Society for Music Theory

Tonal Focus in Atonal Music: Berg's op. 5/3 Author(s): Christopher Lewis Source: Music Theory Spectrum, Vol. 3 (Spring, 1981), pp. 84-97 Published by: University of California Press on behalf of the Society for Music Theory Stable URL: http://www.jstor.org/stable/746136 Accessed: 20/09/2010 17:12
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ucal. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. 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.

University of California Press and Society for Music Theory are collaborating with JSTOR to digitize, preserve and extend access to Music Theory Spectrum.

http://www.jstor.org

Tonal

Focus op. 5/3

in

Atonal

Music:

Berg's

Lewis Christopher
". . . for withoutrulestherecan be no art, any morethantherecan be a house without any doors to conduct you into it." -John Dryden In a beautifully turned metaphor, William Benjamin has spoken of the early atonal works of Schoenberg and Webern as "mysterious houses . . . built. . . without doors."1 Entry must therefore be forcible, effected not with keys, but with tools to aid the dismantling of part of the structure. But even if this is true, it is not, it seems to me, equally valid a point of view from which to look at the atonal music of Alban Berg, which almost always represents an attempt at fusion of tonal and atonal principles of organization, and therefore provides entry by key-in both metaphoric and literal senses-to at least some of its rooms. To be sure, this music is not tonal in the common-practice meaning of being structured around the prolongation of a background triad. Twenty years ago, Roy Travis was applying the term "tonal' to music which "unfolds through time a particular tone, interval or chord."2 If one adopts the position that by
1William Benjamin, "Ideas of Order in Motivic Music," Music Theory Spectrum 1 (1979):23. 2Roy Travis, "Towards a New Concept of Tonality," Journal of Music Theory 3 (1959):261.

"unfolds" Travis means "prolongs"- surely the only possible interpretation of that word applied to a single tone--and that prolongation may be effected by motion between two statements of the tonic sonority, or by motion either from or towards a single statement of that sonority,3 then one may understand as tonal any work in which any pitch or sonority is made the focus of harmonic or linear motion. Travis calls such a procedure "directed motion," and shows its operation in Schoenberg's op. 19/2 and Webern's op. 27 as leading to a "tonic sonority" and a "tonic dyad" respectively.4 It is in this sense that '"tonal" is understood in this paper. The pre-serial works of Schoenberg and his students are sometimes termed "atonal," and sometimes "motivic" music. Neither term is entirely satisfactory; the latter is so general as to be almost meaningless, and the former may be misleading since lack of tonality is not, apparently, an absolute requisite of this music: Schoenberg's own objections to the characterization "atonal" are well known. The most important and extensive attempt at formulation of an atonal syntax is Allen Forte's Structure of Atonal Music, 5 but even though Forte disallows the
3See Felix Salzer,Structural Hearing (New York:Dover, 1962) vol. 1, p. 16. 4See Roy Travis, "Directed Motion in Two Brief Pieces by Schoenbergand Weber," Perspectives of New Music 4 (1966):85-89. The term is borrowed from Salzer, StructuralHearing, p. 11 ff. 5Allen Forte, The Structureof Atonal Music (New Haven: Yale University

Tonal Focus in AtonalMusic

85

it considerationof tonal sets in the analysisof atonalstructures,6 set relations may not has not been shown that his significant operate simultaneouslywith some tonal processes.7 Forte himself has traced Schoenberg's growing setconsciousness-the increasingrelianceon the creationof structural and surface coherence by means of significant set relations- in the works datingfrom about 1904 to those immediatelyprecedingthe ThreePianoPieces, op. 11. His discussion sketchof much of "Lockung," op. 6/7, includesa Schenkerian of the song, and identificationof passages which "althoughnot completely organized in the manner of the fully atonal
piece . . . do exhibit certain structural characteristics that re-

semble those of atonal music."8 can Since the two principlesof organization coexist in such a that set organization is subordinateto triadic tonality, way surely the reverse may also occur, so that vestiges of tonal hierarchy appear in an atonally structuredwork. Forte has analysed Schoenberg's op. 19 as an atonal work; Travis and Westergaardhave given sketches of the second and fourth pieces respectively, showing in each case directedmotions to a tonic sonority.9 Are these different views of the same music
Press, 1973). My analysis presumesthe reader'sfamiliaritywith Forte's theories and terminology. A valuable summaryand commentaryis David Beach, "Pitch Structureand the Analytic Process in Atonal Music," Music Theory Spectrum 1 (1979):7-22. 6Allen Forte, "Sets and Non-sets in Schoenberg'sAtonal Music," Perspectives of New Music 11 (1973):45. 7Forteis careful to define his use of the term "atonal" in terms of what it is characratherthan what it is not: "Any compositionthatexhibits the structural teristicsthatarediscussed [inTheStructureof AtonalMusic], andexhibitsthem as of may be regarded atonal.'"(TheStructure AtonalMusic, p. ix.) throughout, 8Allen Forte, "Schoenberg's Creative Evolution:The Path to Atonality," Musical Quarterly 64 (1978):133-176. 9See Allen Forte, "Context and Continuity in an Atonal Work: A SetTheoretical Approach," Perspectives of New Music 1 (1963):72-82; Roy Travis, "Directed Motion in Two Brief Pieces;" and PeterWestergaard,"Toward a Twelve-Tone Polyphony," Perspectives of New Music, 4 (1966):93.

necessarily incompatible?I think not. Further,I think-to pervertBenjamin'smetaphor-there aremanysuch musicalhouses with bothfrontandbackdoors, neitherof which providesaccess to the whole structure,but which togetherallow one to enterall the rooms. If we are to recognize this duality of pitch organizationas a thana lack of controlof one of truearchitectural principlerather the elements, then it must be shown to be not simply a resultof incidentalsurfacerelationships,but a complex of proceduresso deeply imbeddedin the musical fabricthatit serves to articulate the structureof the piece. It will not be enough to find an occasional symmetrical,linear or cadentialtonal focus, or occasional atonalcharacteristics.But if the beginningof the piece establishesdirectedmotions, or tonal focus, the implicationsof which are developed as the music unfolds, and if at the same time significant set relationsarticulateand unify the structure, then the interaction tonal andatonalelementscan be accepted of as a legitimatecompositionalprocedure,andfurther exploration of this proceduremay well illuminatesome shadowy pages of the atonal literature. The thirdof Berg's VierStucke, op. 5, for clarinetandpiano is a particularlyrewardingexample of the operationof tonal focus in an atonalwork. The following discussionexaminesthis duality of means and considers surface implications and the mannerin which both aspectsof pitchorganization contribute to the structureof the piece. Contrastsof surface features-of tempo, texture, register, density and melodic gesture-divide the movement into four parts (mm. 1-3, 3-8, 9-13, 14-18). Each of the four parts that presentsa single gestureby each of the instruments; gesture may always be considered to have two parts, the second of which is a continuationor an elaborationof the first. Comparison of the clarinetgesturesreveals a similaritybetweenthe final segmentsof PartsI and III andbetweenthose of PartsII andIV. In bothcases, the laterversionis moreelaboratethanthe earlier,

86

MusicTheorySpectrum

Alban Berg, Vier Stiicke, op. 5/3 (1913)

III.
Langsame

Viertel.

I.

OC" epre

,-

Sehr rasch. Klarinette IBpbinB

-c..

PpP legatissimo (poco rit J I I-. p_'-titJ t .spr. p,

tr

~"
__

Ia' "

*;

1PP>--

-f?~

'

5I t ----pp

=-.

Klavienr

O? P1' Gi' rl....

tep

d,n>

Viel Ped.

sehr .-empr,

Rit.

poc rt. ,-

et,E

Etlwas langsamer,

aher immer

-_ '

__

rz'J

I
a tempu

^ .

IV rt

la Inz ' shr

tefmp",-

. 2

4 .., ' C$

S r

^ -P ^

'

4-

'

v7

iIT

l
(p

:w~. -^ :

/'/
lim -__--_--_.so

" ^it^ia G^ ---..-L~ ii, ; PM

rasch unt t

qi^
d leitse

~ nmuglich, als

t.n.
PPP

=
^:t9z ii"

|y.<Lcx

.c '. poco crac. S " i Y- y _PP _ ?

(t

y^

"^'*lQ~~~~ -^_'~ -~~^ir~~~

aeg /-Stt

'

oh*

e Ped.

Ics

mgih

sehr rasch.(. <;.

iK- 9)

rit. pOCO

Pl> , PI),

,
Flatterzungv. . ~~

1 Ppp)

ziemlich

[] |

ni

--I rImmer
PP
.^-?-^1;' h(tt ^

noch rascher

la,PPIPE,,h,,t,,n

~'"~, , :l._l/l~----"~l~

'':'

A.

u ,n i^t) qluasi Flatterzungc. dimin. "P

,I....
Fiz. (.pp))Echoton -: , o P I^ r" ,iemlice

i-s,i Flat,l,7ung,, P'" / ^J PPPP /tu pp

_v1~
-

t
' ? 4', '

o'

dipn
.,,.

f*
-

"
I. H I -

o
-

rw
$

H4 _

4H-;- P7q' "VP` ar_

t ,& . 1(.bAnde

*=d

bA'

-s PPPo.P wea

tj-

r .. _.- ......

: IK

,4
P!iant

ctA_k<

--r

"^ .1

.
J...t"mrr

I 1\
_ . ' '

\
/

I
-

I, 4
-

I..;'I

71

Zsempre -p

Copyright 1924 by Universal Edition, Renewed Copyright 1952 by Helene AmericanMusic. Dist. Corp., sole U.S. Berg. Used by permissionof European agent for Universal Edition.

Tonal Focus in AtonalMusic

87

and reaches a cadence on D. The effect is of resumptionand completion of the earlier, interruptedmotions. This sense of completionis not incidental,for we shall see thatthe focus on D is established in the opening measures and continues until the final resolution. The piano statementof Part I is framed by two augmented triadsbelonging to the same whole-tone hexachord.The material between is, with the exception of the first and penultimate notes, partitionedby rhythm and articulationinto sequential statements of the chromatic trichord(set 3-1). However, the thusformedin mm. 2 and3 unfoldsa whole-tonedescent pattern throughthe same hexachorddefined by the framingtrichords. The bipartite nature of the whole gesture is created by the surface articulationand the eighth rest of m. 2, and by the contrast of a half-step descent in m. 1 with the unfolding whole-steps in mm. 2 and 3 (see Example 1). Continuity is providedby consistentuse of set 3-1 in the upperpartandby the concatenationof transpositionsof that motive to form a chromatic descent from el to d in the lower. This structureis paralleled by that of the clarinet line, the

binary division of which is created by the surface articulation (here, however, it is the firstsegmentwhich is a continuousline, andthe second which is fragmented)andby the firstrest of m. 2 coupledwith the cadentialaspectof the precedingd2.The whole of this line, then, can be understoodas the unfolding of the augmentedtriad with which the piano cadences (see Example 2). It is apparentalready that there is some sense of pitch hierarchy, and that D-especially accompanied by its augmented triad-is of prime importance.We have alreadymentionedthe cadentialimportance the d2in m. 2; this pitch is the of goal of the preceding ascent and is approachedfrom its upper leading tone. Furthermore, rhythmicand articulativeemphasis strengthenthe functionof this note as the mid-pointof the line F andof the lower-level arpeggiation --D-Bb. But the pitchD is also the eventualgoal of the descent in the upperstaff of the piano part: it is at d' that the pattern is broken and the D transferred the bass. One might also takethe view thatthe c#2 to of m. 1 is a registralpreparation the cadentiald2 in the next for measure(see Example 3).

1 Example

88

Music Theory Spectrum

Example 4 Example 2

--A
__

,
Is FfIF --r

1
-

r
I

-.

-,r--

--

_ _ b--

__-

Example 3

I _b__

S4ii_.2~z~
-9:

>?~

The surface motives are almost trivial. The three-notechromatic ascent of m. 1 of course generatesthe whole of the piano part; the first repetitioneffects a half-tone descent (g#1 to gl) which is mirroredin the lower part (cl-c#1) and then occurs directlyor indirectlyfive times in the clarinet(see Example4). An analysisof the primaryset formationsgives a morecomplex picture(see Example5). The melodic lines featuresets 3-1 (not shown in Example 5), 3-2, 3-3, 3-4 and 3-12 (the augmented

triad). The harmonictrichordsin the piano (m. 2) presentthe following set sequence: 3-3, 3-10, 3-11/3-4, 3-11, 3-11/3-5, 3-11, 3-9/3-9, 3-11, 3-12. The first nine trichordsinclude fully half the possible numberof trichord types, butthereis a discernible order. Note the association of the first attack of each sequentialunit with a "small" set contrastingstronglywith the triadicsets which follow. These initiatingchords(3-3, 3-4 and 3-5) themselveseffect an expansionwhich reflects the envelope the of the entire clarinet-piano gesture. Furthermore, nine-note sets formedby all threechordsof each unit are complementsof and trichords; in the case of the last two, the superset important is the abstractcomplementof the initial trichord.Althoughthe by uppervoice descends througha whole-tonescale articulated the three "small" sets (see Example 1), the bass unfolds throughcl, a, f#, d: the set 4-27. When the sequentialpatternis at last broken in m. 3, the concludingtrichordsform an eightnote superset;thatset is 8-27, the abstract complementof the set just arpeggiatedin the bass. Indeed, furtheranalysis shows that these two sets, lurkingjust beneath the surface of the music, permeatethe entirefirst partof the movement(see Example6). Of the eight statementsof 4-27 shown in Example6, two pairs are identical(sets C andG andsets B andF). They arepresented in the music in such a way that set C is associated with set B (they are arrangedin a melodically conjunctsymmetryaround

Tonal Focus in AtonalMusic

89

d2) and set F is associated with set G (both are derived by imbrication -supported by the sequentialpatterning-from the final eleven PCs of the left-handdescent). It follows, of course, thatthe composite sets F+G and B+C mustbe equivalent,andit is surely significant that that set (7-32) appears as a primary segmentas well. Also buriedjust below the surfaceof the music are three interwovenstatementsof set 5-31, a supersetof 4-27 (see Example 7); furtherreference will be made to this set in discussion of Part III.

We have seen thatthe note D is symmetricallyframedby the subcomponentsets B and C, by a backgroundaugmentedtriad and by the cadence chord of m. 3. But these symmetries are foreshadowedas early as the openingtrichord.In addition,there is an even more subtle framing of the D, since the clarinet's semitone cadence on D in m. 2, as well as being the centerpoint of sets B and C, is precededand followed by statementsof set 5-10 (see Example 6).

5 Example

6 Example

A
" <J ^ 10 " , 1 __ . i 1 '

I~ (j0-

~ ~ ~ ~ ~~~_7i:L a -__J V
I

__

/ ___ ___ _^*bl^k^

___H_____

9:

l.. .
_

A,B,C,D,E, ,J, =8-27

4-27 F,G,H 1,K=--7-32

90

MusicTheorySpectrum

7 Example
5-31
A

8 Example

v
'

VP, =I 5-31 5-31 =

II

T-

I_

10~__ _:__J _1-rnZz_74-14 C ' __-9

i~

-x i - 14p - 1

-ii

There are, therefore, at least the following three ways of organizingpitch withinPartI: conjunctmotionsthroughconsistent whole-tone or half-tone segments, tendingtowardsD; formation of primarysegments and subcomponentsin abstractor literal symmetry aroundD; and considerableuse of a limited and numberof important clearly articulated pitch sets (and their complements). Part II, like Part I, consists of two bipartitegestures. The clarinet line is divided by contrasts of direction and motive, at reinforcedby a change of timbre(the fluttertongue) the ff1 of m. 6. The initial segment of this phrase is constructedof two whole-tone motives (A, B, F and C, E, F#); the second segment is a brief chromaticdescent. The piano statesthreechords of diminishing cardinality, ending with the tetrachord4-19. That these chords should be consideredas belonging to PartII ratherthan to Part I, as might be assumed from the tempo and, more importantly, marking,is indicatedby the articulation by a set connectionbetweenthe melodic set which overlapswith the first note of the clarinet entry (and which is stated three times) and the harmonicsupersetformedby the last two of the chords. In addition,thatsupersetplus the firstclarinetnote is, of course, the complementof the three-notemotive of the upper line of the piano chords (see Example 8). 4-19 is a supersetof the augmentedtriad Now, the tetrachord which played an importantrole in PartI. The inclusion (3-12),

relationship is exploited by the separation of the trichord (7,11,3) from the melodic upperpart so that it may assume a subsidiaryfunctionas a separateharmonictrichord.It is at this point that the second part of the piano gesture begins, to be concluded by an inner-voice movement to a symmetricaltetrachord.The linearmotionsof this part,rather moresimple than those of Part I, are sketched in Example 9. The focus on D is now very much less evident than in the first part. Indeed, the only reasonsfor assumingthatthe clarinetdescent from F* to E stops shortof D as a goal arethe similarityof the gesturewith the of pianoline of PartI, andhindsightafterconsideration PartIV, below. But what little linear motion there is in the piano is clearly directed towards the establishment in mm. 7-8 of a tetrachord symmetricalaroundD. Althoughthe D has thus been note in the only weakly implied, Ab is isolated as an important contourof the upperpiano line, and is left unresolved. If one thinksof PartsI andII for a momentonly in termsof theirlinear motionstowardsD, thenin PartI the pianogestureis closed, and the clarinetgesture is open; in PartII, both gestures are open. One of the roles of the remainingtwo partsof the movementwill be to resolve the Abof PartII andthe implicationof D presented by the final tetrachord,and in so doing to provide closure for both lines.

Tonal Focus in AtonalMusic

91

9 Example

I A

'--

111_^~

..^-

__ t

' Xp . r . -

Althoughthe linear motions of PartII are simplerthan those of PartI, the surface set content is not. We have alreadynoted the coincidence of set 4-14 and its complementat the beginning of the section. But these sets, like many of the primarysegments, appearvery few times. Thereis thereforea considerable variety of set types articulatedby the surface motion (see Example 10). The set 4-19 is clearly crucial;it appearsnine times (set Anin Example 10), always with a common subset (7,11,3), and establishes a harmonicpedal for mm. 5-7. Even after its final direct statement, 4-19 continues its influence: first as three subsets ((0,4,7,8); (8,0,3,4); (4,5,8,0)) of set 6-14 (set K in Example lOa), andthenby implicationthroughits complement, which is formed by all the pitches stated after the final direct occursas a subsetof of appearance 4-19 itself. Set 4-19 naturally a numberof primefive-note sets. In fact, the only five-note sets in this passage of which it is not a subset are sets I and J, which occur only once each, at the end of the section, and thereforeas subsets of 8-19. The diversity of set types is thus counteracted between sets C, D, E, by the very strongsimilarityrelationships F, G andH. All except F andH areinterrelated the significant by (strongintervalvector similarityand maximum similarityR2Rp pitch similarity-in most cases stronglyrepresented).The ex-

ceptionsstandin the relativelyinsignificantrelationship to all Rp the other five-note sets, but not to each other. The three sets most musically prominent are the Z-pair 5-Z17/5-Z37 and 5-21. The two members of the Z-pair of course sharethe same vector;5-21 is the only set which is R2Rp related to both 5-Z17 and 5-Z37. These closely related sets in accountfor seven of the nine harmonicpentachords mm. 6-7. m. 6 presentstwo overlappingstatementsof 7-21, Furthermore, so thatjust as the chordal"transition"is unifiedby 4-14 andits complement, so is this partof the gesture unified by 5-21 and 7-21. Example 11 shows the hexachordalsegmentationand a few other sets. The hexachordsof this partare not as significant as the tetrachords pentachords.Thereis relativelylittle repetiand tion of sets; the complementrelationis not exploited(none of the Z-pairsis represented both members);the chartof similarity by relationsreveals minimalconnection. However, because of the persistenceof certainfour-andfive-note sets, the intersectionof adjacent hexachords naturally produces familiar sets as invariants.These are 4-19 as the common subset of C2 and E, 5-21 as the commonsubsetof E andD2, and4-19 as the common subset of D2 and F.

92

Music Theory Spectrum

3-8

A (8-19) C2 C3 D3 /E\ H

. m Example lOa

Fl Di G

F1D2

F2

'C

/ -r

[1
/

iL

; .- -

U -I;
-

F IA_ _,__ |Ul__-

Al

A--

"S-T--

C4-

Example lOb

C (7-21) C(7-21)

X
8-7 \

C (7-21) C(7-21)

A-4-19

B-4-24

C-5-21

D-5-Z17

E-5-Z37

F-5-13

G -5-22

H-5-26

I-5-20

J-5-Z18

K-6-14

Example 11

'J(6-18)

<A(6-34)
-

- .

74V_^i

Tonal Focus in AtonalMusic

93

Intersection of the final two hexachords produces the invariant tetrachord 4-7, andthe precedingtwo measuresconsist of overlappingstatementsof its complement, 8-7. But the four non-invariantPCs representset 4-14, which, with its complementwas seen to controlmm. 3-5. Thus, the final section of this gesturereferssubtlybackthroughstatementof the complements to the eight-note sets associatedwith the transitional chordsand the first part of the gesture proper. Therefore,PartsI and II of the piece contrastnot only on the surface, but also in terms of the type of pitch control most prominent.PartI is controlledmost clearlyby linearmovements towards D and by certain set relations, but Part II exhibits a strong set control with subsidiaryorganizationby symmetrical aroundan impliedD and-very tenuously-by linear structures motions towards D. Part III is dominated by a pitch ostinato from which are derived all the principalmelodic lines. Again one can consider that each line is composed of a bipartitegesture. The partitioning is effected by the change in the ostinato which occurs on simultaneouslyin both instruments the last beat of m. 11;the clarinetdoes not repeatits figure, but continuesupward,while the piano bass ostinato is broken (but does not entirely disappear) at the momentthe upperpartsabruptlychange direction. The minorthirde2-g2statedfirst in the piano in m. 10 and then twice repeatedand expanded,recursin the clarinetand initiates a cadentialline reminiscentof thatof PartI. The correspondence is not precise, but is unmistakable,and even incorporatesthe motive of a falling second (see Example 12). The alterationof the minorsecond to a majorsecond is a resultof the reordering of set 5-31, which now generatesthe ostinatofigure (see Example 13). But although the clarinet line can thus be understood as providingclosure for the open cadence of PartI, the same is not true of the piano gesture, the cadence of which is-in termsof motion towardsD-very much open. The descent of the upper line stops on Eb, which is not permittedto resolve. The incom-

12 Example

[*]
p I Part

-L^Wl^x
,

-x
",

L--'" x .. .. .... t ? .I

X : ....,""'

A bn
Part'I Part II

-=z

_- I
x

I
x

I
x

13 Example
5-31

Lim.10 e m1]O , d

^ bog

5 -31

plete natureof this Ebis obvious when one considersthatevery cadentialresolutionto D thus far has been from either an upper or a lower leading tone. Because so much of the pitch contentof PartIII is controlled by ostinato, the prime set content is limited. Set 5-31 is most common, and is represented its complementas well. But for by the first three measuresthere is little other significantexploitation of set relations. The sole exceptions are the gradualevolution of set 4-Z29 as a subset of 5-25, and severalrecurrences of

94

MusicTheorySpectrum

4-27 (the crucial set of PartI) (see Example 14). The cadential measures, on the other hand, not only present two last statements of 4-27, but do so in such a way that they occur as of subcomponents 8-27, which is given a perfectly straightforward melodic statement.Thus the connection with Part I discussed above as a linearphenomenonis establishedalso through is set design. Even moreremarkable the set connectionwith the sets two most important of PartII: the cadence chordrepresents set 7-21, andthatchordplus the cadentialleading tone gives set 8-19. Both 4-19 and 5-21 occur strongly represented,as the uppermostfour and five notes respectively of the eight-note cadential sonority (see Example 14b). One notes as well the of reappearance the chromatictrichord(3-1) in an innervoice, and the use of 3-5 as initial and final trichordsin the piano bass line. PartIII, then, commences the process of closure of unresolved elements from the first half of the movement. The incomplete cadentialrise towardsd3 (mm. 1-3) is resumedand completed, and the cadentialsupersetincludes primarysets of each

of the first two parts.The elementsstill to be resolvedare(1) the Abof PartII, (2) the incompletedescentfromF# to D, (3) the Eb upper leading tone of Part III, and (4) the implication of D throughsymmetricalformationsin PartsI and II.

14b Example

7-21 (^_ ( ~13

A ----

__<\

4-19

LJ_

A):

Ir^^W^~~1 ;
8-19

14a Example
7-31
4-27

[i.1
-^

ri_5I /4'4

-Z29j

5;25l

--T"^^^-^^^:=^^^-^^^_

7-31
5-31

-31
-4

_
29

-31

' ry~ i;r^~~~ '-1

b4-27 ^~~~~~~ ~
-

8-27

-5
6 4) ====v-^ = r^ __

-25

-3

1 1

A~416
*X
+ s,--_L4

_---__,

i___x

4~ . t~~*-16~

,ft iT -: -l__

Tonal Focus in AtonalMusic

95

gesture Again in PartIV it will be seen thateach instrumental is bipartite.The clarinetdescends througha whole-tone scale to the beginning of m. 17, after which point movement is by semitone. It is significantthatthis change commences with F#, for the following line may thereforebe understoodas a resumpas motionto D enumerated tion and completionof the truncated a repeatedmajorthird, (2) above. The piano gestureconsists of octaves leadingto underwhich the bass dropstwo-and-one-half The firstof these segmentsis tetrachord. a repeatedprecadential an expandedreferenceto the initialthreenotes of the movement, and, like them, establishes a symmetryaroundD (see Example 15). The octave displacements do not disguise the semitone ascent from D to Ab, and the functionof thatline is thereforeto re-establishthe Abof PartsI and II. The final clarinetscale, by preciselyreversingthis line, effects the resolutionof the Abto D (see Example 16).

16 Example

(Clarinet) [6

~(Piano) [l]

Example15

Part II

Part

Part I

The c1 and el of m. 14 frame D as did the symmetrical of structures PartsI and II, and especially the cadentialchordof m. 8. The functionof the evolving pianochordsof mm. 15-16 is to realizethe impliedfocal pointby meansof symmetricallinear movement to a chord which is perfect symmetricalaround a stated d' (see Example 17). The final chord appearsto disturb this symmetry. As Example 17 shows, however, it is derived fromthe penultimate sonorityalmostentirelyby octavetransferral:only the D does not recur,but insteadmoves by semitoneto its upperneighborEb. This pitch has alreadybeen preparedby the clarinet.In effect, it is transferred fromthatinstrument the to piano and then back to the clarinet for resolution as an upper neighbor to D, which may be taken as resolving both the Eb which is the only asymmetricalPC of the cadentialchord, and the Ebwhich has been left unresolvedsince the open cadence of m. 13. Therefore, Parts III and IV have resolved all of those elements and tendencies, whether of set structureor of linear motion, left open by PartsI and II. In view of the summational that of character the final supersetof PartIII, it is not surprising PartIV devotes itself entirely to the resolutionof other aspects and apparentlycontains no significant set relationships. It is immaterialto the musical point of the analysis whether one defines one's terms so as to justify referringto the piece as either "tonal" or "atonal." Since the work is subjectto unification by several systemic procedures,exclusive use of either

96

Music Theory Spectrum

Example 17

I
-

^I----

N I

I
w

l_

---

term obscures the fact that neither set-theoretic analysis nor consideration of tonal or quasi-tonal elements can by itself adequatelyaccount for all aspects of pitch structure. I do not for a moment wish to suggest that all atonally structured music will be illuminatedby attemptsto read into it tonal focuses of one kind or another.As I implied at the beginning of this article, I agree that Webern's music in particular avoids such implications. But neither do I think the piece we have just examinedneed be consideredatypical. One thinksof EdwardCone's remark-made in the context of a discussionof certainlinear,chordalandcadentialaspectsof Schoenberg'sop. 33a-that he cannot"see how music like Schoenberg's,with its can usually clear cadentialstructure fail to arousecertaintraditional associations and responses." 0 I shall invoke two brief examples. GaryWittlichhas given a convincing demonstrationof the pitch-class set structurein Schoenberg'sop. 11/1; he has notedalso thattheredoes seem to be some conscious creationof pitchhierarchy,andthatG, B and Ebare especially important.l1 remarksthat Ebis studiously He
10Edward Cone, "Analysis Today," TheMusical Quarterly46 (1960): 185. 1 Gary Wittlich, "IntervalSet Structurein Schoenberg's Op. 11, No. 1," Perspectives of New Music 13 (1974):41-55.

avoided in mm. 1-11, but when it does appear, it does so prominently,in two octaves. But we hear clearly that the succeeding section, mm. 20-25, contains a number of exposed melodic statements,each of which begins on Eb. Also unmistakableis the cadentialarrivalin the bass in m. 47, precededby severalEb-Emelodic statementsarticulated surfacefeatures. by And mm. 49 and 50 again bringEbinto prominence.(Believers in the Golden Section may calculatethe division for themselves and see what happensat that point!) And, of course, the piece ends with Eb. These observations not constituteeven the beginningof an do but they do stimulateone's curiosity. The Ebis funcanalysis, tioning as an auralpoint of reference;the relativeprominenceof that pitch waxes and wanes in close correspondencewith the formaldivisions of the piece. Might furtheranalysis revealthat this pitch, perhapsin conjunctionwith G and B, is made the of point of linearconnectionsoperatingwithin the strictures the atonal structure? The second piece of the same opus presentsa very different face, for here it is pointsof detail which promptone to examine the broader picture,ratherthanthe otherway around.A preliminary analysis has suggested to me that the piece is as tightly organized in the atonal sense as is the first piece. But I hear the throughout piece clearly defined tonal focuses. Let us con-

Tonal Focus in AtonalMusic

97

siderfor a momentthe openingnine measures.Thereis certainly a sustainedD in the bass, but I agree with Philip Friedheimthat this alone does not make the pitch class a "tonic."'12 However, the upperpart is directedagain and again at Db, often through Eb, and often in conjunctionwith Ab. It is easy to hear both Eb and Dbas semitonedisplacementsof D, especially at the cadential resolution of m. 6. But the succeeding measures again establishandprolongDbandAb,before, at m. 9, thereis another cadential resolutionto D. Is Db displacing D, or vice versa? I thinka carefulstudywill revealthatin fact DbdisplacesD which in turndisplacesEb,andthatthese relationshipsaredealt with at both the backgroundand the foregroundof the piece. In any case, a studyof such connectionsmust be at least consideredin any deep studyof the work. But these points arewell beyondthe scope of this paper. In conclusion, let me draw again upon EdwardCone's lucidthat "the good composiity of expressionandrevive his remark tion will always reveal, on close study, the methodsof analysis neededfor its own comprehension."13To assumethatbecausea piece exhibits a strong atonal structureit may not also depend upon tonaldirectionfor some partof its coherence, runsthe risk of doing an injusticeboth to the flexibility of musical artand to
the subtlety of the minds that create it.14

12Cone, "Analysis Today," p. 187. 13PhilipFriedheim, "Tonality and Structurein the Early Works of Schoenberg," (Ph.D. Dissertation, New York University, 1963), p. 454. 14Afinal footnote is necessaryto maketwo generalacknowledgements.This attemptto considertogether tonal and atonal organizationwas spurredin large partby EdwardCone's statementthat "music whose syntax is primarlytwelvetone may neverthelesslegitimately call upon implicit tonal functionsto clarify its concrete values" (Beyond Analysis). And a considerabledebt is owed, and gratefullyacknowledged,to David Beach, who has readan earlierversionof this paper, and made many invaluable suggestions.