SCHUBEKT

CRITICAL and

ANALYTICAL STUDIES

3d by WALTER FRISC
BOSTON PUBLIC UBRARY
Copley Square
Schubert
 SCHUBERT
Critical and Analytical Studies

Edited by Walter Frisch

UNIVERSITY OF NEBRASKA PRESS

Lincoln and London
Acknowledgments for the use of copyrighted material

appear on pages vii— viii.

Music autography by Helen M. Jenner
Copyright 1986 by the University of Nebraska Press
All rights reserved
Manufactured in the United States of America
® The paper in this book meets the minimum requirements of American
National Standard for Information Sciences — Permanence of Paper for

Printed Library Materials, ANSI Z39. 48-1984.

First paperback printing: 1996

Most
10 987654321
recent printing indicated by the last digit below:

Library of Congress Cataloging in Publication Data

Main entry under title:
Schubert: critical and analytical studies.

Includes index.
Contents: Sonata form in Schubert: the first movement
of the G-major string quartet, op. 161 (D.887) / Carl
Dahlhaus; translated by Thilo Reinhard — Schubert's
promissory note: an exercise in musical hermeneutics /

Edward T. Cone — Dance music as high art: Schubert's

twelve Landler, op. 171 (D.790) / David Brodbeck
[etc.]

1. Schubert, Franz, 1797- 1828 — Criticism and

interpretation — Addresses, essays, lectures. I. Frisch,

Walter.
ML410.S3S2985 1986 78o'.92'4 85-8445
ISBN O-8032-I97I-7 (cl.) ISBN O-8032-6892-O (pa.)
 Contents

Acknowledgments vii

Introduction ix

Carl Dahlhaus, translated by Thilo Reinhard

Sonata Form in Schubert: The First Movement of the G-Major
String Quartet, op. 161 (D. 887) 1

Edward T. Cone
Schubert's Promissory Note: An Exercise in Musical
Hermeneutics 13

David Brodbeck
Dance Music as High Art: Schubert's Twelve Landler,
op. 171 (D. 790) 31

Joseph Kerman
A Romantic Detail in Schubert's Schwanengesang 48

William Kinderman
Schubert's Tragic Perspective 65

Thrasybulos Georgiades, translated by Marie Louise Gollner

Lyric as Musical Structure: Schubert's Wandrers Nachtlied
("Uber alien Gipfeln," D. 768) 84

Arnold Feil, translated by Walter Frisch

Two Analyses
1. Im Dorfe, from Winterreise 104
2. Moment Musical in F Minor, op. 94, no. 3 (D. 780) 116
VI SCHUBERT
David Lewin
Aufdem Flusse: Image and Background in a Schubert Song 126

Anthony Newcomb
Structure and Expression in a Schubert Song: Noch einmal
Aufdem Flusse zu horen 153

Walter Frisch
Schubert's Nahe des Geliebten (D. 162): Transformation
of the Volkston 175

Lawrence Kramer
The Schubert Lied: Romantic Form and Romantic
Consciousness 200

Appendix of Longer Musical Examples

A. Schubert, Moment musical in A-flat, op. 94, no. 6 (d. 780) 238
B. Schubert, Wandrers Nachtlied ("Uber alien Gipfeln," d. 768) 240
C. Schubert, Im Dorfe, from Winterreise (d. 911, no. 17) 241
D. Schubert, Moment musical in F Minor, op. 94, no. 3 (d. 780) 245
E. Schubert, Aufdem Flusse, from Winterreise (d. 911, no. 7) 247

The Contributors 251

Index of Composers and Works 253

 Acknowledgments

This collection would not have come about without the efforts
and talents of many people. I am especially grateful to Thomas
Baker, Ellin Feld, Anthony Newcomb, Ernest Sanders, and
Robert Winter who gave generously of their time to help pol-
ish my translations from the German.
Permission is also gratefully acknowledged
to reprint or
translate the following selections:
Edward T. Cone, "Schubert's Promissory Note: An Exer-
cise in Musical Hermeneutics," © 1982 by The Regents of the
University of California. Reprinted from 19th-century Music 5,
no. 3 (Spring i982):233~4i, by permission of the Regents. Re-
visions by the author for this reprinting.
Carl Dahlhaus, "Die Sonatenform bei Schubert: Der erste
Satz des G-dur-Quartetts D 887," Musica 32 (1978): 125-30.
Arnold Feil, "'Moment musical,' F-moll, op. 94, nr. 3
(D 780)," chapter 6 of Studien zu Schuberts Rhythmik (Munich:
Fink, 1966), pp. 83-87.
Arnold Feil, " Im Dorfe aus der Winterreise" from chapter
3 of Franz Schubert: Die schone Mullerin. Winterreise (Stuttgart:
Philipp Reclamjun., 1975), pp. 34~44-
Thrasybulos Georgiades, "Lyrik als musikalische Struk-
tur (Uber alien Gipfeln)," from chapter 1 of Schubert: Musik
und Lyrik (Gottingen: Vandenhoeck und Ruprecht, 1967),
pp. 17-31.
Joseph Kerman, "A Romantic Detail in Schubert's Schwa-
Vlll SCHUBERT
nengesang," Musical Quarterly 48 (1962)136-49. Revisions by
the author for this reprinting.
David Lewin, '
Aufdem Flusse: Image and Background in
a Schubert Song," © 1982 by The Regents of the University of
California. Reprinted from lgth-Century Music 6, no. 1 (Sum-
mer 1982)147-59, by permission of the Regents. Revisions by
the author for this reprinting.
 Introduction

Among the major composers of the Austro-German tradition,

Schubert has received less than his fair share of enlightening
criticism and analysis, especially in English. Too often he has
fallen prey to the discursive, chatty kind of commentary that
skims the surface of the music, picking up striking details but
failing to delve for any deeper structural features. Perhaps the
only extended studies in English that transcend the limitations
of this style are those by Richard Capell, Donald Tovey, and
(translated from the German) Hans Gal. Denser, more sub-
1

stantial work has emanated from Austria and Germany, be-

ginning at least as long ago as 1928 (the centenary year of
Schubert's death), when there appeared renowned studies by
2
Paul Mies and Felix Salzer. But these and later efforts in Ger-
man have had surprisingly little impact on Anglo-American
scholars. Although the most recent Schubertjahr, 1978, stimu-
lated a considerable amount of publication, the most remark-
able achievement, a full revision of the Deutsch catalogue, was
3
primarily philological, not analytical or critical.
By "analysis" and "criticism" I mean approaches that
concentrate primarily on the technical and/or aesthetic dimen-
sions of the music. Analysis normally focuses on the technical,
internal relationships of an individual work. Criticism broad-
ens the analytical perspective by taking in other forces that im-
pinge on a work — different pieces by the same composer or by
other composers, historical and cultural matters, and source-
related issues. Criticism often involves comparisons and ex-
X SCHUBERT
plicit value judgments. Although they will (and should) be in-
formed by music theory, analysis and criticism will never seek
simply to validate a theoretical approach. The structural and
expressive aspects of the music itself always remain at the heart
of the endeavor. As Edward T. Cone has put it succinctly, "the
critic seeks to impart to others something of his own vivid ex-
perience of the concrete musical values he finds in the works
that interest him." 4
The present volume assembles some of the more stimulat-
ing recent commentary of this kind from both the English and
German traditions. Of the twelve selections, five have been
written especially for this collection by younger American
scholars (Brodbeck, Frisch, Kinderman, Kramer, Newcomb).
Four others appear here in English for the first time (Dahlhaus,
two by Feil, Georgiades). Three additional essays were first
published in American journals and have been revised or up-
dated by their authors for this reprinting (Cone, Kerman,
Lewin). The authors concentrate primarily on individual pieces
(or on a group of related pieces) drawn from the repertory of
songs, piano, and chamber music that many of us regularly
hear, play, and teach.
Perhaps one reason Schubert's works have remained criti-
cally impoverished is that although they form part of the main-
stream of the Viennese Classical/ Romantic tradition, their
"concrete values" are not easily elucidated by the methods de-
veloped for other composers. This point is made in the open-
ing article of the collection, by Carl Dahlhaus. He claims that
Schubert's "lyric-epic" approach to large-scale form demands
different modes of understanding than does Beethoven's "dra-
matic-dialectical" style, even though the two composers share
certain compositional techniques. Dahlhaus demonstrates how
the massive first movement of the G-Major String Quartet
combines a thematic process of Beethovenian logic and continu-
—
ity with a sense of time that is uniquely Schubertian indeed, a
kind of timelessness, or "tendency toward the boundless." To
achieve this synthesis, Schubert draws together techniques nor-
mally associated with two different forms, sonata and variation.
Dahlhaus's essay on the quartet is especially valuable for
being one of the few extended discussions of a single piece by a
scholar known mainly for his historical or philosophical work.
 —

Introduction xi

As in his broader endeavors, Dahlhaus takes a strongly dia-

lectical approach; yet his precise analysis conies closer than
many others to accounting for Schubert's "heavenly lengths."
One special analytical challenge in Schubert studies is pre-
sented by the shorter piano pieces and the Lieder, which devi-
ate even more sharply from the Beethoven/ Classical paradigm.
It is these smaller-scale but equally rich works that occupy most

of the writers in the present volume.

Edward T. Cone and Arnold Feil examine individual Mo-
ments musicaux from very different perspectives. Cone's ap-
proach to no. 6, in A-flat, is pitch-oriented. He demonstrates
how a small, apparently insignificant chromatic detail intro-
duced near the beginning becomes a "promissory note," which
is "redeemed" later when it affects the larger harmonic and
melodic dimensions of the piece. Cone's "congeneric" or inter-
nal analysis is amplified by an "extrageneric" one, in which the
purely musical process is taken as a representation of Schubert's
feelings about his incurable venereal disease. Feil makes little
reference to specific pitches (or ailments) in his account of
no. 3, in F minor. He focuses almost exclusively on rhythm
with rhythm conceived in its broadest sense, as the way a piece
unfolds in time. In Feil's analysis the ostensibly simple phrase
structure of the Moment musical is unveiled as a subtle alterna-
tion and juxtaposition of two different two-bar gestures, or, as
he calls them, "figures of motion."
One potentially significant analytical issue raised by Schu-
bert's shorter piano works (but not treated by Feil or Cone) is
the extent to which they form part of a larger whole. No one
today will seriously accept Schumann's theory that the Im-
promptus of op. 142 (D. 935) are really the movements of a
sonata. But are there in fact musical links between the individ-
ual pieces of a group? In his essay, David Brodbeck argues that,
unlike, many of the random collections of dances, the twelve
Landler of op. 171 (D. 790) cohere by means of numerous har-
monic and melodic relationships. Moreover, the first piece has
certain introductory features and the final one elements of
closure, a fact that suggests Schubert conceived of these dances
as a unified set —
as did Brahms, who saw to their posthumous
publication in 1864.
Two articles explore a tantalizing but underdeveloped
Xll SCHUBERT
area of Schubert criticism — the relationship between the in-
strumental music and the Lieder. Joseph Kerman and William
Kinderman deal not with the kind of self-borrowing found in
the "Trout" Quintet, the "Death and the Maiden" Quartet, or
the "Wanderer" Fantasy, but rather with more a profound trans-
ference of compositional and expressive techniques. Kerman
shows how in the piano introductions to several songs from
1828, Schubert enhances the initial statement of the tonic har-
mony with strongly Romantic and evocative chromatic neigh-
bor chords. Such harmonic devices also assume an important
role in the late instrumental works, he argues: what began as a
literary-poetic inspiration in the Lieder has become a more ab-
stract or (in Kerman's words) "purely musical" phenomenon.
Kinderman suggests that the juxtaposition of minor and major
—
key areas so characteristic of certain songs especially in Win-
—
terreise is a musical representation of a conflict between the

tragic reality of the poem and an illusory world imagined or

remembered by the protagonist. Like Kerman, Kinderman ar-
gues that this technique, developed in the Lieder (he also cites
Erlkonig and Ihr Bild), carries over as "latent symbolism" into
Schubert's mature instrumental works, in particular the Fantasy
in F minor, op. 103 (D. 940).
The five other articles in this volume are concerned exclu-
sively with the Lieder. The analysis by Thrasybulos Georgiades
is taken from his book Schubert: Musik und Lyrik, perhaps the

most important and original study on Schubert to appear since

World War II. English and American readers will probably
never have encountered anything quite like his style of analy-
—
sis at once intensely detailed and rigorous, and at the same
time boldy (even wildly) imaginative. In the introductory chap-
ter excerpted here, Georgiades treats a single brief song, the
Wandrers Nachtlied (D. 768). By a close examination of the syn-
tax of both the poem and the music, he shows that Schubert
does not merely "set" Goethe's lyric or capture its atmosphere,
—
but penetrates beneath its surface its imagery and semantic
—
meaning to the very bedrock of linguistic structure. The
song thus becomes "the musical sounding of language," or
"language as music."
Although too little known in America, Georgiades 's writ-
ings have had some impact on German critics, most notably
 Introduction xiii

his own former pupil Arnold Feil. Feil's discussion likewise

probes well beneath the musical surface of Im Dorfe from Win-
terreise, here to the rhythmic-metrical foundation. The tech-

niques of analysis applied to the purely instrumental F-Minor

Moment musical prove to be highly adaptable to Lieder. Feil
demonstrates how the central image of the Muller poem that —
of the wanderer emotionally bound to the town, and then fi-
nally breaking free —
is transformed by Schubert into "musical

motion" and is actually composed as "musical structure."

Though working outside the German tradition, David
Lewin is the American analyst who comes closest to the tech-
niques of Georgiades and Feil: he too unveils a musical struc-
ture that conveys the text at a very basic level. Through suc-
cessive proportional reductions of the metrical, harmonic, and
melodic framework of Aufdem Flusse, he shows how the deep
structure of the song reflects the two principal, related images
of Miiller's poem, the stream surging beneath its frozen crust
and the protagonist's feelings swelling within his heart.
Anthony Newcomb's essay, written partly in response to
Lewin's, suggests that the expressive and structural crux of the
song isnot the stream/heart image, but the power of memory.
Newcomb traces a musical process in which the protagonist's
gradual recollection of his former happiness is overcome by his
realization of the hopelessness of his present situation. The
song begins with the "frozen motion" of the opening bars,
which gradually thaws and reaches a climax in the animated
rhythm and high vocal tessitura of the final stanza.
The notion of a continuous, developmental musical pro-
cess (perhaps derived from Beethoven, as Dahlhaus argues) is
both attractive and appropriate in analyzing songs that are
through-composed ( Wandrers Nachtlied), or have an A B A' (Im
Dorfe) or modified strophic form (Aufdem Flusse). But what of
simple strophic songs, for which Schubert demonstrated such
an affinity? The question is addressed in my own essay on Ndhe
des Geliebten (D. 162). After tracing the history of other folk-
like settings of this Goethe lyric, I suggest how Schubert trans-
forms that tradition by maintaining the larger strophic design,
but fashioning an individual musical strophe with extraordi-
nary inner tension and development.
Like Dahlhaus's opening essay, the final one, by Lawrence
XIV SCHUBERT
Kramer, elaborates a dialectical viewpoint on Schubert's musi-
cal language. Kramer claims that Schubert songs often mani-
fest "clashing perspectives" between Classical principles of
continuity and more disjunct, unpredictable Romantic tenden-
cies. He analyzes four songs in which the Classical "horizon"
(Husserl's term) of traditional harmonic procedures is obscured
or overwhelmed by a Romantic "presentational" structure,
which creates its own norms and expectations.
It is hoped that the colloquy represented in this collection
of articles will stimulate further detailed study of Schubert's
music. Only from more "close readings" of this kind can we
begin to shape a larger picture of Schubert's musical person-
ality. The best analysis and criticism will always move outward
from the particular to the general, but will never stop striving
to articulate those "concrete values" that form the core of the
music.
Walter Frisch

Notes
1. Richard Capell, Schubert's Songs (London: Benn, 1928); Donald F.

Tovey, "Franz Schubert" and "Tonality in Schubert," both in Essays and Lec-
tures on Music (London: Oxford University Press, 1949); Hans Gal, Franz
Schubert and the Essence of Melody (1974; repr. New York: Crescendo, 1977).
2. Paul Mies, Schubert der Meister des Liedes (Berlin: M. Hesse, 1928);
Felix Salzer, "Die Sonatenform bei Schubert," Studien zur Musikwissenschaji
15 (i928):86-i25.
3. O. E. Deutsch, Franz Schubert. Thematisches Verzeichnis seiner Werke
in chronologischer Folge, rev. Walther Diirr, Arnold Feil, et al. (Kassel: Baren-
reiter, 1978). The major volume in English to emerge (belatedly) from the
Schubert year was Schubert Studies: Problems of Style and Chronology, ed.
E. Badura-Skoda and P. Branscombe (Cambridge: Cambridge University
Press, 1982). There is no mention of analysis or criticism in the title, and too
little evidence of it within the book. A notable exception to the generally

low level of recent critical commentary can be found in the work of James
Webster, who drew on both Tovey and Salzer in his "Schubert's Sonata Form
and Brahms's First Maturity," lgth-Century Music 2 (i978):i8~35; and 3
(i979):52-7i.
4. Edward T. Cone, review of Leonard Meyer, Explaining Music, in
Journal of the American Musicological Society 27 (1974)1335. For a magisterial
account of the distinctions between criticism, analysis, and theory (and a re-
view of the most important literature), see Joseph Kerman, Contemplating
Music: Challenges to Musicology (Cambridge, Mass. Harvard University Press,
:

1985), chapters 3 and 4.

 —

Sonata Form in Schubert:

The First Movement of the G-Major String
Quartet, op. 161 (D. 887)
~&~

CARL DAHLHAUS
Translated by Thilo Reinhard

Despite biographical documentation, the opinion that, as an in-

strumental composer, Schubert stands "in Beethoven's shadow"
is considered folly among the initiated. It is evident that we

should not sacrifice historical fairness to a norm based on aes-

thetics or on compositional techniques — thus, that Schubert's
lyric-epic sonata form ought not to be measured by the stan-
dards of Beethoven's dramatic-dialectic form. Yet it is difficult
to understand the relationship between the theory of sonata
form, which was extracted from Beethoven's oeuvre, and analy-
ses of Schubert's work that also aspire to the realm of theory,
instead of merely describing the musical surface or relying on a
hermeneutic that, by dealing only with the most basic issues,
pays the price of remaining hypothetical and metaphorical.
The first movement of the G-Major String Quartet
(D. 887), composed in 1826 and published posthumously in
1 85 1 as op. 161, may be regarded as epic or novelistic — in the
sense used by Theodor W. Adorno in his book on Mahler
despite the numerous tremolos that characterize almost half
the movement, and whose agitation hardly seems suited to the
idea of epic composure. Both the principal and subsidiary
groups of the exposition comprise a series of variations. Each
variation breaks off more because of the constraint of sonata
—
principles a constraint that Schubert called to mind late
—
enough than because it points beyond itself and pushes ahead
2 SCHUBERT
to the next "station" of the form. The "periods [Zeiten] of the
form," as August Halm calledthem, seem to be dissolved into
a timelessness: the musical moment extends immeasurably.
This occurs most clearly in the subsidiary theme, which in-
vites endless lingering, but also to a lesser degree in the prin-
cipal group.
This is not to deny a functional differentiation among the
variants of the main idea, whose framework is governed by the
chromatic descending fourth G-F#-Fti-E-El>-D, the lament
progression of the Baroque tradition. The four sections of the
principal group can be easily characterized as antecedent (bars
I 5~ 2 3)» consequent (bars 24-32), elaboration (bars 33-53),

and transition (bars 54-63). The melodic gesture of the conse-

quent appears as a reply to the antecedent, of which it forms a
partial inversion. In the elaboration (bars 33-42), the superim-
position of the rhythmic pattern of the "introduction" (bars
1-4) — —
which is in fact thematic, as shall be shown and the
harmonic model of the main idea can be understood as a musi-
cal chain of logic: as an expression of the idea that form is a
process in which later elements grow of necessity from earlier
ones. And in the fourth variant, the replacement of the falling
sequence of the main idea (G— Rt/Fli— E/Et— D) by an ascend-
ing one (G-R/A-Q/B-Atf; bars 54-59) leads to a tonally
remote F-sharp major, from which the D major of the subsidi-
ary group, the traditional key, emerges in sharp relief: conven-
tion achieved by surprise.
The functional differentiation of parts, which obeys the
rule of sonata form, nevertheless alters little in the basic design
of successive variations, which form a cycle insofar as they
draw circles, ever expanding circles, around the theme. The
variation principle as such is not goal-oriented, but rather
resembles a commentary "meandering" about the theme, illu-
minating it from different sides. And in the subsidiary group
which comprises almost a hundred bars (in a "very moderate"
Allegro), thus exceeding by far every normal dimension — the
four appearances of the theme are joined in the form of a circle,
so that the second variation (the third appearance) is diverted to
B-flat major and the third and last returns to D major. Varia-
tions 1 and 2 become longer than the theme because of a devel-
—
opmental extension a sequence based on a five-bar model
 —
Sonata Form in Schubert 3

(bars 90-94) referring back to the beginning of the theme

and an epilogue (100-109), which combines liquidation of the
model with a delayed cadence. And the additional material
after variation 3 (bars 1^4-68) appears on the one hand as an
epilogue to the subsidiary theme (because of the analogy to 1
and 2) and on the other as the closing group of the entire
exposition.
Despite its excessive length, the development section is
littlemore than a paraphrase of the principal group. The ante-
cedent is in E-flat major (bars 180-89), the consequent in E
major (bars 201-9), the elaboration at first in E minor (bars
210-31), and then in A minor (bars 234-55). A developmental
model, derived from blending the conclusion of the exposition
(bar 167) and the lament bass (bars 15-20), serves to introduce
both the antecedent and consequent (bars 168-79 and 189-200)
and is also inserted into the elaboration (bars 218-27 an d
242-51). This model certainly modifies the fundamental char-
acter of the development —
that of a paraphrase of the exposi-
tion by means of modulations —
but does not erase it. In other
words, the development consists of further links in the chain of
variations, now involving changes of tonality. And because the
principle of variation includes the idea of an irreversible pro-
cess, even the recapitulation —
despite its implied tonal restora-
tion and its avoidance of interpolations —
continues to trans-
form the themes. (Since he omits variation 2, which digresses
to remote tonal areas, Schubert begins the subsidiary theme in
the subdominant; in this way he retains the effect of harmonic
surprise in the last variation, where the tonic emerges as some-
thing unexpected.)
The affinity for double variation, essential to the form of
the G-Major Quartet, can be traced back to the early history of
sonata form and continues to play a role in its later develop-
ment. The treatment of the development section as a tonally
extended variation of the main theme, whose contours it delin-
eates, is related to the origin of sonata form in the suite. And in
Haydn's ever more apparent tendency toward thematic con-
centration, thus toward filling the functionally differentiated
stations or "periods" of the form with similar thematic mate-
rial, we can recognize another condition for the proximity

of sonata and variation forms. The practice continues after

 —
4 SCHUBERT
Schubert as well: in Brahms and Mahler we observe the ten-
dency to transform symphonic sonata form, now presented in
an extreme fashion, into a cycle of variations on two theme
groups in regular alternation. It can therefore be said without
exaggeration that the first movement of Schubert's G-Major
Quartet belongs to a tradition of sonata form a tradition, —
however, that deviates substantially from the evolutionary path
dictated by Beethoven.

//

The tendency toward variation cycle, whereby Schubert's quar-

tet movement deviates from Beethoven's model of sonata form,
would nevertheless hardly have been possible without the pro-
found changes in the structure of the variation chain and in —
the very notion of variation —
wrought by Beethoven's epoch-
making "Eroica" Variations, op. 35. (Beethoven spoke of a
"completely new way.") The distance from Beethoven reveals
at the same time a dependence upon him.
In a literal sense it is inadequate — or at least misleading
to speak of "theme" being varied in the principal group of
a
the exposition of the G-Major Quartet. As in the "Eroica"
Variations, the nucleus of the movement is "thematic
really a
configuration." The melody, bass, and harmonic-metrical
scheme of Beethoven's Op. 35 form nearly equal components of
the "theme"; although tightly linked, they are taken up and
varied independently in some of the variations. In Schubert,
too, the thematic configuration embraces a multitude of ele-
ments, whose linkage and separation create a thematic pro-
cess — one that evades the simple notions of statement and
elaboration. The "theme" of the principal group is not a clearly
outlined structure serving as a point of reference for thematic-
motivic development, but rather an embodiment of character-
istics that are "thematic" inasmuch as they represent starting
points of a musical chain of logic. Nowhere, however, do these
combine to form a shape that could be considered "the theme."
Although they function syntactically as a kind of intro-
duction (because of their relaxed construction), bars 1-14 be-
long fundamentally to the exposition. The major-minor shift
in the first chord, the rhythmic pattern of the opening bars,
 Sonata Form in Schubert

Example 1

^^
t

ifa J"' 1
*i
p erase,
f ff
1>

Example 2

ES
10

i > r I V * I
fl

PP

and even the half-step figure of bar 5, which is separated off

from bar 4 —
all these are "thematic" (ex. 1). The isolated half

step then evolves into the ending of the melodic gesture of the
principal theme (the counterpoint to the lament bass) by means
of augmentation and transformation into a whole step (ex. 2).
The stressed upbeat of bar 15 has been anticipated in bar 3.
And the major-minor alternation extends ultimately into both
the second and fourth reworkings of the principal theme: in
bar 25 the minor variant (G-Bt) of the major third (G-Blj),
though still recognizable, is embedded within a diminished-
seventh chord (CU-E-G-B1>); in bar 54 the major-minor shift
emerges openly, once again transferred from the beginning of
the introduction to the principal idea. (Like "introduction,"
the expression "principal idea," actually synonomous with
"theme," is used here for lack of a more appropriate term. It is
justifiedby the musical syntax if not by the substance: al-
though bars 15-23 differ from 1-14 in their tighter organiza-
tion, they form no basic thematic shape, but rather one ele-
ment among others comprising the thematic configuration,
a configuration whose first components have already been
presented in the "introduction," on the other side of the "prin-
cipal idea.")
What distinguishes the modification of a thematic config-
uration from the variation of a clearly delineated theme can be
seen most effectively in the third and fourth versions of the
principal idea. Bars 33-42 represent unmistakably a variant of
bars 15-19, despite the fact that they adopt neither the melodic
6 SCHUBERT
gesture of bars 15-19 nor the descending chromatic fourth
of the bass. All that has been taken over from the principal
idea, it appears, is the harmonic model, the chord progression
G— D— F— C— Et>, which is now stretched to twice its length,
following the example of figural variation. The rhythmic pat-
tern, however, is derived from the introduction (bars 1-4); and
the superimposition of introduction and principal idea creates
retroactively a relationship between the parts, through which
they reveal themselves as elements of a thematic configuration.
Still, one can speak of a variant of the principal idea. On the

one hand, the lament bass is perceptibly implied by the har-

monic progression G-D-F-C-E^, and on the other hand the
rhythm of bar 3 represents the model for bar 15: therefore, the
association between first and third versions of the principal
idea is close enough to justify the notion of a transformed
model.
The fourth version, finally, is based upon a two-bar phrase,
imitated in stretto (bars 54-59). The major-minor alternation
derives, as already mentioned, from; the opening bars; bars 2
and 4 of the rhythmic pattern are contracted, or bars 43-44
(which are in themselves related to bars 2 and 4 through the
mediation of bars 34 and 36) imitated in varied form; bars
43-46 are the model for imitation at the distance of a bar; and
the harmonic design appears as a reflection of the principal
idea, whose falling sequence (the linear progression G-F#-R|-
Eti-El>-D = the chord progression G-D-F-C-E^-G) is re-
shaped into an ascending one (the linear progression G-F#-
A-Qt-B-A|t = the chord progression G-D-A-E-B-Fjt).
(Of the elements into which the lament bass can be divided
the falling half step, the sequential technique, and the direction
—
of the sequence two have been preserved. Thus it does not
seem inappropriate to speak of a variant.)
The wealth of relationships, which renders the principal
theme of the exposition comparable to a tightly knit web, can
be described only if we abandon the notion of a theme as a —
clearly outlined rhythmic-diastematic-harmonic shape for—
that of thematic elements. We must operate instead with such
categories as "alternation between major and minor third" (in-
dependent of the scale degree from which the intervals are
 Sonata Form in Schubert 7

reckoned), "rhythmic pattern" (separated from the diastematic

structure), or "sequence of descending half steps" (without a
specified direction of the sequence). Schubert is a composer
whose musical imagination is to an exceptional degree tied to
the sensuous phenomenon. It is thus surprising to find him
having recourse to "abstract" elements —
disengaged from the
"concrete" theme or motive in which rhythmic, diastematic,
and harmonic structures have grown together.

///

The concept of thematic process, which not coincidentally was

invented for Beethoven's music and only later transferred to
Classical-Romantic instrumental music, embraces two ele-
ments: the "logical" element of motivic-thematic derivation
and the "pathetic" one of a development pressing constantly
forward. Because of their inseparable amalgamation in Beetho-
ven, it could scarcely be believed they were not associated as a
matter of course. The compelling force of Beethoven's sonata
form arises from the fact that the logic of musical discourse,
which extracts later events from earlier ones through motivic
work, developing variation, and contrasting derivation, is of
the same significance as the energy that maintains the music in
a state of nearly permanent intensification by means of con-
tracted phrases, accelerated harmonic rhythm, and concen-
trated accents. (The tension lets up only in the arcadian mo-
ments when the music, as it were, draws a breath. ) Yet Schubert's
G-Major Quartet shows that consistent musical logic the —
weaving of a tight fabric of motivic relaionships is quite —
reconcilable with a relaxed pace and a musical attitude that, de-
spite its agitation, remains devoid of pathos. Although the
concept, of a thematic process normally calls to mind the ho-
mogeneous image of both insistent energy and compelling
logic, these two characteristics are in fact separable. (It would
be mistake to invoke the terminological strategem which
a
claims that a logic spinning its web inconspicuously, instead of
by inexorable syllogisms, cannot be counted as "logic," but
only as mere motivic "association.")
8 SCHUBERT
When Schubert has the rhythm of the subsidiary theme
(bar 65) emerge gradually during the principal group (bars 2,
34, 43, and 51), nothing prevents us from speaking of "con-
trasting derivation" [kontrastierende Ableitung]. The same holds
true when the chromatic descending fourth, which underlies
the principal idea as a basso ostinato, returns in the subsidiary
theme as a middle voice (bars 65-69: D-Q-Q-B-Ajl-Ati).
(In the reprise of the subsidiary theme, bars 344-46, the im-
plied chromaticism stands out clearly as a lyrical counterpoint
in the bass.) To be sure, the contrasting derivation presents it-
self less ostentatiously than in Beethoven's F-Minor Sonata,
op. 2, no. 1, in which the subsidiary theme is at once an inver-
sion of the principal theme and a reversal of its character,
so that the aesthetic transformation is in precise correlation to
the technical one. But there is no reason to withhold from
Schubert's procedure the term "contrasting derivation," which
was coined by Arnold Schmitz for Beethoven's music. In both
cases the thematic ideas display unmistakable interdependence
as well as extreme diversity. And the fact that Schubert adopted
the technical idea from Beethoven (the term "contrasting deri-
vation" refers only to this specific compositional procedure)
points up all the more clearly the aesthetic difference (which
the nomenclature alone is hardly adequate to confirm).
The chromaticism of Schubert's subsidiary theme appears
as a reflection or echo of the principal idea. The connection be-
tween the themes seems to derive not from the principal idea,
whose structure determines that of the subsidiary theme (by
analogy to a consequent), but rather from the subsidiary theme
itself, which seizes upon a trait of the principal idea like a remi-
niscence. In Schubert, unlike in Beethoven, the most lasting
impression is made by remembrance, which turns from later
events back to earlier ones, and not by goal-consciousness,
which presses on from earlier to later. (The preponderance of
one factor should not, however, be taken for an exclusive
status, which is impossible according to the psychology of
form.) The teleological energy characteristic of Beethoven's
contrasting derivation is surely not absent in Schubert, but it is
perceptibly weaker. Conversely, Schubert's procedure gains an
element of the involuntary: the link between the themes is not
 Sonata Form in Schubert 9

deliberately brought about; simply happens. (This distinction

it

is to be regarded as a statement not about the real, completed

compositional process, but about the intentional, present aes-
thetic subject that the listener perceives as standing behind the
music.)
Insofar as the character of reminiscence predominates, the
contrasting derivation is similar in nature to variation tech-
nique, into which the sonata form of the G-Major Quartet dis-
solves. For Schubert's variation principle too is distinguished
less by the dramatic emergence of consequences (like Beetho-
ven's "completely new way" in the "Eroica" Variations) than by
a certain placidity among the relationships. That an "introduc-
tion" proves in retrospect to be "thematic" means something
"Tempest" Sonata, op. 31, no. 2. In
different than in Beethoven's
Beethoven the rhapsodic, prelude-like opening later reveals it-
self to have been a rudimentary expression of a principal and a
subsidiary theme, which grow from the initial shapes with ir-
resistible power. In Schubert the elements of the "introduc-
tion" — the rhythmic pattern, the isolated half step, and the
—
major-minor alternation are related in the way that images of
recollection overlap with one another.
The wealth of motivic relationships in Schubert is as ap-
propriate to a sonata form that tends toward variation cycle as
it is to Beethoven's dialectically developmental forms. In fact,

the affinity to a series of variations can be confirmed by appeal-

ing to the aesthetics of reception: the more intricate the struc-
ture that grows out of motivic relationships, the simpler the
form that constitutes the supportive framework. It is precisely
the simplicity of a succession of variations that renders com-
prehensible remote relationships that would scarcely be dis-
cernible in a rhapsodic development section, because in a varia-
tion one expects certain connections to occur at certain points.
The elements of the fourth version of the principal idea in the
—
G-Major Quartet the major-minor alternation, the rhythmic
pattern of bars 2 and 4, and the sequence model of bars 15-20
appearing in retrograde —
would hardly reveal their origin if
both the principal of variation and the incorporation of charac-
teristics of the "introduction" had not already been impressed
upon the listener by earlier variants.
10 SCHUBERT
IV
The bass of the principal idea, the falling chromatic fourth, ap-
pears in the G-Major Quartet harmonized as 6
G-D
-F-C 6 -
Ek— G, which deviates from the tradition of the passus du-
riusculus, as it was called in the seventeenth century. Whereas
Beethoven's C-Minor Variations present the same lament bass
in a tonally coherent form (see ex. 3), Schubert's version evades
any labeling according to function (the difference between ca-
dential and sequential harmony is independent of that between
minor and major). The chord progression G-D-F-C
origi-
nates in a scheme that is encountered more frequently as a
rising sequence (C-F-D-G) than a falling one. The pro-
gresssion E^-G disrupts the predictable scheme, yanking the
sequence forcefully back from its progression, or regression,
6
into the infinite: E^-B^-D^-A^ . . .

Yet it was precisely the from the infinite

step into or back
to which Schubert, with his genius for letting things run their
course, was drawn. And while in the exposition of the G-Major
Quartet he inserts the chromatic sequence syntactically into the
quadrature of phrase construction —
though not without vio-
lence — and preserves at least tonally the outward appearance of
closure (by means of the abrupt return to G), in the develop-
ment section he ignores syntactical and tonal considerations.
The descending chromatic progression of bars 168-77, gener-
ated by blending the end of the exposition and the lament bass
of the principal theme (and repeated in bars 189-98, trans-
posed up a half step), spans no less than a tenth (if one ignores
the octave transfer). And the impression left by the develop-
ment section is one of a largely limitless continuation, which
breaks off as if by chance, not because a goal has been reached.
The harmonization of the lament bass in the exposition
inspires extreme consequences in the development section:
from the fragmentary juxtaposition of keys in whole-tone

Example 3
 — 1

Sonata Form in Schubert 1

Example 4

m ifJTO 1
rum ft jm a B i u- i

g p. f ^ r l^.fp].|J } /77}-l^JJn. ll

transpositions (G: I-V 6 , F: I-V 6 Ek , I) Schubert extracts a

whole-tone scale, which emerges from the interior of the se-
quential harmony, so to speak, and yet can no longer be justi-
fied by appeal to a traditional harmonic model (see ex. 4).
The whole-tone scale, a perplexing phenomenon in the
1820s (although the G-Major Quartet was not performed in
public until 1850), is partly, though not entirely, disguised as a
chromatic scale; upon close listening it emerges unmistakably
as the framework. And if one reckons from the neighbor notes
in between the whole steps, a pattern of 1 +4 + 4 bars crys-
tallizes in bars 168-77 ( see arrows in ex. 4): D is the final note
of the exposition, and the actual whole-tone scale, which
spans an eleventh, therefore comprises (as a model with se-
quence) two tritones (C-B and E-Bl>). But in distinction to
the chromatic descending fourth from which it originated, the
tritone progression by way of whole tones is an "extrater-
ritorial" phenomenon within harmonic tonality. (The sequen-
tial harmony forms the intermediate link: in part tonally intan-

gible, in part integrated by custom.) Out of the sequential

harmony that Schubert impressed upon the lament bass
and that represents the alternative to the cadential harmony of
Beethoven's C-Minor Variations —
arises finally a succession of
whole tones, which descends into a tonal abyss.
The chromaticism of the development section, the endless
motion that ceases somewhere without making the turning
point its goal, is of the same nature as the succession of varia-
tions into which Schubert dissolves sonata form in the G-Major
Quartet. A tendency toward the boundless is perceptible in
both, no matter whether Schubert yields to it or represses it
under the constraint of formal principles.
12 SCHUBERT
What
appears as involuntary occurrence nevertheless rep-
resents the outward manifestation of precise calculation; there
is no question of unreflective composing. Without intense ra-

tionality, the progression from sequential harmony to a tonally

extraterritorial whole-tone scale articulated in tritones would
be unthinkable as the wealth of motivic relationships, which,
as
embedded within the framework of the variation cycle, vindi-
cate it as the formal scheme of a sonata-allegro movement.
 Schubert's Promissory Note:
An Exercise in Musical Hermeneutics

EDWARD T. CONE

Hermeneutics isdefined by the Shorter Oxford English Dic-

tionary as "the art or scienceof interpretation, esp. of Scrip-
ture." It was toward the end of the last century that Hermann
Kretschmar applied the word to the verbal elucidation of musi-
cal meaning, but musical hermeneutics was an art (or science)
that, under one rubric or another, had long been practiced.
Music, it was generally agreed, had meaning. It was an art of
expression; therefore it should be possible to determine what
it expressed and how. Kretschmar, with his detailed attempts at

exact explanation, thus represented what might be called the

dogmatic climax of a long tradition.
At the very time he was writing, however, a more for-
malistic view, expounded by critics like Gurney and Hanslick
and furthered by theorists like Schenker, was gaining support.
Indeed, for a time the purists were successful in discrediting
hermeneutical methods and results — although their victories
may have been due less to the force of their own logic than to a
widespread reaction against the excesses of ultra-realistic pro-
gram music on the one hand and of literalistic "interpreta-
tions" on the other. As a result, even those who still defended
the possibility and relevance of musical expression were loath
to define its nature in any specific terms.
Today, however, when it is once more acceptable to admit
an interest in the subject matter and iconography of a painting,
it has become increasingly feasible to discuss the putative mean-

ing of a musical composition without evoking immediate deri-

14 SCHUBERT
sion. There have been a number of recent books that try, from
various points of view, to come to grips with the problems
1
involved.
To be sure, not all discussions of musical meaning rely on
the concept of expression. Wilson Coker, following certain
semiologists, conveniently distinguished between two types of
2
meaning: congeneric and extrageneric. The first of these refers to
relationships entirely within a given medium. As applied to
music, includes the significance that each part of a composi-
it

tion possesses through its connections with other parts of the

same composition, and the significance that inheres in the
composition as a whole through its employment of a recogniz-
able sonic vocabulary organized in an appropriate manner.
Congeneric meaning thus depends on purely musical relation-
ships: of part to part within a composition, and of the com-
position to others perceived to be similar to it. It embraces the
familiar subjects of syntax, formal structure, and style. And in
fact, when the lucubrations of the recent school of musical
semiologists are shorn of their pretentious jargon, that is all
they are usually discussing —
syntax, form, and style (and by
no means always originally or even sensibly).
What will chiefly concern us, however, is extrageneric
meaning: the supposed reference of a musical work to non-
musical objects, events, moods, emotions, ideas, and so on.
Here, obviously, we have returned, under another name, to the
realm of hermeneutics, and to the problem of musical expres-
sion. On this subject there is still wide divergence of opinion;
even those who vigorously defend the concept often disagree
as to its nature, its range, and its limits. Whereas the relative
stability of congeneric interpretations has tempted some ana-
lysts to claim that their conclusions are objectively demon-
strable, hard to reach any consensus about the expressive
it is

or other extrageneric significance of even the simplest com-

position, save perhaps in the broadest terms. Writers who are
clear and precise on congeneric meaning often become very
fuzzy when they turn to extrageneric, even while insisting on
its importance.

Coker, for example, discusses the Funeral March from the

Eroica Symphony in terms that, with very slight modification,
 5

Schubert's Promissory Note 1

could equally well be applied to, say, the Funeral March from
Chopin's Sonata in B-Flat Minor:

The first dominated by an unmistakably

part ... is

mournful which the second part

attitude, to contrasts . . .

a prevalent mood of comfort with an alleviation of sorrow

and even moments of exultation. The lament of the . . .

song is conveyed in gestures striving to ascend only to fall

back. as under the weight of sorrow. The tone of
. . ,

mourning is established by a number of effects: the very

slow, fixed pace of the underlying march rhythm; the
minor key and the long initial reiteration of the tonic triad
. . . ; the prevailing lower registers ... in fairly dense
sonorities; and the suppressed dynamic levels. 3
. . .

One could continue further, omitting (as above) only the tell-
tale measure numbers and remarks on instrumentation; for this
account of extrageneric meaning, like many others, deals pri-
marily in surface generalities. A slow, plodding pace; narrowly
circumscribed melodies; somber usually minor —
harmonies —
in a low register: these are the immediately apprehensible char-
acteristics of the musical surface that identify the typical fu-
neral march, whether by Beethoven or by Chopin.
As the example shows, it is easy to assume that one has
explained content when in fact one has only defined genre.
Marches and dances, nocturnes and scherzi the names sug- —
gest gestures, affects, or moods, to which certain musical
characteristics are thought to be appropriate usually broad —
features of tempo, meter, rhythm, mode, and the like. And
such features in turn are, it is true, often accepted as defining
basic types of expression. But generalizations are not enough,
for surely, if a musical composition expresses anything at all,
the importance of the expression must reside in its uniqueness
to that composition, not in what the composition shares with
a dozen others of the same genre. Specificity: that is what
Mendelssohn was getting at in his oft-quoted statement: "The
thoughts which are expressed to me by music that I love are
not too indefinite to be put into words, but on the contrary,
4
too definite."
Surface generalities, then, can tell us only as much about
 —
16 SCHUBERT
the content of a piece of music as the subject of a picture
e.g., aCrucifixion or a still life —
can tell us about its content.
Chardin and Cezanne both painted oranges and apples, but
what each expressed in his pictures was his personal vision
of the fruit. In the same way, Beethoven and Chopin pro-
duced widely differing versions of the same subject, the fu-
5
neral march.
Subject matter again is what is described, in a more de-
tailed way, by the motivic vocabularies of the various types of
Figurenlehre, whether in earlier or in more recent formulations.

—
More detailed, and often more mechanical yet not more spe-
cific, as Coker reveals when he employs the method to explain
the effect of the Trio of the Funeral March, which "clearly con-
trasts to the first [part] by giving a relief from distress and a
more encouraging attitude. Emphasis is on ascending gestures
in triads . . and in steadfast scales rising." 6
.

No; the locus of expression in a musical composition is

completely defined by neither its wider surfaces nor its more
detailed motivic contours alone, but by its comprehensive de-
sign, which includes all the sonic elements and relates them
to one another in a significant temporal structure. In other
words, extrageneric meaning can be completely explained
only in terms that take account of congeneric meaning. If ver-
balization of true content —the specific expression uniquely
—
embodied in a work is possible at all, it must depend in large
part on close structural analysis. 7
That analysis is all most of us need, most of the time. For
surely the best that can be said for the verbalization of content
is what Mendelssohn went on to add, in the quoted letter:

"And so I find in every effort to express such thoughts, that

something is right but at the same time, that something is lack-
ing in all of them." Nevertheless, those of us who do believe in
the existence and relevance of extrageneric meaning ought to
find it uncomfortable to be in the position of insisting on the

validity of a concept that cannot be precisely defined, and in-

stances of which are at best problematic. Once in a while we
should try to derive from the structural analysis of a composi-
tion an account of its expressive content. That is what I pro-
pose to do in the case of Schubert's Moment musical in A-flat,
op. 94, no. 6 (see Appendix, ex. A).
 Schubert's Promissory Note 17

Since a complete analysis of even such a short piece let —

alone a detailed treatment of its expressive content —
would take
much more than the available space, I shall limit my discussion
to salient features that are special to this composition, and I
shall try to show how their congeneric interrelations account
for their extrageneric significance. I shall therefore assume
agreement about certain formal aspects of the piece, taking it
as established that the Moment musical consists of a three-part
song form in A-flat major with a Trio in the subdominant,
D-flat major. Its phrase articulations, harmonic progressions,
and motivic manipulations are clearly defined. The opening
statement consists of two balancing eight-bar phrases, each ar-
ticulated 2 + 2 + 4, the antecedent cadencing on the dominant
and the consequent on the tonic. The soprano line establishes
the fifth degree, Et, reaffirms it by the tonicization of V, and
descends to A^ for an authentic cadence. A pattern of suspen-
sions set up at the outset suggests a meter of \ underlying the
notated \\ moreover it produces a little rhythmic a aba pattern
in each large phrase. All then is in order: a period exhibiting
exact parallelism.
But does it? There is an arpeggiated upbeat in bar 4 that is
not duplicated at the corresponding point in the consequent
(bar 12). This prepares for a much more momentous element
of imbalance: the descent of the bass into a lower register and
the fp that marks the tonicization of V in bar 7. These charac-
teristics of range and dynamics are echoed, not at the corre-
sponding cadence at the end of the consequent, but at its mid-
point, bars 10-12. Here too is a tonicization, this one effected
by two chromatics: a Bt], which, like the earlier D\\, resolves
normally upward, and an Et|. The arrival of the latter, replac-
ing a previous E^ as the resolution of the suspended F, pro-
duces another half cadence, this one on V/vi. But now the Eli
moves down to El>, pulling an implied tonicization of vi back
into the original key.
The result is to make of the prominent Et| what call, if
I

the pun may be forgiven, a promissory note. It has strongly

suggested an obligation that it has failed to discharge — in the
present case, its function as a leading tone. Now, I do not wish
to suggest that incomplete tonicizations represent promises
all

to be kept during the future course of the composition in

18 SCHUBERT
which they occur. The persistence of such an implication
depends not only on the specific context in which the pro-
gression occurs, but also on wider influences of style, both
personal and historical. I suspect, for example, that the devel-
opment of nineteenth-century harmony might be analyzed
largely in terms of the increasing freedom shown by compos-
ers in their dealings with promissory situations — in the devel-
opment of more and more unorthodox methods of repayment,
even in the eventual refusal to recognize the debt. (As we shall
see, that is almost the case with Schubert in the Moment musical.)
Normally, in music of the eighteenth and early nineteenth
centuries, promissory status is demanded, or at least requested,
—
by a note or more accurately, though less paronomastically,
—
an entire chord that has been blocked from proceeding to an
indicated resolution, and whose thwarted condition is under-
lined both by rhythmic emphasis and by relative isolation.
Rhythmic emphasis, of course, results from the stressed posi-
tion of the chord within (or outside of) a phrase, from special
agogic or dynamic inflection, or from a combination of those.
Isolation is effected not only by the motivic detachment that
often separates such a chord from its surroundings, but also by
harmonic novelty. The promissory chord is promoted, so to
speak, by an insurrection that tries, but fails, to turn the course
of the harmony in its own new direction.
Although such chords are temporarily deprived of their
expected resolutions, the result is not, properly speaking, a de-
ceptive cadence. What makes the standard deceptive cadence
work is the fact that the dominant chord is almost properly re-
solved; indeed, the voice leading binds it even more tightly to
its successor than in the typical authentic cadence. A prom-

issory chord, on the other hand, is separated from what fol-

lows by a sudden switch in direction, of voice leading as well
as of harmony — and most often by a break in the rhythmic
flow, too. The combination of emphasis and separation draws
special attention to the unresolved chord and enables it to es-
tablish its influence so powerfully that it seems to require later
attention, the most obvious form of which is a prominent res-
olution so stated as to remind the acute listener of its connec-
tion with the promissory chord.
 Schubert's Promissory Note 19

Closely related to the promissory situation —

and, indeed,
often similarly treated —
is the half cadence, usually on V/vi,

that sometimes concludes the development of a Classical sonata-

form movement, or the entire slow movement of a Baroque
sonata or concerto. Unlike the typical promissory chord, how-
ever, which appears early in a movement or a section, the sec-
ondary dominant in this case normally appears at the end of a
progression that has already clarified its syntactic function. Its
resolution, supplied in advance as it were, is clearly understood
and requires no later confirmation (although, to be sure, that
may be forthcoming anyway 8 ). Look, for example, at the end
of the development of the opening Allegro of Beethoven's Quar-
tet in D, op. 18, no. 3. The reiterated V/vi in bars 154-56 is
the climax of a passage in which the tonality of vi (F-sharp
minor) is explicitly formulated, and in which the responsibili-
ties of its dominant have in a sense already been discharged.
The composer even gives one ample time to think that over, as
he sustains the cello's Q
for two bars before transforming it
into the leading tone of the true tonic, D, and quitting the F-
sharp-minor tonality for good.
Contrast now the opening Allegro of the same composer's
Piano Sonata in F, op. 10, no. 2. Here the smooth progress of
the movement is very soon (bars 16-18) interrupted by a sud-
—
den half cadence on V/iii a threefold statement in an em-
phatic forte, rhythmically and motivically separate from its
surroundings. What follows is a new theme in the dominant
that leaves the unresolved V/iii hanging as an obvious prom-
issory chord. It should therefore cause no surprise to find the
development section beginning with a sudden shift to the me-
diant (almost as if the entire dominant section of the exposition
had been a parenthesis). To be sure, the development might
have started that way in any case; but the virtue of the prom-
issory technique is rarely to elicit a harmony that would other-
wise be missing —
rather, it is to draw temporally separated
sections of a work into more intimate and more interesting
connection.
In the Moment musical no. 6, isolation and emphasis work
together to produce a strong promissory effect. The pattern of
rests has, from the outset, separated each motive from its fel-
20 SCHUBERT
lows. That separation is exaggerated in the case that interests
us, where the subito forte, the octave shift in the bass, and the
octave doubling all draw special attention. (Contrast bars 4-8,
where the descent in the bass is arpeggiated, the doublings are
introduced gradually, and the forte occurs during the rise and
fall of the <C ^> fp
The move toward F minor
.)

is hardly unprepared. That

chord is foreshadowed in the opening: subtly embedded in the

first downbeat, more frankly stated in the vi* of bar 3 But .

both of those are treated as suspensions and fail as functional

harmonies. The half cadence of bars 10-12, with its strongly
dominant-seeking French sixth, uncovers the concealed sub-
mediant influence. So when the concluding phrase member,
dutifully remembering the demands of the true dominant of
bars 7-8 suppresses the tendency toward vi and turns the El;
downward to EK the Et; remains in the ear as a troubling ele-
ment of which one expects to hear more.
It is not surprising, therefore, that Eti dominates the con-
trasting section of the song form. But unlike the development
of the Beethoven sonata, which legitimized the foreign leading
tone by an immediate reference to its tonic, this one is by no
means eager to do likewise. It prefers rather to dwell on the
promissory note and to investigate further its peculiar connec-
tion with Ek The first step is to restate B\ as B. In so doing,
the music enlists the two other previously heard chromatic
tones, Bk; (O) and Dt|, sounding all three together in a German
sixth (bars 16-17). The Et|-El> contrast is thus explained by
reinterpretation as 6-5 in A-flat minor — a detail statedthree
times during the initial phrase of the development: first in the
bass, next in an inner voice, exposed soprano.
and finally in the
That phrase, constructed in the 2 + 2 + 4 model of its
predecessors, terminates on a half cadence in A-flat minor, and
its consequent at first seems to promise the expected conven-

tional balance. But no: the B reasserts its importance. Refus-

6
ing to be drawn back into A-flat minor, it replaces the El> in i
in order to convert it into Vlj (bars 28-29), thus initiating a
modulation to its own key, spelled for convenience as E\\. This

time it is the turn of the E^ to assume the subidiary role.

Spelled now as Djl, it must resolve as a leading tone to B|. The
extension required to produce a convincing cadence results in
 Schubert's Pro m issory No te 21

the first violation of the eight-measure pattern, a phrase with

one additional bar: 2 + 2 + 5. But that is not all. The final E is
extended as a pedal in the bass, over which a new three-bar ca-
dence is twice stated. The melodic and rhythmic flexibility that
has gradually insinuated itself into the preceding phrases is
—
now at its most ingratiating by virtue not only of the three-
bar format itself, but also of the triplet upbeat, the embellishing
grace notes, and the chromaticism of the sinuously introduced
and cancelled All (bars 34-35). The melody, by completing an
octave descent from the original B| (bars 29-33), leads into a
richly sonorous lower register. The restrained, carefully mea-
sured satisfactions of the opening have been gradually trans-
formed by the development into the more sensuous delights of
a berceuse.
The return to the original key is simple: the major chord
on E, by the addition of a seventh, reverts to its status as an
augmented sixth on F\> (bar 41). At the same time, the move
suggests a slight uneasiness with respect to the cadence on E:
was it as firm as it seemed? Was it not perhaps usurping a
tonicization to which it had no right? It is significant that the
key was never clinched by a clear reference to some form of
subdominant. There was an obvious opportunity for a ii 6 in
bar 32 (as in ex. 1), but the offer was spurned: the retention of
B in the bass converted the chord into a dominant ninth.
Moreover, in bar 41, only the momentary doubling of Ql (M>)
keeps it from being heard as a leading tone, and the entire mea-
8~7
sure as V on E preparing for a cadence on A, with a con-
tinuation in that key (as in ex. 2). That temptation is resisted,
however. The chord is interpreted as an Italian sixth, which in-
itiates the return to Ak At the same time, a recall of the original

Example 1

31

w^mm *37

ggfey
22 SCHUBERT

suspension motive induces a reversion to the binary bar pat-

tern. After two four-bar phrases (bars 40-47), the third, on a
climactic forte, is extended to six bars by a carefully written out
diminuendo-ritardando merging without cadence into the
two-bar motive that inaugurates the reprise. Thus the irregu-
laritiesof the development section have yielded to a version of
the original pattern: 4 + 4 + 8 bars (bars 40-55) that overlap the
returning 2 + 2 + 4 (bars 54-61), much as the motives of the
development overlap those of the opening.
That is perhaps one reason why the reprise does not begin
immediately after the arrival of the tonic in bars 46-47. There
is another reason, equally important, in the F-minor harmony
6
that introduces the transitional progression vi -vii
7
-V 6
~5 (bars

47-53)- For here at last is the long-postponed discharge of the

responsibilities of Et} as a leading tone. True, the F minor is not
tonicized, nor is its bass assigned to a root in the lower register;
but there are significant indications of the connection neverthe-
less. The doublings are heavy. The subito forte recalls that of
bars 10-12. The line, leading from F down to BJ>, and forecast-
ing an ultimate AK is a reinterpretation of the original descent
of bars 11-16. To hear the long-range connection I am trying
to establish, play the following in unbroken succession: bars
9
73~92, bars io 3 -i2 2 and then bars 47 3 — 53 2 (ex. 3).
,

At last, then, the promise of Eli as a leading tone has been

kept. Yet how close the music came to forestalling the fulfill-
ment! That can be demonstrated very simply by a performance
that omits the passage comprising bars 47 3 -53 2 The result, su-
-

perficially at least, is acceptable. And is there not perhaps a

touch of irony in the insinuation of the problematic note once
again into the descending line of bars 51-53, immediately after
the emphatic proclamation of F minor? To be sure, the note
 Schubert's Promissory Note 23

now seems docile, forming a passing and passive diminished

seventh instead of a recalcitrant applied dominant. At the same
time, it can be taken as a signal that the road to the final ca-
dence is not quite clear, despite the unusually felicitous re-
establishment of the tonic major.
There remain two other troubling points as well. One is
the nagging doubt left by the E-major cadence in bar 33: was it
a completely satisfactory tonic, or did it exude a faint domi-
nant flavor? Did the development, in giving E to its own head,
perhaps encourage it to incur still further obligations? The other
point looks forward to the course of the recapitulation: now
that the V/vi has been satisfactorily resolved, how can it re-
turn? If it cannot, what will replace it?
What does occur, as Elizabeth Bowen says of the action of
a well-constructed novel, is unforeseen in prospect yet inevi-
10
table in retrospect. The first phrase of the reprise is regular; it
differs from its model only in a cadence that is /rather than fp.
That modification permits a forte consequent that retains the
—
lower bass register and its octave doubling a type of instru-
mentation appropriate to a crucial chord: an alteration of the
7
original IV into the most painful dissonance of the entire piece.
For the B| has returned once more, now as an Bthat replaces
the F (bar 62). This B>, in turn, twice forces the suspended C to
pass through O —
on the way to its resolution on B^ thereby
filling in the one element missing in the chromatic descent of

Example 3
 -

24 SCHUBERT
bars 51-55, and at the same time reversing the situation of bars
10-12, where it was the Bt; that had called forth the El?
The new combination of B> and O
is too strong to be re-

sisted. The of the phrase brings them

fortissimo continuation
both back with a new bass, EH>, here spelt as Dtj. That, of
course, is none other than the dissonant tone of the old German
sixth of bars 16-17. This time, however, it is in the powerful
bass position; and this time the chord insists on being treated as
a dominant, thus confirming our earlier suspicions.
The result is an expansion of the consequent phrase that is
terrifying in its intensity. The phrase is, as it were, broken
wide open by reiterations of the new V 2 and its resolution to a
tonicized Neapolitan. A first attempt to return to the true tonic
fails, interrupted again by an echo of the Neapolitan interpola-

tion. The effect of that interpolation lingers even when the re-
turn to the tonic at last succeeds. The approach is made each
time through the minor; and it is the minor color that remains
in the ear, even though the final resolution is on a starkly am-
biguous octave Ak The harmonic material of the develop-
ment, then, has infiltrated the reprise with devastating effect.
In the same way, the rhythmic irregularity, experienced in the
development as an agreeable loosening of the tight proportions
of the opening, has now almost destroyed the original balance.
Eight bars, 2 + 2 + 4, are answered by sixteen, but not 4 + 4
+ they are 2 + 2 + 5 + 7!
8:

The final empty octave is a neat tactical device. A return to

major would have produced a sudden jar after the minor \\;
moreover, it would have been immediately canceled by the
conventional repetition of the entire section. But a minor chord
would not have led into the Trio, which, construing the final
A\> as a dominant, plunges immediately into D-flat as a new

tonic.
About the Trio I shall point out only that it too plays with
the chromatic neighbors to the fifth degree (now G\\ and B^),
but in an innocuous way, as befits the unproblematical lyricism
of the interlude. (In fact, even the minor third is safe here. It is
— —
heard only once bar 97 and then enharmonically as a pass-
ing Eli, not as a true B). The only moment of ambiguity
occurs during the little digression, when the EH> and its em-
 Schubert's Promissory Note 25

bracing diminished seventh are given a new interpretation (bar

101) in a passage that occasions the only exception to the regu-
larity of the phrase structure (six bars instead of the prevailing
four). The following reprise can now dispense with the B^ al-
together, basing its consequent phrase on the unalloyed major
scale.
The subdominant tonality of the Trio as a whole can be
heard as the expansion of a function embedded in the opening
motive of the entire piece. That initial progression, before its
7
second chord resolves as a suspension, sounds like I-IV The .

impression is even stronger when the first bars, returning after

the close of the Trio, echo the V-I of the D!> cadence. Song
form and Trio are bound together by the A^-O* bass that in-
troduces the Trio and that governs its departure.
The foregoing partial analysis of the structure of the Mo-
ment musical has also been an analysis of its congeneric mean-
ing, for those terms are simply two ways of characterizing the
same body of information. Congeneric meaning or struc- —
tural content, as I prefer to call it, stressing that identity — is

precise and specific, for uniquely defined by a single com-

it is

position. But possible extrageneric meaning —

or what I call
expressive content —
seems to depend on choices from a be-
wildering array of admissible interpretations. I have coined the
phrase "expressive potential" to signify this "wide but not un-
11
restricted range of possible expression." The range is wide
because (pace those who subscribe to the more rigid versions
of Affektenlehre) there is no rule or code by which we can trans-
late musical gestures into exact expressive equivalents cer- —
tainly not in the sense that we can translate words into con-
cepts, or images into objects. At the same time it must be
stressed that the range is not unrestricted; for the expressive
content — the human activity or state of mind adduced as an in-
terpretation of the music —
must be congruous with the struc-
tural content— the musical action itself. In other words, "we
subconsciously ascribe to the music a content based on the
correspondence between musical gestures and their patterns on
the one hand, and isomorphically analogous experiences, inner
12
or outer, on the other." What all such experiences have in
common constitutes what I call the expressive potential.
26 SCHUBERT
What, then is the expressive potential of Moment musical
no. 6? What kinds of human situations present themselves as
congruous with its structure? An no-
astute reader will have
ticed that my analysis has not been wholly objective. have in-
I

sinuated a few leading phrases to suggest to him the kind of

expression I find in the work, and to encourage him to hear it
in the same way. As I apprehend the work, it dramatizes the
injection of a strange, unsettling element into an otherwise
peaceful situation. At first ignored or suppressed, that element
persistently returns. It not only makes itself at home but even
takes over the direction of events in order to reveal unsuspected
possibilities. When the normal state of affairs eventually re-
turns, the originally foreign element seems to have been com-
pletely assimilated. But that appearance is deceptive. The ele-
ment has not been tamed; it bursts out with even greater force,
revealing itself as basically inimical to its surroundings, which
it proceeds to demolish.

That is an account, in as general terms as possible, of the

expressive potential I find in the principal song form of the
Moment musical. When I try to relate this abstraction to a more
specific situation by adducing an "isomorphically analogous
experience" (always, of course, with Mendelssohn's reservation
in mind), I assume the arrival of the "foreign element" to be

symbolic of the occurrence of a disquieting thought to one of a

tranquil, easygoing nature. Disquieting, but at the same time
exciting, for it suggests unusual and interesting courses of ac-
tion. As an old teacher of mine used to say (probably quoting
one of the church fathers), "the first step in yielding to a temp-
tation is to investigate it." That is what happens here. One can
imagine the protagonist becoming more and more fascinated
by his discoveries, letting them assume control of his life as
they reveal hitherto unknown and possibly forbidden sources
of pleasure. When he is recalled to duty, he tries to put these
experiences behind him and to sublimate the thoughts that led
to them. At first he seems successful, but the past cannot re-
main hidden. What was repressed eventually returns and rises
in the end to overwhelm him.
The Trio, of course, tries to forget the catastrophe just —
as one might try to comfort oneself in the enjoyment of art, or
natural beauty, or the company of friends. The identification
 Schubert's Promissory Note 27

of the new D-flat tonic with the subdominant of the opening

phrase might even be taken as symbolizing the attempted re-
covery of a past innocence. No matter: the Trio is doomed to
failure. The memory of the original course of events is bound
to recur, and the -da capo leads inevitably to the same tragic con-
clusion. (Formal repetitions are often best interpreted as repre-
sentations of events rehearsed in memory. A dramatic action
can never be exactly duplicated; yet, as the Moment musical il-
lustrates more than once, we must frequently accept literal rep-
etitions of sections of music usually considered to be highly
dramatic.)
This, then, is the personal contact I make with the psychic
pattern embodied in the musical structure of the Moment musi-
an example of what I have called context: "not the con-
cal. It is

tent . [but] the necessary vehicle of the content." For I be-

. .

lieve that "the content of instrumental music is revealed to each

listener by the relation between the music and the personal
13
context he brings to it." I can go further and suggest a more
specific interpretation of that context: it can be taken as a model
of the effect of vice on a sensitive personality. A vice, as I see it,
begins as a novel and fascinating suggestion, not necessarily
dangerous though often disturbing. It becomes dangerous,
however, as its increasing attractiveness encourages investiga-
tion and experimentation, leading to possible obsession and
eventual addiction. If one now apparently recovers self-control,
believing that the vice has been mastered, it is often too late:
either the habit returns to exert its domination in some fearful

form, or the of the early indulgence have left their indel-

effects
ible —
and painful marks on the personality and frequently, of
course, on the body as well.
I stress this interpretation, not for any moralistic reason,

but because of its bearing on the final step in my investigation.

That consists in an attempt to answer what is possibly a for-
bidden question. What context might the composer himself
have adduced? What personal experiences might Schubert have
considered relevant to the expressive significance of his own
composition?
In dealing with the relation of music to its composer's own
emotional life, I realize that I can put forward only the most
tentative of hypotheses. But I am encouraged to do so in this
28 SCHUBERT
case by the memory of an illuminating passage from Edmund
Wilson's essay on Oscar Wilde. In a discussion of the effects of
the writer's syphilitic infection on his life and work, he wrote:

Read The even the best of the

Picture of Dorian Gray, or
fairy tales, The Birthday of the Infanta, with the Spirochaeta
pallida in mind. In such stories, the tragic heroes are shown
in the peculiar position of suffering from organic maladies
. without, up to a point, being forced to experience the
. .

evils entailed by them. But in the end, in both cases,

. . .

the horror breaks out: the afflicted one must recognize

himself and be recognized by other people as the odious
14
creature he is, and his disease or disability will kill him.

It is now virtually certain that Schubert, too, suffered

13
from syphilis. The
was probably contracted late in
disease
1822; and although it was temporarily ameliorated by treat-
ment, or perhaps just by time, it was, of course, in those days
incurable. Did Schubert's realization of that fact, and of its im-
plications, induce, or at least intensify, the sense of desolation,
even dread, that penetrates much of his music from then on?
(Our Moment musical dates from 1824.) 16 To be sure, a rapidly
increasing emotional maturity was already in evidence wit- —
ness the contrast between the so-called "Tragic" Symphony of
1 8 16 and the "Unfinished" (although that doom-laden score

of fall 1822 may already reflect the composer's early awareness,

or suspicion, of his condition). Later on a cold wind seems to
blow through even some of his sunniest or most placid move-
ments. Listen, for example, to the Andante of the String Quar-
tet in G, to the Adagio of the String Quintet, to the Andante
molto that introduces and interrupts the finale of the Octet.
Listen, above all, to the Moment musical no. 6. Here, if
anywhere, "the horror breaks out." As Shakespeare's Edgar
puts it:

The gods are just, and of our pleasant vices

Make instruments to plague us.

Is it too fanciful to hear a similar reaction musically embodied

in the tonal structure of the Moment musical?
 Schubert's Promissory Note 29

Notes
1. E.g., Leonard B. Meyer, Emotion and Meaning in Music (Chicago:

University of Chicago Press, 1956); Deryck Cooke, The Language of Music

(London: Oxford University Press, 1959); Donald N. Ferguson, Music
and Metaphor (Minneapolis: University of Minnesota Press, i960); Terence
McLaughlin, Music and Communication (New York: St. Martin's, 1970); Wilson
Coker, Music and Meaning (New York: Free Press, 1972); Peter Kivy, The
Corded Shell (Princeton: Princeton University Press, 1980).
2. Coker, Music and Meaning, p. 61.

3. Ibid., p. 172.
4. Letter to Marc-Andre Souchay of 15 October 1842, repr. in Com-
posers on Music, ed. Sam Morgenstern (New York: Pantheon, 1956), p. 140.
5. For further discussion of this analogy, see my "Music, a View from

Delft," in Perspectives on Contemporary Music Theory, ed. Benjamin Boretz

and Edward T. Cone (New York: Norton, 1972), pp. 57-71.
6. Coker, Music and Meaning, pp. 174-75.

7. It has been suggested to me (by my student Kenneth Hull) that the

view expounded here is at odds with the one I defended in "Beyond Analy-
sis" (in Perspectives on Contemporary Music Theory [see n. 5], pp. 72-90).
There is certainly a difference in emphasis, but I consider the two as comple-
mentary rather than contradictory. True, the earlier essay was somewhat
polemical, celebrating the role of "concrete" as opposed to "analytical" val-
ues. But even there I insisted that "artistic expression must involve both con-
crete and analytical values" (p. 85), just as here I shall illustrate, by specific
example, my earlier contention that "expression, by its very definition, im-
plies a relationshipbetween the work of art and something else" (p. 85).
8. The Phrygian cadence that serves as a slow movement for the Third
Brandenburg Concerto is a revealing exception. Whatever this progression
may represent in the way of improvised cadenza or other elaboration, its iso-
lation a.id rhythmic weight are sufficient to ensure the promissory status of
its concluding V/vi, for which there has obviously been no opportunity of

advance payment. Satisfaction comes with the first modulation in the second
section of the binary finale, which introduces an extended passage in vi,
punctuated by an authentic cadence (bar 16) that explicitly recalls and re-
solves the dominant previously left hanging. A standard developmental pro-
gression is thus imbued with wider significance.
9. I.e., from the third beat of bar 7 through the second beat of bar 9.
10. "Story involves action. Action towards an end not to be foreseen
(by the reader), but also toward an end which, having been reached, must be
seen to have been from the start inevitable." Elizabeth Bowen, Pictures and
Conversations (New York: Knopf, 1974), p. 170.
11. Edward T. Cone, The Composer's Voice (Berkeley and Los Angeles:
University of California Press, 1974), p. 166.
12. Ibid., p. 169.
13. Ibid., p. 171.
14. Edmund Wilson, "Oscar Wilde: 'One Must Always Seek What Is
30 SCHUBERT
Most Tragic,'" in Classics and Commercials (New York: Farrar, Straus, &
Giroux, 1950), p. 341.
15. See Eric Sams, "Schubert's Illness Re-examined," Musical Times
121 (1980): 15-22.
16. It first appeared in an Album musical published by Sauer and
Leidesdorf in December of that year. Gary Wittlich has called my attention to
the suggestive title it bore in that collection, Les Plaintes d'un troubadour. Per-
haps the title was the publishers' invention; but one who believes it was really
the composer's may well wonder whether he was not his own troubadour.
See O. E. Deutsch, The Schubert Reader, trans. Eric Blom (New York:
Norton, 1947), pp. 387-88.
 Dance Music as High Art:

Schubert's Twelve Landler, op. 171 (D. 790)

DAVID BRODBECK

In December 1863 Brahms set out to have printed some un-

published dances by Schubert that he had recently acquired.
"I could give you a beautiful collection of 'waltzes,'" he wrote
to his publisher Rieter-Biedermann. "(I would ask 50 fi\, which
I paid for the manuscripts.) The best of all are 12 'waltzes,'

which stand in rank and file on a leaf, with quite the loveliest
1
faces." Brahms's added note that one entire dance and part of
another had already seen print identifies these pieces as the
Twelve Landler, D. 790. The autograph, inscribed "Deutsches
Tempo," is dated May 1823; the second dance and a transposed
version of part of the eighth dance had been published in Janu-
2
ary 1825 as op. 33, nos. 1 and 10b. Henceforth I shall refer
to this source as B. 47, after its registration in Maurice J. E.
Brown's inventory of Schubert's dance autographs. 3
Although Rieter agreed to Brahms's proposal, he met op-
position from a Viennese competitor, C. A. Spina. The latter
publisher eventually triumphed and in the summer of 1864 re-
leased the dances under the title 12 handler (componirt im Jahre
1823) fur das Pianoforte von Franz Schubert. Op. 171. Nachgelasse-
4
nes Wenk. Brahms served — —
anonymously as editor. His ef-
forts did not go unnoticed, however; the Allgemeine musikalische
Zeitung not only disclosed that it was Brahms who saved the
manuscript from oblivion, but also made a pointed reference to
his editing, which, the journal reported, "was scrupulously
limited to [producing] an accurate copy of the manuscript."
That Brahms respected the integrity of B. 47 is worthy of
32 SCHUBERT
note. For most, if not all previous editions were compilations

of dances drawn from an assortment of manuscripts. 6 Brahms

was cognizant of this earlier practice and did not object to it in
principle; on the contrary, in 1869 he compiled just such a
collection himself, issued by J. P. Gotthard as 20 handler jur
7
Pianoforte . von Franz Schubert (Nachgelassenes Werk). Why,
. .

then, did he take the extraordinary step five years earlier of

basing an edition on a single intact manuscript? The answer,
not surprisingly, reflects an apparent accord between the mu-
sic in B. 47 and the editor's own aesthetic sensibilities. In brief,
Brahms seems to have judged the dances in the manuscript
small-scale masterpieces, which when played together make up
a satisfyingwhole.
Rather than address the merits and flaws of Brahms's edi-
tion (the latter of which are surprisingly numerous), here I
shall be concerned with the dances themselves with the crea-—
tive work of Schubert, not the redactional work of Brahms. Yet
from this examination, it is hoped, will come an accounting
for both the editor's high estimation of the Landler and his de-
cision to print them together.

Maurice Brown described Schubert's dance autographs as

"journals" or "notebooks" in which the composer was able to
try out— often during casual evenings spent improvising the at
piano — new ideas and techniques might be developed
that later
8
in more substantial pieces. Dance as sketch: it is a fascinating
concept, but one whose outlines Brown himself only adum-
brated. This is not the place to elaborate his idea, or to dispute
it. The subject arises here only because Brown's solitary example

concerns D. 790, no. 6, which he paired with the scherzo of the

"Death and the Maiden" String Quartet. To be sure, the dance,
written in May 1823, and the scherzo, composed ten months
later, are in very different keys (G-sharp minor and minor). D
Yet long stretches of the pieces are otherwise identical (cf. bars
1-8 and 9-16 of the dance, and bars 9-16 and 23-30 of the
quartet).
Brown probably did not intend the term "sketch" to be
 —
Dance Music as High Art 33

taken literally, for, as he must have seen, the Landler is a fully

achieved work. In it, for example, Schubert skillfully restricts
the statement (but not the strong implication) of tonic har-
mony, and throughout the piece he employs poignant, fre-
quently unprepared dissonances. Thus the opening chords,
iv-V— V/iv-iv, are colored respectively by a 7-6 suspension,
a 4-3 suspension, a minor ninth, and a double appoggiatura
(ex. 1 a). In the analogous passage in the second part of the
dance, this delicate material is transformed to display a con-
trary character (ex. ib). Most obvious is Schubert's replacing
of the quiet dynamics, legato touch, and restricted compass
with loud dynamics, detached mode of performance, and wide
range. More subtle is his reworking of earlier motivic materi-
als. The "circling" motive, appearing at the outset in the inner

parts and set sharply there against the bass line, recurs in varied
form in part 2, where it appears in the lowest voice and actually
incorporates all but the first note of the old bass line (cf. the
lower analytical staves in exx. ia and b). Similarly, the melody
and remaining voice in the right-hand part of bars 1-4 re-
appear in a different disposition in the later passage. Strikingly
enough, this variation is accomplished by invertible counter-
point at the octave (cf the upper analytical staves in ex. 1).
Likewise, the harmony is redefined. The ninth bar corresponds
to the first but is a tonic J chord, not a first-inversion subdomi-
nant chord. Here the Dji, which was a suspended dissonance in
bar 1, is a harmonic tone, and the O, once the note of resolu-
tion, is a neighbor note. By the same token, bar 12, correspond-
ing to the fourth bar, unfolds not a root-position IV chord, but
an applied dominant chord that functions sequentially.
—
Even the form of the dance unique, so far as I know
demonstrates Schubert's studied use of dramatic reinterpreta-
tion. Part 1 comprises two four-measure units (a + b = A),
each o£ which is doubled in length in part 2 to produce a
sixteen-bar period (a' + b' = A'). Most of Schubert's dances,
by contrast, are in more usual binary or rounded binary forms
and present melodies set above simple accompaniments in
stereotypical dance rhythms. In several respects, therefore, the
dance at hand can be looked upon as representing an "advance"
in Schubert's style. It is more daring even than the scherzo later
 — . 1

34 SCHUBERT
Example 1 (a) D. 790, no. 6, bars 1 -4; (b) D. 790, no. 6, bars 9-12; (c)

D-Minor String Quartet (D. 810), Scherzo, bars 1-12.

I tB£
7-6
m r
6

2 b
7 - 6

ruiVLiirvfi
i
^ 1

(to

i
Jl X
2-3 3 2 - 3

tea nrJ Aj
yyihif- p
A / A A
S§ r~* «

P 3 *^i
9*= it *?'/

HH W 33:

L
CO s t £:
-#-'
m p-i 0-
-f- f- W
» 1
# -»
s* L 9 — — B
'

-P*

e
)
"M4
V $—9
4 3
™ -*— f~
>5 r J*
d: i % i

^* ^ m
JUL ±% £
^^
,?> i j J
rt
6
y =*
5
§ ^ 1*
s^-
4
 Dance Music as High Art 35

modeled on Although the remarkable opening mea-

it (ex. ic).
sures of the piano piece were taken over into bars 9-12 of the
quartet, they are preceded there by a passage that establishes
the tonic in the most patent way and thereby robs the borrowed
material of much of its original mystery. The dance, in short,
is a miniature masterpiece wherein Schubert's ambitions were

consummately realized.
The two parts of the eleventh Landler are also related as
statement and varied restatement. Here, however, Schubert's
revision is more extreme, cutting to the harmonic core: chords
that arose as products of contrapuntal elaboration in the first
part reappear beneath melodic structural tones in the second.
Example 2 illustrates this rather complex state of affairs. In the
original statement of the material, the nonstructural etj" and
d\\" are supported by the dyads c'-etT and bti'-dl}', which

merely act as neighbor chords to the structural harmonies (ex.

2a). Both these tones and their supporting dyads recur in
analogous positions within the restatement, but now the me-
lodic tones are structural and the dyads components in applied
harmonies (ex. 2b).
The transformation of nonstructural tones into structural
ones is but part of the tale of Schubert's revision. Actually, the
dissimilar surfaces of the melodies in bars 1-4 and 9-12 result
from different unfoldings of the same set of tones (el>"-eti"-
P'-dlT-eJ'"). Here again the notes et|" and dtf are conspicuous.
Originally chromatic neighbor notes, in part 2 they first occur
as consonances (bars 9 and 11); yet in bars 10 and 12 they revert
to their original form, becoming dissonant and, as before,
leading respectively to f" and e^".

Example 2 D. 790, no. 11 , harmonic sketch: (a) bars 1-4; (b) bars 9-12,
«x) ' (b>

p
, 1 ,

l=*d
,l

Vg pj f^ {
,

f-^
4 i
*®jMm m r r
'^
y pcdai czr/szE) (nynr)
36 SCHUBERT
Example 3 D. 790, no. 11, motivic-rhythmic analysis.

,
b

$m^fttmnm yfW fj&m

wi r H ^^m §1 W^-Wm^^r^
m -h
i'i
i
1
ii
i 1

r r

Of equal importance to this imaginative application of

what might be called "structural variation" is Schubert's use of
what Schoenberg termed "developing variation," whereby a
theme is built by subjecting a small number of motives to con-
tinual modification. In the first phrase each of the note pairs
elf-f" and d\\"-e\>" coheres in a trochaic motive that becomes
the basis of subsequent material (ex. 3, motive a). The figure
recurs in varied form in bars 5-8 (b), where the melodic direc-
tion is reversed; now it embraces three different tones and,
most important, begins on the weak beat and is iambic. The
motivic evolution reaches its final stage in the second half of
the dance, nearly all of whose melody derives from the two
versions of the figure in the first period (c).
One new development is the acciaccatura that precedes
each statement, another the slurring of the three notes; both
contribute to the impression that the downbeat in the right
hand has shifted to the middle of the measure a displacement —
first hinted at when the motive began on the second beat in bar

5. Significantly, this melody binds salient features of both ear-

lier versions of the motive: the rhythm is trochaic (as in bars 1

and 3), but the pattern begins on the middle beat (as in bars
5-8). The accompaniment, however, persists in presenting the
 Dance Music as High Art 37

meter directly, and right-hand and left-hand parts are thus out
of phase, a touch that Brahms would surely have savored.

The charm and graciousness of numbers 6 and 1 1 are very

much in keeping with Schubert's style. What is not typical
about them is also what is most striking: their complexity and
refinement of detail. Why this qualitative difference between
the Twelve Landler and other collections? Probably it derives
from a dissimilarity in compositional circumstances. By and
large Schubert's dances —
more than 400 in number were oc- —
casional pieces, created spontaneously at parties for the enjoy-
ment of friends. In the words of Josef von Spaun, Schubert
"from time to time . surprised us dancing enthusiasts with
. .

the most beautiful Deutsche Tanze and Ecossaises, which were

the fashion at that time." And from Leopold von Sonnleithner
we learn that the composer "was always willing to sit down at
the piano, where for hours he improvised the most beautiful
waltzes; those he liked he repeated, in order to remember them
and write them out afterwards." 9
The by-products of these evenings were hundreds of en-
gaging "written-out improvisations." But it is difficult to be-
lieve that the pieces in B. 47 are fruits of that kind. In the first
place, the subtle manner of the dances implies that Schubert
spared no time and none of his abilities. Moreover, the state
of the autograph suggests that the dances were more soberly
conceived and executed than those in other manuscripts. If,
as seems probable, one sign of the casual origin of many of
Schubert's dance autographs is their lack of title, date, and sig-
nature, then the inscriptions at the head of B. 47 ("Deutsches
Tempo[.] Mai 1823 [.] Frz. Schubertmpia"), before its first
dance (',' Pianoforte"), and after its last dance ("Fine") might
indicate a more formal origin. At all events, the layout of the
manuscript — a bifolium with a single inserted leaf, all of which
—
is filled implies that Schubert knew beforehand how much
space he would be likely to need; it shows, that is, a measure of
planning.
This planning is also reflected in the order of the dances,
 b

Length
(bars)
(NX(N(NXOOOOOO(N(N(N1-

6
<
u
O
tin 3

Scheme

^\ .,_ H W^ h 00 so 00 00 Wrf
*> • • 0000 *
•
• • _|_ r-H .. . . 00 t-h

^ <<<<<<<<<<<
^hOOOOOOOOOOOOOOOOOOOOOO

iii/III i I i III III I

<
C/5

U PQ
a i/I vi VI vi I I VI

c
e2

Q ~ > ~ ~ |
>

&
M Q<QQ5-^ rt fflcq^w

bo
c
^(NcOTtmvor-oooNO^cN O S
 Dance Music as High Art 39

which is by no means arbitrary (table 1). Schubert flanked

numbers 2— II, each of which has the orthodox length of
thirty-two or forty-eight bars, with two formally anomalous
dances. Discussion of the last dance will be put off until later,
but at this point we might briefly consider the opening number.
—
Because of its comparatively large dimensions at seventy-two
bars it is the second most extensive dance in Schubert's oeuvre —
as well as the unusual phrase lengths of its second period (ten
bars + ten bars), this dance does much to establish the com-
poser's high aspirations. But the long-range role of the dance
becomes apparent only as the succeeding ones unfold. When
we ultimately learn that pieces with more common and ordi-
nary dimensions typify the collection, we understand in retro-
spect that the first number stands apart, at least with respect to
form. In a sense, then, this piece, aptly described elsewhere
as an "Invitation to the Dance," sets the stage for the main
event — a series of pieces in more modest dance forms.
10

In view of this hierarchic principle, it is not suprising that

B. 47 displays a high degree of tonal coherence. Numbers 1-4
form a rounded tonal group in D, numbers 5- 11 a group
focused on the keys of B and A-flat (see table i). 11 The first
dance tonicizes the dominant at the end of the first period, thus
anticipating the key of number 2. This relationship is, of course,
common enough; more unusual, and therefore unexpected, is
the tonicization of the submediant G-sharp minor in number
10, which prefigures the key of the eleventh dance (in A-flat)
and underscores the importance of the axis between B and
A-flat /G-sharp. The fifth dance, significantly, couples the two
groups. The tonic of its first period, B minor, is the relative
key of the preceding dance; the tonic of its last period, B major,
is the relative key of the following one. This number also

strengthens the tie between the tonal groups in another way:

the mediant (D) —the key of numbers 1,3, and 4 —
is tonicized

at the end of part 1.

The first eleven dances display motivic coherence as well.

Numbers 3 and 4, for example, constitute an "organic" pair-
ing. Both are formed out of two eight-measure periods, each
of which culminates in a tonic cadence, and both contain promi-
nent chromatic sequences and applied diminished-seventh
chords. In addition to these shared traits —
which are admit-
40 SCHUBERT
Example 4 (a) D. 790, no. 3, bars 15-16; (b) D. jgo, no. 4, bars 1-2,

tedly stylistic commonplaces — there is a more significant rela-

tion. The melody the outset of number 4 is a retrograde ver-
at
sion of that ending number 3, the last three notes of the one
dance being turned around to form the first three of the other
(ex. 4). The eighth-note motive in the penultimate measure of
the third number is similarly echoed in the second measure
of number 4. The later dance thus seems to grow directly out
of its predecessor.
On the matter of motivic relationships we might let the
editor himself speak, through a remark made in a letter to
Adolf Schubring in June 1865. Apparently, Schubring had
asked whether Brahms knew that the second parts of D. 790,
no. 8, and op. 33, no. 10, were identical but for their keys
(A-flat minor and A minor). Brahms exclaimed: "Something
by Schubert to have escaped me! And that I should not have
known the beautiful A-minor waltz in the German dances [op.
33, no. 10]! But this new first part [in my edition] is exactly
12
what isof interest."
What was meant by this remark? For one thing, given the
predilection for motivic connections demonstrated by Brahms
the composer, we can presume that as editor he found gratify-
ing the relation between the patterns of structural tones in the
first part of the eighth dance in B. 47 and its immediate prede-
cessor (ex. 5). (The structural tones in op. 33, nos. 9 and 10,
significantly, bear no relation to one another.) Brahms would
also have appreciated the tonal bond between the seventh and
eighth dances in B. 47 (A-flat /A-flat minor). In other words,
he probably preferred the A-flat-minor version of the dance
largely because of its tonal and motivic echoes of the dance that
immediately precedes it.
 Dance Music as High Art 41

Example 5 Melodic-structural tones: (a) D. jgo, no. 7, bars 1-4; (b)

D. jgo, no. 8, bars 1-4.
(a) (b)

^,, ^ 1 m in

Example 6 (a) D. jgo }

no. 8, bars 1-4; (b) D. 790, no. 8, bars 17-18;
(c) op. 33, no. 10, bars 1-4; (d) op. 33, no. 10, bars 17-18.

But the editor would have had additional reasons for his
choice. The version in B. 47 is superior to the earlier published
version both in its dissonance treatment and its internal motivic
connections. Prominent neighbor notes, which characterize
the melodies in both versions, are given expressive harmonic
support 'only in B. 47 (ex. 6a and c). Thus in the A-flat-minor
setting dfci" is harmonized with a French-sixth chord, and glj"
forms part of a diminished-seventh chord sounded over a tonal
pedal point; by contrast, in op. 33 similar melodic dissonances
merely clash against tonic and dominant chords.
A closer relation between sections is also noticeable in
B. 47. The French-sixth chord, heard in the opening phrase as
 42 SCHUBERT
a neighbor chord, recurs in part 2, now as a true chromatic
preparation for the dominant (bars 11 and 15). Furthermore, in
B. 47 (but not in op. 33) the first phrase of the dance is recast to
form the final phrase (cf. ex. 6a-b with ex. 6c- d). In other
words, the two parts of D. 790, no. 8, are comparatively more
cohesive than those of op. 33, no. 10. In B. 47 Schubert maxi-
mized coherence at once within the dance and between it and
its neighbor.

To point out the various tonal and motivic connections

between dances is not necessarily to demonstrate the "unity"
of the collection. Although the dances follow one another
agreeably, they do not imply a specific conclusion; one piece,
or ten pieces, or some other number, could conceivably follow
the eleventh dance. Accordingly, if the last dance, number 12,
secures closure, it does so not by realizing any long-range im-
plication, but by being saturated, so to speak, with numerous
features having closural force. Put in different terms, should
the final dance be markedly closed, it would bring the entire
13
set to a satisfactory conclusion.
The anomalous tonal center of number 12 (E) obviously
checks the prevailing implication of numbers 5-1 1 that the key
of each dance will be B or A-flat (or the parallel mode of either
key). Yet that says nothing about the direction that the cycle
might take. This new course soon becomes apparent, however.
Unlike the other dances, number 12 begins with an archetypal
cadence, thus signaling not continuation but closure (ex. 7a);
and as the dance progresses the impression of having reached
the end of the series is reinforced. The opening cadence is, ex-
traordinarily, the source of almost all subsequent material. In

Example 7 D. 790, no. 12: (a) bars 1-2; (b) bars 19-22.
(«

pi i
m
..

i
m
^
mmmmmm (b)

ir
r
«) - 3
**--
*Xl
 Dance Music as High Art 43

part repeated in bars 3-4, varied slightly in bars 5-6, and

1 it is

sounded again in bars 7-8; the material in part 2 is similar. The

unusual disjunct melody of the dance also contributes substan-
tially to the growing sense of finality. Instead of a real tune,
Schubert merely provides an arpeggiated animation of the upper
three voices of the cadential harmonies. This melody thus has
no long-range goals; its motion is only local and is played out
against the static background of the reiterated V-I cadences.
As a concomitant of this integrated melodic-harmonic
pattern, the tonal vocabulary of the dance is unusually diatonic.
This simplified stock of pitches also suggests impending ter-
mination, as do two other "natural" signs of closure: the quiet
dynamics (pp) and uniformly thin, uncomplicated texture.
Closure, finally, is enhanced —
settled, actually —
in the last
four measures, wherein the original form of the opening ca-
dence twice recurs (ex. 7b). The loud return of material signals
completion; its quiet echo, rewritten to allow of a tonic chord
on the final downbeat, clinches it.
Still, Schubert's decision to end in E remains something of

a mystery. Of course, the composer could have achieved long-

range closure by ending in the key of the opening number (D).
In any event, the concatenation of closural features in number
12 leaves little doubt that the placement of it last in the manu-
script was a calculated act, one intended in its own way to im-
part closure to the entire collection.

///

In spite of the evident relations among the dances, it might be

argued that B. 47 displays in essence a paratactic, or additive,
structure —that is, that its coherence does not depend upon the
14
sequential arrangement of its components. Each of the dances
is an integral whole, and though the opening and closing num-

bers have features appropriate to their positions —

number 1 is a
preambule, number 12 provides a satisfactory conclusion the —
dances could be interchanged or even taken out of context alto-
gether without risking incomprehensibility. Nevertheless, the
ten inner dances show traces of a sequential structure, compris-
ing linked tonal groups and motivically related pairs. Thus to
interchange dances would be to do violence to the characteris-
 —
44 SCHUBERT
tics of the set most attractive to a mind attuned to "organicism"
in art.
Brahms had such a mind. And thus, itwould seem, he
had no will to violate the integrity of his source —
no reason to
omit, transpose, or reorder any of the dances, or to interpolate
extrinsic ones, all of which earlier editors had done as a matter
of course. Perhaps we can now understand why Brahms de-
scribed the set as twelve dances "in rank and file on a leaf, with
quite the loveliest faces." The subtle style of the pieces must
have prompted the second half of this description. Brahms's
striking metaphor of military order, by contrast, was probably
induced by the "framing" quality of the outer dances, as well
as by the tonal and motivic relations existing among the inner
ones. For though the Landler do not form a unified cycle of
character pieces in the manner, let us say, of Schumann's Car-
naval, they are nonetheless distinguished miniatures displaying
many aesthetically significant interconnections.
our attention again to Schubert and to
Finally, let us turn
the eleventh dance. As Nicholas Temperley has observed, this
piece begins with a striking prolongation of the dominant,
9
comprising the voice leading \~_ 7 over a dominant pedal (i.e.,
a "tonic *" without the tonic note, followed by a dominant-
ninth chord without a third or fifth; see ex. 2). Temperley men-
tions the dance only in passing, as but one example of Schubert's
use of the unusual ® chord, which he attributes to a deep
influence upon the composer of Beethoven's Seventh Sym-
phony. Nonetheless, Temperley 's pointed observation that in
the Landler the Beethovenian chord appears in a "progression
. .intensified to a point of anguish, both by the use of a 7 and
.

by chromaticism," occasions us to take pause. 13 "Anguish"

in a dance?
One of the saddest facts of Schubert's life his contracting —
—
of syphilis late in 1822 may explain this paradox and, at all
events, dispels any notion that the dances might be the prod-
16
ucts of an evening's entertainment. During the month in
which the Landler were written, May 1823, the composer
suffered the throes of an early exacerbated phase of his fatal
disease. It is improbable, accordingly, that Schubert made all
his usual social rounds at this time; indeed, at least part of this
month was spent in the hospital. Very likely, then, the dances
 Dance Music as High Art 45

in B. 47 were not produced casually in the cheerful company

of friends, but were, on the contrary, more "normal" artis-
tic products, thoughtfully written when the composer was,
in more than onerespect, very much alone. Whatever hope
Schubert might have held earlier for a return to good health
seems to have been dashed by his outbreak of symptoms in the
spring of 1823, and on 8 May he composed "Mein Gebet," a
17
feverish poem replete with images of inescapable doom. If
the excruciating tone of Schubert's verse is not matched in each
of the Twelve Landler, this contemporaneous set of dances
nevertheless penetrates feelings seldom hinted at in other col-
lections.Whence, in part, the title of this essay: D. 790 is com-
posed not o( lighthearted improvisations set down on paper,
but rather of deeply felt works of high art.

Notes
1. "Ich konnte Ihnen eine schone Sammlung 'Walzer' geben (50 fl.
erbate ich mir, die ich fur die Manuskripte gab). Vor allem stehen 12 'Walzer'
auf einem Blatt in Reih und Glied, die ganz allerliebste Gesichter haben."
Johannes Brahms, Briefwechsel, vol. 14, ed. Wilhelm Altmann (Berlin:
Deutsche Brahms-Gesellschaft, 1920), p. 82. Brahms's activity as an editor
of Schubert's dances is discussed in my Ph.D. dissertation, "Brahms as Edi-
tor and Composer: His Two Editions of Landler by Schubert and His First
Two Cycles of Waltzes, Opera 39 and 52" (University of Pennsylvania, 1984),
chapters 1, 2, 4, and 5. Parts of the present essay have been adapted from that
study.
2. Brahms left this source, along with most of the rest of his collection

of autographs, to the archive of the Gesellschaft der Musikfreunde (catalogue

no. A 262).
3. Maurice J. E. Brown, "The Dance-Music Manuscripts," in his Es-
says on Schubert (New
York: St. Martin's Press, 1966), p. 238.
4. Brahms's correspondence with Rieter from the winter of 1863—64
(Briefwechsel 14:81-87) depicts the composer as an intermediary between the
two publishers. Although Spina, who at that time claimed the rights to all
Schubert^s music, wrested control of the dances, Rieter and Brahms later col-
laborated in the first editions of the piano score of the Mass in E-flat (D. 950,
1865), and the Drei Clavier- Stiicke (D. 946, 1868).
5. "Seine [Brahms's] Redactionsarbeit beschrankte sich gewissenhaft

auf eine getreu Abschrift des Manuscripts." "Zwei Schubert-Novitaten aus

Spina's Verlag," Allgemeine musikalische Zeitung 2 (1864), col. 872.
6. See Brown, "Dance-Music Manuscripts," p. 219. Schubert's role in

his own editions is uncertain. In preparation for the first published set, which
A. Diabelli and Co. released in late 1821 as the Original Tanze, op. 9, the
46 SCHUBERT
composer set down on the back of the arietta La
the incipits of nine dances
and wrote out the melodies of six others on
pastorella al prato (D. 528, B. 20),
the back of the part song Ruhe (D. 635, B. 31). But only some of the fifteen
dances in these "memoranda" were printed, and most of those in a different
version. By the same token, many of the dances in the autographs containing
complete pieces found in op. 9 were held back from the edition. Some ap-
peared in 1823 in Diabelli's next set, op. 18, but others were left unpublished
in Schubert's lifetime altogether.
Op. 18 occasioned a dispute between Schubert and Diabelli, which led
the composer to seek other publishers. Thus the next edition, op. 33, ap-
peared in 1825 with Cappi and Co. The change in publishers did not alter
editorial policy, however; the dances in this set were likewise drawn from
many sources, including, as we have seen, B. 47. Although Schubert might
have supervised the organization of opp. 9, 18, and 33, it is doubtful that he
played any role whatsoever in the next three sets, opp. 49, 50, and 67; for
these appeared in 1825 and 1826 with Diabelli, with whom Schubert no
longer had any dealings. Since the only dated source for these prints (B. 45)
was written in February 1823, just before Schubert left Diabelli, it is reason-
able to assume that the manuscripts upon which the publisher based his edi-
tions were acquired in earlier years, and that in all likelihood Schubert had
nothing to do with the prints.
Much less can be inferred about opp. 77 and 91, published by Has-
linger in 1827 and 1828; no handwritten sources or any other documents that
might shed light on the question of Schubert's involvement have been pre-
served. For a somewhat elliptical discussion of Schubert's approach to group-
ing dances, made confusing by its failure always to distinguish between the
composer's and his publishers' sets, see Paul Mies, "Der zyklische Charakter
der Klaviertanze bei Franz Schubert," in Bericht iiber den Intemationalen musik-
wissenschajilichen Kongress, Wien, Mozartjahr 1956, ed. Erich Schenk (Graz:
H. Bohlaus Nachf, 1958), pp. 408-11; Mies's conclusion that the editions
published in Schubert's lifetime were not coherent cycles, it seems to me, is
surely correct.
7. On the Twenty Landler, see Brodbeck, "Brahms as Editor and
Composer," chapters 4 and 5, and idem, "Brahms's Edition of Twenty
Schubert Landler: An Essay in Criticism," in Brahms Studies: Papers Delivered
at the International Brahms Conference, The Library of Congress, Washington,
D.C., 5-8 May ig8j, ed. George S. Bozarth (London: Oxford University
Press, forthcoming).
8. Maurice J. E. Brown, Schubert: A Critical Biography (New York: St.

Martin's Press, 1958), p. 230; and idem, "Dance-Music Manuscripts," p. 218.

9. Otto Erich Deutsch, Schubert: Memoirs by His Friends, trans. Rosa-
mond Ley and John Nowell (New York: Macmillan, 1958), pp. 133 and 121.
Alfred Einstein, Schubert: A Musical Portrait (195 1; repr. New York:
10.
Da Capo Press, 1981), p. 216. Schumann later did something similar in Car-
naval, whose lengthy first number —
entitled "Preambule" —
is followed by a

series of shorter character pieces.

11. The final Landler, in E, is not a member of the two interlocking
 . V "

Dance Music as High Art 47

tonal groups. This move to a "foreign" key at the end may, as we shall see,
have significance for our understanding of the set as a whole.
12. "Mir soil bei Schubert etwas entgangen sein! Und ich sollte den
schonen a moll-Walzer in den Deutschen Tanzen nicht wissen! Aber dieser
neue erste Teil ist gerade deshalb interessant." Briefwechsel, vol. 8, ed. Max
Kalbeck (Berlin: Deutsche Brahms-Gesellschaft, 191 5), pp. 206-7. For a
clarification of Kalbeck's erroneous editorial remarks concerning this pas-
sage, see Brodbeck, "Brahms as Editor and Composer," p. 91 (n. 26).
13. As Barbara Herrnstein Smith has observed, many poems are
brought to an end in a comparable fashion, with a final stanza that makes
reference "to any of the 'natural' stopping places of our lives and experi-
ences— sleep, death, winter, and so forth." Such allusions "tend to give
closural force when they appear as terminal features in a poem" (Poetic Clo-
sure: A Study of How Poems End [Chicago: University of Chicago Press,
1968], p. 102). The musical analogues to such images are "closural signs,"
which can involve tonal harmony and tonal pitch (e.g., the harmonic pro-
7
gression Ij— — I, and the melodic motion 3 -2-1), as well as elements like
registral pitch, dynamics, and tempo (e.g., descending lines, decreasing dy-
namics, and slackening tempo). Leonard B. Meyer has termed the former
"syntactic" signs, the latter "natural" signs; see "Toward a Theory of Style,"
in A Concept of Style, ed. Berel Lang (Philadelphia: University of Pennsyl-
vania Press, 1979), pp. 3-44; see also Robert George Hopkins, "Secondary
Parameters and Closure in the Symphonies of Gustav Mahler" (Ph.D. disser-
tation, University of Pennsylvania, 1983), pp. 1-104.
14. For discussion of paratactic and sequential poetic structures, see
Smith, Poetic Closure, pp. 98-139.
1 5 Nicholas Temperley, "Schubert and Beethoven's Eight-Six Chord,
lQth-Century Music 5 (198 1) 149.
:

16. The most thorough account of Schubert's disease is Eric Sams,

"Schubert's Illness Re-examined," Musical Times 121 (1980): 15-22. Edward
T. Cone has drawn a fascinating (and surely provocative) connection between
the tonal structure of the Moment musical, op. 94, no. 6, and the emotional
side effects of the disease ("Schubert's Promissory Note: An Exercise in Mu-
sical Hermeneutics," included in this volume). I hasten to add that nothing
like Cone's specific, extrageneric analysis is intended here.
17. Schubert's poem can be found in O. E. Deutsch, The Schubert
Reader, trans. Eric Blom (New York: Norton, 1947), P- 362.
 A Romantic Detail in Schubert's Schwanengesang

JOSEPH KERMAN

Ihr Bild is one of two Schubert songs to which Heinrich Schen-

ker devoted an entire essay. Schenker was evidently fascinated
by the gigantic simplicity of the piece, of the first two notes:
Two measures serve as introduction:

.. |
ai rAi

Certainly one should not already discern a motif here; so

the question arises: what other goal do these notes have to
fulfill? Do they simply lead into the tonality, or perhaps
prefigure the opening note in the voice, or both? However
this may be, and in any case, one has to ask further why
the Master sounded the same note twice, when it would
have been perfectly possible to have simply held it through
two bars. As a matter of fact, it is only the answer to this
question that brings us to the solution of the puzzle. To
repeat each note in slow tempo, and what is more to re-
peat it in this manner after a rest, amounts to "staring" at
it, as it were; and in doing this, we feel ourselves wonder-

fully transported to the side of the unhappy lover, who

stands "in dark dreams" staring at the picture of his be-
loved. With him, we too stare at the picture. A simple de-
vice, is it not, this placement of a repetition, separated by
a rest, for a note held through two bars. Yet it takes a
 A Romantic Detail in Schubert's "Schwanengesang" 49

genius — it is only granted to a genius — to grasp the dif-

ference between such possibilities, as also (above all) to
plunge himself into the very midst of the spiritual experi-
ence with a force that can yield a device of this kind. Thus
with his very -first stroke Schubert shows himself the true
magician who binds a secret thread around the exterior
situation (here, the staring at a picture), around the soul of
the unhappy lover, and around us —
a thread that invests
the experience with an eternal prospect full of ever-new
l
immediacy, over and above the single manifestation.
Schenker's leap from the two B^s to the notion of "staring" is a
breathtaking one, which many readers will find hard to follow,
even if they consider possible cognate meanings in starren,
stare, stern, and stark. Nevertheless, the fanciful paragraph was
worth writing, and it is worth reading both for what it sug-
gests about Schubert and for what it tells about its author.
When a musical analyst senses that his regular tools will work
no further, it is better for him to stammer out some of his in-
stinctive response to the music, rather than to decree that such
talk is "literally meaningless" and leave the passage unglossed.
The two B^s are remarkable and evocative; clearly Schenker
was on to something. Clearly, he did not get to the bottom of
it. Perhaps this is a case where the analyst's insight can be helped

along by comparative information, from outside the single

piece itself.

More is involved, certainly, than establishment of the

tonic.That function we might call the lowest common de-
nominator of the piano introduction in all but a handful of
Schubert's songs.
2
In his later work — let us say the 250-odd
songs written after 18 19 —
Schubert almost invariably estab-
lished and articulated the tonic by means of a full anticipatory
phrase. What is anticipated is the piano figuration to come, and
something of the initial melodic outline, and in many cases also
some later events in the song, generally of a harmonic nature.
This norm includes all but one of the Winterreise songs, for in-
stance. But it does not stretch very comfortably to include Ihr
Bild, nor several of its companions among the songs of Schu-
bert's last year.
Earlier, Schubert had often done without a piano intro-
 —
50 SCHUBERT
duction altogether — an economy that he lived to avoid stu-
diously, and to regret; after 1821, on the few occasions when
he published an old song lacking a piano introduction, Schubert
would compose one anew. 3 Or sometimes he had written a
small stretch of piano figuration all on the tonic, amounting to
something less than a phrase. (Gretchen am Spinnrade is the fa-
mous example; the only late ones are Willkommen und Abschied
[D. 767], Wohin? [from Die schone Mullerin], Der blinde Knabe
[D. 833], and Hippolits Lied [D. 890].) Or else he had merely
struck a tonic chord or two. This seems a poor excuse for an
introduction, though occasionally it may be granted some po-
etic justification in the suggestion of chords strummed by way
of intonation, before the harpist embarks upon his lay. Is Ihr
Bild some kind of sophisticated throwback to this type of in-
troduction —
which indeed "simply leads into the tonality"?
—
Among the mature songs are a few a very few instances of —
such throwbacks (see ex. 1). Here one would speak just barely
of figuration anticipated, or at least of motion begun; and one
would have to reckon on the air of assumed simplicity that in-
forms the entire Schone Mullerin cycle. Nothing of the kind
characterizes Ihr Bild.
A distant but more revealing precedent for Ihr Bild may be
found in Gondelfahrer (D. 808), an unassuming but deft little
song written in March 1824. The piece is not well known it —
does not appear in the seven volumes of the Peters edition
but it has its modest anecdotal fame as the last song set to
words by Mayrhofer. During 1818-20, when they roomed to-
gether, and even before, the gloomy poet had played a major
role in Schubert's development as an artist; now Schubert was
drawing away.
Asalways, this piano introduction (ex. 2) establishes the
tonic — but witha little more than a chord, and less than a
phrase. A single harmony moves to the tonic, and the progres-

Example 1 (a) Trockne Blumen; (b) Der Miiller und der Bach

(a) *> ,

p p t- f1 4 4
 A Romantic Detail in Schubert's
"
'Schwanengesang" 51

Example 2 Gondelfahrer

atB i f
t-
Y aa

sion is repeated exactly (a pattern that may be diagrammed t —

x-T — x-f). Technically, "x" may be thought of as a type of
plagal cadence, or as an appoggiatura chord, or as a stressed
auxiliary chord, resolving to the tonic. The heart of the matter,
however, seems to be the sensuous quality of the particular
progression. The concept is fully Romantic, the intent "po-
etic": indeed, Schubert's impetus was not purely musical but
illustrative in a very simple way. The poem runs as follows
(italics mine):

Es tanzen Mond und Sterne

Den fliicht'gen Geisterreih'n;
Wer wird von Erdensorgen
Befangen immer sein!
Du kannst in Mondesstrahlen
Nun, meine Barke, wallen,
Und aller Schranken los
Wiegt dich des Meeres Schooss.
Vom Markusthurme tonte
Der Spruch der Mitternacht;
Sie schlummern friedlich Alle,
Und nur der Schiffer wacht.
[The moon and the stars are dancing the fleeting round
dance of the spirits. Who will ever be possessed by
earthly cares?
My boat,
you can now float in moonbeams; and free of
allbonds, you will be cradled by the lap of the sea.
From the tower of St. Mark's chimed the knell of mid-
night. Everyone is sleeping peacefully, and only the
boatsman is wakeful. 1
 — —

52 SCHUBERT
After the introduction Schubert avoided A\> (or anything at all
on the flat side) up to the ninth line, where he abruptly brought

in A-flat triads again, ppp and arpeggiando, dutifully echoing

out twelve times to mark the midnight hour. In the autograph,
4
Schubert actually numbered the A^s from i to 12. This is a
Venice quite in E. T. A. Hoffmann's spirit, with deep, veiled,
Romantic chimes sounding across the canals to Mayrhofer's
meditative gondolier. The same device, in the same key, but
without the forecasting introduction, occurs in Schubert's con-
temporary setting of the poem for male quartet (D. 809).
For four years and nearly a hundred songs, Schubert did
not write another piano introduction according to this pattern.
In 1828, however, he returned to the idea with a much broader
insight into its expressive applicability. The Heine song re-
called most by Gondelfahrer is Am Meer (ex. 3). The
directly
key is the same, the — —
scheme (x t x f ) very similar, with
a dissonant augmented-sixth chord at "x" richer than the l>VI
of the earlier song. In mood, Schubert had refined the rather
obvious misterioso of the midnight bell to an unforgettable,
enigmatic solemnity, which seems to plumb infinite marine
and spiritual depths. What is most interesting, and most un-
usual, is the independence of the introduction here. It does not
signal ahead to a later event in the song, nor does it anticipate
figuration or melody. It simply recurs; the song begins and
ends with an oracle framing or glossing the poetic statement,
rather than playing in to it. From the Classical point of view,
the introduction is nonfunctional; it illuminates nothing. But
from the Romantic point of view it suggests everything
everything in the world that is inward, sentient, and arcane. It
might stand as a prototype for those "unconsummated sym-

Example 3 Am Meer
Sehr iandpam
X.
 '

A Romantic Detail in Schubert's "Schwanengesang" 53

Example 4 Die Stadt

P
t^assig

£p
gesckWmd

eon -pectak tm
4j&&

dim.
? ^
c\

>*iM}n' \

TO f
y I
f
7 mwm
bols" that Susanne Langer has urged us to comprehend in
music.
A third song of water, once again in C, develops the same
kind of introduction in a striking variant. Heine's Die Stadt be-
gins as in example 4. The scheme is t x f if Schubert— ;

meant the dissonant harmony to sound all the way through, his
pedaling indication leaves much to be desired. As will be sug-
gested in a moment, he would have appreciated the ambiguity
of the bare low C. "x," here a diminished-seventh chord,
strikes quite a different note of mystery, but again one that
"plunges into the very midst of the spiritual experience," as
Schenker puts it. Presently the grisly arpeggio is revealed as
illustration of the chill wind rocking the poet's boat, and it
rustles away with almost expressionistic fixity during the entire
center stanza of the song. The whole picture, of course, with
its gray waters, expresses the poet's frame of mind:

Am fernen Horizonte
Erscheint, wie ein Nebelbild,
Die Stadt mit ihren Tiirmen
In Abenddammrung gehullt.
Ein feuchter Windzug krauselt
Die graue Wasserbahn;
Mit traurigem Takte rudert
Der Schiffer in meinem Kahn . . .

[On the distant horizon, like a misty apparition, ap-

pears the towers enveloped in the dusk.
city, its
A dampgust of wind ruffles the gray surface of the
water; with mournful strokes the boatsman rows in my
skiff ... 1
54 SCHUBERT

Schubert in 1828 was no longer thinking like Goethe. But per-

haps Heine's verse woke in him a final intense response to that
impressive mystical incantation of Goethe's:

Seele des Menschen, wie gleichst du dem Wasser!

Schicksal des Menschen, wie gleichst du dem Wind!
[Soul of mankind, how you resemble the water! Destiny
of mankind, how you resemble the wind!]

The lines form the conclusion of the "Gesang der Geister uber
den Wassern," a poem that must have haunted Schubert, for he
set it as many as five times.
The piano introduction
to another Goethe setting, Geistes-
5
gruss (D. 142), evokes another, more amiable ghost, though
his harmonic vocabulary closely approximates that of Die Stadi.
These measures (ex. 5) show how far Schubert interested him-
self in this introduction pattern in 1828: the song dates from
181 5, but the measures in question were added when the piece
was rewritten for publication, in July 1828. As a matter of fact,
all the revisions on this occasion are extraordinarily instruc-

tive. Though few changes were made in the notes pure and
simple, Geistes-gruss was transformed as few works ever were
in their revisions. Originally the first half of the song had em-
ployed the barest recitative (ex. 6). Since he now found re-
citative stiff, Schubert rewrote it in tempo. To make it more
impressive, he slowed it down and punctuated it with broad,
contemplative rests. Since he cared less now for songs that

Example 5 Geistes-gruss

N ic\\X iu Ungfiam
 —
A Romantic Detail in Schubert's " Schwanengesang" 55

Example 6 Geistes-gruss (earlier version)

Rerit: i Hock auf demalt- tnlur-mtsUkt...

change brusquely in the middle, he specifically directed

style
that the second half (a sort of arioso) is to proceed "im ersten
Zeitmass." The most striking improvement came through the
addition of pictorial immediacy to the song by means of the
tremolo. As usual, Schubert seized on a single, central poetic
image to translate into music: water once again, water that car-
ries man's ship of fate, as in Die Stadt an image suggested so —
faintly by the poet that Schubert had scarcely noticed it in 181 5:

Hoch auf dem alten Turme steht

Des Helden edler Geist,
Der, wie das Schiff voriibergeht,
Es wohl zu fahren heisst.
war so stark,
"Sieh, diese Senne
Dies Herz so und wild, fest
Die Knochen voll von Rittermark,
Der Becher angefiillt;
Mein halbes Leben sturmt' ich fort,
Verdehnt' die Halft' in Ruh,
Und du, du Menschen-Schifflein dort,
Fahr' immer, immer zu!"
[High upon the old tower stands the hero's noble ghost;
as the ship passes by, he bids it a safe journey.
"Behold, these sinews were so strong, this heart so
firm and fierce, these bones full of knightly marrow, the
goblet always filled up.
For half my life I tossed like a storm; I let the other half
go by in peace. And you, you little shipload of humanity
there, travel ever onward!"]

And since in 1828 Schubert considered some kind of piano

introduction obligatory, he added three measures according
to his latest pattern (t —
x t). The opening auxiliary chord
 —
56 SCHUBERT
progression, and the quasi recitativo in tempo over the hushed
Romantic tremolo, both recall Am
Meer immediately.
There are two sides to the coin of harmonic experimenta-
tion in the early Romantic period. On the one hand, composers
concentrated on certain rich chords and interesting progres-
sions, deriving new expressive force by means of novel spacing,
texture, instrumental color, tempo, and context. This is mani-
fest in all the song introductions that we have discussed. On
the other hand, composers grew fascinated by possibilities of
ambiguity to be gained by stripping away harmony and tex-
ture: by dealing in paradoxical single notes, unison passages,
melodic fragments open to various harmonizations, parallel
and hollow sounds of all descriptions. This side of the coin,
also, shows up very clearly in the Schwanengesang\ the admirable
low C in Die Stadt is only one of many such experiments,
which date back to the Winterreise and even before. (One need
only think of Der Leiermann, Gefrorne Tranen, and Die Wetter-
fahne.) A Rellstab song from the Schwanengesang, In der Feme,
capitalizes on bare octaves in its piano introduction (d —x d;
ex. 7).
An introduction all on the dominant (d), incidentally, is a
great rarity in Schubert. Besides the present song, the only
other late example is the quite extraordinary Letzte Hoffnung,
from Winterreise. In In der Feme, the semitone step and the sug-
gestion of parallels prefigure inter alia the grating juxtaposition
of B-minor and B-flat-major root-position triads later in the
song, which so exacerbated a contemporary critic cited by
Richard Capell:

If such unseemliness, such insolently placed harmonic dis-

tortions, could find, in defiance of all common sense,

Example 7 In der Feme

Ziemlidi \&ngsa*n
 —
A Romantic Detail in Schubert's " Schwanengesang" 57

their impudent swindlers who would foist them as sur-

plus originality upon patient admirers of extravagance,
—
we should supposing that the egregious thing should
succeed —
soon be removed into the most blissful of all
states, a state of anarchy similar to that of the days of the
6
interregnum.

As often happens with music critics, this man has indulged his
tendency to overwrite so far as to obscure a tolerably good
point.
Stark octaves bring us back to Ihr Bild, whose introduc-
tion can now condensation of the pattern that
be regarded as a
we have been tracing, or as a skeleton of it, an exoskeleton.
Things become explicit only when the two B^s recur later in
the song, just before stanza 3 (ex. 8) —
a progression that re-
calls the beginning of Am Meer (but notice the empty fifth).
Schenker, who remarks that the staring is resumed at this
point, for some reason fails to relate the passage back to the
opening in technical terms. But retrospectively, at least, this
passage helps construe the duality of the notes at the start: by
now, the first Bl> is made to resolve into the other (x t). The
new passage is bodied out motivically as well as harmonically,
deriving its motif from an earlier vocal cadence echoed by a
piano interlude (ex. 9). This piano interlude on the dominant,

Example 8 Ihr Bild

M i
M\
f

Example 9^ Ihr Bild

W%* tkrBiU- ms
1 m
j
y
[pa
Siarrt'

J- >I. > i
mfrfT grxF
:
58 SCHUBERT
with its augmented fourth, sounds very like the introduction
figure itself, in its second, unambiguous manifestation.
Here the reader may be inclined to observe with some
asperity that this long sequence has got him no closer than
Schenker's paragraph to the mysterious intensity and "Tight-
ness" of IhrBild, to the "magician's secret thread." Perhaps not;
and perhaps certain artistic magic will not be very far "ex-
plained" by any course of analysis. But something is gained by
recognizing echoes of the two B^s later in the song; and some-
thing is gained by seeing a context for them outside the song
itself. The context, furthermore, is subject to analysis in terms

of a progression, one that is characteristic of Schubert, and

perhaps of music more widely.
The seed of the idea was literary, nonmusical, even ono-
matopoeic. Actual chimings are echoed, in the piano introduc-
tion to Gondelfahrer. Yet of course the bells of San Marco have
their distinctive Romantic coloring, which already seems to be
the main point of the introduction, in spite of the fact that this
also functions as a signal ahead toward the end of the song. A
similar function occurs in Die Stadt, where again the introduc-
tion illustrates a detail in the poem, directly and now much
more powerfully. In Geistes-gruss and Am
Meer the illustration is
just as clear,though markedly less specific in reference; and in
these works the harmonic cell has so caught Schubert's imag-
ination that it no longer bears an integral relation to the song
proper —as though the spirit has freed itself from the song,
and hovers above it, rapt in its own sibylline evocation. With In
der Feme and Ihr Bild a significant process of abstraction sets in.
Technically, the "x" pulls in to bare octaves, and the repetition
disappears; formally, the introduction is made to grow rather
subtly within the song; and poetically, there is really now no
direct pictorial reference. In the final analysis, then, Schenker's
effort to link the introduction of Ihr Bild to staring or any
other concept from the poem misses the point. This introduc-
tion is more purely musical than any others of the same type.
A device that begins frankly as a reflection of a literary
idea and ends up as a purely musical resource: how often
we see this principle working itself out in the course of music
history. In the century from Schubert to Schoenberg and
— —
Debussy not to speak of earlier times ideas acted con-
 A Romantic Detail in Schubert's " Schwanengesang"
59

stantly as a stimulus in this sense to "absolute music." With

Schubert, finally, the tendency of his introduction pattern to-
ward abstraction and intellectualization can be followed in one
7
or two of the last instrumental works.
Many writers on Schubert have pointed out illuminating
vocal derivatives in the instrumental music. On the basis of
such derivatives, indeed, Harry Goldschmidt has concluded
that Schubert's remarkable stylistic development in 1828 owed
8
its impetus directly to the Heine songs. Hans Koltzsch, in his

study of the piano sonatas, noted that the rumbling trill in the
B-flat Sonata of 1828 (ex. 10) has a parallel in the piano inter-
9
lude of Ihr Bild (cf. ex. o).
The mood is analogous, as is the function, a muffled re-
peated half cadence after the first melodic segment; the chord is
almost the same augmented sixth, in practically the same spac-
ing, with the same A\> in the offing; in both pieces ^ VI serves to
prepare the first modulation. The key is the same (it is the key
of Beethoven's "Archduke" Trio, which must have inspired
Schubert's lyric first subject). Writing a sonata, Schubert was
determined to carry the trill figure on, notably in his retransi-
tion section. But the figure does not develop, certainly not in
any Beethovenian sense. The passage, so handsomely treated
by Tovey, is superb, but the figure remains essentially what it
was at the beginning: a mysterious, impressive, cryptic, Ro-
mantic gesture.
The C-Major Quintet of 1828 opens with a similar gesture
(ex. 11). Since Schubert was writing not a song introduction
but a theme, he found a need to add a four-bar consequent, less
colorful and more businesslike than the pregnant harmonic
cell. How was such a manifest hybrid of a theme to be manipu-

Example 10 Sonata in B-flat Major

X
60 SCHUBERT
Example 11 Quintet in C Major
Allegro ma won troppo v>

*- ^ '
1r
r^L
— Tf .
.
p
-*

=
fo g V gf V g If* y C'. y Utf^ 3 > ' _
fl

Example 12 Quintet in C Major

X

Example 13 Die Allmacht

imxgctm, fzicrikli

lated? Once he had started his great structure moving by repeat-

ing the theme in sequence, Schubert employed it very sparingly
indeed, and in ways that are more akin to thematic transforma-
tion than to Classical development. In the transition, the theme
appears as a vigorous fortissimo in the bass. In the recapitula-
tion, the theme sprouts ecstatic arpeggios. In the coda, it is
violently racked (ex. 12). Here, as the diminished seventh at
"x" explains itself away to the tonic J, we may perhaps think of
the somewhat gross apocalyptic sounds that introduce Die
10
Allmacht (C major once again; ex. 13).
 A Romantic Detail in Schubert's " Schwanengesang" 61

Example 14 Quartet in G Major

m cresc. f V? ^-
5^^
f
?n -
^s * ^p
most impressive theme of this kind had been
Actually, the
composed two years earlier, for the G-Major Quartet of 1826
(ex. 14). Like the quintet theme, but more concisely, this one
comprises a startling harmonic cell followed by more active
material, the whole designed for sequential repetition. The
cell, though not cut in the same pattern as those of 1828, has

strong points of contact with them. However, its great power

—
ambiguity ambiguity more far-reaching than that
lies in its
of the hollow octaves and the bare low Cs mentioned above.
Schubert interpreted the cell in several different ways; alterna-
tion of major and minor triads is only the most obvious way.
On account of its simplicity, perhaps, this is the function that
recurs in the finale, and that emerges in the coda of the first
movement, which has been markedly unstable and tense, and
now receives a splendid firm resolution. But at the start, as
quoted above, the G-minor triad seems to act as a stressed non-
harmonic chord, a passing harmony leading by spasms from G
major to its dominant. As though to force this interpretation,
much is made of the melodic step B^-A (which is at once
repeated, bar 5, then echoed on other degrees, bars 10, 11,
12-13, I 3 -I 4)- At the recapitulation, where the chords are re-
versed so that G minor comes first, the G-minor triad seems to
resolve up into the major tonic. In this transformation, then,
the second, accented chord of the theme is treated as some-
thing functional, rather than as a willfully stressed passing
chord: a fact that contributes with many others to the serenity
of this famous passage. The passing note has been transformed
out of existence. The progression B-B!>-A is gone and even
the residual Bt]-A is hidden by new emphasis, in the first vio-
lin, on the descending fifth E-A.
62 SCHUBERT
No other work, perhaps, hints more excitingly at what
Schubert would have made of thematic transformation, had he
lived into the age of Chopin, Berlioz, Liszt, and Wagner. The
proleptic technique of the well-known "Wanderer" Fantasy,
composed in 1822, seems by comparison elementary.
His Romantic orientation, in the last years, stands out
clearly if the first themes of the G-Major Quartet and the
quintet are viewed in the light of earlier compositions such as
the "Trout" Quintet and the Sixth Symphony (ex. 15). These
themes, too, make a phrase out of an auxiliary diminished sev-
enth plus blander continuations. But Schubert saw nothing
special in the chord: the themes are merely trite, strongly remi-
niscent of a Rossini cliche, and perfectly serviceable for devel-
opment later on. In the quintet, Schubert reinterpreted this
kind of theme in a thoroughly new light, investing the auxiliary
progression with color, mystery, and emotion. Such themes
will not develop in any Classical spirit. They have been crossed
with the brooding introductory sounds of Heine's poems of
the sea.
Der Freischiltz and Euryanthe played in Vienna in 1821 and
1823: the very heralds of German musical Romanticism. The
town fell into camps, one upholding the reigning star of Euro-

pean opera, Rossini, another supporting the composer of the

Wolf's Glen scene. The Schubert circle was on the whole less

Example 13 (a) "Trout" Quintet; (b) Symphony No. 6

P/ZZJ
p

j*P »«f PPW

 A Romantic Detail in Schubert's " Schwanengesang" 63

than enthusiastic about Weber, and Bauernfeld tells a well-

known story about Schubert's own coolness. One is reminded 11

of his equivocal, disturbed statements about Beethoven. To-

ward a less challenging figure, Rossini, Schubert had once been
quick in praise:

You cannot deny him extraordinary genius. The or-

chestration is most original at times. (Schubert to A.
. . .

Hiittenbrenner, 18 19)
Schubert has so much praise for Rossini's Otello; talk
12
with him about it. (Holzapfel to Stadler, 1819)
. . .

But to set Schubert's C-Major Symphony of 1818 next to his

C-Major Quintet of 1828 is to see how emphatically, and how
magnificently, his allegiance shifted.

Notes
1.Heinrich Schenker, Der Tonwille 1 (1 921) .46.
2. The piano introduction in Schubert's songs is well treated by
Edith Schnapper, Die Gesange des jungen Schubert (Bern: P. Haupt, 1937),
pp. 137-51.
3. The best-known example, Die Forelle, is incorrectly printed in the
Schirmer, Ditson, and Peters editions, following Diabelli's edition of 1829:
the first bar of their six-bar introduction is spurious. See O. E. Deutsch,
Franz Schubert: Thematisches Verzeichnis seiner Werke in chronologisicher Folge,
rev. W. Diirr, A. Feil, etc. (Kassel: Barenreiter, 1978), pp. 319-20.
4. Walther Vetter, who points this out in Der Klassiker Schubert (Leipzig:
Deutscher Verlag fur Musik, 1953), takes the sonority as evidence of Schu-
bert's realism: "Der Komponist hat den Glockenklang genau studiert. . . .

Schlagton, Unterton und Obertone sind deutlich vernehmbar" (p. 373).

. . .

I have not been able to consult Paul Mies's article "Zwei Kompositionen zum

Gedicht Der Gondelfahrer von J. Mayrhofer," in Deutschen Sdngerbundes

Zeitung 19 (i927):6i4- 15; it is mentioned in his Schubert der Meister des Liedes
(Berlin: Max Hesse, 1928), p. 379m
5. There are in all five earlier versions of the song; see Deutsch, Schu-
bert Verzeichnis, pp. 99-100. Three are given in the old collected edition as
nos. 174 a, b, and c. No documentary authority exists for the date of the
final, published version — the autograph is lost —
but the internal and cir-
cumstantial evidence seems decisive.
6. Cited in Richard Capell, Schubert Songs (London: E. Benn, 1928),

p. 251 n. Actually the B\> triad functions as a passing harmony between i and
tin (B minor and D minor), an extreme relationship of a kind much favored
by Schubert at this time: see also Kriegers Ahnung, Aufenthalt, Der Atlas, and
(best of all) Der Doppelganger.
 — —

64 SCHUBERT

7. A survey of Schubert's song production reveals at least three other

introductions bearing a technical affinity to those that have been discussed.
Each in its own way differs in spirit so far from the main class that to treat
any of them with that class would be tendentious; but they are interesting
enough to deserve a note.
The reader may have thought of Der Atlas in this connection: as in Gon-
introduction runs t
delfahrer, the — — —
x-t x-t t. The distinction, of course,
comes in the rigorous motivic organization of Der Atlas. It is the powerful
motive that creates the appoggiatura "x," and thereafter it controls the song
measure by measure until the end.
Heimliches Leben (D. 922), another late song (from 1827), opens with
an introductory phrase of the kind used in the C-Major Quintet, which will
be discussed presently. A stressed auxiliary diminished seventh is followed by
more neutral material. Far from foreshadowing the quality of the quintet, the
auxiliary as articulated melodically here turns out to be one of Schubert's
tawdrier creations. The poem was a little warm for Schubert.
Ihr — —
Crab (D. 736; ?i822) begins d x d x d, where "D" is the
dominant of vi (C minor), not of the tonic (E-flat major). This anomaly
relates back to a whole set of harmonic experiments among the songs of
1 8 17- 1 8. What seems to be involved in this introduction is the establishment

of the step G-Ai>, both in the top line and in the dissonance "x." Then the
important motive that begins the song proper ("Dort ist ihr Grab": Ei>-
B^-G) pivots around G by moving from I 6 (G in the bass) to vi (G in the
voice), and leads via another bass G to the key of A-flat. Harmonic ambi-
guity is reduced prior to the final stanza of the song, where the rather delicate
dissonance at "x" is replaced by something closer to the introduction of In
der Feme. While At-G is still in evidence, the emphasis shifts to another
semitone, C^-EH, which is harmonically direct. This "clearing up" recalls
the elucidation of the ambiguous hollow octaves in IhrBild, but the situations
are too far apart to throw much light on one another.
If Ihr Grab remains a cold song, in spite of its mawkish poem, and in
spite of its very great harmonic interest, one reason may be its lack of any
binding musico-poetic image. Even the dissonance "x" is for once promoted
contrapuntally, not coloristically.
8. See Harry Goldschmidt, "Zu einer Neubewertung von Schuberts
letzter Schaffenszeit (1828)," in Bericht iiber den siebenten internationalen musik-
wissenschajilichen Kongress, Koln 1938 (Kassel: Barenreiter, 1959), pp. 118-20.
9. See Hans Koltzsch, Franz Schubert in seinen Klaviersonaten (Leipzig:
Breitkopf und Hartel, 1927), p. 131.
10. Compare also the Fantasy for Violin and Piano of 1827 (D. 934),
still in the key of C.

11. O. E. Deutsch, The Schubert Reader, trans. Eric Blom (New York:
Norton, 1947), pp. 892, 294.
12. Deutsch, Schubert Reader, pp. 117, 120.
 Schubert's Tragic Perspective

WILLIAM KINDERMAN
for Eva

In his article "Prinzipien des Schubert-Liedes," Hans Heinrich

Eggebrecht wrote that "in Schubert's songs major and minor
are often juxtaposed with one another as the illusory world of
beautiful, bright dreams to the real world of banal, wretched,
naked reality. .
."* This suggestive analogy points toward a
.

much larger issue in Schubert's music: the relationship between

overt musical contrast and the dichotomy of external and in-
ternal experience —
perception and imagination —that so pre-
occupied the Romantic poets set by Schubert in his Lieder.
Contrast between the vision of the imagination and a bleak or
threatening reality occurs frequently in the works of Goethe,
Wilhelm Miiller, and Heine, and this aspect of the poetry is
almost invariably seized upon by Schubert. His familiar pre-
dilection for moving between major and minor, his techniques
of modulation, his use of heightened thematic contrast all —
contribute to a duality of perspective rooted in the poetry but
expressed even more vividly in the music.
The first outstanding instances of this dual perspective
among Schubert's songs are the two early masterpieces to poetic
texts by Goethe, Gretchen am Spinnrade and Erlkonig. In the for-
mer, a realistic, ostensibly "external" piano accompaniment,
suggesting the spinning of the wheel as background, serves
also to depict Gretchen's distraught ecstasy: at the climax, the
activity of her accompaniment breaks off, and is resumed only
with difficulty. In Erlkonig, on the other hand, an internal
66 SCHUBERT
viewpoint, only hinted at in Gretchen, provides the central dra-
matic perspective of the song. As Tovey pointed out, the seduc-
tive vision of the Erlkonig presented by Schubert is that of
2
the child, not that of the father. The psychological depth of
the setting is made possible by a sudden shift of perspective ex-
pressed in the music, which enables Schubert to do justice to
both the pictorial dimension of the poem and the inner experi-
ence of the boy.
The turbulent background of the scene, suggesting the
horse's hooves and rushing wind, is expressed by the repeated
octaves and marked rhythmic motives in the piano accompani-
ment. The tonality is prevailingly minor; the major mode is re-
served for the Erlking, whose music stands in sharp contrast to
the rest of the song. He addresses the boy in beautiful, coaxing
melodies; only here does the stormy background of the song
recede. Juxtaposed with this seductive melodic idiom is the
terrified response of the boy, expressed each time by the grating
dissonance of a minor ninth between the voice and piano.
Schubert's tonal scheme, utilizing his favorite device of abrupt
modulation, enhances this effect of a shift in perspective, from
hallucination to harsh reality (see fig. i).
Schubert's distinctive treatment of the Erlking's first two
passages permits him to alter this pattern with powerful effect
in the third and final one (see ex. i). Here the rapidly repeated
chords of the accompaniment finally invade the music of the
Erlking. The playful melodic shape and major mode, both still
present in the first phrase ("Ich liebe dich. ."), give way to a
.

measured declamation, and the Neapolitan E-flat drops to a ca-

dence in D minor. Moments later, a similar but even more em-
phatic cadence in the tonic G minor frames the boy's last words
as he dies: the inward vision expires, and with it the life of the
child.
The achievement of Erlkonig provided a model for the co-
ordination of music with the duality of subjective and objec-
tive experience. Years later, Schubert returned to this musical
procedure in a number of important songs. Most of these
songs, beginning with the setting of Friedrich Riickert's Dass
sie hiergewesen of 1823, are concerned with tragic reminiscence,

with joy remembered but lost. Here, the joy of love is recalled
 Schubert's Tragic Perspective 6j

Figure 1 Erlkonig: The Tonal Plan

Stanza
1 Introduction: G minor - B-flat major - G minor
Narrator
2 Father C minor
Son C minor —» F major
Father —> B-flat major

3 Erlking B-flat major, lyrical melody

4 Son V/G minor; modulates to B minor
Father —> G major, as V/C
5 Erlking C major, lyrical melody
6 Son V/A minor; modulates to C-sharp
minor
Father —»D minor
7 Erlking E-flat major; cadence in D minor on
"Gewalt"
Son V7 B-flat minor; cadence in G minor
on "gethan"

8 Postlude: G minor
Narrator

Note: Thematic juxtaposition is shown by boxes.

as a flashback, as the memory of something that no longer

exists in reality.
Dass sie hier gewesen captures both the sense of unreality of
the reminiscence and subjective force: the song remains
its

poised between the perspectives of external and internal ex-

perience.' The external scene is suggested by the first two lines
of the text, "Dass der Ostwind Dufte/hauchet in die Lufte"
(That the east wind/breathes scents into the air) (see ex. 2). The
first twelve bars assume an eerie, mysterious quality through
the soft appoggiatura chords in the high register, the effective
use of silence, the unsettled tonality, and the narrow range
 .

68 SCHUBERT
Example 1 Erlkonig

*Ick {ie - bedich.mich reiztdeinescko-neGie- stalt; und bist du nicht

$ wit- ft^,
%
so Sraurk ick
^i
ty- wai€
3E
p

Mein Va
}tp

-
f±

ter; mm
mm
.90m

Va-kr,
t
m

ktzt

$km MM mm mm
3 ** I

V frr r
,
.
fr t> * a
I fasst er murk an Erl - feo - nig tat tnir ein f euls & - iWv!

of the melody- especially its persistent emphasis on the in-

terval E-D.
Anextraordinary feature of this setting is that the revela-
tion C major, is delayed until the reminiscence
of the tonic key,
of the beloved at the words "dass du hier gewesen" (that you
were here). These words reflect the internal, subjective per-
 Schubert's Tragic Perspective 69

Example 2 Dass sie hier gewesen

6ekr U«£i
ream

tfc, cU-durch. mut er_ ktmd, dass du titer qe - we-

voe - sen,.

^ m lg| K I s
At s£
li t f^j
** "te^rf:

iipiig P * "r; s
?=p i f

dass du kier^e- we - sen Dass (tier %rcL-nen rin- nen

1 V T .0 -ZZJZ-^ZLl-Z

spective of the song, which is embodied by a melodic fragment

of two bars in the clearest C major, employing only tonic and
dominant chords. The simplicity of this phrase is underscored
by its immediate repetition. The first half of the phrase, com-
prising the descending fourth C-G, is then repeated once
more in the piano before it is abruptly broken off in a bar of
 —
70 SCHUBERT
silence (bar 18). The simplicity and repetitive character of this
phrase make it sound insubstantial, despite its tonal stability
an impression confirmed by the dissolution into silence in the
middle of the phrase. The return of the soft appoggiatura
chords mark the resumption of the external perspective at the
words "Dass hier Tranen rinnen" (that here tears are running).
The musical substance of the entire song is built of these
two contrasting phrases, one associated with the external scene,
the other with the reminiscence of the beloved. The song re-
mains delicately balanced between the two perspectives and
ends in C major, with a final repetition in the piano of the mu-
sical phrase for "dass sie hier gewesen." Unlike in Erlkonig,
a tragic end is forestalled here: the consolation of memory,
even if illusory, compensates for the sense of loss. The song is
poised on this expressive ambiguity derived from the text.
In 1827 Schubert wrote the second of his song cycles to
texts by Wilhelm Muller, Winterreise, whose subject matter is
the quintessentially Romantic theme of the wanderer. Winter-
reise is a particularly grim example: the quest of the disillu-
sioned wanderer leads ultimately to madness and oblivion. As
in Dass sie hier gewesen, Muller's wanderer seeks emotional sus-
tenance in dreams and memories; but here every illusion is
shattered, leaving no escape from the desolate reality that con-
fronts him.
Several of Schubert's Winterreise songs exploit the tragic
pathos latent in illusion, but perhaps none more effectively
than Fruhlingstraum (no. 11) and Tauschung (no. 19). The first
stanza of Fruhlingstraum, which describes dreams of spring, is

set by Schubert as a gently swinging tune in I meter, in the

bright key of A major. The first words of the second stanza
dispel the illusion, as the traveler awakes to the crowing of the
cock, and to bitter cold and darkness. At the same time, the
music shifts into the minor mode, and the melody disappears,
replaced by rhythmic declamation in the voice and dissonant,
syncopated appoggiaturas in the piano. Tauschung, placed much
later in the cycle, employs the same general musical idiom as
3
Fruhlingstraum in the same poetic context. It is also in J meter
and in A major and even employs some of the same turns of
phrase, particularly at the cadences. Unlike Fruhlingstraum,
—
however, it remains within the major mode and hence the
 Schubert's Tragic Perspective 71

sphere of illusion —
until the end of the song. In this instance,
the confrontation with grim reality is postponed until the next
song, Der Wegweiser.
Duple meter and steady movement in eighth notes give
Der Wegweiser a processional character. In this respect it is
reminiscent of Gute Nacht, the opening song in the cycle. But
whereas GuteNacht launches the winter's journey, Der Wegweiser
points to its end: it is a procession to oblivion. Its persistent
—
rhythm and darkness of key G minor after the A major
—
of Tauschung contribute to its overwhelming immediacy of
effect. The climax of the song, heightened by a chromatically
ascending bass, occurs on the words "eine Strasse muss ich
gehen/die noch keiner ging zuriick" (a path must I follow/
from which no one has ever returned).
The suggestion of physical movement in Wegweiser allies
itself with a spatial, or external perspective, while pure lyri-
cism, as in Tauschung, can be identified with the internal and
nonspatial realm of imagination or illusion. Not infrequently,
the evocation of the pictorial in Schubert is associated with
such rhythmic movement, in particular a steady processional
movement. This is true not only of the songs but of instrumen-
tal —
works a striking example is the Andante of the "Great"
C-Major Symphony, composed in 1825-26. After the com-
position of Winterreise Schubert's processional movements tend
to assume a more fateful, and even tragic, character, as in the
slow movement of the E-flat Piano Trio or the Fantasy in F Minor
for piano duet. This reflects a shift in emphasis in the last two
years in Schubert's life from the somewhat naive nature worship
of Die schone Mullerin to a more deeply introspective world,
which is embodied above all in the six Heine songs of 1828.
The second of the Heine songs, Ihr Bild, contains one of
the most sensitive and powerful examples of tragic reminis-
cence in allof Schubert's Lieder. In a sense, the basic idea of
this song the reinterpretation of the single note, BK which
is

acts as a focal point for the voice throughout (see ex. 3). The
song opens enigmatically, with a twofold repetition of an oc-
tave Bl> in the piano. After the entry of the voice, the stability
of the Bk is undermined. In bar 4, it descends to A, in a strong
rhythmic position; similarly, the G> in bar 5 descends to F in
the following measure. In the next two-bar phrase in the piano,
72 SCHUBERT
Example 3 Ihr Bild, first strophe

Lxind&attv

lch stand in dunkitn Trau • men und sfcirrt'ikrBtU- nis

$ m,
to
i t
>
und
J i

r
das
pr
ge-lieb-
r
te
\
\
Ant -
W
iitz

Pm4 ^*=
4 1
1
* &UiKi
l
i

iL
IS
UUl '
,r r

#^ r
taim
r r
tnTTTr
1r.

-tick zu. U -
a a
ben be - cyxnn.

H£)*$ ii g f
'W» f hurT T

the
^
Bl> and G> are heard as unequivocally dissonant, as a double
i i

iiN
appoggiatura to an F-major triad. The interpretation of the Bl>
here as a dissonance, and in the following passage as a conso-
nance, as well as a change in mode from minor to major, em-
body the shift from an external to an internal perspective, from
emotional desolation to the consolation of the imagination, the
memory of the beloved. The harmonic basis for the opening
 Schubert's Tragic Perspective 73

bars is the resolution of the major third B^-O to A-F, while

the simultaneous resolution of both semitones to an F-major
triadis presented by the echo of the second phrase in the piano.

Until the piano echo, in fact, the entire passage consists of bare
octaves; the austere setting embodies the sense of disconsolate
staring at the picture in Heine's poem. 4
Then, secretly, the picture — and the music — come to life.

The voice returns to the same B!> began the song, now-
that
treated as a consonance and harmonized in B-flat major. The
words "Heimlich zu leben begann" are set to a full authentic
cadence, echoed a moment later in the piano. The crux of
Schubert's musical setting is his suppression of the major mode,
and of any vertical sonority, in the opening phrases. After this,
the straightforward tonal progression beginning in bar 9 has
the effect of an awakening of the imagination —
its consonant

harmonies embody the warmth of feeling expressed in the

words. The most subtle aspect of this musical setting is that it
seems to make the major mode dependent for its existence on
the fragile mood implied in the poem. For the beloved is lost.
The emotional consolation of this memory is transitory, and is
dispelled before the end of the song.
The second of the three stanzas in Heine's poem continues
the reminiscence about the beloved, now set in G-flat major,
another harmonization of the crucial opening pitch, Bk Then
the third and final strophe breaks the mood of illusion, bring-
ing a recapitulation of the music of the first strophe. This time,
however, the musical shift from minor to the harmonized
major tonic accommodates a very different sentiment in the
text (see ex. 4). In this final strophe, the warmth embodied by
the harmonized major tonality returns, in spite of a conscious-
ness of the loss of the beloved. There is no need to read an
ironic intention into this passage; such an interpretation actu-
5
ally misses the most profound aspect of the song. For while
the "awakening" of the picture was illusory, the feelings of
consolation and happiness were real; the return of the feeling in
the music of the last stanza speaks to the universal human ca-
pacity to experience happiness despite the pain of loss. The
deeply poignant quality of this setting is intensified, moreover,
by the shift into minor in the piano postlude, which restores
the tragic perspective of the song at the last possible moment.
74 SCHUBERT
Example 4 Ihr Bild, last strophe

Aucn mti ne Xkrarun flos - $erv mir von denWonom her- at

4+h- * Jir pr pp i
V rdPirtrr r ir, ?

M ^P
und ach y
ick kann es vjjni <Aa\k km.dass kk dick vtr- lo - ren (tat'.

^m m i3 _

H « £52=
fell ' ^^

&
4 BS
O

Iff

'^J JJJ J
s^

7/zr B/Wrepresents the last Schubert song to explore the

distinction between external perception and internal imagina-
tion, the territory brilliantly opened up by Erlkonig thirteen
years earlier. It is the most concentrated and unified of these
songs, for in it any direct sense of the pictorial dimension is
withdrawn. 6 The sound of the cryptic Bl> is carried through the
entire song, and the changing harmonic interpretation of this
 Schubert's Tragic Perspective 75

note embodies the change in feeling, or state of being, sug-

gested in Heine's poem. The unity of Ihr Bild is like the unity
—
of the self that residue of consciousness which registers the
inconstancy of experience against its own constant being.

//

As we have seen in several examples from Schubert's songs,

contrast between major and minor may represent one aspect of
a more profound thematic juxtaposition suggesting the dichot-
omy of inward imagination and external perception. This in-
terpretation is strongly supported by the poetic text. Even in
the absence of a text, however, certain of Schubert's late instru-
mental works employ an analogous musical treatment based on
the exploitation of thematic, tonal, and modal contrast. Our
final example is the great Fantasy in F Minor for piano duet,
composed early in 1828. While clearly indebted to the model
of the songs, this composition goes far beyond them in explor-
ing the structural and expressive possibilities inherent in the
controlled juxtaposition of strongly contrasting themes.
Like the earlier "Wanderer" Fantasy, the F-Minor Fantasy
consists of four interconnected movements performed without
a break. In both works, an opening allegro is followed by
a slow movement, scherzo, and final movement employing
fugue. The thematic treatment in the F-Minor Fantasy, how-
ever, has no parallel in the earlier work.
The lyrical opening theme of the Fantasy bears some affin-
ity to the processional themes of pieces like the Andante of the
C-Major Symphony, and Gute Nacht and Wegweiser from Win-
terreise. Its processional character derives from a regularity of

rhythmic pulse in duple meter and the steady octaves in the

bass, repeated twice per measure (see ex. 5). As Eric Sams has
pointed put, the melody itself has an insistent conversational
7
character, suggesting the rhythm and intonation of speech.
The overall quality of the theme is narrative; it seems to evoke
the landscape of ceaseless wandering familiar from the two
Muller song cycles.
After the repetition of the initial thematic statement, the
music shifts into A-flat major, and the lyrical melody passes to
the bass. This section represents the middle part of a ternary
76 SCHUBERT
Exa mp le s Fan tasy in F Minor

NXedfo moito moderate

$&mm m
yrmr
i^ r^y^P
m.
? r t
f

thematic construction. The opening theme returns, however,

in F major, and the brighter sound of the major mode is en-
hanced by richer harmonies in the bass and the emphasis on All
in the melody (see ex. 6). Schubert also exploits the high upper
register of the piano in the last phrases before the melody ca-
dences in the tonic.
—
That cadence brings a shock a contrasting second theme
in F minor, which is utterly opposed to the opening theme in
affective character. The menacing character of the new theme
is due to its stress on I>, the dissonant minor second above the
dominant note, its pointed accents, and its funereal rhythm.
This rhythm, first announced in the bass, consists of the pat-
tern J. J> J J, representing a related but more energetic form of
the rhythm associated by Schubert with death in the song Der
Tod und das Madchen, J J J.
By analogy with Schubert's songs, the statement of the
first lyrical theme in major assumes an air of unreality, of illu-
sion. The illusion is rudely shattered by the plunge into minor
and the threatening second theme. This drastic thematic jux-
taposition then serves as the structural basis for the rest of
the first movement. After the initial statement of the second
theme, the lyrical theme returns in D-flat minor, closing with a
cadence in A minor, where it is once again juxtaposed with the
 —
Schubert's Tragic Perspective 77

Example 6 Fantasy in F Minor

inmo
^m ttm ±±L
m *mm §
Stcondo J' [. ^
PP
mdji mm hlm
F^T^f^p
4
m ^ .,11 ff r

^i3L ffi O v rji «rp y jp ^Pvlt^P

;

Vf f

f
l
f
'

T^T * a *

fT> y
uJ
f.ffr fr
4»,

gfe
«

^
ru
i^ MMm^jpyjp S^ i
m^ FT ff

N
»
N^ W# ?=H?
Eg »
r

theme rhythm. A last statement of the opening

in funereal
theme in F minor completes the series of modulations through
a circle of descending major thirds, F-D^-A-F. This appear-
ance too is juxtaposed with the second theme, which, in a re-
versal of roles, now appears transformed pianissimo, legato,
and in major. The statement of the second theme in major has a
resolving effect, serving to round off the first movement before
the dramatic opening of the Largo in F-sharp minor. The tonal
and thematic plan of the first movement is shown in figure 2.
78 SCHUBERT
Figure 2 Fantasy in F Minor, first movement: The Tonal Plan

Lyrical theme F minor

Melody passes to bass A-flat major, modulates to
V/F minor
Lyrical theme restated F major
Second theme, funereal F minor, modulates to
rhythm

Lyrical theme D-flat minor, modulates to

Second theme, funereal A minor, modulates to

rhythm

Lyrical theme F minor

Second theme, funereal F major (leads to second
rhythm movement)

Note: Thematic juxtaposition is shown by boxes.

The tonal relations and thematic juxtaposition in the first

movement of the Fantasy are similar to those in Erlkonig: in
each, material of strongly melodic character is pitted against
more dramatic, turbulent material in minor, and the contrast is
heightened by means of abrupt modulation (cf. fig. 1). In the
Fantasy, however, this dual perspective reaches its culmination
only in the final bars of the work. The last movement represents
8
a recapitulation and development of the first movement. After a
sudden modulation from the key of the scherzo, F-sharp minor,
the lyrical theme returns in F minor. The entire opening sec-
tion is then restated, in somewhat condensed form, up to the
crucial passage in which the lyrical theme appears in the major
mode. From this point, the work takes a new course.
The dark-hued second subject now becomes the basis for
an extended fugue. In the latter part of the fugue, the principal
rhythmic motive undergoes a series of canonic imitations,
while its free inversion is worked into the rhythmic accom-
paniment as triplets in the bass. The music then builds toward
a tonic cadence in F minor, which is twice avoided before it
appears at the final statement of the fugal theme in the lowest
register. Again the music comes to a climax on a series of
 Schubert's Tragic Perspective 79

diminished-seventh chords, with the expectation of a cadence

in the tonic. This time, however, the cadence is denied: the
fugue simply breaks off on the dominant (see ex. 7). (This pas-
sage is reminiscent of the climax of the Andante of the C-Major
Symphony, where an analogous passage leads to a diminished-
seventh chord, followed by a dramatic silence, and a melan-
choly transformation of the principal theme of the movement
[bars 241-66]. The impact of the climax causes, as it were, the
subsequent transformation of the main theme.)
The conclusion of the Fantasy after this dramatic silence is
one of the most extraordinary passages in Schubert's works,
and merits detailed analysis (see ex. 7). It begins by recalling
the plaintive lyrical theme from the outset of the work; but the
reminiscence lasts only a few bars. The last bars, rising in se-
quence, already pick up the darker coloring of the second
theme; then, ten bars before the close, Schubert steps from one
theme into the other through a subtle transformation of his
material. The melodic continuation of the reminiscence is re-
lated to the second theme, particularly to the rhythmic pattern
at the climax of the fugue, three bars before it is broken off. At
the same time, the bass stresses the crucial semitone D^-C,
and a version of the funereal rhythm returns in the treble in the
bar before the cadence in F minor (see bars 562-63). The clos-
ing eight-bar statement is a further development of the second
theme, employing not only the funereal rhythm, but the me-
lodic stress on D^; the descending triplets in the bass are de-
rived from the fugue. In these final measures, the dark-hued
second theme supersedes the lyrical theme to provide the ca-
dence and resolution of the entire work.
Since the relationship and expressive conflict between these
themes have been key elements in the Fantasy, their juxtaposi-
tion in the coda seems to sum up the whole piece in a single
gesture ./Yet the appearance of the themes here goes beyond any
earlier passage in the work; it achieves a new synthesis. This
final statement is laden with tragic overtones. In context, di-
rectly following the forceful, almost orchestral impact of the
fugue, the lyrical opening theme sounds fragile, insubstantial.
This impression is confirmed in the last bars of Schubert's
coda, a passage that Maurice Brown has described as "the most
remarkable cadence in the whole of Schubert's work.'"
 a

80 SCHUBERT
Example 7 Fantasy in F Minor

Vrimg t0 (iva
,, p-. fT\ Jpk»iiypp Kl

IT cresc.
ens*.

5econdo -#*
4V |. L — fflfl•-V-
^-jjL-
* -61 L
rr'^FSt D #3Ji
t

P*M&« 565

5ys
m
| V>
w^m
»^^p *23wJ* r=F=r
Z Z

S
@
.->

^^ «Ng A-
tfr

a^
*£
n^ JL5.
La_k
Ef
i
IS
r-p~r
I MI '
*
M V

570
 1

Schubert's Tragic Perspective 8

This cadence owes much of its power to the fact that it

serves as the true conclusion of the fugue, after the abrupt in-
terruption and the reminiscence of the lyrical theme. The ac-
tual cadential progression refers back to several cadential pas-
sages in the fugue, in which the dotted rhythmic motive of the
subject is extended by a series of quarter notes. This time the
progression is strengthened by the presence of a descending
chromatic line, doubled in octaves, that highlights the disso-
nant semitone D^-C in the last two chords. The chromatic
line, beginning on F, passes through B|, EK and Dtj, reaching
O* in the penultimate chord, which is emphasized dynamically.
This penultimate sonority may be regarded as a subdomi-
nant minor triad with added sixth, or as a first inversion sev-
enth chord on the supertonic. In either role it might be ex-
pected to resolve to the dominant — as in fact happened at an
earlier point in the Fantasy, at the end of the transition from the
scherzo to the last movement. Schubert's omission of the domi-
nant chord at this cadence illustrates a cardinal principle in the
evolution of nineteenth-century harmony from Beethoven to
Wagner: since the dominant is so clearly implied in the context,
its actual appearance would be neutral and inexpressive. Its ab-

sence therefore greatly strengthens the expressive force of the

cadence. The result is a kind of enhanced subdominant ca-
dence, which combines two crucial motivic elements of the
work: the Dl>-C semitone relationship in the treble, and the
fourth in the bass, the thematic hallmark of the opening theme.
Critics have frequently pointed to a looseness of organiza-
tion in Schubert's work; Theodor W. Adorno even wrote of a
"potpourri," in which themes follow one another without
being organically related on classical lines. 10 The case of the
F-Minor Fantasy reveals that his claim misses the point, for
thematic conflict actually becomes a structural device. Transi-
tional passages characteristic of the Classical style are absent
here; indeed, the dramatic power of the work derives in large
measure from abrupt thematic juxtaposition. The first and last
movements of the piece, themselves interrelated, systemati-
cally exploit the juxtaposition of the two contrasting themes.
Furthermore, there is progress in the relationship of these
themes in the course of the Fantasy. In the coda of the last
 —

82 SCHUBERT
movement, the lyrical theme, recalled in a final brief reminis-
cence, is by the funereal theme, which dominates
obliterated
the entire closing section of the work, just as the lyrical theme
had dominated the beginning.
The overall scheme based on the relationship of these two
evocative themes suggests a latent symbolism analogous to that
of Winterreise. Like the song cycle, the F-Minor Fantasy is
haunted by a sense of progress toward an inescapable destiny,
an idea tied to the universal human theme of mortality. In a
sense, the very structure of the Fantasy is posited on this ap-
propriation of poetic content from the world of Schubert's
Lieder. In this remarkable composition, the expressive content
of the wanderer's tragic journey is transformed, as it were, into
a purely musical structure, absorbed into the sphere of instru-
mental music.

Notes
1. "Prinzipien des Schubert-Liedes," Archiv fur Musikwissenschaji 27
(1970)196.
2. "Franz Schubert," in Essays and Lectures on Music (London: Oxford
University Press, 1949), p. 109.
3. As Maurice Brown has pointed out, the melody of Tauschung was
drawn from Troila's song at the beginning of act II of Alfonso und Estrella,
from 1822. See Brown, "Schubert's Operas," Monthly Musical Record 79
(1949)1126.
4. Heinrich Schenker related the stark octaves of the piano introduc-
tion to the notion of "staring" in his essay in Der Tonwille 1 (i92i):46. A
translation provided by Joseph Kerman in "A Romantic De-
of the passage is

tail in Schubert's Schwanengesang," elsewhere in this volume, pp. 48-49.

5. For an interpretation of this passage as ironic, see Charles Brauner,

"Irony in the Heine Lieder of Schubert and Schumann," Musical Quarterly

67 (1981) 1277— 80. For another recent discussion of this issue, see Gemot
Gruber, "Romantische Ironie in den Heine-Liedern?" Schubert-Kongress Wien
1978, ed. Otto Brusatti (Graz: Akademische Druck- u. Verlagsanstalt, 1979),

pp. 321-32. Gruber distinguishes Heine's sarcastic, self-distancing irony from

Schubert's "Romantic irony," yet in Ihr Bild the source of apparent irony
the discrepancy between the words and music of the last stanza —
is resolved

on a plane of experience combining subjective feeling with an awareness of

external reality.
6. See Kerman, "A Romantic Detail," p. 58.
7. "Schubert's Piano Duets," Musical Times 117 (1976) 121. :
 Schubert's Tragic Perspective 83

8. For a discussion of the form of this work and a comparison with

Schubert's earlier fantasies, see Arthur Godel, "Zum
Eigengesetz der Schu-
bertschen Fantasien," Schubert-Kongress Wien 1978, pp. 202-4.
9. "The Fantasia in F minor, Op. 103," in his Essays on Schubert (Lon-
don: Macmillan, 1966), p. 96.
10. See Adorno, "Schubert," in his Moments musicaux (Frankfurt:
Suhrkamp, 1964), pp. 18-36. This essay originally dates from 1928. See also
Charles Rosen, The Classical Style (New York: Viking, 1971), pp. 454-59.
 —

Lyric as Musical Structure:

Schubert's Wandrers Nachtlied
("Uber alien Gipfeln," D. 768)

THRASYBULOS GEORGIADES
Translated by Marie Louise Gollner

More than six hundred of Schubert's Lieder have come down to

us, including almost seventy on poems by Goethe. The com-
position of a masterpiece such as Goethe's "Wandrers Nachtlied"
—
("Uber alien Gipfeln") raises the question one frequently
asked— whether it is possible, permissible, or worthwhile to
set poems of this rank to music. To be sure, Goethe himself
gave his poem the title of "Lied,"
— "Wandrers Nachtlied,"
and his words of praise for Zelter's setting indicate that he was
not opposed in principle to the setting of his poems to music.
1

Nonetheless we are probably justified in asking to what degree

— —
an outstanding composition like Schubert's is compatible
—
with a poem that is complete in itself like Goethe's. Or, to be
more precise, in what way does the music of a Lied exhibit
characteristics independent of the poem? What is the rela-
tionship of the composition to the poem? We will examine
Schubert's Wandrers Nachtlied (D. 768) in depth, viewing it not
only as an independent work of art, but also taking into con-
sideration the questions we have just raised. In so doing we
hope to clarify what transpired as a result of Schubert's Lied
production, within both the history of music and history of
the lyric (as a category of poetry).
We begin with Goethe's poem:
 Lyric as Musical Structure 85

Rhyme
I. Uber alien Gipfeln a

2. 1st Ruh; b

3- In alien Wipfeln a

4- Spurest du b

5- Kaum einen Hauch; c

6. Die Vogelein schweigen im Walde. d

7- Warte nur, balde d

8. Ruhest du auch. c

[Above all the peaks there is peace; in all the treetops you
feel scarcely a breath; the little birds are silent in the
wood. Only wait, soon you too shall rest.]

Although of varying length, the lines are clearly delimited by

the rhymes. They are also formed in different ways: in lines
1-4 accented (') and unaccented syllables (J) alternate; lines
5 — 8 each have two unaccented syllables between the accented
ones. Lines 1, 4, 5, 7, and 8 begin with an accented syllable;
lines 2, 3, and 6 with an unaccented one. Some of the line end-
ings are feminine (1, 3; &T~7), others masculine (274; s7~%)[ in
the first half of the poem
the succession alternates (1^3; 2, 4),
but in the second half the outer lines enclose the middle ones
(57^8, The number of accents is also variable. Line 2 con-
577).
tains only one accent; lines 3, 4, 5, 7, and 8 each have two.
The first line can be interpreted as having either three accents
(,^,w,w) or on iy two, preceded by two unaccented syllables
(ww,w,w\ Only line 6 is more leisurely; it has three unam-
biguous accents of which it takes full advantage, leaving two
unaccented syllables between each. As a result, it is consider-
ably longer than any of the other lines: the continuously un-
folding triple rhythm gives it a swinging movement and the
feel of a regularly built, songlike form. In this way it differs
from all the other lines, which, with their contours veiled in
86 SCHUBERT
darkness, express so emphatically the idea of night, as well as
the character of the spoken as opposed to the songlike.
Line 6 also differs from all the others in its content. " Voge-
lein" and "Wald" conjure up images from the naive, folklike
sphere, making us realize how completely different in content
is the rest of the poem, with its deep significance, its person-

alized statement. To understand the importance of line 6 more

—
clearly- to realize that it is only the interruption caused by the
change in rhythm and conceptual direction of this single line
that lends the poem its depth and its greatness — we need only
attempt to leave it out. The intellectual content, the reality
of the poem, is destroyed, replaced by a subjective image of
mood that is not binding. The critical role of this line is not
so much in the addition of a third element, the living creature,
to the images of lifeless (lines 1-2) and awakening vegetative
nature (lines 3-5), as in the interruption in continuity which
we have just described.
This can also be illustrated in the structure of the poem.
Although the succession of rhymes when taken by itself re-
sults in a division into 4 + 4 lines (abab — cddc), the rhyme
"Hauch"-"auch" and the poetic content lead us to interpret
lines 1-5 as a unit, as an antecedent, which is then followed by
the statement of the personal sphere, the inner self: aba be, then
ddc. ("Ruhest du audi" is also an echo of line 2, of the Ruh of
nature.) But this statement encompasses only the final two
lines (7-8). Then what of the third line from the end, line 6? It
hangs suspended between antecedent and conclusion, leading a
lifeof its own. According to the structure of the poem (partic-
ularly the rhyme "Walde"-"balde"), it seems to be separated
from the first four lines and bound to the last two; but accord-
ing to the content, it acts as a later and independent supple-
ment to the poetic observations of lines 1-5. This sixth line
functions —from whichever angle we choose to look at it as —
an island within the poem.
Schubert's setting was composed in his mature period,
about 1823 (the exact date is unknown), perhaps shortly before
the cycle Die schone Mullerin. (See the Appendix, ex. B, for en-
tiresong.)
The Lied encompasses a scant fourteen bars in \ time and
remains entirely within the main tonality, B-flat major. Only
 - —
Lyric as Musical Structure 87

the half cadence on the dominant in bar 6 is emphasized by the

preceding diminished-seventh chord on the leading tone B\. In
the following diagram the numbers indicate the length of the
2
individual phrases in units of 4 since it is not the whole but
,

rather the half bar that represents the basic unit of the composi-
tion. We encounter phrases of \ A bars (in the diagram, three
l

half bars); and the literal repetition of the passage "Warte nur
. .audi," consisting of V/2 + 1 bar (bars 9-10 + A = A +
.
l

12-13), appears to be displaced within the boundaries of the \

bar, although musically no metrical displacement occurs.

Vz-bzr units:
2 2 3 2 3 2 3+0 2 2

Bass notes: ,
r_^_^ '
• - „ -•
'

. •
^— —
BK F~BK BK GFE^D E\> E F '
f
'

GFBKF, BK" F, Bt
Structure:
A B A' [setting of' ines 3-7] B B

The two-bar introduction consists of two parts. The first

bar returns, somewhat varied, when the voice enters ("Uber
. .ruh"), and the second bar is identical with both the clos-
.

ing phrase of the song, "Ruhest du audi" (which comes twice,

in bars 10 and 13), and with the single-bar coda. The setting of
lines 3-7 appears, then, to be clamped firmly between the two
(now separated) bars of the introduction (A and B; see diagram
above). And conversely, the introduction acts as a contraction
of the entire Lied. There is in fact a noticeable caesura between
bars 1 and 2; the sudden entrance of the 4 chord (bar 2) after the
tonic triad is accompanied by a separate light impulse. It acts as
a mild thrust that does not emerge from the continuous flow of
harmony in the first bar, but rather creates a firm juncture
point. For a smooth harmonic progression a different chord
for example, one from the subdominant area, as in bar 4
would have been introduced either between the triad and the
4 chord or of the latter. The caesura resulting from the
in place
succession I^ receives even greater emphasis at the close of the
voice part by the fermata that precedes the words "Ruhest du
audi" each time; and the equally unexpected entrance of the
coda confirms the character of this passage. 2
The symmetrical structure of the introduction (1 + 1 bars)
88 SCHUBERT
is not incorporated into the main body of the song (with the
exception of bars 7-8). Bars 3-6 consist of three phrases, con-
taining respectively three, two, and three units of \ ("Uber . . .

Ruh," "In alien Wipfeln," and "Spurest . .Hauch"). The

.

first phrase of the vocal part represents a variation of the first

bar of the introduction and is thus based on the pedal point B^
and built around the tonic triad. Unlike the introduction, how-
ever, it adds notes of the dominant, instead of the subdomi-
nant, on the third beat of the bar and proceeds only after two
quarter notes to the tonic, which now likewise fills out the
space of a half note. The second phrase is constructed in an en-
tirely different manner. The bass descends stepwise through
the fourth G-D, and the upper voice follows in parallel sixths,
embellished by suspensions. And in the third phrase we find yet
another framework. The bass ascends chromatically, proceed-
ing, along with the upper voice of the accompaniment which
r
circumscribes the tone c' (c'-d^'-c'), from the sixth (B>-c )
through the tension-building diminished seventh (£—$') to
the release provided by the dominant fifth (F-c').
Common to the first and third phrases is their formation
around a single tonal center. Through them the two main poles
of the key, tonic and dominant, are placed in direct opposition.
The second phrase, however, presents a progression of sonorities.
Progression and center are two elementary forms of sound
structure; they determine even the oldest forms of notated
polyphony from the ninth century and have retained their fun-
damental significance ever since. 3 Their elemental meaning
shines forth in its original purity from the passage we have just
examined. The initial bar of the introduction also presents the
7
tonic as a tonal center, to which the cadential formula i4-V -I
is added in bar 2. However, this does not yet result in a con-

vincing cadential effect. Only in the further progress of the

composition, through the structural establishment of the domi-
nant (bars 5-6), its extension (bars 7-8), and the resulting ca-
dential progression VI-V-I (bars 9-10) — during which the
voice part continues to hover around the fifth, f" —
only in
light of these proceedings does the second bar of the introduc-
tion appear as a genuine cadence when it returns in its twofold
presentation (bars 10 and 13). Only now does it assume the im-
portance of an event, "Ruhest du audi" (You also shall rest).
 Lyric as Musical Structure 89

The individual character of the three phrases in bars 3-6 is

not, however, confined to their sonorities. The first one brings

the modified pavane rhythm

j nj hj
in the accompaniment, thus linking itself with the introduc-
tion. The secondpresents a calm movement in quarter notes,
and the third a syncopated web of eighth notes:

iSZZP^
Each phrase thus has its own distinctive structure that dif-
ferentiates it essentially from the others, and this within the

closely packed space of just three bars. And in spite of this

the listener has an impression of unity —
so much so that he
has no inkling of the remarkable structure we have just de-
scribed, fashioned, as it were, from individual blocks of gran-
ite. This impression of a unified flow is contingent largely

upon the voice part, although even here the technique of build-
ing with separate phrases is not only present, but in fact reaches
its culmination.
But Schubert's vocal setting is the musical sounding of
language; it is language as music. And language is coherence as
well as articulation, connection as well as separation, unity
as well as distinction. That is to say, it is the sentence —
and the
—
sum of the sentences as well as words, syllables, and sounds.
Music as a natural phenomenon can be compared to a continu-
ous flow; like words taken by themselves it embraces no cen-
tripetal forms that can distinguish among meanings and are
themselves distinct as manifestations. And thus music as some-
thing "natural" can mirror language as unified flow without—
ever penetrating more deeply to the level of language's specific
distinguishing word structure. But music as the bearer of mean-
ing, expressly formed by man, is capable, by adapting itself to
language in different ways in the different epochs, of some-
thing more. From the fullness of its own powers, it can create a
reality which contains a distinguishing principle similar to that
90 SCHUBERT
of language — from the
fullness of its own powers, and yet in
the final analysis based on the model of language as the
still

given factor. The music of the Viennese Classical masters and

Schubert's Lieder are fashioned in this manner, but at the same
time they appear as if they were natural.
And it is thus that the impression of the unproblematic
continuous flow of bars 3-6 is determined by the vocal part. It
mirrors not that progress of the musical structure which artic-
ulates as distinguishing language, but rather the continuous
gesture of expression, the mood invoked by the words, the at-
mosphere emanating from the language. It mirrors, in sum-
mary, not the specific quality of the language but only its bab-
bling aspect. From the extremely peaceful beginning ("Uber
alien Gipfeln/Ist Ruh"), akin almost to psalmlike recitation,
there develops with "In alien Wipfeln" a more animated gesture
—
of intensified expression the treetops ("Wipflen") are not as
rigid as the mountain peaks ("Gipfeln"). At "Spiirest du/Kaum
einen Hauch" the vocal part assumes a more intimate, even
melancholy (d^") tone, determined by the mood one feels—
("spurt"), albeit barely ("kaum"), the stirring of life and—
fades away into the low register.
This unproblematic, "natural" unity, however, is like a
—
mere shadow of another unity this one genuine which —
comes about only as an event. In this case music is active as the
bearer of meaning, in which the distinguishing principle simi-
lar to language is inherent.
Let us proceed therefore from the language. Together the
first two lines form a sentence. However, the verb "is" carries
no weight of its own as far as the statement is concerned, but
rather functions merely as a coupler, joining the two images,
"Uber alien Gipfeln" and "Ruh," in a single image. This is pre-
sented in an almost impressionistic fashion in the sentence; in
translating the static quality of the one image of reality into lan-
guage, however, of necessity it makes use of a succession of
words. In contrast, "In alien Wipfeln" is the first genuine state-
ment of the poem; it contains an active verb ("spiirest du") and
exhibits a goal-oriented, dynamic structure.
The musical setting of "Uber alien Gipfeln ist Ruh" cap-
tures not only the unity of the image, but also both the formal
function of "ist" as a verb and its dependent character as a
 Lyric as Musical Structure 91

coupler. Its function as a verb, that of forming a complete sen-

tence, expressed
is through the dominant sonority (second half
of the bar, in place of the subdominant sonority in the first
bar of the introduction), its coupling, static-impressionistic
character through the pedal on Bl>, which negates the dynamic-
directional tendency of the dominant. And the voice part
merely circumscribes the tone bt>' with the least possible move-
ment. The dactylic pavane rhythm, anchored firmly on the
first quarter note when it appears in the introduction,

fTjr r
1

is transformed into a gently cohesive shape which underlines

the unit of the sentence:

"In alien Wipfeln": after the static character of the pre-

ceding phrase, what an event now confronts us in this all-
enveloping rhythmic, melodic, and harmonic gesture! The
spirit of the wanderer embraces all of the treetops:

J. J>J"JJ
o( - U Wip- fet

The same word, "alien," appeared in the first line, too. But
only at this point, in be grasped in its spe-
the third line, can it

cific breadth, its own weight, its unique structure of imagery.

The setting of the beginning of the sentence "In alien Wipfeln"
as a separate phrase is paralleled by the distinction made be-
tween the following two phrases, "Spurest du" and "Kaum
einen Hauch." This renders the entrance of the first genuine
verb, now even accompanied by the subject "du," musically
convincing as linguistic event, as musical structure. This result
is also contingent on three factors: (1) the new pattern of the

accompaniment,
 —
92 SCHUBERT

(2) the rhythmic impulse accorded to the syllable "Spu-,"

j>T7injV!}!jj
In ai - Un Wip - fctn syix-nstdu

Whereas the rhythm of "In alien Wipfeln" began with an

upbeat

(J* )

and unfolded as a unified, encircling gesture, "Spurest du" be-

gins emphatically with a downbeat, is declaimed rapidly on a
single tone

tftJU

instead of

J. J>J
1 i

analogous to "alien Wip-" —

and has a concisely measuring
on "Spurest du." Whereas the two pre-
quality. (3) the tone c"
ceding phrases, beginning and ending on W, had remained
anchored on the B\> sonority, the c" and its accompanying
sonority now leave the sphere of B^ and move forward into a
new area. Only a genuine verb can speak to us in this fashion.
"Kaum einen Hauch" stands out as a separate phrase by virtue
of the weight given to the word "Kaum" (its length, and the
extension both of the c" to 3" and of the sixth E^-c" to the
diminished seventh E-d^') and by virtue of the entire rhyth-
mic structure,

J.
 Lyric as Musical Structure 93

Melodically, however, it stands in a unified relationship to the

two preceding phrases, "ln-[Wip]feln Spu [-rest] Kaum," supported
by the structure of the accompaniment, which remains the same
from "Spiirest du" onward.
It is thus clear: musical structure when language
is created
is captured as something real, when taken "at its word."
it is

Language, understood in this way as sentence structure and

verb function and the linking of images, stands not only above
music, but also —
and this is the crucial point above poetry as —
a work of art. It appears as the primary phenomenon spending
life and making art possible. Both Goethe and Schubert are
confronted with the fact of language. Goethe too takes as his
point of departure not mere syllables and sounds but rather
preexistent language, his mother tongue. Whereas he, how-
ever, fashions from it a poem, that is lines, Schubert brings
forth musical structure —
based on the poetic language. Both
are committed to language; both create directly out of it, the
one as poet, the other as composer. Yet in the poem the distinc-
tion between linguistic and poetic structure must obviously re-
main hypothetical. The reality of the poem embraces both in
one. Thus the musical composition mirrors the structure of the
poem as well as the language. The rhyme "Gipeln'V'Wipfeln"
appears as melodic correspondence,

^m-^m
"Spiirest du" as a contraction of "Uber alien Gipfeln ist Ruh,"

Syii-rzstdu, ii - berli- (m (rip- fek ist 9uk

J

transposed up a major second from W

to c". Similarly, the
pause on- the word "du" corresponds unmistakably to ("Uber
. . .") "Ruh," and the rhyme "Ruh"/"du" is further empha-
sized by the caesura after both words and by the rhythmic-
melodic contrast

5 tet lU *> Spu-nytdu

Spu-nstd
94 SCHUBERT
In the musical setting of the lines

In alien Wipfeln [t*'-W]

Spurest du [c"]
Kaum einen Hauch [dT-f ']
and poetic structure work together hand in hand.
linguistic
Through musical means Schubert here establishes the coher-
ence of the sentence while simultaneously separating the three
independent linguistic images, thus condensing each line, like
a crystal, into an unmistakable reality that is almost solidly
constituted. The composition of this passage alone is proof of
his creative rank. Its core is formed by the passage "Spurest

du," anchored on the root syllable "Spu-." This entire passage,

however, is brought into relief by the manner in which it is set
against the plane of the unified, undifferentiated setting of the
first two lines. After

Uber alien Gipfeln ist Ruh

come the separate units

In alien Wipfeln Spurest du Kaum einen Hauch.

(Both as a rhyme and in the music "Hauch" remains open, un-

answered; only the close of the final line, "auch," brings the
awaited correspondence.) When musical substance is created in
such a manner from language, it scarcely seems necessary to
point out that the mood appropriate to the poem, the atmo-
sphere resulting from its content, enters into the music almost
of its own accord, or that the vocal part acts as if it were "natu-
ral" (see above, pp. 89-90.)
What a concentration of substance in just four bars! There
now follows the folksonglike sixth line. In the composition it
too forms an island (bars 7-8 fall in the exact center of the
Lied: 6 + 2 + 6). Whereas up to this point the bass has been
weighty, corresponding to the sixteen-foot stop in the low reg-
ister, it now jumps suddenly to the upper octave, in the eight-
foot register, where it sheds its weight and becomes light,
playful, mobile —only to plunge back into the depths at the
following line, "Warte nur, balde." Whereas bars 1-6 con-
tained significant harmonic motion, we now find, over the in-
termittently sounded dominant bass, a primitive fluctuation
 Lyric as Musical Structure 95

between dominant and tonic in the nature of a guitar accom-

paniment. And whereas the independent formation of all the
other phrases results in contours individual to each, we are here
met with a symmetry, brought about by the repetition of bar 7,
which emphasizes the playful character of the passage. Even
the accompaniment figure, while corresponding in its rhythm

imz?t
to the 1/2 bars before bar 7, at the same time separates itself
from them with its playfully varied inversion of the bass figure:

ids
instead of

diS
All of this appears suddenly in bar 7 and disappears just as
abruptly after bar 8. Schubert realizes the songlike, popular
character of Goethe's poetic line with musical means. Through
repetition of the word "schweigen" he creates a musical-song-
like symmetry:

-hi j-
Di* Voa
^jij
kin sdnwd-aen
i/3 j
sebwti- gen
nn>,
im Wal- cU

This passage, then, represents the songlike sphere, achieved in

the poem by linguistic-poetic means, and here as music, by
those means available to the composer. We are thereby specifi-
cally made aware of the unsonglike nature of the remaining
phrases of the Lied.
After this winsome dream, gravity returns; the bass is
once again in the low register, phrases are again formed indi-
vidually, and we again find significant harmonic movement,
now supporting the Lied's statement.
The sentence formed by the final two lines of the poem
Q6 SCHUBERT
is divided, similar to lines 3-5 ("In alien Wipfeln/Spurest
du/Kaum einen Hauch"), into separate images,

Warte nur —warte nur—balde— Ruhest du auch;

and these phrases, in themselves stable, are joined together as a
whole. In the language, and thus likewise in the poetic form of
presentation, the appropriate image emerges from the decla-
mation of "Warte nur" (Only wait). This "Warte nur," how-
ever, is repeated by Schubert: music operates with time as a
material factor; the musical image "Warte" takes on the form of
longer duration by actually lingering. The pause on "balde"
(the fermata) should be similarly understood. The fermata sig-
nifies an extension lasting approximately the duration of a
quarter note, so that the passage "Warte nur, warte nur, balde"
occupies the space of yA bars, 4 + 4 + 4. This variable metrical
form captures the as yet undirected hovering of the line "Warte
nur" and creates simultaneously the necessity of continuing to
the close.
The ww which characterizes the rrrythm of
succession '

Goethe's line "Warte nur, balde" is canceled out by the repeti-

tion of "Warte nur" and by the division into three phrases:

Warte nur — warte nur— balde.

The rhythm of line 6 is changed in a similar manner. Instead of

Die Vogelein schweigen im Walde

Schubert gives it the rhythm
Die Voglein schweigen — schweigen — im Walde.
(The succession '^ w "schweigen-im," is not to be interpreted
,

as a unit here.) The continuous succession >^' is thus main-

tained in the composition only for

Kaum einen Hauch and Ruhest du auch,

that is, for the endings of the first and second parts. They

correspond to one another as half and full cadence (dominant

and tonic), a relationship further strengthened by the musical
rhythm: the open-ended rhythm
J. J> J> J
Kaum einen Hauch
 Lyric as Musical Structure 97

is answered by the closing

J J> i J
Ruhest du auch.

On the other hand, by joining together lines 1 and 2 the

musical setting of "Gipfeln/Ist Ruh" forms ww ', a succession '

not found in the poem. As a result, this passage too is placed in

relation to "Ruhest du auch" so that —
with the exception of
line 6, which stands by itself even this respect all the cadences —
of Schubert's setting correspond to one another:

Gip- feln ist Ruh

Kaum ei-nen Hauch
Ru- hest duauch.

In its rhythm and melodic descent,

"War - te nur, war- te nur

links up with

** sckwei-gen

and

$ WaF de

"Balde" has the same rhythm, but is directed upwards

melodically. The fluctuation in direction of this compound
gesture, mirroring the content of the poetic lines,

warte nur-warte nur-balde,

is supported in part by the accompaniment, which moves in

the opposite direction,
98 SCHUBERT
but in particular by the variable division into three units, as

mentioned above (4 + 4 + 4). The rhyme "Walde'7 "balde" is

rhythmic identity of the two words,

reflected in the

JlJvJlJ

in their melodic inversion

Walde-balde,
and in the sonorous brightening from

i
to

4
In addition, the fermata catches up the entire action from
"kaum" to "balde":

Kaum einen Hauch Die Voglein schweigen,

schweigen im Walde Warte nur warte nur balde

Through this inclusion of "Kaum einen Hauch" the second

section of the poem is joined to the first; and thus the entire
Lied up to the fermata on "balde" appears as a single, albeit
open-ended unit, as a single antecedent, as the meaning-laden
premise. If therefore even this structure, "Uber alien Gipfeln
. .balde," so differentiated in its parts and yet so monu-
.

mental, carries such weight, how heavy must be the closing

block! Yet this consists of a mere two half bars:

*=*
$ Ru - hest du, Mick
 Lyric as Musical Structure 99

Now at last b\>' is reached

the closing tone actually —
reached, whereas beginning of the Lied ("Uber alien Gip-
at the
feln/Ist Ruh") it was merely presented. First, however, the
melody descends once more to P,

^m
summarizes with this F and with the motive d"-c"-f\ which
isreminiscent of

Kauw tintn Uauch

and

5CHwa- qen

— the melodic activity of the entire poem, and brings it to a

close with the step

PB^
The fulfillment of the poetic message, "ruhest du auch,"
7
appears as the cadential completion (I^-V -!) of the Lied; the
weight of this passage is brought to our attention by the sound
of the contra-octave F, which enters here for the first time. The
specific structure of this counterpositioned closing block, how-
ever, comes from the bare, undisguised entrance of the dactylic
rhythm On the downbeat

which now — and only at this point in the Lied — stands all

alone: the heaviest weight, the most static element possible

100 SCHUBERT
within the rhythmic sphere. At the first "Ruhest du auch" the
static quality of this rhythm is still somewhat veiled by the fig-
ure in the accompaniment

r\i~ hestdu cuuk

CJ7

which, reminiscent of "balde," forms a transition to the repeti-

tion of the sentence, "Warte nur, balde/Ruhest du auch." In
this way the statement "Ruhest du auch" becomes milder,
more conciliatory; the contextual emphasis on "auch" is softly
accentuated (pointing to the calm of nature: "Uber alien Gipfeln
ist Ruh"). Only the second "Ruhest du auch" stands completely

by itself. But when the rhythm

iJ JiJ II

appears in the final bar, in the coda, it is com-

stripped bare,
pletely alone, absolutely at rest — like the immovable tomb-
stone above our final resting place.

Schubert has transformed the poem into a composition by

penetrating, as it were, through the poem and beyond it to the

deeper level that sustains it —

to the language —
and by drav/ing
directly upon this. In place of poetic form he has set musical
structure. At the same time, however —
since the linguistic and
poetic layers interact with one another —
his music allows the
poetic work of art to shine through and illuminates it anew. To
be sure, Schubert's music as a natural phenomenon reflects the
continuous flow, the babbling; but as meaning captured in mu-
sical structure it has adopted a method similar to language.
And precisely because the centripetal quality of the linguistic
form is foreign to the nature of music, this factor is expressly
singled out when it is realized in musical structure. Words (and
thus also their elements, the syllables), phrases, and lines ap-
pear as substances fitted firmly together, each with its own
 Lyric as Musical Structure 101

weight. The character of being expressly joined together

(corn-posed) predominates.
Since a poem is also language, it contains the centripetal,
distinguishing factor in its linguistic aspect. But it would be an
error to assume that this factor must always be given concrete
form in poetry by appearing as the determinant of structure at
the level of art. Thus the songlike forms of Goethe's lyric, and
the contemporary poems in general that Schubert preferred to
set to music, exhibit for the most part what we might call
smooth contouring. This finds expression in a uniformly flowing
progression of lines and sentences, in a forward movement
fashioned by the relationships of meaning in the language.
This movement, intercepted at the end of each line by the
rhymes, causes the poem as a whole to vibrate and generates its
tuning, its "mood." This is the atmospheric magic in the songs
of the "singer," the harpist, and Mignon in Wilhelm Meister.
Thus we find Goethe's Lieder "musical," probably because the
generation of mood is considered a specifically musical do-
main. And this is in fact justified —but only to the extent that
music is regarded primarily as something "natural." It would
be more appropriate to call such poems "musiclike" rather
than "musical," not so much because their character as a whole,
related as it is to the nature of music, can be easily borne "on
the wings of music," but rather because in them the centripetal
element of language is not advanced to a structural principle
even in poetry. Because of this the linguistic framework is not
defined in every respect, not completely established from every
angle; and this allows the possibility of composition, the possi-
bility of transforming linguistic lyric into musical structure.
There does exist, however, a poetry that takes up the cen-
tripetal element of language at the level of art and raises it to a
structural determinant. Holderlin's poetry is fashioned in this
manner, By realizing this element that is latent in the language
it renders the words corporeal, firmly joined together, com-
pletely determined from every angle. Precisely for this reason,
however, a poem by Holderlin cannot, in contrast to Goethe's
Lieder, be composed to music. For as a poem it has been
finished, as it were, down to the last detail. A procedure of this
kind expresses itself in the art of poetry as sharp contouring', the
 1 1

102 SCHUBERT
edges of the individual patterns are sharply incised, and rhyme
has no place in a poetic form of this kind. Goethe's descriptive
lyric is suspended within the overall mood, is born of it. But
Holderlin's poetry is characterized by its dispassionate tone. In-
dividual clusters of word patterns, complete in themselves, are
placed next to one another, and unity results from their being
joined together, "composed": the musical term now trans-
ferred to this kind of poetry and applied in its literal sense. The
first strophe of Holderlin's "Brot und Wein" is dominated by
images similar to those in "Wandrers Nachtlied": rest, night,
stillness, the peaks of the glade, mountain tops. But how
"singing" are Goethe's lines, how warmly flowing, bathed in
4
sentiment, and intimate; and by contrast how corporeal, how
mercilessly dispassionate, how real —
and thereby still glow-
ing —
are Holderlin's forms! They too suggest the mood in an
extremely forceful way; but this happens only incidentally, as it
were: "Ringsum ruhet die Stadt;/ Still wird die erleuchtete
Gasse."
If we attempt a recitation of "Wandrers Nachtlied," espe-
cially of lines 2-4, based on Schubert's setting in such a way,—
that is, that not only the rhythm but also the reality of Schubert's
Lied becomes apparent —
then Goethe's lines will appear as
1

though transformed into the language of Holderlin: "In al-len

— ;
1 1

Wipfeln Spurest du Kaum einen Hauch." Schubert's technique

1

in Wandrers Nachtlied of joining together through separation is

related to that of Holderlin. The one discovers it as music, the
other as poetry. Both are committed to the language, albeit in a
manner different from Goethe. Schubert is related to Goethe
by the warm uniform flow, the surfacing of his Lieder out
of the whole of the mood: through that aspect, then, which
is particularly encouraged by the "nature" of music, as we

have seen.
The free verse form of "Wandrers Nachtlied, " which avoids
a specifically songlike character,enabled Schubert to compose
it manner described above. He does not, however, realize
in the
the linguistic model as musical structure in the same way in all
of his Lieder. It is characteristic of Wandrers Nachtlied that the
—
accompaniment with the exception of the songlike passage
in bars 7—8 —
is not distinguished from the vocal part by means
 1

Lyric as Musical Structure 103

of an independent pattern: the individual phrases, their "sharp

contours," originate simultaneously in the voice part and ac-
companiment. There are no independent accompaniment fig-
ures running parallel to the song melody, as, for example, in
Das Wandern or Wohin? from the cycle Die schone Milllerin. But
even those Lieder with markedly songlike, usually strophic
3
texts are only apparently artless-natural.

Notes
1. See the letters of 22 April 18 14, 2 May 1820, and 4 September 183
in Der Briefwechsel zwischen Goethe und Zelter, ed. Max Hecker (191 3; repr.
Bern, Lang: 1970), vol. 1, p. 386; vol. 2, p. 57; and vol. 3, p. 470.
In Georgiades's original essay the first word of the title of Schubert's
song was spelled with an "e," Wanderers Nachtlied, as it appeared in the first
edition of 1828. In this translation the word is spelled without the initial "e,"
in conformity with Goethe's original poem and with more recent editions of
the song, including the old Gesamtausgabe and the Neue Ausgabe samtlicher
Werke [Ed.].
2. The 4 chord, introduced as a suspension of the dominant chord im-

mediately after the tonic triad in order to create a marked dividing point,
occurs frequently in the works of the Classical masters, and subsequently in
those of Schubert. (Compare, for example, Mozart's An Chloe, bars 3-4 and
14—15, or no. 27 from Le nozze di Figaro, "Giunse alfin il momento," bars
25-26, and many others.)
3. See Thrasybulos Georgiades, Music and Language, trans. Marie
Louise Gollner (Cambridge: Cambridge University Press, 1982), pp. 34ff.
(also i8ff. and 25).
4. more characteristic of Goethe's approach are the strophic poems,
Still

such as "Sommernacht" (from West'ostlichen Divan), which has a similar un-

derlying mood.
5. The cycle Die schone Miillerin is analyzed in detail from this point of

view in part 2 of Georgiades's Schubert: Musik und Lyrik, pp. 215-381 [Ed.].
 ——

Two Analyses

ARNOLD FEIL
Translated by Walter Frisch

I. Im Dorfe, from Winterreise

The principal difference between the two stanzas of Wilhelm

Miiller's poem is that the first comprises an external descrip-
tion of a sleeping village on a winter night ("Es bellen die
Hunde . .while in the second a subject steps forward and
."),
speaks, beginning abruptly with the imperative voice ("Bellt
mich nur fort !"). This person is an outcast, standing
. . .

apart from the village and its sleeping inhabitants. A wanderer,

he must continue on his way:

Es bellen die Hunde, es rasseln die Ketten.

Es schlafen die Menschen in ihren Betten,
1

Traumen sich manches, was sie nicht haben,

Tun sich im Guten und Argen erlaben:
Und morgen fruh ist alles zerflossen.
Je nun, sie haben ihr Teil genossen,
Und hoffen, was sie noch iibrig liessen,
Doch wieder zu finden auf ihren Kissen.
Bellt mich nur fort, ihr wachen Hunde,
Lasst mich nicht ruhn in der Schlummerstunde!
Ich bin zu Ende mit alien Traumen
Was will ich unter den Schlafern saumen?
[The dogs are barking, their chains are rattling.
People are sleeping in their beds, dreaming of much that
they do not have, delighting in good and evil. And to-
morrow morning everything will have vanished. Still,
 Two Analyses 105

they have enjoyed their share, and hope to find once more
upon their pillows whatever remains.
Let your barking send me on my way, you wakeful
dogs; do not let me rest in the hour for sleep! I am
through with all dreaming. Why should I tarry among
sleepers?]

From the two-part poem

Schubert fashions a three-part
song, of which the third corresponds musically to the first (see
the Appendix, ex. C). The more idyllic middle part, oriented
toward the sleeping inhabitants, is set in opposition to the
outer ones. But can this contrast be described more precisely?
The middle part has many distinctive features: (1) It lies in the
subdominant, which creates a distinctly softer effect. (2) Voice
and piano relate to each other in the manner of a traditional
Lied. (3) The vocal line is divided into shorter segments, whose
brevity is reinforced by the repetition of "Je nun" and "und
hoffen." (4) But above all, the agitated sixteenth-note motion
characteristic of the piano in the outer parts is abandoned. But
is the characteristic tension of this song due solely to these
rather obvious kinds of contrast? When we describe the con-
trast between the middle and the outer parts, have we said
everything there is to say —
or even the most important things?
The question cannot be answered until we undertake to
describe with more precision what happens musically in all
three parts —
until we examine compositional technique and
poetic content. Only then can it be determined whether the
opposition we have remarked was important to the composer
himself. Indeed, Schubert might have intended to set another
element in relief, for example the atmosphere of the winter's
night, with the sleeping village and the barking dogs. Analyses
of this song always suggest that the sixteenth-note figure of the
piano accompaniment is inspired by the rattling of the chains,
or even the barking of the dogs —
thus by an element of the
poem's atmosphere. I repeat: is that really important? There is
little doubt that the sixteenth-note figure is significant. But we

must consider further whether its primary function is to cap-

ture the atmosphere of the poem.
Listening to the piano prelude in relation to the whole
song, we discover that it is not a conventional, dispensable
piano introduction, placed at the beginning merely to pro-
106 SCHUBERT
vide a starting point, or to prepare the entrance of the singer.
Schubert eliminates the distinction between "prelude" and
"song" proper. Whether we can legitimately say that the pre-
lude runs into the song remains to be seen. At any rate, what
we hear is a compact succession of repeated chords, with an
almost trill-like figure in the bass, flowing into the same chord
in open position. At this point the music breaks off abruptly,
and there is a long pause before the succession of repeated
chords reappears in the second bar. Only in the following mea-
sures does the harmony become active; its goal is to move from
the dominant degree to the tonic, thereby leading back at bar 7
to a repetition of the opening measures. The harmony, how-
ever, is less striking than the persistence, indeed the obstinacy,
of the repeated pattern: agitated chords-long pause/ agitated
chords-long pause. Again we must ask what it is we really
hear, what stands out.
To express this with words requires the help of musical
notation, and of the appropriate description of that notation.
Let us note first Schubert's tempo indication, Etwas langsam,
which was in fact originally langsam. Etwas was added later by
Schubert: apparently langsam was too slow. What does Etwas
langsam really connote? What is it that really goes "somewhat
slowly" in this piano part? Even at a slow tempo the chords,
*2
thus the eighth notes of the 8 measure, follow each other
swiftly, the sixteenth notes still faster. In rehearsing and per-
forming the song, one comes to understand that langsam im-
plies only a regular ebb and flow within the overall motion; the
unit of breath, so to speak, is the half bar, not the twelve indi-
vidual eighth notes. If we attempted to conduct this, the result
would be a gentle motion in half bars, swinging between down-
ward and upward strokes; a strong attack comes on each down
stroke —thus on the downbeat. (By placing accent marks on
the chord in the middle of the bar in bars 1-4, Schubert sought
to bring into relief both the tiny melodic motion in the upper
voice and the change of chord from close, root, to open, in-
verted position; but by no means did he want to introduce a
pattern of accents different from that implied by the notated
meter.)
The motion pushes off from the downbeat, from the first
 Two Analyses 107

Example 1

Example 2

beat of the bar. The upbeat out of the long pause in the
arises
second half of the bar, then sinks down again on the first beat
of the next bar, onto the disquieting, agitated chords that
present the sharpest possible contrast to the basically tranquil
pulse of the piece. This contrast between surface rhythm and
underlying pulse is of deeper significance than the one ob-
served earlier between the three parts of the song.
An experiment reveals still another dimension of contrast.
If we play the prelude (either on the piano or on a recording)
without looking at the notes, and then attempt to bring in the
voice part, our natural tendency will be to follow the rhythmic
structure of the prelude: we will, I think, perform as in ex-
ample I. Should we continue to sing and play naturally, our
108 SCHUBERT
Figure 1

Upbeat metrical motion

t i t *

Voice
; H z^ZT_ -H
Piano i » i »

Downbeat metrical motion t *

error will become apparent only at "traumen sich manches,"

where voice and piano no longer fit one another.
What do we learn from this experiment? Schubert brings
in the voice where the listener neither expects it nor finds it
natural; it enters not as the "prelude" implies, but as in example
2. Schubert's intention will become clear to anyone who at-
tempts to begin in a "naturally" incorrect manner. Indeed, it is
very difficult to sing correctly, to make the proper attack, for
there is a very real opposition built into the musical structure.
The piano prelude suggests a downbeat orientation by the al-
most stereotyped repetition of its down-and-up motion; we
expect the entrance of the voice to continue this pattern. But
—
when the singer does not conform and Schubert wants it
so— the voice establishes a tendency of motion in direct op-
position to that of the piano. We can represent the musical ex-
perience graphically as in figure i and verbally as follows: al-
though united by the restful Etwas langsam of the music, voice
and piano articulate this motion in different, contradictory
fashions. Thus arises the impression of distinct levels of mo-
tion, in which one kind of rhythmic structure runs in opposi-
tion to another.
The whole first section of the song is dominated by the
opposition between a voice part that enters continually on an
upbeat (and an upbeat half a bar long!) and a piano part that
insists on the downbeats. The contradiction is not resolved but
gives in to compromise, so to speak, in bars 17-18, where the
right hand creates from the sixteenth-note figure(now lying in
the upper part) a melodic closing figure that appears to accom-
modate the voice.
As a consequence of this compromise, a certain rhythmic
 —
Two Analyses 109

tranquillity prevails in bars 18-19, which lead from the first to

the second part of the song. Both the underlying motion and
its surface divisions appear to be reduced to a rhythmic zero

point —
if such an extreme image may be permitted. The six-
teenth notes in the first of the two bars and the eighth notes in
the second flow unarticulated, in a vacuum, as it were. Thereby
emerges the possibility of introducing a new articulation, and of
allowing it to be perceived as really new. A new key, G major,
blossoms forth suddenly in bars 18-19 from the pedal tone D,
the tonic of the first part. In a similar way a measure emerges
jj

from the compromise between piano and voice (who now

seem reconciled). Although no change of meter is notated
it would not have been customary in Schubert's time it is —
clearly felt. measure (as against the long
In this short-winded jj

2
g
of the first part) the voice seems to be cradled within the
melody. The piano, having now given up its resistance, like-
wise revels in the simple accompaniment. In more technical
terms, this moment signifies a change in the relationship be-
tween piano and voice: the obbligato accompaniment of the
first way to a homophonic texture, in which the "ac-
part gives
companiment" has merely to fill out the harmonies implied by
the voice.
We have already remarked on the three-part form of this
song, a design that Formenlehre customarily identifies with the
formula ABA'. Had a closer examination of the A and B
parts established nothing more than the fact that they are dif-
ferent, wecould expect that the third part would be a simple
reprise of the first, possibly with a few deviations. But the dis-

tinction we have observed affects the musical structure so pro-

—
foundly that no repetition not even a varied one is possible —
at the conclusion. What is required is a careful musical working
out and resolution of the problems. If we are to take seriously
the experience of listening and our analysis of the results, the
characteristic tension between voice and piano, which are re-
lated to one another as obbligati, cannot be resolved as easily as
in the "compromise" of bar 17. The contrast of the middle part
is less one of atmosphere than of technique: its powerful effect

is based on a different texture (Liedlike accompaniment of a

tuneful vocal part), a different quality of motion (jj instead of

110 SCHUBERT
2
g ),
and (abandonment of the division into
a different structure
different "levels"). Moreover, at the beginning of the third part
tension once again permeates the music, the same tension we
encountered in the first part. The listener asks himself no —
matter whether consciously or unconsciously — how a piece of
music like this could really reach a conciliatory conclusion.
Although the poet avoids such an ending, Schubert at-
tempts to resolve the conflict by uniting the different elements.
—
He wants to indeed, he must, to assure the musical unity of
the composition —
bring into agreement the opposing types of
motion represented in figure i. He must therefore bring about
the result by a decisive act of will, as it were. Schubert achieves
his goal in this way: in bar 37, at the first "was will ich unter
den Schlafern saumen," the voice enters unexpectedly at the
tenth eighth note in the bar and in the middle of the upbeat.
Neither voice nor piano has hitherto begun at this point or
with this rhythmic pattern. It is as if the wanderer has torn
himself free from the picture, as if by contemplating the sleep-
ing village he has realized his situation, and now wants sud-
denly to change it. For the first time (except in the middle part)
the piano does not insist on its stereotyped sixteenth-note fig-
ure; for the first time
it is subordinate to the voice; for the first
time voice and piano are united within the obbligato accom-
paniment, and indeed, within a common, broad gesture of
motion. This moment arrives so suddenly — deviating so for-
cibly from the unity that has prevailed until this point — that it
suspends the metrical arrangement and disrupts the rhythmic
continuity. Schubert not only renounces the subdivision of the
bar into eighth notes; he changes the even arrangement of
2
pulses into an uneven one, by replacing the g bar, which is di-
vided structurally into eighth notes, with an uneven g8 bar,
whose unit value is the dotted quarter. The Etwas langsam mo-
2 8
tion is thus changed from 2 J. (g ) to 3 J. (g ) (see ex. 3).
A digression is necessary in order to answer a question
that has no doubt occurred to some readers: "Change of meter?
No change is notated in my edition!" In Schubert's time the law
of regular recurrence of stressed beats was still in effect. For ex-
ample, in a succession heavy-light/heavy-light (downbeat-
2
upbeat/downbeat-upbeat), thus in a 4 meter (or in our slow g
as well, in which the half bar provides the unit of pulse — there
 1

Two Analyses 1 1

Example 3

P
*Z- 2 J- 3J.

t Tj- I
J- 1" ^ PP
-was -will tcit un - Ur den&chla- fern sou -

is no question here of an uneven bar, or of a four-pulse bar),

the downbeats succeed each other at constant intervals; between
the strong "ones" there is always a weak "two." This funda-
mental law of musical meter developed to full strictness in the
eighteenth century. Schubert was among the first composers to
call it into question, as our Lied demonstrates. But notational
practice could not easily accommodate changes of meter. In
order to realize a certain unconventional rhythmic configura-
tion within the notated meter, Schubert would have had some-
how to ignore the immovable barline and make the new meter
apparent by other means. (For it is the barline that stands as the
visual representation of the metrical pattern of stresses.)
In order to indicate stresses that do not coincide with the
notated meter, Schubert uses either an accent mark or the dy-
namics fp or fz, or both together, prepared by a hairpin or a
written crescendo. So far so good. But neither every accent
nor every fp preceded by a crescendo suggests a change of
meter, though admittedly many serve to highlight a particular
rhythmic configuration that still remains entirely within the
notated .meter. In published versions of Schubert songs (even
in contemporary prints) Schubert's accent mark his favorite —
sign, which appears thousands of times is normally repre- —
sented as a decrescendo hairpin, which suggests a swelling and
subsiding of the dynamics. Nevertheless, Schubert's intentions
should become apparent, especially to those who observe and
listen carefully. What he had in mind was a particular rhythmic
phenomenon transcending the notated meter.
 —
112 SCHUBERT

The crescendo hairpin in the piano at the end of bars 37 and

42 (accompanying the word "was") is by itself meaningless: it
appears beneath an eighth note followed by rests. But in con-
nection with the corresponding signs of the next bar, it be-
comes an indication of a metrical shift; the rhythmic structure
of this passage thus becomes clear despite the fixed barline.
When other such places in Schubert raise some question about
whether the printed decrescendo hairpin is to be read as an ac-
cent mark or as a real decrescendo, the performer should let
himself be guided by the music into passive conducting mo-
tions, simple motions that reflect visibly what is being heard
2
or sung. The composer's intention will then immediately be-
come apparent. In our example, correct declamation is impos-
sible within the notated meter:

t 1 T i t 4
was will ich unter den Schlafern sau men?
If we disregard the barlines, however, and observe instead
Schubert's accent marks, there emerges another meter, one that
contradicts the barline but yields a proper declamation:

t 1 - t i I
was will ich unter den Schlafern sau — menr
The second "saumen" (bars 44-46) is, of course, surprising;
and in order to understand it we must return to our discussion
of the song.
All the rhythmic levels or surfaces intersect in the upbeat-
like 3 J- motion at "was will ich unter den Schlafern saumen?"
The union of the different levels, which contradicts the prevail-
ing metrical design, makes this a crucial moment in the overall
structure of the song.
To reinforce its importance, Schubert repeats the segment,
but not without further clarifying its original purpose. The
first time around, in the second 3 J- bar (bar 40), the stereo-
typed sixteenth-note figure breaks in, bringing with it the
earlier metrical arrangement, as well as the conflict between
rhythmic levels (see ex. 4). It is as if the action has not been
bold enough, its power insufficient to resolve the painful op-
position —as if the wanderer's determination to tear himself
away has not been decisive enough. After the downbeat, the
3 J- motion is broken off abruptly by a new downbeat, a new
 Two Analyses 113

Example 4

a S?W
p! J-

-W06
1J-
will
r
kit
^
un-
r
tor
pj- J -i
J
din&Maftrn sou
-

-
1
J
men?
** J
left
u
bin
J

$ jj.

p*£=fr 3±
±r
wm ?
«w mm
—^^SHSS^^^EE
^ i
vtj i

which appears in place of the proper upbeat or sec-

first beat,
ond beat. With one stroke Schubert brings back the former
metrical/rhythmic structure and cancels out the anomalous
segment (bars 37-39). The force of this reversal is enhanced
by the juxtaposition of two downbeats, violating the law of
musical meter, which prescribes regular alternation between
stressed and unstressed events. The listener experiences this as
the intrusion of a free, essentially nonmusical rhythmic ele-
ment into the prevailing articulation of musical time. This mo-
ment reveals to us both the individual strands of the rhythmic
fabric that binds them. The listener can
and the larger structure
thereby experience the musical process as a physical reality, as
musical motion.
We are made vividly aware that the conflict has failed to be
resolved, and that another, more forceful action is necessary to
bring about resolution.
The second time, the broader gesture of two complete 3 J-
bars (beginning with the notated bar 43 and the repeated "was
will ich") fuses the disparate elements, cancels out the conflict,
and brings about resolution. The wanderer has at last broken
free (ex. 5).
In order to create such a decisivemoment, the composer
draws on all the means at his disposal including, of course, —
melody and harmony. We experience all dimensions of the pas-
sage, even if it is analyzed from only one viewpoint specifi- —
cally from that of rhythm, or, more precisely, motion. We hear
the entire work even if our perception sometimes singles out
114 SCHUBERT
Example 5

^j.)

,J. | J. J. * ' IE
gjj f y p
« < •

ivas wiK kk un - ttr denStlnfo-hrn sau.

one element from the others. Thus, we would already have

heard the melodic change between bars 37 and 42, from G to
Qt (on the upbeat, "was"), as an expression of the most painful
anguish. The sensation becomes so intense that it not only
yearns for resolution, but appears literally to bring it about by
its expressive power. Nor will we have escaped the effect of the

color contrast between the B-flat/G minor of bar 37 and the

D major of the following bar, where "will ich" seems to be
flooded with bright light.
Indeed, there are few places in the literature of Classical
music like bars 43-46, in which the listener can feel a full ca-
dence as both a resolution of long-range tension and as a com-
plete harmonic closure. In this cadence even the suspensions
are savored, as it were. To be sure, this effect presupposes that
we have already heard the cadence once, in bars 38-40, where,
although harmonically complete, it was abruptly broken off.
We should also be aware that in the first part of the song and in
the portion of the third part corresponding to the first, the har-
mony changes only with each bar, and thus scarcely makes ap-
parent any progression. But here a succession of chords leads
swiftly to a cadence within only two bars: harmonic motion
comes all at once, and the progression sets in relief the rhythm
we have already examined. We can truly say that "harmonic
rhythm" determines the musical structure.
Tranquillity prevails at last. The rhythmic motion seems
to have slowed down, especially in the second cadence with the
3 J- bars, where the voice sustains an infinitely long A (actually
 Two Analyses 115

lasting a complete 3 <J- bar). All compositional elements seem

now to reflect this deceleration. For the first time the root of
the D-major chord is of the right hand (bar 46); for
in the top
the first time the chord remains in the same position for the
second half of the bar. The stereotyped sixteenth-note figure
of the left hand has lost all its restless agitation. The F# in the
bass on the tenth beat of bar 47 seems only a pleasant reminis-
cence of the upbeat articulation that had created such a discrep-
ancy in rhythmic structure.
Let us return to our starting point and to the problem of
song composition in general. Im Dorfe is not a strophic song.
But is it through-composed? Goethe disapproved of "so-called
through-composing" because it "completely destroys the gen-
eral lyrical character, and it necessitates and encourages a false
preoccupation with individual elements." According to these
criteria, our Lied is not through-composed, since a preoc-
cupation with many different details —
which is clearly what
—
Goethe had in mind is neither encouraged nor necessitated.
Rather, the whole song appears to be composed from a single
viewpoint.
The wanderer, at once fettered to and ostracized from the
village, finds his way out of the conflict by tearing himself
away. Schubert's song is constructed from this viewpoint; it
culminates in the moment of tearing away, that is, in an actual
event. The song is based on the musical realization of an occur-
rence (in the sense of an act rather than an action), on the com-
positional techniques that correspond to the singularity of the
wanderer's deed. The "general lyrical character" is of signifi-
cance insofar as it gives rise to the conflict we have described
between voice and piano; this presses toward decisive resolu-
tion in the act. To be sure, Schubert sees this tension not as a
prevailing condition; he composes it as a musical occurrence.
That is, he does not depict a "general lyrical character"; nor
does he "necessitate or encourage preoccupation with individ-
ual elements"; nor does he paint any specific "atmosphere" or
frame of mind. Rather, he develops the tension as an event in
the music, as reality in a narrower sense, by means of composi-
tional structure.
When Schubert realizes a lyric as amusical structure, he
interprets the poem from a definite point of view. But this
 "

Il6 SCHUBERT
is something from the "focal point" described by
different
E. T. A. Hoffmannconnection with Goethe's "general lyri-
in
cal character": "Inspired by a poem's fundamental meaning,
the composer must capture all its emotional elements as if in a
focal point, from which the melody radiates forth ["melody"
here synonymous with composition]. The tones become the
is

external symbol of all the different aspects of the inner emo-

3
tions contained in the poem."
As a poet, E. T. A. Hoffmann is thinking about the gen-
eral atmosphere in which the musician must bathe the whole:
themedium of music realizes what lies behind the words of the
poem. The basic mood must be developed to physical percep-
by the composer.
tibility
Such settings are rarer among Schubert's Lieder than we
might expect. Indeed, by these standards we might admit that
other composers —
perhaps Robert Schumann and Johannes
Brahms, and later Hugo Wolf and Richard Strauss wrote —
much more beautiful songs. But our actual musical experience
makes us ask: More beautiful? Now that is a different matter!
Consider Schumann's Mondnacht (op. 39, no. 5 [1840]; to a
poem by Eichendorff ) or Strauss 's Traum durch die Ddmmerung
(op. 29, no. 1 [1894-95]; to a poem by Otto Julius Bierbaum),
whose texts are comparable to that of Im Dorfe. These songs
are more concerned than Schubert's with depicting a nocturnal
atmosphere and the attuning of a soul. They are more color-
istic, and the musical procedure is thus entirely different. That

these Lieder differ in this way from Schubert's is clear; that they
are thereby more beautiful no one would seriously consider.
Schubert had a different goal from Schumann and Strauss:
therein lies the distinction. He was inspired neither by Goethe's
"general lyrical character" nor by Hoffmann's "focal point,"
which contains "all the aspects of the inner emotions" and
from which the Lieder of Schumann and Strauss "radiate forth.
Schubert, however, seeks to transform the lyrical element into
musical motion, to compose it as musical structure.

IT Moment musical in F Minor, op. 94, no. 3 (D. 780)

As a composer for the piano, Schubert "had his final say in the
short pieces of his last years. . . . It is their deeper, Schubertian
 Two Analyses 117

originality that distinguishes these pieces." Alfred Einstein,

who accords the Impromptus, op. 90 (D. 899), and the Moments
musicaux, op. 94 (D. 780), such significance, believes at the same
time that "it is easy to understand why they are his 'last word.'
For he was an inventive spirit, a composer of the spontaneous,
striking inspiration, and not one, like Haydn or Beethoven,
4
who could fashion something great out of a trivial idea." Yet if
we accept this latter statement, the first one will seem strange.
Is it really easy to understand just how these pieces achieve
greatness? Are they in fact distinguished above all by spon-
taneous, striking inspiration? Of what might this consist? We
shall investigate this question with regard to the Moment musi-
cal in F Minor, that "musical epigram, or 'Divertissement in
miniature,'" as Einstein called it. This work appeared as "Air
russe fur das Pianoforte" in an Album musical published by
Sauer and Leidesdorf in December 1823; early in 1828 it was
reissued by the same firm, together with other short pieces,
under the title Moment musical. (See the Appendix, ex. D, for
entire piece. In the following discussion, all bar references will
be to the score as printed in ex. D.)
In his investigation of meter and motive, Jammers has em-
phasized that a strong beat must be accented perceptibly at the
beginning of a composition so that a listener can learn how to
measure the passage of time. "His attention must be drawn pri-
marily to the meaning of the musical event. And the question
of meaning leads us to the matter of motive." 5 At the be-
ginning of our Moment musical no part of the bar stands out
strongly; nor at first is there any motive to provide temporal
articulation. The left hand begins alone, presenting an accom-
paniment in which a quarter-note pulse is articulated by eighth
notes, a pattern encountered frequently in movements marked
Allegro moderato. At this point the listener can discern neither
mode nor meter: he awaits the entrance of an upper voice to
clarify the indeterminate harmony and divide up the continu-
ous rhythmic motion.
The key of F minor and a 42 meter are established only in
bar 3, when the upper voice begins on the third of the triad
with a rhythmic figure that defines the bar. A grace note dis-
tinguishes the first eighth note from the third as the strong
beat; the second eighth note is defined as weak by a melodic
Il8 SCHUBERT

division into sixteenth notes, the fourthby a dominant har-

mony. Although projected unambiguously in the third bar, the
4 meter becomes obscured in the fourth. Here the two quarter

notes are rhythmically, melodically, and harmonically identi-

cal; and the accent marks on both prevent any dynamic differ-
entiation. Because of the accent marks, the stress that falls
naturally on the first of the two quarter notes must fall likewise
upon the second. Thus, while the third bar conforms to the
down-up motion of a \ meter, the fourth does not, because nei-
ther of its quarter notes can be subordinated to the other. Nor
does the accompaniment pattern delineate a \ metrical scheme.
Even if we were to represent the metrical stresses as in example
6a (where downward arrows indicate perceived downbeats, up-
ward arrows upbeats), we would scarcely capture the musical
gesture, because the two quarter-note downbeats are different
in kind from the downbeat of the normal \ bar that follows: the
two quarter notes of bar 4 are at once peculiarly rigid yet elastic.
Although they exhibit a remarkable precision of gesture and
movement, they do not join together to form any standard
metrical pattern. It seems as though they attain their identity
only as part of a two-bar group, which thereby establishes it-
self as the smallest unit of musical meaning.
More clearly than is commonly the case, the upper voice
is organized in distinct two-bar units; until bar 62, these are
 —
Two Analyses 119

organized into groups of four. Yet these groupings do not join

to form eight-bar units; rather, the groups of two function
so independently that the larger grouping appears neither as
an integral building block nor as a musical unity of a higher
kind, but actually as the sum of smaller discrete segments. The
eight-bar groupings maintain a certain degree of closure solely
because of the repetitions that are indicated for the first three.
At first we are inclined to interpret the small two-bar unit
as a motive. But this label is evidently inadequate. For, as
Riemann says, a motive is the individual gesture of musical ex-
6
pression; and our two-bar group comprises two different ges-
tures. According to the criterion of expression Riemann ad-
duces in defining motive, we would scarcely be tempted to
give each bar its own meaning; but according to the criterion
of gesture or motion, we would. The two gestures within a
two-bar group are movement: that of the
essentially distinct in
two quarter notes is fundamentally different from that of the
rhythmically subdivided bar. Moreover, the gesture of what
we may call the "quarter-note bar" appears to remain more or
less unaltered throughout the piece, while that of the bar with
different rhythmic values undergoes changes. Nevertheless
or perhaps precisely on that account —
we have the impression
that the two gestures complement each other, that the two bars
inevitably belong together. We tend to "hear" both gestures
together, if not as a motivic unit, then as a coherent figure
of motion, because the element of motion is so prominent.
One gesture seems to determine the other, just as a motion
in one direction creates a corresponding countermotion, and as
one step in a dance figure determines a second one correspond-
ing in space.
The whole period of the Moment musical follows this
first
pattern. Indeed, although the two-bar unit is continually var-

ied, one-can scarcely speak of motivic development. The com-

positional procedure is that of a series of units, or rather, a
series of different shapes derived from a single unit. Thus, de-
spite the element of variation, the repetitions continually fuse
the quarter-note bar and the rhythmically differentiated bar
into a single figure. The two bars of the
repetition of the first
melody (3-4) with scarcely a change in melody
in bars 5-6,
and rhythm, emphasizes that they belong together. So does
120 SCHUBERT
the following two-bar group (7-8), which is the only one to
—
vary the quarter-note bar thereby confirming its particular
character. The cadential motion in bar 9, as well as the rhyth-
mic activity,which imparts the quality of an upbeat, similarly
link bars 9 and 10, and at the same time all the corresponding
two-bar groups that preceded.
The constructed entirely from two-bar groups
piece is

that behave like the first two bars of the melody. The quarter-
note bar gives each group its characteristic motion. The ges-
ture, the direction of motion, differs depending on whether
the quarter-note bar comes first or second within the two-bar
group (see exx. 6a and b). One figure can even function as the
inverse of the other, particularly when harmony and melody
vary or reinforce the direction of the movement (compare bars
3—4 with 9-10 and 61-62). The different forms of the two-bar
unit, which are suspended along the thread of the continuous
quarter-note pulse of the left hand, function as a succession of
figures of motion: in the figures of bars 3-10, the second bar
contains the quarter note; in bars 35-40, it is the first bar. Each
segment thereby retains a definite, pronounced tendency of
motion (ex. 7). Their juxtaposition creates the effect of an
abrupt shift, a change in movement —
as, for example, in bars

18-19, where for the first time two quarter-note bars are placed
side by side (ex. 8), or in bars 39-42, where the succession of

Example 7

(*> 3

f\» >r$n if 4 i>jij1

n
s £ **

->!-
*JfL

pa
k *
ag Aft*
\<-
 Two Analyses 121

Example

pilgP w s@
SB ^
t
M i*^* ifp
1

Example 9

j"i ''
ir^ijii p*
»s ttc

Pi pp
Example 10

$
stfe
s^w pp ^
''^OTTOJJJi
25 > .>
Mm j^ pp
H 1p£=£
m * t
Psrt
PPP P uTJhrirj
122 SCHUBERT
units beginning with a quarter-note baris concluded by a unit

with the quarter-note bar in second position (ex. 9). Further-

more, when figures with different directional tendencies are
joined together, something new develops: in the second part of
the melody (bars 19-34), "opposed" two-bar figures are linked
together to form four-bar ones (ex. 10).
The result is a kind of game played on many levels with
musical-rhythmic figures; from this arises the dancelike quality
of motion in the piece. If we were to realize the quarter-note
bar in dance as emphatically stamped, and the other bar as
dissolved in movement, we would understand the particular
quality of motion in both the individual figures and the whole
composition: although the gestures change, equilibrium is
maintained by the larger flow.
The composition thus appears to be constructed so that
both the bar and rhythmic-motivic development are super-
seded as indicators of temporal articulation by a third factor:
the bar group, which divides up the continuous quarter-note
pulse of the rhythmic background and thereby cancels the in-
dividual bar as the unit of time and as the principle of articula-
tion at the lowest rhythmic level. It is the group that binds
these segments into musical-rhythmic figures. The construc-
tive element of the bar group thus fulfills simultaneously a
time-measuring and a time-articulating function. Because of
the compound gesture and because of the way it is transformed,
the bar group has a distinct character of motion as a rhythmic-
musical figure in the piece. Thus the structure of the composi-
tion seems to be determined in a particular fashion by the ele-
ments of motion, and the musical rhythm by the rhythm of
motion.
This aspect becomes especially apparent toward the end.
In the coda (bars 59ff.), Schubert avoids linking the more ac-
tive bar to the quarter-note bar. He repeats the first bar by it-

self; thathe repeats in isolation the gesture which up to now

is,

was completed by a second one. As great as is our expectation

—
of the closing gesture, still greater when it in fact arrives is —
our impression of how different the two gestures are and, at the
same time, how they belong together in a single figure. Bars
65-68 appear neither as a single four-bar unit nor as two-plus-
 Two Analyses 123

—
two; rather, they consist one can actually hear this taking
shape —of a single bar repeated three times, plus an additional
bar necessary for completion. This reveals that both bars are
"single." But that they nevertheless, indeed necessarily, belong
together is shown by the subsequent repetitions of the two-bar
group they form together in bars 69-70, 71-72, 73-74, and
75-76. The composition then concludes with the second, an-
swering gesture: now that our ears have had a sufficient dem-
onstration of both independence and interdependence, the sec-
ond gesture is also detached and repeated (bar 77). This leads
back again, as it were, to the easily flowing, undifferentiated
pulse of the beginning, which had established the rhythmic
foundation.
We have seen that the most immediately striking aspect of
inspiration in this composition is to be sought more in its
qualities of motion, in its gestural dimension, than in the realm
of melody and rhythm, where the word "inspiration" is usually
applied. To be sure, the individual gesture is also rhythmically
and melodically significant; but more important are its element
of motion, its terseness and plasticity, and an insistence that
yields an almost physical effect. This effect apparently presup-
poses the composer's division of the rhythmic dimension into
different levels, since in that way it is possible for one kind
of motion genuinely to stand out against the others. Thus
Schubert begins by laying a rhythmic foundation: the left hand
commits itself to neither a clear meter nor a definite kind of
motion. One might say that the broken quarter notes have the
quality of pure process; the basic rhythm is so severely limited
that its uninterrupted pulsation is not actually perceived as a
basic "motion" at all. The continuous pattern in the left hand,
avoiding any rhythmic profile and articulation, as well as any
distinct quality of motion, thereby constitutes a thread run-
ning through the composition, a "filo," along which the com-
ponent phrases appear to be strung.

Until the means of analyzing musical structure are refined,

it seems appropriate to restrict oneself to works of simpler
design, and to adduce complicated structures only if a par-
ticular characteristic can be easily perceived. The experience
124 SCHUBERT
gained through simpler works can later though still with —
great caution —
be applied to larger ones. Nevertheless, let us
risk casting a glance from our Moment musical to the Andante
con moto of the "Great" C-Major Symphony (D. 944). Al-
though the structure is incomparably more complicated here,
we can still see an affinity between certain elements. The re-
markable "introduction," the quarter-note bars that permeate
the melodic-rhythmic structure, and Schubert's way of using
bar groups as building blocks —
all clearly show (despite the

higher level of formal organization and the difference in genre

and status) that the two pieces share premises of phrase con-
struction and presentation. Although they differ in tempo, the
works correspond in their manner of motion. As Grove re-
marked in his renowned dictionary article on Schubert, "the
symphony was written as an absolute impromptu." 7

Notes

1. Muller's original line 2 reads, "Die Menschen schnarchen in ihren

Betten."
2. Such experiments, in which one learns how a piece "goes" by using

motions like those of conducting, were proposed by the great pedagogue

Leopold Mozart as early as 1756 (the year of his son's birth): "Therefore no
pains must be spared when teaching a beginner to make him understand time
thoroughly. For this purpose it will be advisable for the teacher constantly to
guide the pupil's hand according to the beat, and also to play to him several
pieces of different meters and varied speeds, allowing him to beat time him-
self, in order to prove whether he understands the division, the equality, and

finally the changes of speed" (A Treatise on the Fundamental Principles of Violin

Playing, trans. Editha Knocker, 2d ed. [London: Oxford University Press,
1951], p. 33). Similar advice comes from the important eighteenth-century
theorist Johann Philipp Kirnberger: "Therefore the inexperienced composer
is advised first to sing or play the melody that he has in his head and wants

to write down, and to beat the time with his hand or his foot. In this way
he will not miss the principal notes that fall on the downbeat, provided that
the melody is metric" (The Art of Strict Musical Composition [1776], trans.
David Beach and Jurgen Thym [New Haven: Yale University Press, 1982J,
pp. 388-89). See also my Studien zu Schuberts Rhythmik (Munich: Fink,
1966), pp. 17-21.
3. In a review from 1814. See E. T. A. Hoffmann, Schrifien zur Musik,
ed. Friedrich Schnapp (Munich: Winkler, 1963), p. 238.
4. Alfred Einstein, Schubert: A Musical Portrait (London and New York:
Oxford University Press, 195 1), pp. 288-89.
 Two Analyses 125

5. Ewald Jammers, "Takt und Motiv: zur neuzeitlichen musikalischen

Rhythmik," Archiv fur Musikwissenschaft 20 (1963): 199.
6. See Hugo Riemann, System der musikalischen Rhythmik und Metrik

(Leipzig: Breitkopf und Hartel, 1903), pp. viii and 13-18.

7. George Grove, "Schubert," in Grove's Dictionary of Music and Musi-
cians, 3rd ed., ed. H. C. Colles (New York: Macmillan, 1928), vol. 4, p. 613.
 —

Aufdem Flusse: Image and

Background in a Schubert Song

DAVID LEWIN

I propose here to explore the relation of musical structure to

textual imagery in Schubert's song Aufdem Flusse, from Win-
terreise. The
exploration will have several parts. Section I devel-
ops a general critical stance toward the relation of music and
text in Schubert's songs, a stance that will underlie the subse-
quent critical and analytic discourse. Section II offers a reading
of the text for this specific song. According to this reading
which I am of course claiming to be Schubert's, on the basis of
his setting —the text is in a sense "about" the creation and eval-
uation of a poetic image; the essential point is to take the two
concluding questions, "Mein Herz, in diesem Bache/Erkennst
du nun dein Bild?" and "Ob's unter seiner Rinde/ Wohl auch so
reissend schwillt?", as undecided, not rhetorical. The climactic
musical events over the second half of the song, which sets
those questions, reflect and project the tensions experienced by
the singer as he contemplates the answers.
Section III examines some of the ways in which various
aspects of the musical texture combine to suggest and elaborate
the reading of section II. From a theoretical view, the aspects
touched on are of a traditional sort: the length of musical sec-
tions vis-a-vis text sections, tonality, modality, the relations of
the vocal line to the soprano and bass lines of its accompani-
ment, and motivic rhythms. Novel, perhaps, is the considera-
tion of such matters in connection with the critical stance of
section I, and the sort of reading asserted in section II.
Section IV goes systematically deeper into the tonal and
 "Aufdem Flusse" 127

rhythmic structure of the music. This exploration gives rise to

reductive sketches which present in themselves striking im-
ages, images that appear to relate forcefully to the imagery of
the text and the questions concerning it. The relation of my
reduction technique to Schenkerian theory, and of my results
to Schenker's published sketch for this song, are taken up in an
appendix. Section V
attempts a consistent and dramatically
cogent interpretation of the images and questions raised by the
sketches of section IV, as relating to the images and questions
of the text.

I. A Critical Stance:
The Composer as Actor

Whatever filled the poet's breast Schubert faithfully represented and

transfigured in each of his songs, as none has done before him. Every
one of his song compositions is in reality a poem on the poem he set
to music.
1

Josef von Spaun's eulogy of 1829 fairly represents the critical

style of Schubert's sympathetic contemporaries. Today we feel
more comfortable with critical discourse of a kind that points
to networks of specific events in specific compositions. Still, it
would be a mistake to dismiss Spaun's prose as unsubstantial
and unhelpful. If we read it closely, we will find him asserting
propositions that bear directly on the methodology of analysis
and criticism.
One might put these propositions as follows. A Schubert
song takes as structural premises not only musical syntax, as it
was understood at the time, but also the structure of the indi-
vidual text at hand. The world of the song, then, is not simply
a musical world. On the other hand, it is also not simply the
textual world translated into music: it not only "represents"
this world, says Spaun, but also "transfigures" it. So, if we
have as text a poem on X, we should not consider the song to
be another, related poem on X. Rather, the song should be
considered a poem on the poem-on-X.
Hence we can understand the song as a poetic "reading"
of the poem-on-X that is its text, a reading that employs a par-
ticular mimesis of X
as a representational means. From this
 "

128 SCHUBERT
point of view, I suggestive to conceive the relations of
find it

composer, text, and song

as analogous to the relations of
2
actor, script, and dramatic reading.
To exemplify the utility of this analogy, consider the fa-
mous pause in Die Post. The opening text reads: "Von der
Strasse her ein Posthorn klingt. /Was hat es, dass es so hoch
aufspringt, / Mein Herz? / Die Post bringt keinen Brief fur dich.
Schubert extends the second sentence by repetitions of the
text, going through a substantial harmonic excursion and re-
turn, using an ostinato springing rhythm throughout. There
then follows an abrupt silence for one bar, before the song con-
tinues, "Die Post bringt keinen Brief," in minor and with the
springing rhythm gone. These gestures involve a mimetic
reading of the text by the composer, a reading that is far from a
simple musical translation or mere "representation," as Spaun
puts it. Imagine an actor performing the text in question as a
script. According to Schubert's reading, the actor would elabo-
rate and extend the action associated with the second sentence.
He might pace back and forth, rush to the window and back,
and so on. Then, as the postman approaches the door or the
mailbox, the actor would stop moving during Schubert's pause,
tensely cocking an ear to listen for the sound of the letter drop-
ping. He would relax his features in dejection as the postman
leaves, and continue the recitation: "Die Post bringt keinen
Brief."
Contrast this performance with that of another actor, who
is seated at a desk on stage, immersed in reading, writing, or
composing. He looks up, momentarily distracted, and notes,
"Von der Strasse her ein Posthorn klingt." He resumes his
activity for a while, then breaks off in ironic amusement and
says to himself, in one phrase, "Was hat es, dass es so hoch
aufspringt, mein Herz? —
Die Post bringt keinen Brief fur
dich" This second reading of the text is surely as plausible as
the first. In fact, it would not be difficult to argue that Schubert's
setting involves reading into the text a good deal more than the
second actor has to. 3
So while it would be accurate enough to say that Schubert's
reading "represents" the text, one cannot go very far critically
until one investigates how this particular representation, from
among a number of plausible readings, interacts with musical
 "Aufdern Flusse" 129

structure to project an overall poetic conception of the poem

that is the text. In this regard, for instance, one thinks of the

rhythmic complexity of Schubert's composition, with its con-

trasts of expansion and contraction, of regularity and irreg-
ularity, of ostinato clock time, musical-phrase time, and text-
line time. One notes the importance of exact rhythmic and
metric proportion in the effect of the first actor's scene: his "in-
ternal clock," representing his heart in the text, enters into
complex relationships with external clocks of the actor's outer
—
world that is, the postman's springing horse, his rounds, etc.
The second actor's scene has much less of an exact mensural
character; it could be delivered much more "in its own time."
view I have just proposed, the relation between
In the
Miiller's poem and Schubert's setting is formally analogous to
that, say, between Shakespeare's Hamlet and Henry Irving's
Hamlet. One could not sensibly analyze or criticize Irving's
Hamlet without referring to Shakespeare's, but it is important
not to identify or confuse the distinct artworks.

77. Schubert's Reading of the Text

The text of Aufdern Flusse follows the poet's creation of a cen-

tralimage, and his reaction to it. One can fairly say, as I put it
in my preliminary remarks, that Schubert's reading is in a sense
actually "about" the poet as image maker and image ques-
tioner. A condensed synopsis of the poem will help clarify the
point. The text of the poem can be found in table 1, on pp.
132-33. The song is printed as example E in the Appendix.
In stanzas 1 and 2, the poet observes that the stream,
which used to rush in a wild bright torrent, has become still,
cold, and rigid, with a hard stiff cover ("Rinde"). In stanzas
3 and 4 the poet, as if idly, scratches upon the ice the name of
his lost teloved, along with the dates of their first meeting
and their final separation. He circumscribes the whole with a
broken ring.
In stanza 5 the poet is struck by theimage he has "inadver-
tently" created and investigates its pertinence by posing two
questions, rhetorically. The first ques-
which might be taken
tion, insofar as can be taken rhetorically, points out in the
it

picture of the broken ring, guarding within it the name and the
 —

130 SCHUBERT
dates, the image of the poet's broken heart. The second ques-
tion, insofar as can be taken rhetorically, points out a further
it

and potentially optimistic aspect of the image. Just as the ver-

nal torrent of stanzas 1-2 is still rushing on beneath the icy
surface of the stream, so the poet's heart is still swelling tu-
multuously beneath its frozen crust.
I contend, however, that Schubert's reading takes the two

questions at face value, rather than rhetorically. The first ques-

tion asks, "Mein Herz, in diesem Bache/Erkennst du nun dein
Bild?" Read rhetorically, this does indeed assert the pertinence
of the image; taken at face value, however, it questions the perti-
nence of the image. Is the inscribed stream truly the image of
my heart? Taken at face value, the question also calls into issue
the poet's capacity to judge the propriety of the image. Does
my heart recognize and credit the image, perceiving or judging
it as apt? (These nuances can fairly be read into "Erkennst du

nun." N.B.: Erkenne dichl = "Know thyself")

The second question asks, "Ob's unter seiner Rinde/Wohl
auch so reissend schwillt?" An attempt to take this question on
face value seems at first ridiculous: the poet, presumably a stu-
dent of natural science in the spirit of Goethe, must know that
rivers flow beneath their ice in wintertime. Perhaps the stream
is small enough to cast the issue in doubt. But this is beside the

point. It is clear that neither the geological structure of the

stream nor the biological structure of the heart is essentially at
issue.
Having said that, one becomes struck by the use of the
word "Rinde." The word is in fact a common biological term,
used to denote the cortex of the heart. So it is the use of "Rinde"
in connection with the stream, not the heart, that is meta-
phorical. Through this metaphor, the relation of subject and
image is inverted. In the first question, the stream was the
putative image of the heart; now the heart with its "Rinde" is
the putative image of the stream. This device was prepared by
the earlier, more descriptive, reference to the "Rinde" of the
stream in stanza 2.
At that time, the stream was "du"; the very opening of
the poem, in fact, calls attention to the stream in the as a —
second person. By the time we get to the first question in
stanza 5, however, there has been a transformation of persons:
 '

"Aufdem Flusse" 131

the heart has become the second person. When we get to

the second question, we have just heard "Erkennst du nun
dein Bild?" addressed to the heart. Then, since the Rinde, in
usage, belongs to that second-person heart, it is properly
literal
"deiner" Rinde, not "seiner." From this point of view we can
regard the inversion of subject and image, discussed earlier, as
a means of transforming and hence avoiding the implicit ques-
tion, "Ob's unter deiner Rinde/ Wohl auch so reissend schwillt?"
The transformation of second and third persons over the poem
also abets this interpenetration of subject and image.
In Schubert's reading, then, I say that the second question
is taken to ask: Is there any capacity for flowing torrential

warmth left under the frozen exterior of the heart? Or is it

frozen solid, through and through? A question not to be asked!
(And it is not asked . . yet it is.) The stream will melt next
.

spring, returning to the state described in the opening line of

the poem. The subglacial flow of the stream, if existent, is a
portent linking its future with its past. But will the poet's heart
ever thaw, returning to a state "des ersten Grusses"? Is there
some subglacial flow within it that portends such an eventual
thaw? And thus, picking up the first question again, is the
stream, which will thaw next spring, a true image of the heart?
It is the tensions underlying this reading of the questions

that, as I suggested earlier, force the expansion of the setting of

stanza 5 over the entire second half of the song. In this connec-
tion, it is interesting to note that Miiller's text contracts at this
point rather than expands. Stanzas 1 and 2 are paired by their
subject matter, and so are stanzas 3 and 4; stanza 5, however,
which ends the poem, has no partner. Instead it asks two ques-
tions. The pair of questions thus substitutes, as it were, for a
pair of stanzas; this contracts the time involved by a factor of
two. One might say that Miiller's text ends with an unresolved
systole, for which Schubert's song substitutes an enormous
diastole.

III. The Song: Mimetic Techniques

This diastole can be observed in table 1, which plots some of
the easily perceivable aspects of the musical setting against the
coextensive text. In clock time, the crucial point at bar 41
132 SCHUBERT
Table 1

Stanza Strophe
Bar Text Number Number
5 Der du so lustig rauschtest, 1 1

du heller, wilder Fluss,

wie still bist du geworden,
giebst keinen Scheidegruss!

14 Mit Rinde
harter, starrer 2 2
hast du dich uberdeckt,
liegst kalt und unbeweglich
im Sande ausgestreckt.
23 In deine Decke grab'ich 3 3
mit einem spitzen Stein
den Namen meiner Liebsten
und Stund' und Tag hinein:
31 Den Tag des ersten Grusses, 4 4
den Tag, an dem ich ging:
um Nam' und Zahlen windet
sich ein zerbroch'ner Ring.

41 Mein Herz, in diesem Bache 5 5

erkennst du nun dein Bild?

48 Ob's unter seiner Rinde (5) 6

wohl auch so reissend schwillt?
54 Mein Herz, etc. (5) 7

62 Ob's unter seiner Rinde, etc. (5) 8

comes more than halfway through the song. But in strophic

time, the four strophes that set stanza 5 after this point are
equivalent to the four strophes setting stanzas 1-4 before it. In
a sense, this musical diastole is consistent with Muller's textual
systole. For one can say that the music expands with respect
 "AufdemFlusse" 133

Relation Right-hand
of Voice Rhythmic
Duration Key to Piano Motive
9 bars E minor Voice
doubles
v y J)
J)
bass
(lowE).
9 bars

8 bars E major Voice (G#)

takes over
y
from RH ^ nv<
(Gli); now
10 bars
RH doubles
voice.
(8 + 2)
*J* JEJHm

7 bars E minor Voice (B)

y v
J^ $
rises above

6 bars G-sharp both hands.

minor
y -5S y S W
8 bars E minor Voice
reaches y ^ y J)

high E;
continues
8 bars
in highest
register, y J3? JH
above both
hands.

to the text it sets or, taking a relativistic view, one can say that

the text contracts with respect to the amount of music that sets
it. The latter formulation is consistent with the reading of sec-

tion II, which had stanza 5 contracted with respect to its emo-
tional and structural implications.
134 SCHUBERT
In any case, Schubert, as he approaches the crucial text
questions, makes use of temporal contraction in another way.
I am referring to the rhythmic scheme that governs the dura-

tions of successive strophes. As appears from the "duration"

column in table i, the process of contraction leads directly and
unambiguously to strophe 6, the climactically compressed
strophe that first sets the climactic final question of our read-
ing. (The two-measure extension of strophe 4, as the maggiore
ends, does not disturb the sense of the rhythmic scheme.)
Strophe 6, as goal of the contraction process, is consistent with
what I have asserted to be Schubert's literal, rather than rhetori-
cal, reading of the text questions.
Strophe 6 takes another kind of strong accent because of
the key of its opening. In this sort of loose chaconne/pas-
sacaglia/variation structure, strophe 6 is the unique strophe
that does not begin with a strong downbeat on E minor or
E major. It begins with the usual strong downbeat, but in a
foreign key. The relation of G-sharp minor to E major is clear
enough: the G-sharp harmony represents iii of E major and as
we shall see later, the Gjl in the bass at bar 48 can fit very con-
vincingly into a structural arpeggiation of the E-major (!) triad
over the bass line of the song as a whole. These functions put
strophe 6 into the realm of E major, the key associated with
happy memories of earlier times, perhaps springtime, in stanzas
3-4. This tonal function for G-sharp minor thus supports a
potentially optimistic answer for the concomitant second text
question. If indeed a secret E-major deep structure lies unterder
Rinde of the E-minor surface structure, then the poet's heart
does indeed preserve its capacity for warmth and the return of
a vernal state.
Of course, the song is not in the major mode, which is to
say that the optimistic answer to the questions is ultimately un-
tenable in Schubert's reading. Though, as we shall see pres-
ently, it is not so easy to show convincingly why the song is not
in fact in the major mode once one begins to examine its
deeper tonal structure, which hinges not only on the bass of Q
bar 48, but also on the powerful vocal Qt at bar 23. To answer
that "why," we shall have to explore what is wrong with the
image the poet has constructed, what militates against the
E-major image suggested by the deep-level structure.
 "Aufdetn Flusse" 135

But we must defer study until sections IV and V. Here

this
we can note that while the Qf does indeed put the second ques-
tion into the realm of the happy, warm E-major memories, the
tonal material of strophe 6 also, and ambivalently, refers back
to the icy, immobile E-minor world of the opening strophes.
The G-sharp minor world of the sixth strophe specifically re-
calls the icy harmonies of bars 9-12 and 18-21. Those mea-
sures encompass a most desolate part of the E-minor world;
the coextensive text includes the words still geworden, halt,
. . .

and unbeweglich. Here (in bars 9-12) the tones AH and D# are
locally diatonic, as they also are in bars 48-51. So, as it were,
we have been warned that those locally diatonic tones can very
easily elaborate and return to E minor in this song. With a bit
of stimulation, in fact, one can hear how the melodic structure
of the bass line in strophe 1 is transformed into the melodic
structure of the vocal line in strophes 5 and 6. Figure 1 provides
the stimulus. The
reader should not be distracted by the har-
monic will be discussed soon. The main point
notation, which
of the figure is to suggest the melodic transformation of the re-
duced bass line, bars 5-14, into the reduced vocal line, bars
41-54.
From this point of view, the bass of strophe 6 does not Q
function as a root representing the major third degree of E.
Rather, it functions as a means of providing consonant support
for the structural tones B and Djf in the vocal line. And figure 1
makes it clear that B and Dff strongly project the dominant
function in E minor. Riemann's notion of "dominant paral-
lel"
— —
"(D + ) p " furnishes a good label for the G-sharp chord
in this connection, as figure 1 shows.
The (D + )p
analysis of the harmony, in asserting dominant
function, implicitly rejects an alternate Riemann analysis of the
harmony as ^+ , an analysis that would assign the harmony

Figure 1

J=bar
5

y fJrwtJ i i
r "''fr^rr WcJcTiuir
v +
e: T D D T td:
136 SCHUBERT
tonic function. The rejected analysis would be the way to assert
in Riemannian terms the idea that the bass Qt represents the
+
third of a structurally prior E-major tonic. Instead, the (D ) p
analysis asserts for the bass Qt a function as the under-fifth of
the voice's Djt, the tone that is in fact the Riemann root of the
harmony. And that root Djt of the asserted (D + ) p functions
as the third of a structurally prior B-major harmony, a har-
mony that is in turn the major dominant of a structurally prior
—
E-minor tonic understanding "E minor" in our sense now,
notRiemann's.
The use of G-sharp minor in strophe 6 is strongly quali-
fied, then, not only by a potential optimistic E-major arpeg-
giation lying unter der Rinde of E-minor surface events, but also
by the recollection of the icy Djt and its ultimate root B that lay,
halt and unbeweglich, unter der Rinde of the right hand, during
the E-minor events of strophe 1. In this way, Schubert projects
and reinforces his ambivalent reading of the second text ques-
tion: What does lie unter deiner Rinde — the E-major world of
strophes 3-4 or the E-minor world of strophes 1-2? The use
of tonal and modal ambiguity here, as well as the progressive
contraction of strophe lengths, makes it impossible to read the
text questions rhetorically. In this connection, note how the
forte dynamic at bar 48 bursts out from a song that had been
pianissimo and pianississimo up to bar 41, and had risen only to
piano at that point. One does not sing such a forte rhetorically.
We have so far observed that strophe 6 is a climactic goal
for a number of musical processes. We have observed in this
connection its ambivalence in referring both to the E-major
and E-minor worlds, and we have associated this ambivalence
with a structural indecision as regards optimistic or pessimistic
answers to the questions of the text.
These views of strophe 6 are further confirmed by the de-
velopment of the right-hand rhythmic motive, which is sum-
marized in the last column of table 1. Once again, strophe 6 is
the climactic goal of a process. And, as with the G-sharp to-
nality, the motive form of strophe 6 has ambivalent references
to the two worlds of the song. On the one hand, the thirty-
second notes pick up and continue the process of quickening
that began with the sixteenths of strophe 3 and continued with
the sixteenth triplets of strophe 4; hence we could read the mo-
 "Aufdem Flusse" 137

tive form of strophe 6 as even more torrentially flowing, be-

fitting an optimistic answer to the question. On the other
hand, the thirty-second notes are interrupted by a syncopating
rest on the third eighth of the motive: in this respect the mo-
tive form resembles the shuddering forms of the pessimistic
strophes 1 and 2, just recalled in strophe 5, rather than the
smoother, more flowing forms of strophes 3 and 4. The poet's
heart is certainly pumping furiously over strophe 6, even faster
than it was during its warm beating over strophes 3 and 4; yet
the heartbeat here is perhaps more a raging palpitation than a
4
swelling flow. The return of the thirty-second notes in the
final strophe seems to confirm a pessimistic reading for the
motive form, though this outcome is, of course, as yet unheard
during strophe 6.
Finally, it is illuminating in this context to examine the
doubling relations of voice, right hand, and bass line. These
relations are sketched in the penultimate column of table 1.
During strophes 1 and 2, the right hand incessantly reiterates
3-2 in E minor, kalt und unbeweglich. Meanwhile, the voice
doubles the bass line, which avoids prominent third degrees.
The singer seems to be saying, "I am, at this point, only a bass
line, unter der Rinde; I have nothing to do with that terrible
minor Kopfion in the right hand."
Consistent with that attitude, the singer thereupon takes a
sharp stone and scratches a Q
on the icy surface of the right
hand, at bar 23. Using that Qi, he substitutes a reiterated major
3-2 for the earlier minor 3-2 over the next two variations
of the strophe. The right hand, transformed by the image-
making activity of the poet, momentarily follows him along,
abandoning its own preferred minor Kopfion while, so to speak,
the G# image is warm. (It will freeze over again presently.) The
singer has insisted on G#, not Gti, as the structural Kopfion for
the melody, a feature that all the doubling relations so far em-
phasize. (While the melodic activity of the third and fourth
strophes is more florid than the unadorned G— Fjl gestures of
strophes 1 and 2, it is not hard to hear the underlying Qt-FH
gestures that form the basis for that activity. The underlying
3-2 will be made clear in the reductions of section IV.)
At bar 38 the voice, detaching itself from the right-hand
part, leaps up to B at the cadence, rather than settling on Fjl
I38 SCHUBERT
once more. The rising pitch, coming at the end of all this mate-
rial,implies a question; it thus foreshadows the explicit ques-
tions coming up in the text. The poet silently contemplates the
image he has made, as we listen to the extended fluttering of
his heart in the rhythmic motive of the piano. Then every-
thing, even the heart itself, is silent for one measure.
Despite the tremendous musical climaxes later on, bar 40
can be taken as the dramatic climax of the composition. When
E minor returns at bar 41, with the right-hand motive of the
opening, one senses a tragic catastrophe ahead. It is as if the
singer, during the silence in bar 40, had already asked himself
the questions and answered them negatively. This sense is
made all the stronger by the fact that, while the left hand be-
gins to sing like a cello unter der Rinde of the right hand at bar
41, the singer himself temporarily remains silent. He is still
geworden and unbeweglich; he cannot double this bass line, as it
flows warmly on beneath the icy surface of the right hand. His
heart ( "mein Herz") is not like the riverbed flowing on beneath.
ASchenkerian approach to the large-scale tonal activity
hereabouts is revealing. If at bar 41 the singer were indeed to
—
double the bass for example, by singing "mein Herz" on
low B and E across the barline of that measure —
then the tonic
degree E could be heard as closing a Schenkerian Ursatz, and
furthermore closing that structure in the major mode. Each
strophe so far has elaborated the essential melodic gesture 3-2,
in minor or major, with tonic-dominant support. A vocal E,
then, at the barline of bar 41 after the vocal Gtf-F# gestures of
strophes 3 and 4, would provide a structural melodic first de-
gree with tonic support; the Urlinie and a concomitant Ursatz
would essentially close. Given the relations of the voice to the
right hand, and to the modal character of the music, such a
closure would make musical sense only in the major mode
here. This would certainly provide a conclusive and optimistic
answer for the questions of the text and the subtext as we have
discussed it, questions that in fact are only starting to be pre-
sented at bar 41. This situation would indeed render the ques-
tions purely rhetorical.
This Schenkerian view, then, gives valuable insight here
into the tragic structure of the song, in particular into the in-
ability of the voice to double the bass line at bar 41. Still, the
 —
"Aufdem Flusse" 139

singer, while unable to double the bass, does refuse to submit

to the right hand's ominous Qi. Rather than double the Q,
abandoning his major Kopjion, he continues his quest for a
tenable position in the context. He thus sings neither E nor G\\;
rather, he holds onto the questioning B of bar 38 as the one me-
lodic tone that is available for him to take into bars 41-42 the —
B both the broken ring and his heart. That question-
that sets
ing B, in fact, will sound on after the cataclysm of strophe 8
has passed. It sounds in the piano at bar 74, the very last event
of the piece, as the poet moves on and away, taking his heart
and his questions with him.
From the vocal B of bars 41-42 to the end of the sing-
ing —that is, over the entire questioning half of the song
bass, right-hand melody, and vocal line are essentially distinct,
despite some partial doublings. In this respect the situation re-
sembles neither strophes 1-2 nor 3-4. Instead a new mode of
relationship develops: beginning at the crucial B of "mein
Herz," the voice rises above both the other lines, dissociating
itself from both the earlier couplings.
Such behavior is not consistent with a supposition that the
poet is confirming rhetorically the validity of an image that has
been well established. Rather, it supports again, and in a new
way, the notion that he is examining, questioning, and criticiz-
ing the image he constructed during strophes 3 and 4. Just as
he can be imagined getting up from the ice, rising up physi-
cally from his position at surface level to stand above the entire
accompaniment, surface as well as bass. The rising tessitura of
the vocal line, in fact, is a feature that persists continuously
throughout the entire song from its beginning, not just from
bar 41 on. Expanding our interpretation of the rising B as a
question mark, at bar 38 and following, we can then interpret
this continuous rise in the vocal line, up to the final climax, as
the mimesis of a giant structural qestion mark. And this suits
very well the reading of the text that I proposed above, in sec-
tion II.

IV Deeper Structure

In section III, we investigated a number of musical gestures by

which Schubert, in the words of Spaun, "represented and
140 SCHUBERT
transfigured" the images and questions of the text. Some of
these matters involved phenomena progressing over consider-
able spans of time and bearing strongly on the central aesthetic
content of the song. We were able to discuss them without in-
voking very deep levels of tonal structure in a technical sense.
I should now like to show that such representation and trans-
figuration also permeate events on those deeper levels.
Some preliminary remarks are in order. It will be wise to
state explicitly what I hope the analysis so far has made clear:
my personal belief that the aesthetic significance of a musical
phenomenon in a hierarchic tonal or metric structure should
not be correlated a priori, either directly or inversely, with
the depth of the structural level at which the phenomenon is
manifest.
I do not wantto interrupt critical discussion of the song
more than isnecessary to introduce the reductive sketches I
shall be using. As I have already said, technical commentary on
those sketches is relegated to an appendix; there, too, I discuss
various differences between Schenker's diagram of the song in
Der freie Satz and my own readings. But the reader who has
not yet glanced ahead to that appendix should still be provided
with a certain minimum of background information for my
sketches. Each note on a sketch testifies that I hear the tone of
the indicated line (voice, right hand, left hand) as an essential
participant in a harmony that essentially governs the rhyth-
mically symbolized span of the piece. The reader, while doubt-
less disagreeing with some of my assertions, and with some
more than others, will nonetheless quickly get the sense of the
—
method by actually performing the sketches they are explic-
itly designed for performance by keyboard and possibly voice,
in tempo and in the indicated meters. Within reason, some fill-
ing in of the harmony would be unobjectionable. The point is
to check the harmonic, linear, and rhythmic-metric assertions
of the sketches by ear, both as plausible tonal syntax in them-
selves and as accurate reportage of tonal and metric structuring
within the piece. Time spans analyzed as expanded or con-
tracted at a given metric level are adjusted to the asserted norm
at the appropriate deeper level.
These preliminary observations out of the way, let us turn
to figure 2, which brings into clearer focus a number of fea-
 'Aufdern Flusse" 141

'^rJiJlfifJ l

t
i
i'» filfifil |
i
lyi lr-iifl^

6= 5*1 or 1x3
3 for 4 ^j for8 = 4X£
'JUlL.i
H^J^I , P 33E
,

$ f
l

frVr i
'^cjr ciid cid
T
Lir

i (g J i
r"r Hi
"T
^ e p #a gi ^
at

NB!
f
j
f
P

^^^ }L '
J? ^ J J \
m **? mtm
.mr f.r
yftrrp
ffrrWi^rrpi [ J Ff

J)
J J J 1 Jj,J |s & P
(W

^^ W §^ m * "ii

is the technique of
tures discussed earlier. Particularly clear
rhythmic expansion and compression within the large hyper-
measures that demarcate the strophes. In this connection, it
is curious how the six-bar group of bars 35-41 expands a

four-bar group on one metric level (that of the half-strophe),

 >

142 SCHUBERT
while the six-bar group of bars 48-54 contracts an eight-bar
group at a higher metric level (that of the entire strophe). The
bass structures of the two passages in figure 2 are remarkably
similar.
Overall, the bass line of figure 2 clarifies the sense in
which the strophes constitute a loose chaconne or set of varia-
tions. The right-hand line of the figure displays clearly its

characteristic repeated 3-2 gestures at this level.

Figure 3, which reduces figure 2 one metric stage further,
also shows repeated 3-2 gestures in the right hand and varia-
tional structure among the strophes on that metric level. In ad-
dition, it begins to bring into view some curious phenomena.
Its bass line descends an octave, from the upper E at the begin-
ning to the lower E at bar 54. The descent takes place using
degrees of the major rather than the minor mode of E. After
reaching the low E at bar 54, the bass of figure 3 does begin to

Figure 3

J = a of Figure I
(doubling)
O

$ ^S 1«5
43-

P m
$*
XT m
ilrrk*P'rr <l**i
ap
l

m mm
,r-J for 4—
48, «
 "Aufdem Flusse" 143

Figure 4

J - J of Tigurt 3, J - strophe
H

$
u S

' l)
i
j >J
|J
sB
'<J l )J
tj
41

g=F«
^
i m
^ ^ £ *^=*

70

J' I
LI I
I

J
ft hJ
V P^
S PP
sound in the minor mode, but only to confirm a melodic goal
already attained, as it were, in E major. Meanwhile the vocal
line of the figure, which begins essentially by doubling the bass,
rises up the octave to the high E at bar 54, in contrary motion to
the bass line. This ascent also takes place "in E major"; that is,
the vocal line of figure 3 sounds firmly in that key when played
or sung.
This phenomenon comes into even sharper focus in figure
4, in which the reduction is carried one stage further. The re-
duced bass line and vocal line, in their mirror relationship, now
both sound clearly in the major mode throughout. One notes
the structuring force of the G$ in the voice over strophes 3 and
4, and the Gfl in the bass at the beginning of strophe 6. The
vocal Ql that begins strophe 3, it will be recalled, "corrected"
the earlier icy G\\ of the right hand in strophes 1-2. At this
metric level the G\\s that occur within the bass line during
144 SCHUBERT
strophes 7 and 8 disappear, leaving the field completely clear
for the major mode as regards the overall structure of the bass
line at this level. In the piano part of figure 4 one can hear very
clearly the essential variation structure underlying the succes-
sion of strophes —
strophe 6 being exceptional in this respect,
as in so many others.
The mirror arpeggiation of voice part and bass, both pro-
jecting the major mode, is even more starkly portrayed in
figure 5. In this final reduction, only the right hand of the ac-
companiment maintains the minor mode, with its halt und
unbeweglich insistence on G], Gt}, Q|, beating on grimly and in-
cessantly save when momentarily scratched by G# in strophes
3—4 and strophe 6.
What a picture! That remark indeed gets back to the point:
we are dealing here precisely with the singer as a maker of im-
ages, and a critic of images. We have already noted, in section
III, how the imagery and its questioning relate to matters of

essential melodic structure, modality, and the relation of the

vocal line to each of the hands in the piano. Now we have to
confront the emergence of these matters on a very deep struc-
tural level, and interpret the image that figure 5 conveys to us.

V. Under and Over, Inside and Outside

The vocal part, right-hand part, and bass part of figure 5 can
be taken to represent respectively the poet, the surface of the
ice, and the warm flow beneath the ice. The poet (vocal line)

Figure 3

J--o o{ Figure 4 = strophe

j' 1 I
It' HJ 1

U T UP!!
J J"] kJnji ia
ftx

ff pf #
 "Aufdem Flusse" 145

rises over the static, frozen surface of the ice (right hand); the
riverbed (bass line), reflecting the poet mirrorwise, descends
beneath that surface. Thus unter der Rinde, with its E-major ar-
peggiation downwards, is the mirror of uber der Rinde, with its
E-major arpeggiation upwards; uber der Rinde, or even ober der
Rinde, is the vantage from which the poet as critic hopes to
exert mastery over the image he has created.
In order to interpret figure 5 in all its complexity, it will be
helpful to distinguish a "false image" and a "true image" that
can be read from it. The false image develops from the appar-
ent structural priority enjoyed by the outer voices. "I am like
the riverbed, which I reflect," imagines the poet. "As melodi-
cally active outer voices, we control the deep structure of the
piece in our optimistic E-major mode. The static frozen Q of
the right hand, prominent though it may be, does not affect
—
our deep structure, where it is only a blue note and in an
inner voice, at that. Unter der Rinde lies the E major of the bass,
which reflects me and which I reflect, latent though this secret
may be in the surface structure of the piece." In this way, the
singer reasserts his association with the bass line, and his con-
trast with the right-hand line, which we discussed earlier in
connection with doubling relations. After the B of "mein Herz"
the singer can no longer double the bass line, but he hits in-
stead upon the ingenious expedient of mirroring it, a relation
appropriate to his physical action —standing up — at the begin-
ning of stanza 5.
What makes this optimistic image musically false is the
implication that the E-major outer voices of figure 5 really
control the deep structure of the music. If they did, we would
hear the piece in E major; and since we clearly hear the piece in
E minor, the image cannot be valid. What makes the image po-
etically false is its misplaced obsession with the categories of
"over" and "under." While these categories are appropriate to
characterize the relation of icy surface to riverbed, and of right
hand to left hand, they are not appropriate to the cortex of the
heart and its nor to the relation of vocal line
interior chambers,
and bass line combined to the right-hand line of figure 5.
The correct categories in both the latter cases are not
"over" and "under," but rather "outside" and "inside." It is
this misfit of categories that provides a negative answer for the
I46 SCHUBERT
question, "in diesem Bache erkennst du nun dein Bild?" The
stream, that is, is not a valid image of the heart. We have just
discussed some reasons why. The poet senses such reasons
with tension and alarm but is unable to analyze them; pre-
sumably the analysis would be too painful.
The true image of figure 5, then, proceeds, on the basis of
"outside" and "inside." The poet, by entering into relation
with the stream, has coupled his vocal line with the bass line to
form a hull (Rinde) of "outer voices," ostensibly active and
in E major. But the static "inner voice" that the right hand
projects as the kernel at the heart of this Rinde tells us better: its
innermost secret is G\, G\\, G\\, as indeed was foretold by the
opening motive of the right hand in the piece. Within the
elaborately constructed exterior show of motion and warmth,
the poet's heart is frozen solid forever. This must be so, since it
is only and preeminently the right hand line of figure 5 that can

make us hear the piece with a deep structure in the minor

mode. 5 Ironically enough, it is the poet's very creation of the
false image for figure 5, with its pairing of outer voices against
an inner voice, that enables the true image to assert itself.

Afterword

On p. 130 I discuss the word Rinde, saying that it "is in fact

acommon biological term, used to denote the cortex of the
heart. So it is the use of Rinde in connection with the stream,
not the heart, that is metaphorical." These remarks have gener-
ated some powerful and interesting criticism by a number of
readers, including Anthony Newcomb, whose splendid article
follow mine in the present volume. Still, I am disposed to stick
to my guns on the specific point. Rinde is in fact a common
biological term. used for the bark of a tree, the shell of a
It is

crustacean, the crust on a pastry or a loaf of bread, and the rind

of a cheese. In addition, it is used in specifically anatomical
contexts to mean the cortex of an organ: rindenartig and rindig
translate as "cortical" and "corticose" in those contexts.
In all these usages, the Rinde has three salient features.
First, it is a curved surface surrounding completely the entire
volume of a three-dimensional object; it is not a flat planar sur-
face extending over the top of a spatial region, as ice is when it
 "Aufdem Flusse" 147

covers a stream. Second, the object that is surrounded by the

Rinde is alive, or recently alive, or at least characteristically or-
ganic. (Perhaps we should consider French cheese rather than
German in this connection.) Third, the Rinde protects the ob-
ject from injury, in the manner of armor.
In my personal dictionaries (Heath and Muret-Sanders,
1910), I find no usage for Rinde that is not consistent with each
of the three features discussed above. In contrast, the use of
Rinde for the ice does not fit easily with any of the features. As
already noted, the ice is a flat planar surface, not a curved sur-
face bounding an enclosed volume. The stream is not literally,
but only metaphorically, alive; it was only metaphorically re-
cently alive; it is technically organic in the sense that water is
composed of the organic elements hydrogen and oxygen, but
it is not characteristically organic beyond that, except meta-

phorically. Finally, though a stream does develop its ice as the

crayfish develops its shell, we do not normally think of the ice
as having a protective function, defending the stream from ac-
cident or attack. (The American Heritage Dictionary [New York,
American Heritage, 1969], pp. 1536-37, gives rendh- as the
Indo-European root for English "rind," and specifically brings
out the issue of attack: "rendh-. To tear up. ... 2. Germanic
* rind- in Old English rind(e), rind (< "thing torn off"): rind."
This way of regarding the Rinde, from the viewpoint of the ag-
gressor, seems all too depressingly Indo-European.)
In sum, I am still very comfortable with the idea that "it is
the use of Rinde in connection with the stream, not the heart,
that metaphorical." The word appropriate for the ice in lit-
is

eral discourse would be Decke, not Rinde. The Rinde metaphor

does not lie far from common speech of a flowery sort, but that
observation makes me feel the force of the metaphor all the
more strongly. Here, though, I must draw back from explor-
ing the fascinating contentions which that idea suggests, and
point out that I really have nothing to "prove" with all my lin-
guistic commentary, except that it would be reasonable to
imagine Schubert finding in Miiller's text an interpenetration
of object and image over the last stanza, a relationship project-
ing a subtle misfit of "underneath (the ice)" with "inside (the
heart)."
I am sure that Muller was sensitive to that metaphorical
I48 SCHUBERT
dissonance; no mere adventurer, he assumed an official post as
teacher and ducal librarian at Dessau in 18 19, the year after he
completed the Muller-Lieder. My
belief that Schubert was also
sensitive to these matters must rest upon the manner in which I
read various aspects of his musical structure. I do not really
care, of course, to what extent Schubert was a fine critic of
Muller's text. I do care how well he "acted" that script per mu-
sica, and so I have only to show that the ideas about "under the

surface" and "around the kernel," ideas that emerge in my mu-

sical analysis, do engage things that can be legitimately (if not
necessarily) read from the text. The musical analysis itself is
amply problematic.

Appendix
As mentioned in section IV, the basic method of the reductions is to proceed
from one metric level to the next, setting down at each stage as few notes as
possible to represent an essential harmony governing each rhythmic unit at
the pertinent metric level. At each stage, I find it important to make the
sketch performable, give or take a few figures and/or some continuolike har-
monic realization. In general, one can usually proceed in a straightforward
way from one metric level to the next.
At times, one must adjust expansions or contractions to an asserted
metric norm at the next level. The technique can be inferred from the perti-
nent relations of figure 2 to figure 3, etc. Interpretations as to what is ex-
panded and what is contracted, at which metric level, are not always as clear
in other pieces as they are here, where the basic strophe model provides clear
large downbeats and a referential rhythmic matrix.
At times, also, one must juggle several rhythmic levels at once. The
systematic elimination of accessory tones, proceeding from the more de-
tailed to the broader metric levels, will generally lead to a clear choice of es-
sential tones at the broadest levelunder consideration. Figure 6 illustrates the
technique, using a tricky passage of the song, bars 27-28. The harmony at
the barline of bar 28 indicates that the high E is an appoggiatura to the Dt| at

Figure 6

p^g pp
mm
 '

"Aufdem Flusse" 149

this metric level, and not an essential constituent of the E 6 harmony on a

larger level. An alternate reading, to the latter effect, might be worked out; it

would lead to the same end result.

As in most reductive methods of tonal analysis, a given event may
project a variety of functions with respect to a variety of nested and/or over-
lapping contexts in which it is embedded. One is free to read, or not to read,
any a priori aesthetic value into the largest or most final of such contexts,
beyond its formal status as the largest or most final. In Aufdem Flusse, for
example, my method yields a variety of functions in a variety of contexts for
the J chord in bars 9-10. Considering the context spanned by the B triad at
the end of bar 8 and the Gf chord at the beginning of bar 11, for instance,
one can hear the $ chords as passing through. This context is certainly not the
most powerful one functioning over strophe 1; still, the notion that the Alt in
the bass of the J can pass down from B to GU is clearly suggestive in connec-
tion with the approach to the climactic Gfl in the bass of strophe 6. Note the
disposition of registers in the bass line from bar 44 to bar 48, the latter pick-
ing up in register the AH from bar 45. We have earlier discussed the G-sharp
harmony of bar 48 as, inter alia, a substitute for the B harmony; this consid-
eration lends support to the idea of passing from one to the other.
An overlapping context, that of bars 9 through 12 in isolation, suggests
7
that the Gtf chord is a returning neighbor to the $ chord, which then resolves
in D-sharp minor following the standard cadential formula. Figure 7 displays
the analysis. In this context the AH of the bass is essential, not passing, and
the Ftf and Djf above are suspensions, not essential tones.
The implications of figure 7, however, disappear in a larger context
that includes all of strophe 1 and the beginning of strophe 2. Here the overall
harmony of the metric unit defined by bars 5-6 is clearly E minor, as is the
overall harmony of bars 7-8, and that of bars 14-15. In addition, there are

Figure 7

P P ¥ id-
—p— ?

* 3= 1
»J- \^=
-to

Figure 8

& m <*

m ^^
150 SCHUBERT
Figure g

S=^j8= ^

KB: & -F* Or- F'intdeRH

Figure 10

large pulses at the barlines of bars 5 and 14, along with a subordinate pulse at
bar 9 and still lesser pulses at bars 7 and 12. An attempt to fit the reading of
figure 7 into this larger context using my method would result in the syntac-
tic impossibility of figure 8.
In this context, the weight of the measure-pair 12-13 must clearly be
carried by the B
chord, functioning as dominant of the following E, and not
by the Djf chord. Then the AH J can be heard, in this (!) context, to inflect the
structural B chord that follows it, as in figure 9.
My methods and symbology suffer here, I think, in comparison with
Schenker's. Because of the priority I assign metric hierarchies in the reduc-
tion process, I am unable (so far as I can see) to assign the Djj cadence the
tonal (as opposed to metric) weight I hear it bearing in the larger context.
Schenker's symbology, not so strictly bound to metric hierarchies, does a
much better job of demonstrating the simultaneous functioning and inter-
penetration of the disparate contexts portrayed in my figures 7 and 9. His
6
sketch is reproduced as figure 10.
I have already discussed in note 5 my dissatisfaction with Schenker's
Urlinie, which begins at bar 53 and closes in bar 54. As I said there, it seems
dramatically essential that an Urlinie, if there is to be one, should begin ex-
actly with the opening Gti in the right hand. (This would destroy the theoreti-
cal point of Schenker's figure, which is to exemplify arpeggiation in the
structural upper voice before the onset of the Urlinie proper.) A Kopjion in
 "Aufdem Flusse" 151

bar could, of course, later be transferred into the voice at bar 53. The vocal
1

high 54 could then transfer back to the upper right hand E of bar 70,
E of bar
finishing the proposed Urlinie in the proper register (and instrument). The
advantages of listening for an Urlinie, particularly around bar 41, have al-
ready been discussed. On the other hand, the immobile frozen G; projected
by the right hand of my figure 5, a G^ that does not descend, has a symbolic
and dramatic value too. Figure 5 is, of course, utterly un-Schenkerian.
The other substantial divergence of my musical results from Schenker's
involves the bass Gjl at bar 48, governing the beat at the onset of strophe 6.
Almost all my discussion, particularly around table 1 and figure 5, has

pointed to the necessity of including that Qt in the large structure of the bass
arpeggiation over the song, given my reading. Without the Gtf there is no
"secret" E-major bass structure to mirror the E-major vocal structure in fig-
ure 5, and the dramatic force of the various processes converging climac-
tically onto strophe 6 is largely dissipated.
The priority I attribute to metric hierarchies would force me in any
case to write the bass G# at bar 48, since it carries the downbeat of its strophe.
Schenker's use of the Q
in the bass of bar 53 for his Bassbrechung is incon-
ceivable in the context of my method. I think Schenker was repelled on prin-
ciple by the idea of mixture in the Bassbrechung. He also evidently wanted
good local bass support for the third degree of his Urlinie. Perhaps, too, he
heard the G-sharp harmony very strongly as a "dominant parallel" in the
sense of figure 1. Be that as it may, my methods are well suited to throwing
light on the dramatic meaning of bar 48 and its environs, while Schenker's
sketch jumps uncomfortably, to my critical taste, from bar 41 to bar 52.

It goes without saying that any reductive method of analysis for tonal

music owes an inestimable debt to Schenker's work. I have felt, therefore,

some obligation to clarify my departures both from his method and from his
reading of this piece. My own method, which I have been using for some
time, combines aspects of Schenkerian technique with metric considerations
first suggested to me by Andrew Imbrie at Berkeley in the early 1960s. The
published sketches that mine most closely resemble, I think, are those pre-
sented by Imbrie in his article " 'Extra' Measures and Metrical Ambiguity in
Beethoven," in Beethoven Studies [I], ed. Alan Tyson, (New York: Norton,
1973), PP- 45-66.
I do not consider my method worked out into a theory. It does have

strong theoretical implications. I do not feel completely comfortable with all

of these, nor do I always find the analytic readings produced by my harmonic-
metric consistency at all levels more suggestive than alternate readings as-
signing more priority to higher-level voice leading. I find it important to
make my sketches "performable," as opposed to conceptual, but I am not

sure why I feel that way.

Important publications that bear on the theoretical assumptions I seem
to be making include Edward T. Cone, Musical Form and Musical Performance
(New York: Norton, 1968); Arthur J. Komar, Theory of Suspensions (Prince-
ton: Princeton University Press, 197 1); Maury Yeston, The Stratification of
152 SCHUBERT
Musical Rhythm (New Haven: Yale University Press, 1976); and Fred Lerdahl
and Ray JackendofF, A Generative Theory of Tonal Music (Cambridge, Mass.:
MIT Press, 1983).

Notes
1. O. E. Deutsch, The Schubert Reader, trans. Eric Blom (New York:
Norton, 1947), p. 875.
2. In composer a mimetic actor rather than a more general
making the
poetic reader of the text,I go beyond the stance adopted by Edward T. Cone

in his heartwarming study The Composer's Voice (Berkeley and Los Angeles:
University of California Press, 1974).
3. An acquaintance of mine once placed a telephone call and was
greeted by an unfamiliar voice saying without preface, "You have the wrong
number." "Yes, I do," said my friend, "but how did you know?" "Nobody
ever calls me," said the person at the other end, and hung up.
4. It would be helpful to have some indication as to whether the thirty-
second-note groups are to be played slurred, staccato, detached, or even mar-
tellato. The resulting effects would heighten the associations of strophe 6,

more or less accordingly, with the optimistic or pessimistic worlds of the

music. The autograph does not help here: its attack symbols (dots plus slurs)
stop abruptly, as do those in the published scores, in bar 28. That is why I
have written sic here and there on table 1.
5. Schenker's analysis (see appendix to this article) asserts otherwise:

he hears the structural Kopfton for the Urlinie only at the vocal Gl; in bar 53.
I am convinced that he errs here. The whole discussion so far indicates why I

feel it is essential to attach basic structural weight to the opening G\\ in the
right hand (which Schenker omits from his sketch). Figure 5 treats the Gt| as
a Kopfton, beginning an Urlinie that will eventually descend and close. That
hearing, the reader will recall, made excellent sense of the failure of the voice
to double the bass line at the return of the minore at bar 41. What is at issue
here is not the presence or absence of an Urlinie, but rather the function of the
Gi in the right hand at the opening. If there is to be an Urlinie, it ought to
begin there.
6. Heinrich Schenker, Free Composition, trans, and ed. Ernst Oster
(New York: Longmans, 1979), vol. 2, figure 40, 2.
 Structure and Expression in a Schubert Song:
Noch einmal Auf dem Flusse zu horen

ANTHONY NEWCOMB

This essay will explore the relationship between structure and

expressive meaning in music. Most who have thought about
this subject have realized that the expressive meaning of music
cannot be translated into words. But it does not follow that
musical expression should be excluded from those areas about
which serious, informed words may be spoken. True, the ex-
pressive meaning of a musical work cannot be translated into
words, or into anything else for that matter; no more can the
expressive meaning of a painting, or a poem. To translate, in
the normal sense of the word, is not the goal of expressive
analysis. The goal is to interpret— not only to isolate aspects of
structural function, but also to ask what it might mean that the
function is fulfilled in this particular way rather than another.
Clearly such an enterprise cannot set out to prove; it sets out to
persuade — to enhance understanding by illuminating particu-
lar aspects of the work in a particular way. Abandonment of
!

the ideal— or the chimera — of proof does not entail abandon-

ment of careful reasoning and of thorough factual and tech-
nical preparation. On the contrary, expressive analysis can be
done with considerable precision, avoiding the loosely sug-
gested or entirely unspecified connection between expressive
interpretation and musical-technical analysis. And it benefits
from a full fund of historical information.
With this in mind, I have limited myself here to one song,
and one for which there is already a considerable foundation
154 SCHUBERT
of careful recent criticism on which to build: Auf dem Flusse
figures in recent studies by Georgiades (1967), Schwarmath
(1969), Feil (1975), and Lewin (1982), and I shall refer to these
below. Lewin is the most recent and most detailed of my prede-
2

cessors here, and the degree to which his thoughts have stimu-
lated mine will be obvious to those who have read his essay,
which precedes mine in this volume. I shall stray farther from
purely structural matters than does Lewin, but I shall try to
demonstrate a continuous connection between the structural
and the expressive interpretation.
Like Lewin, I shall start where the composer of virtually
every song has presumably started, with a reading of the poetic
text. This will then lead to the interaction of music and text in
the single song, then outward in turn to a consideration of the
larger whole to which both belong.

//

Muller's text contains five stanzas, each with the following ac-
cent pattern and rhyme scheme (save for the first line of stanza
1, which begins , ^ J):

t
w b

f
<y J
W f
KJ C

Der du so lustig rauschtest,

Du heller, wilder Fluss,
Wie still bist du geworden,
Giebst keinen Scheidegruss.

Mit harter, starrer Rinde

Hast du dich uberdeckt,
Liegst kalt und unbeweglich
Im Sande ausgestreckt.
In deineDecke grab' ich
Mit einem spitzen Stein
Den Namen meiner Liebsten
Und Stund' und Tag hinein:
 Structure and Expression in a Schubert Song 155

Den Tag des ersten Grusses,

Den Tag, an dem ich ging,
Um Nam' und Zahlen windet
Sich ein zerbroch'ner Ring.

Mein Herz, in diesem Bache

Erkennst du nun dein Bild?
Ob's unter seiner Rinde
Wohl auch so reissend schwillt?

[You who used to ripple so happily, bright, wild

stream, how still you have become: you give no parting
greeting.
With a hard, stiff crust you have covered yourself. You
lie cold and motionless, stretched out in the sand.
In your shell I engrave with a sharp stone the name of
my beloved, and hour and day:
The day of our first greeting, the day on which I went
away; around name and numbers runs an uncompleted
circle.
My heart, in this brook do you now recognize your own
image? Is it, underneath its shell, seething too, and near
to bursting?]

In stanzas 1 and 2 the lyric protagonist addresses the

stream. He does so largely as a way of giving information
about the stream: it is at least frozen over, perhaps frozen
3
through. In stanzas 3 and 4 (as in the middle sections of so
many of the Winterreise poems) he turns his glance backward in
time. Here his memory is, rather exceptionally for this cycle,
stimulated by a present action, which this middle section de-
scribes: with a stone he is engaged in cutting into the ice the
name of his beloved, and the dates of their first and last meet-
ings. Although when the protagonist begins the middle sec-
tion he is still addressing the stream, the action and its stimulus
to his memory turn his attention inward, toward his own heart.
He then addresses his heart directly in the last stanza, the first
one in which he makes a statement about his own present emo-
tional condition (as opposed to observing and describing a sit-
uation from without).
He makes this statement through two rhetorical ques-
tions— that is, two questions asked not to elicit information
I56 SCHUBERT
but for other reasons, here to give information. With the first
question, he directs his heart to the potential similarity be-
tween itself and the stream, perhaps including both the frozen
exterior of stanzas 1 and 2, and the scratched, scarred exterior
of stanzas 3 and 4. (The substitution of "erkennst du wohl" for
Miiller's "erkennst du nun" in the printer's clean copy and in the
first edition of the song seems to emphasize the proposed affir-
mative answer to this first rhetorical question.) Then he uses
this proposed similarity to assert that his heart, though cold
—
and scarred outside, is still seething beneath that he is still
warm with passion for his beloved. He asserts this by asking if
the stream also swells with motion beneath the surface. For the
heart to recognize its image this would need to be true.
Lewin's interpretation of this last stanza especially (pp.
129-3 1) is quite different. It rests heavily on a double meaning
of the word Rinde, as not only the frozen shell of the river but
also the cortex of the heart. In fact, he asserts this "common
biological" meaning to be the primary one here, and goes on
to claim that the final question of the poem is a real, not a rhe-
torical one— a real one asking if the poet's heart, now frozen
like the river, will warm with emotion again in the spring. But
to take the appeal here to be to the biological term meaning
cortex of the heart is certainly too literal. Neither the poet nor
Schubert wants to call attention to the heart of biology treatises
—
and anatomical diagrams tubes and muscles for pumping
4
blood. Both use the conventional metaphor of a psychological
not a physical heart, one that is the seat of the emotions. To
insist on the cortex in this metaphor would certainly be out of
place. Nor is there a mention in the poem of the idea of the
river's thawing in the spring. On the contrary, the idea of an
approaching spring is excluded from the expressive world of
5
Winterreise. It seems a questionable source for the true mean-
ing of the climactic question of this poem.
The evident tension in the two questions of the final
stanza of the poem is not the arcane one Lewin proposes, but
the relatively straightforward one between, on the one hand,
the mounting of internal passion excited by memory, and,
on the other, the denial and repression of that passion necessi-
tated by the external situation in which the protagonist of the
poetic cycle finds himself.
 Structure and Expression in a Schubert Song 157

///

poem starts at a nadir of rhythmic

Schubert's setting of the
activity. (See theAppendix, ex. E.) Georgiades has noted that
the beginning of the song is unique in Winterreise
6
pure accom- —
panimental figuration, stripped of all ornament. It is in fact less
a rhythm than emitted in neutral four-bar chunks. The
a pulse,
opening four bars of the vocal line, which circle around the
tonic triad without real melodic or harmonic motion, do little
to relieve this static, frozen atmosphere. The literal content of
the first two lines of text —
the "lustig rauschen," the "heller,
wilder Fluss" —
is reflected only by strong irony. While one can

easily imagine using the inherent contrast in the text to build a

strong musical contrast in the opening of the song, Schubert
here sees only a motionless, monotone gray. With "wie still
bist du geworden" comes the first distinctive movement of the
song: a chilling downward slip by voice and bass in parallel oc-
taves to an unprepared second-inversion triad on the raised
seventh degree. Though violent in its unconventionality, this
first move remainstiny, and even this tiny move is downward.
H. H. Eggebrecht has proposed that "numerous Schubert
songs are . reducible to a kernel that sounds out literally at
. .

the beginning of the song, and that lies at the basis of the the-
matic development of the entire song." 7 This distinctive ges-
ture at the beginning of bar 9 is such a kernel, both verbal and
musical. The verbal element is, initially at least, the word
"still,"but it is the still —
of frozen motion of intrinsic motion
—
temporarily restrained not a genuine still (cf. the correspond-
ing place in stanza 2: "kalt und unbeweglich"). The ironic ten-
sion between motion and its apparent absence, intimated in the
setting of the opening lines, thus continues with this musical-
verbal kernel. Though as muted as everything else so far, the
kernel will not remain so.
This kernel is the first of three particular features that I
shall trace throughout the song, for the three together give a
highly distinctive shape to a relatively conventional schematic
design. The second announces itself immediately in bars 1 1 - 12,
where what I have called the almost nonrhythmic pulse of the
opening is animated ever so slightly by the vocal line, at the
words "keinen Scheidegruss." In another instance of ironic
I58 SCHUBERT
tension between textual image and underlying musical mo-
tion, the vocal line breaks perversely out of its lethargy to imi-
tate, even down to the cheerful little inverted mordent, the
parting wave that the text is simultaneously denying. The ac-
companiment then picks up and extends the sixteenth-note
motion of the voice, in the process breaking the rigid four-bar
mold of the previous phrases with a fifth bar containing a small
surge of motion that will be the direct model for the larger
8
ones to come.
The third feature, involving the treatment of range in the
vocal part, also announces itself in bar 11. Here the restricted
circling gestures of the opening are expanded: the voice covers
an eleventh in little more than a quarter note, again in seeming
ironic portrayal of the motion denied in the text. A similar
tension between text and musical motion, now amplified by
the high djt' on the "un-" of "unbeweglich," exists at the corre-
sponding place at the end of strophe 2. Admittedly, the me-
lodic motion in both places is still the same circling motion
that had marked the opening phrases of the song, now deco-
7
rating a simple cadential progression (ii* or iv -V-i). But
the gestures themselves are becoming more concentrated and
energetic.
The preservation of the draft of much of the first part of
Winterreise enables to us to observe Schubert as he refines his
9
initial ideas. In the particular instance of bars 11-12, we can
see him doing a bit of what might be called expressive fine tun-
ing, by which he turns a vague, generically expressive musical
gesture into a more sharply defined and specific one. His initial
version of the voice part in bar 1 1 went up only to a c|' on the
fourth sixteenth note, then down only to a c* on the last eighth
note. In reworking the passage he considerably expanded the
energy of the gesture simply by changing its extreme notes.
He also sharpened the musical expression of swelling, but sup-
pressed motion in the piano interlude of bar 13 by revising the
upper voice of the firstfour sixteenth notes, whose original
static version is given in example
1. Again, the tiny initial de-

tail grows subsequent variation. The rise of a minor second

in
in the upper voice at bar 13 becomes a major second upon repe-
tition in bar 22, providing the starting note, gU, for the lyrical
upward expansion of the vocal part in the ensuing bars, just
 Structure and Expression in a Schubert Song 159

as the momentary rhythmic activity in the bar of extension-

interlude becomes the source of the more consistent motion at
10
the beginning of the next section.
A contrast of motion, mode, and lyric style define this
next segment as a contrasting middle section. The textual mo-
tivation for this contrast is the protagonist's switch at this point
from description of an external situation in nature to descrip-
tion of a present action of his —
own an action that he still,
however, describes impassively, as if from the outside. In fact,
the tone of the poem remains constant and rather cold across
stanzas 3 and 4. The scratching of a Rinde with a spitzen Stein is
neither a warm sound nor a friendly image, and the unpleasant
aspects of this memory in Muller's poem (the "Tag an dem
ich ging" and the "zerbroch'ner Ring") almost outweigh the
pleasant ones. In Schubert's middle section, however, an in-
creasing warmth comes over the protagonist as action stimu-
lates memory.
Although the turn to the major mode announces this
change in conventional way, the development of the three fea-
a
tures to which I have referred above gives individuality and
evolving shape to the emotion expressed. First, the sixteenth-
note rhythmic motion of the piano interlude is converted into
a more consistently used and clearly directed motion of up-
beat-loaded second beats, pressing across to the following
downbeat. 11 Simultaneously the vocal line begins to rise in
range. Moving upward in a conjunct lyrical style that replaces
the triadic circling of the opening stanzas, it passes from the
third, through the fourth and fifth, to the sixth degree of the
scale (with harmonic support for each), at which point (bar 28)
it frees itself from the accompaniment to overreach the d' with

an appoggiatura e', the highest note of the piece so far. This

high d' with upper appoggiatura occurs in the spot in the po-
etic and musical stanza (the third line and phrase) that corre-
sponds to what I have called the musical-verbal kernel in the
 —
160 SCHUBERT
firsttwo stanzas: the sinking motion to the D-sharp triad. The
transformation of this kernel in the middle section gains expres-
sive meaning not only from the rising swell of the musical
gesture that it culminates, but from the contrast with the analo-
gous moments in the previous stanzas. The frozen motion-
emotion of stanzas i and 2 is warming and pressing to the
surface. This expressive meaning is reinforced by the simulta-
neous failure of the rhythmic motion to stop on the downbeat
of bar 28, as it had on those of bars 23-27. Instead, the six-
teenth notes flow right through the first half of the bar, stop
only momentarily on the second half, then flow continuously
across bar 29 and the first half of bar 30, before breaking into
the triplets at the beginning of stanza 4.
Though Schubert's setting of stanza 4 follows closely the
model of 3, several details reveal that the protagonist's emotion
continues to grow as he pursues his memory. As before, his
vocal line rises in a single gesture to the high e', although to
do so it must now ignore the emotional downward turn of
Muller's "den Tag, an dem ich ging." With a subtle variation of
this repeated musical climax, the coolness and distance of the
beginning of the song disappear entirely: it is as if we were
present at the action described. As the protagonist draws the
ring around the names ("windet/Sich"), he must suddenly
confront the realization that he cannot complete the action: the
vocal line, after circling back warmly to the climactic high e',
suddenly breaks, plunging down a gaping seventh. The at-
tempted gesture has failed, and with it breaks the dream.
Although the dream has broken, Schubert's setting goes
on to express through musical metaphor the idea that remem-
bered passion cannot be so easily denied. The gap between
stanzas 4 and 5 is full of unresolved tension. This tension is ex-
pressed by the bass line, where the move to the low E implied
by the descent in bars 38-39 is interrupted at the beginning of
stanza 5 in bar 41, and resolved only at the cadence in bar 54.
Likewise, the rhythmic motion of the end of stanza 4 — like
that of stanza 3, more continuous than that of the beginning
now runs on even longer, across two full bars of piano inter-
lude, before stopping abruptly at the rest in bar 40. This en-
ergy is not released in a downbeat accent, not dissipated or
wound down; it is interrupted. And it bursts out again imme-
 Structure and Expression in a Schubert Song 161

diately in the jerky interjections of the voice in bars 4 iff., then

erupts with unprecedented force in bars 47ff; here the phrase
rhythms lose their four-bar regularity, not through extension
as in stanzas 1 and 2, but through truncation, which imparts an
impatient forward thrust.
The clearest musical expression of this unresolved tension
is the handling of range in the vocal line. The failed gesture of

bars 36-37 will not fully accept its failure. The vocal line in bar
37 rises once more to c|T, breaks once more ("zerbroch'ner
Ring"), then rises again to b in bar 38 for its final note of the
stanza. After the interlude and general pause, the accompani-
ment proposes a return to the repressed motion and emotion of
the beginning. But the heart of our protagonist, the psycho-
logical heart of the literary convention, says no. He separates
himself both from his own original circling melody and from
the downward g-ft pull of the top voice of the accompani-
ment, and, beginning stanza 5 at the ty where he had ended the
fourth, moves directly to the high e' of his previous climax.
A tension had already been introduced between accompani-
—
ment and vocal line voice pulling upward against downward-
pulling accompaniment —
when the voice overshot the accom-
paniment for its previous climax (bars 28 and 36). It became
more striking when the voice refused to return to the g-ft ca-
dential area of the accompaniment in bars 37-38, insisting on
its own c#' to b. voice now stages a full metaphorical
The
rebellion against the return proposed by the accompaniment
and by the conventional ABA'
form.
12

Here again an examination of Schubert's draft reveals how

he sharpened the expressive musical metaphor. In revising the
draft Schubert changed the rhythmic figure of the accompani-
ment of bars 31-39 from sixteenth notes to triplet sixteenth
notes, making more vivid the expression of mounting excite-
ment across the middle section. In the draft, a c|T on "sich" in
bar 36 — instead of the high e' of the final version —
falls short
of the wonderful immediacy created in the final version by the
continuing upward swing of the climactic gesture right through
bar 36 until the sudden break at the beginning of bar 37, the
primary component in making us feel that we are present at the
moment of the breaking of the dream. The vocal line of bars
37—38 originally proceeded as in example 2, which virtually
162 SCHUBERT
Example 2

s—v—#^*

fails to express the continued upward pull of the voice part

against the descending accompaniment, embodied in the c|T
13
and b of the final version.
As we have noted, the vocal part of the beginning of
stanza 5 refuses the proposal of the piano part to return to the
suppressed, frozen motion of the beginning of the piece. In-
stead, it pursues across the entire closing section the curve of

mounting passion begun in the middle section, thus ending up

in full tension with the conventional shape of the song, with
the requirement posed by the narrative of the poetic cycle, and
with the initial musical image of the song as proposed and re-
asserted by the accompaniment. The course of this curve of
mounting passion is traced especially by the continuing devel-
opment of the three elements already isolated: vocal range,
rhythmic motion, and the kernel of bar 9.
After beginning the closing section in the same b-e' range
occupied at the climax of the previous section, the voice climbs
gradually higher across the remainder of the song. Annexing
ftf' at the first question (bar 46), g' at the second (bar 53), and

finally a' at the repetition of the second question (bar 59), it ends
on the tonic an octave above its starting point. 14 After proposing
a return to the rhythmic style of the opening of the song, the
accompaniment is soon swept into bursts of thirty-second
notes, completing the progression of the whole song from
eighths to sixteenths to triplet sixteenths to thirty-seconds.
Simultaneously with this climax of rhythmic motion in
the accompanimental figuration comes a wondrous revision
of the functional-harmonic and the metaphorical-expressive
meanings of the kernel of bar 9. The stage is set for this revi-
sion by an expansion in total range, as the voice part moves to
the top of its range and the piano drops an octave to its lowest
extremes, and by a tightening of the phrase rhythm, as this oc-
tave drop in the piano pushes to the subdominant of D-sharp
minor a bar early according to the model of the opening stanzas,
 Structure and Expression in a Schubert Song 163

which has so far been followed. All this prepares a release of

energy from the second beat of bar 47 to the downbeat of bar
48. Here the protagonist, as Schubert portrays him, takes his
resolve and makes explicit his continuing passion. And at just
this point the kernel of bar 9 is reinterpreted. Just as the "still"
and "kalt" of stanzas 1 and 2 no longer convey the true internal
situation of the protagonist (and perhaps never did), the ker-
nel, as developed in bars 45ff., is no longer a tiny sinking ges-
ture to the leading tone, made only to be revoked by a quick
return to the tonic. Instead it becomes a real point of depar-
ture, a platform from which to vault upward by fourth to
G-sharp minor. 15
Here I part from Lewin's interpretation, which sees in the
move to G-sharp minor a reference to the parallel major of the
middle section and a ray of bright optimism in the bleak musi-
cal landscape. This is, of course, a possible interpretation of
G-sharp minor in the abstract; but that is not the way the key
area is presented here. Nowhere in this passage, or indeed in
the entire song, does Schubert call attention to the possible
mediant relationship between G-sharp minor and E major.
The preparation and presentation of G-sharp minor here stress,
and hence express, two qualities: first, the vigorous, overreach-
ing quality of Gil, which lies above the expected G^ of the E-
minor scale and is approached by a leap of a full fourth from
the leading tone, instead of by half step as in the model pro-
posed by the opening stanzas; second, its quality of being
minor, rather than the conventional relative major on G. Both
these qualities will be reinforced in the following repetition of
the entire poetic stanza.
At the beginning of this repetition the accompaniment
pulls back, once more urging a return to the opening image.
The answer this time is a still more extreme interpretation of
the kernel. Schubert for the first time modulates up within the
first phrase of the stanza, reminding us of the standard upward
modulation in an E-minor piece: to G major not G-sharp
minor. Then he uses the sinking half step —
now to FH as a —
platform to vault up not one but two fourths. The voice then
modifies its previous melodic line to pick up these upward-
leaping fourths and move in sequence to an astonishing G
minor, emphasizing with this second question of the repeated
164 SCHUBERT
stanza the minor-ness of the gesture at the first setting of the
question.
After the repetition of this last line —
simultaneously the
—
end of the vocal line and its climax in range the accompani-
ment snaps back at once to the opening image of frozen mo-
tion. The hints of ironic tension between surface textual detail
and musical representation in the first section of the song have
developed into a wide gap between the denial of emotion re-
quired by the external situation and the warmth of emotion
within, a gap never wider than at the juxtaposition of the last
vocal phrase with the piano postlude.

IV
The accumulated tension flashes across this gap and is released
only at the beginning of the next song, Ruckblick, to which
Aufdem open-ended and unresolved, serves
Flusse, in this sense
16
as a large upbeat. There are other connections drawing to-
gether these adjacent songs. The stepwise ascent from the first
to the fifth degree that opens the vocal line in Aufdem Flusse
(and that returns throughout its final section) takes on two
chromatic degrees to become the opening of the piano intro-
duction to Ruckblick. The shocking G minor in the last strophe
of Aufdem Flusse (bars 64-65), and the final rocking back and
forth between E and G in the piano postlude (bars 70-71) pre-
pare the G minor of Ruckblick. The broken octaves in the last
bars of Aufdem Flusse ( y <FS ) are transferred to the open-
ing of Ruckblick (y TfiyFfi )•

The connections between Aufdem Flusse and Ruckblick are

by no means the only such close structural connections be-
tween adjacent songs in Schubert's Winterreise. It is usually pro-
posed that Winterreise is not a true cycle, even in the loosely

connected, narrative tradition of the Liederspiel, such as Die

schone Mullerin. I should like to propose that at least the first
part of Winterreise is indeed such a cycle, or was read as such
by Schubert when he first encountered it as an independently
published unit of twelve poems, and went about setting it to
music. Myarguments to this effect will proceed from the his-
torical information surrounding Winterreise, to the cycle of
poems, to the cycle of songs. 17
 Structure and Expression in a Schubert Song 165

1. history of Schubert's encounter with the poems of

The
18
Muller's Die Winterreise has been told many times. Schubert
encountered the twelve poems that were to become his part I
for the first time (we do not know precisely when) in a peri-
odical of 1823. They are printed there as a group and in the
19
order and the versions in which he set them. Only later (again,
we do not know precisely when) did he encounter a publica-
tion of 1824, in which Miiller had interspersed twelve addi-
tional poems among the dozen original ones.
Schubert may have encountered the larger version only
after he had drafted his setting of the twelve-poem version.
This seems to be indicated by the presence of the word Fine
after the last song in Schubert's autograph of the first twelve
songs. (It also appears at this point in the non-autograph clean
copy used as printer's model; the first edition is the first source
20
to delete this word.) This autograph bears the date February
1827, but it is on the clean copy of the revised first song. The
date thus may well indicate when Schubert replaced his origi-
nal draft of this song with this recopied revised version, after
the main work on all twelve songs had been completed. Robert
Winter's investigations into the paper types used by Schubert
show that the paper on which he wrote both the draft of the
first twelve songs and all of the inserted revisions, save that of
Ruckblick, was a paper used by Schubert from October 1826
through May 1827. 21 (The revision of Ruckblick is on a paper
used in June through September 1827.) All this evidence con-
spires to indicate that the bulk of the work on these twelve
songs was done in the winter of 1826-27, probably before
March 1827.
The draft for the complete second part has disappeared,
and only a clean copy in Schubert's hand, dated October 1827,
survives. Drafts survive for Mut and Die Nebensonnen, how-
ever; that of Die Nebensonnen can be associated with events of
September 1827, 22 and is on a paper used at the end of the sum-
mer of 1827. 23 In these drafts the two songs bear no number,
and hence give no evidence of being yet firmly placed in a
24
cycle. All this indicates late summer and early autumn of
25
1827 for the major work on part II. Schober, as reported by
Kreissle, claimed that Schubert had found Muller's Die Winter-
26
reise in the little library that Schober had "set up for" Schubert.
166 SCHUBERT
How can we reconcile this with the Schubert had
fact that
27
moved in with Schober only in March Perhaps it was in
1827?
the little library provided by Schober that Schubert found the
second volume of Muller's Gedichte aus den hinterlassenen Papieren
eines reisenden Waldhomisten (1824), containing the full twenty-
four-poem version of Die Winterreise. Then in the summer
or early fall of 1827 he set the additional twelve poems, thus
modifying his original intent. (Whether the twenty-four-song
set becomes a new whole is a question that I shall not take
up here.)

2. The twelve poems published in the periodical of 1823

do indeed describe simple actions and have references back and
forth that weave the group into a narrative sequence, however
28
loose. The dream of Fruhlingstraum, for example, takes place
in a building (cf. the reference to the "Fensterscheiben"), pre-
sumably the "Kdhlers engem Haus" in which the protagonist
has lain down to rest in the preceding Rast. The rest hoped for
in Rast certainly leads to the dream of Fruhlingstraum, and
thence to the resumed walking (though "mit tragem Fuss") of
Einsamkeit —
walking that refers back to the beginning of the
sequence and rounds it off. In Der Lindenbaum the protagonist
refers to having passed the linden tree in deepest night and to
being now many hours from it, establishing a relationship both
with the action and with the time passed in numbers 1-4.
One can even construct a chronology along the following
lines:
— The protagonist leaves at night (song no. 1); stands before
his beloved's house (no. 2); weeps, perhaps at this time, perhaps
later (no. 3).
— Theprotagonist escapes —
he runs away from where he is
(no. 4). is seeking footsteps in the snow seems to im-
That he
ply an external source of light, either moonlight or the light of
the next day. In any case, since no. 5 mentions that the linden
tree of the departed village is now several hours behind him,
some considerable time has elapsed since no. 1.
— Two static songs follow: the protagonist weeps again
(no. 6 —
cf. no. 3) and establishes the image of the frozen

stream; he interrogates the frozen stream (no. 7).

— The protagonist escapes again (no. 8), now probably from
 :

Structure and Expression in a Schubert Song 167

an imagined view of the towers of the village called up by his

acts of recall in no. 5 and in the middle section of no. 7: the
text of no. 8 is a complex mixture of memory and action, past
and present, fleeing and returning. The protagonist next fol-
lows a false image of nighttime light (no. 9). He then lies down
to rest (no. 10), on what would be the night after he had left
the village, after walking and brooding for close to twenty-
four hours. He dreams and wakes (no. 11).
— Subdued, dragging, he sets off in the clear morning of the
second day (no. 12).
The whole, thus interpreted, is a unified and rounded slice
of time and experience. But it is not a closed one; it circles back
to resume the action of the beginning. For that very reason it is
perhaps more unusual than Die schone Mullerin.
A number of distinctive recurrent images tie the poems
together as well:
— The frozen tears and melted stream of no. 3 connect with
the hot tears and melting snow of no. 4, which connect in turn
with the tears melting the snow that becomes a stream in no. 6,
which connects in turn with the (frozen) stream of nos. 7 and 9.
Recurrent images are especially frequent in nos 5 through 8
.

— The linden tree of no. 5 connects with those of no. 8.

— The images of affection carved on the bark in no. 5 connect
with the images carved on the ice in no. 7.
— The image of the stream's return to the beloved's house in
no. 6 connects with the image of the protagonist's return in
no. 8.
Images of memory-in-dream accumulate across the sec-
ond half of the group, from no. 5 onward. In no. 5 comes a
large block of memory, unmotivated by any particular event,
and following a sharp break from the mood of no. 4; in the
middle section of no. 7 the action of carving in the ice stimu-
lates memory; in no. 8 the memory of the middle section is
powerful enough to lead to the exceptional transformation of
the original material at the close; a hallucination (a kind of
waking dream) in no. 9 finally leads to an explicit dream-
memory poem in no. 11.
One can also see an alternation of static poems with poems
of motion, giving a distinct dynamic shape to the whole twelve-
poem set.
168 SCHUBERT
— A poem of motion by the protagonist (no. i) is succeeded
by one of violent motion that is not his (no. 2), then by a
static, almost timeless poem of weeping (no. 3).
— This is followed by another poem of vigorous and sus-
tained motion by the protagonist (no. 4), which is followed by
a sharp break leading to the three static poems of accumulating
memory-dream images to which we have referred (nos. 5-7).
— The tension of this longest group of poems without mo-
tion by the protagonist snaps with the violent motion of the
beginning of no. 8, which is followed by a poem of gentler,
erratic motion without clear direction (no. 9), slowing toward
the rest foreseen in no. 10, and the dream of no. 11.
— The whole closes with a reluctant return to motion (no. 12).

In poems, one must reconstruct types of physical mo-

3.
tionfrom the verbal images. In songs, especially in songs
by Schubert, types of motion are presented directly to the
imagination, and are among the fundamental elements of the
29
artwork. It is thus no surprise that Schubert's music makes
clearer and stronger the design of motion types implied by
Muller's poems. He creates a clear growth in the songs of mo-
tion —
from no. 1, through no. 4, to a climax in the violent,
irregular, forward-stumbling motion at the beginning of no. 8.
This song, for the first time, does not end in the style of mo-
tion with which it opened, but quiets markedly toward the
end. Then the slower, erratic motion of no. 9 leads to a return
in no. 10 of the regular eighths of no. 1, now with a gently
upbeat-leaning, cross-slurred articulation. The dream song of
no. 11 gives in microcosm the various motion types of the
cycle. Finally, the regular eighth notes, now with dragging
afterbeat slurs, recur in the main musical image of no. 12. The
songs creating this archlike design of motion types interlock
with those elaborating the principal static image of the cycle,
the tears-snow-stream image of songs 3, 6, 7, and 9.
Discussion of overall tonal design in Winterreise is made
exceptionally difficult by questions concerning transposition
from the pitch level at which the songs were originally written.
Songs 6, 10, and 12 from part I were transposed (see n. 28), as
were songs 22 and 24 from part II. We have authorization in
Schubert's hand only for the transposition of no. 10. Nos. 6
 Structure and Expression in a Schubert Song 169

and 12 appear in new keys in the nonautograph printers copy;

nos. 22 and 24 appear in the autograph fair copy in A
minor
and B minor respectively, with directions for transposition in
the publisher Haslinger's hand. There seems little doubt that
Schubert accepted these transpositions, but there is also little
doubt that he was forced to do so by a prohibition issued from
some quarter against requiring a high a' in the vocal part. In
nos. 6, 10, and 12, for example, he rewrote the vocal line, or
decided to eliminate an alternative version when he revised his
draft; these revisions resulted in requiring a high a', which in
30
turn led to transposition. If we accept that these transposi-
tions were forced upon Schubert by the necessity of avoiding
the high a', we should perhaps then regard them as compro-
mises, potentially as far from the ideal as the sorry alternative
ending of Aufdem Flusse.
Curiously enough, the transpositions do not much change
the overall effect of tonality in the cycle, for in both Die schone
Mullerin and Winterreise Schubert uses tonality more to group
and interrelate songs than to make overall architectural shapes. 31
The sharp breaks in part I of Winterreise remain between nos. 2
and 3 (which suggests that no. 3 is seen as happening at a sepa-
rate time and place from no. 2), and between nos. 4 and 5
(inaugurating the dream-memory series to which I referred
above). The apparent break between nos. 7 and 8 is bridged by
the tonal preparation of G minor in no. 7, as we have seen.
Whether no. 10 is in C minor or Dminor, no. 9 is isolated on
both sides, as befits its hallucinatory subject. 32 The tonal con-
nections across nos. 5 and 7, on the other hand, are particularly
strong. Thus the largest group of interrelated songs stretches
from nos. 5 through no. 8.
Just as there are particular types of motion in Winterreise,
so there are particular types of vocal range. The most striking
of these is that which covers rapidly, often in a single gesture,
an abnormally wide span —
a tenth to a twelfth. In part I, nos.
6, 9, and 10 belong to this group, and are prefigured by the
violent interval vocabulary of no. 2. Of the songs covering
a range of more than a twelfth, only no. 7 does not handle a
wide range in this way. Rather, as we have seen, it expands
upward pitch by pitch, reaching a climax only at the last line
of the song, and finally covering the widest range of any song
1 70 SCHUBERT
—
of Winterreise one half step short of two octaves. 33 Thus the
songs with abnormally wide range (nos. 6, 7, 9, and 10) stand
grouped around no. 8 in the center of part I, with no. 7, at
their apex, unique among them.
Local motivic connections in this central group of songs
tell the same story. There are close motivic connections be-

tween nos. 4 and 5 (thus bridging in an almost subliminal fash-

ion across the clear break created by tonality and by type of
34
motion), and between nos. 5 and 6. I mention this only as a
last item in this discussion of the many approaches by which
part I of Winterreise emerges as a unified structure. All of these
approaches place nos. 7-8 at the apex of this structure. It
seems fair to conclude that what has caused several commen-
tators to focus on the expressive intensity of Aufdem Flusse de-
rives from its structural position in a musical whole larger even
than the song itself.

Notes
1. I have discussed these matters in greater detail in "Sound and Feel-

ing," Critical Inquiry 10 (io84):6i4-43, including a review of recent literature.

2. Thrasybulos Georgiades, Schubert: Musik und Lyrik (Gottingen:
Vandenhoeck und Ruprecht, 1967); Ermudte Schwarmath, Musikalischer Bau
und Sprachvertonung in Schuberts Liedern (Tutzing: Hans Schneider, 1969);
Arnold Feil, Franz Schubert: Die schbne Miillerin, Winterreise (Stuttgart: Philipp
Reclamjun., 1975); David Lewin, "Aufdem Flusse: Image and Background in
a Schubert Song," this volume.
3. I disagree with Lewin (p. 130) that the latter possibility has no bear-
ing on our interpretation of the poem. That the stream may be frozen through
(a matter not of science, as Lewin proposes, but of practical observation)

gives some meaning to the last question of the poem. As we shall see, how-
ever, it does not make the question a real, as opposed to a rhetorical one.
4. Lewin nowhere gives any evidence in this section that this is "Schu-
bert's reading" of the text, as he claims on p. 131. His evidence is purely
verbal, divorced from any discussion of the setting.
5. One may find a hint of the idea in the second stanza of Miiller's

Wasserfluth, but Schubert's setting takes no notice of this possible meaning.

The music remains wintry and static throughout.
6. Georgiades, Schubert, p. 360.

7. H. H. Eggebrecht, "Prinzipien des Schubert-Liedes," Archiv fur

Musikwissenschafi 27 (1970): 89.
8. Were it not that the composer makes clear the connection between

the successive stages in the growth of rhythmic motion across the song (bars
20-22 are the source of the motion of bars 231!., which in turn is the source
 Structure and Expression in a Schubert Song 171

of that of bars 3 ifF. which in turn bursts out intensified in bars 48ff.), we
,

might pass over this first instance. Since he does make each image of motion
grow from this gesture, we must, at least in retrospect, accord it the proper
weight.
9. The surviving primary sources for Winterreise and the variants
among these sources are described in Franz Schubert, Neue Ausgabe samtlicher
Werke (Kassel: Barenreiter, 1964-), series 4, vol. 4b, pp. 2991T. The auto-
graph draft of part I, together with the autograph fair copy of part II, has
been published in facsimile (Kassel: Barenreiter, 1955).
10. The transcription of the revised draft of Aufdem Flusse is included
in Schubert, Neue Ausgabe 4:4b, 2421!. It correctly records that Schubert did
not write a trill in the vocal line of bar 11. The trill was present in the drafted
vocal line in D-sharp minor and disappeared, presumably by careless omis-
sion, only when Schubert hurriedly renotated the passage in E-flat minor. It
reappears in the fair copy and in the first edition.
11. This conversion of the neutral, undirected pulse of the accompani-
ment of the first section into a more forward-pressing figure is aided by an-
other detail. From bar 23 onward Schubert writes an accent on the second
half of the first beat of each bar, which prevents one from hearing it as a
passive afterbeat, a simple second half of the beat, and tilts it forward toward
the upbeat grouping of the second beat of the bar.
12. I do not mean to imply that the accompaniment represents always
one dramatic character or force, the voice another. This claim would over-
specify the musical metaphor. The dramatic relationship between voice and
accompaniment has been explored with great sensitivity by Edward T. Cone
in The Composer's Voice (Berkeley and Los Angeles: University of California
Press, 1974). I use Cone's phrase when I say that the "expressive potential" of
the relationship between voice and accompaniment in this song is broad and
changes frequently. At the beginning of stanza 5, for example, the accom-
paniment pulls against the passionate rebellion of the voice by asserting the
initial image of the song — metaphorically the force of convention and exter-
nal situation. But only a few measures later it is in the lunge of the accom-
paniment that the suppressed rhythmic motion breaks out anew, prefiguring
the decisive gesture of the voice.
13. While Lewin notes the upward leap of the voice to b in bar 38, he
fails to see its expressive significance. This failure is bound up with, and may

even have led to, a small musical misreading as well. I quote Lewin (pp. 137—
38): "At bar 38 the voice, detaching itself from the right-hand part, leaps up
to B at the, cadence, rather than settling on FH once more." But the detach-
ment of the voice from the right-hand part is not the remarkable thing here.
The voice has already detached itself from the right-hand part, first momen-
tarily in bar 28 and then definitively in bar 36. In bars 37-38 it occupies its
own tetrachord, from e' down to b. From now on its area of activity will be
separate from that of the piano part. Thus the leap — the detachment in bars
—
37-38 of which Lewin speaks is the one downward from its own tetra-
chord to rejoin for a fleeting moment the area of the right-hand part, first
with the at the beginning of bar 37, and then again with the e at the end of
fjt
172 SCHUBERT
the same measure. Schubert's change of the last note of bar 36 from cfl' to e'
emphasizes this point.
About another change between draft and final version we shall proba-
bly never be entirely certain. In the accompaniment of bar 37, Schubert draf-
ted a d] in the right hand on the second eighth note, as required by the origi-
nal vocal line of example 2. Then he changed this vocal line, but, either
intentionally or by oversight, did not change the piano part on the second
eighth. (He did, of course, change the ensuing eighths, having first written a
V 7 chord on the third and fourth eighths of bar 37 and a tonic on the first of
bar 38.) Thus the revised autograph has a di here, producing an even more
violent sense of reversal at the beginning of the bar and a heightened sense of
tension between voice and accompaniment as the voice produces a striking
cross-relationby insisting anew on the c|t' on the third eighth. The non-
autograph clean copy used by the printer retains a cH before the right-hand
chord on the fourth eighth note, but the implied natural before the c on the
second eighth, originally present in this copy as well, was erased by someone
("Schubert?," as the editors of the Neue Ausgabe say), and does not appear in
the first edition.
14. The which appears for the first time in the first
alternative ending,
edition — not —
autograph or the clean copy seems an unacceptable
in the
compromise for the sake of avoiding a high a'. Schubert was willing to trans-
pose entire songs rather than sacrifice the contour of the vocal line once he
had settled on it (see the discussion of range in the entire cycle below). This
alternative ending utterly changes the climactic gesture and the end of the
song; it is hard to imagine that it comes from Schubert's hand. I have never
heard a performance that incorporated it.
15. This same expression of a firming of resolve in the third line
of stanza 5 is heightened by the declamation of this stanza, especially as
Schubert revised it. The first line of the stanza is declaimed in breathless
snatches. The second is at least delivered whole, although it enters late in the
phrase and forces the phrase to finish early (according to the model of stanzas
1 and 2). The third was initially to be declaimed in a rushed snatch like the

first:

ovs un -ter sd -ner Rin - de

But Schubert revised this to return to the measured declamation of the pre-
vious stanzas, now changed in meaning (as is the original melody) by the
vaulting modulation to G-sharp minor, by the new rhythmic motion of the
piano part, and by the first forte in a whispered song. Variation in the rate of
text coverage expresses the same resolve. Whereas at the beginning of the last
stanza Schubert had halved this rate, covering only one line of text per four-
bar phrase of the model (the second phrase is shortened to three, as the re-
solve of the protagonist takes shape), at the beginning of the third line of the
stanza he moves back to covering one line per two-bar phrase. This impres-
sion of a decision taken is reinforced when the last line of the stanza is re-
peated — the first repetition of a textual unit in the song.
 Structure and Expression in a Schubert Song 173

16. A linear analysis should, I think, acknowledge the exceptional

nature of the case when a single, continuous vocal line resolves to a tonic an
octave higher than the one from which it started. Schenker, in the analysis
quoted by Lewin, has to concentrate the entire 3 — 2—1 Urlinie of the song in
bars 53-54 in order to get it in the correct octave!
17. Jacques Chailley also proposes that parts I and II be seen as inde-
pendent cycles, but scarcely concerns himself with defending or drawing
conclusions from the hypothesis as regards part I. His concern is with what
he sees as the revolutionary nature of part II. See Jacques Chailley, Le voyage
d'hiver de Schubert (Paris: Leduc, 1975).
18. See Georgiades, Schubert, pp. 357-59; and Feil, Franz Schubert,

pp. 26-29.
19. The last two lines of the first stanza of Erstarrung, for example, are
quite different in the twenty-four-poem version of 1824.
20. Schubert's close friendjosef von Spaun reports in his Aujzeichnungen
iibermeinen Verkehr mit Franz Schubert of 1828 that Schubert presented to his
friends what Spaun called "die ganze Winterreise" (see O. E. Deutsch, Franz
Schubert. Die Erinnerungen seinen Freunde, 2d ed. [Leipzig: Breitkopf und
Hartel, 1966], pp. 160-61). The editors of the Neue Ausgabe (4:4a, XX) refer
this to spring 1827, without stating their reasons. I can find no basis in
Spaun's chronicle for fixing the date more closely than at some time after
March 1827.
21. Robert Winter, "Paper Studies and the Future of Schubert Re-
search," in Schubert Studies, ed. E. Badura-Skoda and P. Branscombe (Cam-
bridge: Cambridge University Press, 1982), pp. 240-41.
See Schubert, Neue Ausgabe 4:4a, XXI.
22.
See Winter, "Paper Studies," p. 246.
23.
24. Is it significant in this connection that Schubert set the new poems
of the twenty-four-poem 1824 publication in the order in which they oc-
curred in that publication, simply skipping the ones he had already set from
the 1823 publication —
with the exception of Mut and Die Nebensonnen, whose
order he finally decided to reverse?
25. Cf. Schubert, Neue Ausgabe 4:4a, XX-XXI.
26. Heinrich Kreissle von Hellborn, Franz Schubert (Vienna: C. Gerold's
Sohn, 1865), p. 482.
27. See O. E. Deutsch, The Schubert Reader, trans. Eric Blom (New
York: Norton, 1947), p. 931.
28. By way of reminder, I list here the twelve poems in the 1823 pub-
lication and the twelve songs of Winterreise, part I:
I. Gute Nacht D minor
2. Die Wetterfahne A minor
3- Gefior'ne Trdnen F minor
4- Erstarrung C minor
5- Der Lindenbaum E major
6. Wasserflut E minor (orig. F-sharp minor)
7. Aufdem Flusse E minor
8. Ruckblick G minor
174 SCHUBERT
9. Irrlicht B minor
to. Rast C minor (orig. D minor)
11. Fruhlingstraum A major
12. Einsamkeit B minor (orig. D minor)
Georgiades and Feil
29. have drawn attention to the variety and the im-
portance of types of motion in Schubert.
30. The same is true of no. 22. The only case in which transposition
was caused by something other than a required high a' is no. 24. One can
only speculate that this may have been a result of the transposition of no. 22
away from A minor, or perhaps a positive second thought that nos. 23 and 24
should have the same tonal center. (See n. 9 above for the various sources and
variants.)
31. Georgiades {Schubert, pp. 194-99) and Feil (Franz Schubert, p. 30)
discuss transposition in Winterreise. Georgiades advances the hypothesis that
in Lieder, as opposed to opera or oratorio, the actual pitch location of a
—
song by which he means its particular place in a particular voice type has —
no importance. The evidence often given for this is that Schubert copied
songs from Die schbne Mullerin into different private song albums in different
keys. Although this evidence may support Georgiades's hypothesis, it does
not touch the matter of tonal interrelations in a larger whole, a matter with
which Georgiades is not concerned, for individual songs copied into private
albums are expressly detached from any larger whole. I shall have to put off
until another occasion a development of the argument that the function of
tonality within Schubert's cycle is especially to draw together some songs
and separate others.
32. The new transpositions produce a rather different effect in nos. 10
through 12, which were originally related as i- V-i. After the transpositions,
one can hear the G major of the end of no. 8 as the dominant of no. 10,
functioning across the parenthesis of no. 9. The A major of no. 11 now ap-
pears even brighter, more dreamlike, and more sharply contrasting than in
the original arrangement of keys. But Schubert seems deliberately to have
avoided the subsequent return to the key of no. 10 in no. 12. Nothing re-
quired him to transpose no. 12 to B minor. He might have taken it to C
minor and still avoided the high a', while retaining the tonal connection with
no. 10 that existed in the draft version.
33. This is undoubtedly why Aufdem Flusse could not be transposed as
a way of eliminating the high a'.
34. The opening motives of the piano parts of nos. 4 and 5 are the
same, save for the modal change; the arpeggios in barcarolle rhythm of
stanzas 3-4 and 6—7 of no. 5 are slowed down and transferred to the intro-
duction of no. 6.
 Schubert's Ndhe des Geliebten (D. 162)

Transformation of the Volkston

WALTER FRISCH

The Lied should be the simple and comprehensible musical expres-

sion of a precise feeling, so that it can thereby encourage the partici-

pation of every voice capable of natural singing. As a small artwork,

easily graspable at a glance, it must necessarily be a correct, complete
whole, whose particular effect consists of the unity of the vocal part,
and whose instrumental accompaniment, if not dispensable, should
nevertheless serve only to support the voice. 1

In 1786 Johann Friedrich Reichardt summarized thus the aes-

thetic idealsof the German Lied. He and the other composers
of the Berlin school, including Krause, Schulz, and Zelter, at-
tempted to practice what they preached, aiming for Sangbarkeit
in their ownsongs. It is therefore not surprising that they
turned frequently to the folk style, the Volkston, as their model.
As Reichardt explained his preface to Frohen Liedern fur Deutsche
Manner of 1781, "folksongs are truly that upon which the
genuine artist relies when he
begins to suspect his art is on the
— —
wrong track as the sailor relies on the north star and which
he studies the most for his profit." 2 Most folksongs are, of
course, simple strophic structures, in which one strain of mu-
sic serves for all stanzas of the poem. The majority of the
3
Berlin school songs follow that plan.
In the late eighteenth century there was no shortage of
poems to set in this style. Herder published two anthologies of
collected Volkslieder in 1778-79. Major poets also began to
cultivate the folk idiom: as early as 1773 Goethe published
I76 SCHUBERT
"Heidenroslein," one of his most renowned and artful crea-
tions in the (ostensibly) simple style.
To a young, ambitious Austro-German composer in the
firstdecades of the nineteenth century, the folk idiom would not
have seemed an especially promising one to pursue. Schubert,
who began composing songs in about 181 1, largely avoided it
at first. He concentrated instead on long, narrative ballads,
in which he could explore the dramatic styles of recitative,
arioso, and aria, derived from the more prestigious tradition of
opera seria. But three years later his allegiance shifted; between
1814 and mid-1816 he composed almost 100 simple strophic
songs, many of which capture the folk tone admirably.
In a work like Heidenroslein (D. 257) formal design and
melodic and harmonic content seem perfectly coordinated.
The song manages to attain sophistication without ever over-
stepping the aesthetic bounds of its folklike idiom. The same
could be said for many of the strophic songs of 18 14- 16. But
Ndhe des Geliebten (D. 162), a strophic Goethe setting com-
posed on 27 February 181 5, manifests a compelling dialectic
between form and content. The harmonic and melodic aspects
of the song utterly transcend the folk style associated with a
strophic framework. Ndhe des Geliebten is an appropriate sub-
ject for a critical-analytical study because its poetic and musical
background can be traced quite precisely and because its auto-
graph sources reveal something of Schubert's endeavor to recon-
cile the conflicting demands of outer form and inner content.

II

The story of Ndhe des Geliebten begins with neither Goethe nor
Schubert, but with "Andenken," a four-stanza poem written
in 1792 by Friedrich Matthisson:

Ich denke dein,

Wenn durch den Hain
Der Nachtigallen
Akkorde schallen!
Wenn denkst du mein?
Ich denke dein
Im Dammerschein
 y
Schubert's "Ndhe des Geliebten' 177

Der Abendhelle
Am Schattenquelle!
Wo denkst du mein?
Ich denke dein
Mit siisser Pein,
Mit bangem Sehnen
Und heissen Thranen!
Wie denkst du mein?
denke mein,
Bis zum Verein
Auf besserm Sterne!
In jeder Feme
Denk'ich nur dein! 4
[I think of you, when the chords of the nightingales

sound through the grove. When do you think of me?

1 think of you in the twilight of the clear evening, at

the shadow's source. Where do you think of me?

I think of you with sweet pain, with anxious longing,

and with hot tears! How do you think of me?

O think of me, until we are united on a better star! In
every distant place, I think only of you!]

The poem's device is clear: each stanza describes how the poet
reacts to, or reminded of, his beloved. The first three begin
is

with "Ich denke dein"; the final one inverts the syntax with the
plea "O denke mein." The last line of each stanza shows a simi-
lar process working the other way around: the first three stanzas
conclude with "Wenn," "Wo," or "Wie" "denkst du mein?";
the last one with "Denk'ich nur dein!" Accompanying the
when-where-how progression is a kind of phenomenological
or sensuous sequence. The first stanza deals with aural percep-
tion (nightingales' chords), the second with visual perception
(twilight and shadows), the third with physical feeling (pain
and tears). Stanza 4 culminates this progression as the poet
imagines being united with his beloved. Yet in the final couplet
he acknowledges the real distance that separates them.
Matthisson's poem appeared in 1802. It was given modi-
fied strophic settings by Beethoven in 1809 (WoO 136, pub-
lished 1 8 10), and by Schubert in April 18 14 (D. 99). Schubert's
first strophe (ex. 1) will suffice to give an idea of how well he
I78 SCHUBERT
Example 1 Schubert, Andenken (D. gg)

m f^ft
Ich den fee dem, wnn

*u >p5
$
fc*
^ PPI
fp:

» #
i
EfcEli:
^
/p

^Ji ji
durch den
i
LP
Haiti ckr
i

r
fach
i'

tl -
l
JUpi^ni
aai - [en Ac- cor- de — ^
scnai - ien\

1 fJ h
i '

\AUmm
Warm AliAjkSaci'
denkst
PMUrf
Ait
du vH<Vr*i
tuein:
.'

could capture the naive folk style. (Throughout this article I

shall use "stanza" to refer to the unit of the poem, and "strophe"
to refer to the musical unit.) The brief piano introduction
evokes horn calls, a sonority often employed in the music
of this period to connote memory and/or distance. (I return
 Schubert's "Ndhe des Geliebten" 179

to this topic below.) The vocal melody consists of two sym-

metrical, eminently sangbar four-bar phrases, the second mov-
ing to the dominant. There follows a one-bar piano extension,
then the two-bar "punch line," where the inquistive dominant
seventh underpins the poetic query, "Wann denkst du mein?""
The changes in strophes 2 and 3 involve principally piano
figuration. The final strophe begins like the others but be-
comes considerably expanded at the final couplet "in jeder
Ferne/Denk'ich nur dein." This climax is beautifully achieved
by Schubert, chiefly through melodic breadth and repetition.
Yet it must be admitted that his setting does not quite capture
the phenomenological progression of Matthisson's poem, espe-
cially the final acknowledgment of "Feme," where we would
expect something quieter, more inward. Andenken remains
firmly— —
and charmingly rooted in the Volkston; introspec-
tion and deeper reading of a poem are not part of its world.
Well before it attracted the attention of composers, Mat-
thisson's poem inspired other poets. In 1792, very soon after its
creation, it became the model for "Ich denke dein," a five-
stanza poem by Frederike Brun, a friend of Matthisson's and
herself a minor poet:

Ich denke dein, wenn sich im Bliitenregen

Der Fruhing malt,
Und wenn des Sommers mildgereifter Segen
In Aehren strahlt.
Ich denke dein, wenn sich das Weltmeer tonend
Gen Himmel hebt,
Und vor der Wogen Wut das Ufer stohnend
Zurucke bebt.
Ich denke dein, wann sich der Abend rothend
Im Hain verliert,
Und Filomelens Klage leise flotend
Die'Seele ruhrt.
Beim triiben Lampenschein in bittren Leiden
Gedacht ich dein;
Die bange Seele flehte nah am Scheiden:
Gedenke mein!
Ich denke dein, bis wehende Cypressen
l80 SCHUBERT
Mein Grab umziehn;
Und auch in Tempe's Hain soil unvergessen
Dein Name bliihn. 6

[I think of you, when spring is reflected in the bursting

forth of blossoms, and when the gently ripened yield of
summer shines upon the ears (of corn).
I think of you when the ocean rises sonorously toward
the sky, and when the shore, moaning, recoils from the
violence of the pounding waves.
I think of you when the reddening evening dissolves in

the grove, and when Philomel's lament, gently fluting,

stirs the soul.
By the dim lamplight, in bitter suffering I thought of
you; my anxious soul implored you to the point of dying:
think of me!
I will think of you, until the swaying cypresses sur-

round my grave; and even in Tempe's grove shall your

name bloom unforgotten!]

Though she draws upon both the general design and specific
features (but not the metrical scheme) of the Matthisson poem,
Brun has made a muddle of his Even granted po-
elegant lyric.
etic license, her syntax is clumsy and awkward. Matthisson's
logical progression of imagery has become a chaos of meta-
phors. Each stanza now presents two different images (one in
each couplet), weighed down by excessive adjectives and par-
ticiples. The references to Philomel and Tempe seem especially
gratuitous amid the nonspecific natural and physical imagery.
The shift to past tense in stanza 4 also makes little structural
sense; it is not an effective preparation for the future-oriented

thoughts of the final stanza.

None of these flaws prevented composers from setting
the poem. A simple strophic version of it by C. F. Zelter ap-
1
peared in 1795 (ex. 2). Schubert had conveyed the charm of
Matthisson's poem, even if he had not risen to its subtleties;
but Zelter descends directly to Brun's level of insipidness. The
chromatic neighbor and passing notes (All and Bt| respec- ,

tively) and the little roulades on "regen" and "Segen" seem as

superfluous as much of Brun's imagery. Neither "natural" nor
"simple," they violate the Volkston supposedly at the basis of
this style.
 Schubert's "Ndhe des Geliebten" 181

Example 2 Zelter, Ich denke dein

Ausdruckswll

$*i j
J J i

r
J r y r r ir ^r r r
i

ltd den- fee dcin; wenn sick tm "Btu - tW - re qen der frvk-lind

pm^ PiNiN=
i -UJU fejiti

gpps I
r f=f
^ f r r

PS
tr

> i>

?r
molt,
/ 1
r r m'r &
iind-^enndesSornv/uramiU u-retfter <5e
•"">*#

-
f r fti

qin w, Aek-rtn
p

strahlt.
a

The of Zelter's ic/z denke dem would

historical significance
be minimal song caught Goethe's at-
if not for the fact that the
tention. It is well known that Goethe admired Zelter's songs;
indeed, he often praised Zelter's settings of his own poems,
in which he found the "total reproduction of the poetic inten-
8
tions." In June 1796 he wrote to the wife of his publisher,
"Zelter's melody to the poem 'Ich denke dein' had an incon-
ceivable charm for me," adding that the song had directly in-
spired him to write his own poem, entitled "Nahe des Gelieb-
9
ten," to fit Zelter's melody:

Ich denke dein, wenn mir der Sonne Schimmer

Vom Meere strahlt;
Ich denke dein, wenn sich des Mondes Flimmer
In Quellen malt.
Ich sehe dich, wenn auf dem fernen Wege
Der Staub sich hebt.
182 SCHUBERT
In tiefer Nacht, wenn auf dem schmalen Stege
Der Wandrer bebt.
Ich hore dich, wenn dort mit dumpfem Rauschen
Die Welle steigt,
Im stillen Haine gen' ich oft zu lauschen,
Wenn alles schweigt.
du seist auch noch so feme,
Ich bin bei dir;
Du mir nah!
bist
Die Sonne sinkt, bald leuchten mir die Sterne.
O, warst du da!
[I think of you when the sun's luster shines from the
sea; think of you when the moon's glimmer is reflected
I

in the fountains.
I see you when on distant paths the dust is raised, in

deepest night when on the narrow bridge the wanderer

trembles.
I hear you when with a dull rushing the wave rises. In

the still grove I often go to listen, when all is silent.

I am with you; however far you might be, you are close

to me! The sun is setting, soon the stars will shine for
me. O, that you were here!]

The poem first appeared in 1795 in the Arienbuch der Claudine

von Villa Bella.A few months later it was printed in Schil-
ler'sMusenalmanach for 1796, where a setting of the poem by
Reichardt was added as a musical insert. In 1809 Reichardt
published a revised setting in a collection of Goethe songs
(ex. 3).'°
Reichardt follows the Zelter/Brun model closely. There
are two
parallel musical phrases, corresponding to the paired
lines of each stanza. As in Zelter, the first phrase cadences on
the dominant, the second returns to the tonic. But Reichardt
improves upon his predecessor's rather cramped song, allowing
the poem more time to unfold. Zelter had virtually ignored the
caesura (in Brun's poem) between the opening verb phrase and
the dependent "wenn" clause. Reichardt creates a pause at this
point and also wisely shifts the musico-poetic stresses. Zelter
had placed "dein" on a downbeat; Reichardt moves the down-
beat forward onto the verb "denke," and thus onto the new
verb at the analogous spot in each succeeding stanza. (The
verbs change only in Goethe's poem, not in Brun's.)
 Schubert's "Nahe des Geliebten" 183

Example 3 Reichardt, Nahe des Geliebten

fil
Inni£

Ick
i
ng
den - lee dein
^iNyp
iwnn mtrdfriStfn-ne
i
LW
6cfom-
'
f
•m^rvotaJ^-
33=*
re
-
strand ;
=Fff

ie(t

dm- fee debt

^^ S
, wmn«siehcksMon-d&5 Him
&*
-
^jiwin Quel -
mm
Un rndt

t^m
n a
it y J £2* i
PT
fe
SB I P?
Reichardt also captures the folk idiom more effectively
than Zelter. Even the chromatic Etis and the roulade on
"Schimmer" do not intrude upon the basic Sangbarkeit of the
vocal part. But this is hardly a great song; indeed, the idea of
greatness in such a context would have seemed inappropriate to
Reichardt and his Berlin colleagues. As Wiora has noted, they
were creating essentially Gebrauchsmusik, not art songs. 11 Al-
though Reichardt prided himself on reading a poem aloud over
and over before setting it to music, so as to catch every nuance, 12
in 1797 he thought nothing of publishing his original (1796)
music for Nahe des Geliebten with the Brun poem substituted
13
for the Goethe!

///

When Schubert sat down to compose Nahe des Geliebten on 27

February 181 5, he thus had behind him, and probably in his
184 SCHUBERT
mind's ear, a fairly extensive poetic and musical tradition an —
"Ich denke dein" tradition, so to speak, to which he had him-
self contributed Andenken. It was a tradition he had already be-
gun to transform four months earlier, with his first Goethe
settings. There is, of course, little of the Volkston in the very
first, Gretchen am Spinnrade, of 19 October 18 14. But he soon

turned to shorter Goethe lyrics that were suited —

and intended
—
by the poet for the simple style. First came the strophic
Nachtgesang and Trost in Tranen (both composed on 30 Novem-
ber 1 8 14), in which Schubert achieves a synthesis of the naive
and the sophisticated. In Schafers Klagelied, composed on the
same day, the folk style is skillfully adapted to the discursive
ballad or scena format.
There would have been every reason to expect that when
Schubert turned his attention to Ndhe des Geliebten later that
winter, the resulting song would exhibit a similarly refined folk
idiom. Indeed, on the same day in February he composed just
such settings of Goethe's Am Flusse (D. 160) and An Mignon
(D. 161). But what emerged in Ndhe des Geliebten was far re-
moved from any song he had yet composed, and light-years
beyond the "Ich denke dein" tradition (see plate 1).
The tempo indication alone suggests that Schubert is at-
tempting to capture a special mood: Ndhe des Geliebten is the
first song for which he used the phrase langsam, feierlich mit
Anmut (slowly, solemnly, gracefully). (The metronome mark-
ing is his own, or was at least approved by him; such figures
appear for all the songs of op. 5 in the first edition.) The con-
junction of langsam and feierlich appears in a number of later
songs; yet on only one occasion, in Die Allmacht (D. 852,
1825), is it specifically associated with the accompaniment pat-
tern of full, repeated chords we find in Ndhe des Geliebten. Die
Allmacht is, of course, a fervent (if somewhat overblown) ex-
pression of religious feeling, for which the notion of "feierlich"
and the grandeur of thick chords seem justified. But it is in-
deed remarkable that a small love lyric like Ndhe des Geliebten
called forth this style from Schubert. He makes of it a kind of
secular Allmacht, an impassioned hymn to a beloved of human,
rather than divine, dimensions.
Schubert wrote two drafts of Ndhe des Geliebten on the
same day. 14 Both have the same tempo /mood indication and es-
 Schubert's "Ndhe des Geliebten" 185

Plate 1 Schubert, Nahe des Geliebten (D. 162). Facsimile of first

edition (Vienna: Cappi & Diabelli, 1821), with bar numbers added, in
italics. By permission of the Music Division, The New York Public
Library at Lincoln Center, Astor, Lenox, and Tilden Foundations.
186 SCHUBERT
sentially the same musical design. But the first is cast in I
meter, and the accompaniment is subdivided into two chords
per beat. As Schubert realized almost immediately, this pattern
was too fussy, too busy to convey the grandeur and spacious-
ness he sought. He wrote "Gilt nicht" on the page, struck it
through with several lines and took up another leaf, writing
out the song in the version we know today (with a few small
but significant exceptions, to be discussed below). This was
published as op. 5, no. 2, in 1821.
Nahe des Geliebten is one of the earliest Schubert songs to
display the special kind of intimacy between music and text
analyzed by Georgiades in the 1823 Wandrers Nachtlied, a Goethe
poem which Schubert "transformed into a composition
. . .

by penetrating, as it were, through the poem and beyond it to

the deeper level that sustains it— to the language —
and by draw-
ing directly upon this."
13
Of course, "Nahe des Geliebten" is
a folklike poem comprising four regular four-line stanzas, a
very different kind of lyric from the terse, epigrammatic eight-
line "Wandrers Nachtlied." Yet, as Georgiades has suggested, it
too has a compelling deep structure:

The content of the language, which in Matthisson and

Brun contained only the latent possibility of becoming
poetry, is here transformed into genuine poetry. The repe-
titions of "Ich denke" are replaced in each stanza with a
new verb: "Ich denke dein" becomes "Ich sehe dich" and
"Ich hore dich," each one linked with a physical realiza-
tion. Matthisson's final strophe began with the inversion
"O denke mein" instead of "Ich denke dein"; this is no
more than a formal trick, a sentimental closing formula.
Since in Goethe's version the sense of dialogue is elimi-
nated entirely, there is no such inversion; but in its place in
the final stanza Goethe brings about a synthesis of "ich"
and "du," a deepening. No longer "denke"; no longer
an image projected in the outer world, "sehe," "hore";
but the inner realm, "ich bin . .du bist mir nah" and
.

"O, warst du da!," whereby the monologic character of

the poem is stressed. This is also reflected in the new
title: "Nearness of the Beloved," in spite of the actual

distance. Thus in Goethe, as already in Brun but not—

 Schubert's "Nahe des Geliebten" 187

inMatthisson —
the "Ich denke dein" remains without
16
a
rhyme, without an answer from the beloved.
Georgiades is perhaps too hard on Matthisson's poem (and
not hard enough on Brun's). Although he does not change the
opening verb in each stanza, Matthisson does, as I have sug-
gested, trace a kind of phenomenological process, culminating
in a "synthesis" in the first couplet of stanza 4, then an accep-
tance of the actual Feme is the final couplet. He surely deserves
credit formore than a "formal trick" ("formales Spiel"). Brun
ignored this progression entirely; but Goethe builds upon it,
makes it explicit. He absorbs several natural images from both
his poetic sources —
the stars, the grove, the twilight from
Matthisson (the latter two also were also borrowed by Brun),
and the sea and waves from Brun. But as he had done with the
verbs, he fashions from these images a consistent pattern in-
volving a contrast between the two couplets of each stanza.
The opposition is perhaps clearest in stanza 1: in lines 1-2 the
17
sun radiates, in 3-4 the moon shines. Stanzas 3 and 4 show a
similar pattern. In stanza 3 the rising wave of the first couplet is
opposed to the still, silent grove of the second; in stanza 4 the
immediacy and intensity of "Ich bin" and "Du bist" are op-
posed to the sinking sun, the shining stars, and the less em-
phatic conditional "warst du." Even in stanza 2 one can detect
something of this contrast, between the active image of the
dust being raised and the darker, more passive one of the wan-
derer trembling in the dark night.
This device is Goethe's own inspiration; there is no hint of
it in Matthisson or Brun. Nor is it captured in Reichardt's set-

ting, where the two musical phrases of each strophe are paral-
lel,and virtually equal in importance. But Schubert seizes on
the poetic pattern of strong /weak or active/ passive, and on the
presence of the forceful verb phrase at the beginning of each
stanza. The result is a musical strophe divided into two phrases
of very different weight and shape. Schubert rejects the rather
conventional Zelter-Reichardt plan of having the first phrase
cadence on the dominant, the second on the tonic. Instead,
both phrases close on the tonic (bars 5 and 8), but approach it
in strikingly different ways.
Close in on — rather than close on — might be a better
 —

188 SCHUBERT
term for how the tonic is first reached in Nahe des Geliebten.
Most Schubert songs begin by defining the tonic and /or sug-
gesting the principal thematic material. The piano introduc-
tion is usually harmonically rounded, ending with a half or full
18
cadence. Andenken (see ex. i) is typical of many of the folk-
like songs: in the first four bars the piano firmly establishes the
tonic (though not, in this case, the thematic material) before
the voice enters in bar 5. Nowhere in Schubert's Lieder, how-
ever, do we find an "introduction" quite like that of Nahe des
Geliebten, which begins well away from the eventual tonic G-
flat and offers no cadence. Indeed, the voice has virtually to

force its way in at the end of bar 2.

The
introduction begins over a pulsating dominant pedal
in the key of E-flat minor, a dark and unstable sonority that is
especially striking in the context of this ostensibly simple love
lyric. (See ex. 4 for harmonic reduction.) The \\ chord in the
second half of bar 1 refuses to function cadentially. Instead the

bass rises a half step to the flatted sixth degree, O, and the har-
monic pace doubles as the right-hand chords ascend chro-
matically in parallel sixths.
19
On the final three beats of bar 2
the harmony We expect a
coalesces into a German-sixth chord.
cadence returning to and resolving the i* chord
(as in ex. 4)
abandoned in bar 1, and followed by the entrance of the voice.
Instead, the voice bursts in on the last beat of bar 2, above
the dissonant augmented-sixth chord. It is near the very limit
of its range (a difficult beginning, as any performer can attest),
doubling the top line of the piano part. On the downbeat of
bar 3 the bass Bf>, the proper linear resolution of the O, is in-
terpreted not as the dominant of E-flat minor, but as the third

Example

1 Abu
u k
f
——
— —-—^—
' k- k< ifo (%
—
•we expect

yj r\
flp p
"g *
— * -~
*r-4

m ^* -)
m
; \
*
fc
1?V
J-
ek T i$ 3ZT (fasewgi Gem. 6^(1$ V i) nr
fe

& • I
 V

Schubert's "Ndhe des Geliebten" 189

degree of G-flat major. As if stunned by its own arrival —

by its
own bold usurpation of the bass Bl> —the first-inversion chord
resonates for six full beats before descending (rather cursorily)
toward the root. Above this the voice holds the G? on "den-"
for five beats. As Georgiades points out, this remarkable ex-
tension of the first syllable of "denke" creates a durational pro-
portion of 5:1 between the two syllables, considerably larger
than the 3:1 at the opening of Andenken (and also much greater
than the 2:1 of Reichardt's [ex. 3] and the 1:1 of Zelter's [ex. 2]).
One could scarcely imagine more musical emphasis on the
verb phrase.
Throughout the first three bars the bottom line of the
piano part does not sound like a functional bass line; it is rather
the lowest element of the throbbing chordal texture. But on
the downbeat of bar 4 it begins at last to behave more con-
ventionally. It detaches itself from the pulsating eighth-note
motion, moves into the genuine bass register, and leads pur-
posefully to a cadence on G flat through the first normal pro-
7
gression of the song, ii|— 1^— — I. In doing so, the bass also
returns to and "explains" the O, the pitch that had proved so
perplexing in bar 2. There it had dropped down a half step to
the "false" dominant note and had failed to generate a satisfac-
tory harmonic resolution; now it rises smoothly by half step
toward the true dominant, Dk
Like the bass line, the voice establishes its independence in
bar 4. In the preceding two bars it too has been part of the
dense texture, doubling the top line of the piano. But after the
A\> of "Sonne" it goes its separate way, arpeggiating up the

ii triad to Et, sitting on the dominant D!> for five beats (by

analogy to the tonic G> sustained for five beats at "den-[ke]")

and rejoining the piano at the cadence on beat 7 of bar 4. The
vocal phrase thus traces essentially an octave descent between
the two G^s.
The now-independent voice and bass line of bars 4—5
are also virtually mirror images, forming
splendid musical
a
analogy to the sun's reflection in the sea. In bar 4 both parts
move to Df>, the voice approaching from above (the appog-
giatura Et) the bass from below (O— Q—
0»). In the next bar
they both resolve the 0> to the tonic, the voice again from
190 SCHUBERT
above, the bass from below. (At "Mee-re" the Dl> certainly
bears more weight than the leading tone; the temporal propor-
tion is again 5:1, as in the opening "denke.")
Though it comes only halfway through the song, the ca-
dence of bar 5 imparts a feeling of finality, of closure. It firmly
resolves the harmonic ambiguity and marks the endpoint of
octave descent in the voice. Where Reichardt and Zelter made
of the first musical phrase a conventional antecedent, Schubert
has thus fashioned a musical whole corresponding to the po-
etic sense of the active verb— a phrase in which enormous ten-
sion and dissonance are created and resolved within the space
of five bars. This kind of compressed intensity is rare in early
Schubert. What is unique is the placement of the musical cli-
max within the piano introduction, or more precisely, at the
moment where the introduction is overtaken by the entrance of
the voice.
Everything that follows the cadence of bar 5 functions
musically as a kind of coda, confirming the tonic. First, in bar
6, comes another cadence over a tonic pedal; then in bars 7-8
Schubert moves away to V of vi, coming back to the tonic
7

through vi, IV, 1% V 7 Nothing unusual in common-practice

.

music, this progression here has a very special function: it

serves to reintegrate into the tonic region of G-flat major pre-
cisely the harmonies that had been made to seem so remote at
the opening of the song: vi (E-flat minor) and IV (C-flat
major). In bar 2 the C^flat harmony had been chromatically
obliterated. In bar 4 it had returned and explained itself as ii^ of
G-flat major. Yet even there it had been approached abruptly,
without preparation or smooth voice leading. Now at last the
C-flat harmony is placed fully in context, embedded in a strong
functional progression. The progression is given rhythmic em-
phasis by the accompaniment, which in bars 7-8 momentarily
abandons the continuous, pulsating eighth-note motion. The
first beat of each group of three is articulated only by the bass
note. This waltz-like rhythm lasts only as long as the actual
cadence; during the piano postlude Schubert gradually re-
introduces the original pattern (thus preparing the next musi-
cal strophe).
Phrase 2 of the song, then, has retraced and clarified the
harmonic course of phrase 1. In a similar way the vocal line of
 Schubert's "Ndhe des Geliebten" 191

bars 6-8 retraces the octave descent of bars 3-6, almost as if in

dreamy remembrance. As in bar 3, the voice moves at first
from O to DK but now the stark initial leap of a fourth is filled
in by gentler stepwise motion. In the first phrase the unex-
pected Qe had been articulated on both upbeat and downbeat
(bars 2-3). Such emphasis is no longer needed in the second
phrase; the pitch and tonality of G> have been established. The
upbeat thus yields to an Flj on the downbeat of bar 6. Despite
the similarity of melodic profile, there are significant differ-
ences between the two phrases. The first is harmonically un-
stable, the second firmly in the tonic. The vocal line of the first
phrase had refused to rest on the D\> of "dein," pushing impa-
tiently down the scale. Now
it is more content to linger: it

touches back to the high G>, then rests before resuming the
descent. At "Mondes Schimmer" the phrase regains the high G>
one last time, then moves to the third degree, before cadencing
from the fifth to the tonic.
The final note of the vocal part was a source of indecision
for Schubert: it is different in each of the three autograph
sources for the song. In the earliest (the first draft in I meter)
the voice concludes on the high D!>, which is sustained for five
beats (ex. 5a). In the second autograph (the first copy of the
final version, made on the same day) the vocal line descends to
the third, Bf> (ex. 5b). Only in the third autograph (a copy of
the final version, which Schubert made in 18 16 for a volume of
autograph songs sent personally to Goethe), and in the pub-
20
lished version (plate 1) is there a conclusion on the tonic. Each
of these melodic cadences is plausible. Although it weakens the
melodic parallelism between the two vocal phrases, a conclu-
sion hovering on the dominant pitch Dl> suits the tranquil im-

Ex ample 5'

[en molt:.
192 SCHUBERT
agery of the second couplet. The Bl> strikes quite literally (and
effectively) a medium between openness and closure.
The G> Schubert ultimately selected does, however, seem
the most effective conclusion. It confirms the parallelism of
melodic profile between the two phrases and creates an elegant
analogy between the emergence and departure of the vocal
part; it brings the voice down to the same pitch as the top line
of the piano (and the same register, if the song is performed by
a soprano). The voice thus rejoins —
blends back into —
the ac-
companiment an octave below their original meeting point.
More significantly, as Georgiades points out, the G> creates a
direct motivic association between the opening and closing vo-
cal gestures: the initial G»-DI> descent is completed by its in-
version, the cadential 0>-G>. The two complementary mo-
tives/intervals are then juxtaposed more immediately in the
brief piano postlude, where the G>-Dt> stepwise descent in the
right hand is answered by the Dl>-G> of the left. The postlude
also serves thereby to prepare the opening vocal statement of
G>-D> in subsequent strophes.
Distance: this is another important quality of the second
phrase. If by its intensity the first phrase strives to bring the
beloved closer, the tranquillity of the second suggests the poet's
acceptance of her actual remoteness. Schubert conveys this not
only with the harmonic and melodic procedures already exam-
ined, but also with a specific musical image —
horn fifths. In
the second half of bar 5 the texture of the right hand of the
piano, which has up to this point consisted mostly of full
triads, thins out to two notes and assumes the unmistakable
sound of a pair of horns. The sonority is evoked again in the
second half of bar 6, then is absorbed once more into the
thicker chordal texture.
Originally associated with hunting, of course, horn calls
in early Romantic opera and Lieder often evoke the remote
world of the German forest, as in Weber's Der Freischutz or
Schumann's song Waldesgesprach from op. 39. In nonvocal and
nonorchestral music, the musical metaphor becomes more
generalized, suggesting separation and/ or departure. The first
movement of Beethoven's "Les Adieux" Sonata, op. 81 a, is an
21
explicit example. In the finale of Papillons Schumann presents
the Grossvatertanz tune in horn fifths. Papillons, of course, has
 —
Schubert's "Ndhe des Geliebten" 193

nothing to do with hunting or forests; here the gradual dimin-

uendo and fragmentation of the tune depict the departure of
the dancers from the ball. Schubert's own D-Major Piano
Sonata (D. 850) is full of horn sonorities (which is surely what
provides the "country" feel that has caused commentators to
associate the piece with Gastein, where it was composed in
1825). They are also prominent in the 1823 Wandrers Nachtlied
(D. 768), analyzed by Georgiades elsewhere in this volume,
where they are again associated with the forest world.
Schubert is undoubtedly drawing upon these kinds of
association in Ndhe des Geliebten, where the horn fifths appear
in remote dynamics pianissimo in the printed edition and an
even softer triple piano in the autographs —and introduce the
poetic couplets dominated by images of distance or darkness:
the flickering moon, the deep night, the silent wood, sunset
and starlight. In 181 5 he would have known, of course, only
the Beethoven sonata from among the pieces mentioned. But
he need have looked no further than his own earlier works for a
precedent for using horn fifths in this context. It occurs in an-
other song about absence, Andenken, the precursor of Ndhe des
Geliebten already discussed. Here (see ex. 1) the horns lie in a
sense outside the song, appearing only in the four-bar intro-
duction. They somewhat gratuitous, contributing to
are thus
the general Volkston, but bearing no significant relationship to
the song or the poem. In Ndhe des Geliebten, however, the horn
fifths literally grow from and blend back into the piano texture
and are intimately linked with the text.
They also serve to isolate the strongest motivic/intervallic
feature of the song, the fourth, which here (bars 5-6) ascends
stepwise through G>-Al>-BI>-Ck (The motive is at a different
pitch level than the initial and concluding G>-0>, but the asso-
ciation is, I believe, clearly audible.) In this stepwise form the
motive first appears distinctly at the end of bar 3, descending
in parallel tenths between outer parts. It is heard twice again in
the two subsequent bars, descending and ascending (respec-
tively) in parallel sixths in the right hand of the piano part. In
the third statement (bar 5), the motive ascends through the
notes of the dominant chord, D-flat. At the long-awaited tonic
resolution in the second half of this bar we would expect an-
other statement in parallel sixths. But it is precisely at this
194 SCHUBERT
point (beat 7) that the right-hand texture thins to support the
ascending fourth with bare horn fifths instead of a thicker triad
and parallel sixths. The motive is given further emphasis by
the canonic echoes of the horn calls in the left hand of bars 6
and 7. What was a banal effect in Andenken, then, has become
both evocative and structural. In Ndhe des Geliebten the horn
call serves purely musical functions: both to clarify the har-
monic design (by coinciding with the tonic arrival) and to cul-
minate the motivic process (by uncovering the basic motive).
To summarize, then: the musical design of Ndhe des
Geliebten comprises a piano introduction beginning off the
tonic, which is then reached suddenly and unexpectedly at the
same time the voice enters; a vocal part that begins in the ex-
treme of its register on the highest pitch of the song and de-
scends through an octave; a decrescendo leading to a strong,
functional cadence halfway through the song; and a second,
codalike phrase serving to confirm or stabilize the tonic. It was
the shape of Goethe's stanza, as described above, that moti-
vated this extraordinary plan. Schubert understood that each
stanza has its moment of greatest emphasis at the very begin-
ning, with the verb phrase. To capture this he needed to build
musical tension before the entrance of the voice. The normal
tonic-establishing piano prelude would not suffice: hence a
piano introduction that builds up enormous musical tension,
culminating in the entrance of the voice. To convey the softer
or darker imagery in the second half of each stanza, Schubert
creates for bars 6-8 a kind of echo of the first musical phrase,
one that confirms the tonic and reintegrates the anomalous
harmonies of the opening phrase.

IV
The preceding discussion has implied that the single musical
strophe of Ndhe des Geliebten is well suited to all the poetic
stanzas. Not all critics are agreed on that point. Hans Gal has
argued:

The opening phrase, "Ich denke dein" . expands into a

. .

wonderfully noble melody. . .This simple phrase has

.

the same magical effect when it returns in the second and

 Schubert's "Nahe des Geliebten" 195

third stanzas to the words "Ich sehe dich" and "Ich hore
dich." But anyone who
has a feeling for such matters will
be disappoined by this phrase in the last stanza. Here the
words are "Ich bin bei dir," and the false accent on "bin"
22
can hardly be camouflaged; the spell is broken.

There is some justice to Gal's broader point, though not to

his specific reasoning. The vowel sound in "bin" may not be as
broad "denke," "sehe," and "hore" (and thus not lend it-
as in
self as easily to extension). But the accent is surely not "false":
the word would be stressed in normal speaking, and in the
poem it occupies the proper place in the iambic pattern. Gal
might reasonably have objected to the less plausible stress on
"noch" in this stanza (bar 4). He might also have claimed that
Schubert failed to capture the different phenomenological status
of the phrase "Ich bin bei dir, " whose immediacy, as Georgiades
points out, significantly alters the pattern of the preceding verb
phrases. And yet, taking Schubert's side, we might argue that
by retaining the same strophe for the final stanza he demon-
strates musically that despite the apparent nearness — despite the
"bin" — the beloved remains equidistant throughout the
really
poem. Indeed, the conditional "O warst" in the last line im-
plies that the nearness of the beloved is only in the poet's mind.
Good performers can (and should) compensate for some
of the differences in nuance between the poetic stanzas, for ex-
ample by softening the opening attack somewhat in strophes 2
and 3, then returning to a much grander forte for the "bin" of
the final strophe. But the real strophic "problem" in this song
cannot be solved by any performer, for it lies not in the rela-
tion of the musical strophe to the poetic stanza, as Gal would
have it, but in the status of the "introduction" and "postlude."
It is here that form and content come genuinely into conflict.

The introduction, one of the elements most indispensable

to the shape of the musical strophe as I have described it, is not
repeated in each strophe: bars 1-2 are heard only once. The
postlude of bars 8-10 does service as the introduction for
strophes 2-4, in which the high G\> on the last beat of bar 2
("Ich" in all stanzas) is to be sung without any accompaniment.
The absence of the original introduction alters radically the re-
lationship between the first and all subsequent strophes. No
196 SCHUBERT
longer entering of a musical wave initiated by the
at the crest
piano, the firstvocal phrase of strophes 2-4 must instantly
create — or re-create —
its own climax before subsiding to the

cadence. Despite its motivic relationship to the opening vocal

figure (both share the descending fourth), the postlude does
little to prepare, or to mitigate, the rather violent attack on the

high Gk
Schubert himself was well aware of this situation. The
status of the introduction is the most significant and suggestive
issue raised by the autograph sources of Nahe des Geliebten. In
none of the autograph manuscripts is bar 3 preceded by the be-
ginning indication of a repeat sign: bars 1-2 are thus to be re-
peated with each strophe. This is also suggested by the last
bar, bar 10, in which there is no separate upbeat for the voice
("Ich"); in the autographs the closing repeat sign is preceded in
the vocal part by two complete half rests, the second with a
fermata that corresponds to that over the G> chord of the piano
(ex. 6). (Schubert did not use any dotted half rests in these au-
tographs; the half rest thus can be interpreted as indicating a
full half measure.) These notational features clearly imply a
repetition of the full introduction: each strophe is to have its
own introduction and postlude. Only in the published version
of the song (as in plate 1), did Schubert (or the publisher, pre-
sumably with Schubert's approval) place the repeat sign before
bar 3.
This change brings Nahe des Geliebten into line with other
strophic songs that have a distinct introduction and postlude.

Example 6
 Schubert's "Nahe des Geliebten" 197

For example, in An Mignon, a Goethe setting composed on the

same day as (and on the verso of) Nahe des Geliebten, Schubert
omits the introduction after the first stanza; the postlude then
functions as prelude to subsequent stanzas. An apparent excep-
is Tranenregen, no. 10 of Die schone Mullerin,
tion to this practice
a modified strophic song which also has a separate prelude and
postlude. No autograph source for the song (or the cycle) sur-
vives, but in the first edition —
as in the autographs of Nahe
des Geliebten — there is no repeat sign at the end of the piano
prelude, which is therefore to be repeated before strophes 2
23
and 3. In Trdnenregen, An Mignon, and other similar songs,
however, the introductions are all self-standing harmonically
and melodically. Their presence or absence is thus not as cru-
cial a matter as in Nahe des Geliebten, where the "introduction"
surges directly into the first strophe.
Although we should probably take the printed version of
Nahe des Geliebten as representing Schubert's final thoughts on
the matter, we can still ask which version is preferable with —
or without the repetition of the introduction? On the one
hand, bars 1-2 seem too magnificent, and too closely bound
up with the architecture of the song, to be abandoned after
only one hearing. On the other hand, their successive repeti-
tion could never re-create the initial harmonic ambiguity or the
surprising emergence of the voice from the piano part.
That there is (or was) no fully satisfactory solution to
the introduction "problem" points up the special nature of
Schubert's Nahe des Geliebten. Reichardt experienced no diffi-
culty in shifting his music between texts; and Goethe happily
wrote his poem to fit a preexistent melody. But for Schubert,
re-creating Goethe's poem in musical terms was not as easy. In-
deed, it was something of a Procrustean effort, an attempt to
fit into the conventional folklike strophic structure, implied by

the poem's broader design and by the musical tradition, some

very unconventional music, inspired by the content of the
poem. To reiterate my initial point, then: the form and content
of Nahe des Geliebten engage in a dialectic. That the dialectic
remains unresolved is not a flaw; it is, rather, the reason why
Nahe des Geliebten stands as one of Schubert's most compelling
early songs.
 . ""

I98 SCHUBERT
Notes
1. Cited in Ernst Biicken, Das deutsche Lied (Hamburg: Hanseatische
Verlagsanstalt, 1939), p. 43. (Translation mine. In this article, all translations
from the German are my own unless otherwise noted.)
2. Cited in Max Friedlander, Das deutsche Lied im i8.Jahrhundert (1902;
repr. Hildesheim and New York: G. Olms, 1970), vol. 1, part 1, p. 196.
3. The best modern survey of the aesthetics of the Lied in the late eigh-
teenth and early nineteenth centuries (with an extensive bibliography of pri-
mary sources) is Heinrich W. Schwab, Sangbarkeit, Popularitdt, und Kunstlied
(Regensburg: Bosse, 1965). On strophic songs, see esp. pp. 51-54. See also
Walter Wiora, Das deutsche Lied (Wolfenbiittel und Zurich: Moseler, 1971),
pp. 22-44.
4. The poem can be found in Friedrich von Matthisson, Schrifien
(Zurich: Orell, 1825), vol. 1, pp. 191-92.
5 In line 5 of stanza 1 Schubert altered Matthisson's " Wenn" to " Wann.
Schubert's Andenken is discussed and compared with Beethoven's setting and
with his own Nahe des Geliebten in Thrasybulos Georgiades, Schubert: Musik
und Lyrik (Gottingen: Vandenhoeck und Ruprecht, 1967), pp. 40-55. I re-
turn to Georgiades 's ideas below.
6. Stanzas 1, 2, 3, and 5 of the poem are given in Friedlander, Das
deutsche Lied 2:459-60. The complete poem can be found in the original
printing, Musenalmanachjursjahriyg^, ed. J. H. Voss (Hamburg: C. E. Bohn,
1795), P- 177.
7. The song is printed in Friedlander, Das deutsche Lied 1/2:212.

8. Goethe's remarks to or about Zelter are discussed in Jack Stein, Poem

and Music in the German Lied from Gluck to Hugo ^//(Cambridge, Mass.:
Harvard University Press, 1971), pp. 41-42.
9. J. W. von Goethe, Goethe's Letters to Zelter, trans. A. D. Coleridge
(London: G. Bell and Sons, 1892), p. 1. The original German is cited in
Friedlander, Das deutsche Lied 2:201-2.
10. The 1809 setting is included in J. F. Reichardt, Goethes Lieder, Oden,
Balladen und Romanzen mit Musik, ed. Walter Salmen. Das Erbe Deutscher
Musik, vol. 58 (Munich: Henle, 1964), p. 22. The history of Goethe's poem
is given by Friedlander (see n. 9).

11. Wiora, Das deutsche Lied, pp. 35-37.

12. See his remarks cited in Friedlander, Das deutsche Lied 1/1:190, and
partially translated in Alfred Einstein, Schubert: A Musical Portrait (New York:
Oxford University Press, 195 1), p. 25.
13. See Friedlander, Das deutsche Lied 2:459. There were (and would
continue to be) many other settings of Nahe des Geliebten, which I do not
treat in this study. See the inventories by Willi Schuh, "Goethe Vertonungen,
in J. W. von Goethe, Gedenkausgabe der Werke, Briefe und Gesprdche (Zurich:
Artemis, 1948-52), vol. 2, p. 700; and by Friedlander, Das deutsche Lied
2:201. Zelter's setting of 1808, which bears no relationship to his Ich denke
dein, remained unpublished in his lifetime. It appears in C. F. Zelter, Funfzig
Lieder, ed. Ludwig Landshoff (Mainz: Schott, 1932), p. 32.
 Schubert's "Ndhe des Geliebten" 199

14. The included in both the old and new collected edi-
first draft is

tions of Schubert's works. The two drafts are reprinted from the old edition
in Schubert's Songs to Texts by Goethe (New York: Dover, 1979), pp. 47-48. In
the new edition, see Franz Schubert, Neue Ausgabe sdmtlicher Werke (Kassel:
Barenreiter, 1964-), series 4, vol. 1, p. 276. See also pp. 14-15 of the critical
report (Quellen und Lesarten).
15. See Georgiades, "Lyric as Musical Structure," included in this vol-
ume, p. 100.
16.Georgiades, Schubert, pp. 49-50.
17.Schubert changed Goethe's "Flimmer" (stanza 1, line 3) to "Schim-
mer," thus reproducing the word already used in line 1. This seems to me an
unfortunate (and perhaps unconscious) alteration, which obscures (though it
does not eliminate) the sun-moon opposition. In stanza 2, he similarly sub-
stituted "Wege," the last word of line 1, for the original "Stege" in line 3.
18. For a sensitive discussion of the musical and poetic aspects of
Schubert's song introductions, see Joseph Kerman, "A Romantic Detail in
Schubert's Schwanengesang," included in this volume.
19. In the first draft of the song (see n. 14) there is no quickening of the
harmonic rhythm; each chord of the introduction lasts three full beats (in \
meter). In this respect, as in several others, the final version is much more
effective.
20. For a discussion of the variants among these sources, see the criti-
calreport (see above, n. 14). Both autographs of the final version of the song
have been reproduced in facsimile. The first one (dated 27 February 181 5) is
in Paul Gottschalk, A
Collection of Original Ms. of the World's Greatest Compos-
ers (Berlin:Gottschalk, 1930), plate 21. The 18 16 copy is included in the fac-
simile edition of the entire volume sent to Goethe, Lieder von Goethe kom-
poniert von Franz Schubert, ed. Georg Schiinemann (Berlin: Albert Frisch,
1943), pp. 18-19 (of facsimile proper). The introduction relates the story be-
hind the preparation of this manuscript.
21. See Charles Rosen, Sonata Forms (New York: Norton, 1980), p. 183.
22. Hans Gal, Franz Schubert and the Essence of Melody (New York:
Crescendo, 1974), pp. 78-79.
23. The editors of the Neue Ausgabe take a different position on
Tranenregen, claiming that the publisher inadvertently out the repeat sign
left
before the last beat of bar 4 and that the error was not caught by Schubert's
brother, who proofread Die schone Mullerin. (See Schubert, Neue Ausgabe
4:2, p. 303.) There is perhaps some justice to their claim, since the song has a
modified strophic form. The final strophe is written out separately and is not
preceded by the original prelude. In the case of the Ndhe des Geliebten auto-
graphs, however, as in most simple strophic songs, Schubert wrote out only
one musical strophe, with the words of the first stanza underlaid. In the vol-
ume prepared for Goethe (see n. 20) he wrote out the poetry for stanzas 2-4
on the following page.
 The Schubert Lied:
Romantic Form and Romantic Consciousness

LAWRENCE KRAMER

I. The Schubert Lied

In Schubert's hands the German Lied became the first fully de-
veloped genre of Romanticism in music. As Charles Rosen
affirms, without mincing words, Schubert's songs "are related
to the Lieder of the past only by negation; they annihilate all
that precedes. The classical idea of dramatic opposition and
resolution is completely superseded: the dramatic movement is
simple and indivisible." The traditional way to explain why
1

the Lied, in particular, should form the breakthrough genre is

to observe that structural looseness and harmonic irregularity
can be persuasively justified as expressions of a text. Schubert
could support a Romantic style, as Schoenberg and Berg later
supported an atonal one, on fluctuations of feeling enforced by
2
poetry. Traditional descriptions of Schubert's songs accord-
ingly tend to focus on the new expressive resources of this mu-
sic: the richness of the piano accompaniment in contrast to the

anemic chords and arpeggios of earlier Lied composers like

Reichardt, Zumsteeg, and Zelter, or Schubert's gift for lush
3
sonorities and inventive modulations.
My purpose in this essay is not to declare these explana-
tions wrong, but to get beyond the limits imposed by their
static, merely descriptive rightness. As this comment sug-
gests, it should be possible to approach Schubert's Lieder in
terms of musical processes rather than of musical traits: to
place them in contexts of tension between tradition and inno-
vation, structure and texture, musical form and musical Ian-
 The Schubert Lied 201

guage. Only in that way can we find convincing motives for

Schubert's practice and open the possibility of a criticism that
will not substitute the admiration of his innovative textures
and harmonies for an understanding of how they work.
As a genre, the Lied as developed by Schubert seems to
rest on two imperatives. The first is to break away from the
harmonic contour of the Classical style, and especially to break
the grip of the dominant and its dominants — the cadential
—
circle of fifths on musical structure. As Rosen notes, Classi-
cal music was based on patterns of "dramatic opposition and
resolution." These were realized harmonically by taking tonic-
dominant tension as the foundation of both local and large-
scale form. Until fairly late in his career, Schubert could not
depart from this model convincingly within the privileged
framework of the Classical sonata. Before 1823-24, the years
of the A-Minor Piano Sonata (D. 784) and the A-Minor and
D-Minor Quartets, he filled out Classical forms with indi-
manner, at the cost
rectly related tonalities in a straightforward
of harmonic urgency. As he eventually realized, the use of post-
Classical harmony required a thorough recasting of Classical
structure — a task accomplished in works like the late B-Flat-
Major Piano Sonata, the C-Major String Quintet, and the last
symphony (in C). 4 But from the very beginning, Schubert was
essentially un-Classical. As Martin Chusid has remarked, he
seemed unable to tolerate long stretches in the tonic without
some intriguing digression. 5 At the start of his career, his har-
monic sense demanded a form that was burdened by neither
the prestige nor the rigor of the Classical genres. And this he
found in the Lied, which in the hands of its earlier practitioners
was an empirical, indeed somewhat haphazard affair, loosely
positioned between invocation of the Volksgeist and dramatic
recitative.
The other imperative of the Schubert Lied is to align mu-
sicwith the widespread effort of literary and philosophical Ro-
manticism to represent subjectivity in action. This expressly
means that the purpose of Romantic song is not simply to en-
hance the emotional force of the text, nor even, as writers on
the Lied traditionally claim, to evoke the meaning of the text,
whether directly or ironically. The purpose is to represent the
activity of a unique subject, conscious, self-conscious, and un-
202 SCHUBERT
conscious, whose experience takes shape as a series of conflicts
and reconciliations between inner and outer reality. According
to Wordsworth, this "mighty Mind" is the subject, in all senses
of the term, of Romantic poetry. "Ever on the watch, /Willing
to work and to be wrought upon" (The Prelude [1805], 13.
102-3), such a mind "feeds upon infinity," "exalted by an
underpresence,/The sense of God, or whatsoe'er is dim/ And
vast in its own being" (13. 70-73). Schubert's songs literally
give a voice to this historical expansion in the concept of the
self. Perhaps that says something about song as a form: after

all, the development of expressive monody in the Renaissance

did the same thing, though the model of self involved was very
different.
We start with a voice for the self, then. But whose voice?
Edward T. Cone, in considering Schubert's Erlkonig, calls it the
composer's voice: a presiding personality or "persona" that ex-
presses itself through the music, "speaking" for the composer
in the same indirect and partial way that the "I" of a lyric poem
6
speaks for the poet. Cone also makes the important point that
this musical persona stands apart from both the vocal line of
a song and the accompaniment: we hear it in the song as a
whole. Beyond this, however, the persona of the song stands
apart from the persona of the text —
a difference that Schubert
expands into a generic distinction.
Like any song, a Schubert Lied evokes the persona of its
text, which comes to life in the broad expressive character of
vocal melody and accompaniment figuration. But Schubert's
Lieder also negate that textual persona in favor of its musical
counterpart. The musical persona affirms itself by recasting
the rhetoric, rhythm, and imagery of the text, reinvesting the
textual material with a new subjectivity. It is as if the music
were exercising the "magical power" of imagination described
by Samuel Taylor Coleridge, which "dissolves, diffuses, dissi-
7
pates, in order to re-create." The movement of the song corre-
sponds to this re-creative process, which can range in its aims
all the way from the musical persona's identification with, to its
8
repudiation of, a textual alter ego. Romantic poetry in both
German and English is full of similar encounters between selves
and personified self-images.
 The Schubert Lied 203

Schubert's approach to song through the Romantic subject

is closely allied with his treatment of a larger musical issue:
harmony. Schubert's music recurrently incorporates a conflict
between Classical tonality and harmonic innovation, a conflict
found in its most direct and concentrated forms among the
songs. In its progressions, key relationships, emphasis, and
chordal texture, Schubert's harmony repeatedly steps beyond
Classical boundaries. Yet the music invokes the Classical con-
text of these Romantic harmonies so strongly that the Classical
style retains a substantial lingering authority. Harmony thus
becomes a matter of clashing perspectives. Harmonic structure
becomes dialectical rather than systematic, as Schubert's Ro-
mantic idiom evolves by negating, but never escaping, its
Classical origins.
For Classical tonality, this dialectic entails a demystifica-
tion at the hands of the musical persona. Schubert's mingling
of harmonic idioms undermines the naturalness that Classical
harmony seems to possess on its own ground, and exposes that
harmony as a historically determined language that may at-
tract, but cannot define or delimit, a Romantic subject. Both
Mozart and Beethoven had earlier taken steps in this critical or
9
"deconstructive" direction. Yet if the dissonant introduction
to Mozart's C-Major Quartet (K. 465) stretches Classical har-
mony to the limit, it still stays within that limit; and if Beetho-
—
ven goes beyond the limit one-upping Mozart, say, in the
—
C-Major Quartet, op. 59, no. 3 he does so only to recommit
himself to the Classical style more decisively. 10 Schubert is the
first composer to set a Romantic harmony into play without
assuming in advance its subordination to a higher Classical
power.
Many of Schubert's songs seem to regard the affirmation
of the musical persona and the dialectic of Romantic and Clas-
sical harmony as concurrent versions of the same general pro-
cess —that is, of the representation of subjectivity in action.
Sometimes, especially in strophic songs, Schubert forms blocks
—
of contrasting harmonies and juxtaposes them either to dis-
place the structural effect of Classical fifth relationships with
the coloristic effect of third relationships, or to spell out an
antithesis between Classical and Romantic idioms. So Der
 —

204 SCHUBERT
Musensohn proceeds by the impetuous juxtaposition of the
tonic and the mediant major. The keys alternate without mod-
ulation as the vocal line moves from an upbeat in one to a
downbeat in the other; Classical movement by fifths occurs
locally,but not structurally. At the same time, Schubert writes
recurrent hesitations into the lightly skipping \ melodic line, so
that his musical persona admits to a slight but telling plain-
tiveness that Goethe's son of the Muses, who rapturously cele-
brates transience, tries to leave out.
More radically disjunctive is the collagelike design of Am
Meer, from Schwanengesang. The accompaniment of this song
alternates between doubling the voice over soft, gently rocking
tonic-dominant harmony, and splitting off from the voice in
dramatic tremolo swells. As the swells rise and fall, a plain
circle-of-fifths descent in the tonic C major, Wii-ii-V^-I, is
so distorted by dissonant intrusions that its cadential impact all
but vaporizes (ex. i; note especially the alteration of ii at a, and
the unusual treatment of the dissonant F of the V5/I, which
disappears without resolving, then reappears as an appog-
giatura to the tonic triad).
This clash of harmonic perspectives echoes the more radi-
cal one produced by the famous elliptical progression, German
sixth to tonic, that both opens and closes the song. Joseph Ker-
man describes the frame passage as an "oracular" suggestion of
"
"everything in the world that is inward, sentient, and arcane."
One thing that makes it so is a dim feeling that the German sixth
has not really been resolved at all, that the elided sonority
the dominant —
is obscurely missed. Schubert transforms the

textbook movement from German sixth to tonic J by doubling

the C of the German sixth as a pedal tone, which turns the sub-
sequent tonic chord into a root-position triad. But since the
root in the bass is not reattacked, the upper tones act together
as a kind of residual or pseudo J chord, more a point of transi-
tion than one of rest. The pedal tone puts a familiar progres-
sion into a strange, even an uncanny, light. By framing the
song with this subtle mystification, Schubert both extends and
parodies, salutes and betrays, the circling movement at the heart
of Classical tonality.
With the through-composed songs, Schubert often works
by splitting the composition into disjunctive halves that are
 The Schubert Lied 205

Example 1

Wit - der: out <ki - mnAu- am lit - be-v<A

tenuously integrated at the close by a few perfunctory mea-

sures of tonal circling. Einsamkeit, from Winterreise, begins
pianissimo with a bleak, detached texture; the harmony is con-
fined to the tonic, dominant, and submediant areas of B
minor. In the second half, there is a jagged and sonorous dis-
play of shifting dynamics, framed by swells on tremolos and
heavy chordal triplets. Twice the music drives to a plateau on
the flatted supertonic, C, through a jangle of diminished chords
and fragmentary circle-of-fifths progressions. The tonic and
dominant are submerged, all but obliterated, as the submediant
G-maj or. sonority that was naively prominent in the first half
suggestively redefines itself as the dominant of C major/minor,
the G-C progression is less an event within the orbit of B
minor than it is the articulation of a competing tonic-dominant
axis. Schubert underlines the point by refusing to represent a
single C-major chord as a Neapolitan harmony in B minor.
Overall, the large-scale resolution of one tonic-dominant
polarity into another can be said here to replace the Classical
206 SCHUBERT
resolution of dominant to tonic. In particular, the setting of
the final phrase, "so elend nicht," turns on the reinterpretation
of G 7 as a German sixth of B minor. But the resolution itself is
dubious. The closing return to the tonic, only four measures
long, depends on a rhythmically feeble cadence and quickly
peters out on the hollow sound of open fifths and sixths.
Aufdem Flusse, also from Winterreise, is similarly if more
subtly designed. In the first half, the major harmonic event is
the first modulation, a startling move from the tonic to the
leading tone, E minor to D-sharp minor, after only eight mea-
sures; Schubert seems to seize on Djl, the third of the dominant
major, as a kind of subversive secondary dominant. The sec-
ond half of the song is antithetical to the first both in texture
and harmonic design. The vocal melody that begins the work
now disappears from the voice and migrates to the bass as an
ostinato. As a concomitant, the Djl dissonance, often in op-
position to Dti, is presented on a wide variety of harmonic and
melodic planes —
as a tonic, a dominant, a mediant, a false rela-
tion, a leading tone —
so that the closing return to the tonic co-
incides with the exhaustion of the imperious, heightened dis-
cord. This process becomes explicit in the closing phrase of the
voice, which is nothing more than the melodic elaboration of
Djl — E resolution. In its last incarnation, Djl appears as part of a
dominant-seventh chord, but the cadential resolution that fol-
lows is less significant in itself than as a medium for the long-
term resolution of the polarized leading tone. 12
Of course, not all of Schubert's more than six hundred
songs are as disjunctive as these. Am Meer and Einsamkeit in
particular represent only an expressive limit within the broad
generic contour of the Schubert Lied. But a simple diatonic
song like Heidenroslein does the same thing; its Classical quali-
ties represent a limit, not a norm. The diatonic naivete of the
song is sophisticated, not innocent; it belongs to a musical per-
sona who, like Goethe's narrator, declines to make anything of
the irony implicit in his little fable.

II. Presentational Structures

In order to makeRomantic form and Romantic conscious-

the
ness of the Schubert Lied available to criticism, two working
 The Schubert Lied 207

principles will be necessary.The first is that the relationship of

the text to the music must always involve a critical interpreta-
tion of the poetry; merely to recount what the poem "says" is
13
not enough. The persona of a Romantic song is rarely content
just to echo a text*. On the contrary, it actively appropriates the
text, which is to say that it and substantially re-
both identifies
vises the poetic persona's pattern of consciousness. The famil-
iar techniques by which the music mimics, emphasizes, or al-
ludes to the text cannot automatically be taken as parallels;
they must be thought through as parts of the musical persona's
subjective drama.
The second principle is that Schubert's songs are likely to
complement their movement around a central tonality in —
—
some cases even to replace it with harmonic structures of a
different kind. I have suggested elsewhere that the Romantic
style characteristically presents such independent harmonic
structures in tonal contexts of varying strengths and promi-
14
nence. Edmund Husserl's phenomenological concept of "hori-
zon," the tacitly apprehended context of lived experience, is
helpful on this point. Classical music can be said to present
harmonic patterns, both local and structural, that exemplify
Classical tonality itself —
the system of tonal relationships and
musical forms that constitutes the horizon of the harmonies
presented. Romantic music, in contrast, depends on the ever-
present possibility of presenting harmonic patterns that sound
within a tonal, even a Classical, horizon, but do not exemplify
15
it. In Schenkerian terms, this could be put as the quite un-
Schenkerian proposition that Romantic music customarily es-
tablishes foreground or middleground structures that are de-
tached from the composing-out of a background.
Die Stadt, from Schwanengesang, offers a skeletal, almost
schematic example of Romantic presentation in Schubert. The
song follows an a b a' b' a design. In the a sections, the piano
creates an impressionistic haze by obsessively arpeggiating
a functionless diminished-seventh chord. A nominally tonic
pedal point pulses along underneath, but its effect is to cloud
the sonority, not to clarify the harmony. The b sections answer
this ambiguity with rigidity: they confine themselves to rudi-
mentary tonic, dominant, and subdominant harmonies, as if
to intimate that anything more would risk a tonal collapse.
208 SCHUBERT
This antiphony of harmonic idioms becomes a dramatic con-
flict when the voice, which enters the song with the first b sec-

tion, is accompanied —
overtaken —
by the shivery arpeggio
during the second a section. For two measures the vocal line
retains its original orientation to the tonic, C minor. But with
the expressive entry of an h\ that embitters the phrase "an
graue Wasserbahn," the vocal line commits itself to an ex-
tended falling articulation (C-Alj-Fjl-B'-C) of the ambigu-
ous diminished-seventh chord that ripples beneath it.
The only significant harmonic event in Die Stadt arises
unexpectedly as the vocal line draws to a close. Schubert re-
6
capitulates an earlier progression, ii -i*-V, by altering the su-
pertonic chord to a Neapolitan sixth and adding a subdomi-
nant: ^II -iv -i4-V. A small change
6 6
—
but it suddenly expands
the harmonic frame of reference and proclaims, though be-
latedly, that Classical tonality is not dead, only sleeping. Yet
this, the point of greatest tonal integration, is also the point of
emotional disintegration. The Neapolitan progression marks
the moment in the poem when the rising sun illuminates the
spot where the speaker has lost his beloved. The poem invites
us to recognize from this that the speaker has been making a
kind of pilgrimage to that very spot, where the visual illumi-
nation that greets him ironically takes on a psychological paral-
lel. What the light of day reveals is that his homecoming is an

exercise in self-torment, a neurotic reenactment of his failure

in love. By ending the poem with the revelation of his loss, the
speaker imbues his ritual of return with pathos and nostalgia:
the force of his obsessiveness is blunted by the conventional
posture of the disappointed lover. Schubert echoes this effect
with the traditional pathos of his Neapolitan sixth, but he will
not let the pathos stand. The song closes with the solo piano
once more tossing up its haze of arpeggios, a move that re-
verses the speaker's effort to expose an obsession and explain it
away in the same breath. Instead of being glossed over, the feel-
ing of psychological disturbance intensifies as the song ends
in vocal silence amid the tonal uncertainty of its gray water-
music.
A much earlier song, Geistes-Gruss (D. 142, 181 5- 16), is a
match for Die Stadt in its dissociative form. Geistes-Gruss be-
gins with soft tremolos in E major, moves to the dominant, B,
 The Schubert Lied 209

then turns without modulation to the lowered mediant, G,

where it adopts a declamatory texture and moves to a close.
The skewed key design is an extravagant gesture, not a radical
one, but it has an expressive significance that is lacking in the
tonal ramblings of Schubert's longer, balladlike early songs.
The G-major section begins with a reiterated I-V5 progression
that is sequentially restated on vi. The sequence represents a
moment of self-conscious reflection on the harmonic structure
of the song. The chord on the sixth degree of G major is
E minor, so that the sequence forms a diatonic version of the
chromatic third relationship that shapes the piece as a whole.
The E-minor triad and the E-major tonality even behave in the
same way, each moving to its own sixth degree. Finally, the
harmonies involved in the initial G-major sequence reappear
near the close of the song (at "und du, und du") in retrograde
order. The result is a resolution in G major via vi-V — I.
7

The song closes by echoing its final cadence over a tonic

pedal. By thickening the sonority and sinking quickly from
forte to pianissimo, the little epilogue mutes the feeling of ar-
rival without disturbing the harmony. This both gives the
spirit-greeting celebrated by the text, an injunction to fare ever
onward, an appropriately open-ended feeling, and turns the
greeting into an oracular justification for the split tonality of
the song. Once again, the music becomes an allegorical reflec-
tion on its own design.
Alook at the multiple versions of Geistes-Gruss, four of
which appear in the old Gesamtausgabe, suggests that Schubert
hesitated over just how disjunctive the song should be.
16
The
idea of bluntly juxtaposing chromatically related key areas ap-
pears in the first version, but not in the second and third,
where Schubert smoothes the transition to the closing key by
moving through its dominant-seventh chord. In the final ver-
sion, Schubert returned to his original idea and added the
tremolo' figuration. As the song now stands, the spirit-voice
appears to enter through the breach in tonal nature represented
by the abrupt appearance of G major. The shift in harmony be-
comes an act of conjuration.

Four full-scale critical interpretations will take up the rest

of this essay. They are intended to give a feeling for the variety,
210 SCHUBERT
complexity, and consistency of Schubert's presentational struc-
tures, and so to give a sketch —
only that, by necessity, but one
—
ample in particulars of the Schubert Lied as a Romantic genre.
The songs in question, Meeres Stille, Pause, Der Doppelganger,
and Ganymed, span Schubert's career, and all represent his song
writing at its best and most inventive.

III. Meeres Stille (D. 216, 1 815)

Marked Sehr metronome mark of J =

langsam, angstlich, with a
27, Meeres Stille music that barely moves. Schubert's tone
is

painting of Goethe's calm sea involves immobility in every ex-

pressive dimension. Both the vocal line and the accompani-
ment are depressed in register; the highest note in the song is
c'. There is no change in dynamic level and, in the accompani-

ment, no change in rhythm; the voice suspends itself over dull

arpeggios, one to a measure, in steady pianissimo. Even the
words fail to move in the usual sense; the singer's long notes
make taffy of the text at the tempo indicated. Only the har-
mony is not but while the arpeggios vary in tonal func-
static,
tion their sonority monotonous, an almost unbroken succes-
is

sion of close-position chords and octave basses.

Meeres Stille is very nearly a continuous fabric of sound,
but Schubert gives it a tripartite design by means of a short rest
and a fermata. The work is constructed symmetrically. The
first section consists of a modulation from the tonic C major to
the mediant major, E. The middle section slips into the sub-
mediant, A
minor, cadences on F, then initiates a return to the
mediant major. The section closes on V of III, and the third
section reverses the harmonic direction of the first by moving
from the mediant back to the tonic, closing on a full cadence.
The harmonic design, then, is based on a recursive principle of
enclosure: C major surrounds E major, which in turn sur-
rounds A minor. The rigidity of the pattern, with its concep-
tual image of envelopment, is quite possibly a piece of Bach-
like musical symbolism on Schubert's part, a schematic portrait
of the ship that the poem imagines enveloped by the "mon-
strous breadth" ("ungeheuren Weite") of ocean.
Considered abstractly, these harmonies are straightforward
 V

The Schubert Lied 211

Classical progressions. Heard concretely, they are tenuous and

self-subverting. In the outer sections, the tonic is slackly pro-
filed and the mediant major is surprisingly feeble. As the open-
ing C-major chords are arpeggiated, their spacing divides them
into a pair of distinct perceptual units, a tonic octave followed
by a pseudo J chord. The result is a tenuous quality that dis-
appears if the chords are played as simultaneities. More im-
portant, the mediant major starts off strongly and then dis-
integrates. The modulation that introduces it at the end of the
first section (bars 6-8) includes a substantial cadence in E/e:
6
ii° —
7
— I. The diminished ii chord points to E minor; E major
emerges instead. It is rather bemusing to hear this new key dis-
solve instantly, but so it does; the middle section begins by dis-
pelling it with a diminished-seventh chord from which A minor
tentatively emerges.
Later on, the peculiar instability of the mediant major
comes to make sense. The middle section draws to a close
by initiating a new mediant modulation, which is held up by a
fermata on V/III. The third section begins by completing the
full cadence, but with E minor replacing E major. The outcome
of this cadence both recalls and intensifies the major/ minor
clash implicit in the first section, and Schubert at once brings
the conflict to the surface. He repeats the original mediant
modulation exactly, then immediately corrects III to hi. In
other words, the chromatically tonicized key of E major is
twice reduced, demoted, to a diatonic step within C major.
This permits the tonic to reemerge without modulation, as if it
had never been left at all. The cadential descent to the mediant
major is never answered; it is elided, left as a kind of concealed
fissure in a surface that is seemingly all tonic.
Schubert's implicit and explicit juxtapositions of the
major and minor triads of E call attention to an ambiguity that
is ever-present in Classical tonality, the uncertain distinction
between a scale degree and a key. The sense of ambiguity turns
the song, like Geistes-Gruss, into a projected form of self-
consciousness, a meditation on the slipperiness of harmonic
meaning. The odd tenuousness of the seemingly firm mediant
major bears out the difficulty, and helps to disturb the tonal
clarity of the work as a whole.
212 SCHUBERT
Example 2

The middle section goes even further. Putatively in A

minor with a hintof F major, it develops a Romantic presenta-
tion that reorients the song around an independent harmonic
process that is antithetical to tonic resolution. The episode
sounds indefinite and confused, and it should; its role is to
create a dissonance of harmonic perspectives within a fuzzy,
generalized tonic horizon.
Here again, an ambiguity inherent in Classical tonality
isheightened and drawn into a pattern that comments on its
own uncertainties. Once introduced, the submediant A-minor
sonority shows an exclusive affinity for its own submediant, F,
which gradually establishes itself as a new key area. At bar 17,
the addition of a dissonant tone to the F-major triad produces a
chord that is spelled as a German sixth of A and progresses to
an A-minor \ chord. The progression forms a partial reso-
lution that would normally be completed by a move to the
dominant of A minor. Instead, the augmented sixth reappears,
abandons all thought of resolving, and leads to still another
dissonance of the same type (ex. 2). This new chord is com-
pletely unanchored: it could be either a dominant seventh of G,
which is how it is spelled, or an augmented sixth of F-sharp.
For a moment, all tonal bearings are lost, and in that lies the
crux of the song. As V /V of C, the dissonant chord seems to
7

gravitate toward the tonic. As an augmented sixth, it points to

the axial tonality of the tritone. So, in one sonority, maximum
 The Schubert Lied 213

stability and maximumtension float suspended. As it happens,

Schubert takes the chord as a function of F-sharp, which makes
the asyntactic arbitrariness of its arrival a point of collision be-
tween harmonic idioms.
Two independent presentational patterns are exposed by
this. First, a look at bars 17-21 shows that the F-sharp-minor
triad isproduced by a continuous semitonal modification of
the initial augmented sixth of A, the semitones falling in the
bass and rising in the upper voice (ex. 2). The process subordi-
nates and skews tonal syntax, as its strangest product shows;
the problematical chord of bar 20 is a kind of enlarged pass-
ing dissonance, but one with a tonal identity too strong to be
submerged.
—
Behind this pattern essentially one in which voice lead-
ing bluntly overrides harmonic coherence — there is a larger
one. As it turns out, the presence of the F-sharp-minor chord
clarifies every harmonic event in the song. The mediant modu-
lations that frame the middle section begin with a diminished
triad, the pivotal Fjl-A-C, because the long-range reference
point of the progressions is not the mediant (and indirectly the
tonic) but the resolved form of that triad, Ftf-A-Q. Likewise,
the mediant major evaporates into the submediant, and the
submediant moves only toward developing its augmented sixth,
because the submediant augmented-sixth chord is needed to
provide the point of origin for the semitonal "osmosis" that
generates the F-sharp-minor sonority. The Ftf can even be heard
as a destructive "alter ego" that negates the harmonic authority
of the F-major triad that comes to prominence in the middle
section, whether as the subdominant of the song's tonic or as
the submediant of A minor.
17

In sum, all the lines of presentation are oriented toward

the tritone rather than toward the tonic. And though the final
mediant ^modulation tries to cover the breach thus opened by
reclaiming the F-sharp triad as a predominant chord, no return
from the irrational yet overdetermined central tritone can be
fully convincing. It is no accident that the word set to the
F-sharp-minor triad is Todes. Spelling out the ultimate suspen-
sion of vitality, the ultimate enclosure, the word marks both
the point of maximum estrangement in the poem and the pivot
for Schubert's subjective appropriation.
214 SCHUBERT
Goethe's "Meeres Stille" evokes a feeling of oppressive
stillness by repeatedly making the same observation from dif-
ferent perspectives. The language of the poem, like its be-
calmed sailor, is held fast to the empty sea. Of the two stanzas,
the first repeats itself and the second makes matters worse by
repeating itself and the first stanza, too. Even the grammatical
constructions are parallel:

Tiefe Stille im Wasser,

herrscht
Ohne Regung ruht das Meer,
Und bekummert sieht der Schiffer
Glatte Flache rings umher.
Keine Luft von keiner Seite!
Toddesstille furchterlich!
In der ungeheuren Weite
Reget keine Welle sich.
[Deep stillness rules the waters, the sea rests without
motion; and with anxiety the sailor sees a smooth surface
surrounding him.
No breeze from any side! Terrible death-stillness! In the
monstrous/enormous waste not one wave stirs.]

Between the two stanzas, however, there is a significant

shift in narrative style. The speaker of the is de-
first stanza
tached, impersonal, and uncharacterized —
more a function
than a persona. The second stanza carries out a repeal of this
neutrality. Description turns to exclamation; the first two lines
of the second stanza sympathetically assume the perspective of
the anxious sailor. And this movement from detachment to
sympathy, from observation to identification, invests the op-
pressive seascape with a new meaning. Seen from within the
sailor /speaker's anxiety, the scene becomes a metaphor. The
stillness ("Stille") of the first stanza reappears as a death-
stillness ("Todesstille"), while the connotation of monstrosity
in "ungeheuren Weite" suggests a rift in nature from which
neither the sea nor the sailor can emerge.
Schubert's song deletes the moral and imaginative dimen-
sion of sympathetic identification from the poem. The sec-
tional division of the song pointedly ignores the stanzaic divi-
sion of the text, the unvarying accompaniment reduces all
difference to sameness, and the vocal line does nothing to dis-
 The Schubert Lied 215

tinguish between the abstract voice of the first stanza and the
humanized voice of the second. The persona behind this music
is a victim of his own anxiety from the start. Unable to make

room for others, his subjectivity isolates him and fixates him
on what is literally adead end. All bekummert, the song is also
all metaphor: it forms a leaden epiphany of death that consum-

mates with the F-sharp-minor triad and pronounces its vision

to be unendurable with the succeeding three chords, which
come to rest in unresolved dissonance. When the voice closes
the bleak annunciatory phrase "Todesstille furchterlich," it falls
to its registral nadir, a low B, the dark sound of which marks a
18
paralysis motivated by unconditional mortal terror.

IV. Pause (D. 795, no. 12; 1823)

Like Einsamkeit, Pause from Die schone Mullerin is divided into

antithetical blocs of harmony. The song follows an a b a' c pat-
tern with a short coda; the harmonic plan can quickly be sum-
marized. The a section keeps to the tonic, B-flat major. The b
section begins in the submediant, G minor, includes a strong
cadential progression to C major/ minor, proceeds through
tonic-dominant harmony into the flat mediant, D-flat major,
and ends with a half cadence on the dominant. The a' section
recapitulates the harmony of a and the first part of/?. It begins
in the tonic, then moves quickly through G minor to a pause in
C minor. The c section begins in A-flat major, from which it
twice moves chromatically onto Ft, then to the tonic.
This bare description alone brings to light a Romantic
presentation based on tritone relationships. The submediant
passage at the beginning of the b section lies a diminished fifth
away from the flat mediant passage to which it leads; and the c
section uses a flatted dominant, B>, as a harsh and perplexing
avenue tp the tonic. Even if the song were worked out with
impeccable Classical syntax, this design would put it under a
strain. But the tonal syntax is a dialectical element here. Im-
peccable enough, even ingenious, through the close of the a'
section, it becomes eccentric —not to say bizarre — in the c
section.
The turning point comes as the a' section closes. After the
cadence to G minor, C minor is approached abruptly via an
2l6 SCHUBERT
A-flat-major triad whose incongruous brightness seems to
mock the word "bange" that it accompanies. This is not a mere
casual detail, but a kind of tear in the Classical fabric of the
music. For no apparent reason, the c section seizes on A-flat as
a tonality and uses it to reprise the main motive of the song,
which comes out sounding strangely remote and hollow.
Nothing is done to tonicize this intrusive chromatic re-
gion, and nothing to rationalize its materialization out of a
stray wisp of dissonance. Nothing, in fact, is done to it at all.
The A-flat-major triad remains an entirely static tonal level for
seven measures, then turns fleetingly to the minor and disap-
pears, along with the melody, into a sustained triad of F-flat.
A painful, mystifying dissonance then follows, the upper tone
of which is resolved as an appoggiatura to produce a lingering
O5 chord (ex. 3). This new dissonance is overtly irrational,
like everything in its surroundings, but its ambiguity is "decid-
able." By respacing the chord, Schubert forms an augmented-
sixth chord of B-flat, the song's tonic. The tonic obligingly re-
turns, but it has barely been established before the music lapses
back into the tritonal area. The A-flat episode is reprised, but
condensed into a mere three measures; A-flat is represented
only by its minor before the F-flat swallows it up. When B-flat
major returns again (via the triple dissonance) it peters out
quickly, rattling off the much-repeated main motive in a weakly
closural coda.
Like the F-sharp-minor tritonal chord in Meeres Stille, this
remarkable passage forces Classical harmony to become in-
coherent by insisting on a counter-coherence all its own. It is
easy enough to move up the circle of fifths from A-flat to

Example 3

4 m PP?
1st esdertfack-klawi
m twv
matter Lie - iyes -
r,pp |f
vein? Sovl a das Vor
5
 The Schubert Lied 217

B-flat. If Schubert's route is tortuous, even fantastic, the reason

lies with a Romantic presentation that strips the tonic of its

centralizing power. In particular, A-flat major intrudes on the

song after a cadence to C
minor; the A-flat harmony therefore
represents a move to the local submediant area. Schubert then
applies this movement recursively: the same distance is covered
in the chromatic movement from A-flat to F-flat. And to go a
step further, the dissonant tone that forms the appoggiatura to
the pivotal f chord is C, so that the movement away from F-flat
also involves a chromatic submediant relationship, here re-
—
duced to a melodic dissonance one shared by the voice and
piano.
The overall structure of the c section is determined by this
preference for the lowered-submediant sonority. The chain of
descending thirds, C- A-flat -F-flat, forms an independent,
harmonically indefinite pattern. Still more important, the chro-
matic submediant relation A-flat- F-flat forms an antithetical
exaggeration of the diatonic one, B-flat-G, that is prominent
earlier. The condensed repetition of the A-flat passage focuses
the antithesis: the movement from V to I to vi in B-flat just
7

before the last entry of the voice is immediatelyjuxtaposed with

a movement from 6
V
to i to VI 6 in A-flat. And the fact that the
tonic is ultimately regained through a chord of its augmented
sixth assimilates it, too, to the grid of minor-sixth sonorities
that governs the c section. In this, as in every other respect, the
end of the song parodies and deconstructs its beginning.
Pause marks the turning point in the drama of Die schone
Mullerin, and the disjunction between the Classical and Ro-
mantic materials of the song draws the boundary line between
the miller's illusion of happiness and his eventual misery and
death. Since falling in love, the poetic speaker has been too
joyous to sing, but the wind sometimes sighs through his ne-
glected lute, and the sound disturbs him:

Oft um die Saiten mit seufzendem Klang,

fliegt's
1st esder Nachklang meiner Liebespein?
Soil es das Vorspiel neuer Lieder sein?

[Sometimes a sighing flutters at the strings. Is it the

echo of my love-torment? Will it be the prelude to new
songs?]
2l8 SCHUBERT
Theclosing question is twice set as a hushed antiphony,
the half suspended over the tritone, the second moving
first
into the tonic. The distance between the harmonies forms the
measure of self-estrangement. Unwilling to answer, the musi-
cal persona dwells on the dread that disturbs him in the form of
an external sighing, the outward projection of a painful antici-
pation that he cannot suppress but wishes to disavow.
The texture of the whole final section, with its paradoxical
combination of harmonic stasis and abrupt shifts of sonority,
reinforces this suggestion of an anxiety that the speaker is
unwilling to confront. The hopeful but unconvincing resolu-
tion for the final line of text is imbued with a feeling of self-
deception, the nervous expectancy of a divided self in whom
the wish to disentwine Romantic passion from suffering is
contradicted by the sound of the word "Liebespein" set to the
most painful dissonances in the song (see ex. 3).

V. Der Doppelganger (D. 957, no. 13; 1828)

Der Doppelganger belongs in a unique expressive category with

the great concluding song of Winterreise, Der Leiermann. Both
pieces find the extremity of despair in a confrontation between
the halves of a divided self: the subject who desires, and a
double who represents that subject's worst aspects — anxiety,
self-torment, self-contempt. Both songs follow what must be
called a minimalist aesthetic; they rest on a bare scrap of mel-
ody, a brief, dully repeated accompaniment figure. In the case
of Der Doppelganger, this minimalism forms part of Schubert's
response to Heine. It is shared by at least two of the other
Heine Lieder, Die Stadt and Am Meer, and it is reflected in both
the bare octaves that form part of the accompaniment in Ihr
Bild and in the texture of octaves and tremolos in the outer sec-
tions of Der Atlas. The sense of the text embodied by this sty-
listic choice is harsh. Heine's early lyrics, with their combi-
nation of willed naivete, extravagant sentiment, and jeering
irony, suggest a persona in whom the self has thinned out to a
kind of linguistic diagram, a raggle-taggle bunch of cliches
strung together by a self-consciousness that can neither accept
nor exclude them. Schubert's minimalism poignantly exposes
this feature of the poetry, and at the same time tacitly resists it.
 The Schubert Lied 219

By constructing his minimal forms from simple cadential ma-

Schubert suggests a last-ditch, antithetical force of co-
terial,
herence working against the dissociation set loose in Heine's
poems.
Aside from this, the spare repetitiveness that Schubert
associates with the spectral double has a generic basis. The
speakers in the Heine poems that Schubert set in 1828 are vic-
tims of compulsive repetition who return endlessly to the
scene of their worst loss. Similar compulsives, haunted by sex-
uality or guilt, are frequent in the music and literature of the
Romantic period. A short list would include Coleridge's An-
cient Mariner, Berlioz's persona in the Syrnphonie fantastique,
Byron's and Schumann's Manfred, Wagner's Flying Dutchman,
and Nathanael, the hero of E. T. A. Hoffmann's "The Sand-
man." 19 If Maurice J. E. Brown was right to suggest that the
appropriate sequence for the Heine songs is not the post-
humously published one, but the order in which the poems
appear in Heine's Die Heimkehr, then the second half of
Schwanengesang forms a coherent little song cycle that portrays
the tragic lapse of the self into obsessive repetition as the result
20
of sexual grief. Schubert's concentrated, insistent song struc-
tures accept this repetition compulsion, as Freud called it, as
their imaginative milieu. In this context, a double becomes the
reductio ad absurdum of self-consuming repetition, and Schubert's
musical texture in both Der Doppelgdnger and Der Leiermann
goes to the same limit.
In general, the impulse of Schubert's Heine songs is to
bring musical structure almost to the point of disintegration.
Die Stadt, which ends with a solitary C that stands for the
tonic C minor but is not even securely tonal, is the extreme
case; Der Doppelgdnger is formally less desiccated, but far more
insidious in its irony. The basic process in Der Doppelgdnger is
the steady erosion of Classical harmony as a source of musical
meaning, the gradual exchange of a Classical for a Romantic
presentation. Like Pause, the song unfolds by deconstructing
itself, but continuously rather than abruptly. Two overlapping

perspectives are involved in this, one centered on the depletion

of the tonic, the other on a breakdown of harmonic syntax.
The essential element in each case is the accompaniment,
which consists entirely of variations on a four-chord progres-
 V

220 SCHUBERT
sion, "tolling," as Richard Capell long ago observed, "in the
21
half-conscious fancy of the speaker." The content of the pro-
gression is i-
6
—
-III-Vj (or Vj) in B minor an extended half
cadence. As example 4 shows, the texture of the chords is curi-
ous. All of the tonic and dominant triads are incomplete; the
tonic and closing dominant, missing their thirds, sound espe-
cially barren in empty fourths and fifths. Only the dissonances,
the dominant-seventh chords and some of the mediant triads,
are filled out. Eventually, a few scattered B-minor triads do ap-
pear in complete form, but with steadily failing power to direct
the harmony; the dominant triad remains incomplete except
when the seventh is added.
Schubert's stilted chording is enough by itself to give
B minor a spooky, desiccated feeling, but there is something
more. The song leaves the tonic for only five measures, just
before the close; the move, a wrenching nonmodulatory shift
to the raised mediant minor, D-sharp, coincides with the point
of crisis in the poem, where the speaker accuses his double of
aping his love-torment (Liebesleid). As example 5 shows, the
spell of D-sharp minor consists only of a pair of half cadences,
i-V-i-V, a varied transposition of the obsessive B-minor
progression. The apparent continuity actually marks a disin-
tegrative gesture. As a concentrated and repeated half cadence,
the D-sharp-minor progression is stronger and more focused
than its B-minor counterpart. Moreover, all of the D-sharp-
—
minor chords are complete, richly sonorous triads a dis-
sonant fulfillment, by chromatic heightening, of the full medi-
ant triads that have been heard (with increasing emphasis) since
the beginning. Add to this the tone weight of the longest

Example 4

IS m
SP
1
 The Schubert Lied 221

Example 3

f^ ^V was
\ i
itffst
i

dunachmein
^'pJ
iie-besleid,
m^Mjj^ p
das ynkk^ua^tcu^dieser Ski- it

g
ff

m m #'
n *
p
stretch of fortissimo in the song, and it becomes clear that the
raised mediant in Der Doppelganger is a far more forceful to-
nality than the tonic. As in Pause, the structural dissonance of
the music cannot be contained by the tonal architecture; it
breaks the frame.
The brief remainder of the song, fading from piano to
triple piano, helplessly concedes the point. Initiated by an eerie
altered chord, the close is almost astonishing: the long un-
broken chain of half cadences is not resolved by an antithetical
perfect cadence, but by a tepid plagal one. More a surrender
than a conclusion, the final cadence doubles its failings by
moving incongruously to the tonic major. The Picardy third
seems like a solecism in music so fixedly somber. Its DH, sub-
liminally echoing the dissonant D-sharp tonality, sounds the
wrong note in both a figurative and a literal sense. This way of
ending represents both a strained attempt to assert the integrity
of the tonic and a confession that long-term dissonance resolu-
tion is impossible here. Impossible: not so much for the music,
taken abstractly, but for the subject whose music this is.
The other harmonic contour of Der Doppelganger involves
the disintegration of the i-V 6 -III-V4 pattern by the alteration
of its closing chord. The focus of the alterations is the estab-
lishment of Cl| as an unresolvable long-term dissonance inde-
pendent of tonal orientation. This process occupies the second
half of the song, where it grows more disruptive by degrees.
The first chord affected is simply a dominant seventh with its
fifth flattened from Q
to Q; left unresolved, the seems to Q
222 SCHUBERT
Example 6

be an expressive inflection evoked by "Schmerzen," the word

it accompanies. With the second affected chord, things get
more problematic. Introduced triple forte as the speaker first
recognizes his double, the chord is the seventh Q-E-G-Ajl,
reinforced with a Q in the bass. The spelling indicates a Ger-
man sixth of the subdominant, E; the other possible function
of the sonority, as a dominant seventh of the tritone F, is
penumbrally suggestive. The chord, however, is resolved ac-
cording to neither of these functions; its G is treated as an
appoggiatura and resolved to produce another V* with a re-
calcitrant flattened fifth (ex. 6). The tonic then returns to begin
another extended half cadence, but its initial movement to the
dominant comes to grief against the next incarnation of the Q
dissonance: a glassy measure filled by a bare Q-Fjf tritone that
announces the dread term "Doppelganger." This is resolved to
the bare dominant fifth Fjt-Q, but the failing progression is
then disturbed again by the clangorous substitution of the raised
mediant, D#, for the closing dominant chord. For its final visi-
tation, the "wild" C\\ returns with uncanny effect in a root-
position C-major triad. Unlike the earlier C\ seventh and
Q-F# tritone, this chord fails to lead to the dominant that it
displaces. As if to parody the tonic, it moves instead to its own
—
V 7 /iii which, as the song's V 7 /iv, ushers the music to its
nerveless plagal close. The final tonic triad remains polarized
against the C\ dissonance that it has failed to resolve on either a
large or a small scale. It is less of rest than a disavowal of
a point
the paralysis that has overtaken the Classical harmony.
 The Schubert Lied 223

Like Einsamkeit and Geistes-Gruss, Der Doppelganger is

built from the confrontation of antithetical halves, though
without a collision of textures. And here the confrontation has
meaning: like the poetic speaker, the song is split into
a special
a"normal" half and a mirror image that mocks and distorts it.
The music, too, has its doppelganger. But Schubert's expres-
sive design does more than simply absorb the self-punitive pat-
tern of the text. When Heine's speaker encounters his double,
he desperately resists acknowledging his identity with this other
self. The crux of the poem is a sudden shift from description to

direct address: the speaker turns to, turns against, the double
with a question that is an accusation, an act half of defense and
half of defiance:

Da steht auch ein Mensch und starrt in die Hohe,

Und ringt die Hande vor Schmerzensgewalt;
Mir graust es, wenn ich sein Antlizt sehe,
Der Mond mir meine eigne Gestalt.
zeigt
Du Doppelganger, du bleicher Geselle!
Was affst du nach mein Liebesleid,
Das mich gequalt auf dieser Stelle
So manche Nacht, in alter Zeit?
[There stands a man and
stares on high, and wrings his
hands in violent shudder when I see his face the
grief; I —
moon shows me my own form.
You doppelganger, you pale companion! Why do you
ape my love-torment, which tortured me in this place so
many a night, in the old times?]

— —
The difference the saving difference between the self and its
shadow is manifested by the muteness of the double, which
Heine's imagery stresses hard. Schubert's musical persona, in
contrast to all this, accepts his identification with the specter
passively and half-nostalgically. Most of the text is set in an
overwrought, declamatory style, but at the thought of old
times, with their many nights of tortured love, the musical
22
persona breaks into eloquent, melismatic song (ex. y). With
this gesture, he repeats and travesties his earlier grief for the
ear, as the double has done for the eye: he gives the mute
double a voice. The Liebesleid returns as a distorted Liebeslied,
in an animation of Heine's submerged pun. Like the other vie-
224 SCHUBERT
Example 7

so man - 6m, Wacfit, in

p^
al fcr
P
Zeii?

tims of Romantic repetition, the musical persona surrenders

moment of erotic failure in which his inner
his allegiance to the
life was formed.

VI. Ganymed (D. 544, 18 17)

In an effort to capture a moment of supreme ecstasy, Ganymed

carries presentationalindependence to a giddy, not to say reck-
less extreme. The song begins with the most elementary tonic
and dominant harmony in A-flat major. By the time it ends, it
has evolved into F major by means of a harmonic process that
is little more than nonsensical from the Classical standpoint

evoked by the opening measures. This melting away of tonal

coherence follows closely from an appropriation of Goethe's
poem that is both radical and emotionally urgent; so for this
analysis, we have to begin with the text.
Goethe's Ganymed is an example of an image frequent in
—
Romantic poetry, a figure who attracts seduces the divine —
through the richness of his consciousness. 23 Within the poem,
he participates double transcendence. Hrst, there is an ec-
in a
static immersion of the self in nature; then, when this proves to
awaken more desire than it can satisfy, there is a transcendence
of nature into the presence of a divine principle. Both nature
and the divine are personified, one in feminine form, as Spring,
the other in masculine form as the "all-loving Father," Zeus,
or, as Holderlin would call him, "Vater Aether," the fertilizing
origin of spring. Both personifications are also the objects of
intense erotic longing —
for that matter, are the imaginative
products of that longing, which is Ganymed's only mode
of consciousness. The transfer of desire from a female to a
male love object, from nature to divinity, suggests a poem of
radical disjunction, and Goethe marks the turning point when
Ganymed recognizes that his ever-burgeoning desire is leading
him somewhere, and asks where: "Wohin? Ach, wohin?" But a
 The Schubert Lied 225

look at the poem reveals that this discontinuity is actually the

articulationof an underlying continuity. Nature and the divine
are woven together into a larger unity that humanizes both in a
vision of answerable desire:

Wie im Morgenglanze
Du rings mich angliihst,
Fruhling, Geliebter!

Ach, an deinem Busen

Lieg ich, schmachte,
Und deine Blumen, dein Gras
Drangen sich an mein Herz.
Du den brennenden
kiihlst
Durst meines Busens,
Lieblicher Morgenwind!
Ruft drein die Nachtigall
Liebend nach mir aus dem Nebeltal.

Ich komm, ich komme!

Wohin? Ach, wohin?
Hinauf! Hinauf strebt's.
Es schweben die Wolken
Abwarts, die Wolken
Neigen sich der sehnenden Liebe.
Mir! Mir!
In euerm Schosse
Aufwarts!
Umfangend umfangen!
Aufwarts an deinen Busen,
All liebender Vater!

[As in the glow of morning you beam all around me,

Springtime, beloved! . . .

Ah, on your breast I lie and languish, and your flowers,

your grass, press on my heart. You cool the burning thirst
in my breast, lovely morning wind! In you the nightin-
gale lovingly calls to me from the misty valley.
I come! Where? Ah, where?
come, I

Upwards! I am urged upwards! The clouds float down,

 —

226 SCHUBERT
the clouds bend down to my yearning love. To me! To me!
In your lap, upwards! Embracing embraced! Upwards to
your breast, all-loving Father!]

As the poem proceeds, all the leading images associated

with Ganymed's springtime rapture evolve into complemen-
tary images that express his union with the Father. The earthly
rapture so to speak, transposed to a higher register. The
is,

breast of the beloved spring becomes the Father's bosom; the

misty valley becomes the floating clouds; the morning wind
that answers to desire becomes the bending clouds that re-
spond to "yearning love"; Ganymed's repose on the grass be-
comes his upward striving; and the passive, heteroerotic desire
for spring, which begins with being enveloped and ends in a
responsive caress, turns to the active, homoerotic desire for the
Father, which posits a reciprocity of active and passive love
with the key phrase "embracing embraced!" The heavenly bliss
even subsumes the feminity that is left behind on earth, as
Ganymed yearns within the "Schoss" of the clouds: "lap"
in reference to a male, but also "womb." The poem achieves
closure as Ganymed's longing, a conflation of the erotic and the
metaphysical, reaches the threshold of both fulfillment and
quite literally —
sublimation in his ascent to the Father.
Schubert's song takes over the purposeful fluidity of the
poeirTwithout fully matching Goethe's extravagance and am-
bivalence. The music depends on two extended Romantic
presentations flf disjunctive one that evokes the overt gap be-
tween earth and heaven, female and male, and a continuous one
that weaves the divided forms into a larger unity. These pat-
terns unfold simultaneously. Overlapping rather than colliding
with each other, they blend together in a kind of structural
polyphony.
The disjunctive presentation involves a contradiction be-
tween the ultimate establishment of a tonic key and an impres-
sionistically conceived harmony that floats unpredictably from
tonality to tonality. The song moves through a series of four
primary harmonic areas, each of which is associated with its
dominant: A-flat-C-flat-A-F. From a Classical standpoint,
the tonal sequence is arbitrary, even a little bizzare. There is
neither a harmonic descent nor a harmonic circle, neither a ca-
 —
The Schubert Lied 227

dential contour nor a central key. But the harmonic texture of

the music is paradoxical. Though the series of primary areas is
virtually phantasmagorical, the harmony within each area is
unproblematical in Classical terms. In other words, long-range
and short-range syntax have been hopelessly sundered. The
harmony, as a result, has the moment-to-moment effect of a
curious fragility, a crystalline or deliquescent quality, almost
given the lack of a tonic —
a sense of illusoriness.
Yet a tonic is established: there is no sense at the end of the
—
song that closure has not been achieved quite the contrary. In
part, Schubert manages this by weighing his proportions. Of
121 measures in the song, 46 are in F major, so that when
the music cadences in that key and stays there till the close,
the intention of tonic resolution, if not its Classical rationaliza-
tion, is achieved. But something more radical is at work, too.
As the four tonic-quality areas succeed each other, the syn-
thesizing and organizing power of tonality is called on more
and more fully. Each new area uses more dissonant, more
complex harmony than the last, and each, as it emerges, re-
solves a greater degree of harmonic uncertainty. The eventual
tonic thus evolves through a gradual, indeed seamless, process
of intensification, rather than through the clearly directed,
highly articulated harmonic movement demanded by the Clas-
sical style. In particular:
1. The A-flatharmony of the opening remains placid un-
til the first big crescendo, when a jarring chromatic chord
brings an interrupted cadence. Schubert takes the dissonance as
a pivot for a cadential progression to C-flat (ex. 8).
2. G> (V of C-flat) is respelled as Fit and the music turns
or rather wanders, since the harmonic foreground is fairly
volatile —
down the circle of fifths to E. The new tonality takes
charge for several measures; then a strong cadential progres-
sion leads into A major/ minor.
3. The tonic F major is presented as a specific negation of
the preceding E and A. Schubert announces that A is a transi-
tional tonal area by restricting its triads to their first inversion,
but it is not yet clear where the music is heading. E briefly

seems to be a possibility again, but the allusion to it only con-

firms its inadequacy as a point of arrival. With dramatic abrupt-
ness, a I chord of E minor rises into a forcefully attacked
228 SCHUBERT
Example 8

C
b
: u6 vf/ii a* Vl/a

C-major triad. After harping on the C-major sonority for sev-

eral measures, Schubert interprets C as a dominant and ca-
dences into F major. The strength of this final tonality is estab-
lished by its referential control over the harmonies that follow,
the liveliest and most wide-ranging in the song. F major is
never left, but there is not another cadence on it for over thirty
measures. Its presence is contextual, centralizing —
which is to
say that it acts like a strong tonic.
The overall shape of the fifty-odd F-major measures con-
sists of a movement toward three progressively longer ca-
dences — the first dominant, the others tonic —
on the closing
phrase of the text, "All liebender Vater." The cadences con-
summate the song by calming its volatility; they all slow the
pace of the accompaniment and float the voice on a long, lei-
surely melisma. At these moments, too, a rhythmic resolution
supports the harmonic one. The sequence of primary tonal
areas is graduated rhythmically as well as harmonically. Each
new area brings with it an increase in rhythmic agitation, the
result of increasingly sensitive syncopation in the accompani-
ment and an increasing tendency for the voice to run phrases
of text breathlessly together. The F-major section carries this
—
rhythmic tension to a climax, but it also resolves it precisely
in the serene contours of the three cadences.
The arrival at F major, then, splits the song harmonically
into a wandering half and a centered one, and so articulates
the gap between mundane and divine rapture that Ganymed
quickly, inexplicably, orgasmically crosses. In the poem, move-
ment across this threshold is marked by an evaporation of self-
 The Schubert Lied 229

consciousness, but the song —

perhaps defensively
24
takes the —
opposite tack and chooses this moment to reflect on itself with
a text-music parallel. The first cadence to F major occurs dur-
ing the setting of Goethe's transitional question-and-answer,
"Wohin? Ach, wohin?/Hinauf!" The repeated "wohin" is set to
a dominant harmony; the resolving "Hinauf " divides its syl-
lables between V 7
and I of F major. The cadence to the long-
awaited tonic thus confirms Ganymed's discovery of his divine
destination, while at the same time Ganymed's cry of discovery
proclaims that the musical destination has been reached. Clo-
sure does not come within the framework of inherited dis-
course, either perceptual or musical: it comes where desire is
quenched by the breaching of such discourse.
Schubert's realization of this consummating breach is dis-
tractingly effective, but it is, as we know, only half the story.
The song also presents a realization of the seamless continuity
that joins the mundane and the divine, the empirical and the
transcendental ego. For this, it relies on a structural symmetry
that both cuts across the divisions of the text and leaves the
evolution of the tonic out of account.
The contradiction between immediate harmonic coher-
ence and large-scale harmonic drift that governs the song as a
whole also operates at a more intermediate level. Since each of
the four primary harmonic areas is linked with its dominant,
the overall fluidity of harmonic movement is locally contra-
dicted by the strong contour of tonic-dominant harmony.
What Schubert does with this situation is to create what might
be called a denatured (in terms of the text, de-natured, tran-
scendentalized) cadential pattern. The first two primary areas,
A-flat and C-flat, move second two,
to their dominants; the
A and F, are produced from their dominants. In other words,
the sequence of harmonies resembles —
at a distance the most —
basic pattern in tonal music: a large half cadence followed by a
large full cadence, I-V-I-V//V-I-V-I. Conjoined by the
fluid transitional movement from F-sharp major to the first
E-major cadence, the half- and full-cadence groups comple-
ment each other as structural upbeat and structural downbeat.
This pattern of disembodied large-scale cadences is one
among several sources of harmonic continuity in Ganymed,
each of which uses a denatured fragment of tonal syntax to
230 SCHUBERT
re-create the movement of transcendence found in the poem.
A second pattern depends on the fact that each of the large half
cadences lies a semitone apart from one of the full cadences:

At-Et E-A
O-GKFtf) C-F
I V I V // V I V I

From this standpoint, Ganymed can be taken as a single enor-

mous chromatic appoggiatura, Gl>— F^, once again marked off
as such by the transitional area between Gl>/F# and E. This
relationship not only matches the transposition of images in
Goethe's text —
mist to clouds, mother earth to Father Aether,
and so on — but also suggests that the tensions of Ganymed's
earthly desire can only be resolved by restatement in a celestial
embrace. The sense of longing associated with semitonal voice
leading in both Classical and Romantic contexts becomes a
diffuse presence in the musical texture, which imperceptibly
transforms the longing into consummation.
Chromatic relationships also give a plausible meaning to
the seemingly random sequence of primary tonalities. The first
half of the song articulates a rising chromatic third-relationship,
A-fiat major to C-flat major. In the second half, this dissoci-
ative movement is balanced out by a falling diatonic third-
relationship, A minor to F major. The means for this move is C
major, the resolved form of the structural upbeat, O. Com-
plementary movement by thirds thus replaces the same sort
of movement by fifths as a cadential contour, and the resolu-
tion of large-scale harmonic tension comes to depend on the
retrospective definition of C-flat as a chromatically lowered
dominant.
The line of relationship between C-flat and F is drawn by a
beautiful and unexpected melodic cross-reference. Such allu-
sions are rare in this song, which makes a point of flouting sur-
face coherence; until the repetitions at the close, melodic figur-
25
ations melt away soon
they have established themselves.
as as
The fruitful exception to this is the melismatic setting of the
phrase "unendliche Schone," which anticipates in texture the
pair of full cadences on "All liebender Vater" at the close of
the song. As it happens, this early melisma occurs just as the
music cadences on C-flat; the tone O that emerges as the me-
 The Schubert Lied 231

Iodic goal of the voice belongs to the initial C-flat-major triad

(ex. 8). The concluding F-major melismas, especially in their
predominant Bl;(=0)-Ct| appoggiaturas and in the closing
melodic fall C-F, seem both a reminiscence and a correction of
the critical tritone relationship on which this harmonic design
turns (ex. 9).
The
pattern of complementary thirds does not have the
immediate clarity of a large-scale Classical progression. But its
presence is established, more or less subliminally, when we ac-
cept the continuity of the song and hear the F-major ending as
a full release o£ tension. The production of the tonic by inten-
sificationis supplemented by a new tonal syntax, formed for

the moment, which depends for its effect on its dialectical rela-
tionship with the Classical harmony that it violates.
Overall, the impact of Ganymed depends on its smooth
integration of seemingly irreconcilable elements. By keep-
ing the harmony within each tonal area simple and Classical,
Schubert makes the listener feel at home, yet by making the
large form a kind of sublime non-sense, he makes a transcen-
dental argument. There is a perceptual flux, a constant blur-
ring and evaporating of relationships, in the music, yet it is
pervaded by a peculiar feeling of direction, as if toward an un-
recognized object of desire. This ambiguity is enhanced by the
intricate overlapping of harmonic presentations, which co-
alesce in the presence of a tonic that comes from nowhere, po-
tent but a little unreal. Perhaps this is felt more strongly after
the closing cadence, which is not allowed to evoke any sense of
232 SCHUBERT
Example 10

piano moves into

rest or fixity. After the voice falls silent, the
the upper registers for the time in the song. Three tonic
first

triads and a subdominant-quality diminished triad sound and

sound again, slowly ascending; then the song concludes with a
dominant-seventh chord resolving over a tonic pedal (ex. 10).
The strong resolution melts into air, as Ganymed's conscious-
ness dissolves into the presence of the all-loving Father.
All of this represents a highly selective response to the
text. It heightens the perceptual continuity of the poem while
demoting its eroticism, though the song represents the erotic
turning point of the poem with real intensity. The emphasis on
perception is continued even after the dissolution of the final
cadence when a fermata appears over a rest. The notation tells
the performers to prolong musical silence by not relaxing for
applause; the listener's musical perception must be allowed to
fade slowly into an echo or blank.
What this amounts to in relation to the erotic material is a
partial evasion, or more exactly a sublimation. And since sex-
ual desire in song hardly needs sublimating as such (many
songs actively ^sublimate it), we seem led to conclude that
what is being sublimated here is the homoerotic element in the
poem. In saying this, I do not mean to suggest that Schubert
is disavowing any homoeroticism of his own
26
— nor, for that
matter, that he isn't; I simply don't know. But his song wants
to idealize its rapture, and therefore declines to give direct ex-
pression to the erotically charged yearning of a male persona
toward another male. Instead, it defuses the sexual radicalism
of the text by assimilating it to a perceptual deliquescence
so extreme that all ordinary connections —
tonal, emotional,
erotic —are abstracted and etherealized. In a sense, this means
 The Schubert Lied 233

only that Schubert is less erotically sophisticated than Goethe,

which hardly comes as a surprise. Goethe uses homoerotic de-
sire in "Ganymed" to reinforce the presentation of an ecstasy
that transcends the presumed boundaries of nature. Later, in
Faust 2, he does so again, when Mephistopheles, gloating over
the dead Faust, is suddenly overwhelmed with lust for the boy
angels who come to fetch Faust's body away. "Der Geist der
stets verneint" forced into an affirmation of being through
is

the desire awakened by the boys.

But Schubert is not working toward that kind of affirma-
tion. What he is after instead is perhaps best suggested by
Schoenberg's remarks on the finale of his own String Quartet
no. 2 in F-sharp Minor, which contains his first stretch of fully
atonal writing. The music is supposed to represent a move-
ment of transcendence that purifies the self of earthly cares,
of the gravity of life and the "gravity" of tonality: "passing
through the clouds into thinner and thinner air, forgetting all
27
the troubles of life on earth." Schubert, presenting an ab-
stracted form of harmonic motion in which the tonic is free-
floating and the cadence is free of content, seems bent on evok-
ing the same kind of release. The eroticism at the margins only
heightens the feeling.

VII. Conclusion

As aRomantic genre, the Schubert Lied articulates a multi-

plicityof conflicts: between harmonic languages, between the
—
Romantic subject and the force the embodiment, direction,
transformations —
of its own desires, and between the musical
and poetic personae. Perhaps the most striking thing about this
process is its spontaneous quality; it seems to reflect the pres-
sure of a subject working from within the expressive world of
Classical tonality rather than the pressure of a subject intruding
from without. Schubert accomplishes this by confining his
harmonic innovations to an almost covert erosion of musical
syntax. He does not often indulge in functionless symmetrical
sonorities, chromatically overladen chords, mystification by
semitonal part writing, or the suppression of root progression
as an audible principle of structure. His discontinuities are
produced with the Classical palette of dissonances and exotic
234 SCHUBERT
chords. Dominant sevenths, diminished sevenths and triads,
Neapolitan and augmented sixths are all put to uses that expose
the rationality of Classical syntax as historically contingent
rather than natural. Concomitantly, the static, reiterative ac-
companiment figures that are basic to Schubert's style help to
isolate planes set them in opposition to each
of sonority and
other as diatonic progressions are simplified and stripped of
the submerged traditions that tie them with seeming inev-
itability to certain standard dissonances. Schubert's habitual
fondness for chromatic third relationships has the same effect.
By loosening the specificity of harmonic tension, as embodied
by the demand of the dominant for the tonic, they isolate dis-
parate sonorities and make dialectic audible.
The history of the Lied after Schubert suggests the impor-
tance of positioning the music on the edge between paradigms.
Schumann often follows Schubert's model in his songs, but
with Wolf the authority of the Classical style is often so attenu-
ated that the music — —
movingly enough dissolves rather than
reinterprets its bonds to the past. One
dissolution, moreover,
seems to lead to another. Anxious and ego-laden, the songs
substitute the perfection of declamation for a transforming re-
sponsiveness to the text. Part of the great outpouring of vocal
—
music in the early twentieth century by the early Schoenberg
—
and Berg, by Mahler, Debussy, and Ives reflects the oppor-
tunity of renewing an exhausted argument between Classical
and Romantic tonality on the borders between tonality and
atonality. The Lied, at its most characteristic, is apparently a
twilight form, with a peculiar eloquence that comes from a
changing language. The Schubert Lied is its archetype.

Notes
i. The Classical Style (New York: Norton, 1972), p. 454.
2. See, for example, Ernest Porter, Schubert's Song Technique (London:
Dennis Dobson, 1961), p. 34.
3. See the essays by Alec Robertson (esp. pp. 15- 5 5) an d T. C. L.
Pritchard (esp. pp. 243-46) in The Music of Schubert, ed. Gerald Abraham
(New York: Norton, 1947).
4. See Charles Rosen, Sonata Forms (New York: Norton, 1980),
pp. 281-91.
 .

The Schubert Lied 235

5. Schubert, Symphony B Minor ("Unfinished"), Norton Critical

in
Score, ed. Martin Chusid (New
York: Norton, 1971), p. 75.
6. Edward T. Cone, The Composer's Voice (Berkeley and Los Angeles:
University of California Press, 1974), pp. 20-40.
7. Samuel Taylor Coleridge, Biographia Literaria, ed. George Watson
(London and New York: Everyman Editions, 1965), p. 167.
8. I will risk throwing out the suggestion that pre-Romantic song

keeps largely to various forms of identification, though often in terms that

alter the textual persona.
9. By "deconstruction," I mean the effort to show that naturalness,
immediacy, and self-evidence are effects produced by language, not proper-
ties of truth or presence, and are therefore of only limited authority. De-
constructive activity is not new, though the term, and a certain characteriza-
tion of the activity it describes, has come to prominence recently through the
controversial work of Jacques Derrida. The best short treatment of the sub-
ject is Christopher Norris, Deconstruction (London and New York: Methuen,
1982).
10. For a full analysis of the Beethoven quartet in this context, see my
Music and Poetry: The Nineteenth Century and After (Berkeley and Los An-
geles: University of California Press, 1984), pp. 57-75. On the Mozart, see
William DeFotis, "Mozart, Quartet in C, K. 465," lgth-Century Music 6
(i 9 82): 3 i-38.

11. "A Romantic Detail in Schubert's Schwanengesang," this volume,

p. 52.
12.See the discussions of this song by David Lewin and Anthony
Newcomb, included in this volume.
1 3 Such recountings or paraphrases are also interpretations disguised —
ones — that are based on the view that one kind of meaning, identified as "lit-
eral" or "referential," takes precedence over all others. This view of meaning
is a source of considerable dispute in contemporary literary theory.

14. In "The Mirror of Tonality: Transitional Features of Nineteenth-

Century Harmony," 19th-century Music 4(1981): 191 -208.
15. I borrow the term "exemplify" from Nelson Goodman's Languages

of Art (2d ed., Indianapolis: Hackett, 1976); for discussion, see Monroe C.
Beardsley, "Understanding Music," in On Criticizing Music, ed. Kingsley
Price (Baltimore: Johns Hopkins University Press, 1981), pp. 55-73; and
Anthony Newcomb, "Sound and Feeling," Critical Inquiry 10 (i984):6i4~43.
16. Franz Schubert's Werke (Leipzig: Breitkopf und Hartel, 1884-97),
vol. 20.*3 pp. 189-92. Repr. in Schubert's Songs to Texts by Goethe (New York:
;
Dover, 1979), pp. 85-88. Two additional versions or transpositions are men-
tioned in O. E. Deutsch, Franz Schubert: Thematisches Verzeichnis seiner Werke,
ed. and rev. W. Durr, A. Feil, etc. (Kassel: Barenreiter, 1978), pp. 99-100.
17. I am indebted to Walter Frisch for pointing this out.
18. Some comparison between Schubert's setting and the one that ap-
pears in Beethoven's cantata Meeresstille und Gluckliche Fahrt, also composed
in 1 81 5, seems obligatory. Beethoven's piece falls squarely within the Classi-
cal style. The music evokes a feeling of unrest and alienation through a vari-
236 SCHUBERT
ety of expressive devices: pitch repetition in all voices, liberal use of sus-

pensions and anticipations, reiterated harmonies. There is also a certain

avoidance of the tonic in the first part of the setting. Beethoven begins with
an ambiguous third, D-B, which is only established as a tonic triad through
7
a deceptive cadence that moves away from it: I-V -vi. This is followed by
another, more elaborate slippage from the tonic to the relative minor. There
is no cadence to a tonic triad until bar 31. All of this is highly effective, but it

is not problematical. Incidentally, Beethoven's appropriation of the text also

differs quite strikingly from Schubert's. Its emphasis, carried mainly by the
shocking forte attacks on "Weite," is on physical and spiritual isolation.
19. For a full discussion of Romantic repetition, see chapter 2 of my
Music and Poetry.
20. Schubert's Songs (1967; Seattle: University of Washington Press,
1969), pp. 59-61. See also Richard Kramer, "Schubert's Heine," lgth-Century
Music 8 (1985) :2i3~25. The original order of the poems is "Das Fischer-
madchen," "Am Meer," "Die Stadt," "Der Doppelganger," "Ihr Bild,"
"Der Atlas."
21. Schubert's Songs (1928; repr. New York: Basic Books, 1957), p. 256.
22. As Thrasybulos Georgiades points out, this shift from declamation
to song is articulated with a subtle rhythmic shift. Prior to "Was affst du,"
the vocal line always begins after the accompaniment. From this point on,
each vocal phrase anticipates the accompaniment. Schubert: Musik und Lyrik
(Gottingen: Vandenhoeck und Ruprecht, 1967), p. 76.
23. For an account, see my "The Return of the Gods: Keats to Rilke,"
Studies in Romanticism 17 (1978) ^83- 500.
24. See pp. 232-33 below.
25. Another exception, more felt than heard, is the motivelike use of a
falling whole tone in the voice, which tends to retard phrases as a mark of
yearning. This motive always appears as or in a dissonance —
another demand
for requital. At the close, the whole tone is expanded into the third A-F (re-
capitulating the resolving third relationship), then expanded again into the
cadential fifth C-F.
26. On the question of Schubert's sexuality, see Maynard Solomon,
"Franz Schubert's 'My Dream,'"
American Imago 38 (i98i):i37-54.
27. Arnold Schoenberg, "Notes on the Four String Quartets," in
Schoenberg, Berg, Webern: The String Quartets. A Documentary Study, ed. Ur-
sula von Rauchhaupt (Hamburg: Deutsche Grammophon, 1971), p. 48.
 Appendix of

Longer Musical Examples

All the examples in the Appendix are reproduced from Franz Schubert's
Werke. Kritisch durchgesehene Gesammtausgabe (Leipzig: Breitkopf
und Hartel, i884~g7).
 ; | —

238 SCHUBERT
Example A Schubert, Moment musical in A-fiat, op. 94, no. 6
(D. 780)

Allegretto.

JO ^
,
>T
1,
L"|
L
t>
^
1

i j
. ,0
h&~ !
r
—^= rlr^j — ij — —P-M—e ^
**? '* ^•—-
1

nil 3 ^
1

O' 4 1
-

1
p
- / kf
>K l> L | j» J— H £ --\yV 1 i » L
bF
\ A- » r

38

3r '"
Vp'f crese. P^
|» "}a

«
r-

aV i-,!)] j^TT
ri . 1

/
1

Jig
ju />/>

j-
 — . =

68
l
^ ilt ?'
00
^f
P fp — PP

»'

(77) Trio.

w *
.— ^ JJ

-6
PP
k):
i^Ha
\, i

—4H-ls -
p«
J*- -f^ -a f
r J
—
£:

rN~ ir
• 1

Allegretto D.C.
240 SCHUBERT
Example B Schubert, Wandrers Nachtlied ("Uber alien Gipfeln,
D. 768)

Langsam
Singstimme.

Pianoforte.
 Appendix of Longer Musical Examples 241

Example C Schubert, Im Dorfe, from Winterreise (D. 911, no. 17)

Etwas langsam.

Singstimme

Pianoforte

»') JJjJJJJJJTJ'JBf v i J Hi a in i jT] !jp 1 , <

 s ^

9 -

Men - schen in ill ren Bet - ten,

( cresci
L- 3* 3» 3» J* > > 3» 3» 3* -J»-

[V^r-ff^ '— ^_^

11

trau men sich Man ehes, was sie nicht

1 1 4 ^ ^
P jftP

—
fc):

IZ_1_^
ft
'

1 I H-J i-

XJ <Of i i i

13

ha ben, thun sich im Gu - ten und Ar . gen er .

I
* m • * * J ^ft ^ dftjdjdjdjujtj Jjfj 7 7 & 7
creac.
-*—*=-
h2 J l i
f=
*
1 1 *=
V

f F I I
i L
15

la . ben; und mor. gen

- » - rrvr^f |l)^^" • m f
>-»- » «< -A*
1

1 1 i
^— J

J ' 7 7 * 7

i
PP
±\ ft
— ff- i 1 a a a a 9 j

ritard.
 Je nun, je nun, sie haben ihrTheil genossen, und

hoffen, und ho ffen, was sienochub.rigliessen, doch wie . der zu fin . den,doch wie . der zu
47 f?s

i
4 4' t
4
I
j
1
'
M
1
- 4 * i 4 i i'
i t t 1
11 •

TT-
 — — — — H

Appendix of Longer Musical Examples 245

Example D Schubert, Moment musical in F Minor, op. 94, no. 3

(D. 780)

Allegro moderate*

±* 3-
W F3?H flBJP
fr «?l
^ ^^ 2*2* ^g
y^ J if I

JJ M
n rn r m i rn r i r.
i
1 nw bw rj 1

£ R^^ > >

1^ i^jTTf? 1

^ '
.

^-
r#

f-
»

r
1 ^ r —
* 4
1

'*j¥^
•' t> 1/
—
1
F^; P~-
r
J f fi 1

itit
1 1

j
1

1
iii i 1

22 >• =»
r*
^ rt-
^F«=,pi
P ££J
,b

£/
>- :>-
F ##*» all

t
|ai^ r
*'«; '
i< ij-i. ^1 =¥* [
1

it it
29
^ r
r
zfc tic 1 1 |

\p±_ 9 Pj

liAlp ,_ 11
j id
* y~lFJT4-*^£~i—
1
1
1 J
lr
Lfc b
l
'g
i»i l D

^i^- >» ,
>
>"^J^TJj Wj
— 1
jq.._TTTh
(Jf^f-n
t t -¥-*-
[tyii'.lr ,1-1 f^ If
» — |

— U 4 V v '
 — • — ' 1

p-4-4-
42 , > »
;?^i >• J
-Pi
•

T? ^ i
.Tr.n:

/.» t
'* *
m m.
^ f
t
('>>.'.'.•
E-i E5a 1

B B 1

1J
I

^
f
1 ; ;
tp-j r^—

49
gfl ^» m !

| J J, | pj 1 J
rH=F3-n
—* <-f j H_ K '

— f=n— R
r

— — »g 4=a •
^ 1

c «L*J>» 1 2- r r
pp
» » * f »
f
-*— : — * :
\
;
f ; p ;

^ — e^ *j>/TJ
p|
£ ;
?
--
» * • -»—
yJ
=
f

# *
-

*
»

*
»

*
1
-, f |

t
•*•«
f

* It
([•)••',. . T— 1. ^=
ft

r , J==
f

61 _^~~^ ^* —
4
; ~ »J—
-# » -
-f |
« | _• • ..I. ,
^y-

s -=
^ "if-
9- <0-
PPP
=^ dim.
i it

'

-J — lu* — :
-J ^_ 1— c3
1

—cd l=j—
1

uJ

^
72 .
n^p] ""^^1 o
ih
-6* • —5*— *H
(«- :
j

r
^ * ^ ^ 5

[-^ 1

o
 Appendix of Longer Musical Examples 247

Example E Schubert, Aufdem Flusse, from Winterreise (D. 911, no.

7)

,angsam

Singstimme.

Pianoforte.

6 sehr 'eise

rauschtest, du hel ler, wil . der Fluss, wie still bist du ge . wor.den, giebst

f $ '

ppp
rl i~
\-pg-i —2 Jj—2 —^ 1
7
J)
7
=d

du dich ii . ber. deckt, liegst kalt und un . be . weg. lich im San.de_aus . ge

T;ig, an dem ich_ ging; urn Nam' und Zah . len win . . det sich
 auch so roi . ssend schwillt, ob's wohl auch so rei . ssend

P^ft
 The Contributors

Edward T. Cone, professor emeritus of music at Princeton University, is the

author of numerous critical and analytical studies, including two books, Mu-
sical Form and Musical Performance (1968) and The Composer's Voice (1974).

Arnold Feil, professor at the University of Tubingen, has written two books
on Schubert, Studien zu Schuberts Rhythmik (1966) and Franz Schubert: Die
sch'one Mullerin, Winterreise (1975). He is also an editor of the Neue samtlicher
Ausgabe of Schubert's works.
David Brodbeck, assistant professor of music at the University of Southern
California, completed his Ph.D. at the University of Pennsylvania in 1984 on
Brahms's editions of Schubert dances, and the influence of the dances on
Brahms's own music.
The most recent works of Carl Dahlhaus to appear in translation are Aesthet-
icsof Music (1982), Analysis and Value Judgement (1983), Foundations of Music
History (1983), and Music and Realism (1985). His major study of nineteenth-
century music, Die Musik des ig. Jahrhunderts, was published in 1980 and is
due in translation from the University of California Press.
Walter Frisch, assistant professor of music at Columbia University, is the au-
thor of Brahms and the Principle of Developing Variation (1984) and a coeditor
ofthejournal lgth-Century Music.
Thrasybulos Georgiades, who died in 1977, was professor of music at
Heidelberg, then at Munich. He was the author of wide-ranging studies on
music, including Greek Music: Verse and Dance (1949; English translation,
1956) and Music and Language (1954; English translation by Marie Louise
Gollner, 1982).
Marie Louise Gollner, professor of music at the University of California,
Los Angeles, is a specialist in medieval and Renaissance music. A former stu-
dent of Thrasybulos Georgiades, she translated his book Music and Language
(1982).
252 SCHUBERT
Joseph Kerman, professor at the University of California, Berkeley, has been
closely involved with both the theory and practice of criticism. His writings
include Opera as Drama (1956), The Beethoven Quartets (1967), and Contem-
plating Music: Challenges to Musicology (1985).
William Kinderman, who teaches at the University of Victoria in British Co-
lumbia, is an accomplished pianist and has published analytical articles on

Beethoven and Wagner. His monograph on the Diabelli Variations is forth-

coming from Oxford University Press.

Lawrence Kramer, who teaches in the Humanities Division of Fordham Uni-

versity, is a composer and the author of Music and Poetry: The Nineteenth Cen-
tury and After (1984).
David Lewin, professor of music at Harvard, has published many articles on
music theory, as well as analyses of Schoenberg, Brahms, and Wagner.
Anthony Newcomb, on the faculty of the University of California, Berkeley,
works in both Renaissance and Romantic music. He has published articles on
Wagner, Schumann, and on more general topics involving aesthetics and
analysis.

Thilo Reinhard, who holds an M. A. in composition from the University of

California, Berkeley, is preparing The Singer's Schumann, a volume of transla-
tions and commentary intended to aid singers and vocal coaches.
 1

Index of Composers and Works

Boldface indicates a more extended analysis.

Bach, Johann Sebastian: 29 n Trio, piano and strings, B-flat

Beethoven, Ludwig van: 1, 7, 8, major, op. 97 ("Arch-
117 duke"): 59
WORKS Variations, piano, C minor,
Andenken, WoO 136: 177, WoO 80: 10- 1
io8n Variations, piano, E-flat
Meeresstille und Gluckliche major, op. 35 ("Eroica"):
Fahrt, op. 112: 235~36n 4, 9
Quartet, strings, D
major, Berg, Alban: 200, 234
op. 18, no. 3: 19 Berlioz, Hector: 219
Quartet, strings, C major, Brahms, Johannes: 4, 31-32,
op. 59, no. 3: 203 40-41, 44, 45n
Sonata, piano, F minor, op. ;

no. 1:8 Chopin, Frederic: 15-16

Sonata, piano, F major,
op. 10, no. 2: 19 Debussy, Claude: 234
Sonata, piano, D
minor,
op. 31, no. 2 ("Tem- Haydn, Franz Josef: 117
pest"): 9
Sonata, piano, E-flat major, Ives, Charles: 234
op. 81a ("Les Adieux"):
'
192, 193 Krause, Christian Gottfried: 175
Symphony no. 3, E-flat
major, op. 55 (Eroica): Mahler, Gustav: 1, 4, 234
14-16 Mendelssohn, Felix: 15, 16, 26
Symphony no. 7, A major, Mozart, Wolfgang Amadeus: 103 n,
op. 92: 44 203
 n n

254 SCHUBERT
Reichardt, Johann Friedrich: 175, F minor (D. 940): 71,
182-83, 187, 189, 190, 197, 75-82
200 Fantasy, violin and piano,
Rossini, Giaocchino: 62-63 C major (D. 934): 64n
Forelle, Die (D. 550): 63
Schoenberg, Arnold: 200, 233, 234 Ganymed (D. 544): 224-33
Schubert, Franz Geistes-Gruss (D. 142): 54-56,
WORKS 208-9, 2ii, 223
Alfonso und Estrella (D. 732): Gesang der Geister iiber den

82n Wassern (D. 484, 538, 705,

Allmacht, Die (D. 852): 60, 184 714): 54
Am Flusse (D. 160): 184 Gondeljahrer (solo song,
Andenken (D. 99): 176-79, D. 808): 50-52, 58, 64n
188, 189, 193, 194, I98n Gondeljahrer (part song,
An Mignon (D. 161): 184, 197 D. 809): 52
Blinde Knabe, Der (D. 833): 50 Gretchen am Spinnrade, op. 2
DANCES (D. 118): 50, 65, 66, 184
16 Deutsche and 2 Ecos- Heidenrbslein (D. 257): 176,
saises, op. 33 (D. 783): 206
40-42, 46n
31, Heimliches Leben (D. 922): 64n
Galop and 8 Ecossaises, Hippolits Lied (D. 890): 50
op. 49 (D. 735): 46n Ihr Grab (D. 736): 64 n
12 Grazer Walzer, op. 91 Impromptus, op. 90 (D. 899):
(D. 924): 46 117
16 Landler and 2 Ecossaises, Landler. See Dances
op. 67 (D. 734): 46n Mass, E-flat major (D. 950):
12 Landler, op. 171 45 n
(D. 790): 31-45 Meeres Stille (D. 216): 210-
36 Originaltanze, op. 9 15, 216
(D. 365): 45-46n Moments musicaux, op. 94
12 Valses nobles, op. 77 (D. 780): 117
(D. 969): 46 No. 3, F minor: 116-24
34 Valses sentimentales, No. 6, A-flat major: 16-28,
op. 50 (D. 779): 46n 47 n
12 Waltzes, 17 Landler, 9 Musensohn, Der (D. 764):
Ecossaises, op. 18 203-4
(D. 145): 46n Nachtgesang (D. 119): 184
Dass sie hier gewesen (D. 775), Nahe des Geliebten (D. 162):
66-70 175-97
Drei Clavier-Stiicke (D. 946), Octet, winds and strings,
45 n F major (D. 803): 28
Erlkonig, op. 1 (D. 328): Pastorella al prato, La
65-67, 74, 78, 202 (D. 528): 46n
Fantasy, piano, C major, Quartet, strings, minorD
op. 15 ("Wanderer"; ("Death and the Maiden";
D. 760): 62, 75 D. 810): 32, 33-35, 201
Fantasy, piano four hands, Quartet, strings, A minor,
 n

Index 255

op. 29 (D. 804): 201 finished"; D. 759): 28

Quartet, strings, G major, Symphony, C major ("The
op. 161 (D. 887): i-ii, Great"; D. 944): 71, 75,
28, 61-62 79, 124
Quintet, piano and strings, Tod und das Madchen, Der
A major, op. 115 (D. 531): 76
("Trout"; D. 667): 62 Trio, piano and strings, E-flat
Quintet, strings, C major, major, op. 100 (D. 929):
op. 163 (D. 956): 28, 7i
59-60, 61, 62, 63, 64T1, Trost in Tranen (D. 120): 184
201 Waltzes. See Dances
Ruhe(D. 635): 46n Wandrers Nachtlied (D. 768):
Schdfers Klagelied (D. 121): 184 84-103, 186, 193
Schone Mullerin, Die, op. 25 Willkommen und Abschied
(D. 795): 50, 71, 103 n, (D. 767): 50
164, 167, 169, I74n Winterreise, op. 89 (D. 911):
1. Das Wandern: 103 46, 59, 69, 82, 156, 157,
2. Wohin: 50, 103 164-70, i73-74n
10. Tranenregen: 197, I99n 1. Gute Nacht: 70, 75, 166,
12. Pause: 215-18, 219 168
18. Trockne Blumen: 50 2. Die Wetterjahne: 56, 166,
19. Der Mailer und der Bach: 50 168, 169
Schwanengesang (D. 957) 3. Gefrorne Tranen: 56,
2. Ahnung: 63 n
Kriegers 166—69 passim
5. Aufenthalt, 63 n 4. Erstarrung: 166-70 pas-
6. In der Feme: 56, 58, 64n sim, 173 n, I74n
8. Der Atlas: 63 n, 64 n, 218 5. Der Lindenbaum: 166-70
9. IhrBild: 48-50, 57~59, passim, I74n
71-75, 82n, 218 6. Wasserflut: 166-70 pas-
11. Die Stadt: 53, 54-56 sim, I74n
passim, 58, 207-8, 7. Aufdem Flusse: 126-52,
218, 219 153-70, 205-6
12. Am Meer: 52-53, 56- 8. Riickblick: 164-70 passim
58, 204-5, 206, 218 9. Irrlicht: 166-70 passim

13. Der Doppelganger: 63 n, 10. Rast: 166-70 passim,

218-24 I74n
Sonata, piano, A minor, 11. Fruhlingstraum: 70, 166,
op. 143 (D. 784): 201 167, 168, I74n
Sonata, piano, D
major, 12. Einsamkeit: 166-69 pas-
op: 53 (D. 850): 193 sim, I74n, 205-6, 215,
Sonata, piano, B-flat major 223
(D. 960): 59, 201 13. Die Post: 128-29
Symphony no. 4, F minor 16. Letzte Hoffnung: 56
("Tragic"; D. 417)128 17. Im Dorfe: 104-16
Symphony no. 6, C major 19. Tauschung: 70-71, 82
(D. 589): 62-63 20. Der Wegweiser: 71, 75
Symphony, B minor ("Un- 22. Mut: 165, 168, 169, 173 n
2$6 SCHUBERT

Winterreise (continued) Wagner, Richard: 219

23. Die Nebensonnen: 165, Weber, Carl Maria von: 62-63, J 92
I73n Wolf, Hugo: 234
24. Der Leiermann: 56, 168,
169, 218, 219
Schulz, Johann Abraham Peter: 175
Schumann, Robert: 44, 116, Zelter, Carl Friedrich: 84, 174,
192-93, 219 180-81, 183, 187, 189, 190
Strauss, Richard: 116 Zumsteeg, Johann Rudolf: 200
 BOSTON PL BLIC LIBRARY — «• 1% U

1 11111111111
3 9999 03134 805 3
v.
•
\l

\N
v^s^**^
BOS'

sate
BAKER S TAYLOR
"The excellence of this collection stems from the brilliant sch<

and critical attitudes of its contributors. The importance of the

collection is threefold: it introduces in English translation the
work of three German scholars noted for their contributions to
nineteenth-century music studies; it provides provocative schol-
arly-critical interpretations that far outstrip the conventional
'impressionistic' approaches of the past; it demonstrates a new
approach to critical interpretation of Schubert's music. It is, in

sum, a contribution of outstanding quality. " — Richard Swift,

University of California at Davis

Addressing a wide range of topics— from Schubert's approach to

large-scale musical form to his innovations in instrumental forms
and Lieder— Schubert offers a diverse, illuminating portrait of the
composer and his music. Contributors to the volume include dis-
tinguished scholars Joseph Kerman, Edward T. Cone, David
Lewin, Walter Frisch, David Brodbeck, William Kinderman,
Lawrence Kramer, and Anthony Newcomb. There are also trans-
lations of writings by the influential German critics Carl
Dahlhaus, Thrasybulos Georgiades, and Arnold Feil.

Walter Frisch is a professor in the Department of Music at

Columbia University. He is the author of The Early Works of

Arnold Schoenberpf.

ISBN 0-8032-6892-0

9 780803 268920
ll
9 0000