Wiley Music Analysis: This Content Downloaded From 128.179.146.121 On Wed, 10 Jan 2018 13:14:33 UTC

Scriabin's Self-Analyses
Author(s): George Perle

Source: Music Analysis, Vol. 3, No. 2 (Jul., 1984), pp. 101-122
Published by: Wiley
Stable URL: http://www.jstor.org/stable/854313
Accessed: 10-01-2018 13:14 UTC
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/854313?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
http://about.jstor.org/terms
Wiley is collaborating with JSTOR to digitize, preserve and extend access to Music Analysis
This content downloaded from 128.179.146.121 on Wed, 10 Jan 2018 13:14:33 UTC
All use subject to http://about.jstor.org/terms
GEORGE PERLE
SCRIABIN'S SELF-ANALYSES
What authority does a composer have as analyst of his own music?

question that we are often constrained to consider, since compo
offer us anything that can be construed as an analysis when i
explaining how their music is put together. And we cannot alwa
relevance of the vague and spacious generalities they give us ins
motivations behind these. When Schindler asked Beethoven why his
sonata had only two movements the composer replied that he hadn't
write a third, a reply that was quite good enough for Schindler
extended exposition of the twelve-tone system Schoenberg makes t
assertion about the use of inversional forms of the series: 'While a
begins with the basic set itself, the mirror forms . . . are applied o
Almost every twelve-tone piece of Schoenberg's controverts this as
opening bars. Webern, himself a conductor, whose works are m
marked with explicit and exhaustive metronomic indications, se
had no idea at all of their time-span. The durations given in the scor
letters are as much as two or three times longer than the durations im
metronomic indications. Though Scriabin admitted that there was 'n
accident' in his music, that he composed 'according to definite p
refused to say what the principle was. Faubion Bowers quotes a cont
'Scriabin always said that everything in his later compositions w
according to "law". He said that he could prove this fact. However, e
seemed to conspire against his giving a demonstration. One day
Taneyev and me to his apartment so he could explain his
composition. We arrived and he dilly-dallied for a long time. Finally
had a headache and would explain it all another day. That "another d
came'.2
Yet even though they did not give us analytical surveys of their compositions,
the composers of an earlier age were constantly making explicit and detailed
analytical assertions, in the very act of writing the notes down. When
Beethoven, in the slow movement of the Fifth Symphony, spells Forte's pc set
4-27 as Ab C Eb F# in b. 29 and as Ab C Eb Gb in b. 206, he is engaged in
an act of analysis, as well as an act of composition. For Scriabin, as for
MUSIC ANALYSIS 3:2, 1984 101
GEORGE PERLE
Schoenberg, and almost at the same time, a diatoni

differentiated notes was replaced by a semitonal scale
ferentiated notes. The notational system was not rep
new music, based on the material of the universal set of
to make do with only seven degree-names and the 'accid
us to modify their signification. The discrepancy be
language and the means of notating it troubled Scho
invented and, without success, proposed the adoption of
Scriabin, on the other hand, tried to establish consiste
for the continued employment of the traditional no
superficial glance at some of the late piano pieces sh
something of the sort. Forte's set 4-17 [0,3,4,7] plays an
the Seventh Sonata and the fourth of the Five Preludes
opus. In the former the corresponding notes are invaria
Db Fb (Ex. 1), and in the latter as A C C# E (Ex. 2). I
structure is subjected to curious notational revisions whe
Ex. 3, from Op. 74, No. 3.
Ex. 1 Ex. 2 Ex. 3
A, " Id
i.L. .,4e I
The revised spelling, unlike our example from Beethoven, implies no

distinction in function between the two chords. What it tells us is that both
derive from the same set, an octatonic scale which contains its own tritone
transposition and which - if we lay it out so that the two notes that employ the
same letter-name, A# and A, occur as boundary elements - may be spelled
'diatonically', that is, so that successive notes unfold successive letter-names as
in the diatonic system (Ex. 4).
Ex. 4
A, . ,. R'
WD*wl l 3i q'
vi .
Since this scale consists of two 'diminished 7th' chords, the

collection is repeated at t3 and t9 transpositions, as well as at t6.
102 MUSIC ANALYSIS 3:2, 1984
SCRIABIN S SELF-ANALYSES
the same spelling, regardless of transposition (Ex.

harmonic' element is introduced in the piece, a pass
example, that is repeated in exactly the same context w
Ex. 5
pp
Ex.
we m
reso
Ex.
But
chor
segm
'mas
repr
Ex. 7
IAP v.j
vwAl's
are all derivabl

Scriabin prefer
transpositions
The derived he
to the octato
GEORGE PERLE
are identical in pitch-class content, there are only three

collections: no hierarchical structuring is possible, since
class content of each with either of the others is alway
component 'diminished 7th' chords is retained, the othe
heptatonic scale, however, has twelve independent form
master scales, and each form shares a different collec
with each of the others derived from the same maste
collection of three pitch classes with each of the forms
two master scales. The master scales and the derive
illustrated in Ex. 8.4
Ex. 8
C3
0,2
C31
IF of I I
C31,3
The tonal vocabulary is still frustratingly restrictive, however, for a musical

language that makes no distinction between chord and scale, in which the linear
dimension unfolds as a series of arpeggiated segments of paired 'diminished
7th' chords. Scriabin provides a source of contrast through a variant form of the
derived scale, the final degree of which is occasionally raised by a semitone, a
revision that results in a striking change of harmonic colour by converting a
five-note segment of the scale into a whole-tone collection (Ex. 9).
Ex. 9
9JrI
That this is the primary, though not the sole function of t

is made abundantly clear throughout the work. When the
it entirely converts the tone material given at this point
segment of the scale (Ex. 10, b.30). The seventh degree is
form, and then immediately raised a second time as p
progression in the middle voices (Ex. 10, bs 33-34).
Ex. 10: Sonata No. 7

avec une celes te volupt
fres pur, avec une profonde douceur
.. m-,, " '--"
Each of the Five Preludes is a

unproblematical piece whose ton
4 and 5. A single octatonic pit
compositional representations
equivalent pitch levels. Prelude N
finest of Scriabin's late pieces. A
represented through its derived
(minim),5 is based on master s
C30,2 and returns in b. 6 (crotc
bs 8 (crotchet) to 12 (minim) -
codetta (Ex. 11) commences wi
recapitulation, followed by a cad
D# of the preceding chord to D
the whole-tone variant of the
and neighbouring notes occu
function as elements of melodic
7th' components of the maste
'dissonant' notes that cannot eas
GEORGE PERLE
Ex. 11: Op. 74, No. 1

appP
cq app
.d3amp,3a
"-'--s ' "' ' ' fr
C31.3
AVIo I-
traditional harmony, and both are instances of th

instance is the climactic note at the midpoint of
of the middle section (Ex. 12). The effect is that
whose progress is resumed and completed at the
tion in b. 10 (crotchet), again at the inception of
last time in the lower octave in the concluding p
Ex. 12
6-
API I
3I
C31,3
second lower octave displacement, finally takes its plac

component of the whole-tone variant of the hepta
principles exemplified in the Sonata are subtilized a
ordinary degree in these sixteen bars.
The distinction between the two forms of the derive
implications for harmonic differentiation and struc
realized in Prelude No. 5. Transpositional relations
Interval-2 as well as the Interval-3 cycle. The pitch-cla
shown in the following table:
PART ONE, SECTION ONE

Bar(s)
1: G A Bb C D6, E6 F G A Bb (C) D, E6 F IA B (C) D E, F G

2: F G (Ab) Bb C, D6 E,/Ebb
3: D# E# F# G# A B C# I D# E# F# (G#) A B C# I E# Fx (G#) A# B C# D#
4: C# D# (E) F G A B/B,
PART ONE, SECTION TWO
5: C# D# (E) F# G A Bb
6: C# D# E F# G A Bb
7: E F# G A Bb C D6
8: G A Bb C D6 E6 F6 G6
PART TWO, SECTION ONE

9-10: = Bars 1-2
11: ABC DEbFG ABC(D)Eb FG I BC#(D)EFGA

12: GA (B6) C D6 E6 F/F6
PART TWO, SECTION TWO

13: GA (Bb) C Db Eb Fb
14: GA Bb C Db Eb Fb
15: A# B# C# D# E F# G
16-17: C# D# E F# GA Bb C GABb C Db Eb Fb G
Scriabin's own bar lines and rhythmic groupings within each bar provide an
unproblematical basis for partitioning the composition into the pitch collec-
tions that are here represented as scales. In representing them thus I have
simply ordered the pitches so that successive letter-names correspond without
exception to the composer's own spelling. The only liberty I have taken is in
supplying, in parentheses, the single note that is missing from some of the
scales, a note unambiguously implied, in every instance, in the invariant scale
patterns of the movement. If we eliminate literal repetitions and the final
cadential chord, we find the whole movement contained in only five bars: bs 1-2
return at t8, tO and t2 respectively in bs 3-4, 9-10 and 11-12; b. 5 returns at t6 in
b. 13; b. 6 returns at t3, t6 and t9 respectively in bs 7, 14 and 15; b. 8 returns at
t6 in b. 16. The second half of the piece, bs 9-16, is a repetition of the first,
GEORGE PERLE
commencing at the original pitch level and shifting after

Three scale-types appear. The complete octatonic sc
close of each part. The opening is based on what we call
the Seventh Sonata, a variant form of the seven-note
scale, in which a five-note segment of the whole-to
through the raising of the seventh degree (Ex. 9). This
is indicated in italics in the above table. The whole-to
plays such an important structural role in Prelude N
appropriate to define it as a variant form of some othe
occur at every even pitch-level, so that every five-n
whole-tone scale, C21, is represented. In every instan
component of a whole-tone collection (circled in Ex.
foreground.
Ex. 13: Op. 74, No. 5

fier, be/li uLeux-
bow -- ... .. .. .... .......................
~-~--b------
i ..... r7 1 3
Ib I. bb ~,i
L%- Y. L iI
ilf a i,l [Ji . -Am
irrl' .mDO 6m-,6"
In b. 2 and at the corresponding point

'normal' octatonic version, so that the
of the octatonic scale. In b. 4 thi
contrasting episode, bs 5-8, based on the unaltered form o

Here, and at the return of this music in the last four bars
are determined by the Interval-3 cyclic structure that
scale: the 'normal' heptatonic scale of bs 5-6 returns
explicit unfolding of the master scale in b. 16 embraces a
apparent tritone transposition at the close is only an enha
notational way of referring back to the very beginning o
chord, Eb-Db-G, in association with the bass line, A-E
two cyclic structures, C21 and C31,3, on which the piece is
itself to those notes which are common to both. The u
cycles is illustrated in the following outline, which embr
segments and shows the source of Scriabin's notation for
Bars
1-4: Tonality C21 (Cb Db Eb F G A B C# D# E# Fx)

5-8: Tonality C31,3 (C# D# E F# G A Bb C Db Eb Fb Gb)
9-12: Tonality C21 (Cb Db Eb F G A B C#)
13-17: Tonality C31,3 (A# B C# D# E F# G A Bb C Db Eb F
The first of the two Preludes of Op. 67 is a precursor of
piece. The whole-tone collection shown in Ex. 14, or a
same, occurs on the downbeat of fourteen of its thirty-fiv
Ex. 14
The only note that is not a component of the C31,3 octatonic scale in the first
twelve bars is the Ab of Ex. 14. Consecutively ordered degree-names in the
composer's own spelling unfold as follows (the non-octatonic component is
represented by a lower-case letter):
D# E F# G ab A Bb C Db Eb Fb Gb
The scale is shifted to C32,4 in bs 15-16:
G# A# B C D E FJ t G
There is a return to the 'home key' in bs 19-20, a tll transposition of bs 15-16:
GA Bb C Db Eb Fb fGb
GEORGE PERLE
The C31,3 octatonic scale is maintained for the remainder

the non-octatonic element appears as d in b.21 and as cb
is transposed at t6 and t3 respectively before returning t
in b.27. In the last four bars the non-octatonic element
E F# G A Bb C Db Eb
The modulation from C31,3 to C32,4 is effected in bs
Ex. 15: Op. 67, No. 1

~ ~C.=.
C31 (E G Bb Db) is shared by

scale is displaced by C32 (F Ab
the shared component of the s
collection. In the following bar
prominent melodic figure of t
so, in spite of the retention of
of the first scale. The missing
in the second chord that acc
downbeat of b. 15. This overlap
interpreting Ex. 15 as a simult
Ex. 16 Ex. 17
----C 3 ----
C ,
C 32,4
A tll transposition of bs 13-14 (bs 17-18) effects the return from C32,4 to
C31,3.6
At first sight the musical orthography of Prelude No. 2 of Op. 74 seems
bizarre and eccentric to an extreme, with its spelling of a

motive as inflections of a single scale degree, of octave re
7ths or diminished 9ths, of perfect 5ths as doubly
diminished 6ths (Ex. 18).
Ex. 18: Op. 74, No. 2
Tres lenf, contemplatif
2Pp4
But h
especi
C31,3
repre
(passin
A# B
A seco
B# (C
The ba
of the
interr
scale a
piece
largel
motio
follow
A# B
Strict
A# to
not ca
substa
inclus
the pr
GEORGE PERLE
Ex. 19
Ex. 20
In Prelude No. 5 the infiltration of a non-octatonic element is exploited for its

extension of whole-tone relations. In Prelude No. 2 it is exploited for its extension
of semitonal relations. Prelude No. 4 investigates still other implications of this
variant of the octatonic set. Through the semitonal inflection of any one note of
the set we derive a second complete cycle, the 'augmented triad', and this
provides still another means of harmonic differentiation and another basis for
progression among the three octatonic scales. The t4 and t8 transpositions that
occur throughout Prelude No. 4 are as integral to its musical language as t3, t6 and
t9 transpositions are integral to the musical language of Prelude No. 3 and the
Seventh Sonata.
Every simultaneity, without exception, is derivable from one of the three
non-equivalent octatonic scales, or from a variant in which one or another
element of the scale is semitonally raised or lowered through chromatic
inflection and thereby converted into an element of the C3 collection that is not a
component of the given octatonic scale. Though there is never more than one
such chromatically inflected note in any simultaneity, the aggregate result is a
twelve-tone master scale, each of whose constituent sub-scales comprises one
complete Interval-3 cycle, one or another three-note segment of a second
Interval-3 cycle, and one or another note of the remaining Interval-3 cycle.
Through chromatic inflection each of the three non-equivalent octatonic scales
is convertible into such a twelve-tone master scale. The three octatonic scales
and their aggregate variant forms as represented in Scriabin's own notation are
shown below (letter names of variant elements are italicized; letter names of
missing elements are in parentheses):
C31,3: A B/B C D/D# E F/F G/G# A Bb/B C Db Ebb/Eb (Fb) Gb (Abb)/Ab Bbb
C30,2: D# E F G A Bb/B C/C# D (Eb) F Gb
C32,4: B (C#) D Eb/E F Gb/G Ab/A Bb Cb/C Db Ebb Fb
In the totality of their variant forms all three octaton

'augmented triads'. For example, the first and third of
unfold all twelve pitch classes and respectively represe
as follows:
C31,3: B D# GB Eb, B E G CAbC~, F A Db, D F# Bb Ebb Gb

C32,4: B Eb G Cb, E Ab C Fb, FA Db, D Gb B Ebb
The aggregate variants of C30,2 that are exploited in Prelude No. 4 do not
unfold the complete twelve-tone master scale. Only two 'augmented triads'
are represented:
C30,2: E# A C# F, F# Bb D Gb
The shared Interval-4 cycles result in a much closer interrelatedness of the
three octatonic scales. This explains why there is a more rapid and continual
interchange of scales than in any of the other pieces we have examined.
Compositional implications and characteristic notational procedures are both
demonstrated in Ex. 21 and in its t4 transposition, Ex. 22. The same sequence
of 'augmented triads' unfolds at both pitch levels: C40, C43, C42. The
following scale segments are represented respectively, in the two examples:
C31,3: A# B/B# C# D/D# E F# G/G# A (Bb) C
C32,4: D Eb/E F Gb/G Ab Bb Cb/C Db (Ebb) Fb
Ex. 21: Op. 74, No. 4
?TT I Io--%
nl
LL!
Ex. 22 Ex. 23
r_ A !
,/J
GEORGE PERLE
In Ex. 23 the apparent notational inconsistencies (C# and C

Db and B# on the second) are explained by the conflict be
and harmonic considerations. The outer voices progress in
their adjacent semitonal neighbours, thus expanding the in
boundary elements of the scale:
C31,3: B CQ (D#) E F G# A (Bb) C Db

The opening bars and their strikingly modified recapi
are shown in Exs 24a and b. In the last phrase the in
Ex. 24a
lent, vague, ind cis
poch iss.
C 31,3 C 30,2 133,3 C30,2 C 31,3

Ex. 24b
r"al @ I L I-
Ad , - - L II , . -
u' llW-----------------1----- -
C31,3
C C30,2
3,3 C 302 C31,3
C3 13 C30,.2
C30,2
C2,4 C31,3
C32,4 1,~3
transposition of b. 2 and rhythmic augmentation postpone the final ret

the pitch-class content of the initial chord. Linearly unfolded Int
cycles in the closing progression restore the basic tonality, C31,3,
interpolation of E in the last chord restores the initial pitc

25).
Ex. 25
C33
We must distinguish between Scriabin's literal use of the octatonic scale in

Op. 74, No. 3, and antecedent examples like the first four bars of Liszt's Sonetto
104 del Petrarca, in which the same pitch-class collection is engendered in a
tonal context through the recurrent neighbouring-note juxtaposition of two
diminished 7th chords.7 But in making this distinction we must recognize that
the example from Liszt represents an important late stage in the development
toward the integral, autonomous exploitation of cyclic interval relations that we
eventually find in Scriabin. On the other side of the example from Liszt we find
the interval cycle partitioning musical space in the chromatic prolongation
techniques of other nineteenth-century composers. Salzer and Schachter cite an
example from Schubert in which the tonal prolongation unfolds an Interval-2
cycle, from Chopin in which it unfolds an Interval-3 cycle, and from Wagner in
which it unfolds an Interval-4 cycle.8 In the music of Rimsky-Korsakov cyclic
progressions play so extensive and pervasive a role that they tend to replace,
rather than prolong, the traditional harmonic functions of the diatonic tonal
system. In his memoirs Rimsky-Korsakov cites Liszt and Glinka as influencing
his own discovery of the octatonic scale: '[In Sadko] Glinka's scale, descending by
whole notes, has been replaced by another descending scale of semitone-whole
tone, semitone-whole tone, a scale which subsequently played an important role
in many of my compositions'.9
Nothing so surely demonstrates Schoenberg's profound and productive
conservativism ('I personally hate to be called a revolutionist, which I am not',
he wrote, in a letter discussing the origins of the twelve-tone system)'0 as his
extraordinary extension of tonal prolongation techniques in the Chamber
Symphony, Op. 9, a work in which cyclic collections - the whole-tone scale,
augmented triads, series of perfect fourths - are integrated into the
hierarchical relations of the diatonic tonal system. This continued develop-
ment, in an increasingly perilous harmonic situation, of chromatic prolongation
procedures of the sort described by Salzer and Schachter comes to an end with
the first atonal pieces of 1909.
GEORGE PERLE
Scriabin's parallel evolution, on the other hand, leads h

but rather into a new kind of 'tonality' in which symmet
semitonal scale by means of interval cycles generate
referential harmonic structures. Whatever cannot be int
of cyclic relations that defines the pitch material of
eliminated. Scriabin's compositional exploitation of
relations and new referential harmonic structures sometimes tends to be literal
and mechanistic. There are passages that are hardly more than routine demon-
strations of the pitch-class invariance maintained under successive t3 trans-
positions of the octatonic scale. It is almost as though he were so intoxicated with
the excitement of his discovery of a new tonal system that he sometimes forgot
that to compose means something more than the literal surface restatement of
background structural relations. But the Five Preludes show a growing aware-
ness of the limitations of his musical language and an increasing subtlety and
sophistication in his compositional technique. In his careful and sensitive use of
'non-harmonic' tones in Preludes Nos 1 and 2 he confronts the problem of voice-
leading in a system that makes no distinction between the linear and harmonic
implications of the scale. The intersection of different cyclic systems in Preludes
Nos 4 and 5 enlarges transpositional and formal possibilities. In Prelude No. 4
there is an enormous enrichment of the harmonic vocabulary. Above all, there is
the extraordinary diversity in unity of the Five Preludes as a group. Though all
five movements are based on the same principal master scale, C31,3, each has,
within that basic tonality, its own distinctive tonal and harmonic identity. The
central movement, in its simple and straightforward surface unfolding of the
invariant relations inherent in the basic octatonic scale, has the character of an
axis of symmetry in the context of the work as a whole. In sum, the Five Preludes
hold great promise for what Scriabin might have achieved in the evolution of a
comprehensive and coherent post-diatonic tonal system, had death not brought
his work to an end, in his forty-fourth year, so soon after their completion.
Elsewhere I have stressed the role of cyclic interval structures in the atonal and
twelve-tone music of Schoenberg, Berg and Webern, in the string quartets of
Bart6k, and in Stravinsky's Le Sacre.11 The connections between Scriabin and
Berg are particularly striking. Ex. 26 from Prelude No. 4 is characteristically
Bergian in its simultaneous unfolding of different cyclic progressions. We find
numerous similar passages in Berg's oeuvre, from the Piano Sonata, Op. 1, to the
Violin Concerto and the third act of Lulu.
Ex. 26
ALjJ~bb
clf#~t
The same pentachordal partitioning of the whole-t

increasingly significant role as we progress from the Se
first Prelude of Op. 67 to the fifth Prelude of Op. 7
language of Wozzeck.12 The pentachordal whole-tone
note which brings the first of the Five Preludes to a clo
its t7 transposition, as the closing chord of Wozze
referential collection of the work as a whole (Ex. 28).
Ex. 27 Ex. 28
VJ 0p T
The famous 'mystic chord' of Prometheus gives us the same collection in its
inversional form (Ex. 29). Faubion Bowers reports that on 27 June 1910 'the
Ex. 29
Ai~
newspapers carried an announcement that the Mysterium will soon be finished

and as Prometheus was built from six tones like the Pleiades, the Mysterium will
be hitched to a constellation of nine'." Surely what Scriabin had in mind was
GEORGE PERLE
the octatonic scale plus the raised scale degree that conve
of that scale into a whole-tone collection. This 'constellat
'mystic chord' of Prometheus plus the normal octatonic
basic set of bs 1-12 of Op. 67 is equivalent to a tritone tr
'constellation of nine' (Ex. 31).
Ex. 30 Ex. 31
L -9- Fm
A
lI r PT% U M vw P'1IF
If cyclic interval structures play such an impor

many different composers, an explanation in term
another is surely inadequate. The interval cycl
partitioning, and thus imposing an ordering u
ferentiated pitch classes of the twelve-tone sc
'naturally' and 'intuitively', as well as through 'in
the weakening and eventual elimination of tradit
the replacement of a diatonic scale of unequal de
semitonal scale. But the interval cycle is not the
partitioning the tone material. Inversional com
music based on a twelve-tone scale inversionally
(Exs 27/28 and 29 are so related) are functiona
with Schoenberg's formulation of the twelve-t
immediately preceding this, beginning in 19
inversion was axiomatically recognized in Schoen
come to recognize it as a theorist of the new mus
recognized in atonal music from the very beginn
basic cell and in the special role assigned to s
chords which are their own literal inversion.'5 It
that strict inversional complementation comes to
in the music of Bart6k.'6 At the time of his
discovered the meaning of inversional comple
tonal system, but it seems likely that he would so
symmetrical - that is, self-invertible - pitch-c
the pitch material of his late music. In fact, he fr
of the precompositional pitch structures as a bas
as in Ex. 32, in which the two upper parts and the t

through their simultaneous transposition in opposite d
t3, that maintains the pitch-class invariance of the oct
Ex. 32: Op. 74, No. 3
1 ,' 11
' l w -W w i-I w
#9I
What he fails to recognize is the fact, or at least the impl

the whole movement may be literally inverted witho
pitch-class content of the same octatonic scale, simply b
note of the scale its symmetrical complement, a sub
realized at four different transpositional levels (Ex. 33).1
Five Preludes must surely have been well on the road to
meaning of inversional equivalence.
Ex. 33
Sum 7 Sum 10 Sum 1 Sum 4
Would Scriabin's experience in a post-diatonic musical l

led him, as Schoenberg's eventually did, to the concept of
structure embracing the totality of pitch classes - the twelv
No. 4 suggests that he may have been moving in this dir
sketches for his projected, but never realized, 'Prefator
Scriabin originally conceived as a prologue to what was inten
mankind's, ultimate aesthetic experience, the 'Mysteriu
direct evidence.'" (Preludes Nos 1, 2 and 4 appear to b
GEORGE PERLE
projected work, since fragments of them appear among

we find a vertical statement of the twelve pitch clas
striking out octave duplications in a 16-note chord (Ex.
'major 7th' chords at successive t3 transpositions (pp. 6,
Ex. 34 Ex. 35
| t1
-00
On p. 14 there is another vertical statement of the twelve pitch classes (Ex. 36),
one that is exactly analogous to a twelve-tone set devised by Berg some fifteen
years later for his opera Lulu. Reading upward from the lowest note we find an
octatonic collection partitioned into two 'French 6th' chords plus the 'dimin-
ished 7th' chord that is not comprised in that collection:
Ex. 36
A5. "
9s
A =
What I call 'Trope III' in Lulu (Ex. 37) comprises a basic cell that correspond
to the 'French 6th' in that it also consists of two tritones,20 the same cell at the
(or t3) transposition - the two together give the octatonic collection - and the
'diminished 7th' chord required to complete the twelve-tone aggregate.
Ex. 37
A t i~
~) r
The messianic obsessions of Scriabin's

question as to what he might still have ac
deal, in spite of his quirks and delusions. I
the evolution of an autonomous and coh
long delayed because of his early death.
NOTES
1. Arnold Schoenberg, Style and Idea (London: Faber, 1975), p. 227.

2. Faubion Bowers, The New Scriabin (New York: St Martin's Press, 197
3. The consequent compositional implications are discussed in my boo
Composition and Atonality (Berkeley: University of California, 1981, fifth
pp. 41-3.
4. The octatonic scales are named in Ex. 8 according to the terminolo
identification of interval cycles proposed in my book, Volume Two of The
Alban Berg (Berkeley: University of California, 1984), pp. 199f. The
followed by an interval-class number identifies the cycle; a subscript ide
transpositional level of the specific cyclic collection by the pitch-class num
C, 1 for C#, etc.) of one of its elements. Thus 'C30' designates the 'd
7th' chord that contains pitch-class C, 'C31' designates the one that conta
class C4, and 'C30,1' designates the octatonic scale that contains both
may also be represented as 'C31,3' since the same 'diminished 7th' ch
contains C also contains Eb. I have preferred to show a difference of a w
rather than a semitone between the pitch-class numbers because thi
correspond to Scriabin's usual conception of the relation between t
components of the scale.
5. The durational term indicates the portion of the bar comprised in the cit
6. The Romanian musicologist Adrian Ratiu, in 'Sistemul armonic al lui
Muzica, Vol. 22, February and March, 1972, also describes the F in b.
anticipation of the second octatonic scale; see Roy James Guen
Examination of Analytical Approaches to Harmonic Organization in
Piano Works of Alexander Scriabin', M.A. diss., The Catholic Univer
America, 1974, pp. 123-4.
7. See Paul Lansky and George Perle, 'Atonality', in The New Grove Dic
Music and Musicians (London: Macmillan, 1980).
8. Felix Salzer and Carl Schachter, Counterpoint in Composition (N
McGraw-Hill, 1969), pp. 215-19.
9. N. A. Rimsky-Korsakov, My Musical Life (New York: Tudor, 1936), p
10. In Nicolas Slonimsky, Music Since 1900 (New York: Coleman-Ross, 19
edition), pp. 680-1.
11. Perle, op. cit., pp. vii-ix, 15, 38-9, 49-50, 121 (n.6); 'Berg's Master Ar
Interval Cycles', The Musical Quarterly, Vol. 63, No. 1, January 1977,
GEORGE PERLE
Twelve-tone Tonality (Berkeley: University of Californ

162-72; The Operas of Alban Berg, Vol. 1: Wozzeck, pp.
Lulu, pp. 161-6 (Berkeley: University of California, 1980 an
12. Perle, Serial Composition and Atonality, pp. 38-9; The Ope
pp. 155-8.
13. Faubion Bowers, Scriabin, Vol. 2 (Tokyo: Kodansha Int
14. See Perle, review of Schoenberg, Style and Idea, in MQ, V
pp. 435-41.
15. Perle, Serial Composition and Atonality, Chapter 2, passim
16. See Elliott Antokoletz, The Music of Bjla Bart6k (
California, 1984).
17. The given inversional relation is identified by its 'sum of
fixed sum, mod. 12, of complementary pitch-class num
Master Array', p. 7; and Twelve-tone Tonality, p. 2.
18. See the brief but important article on the sketches
esquisses musicales de L'Acte Pr alable de Scriabine', Re
57, 1971, pp. 40-8. I am indebted to Faubion Bowers for br
the 'Prefatory Action' to my attention and for his generous
19. The brackets in Exs 35 and 36 are my own addition. Each
Ex. 35 unfolds an octatonic collection. This interpretat
where we find Ex. 35, in an enharmonically equivalent
second chord comprising its alternate elements spelle
Db(b), Abb, E,, Bb, F#, C#, Gx, Dx. That the cont
pitch classes in the 12-note chord of Ex. 34 is not coinc
Scriabin's annotation, '12', below the chord. It is follo
which is evidently conceived as its resolution, a 12-note cho
(not the 'octatonic collection', however).
20. This basic cell, the chief motive and pitch-class structu
well-known to Webern and Bart6k. See Perle, Serial Compo
ed., pp. vii-ix, and Twelve-tone Tonality, pp. 11-2 and 1

Wiley Music Analysis: This Content Downloaded From 128.179.146.121 On Wed, 10 Jan 2018 13:14:33 UTC

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Wiley Music Analysis: This Content Downloaded From 128.179.146.121 On Wed, 10 Jan 2018 13:14:33 UTC

Enviado por

Direitos autorais:

Formatos disponíveis

Scriabin's Self-Analyses

Author(s): George Perle

What authority does a composer have as analyst of his own music?

MUSIC ANALYSIS 3:2, 1984 101

Schoenberg, and almost at the same time, a diatoni

Ex. 1 Ex. 2 Ex. 3

The revised spelling, unlike our example from Beethoven, implies no

Since this scale consists of two 'diminished 7th' chords, the

102 MUSIC ANALYSIS 3:2, 1984

the same spelling, regardless of transposition (Ex.

are all derivabl

MUSIC ANALYSIS 3:2, 1984 103

are identical in pitch-class content, there are only three

The tonal vocabulary is still frustratingly restrictive, however, for a musical

104 MUSIC ANALYSIS 3:2, 1984

That this is the primary, though not the sole function of t

Ex. 10: Sonata No. 7

fres pur, avec une profonde douceur

.. m-,, " '--"

Each of the Five Preludes is a

MUSIC ANALYSIS 3:2, 1984 105

Ex. 11: Op. 74, No. 1

"-'--s ' "' ' ' fr

traditional harmony, and both are instances of th

106 MUSIC ANALYSIS 3:2, 1984

second lower octave displacement, finally takes its plac

PART ONE, SECTION ONE

1: G A Bb C D6, E6 F G A Bb (C) D, E6 F IA B (C) D E, F G

PART TWO, SECTION ONE

11: ABC DEbFG ABC(D)Eb FG I BC#(D)EFGA

PART TWO, SECTION TWO

MUSIC ANALYSIS 3:2, 1984 107

commencing at the original pitch level and shifting after

Ex. 13: Op. 74, No. 5

bow -- ... .. .. .... .......................

irrl' .mDO 6m-,6"

In b. 2 and at the corresponding point

108 MUSIC ANALYSIS 3:2, 1984

contrasting episode, bs 5-8, based on the unaltered form o

1-4: Tonality C21 (Cb Db Eb F G A B C# D# E# Fx)

The scale is shifted to C32,4 in bs 15-16:

There is a return to the 'home key' in bs 19-20, a tll transposition of bs 15-16:

MUSIC ANALYSIS 3:2, 1984 109

The C31,3 octatonic scale is maintained for the remainder

The modulation from C31,3 to C32,4 is effected in bs

Ex. 15: Op. 67, No. 1

C31 (E G Bb Db) is shared by

110 MUSIC ANALYSIS 3:2, 1984

bizarre and eccentric to an extreme, with its spelling of a

Ex. 18: Op. 74, No. 2

Tres lenf, contemplatif

MUSIC ANALYSIS 3:2, 1984 111

In Prelude No. 5 the infiltration of a non-octatonic element is exploited for its

112 MUSIC ANALYSIS 3:2, 1984

In the totality of their variant forms all three octaton

C31,3: B D# GB Eb, B E G CAbC~, F A Db, D F# Bb Ebb Gb

C31,3: A# B/B# C# D/D# E F# G/G# A (Bb) C

C32,4: D Eb/E F Gb/G Ab Bb Cb/C Db (Ebb) Fb

Ex. 21: Op. 74, No. 4

In Ex. 23 the apparent notational inconsistencies (C# and C

C31,3: B CQ (D#) E F G# A (Bb) C Db

C 31,3 C 30,2 133,3 C30,2 C 31,3

transposition of b. 2 and rhythmic augmentation postpone the final ret