Liszt S Androgynous Harmony

Liszt's Androgynous Harmony
Author(s): Zdenek Skoumal

Source: Music Analysis, Vol. 13, No. 1 (Mar., 1994), pp. 51-72
Published by: Wiley
Stable URL: http://www.jstor.org/stable/854280 .
Accessed: 16/05/2013 04:07
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.
Wiley is collaborating with JSTOR to digitize, preserve and extend access to Music Analysis.
http://www.jstor.org
This content downloaded from 147.91.236.58 on Thu, 16 May 2013 04:07:57 AM

All use subject to JSTOR Terms and Conditions
ZDENEK SKOUMAL
LISZT'S ANDROGYNOUS HARMONY
The weakeningof tonal syntaxthat characteriseslate nineteenth-century

music was broughtabout in a varietyof ways duringthe course of the
century.* The established conventions of the preceding style were
transformedby such means as increasing levels of dissonance, non-
functionalharmonicprogressions,non-diatonicscales and the attenuation
of the tonic-dominantpolarity.'The late worksof Liszt demonstratemany
of these characteristicsassociated with the move towards atonality.In
particular,theyshow a weakeningof the traditionaltonic-dominantaxis by
minimisingthe contrastbetweenits two constituentelements.The music
featuresa harmonythat,in a sense, is tonic and dominantsimultaneously,
creatinga synthesisof function.I term such harmonyandrogynous, i.e.
having the characteristicsand nature of both tonic and dominant.2
I first
discuss the phenomenon with relativelyobvious examples from Liszt's
music writtenbefore 1880, then examine threelate piano pieces in more
detail:Nuagesgris,La lugubre gondolaI, and R. W.- Venezia.3
The V-I motioninvolvesthreedistinctelements:1) the bass leap 5 to 1;
2) the step 2 to 3 (or 2 to 1); and 3) the semitonemotionI to 1. The V-I
motion is weakened to various degrees by alteringone or more of these
elements,forexample,by inversionof one or both of the harmonies,or by
the use of anticipationsand elisions. AlthoughLiszt employed the V-I
gesturethroughouthis music, the laterworksweaken it frequentlyand at
times significantly.The finaltonics and dominantsare ofteninvertedand
in some cases partiallyhidden,usuallythroughsubstitutionor additionof
non-harmonictones. However, of the three contributingelements,Liszt
takes care to preservemost carefullythe 7-1 motion.The resolutionof the
leadingnote is sometimesone of the onlyindicationsofthe tonal centre:
* This
studywas completedwiththe help of a post-doctoralfellowshipfromthe Social Sciences and Humanities
Research Council of Canada. WilliamE. Benjamin and JohnRoeder made helpfulsuggestionson an earlierdraftof
the article.
MUSIC ANALYSIS13:1, 1994 51

? Basil Blackwell Ltd. 1994. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and
238 Main Street,Cambridge,MA 02142, USA.

ZDENEK SKOUMAL
Outsideofthoserareinstancesin whichkeyorientation mayonlybe

ascertained
by subjectivespeculation(e.g., Unsternand Bagatelleohne
Tonart),the resolution
of the leading tone to the tonic is usually
sufficient
meansforestablishing tonalidentity... 4
In this way Liszt reduces the most tonal of all harmonic events, the
dominant-tonicprogression,to the smallest possible event: a single
semitone shift.In his earlierworks this semitonemay be only a surface
detail,but it gains importancein the late style.
ANDROGYNY IN THE MAJOR
In addition to chord V, the leading note belongs to VII and III. The
diminishedtriadon VII is typicallyunderstoodas a typeof dominant,an
incompleteV7. III, on the otherhand, is a more interestingstructurein
thatit may functionas both a tonic and a dominant.The reason is thatit
shares two notes with each chord.5 Usually the context clarifiesthe
function(Ex. 1).
Ex. 1 Two FunctionsofIII
d d d
In b. 1 of Ex. 1 the III chord functionsas a tonic,arisingfromthe passing

note in the soprano, while in b.2 it is a dominant, arising from a
substitutionof 6 for5 withinthe dominantchord. Nevertheless,this kind
of clarityof functionbecomes less common in Liszt's late works.In certain
cases the III functionsas both I and V simultaneously.
The 'Agnus Dei' from Liszt's Organ Mass (1879) ends with the
followingprogression(Ex. 2):
Ex. 2 OrganMass, 'AgnusDei'
40
pV III
52 MUSIC ANALYSIS 13:1,1994

? Basil BlackwellLtd. 1994

The V is prolongedthrougha neighbouringIII and thenresolvesto I. The

bass subsequentlyrepeatsthe cadentialgesture,but a pedal D in an upper
voice makes the chord at b.36 a IIPI, not a V. Of course it does not
functionas a mediant (it is an 'apparent' III),' but ratheras an altered
dominant (6 replacingS) that prolongsthe finaltonic. Nevertheless,the
pedal D and the common note F give this dominant a strong tonic
character.In termsof pcs, only a single semitonedifferentiates the V and
the I. The reason we hear a relativelystrongmotion to V is the leaping
bass; withoutit we would be more likelyto perceivethe chordas an altered
I (like the firstIII in Ex. 1). As it is, both harmoniescan be heard; it is
both tonic and dominant- it is androgynous.In the followinganalyses I
representsuch harmonicconstructswiththe symbolV. Like the harmony
itself,it is a combinationof I and V (as well as a distortedIII).
The ambiguityof I and the alteredV becomes particularly notablewhen
the contextitselfis ambiguous; the true meaningof III is ambiguous and
its functionuncertain.Two othermovementsof the Mass end in a similar
fashion,thatis, III to I, the III betterlabelled as V. However,in the other
movementsthe bass does not leap 5 to 1, and hence the functionof these
dominantsis less certain(the strongerclose of the 'Agnus Dei' is clearly
necessitatedby its functionas the close of the entireMass). Ex. 3 shows
these two endings, as well as the prototypicalharmonic patternsthey
represent.
The 'Kyrie' followsa non-traditionalharmonic structure(to which I
returnlater) and ends witha shiftfromVIITj up a semitoneto the tonicB.
The shiftoccurs via T. More conventionallythis chord would be V6, but
the third of the tonic, the D, arrivesearly. The 'Sanctus' ends with a
varianton the familiarprogression?VI-V-I (NVIrespelledforconvenience).
Liszt firstaltersthe mode of the NVI- actuallyanticipatingthe thirdof the
next chord - then anticipatesthe fifthof the tonic, again producinga V-I
ending.In both cases the finaltonic arrivesas a resultof a singlesemitone
motion upwards; the cadential close becomes more of an appoggiatura
gesture.The penultimatechord already seems like the tonic. Yet it also
representsthe dominant;thereis no otherharmonyto take the dominant
function and the ending is unsatisfactorywithout it. The chord is
androgynous.7
The use of V is not limitedto Liszt's churchmusic. The songs depart
from conventionaltonal structuresto varyingdegrees, but their final
gesturestypicallyintroducethe leading note-tonicmotion as a dominant-
tonic reminder.A simple example is 'Ihr Glokken von Marling' (1874)
(Ex. 4). The song is not exceptionallychromaticand bringsa relatively
clear structuralV in b.41 (in second inversion). Although a EVI-for-I
substitutionin b.42 delaysthe tonic arrival,the I does appear in b.45. Like
the dominant,the tonic is inverted,somewhatweakeningits effect.It is
prolongedwith subdominantand supertonicharmoniesfromb.45 to the
end of the song, settingthe singer'stripleplea 'ihr Glokkenvon Marling,
MUSIC ANALYSIS 13:1, 1994 53


ZDENEK SKOUMAL
Ex. 3 OrganMass, endingsof 'Kyrie' and 'Sanctus'
'Kyrie' from and

35
I VII I V I
Bb6 VII V1
'Sanctus'
-P-0
Db: bVI- 16
from and
,,it R C
bVI1 V 1 6 bVI V 16
behiitetmich gut!'. The final three bars decorate the tonic with the
neighbouringmotion E-D#-E.The D#is the leading note and produces a
T. This is the only7-i motionsince the structuralclose in bs 41-2, a subtle
but characteristicconfirmation ofthe tonic.
In contrast,the song 'Sei still' (1877) employsthe 7-I gestureas one of
the importantstructuralideas. The text poses questions about life and
death and suggeststhatboth become less difficult when our heartis calm,
as God wishes.The song is in D major,but the firsttonicharmonyand the
firstimportanttonic pitch do not appear until b.27, the exact midpoint.
This bar sets a crucialword of the text- 'Leben' - and containsthe song's
climax,both agogic and dynamic.It is the culminationof a graduallyrising
chromaticline: C (bs 10-11), C#(bs 22-3, 26), D (b.27). Despite a melody
that for the most part calls for I and V7, a C#pedal makes the primary
harmonyin the firsthalf of the song IIII alternatingwith VII7. III6 also
precedes the firsttonic at b.27. Thus III replaces I, the resultof a 7-for-i
substitution.In thiswayLiszt saves the D untilthe climacticmoment.
The second halfof the song focuseson P6; it containstwo II6-VII4-V3-I6
progressionsthatbringthe tonic in a more convincingmanner.The final
54 MUSIC ANALYSIS 13:1, 1994


Ex. 4 'Ihr Glokkenvon Marling'
41 f _ p
ihr Glok - kenvon Mar-ling, be - hii - tetmich gut, ihr
Sf -p
=
46fA
Glok - kenvon Mar-ling, be - hii - tetmich gut,
A ii
pb g I a. p
51 p
. tI
ihr Glok - kewrvon Mar- ling, be - h - - - - tet mich gut,

poco a poco nt.
be - hu - tetnich gut!
perdendo
s. Cb* * . *
*cb.
*
16 6
16


ZDENEK SKOUMAL
Ex. 5 'Sei still'
Langsam
Ach, was ist Le - ben doch so schwer, wann,was du lieb hast,
Ip
IV
D ist nichtmehr: a- bersei still, sei still; weil Gott es will!
-iD
13
Ach, was istSter-ben doch so schwer, wann,was du lieb hast,
18
weint
umher: a-bersei
still, sei still; weilGott es will!
P m
a
_ . -__ oo -
--


Ex. 5 cont.
24 >
Ach, Le - ben, Ster - ben wirnichtschwer, wenn
if 1?
-
V III6 16
31 - rit. lang
I JI. ,I I i L " I
un - ser Herznur stil- lewlr, wennun - ser Herznur stil - le Dar - um sei
wir.
37 39
wel
still,Gott
eswill,
dar
-um
seistill,
welGott
eswill,
S, I- -
IV16 VV vV I --_-- -0 116 VII V 6 I:
1
44
sei still, weil Gott es will.

cf.bs 1 + 39
i6 I I' I 6 -
?
16 P 16 Y6 16 16


ZDENEK SKOUMAL
tonic is prolongedwith a T-I6 alternation.8Thus, in 'Sei still',T is more

than a briefsurfacedetail; its appearance in the closing bars representsa
reflectionofthe song's overallstructure.This is especiallyevidentin bs 50-
1 whereT appears withC#in the bass, just like the prolongedsonorityof
the firsthalfthatpreparedthe climacticvocal sigh.
ANDROGYNY IN THE MINOR
In minor keys the phenomenon is similar. The main differenceis the

quality of the III: in a minor key, a III with the leading note is an
augmentedtriad. Because of its symmetricalstructurethe augmentedIII
containsthe same pitchclasses as an augmentedV and usuallyfunctionsas
a dominant.The resolutionofthischordto a minortonicis a phenomenon
thatparallelsthe precedingdiscussion:two of the pcs are the same, while
the thirdpc is the leading note thatrises a semitoneto the tonic. In other
words, the differencebetween the tonic and the altered dominant is a
semitone in one of the voices. I shall again employ the symbol T to
representthisharmonicfunction(Ex. 6).
Ex. 6 Augmenteddominantresolvingto minortonic
0k069
The resolutionof an augmented dominant triad to a minor tonic is
generallya weakerharmonicmotion than a similarresolutionto a major
tonic. In a major key the alterednote behaves as a second leading note,
resolvingto the thirdof the major triad.In a minorkeythe alterednote is
the same as the thirdof the tonic and hence cannot resolve in the same
way. The strong dissonance of the altered dominant becomes a con-
sonance withinthe tonic. The resolutionis strongif the augmentedtriad
resultsfroma suspension,an escape note or melodic arpeggiation;thatis,
when a downward stepwise resolutionis assumed (Ex. 7a). If the aug-
mentedtriadresultsfromanticipationor a pedal, the resolutionis weaker
(Ex. 7b). In such cases the dominantelementhas a strongtonic character
and may be perceivedsimplyas an inflectionof the tonic. It is just such
situationsthat become common in Liszt's later music; hence the tonic-
dominantcontrastis weakened.
In a major keyan alteredV invertswithoutdifficulty. In the minor,the
inversionsare quite different.Although the augmented triadsare the same
ones that function as altered dominants in the 6 inversion,in other
inversionsthe impressionof dissonance changes.The most dissonantnote
withinthe 6 is stable in the ' and 6 (Ex. 8). In the ' inversionthe chord
typicallyresultsfromthe chromaticpassing motion -#~5-6, III-I6, rather


Ex. 7 Originsof the augmenteddominantin the minor
usp. IN arp.
ant.. pedal
b)
than froman altered-V-Icadence. In the 6 inversionthe chordresultsfrom

a simple neighbouringmotion of the bass. That note is the perceived
dissonance;the othertwo forma consonantpedal. While thisis clearlythe
largermeaningof the harmony,in both cases the pc set is equivalentto the
augmenteddominant.
Ex. 8 Inversionsofthe augmenteddominant
6 56
3 34a
Liszt employedall inversionsas dominants.In the PetrarchSonnetNo. 104

the augmentedtriadfunctionsas an applied dominantto VI (Ex. 9, b.8).
Nevertheless,note the E/G#pedal; even though the chord is an applied
dominant,it is also an inflectedVI. Like the III chords in the major, it is
androgynous.
Ex. 9 Petrarch
SonnetNo. 104, bs 7-9
E 7'D I I
riten,
I g
E: I
The symmetry of the augmentedtriadallows easy reinterpretation

of its
root. In II PenserosoLiszt uses this propertyto modulate fromE major to
G major (Ex. 10). The augmentedtriad in b.9 is a neighbouringaltered
dominantto E minor.G and B remainas a pedal, while D#is the leading
note that rises back to the tonic. In b. 11 a respelling of the same


ZDENEK SKOUMAL
augmentedtriadproduces an altereddominantof C minor.It resolvesto a

C minorchord in b. 12 by keepingE? and G as a pedal and treatingthe B
as a leadingnote. The C minorchord thenfunctionsas IV3 of G, the goal
of thismodulation.Thus, the augmentedtriadacts as an altereddominant
to two differentharmonies;it changes its meaning by alternatelyraising
one, thenanotherchordmember.Even thoughthe largereffectis relatively
dramatic(a shiftup a third),the local effectis subtle:the augmentedtriad
is perceivedonlyas a chromaticinflectionofthe two tonics.
bs 8-13
Ex. 10 II Penseroso,
E:
ap
l A- rinforz. 2I
G: VI
S ,UI I t
,.J ,i
IL 4>*
IV6 V7;
The opening of the Faust Symphony(1854) is a celebratedexample of

'dissonant prolongation';it prolongs an augmentedtriad for twenty-two
bars.9The listenerbegins to feelthatlocallythe augmentedtriadis stable,
the referentialpoint. This is, of course, a slow introduction,like manyby
Haydn and others. Whereas in the Classical style a slow introduction
typicallyprolongs the dominant which resolves to the tonic at the
beginning of the Allegro, here the mysteriousopening leads into the
subdominant.At b.24 the prolonged augmented sonorityresolves to F
minor6, the IV of the tonic C minor (Ex. 11 la). The primarybass note of
the introductionis Al and it remainsas the bass of the chord of resolution.
The means ofresolutionis a singlesemitonemotion:E risesto F."o
In bs 359-81 (within Rehearsal Letter W) the introductorypassage
returns. This time, however, the prolonged augmented triad is
reinterpreted as an altereddominantof A minor and resolves as such in
b.382. Abbecomes G#and rises a semitonewhile the othertwo notes stay
as a pedal. An E-F motion follows,now representing3-6 of the local A
minortonic (Ex. 1ib). In both partsof the movementit is difficult to say


introductionand itsreturn
Ex. 11 Faust Symphony,
1 4 6 11 15 17 22 24
Cm:-- IV6
359 382 384
b)5 6 5 6 6
Am: I 6
unequivocallythatthe prolongedaugmentedtriadis a dominant;thereare

two notes from the dominant but also two notes from the following
tonicisedtriads.The harmonicfunctionis best labelled as V.
Liszt's song 'Die tote Nachtigall' (1878) involvesseveralshiftsfromF#
minor to F# major, correspondingto the textual images (generally
associating minor with the ill-fated nightingaleand major with the
awakeningof spring). The closing section, which returnsto the minor,
('... und ach! keinRuf erweckt dich wieder,du arme,kleineNachtigall!') is
fourteenbars long and containsonlytwo harmonies(Ex. 12).
The augmentedtriadis an inflectedtonic,but at the same timeit is the
only candidate for a dominant.There is no otherdominantto close the
song, even though one is necessaryafterthe diminishedseventhof the
precedingbars. When the augmentedtriad firstappears it is heard as a
dominantbut by the time it resolvesit is feltmore as a tonic. The only
eventof harmonicimportwithinthe last fourteenbars is the resolutionof
the leading note. That whichwas an almost insignificant detail at the end
of 'Ihr Glokken von Marling' becomes an importantevent here. The
leading-noteresolutionembodiesthe moretraditionalV-I close.
The pieces I have discussed thusfaralwayssubsumethe ambiguityof V
withina larger,tonallyclear context.The remainderofthisstudyexamines
threelate workswherethe contextitselfbecomes increasingly uncertain;in
thesethe focusbeginsto shiftfromI to V. It is herethatLiszt appears to be
moving in the direction of atonality. Nevertheless,my analyses still
interpretthe music in terms of the traditionfromwhich Liszt arrived.
Leading note activityformsthe focus of the discussion; the leading-note
MUSIC ANALYSIS13:1,1994 61

ZDENEK SKOUAMAL
Ex. 12 'Die toteNachtigall'

Ir 4 ad lib.
ftr
-L Lr I .tr L.tr
Nach - ti-gal - - len Lie der,
56
schlafst
Dann schlfst
Dann du
du still
still in Gra -
in Gra bes Nacht,
bes und
und ach!
Nacht,
ach!-
I
TrIiten.
TI kein Ruer - weckt dich wie - der, du
LL-
ar - me, klei - ne Nach

-r- - -
ti
-
gall!


resolutionspawnsothersemitonemotionswhichtake on a motivicrole.
NUAGES GRIS (1882)
Nuages gris is a model example of the minimal contrastbetween the

augmentedV and minorI; the basic structureof the piece consistsentirely
of theiralternation.Bars 41-2 contain an arguablepre-dominantsonority,
while otherchords resultfromeitherneighbouringmotionsor chromatic
passing connections that prolong the augmented V. As a related
phenomenon,the semitonebecomes the primarymotivicidea.
The G minor tonic is relativelyclear: the opening melodic gesture
beginson the dominantnote and thenoutlinesthe tonictriad,the melodic
fragmentsof bs 25-32 are stereotypicalG minor gestures,and the final
melodic motion is from F? to G. The key signatureof two flats also
indicatesG minor.However,the G minorsound is disturbedby numerous
chromaticpassingchords,an ambiguousand aimlessbass, and the famous
whole-toneending.
The altered V of G minor is D-F?-Bb. This androgynousdominant
shares D and B% with the tonic; the only differencebetween the two
harmoniesis the semitoneF#-G.Liszt furtherminimisesthe contrastof
tonic and dominantby employingone of the commonnotes,the B%,as the
primarybass. The background harmonic progressionthat controls the
piece thereforelooks as follows(Ex. 13):
Ex. 13 Harmonic designofNuagesgris

1 9 21 25 33 47
Gm: [6 ?
?r [I I
G m: 16 ? 16 16
I "
The tonics expand throughsimple repetition,while the androgynesdo so

throughchromaticlines. Bars 21-4 emphasise Bb/Dand avoid F0 or G;
hence theyremainintentionally ambiguous.The last harmonyin the above
is
example hypothetical, as Liszt leaves the endingpartlyunresolved.The
reductionclearlyshows thatthe dominantelementhas been reduced to its
absolute minimumand hence may be perceived as an inflectionof the
tonic. This is the same phenomenonwe saw earlierbut on a largerscale,
involvingthe entirepiece.
F? to G - the only differencebetween the tonic and the altered
dominant- is a crucialeventand formsthe climacticresolutionat the end
of the piece. Other semitones that surround the tonic triad become
motivic:CO-D, E&-D, A-B%(C&B%appears only coincidentallyas part of a
chromaticline, and then reappears as a Picardy B&-B?at the end). In

? Basil Blackwell Ltd. 1994

ZDENEK SKOUMAL
bs 9-20 and 33-43 the repeated bass neighbour alternatesthe

B,-A-Bkmotivesgrow
altered and fundamentalformsof V. From the chromatic
chromaticlines: R.H. 9-20, L.H. 33-42, R.H. 33-48. Ex. 14 showsNuages
in greaterdetail. The example includesthe large-scaleunfoldingof the
gn'ris
T- thatAllen Forte revealsin his analysisofthe piece."
I
Ex. 14 ReductionofNuagesgris
'Y-I
2I N'33--------
21
Gm: I ? I , I
In the concluding bars the bass remains frozen on A and forms a

dissonance with the final tonic of the melody. Likewise, an inner-voice
neighbouringE remains unresolved.These dissonances are necessitated
by the precedingbars; the highlevelof dissonanceat the specifiedAndante
tempo engenders a banal, almost ridiculous effectin a resolutionto a
simple triad. The Picardy thirdonly accentuatesthe problem. However,
since the crucialleading-noteresolutiondoes end the piece, Liszt is freeto
retainotherdissonances. His solution is ingeniousin that it providesthe
necessarytonal closure and yet retainsthe aura of mysterythat pervades
the entirepiece.
LA LUGUBRE GONDOLA I (1882)
While Nuages griscontains two significantprolongationsof , La lugubre

gondolaI is in its entiretyone large statementof 1, with only incidental
appearancesof the tonic.The music leads to a pointwherea decisivetonic
could appear, but instead the initial T returnsand the piece ends as
ambiguouslyas it began. In a way La lugubregondolaI emulatesthe Faust
Symphony introduction, but withouta definitive
resolution.
The piece has been somewhat misunderstoodin published analyses.
The main problemis the key;it is alwaysgivenas F minorwhen in factthe
tonic is D6 minor. The confusion is understandable. The prolonged
sonorityis the augmentedtriad E-A&VC,and hence can implyan F, A or
Db tonic. A recognitionof the D6 as tonic is essential for a proper
understandingof the piece and its large-scaleuse of T. The reasons for
64 MUSIC ANALYSIS13:1, 1994


preferringF minormightbe as follows: 1) the opening key signaturehas

fourflats;2) the opening melodic intervalC-A is typicalof an F minor
piece; and 3) the bass effectsa standardF minorgesturein its firsttwenty
bars- E?-F(leading-note
to tonic,Ex. 15). It has also beenremarked
that
F minor appears to be Liszt's funeralkey, acting as the tonic in other
funerealworks.'12
Ex. 15 La lugubre
gondolaI, bs 1-23
Andante marcato
semprelegato
una corda
13
19 semprelegato
:' Lll UWUFWL.W r uiw I -)do

I,
MUSIC ANALYSIS13:1,1994 65
C Basil BlackwellLtd. 1994

ZDENEK SKOUMAL
The primary problemwithF minor,however,is the melody;afterthe

openinginterval, F minoris possibleonlythrough chromaticism. It moves
to A?(possiblyexplainable ?s modalmixture), thenG, FK,E~ - hardlyan F
minorgesture.JamesBakerhas observedthatthe 'melodyunfoldsa sixth
descendingfromA? to C9, althoughthe exactcourseof thisunfolding is
difficult
to analyse'."It is,however,mucheasierto analysethemelodyout
of context;it is movingdiatonicallytowardsDb minor,thoughconstantly
avoidingthe tonic byjumping to theleading-note C. It finallydoes landon
Db in b.19, the end of this melodic section (a new melodic idea
immediately follows).The harmony in s
in b.19 D6 - not an F minor tonic
withan added sixthbutrathera majortonic withan added seventh.Here
is wherethe modal mixtureoccurs,not bs 6-7.'" The A?ofbs 6-7 is not the
resultofmodal mixture;it is a misspelledBb.
The main problem with the DB minor reading is the F minor key
signature.There are two possible reasons. First,thereis no key signature
forDB minor(it would requireBb%);hence Liszt uses the firstkeysignature
that includes the tonic Db (especiallysince that note appears in everybar
from 1 to 28). The second - more contentious- possible reason is that
Liszt himselfinitiallyfelt F minor, with the leading-noteE? and the
openingmelodic gesture."
The reading of a DB minor tonic is supportedby subsequent events.
Bars 1-34 are immediatelyrepeated at Tio, with only minor variations.
Apart fromthe transposition,there are two significantchanges: the key
signatureis two sharps,and the second halfof b.44 containsa G?, not Fx.
F minor at Tio produces D# minor or Eb minor, whereas Db minor at
Tio produces B minor- two sharps.The melodic line now clearlyshows a
B minor scale, focusingon the leading-noteA#,then finallyreachingthe
tonic B in b.57. Beginningat b.77 the openingmaterialreappears,greatly
abbreviated,at Ts. The keysignaturehas no sharpsor flats.Db minorat Ts
producesA minor,no sharpsor flats.
Treating DB minor as the actual tonic of the piece, we can make the
followingobservations.The tonic firstappears withinthe openingpattern;
the Ab-C-E?arpeggiowitha Db neighbourrepresentsan altereddominant6
alternatingwitha tonic 6.16 There is littlesense of harmonicchange; only
one harmonyis projected and it fuses the two harmonicfunctions.The
tonic returnsmore clearlyin the parallel mode (bs 19ff.),it is strongly
expected in bs 101-3, and it also returnsin the finalbars (especiallyin
b.116). Yet it is not established as the primarysonority;overall it is
subsumedwithinthe prolongedT.
The overallbass descent outlinesthe same augmentedtriad; it moves
down fromthe fifth(b.1), throughthe third(b.77) to the root (b.95). This
is shown in Ex. 16 (I have changed the key signature and respelled some of
the notes to emphasize the DE tonality). Once in root position the
augmented triad appears ready for a resolutionto the DB minor tonic
(perhaps in b.101 or b.103). The dynamics emphasise that moment, as a


crescendobeginswiththe arrivalof the bass A? in b.95. Nonetheless,in the

present context such a resolutionwould seem to be anachronisticand
somewhatbanal (thoughnot as disturbingas a major triadat the end of
Nuages gris). The point of this piece is not to prepare a glorious tonic
resolution but to highlightthe ambiguous nature of the augmented
dominantin a minorcontext.Instead of the tonic,b.101 bringsback the
bass E and the androgynousharmonycontinuesto the end of the piece.
The finalbars containonlytwo pcs in the lefthand tremolowhilethe right
hand adds the third,determining note. Bars 115-17 waverbetweenC and
Db (a neighbouringfigure taken directlyfrom the opening left-hand
gesture),and it almostdoes not seem to matterwhichone willbe last.
Ex. 16 ReductionofLa lugubre

gondolaI
1 19 39 57 77 87 95 101 116
D
,m:
Iy 15 VII /l /l
/VI
s /VII
"
R. W. - VENEZIA (1883)
Of the threelate piano pieces I analyse here, R. W. - Veneziais the most

ambiguous in termsof its large-scaletonal orientation:'R. W - Venezia
demonstratesthat on occasion Liszt found the confidenceto dispense
altogetherwiththe framework of tonality'."There is verylittleto help the
ear. Nevertheless,the structurallogic becomes less of a
tonality-seeking
mystery once we are aware ofthe earlierworks.
The tonic is B%, though its existence is precarious. It seems to be
establishedonly throughone augmented-Vto I6 motion (bs 30-4) and is
supportedby the key signatureof two flatsin bs 31-8. However, the true
confirmationof B%comes withinthe outer sections of the piece. These
expand the T of B%minor and hence point to a rathertraditionalminor-
major-minorternarydesign:
A (bs 1-30) - expandsV of B minor(C? pedal)
B (bs 31-42) - presentsthe tonicB%major
A' (bs 43-9) - returnsto V (C#pedal)


ZDENEK SKOUMAL
The C#pedals representboth C#and Db, dependingon immediatecontext;

the pitchis eitherthe alteredfifthof the dominantor the thirdof the minor
tonic.Ex. 17 summarisesthe piece in a graphicformat.
Ex. 17 SummaryofR. W. - Venezia

A B A'
31 42 43
As this summarysuggests,the revolutionary nature of the piece depends

not upon its large-scaleplan but on the way its sectionsexpand. None of
the threeprolongsa traditionallystable harmonyin a conventionalway.
Section A prolongsT througha chromaticrisingoctave; B moves away
fromthe tonic in risingminorthirdsand neverbringsthe tonic back; and
A' vacillatesbetweenI and T throughout.
Some of the coherence that would resultfroma more stable tonality
shiftsto the motivicstructure.As in Nuages Gris,the semitonebecomes
the fundamentalmotivicidea, primarily the motionA-Bb.It firstappears in
bs 4-5, 6-7 and 8-9, returnsin bs 30-1 and is thenreversedin the closing
moments,bs 44-7 (Ex. 18). As was evident in Ex. 17, the large-scale
designof thepiece projectsthe same motivicgesture.
Ex. 18 ReductionofR. W - Venezia
1 5 9 15 19 26 30 31 35 39 42 43 48
A L A!
""B
.-AV-
B: Y
4
6:VT s
16
"
16
1 6116
16 III6 IbV6 VII& Y
The A sectionbeginson T and resolvesto a (misspelled)B%minorchordin

b.5.'8 Aftera sequential repetition,fasterharmonicrhythmand rhythmic
fragmentation, the sectionends on the same augmentedtriad,again withA
as the melodic note. The pedal C# (the altered note of the dominant)
68 MUSIC ANALYSIS13:1, 1994


finallyresolvesto D in b.31, the beginningofthe B section.This resolution

parallelsthe one in bs 4-5, exceptforthe mode shiftfromminorto major.
The B sectionmoves throughthe risingthirdsBV-DV-E?, and then turns
to - and ends on - A. The turnto A accomplishestwo goals: the framing
harmoniesof the B sectionproduce a large-scalemotivicstatementB%-A,
and the A harmonypreparesthe augmentedtriadthatbeginsand ends the
piece. All thatis needed is the motivicsemitoneE-F.
A' returnsto the openingaugmentedtriadbut now withthe additionof
a decorating Bb. The A-B%motive has become B%-A,and the overall
arpeggionow only descends. This arpeggiois clearlyandrogynous:it is a
combinationof the tonic minorand the altereddominant.We cannot say
whetherthe finalharmonicimpressionis that of tonic or dominant.The
finalnote is C#(=Db), whichis commonto bothharmonies.
To reinforcethe view thatLiszt consideredB%as the tonal centre,it is
usefulto make a comparisonbetweenR. W. - Veneziaand the OrganMass
'Kyrie' discussed earlier.The Mass as a whole centreson B%(compare the
beginningof the 'Kyrie' withthe end of the 'Agnus Dei'). Similarly,the
'Kyrie' itselfbegins and ends in Bb. However, apart fromthe framing
tonics,the movementdoes not confirmthe keyin a traditionalway. It in
factfollowsthe identicalharmonicpatternfoundin the B sectionofR. W. -
Venezia. The main structuraldifferencebetween the two compositions
lies in the two A sectionsthatsurroundthe centralpartof the piano piece.
Ex. 19 summarisesthe 'Kyrie';comparethiswiththe B sectionofEx. 18.
Ex. 19 Reductionof the 'Kyrie'fromthe OrganMass
1 12 24 35 37 38
6 vII16 V6 VIIp I
1K
B.: -
CONCLUSION
The vagueness pervading these late pieces - and the vagueness of

androgynousharmonyin general- is a directexpressionof Romanticide-
ology,an ideologywhichvalues openness and progressabove closure and
stability(the stateof BecomingratherthanBeing,as Leonard Meyermight
put it'9).The tentativeopeningssuggesta process alreadybegun,whilethe
open endingsrepresentcessationratherthan closure.20Both, of course,are
characteristicof the nineteenthcenturyand have numerous precedents.


ZDENEK SKOUIMAL
The former is usually traced to the opening of Beethoven's Ninth

Symphony;Liszt himselfanticipates the latter in tone poems such as
Orpheus(1854) and Hamlet (1858), wherethe endingsreflectthe implied
programmes.Meyer suggestsa reason forour naturalacceptance of these
endings:
There is something deep withinour primordial
affective
beingsthat
leads us to identify
withsuch abatementprocesses..., sensingin
themthenaturalcyclesof existence- thetimesof day,theseasons,
and thecourseoflifeitself.2'
In that sense, Liszt's last tone poem, Von Wiegebis zum Grabe (1882),
stands as an almost prototypicalexample. La lugubregondola I and
R. W. - Veneziapracticallydemand such closure.The endingsof the earlier
tone poems are indefiniteonly in theireffect,though,throughsecondary
parameters;tonallytheyare quite clear. It is howevera naturalextensionto
accentuatethe finalvagueness througha vagueness in the harmonyitself,
once theharmoniclanguageallows it.
Richard Taruskin writes that 'Wagnerian harmony is essentially
dominant harmony.... Lisztian harmony,on the other hand, with its
circlesof thirdsis harmonythat seems, at times,to deny the existenceof
the dominant'.22This statementcould be clarifiedby specifying thatLiszt's
harmony is still about dominants, but at times they are hardly
distinguishablefromthe tonics and hence are ofteninterpretedas tonics.
Careful analysis shows that in theirproper interpretation as dominants
relativelyfamiliarharmonic structuresemerge. This is, of course, not
meant to deny Liszt's great advances in the harmoniclanguage; theyare
obvious, even on casual acquaintance withthe music. These, however,did
not appear spontaneously.They developed as transformations of tradi-
tionalstructuralelements.
In a thought-provoking discussionof Liszt's experimentalidiom, Allen
Forte indicatesset 4-19 (0148) as one of the 'fundamentalcomponentsof
much of thatmusic'.23 The factcan hardlybe disputed;the set is foundin
numerous and diversecontexts.However, it is significant that 4-19 has a
rathersimpletonal meaning:it is the combinationof a minortonic and its
augmenteddominanttriad.The androgynousharmonythatLiszt explored
in his last yearscreateda constructthatbecame entirelyat home withinthe
radicallytransformed musicallanguageofthe earlytwentiethcentury.
NOTES
1. Some recentdiscussions
oftheseissuesincludeBryanR. Simms,Musicofthe
TwentiethCentury:Style and Structure(New York: SchirmerBooks, 1986),
Chapter 1; Joel Lester,AnalyticApproachesto Twentieth-Century
Music (New


York: Norton, 1989), Chapter 1; and Leonard B. Meyer, Styleand Music:

Theory,History,and Ideology(Philadelphia: Universityof PennsylvaniaPress,
1989), especiallyChapter8.
2. Apart fromthe generalidea of 'characteristics of both', it seems unnecessary
to suggestany furtherparallels between male/femaleand tonic/dominant in
this context.The harmonycould also be termed'synthetic',but this creates
otherproblems: apart fromthe more recentassociationwith 'chemical' and
'plastic', it has a specificmeaning in nineteenth-century theorybased on
Hegelian dialectics and also appears in nineteenth- and early twentieth-
centuryRussian theoreticalwritings.The term'dual' seems bland, while the
coined terms'domic' and 'toninant'appear absurdin a scholarlycontext.
3. Recent articlesthat relate directlyto the present study include: Lawrence
Kramer, 'The Mirror of Tonality: Transitional Features of Nineteenth-
CenturyHarmony', 19th-Century Music, Vol. 4, No. 3 (1981), pp.191-208
(specifically pp.203-6); Howard Cinnamon, 'Tonic Arpeggiation and
Successive Equal Third Relations as Elements of Tonal Evolution in the
Music of Franz Liszt', Music TheorySpectrum, Vol. 8 (1986), pp.1-24; James
Baker, 'The Limits of Tonalityin the Late Music of Franz Liszt', Journalof
Music Theory,Vol. 34, No. 2 (1986), pp.145-73; Allen Forte, 'Liszt's
ExperimentalIdiom and Music of the Early Twentieth Century', 19th-
CenturyMusic, Vol. 10, No. 3 (1987), pp.209-28. Forte's article includes
numerousotherusefulbibliographicreferences.
4. Bernard C. Lemoine, 'Tonal Organizationin Selected Late Piano Works of
Franz Liszt', in Liszt-Studien2 (Munich: Musikverlag Emil Katzbichler,
1978), p.130.
5. This idea formsan importantpart of Hugo Riemann's harmonictheory.He
derivesthe secondarychords of a key throughsubstitutionwithina primary
chord: either6 for3 (Parallelklang)or I (leading-note)for i (Leittonwechsel-
klang).In C major an E minorchordis eitherthe dominantParallelklang(Dp)
or the tonicLeittonwechselklang (T), dependingon context.WilliamMickelsen
summarisesRiemann's positionon chord III as follows:'The III chord is an
unhappychord which,havingtones of both the tonic and dominant,can be
eitherthetic [tonic] or synthetic[dominant]according to use'. William C.
Mickelsen, Hugo Riemann's Theoryof Harmony (Lincoln: Universityof
Nebraska Press, 1977), p.29. See also pp.64-6 and pp.221-2.
6. As Aldwell and Schachterdescribeit in Harmonyand VoiceLeading,2nd edn
(New York: HarcourtBrace Jovanovich,1989), p.280.
7. The 'Gloria' also has a similarending,thoughless obvious. Aftermanybars
that implyF#major and B major, the finalD tonic arrivesvia an
F#minor
harmony(althoughonlytwo notes appear on the downbeatof b.77, the C#is
clearly implied). The immediate repetition of the passage omits b.76,
reinterpreting the V as 16.Here the finaltonic is prolongedwith a I-V-IV- 16
progression.
8. The IIF chord in the penultimatebar is actuallya combinationof V and V,
both A major and F minorappear simultaneously.It resultsfroma V and an

C Basil BlackwellLtd. 1994

ZDENEK SKOUMAL
FOpedal. In effectit is similarto theV and hence I label it the same way.
9. See Robert P. Morgan, 'Dissonant Prolongation:Theoretical and Compo-
sitionalPrecedents',JournalofMusic Theory, Vol. 20, No. 1 (1976), pp.49-91.
10. Liszt anticipatesthis eventwithinthe introductionitselfby movingto the F
several times. These Fs, however, are usually harmonised with another
augmentedtriadand hence createtensionratherthan servingas resolutionsof
the precedingdissonance.In bs 16 and 18 the F is suspended intothe second
beat, where it createsa brief'apparent' consonance beforeresolvingback to
the E and the main augmented triad. Those two points encapsulate the
relationshipof the introductionto the openingof theAllegrobut in reverse.
11. Forte, 'Liszt's ExperimentalIdiom', pp.225-7.
12. Alan Walker, 'The Music of Catastrophe:Liszt's "Wagner Elegies"', Piano
and Keyboard,No. 142 (May/June1993), pp.19-27.
13. Baker,'The LimitsofTonality',p. 156.
14. This readinganswersthe question thatBaker poses afterthe above quotation:
'Is Eb in b.9 a main element,or is it a lowerauxiliaryto Fb?' (p. 156).
15. By way of comparison,La lugubregondolaII uses the same harmonyand
opening melodic gesturebut is unambiguouslyin F minor. However, the
melodyis significantly altered,employingG? and focusingon F as the tonic.
16. The augmentedtriadis arpeggiatedin sixthsand hence the second and third
notes of the lefthand anticipatethe opening of the right-handmelody. This
intervalthenbecomes the tremolandobass in bs 77-85.
17. Baker,'The LimitsofTonality',p.171.
18. Accordingto its notationthe chordis an 'apparent' consonance,includingC#
and Bb. However, the sequential repetitionin b.15 spells the minor chord
correctly.Linear considerationsand avoidance of enharmonicnotation(CO to
Db) outweighlocal verticallogic. The chordsare part of a 0-6 sequence; the
relativeimportanceof the alternating harmoniesis locallyquestionable.
19. Meyer,Styleand Music,see especiallypp. 199-200.
20. Terms used by Meyerto describethe Tristanending,Styleand Music,p.324.
21. Ibid.,pp.323-4.
22. Richard Taruskin, 'Chernomor to Kashchei: Harmonic Sorcery; or,
Stravinsky's"Angle"', Journalof theAmericanMusicologicalSociety,Vol. 38,
No. 1 (1985), p.125.
23. Forte, 'Liszt's ExperimentalIdiom', p.214.



Liszt S Androgynous Harmony

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Liszt S Androgynous Harmony

Caricato da

Copyright:

Formati disponibili

Liszt's Androgynous Harmony

Author(s): Zdenek Skoumal

This content downloaded from 147.91.236.58 on Thu, 16 May 2013 04:07:57 AM