a theoretical view-
point that, even in the earliest stages of the movement, Schoenberg deviates from what,
according to the sketches, was a laboriously devised scheme. The first phrase of the Adagio

9 Ibid., p. 142.
'0 See, in particular, Martha Hyde, Schoenberg's Twelve-Tone Harmony: the Suite Op. 29 and the Com
tional Sketches, Ann Arbor, 1982.

64

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
unfolds three complete sets plus the final Eb played by the bassoon, which is the first pitch-
class of the following set and which Haimo's account ignores. In this way, the carefully
achieved symmetry is broken. This deviation may indicate that other criteria besides strict
twelve-note ones are in operation, as I attempt to show in the following analysis.
The changes in the three preliminary drafts for this phrase provide evidence for
reconstructing the lines of thought underlying this compositional decision. In the earliest ex-
tant draft, Arnold Schoenberg Institute MS 508, dated 10/5[1923] (reproduced in Haimo,
Ex. 5.10), the lower voice starts with the whole-tone segments, as in the final version, but the
top voice contains pitch-class repetitions. Besides being less accomplished in its construction
than the final scheme, the first version is also less effective in terms of its voice-leading and
motivic treatment. Schoenberg's changes in the two subsequent drafts included in MS 525,
dated 1/6/1923 (Haimo, Ex. 5.12), considerably improve the linear profile of the first draft.
In MS 508 the lower line of the first phrase, bars 1-8, reaches its lowest points at bars 4 and 6
with d and c# respectively, before closing on eb at bars 7-8; while in the second draft in MS
525, the descent to Eb is more convincingly attained by gradual downward motion followed
by a semitonal ascent to Eb:

bar 1 2 4 5 6 8
co GO E G C D Eb

In the first draft the top voice starts on Eb and closes on G, and the bottom one does the
reverse. In the third draft the top voice closes on FO (bar 7), and the bottom one closes on Eb
(bar 8) while the top voice has a rest. The sounding of the Eb against the FO occurs only in
the final version (bars 7-8). Eb and FO perform an important articulative role in the Adagio,
comparable to the function of those pitch-classes and pitch-levels in the atonal works which
'serve to differentiate between sections which return to their starting-point and those which
move away from it'. " Moreover, the lower voice in the final version tightens the number of
intervening shapes by repeating several times the contour of the first four notes. The various
stages in the sketching of this theme indicate a concern for constructional consistency,
motivic tightness, balanced voice-leading and, finally, the establishment of referential
pitch-classes.
The importance of the addition of the Eb at bar 7 becomes apparent when seen in the con-
text of the pitch hierarchy of the movement. Schoenberg gives prominence to certain pitch-
classes by placing them at extreme points of the two principal parts. For instance, in the
opening phrase the Hauptstimme starts on Eb and closes on FO, while Eb is also the final
pitch-class of the Nebenstimme (bars 1-8). The Hauptstimme of the second phrase ends on
c' (bar 15) and that of the following phrase on a' (bar 19), which was also prominent in the
flute at bar 15, c' is also the final pitch-level of the clarinet and oboe Hauptstimme in the
final phrase (bars 20-21), and f '-gL' becomes the point of arrival of the flute's cadential
figure. Thus, these points of arrival and departure in the Hauptstimme - eL", f '-gL", c'
and a'-which are further associated by being within the same tessitura, outline a dimini-
shed seventh. Besides these pitch-levels, only Db receives special emphasis as the final note in
the Nebenstimme of the second phrase (bar 14), the beginning of the Hauptstimme of the
third (bar 15), and finally in the closing flute melody, where it gives fifth support to Gb (bar
21). This prominent linear cadence on gb', which is followed by a long pause in all parts,
recalls the f ' which closes the opening phrase (bars 7-8). The Hauptstimme of the final
phrase of the first subject (bars 22-30) initially expands the registral span of the treble to
close eventually onf ', preparing for theJ ' which opens the first subsidiary subject (bar 34,
bassoon). All these referential pitch-levels, except for the bassoon's EL at bar 8, are further
associated by being within the same tessitura.
Haimo's commentary on the sketches tells us that the melody of the first draft consists of
more than twelve notes, that it does not include all pitch-classes, and finally that the second
version is very close in every other aspect but pitch to the final version. His avoidance of mat-

" Walter & Alexander Goehr, 'Arnold Schonberg's Development towards the Twelve-Note System', Euro-
pean Music in the Twentieth Century, ed. Howard Hartog, London, 1957, p. 91. The authors refer here only
to Erwartung.

65

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
ters of motivic development, voice-leading and tonal motion in his interpretation of these
drafts is consistent with his view of Schoenberg's twelve-note works as autonomous pitch-
structures.
Haimo's treatment of the Suite Op. 29 follows similar lines. For instance, in the discussion
of the Ouvertiire's form he concludes that the choice of transposition level for the inverted
recapitulation of the second subject is determined by a particular property of the set
(p. 125). Yet the property in question (hexachordal content retention) holds not only for the
axis of inversion used (index-number I 1) but also for index-numbers 7 and 9. Even if Haimo
is correct in concluding that the determining criterion was the preservation of hexachordal
content, what makes Schoenberg invert the second subject at this particular index number?
Perhaps this question lies beyond analysis, but if we are prepared to put forward hypotheses
about compositional decision it seems appropriate to consider also matters of pitch hierarchy,
as I have attempted in the following analysis.
In the exposition the general linear pattern of the second-group Hauptstimme is the
movement away from FO via B, while in the recapitulation the reverse process (i.e., the
movement from the pitch-class centre B via FO) is systematically accomplished by virtue of
the properties of twelve-note inversion, as shown in Ex. 1. In the exposition FO-B functions
as the main referential configuration of the second group in terms of points of arrival and
departure in the Hauptstimme, while F-C is the secondary referential dyad. An examina-
tion of the following diagram shows that the dyads C-F and FO-B both invert as interval-
class 5. Since the inversion in the recapitulation takes place across a single inversional
axis, there is only a single set of inversional dyads, that is, each pitch-class has an inversional
counterpart, as follows:

inversion at index number 11

D Co C B Bb A

interval-class 1 3 5 5 3 1

Eb E F FP G GO

Since FO becomes B, F becomes C, and vice versa, the recapitulation reverses the linear
movement of the exposition by reaching FO via B and C via F (Ex. 1). 2
The function of FO as a prominent pitch-class centre in the second group is prepared in
the opening three bars of the Ouvertiire, which contain in embryonic form the harmonic
relations of the movement and the elements of its prolongational syntax.'3 The harmonic
discourse involves the establishment, prolongation and shift of pitch centres which are pitch-
levels or pitch-classes, acting as points of departure, frames or goals of linear gestures within
hierarchically structured lines. In these three bars the pitch-class Eb is predominant, but
FO also functions to a lesser degree as a centre, as shown in Ex. 2.

12 The other two inversions which preserve the hexachordal content constant take place around the following
axes:

inversion at index number 9

B A Ab G FO F

interval-class 1 3 5 5 3 1

Bb C C D Eb E

inversion at index number 7

C B Bb A Ab G

interval-class 1 3 5 5 3 1

Db D bl E F Fk

'3 My use of the term 'prolongation' in relation

and n. 22.

66

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
Ex. 1
Exposition
bars 32 37 39 40 41 47 48 49

violin clar4 E6 clar4 piano

Recapitulation
bars 144 149 151 152 153 159 160 161

violinpan

ILL7
cello viola piano
piano pan p

Local hierarchies are determined by the following means:

(a) The grouping of notes into triads: the set divides into two inversionally related hexachords, each of which is
partitioned into two trichords through the vertical spacing of its elements, yielding in both cases a major and
a minor triad related by a minor sixth, as in the upbeat to bar 1:

B minor Db major

P3 Eb G F Bb D B C A GO E F Db
I I I I I
Eb major A minor

(b) The emphasis on pitch-class centres which function both as members of triads and independently of them
(Ex. 2). This is achieved by:

(i) registral disposition: the predominant lines of chordal textures being the highest and lowest;
(ii) the bass line supporting the overall tonal movement in a manner analogous to that of tonality;
(iii) the functional deployment of an implied 5-1 linear progression to support pitch-class centres;
(iv) the borrowing of the bass line idiom of tonal cadence; for example, the bass progression IV-V-I on
strong beats reinforces Eb, while the same progression on weak beats strengthens F#, the secondary
pitch-class centre;
(v) the consistent deployment of the hierarchical relations of metric structure; for example, those
elements which support the establishment of Eb as a pitch-class centre occur on the beat while those
which tend to a lesser extent to reinforce FO are off the beat;
(vi) the functional use of timbre; for example, at bars 1-2 Eb elements are played by clarinets and strings,
FO elements by the piano;
(vii) the functional deployment of textures consisting of contrasting components; for example, at bars
1-2, chord progressions support Eb while arpeggios articulate Fo elements;
(viii) placing pitch-class centres at the extreme points of groupings; for example, in bars 1-3, Eb and FO ap-
pear respectively both as the first and final notes of the top and bass lines.

The three predominantly contrapuntal passages of the first group, which can be classed
as strettos in view of their combinatorial treatment of motivic material," contribute to
articulate the displacement of Eb by FO. In the first stretto (bars 7/ii-14), the predominant
placement of Eb and Bb at the beginning and end, and/or at the highest or lowest points of
rhythmic groupings, occurring in more than one octave, results in the saturation of the tex-
ture by this dyad. The prolongation of Eb is primarily achieved through a process of rhythmic
and textural animation. In the second stretto (bars 17-21), though entries of the motif occur
on seven different pitch-classes, the Eb-Bb pair becomes predominant as a consequence of its
registral placement at the highest points of the Hauptstimme (violin bars 17-18, and E flat
clarinet and clarinet bars 19 & 21). Furthermore, EL is given prominence by the articulation

'4 The characterization of these passages as strettos has previously been made by Hugh Wood in his thematic
analysis of the Ouvertiire included in the miniature score of the Suite published by Universal Edition.

67

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
 Mr~~~~~~~~~~~~~~r I

_xr- _X- , .t )

I ~~~~~~~mI

*
_M-
r- X 3r-
-
40 _137
*
I- r -

I I~~~

_3 _l 0 = I

- IIZ

Co I~~~~~~~J

68~~~~~~~I -

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
of D-Eb as a leading-note relation between the last note of the Hauptstimme of the chordal
section and the first one of the second stretto (bar 17, E flat clarinet & violin), while the
third stretto (bars 21-27) focuses on the movement from Eb to F# and in this sense functions
similarly to a transition section in classical sonata form, preparing for the establishment of
F#-B as important pitch-class centres in the second group. From this we may conclude that
the question of pitch hierarchy -the retention of F# as an important pitch-class in the linear
unfolding of the second group -may have been the criterion which determined the choice of
inversion at index-number 11.

One of the biggest problems of . . . [the early twelve-note] compositions-the obscurity of harmonic relation-
ships-results from the piling up of set forms without a systematic basis for that combination. (p. 150)

The question of 'harmony' is central to much recent work on Schoenberg's dodecaphonic

music. Although Haimo claims that his ideas on the subject draw directly on Babbitt and
Perle, he defines his position against the background of Hyde's work.
Hyde argues that in Schoenberg's twelve-note compositions harmony is entirely regulated
by the basic set. l With the aid of Schoenberg's sketches, she is able to illustrate
Schoenberg's well-known preoccupations with constructional consistency and his fascination
with symmetrical pitch formations, particularly in the detail of local thematic statements.
Hyde concludes that

Schoenberg uses harmonies of the basic set to integrate all dimensions of harmonic structure. By expanding the
definition of harmony to include nonsimultaneous events occurring in both horizontal and vertical dimensions,
by allowing harmonic events to overlap, and by identifying the relating harmonies primarily by total intervallic
content, Schoenberg legitimately asserts that his twelve-tone compositions are totally integrated, because every
feature is derived from a single source, the basic set.'6

Haimo is critical of Hyde on the following grounds:

First, by attempting to account for all of Schoenberg's twelve-tone music, from Opp. 25-50, Hyde does not take
into account the enormous changes that took place in Schoenberg's twelve-tone method during this period.
Significantly, most of her analyses are from the earliest compositions: Opp. 23-9. The changes that took place
during this period . . . had an enormous impact on Schoenberg's harmonic vocabulary. In particular we see
Schoenberg moving from harmonic relationships dominated almost exclusively by local, associative harmony,
to a more consistent and coherent 'combinatorial harmony'. Second, it is easily demonstrated that the surface of
Schoenberg's twelve-tone compositions is covered with harmonies that are in no way equivalent to embedded
harmonies of the set, even if one extends those harmonies (as Hyde does) to include 'around the corner' har-
nionies ... Hyde tends to deal with this issue by marking for attention only those pitch-class sets that are in fact
equivalent to embedded harmonies of the set (pp. 31-32 n. 32)

Yet Haimo does not question Hyde's notion of 'harmony' itself. To me 'harmony' in
Schoenberg seems always intertwined with notions of hierarchy, prolongation and tonal
motion. 'Harmony' is generally understood to imply a hierarchical relation between struc-
tures in a composition, so even if, as Hyde claims, we were to establish that all pitch events
are reducible to the source sets contained within the set, we would have described only a
method for producing intervallic patterns. The concept of 'harmony' is of value when it im-
plies the existence of configurations which can be systematically or contextually recognized
as being 'non-harmonic'.
Haimo views the introduction of 'combinatorial harmony' as a qualitative improvement in
terms of consistency and coherence, which eliminated 'the obscurity of harmonic relation-
ships' evident in the early twelve-note works. Summarizing several of Babbitt's ideas, he
explains that

the basis of Schoenberg's mature twelve-tone harmony lies in the hexachordal division of the set, its IH-
combinatorial structure, and the opposition of order corresponding hexachords of IH-combinatorially related
set forms. In this way the structure of the set becomes the critical determinant of harmonic relationships.
This is so because the intervals most common within the hexachord are those most infrequent between the hex-
achords (and vice versa). (p. 32)

5 For instance, see Schoenberg's Twelve-Tone Harmony, passim.

16 Ibid., pp. 17-18.

69

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
However, Babbitt nowhere claims to be dealing with 'Schoenberg's twelve-tone harmonic
vocabulary' (p. 32 n. 33), but only with matters of set association. Besides, in what sense
does a systematic basis for associating sets 'solve' the problem of harmony? Perhaps even the
notion of 'twelve-note harmony' itself should be questioned and replaced by that of 'har-
mony generated with the twelve-note method'.
The new method provided Schoenberg with a concise system of 'self-defined' musical rela-
tions which, in a constructional sense, could replace those formerly given by tonality. Yet,
as Babbitt explains, unlike tonality the twelve-note method has no recourse to generalized
functionality, in the sense that it does not prescribe functional relations which are constant
for all compositions, such as the hierarchy of tonal distance provided by the circle of fifths in
tonal music. " But progression is 'associative' and 'non-functional' since no tonal motivation,
hence no tonal function, can be inferred without recourse to the connotations of a generalized
system of functions, such as tonality.'8 On the contrary, in twelve-note composition 'the nor-
mative factor is determined without any reference to means of its being so recognized other
than by internal structure, which is not true in tonal music, and by priority, which is not
necessary in tonal nmusic'. I
I believe that in many of Schoenberg's atonal and twelve-note works, tonal function is not
abandoned completely, but single pitch-classes or pitch-levels, rendered prominent by
virtue of their position as boundaries of groupings, are often made to bear implications for-
merly pertaining to tonal regions or keys and therefore function as true tonal centres,
displaying centricity within a given context without necessarily carrying all the implications
of the tonal system.20
In Schoenberg's twelve-note works the structural importance of tonal centres and the
manner in which they are used varies not only among different works but also among dif-
ferent sections of a single work. However, many of Schoenberg's compositions share certain
devices for creating hierarchies: the structure of regular metre and the prominence of those
pitch-levels or pitch-classes that appear at the boundaries of groupings, often reinforced by
leading-note and appoggiatura-like semitonal figures, and frequently supported by perfect
fifths and by the use of idiomatic cadential gestures. While the nature of conventional
metric structure remains essentially unaffected by the new situation, the retention of
pitch-class centres or pitch-level centres unsupported by tonal progression creates a new syn-
tactical context in which tonal motion is achieved through the polyphonic unfolding of
hierarchically structured lines. Schoenberg's twelve-note music relies mainly on associative,
as opposed to functional, harmony, even when it involves triads, as in the Ouvertiire of the
Suite Op. 29 (discussed above) and the Ode to Napoleon. In Schoenberg, 'combinatoriality'
is not necessarily associated with non-hierarchical pitch organization, nor does triadic
material necessarily imply tonal centricity.
The notion of 'tonal centre' implies that of prolongation, though not necessarily in terms
of the Schenkerian Ursatz. In this context prolongation refers to pitch-levels, pitch-classes,
or referential configurations whose position of priority within the structure is exercised even
when they are not literally present. Tonal prolongation involves a small number of pro-
longational types (the passing note, neighbour note and arpeggiation), which provide a con-
sistent set of relationships between notes of lesser and greater structural weight.2' Although
such simplicity and consistency of prolongational types has no equivalent in Schoenberg's
twelve-note music, this does not imply that prolongation as embellishment or diminution is
totally absent. On the contrary, the music abounds in gestures reminiscent of tonal melodic
modality. Often contour, rhythm and articulation contribute to evoke conventionalized

'' See Milton Babbitt, 'Quatrieme Cahier (n.d.): Le Systeme dodecaphonique', Journal of the American
Musicological Society, iii (1950), 265.
18 See idem, 'The String Quartets of Bart6k', The Musical Quarterly, xxxv (1949), 377, 380.
'9 Idem, 'Quatrieme Cahier (n.d.)', loc. cit.
20 Arthur Berger deals with a similar concept, which he calls 'pitch-class priority': see 'Problems of Pitch
Organization in Stravinsky', Perspectives on Schoenberg and Stravinsky, ed. Benjamin Boretz & Edward T. Cone,
Princeton, 1968.
21 See Joseph N. Straus, 'The Problem of Prolongation in Post-Tonal Music', Journal of Music Theory, xxxi
(1987), 4.

70

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
formulas, such as cadential leading notes, appoggiatura gestures and nota cambiata figures.
The hierarchies delineated by these figures are, however, unsupported by a consonance-
dissonance syntax.22
In his discussion of the Wind Quintet, Haimo states that 'where Schoenberg states the set
as a succession of chords, the resultant lines often seem to have only local significance.
Where the polyphony is created by layered set statements, the harmonies lack focus'
(p. 115). Admittedly, a distinctive feature of the Wind Quintet is the high degree of dif-
ferentation between its textural components, which is not necessarily equivalent to 'har-
monies lacking focus'. The extreme diversity of individual lines, which coincide only
momentarily, immediately to diverge, makes the few chordal passages stand out as caesuras
within the continuous linear flow. For this reason, sustained simultaneities, such as the
closing quartal chord accompaning the flute cadence figure at bar 21 of the Adagio,
become prominent events. The presentation of this material towards the end of the
recapitulation, at bar 103, performs an important associative function. It refers to the event
which marked the beginning of the large-scale shift towards the pitch-class centre of the sub-
sidiary subjects, that is, the linear cadence onfX ' at bar 21. Yet in this case eb"' is held over
the quartal chord (EL being the tritone above the bass note, while previously FO was the
tritone above the top note of the chord) instead of FN:

bar 21 A D G F
bar 103 D G C Eb

The originalf ' is not sounded until the very end of the bar and is immediately followed by
the reprise of the opening theme at its original pitch-level (starting at bar 104).23 After the
first note of the Adagio, no structurally significant Hauptstimme event either starts or ends
on eb' until this point. In terms of the large-scale structure of the Quintet, Eb-C-BW-F, in
combination with the whole-tone segments, also perform a powerful associative function,
appearing at major articulative points.24
These prominent chords function as landmarks associating distant events. The reason
for these quartal chords functioning so effectively as associative harmonies possibly has

22 In view of the absence of many of the fundamental features of tonal prolongation from much post-tonal
music, Straus proposes that a more reliable basis for describing its voice-leading is that of contextually associated as
opposed to prolonged pitch-levels (ibid., p. 13). The 'associational model' makes no a priori claims regarding the
pitches that intervene between the associated ones (ibid., p. 15) and can operate in music which makes no reference
to any common practice of harmony or voice-leading (ibid., p. 8). Straus's model can deal with procedures which
have their roots in two different areas of traditional practice: tonal prolongation and motivic association (ibid., pp.
15-17). Yet since this study is particularly concerned with matters of historical derivation, and since in
Schoenberg's twelve-note music, remnants of tonal voice-leading (primarily in the form of characteristic gestures,
large-scale semitonal leading notes, and fifth support to tonal centres) play such an important part in defining
structures, I prefer to retain the term 'prolongation', stripped of its strict technical characteristics, for discussing
procedures concerning hierarchical pitch organization, in order to distinguish them from those that have their
origin in motivic association.
23 According to MS 566 and to serial ordering, the last note of the bassoon at bar 110 should be F and not Ah as it
appears in the Universal edition.
24 (a) The coda of the first movement (bars 205-27) first comes to a close on a sustained C-Bb-F-Eb chord
accompanying a flute melody consisting of whole-tone segments (bars 209-12).
(b) The second movement is punctuated by two appearances of this tetrachord accompanying a cadenza-like
rendering of the whole-tone segments by the E flat clarinet (bars 142 & 359). The tetrachord is also present at the
climactic sequential passage at bars 400-405, where the whole-tone material is played as an ascending scale by the
horn and then by the flute. In addition, this tetrachord appears at less prominent cadences such as bars 188-90,
where it is followed by its linear statement, played by the horn (bars 191-2).
(c) The second movement closes with a chordal presentation of the tetrachord and a whole-tone E flat clarinet
descent; the same pitch collections feature in the opening two bars of the Adagio. At the beginning of the Adagio
the minor ninth descent of the Nebenstimme is reminiscent of the flute melody in the coda of the first movement
(bars 209-12). The association was stronger in the earliest extant draft for this phrase, MS 508 (Haimo, Ex. 5.10),
which shows a twofold statement of the flute shape from the coda of the first movement.
(d) The sequential passage towards the end of the second movement involving these referential collections
(bars 400-405) reappears in the fourth movement at bars 223-5. The connection is evident despite the fact that in
the fourth movement the direction of the scalic figures is reversed as a result of the derivation of the referential
tetrachords from an inversion of the set (Io, i.e., inversion starting with C).

71

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
something to do with the fact that they are segments of the set but also with the more general
fact that a single and distinct intervallic combination is consistently reiterated at prominent
places, a procedure which Schoenberg already knew from his use of quartal harmonies as
landmarks articulating the first three movements of the Second String Quartet.
For Haimo, 'throughout the Quintet Schoenberg came up against this paradox: if he
made the lines conform to the set, then the simultaneities lacked focus, but if he formed the
simultaneities from segments of the set, the lines lacked direction' (p. 117). Haimo thinks
that Schoenberg found a solution to this in the procedures used in the opening of the third
movement (discussed above), that is, the partitioning and association of sets determined by
the structure of the set itself. I would like to suggest that the paradox is not in the Wind
Quintet but in Haimo's mode of reasoning: for the lines lack direction, and the
simultaneities focus, if we a priori equate 'direction' and 'focus' with the intervallic content
of set segments, but not necessarily otherwise.

A number of comments in . . . [Schoenberg's] writings indicate his belief that twelve-tone music 'unleashed
the potential of absolute music' and provided solutions not found in his earlier, atonal, compositions . . . He
had noticed that his more extended atonal works were text settings and that he tended to articulate the formal
structure not by internal pitch logic but, rather, by means of the differentiation provided for by the various sec-
tions of the text. (p. 29)

Schoenberg's misgivings regarding the dependence of his tonal works on extra-musical

elements and the supposed difficulties in composing 'absolute music' in the 'style of the
freedom of the dissonance' have given rise to the idea, to which Haimo subscribes, that
Schoenberg found in the twelve-note method the answer to the problem of atonal form. The
fusion of the twelve-note method with Baroque and Classical forms furnishes a composition
with a repertory of fixed relations existing prior to the particular composition, thus pro-
viding a more stable context of reference and expectation than that which existed in the
atonal works. However, Schoenberg solved many problems of atonal form before his concep-
tion of the twelve-note method. Some of these solutions have to do with free association as in
Herzgewdchse, and others with the expectations of formal prototypes as in some pieces in
Pierrot lunaire. Admittedly, many of the atonal works involve text, but these do not include
the Five Orchestral Pieces nor the Piano Pieces Opp. 11 and 19.
In Pierrot lunaire the use of formal prototypes and established contrapuntal textures is
generally motivated by the imagery of the poems. Imitative textures, as in 'Der Mondfleck',
and ground-bass techniques, as in 'Nacht', are particularly well suited to a compositional
method largely based on motivic cohesion. Yet the integration of pre-Classical and Classical
idioms and forms with material which originally operated in the context of virtually
unrestricted free association is also symptomatic of a shift to a more conscious compositional
attitude.
Only in the 1 920s did Schoenberg come to regard the dependency of this type of music on
text and free association as a serious limitation.25 Such constructional qualms would have
been utterly unacceptable to him in the previous decade, when he criticized Liszt for not
having 'the courage to explore the dark region of the unconscious, in order to bring up
content and form as a unity'.26 At that time Schoenberg defiantly advocated unmediated
comprehension and believed that through artistic intuition it was possible to attain complex
formal continuity in an undeliberate manner. A letter to Kandinsky dated 24 January 1911
reads:

Every formal procedure which aspires to traditional effects is not completely free from conscious motivation.
But art belongs to the unconsciousl One must express oneself! Express oneself directlyl Not one's taste, or one's
upbringing, or one's intelligence, knowledge or skill. Not all these acquired characteristics, but that which is
inborn, instinctive. And all form-making, all conscious form-making, is connected with some kind of

25 See Schoenberg, 'Analysis of the Four Orchestral Songs Op. 22', trans. & annotated by Claudio Spies, Perspec-
tives on Schoenberg and Stravinsky, p. 27.
26 'Franz Liszt's Work and Being', Style and Idea, p. 444.

72

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
 mathematics, or geometry, or with the golden section or suchlike. But only unconscious form-making, which
sets up the equation 'form = outward shape', really creates form; that alone brings forth prototypes which are
imitated by unoriginal people and become 'formulas'.2"

Schoenberg's radical change of rhetoric in the 1920s, with its focus on construction, com-
prehensibility and objectivity (Sachlichkeit) is the counterpart to striking changes in
musical style. According to 0. W. Neighbour, 'In January 1915 Schoenberg wrote to
Zemlinsky that his new symphony would be "worked" ("ein gearbeitetes Werk") in contrast to
his many "purely impressionistic" recent works'.28 The thematicism of Die Jakobsleiter
reflects the change of emphasis from 'one must express oneself directly' of the letter to
Kandinsky to an uncompromising attitude towards clarity, coherence and directness, cor-
responding to a time when Schoenberg was consolidating his personal, somewhat
theosophical outlook, and considered it his task 'to define his thoughts and to compose
coherently out of them'.29
Webern's account of the path that led to the twelve-note method points to the connection
between Kraus's and Schoenberg's preoccupations: 'Finally I must point out to you that this
is so not only in music. We find an analogy in language . . . Karl Kraus's handling of
language is also based on this; unity also has to be created there, since it enhances com-
prehensibility'.30 Schoenberg's statements in 'Composition with Twelve Tones (1)' similarly
places the question of 'comprehensibility' as the central issue:

Form in the arts, and especially in music, aims primarily at comprehensibility. The relaxation which a satisfied
listener experiences when he can follow an' idea, its development, and the reasons for such development is closely
related, psychologically speaking, to a feeling of beauty. Thus, artistic value demands comprehensibility, not
only for intellectual, but also for emotional satisfaction. However, the creator's idea has to be presented,
whatever the mood he is impelled to evoke.
Composition with twelve tones has no other aim than comprehensibility.3

The commitment to 'carry on a logical and comprehensible argument in which Idea gives
rise to related ideas through the agency of musical structures'32 resulted in the integration of
the 'style of the freedom of the dissonance' into a pre-established hierarchy of formal
prototypes and balancing phrase-construction. Catalogued forms, such as dance forms, rondo
and sonata, as well as established textures, such as fugue and passacaglia, provided not only
a set of patterns of expectation but a repertory of idioms traditionally associated with them.
The interaction between tonally evasive material, associative harmony, single notes carrying
the function formerly attributed to keys, and the idiomatic gestures and formal expectations
of the quoted historical model predates by many years the conception of the twelve-note
method. Viewed in this context the twelve-note method emerges as a refinement of tech-
nique complementing ideas about musical logic and form which predated and motivated it.

Haimo's book places great emphasis on the elements of coherence and continuity in
Schoenberg's development. As Foucault points out,

this law of coherence is a heuristic rule, a procedural obligation, almost a moral constraint of research: not to
multiply contradictions uselessly; not to be taken in by small differences; not to give too much weight to
changes, disavowals, returns to the past, and polemics ... But this same coherence is the result of research ...
In order to reconstitute it, it must first be presupposed and one will only be sure of finding it if one has pursued
it far enough and for long enough.33

27 See Jelena Hahl-Koch, Arnold Schoenberg, Wassily Kandinsky: Letters, Pictures and Documents, trans.
John C. Crawford, London & Boston, 1980, p. 23.
28 Schoenberg is here referring to his plan for a choral symphony incorporating texts from Dehmel, Tagore and
the Old Testament, which included DieJakobsleiter in its final section. Quoted in 0. W. Neighbour, 'Schoenberg,
Arnold', The New Grove, xvi. 713.
29 Alexander Goehr, 'The Idea behind the Music: Schoenberg and Karl Kraus', Music Analysis, iv (1985), 69.
Quoted in Willi Reich, Schoenberg: a Critical Biography, trans. Leo Black, London, 1971, p. 135.
'Composition with Twelve Tones (1)', Style and Idea, p. 215.
32 Goehr, 'The Idea behind the Music', p. 67.
Michel Foucault, The Archeology of Knowledge, trans. A. M. Sheridan Smith, London & New York, 1989,
p. 149.

73

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms
Haimo's reading of Schoenberg's twelve-note music exhibits a systematism far more com-
prehensive than that of previous commentators. While weary of the magnitude of the task
(p. 8), he seems undeterred by the general rule that the more 'ideal' the 'architecture', the
greater the risks of betraying the truth.
The form of the book reflects the teleological nature of the enterprise. In order to present
'Schoenberg's compositional choices in context, as components of a developing critical pro-
cess, and not as a succession of unrelated decisions' (p. 7), it starts with a general discussion
of the mature twelve-note style (the period starting with Von Heute auf Morgen, Op. 32,
and including the remaining twelve-note compositions). This is followed by a detailed ex-
amination of works written between 1914 and 1928, the 'transitional period'. The final
chapter involves extensive discussion of both the Third String Quartet and the Variations for
Orchestra. Haimo regards the latter as the 'Art of Fugue' of twelve-note composition, in the
sense that it exhibits masterly command of those techniques that would be favoured in the
'mature style'.
The final chapter contains several generalizations about the works written after Op. 31
which I find are not always fully borne out by the music and which fail to take into account
some of the developments discussed in the Epilogue, subtitled 'Op. 32 and beyond'. In this
chapter Haimo claims that once Schoenberg discovered ways of partitioning the set into
various voices in such a way that

the melodic structures created within those individual lines could be derived from the structure of the set . . .
[he] no longer felt it necessary to use the primitive techniques of rotation for melodic variety or layering to
generate polyphony. These techniques now disappear from Schoenberg's vocabulary, never to return. (p. 153)

While this seems to apply to rotation, in the late works Schoenberg used many means of
achieving melodic variety besides partitioning, such as the use of multiple orderings of the
source hexachord in the same composition (which Haimo mentions on page 181). For in-
stance, in the Ode to Napoleon Schoenberg usually treats the hexachords as motifs which he
presents linearly (as in the principal part in bars 86-96), and polyphony generally results
from layering of the hexachords; and in the Phantasy, Op. 47, linear presentation of sets is
at least as pervasive as partitioning.
Similarly, Haimo's statement that after the Third String Quartet Schoenberg abandons
retrograde combinatoriality as a compositional procedure (p. 161) ignores the various in-
stances of this form of combinatoriality in many late works, such as the Fourth String
Quartet (see, for example, bars 638-40 and 686-8). While the 'near-octaves' or 'voice ex-
change' which result from retrograde combinatoriality may have been one of the reasons
why Schoenberg did not favour it (loc. cit.), we should not forget that in some of the late
works octave repetitions involving 'voice exchange' are a prominent feature (see, for ex-
ample, the String Trio, bars 23-30).
Finally, I find the claim that after the Third String Quartet 'no longer are the sources of
developing variation to be found in the strict linear ordering of the set but, rather, in the
transformation and relationship that can be brought out by isomorphic partition' (p. 159)
unnecessarily limiting. Developing variation is a pervasive feature which finds its source
both in the linear set and in multifarious forms of partition and which often progresses
according to principles that have no relationship with the twelve-note method. Admittedly,
many of the generalizations put forward in this chapter correctly refer to well-known
tendencies in the late compositions; yet by presenting trends as firm rules, they conspire to
dissipate the contradictions and ruptures which, far from being accidents, are at the very
basis of creative development.
Briefly, as a revision of almost half a century of theoretical speculation on Schoenberg,
Haimo's book is an ambitious, though not unproblematic, contribution. Given its remark-
ably clear style, devoid of unnecessary jargon, and the wide range of subjects covered in such
a short space, this book will be of value to readers wanting an introduction to a most com-
plex subject.

74

This content downloaded from

167.98.204.179 on Tue, 06 Apr 2021 20:30:08 UTC
All use subject to https://about.jstor.org/terms