===       ===     =============        ====
             ===       ===           ==            ==   ==
            == ==     ===           ==           ==      =
           ==   ==== ===           ==           ==      ==
          ==     ==  ==           ==            =      ==
         ==         ==           ==             ==   == 
        ==         ==           ==               ====
       M U S I C          T H E O R Y         O N L I N E
                     A Publication of the
                   Society for Music Theory
          Copyright (c) 1994 Society for Music Theory
| Volume 0, Number 7      March, 1994      ISSN:  1067-3040   |
  All queries to: mto-editor@husc.harvard.edu
AUTHOR:  Alphonce, Bo, H.
TITLE:  Dissonance and Schumann's Reckless Counterpoint
KEYWORDS:  Schumann, piano music, counterpoint, dissonance, rhythmic shift
Bo H. Alphonce
McGill University
Faculty of Music, Department of Theory
555 Sherbrooke Street West
Montreal, Quebec H3A1E3
ABSTRACT:  Work in progress about linearity in early romantic
music. The essay discusses non-traditional dissonance
treatment in some contrapuntal passages from Schumann's
*Kreisleriana*, opus 16, and his *Grande Sonate* F minor,
opus 14, in particular some that involve a wedge-shaped
linear motion or a rhythmic shift of one line relative to  
the harmonic progression.
[1] The present paper is the first result of a planned  
project on linearity and other features of person- and  
period-style in early romantic music(1).  It is limited  
to Schumann's piano music from the eighteen-thirties and  
refers to score excerpts drawn exclusively from opus 14  
and 16, the *Grande Sonate* in F minor and the  
*Kreisleriana*  -- the Finale of the former and the first  
two pieces of the latter.  It deals with dissonance in  
foreground terms only and without reference to expressive  
connotations. Also, Eusebius, Florestan, E.T.A. Hoffmann,  
and Herr Kapellmeister Kreisler are kept gently off  
1. While Schumann's dissonance treatment is not a new  
topic, the response "no matches were found"  from  
smt-search to my request for "Schumann and dissonance"  
served as an encouragement for this analytical caprice.
[2] Schumann favours friction dissonances, especially the  
minor ninth and the major seventh, and he likes them raw:  
with little preparation and scant resolution.  The  
sforzato clash of C sharp and D in mm. 131 and 261 of the  
Finale of the G minor Sonata, opus 22, offers a brilliant  
example, a peculiarly compressed dominant arrival just  
before the return of the main theme in G minor.  The  
minor ninth often occurs exposed at the beginning of a  
phrase as in the second piece of the *Davidsbuendler*,  
opus 6: the opening chord is a V 4/2 with an appoggiatura  
6; as 6 goes to 5, the minor ninth enters together with  
the fundamental in, respectively, high and low peak  
registers.  Or it occurs at the accented culmination of a  
fast melodic wave as in mm. 136 and 138 of "In der  
Nacht," the fifth piece from *Fantasiestuecke*, opus 12  
(an almost identical situation is found in m. 11 of  
Variation VIII from the *Symphonic Etudes*, opus 13).   
There is also the pivot function, as in Number 15 of the  
*Davidsbuendler*, where in mm. 43ff the C minor leading  
tone B is held over from the dominant seventh chord,  
itself equipped with minor ninth, and is intensified as  
the minor ninth Cb of the dominant chord of Eb major.  Or  
the use as opening melodic interval, as in the Scherzo  
variant of the Clara Wieck theme in opus 14 or the  
show-off imitative counterpoint in mm. 2ff of the first  
*Intermezzo*, opus 4.
[3] To a great degree Schumann's dissonance treatment is  
a reflection of contemporary practices.  External  
dissonance -- linear components entering simultaneously  
with and conflicting with the chord -- occurs as metric  
suspension or appoggiatura, submetric appoggiatura, or  
simply as accented passing note, very often as accented  
chromatic lower neighbour.  Internal dissonance favours  
the accented passing and neighbour notes, especially the  
chromatic lower neighbour.  In these situations, however,  
Schumann characteristically does not work around harsh  
dissonances but tends to seek them out; typically,  
resolution notes clash with non-chordal notes.  In most  
cases tempo and textures keep the degree and intensity of  
dissonance within the bounds of refined spicing that is a  
distinctive feature of Schumann's piano music.  But  
sometimes the intensity is heightened. This paper looks  
at three contrapuntal situations where this happens: the  
linear wedge and two kinds of rhythmic shift.
[4] The right hand part of Example 1 (*Kreisleriana*,  
opus 16, first movement, mm. 1-4) analyses into two  
chords per measure, each chord built by two triplets.   
The pattern is enriched by the half step motif that  
begins the piece and connects each chord with the next.   
The function of the motif as chordal or non-chordal is  
ambiguous and subject to interpretation -- no particular  
reading is consistent throughout the phrase. Now imagine  
the left-hand part pulled back an eighth note so that  
each of its notes supports  the right-hand chord at the  
strong beat.  Apart from the ninth on the first downbeat  
and the suspension chord at the cadence, linear  
dissonance is not a propelling force in this  
configuration. But when the left hand trails by an  
eighth, as it does in the actual music, the successive  
dissonances on strong beats: the ninth in m. 1, the  
seventh and diminished octave in m. 2, push forward more  
strongly than the harmonic progression alone (note that  
as harmonic dissonance takes over at the end of m.2,  
dissonances on strong beats become less pointed).  Thus,  
the rhythmic shift changes an essentially homophonic  
passage into a kind of two-part counterpoint, not  the  
classical kind but a counterpoint where one of the two  
parts is a composite of chordal arpeggios and the  
left-hand "suspensions" resolve by joining the new chord  
without concern for interval or direction.  
[5] In the consequent phrase the left-hand part too  
becomes chordal and the conflicts become more complex.   
It could be argued that the performer can reduce  
complexity by articulating the left-hand part away from  
the strong beats.  But whatever the performer does, the  
listener is likely to hear what auditory stream theory  
predicts: two streams where dissonance conflict helps  
rather than hinders segregation. As in any tonal  
contrapuntal situation, the dissonances support the  
independence of the lines and the quasi-imitation brought  
about by the left-hand syncopation binds the two lines  
together enough to allow the common harmonic progression  
to remain intact. In other words, the "reckless  
counterpoint" caused by the rhythmic shift strengthens  
stream segregation and thus is a powerful way to  
reinforce the linear dimension of the harmonic  
[6] The trailing left hand technique occurs in others of  
Schumann's piano works from the thirties, for instance in  
mm. 5ff of opus 10:2, but apart from experimentation with  
rhythmic shift and dissonance clashes in *Carnaval*, opus  
9,(2) the discovery of its potential for forceful  
dissonance treatment appears to have come with the  
*Kreisleriana*. Striking ways of creating conflict  
between elements of different chords, however, are at the  
center of attention in the Finale of opus 14 which  
antedates *Kreisleriana* by about two years; in the  
discussion of Examples 6 and 7 we shall see chord and  
dissonance conflict at the full measure level.
2. See #18, the piece called "Paganini."
[7] Example 2 (*Kreisleriana*, first movement, mm. 24-26)  
is included to show in the first place the remarkable  
contrast between the first section of the movement and  
the second with its downward arpeggios and pedal tones  
and its absence of chordal conflict even while the motion  
continues to be carried by the same triplet rhythm.   
Refined dissonant spice here results from the chords of  
the harmonic progression shining through a veil of pedal  
tones, Bb and F.  But the passage also hints at a linear  
relationship that is exceedingly common in Schumann's  
piano music. While the upper voice (the slurred sixteenth  
pairs) makes oblique motion against the pedal, a couple  
of measures after those shown in the example a descending  
bass motion develops below the pedal Bb, the pedal  
gradually dissolves, and the upper voice maintains its  
register in short contrary motions against the bass until  
just before the cadence. The result is a wedge-like  
motion where one voice functions as a simple or  
ornamented pedal.
[8] Even more pervasive than the oblique wedge is the  
contrary motion wedge.  Every section of the second  
movement of *Kreisleriana* offers examples of some kind  
of wedge, in oblique or contrary motion or a combination  
of both; one of the most striking wedges begins in m. 17  
where three voices engage in contrary motion extending  
from the interval of a sixth to the distance of a full  
three octaves, all of this happening over a pedal.  In  
stepwise diatonic contrary motion there is of course  
always a point where a seventh and a ninth  occur in  
succession (separated by a major or minor second -- and  
thus belonging to interval class 2 or 1 with their  
difference in dissonance quality -- depending on whether  
or not the octave/unison of the pair occurs on one of the  
notes of the tritone in a diatonic scale).  The strength  
of the respective lines takes precedence over any  
resolution rules that may be embedded among the style  
criteria so that the notes are heard as passing notes.  
The effect of the successive dissonances then is a sense  
of momentary friction that adds to the vigour of the  
passage.  With three voices engaged, Schumann takes  
advantage of this effect in characteristic ways -- and  
since this example was an afterthought and has not been  
included as a GIF, I must ask the scoreless reader to  
imagine the texture.  In m. 17 the upper and the lower of  
the three moving voices make an exchange in stepwise  
eighth-note motion, the upper voice running from Bb4 to  
G5, the lower from G3 to Bb2 within the harmonic  
framework of a dominant seventh chord built on the pedal  
tone C.  The middle voice runs essentially in parallel  
tenths with the lower voice but changes register twice:  
between the first two notes, Bb3 to A4, and the last two  
notes, E4 to D5.  At the moment the third and fourth step  
of the outer scale segments have gone from minor seventh  
to major ninth, the middle and upper voices reach their  
seventh and go on to the ninth -- in this case the major  
seventh and the minor ninth.  The leap in the inner voice  
after the ninth weakens the line and makes the dissonance  
stand out.  Despite the brevity of the impression, this  
clustering of passing dissonances on the way to the  
culmination of the wedge (added to the vertical  
coincidence for a short moment of the pedal C3 with D3,  
E5, and F4) creates an exquisitely calculated contrast to  
the consonance and mild dissonance of the section as a  
[9] A chromatic wedge either has two unison/octave points  
a tritone apart, surrounded by intervals of interval  
class 2, or has no unison/octave points but instead has  
successive intervals of interval class 1 at points a  
tritone apart.  The first kind typically occurs in the  
omnibus progression -- and there are numerous examples of  
omnibus fragments in Schumann's music (the section  
beginning in m. 119 of the second movement of  
Kreisleriana uses the same kind of chromatic contraction  
and expansion as one finds in the omnibus progression).   
In fact, Schumann avoids the second kind and often also  
manipulates the diatonic wedge by some chromatic motion  
so as to avoid the seventh-ninth succession.  Example 4  
(Kreisleriana, first movement, mm. 15-17) shows a wedge  
with mixed diatonic and chromatic motion in both outer  
voices where the outer voices avoid interval class 2  
dissonance; this, however, finds its way into the  
progression by a different technique -- see below.   
Clearly, if the upper voice in m. 16 had continued  
chromatically down over the chromatic bass, the minor  
tenth would have been followed by a minor ninth and a  
major seventh, but then one purpose of the wedge would  
have been lost:  to retrieve D minor after a brief  
excursion into the Neapolitan region.   The bass is  
chromatic for harmonic purposes: it includes the leading  
tones of IV and V in D minor;  the right hand completes  
the harmonies. The effect is a clarification of harmonic  
direction as if coming from "somewhere out there" and  
gradually focusing on the goal.  The use of this  
technique in chromatic harmony would be worth its own  
investigation; Max Reger, for instance, uses it  
[10] Even though the wedge construction in Example 4  
itself bypasses sharp dissonances, they are there anyway.   
In contributing to the harmonic progression, the  
right-hand part is at odds with the left-hand part in a  
way that reveals another type of rhythmic shift; for want  
of a better term let me call it "chordal anticipation."   
This technique places a chord, or the better portion of a  
chord, rhythmically before the beat to which it belongs  
harmonically.  A simple version consists of anticipating  
the entire chord immediately before playing it on the  
beat, as at the opening of the first piece of the  
*Davidsbuendler*.  Rhythmically more elaborate versions  
tie the right-hand anticipation to the beat and  mark the  
beat by the left hand as in the C major *Fantasy*, opus  
17, second movement, mm. 92ff, or tie both hands into  
chordal syncopation as in mm. 84ff of the last movement  
of the same piece.  Variants of chordal anticipation are  
numerous in Schumann's work, but the most intriguing  
version from the point of view of dissonance treatment is  
the one shown in Example 4: one linear strand arpeggiates  
before the beat and completes the chord on the beat while  
the other line joins the same chord on the beat.  As a  
consequence, chordal elements are again out of sync so  
that the arpeggio towards the next beat creates a chordal  
conflict with the chord established on the current  beat.   
Imagine the right-hand part pushed forward by two  
sixteenths so that each arpeggio begins on the beat and  
fills the duration of the beat; then the progression is  
just a normal harmonic wedge approaching the tonic.   
Instead, the chordal anticipation technique fills each  
beat -- traditionally the space for the chord supported  
by the bass line -- with chords that more or less  
contradict both the progression and the expected  
voice-leading.  Obviously,  careful articulation and  
weighting of the arpeggios in performance will guide the  
listener's understanding of the harmonic progression, but  
even so, the diminished octave E-Eb in m. 15 and the  
minor second G sharp-A in m. 16 will lend a distinctive   
character of linear dissonance to the whole passage.
[11] What goes before the passage shown in Example 4 uses  
the same triplet rhythm but does not involve chordal  
anticipation; what follows immediately after it is a  
recapitulation of the opening material shown in Example  
1.  Thus, with its chordal anticipation shift the wedge  
passage mediates between a section without rhythmic shift  
and the recapitulation of the trailing left hand shift.   
The two types of shift are closely related and tend to  
have similar effects on dissonance treatment.  As we  
shall see, things get more intricate when more than one  
level of harmonic progression is involved. But first let  
us have a look at a wedge with a different formal  
[12] Using a linear wedge to close a section or prepare a  
recapitulation is not unusual in Schumann's oeuvre.  As  
an early example, consider the second *Intermezzo*, opus  
4, where the second repeat section prepares both its own  
return and the return of the initial material in a new  
key by means of an omnibus-type chromatic wedge  
prolonging a dominant seventh chord.  In the first  
movement of the Piano Concerto practically every return  
of the main theme is preceded by some kind of wedge;  
likewise, in the Finale of the F minor Sonata both  
returns of the main theme are prepared by a  
contrary-motion wedge.  But wedges are used in a number  
of different functions, on a smaller scale for voice  
exchange within a prolonged harmony anywhere in a formal  
section, on a larger scale for theme construction, as for  
instance in the Finale of the F sharp minor Sonata, opus  
11, where the wedge is one of the main voice-leading  
techniques throughout.
[13] From the dissonance point of view an especially  
intriguing wedge is shown in Example 5 (*Kreisleriana*,  
second movement, mm. 100-106 -- beginning of the second  
section).  Rather than closing a section, it comprises  
the contrasting middle section of a small ternary form  
and achieves the unlikely combination of decreasing space  
and increasing energy.  Based on Litzmann's book about  
Clara Schumann(3), Peter Ostwald quotes Robert Schumann's  
word to Clara that "there is a thoroughly wild love in  
some of the movements" of the *Kreisleriana*(4).  Without  
speculation on expressive connotations (I promised!), in  
terms of counterpoint this certainly is one of the wilder  
moments.  Its basic material is given in Example 3 (mm.  
92f -- the beginning of the first section of the same  
movement).  This has some of the character of the  
trailing left hand but not unambiguously; the harmonies  
work either way, and the bass appoggiatura in m. 93 of  
course represents a dissonance treatment Schumann shares  
with most other romantic composers.  The left hand  
imitates the upper line at the lower fifth in a stepwise  
descending sequence up to m. 96 but skips every other  
dotted rhythm.  The only real argument in favour of  
understanding this passage as an example of trailing left  
hand shift comes on the downbeat of m. 96 where the bass  
line has reached a C over which the right hand plays a D  
major arpeggio.  Even though the harmonic context  
suggests the function of the harmony as V 4/2, the bass C  
is simply abandoned for a D and the phrase ends on an  
emphatically tonicized D major harmony.  Abandoning an  
exposed harmonic dissonance is unproblematic if the  
left-hand part is understood as trailing the harmonic  
progression by a beat, but this reading is ambiguous at  
3. Litzmann, Berthold. 1925. *Clara Schumann, Ein  
Kuenstlerleben*. 7th ed. 3 vols. Leipzig: Breitkopf &  
4. Ostwald Peter. 1985. *Schumann, Music and Madness*.  
London: Victor Gollancz Ltd.  The quotation is from vol.  
1, p. 224 of Litzmann's book.
[14] In Example 5 questions of rhythmic shift become  
entirely irrelevant since here linear progression  
supersedes harmonic progression during the ascent until D  
minor is retrieved at the climax in m. 106.  The linear  
framework exhibits rhythmic imitation and  
quasi-inversional melodic imitation.  But the upper line  
moves up by step from measure to measure so, remarkably,  
the wedge is shaped by two ascending lines: the upper  
line moving mostly by whole steps while the lower line  
catches up by minor thirds (filled in by chromatic  
motion) and the interval between the two lines becomes  
gradually narrower.  From one accented second beat to the  
next the interval shrinks from diminished twelfth over  
diminished eleventh and minor tenth to major ninth; at  
this point the approach is halted and an exchange of  
major ninth and minor seventh moves up by whole step to  
the climax.  This indeed is dissonance with little  
preparation and scant resolution and would be passed off  
in any traditional text as reckless counterpoint.   
Dissonance here seems to function at a new and different  
level: it is not a moment-to-moment event but a gradually  
increasing intensity over the length of the phrase  
followed by a gradual decrease (mm. 107 and 108 echo the  
minor ninth appoggiatura from m. 93).  Resolution at this  
level occurs only in m. 110 where the two lines move in  
parallel octaves down to a recapitulation of the first  
[15] Now, what happens when chordal anticipation in  
oblique motion is imitated, so that  two compound lines  
create a counterpoint that moves in sequence over a  
pedal?  One of my earliest acquaintances with Schumann  
was his *Grande Sonate* in F minor, opus 14,  which I  
stumbled through as a teenager and whose last movement I  
found terribly boring and in part ugly.  A little bit of  
the "ugly" stuff is shown in Example 6 (*Grande Sonate* F  
minor, opus 14, last movement, mm. 46-50).  The material  
on which this passage is based comes from mm. 9-10 where  
it follows immediately after the main theme. It is  
identical to the arpeggio figures in mm. 46-47, except  
that the last two arpeggios in the second measure are C3  
G3 E4 and F3 Ab3 with F4 on the downbeat of the third  
measure ending the phrase.  In performance it should be  
articulated as chordal anticipation so as to avoid  
harmonic confusion at the major seventh leap E4 F3.  Yet,  
the dissonant leap will make its impression and will  
foreshadow the character of those later sections where  
this material is used.  Taken separately, the two entries  
in mm. 46 and 48 seem to move in Eb major and C minor,  
respectively, each ending on a half-measure suspension,  
although it is only when the third entry comes in on Ab  
that the second ending is made dissonant.  The effect of  
the second entry is to change the Eb harmony into first  
inversion C minor -- unless one is capable of hearing the  
two strands separately as a bitonal or bichordal  
progression.  The voice-leading B to C in the first two  
right-hand arpeggios in m. 48 against the left-hand  
suspension F to Eb promises a resolution to C minor so  
strongly that I hear the G-major chord as a large  
suspension chord rather than as an independent harmony.   
In either case there is a chordal conflict due to the  
late arrival of the C minor triad in the right-hand part.   
It seems then that two rhythmic shifts are at work  
simultaneously, but at different levels: chordal  
anticipation at the beat level and the displacement of  
the expected harmony in the right-hand part at the  
measure level. The effect of the third entry is to alter  
the expected F minor harmony so that it functions as part  
of a Bb major ninth chord, supported by the large-scale  
pedal note, Bb2.
[16] When the two compound parts -- the left hand and the  
right hand -- sound together, the amount of dissonance is  
considerable, but dissonance treatment here stays well  
within traditional practice.  The double-entry  
progression is sequenced a third up as indicated by the  
entry on Ab in m. 50 and again another third up in m. 54  
while the pedal Bb is refreshed regularly. By starting on  
successively higher scale steps the six entries together  
unfold a major sixth from F to D, i.e. the fifth and  
third of the overall prolonged Bb harmony.  When the  
upper line reaches D6 over the pedal note, the third  
sequence is curtailed and the line descends, supported by  
parallel sixth chords, through the dominant seventh note  
Ab to resolve by suspension Ab to G over Eb at the point  
where Example 7 begins.
[17] In Example 7 (*Grande Sonate* F minor, last  
movement, mm. 60-63) the chordal conflict extends over  
not just one but two measures.  While the triplet motion  
now is fragmented and repeated, creating shorter units,  
the two measures are held together by the eighth-note  
arpeggio figures that spell out a C minor chord with  
downbeat appoggiatura dissonance (craftily supported by  
the "wrong" harmony in the chordal anticipation part).   
Again the expected Eb harmony is transformed into first  
inversion C minor; imitation now takes place in  
invertible counterpoint a fifth down, spelling out a root  
position F minor chord; thus, the underlying structure is  
a 6 5 linear intervallic pattern -- not the figured-bass  
motion where the upper line descends but the "inverted"  
kind where the bass ascends.  This is again sequenced  
upward and is curtailed when it has reached the Eb major  
sixth chord.  
[18] While in these two passages the counterpoint is  
perfectly passable as far as dissonance treatment is  
concerned, it is the harmonic conflicts that make them  
sound like reckless counterpoint.  At this level, then,  
the sense of heightened dissonance intensity is not so  
much a question of dissonant intervals as of conflicting  
chords.  By techniques that are bold and advanced for his  
time, Schumann has achieved a sound that is not only  
uncharacteristic of early Romantic music but even  
un-Schumannesque.  While I no longer find these  
progressions ugly but instead exciting and refreshing,  
their particular sound seems more akin to that of some  
works in the 20th-century neoclassicist and "neue  
Sachlichkeit" tradition, the environment where the phrase  
"reckless counterpoint" was originally coined.
[19] This essay has discussed a few style features from  
Schumann's early piano music and, in particular,  
different combinations of features.  The fact that some  
combinations occur in a work that stands out as the  
quintessence of Schumannesque sound, the *Kreisleriana*,  
while others contribute to a sound that strikes the  
listener as peripheral to Schumann's style, brings up a  
central question that the ongoing project must address:  
the question of what features a given composer shares  
with what groups of other composers over what spans of  
time and within what regions of space, and what features  
are exclusive to that composer.  It also serves as a  
reminder that questions of style delimitation are highly  
complex and probably have to do with intricately combined  
and continuously recombined features of both distant  
heritage and spontaneous creation.
Copyright Statement
[1] Music Theory Online (MTO) as a whole is Copyright (c) 1994,
all rights reserved, by the Society for Music Theory, which is
the owner of the journal.  Copyrights for individual items 
published in (MTO) are held by their authors.  Items appearing in 
MTO may be saved and stored in electronic or paper form, and may be 
shared among individuals for purposes of scholarly research or 
discussion, but may *not* be republished in any form, electronic or 
print, without prior, written permission from the author(s), and 
advance notification of the editors of MTO.
[2] Any redistributed form of items published in MTO must
include the following information in a form appropriate to
the medium in which the items are to appear:
	This item appeared in Music Theory Online
	It was authored by [FULL NAME, EMAIL ADDRESS],
	with whose written permission it is reprinted 
[3] Libraries may archive issues of MTO in electronic or paper 
form for public access so long as each issue is stored in its 
entirety, and no access fee is charged.  Exceptions to these 
requirements must be approved in writing by the editors of MTO, 
who will act in accordance with the decisions of the Society for 
Music Theory.