Theorists' Views on Bartók from Edwin von der Nüll to Paul Wilson

Author(s): Iván Waldbauer

Source: Studia Musicologica Academiae Scientiarum Hungaricae, T. 37, Fasc. 1, Proceedings of
the International Bartók Colloquium, Szombathely, July 3-5, 1995. Part II (1996), pp. 93-121
Published by: Akadémiai Kiadó
Stable URL: http://www.jstor.org/stable/902133 .
Accessed: 07/04/2014 11:45

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
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.

Akadémiai Kiadó is collaborating with JSTOR to digitize, preserve and extend access to Studia Musicologica
Academiae Scientiarum Hungaricae.

http://www.jstor.org

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 Theorists'Views on Bartok
fromEdwinvon derNull to PaulWilson*
IvanWALDBAUER
Oberlin, OH

The paragraphI would like to quoteby way of a preambleis from

Bartok'sHarvardlecturesof 1943, an interjectionin his explanationsof
polymodalityand his new chromaticism.
Now, the frequentuse of modal chromaticismquite graduallygave me the idea to try a
kind of melodic new chromaticism,developed quite subconsciously and instinctively.
By the way, the working-out of bimodality and modal chromaticism happened
subconsciously and instinctively, as well. I never created new theories in advance, I
hated such ideas. I had, of course, a very definite feeling about certain directions to
take, but at the time of the work I did not care about the designations which would
apply to those directions or to their sources. This attitude does not mean that I
composed without ... set plans and without sufficient control. The plans were
concerned with the spirit of the new work and with technical problems (for instance,
formal structureinvolved by the spirit of the work), all more or less instinctively felt,
but I never was concernedwith generaltheoriesto be applied to the works I was going
to write. Now that the greatest part of my work had already been written, certain
general tendenciesappear generaltormulasfrom which theoriescan be deduced. But
even now I would prefbrto try new ways and means insteadof deducing theories.'

Bart6k'sstancetowardsmusictheoryand his notoriousreticenceto

talk abouthis music are squarelyin perspective.He has little interestin
theorygenerally,and he is outrighthostile to any preconceivedtheory.
Generaltendenciesthatmay have becomediscernibleover the yearsare
* This studyis a slightlyexpandedversionof a papergiven at the International BartokColloquium,
Szombathely, Hungary, July3-5, 1995.Researchforthispaperwassupported by a grantfromtheInternational
Research& ExchangeBoard,with fundsprovidedby the U.S. Department of State (Title VIII) and the
NationalEndowment for the Humanities.
Noneof theseorganizations is responsiblefor the viewsexpressed.
The authorwishesto expresshis gratitudeto theseorganizations
at this place.
I Bela Burtok Essays, ed. BenjaminSuchoff(New York:St. Martin'sPress,1976),376 [Essay.v].

Shditl Mlexicolostic(l
Ac(ldesni(leScientiaralmHttngt(lricae37/1, 19%/7/7. Y3-121
(0SW326664 i.()() o % At(ldenti(liKiad(9.Bud(lpext

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
94 I. Waldbauer:
Theorists'Viewson Bartok

just that,generaltendencies,possiblyseveralof them.For they are the

result of an ever recurringsearchfor new ways and means, for new
techniques,for new construction.The new ways are dictatedby instinct
andby "thespirit"of each new work.Stylisticandtechnicaldiversityin
the Bart6koeuvreare accountedfor in largemeasure.
Since unlike his contemporaries,Bart6k provides little help to
theoristsfor choosingtheiranalyticalendsandmeans,they areleft largely
to their own devices.2Faced with the task to sort out some common
threadsin this dauntingdiversity,they become ever more selective in
choosingparticularaspectsto investigate.Pitchorganizationseems to be
the greatestchallengeto mostof them,andhas in factbecomethe primary
if not the only concernof manyBart6kstudies,increasinglyso fromthe
1960son. Systematictreatment of suchotheraspectsas rhythmandmeter
or thematicprocesseshas beenpushedlargelyinto the background. None
of us is happywiththis stateof affairs,but for the purposesof this study
I feel compelledto acceptthis limitationandwill confinethe discussion
to pitchorganization.

PartI

In the firstpartof this paperI proposeto surveywhatI see as four

phasesin the developmentof theoreticalthoughton Bart6k.The fourfold
divisionis guidedby measurableshiftsin interestandmethodratherthan
by chronology.Forpurposesof reference,I dub themas the Null phase,
the Lendvaiphase, the post-Lendvaiphase, and the phase of recent
Americantheorists;and the style of treatmentwill becomeincreasingly
telegraphicas we are comingcloserto the present.In the secondpart,I
will presenta so-calledvoice-leadingchartof the expositionof the first
movementof Bart6k'sPiano Sonata.It is to serve as a basis for com-
parisonbetweenfour differentanalyses,each coming from a different
2 Peter Petersen,Die Tonalittitim Instrumental-Schuffen von Bela Bunok (Hamburg:Karl Dieter Wag-
ner, 1971 [Petersen 1971] managesto reproduceon just 16 typewrittenpages everything that Bartokever said
or wrote on pitch organizationand tonality ( 15-30), with the exception of only the few rathersketchy analyses
of his own works and the full text of the HarvardLectures(Es.say.s,354-392). The latter was not yet available
in 1971, but Petersencites relevant passages from it after John Vinton, "Bartokon his own Music," Journul
of the American Musicological Society 19 (1966), 232-243.

Shclia Ml$V;CFJ/(Je;C(l
Academi(leScientiarlemBlinge(lricae.S7/1.19%

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 l. Waldbauer:
Theorists'Viewson Bartok 9s

phaseof the analyticalspectrum,with additionalcommentsbasedon the

workof yet otheranalysts.
Edwinvon derNull's 1930monograph on Bartok'spianomusicfrom
1908 to 1926 is not only the firstmajortheoreticalstudyon Bart6kSbut
the only one to have at least some inputfromBartokhimself.3Null sees
Bartokas an evolutionaryphenomenonin mattersof form and pitchor-
ganizationalike, who, in contrastto Schoenberg,invariablystartsout
fromprecedentandgoes on fromthereto unfoldfurtherpossibilities.As
regardsBartok'sconceptionof form, he considersthis "dynamic"on
smaller and larger scale alike, this dynamismmanifestingitself in
thematictransformation technique,in controlof risingandfalling"inten-
sity curves,"andin progressivevariationtechnique.4 Thepassagesdealing
with these aspectsof Bartok'smusic are the best partsof the book and
its analyses.
As regardspitch organization,Null assertsthat it is tertian,tonal,
anddiatonic.Inasmuchas thesequalitiesarenot alwaysimmediatelyap-
parent,this is becauseBartok"extends"the conceptof tonalityby four
specificmeans.Twoof these,the added-notetechniqueandthe technique
of ellipsestend to adumbrate the tertianbase.5Most of us would see no
difficultywith addednotes, and would find simplerellipses also accep-
table.A few of Null'sellipses6are,however,too strainedto be accepted,
3 Edwinvon der Null, Bela Burtok. Ein Beitrug zur Morphologie der neuen Mu.sik(Halle [Saale]:
Mitteldeutsche Verlags-Actien-Gesellschaft,1930[Null, 1930].Thecorrespondence betweenBartokandNull
is lost, only shortpassagesof Bartok'slettersare cited in Null's monograph and in anotherof his books,
Moderne Hurmonik(Leipzig: Fr. Kistner& C.F.W.Siegel, 1932).AndrasWilheimis currentlypreparing a
studyon the relationof BartokandNull.
4 Null notes thematictransformation techniquesas early as in the Two Rumunian Dunces, op. 8a,
1909-10 (Null 1930, 19-23). His fullestdiscussionof the subjectis in connectionwith the Finaleof the
Sonatafor pianoof 1926(Null, 1930,93ff). A revealingdiscussionof risingandfallingintensitycurvesis
in connectionwith the Suiteop. 14 of 1916(Null 1930,37ff). Null sees furthermanifestations of Bartok's
dynamicconceptionof formin the developmental transitionsin smallerforms,and the mannerof theme
presentations in bothsmallerandlargerforms.Ratherthanappearing in staticandbalancedfour-andeight-bar
periods,themestendto developgradually. '4DasThemaist nicht,es wird,"he remarksof the openingtheme
of the Sonata.Partof his analysisof the latterworkwill be discussedin some detailin the secondpartof
the presentessay.It is unfortunate thatsincethe emphasisof thatdiscussionis on pitchorganization, it tends
to addressNull'sweaknessesratherthanhis strengths.
5 Null identifieschromaticnotessoundingtogetherwiththeirresolutionsas earlyas in the Fourteen
Bugutelle.s,op. 8, 1908(Null 1930,chapter1) andhe singlesout a specialcase of addednotesderivedfrom
martellato technique,callingit Tonvedluchip,}un,z,}
(Null, 1930,p. 68, andchapter7, 70-87, particularly77-79).
6 A particularly strainedellipsisis foundin Null 1930,p. 4. Null contendsthatthe Bagatelleno. 2
op. 8 is in G major.Its finalchord,DS-AS-B;is a misspellingof CtWA;-AX, thatis IIXin G withbothlowered
andraisedthird,andthatthe D G-B resolutionis "self-explanatory for Bartok.The listenermustcomplete

StutliaMl. ico/o,ic(l AcatlemiaeScientiarumHungaricae S7/1. I9RS

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
96 l. Waldbauer:Theorists'Viewson Bartok

and his explicitdismissalof quartalharmoniesas rareand insignificant

in Bartokmustalso be rejected.
Moredirectlytouchingupontonalityand diatonicismare two other
techniques,the use of unresolvedchromaticneighborsand suspensions
and the use of whatNull calls modalmixtures.7The first of these leads
to occasionalbitonality,but only in that sense of the term that would
probablybe acceptableto Bart6k,for in all cases the listener'sear can
reducethe resultingsoundcomplexto a singletonality.Thesecond,Null's
observationthatBart6kuses the notes of all diatonicmodesbasedupon
the same tonic simultaneously or in close proximityappearson the face
of it a genuinetrouvaille.8 It goes a long way towardsBart6k'sown later
formulationof polymodalchromaticism. Attractiveas Null's propositions
are,the two sourcesof chromaticism presentseriousproblemsin his own
actualanalyses.DespiteNull'scarefullycraftedguidelines,the determina-
tion whethera given pitchis a modalprincipalnote or a merechromatic
neighboris inordinatelydifficultin Null's system, sufficientcause for
PeterPetersento rejectit altogether41 yearslater.9My own objections
areon somewhatdifferentlines. As Null contendsthatthe importanceof
extendingthe chromaticresourcelies chiefly in vastly increasedpos-
sibilitiesfor modulationby commonchord,his analysestendto exhaust
themselves in mere chord grammar,at the expense of larger linear
relationships.l°Since this pointhardlyshows up in the sampleanalysis
for the secondpartof this paper,I presentExample1, with a minimum
[theprogression] for himself."In 1945,Bartokhimselfidentifiedthe key as Dlvmajor,in his introduction to
a volume,Bekl BflrtokM(lstepieces for Piclolo,plannedby E. B. MarksMusicCorporation, New York.The
volumewas neverpublished,butthe introduction is printedin Es.sclys,
432-433. Quartalharmoniesare
discussedin Null, 1930,84-87 in connectionwithchordsof Sthsandpclssim.
7 Null'stermis CeschlechtsvermischunA,}
dicltonischerSkfllendesselben Crundtons(Null 1930,p. 75).
x Null 1930, p. 74 presentsa table of the twelve chromaticnotes fromc' to b', with each pitch
identifiedas to whichdegreeit representsin the six dintonicmodesbasedon c'. The pentatonicscaleis not
included,becauseNull considersthis merelyan incompleteformof diatonicism,nor does Null take non-
diatonicheptatonic scalesintoaccount,suchas the one thathasits five tonesandtwo semi-tonesdistributed
in a T-T-T-S-T-S-T pattern(acousticalscaleor heptatonia secunda)or Petersen'sS-S-T-T-T-T-T pattern
(cf. note 18 andpp. 101, below)or any of the otherartificialscales(harmonicminor,Gypsyminor,whole
tune,octatonic,etc.).
9 Petersen1971,5-8
t()Null 1930,p. 81 presentsanothertable,whichplacesthe C-EW-EIeG major-minor triadinto the
contextof 14 differentkeys,andsinceNull does not list all the doublerolesthe chordcan playin a given
key, e.g., upperneighborchordof IV or lowerneighborchordof V in F1,the tableis not even complete.
Null'spurposeis to showhowmuchthepossibilities of commonchordmodulation arein his extendedtonality.

Stl>diaMl.rico/os,wica
Act>z:Iemible
ScientiarilmHIans,zlricae
S7/1. 1'9%

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 I. Waldbauer:Theorists' Viewson Barto'k 97

of commentary.Example la shows nothingabout how the composer

workedandvery little abouthow the piece works.In contrast,lc and ld
show the descentof minortriadsby steps betweenthe two B; major
triadsin mm. 1-21, andthe descentof majortriadsby minorthirdsfrom
BS to G in mm. 21-36. The formunfoldsgraphically,or as Null would
put it, the intensitycurverises dynamicallybeforeour eyes.
The secondphaseis assignedto ErnoLendvaialone,partlybecause
in fact he workedalmostentirelyalone, and partlybecause for nearly
three decadeshis influencewas enormous,sufferinga gradualdecline
only from the 70s on." The threepillarsof his theoriesare familiarto
all of us. They are (1) his axis systemwith its complementin the acous-
tical scale;'2(2) his chordmodels based upon the alphachord and its
beta, gamma, delta, and epsilon segments, and his scale-cum-chord
modelsbasedon divisionsof the octave,most importantlythe 1:2, 1:3,
and 1:5 divisions;and (3) the golden section as governingprinciplein
both pitch and formalorganization.Only the secondpillarstandstoday.
Its conceptshavebecomethe commoncoin of theoreticallanguage,albeit
undera varietyof differentnames.The restof the theorylives on not in
its specific conceptsbut ratherin a plethoraof well-observedmusical
detail containedin them, and in the form of generalideas concerning
importantmatters,includingsymmetries,correlationbetweenlinearand
verticalelements,even the generativeinfluenceof bothfolkloreandpar-
ticularartmusictraditions.
Criticismcenterson the golden section and the axis system. Two
objectionsemergein regardto the goldensection.First,it is seen not so
muchan analyticaltool but a style featureoccurringin some of Bartok's
formalproportions, most likelythe productof refinedinstinctratherthan
calculation.l3Second, the Fibonacciseries, which is the additive ap-
11Lendvai's publications begin with the article, "Bartok 'Improvisations'sorozatarol (1920)," Zenei
Szemle 3 (1947), 151-167. Among his other works in Germanor English, the most importantare: "Einfuhrung
in die Formen und HarmonienweltBartoks,"Belfl Bclrtok Wep,} und Werk,Schriften und Briefe (Budapest:
Corvina 1957), 91-137; "Duality and Synthesis in the Music of Bela Bartok,"New HungflriflnQuflrterly
(1962), 91-114; Belfl Bclrtok,AnAnfllysi,sof His Mu.sic(London: Kahn & Averill, 1971, revised ed. 1979);
The W(wrk,shop of Bflrti)kfmd Kl)dcily(Budapest:Editio Musica, 1983). Theoreticalobservations by Colin
Mason and Halsey Stevens are contemporaneouswith Lendvai's, but by orientation, they belong into the
post-Lendvaiphase.
t2 See note 8, above.
i3 George Perle, "The StringQuartetsof Bela Bartok,"in Belfl Bclrtok,programnotes for the recordings
performedby the TatraiString Quartet(New York:Dover, 1967, repr.in A Mu,sicfllOf.>erin^,: E,sscly,in Honor

Stlwflia Mlwsi{*lostica Actlfiemitle Stientiarum HIan,,ricae 37/1. 19%

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 mm. 1-36
Example1: Bart6k,Suite, opus 14, I,
1-20 traslatedinto conventionalsymbols
Null's analysis of mm.
(a) indicatedby parentheses disca
The chords,with pitch duplications all discant notes; lower stave, top voice shows essential
(b)
stave contains
Firstreduction;upper
(c)
showaccompanyingchords
Second reduction
(d)

Abbreviations:
N=neighbornote;
IN=incompleteneighbornote;
P=passing note
85
;1V(4
C#V#3
FIV =el v
=C#v C#1V
[=1VI .9] d#>Xy MMI
a B;I=FIV FVIIr

;Wg , ###8 (§, ###8 (ttt,

b
v Wo }t b°
17 8
15 12
79 12
,5 t

;;
21 ' ;We;+W;^W
i _ _ |.A;1

c)
t wr (; }, b;n 4
g IN=mistilnedV

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 -

I. Waldbauer:
Theorists'Viewson Bartok 99

oo

oNt

oo

StlldiaMlxxicoloswila
At(ldemi(leScientiarllwrl
Hung(llicoe .fl7/1.19%

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 100
1. Waldbauer:Theorists' Viewson Barto'k

proximationof the golden sectionby integers,is an inappropriate tool

for measuringpitchintervalsbecausethe latterrepresenta geometric,not
an additive,series.As regardsthe axis system,thereare boththeoretical
and practicalproblems.The formerhave been dealt with by various
theorists,most recentlyand most comprehensively by PaulWilson.l4Of
practicalproblemsI mentiononly two. One is the difficultyor near-im-
possibilityin assigninggiven chordsto one or anotherof the axes, not
to speak of the usefulnessof such an exercise.The otherbecomes ap-
parentwhen certainanalyticalresultsare confrontedwith actualmusical
perception.For instance,if we returnto Exampleslc and Id, we can see
thatin mm. 21-36 of the Suite, op. 14, all the chordsareon the B; tonic
axis. In Lendvai'ssystem,this spells stasis.In actualperception,at least
my perception,it spells the opposite.
Predictably,the post-Lendvaiphase is the largestof the four.Con-
tributionsstretchfrom the 1960s to the presentand come from three
continents.15 This is the phase in which interestin pitch organization
becomesparamount, with only a numberof Hungarianscholarsexempt
from this restriction.They are includedhere nevertheless,becausethey
of MartinBernstein,New York:Pendragon Press,1977),notesthatgoldensectionandFibonacciproportions
occurin Bart6k'smusicwitha frequencybeyondstatisticalprobability, but he does not seem to regardthis
asproofof all of Lendvai'stheories.RoyHowat,generallyfavorably disposedtowardsgoldensection,Fibonac-
ci, and otherrelatedproportions (see his Debu.vvyin Proportion, Cambridge: CambridgeUniversityPress,
1983)takesissuewithsomeof Lendvai'sfindingsin his reviewarticle,"Bartok,Lendvaiandthe
Principles
of Proportional Analysis,"Mu.sisAmlly.si.s2 (1983),69-95. In his "Masterwork (II):Sonatafor Two Pianos
andPercussion," The Bllztok C)mpnion, ed. MalcolmGillies (Portland,Oregon:AmadeusPress, 1994),
315-330,he statesthatoccurrences of suchproportions in Bart6k'smusicsuggest"nota strictmethodbut
rathera dramaticdevicethatcanbe used(or thatoccurs)whenappropriate" (p. 322). Howat'scitedreview
articlein Mu.sisAnuly.sisdrewa rebuttalfromLendvai,'sTheLimitsof MusicalAnalysis,"The New
Hungllrilln
Qullrterly26, No. 97 (Spring1985),201-207.
14PaulWilson,The Mu.sisof BelclBclrtok(New HavenandLondon:YaleUniversityPress,1992),S8
and203-208.
ls A completelist of publications wouldbe in excessof severalpages.In view of the generalacces-
sibilityof Elliott Antokoletz'sreferencework,Belcl Bclrtok,a Guide to Re.vellrLh(New York:
Garland
Publishers, 1988),withits enlargededitionalreadyin preparation, I mustcontentmyselfby presentinga list
ofwriters,whosecontributions aretheoreticalin full or in part.The list mustbe headedby threescholars
roughly coevalwithLendvai,MiltonBabbitt,HalseyStevens,andColinMason.Fromthe U.S. comeJames
Baker, Jonathan Bernard, WallaceBerry,AllenForte,RichardParks,GeorgePerle,Felix Salzer,RoyTravis,
LeoTreitler,and JamesE. Woodward; fromBritainDonaldChittum,GilbertG. French,MalcolmGillies
(actually fromAustralia),DavidGow,Roy Howat,RobertSmith,ConstantVauclain,and ArnoldWhittall;
fromBelgiumYves Lenoir,fromDenmarkIanMaegaard, fromSweden Ove Nordwall;fromthe German
speaking landsReinholdBrinkmann, Hartmut Fladt,JurgenHunkemoller, ErichKapst,FrankMichael,Ilkka
Oramo (actuallyfromFinland,butfrequently publishing in German), PeterPetersen,andGunterWeiss-Aigner;
andfromHungaryLaszl6Dobszay,JanosKarpati,SandorKovacs,Laszl6Somfai,andAndras
Wilheim.

Stldi(l
M{.vicoloswica
Al(ldemi(leSlientiarllmHflng(ltic(le37/1. 19%

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 1. Waldbauer:
Theorists'Viewson Bartok 101

sharemanyassumptionswiththe group.But despitesharedassumptions,

a groupof Americantheoristsareexcludedon accountof theirspecialized
methodology.A convenientwindowon these assumptionsis providedby
PeterPetersen's1971 studyon tonalityin Bartok'sinstrumental music.
It is comprehensiveto the point of being encyclopaedic,and it is the
morecredible,becausePetersenstartsoutby reproducing a near-complete
recordof all Bartok'swritingsrelevantto pitchorganization.l6
PetersenconsidersBartok'sespousalof atonalityaround1920 as no
morethanan expressionof solidaritywith the avant-garde.'7 He goes on
to demonstratethat both the laterBartokwritingsand the music itself
develop in the directionof a new conceptof tonality,and he identifies
threecomponentsas its principalingredients.First,tonalityis no longer
based on the traditionaldiatonichierarchy,but on the total chromatic
resourceas affordedby polymodality.Second,dominatingpitchesare no
longer determinedby their positionin the gamut,but by a varietyof
other criteria.And third,the configurationof given chords no longer
indicatestheirtonal functionper se, partlybecausetheir very structure
is equivocal,partlybecausethis structuremaybe no morethanthe result
of horizontalmotion. He then proceedsto discuss these components
separately,tonal systems and modes first, chord formationssecond,
criteriafor establishingpitchhierarchylast.
As regardstonalsystemsandmodes,Pete.sen addsa new heptatonic
scaleto the familiararray,theone withthe two availablesemitonesplaced
next to each other,and he also includesthe harmonicminor and the
chromaticaggregateon theirown right.'8An openingwedge for intro-
ducingthe importantnew conceptof competingtonalcentersis afforded
to Petersenby his proposalto divide the diatonicmodes into authentic
and plagalgroups.He firstobservessuch competitionbetweenpitchesa
4th apartin certainplagalfolktunes,but the principleitself reappearsin
1(5Petersen1971, 15-30,cf. note2 above.In additionto thisstudy,Petersenalso publishedthe article,
"Rhythmik undMetrikin BartoksSonatefurzwei KlaviereundSchlagzeugunddie KritikdesjungenStock-
hausenan Bartok,"Mu.siktheorie9/1 (1994),39-48.
17 JanMaegaard, "BelaBartokund 'Da, Atonale',"JullrZ7uch Peters 1981-1982(1985),3032, bears
outthisstatement.Thearticleis a thoroughinvestigation
of theprofessionalandpersonalrelationshipbetween
Bart6kandSchoenberg.
IxCf. note 8 above.Petersenjustifiesthe inclusionof the chromaticscale on its own righton the
groundsof manyBartokmelodiesappearingin two forms,one with largerintervals(mostlydiatonic)the
otherin chromaticcompression (Petersen,1971,49-50).

StZ#C/i(lMl/.vicol(JRic(tAcciclemi(leScienti(wrZJEt1Hling}ClEitcle S7/1. 19%

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 102 I. Waldbauer:Theorists'Viewson Barto'k

his study in many different foI*ms.l9 AS regards chord formations, his

discussion iS remarkable on four counts. He includes tertian chords. He
provides characteristically sober criteria for determining the roots of cer-
tain dissonant as well as quartal harmonies. He singles out the nature of
final chords for special attention. And, as in all other matters, he supports
his findings with solid statistical evidence.20 HiS conclusion suIprises no
one today: the tension-resolution principle iS still valid for Bartok. AS
regards criteria for determining the role of pitches (and by implication
chords) as either competing or as primary and secondary centers, Petersen
offers four of them, the four overlapping with each other in varying
degrees. They are (1) circle stI Uctures by Sths (or 4ths) with established
initial and goal tones,2 (2) mirror StlUCtUreS and their axes;22 (3) held
and repeated notes, including ostinati with their upper boundaries station-
ary or changing;23 and (4) single pitches exposed by either dynamic and

19Petersen 1971 (51-55) misreadsBartok on plagal modes, but the point he raises in
connection with
them has some validity. In the Germanversion of his book, DcesungclrischeVolkslied,46-47,
Bartokdiscusses
tunes that have variant forms ending on the tonic as well as variantsending a 4th below the
tonic, and he
considers the latter the plagal form of the former.On this ground, Petersenproposes that all tunes
that have
pronouncedemphasis on the perfect 4th above the fitlulis should be consideredplagal. His example is a tune
from For Children (no. 41 in the first edition, no. 39 in the revised edition) generally
classified as being in
A Aeolian. He would call the mode of this tune D hypo-Aeolian, disregardingthe fact
that all variants of
this tune and all variants of many similar Aeolian, Mixolydian, and Phrygian
tunes-invariably end on A
(or ratheron the pitch a 4th below the emphasizedpitch). The issue would be merely a matter
of nomenclature,
were it not for the fact that in his compositions Bartok occasionally appears to justify
Petersen. A case in
point is a tune appearingtwice in the exposition of the Sonata for piano of 1926. The
reader may turn to
Example 2 and verify that the melodies in mm. 76-92 and 93-114 are modeled on Petersen's
example (and
on its many relatives) and that their harmoniesas well as the sequence in which they occur
vindicates the
Petersenianassumption of a key center a 4th above the final note. These tunes seem to be better
examples
for the phenomenonin question than the debatableones chosen by Petersen himself on his
pp. 54-55.
2() Petersen 1971, 46-49 lists 42 cases identifying various passages as
to scale system and mode, and
chapter5 (57-78) contains no less than four similarly exhaustive statistical tabulations.
21 Petersen 1971, 82-96 discusses five circle fugues and eleven
circular sequences.
22 The point is valid in general terms, but some of Petersen's
examples are debatable. One case in
point: Petersen (114-119) regards the A;-Ah augmented prime as the key center of
Mikrokosmosno. 144,
"MinorSeconds and Major Sevenths," on the ground that it is the axis of the most frequently
occurring
positionof the symmetricalset of perfect 4th + minor 2nd + perfect 4th (ES-AS-AWD).He
identifies other
non-symmetricalelements in the piece but says expressly that they can be disregarded.My study
"Interplay
of Tonality and Nontonal Constructsin Three Pieces from the Mikrokosmosof Bartok,"
Music und Context,
Esscrys.forJohn M. W(lrd,ed. Anne Dhu Shapiro (Cambridge:Departmentof Music, Harvard
University,
1985), 418-440, shows that the elements disregardedby Petersen spell ES. Two other analysis
in the same
study,those of MikrokosmosNos 140 and 143, can also be compared with Petersen's
analyses on his pp.
105-107 and 109-111.
23 A particularlyenlighteningdiscussion of ostinati is found in
Petersen, 1971, 145-146 with reference
to the different opinions expressed by Rudolph Reti (in his
Toncllity-Atoncllity-Pcontoncllity,
London, 1958,

Stuclia
Mxicolo^,Aic(J
AcalulemileSsientitlrm Hn,l,lricae 37/1. 19%

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 I. Waldbauer:
Theorists'Viewson Bartok 103

registralmeans,or by being verticalor horizontalboundarynotes, or by

signalingthe end of a constant-interval
cycle by changeof eitherinterval
or direction.24A fifth criterionemergesfrom Petersen'sdiscussionson
othermatters.A concealedtonalcenterwithinan otherwisetrulyatonal
passagecan be surmisedby the cleartonaldefinitionof the surrounding
passages.25 This last seems to be morethanjust an extensionof the in-
itial-tone-goal-toneprinciple.
The pointssummarizedabove keep recurringin the work of most
scholarsof the post-Lendvaiphase,often in a moreelaboratestate.But
Petersenmakestwo additionalpointsfound much less frequently,if at
all. Instead,they connecthim on the one handwith certainHungarians,
and on the otherhand,with some of the Americansof the fourthphase.
One is his emphasison certainspecifics of folk modes and folklore
forms.26 The otheris his adoptionof particular
chordformations he calls
themGrundkonstellationen as principalcharacteristics
of entiresections
or entirepieces.27Thisconceptformsthe kernelof muchrecentAmerican
theory,alongwith the initial-tone-goal-toneconceptit leadsto PaulWil-
son, andalongwithmirrorstructures andconstant-interval cycles to Eliott
Antokoletz.
As mentionedbefore,sharedassumptionstie the two Hungarians,
LaszloSomfaiandJanosKarpatito the post-Lendvaiphase,even if their
preoccupation with the generativeinfluenceof folkloresets them apart.
Somfai rarelyconcernshimself with pitch organizationper se, but the
tonal-harmonic assumptionsunderlyinghis exemplaryanalysesof form,
of melodicconstruction,andof the relationof theseto particularspecies
76-79) andCarlDahlhaus(in his "DerTonalitatsbegriff in derneuenMusik,"Terminologieder Neuen Musik,
des InstitutsfierNeue Musik und Musike)ziehung,Darmstadt,
Ver(jthentlietiungen vol. 5, Berlin,1965,83-88).
24 The discussionof thesecriteriaare in chapters6 to 9, Petersen1971,79-144.
25 Petersen's
exampleis thefirstportionof thedevelopment sectionin thefirstmovementof theSonutu
for Two PiconoscondPercilssion, the segmentcalledby Bart6k"a shorttransition of superimposed layersof
fourths"("Aboutthe Sonatafor Two PianosandPercussion," E.sselysp. 417) placedbetweenthe E ending
of the expositionandthe E ostinatoof the continuation.
26 Petersen1971, 147-153is a well-informed discussionon the four-part peasant-song form,with an
impressivelist of wheresuch formscan be foundin Bartok'smusic(apartfromactualfolk songs).To my
knowledge,he is thefirstto discover,thatin the first31 measuresof theSecond Picino Concerto the material
playedby the solo addsupto folk patternA-A5-B-A,even thoughthe phrasesareseparated fromeachother
by othermaterialsplayedby the tutti.A similardiscoveryin the thirdmovementof the Musie for String.s,
Pezeussion, cmd Celestbltestifiesto Petersen's
thoroughinsights.
27 Petersen1971, 157-164 is an analysisof the Sonatafor piano, 1926. Crundkon.stellcitionen are
discussedon pp. 161-163.Part2 of the presentstudydealsextensivelywith this analysis.

Stcli(J Mw.vicolos,tictzActzC/e'1t1i(/e
Ssienticirsm Hn,z,zciric(Je 37/1. I YYd

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 1. Waldbauer:Theorists'Viewson Barto'k
104

of the folklore traditionseem to indicate

sharedground.28Kaxpatiis
probablythe most theory-minded amongthe Hungarianscholars.29
greatestdebt music theoryowes him is for his The
developingthe concept
of mistuningand his refinementof the
conceptof bitonality.He sees
mistuningas originatingin folk practicebut then
goes on to show it as
a tool in Bartok'shandsfor relaxingthe hold
of the momentarily prevail-
ing tonalsystem.
The fourth and most recent phase consists
of the work of four
Americantheorists,Allen Forte, Elliott Antokoletz,
Paul Wilson, and
RichardCohn,the last threeBartokspecialistspar
excellence. The com-
monthreadbetweenthemis relianceon
theoreticalconceptsand systems
that were developedoriginallyto deal with
music generallylabelled
atonal.30 Cohnindeedgoes so far as to stateexplicitly
thatwhile tonality
doesexist in Bartok,meaningfuldiscourseof
his compositionalequip-
mentis quite possibleapartfrom it.3'And
Allen Fortehimself,the for-
mulatorof the set theoryon which Cohnrelies,
seems to have come to
2x Somfai'sanalyticalwork is best
representedin his 18 Bcortok
(Budapest: Zenemukiado, 1981).Analyticalinsightsare in abundance Tcinulm(iny[18 Bartokstudies]
"Analizis-jegyzetlapokaz 1926-oszongorasesztendorol," throughout, and threeof the essays,
1926," SM 26 (1984),"Statikai "Analyticalnotes on Bartok'sPianoYearof
tervezeses formaidramaturgia a 2.
and formaldramaturgy in the PianoConcertono. 2], and"Variacioszongoraversenyben" ['Statical'planning
"Strategies of Variationin the SecondMovementof strategiaa Hegeduverseny II. teteleben,"
directly Bart6k'sViolinConcerto1937-1938,"SM 19
addressthe problemof analysis.Timehonored (1977)
areintroduced approaches find new application,novel approaches
in exemplaryfashion.His two contributions
note13),thoughnottheoretical in Gillies,ed. The Burtok Cl)mpunion(cited
per,se, arepertinent
to theory.His"Soktemas in
in
an expositionof manythemes],partVlll of expozicioorganikaja" [Organism
the "AnalyticalNotes"will be discussedin
part2 of this study. some detailin
2'1JanosKarpati, B(lrtok'sChumberMu,sic(Stuyvesant
version of his B(lrtokString Qu(lrtet.s(Budapest: N.Y.:Pendragon Press[1991]),is an
is CorvinaPress,1975).Thefirstpartof thebook,pp.enlarged
devotedto theoryof variouskinds,pp. 169-235to 21-235
pitch
othertheoreticalworks"TonalDivergencesof Melodyand organization, includingmistuning.Amonghis
Musical Harmony: A Characteristic Device in Bartok's
Language," StudiclMu,sicologiccl24 (1982),373-380 is
inGillies,ed. TheBclrtokCompclnion(see note13)are particularly important, andhis contributions
to
themusicallyeducatedlayperson. repletewiththeoretical observationsin a formaccessible
3()GeorgePerle,Twelve-ToneToncllity(Berkeley
iswitness andLos Angeles:Universityof CalifomiaPress,
to thegeneraldislikeof thetermatonal.His 1977)
ofBelaBartok,"The Musis Review 16 (1955), articles,"Symmetrical Formations in theStringQuartets
300 312 and"Berg'sMasterArrayof the
Mu,siccll
Quclrterly63 (1977), 1-30, aregerminalto the IntervalCycles,"
Sl RichardCohn,"Inversional theoriesof ElliottAntokoletz.
SymmetryandTranspositional Combination in Bartok,"Mu.sicTheory
Spectrum 10 ( 1988),1942; and"Bartok's OctatonicStrategies:A MotivicApproach," Journcll*)f'theAmericcln
Mu,sicologiscllS)ciety 44 (1991), 262-300.Thereferencein
work,
as is trueof thatof mostof his American thetextis on p. 264 of the secondarticle.
Cohn's
reference colleagues,AntokoletzandWilsonamongthem,containsno
to eitherNull or Petersen.

StlJzlia
Mu.vicolof,wictl
AcazlemiaeScientiarllmAflmf,lricaleS7/1. ISJX

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 I. Waldbaaer:Theorists'Viewson Barto'k 105

the sameopinion,as shownby his recentanalysisof an earlyBela Bart6k

miniature.32
Antokoletz,who startsout from the writingsof the dodecaphonist
GeorgePerle, is considerablymore cautious,as befits a theoristintent
uponconstructinga theorythatwouldbe applicableto the whole of the
Bartokoeuvre.33The strandhe sees runningthroughthe entirecorpusis
Bartok'spreoccupation withactuallyor potentiallysymmetricalstructures
in conjunctionwiththe completeset of intervalcycles. Accordingto him,
this tendencybeginsin 1908,becomesgraduallystronger,andculminates
in the StringQuartetno. 4 of 1928, where it becomesthe dominating
factorin generatingboth sense and substanceof tonality.Thereafter,it
sharesthis role in varyingmeasurewith more traditionalmeans of es-
tablishingmodalcenters.Beyondstatingthathis tracingof the so-called
cell-Z, thatis, the doublysymmetricalperfect4th + minor2nd + perfect
4th chord,surpassesin excitementthe best of detectivenovels,no further
commentaryseems necessaryhere,for this is providedin his own study,
"OrganicDevelopmentandthe IntervalCyclesin Bartok'sThreeStudies,
op. 18" of the presentvolume.
Wilson'spoint of departure,like Cohn's,is also Forte'sset theory.
He is, however,even morecautiousthanAntokoletz,when he explicitly
professes skepticismconcerningthe possibilityof any single general
theoryto fit all of Bartok'smusic.34In his book, a generaltheoretical
frameworkprecedeshis analyses,and in this he includesboth tonaland
32 Allen Forte, "Bart6k's'Serial' Composition,"Mu.siccllQuclrte)ly46 (1960), 233-245; "Foreground
Rhythmin Early Twentieth-CenturyMusic,"Model.sof Mu.siccllAnclly.si.v. EclrlyTwentieth-Century Mu.sic,ed.
JonathanDunsby (Oxford and CambridgeMA: Blackwell Publishers, 1993), 132-147. The latter study is an
analysis of Bart6k, Bagatelle no. 8 of the Fourteen Bclgcltelle.s,op. 6. 1908. His book, The Structureof
AtoncllMu.sic(New Haven and London: Yale University Press 1973) is germinal in the work of all scholars
using set theory.
33 Elliott Antokoletz, The Music est Belcl Bclrtok:A Study est ToncllityclsldProgre.s.sionin Twentieth-
CenturyMu.sic(Berkeley and Los Angeles: University of CaliforniaPress, 1984). "The Musical Language of
Bart6k's 14 Bagatelles for Piano,"Tempo137 (June 1981), 8-16."Principlesof Pitch Organizationin Bart6k's
FourthString Quartet,"In Theor Only 3 (1977), 3-22. ProfessorAntokoletz was kind enough to let me have
the abstractof his article in preparation,"Theoriesof Pitch Organizationin Bart6k's Music: A Retrospective
View."
34 Paul Wilson, op. cit. (see note 14). "Conceptsof Prolongationand Bart6k'sOpus 20," Mu.sicTheoty
Spectrum6 (1984), 79-89. Review of The Mu.sicof Belcl Bclrtok...,by Elliott Antokoletz, Journcllof Music
Theory 30 (1986), 113-121. The third chapter of his book (55-84) is devoted to the Bart6k Sonata, and the
part of it concerning the exposition of the first movement will be discussed in some detail in part 2 of the
present study.

StllCI;U
MllXj(O/08jCa ScientiarlemBlingaricae S7/1. I!JR5
AL(JC/emj(Je

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 I. Waldbaaer:
Theorists'Viewson
106
Bartok
nontonalconcepts,and then goes on to
requiredby the particularsof each roundthese out separately,as
is both too long and too subtle analyzed work.The list of concepts
for simpleenumeration.It must
here to say four thingsaboutit. It suffice
includesalmostall conceptshitherto
encountered, mostof themin morerefinedform,
Second, it includes concepts somewithqualifications.
concerningboth the precompositional
resourceandcontextualrelations.Third,
one of his new concepts,thatof
projectedsets, will be demonstrated in connectionwith the second
of this paper.And fourth,another part
new concept,thatof structural
is importantenoughto be overlay,
identified
is that severalpossibly disparate here,howeverbriefly.Its substance
strandsunfold simultaneouslyin the
courseof one composition,possibly
but not necessarilymakingcontact
witheach other at specific or
predictablepoints.The test is, as in all
othermatters,recognizability,and one strandwill be easily
inthe analysisfollowingin the recognized
next portionof this paper.

Part2
The voice leadingchartof Example
tonal 2 presentsmy own analysisof
directionin the first movement
1926.35 expositionof Bartok'sSonataof
It is drawnwith sufficientdetail
four to providea basisfor evaluating
otheranalysesof the complete
order)by Null, Petersen,Somfai,and movement,those (in chronological
Wilson.
analysesis centeredon four different The discussionof these
five
Theyare (1) the progressive aspectsof the composition.
development of thematic
use pitchsets;(3) tonallymeaningful material;(2) the
of characteristic
key
centers,andvaryingdegreesof chordprogressions,
bitonality;(4) the role of one sharply
35The chart could be
considered Schenkerianonly in the
technique
of reductionto make certain sense that it borrows from
patternsmore readily recognizable, Schenker the
tonal
hierarchy.But even these rhythmic and uses rhythmicsymbols to
symbols are not used with show
to a
pre-existent tonal hierarchy.Rather,they are Schenker'sprecise meaningswith
ad hoc (and I hope self-explanatory) reference
the
relative
position of pitches within hierarchies attempts to indicate
ofBartok's that keep developing from one
music closer to Schenkerian situation to the next. Charts
York: orthodoxy can be found in Felix Salzer, Structurul
Dover Publications, 1952, rev. ed., Heuring (New
1962), Examples 186, 241, 441,
"Toward
a New Concept of Tonality," 452, 480, and 504; in Roy Travis,
Journul ot Music Theoty 3, (1959),
Mikrokosmoss
No. 124; in Travis,"Tona} 257-284, containing a chart of
The Coherencein the First Movement of
Musis
Forum 2 (1970), 298-371; and more Bartok'sFourthString Quartet,"
of
Music(ll recently in James Baker,"Post-Tonal
Awlulysis(see note 32), 2(F41. The
exemplary analysis of Bartok, Suite op. 14, Voice-Leading,"Models
both in regardto musical Scherzo on pp. 34-39 is
substanceand to Schenkerianform.

Stllzli,,
Mll.vico/of,^ic(l
Ac(lclemi(le Ssienti(llllm Aflmf,^rllicae
37/1. IS)Yd

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 l. Waldbauer:
Theorists'Viewson Bartok 107

bitonalphrase(mm. 57-67 and ff.) in the formal constructionof the

. .

exposltlon.
Sincenoneof the analystsaddresseseachof theseaspects,it is hoped
that the followingdiscussionwill yield a pictureof this expositionthat
is betterroundedthan any one pictureemergingfrom any one of the
analyses.Onepracticalpointbeforeproceedingto substance.The division
of the expositioninto the segmentsshownin the chartrepresentsagree-
mentbetweenall analysts,the segmentsbeing,mm. 1X3, 44-56, 57-75,
76-92, 93-115, and 116-134.

Progressive Developmentof ThematicMaterial

The discantline of the chartmerelyelaboratesSomfai'sexemplary

analysisof the thematictransformation process in the movement,with
just a few minoradditions.Since it will becomeapparentin the course
of the analysisthatthroughout most of the expositionthe discantmelody
generatesthe tona]ity-defining harmonies,it will be well to summarize
at this point the pitch relations within this thematictransformation
process.The nucleusof the themeconsistsof the minor3rd motiveex-
posed in mm. 1-13 and the major2nd addedabovethis in mm. 14-17,
as seen in the E-(F#)-G-Apattern(also in the C-(D)-ES-F patternof
mm. 26-30). This minor 3rd + major 2nd patternremainsconstant
throughoutall transformations of the exposition.The first two steps of
the transformation come in mm. 44-54, wherethe nucleusis stretched
upwardsto includethe ESa minor3rd abovethe A-C minor3rd, andin
mm. 57-67, wherea furtherstretchreachesthe F#a perfect4th above
the ASC#minor3rd and producesits own inversion(ASCFD#inverted
as CFD#-F#).In mm. 76-83 the directionof the stretchis reversed.The
G# B-C# nucleusis stretcheddownwardsto D#a perfect4th below G#,
andthis perfect4th is thenfilled in by stepwisemotion.This new pattern
is repeateda perfect5th higherin mm. 93-110 (beginningon ES),after
which the final transformation in mm. 116-25 takes the form of a free
inversionof the newestpattern.This summaryshows the minor3rd +
major2nd nucleusas a set on its own right,but a set thatis verticalized
only in the first segment(mm. 143): GSB-C# (mm. 7, 14 and ff., 39
and ff.), D-H-E (m. 7), B-CSE (m. 13), and CSE-F#(mm. 26 and ff.)
Of the threeanalystsotherthanSomfai,Petersenconsidersdiscantele-
Sttdi., Mll.ticvlv^icaAtdemi.le ScientiarllmHlxngtaricale
S7/1, 19%

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
Xt, 1 X<:_r,.l;,
A I- t -I . *t L

Example2. Bartok,Sonata (1926), I, exposition, mm. 1-134

Diamond shapednotehaeads(O or +) indicatebitonal elements
-
14
5 6, 7
1- 12,13

25 26 30
21

47 48 49 50
44 45 46

o"z 4s#o . ; ,> ,f

4J. 1Z. ;# o 4
.
8

--- -- -- -- -- -- -- -
F16

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 # * , - -

I. Waldbauer:
Theorists'Viewson Barto'k lo9

|b
J -

tt
T

:,1 rs

*1
qWfi
's

#
|i
:; lfi
- 2
-ifi
---l

t,s
l

t t- t

rr

**

t rW

|t

oo
l tbQ
w

*v "\

/*
' ^
. _b

(+* (; R
t

Ir X > >
#
>
L
,
*-

Acclchemi(le
Stl{lja MlJVjCO/O},TjC(I Scienti(lrlJn1
Hlms,(lriccwe
@7/119%

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 oX ' # * m 1: .o o * .X . £ tcr + +

110 1. Waldbauer:Theorists'Views on Bartok

+ % *X1
N , i '', m * E
_, *+#,.,,,_ 1+ X
,

*< ', + rb
_ b
_ ' **

o '' * +e -
v ', +
" '
+

" :S W - , " # , ¢

, wH. t

X i. ¢ X

S
i-
_
r r,
t
fi
t
_-ze ., N . N
,
s a ' W i
X W X X X

XF, o ++ fi

_ k " R t a e sl_ S Wf

S7/1. 19%
StualiaMuxisf)lf),;saAcalemiae Scientiarm Huns,waricae

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 1. Waldbauer:Theorists'Viewson Bartok
lll

mentsonly rarely.Null notesthatmm. 14-18 anddv116 havetheirperfect

4ths filledout in like fashionandsees the similaritiesbetweenall themes
of the expositionas well. Wilsoncomesclosestto Somfai'sanalysiswhen
he says that "the three successivethemes [fromm. 57 on] sound like
variantsof a single idea"(p. 57 of his book, referredto in note 14).

Characteristic
pitchsets
The primaryrole of pitch sets in this movement,or any other in
whichthey are employed,is to providethe compositionwith its charac-
teristicverticalsound.It seems advisableto prefacetheirdiscussionby
repeatingthatthe sets to be considerednext are differentfrom the one
producedby the discant,the two operatingindependentlyfrom one
another,and by remindingthe readerof a basic tenetof set theory.This
is that the identityof a set and any of its transpositions dependson its
intervalsandthe directionin whichthe intervalsfollow one another.The
pitches of a given set must outlinethe same intervalsin the same as-
cendingor descendingorder(rectoandinverso),buttheindividualpitches
may be placed in any octave register.Thus, the set of a minor2nd +
minor3rd can take the form a#'-b'-d" (or any of its eleven transposi-
tions), and it remainsthe same in any of its possiblespacings(e.g., d'-
b'-a#",b'-a#"-d"',etc.) andin any of its mirrorinversions(e.g., F#'-a'-
b;', or a'-c"-c#", c#'-a'-ch",etc.) The identityof the set wouldchange,
however,if only the intervalswould remainidenticalbut the direction
wouldchange.Thus, a#'-b'-d"does not equalg#'-b'-a#'(or g#'-a#'-b'),
thatis, a minor3rd up followedby a minor2nd down.
Both Petersenand Wilsonhave discoveredthat characteristic pitch
sets are to be found in the Sonata.Wilson emphasizesthat these sets
operateindependently from,andwithoutdirectreferenceto, discant,tonal
direction,or anyotherelementin the piece even whenthey intersectwith
those elements.In somewhatsimplifiedterms,he identifiesthe sets of
minor2nd + minor3rd(hereafterthe subset)andthe minor2nd + minor
3rd + minor 3rd (hereafterthe superset)as the twin sound complexes
dominatingthe exposition,indeed the entire movement.His analysis
shows the unfoldingof the twin sets as one of the movementsstructural
strands.It is convincing,particularlysince the use of these sets strikes
mostobserversas reflectingprobableconsciousintentionon Bartok'spart.
Stula Mu.vicolol,;cuAcu(letrue ScrentFurum
Htn^,}uricue
37/1. 1YX

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 112 I. Waldbauer:Theorists'Viewson Bartok

As the sets can be identifiedin the chartwith relativeease, it did not

seem necessaryto markthemby specialsymbols,nevertheless,in mm.
dv1'16,76-83, and93-99 slursareused to connectthe firstthreepitches
of the set withthedelayedfourthpitch.Thefirstoccurrencesof the subset
arein the bassof mm.2-13 and3943, as D-ASB; the firstoccurrences
of the superset,in the bass of m. 14 as EFG#-CFD,and in m. 26, as
ASC#-FFG.The invertedsubset,A-C-C#, first occurs in m. 36, then
againin m. 44. Fromhereto the end of the expositionboth subsetand
supersetare inverted,resultingin the subsetD-F-F# in mm. 49 and 51,
and F-G#-A in m. 67, then the supersetE-G-G#-C#in mm. 55ff. and
76ff. and B-D-ES-ASin mm. 93ff. The final segmentbegins at m. 116
with FFA-ES-BS,but almostequallyconspicuousare two recto forms,
F1WA-D-E; andES-EFG-BS;andthe highlevel of saturationof the pitch
content(the octatonicscale beginningwith ES-Et,with ASandD natural
added)impliesyet otherforms.
In additionto Wilson'ssets, thereis one otherset operatingwith a
degree of consistencyin the exposition,the major-minortriad, or as
Lendvaicalls it, the gammachord.36It is first projectedhorizontallyin
mm. 1-14 and7-29 (markedby beamsin the chart)andrecursin vertical
form at mm. 67ff., 76ff., and 93ff. At the last threepoints it intersects
with Wilson'ssets. Its functionin the expositiondiffersfromhis sets in
two respects.Its role in shapingthe characteristicsound is decidedly
secondary.At the sametime,unlikeWilson'stonallyneutralsets, this one
is instrumental in definingthe rulingharmonyin all threeof its recur-
rences.
Tocompletethediscussionof sets,it mustbe mentionedthatPetersen
is in agreementwith Wilson as regards the identity of the set
(Grundkonstellation in his words)dominatingthe five segmentsin mm.
44-134, buthe assumesa differentset for the firstsegmentin mm. 143.
Hischoice, the set of perfect4th + minor2nd + perfect4th, seems less
fortunate thanWilson's.At mm. 14 and 16 he identifiesG#-CFD-Ghand
CFFFG-C4.These are actuallyWilson'ssets withoutthe bass notes Et
andA#but with the discantnotesG andC added.They may seem to be
verticalsoundsof importanceat these points, but Petersen'sargument
36 In a privatecommunicationProfessorWilson expressed agreement
with my analysis of this set, even
thoughhis book does not recognize the chord as a recurringset.

Sttlulia
Mu.icolo,s,wica
Afclulemi(Je
Sf ientiartimBun,s,scirif
cle37/1. IS*9t5

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 I. Waldbauer:TheoristsXViewson Bartok 113

thatthe pitchesD-E[?]-ASB in mm. 2ff. actuallyrepresentthe sameset,

only with D substitutedfor E#is certainlystrained.Nor does Petersen's
set recuranywherein the furthercourseof the movement,except in the
in mm. 187-202.
almostliteralrecapitulation

Tonallymeaningfulchordprogressions,keycenters,andbitonality
For identifyingharmoniesand key centers,the chartis guidedfirst
andforemostby the harmonicimplicationsof melodicmotionover large
timespans,ranginganywherefromtento twenty-fivemeasures.Particular
verticalchordformationsplay a lesserrole, sometimesnone at all. Null's
approachis in completecontrast,as he concentratesalmostexclusively
on verticalchordformations,whilePetersenandWilsonoccupya middle
groundbetweenthese two positions.The chartreduceslarge-scaletonal
motionto threekey areas,E in mm. 143, F in mm. 44-75, and ESin
mm. 76-34, and indicatesthe presenceof bitonalelementsfrommm. 44
to the end of the exposition.The termbitonalelementseems appropriate
here,becausethe key of theseelements,markedon the chartby diamond-
shapednoteheads,is recognizablysubordinated to the prevailingprimary
key, quitein keepingwith Bartok'sconceptionof bitonality(see Essays,
p. 366). Further,since the key of the bitonalsecondaryelementis better
defined in one segmentthan in another,there are differingdegreesof
bitonality.Nevertheless,the chartuses no symbolsto distinguishbetween
higheror lower degreesof bitonality,because,on accountof the many
variables,everycase needsto be evaluatedseparately,andin conjunction
with bothlarge-scaleandsmall-scaleharmonicmotion.This is best done
by proceedingfromone large-scalekey areato the next. This procedure
has the addedadvantageof affordingthe best opportunity to discussdif-
ferencesin interpretationarisingfromthe differentapproachesof the four
analysts.
Thesegmentof mm. 143. All analystsagreethatthe segmentis in
E, but local motionwithinthis E gives rise to differentinterpretations.
Petersenand Wilson concentrateon their respectivecharacteristicsets,
thussidesteppingthe issueof interpreting harmoniesfroma tonalperspec-
tive. But both single out m. 26 as the pivotal point where the discant
melody moves up a 6th. In contrast,Null's tonal analysis is overly
detailed.He assumesseveralchordprogressions,amongtheman uncon-
A((JTIemSae
St/clia Mtlxicolos,^i(l ae S7/1 IS}S*rS
Sf ientiartlmH/n,$^arif

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 114 I. Waldbauer:Theorists'Viewson Bartok

vincingone to VI and IV of E in m. 14 and a bizarremodulationto F#

at m. 27 [recte26]. The chartidentifiesone largertonalmotionof local
significancefromI to N(eapolitan)6 andbackto I in E, butit wouldseem
that assuming a functionallyequivalentIV¢3#would be an equally
plausibleinterpretation of the N6. The point would be of little conse-
quence,were it not for Bart6k'sletterto Null (knownonly frombeing
cited in Null 1930, pp. 99-100) thatidentifiesthe chordat m. 36 as F6
and states that the same chordat m. 44 representsa modulationto F
major.Underthe circumstances, the pointcalls for discussion.
Fromthe littleNull tells us aboutBart6k'sstatement,it appearsthat
Bart6kequatesthe chordsof m. 36 andm. 44. Thiswouldseemto imply
thatfroma purelytechnicalpointof view mm. 3844, whichrepeatthe
E to F progressionin mm. 1-36 in shortenedform,arenot strictlyneces-
sary,the new segmentof mm.44ff. couldbeginshortlyafterm. 36. Now,
in mm. 44ff. melodic motionestablishesF major(or F Lydian,or the
acousticalscale on F) with clarity,but the perceiveris not yet awareof
this in m. 36. An analysisthatcouldevaluateBart6k'sclaimon the basis
of mm. 1-36 alone,is therefore,welcome.Such an analysis,althougha
littlecumbersome, is available.AssumingN6(= F6)at m. 36 andassuming
that this chordis presentalreadyat m. 26 can be arguedin two steps.
The bass chordsin mm. 14 and26 arebothneighborchords,the firstin
relationto the precedingand followingE chord,the second an appog-
giaturain relationto the chordover the A bass note. In this case, the
appoggiatura is soundedsimultaneously withthe upperchordtones of its
own resolution,and the boundarynotes of the discantmelody in mm.
26-35 show these upperchordtones are C and F. Thus,the chordis F6
(= N6) as Bart6kclaims,and it is therealreadyin m. 26, only its bass
is delayeduntilm. 36.
Thesegmentsmm.44-56 and 57-75. The two phrasescontainedin
mm. 44-75 are crucial.The tonalplan,the formalconstruction,andthe
role of bitonalityin the entireexpositionall hinge on them. The chart
shows that the phrasein mm. 44-56 is a full composingout of the F
majorchord,arrivingin mm. 55-56 to a half cadenceon V6, the tradi-
tionaldominantof F. The chromaticneighbornotes,C#andF#in the top
voice, producea low degreeof bitonality,which is often the case with
similarbagpipeeffects in Bart6k'smusic.This antecedentphraseis fol-
stlJulia Mlexifolos,sica
AcaclemiaeScientiarilm[flsn^,sarifae
.S7/1,19#

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 1. Waldbauer:TheoristsXViewson Bartok 115

lowed in mm. 57-67 and ff. by a highly bitonalconsequent.The bass

continuesin F with a composingout of the dominant(either V4 or
VIIX,a distinctionwithouta difference)and a returnto I of F in m. 67.
The discantis a well articulatedspinningout of a pentatonicphrase-
perhapsKodaly'stermhalf-tune(feldallam)wouldbe a betterdesignation
for it withC#as its recitationtoneandA#as itsfinalis, the latterreached
by the standardpentatonicclosingformulad#"-c#"-g#'-at', markedin the
chartby a horizontalbracket.The primacyof the bass in this bitonal
combinationis establishedby two factors.First,it is the expectedcon-
tinuationof the I of F in mm. 44-54, and it resolvesto the expectedI
of F in m. 67. Second,the tonalityof the discantis not fully determined
at this point. It could be pentatonicwith A# = BS as la (finalis), as
describedabove,but also F#= GS(major),D# = ES(minor),even C#=
DS (majoror minor).Moreover,at the cadencein m. 67 the discantap-
pearsto rejointhe I of F chord.Its G#= ASprovidesthe minor3rd of
the major-minortriad(gammachord)andthe A#= BS seems to lose (at
least momentarily) itsfinalischaracteralmostas soon as it appears.The
gamma chord at m. 67 is, of course, the first recurrenceof the set
projectedat mm. 1-14 and 7-29.
It is too soon at this pointto assess the significanceof the bitonal
phrasefromthe view point of formalconstruction,but the processthat
bringsbitonalityintoexistencecan be identified.It is an extensionof the
principleoperatingin the phenomenoncalled mistuningby Karpati.All
propertiesof the antecedentphrase(mm.44-56) suggestthatthe conse-
quentphrasewill begin with the discantnote C over the V of F chord
of the bass, but the actualcontinuationdeflectsone partof the expecta-
tion. Insteadof C, the discantis madeto slide up to C#,only the bass
continuesat its expectedlocation.The pentatonichalf-tuneis dislocated
in relationboth to the melodyprecedingit and to the harmonyaccom-
panyingit. It soundsas if it were playedon the mistunedstringsof a
violin. The role of this mistuningprocessis to endowthis passagewith
a very high degree of bitonality.It makes bitonalityeven more con-
spicuousthan the individualprofileof the melody and its exposurein
the top register.
None of the threeotheranalystsinterpretthese two segmentsin F,
whichis particularly surprisingin the case of Null andPetersen,who are
Stl/zli(lMt/xicolo^8^ic(/
AcatlemicleScientiClrllm
Htxn^,^clrifcle
37/1, 1S*S26

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 116 l. Waldbauer:
TheoristsX
Viewson Bartok

bothawareof Bart6k'sidentification of thatkey. Severalreasonsfor this

maybe conjectured, butI wouldlike to pointto a singleissue mostlikely
to misleadanalysts.TheLydianmodein general,andthe nearlymechani-
cal third-doublings in this passagein particular,tendto createa sense of
headlongrush,whichmakesit difficultfor the earto findthe appropriate
restingpoint.The temptationis strongto find a morefamiliarsounding
finalis, such as the thirdor the fifth degreesof the Lydianmode.Indeed,
both Null and Petersenchoose A, the Lydianthirddegree,as the tonic
at the beginningof the segmentof mm. 44ff., andWilson'sgraph(on p.
60 of his book) also begins with an A bass note at this point, but he
makes it clear that he does not considerthis A to have any tonal sig-
nificance.In the case of Null, his generalfixationon verticalelements
has also muchto do withthe choiceof A tonic,as it does withhis further
analysisof the two segmentsin mm. 44-75. He arguesthat Bart6k's
designationof F tonic is an error,becausefor him (perhapsalso for
Petersen?)the A with C-C#3rd aboveit (in mm. 36 and 44) impliesan
E a 5th aboveA, not an F a 6th aboveA. Thushe designatesA as tonic
in m. 44 andby the samelogic he sees a D chordfive measurelater.At
the end of the segment(mm. 55-56) he identifiesC as the new tonic,
on the groundthat the discantC# is a substitutionfor C natural,and
considersthe whole followingsegment,mm. 57-75 to be in this key.
Thus,m. 67 is IV of C for him andthe lastmeasure(or ratherthe second
chordof m. 76 over the D bass),II of C, whichhe thenequateswith V
of G to lead him into the next segment,whichhe considers,as we shall
see, to begin on IV of G. Petersen'sanalysisof these two segmentsap-
pearsin the form of an annotatedtable of his book (pp. 163-64). The
annotations give no rationalefor his choice of keys. He choosesA as the
primarycenterin m. 44, possiblybecausehe considersthe Lydianmode
on F withoutsignificance,or possiblybecauseA is the bass note of his
characteristic set, and designatesC# as a secondarycenter,noting its
bitonalimplications.For the sake of consistency,he mustthenchoose E
as the primarycenterat m. 55, but therehe can retainC#as the bitonal
secondarycenter,which is one of four possible interpretations for the
half-tunein mm. 56-67. His apparentsilence on m. 67 is causedin all
probability by a typographical error.Wilson'sprimaryconcernin these
segmentsis withthe operationsof his sets, he does not addressthe ques-
SltliaMuxic*lsif a AfatlemiaeSf ientiarilmAItm^arile S7/1. I fJ96

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 1. Waldbauer:Theorists'Viewson Bartok 117

tion of tonalityas such.But two graphsof his (on pp. 60 and61) display
sometonalimplications. Thefirstof these,a graphof mm.46-85, implies
an F tonic in m. 67, but as mentionedearlier,Wilsonexplicitlydisclaims
anytonalsignificancefor theA (m. 46)-D (m. 49)-A (m. 53) bassmotion
at the beginningof this graph.37
Thethreesegmentsin mm.76-92, 93-115,and 116-134.Petersen,
Wilson (by implication)and the chartare in agreementconcerningthe
threechordscontainedin the last threesegments.They are G#= AS in
m. 76, ESin m. 93, both of these outlinedby the discant,and ESagain
in mm. 116ff., identifiedby the ES-G-BSpedal and the tenormelody.
The chartinterpretsthem as IV6, I6, and I in ES.In addition,the chart
also shows thatthe bass ostinatiof mm. 76ff. and mm. 93ff. are bitonal
in thatthey fill out the Lydianpentachordon C and G, respectively,and
that the bitonalelementin the final phrase,mm. 116-34 shifts to the
discant,or ratherto thehighestmelodicvoice.OnlyNull'sanalysisdiffers
from the above, and it will be convenientto considerit at this point,
before any furtherdiscussionof bitonality.Accordingto Null, it is the
bass ostinatoratherthanthe discantthatis tonallydecisivein mm. 76ff.
and 93ff. He also shows his aversionto the Lydianmode when he calls
the ostinatoof mm. 76ff. a pedalon IV of G ratherthana pedalon the
Lydianpentachordon C. Althoughhe notes the G#minortriadof the
discant,he contendsthatthis is the upperneighborchordof G natural,
and as such, it confirmsG naturalas tonic. By the same rationale,he
designatesD as the tonic in mm. 93ff., consequentlyB7, the last note of
the discantmelody is for him the upperneighborof A, the 5th of the
impliedD chord.In whatfollows, Null glimpses,howeverobliquely,an
importantfact. He sees a role changebetweenthe A and the B; in mm.
112-115:A becomesneighborto BS,andBS,now a principalnote,serves
as dominantof E7,the tonicfromm. 116 on. An exchangeof roles does
indeedtake place here, as we shall see shortly,even if not exactly as
Null describesit.
"TheProblemof Prolongation
37 Wilsonrefershereto JosephN. Strauss, music,"Journcll
in Post-Tonal
of Music Theoe 31 (1987), 1-21, whichstates(on p. 1:S)thatthe A-D-A bassmotionis a "middleground
pun"in thatit appearsas if it wouldbe tonalmotion,whenin fact the significanceof thesepitcheslies in
theirbeingpartsof the A-DX andD-F-FW subsets.Clearly,neitherStraussnorWilsontaScesthe F major
outlinedby melodicmotionintoconsideration here.

.q7/1.199¢5
AcaulemiaeScientiarllmHIms,^aricae
Stl/ulitlMll.vicolo,Pica

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
118 l. Waldbauer:TheoristsXViewson Bartok

Returningto the chartandto thebitonalityin the lastthreesegments,

thereare severalreasonsfor hearinga low degreeof bitonalityin mm.
76ff. and 93ff., a high degree in mm. 116ff. In the first two of these
segmentsthe left hand ostinatirepresentthe secondaryelement.Their
separatetonal identityis barelynoticeablefor two reasons.First,their
respectivebass notes, C and G, are also the bass notes of the chords
impliedby the discantabovethem,turningthe latterintothe major-minor
triadsprojectedin mm. 1-14 and 7-29. Second,the melodicprofileof
the ostinati,none too distinctto begin with, is all but concealedin an
innervoice. Evenat the end of the tow phraseswhena stationarydiscant
allows the secondaryelementto achieve a degree of prominence,the
secondaryelement assumesa new scalarform, which loses whatever
Lydianidentitythe earlierostinatoform had. The scales in mm. 90-92
and 112-115 merelyoutlinethe diatonicgamutsof one sharpand two
sharps,respectively,enoughto differentiatethemfromthe gamutsof their
respectivediscants,but not enoughto determinetheirrespectivemodes.
In contrast,the top melodyof mm. 116ff.has a highly individualand it
may be added a well-remembered profile; it is exposed by its high
register;and its E-A-D scaffoldingprovidesmaximumconflictwith the
ES-BS-ESscaffoldingof the surrounding pedal.
Thesegmentof mm. 116ff. is thefinalphraseof the entireexposition,
and its climacticqualityowes muchto the two processesthatbringits
bitonalityinto existence.One is the species of mistuningencountered
earlier,the othera species of voice exchange,or as medievalistswould
call it, Stimmtausch.To see bothmistuningandStimmtausch, the phrase
must be consideredin conjunctionwith its antecedentphrasein mm.
93-115. The antecedentis in ED,andits discantendswiththe descending
skip fromESto BS.Precedent(of mm. 86-93) andtonaltraditiondictate
that BS be the first discantnote in the consequent(m. 116), folklore
traditionwould allow eitherBSor AS.Instead,it is A natural,and only
the fabric surrounding the melody producesthe expectedES harmony.
The melodyis dislocated(or mistuned)in relationboth to its antecedent
and its accompaniment, and makestherebybitonalityas conspicuousas
it was in mm. 57ff. This time, however,Stimmtausch(of sorts) adds a
new dimensionto the process.To be sure,the exchangehere is not the
traditional one, exchangingthemelodiesof the bassandthe discant.What
Stlwcli( Mllxicolo,s,}ica Acac/emiae Scientiarlwm Hlwn,s,(ricae .s7/1. 195F

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 I. Waldbauer:Theorists' Viewson Bartok 119

happensis that the two-sharpgamutof the antecedentbass rises to the

top to becomethe gamutof the consequentdiscant,while the three-flat
gamutof the antecedentdiscantdescendsto become the gamutof the
consequentbass. The presenceof GS alongsidewith G naturalin the
antecedent,and the spreadingout of the EDchordover five octaves in
the consequentdoes not alterthe prevailingsense that an exchangeof
gamutshas takenplace. Andjust as Stimmtausch in medievalmotetsor
Baroquecounterpoint, theexchangeof gamutshas the effecthereof equi-
libriumachieved,of logical processcompleted.Togetherwith the high-
profilebitonalityof the final phraseparallellingonly the equallyhigh-
profilebitonalityof mm. 57ff., the exchangeof gamutsimpartsa sense
of finalityto this phraseendingthe entireexposition.

Theroleof thebitonalpassageinmm.57-75
intheformalconstruction
Withthe tonalplan of the exposition,E (143)-F (44-75)-ES (76-
134)laidbare,the role of the bitonalpassageof mm.57-75 in the formal
constructioncan now be assessed.It is uniquein thatthe F, the key of
the bass andthe primarykey in the entirepassageof mm.44-75, belongs
to the firstpartof the exposition,andthe key of the pentatonichalf-tune
in the discant,althoughnot yet fully determinedat this point, belongs
alreadyto the E; of the secondpart.It is as if two segmentshad been
telescopedinto one, the bass finishingoff whatwent before,and simul-
taneouslythediscantbeginningwhatis to come.To his credit,Null argues
this Janus-likecharacterof the passageon rhythmicandmelodicgrounds
(Null 1930,p. 90), buttwo arguments on tonalgroundsarealso available.
The firstargumentis directedtowardsthe sensibilitiesof the normal
western-trained musicalear.Afterhavingheardthe entireexposition,this
westernear will summarizeits large-scalemelodic motion as the des-
cendingsequenceof Db(m. 57)-B; (m. 67)-AN(m. 76)-EW(m. 93). The
sequenceis a recognizablehallmarkof the EStonality,and it has also
been well preparedin that it containsboth the rectoand inversoforms
of the nucleusof all expositionthemes.Bitonalityin mm. 57ff. alertsthis
westernear that somethingmomentousis happeningin mm. 57ff., but
that this is an accelerationof the pace at which the form developsand
that it comes aboutby the telescopingof the F and ES tonalitieswill
St",li, Mt..ic,,l,,ltwic(
Ac(tdemi(eScienti(l-limBuns2(ric(te@7/1.19%

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 120 I. Waldbauer:Theorists' Viewson Bartok

become apparentto it only afterthe whole expositionruns its course.

The secondargumentconcentrateson the local events withinthe large-
scale melodicoutline,andby showingfolkloreinfluencein thesedetails,
it also shows that the folklore-trained Hungarianear has a certainad-
vantageoverthe westernear.Thecrucialfactoris thepentatonichalf-tune
in mm. 57ff. The Hungarianear dismissesthe possibleDS and GSinter-
pretationsof the tune. It fastensonto BS as tonic. Moreover,since it is
accustomedto hearingsuch half-tunesansweredby theirtransposition a
perfect5th lower- in reduction,the antecedentof dS"4"-dS"-aN'-bS'
answeredby g;'-aN'-gS'-dS'H'consequent it regardsBS as a merely
temporarytonic andexpectsalreadyat m. 67 thatthe final tonic will be
ES.Whatactuallyfollows in mm. 76-92 is not a literalrepetitionof the
precedingphrase,but it does neverthelesslead to the EStonic of m. 93.
The advantageof the Hungarian ear is thustwofold.It anticipatesthe ES
tonic early in the proceedings,and it is in a positionto perceiveim-
mediatelythe dual role of mm. 57-75 as consequentof the preceding
phraseand antecedentto the next one. It is likely to become awareof
the telescopingeffect rightas it is takingplace.

Conclusion

Fromthe viewpointof theory,the mostimportant conclusionoffered

by the foregoinganalyticalexerciseis thatset theoryand the conceptof
structuraloverlay are necessaryto Bart6kanalysis.Wilson's sets are
recognizable, perhapsmoreeasilyby the earthanby the eye; theyunfold
into a structureof theirown; and the structureoperatesin simultaneity
withotherstructures,frequentlyintersectingwith them, but completely
independent from them.Two questionsarise of theirown accord.How
muchBart6kmusiccontainssuch independently operatingset structures?
Theinformedguessis, quitea few,fromtheearlyBagatellesto the Sonata
forSolo Violin.The other:Is therea discoverableprincipleunderlying
allcases of structural
overlay?In view of the infinitepossibilitiesandin
viewof Bart6k'savowedcase-by-caseapproach,this is not likely.
The analysisof the melody,harrnony, and tonaldesign,contributes
moreto the understanding of the expositionof this particularpiece than
to theoryas such. Theoryhas knownfor a long time that one way or
Sttttli(l
Mttxicolss,0ic(
Ac(ttlemi(teScientiartJm9ns,(triczte 37/1. 19X

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions
 I. Waldbauer:Theorists' Viewson Bartok 121

another,tonaldirectionis alwaysdiscerniblein Bartok'smusic,and also

thatthe meanswherebyit is achievedandthe patternsin whichit results
are an ever varyingamalgamof westernand folkloretraditionsand of
novel non-tonalor atonalprocedures.In this piece, the structuralstrand
of Wilson'ssets representsa fully non-tonalprocedure;the melodiesand
the tertianharmoniesthe melodiesgenerateare almostfully traditional;
the tonaldesignresultingfromthe harmoniesis in parttraditional, in part
non-traditional.Eachhalf of the expositionis easily understoodin terms
of westerntonaltradition-the E I-F I-F V-F I progressionin mm. 1-75
and the ESIV6-ESI6-ESI progressionin mm. 76-134- but the way the
two halvesare telescopedinto each otherin the middleis not only non-
traditional,but, as far as I can tell, uniquein the entireBartokoeuvre.
If thereis a moralfor musictheory,it is thatany analysisthatdisregards
tonaldirection,particularly traditionaltonaldirection,will be incomplete,
at least to thatextent,possiblyincompletebeyondthatextent in that it
will miss importantoperations,for instancethe telescopingof tonalities,
thathinge on a traditionaltonalexpectation.The same caveatappliesto
folkloreinfluence.The listeneror performercan takethejust measureof
a given piece withoutconsciousawarenessof the role of folklore,just
as he or she can do so withoutconsciousawarenessof tonaloperations.
The analystintentupon accountingfor all forces at workin shapingall
particularsof the analyzedpiece cannotaffordthis luxury.

Stttdi, Mlwxicoloswic(
Ac(Xder1?iae
Scienti(ritmHltngyaricaeS7/1. 19X

This content downloaded from 212.183.209.30 on Mon, 7 Apr 2014 11:45:25 AM

All use subject to JSTOR Terms and Conditions