der tonwille
volume ii

This page intentionally left blank

Der Tonwille
Pamphlets/Quarterly Publication
in Witness of the Immutable Laws of Music,
Oered to a New Generation of Youth by
HEINRICH SCHENKER
Semper idem sed non eodem modo
VO LU M E I I : Issues ()


E d i t e d by Wi l l i a m D r a b k i n
t r a n s l ate d by
ian bent
william drabkin
j o s e ph d u b i e l
j o s e ph lu b b e n
w i l l i a m re nw i c k
ro b e rt s na r re n b e rg

Oxford New York

Auckland Bangkok Buenos Aires Cape Town Chennai
Dar es Salaam Delhi Hong Kong Istanbul Karachi Kolkata
Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi
So Paulo Shanghai Taipei Tokyo Toronto

Copyright by Oxford University Press, Inc.

Published by Oxford University Press, Inc.
Madison Avenue, New York, New York,
www.oup.com
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.
Library of Congress Cataloging-in-Publication Data
Schenker, Heinrich, .
[Tonwille. English]
Der Tonwille : pamphlets in witness of the immutable laws of music /
Heinrich Schenker ; edited by William Drabkin ; translated by Ian Bent . . . [et al.].
p. cm.
Includes bibliographical references and index.
ISBN ---
. Schenkerian analysis. . MusicHistory and criticism. I. Drabkin, William.
II. Tonwille. III. Title.
MT.S T
dc

Printed in the United States of America
on acid-free paper

General Preface

History of the Tonwille Project (Continued)

The general preface to volume traced Schenkers dealings with Universal Edi-

tion from their beginnings in through to early , with particular attention

to those from onward concerning Der Tonwille. As shown there, conflict between Schenker and Emil Hertzka, the firms director, concerning this publication had already arisen by May , and relations had come under severe strain
in February , with Schenkers allegations of censorship and terrorizing.
In April , Schenker evidently questioned his royalty receipts; for Hertzka,
drawing attention to the fluctuations in currency exchange rates of the time, and
the imposition of numerous extra taxes and duties, suggested a meeting at which
a new financial arrangement could be discussed (OC /).
On October , , a meeting took place at which, although no minute
seems to have been kept, the two men agreed to move from the loosely scheduled publication of Tonwille issues (Hefte) to a regular quarterly schedule by
annual volume (Jahrgang), each comprising four issues. A change of subtitle
was approved on December (OC /), Flugbltter simply being replaced
by Vierteljahrschrift:
Der Tonwille / Vierteljahrschrift / zum Zeugnis unwandelbarer Gesetze der Tonkunst /
einer neuen Jugend dargebracht von / Heinrich Schenker
(Der Tonwille: quarterly publication in witness of the immutable laws of music,
offered to a new generation of youth by Heinrich Schenker)
At that point, issues had already been published, and and were in production; thus the quarterly (as Schenker now liked to abbreviate it) was to
begin with what was, by the old numbering, issue and, by the new, issue of annual volume IV, on the basis that the first six issues were published during a
three-year period, . The old numbering was preserved alongside the new

on both the wrapper and the title page: for example, annual volume IV, issue
(seventh issue of the complete series). We can reconstruct the overall pattern of
publication as follows (the middle column shows, where known, the date on
which Schenker was notified of publication by Universal Edition):
Numbering

Publication Date

Title-page Date

Tonwille

went to press: May ,

(OC /)

Tonwille

June ,
(OC /)

Tonwille

January ,
(OC /)

Tonwille

June ,
(OC /)

Tonwille

before February ,
(OC /)

Tonwille

April ,
(OC /)

year IV, issue

(Tonwille )

before August ,
(OC /)

JanuaryMarch

year IV, issues

(Tonwille /)

before November ,
(OC /)

AprilSeptember

year IV, issue

(Tonwille )

before January ,
(OC /)

October

A single exchange of letters in September helps gauge the temperature between

the two parties. On September , Emil Hertzka wrote to Schenker (OC /):

gener al preface
Relations deteriorated sharply in Spring over Hertzkas refusal to publish
issue intact, his grounds being that (a) the issue far exceeded the two-gathering
limit stipulated in the April/July , , contract; (b) much of the content of the
Miscellanea was unacceptable to him because it did not concern music, therefore
lay outside the terms of the contract; and (c) he found the polemical material
offensive, as it included a sustained attack upon a personal friend of his.
After the meeting with Hertzka at which quarterly publication is agreed (Oct.
, ), Schenker now believes that the old contract will be replaced by a new
one. However, no new contract is forthcoming. He staunchly maintains that the
provisions of the old contract, especially the limit of two gatherings (thirty-two
pages), do not apply from year IV, issue onward, as he writes here, probably in
October (OC / ):

Dear Professor,
I have just been informed that you maligned us in a wholly unjustified manner to one of the employees of the Gutmann Music Store. May I
ask you please to comment on this. I cannot credit this report, although
it comes from someone accustomed to speaking the truth.
In awaiting your most immediate and detailed comments, I remain
Yours respectfully
Universal Edition
Hertzka
This is Schenkers response of September (OC /, WSLB ):
I have today received your letter of the th; I freely admit that, in keeping with my character (with which you are familiar), a few days ago I gave
ventindeed full ventto my justified displeasure concerning your observed actions against me.

The first year of the quarterly [no longer?] had any contract whatsoever
as its basis. To make the first contract, designed for the irregular release
[of issues], apply to the quarterly, as the publishing house has done, is a
contradiction of the contract as well as of the very idea of a quarterly publication, not to mention also a contradiction of the word of the publisher
himself, but one which he later declared invalid with the excuse that his
word, dropped in a meeting, was not included in the contract.

Schenkers spontaneously defiant reaction to Hertzkas question betrays a

suspicion that Universal Edition is working not on his behalf but against his interests. This feeling will grow over the coming year: for example, in a draft response of December , , Schenker speaks of the publishers very desire to
sabotage my work and the dissemination of the periodical (/).1
Schenkers discontents in the fall of can be subsumed under two broad
headings: () contractual matters, including authorial autonomy, a publishers
freedom to cut, edit, and impose limitations, what constitutes the realm of
music, and the publishers imprint; () administrative matters, including publication schedules, accounting, and marketing.

Authorial independence. An unsigned letter of September , , from Universal Editiona communication that marks a watershed in relations with
Schenkerannounces that issues , , and , which will appear by the end of the
year, must be limited to two gatherings each. This means that Schenkers study of
Brahmss Variations on a Theme of Handel, Op. , must be split between issues
and , and that both issues must be padded out with shorter, make-weight articles (OC /). Schenker is incensed:
. at the re-imposition of the two-gathering limitation, which he believes has
been superseded by the change to quarterly publication;
. at the reediting that Universal has been doing to his material, including adaptation of the Miscellaneahe remarks in an ill-tempered first-draft response
that is subsequently softened: U. E.s role is merely that of the publishing
house, and it is its damned duty to ask for my consent with all due courtesy.
But without my consent the publishing house may not lay a finger on the
content of the volume.;
. at the very idea of make-weight articles, and the disruption of his workflow
that having to break off and write such items would represent. He makes a

Contractual Matters

On contractual matters, the conflict between Schenker and Hertzka dates back to

December , , when the latter first hinted at concerns over the opening essay
of issue , The Mission of German Genius, then in galley proofs (OC /).
1
Suspicion of sabotage by Universal Edition goes back two years earlier. Schenkers diary for
Sept , , reports: [Wilhelm] Furtwngler . . . expresses the opinion that Hertzka is sabotaging
me, and says he is willing to make inquiries with Peters and Breitkopf & Hrtel in Leipzig; also Walter
Dahms alludes to it in . See Hellmut Federhofer, Heinrich Schenker, nach Tagebchern und Briefen
in der Oswald Jonas Memorial Collection (Hildesheim: Olms, ), pp. , .

vi

gener al preface
forceful claim for integrity: each issue, as a defined whole, conforms to a
plan, and I cannot allow anyone in a spirit of mischief-making to compromise this plan, and it is crucial to offer subscribers well-rounded issues, each
of which addresses a particular theme. (OC /, ca. Sept. , ).
In the same letter, Schenker sounds an ominous note that issue is getting
out of the publishing houses control, just as issue had done in the spring of
, and makes a fascinating allusion to the enforced rearrangement of paragraphs in the foreword to the second volume of Kontrapunkt (published by Cotta
of Stuttgart 2 in , but under an arrangement with Universal). Universals response (OC /, Sept. ) alleges the technical impossibility of producing more
than a total of six gatherings in the coming three months, but then reluctantly
offers Schenker a way out: he may publish issues and together as a double
issue of four gatherings, which will thus accommodate the Brahms essay intact.
Schenkers response of September (OC /) is grudging, and is cast in the
form of a countersuggestion that issues and be released simultaneously and
compensate for one another in size:

Tonwille Pamphlet Press

Vienna, I., Opera House
(Albert J. Gutmann & Co.)
Leipzig, Karlstrasse
(Friedrich Hofmeister & Co)
Schenkers legal mind compels him to offer proof: the directors of Gutmann (a
music retail and hire store located in the Opera House), Josef Simon and Emil
Hertzka, are also president and chairman respectively of the board of Universal
Edition, and because his own business has always been conducted with Universal, therefore Gutmann must be merely a front for Universal Edition (OC /
, /, / ).3
Schenker is convinced that, with the conversion of Tonwille to quarterly
publication, Hertzka has given his word to changing the imprint to Universal
Edition. Indeed, Schenker demonstrates, by textual analysis of his list of abbreviations on the verso side of the title page, that the latter imprint had actually appeared on the title page of issue ( IV/) only to be expunged at the last moment and replaced by Tonwille Press (Tonwille-Verlag: the word Flugbltter had
meanwhile disappeared from the title: OC /). Thus, the imprint Tonwille
Press appears on the covers and title pages of issues , /, and (IV/ ).4
However, Universal has an answer to Schenkers allegations (OC /, Sept.
, ):

I should like, however, to make clear right away that it has come only after
much soul-searching, because my conscience as author and artist balks at
striking bargains at the cost of my creativity. . . . This bargain constitutes
the last sacrifice that I will make in order to facilitate the appearance
within this year of the three outstanding volumes.
Universal, however, adheres to the double-issue plan, permitting up to sixtyfour pages. The material on hand fills only fifty-five pages, and Universal stresses
that it would prefer the volume to be visibly of double size (OC /). Schenker,
however, remains adamant that he will not provide make-weight items, and issue
/ is released with fifty-five pages plus a final, blank page, with an understanding that issue may proportionally exceed the two-gathering limit and extend
to between forty and forty-two pages (it ended up as forty-two pages exactly).

You have forgotten a conversation that we had before the [] summer

vacation. When we insisted on tying the change-over of Tonwille to being
the published work of U. E., along with full artistic freedom for you as editor, to a certain right of censorship when it comes to personal and national-political attacks, you made very clear that under these conditions
you would prefer Tonwille Press to continue to appear as the publisher.
And we took note of this at the time, and accordingly retained it as the
publishers imprint.

Publishers imprint. The choice of publishers imprint for Tonwille was long a
bone of contention. We gave evidence in the general preface to volume that, as
early as the fall of , the use of the fictitious Tonwille-Flugbltterverlag rather
than Universal Edition aroused Schenkers suspicions. The full imprint for issues had been:

Universal Edition had bought up Gutmann in ; see Federhofer, Heinrich Schenker, p. .

Two proof copies of the cover and title page of Tonwille have been kept with the correspondence from Universal Edition. On one, the journal is still described as a series of Flugbltter, published by Tonwille-Flugbltterverlag (OC /); on the other, it is indeed called a Vierteljahrschrift, published by Universal-Edition A.-G. Wien (OC /). A reprint of issue carries
the imprint Universal Edition / Vienna.
3
4

On Schenker and Cotta, see Federhofer, Heinrich Schenker, pp. , .

vii

gener al preface

Administrative Matters

Accounts and subscriptions. As we have seen earlier, Schenker was already questioning the accuracy or fairness of Universals accounting by April . Prophetic
of the eventual outcome, he comments on March , : I am ready to relinquish to you all the books of mine that are covered by royalty agreements for a
corresponding cash sum, just so as not always to have to be rapping the knuckles
of those who falsify their accounts.5 Something infuriates Schenker in the first
half-year accounts for , for he speaks of an (unspecified) utterly shameless
attempt in July, involving Universals commercial sharp practices in fulfilling
contracts or in accounting methods (OC /, draft, ?Sept. ).6 In a remark
primarily concerned with the integrity of Tonwille issues, he makes an aside that
reveals a certain amount of envy (OC /, ca. Sept. , ):

Administrative matters are closely interrelated with contractual ones. Schenker

comes to believe that Universal is deliberately dragging its feet on publication
schedules. As he later puts it: Despite my having delivered manuscripts and returned corrected proofs punctually, the volumes have appeared after many
months delay, with the result that despite his vigorous urgings over five years
ten rather than around sixty envisioned issuesthe maximum figure, admittedlyhave been published (OC /v, Nov./Dec. ; OC /v, Apr. [?],
). Thus, in Schenkers eyes, minimizing the number of published issues
formed just another part of the strategy that included hiding those that were
published behind the Tonwille Press imprint so as to minimize damage to Universals international reputation. To these can be added other concerns: his attorney points out (OC /, Mar. , ; bullet points added for clarity):

that Tonwille has not appeared (more correctly: has not been allowed to appear) openly under your masthead, but has been obliged by you to hide behind a publishing pseudonym. I should mention that it was not advertised in
Die Musik until after it had been in existence for 1/2 yearsand then only
once, never again;

that the conversion from occasional to quarterly publication of its issues went
completely unpublicized;

that not the slightest effort has been made to propagate the publication;

that the retail outlets are not supplied with copies;

that stocks are merely piled up in the distribution center and are consequently kept secret;

that not a single copy is to be seen in the display area of Gutmann (Opera
House);

that issues already published are declared there to be not yet in print;

that the employees of that firm never know where on the premises to look
forlet alone findissues that have been asked for;

that again and again orders insistently requested from other countries are
marked as having been fulfilled, so that repeated visits by the ordering customer to the relevant bookstore are inevitably fruitless;

that, while new releases by my client are admittedly occasionally advertised
in U. E. (e.g. in the issue of Die Musik this March), the works published [i.e.,
analyzed] in Tonwille . . . are denied advertisement, despite their being attributable in an ancillary sense to Beethoven (Fifth Symphony, Op.!).

If the publishing house, for example, is doing everything possible to increase the number of subscribers to Anbruch,7 then it must surely appreciate, aside from matters of planning and substance, how important it
must be to me to offer the subscribers whom I myself am recruiting without the slightest cooperation from the publishing house, indeed in the
face of its machinations, at least well-rounded issues.
However, once Schenker receives the accounts for the second half of on
February , , he excoriates Universals bookkeepers (OC /) and demands a thorough inquiry. At the same time, he writes to pupils and friends, soliciting their experiences in subscribing to Tonwille. His findings are that as many
as seventy subscriptions have been placed and paid for that do not appear in the
twice-yearly statements. These subscriptions were taken out and paid for on behalf of the music departments of the universities of Berlin, Bonn, Breslau, Frankfurt, Freiburg, Gttingen, Halle, Knigsberg, Leipzig, and Munich during or
early . Many other would-be subscribers or purchasers, he reports, have approached Gutmann or Universal, and have been ignored.
Draft letter to Universal Edition, quoted in Federhofer, Heinrich Schenker, pp. .
Schenker was philosophically antagonistic to the outlook that motivated all trade and commerce; see Federhofer, Heinrich Schenker, pp. .
7
Musikbltter des Anbruch was the house journal of Universal Edition from to , in which
had appeared offensive reviews of Schenkers Erluterungsausgaben and the Beethoven Moonlight
Sonata facsimile by Paul Bekker (a personal friend of Emil Hertzka) in the February and April issues
of , and a favorable two-part article about Schenkers work by Otto Vrieslander in the February
and March issues of . It was Hertzkas attempts to excise two outspoken passages from the latter
(one against Bekker, the other against Hermann Kretzschmar) (OC /: Jan. , ) that prompted
Schenker to call Hertzka unjust, partisan, and terrorizing (OC /: Feb. , ).
5
6

viii

gener al preface
of issue (quarterly issue ). I have no further binding obligations to U. E.,
other than the Tonwille contract, and anyone who cares to go through the
letters and events and see all that has happened will appreciate that I have
no further desire to work for the publishing house.

Amid the furor of charges and countercharges, a meeting takes place (after
much foot-dragging, says Schenker, and with the intermediacy of his brother
Moritz8) on March , , between Schenker, his wife, Jeannette, his attorney, and
Hugo Winter, chief of bookkeeping at Universal Edition. The subscription book
that Winter presents arouses immediate suspicion because many pages have been
torn out; Schenker dubs it pejoratively a Hilfsbuch [manual] and a Schmierbuch
[scribbling-book] and reports that the vast majority of the subscriptions already
paid in the year are absolutely nowhere recorded in the subscription book
andeven more remarkablycannot be particularized by U. E.! (OC /,
Apr. [?], ). Universal does eventually locate the seventy missing subscriptions,
claiming negligence on the part of an employee (OC /, /, June , ).

Hertzka at first merely reserves his response (OC /, Sept. , ).

Schenker declares, probably in the following month, With the release of the next
issue [i.e., IV/, Oct ], I shall sever all connection with U-E as collaborator
(OC /). On November , , Hertzka makes a proposal whereby
Schenker would continue to contribute issues of Tonwille for a new fixed payment, and then goes on to apply pressure (OC /):
The continuation of the series (Sammlung) appears to us desirable in the
interests of musical education. If the series were to be continued by another
publisher, that could happen under the title Tonwille only if the publisher
were to take over the publishing rights, stock, plates, etc. in their entirety.
We doubt if a publisher could nowadays be found who would declare himself ready to do that. . . . The discontinuance of Tonwille altogether would
surely be of great disadvantage for the ten already existing volumes, of
which large stocks are in hand. This discontinuance would, moreover, not
make a favorable impression as regards you personally as an editor.

Change of Publisher

t is hard to determine without a fuller survey of the publishers records whether

Schenkers rapidly proliferating complaints arise out of an accurate assessment of
incompetence and subversive action on Gutmanns or Universals part, or out of
paranoia on his part at an honorable commercial enterprise that he finds antipathetic. The strength of Schenkers feelings is unmistakable: For over years, it
has been like going through a hell [with Universal] (OC /, draft, May ,
). By historical circumstance, he had found himself working with a publisher
that was the embodiment of all that he detested: a cosmopolitan Jewish company
that promoted the work of French, Italian, Hungarian, and Czech composers
alongside Austrians and Germans, and advocated progressive styles in preference
to the Austro-German classical repertory that he so revered. The first deliberative
intention to break with the publisher may come in a draft of a letter of circa September , , to Universal (OC /):9

By this time, Schenkers pupil Otto Vrieslander, who had contributed a laudatory two-part article on Schenker to Musikbltter des Anbruch of February and
March , is acting as intermediary with the Munich publishing house Drei
Masken Verlag,10 for Schenker receives a letter dated November , , from
10
Drei Masken Verlag was established in exclusively to publish works related to the theater
(hence its name: three masks), especially that of Munich. It acquired an office in Berlin, and in the
s expanded its program ambitiously to include politics, music literature, memoirs, and expressionist literature; by then it was among the top percent of the publishing houses in Munich, a
significantly larger publishing center than Vienna.
Drei Masken Verlag produced high-quality autograph facsimiles, including two Beethoven piano
sonatas and orchestral works by Schubert and Wagner. Among music books, it issued the second edition of Guido Adlers Richard Wagner: Vorlesungen (); but most particularly, it had already published two volumes of the Mozart-Jahrbuch ( ) and four volumes of the Sammelbnde fr vergleichende Musikwissenschaft ( ). The latter, which involved complex setting and unusual
music examples in acoustics and ethnomusicological transcriptions, must have allowed the publisher
to feel they could handle the intricacies of Schenkers work. Indeed, there are similarities in the layout and formatting of Bartks Volksmusik der Rumnen von Maramures (vol. of the Sammelbnde),
and Das Meisterwerk in der Musik. With Schenkers agreement, Drei Masken Verlag modeled Meisterwerk directly on the Mozart-Jahrbuch (OC / and /).

I therefore have no alternative but to face the consequences and to propose that we dissolve our contract by mutual agreement after publication
8
Moritz (or Maurice, or Moses) Schenker, Heinrichs younger brother, born August , , who
became General Director of the Vienna branch of the Zentral-Europische Lnderbank, and who
oversaw Heinrichs financial affairs.
9
However, in his diary for September , , Schenker, regretting the failure of recent contacts
with Peters Edition, wrote: I shall extol the day on which I manage to do my last volume in Vienna
and can let my connection with U. E. go (Federhofer, Heinrich Schenker, p. ). Four years earlier still,
he must have said something disparaging, causing Hertzka to write When I get back, I will relieve
you of your pessimism about U. E. (OC /, Jan. , ).

ix

gener al preface
two representatives of that press, Alfred Einstein and a person who signed all correspondence Demblin,11 expressing cautious interest in working with him.
They ask Schenker to supply details of production costs, print-run, and sales of
the issues of Tonwille that have appeared so far and (so confirming Hertzkas
surmise) indicate that, if it proved impossible to acquire the whole of Tonwille,
they could contemplate only a new serial with a new title. The authors make it
abundantly plain that they want to have nothing to do with Hertzka and will
leave all negotiations to Schenker (OC /).
Schenker transmits the request for information to Universal only in a draft
dated December , in which he rebuffs Hertzkas fixed-payment proposal and
warnings of November , remarking: I have several publishing houses in mind,
and am confident that I shall succeed in finding a home for Tonwille (OC
/). Applying his own pressure, he offers to forego a claim for material
and moral damages against Universal if they will facilitate the release of all ten
existing issues to the future publisher. Hertzka replies curtly that he has no intention of surrendering the ten issues, and refuses to supply the requested information (OC /, Dec. , ).
A second letter from Drei Masken Verlag, of January , , assesses Tonwille as not a truly commercial venture, saying:

Masterwork in Music: A Yearbook), all material to be submitted by the July of

each year to allow publication by Christmas, Schenker to receive percent of the
retail price.12 By early June, Universal knows that Schenker has been successful in
procuring another publisher, and reacts accusatorily (OC /, June , ):
Your client, Dr. Schenker, has informed us today that he intends to continue the publication that began as Tonwille with another publisher under
another title. I would point out that in clause of the contractual letter of
April , [recte ] Dr. Schenker acknowledged his commitment not
to transfer the publication within the terms of Tonwille to any other publisher. If he did this, your client would be guilty of a breach of contract.

Legal Proceedings

But this is mere lawyerly bluster, for legal proceedings have been in the air for

some months. On September , , Universal suggests that Schenker nominate

a representative to conduct future discussions (OC /). Within a few weeks
he is communicating with an attorney, Dr. Theodor Baumgarten, who on February , ,13 prepares the first draft of a legal document (OC /) headed:
Dr. Heinrich Schenker, writer on music, in Vienna III. Keilgasse , via:
Re: Universal Edition & Co in Vienna I. Karlsplatz
Presentation of the facts in the case
To the police headquarters in Vienna (Police for Economic Affairs)

Tonwille appears in an edition of only copies. Because of that, we

have to concede to Director H. that with Tonwille he is making a sort of
gesture of tribute [Ehrengeschenk] to you that only he as publisher of your
great theoretical writings and your Bach and Beethoven editions can
make. (OC /)

This is an account of the pertinent events combined with allegations against Universal, and is annotated in both Baumgartens and Schenkers hands. On April ,
Schenker agitates that they have in the meanwhile significantly lost ground and
made themselves look weak, pliable, and vacillating, and urges repeatedly that
they now put their complaints before a judge (OC /). Baumgarten disagrees, but produces a second version of his document on April (OC /).
By chance, Hertzka writes on the same day to Baumgarten, remarking that
Universal has never yet had to take legal proceedings against any of the hundreds of its authors. He goes on, with a veiled threat (OC /):

(If this is to be trusted, it tells us that the print run for Tonwille was reduced by
Universal from two thousand copies, as specified in the contract for the first four
issues, to eight hundred by the end.) Einsteins and Demblins letter proposes,
instead, a yearly publication, equivalent to Tonwille IV/ , that is, roughly fifteen gatherings, with the title Das Meisterwerk in der Musikein Jahrbuch (The
11
Alfred Einstein (), distinguished German musicologist, editor of Riemanns Musiklexikon (ninth to eleventh editions, ), of the Zeitschrift fr Musikwissenschaft , and of
Kchels catalogue of Mozarts works (rd edition, ). He was the author of Geschichte der Musik
(), Heinrich Schtz (), Gluck (), and articles on Medieval, Renaissance, and Baroque
music. Settled in the United States in , he went on to write a biography of Mozart (), and his
two most celebrated studies, Music in the Romantic Era () and The Italian Madrigal ().
The other signatory may be that of August Demblin. A short book of his, Czernin und die SixtusAffaire, is listed on p. of the catalogue of publications by Drei Masken Verlag (OC /).

We utterly deplore the fact that your client has for several months now
adopted an attitude that is totally incomprehensible to us and diametriReported in Schenkers diary for January , : see Federhofer, Heinrich Schenker, p. .
For this date, see the two later drafts: OC /, and OC / (paragraph ).

12
13

gener al preface
cally opposite to normal conduct in the light of some years of dealings
between your client and the publishing house. . . . Your client seems bent
on being the first author with whom Universal Edition goes to court [on
grounds of accounting]. . . .
We are in complete agreement that the affair be submitted to the appropriate forum on the issues and legal points involved, and so as to bring
light to the darkness we will lay before this forum for purposes of characterization the interesting correspondence of your client.

Summer and early fall go by; then on October , with no response still from
Universal, Baumgarten begins work again on the legal document and warns Universal that his client is not afraid to go to court (OC /; OC /, Nov. ,
). It is evident, however, that he has set his sights on a face-to-face meeting
between Schenker and Hertzka, with the two attorneys in attendance, so as to engineer a financial settlement; thus he applies pressure by issuing on November
a further warning of legal proceedings (OC /). Nine days later, Baumgarten
invites Schenker to his office, bringing all the relevant documents so that he can
draft the formal complaint that is to go to court (OC /). Such a draft exists:
it uses much of the wording of the previous two drafts, updated, and more tightly
drafted in fifteen clauses. It covers the publishers imprint, neglect of contractual
obligations, accounting errors, and mishandling of subscriptions. Two passing
remarks seem to imply still that nothing has so far been submitted to the police
(OC / , Nov.Dec. , clauses and ):

But now Hertzkamagician that he ispulls a deal out of the hat that
changes the complexion of the whole affair: () a royalty of two thousand German
marks on the seven thousand unsold issues of Tonwille , to be paid to Schenker
in annual installments of four hundred marks over five years; () a half royalty on
any new printings; and () the deal being extensible to Schenkers other works.
From this point on, Schenker and Baumgarten develop a two-pronged attack,
arguing the settlement figure upward, while building the case for court proceedings. Thus Baumgarten broadens the royalty offer to cover Tonwille spinoffs
monographs, Erluterungsausgaben resulting from the analyses, and the republication of all ten issues as a single volume;14 he throws in an element of damages
from previous breaches of contract and mismanagement, and asks that the collected edition of the Beethoven piano sonatas and the Neue musikalische Theorien
und Phantasien be included (OC / ). A month later, he proposes a one-off
payment of twenty thousand Austrian shillings15 for Tonwille, or alternatively a
severance (Ablse) payment of fifty thousand shillings allowing Universal to continue to market the existing issues of Tonwille, with full rights for translations,
new editions, spinoffs, or collective publication, which Universal dismisses out of
hand (OC /, May , ).
After two months of foot-dragging by Universal Edition, Baumgarten reverts
to the legal route, writing to Schenker (OC /, July , ):

However much I [i.e., Schenker] now regret itthough, given my hardworking and peaceable nature, it is admittedly really not surprising
I have been persuaded by my attorney . . . to persevere with seeking an
amicable settlement.
My attorney, who is conciliatory to a fault, and who believed that by this
means all shifting differences can be reconciled, became convinced by [a
meeting on June , , with Universal Editions attorney, Gustav Scheu]
that he should abandon the position he has adopted persistently up to now.
In parallel, Baumgarten continues to urge a meeting: Hertzka is apprehensive
at the presence of the attorneys, but Baumgarten pleads with Scheu (OC /).
Finally, the meeting takes place on December with Schenker, Hertzka, and Winter (no mention of the attorneys), and a transcript survives in Jeannettes hand
(OC /), and agreements (Abmachungen) are discussed. Schenker comments that he will receive even less than was envisioned in the original contract.
The agreement, which seems not to survive, is finalized on December , and provides for four annual installments of , shillings for Tonwille (i.e., , shillings in total), plus ongoing annual installments of , shillings for the collected
edition of the Beethoven piano sonatas.
Payments are to be made to the Vienna branch of the Central European Regional
Bank, credited to the account of Moritz Schenker, the general director of the bank.
The first payment is made on December , (OC / ), and payments
continue in the last few days of each year following that (OC /, /, /

If, despite our already overly drawn-out correspondence with Universal

we arrive at no acceptable outcome, should we not go to an arbitration
court? It works faster than the state court [staatliche Gericht], and leads,
if intelligent and competent judges are chosen on both sides, to a satisfactory arrangement.
14
Universal never reprinted Der Tonwille as a single volume, but did reprint issues , , and
collectively as Jahrgang , Jahrgang , and Jahrgang
15
The exchange rate in March was German mark . Austrian shillings.

xi

gener al preface
). After , the Tonwille payments stop, while those for the Beethoven Sonatas continue (OC /, /). Significantly, a week after the agreement is
reached, Universal mails a packet to Baumgarten enclosing all the physical assets
of Tonwilleplates and matrices (OC/, Dec , ).
Whether or not legal proceedings were ever initiated we cannot be sure; all that
survives is a postal receipt (OC /) for the dispatch of a letter on December ,
, to the police administration (marked Polizei D[irekt]ion), to be forwarded
to the Pressepol[izei], that is, the department concerned with newspaper and
magazine libel actions, and to anything else related to published material. There
seems not to have been a response from the authorities. It may be that a legal complaint had previously been submitted, and that the letter in question withdrew it.
Thereafter, correspondence between Schenker and Universal continues without
break. Schenker regularly receives accounts and notifications as well as royalties, and
in time Hertzka resumes normal civilities. In Das Meisterwerk in der Musik
ceases with the third volume, and Universal once again becomes Schenkers publisher, producing his commentary on the Brahms manuscript concerning consecutive octaves and fifths (Johannes Brahms, Oktaven u. Quinten u.a., ) and the final
volume of the Neue musikalische Theorien und Phantasien, i.e. Der freie Satz ().

The second postscript is intriguing. In April , Schenker proposes to Universal that the Urlinie-Tafeln in Der Tonwille, that is, the large sheets of graphs at
the back of each issue, be published together as a single small volume, and marketed in that form. (Kalmus expresses some concern as to how the graphs, being
of different sizes, could be combined; and Hertzka is puzzled as to how they
could stand on their own, shorn of their accompanying text, and asks for further
guidance; see OC /, WSLB .) This proposed venture is consistent
with the concept of the stand-alone graphic analysis, first realized in published
form in the Fnf Urlinie-Tafeln of but already familiar to Schenkers circle of
students by early .17

Survey of Issues

In general, the sort of material that went into Der Tonwille did not change much
during its four-year history. Schenker was consistent in mixing analyses of
canonic works of the eighteenth and nineteenth centuries with a small number
of short essays on general themes; the Miscellanea (Vermischtes) at the end of
most issues comprises an assortment of political, cultural and musical polemics.
The principal achievements in issues are the completion of the long
study of Beethovens Fifth Symphony, and essays on two of Schenkers favorite
teaching pieces, Beethovens Appassionata Sonata and Brahmss Handel Variations, Op. .18 The analysis of the symphony is wide-ranging, and goes considerably beyond the explication of the graphs of the Urlinie, which give an outline
of the basic counterpoint and harmony; several features of Schenkers analysis of
the form are particularly noteworthy. The first is the indivisibility of the third
movement (the main scherzo section), alluded to by the Goethe epigraph It is as
if all the things were wedged into one another. By this, Schenker means that there
is no contrasting middle section, or contrasting key, and the B
chords in bars
must be understood as a transitional harmony, part of the descending fourth
in the bass, CB
A
G. (This is clarified at the start of the essay, with the aid of
more detailed, multi-layered graphic illustration.)

There remain two curious postscripts to this account of the relations between
Schenker and Universal Edition. The first must have struck Alfred Kalmus as
ironic. In November , the Free Student Brotherhood of Kiel requests from
Universal permission to reprint the opening essay of Tonwille, The Mission of
German Genius, indicating that Schenker had already given his consent. This is,
of course, the essay that had caused Hertzka so much offense, triggering the deteriorating relations between author and publisher. The statement of account notes
Mission to the Schleswig-Holstein Hochschulbltter, that is, the newspaper of
the conservatory in Kiel (OC /, /, /). Perhaps Schenker thought his
dream of educating a new generation of German youth had begun to be realized!16
16
The request was probably made at the suggestion of Reinhard Oppel, who was then a professor
of music theory at the conservatory.
Schenkers diary for January , , reports Hertzkas remarks about Mission: it is sacred,
magnificent, but protokolliert in New York: he would not dare to present it to foreign readers until it
had first been discussed at a meeting of the board of Universal Edition in New York. [Hertzka] suggests that the article be published separately, . . . then hits upon the idea of bringing a fictitious publishing house into existence . . . now you can write whatever you want. See Federhofer, Heinrich
Schenker, p. .

17
In a letter written in January to a former pupil, Felix-Eberhard von Cube, Schenker writes
of a Bild, that is, an analytical graph, of a Bach prelude that can speak even without an accompanying text. See William Drabkin, A Lesson in Analysis from Heinrich Schenker: The C Major Prelude from Bachs Well-Tempered Clavier, Book I, Music Analysis (), pp. .
18
The lesson books for the years and following, preserved in the Oster Collection (OC ),
show that many of Schenkers pupils studied one or the other of these works, or both, and that
Brahmss Op. also was used as a model for the composition of variations.

xii

gener al preface
voice-leading graph that appears early in the essay (called Urlinie-Satz, and distinct from the Urlinie-Tafel at the back of the issue). These two essays on largescale piano pieces include as much detailed commentary on performance as we
find anywhere in Schenkers writing: the remarks, which are designed to show
how difficult works may become more accessible to the performer once their
content is understoodindeed, an important role of analysis is to provide insight into how to perform a workusefully complement the more theoretical
approach taken in Schenkers fragmentary treatise on performance.19
The final issue of the series, like Tonwille , is made up entirely of short analyses; most of these are of popular piano pieces of moderate difficulty by composers whose keyboard music had not been discussed in earlier issues: two Mendelssohn songs without words, two pieces from Schumanns Scenes of Childhood,
and an impromptu and a moment musical by Schubert. The inclusion of Haydns
Emperor Hymn gives Schenker an opportunity to combine analysis with politics without having to invoke extramusical factors: he appeals to the Austrian nation that they should at least know what a treasure they possessed in their proud
past, so that they may forever be edified by it in a more troubled future. The lead
essay, on the opening chorus from St. Matthew Passion, is the second of a pair of
Tonwille analyses of excerpts from this work. Schenker knew the St. Matthew Passion intimately, and it remained close to his heart to the end of his life; it was, according to Jeanette Schenker, the subject of his dying words.20
Schenkers interest in the accuracy of transmitted musical texts continues to
play a prominent role in the later issues of Tonwille. In the lead essay in Tonwille ,
on Gretchen am Spinnrade, Op. , he undertakes a thorough comparison of the
text of Schuberts autograph score with that of the first edition. In the next issue,
he follows through the implications of a dispute over a single note in Beethovens
String Quartet Op. , which has taken place among musicians in the service of
the works dedicatee, Prince Nikolai Galitzin, at St. Petersburg in summer .
Especially here, but also in other discussions of textual problems, it is music
analysis, rather than philological study of the sources, that guides Schenker to
what he perceives as the correct reading of a passage; this is entirely consistent
with the stand he took in earlier essays, notably on Beethovens Sonata Op. ,
No. (see Tonwille , pp. /I, p. , and the general preface to the first volume

A second striking feature of the study of Beethovens Fifth is the merging of

the graphs of the Urlinie for the third and fourth movements into a single graph,
with a broad linear ascent (Anstieg) in the last fifty bars of the scherzo (prominently bracketed in the graph) leading directly to the start of the finale; the principal theme of this movement, too, is based on the same ascent. Characteristically,
he does not call the end of the scherzo a coda but a transition to the last movement, and again he provides a more detailed graph for clarification.
In the analysis of the finale, Schenker shows how the development is inseparable from the closing subject of the exposition and, especially, how purposefully
the motivic development clarifies the design of the coda and how this section
amounts to far more than a mere bombardment of VI cadences. Again, each
of these points is clarified by a multilayered graph.
Despite deteriorating relationships between publishing house and author, the
three installments of Beethovens Fifth Symphony were published in book form
in September . From the correspondence with Universal Edition we can see
that Schenker had seen the proofs of this reissue (WSLB ; OC /, ).
But, whereas the pagination is new, and the list of abbreviations at the beginning
of the volume has been changed to reflect the latest version of Schenkers bibliography, the text and music examples have been left untouched, so that numbering of the music examples begins from a Fig. no fewer than five times: in the
commentary on the first movement (analysis, remarks on the autograph, performance issues), in the discussion of the literature on the movement, and in the
commentary on each of the second, third, and fourth movements. The book format also shows plainly that the study of the symphony divides broadly into two
parts, one on the first movement, and one on the remaining three.
The Brahms essay in Tonwille /, Schenkers only published study of a work
by the composer he had earlier eulogized as the last master of the German music,
is likewise marked by a concern for showing how a long series of variations is
much more than a set of quasi-independent statements, like scenes in a panorama, but rather that each grows from the previous one. In summing up his analysis of the variations, Schenker invokes the genealogical successions recounted in
the early books of the Old Testament to argue that, in Brahmss variation technique, nothing . . . is too small, too insignificant, that it might not be called upon
for a new act of procreation. The essay on Beethovens Appassionata is unusual
in that motivic and voice-leading analysis are closely brought together: Schenker
identifies a neighbor-note figure as the Urlinie motive, then proceeds to parse
the first movement as a series of variations on this figure, as shown in a special

19
Schenkers notes for this treatise were published as The Art of Performance, ed. Heribert Esser,
trans. Irene Schreier Scott (Oxford: Oxford University Press, ).
20
See Hellmut Federhofer, Heinrich Schenker, p. .

xiii

gener al preface
of this translation). The short essay on Op. also enables Schenker to identify
with Beethoven, by showing how the composers reasoning is entirely consistent
with the principles of voice-leading he has developed. In the Beethoven letters
that are quoted in the Miscellanea immediately following, he takes this identification further, highlighting the theme of composer/theorist as misunderstood
genius.
The general idea that what is closer to a composerfor example, an autograph manuscript, or letteris more authentic also resonates in the second of
two Miscellanea entries on Beethovens Ninth Symphony, whose centenary was
celebrated in , the year in which the last issues of Der Tonwille appeared. The
apparent success of a tribute performance of the symphony by the Vienna Philharmonic conducted by Paul von Klenaua hundred years to the day after its
premiere, in the same theater and with the original scoring, and attempting to
replicate all the composers performance indicationsled Schenker to suggest
that his home town establish a festival of Viennese Classicism, along the lines of
Wagner at Bayreuth, as a way of preserving tradition. These ideas echo a sentiment expressed in a polemical text, written a few years before but suppressed
from an early issue of Der Tonwille, to the effect that rather little was being done
to celebrate the work of Beethoven, compared to the attention lavished upon
Wagner and Richard Strauss. (This text is published for the first time, in English
translation, in the Appendix to this volume.)
There is in general, however, a marked reduction in the polemic content of
each issue. Three sections of Miscellanea in Tonwille together take up a mere
nine pages, that is, less space than any of the Miscellanea in issues , , and , and
less than half the space occupied by the lead article of the series, The Mission of
German Genius. Moreover, the emphasis of the polemic is more cultural than
political, and is often related to the composers whose works are discussed in the
essays in the volume in which they appear: thus, for instance, a Schubert theme
dominates the Miscellanea of issue , which was headed by the Gretchen essay;
and the gloss on Beethovens letter on Op. is followed by further extracts of
Beethoven letters in issue . The Miscellanea of Tonwille / is largely devoted to
the theme of Beethovens Ninth Symphony in the works centencry year. For the
last issue of Tonwille, Schenker suppressed the Miscellanea column altogether,
perhaps intending to make good an earlier intention to keep an issue devoted
mainly to short didactic pieces free of polemic (see the general preface to the first
volume, p. and p. , note ).
A similar trend toward concision or suppression also may be discerned in the

surveys of the literature on the pieces Schenker analyzed in the later volumes of
Tonwille. Thus, for example, Tonwille had included about seventeen pages of
critique, in fine print, of the literature on the first movement of the Fifth Symphony from E. T. A. Hoffmann () to Paul Bekker (), which was as much
space as he had allotted to the entire musical commentary; but the literature on
the remaining three movements is dispatched in just four pages, compared to
forty-five of musical commentary spread across issues . And while each of the
major essays on Classical piano sonatas in issues had included substantial
discussions of the secondary literature, amounting in total to about one quarter
of the text, the literature on the large-scale piano works discussed in issues and
/ is summarily dispatched: on Beethovens Appassionata Sonata, pens wrote,
not ears, while critiques of Brahmss Handel Variations are scanty and insignificant.
Tonwille / marks the entry into Schenkers writings of the Elucidations
(Erluterungen), a three-page prcis of his theory as it had developed during the
early s, in which many of the terms we associate with his theories are concisely explained and illustrated: Urlinie, degree (Stufe), prolongation, the transformation of dissonant tones into consonances, unfolding, ascending and descending register transfer, background and foreground. It was reprinted in Tonwille
and, with minor changes, in the first two volumes of Das Meisterwerk in der
Musik. The title refers back to the earlier Erluterungsausgabe project, while the
accompanying note (from Freier Satz) points toward the formulation of the
theories in the final part of the Neue musikalische Theorien und Phantasien, although this was not to be published for another ten years.

The Early Reception of Tonwille

ive reviews of Der Tonwille, including short notices, are preserved in the
Schenker scrapbook catalogued as File of the Oster Collection. The first is by
Wilhelm Altmann, director of the music division of the Prussian State Library; it
describes Schenker as a profound and extremely knowledgeable Viennese music
researcher and underscores the importance of the Erluterungsausgaben (for
which Altmann had supplied Schenker with reproductions of original manuscripts). But although he mentions the Urlinie and the works to which this concept is applied in the first issue of the new journal, the review is dominated by
quotations from the lead article, an essay that he believes is likely to attract interest not only in Germany but also far beyond its borders (OC /).

xiv

gener al preface
Bernhard Paumgartner, like Schenker, studied law at the University of Vienna
(he received a doctorate in ), but made a career in music: initially as a conductor and, in and , as director of the Mozarteum in Salzburg.
His review of Tonwille for the monthly magazine Der Heimgarten () was one
of a series of enthusiastic notices he wrote about Schenker, with whose work he
came into contact. He had issue to hand when he wrote his review, and describes
how a more literary section . . . leads from the purely musical into matters of
more general concern to humanity, in which the profoundness of German music
finds words as they have seldom been written. Paumgartner acknowledges the
polemical tone of Schenkers essays and is aware that his tactlessness has made him
many enemies, but he acknowledges his absolute sincerity and praises him as a
musician of great and pure heart with an extraordinary critical faculty (OC /).
A brief, descriptive notice of Tonwille , unsigned, appeared in the Halbmonatsschrift fr Schulmusikpflege [Bimonthly Journal for School Music Practice]
for December , (OC /); the other reviews were published in in Die
Musik, a journal to which Schenker had himself contributed an article two years
earlier.21 The first and longer of the two reviews, by Max Broesike-Schoen, appeared in January and is concerned entirely with the ideological propositions set forth by Schenker, that is, that there are immutable laws of music that
are to be found in the works of the classics (of whom Bach, Mozart, and Beethoven are representatives), invalidating all efforts that do not conform to these laws:
Broesike-Schoen mentions two repertories, Netherlands polyphony and Wagnerian music-drama, that can be invoked as counterexamples to Schenkers metaphysics and dogma formation; and although he concedes that thorough and
penetrating analyses show that there is a positive side to Schenkers work, he regards the theorist less as a prophet and canonic figure than as a cranky pedant
who will be remembered as one of the curiosities of music history. BroesickeSchoens review was followed six months later by a shorter notice of Tonwille
and by the Cologne-based critic Willi Kahl, who is quick to applaud the extremely conscientious analyses of Beethovens Fifth Symphony, the Bach preludes, and other works, and to single out for special praise a valuable text-critical
investigation of Gretchen am Spinnrade, but laments the fact that not only does
Schenkers temper continue to flare up in the Miscellanea but that he has initi-

ated a trend for this kind of writing, as can be seen from Otto Vrieslanders recent book on C. P. E. Bach.22 Kahl quotes a passage from A Bach Prelude from
the Miscellanea from Tonwille as a particularly grotesque example of Schenkers furor teutonicus.
Although the number of actual reviews of Der Tonwille is small, the early
s saw the appearance of several newspaper and journal articles mentioning
the achievements in Tonwille in the context of Schenkers overall achievements as
a scholar and theorist. A few of these are highly partisan pieces by Vrieslander,
and several more are by the writer and critic Walter Dahms, who studied for a
time with Vrieslander and in whom Schenkers stand against contemporary
music and music journalism struck a sympathetic chord. (Some of Dahmss expressions, such as the general decline of musical production since Brahms and
the cries, falsifications and slander of journalists who try to drown out the voice
of truth, might easily be mistaken for Schenkers own; see OC /.) Ludwig
Moormans brief survey of Schenkers work as editor and theorist, written for Die
Musikantengilde in , is the only writing preserved in the scrapbook that highlights a specific analytical idea in Der Tonwillethe growth of the entire first
movement of Mozarts Sonata in A minor, K. , from its opening grace-note
to illustrate the theorists insight into music (OC /).

21
This is the analysis of the Prelude in C minor from Bachs Well-Tempered Clavier, book , which
was reprinted at the end of Das Organische der Fuge, the essay on its companion fugue, in Meisterwerk .

22
This monograph had been published in and would have been well known to Kahl, who
completed his Habilitationschrift, a history of eighteenth-century keyboard music, for the University
of Cologne that year.

A Note on the Translation

n account of our editorial procedures is given in the general preface to the first
volume of this translation. As before, we have consulted materials in the Oster
Collection of the New York Public Library; we have indicated changes made by
Schenker in his personal copies of Tonwille, for instance, to reflect the developments in his notation of musical structure, or to add a further thought about the
composer or work under discussion, although none of these changes was adopted
in the reissue of Tonwille in three annual volumes, or when the three installments
of Beethovens Fifth Symphony were reprinted as a separate monograph.
The Miscellanea in the later issues of Der Tonwille seem not to have been subjected to Universal Editions censorship at proof stage, as had been the case with
that published in issue ; their relative brevity and less caustic tone must have

xv

gener al preface

been cause for relief, if not celebration, on Hertzkas part. There remained, however, the essay on music criticism attacking Paul Bekker, which Hertzka had rejected before Tonwille went to press and which was not published in any of the
subsequent issues. This polemic, together with the canceled page proofs from the
Erluterungsausgabe of Op. on which it is partly based, survives in Schenkers
folder of censored material, which Schenker agreed to cut only under pressure,
and is reproduced as the appendix to this volume. For this essay, the numbers in
enclosed in curly brackets, from {} to {}, refer to the sheets of paper in the
Oster Collection, file , on which the full text is preserved.
The text of the Beethoven letters quoted in Tonwille and has been checked
against the new collected edition of Beethovens correspondence, Beethoven:
Briefwechsel: Gesamtausgabe, edited by Sieghard Brandenburg and published in
the late s. The new edition is in all respects vastly superior to Alfred Kalischers
five-volume edition of the letters, which was Schenkers source, and to all other
editions that were current at the time. It is of special importance for the essay on
the Quartet Op. , not only because it corrects the poor reading offered by
Kalischer (who was not even able to give the correct identification of the piece
under discussion) but also because it provides additional background information about Beethovens letter and some of its earlier drafts.

This volume marks the completion, in English translation, not only of Der Ton-

wille but of all of Schenkers major writings of the s, a project that has occupied us for a number of years; we are grateful for the help and encouragement
of scholars, editors, and production teams throughout this period, and also for
the imagination and the efforts of several teams of theorists and analysts who are
gifted in the art of translation. We wish to record our special thanks here to
Kimberly Robinson and Robert Milks of Oxford University Press for the care they
have exercised in editing and producing this book, and to Barry Cooper (University of Manchester), K. M. Knittel (University of Texas at Austin), Michael Musgrave (Goldsmiths College, University of London), and Robert Pascall (University of Bangor) for their help in resolving specific questions.
As ever, the translators and editors are indebted to Andrea Reiter of the University of Southampton, not only for help in clarifying several passages in the text,
but more particularly for her exceptional insight into Schenkers use of language.
Ian Bent, Columbia University
William Drabkin, University of Southampton

xvi

Contents

German Words, Phrases, Technical Terms, and Abbreviations Used

in the Music Examples xix
Bibliographical Abbreviations

xxi

Tonwille 8 9
Brahmss Variations and Fugue on a Theme by Handel, Op.
Genuine and Sham Effects
Elucidations
Miscellanea

Tonwille 10
Tonwille 6
Schuberts Gretchen am Spinnrade: Latest Results of a Manuscript
Study
Beethovens Fifth Symphony (conclusion)
True Performance
Miscellanea

Tonwille 7
Beethovens Sonata in F Minor, Op.
The Recitative Erbarm es Gott from Bachs St. Matthew Passion
Beethoven on His Quartet Op.
Miscellanea

The Opening Chorus (First Chorale Fantasy) from Bachs

St. Matthew Passion
Haydn: Austrian National Anthem
Schuberts Impromptu, D. (Op. ), No.
Schuberts Moment musical in F Minor, D. (Op. ), No.
Mendelssohns Venetian Gondola Song, Op. , No.
Mendelssohns Song without Words, Op. , No.
Schumanns Scenes of Childhood, Op. , No. Of Foreign Lands
and Peoples
Schumanns Scenes of Childhood, Op. , No. Trumerei

Appendix
Music Criticism
Index

This page intentionally left blank

German Words, Phrases, Technical Terms, and

Abbreviations Used in the Music Examples

This is an alphabetized list of all the analytical labels and explanatory text

Ges
Grundton
H
I, II
Hherlegung
Hrn [Horn, Hrner]
linke Hand wie bei a)
moll
Md. [Modulation]

found in the music examples and Urlinie graphs of Tonwille . In general, an

abbreviation is followed by the full form of the word in question, either after a
comma (if that word also appears somewhere in the examples), or in brackets (if
it does not). To obtain a translation of a short phrase, it is sometimes necessary
to look up the component words or abbreviations: thus Des dur D
major.
As
B, b
Br. [Bratsche(n)]
Brechung des G-Klanges
C.B.
Ch.[Chor]
Des
Df. [Durchfhrung]
Dg. [Durchgang]
Dg. mit -Auswechslung
dur
durch bergreifen von Stimmen
eine Oktave tiefer
eine Stimme
Fag. [Fagott(e)]
Fallen durch Hherlegung und
bergreifen
Fis
Fl. [Flte(n)]
fr
Ged [Gedanke]

A
(note)
B
(note)
viola(s)
arpeggiation of the G (major) chord
contrabass(es)
choir
D
(note)
development section (in sonata form)
passing note, transitional harmony
passing tone [progression] with
alternating fifths and sixths
major
by the reaching-over of voices
one octave lower
one voice
bassoon(s)
descent by means of ascending register
transfer and reaching-over

F (note)
flute(s)
for
subject, group (in sonata form)

N. S. [Nachsatz]
Nbn. [Nebennote]
Nbn.-Hm. [Nebennotenharmonie]
nicht
Ob.
Oberquintteiler
Oberquint als Teiler
oder
Orch.
Pos. [Posaune(n)]
Quart
Quartzug
Quint
Quintzug
Quintzge als springende
Durchgnge
Rp [Reprise]
Rckung

xix

G
(note)
root
B (note)
first, second
ascending register transfer
horn(s)
left hand as in [Fig.] a)
minor
modulation (in sonata-form
exposition)
consequent phrase
neighbor note
neighbor-note harmony
not
oboe
divider at the upper fifth
upper fifth as divider
or
orchestra
trombone(s)
fourth
fourth-progression
fifth
fifth-progression
fifth-progressions as leaping
passing tones
reprise, recapitulation (in sonata
form)
rhythmic shift

ger man words, phr ases, te chnical ter ms, and abbrev iations used in the music examples
Sept
so
T. [Takt]
Teiler der Oberquint
Terz
Terzsatz
transponiert
u.s.w. [und so weiter]
und
Unterquintteiler
Unterquint als Teiler
Urlinie-Motiv

seventh
thus
bar, bar number
divider at the upper fifth
third
counterpoint in (parallel) thirds
transposed
and so forth
and
divider at the lower fifth
lower fifth as divider
motive of the Urlinie

Urlinie-Satz
V. S. [Vordersatz]
Vcl.
Verwandlung von dissonanten
Durchgngen in konsonante
Vl. [Violine(n)]
Wdhg. [Wiederholung]
wie
wirkliche Lage der Oberstimme
wirkliche Lage der Unterstimme
zwei St[immen]

xx

contrapuntal setting of the Urlinie

antecedent phrase
violoncello(s)
transformation of dissonant passing
tones into consonant ones
violin(s)
repetition
as, the same as
actual register of the upper voice
actual register of the lower voice
two voices

Bibliographical Abbreviations

Unpublished Materials
OC

WSLB

Beethovens neunte Sinfonie Beethovens neunte Sinfonie (Vienna: Universal

Edition, )
English translation: Beethovens Ninth Symphony,
trans. John Rothgeb (New Haven, CT:
Yale University Press, )
Erluterungsausgabe
Beethoven, Die letzten fnf Sonaten: kritische
Ausgabe mit Einfhrung und Erluterung
(Vienna: Universal Edition)
of Op.
Sonate A dur Opus ()
of Op.
Sonate E dur Opus ()
of Op.
Sonate As dur Opus ()
of Op.
Sonate C moll Opus ()
(Opus not completed or published)
abbreviated second edition: Beethoven, Die letzten
Sonaten: kritische Einfhrung und Erluterung ed.
Oswald Jonas (Vienna: Universal Edition, )
Tonwille , etc.
Der Tonwille, ten issues (Vienna: TonwilleFlugbltterverlag [ Universal Edition], )
[Tonwille / is a double issue]
English translation: the present publication (in two
volumes)
Meisterwerk i, ii, iii
Das Meisterwerk in der Musik, three vols. (Munich:
Drei Masken Verlag, , , )
English translation: The Masterwork in Music,
trans. Ian Bent, Alfred Clayton, William Drabkin,
Richard Kramer, Derrick Puffett, John Rothgeb,
and Hedi Siegel, ed. William Drabkin, three vols.
(Cambridge: Cambridge University Press, ,
, )

The Oster Collection: Papers of Heinrich Schenker,

New York Public Library (New York)
A Finding List of this collection, compiled by
Robert Kosovsky, is dated May , , and issued
by the New York Public Library
Wiener Stadt- und Landesbibliothek [Municipal
and Provincial Library of Vienna]: collection of
letters from Schenker to Universal Edition,
Vienna (on loan from Universal Edition)

Schenkers Published Writings

Harmonielehre

Ornamentik

Kontrapunkt i, ii

Harmonielehre Neue musikalische Theorien und

Phantasien, part (Stuttgart: Cotta, )
Abbreviated English translation: Harmony, ed.
Oswald Jonas and trans. Elizabeth Mann Borgese
(Chicago: University of Chicago Press, )
Ein Beitrag zur Ornamentik, rev. nd ed. (Vienna:
Universal Edition, )
English translation: A Contribution to the Study
of Ornamentation, trans. Hedi Siegel, Music
Forum (), pp. .
Kontrapunkt Neue musikalische Theorien und
Phantasien, part ; vol. i (Stuttgart: Cotta, ),
vol. ii (Vienna: Universal Edition, )
English translation: Counterpoint, trans. John
Rothgeb and Jrgen Thym, ed. John Rothgeb,
vols. (New York: Schirmer, )

xxi

biblio g r aphical abbrev iations

Der freie Satz

Der freie Satz Neue musikalische Theorien und

Phantasien, part (Vienna: Universal Edition, ,
rev. /)
English translation: Free Composition (Der freie
Satz), ed. and trans. Ernst Oster (New York:
Longman, )

work is extensively documented in the Oster Collection. Schenker worked out a

plan for it during World War I, and had completed an initial draft by August ,
to which additions and emendations were made.1 Sometimes he refers to sections
of Freier Satz that can be specifically identified in the early draft; other references show that his conception of this work was changing, although the final version, published posthumously with the title Der freie Satz as the third and final
part of the Neue musikalische Theorien und Phantasien, was to turn out very different. We have attempted to trace these references back to the Freier Satz as
preserved in the Oster Collection, and ahead to the published form of this work.
Our references to Der freie Satz use paragraph numbers (), which are the same
in German and English editions.
For the Erluterungsausgaben of the late Beethoven sonatas, the italicized
page numbers refer not to an English translation (at present, none has been published) but to Jonass revised German edition.
Unless otherwise stated, references to material in the Oster Collection are by
file and item number. Thus, for example, OC / refers to file , item , in
the collection.

To facilitate reference to Schenkers original writings and the current standard

translations, a system of double page references is used in this translation. Thus,
for example,
(see Kontrapunkt i, pp. ff/pp. )
indicates that the Schenker original referred the reader to pp. ff of the first volume of his Kontrapunkt, and that we have in addition supplied the corresponding
page numbers, , in Rothgeb and Thyms English translation.
Schenkers references to II3 are to a projected third volume of Kontrapunkt;
which used the title Freier Satz throughout the publication of Der Tonwille; this

1
For an introduction to the genesis of the final volume of the Neue musikalische Theorien und
Phantasien, see Hedi Siegel, When Freier Satz Was Part of Kontrapunkt: a Preliminary Report,
Schenker Studies , ed. C. Schachter and H. Siegel (Cambridge: Cambridge University Press, ),
pp. .

xxii

Tonwille

This page intentionally left blank

Schuberts Gretchen am Spinnrade: Latest Results of a Manuscript Study

Franz Schubert: Gretchen am Spinnrade. Neue Ergebnisse einer Handschrift-Studie {Tonwille , pp. }
t r a n s l at e d b y i a n b e n t

first one thinks it identical with the familiar text, only to be astonished on closer
comparison to find oneself bombarded with a wealth of new features until one is
finally driven to admit to never having been more surprised in ones life than at
this manuscript.
In this publication the manuscript is preceded by a reprint of the first edition,
which was issued as Op. on April , , almost six years after the composition of the song (the manuscript is dated October , , in Schuberts hand in
the top-right corner).3 Since the first edition provides the basis for all editions
currently in use, I need present only the artistic facts of the autograph here in
order to convey to the reader the differences between the two versions.
The accompanying chart (see p. ) shows the differences between the dynamic markings.
{} It is not even necessary to examine the form, modulations, or harmonic
progressions to see how greatly Schuberts handling of dynamics surpasses those
of the first edition in truth to art and nature. (On the treatment of prosody in
particular, see Tonwille , p. /I, p. .4) True, the real profundity of his markings
can be appreciated only by someone who hears them in relation to the formal
and harmonic events. Such a person will immediately agree with me thatto anticipate my resultsthe first edition cannot possibly have been edited by Schubert
himself, but by someone devoid of musical genius whose aim was merely to gratify the cravings of the mediocre people close to him, and to curry favor with
someone who prefers a musical work gripped by the chaotic fevers of human
mediocrity to a work of art that resolves on to a divinely pure harmony of artistic expectations and fulfillments.

For whom has this musical notation not been an old friend since childhood!
On closer examination, however, we find significant differences when compared
with the familiar reading, and can only wonder that it has taken us any time at all
to notice them. The new reading is taken from Schuberts autograph manuscript.
Previously owned by Nikolaus Dumba,1 and now the property of the Vienna City
Council, it has been published for the first time by the noted Schubert scholar O. E.
Deutsch in volume of Musikalische Seltenheiten (Vienna: Universal Edition).2
The remainder of the text has the same effect as the beginning of the song. At
1
Nikolaus Dumba (), Viennese industrialist in cotton and spinning, major patron of
the arts, especially music and sculpture. Dumba, who studied music himself, and sang tenor, took a
special interest in Schubert, and assembled the great majority of the latters autograph manuscripts
in his private collection, later donating his manuscripts to the Vienna Municipal Library.
2
[S.] [Musikalische Seltenheiten/Wiener Liebhaberdrucke] vol. contains: Franz Schuberts fnf
erste Lieder: Am Erlaf-See, Widerschein, Die Forelle, Erlknig, und Gretchen am Spinnrade, nach
den fnf Erstdrucken und einer Handschrift [in Faksimile-Reproduktion von Otto Erich Deutsch
(Vienna and New York: Universal-Edition, ); Gretchen am Spinnrade: pp. first printed
edition, Cappi and Diabelli, Vienna; pp. autograph manuscript. It was as vol. of this series that
Schenker published the facsimile of Beethovens Moonlight Sonata, Op. , No. in , the introduction to which was hostilely reviewed by the Frankfurt-based music critic, Paul Bekker: see infra,
General Preface to vol. .]

The song was in fact published the following year, April , .

I.e. Franz Schubert: Ihr Bild (Heine), in which Schenker argues that by favoring word-stress
over natural prosody, modern composers merely reduce genuine poetry to prose.
3
4

tonw i l l e 6
while on the other handwhat a fortunate stroke of synthesis!he links it motivically to the line that follows it on each occasion.

Hence the markings in the first line relate exclusively to the two word-repetitions, whose special emphasis they reinforce through their own means. (Schubert
himself is responsible for the first word-repetition, Ich finde.) The first edition,
contrary to that, places the cresc. too early, clearly unrelated to the first repetition,5 and it evidently leaves the second one in the lurch, since it deletes the cresc.
in bar , as a result of which, incidentally, the > sign in bar no longer makes
sense.
In terms of dynamics, Schubert conflates the second and third lines: they rise
by way of a cresc. (bar )the point at which the modulation occursto mf
(bar ) and by a second cresc. (bar ) to f (bar ). Here again the harshly dissonant so-called chord of the major seventh over an F appears; its sixth, d
, is later
transformed so beautifully into the leading tone, c .

All the remaining dynamic markings reinforce chromatic tones (bars , ), as

well as the prosody6 (bars , , ). The first edition pays no heed to modulation or cadence, allows the mf to crop up arbitrarily, and proceeds via a cresc. to
the f in bar , but, contrary to harmonic logic, merely for the sake of a single
word-emphasis (armer [poor]), as a result of which the markings that follow,
cresc. >, also lose all consistency and clarity.
{} In the fifth line,7 Schubert is clearly portraying urgency, breathlessness
(nach ihm . . . , nach ihm . . . [in search of him . . .]) with the two cresc. p in

First, I should point out that each line of the chart corresponds to one stanza
of the poem.
Schubert reserves a certain independence for the first line of the song, both
with respect to the poet and in musical terms, for it must not only recur several
times and introduce the stanza-groups but also bring the whole song to a close,

Tonwiederholung; but word repetition (Wortwiederholung) is surely intended.

I.e. the alliteration and assonance in mir vergllt . . . mir verrckt . . . mir zerstckt.
7
In der vierten Zeile. In reckoning the lines of the song, Schenker has omitted the fourth stanza
of Goethes poem, which is an exact repetition of the first, but he does include the eighth, which is
6

Schuberts Gretchen am Spinnrade: Latest Results of a Manuscript Study

bars and , and in so doing paves the way so perfectly for the big climax
that follows. Thus we may say that all three lines will be bound together in a single
dynamic unfolding. The first edition, however, is oblivious of these indications.
With the composition of the sixth and seventh lines, Schubert accomplishes
a veritable miracle. What a gift the seventeen-year-old Schubert has brought the
poet Goethe here! Fausts features come to Gretchens mind, sein hoher Gang,
sein edle Gestalt . . . [his lofty gait, his noble figure . . .]; she goes into ecstasies,
becomes ever more entranced and confused, her pulse grows faint, her speech falters, and the wheel, too, begins to falter, untilfinallyit stops altogether and
she must first recover her senses if she is to set it turning again. With this bold
tone-painting, audaciously interjected, he not only assists the poet, for whom
such a thing is not available within his lyric poem, but in so doing also procures
advantages of a higher order for the musical composition itself: he protects it
from the monotony that might have crept in at this point had he merely replicated the return modulation of bars ; what is more, he succeeds in creating
a two-part division for the whole that is more effective than the three-part division inherently suggested by the poem. (Admittedly, he was able to venture the
two-part division only because he had a good sense of equilibrium and extended
the third group on his own initiative, by the repetition of lines.)8
Now to Schuberts dynamic markings in this section. In accordance with the
two VI progressions in bars and , he locates the f precisely at the onset
of the third VI progression, that is, bar . In bar , thus just where the first
augmentation of the spinning-wheel motive9 occurs to signify that Gretchens
foot is about to stop, precisely at this point he combines the first sf with the B
/b

octave in the bass. And then again, from bar on, just where the double augmentation of the spinning-wheel motive commences, signifying the mounting
bewilderment that Gretchen has been undergoingthe a1 is lost between b
1 and
g 1 in the midst of so much agitation, but in return the a1 literally swoons in bar
before the b
1 that begins the next motive repetition (as an anticipation)

precisely from this point on he transfers the sf up into the right hand and depicts
her swooning in bar in such a way as to bring the last b
1 in clearly beyond the
range of the > sign.
The first edition, however, reflects none of this. It displays not the slightest
appreciation for the rightness of a relationship between dynamics and VI progression or motive, and so on that is felt and dictated by art. Operating only on the
most superficial level of prose, it goes after emphasizing individual words, thumping, so to speak, the principal words Rede (speech) in bar it must have
thought Schuberts f on seiner (his) an abomination!and Hndedruck
(handclasp) in bar . It molds the first motive augmentation with cresc.sf
accel.ff, and from bar [to bar ] places the sf roughly in the middle [i.e., between the two staves of the piano part] without any bearing upon the motive.
In the final group of lines, Schubert presses forward to the f in bar , then by
way of the two > 10 in bars and 11 (upsurges merely reflecting neighbornotes) to the ff in bar here at last we get the long-awaited dominant, crucial
to the form!which, from this point on, continually reinforced by sf 12 accents
bar after bar, applies also to the lines added by the composer. But once again, the
first edition has it differently: It introduces the ff as early as bar at any rate,
kssen [to kiss] matters more to it than the dominant in bar ; on the other
hand, it is quite unable to follow aurally the synthesis whereby Schubert, true to
art, makes the two rhyme.
If we now put together all the dynamic fluctuations in Schuberts autograph,
we see that there is one culminating f in each of the first two groups,13 and in the
last group {} one concerted climax rising through f to ff, which at the same time
forms the peak of the songs entire unfolding.
Moreover, all the other divergences in the first edition betray a total failure to
recognize purely artistic necessities:
The autograph manuscripts reading of the first six measures set forth the
motto. The first repetition, bars , still maintains exactly the same reading.
Not until the second repetition, bars , is there a variant, but one that we

also an exact repetition of the first. To make it easier to use the chart accompanying this essay, I have
changed Schenkers numbering of the lines, so that it conforms to the stanzas of the poem.
8
I.e. the repetitions of lines in Goethes tenth stanza, including a modification to the first line:
Und kssen ihn becomes O knntich ihn kssen. At the end, Schubert reduces the final stanza to
its first two lines.
9
Schenker is presumably referring to the motive b
1 a1 g 1 a1 in dotted quarter notes, which he
derives from the right-hand figuration, and which is a rhythmic augmentation of part of the original
sixteenth-note figure, (f 1 a1 )f 1 e1 d1 e1. This is illustrated in the music example given in the chart.

>

10
Schenker fails to note, or record in his table, that the autograph manuscript has both a
and a cres sign in the first part of the bar.
11
T. und : in counting the bars in this song, Schenker used the same number, , for both
bars and ; all bar references from this point onward have been tacitly corrected here.
12
Schuberts markings are all fz in bars , rather than sf, as elsewhere.
13
The three groups of lines are based upon the repetition of the opening stanza, and correspond
to stanzas , , and .

tonw i l l e 6
recognize as the version familiar to us hitherto. And the ending actually reads as
follows:

Here are some additional features of the Dumba manuscript:

In bars and , as also in all later repetitions, bars , , the voiceleading of the chords in the left hand is much more careful:

{} In bar , in the left hand:

avoids a succession of bare octaves (being used for reinforcement, they would not
have qualified as consecutive octaves).
In bar , in the vocal part:

I surmise that Schubert expressly wanted to avoid a symmetry between bars

and in the first of these three versions [see Fig. ]. On the one hand, two
leaps of a fourth in quick succession better suit the word painting (two spins of
the wheel); on the other hand, the dotted half notes in bars lead more effectively into the dotted half notes in the bars that follow. The middle version, bars
,14 provides a transition to the closing version: the leaps of a fourth are
placed further apart and effect a symmetry between bars and . Finally, at the end of the song, Schubert repeats this selfsame symmetrical arrangement (see Figure ), adding to itas if it were only a modelthe replica15 of the
more compact symmetry, bars (see the brackets).16 The first edition has
discarded all these differences in favor of the middle version.

obeys the beautiful and rightful rule whereby an ornament belongs only the second time around.
In bar , in the accompaniment:

14
Die mittlere Fassung, T. : although he is clearly referring to the return of the opening after
the climax of the song, Schenker erroneously gives the bar numbers that apply to the previous occurrence of this music. The other bar numbers in this sentence have been tacitly corrected.
15
VorbildNachbild: Schenker uses these terms almost as equivalent to antecedentconsequent
(VordersatzNachsatz). Yet, Nachbild der kleineren Symmetrie evidently refers to the closer symmetry
in bars , discussed at the beginning of this paragraph, of which bars are now an echo.
Thus what initially came second (the wider symmetry of bars ) is now the model, and what
came first is now the replica.
16
[S.] It would be helpful to inspect the fragment of a second autograph manuscript containing
just the first sixteen bars, which survives in the Prussian State Library in Berlin [now the Staatsbibliothek zu Berlin, Preussischer Kulturbesitz]. In the supplementary volume to his Schubert edition
(Edition Peters ), Max Friedlnder remarks: The Berlin manuscript, which probably served as
the exemplar for the printed edition, has no variants. But since Friedlnders critical commentary is
by no means completehe never once mentions the different versions that are so striking in the repetitions of the first lineI do not know what value to place on these words. I am thus of the opinion
that if the opening measures of the fragment correspond to those in the Dumba manuscript, that
would constitute compelling evidence of Schuberts artistic intentions.

prepares for the register that follows.

In bar , in the left hand:

is in deliberate contrast to bar .

[In fact the autograph fragment, which is part of a collection of Goethe settings, corresponds in
all particulars with the first edition; the editor of the first volume of songs for the new Schubert edition, Walther Drr, was likewise of the opinion that Schuberts second autograph served as the engravers copy (Stichvorlage) of the first printed edition and censured Schenker for being so vehemently
opposed to the possibility that Schubert had a hand in the changes (Franz Schubert: Neue Ausgabe
smtlicher Werke, ser. IV, vol. a [Kassel: Brenreiter, ], pp. xvii, xxxxi). See also note here.]

Schuberts Gretchen am Spinnrade: Latest Results of a Manuscript Study

To these should be added the differing note-values at the cadences of the individual poetic lines (bars , , , , , , , , , , , , , , , , ,
), of which bars and are especially important:

Finally, the distribution of the sixteenth-note figures between the treble and
bass staves is carefully gauged. (Mozart, Beethoven, and other composers observed this same notational practice).17
Let us now draw conclusions from all of this. We might perhaps consider the
cadences of certain poetic lines to be in need of improvement. However, the first
edition has clearly rendered them too uniform; for example, at the cost of the intended contrast in bars and . Indeed, anyone who presumes Schuberts hand
here would have to concede that none of the other variants could possibly be attributed to him, since he has left the earlier musical readings untouched; for example, if the variant at bars had occurred to him previously, what would

have prevented him from improving on the first versions in the manuscript
themselves, that is, regularizing all repetitions, as the first edition does? In other
words, if the augmentation of the motive in bars ff was a conscious process
when he first wrote it down (and who {} would doubt that?), then he could not
possibly have repudiated later the signs that held good for it. In a word: the person behind the first edition was someone other than Schubert.
The question as to who carried out the changes in the first edition I must
leave to others to decide. My guess is that one of his friends, most likely that perpetual know-all Diabelli, his publisher, simply foisted them on him. We know for
a fact that Schubert was always being given such pieces of advice, and that, in his
straitened circumstances, and at the beginning of his career perhaps also out of
shyness and lack of self-confidence, he believed himself obliged to endure them.
Thus, at the very least, his manuscripts testify to his artistic wishes. In fact, for us
today, they must be considered as his own assertion of his artistic intentions, in
the face of patrons and friends whohowever well meaning toward their protg and friendin a higher sense nevertheless mutilated his art, indeed art as a
whole, by putting themselves in Schuberts place.18

17
Schenker does not point out that the first edition abandons this device, placing all the sixteenth
notes in the upper staff.

18
The autograph fragment referred to in note forms part of the Berlin library manuscript
mus.ms.autogr. Schubert , the first of two collections of Goethe settings, all in Schuberts hand, which
were sent to the poet in in hope of winning his support for a projected eight-volume collection
of settings of German poetry, arranged by author (Goethe in volumes , Schiller in volume , Klopstock in volumes , and so on). Goethe returned Schuberts manuscript without comment, and the
project was abandoned; but other songs from the autograph collections were used as the basis of earlier printed editions. That Schubert was himself responsible for the differences between the Dumba
autograph and the first edition is probably beyond doubt.
The discrediting of Schenkers music philology here, however, does not necessarily invalidate his
insights derived from this study: Schubert may have written out the second Gretchen autograph from
memory, or in great haste, without giving the placement of the dynamics and other details the attention that Schenker felt they surely deserved.

Beethovens Fifth Symphony (conclusion)

Beethoven: V. Sinfonie (Schlu) {Tonwille , pp. }
t r a n s l at e d b y w i l l i a m d r a b k i n
It is as if all the things were wedged into one another.
Goethe, on his Pandora

Andante, at bar , whereas here at bar it is somewhat less clear, on account of

the reaching-over technique. On the other hand, there is a fundamental difference between the bass parts in the two movements: in the Andante the bass ascends in accordance with the basic plan of making its way via III to V; here it descends linearly through a fourth to the root of the upper-fifth divider, which
effects a completely different voice-leading.
If, then, the plan excluded a modulation, its realization required even more so
a corresponding increase of content, a broader span. The increase of content is
served here by a special type of repetition, as is shown in Figs. b and c: a mood
of fantasy inspires the master to have the component phrases lead, in a questioning manner, to half cadences or plagal cadences, either genuine or copied1 (bars
, , , , , , ), and in addition to avail himself of expansions, be they
clearly indicated by fermatas (as at bars , and ), {} or suggested by extensions to the ends of phrases, as at bars and bars , which amount to
the same thing. (This use of fermata is, however, different from Haydnssee the
first essay in Tonwille which are more often like taking a breath, as for example in chorales, and serve to mark formal divisions less frequently.)
It was these repetitions, together with their half-cadence-like endings, that
above all pushed the movement along the path of the fourth-progression. For
could the in bar have been supported by any root other than VII, given that
the effect of a half cadence would have been altogether impossible with II (di-

Third Movement (Allegro)

n contrast to normal practice, the first part of the Scherzo has no modulation
and thus neither a two- nor a three-part song form. Keeping its tonality intact, it
proceeds in a single sweep, driving upward from to in bars and then
falling in bars from to . The piece may be reckoned among those whose
form cannot be determined without knowledge of the Urlinie. The following illustration offers insight into the origin and development of the voice-leading that
the master has chosen for the construction of :

Just as with of the theme of the Andante (bars ), the ascent (see
Fig. a) here, inclusive of , should also take place entirely above the I. Although
the simpler adornment with neighbor notes in the earlier movement is replaced
here with the more artistic technique of reaching over, the upper-fifth divider of
I is also present, exactly as in the earlier movement; the divider is clearer in the

1
in wirkliche oder nachgebildete Halb- oder Plagalschlsse: Schenker is making a distinction between half cadences in relation to the principal tonality of C minor (as, for instance, at bars and )
and those applied to other scale degrees. By plagal cadences, he probably means the harmonies in


bars , which are to be read in C minor as I IVI , and so on (and not as VIV in F minor).

Beethovens Fifth Symphony (conclusion)

minished triad) or IV (minor triad)? And if the (see Fig. a) had indeed to be
supported ultimately by I (bar ) but, in conformity with reaching-over technique (Fig. b), had to be reached in advance of f 2, could the V have been approached by any chord but VII?2

provides the impetus for the lower neighbor note b1 leading to c2 across bars |
and d2 leading to e
2 across |. Innate lower neighbor notes are thus what fill out,
as passing tones, the arpeggiation in bars ; they assist in the broadening, in a
sort of ritenuto that, the more it contrasts with the quick arpeggiation of the upbeat in bar , the more effectively it lifts up the . In fact, the extended arpeggiation in the upper voice is accompanied by an arpeggiation also in the lower voice,
in bars , one based on the metrically regular move between the tonic and the
upper-fifth divider, the latter being transformed to produce descending leaps of
a fourth. {} Finally, Fig. d illustrates the tightly stretched form of the motive,
which will play an important role in the course of the bass part.
An important decoration of the motive (see the graph of the Urlinie) is provided by the upper neighbor note, which leads to the individual [Urlinie] tones:
e
1 to d1 across bars |, and f 2 as a long appoggiatura to e
2 in bar . Such neighbor notes, which should not be confused with reaching-over tones, continue to
be applied to the motive in the further course of the movement. How enlivening,
how logical is the interplay of upper and lower neighbor notes!
The repeat in bar (with upbeat in bars ) introduces an expansion in
bars . This expansion arpeggiates the upper-fifth divider, whereby the succession BC in bars is annexed, in a wonderfully organic way as a rhythmic
augmentation, to the preceding neighbor-note patterns, F  G and C  D (see the
graph of the Urlinie). What elegant logic! As stealthily as the expansion enters
during the repeatin bars it would have been unthinkableit prepares
with natural force a greater expansion to come, in bars .

Or, considering the whole, could any other chord have been used along the path
of the fourth-progression from I down to V?
The meter is based on a length of two bars: thus the first four notes should
have merely the value an upbeat (see Nottebohm, Zweite Beethoveniana, p. );
nonetheless, I have of course numbered the bars in the graph of the Urlinie
(shown on p. ) in the usual way, five bars at a time.
So much for general matters.
Bars . The first group of bars (marked A in Figs. b and c) is governed
by , that is, by the third of the chord embellished by its lower neighbor; yet the
is not set directly above the I but is gained by a compositional elaboration [Auskomponierung] whose wondrous paths are set out in the following example:

Bars . The second group of bars, marked B in Figs. b and c, is governed

by the . For the purpose of the reaching over, this is served by the upper neighbor note, g2, which begins as early as bar (one must merely imagine a g2 here,
rather than g1, above d2) and thus indicates, at one stroke, both the reaching over
and the status of the bar as an upbeat. (Regarding the bass motive in bars ,
see Fig. d.) In the bass notes BC in bars and A
B
in bars we have
the continued after-effect of the neighbor-note principle (F  G in bars ,
BC in bars ), irrespective of the fact that the bass note A
in bar at the
same time helps to avert the consecutive fifths in bars : cgf
b
. Basically, an in1


sertion of a above a stationary C would have been sufficient, along the lines of
a exchange, had not the propagation of the neighbor note cried out for an
A
in the bass.
In bar , the neighbor note b
1 appears before the a
1 that is essential for the

The c1 in bar is followed by d1 as early as bar , not bar (in the form of a
metric shift, that is, ahead of time); see Fig. a. Bars (see Fig. b) then appear
to give the motive in compressed form, but at the end it is complete (up to e
2).
This impression is certainly not far off the mark; but closer inspection tells us
that Beethoven has perceived the motive only in the broader form shown in Fig.
c. Here is an account of his thought processes: The quarter note F  that leads to

G across bars | (see the graph of the Urlinie), by which F  may be heard as a II[ ],
2
[S.] See Tonwille , p. /p. , Fig. ; Tonwille , p. /p. , Fig. ; p. /p. , Fig. ; and p. /
p. , Fig. c, and so forth.

tonw i l l e 6

Beethovens Fifth Symphony (conclusion)

voice-leading. These two notes, still united with g1 in bar , represent an ornament that recalls the higher-level neighbor note in bar (and ) and the long appoggiatura in bar (and ); in addition, the entire progression from b
1 to f 1 is
the model for the expanded progression from b
2 to f 2 in bars .
Now for the repeat: the divider at the lower fifth, E
, makes this possible. With
the minor-mode mixture at the lower fifth (a second-degree mixture: see Harmonielehre, pp. ff/pp. ff ), however, urgent requirements are associated: it
replies to the minor triad at bar (motivic parallelism: see Harmonielehre,
and /pp. , ) and provides a model with g
2 f 2 more strictly than
would have been possible with the diatonic g2 f 2 in C minorthe crucial diatonic succession a
2 g2 in bars . What fore-sight!3 In the repeat, too, the
motivic third-progression returns in the bass, at bars , and the neighbornote constructions (there are several of them here, of course) applied to the expansion of b
2 f 2 described above likewise contribute to the avoidance of consecutive fifths (see the graph of the Urlinie). It is striking that, in the inner voice, the
transitional harmony is lacking the seventh, which had been used in the previous
group (f 1 in bars and ). The slurring of f 2 to g
2 in bars also, it may be
noted, fulfills the intention of freeing the reaching-over g
2 from its role as neighbor note and thereby allowing the to gain prominence.
That bars provide the descent to the bass note G, that is, the completion of the fourth-progression, has been already mentioned (see Fig. ). Beethoven of course plays tricks with this at first, by transposing the opening bars, including the fermata, to the chord of VII in bars ; {} this actually suggests
to the listener a kind of consequent phrase for bars ff. But he does not extend the
parallelism beyond bar , so that its only purpose is to present the motive of a
third first in the minor version, b
cd
, so that its chromatic transformation,
b
cd  in bars (a result of the fifth-progressions: see Fig. c), can help
confirm the transition to the G major chord even more effectively. How well the
apparently tangled, enigmatic game, the perplexing alternation of lower and
upper neighbor notes, is thus clarified!
In bars ff, the G major chord is expanded even more than in bars
(note the descending arpeggiation, GDB, in bars , , and ), and the op-

Bars . We finally reach the third group, marked C in Figs. b and c, which
is governed by the . Only one who understands how g2 reaches over f 2 in bar
will be able to perceive the tighter connection between sections B and C,
which will otherwise go unnoticed across the broad expanse of the intervening
bars . Finally g2 in bar is brought into alignment with I, whereby the
basic plan (see Fig. a) is fully satisfied. The impression made by the parallelism
between bars and bars , which easily leads one on the wrong track, is
justified by the use of the lower-fifth divider in bars , which serves the
neighbor note, a
2.

3
[S.] See Tonwille , p. /I, p. . [ Schenker begins the Miscellanea with an explanation of the
greater visionary powers of geniuses, to see and comprehend far beyond what the ordinary eye can
perceive, and thus to be not just ahead of their own time, but ahead of all times. His orthography
herein the exclamation Welche Vor-Sicht!also hints at the normal meaning of Vorsicht (caution), which refers to geniuses never overstep[ping] the boundaries of their art.]

4
That is, the seventh (F) and the root (G) exchange voices; this is shown by the criss-crossed
dotted lines in the graph of the Urlinie, bars .
5
Schenker refers here to similar exchanges between a seventh and its root in Op. , third movement, bars (fourth and fifth bars of the second half of variation ), and Op. , third movement, bar .

portunity is perceived to take the Urlinie tone f 2, which has additionally undergone the gradual transition to a seventh by way of the interval succession
through the progress of the bass (see Fig. c), and to reveal it as a seventh also
along motivic paths, with def (see Fig. c and Fig. c). (At any rate, the succession e2 f 2 is the immediate result of the neighbor-note drive; see especially C  D
in the bass across bars |.) From the fifth of the chord, D in bar , to the third,
[B  ] in bar , the bass takes the longer route of an ascending sixth instead of a
the shorter path of a third downward; of course, the seventh in the upper voice
must yield (bar ) as soon as the lower voice prepares to take it;4 see the Erluterungsausgabe of Op. , p. /pp. ; and the Erluterungsausgabe of Op. ,
p. /p. ;5 and so on.

Bars . The realization of the Urlinie descent, , is clarified by the following illustration:

If here, too, are placed above IVI, then the V is again likely to be
understood as an upper-fifth divider, no matter how extended its elaboration.
Thus the descent represents a mirror image of the initial ascent (see Fig. a).

tonw i l l e 6
The new section returns to the motivic material from the beginning of the
movement, and for this reason the motive from bars must not be omitted.
Used three times in succession, it no longer develops the tonic chord, but rather

that of the dominant, with

, thus imparting to the seventh ( ) a special emphasis. In bar , the ninth is added, whose pronounced neighbor-note effect

proves highly useful for the -chord that finally arrives in bar . The upper
neighbor notes, too, are found again with the motive: f 2 to e
2 in bar , g2 to f 2
in bar , and a
2 to g2 in bars and . {} In the last instance, it appears that
the actual intervallic significance of a
has been displaced: as the ninth above the
bass it signifies an appoggiatura to g2; but this note, although it is the octave
above the bass, is itself again just an appoggiatura to the seventh, f 2 (see Harmonielehre, Fig. ).6 In bars , the following would have indeed been a more
exact imitation of the motive:

facts of the matter in the fivefold sequence) effects the mighty division of I by its
upper fifth. What a splendid fulfillment of the collective presuppositions all of
this signifies! The rhythm of the accompanying parts, two quarter-note rests and
a chord struck on the third quarter, had expressly been introduced as early as the
upbeat in bars .
There now follows the actual cadence in bars . Looking at the
bass, one is immediately struck by the deceleration of the arpeggio (at half
speed), which amounts to the same as a written-out ritenuto (see the dotted slurs
in the graph of the Urlinie). Strangely, the relationship of I (bars ff) to its divider in bars ff is actually strengthened further by the fact that the beginning
of the motive introduced by the cellos in bar , g1 f 1 e
1, is now applied to the
bass part beginning from c1, in the succession c1 b
-a
(and so on). The group of
bars (with bar as upbeat) provides the last reinforcement, in that it
dispatches the in bar . In reality, the conduct of the upper voice should be
understood as follows:

But as this would have invoked, above all, the impression of a -chord, it had to
be avoided and replaced by a freer version, one that better guarantees here the required coincidence of b and f.
The expanded V, under , is followed by an equally expanded I, under , in
bars . Strange mysteries weave their way through the motivic material. It
is not merely that one may perceive in itself an expansion of the constructions in bars , , and ( ); nor, moreover, that the apparently
new motive in bar (complete with tiny neighbor notes!) derives from these
motivesthe eighth-note figuration here is initially deceptive: the threefold appearance of f 2 to e
2 ( ) in bars , , and appears to present
the imitation of the neighbor note in a significant form. Here, again, the principle
of the broad upbeat, which runs through the entire movement, offers assistance,
as f 2 always appears only in the upbeat, as is fitting for this neighbor note. And in
addition we have the bass motive of bars , which has already been introduced
as an upbeat under and by whose quasi-ostinato behavior (it is only the ascending register transfer, shown in the graph of the Urlinie, that reveals the true

Of greater significance, however, is the animation in the bass: the motive of a

third, cde
, from bars (see Fig. d) is extended to the fifth, g (the insertion of IV in bar , which is required by the cadence, does not contradict this),
and so gains the form with which the main section of the movement slides into
the trio section!
Bars ff. {} The trio section has a three-part form: a1 bars , b bars
, a2 bars .
The motive of a fifth in the bass at the end of the main section, bars ,
provides the impetus in the trio for an Urlinie fifth-progression in C major, .

Nevertheless, the process of tonicization, II V (see Harmonielehre, ff/
pp. ff ) makes its influence felt and gives the impression of ( ) in G
major. It is, at any rate, easier to recognize the relationship of the earlier ascending fifth to the beginning of the actual theme of the trio: two motives of a third,
whose derivation from the motive of a third in bars is immediately recognizable, initiate the eighth-note motion in bar . And now in bar , the motive of a fifth, affirming the major key, storms in, as if in the spray from a wave
breaking on the shore. In retrospect we understand that the motive of a fifth here

6
Beethovens Sonata in A, Op. , first movement, bars . The e2 in bar , lying an octave above
the bass, is nevertheless dissonant against an inner-voice B and must therefore be treated as a dissonant suspension (resolving by step to d2). In the English edition of Harmonielehre, Jonas deleted the
footnote containing Fig. .

Beethovens Fifth Symphony (conclusion)

has even included the upbeat, since bar refers back to the B of bar . It is another matter that the B in bar , coming between the Cs in bars and ,
gains the meaning of a neighbor note, which can be raised to the status of an accented dissonance7 only if one considers bar as metrically strong.
The continuation of the theme beyond the motive of a fifth is the result of a
very common unfurling of the two semitone steps in the diatonic scale, which appear initially in succession, in bars and (see also the brackets in
the graph of the Urlinie):

bars now becomes, by the addition of counterpoint, a simultaneous event,

and indeed in both inversions: in bars and in bars . These
relationships are not even impeded by the actual roots of the chords, which are
additionally brought in for support in bars . {}
Bars ff. In the b-section, the modulation that returns to the principal key is

achieved by means of V 7.
Bars ff. The a2-section, too, introduces a series of imitations, but now the
theme is reduced to a mere four bars (the fourth-progression is dropped) and,
what is more important, the sequence of entries proceeds throughout by way of
the lower fifth (IV). Accordingly, the Urlinie gains the linear progression (see
the graph of the Urlinie), as a continuation of in the a1-section. In bars
the B, reached by the downward arpeggiation of the dominant chord, appears
to be already the first note of the theme (see bar ), something which without
doubt promotes the connection between the b- and a2-sections; on the other
hand, the concern for complete clarity required the composer to present the
theme exactly as in bars once more (played by the violas). The elaboration of the dominant in bars does not signify a pedal point. The descent
of the figure in bars ff, in contrast to the ascent in bars , is also worthy
of note, as is the hemiola in bars .
Thus the trio, considered as a whole, expresses a scale with the range of an octave, from C to c. In this, there is evidently a relationship with the fifth-progression in bars , which is now outbid by the octave-progression. In retrospect,
we also understand why the fifth-progression, already in the a1-section of the trio,
always pushed beyond its boundary to the octave: the small octave-progression
marks out the large one, which is realized over the entire course of the trio!

The torrent of the fifth-progression also carries with it the fourth-progression

from g to c (bars ), which is accompanied by the third-progression ce in
the lower voice. The balance between the two progressions (see the graph of the
Urlinie) is completed by the exchange. The characteristics of the theme are,
thus, as follows: the path from the root to the fifth and the fourth-progression
from the fifth to the octave, orwhat amounts to the same thingthe division
of the chord of C by the upper fifth and the interlinking of both parts by the
reversal [of intervals] shown in Fig. . From the indivisibility of the fourthprogression it follows that the total of six bars for the theme can by no means be
partitioned in any other way. The theme is then used in imitation in such a way
that the entries, as the graph of the Urlinie shows, again proceed by way of the
upper fifths: CGCGC in bars . A fugal development is
excluded. Already the second entry, bars ff, rebuffs the fugal principle: for in a
tonal fugue the dominant tone in bar would have had to be answered by the
tonic in bar . But even in a real fugue, the entry on the third quarter of bar
would have required g, instead of a, for its first note. Thus, the imitation here exceeds both types of answer, and only the retransition by means of the seventh, f,
in bar recalls the principle of fugue. And yet this note, too, is used less to fulfill the fugal principle than for the sake of the contrast in bars , in which

the f  is already hinting at the chromaticism that is to come (with II ).
The counterpoint at the second entry refers, as is shown by the brackets in the
graph of the Urlinie, to Fig. : what had been represented as successive events in

Bars ff. The repeat of the b- and a2-sections is written out and creates expressly
with f, diminuendo, p, sempre pi p, pp, and finally with the legato (beginning in bar
) a fitting retransition to the repeat of the a1-section. The quarter notes with the
quarter-note rests that follow them in bars prepare the pizzicato of bars
. The resumption of arco in bar merely underscores the beginning of the repeat of the main section;8 for, as early as bars , rests are used to weaken the
connection between the notes, so that they may simulate the sound of pizzicato.
8
Here, and at the beginning of the next paragraph, Schenker inadvertently calls the reprise of the
C minor part of the movement a repeat of the a2-section, having forgotten that he had said that its
form was indivisible (and having restricted the labels a1, b and a2 to the trio).

zur Wechselnote schrft: Schenker normally uses Wechselnote for accented passing tone; here he
is suggesting that the term can be applied to a neighbor note.

tonw i l l e 6
Bars ff. The reprise of the first part of the movement is played pianissimo, at
the same time with pizzicato in the string parts. It is as if the main section had
cast its shadow! In the expansion of bars , the pizzicato requires a repetition of the notes c  and d. Otherwise, the repeat is made as short as possible,
moving immediately to the IV (cf. bars ff)

nally f 2 (here as seventh of the dominant) leading to e2. Fig. c shows the transfer
of d2 to the higher octave, and also the voice exchange at the resolution of the suspended fourth. In Fig. d one sees the transition from a fifth to a sixth [between
the two upper voices], via , which is required for the register transfer.
The timpani in bars continues the rhythm of the preceding group of
bars, and leads to the next four-bar units, bars and , in such a way
that these are identified expressly by the change of rhythmic pattern. These three
four-bar units make up a group, the first.
There follows a second group, which is likewise put together from three fourbar units. In the first of these the timpani announces, with its quarter-note pulse,
the quarter-note rhythm of the first violins in bar . In the two four-bar units

that follow, bars , the II 3 is elaborated.
The totality of bars make up the third and final group, which, as the
graph of the Urlinie suggests, likewise represents what is essentially bars:
bars , , and . The first four-bar unit presents what may be
thought of as a tied-over chord (syncopation), whose significant and beautiful effect might, on the contrary, be initially deplored as an ugly effect if bar were
followed immediately by bar which is, actually, possible. Even in what followed, it would have been possible to observe the four-bar constructions as, for
example:

Bars ff. The following illustration clarifies, in compressed fashion, the remarkably bold plan of the transition to the finale:

{} The harmonic progression shows a reinforcement of the close, beginning


with the deceptive cadence: VIII 3 VI. But while such reinforcements of the
close usually add no more than to the Urlinie, here the opening motive of
the scherzo is drawn from those harmonies to an unparalleled extent, significantly transformed into the major at the midpoint (bar ), and in this form it
is used once more to set things in motion since the last movement also begins
with the motive of the [rising] third. The repeat of the two third-progressions
should not be seen merely as a superficial conjoining of scherzo and finale, rather
it repeats that repetition that had already governed the beginning of the first part,
in bars : as we had twice cde
, so here we have twice cde! Now tell me
if these are not organic forces that can compete with those of nature! In Fig. a,
the basic plan may be seen, including the suspended fourth above V. Fig. b shows
the motivically innate neighbor notes: d2 initially preceded by e
2, in conformity
with the tonality of the scherzo, then by e2 in anticipation of the major, and fi-

But the register transfer demanded an expansion. This begins with bar , where
for the first time e
2 actually appears on the downbeat of the bar! The first half
note of the three-note motive, which from this point moves in a circle, will not
fall on a downbeat for another three bars; within this space, however, the motive
is used only twice. The fifteen bars of the expansion may thus be explained as
comprising five of these three-bar units. But what is special about this effect (and
not previously understood) lies in the fact that, in spite of the three-bar grouping, the basic duple pulse is carried through; see the dotted slurs in the graph of
the Urlinie. The six-bar group, , serves as a transition to the eight-bar
group, . Beethoven strengthens the four-bar construction that is secured
in the expansion even further, by a special application of slurs in the {} firstviolin part: the first slur embraces bars ; the second (present in the auto-

Beethovens Fifth Symphony (conclusion)

graph score!) likewise only four bars, ; and if he draws the third slur over
seven bars, instead of six, he does so only that the last element of the motive does
not stand alone, outside the slur. Note however how, to indicate the beginning of
the eight-bar group, he also begins the eighth-note motion in the basses, and at
the same time in the bassoons, precisely in bar !

In the drafts of bars ff (Nottebohm, Beethoveniana,10 p. ), the expansion

in bars ff is not yet present, and the reprise of the first part of the movement
follows immediately upon the perfect cadence.
Originally Beethoven had specified, with a prima volta at bar , the repeat
of the entire first part and the trio. He later dropped it; but by an oversight the
two bars of the prima volta were retained in the fair copy and the first edition,
without repeat marks. Thus a prima volta and a seconda volta stood next to one
another, which repeated the series of tones in bars , with the only difference
that the former was legato and the latter had quarter-note rests. The pure nonsense that resulted from this oversight had its enthusiastic panegyrists, who regarded it as a genuine Beethovenian touch until the year . (On this matter,
see the highly informative passage in Beethoveniana, p. , especially the polemic
against Anton Schindler; then read the account in Thayer-Riemann, vol. ,
pp. , which includes Beethovens clarifying letter and singles out Mendelssohn for the credit that is due to him.)
For the transition in bars , one also finds several sketches transcribed
by Nottebohm. In one (Beethoveniana, p. ), Beethoven deals very briefly with the
motive of bars in the course of the harmonic progression VIIIV(I) and
closes with the motive of bars . In two other sketches (Beethoveniana, pp.
and ) he has already introduced the timpani, and in a further sketch (Zweite
Beethoveniana, pp. ), he makes use, in a clear effort to expand, routine
progressions in the bass and the upper voice, but not the motive of bars .
{} I should also like to say something about one of the sketch-leaves preserved in the Gesellschaft der Musikfreunde in Vienna, which has some drafts for
the bass at bars ff (see earlier), and which Nottebohm did not publish:

From the sketches for the scherzo, Nottebohm (Zweite Beethoveniana,

pp.
ff) tells us that Beethoven, in order to establish the two-bar unit, preceded the actual start of the motive with a sort of introduction, in which the fourth, CG, was
sounded twice in bars, that is, in four bars. He comments:
9

Beethoven later removed the introductory bars in the sketch, probably on

the grounds that, as unnecessary ballast, as something not really belonging to the theme and standing apart from it, they would have impaired its
melodic design.
On the contrary, I think that those introductory bars caused damage not so
much to the melodic design as to the rhythmic design of the principal motive
(that is where the mistake lay), to which they nevertheless ought to have been of
service: bars of the final version, in which c occurs at the high point of the
motive of a third, represent unambiguously the first two-bar unit, which is metrically strong. In the sketch, they occur in the fourth two-bar unit, which is weak,
with the result that the c of the first and third two-bar units would have fallen
into conflict with the c1 in the fourth two-bar unit, and thus would have been deprived of a solid point of support and departure. But it cannot be determined
whether the final form of the upbeat arpeggiation, by which the two-bar unit has
been shown to be more secure than with the introductory bars, actually was determined by the principal motive from the finale of Mozarts G minor Symphony,
even though it is clear that Beethoven copied out twenty-nine bars of this symphony on a bifolio belonging to the Fifth Symphony. Thus Nottebohm, who
communicated this observation (p. ), is justified in posing the question: Did
Beethoven notice the similarity? At any rate, the Urlinie follows a different
course in Beethoven from that in Mozart.
9
Gustav Nottebohm, Zweite Beethoveniana: nachgelassene Aufstze (Leipzig: Rieter-Biedermann,
).

10
Gustav Nottebohm, Beethoveniana: Aufstze und Mittheilungen (Leipzig and Winterthur:
Rieter-Biedermann, ).

tonw i l l e 6
One can see how the master struggled with the slowed-down arpeggiation. The
first line shows no fewer than four complete IIVVI cadences; there are two
cadences in the second line, which is two bars longer; the third represents a regrettable step backward;11 and the fourth expands the IV.
Now for the sketches that are still contained in the manuscript:12

In bars , the passing B
is still missing from the bass; by its insertion, the
horn figure in bars is anticipated.

In bars ff, the horn originally took the bassoon part, and vice versa; the improvement gives the horn a more suitable continuation of the soloistic content of bars .

It is worth mentioning that the slurs in the bassoon part in bars and initially encompassed the half note and the third beat, which suggests that Beethoven sought to have the slurring conform with that of the other instruments in bars and ; but in the clarinet part, which plays the same thing
an octave higher, he places a slur from the third quarter of bar to the dotted half note of bar and, moreover, writes the parallel passage in bars
and in the same way. In fact, there would have been no justification for
making the slurring at bars and uniform in all the parts, since the bassoons and clarinets do not play their sf until the third quarter, whereas in all
the other instruments it occurs at the start of each two-bar unit.

In bars , the first violins were doubled by the first flute, and the movement in thirds between the second violins and the violas was doubled by the
two clarinets.

From among the numerous corrections found in bars ff, I shall mention
only that the flute and oboe originally entered right at bar , timpani right
at bar , and that in bars ff the timpani (which is silent in the final version) originally took part by playing the rhythm of the woodwind instruments (see earlier), sempre pp.

11
This is not necessarily the case: in jotting down a sequence of sketches for the same passage,
Beethoven sometimes attempts merely to improve part of the previous entry, rather than replace it altogether. In the four entries shown here, taken from MS A in the Gesellschaft der Musikfreunde,
line may have been intended as an alternative to the last five bars of line (or even the last four bars
of line ); similarly, line may be an alternative version of the end of line , which would have resulted in a two-bar expansion.
12
Schenker avoids calling these changes to the score corrections, possibly to avoid giving the impression that an earlier reading of a passage could be part of a different version of the piece. In his
essay on the first movement (Tonwille , p. /I, p. ), this compositional stage was discussed under
a separate rubric: Autograph.

In bars , the bassoon part still went with the basses.13

The scoring of the contrapuntal parts in bars ff, second violins and violas,
created difficulties. The first violins in bar were accompanied in sixths not
by the violas, but by the second violins.
In bars ff, the bass counterpoint was syncopated; and even despite the thematic statement in bars , the syncopation was continued in bars ff.
Conversely, however, the bass remained stationary on G from bar onward.
In bars ff, the manuscript shows the attempt to use the main theme of the
trio as a counterpoint in the first violins (in the register of the flutes in bars
, and always fitting with the transitional harmony.
In bar the bassoon originally had g1, and also e
1, of course, in bar .
In bars ff, the cellos had rests on the second and third quarters.
In bars , and likewise in bars , the violas replied {} in reverse
order, with f 1 and a
1, to the notes a
2 and f 2 in the first violins.
In bars ff., the second violins did not yet have a rest on the first quarter.
Even in bars ff, the bassoon continued along the path of the basses: how
fortunate that, as a result of the improvement, the bassoon was taken out of
play before the deceptive cadence!
In an apparent effort to prepare the crescendo, but mainly to indicate the fourbar unit (in the final version, this is left to the timpani alone), the bassoons
entered in octaves, c1 c, as early as bar .
In bars ff, the bass proceeded in dotted half notes; from bar onward,
however, the G remained stationary.
The rising transitional passage of the first violins was accompanied by the
second violins in such a way that they held on, in long note values, to the
lower boundary of the sixth which the first violins embellished; this actually
resulted in rhythmic shifts, for instance, b1 as early as the third quarter note
of bar . The violas, however, accompanied the first violins in sixths throughout. Altogether, one sees the master struggling mightily with a polyphonic accompaniment for the first violins, whose part caused him much trouble in
any event. Finally he was obliged to leave it to the first violins to articulate the
sixths, on their own and in a concealed manner, instead of actually calling
upon other instruments to help out.
In bar , the oboes already entered with the sixth, f 2 a1, and in bar the
second violins were an octave lower.
The bassoons actually follow the violas, not the cellos and basses, in bars .

13

Beethovens Fifth Symphony (conclusion)

. In bar , there should be a p, not pp, in the cello part (according to the autograph and first edition); likewise
. In bar , there should be just a p in the bassoon.
. In bar , there should be a pp in the horn (according to the autograph: the
first edition has p).
. In bar , a new slur should begin in the first-violin part, which continues
to bar . (The first edition divides this, against the clear wishes of the composer.)

ere is a list of passages that have been incorrectly transmitted in the printed
scores:
. In bar , the bass part should have an sf > (thus in the autograph; in the
first edition there is an sfp).
. In bar , the bass slur should begin right from the half note on the downbeat. (This is correct in the first edition.)
. In bar , the bass slur should begin right from the f.
. In the viola part, a new slur should begin in bar ; it prepares the slurring in
the lower string instruments in the bars that follows. (This is correct in the
first edition.)
NB: In bars and , Beethoven does not mark any slurs in the bass part, even
though he had provided them with slurs in bars and . (The first edition follows the autograph.) Without doubt, the string dotted half-notes in bars ,
when related to the quarter notes in bars , and then the ff in bars ff and
ff should approximate a non legato.
. In bar , the trumpet part should have a p and nothing else. (In the autograph, the original diminuendo is crossed out and piano added; but the first
edition already has p dim. pp, apparently to make the trumpet dynamics the
same as those of the other instruments.)
NB. In bar there is no dynamic mark at all in the cello (this is also true of the
first edition). Yet it is self-evident from the f > p in bars and marked in
the autograph (which the first edition simplifies to fp) that a p is preferable to the
pp from bar .
. In bar , a p should be added to the viola part (according the autograph and
first edition).
. The slur that starts in the cello part in bar should continue uninterrupted
until bar . The same thing applies self-evidently to the bassoons! (The autograph leaves no doubt about this; nevertheless, the first edition already introduced a division of the slur.)
. In bar , according to the autograph and first edition, there should be a ff
for the entry of the first violin; likewise {}
. In bar there should be a ff in the first flute.
. From bar onward, the instruction sempre pp applies only to the instrument that has the broken chords, that is, the violin in bar , and the viola in
bars , , and (as is indicated in the autograph).

For the performance of this movement (  ), it is essential to have a secure

feel for the two-bar units, all the more so in those places where they grow into
large expansions (see earlier). The poco ritardando in bars should be treated
only as the gentlest departure from the tempo, of the sort that would voluntarily
occur merely on account of the fermata, even without a prescribed ritardando.
(Compare, for instance, Beethovens Piano Sonata Op. , No. , first movement,
bars ; in order to convey the fermata, one must make the eighth-notes somewhat slower toward the end, even though this is not prescribed.)
The technique of reaching over, described earlier, can be conveyed in performance, and it is even easy for the conductor to achieve. He should introduce
the horn even before the fermata in bar has finished. If, further, he introduces
in bar the reaching-over g2 as if it were a long-expected note, perceived in advance, and verily concentrates on this note alone until bar , finally letting the
tension reach its climax in the ff in bars ff, then he will have conveyed the
reaching-over design of as the composer understood it. But he should not
forget to allow the motive of a third to become prominent, wherever it surfaces
(bars , and so forth).
It is equally possible to convey the unity of the descent from to . The conductor should concentrate first on bars , where the composer himself
underscores the that has been reached by repeating it twice. Then from bar
onward, in following the undivided Urlinie segment , he should proceed with
a single impetus to the end, taking care, however, not to play the motive of the
new group of bars as if it were a new idea, or to make a climax there by means of
an overly powerful entry. Finally, the fifth-progression, bars ff, will crown the
whole.
In the trio, the eighth notes of the fifth-progression (the second bar of the
theme) should be in the foreground; this applies to every entry. An imperceptible

tonw i l l e 6
to this e3, in accordance with the law of obligatory register (see Tonwille , p. /I,
p. )in what a wondrous light the ascending register transfer appears there!
but also the smaller arpeggiations in bars and answer one another within
the large arpeggiation. The construction of the theme is secretly governed by the
number , which has influence over the groups of bars and the principal harmonies. Thus and I govern bars , grouped as bars (note the change
of rhythm); and IV govern in bars , likewise grouped as ; finally
(or ) governs bars , again as , whereby the last bar, without relinquishing its basic significance as a sixth bar, at any rate expands to a special
eight-bar group, bars . This group does not actually contribute anything to
the harmony, or to the Urlinie, and contains only a descending arpeggiation that
restores the original register of bar . That does not change the fact that the eightbar group shows a well-rounded organization, that is, the alternation
in the
rhythm of   in bars and by foreshortening (i.e., acceleration) the same alternation in the rhythm of    . Nor does it change the circumstance that, in
bars , high and low registers are invoked for a mighty crescendo to increase
the sonorous effect of the unison passage.The internal principle of construction in the individual six-bar groups is striking, in that the greater metric values
fall on the weak bars: see the dotted half notes g2 in bar , c2 in bar , e2 in bar ,
c3 in bar , and so on.

softening after the f at the beginning of each entry, together with a crescendo from
the third eighth of the second bar, will lead to the desired goal.
In the transition to the last movement, the conductor should follow the composer above all in the organization in units of four bars (see earlier); he should,
in particular, stress bars and as [rhythmic] nodal points.

Fourth Movement (Allegro)

In my instrumental music, too, I always have the whole before my eyes.
Beethoven to Treitschke14

The sonata form of the first movement proceeds as follows:

first subject:
theme A
theme B and modulation
second subject
third subject
antecedent
consequent and transition to exposition repeat
development
recapitulation
coda and stretta

bars

Bars ff. The continuation from bar nevertheless does not resolve the half cadence of the first subject; that is, the preceding Urlinie progression is by no means
finally resolved by the in bar ; on the contrary, it expires at the (i.e., ), and
in bar now marks the start of the second theme, B, which likewise strives
after and which, in the manner of a consequent phrase, also executes the modulation. {} (One need only imagine in bars ff the usual succession of repeated
chords, instead of the arpeggiation, in order to recognize that here, too, as is so
often the case with antecedent phrase constructions, the harmony concludes on
V. The characteristics of bars ff as a consequent phrase are compromised only
by the fact that the Urlinie ascent, despite remaining the same, in the manner of
a reply, is now realized by a different motivic content.)
Groups are formed of four bars each: bars , bars . But the group
of bars are ultimately derived from a four-bar construction, insofar as bars
follow bar and bar as an expanded third bar, bars as a likewise
expanded fourth bar.
What is crucial is the rhythmic shift in the first and second four-bar groups:

Bars ff. The Urlinie (see the graph, p. ) shows the ascent in theme A from to
, to which the neighbor-note is attached at bar for the purpose of strengthening the goal (as is so often the case when a is attached to ). In the descent,
the Urlinie reaches the (bar ), unless one regards the succession merely
as an elaboration of .15 The e3 in bar is approached by a large arpeggiation,
which begins right at bar , as is shown by the dotted upper slur in the graph of
the Urlinie. Not only does the last high f 3 in the retransition (bar ) thus reply
14
Georg Friedrich Treitschke (), playwright and theater director. Beethovens remark
appears in a letter of March concerning some vocal pieces, including a new version of Fidelio (for
which Treitschke was to revise the libretto) planned for the Krntnertor Theater, Vienna.
15
In later writings, the would be described as the principal goal of such a progression, with
(often noted ) attached as a third-progression of the first order; see Der freie Satz,
and Fig. a.

Beethovens Fifth Symphony (conclusion)

d2 arrives as early as bar (instead of bar ) and e2 as early as bar (instead of

bar ). The motive in bars is thus created in such a way that, if one disregards the reminiscence of the fourth-progression in bars (see the downward movement of the final sixteenth, c2 b1 across bars | and d2 c2 across
bars |), it embraces the notes of the rhythmic shift like a bracketing third, as
if the notes comprising the third were squeezing out the rhythmic shift with unified strength. The same bracketing third of the motive then pushes with its continuing procreative force once more from e2 (bar ), and again upward to f 2
(bar ), instead of downward to d2, the note to which the Urlinie might nevertheless have wanted to move immediately, by reinterpreting its as in G major
(see the graph of the Urlinie). All that remains for the composer is to find a way
out of the push upward andthis is the very purpose of the first expansion, bars
to strive for the in the form of d3, by means of a quasi-obligatory ascending register transfer. But if here, in order to avoid having to move from f 2 up
to d3 in a step-by-step manner, a shortcut had to taken to assist him, then this
would have had to be taken care of in good time. This occurs in bars : since
Beethoven repeats the falling third in bar (see the graph of the Urlinie) again
on the downbeat of bar , he gains with the new construction in bar a onebar abbreviation of what was formerly a two-bar motive (bars and ),
without in the meantime losing the two-bar organization of bars . Not
until new content of these bars is confirmed by a repetition in bars thus
Beethoven allows laws of nature also to govern his art!has the time also come
to free the abbreviation (bar ) from the two-bar construction and to let it become a new seed: see bars , , and . The more detailed organization of the
five bars, , is indicated by the voice-leading (see the graph of the Urlinie),
which uses the neighbor-note motion of the third (see Kontrapunkt, ii, pp. ff/
p. ): for the | | construction, bars are related to bar as bar
is to bar . That, moreover, the composer himself is also conscious of these
rhythmic relationships can be deduced from the fact that he introduces the continuous eighth-note movement precisely in bar .
In bars , where the d3 awaited since bar and finally reachedcancels the obligation for a further rhythmic shift, the second motivic component
immediately takes advantage of the opportunity and straightens itself once more,
as in bars ; the summative fourths appear first in the bass and then in the
upper voice. (This compensates for the difficulties that the fourth-progressions
in bars had first to endure on account of the rhythmic shifts.
One can additionally say that the first four-bar group is connected to the sec-

ond by the bass figure, and that the second group is connected to the third by the
eighth-note motion that continues onto the downbeat of bar , and that, finally,
from bar onward the bassoon enters with an inversion that counterpoints the
second motivic component, which is significant for what will follow. {}
Bars ff. After the modulation has brought into a new diatonic setting (with
merely as a neighbor to ), the second subject, bars ff, now seeks to firm up the
descending Urlinie progression: the antecedent (bars ) provides , the
consequent (once more with the insertion of in bar as a neighbor note)
the descent from to the in bar .
In bars we have, in the service of the descent, the motive of a
fourth from the modulation (bars ), made up of triplet figures in a quasisyncopated context. The intention was, apparently, one of rhythmic diminution,
which the motive of the fourth-progressions suggested to the composer even for
purposes of elaboration. But the triplets made a better effect than the straight
eighth notes, which would have been far too lame:

And thus the composer might perhaps have applied, from this point onward, the
triplet form also for all the other figures of the upper and lower voice, regardless of
whether they descended, and thus appeared to be linked to the fourth-progression
of bars , or ascended, and thus recalled the first-bassoon and viola counterpoint in bars , , and so on. In the second subject, the characteristic of the first
subject returns, namely, the longer note values (dotted half notes) are placed on
the metrically weak bars. For this reason, the V is perceived more as arriving in
bar , not as early as bar (compare in the development section bars , ,
and , and in the transition to the coda bars , , , , and so on).
In the consequent, bars ff, the ascending register transfer (see the flutes)
must be heard as significant and, accordingly, the entire fifteen-bar group should
be perceived as basically only two bars longer than the antecedent, that is, as a
six-bar group: bars are the first and second bars; bar (where IV enters,
in support of ) and bar are the third and fourth bars; and bars are
the fifth and sixth bars. The expansion of the second bar, that is, the eight-bar
group comprising bars , is filled out by a transition in the string parts, beginning with d2 in bar and also ending in bar with d2, thus confirming that

tonw i l l e 6
Bars ff. In bar , the second subject concludes with , and at the same time the
closing subject begins. There are strong grounds for regarding the [end of the]
second subject as a perfect cadence: not only does the [previous] V (bar ) lack
the necessary expansion to become a half cadence (apart from the fact that, in ad
dition, the preceding II, in bar , lacks the tonicizing third, II 3V), but also the
c  itself, which is hurled up into the upper voice, swings mysteriously from the
II across to the V and forms a seventh above this chord; its intrinsic force as a
passing-tone leads to the cadence only at the I that follows.
The antecedent phrase of the closing subject (bars ) begins from , in
order to overtrump the Urlinie progressions from belonging to the second
subject and, in so doing, accentuate its closing character. But here, too, the motive of a fourth, descending and ascending, continues to operate. For the first
time, the descending motive falls on the strong beat, and it is accompaniedin
fact, it is crossedby the ascending motive (see the crisscrossing slurs in the
graph of the Urlinie). But the ascending motive does not merely round off the
descending fourth-progression as a fifth-progression, as earlier in bars ;
instead, the fifth-progressiondespite the repeat in bars comes into
being only when the c2 (standing for c1) is attached in bar . For this reason, the
harmony at this point should be interpreted as IV. The progression that now descends from c2, still always accompanied by ascending fourth-progressions, already follows the harmonic progression of the cadence, IVIIVI, a special
feature of which is the tonicizing g  in the second half of bar that actually
underscores the move toward II.The change of chords from one bar to the
next is influenced by the rhythm of the closing bars of the second subject (bars
): thus we recognize the significance of the model only in the copy that
follows! The graph of the Urlinie shows how g1 ( ) is drawn by way of c2 to f 2
(  ); by substituting g2 for g1, one obtains the two-line octaves as the actual register of the Urlinie tones. The consequent phrase then provides the ascending
register transfer of the Urlinie tones into the three-line octave, which is also retained in the development section.

the Urlinie is at rest. But with how many naturally born powers is this transition
supported! Whether the first-violin motive begins on d2, e2, or g2, the composer
always leads it down by only a fourth, as previously. But in order to comply with
the no less justified demands of the rising fourth-progression, he interjects this
each time in the remaining string parts, playing in unison. In this way he succeeds
in augmenting the [first] two fourth-progressions, which had remained motivically faithful, to form the fifth-progression that is indispensable for the voiceleading here! If, in the last quarters of bars , , and , the fifths appear less
pronounced, this is a result of the more prominent effect of the thirds (see the
bassoons!), which is demonstrated by the following illustration:

The expansion in bars is filled by a sixteenth-note run that storms upward.

It not only answers and surpasses the rising fourth-progressions in bars , ,
and , but beyond this its arpeggiation of a third, from c2 to e2 on the strong and
weak beats of bar , is similarly related to the arpeggiation b1 d2 on the downbeat of bar (see the graph of the Urlinie).16 The crossing of the first flute and
the first violins in the {} downbeat of bar confirms the leading role for the
flute (see earlier). It is only after this g3 has been reached that the descent occurs in its proper place: see especially the eighth notes in the oboes! This is correct, despite the fact that the first violins also reach the same goal, only an octave
lower than the first flutes, corresponding to the lower register that was taken from
bar onward: still, the afore-mentioned reply with the third c2 e2 in bar has
already striven to preserve precisely the lower register, so that the high-arching
sixteenth-note run proves to be merely a detour, which does not in the least displace
the original register of the violins. Note above all, however, that the two expansions,
bars and , are strongly bonded to form a unity, which is expressed
see the large bracket in the graph of the Urliniefirst of all by the closed harmonic cycle IIVVI, second by the corresponding internally closed bass
movement GG, and third by of the Urlinie, which basically signifies
nothing more than a stationary . The semitone progressions in the bass in bars
continue the bass neighbor-note motions of bars , , and .

Bars ff. Attached to the consequent phrase is also the return modulation to C
major, which has to prepare the way for both the repeat of the exposition and the
development. For this reason, the linear progression to IV in G major is avoided
and the I in bar is clearly continued (see the bass), so that the downward elaboration of the upper voice from the octave down to the third is understood un-

16
The connection between the two thirds is indicated in the graph by a dotted line from the b1
to the c2.

Beethovens Fifth Symphony (conclusion)

ambiguously in terms of I overall. It is thus the conduct of the bass which, in bar
, requires us to interpret c3 (instead of c2) no longer as the conclusion of a fifthprogression and a sign of a change of harmony, as in bar : we perceive that the
repeat of the earlier cadential progression has become impossible. {} Since the
II is thus omitted, the chromatic note, too (recall the g  in bar ), becomes immaterial; thus the master reinterprets g  as a
in bar and proceeds, in terms of
V in C major, to f 3 in bar , by which he confirms the return modulation also
in the realm of the Urlinie, that is, gfe as in C major. The rare voiceleading in the bass is clarified by the following illustration:

{} In Fig. a, we see that the development section, in spite of the seconda

volta, has actually been active since the [start of the] return modulation in bars
ff, since indeed bars already provided the Urlinie with . The composer actually has his goal in mind, in the distant future: to affirm, by division at
the upper fifth, the significance of the dominant chord leading to the recapitulation as that which holds sway over the entire elaboration in bars . Understood in this way, the lower voice is brought from the root G to the divider at the
fifth, D (not until bar !), and juxtaposed with its third, the neighbor note f  in
the upper voice, for the sake of making a consonance. In Fig. b, the path taken
in bars , the succession gf, is included; and we see how the e in bar , the
note to which the seventh (f) has resolved, is already greeted by the second note
of the fifth-progression, the root A. Fig. c shows the purely diatonic fillingout of the fifth-progression with BCD the lower voice, to which the upper
voice replies with the neighbor-note figure efe in bars , , and , again in
order to make consonant chords. Fig. d shows the fifth-progression ornamented
with semitone steps, which had been introduced by the raising of the root to G 

In Fig. a, where the original sequence of chords is reproduced. We see the passing-note motion in the lower voice from the third B by way of A
to G, and in the
upper voice the entry of the seventh, f, which must be understood as a passing
note, precisely above the passing A
in the bass; this leads at this point to what appears to be a chord. The elaboration of this very chord, a
2 c3 f 3 in the upper
voice (see Fig. b), compels the composer to exchange the voices in bar and to
give c (see the inner part of Fig. a) to the bass. In the overlapping progression
from g2 (bar ) to c3 (bar ) to f 3 (bar )see the dotted slursis expressed
the will to the high register, which the composer must achieve once more in accordance with bar and bar , respectively.
Bars ff. In bar , seconda volta, the root of the dominant chord of G major is
affirmed as even more emphatically than in the prima volta, and underscored
expressly by the new rhythm in the clarinets, all the brass instruments, and the
timpani:

back in bar ! Consecutive fifths between bar and bar (see Fig. c) are
avoided by the exchange in bars , , and (see Fig. d), which at the
same time places the necessity of the neighbor-note motion in the upper voice in
its proper light. Fig. e, finally, shows finally how the lower voice, instead of proceeding simply from B
(bar ) to B  (bar ), takes the more colorful path of
elaboration, B
CDD
C; it goes without saying, however, that this route does
not invalidate the actual movement of the bass shown in Fig. d.
Some further remarks on the details:

Now the bass enters with all the more justification in bar with the raised note,
G , which leads after four bars to the chord of A major. The great thrust of the
development section is deciphered in the following illustration:

tonw i l l e 6

The repeat of the bass figure in bars is intended to verify the raising of
the root merely as a chromatic passing tone; it is only the complete change of
motivic content, by the use of the second subject in bars ff, that points to
the weighty diatonic bass note A; see Figs. b and c.
The inversion in bars is self-evident; note here especially the insertion
of the rising fourth-progressions in the oboe and flute.
The transition to the chord on A in bar is achieved by imitations that are
built upon the exchange of diatonic semitone steps (for the purpose of a
change of register). This is, at any rate, concealed by the numerous leaps of a
third in the bass, which appear to result in a cadence in F major.
What follows in bars is by no means an exact repetition of bars ,
which would similarly have had to introduce a chord on B
in bar , but a
merely imitative treatment of the upper and lower voices, which at the very
last moment (across bars |) lays out a B
major triad, in order that the
f in the upper voice can be held onto as the fifth above the root B
.
After the insertion of these two groups of bars, the play with the motive of the
second subject is repeated: bars correspond to bars , and bars
correspond to bars . But here the end-notes are subjected to expansions, which prepare the more significant expansions at the root notes C (bars
) and G (bars ff)! The expansion in bars ff serves primarily a
threefold, quasi-stretto presentation of the motive in the first and second violins, horns and trumpets, violas and cellos. But even before the conclusion of
the last imitation, the first and second violins begin anew a (fourth) entry,
which now continues in augmentation (and, at the same time, with an ascending register transfer) to f 3 in bar (see Fig. e). This transitional passage could
certainly have been elaborated in its entirety as a succession of thirds between
the outer voices; nevertheless, the semitone motion, e2 f 2, together with the
flutes ascending fourth-progressionwhich leads to the semitone successions
g2 a
2, b2c3, d3e
3, {} and finally f 3g3 (in bars , as an augmentation
of the previous semitone motions!)organizes the transition more tightly,
with ||||. What logic there is in these events, which seem to
drive the music onward, though they themselves are driven!
In bars , only the dominant is elaborated (not as a pedal point!): the
space of a fourth, dg, in bars ; thenafter the octave has been overshotthe space of a third, gb, to bar , and finally the arpeggiation to the
next octave, bar .

Bars ff. In order to lead back to the recapitulation, it was necessary to gain the
succession gf(e) in the upper voice. Now, Beethoven would not have been the
very master that he was, not the ear and the voice of artistic and natural necessities, had he not used this opportunity to recall the scherzo, and thus establish a
parallelism with the transition to the finale. And how much stronger is the effect
than it would have been if the fortissimo recapitulation had been directly attached
to the sempre fortissimo of bars ! The neighbor note f 2 in the oboe in bar
prepares the descent from g2 to f 2. This f 2 appears for the first time in bar ,
a metrically weak bar, which leads to a string of syncopations in the course of the
arpeggiation f 2 d2 b1 d2; even the flute and bassoon, which join the oboe in bar
, enter accordingly in a weak bar. Thus the arpeggiation proceeds, twice
downward and upward, in syncopation; not until the d of the third arpeggiation
(bar ) is the tying-over omitted, resulting in a kind of abbreviation from bar
onward, whereby notes and bars follow one another in the regular alternation
of strong and weak. By means of the very last abbreviation, which takes the form
of a leap of a fifth fb in bars , the passage is expanded to a six-bar group,
which however should be regarded in terms of the two-bar units in the scherzo
as ; in this threefold ordering, the group should thus be understood as an
elaborated ritenuto just before we jump back into the finale.
Bars ff. The recapitulation may be passed over. Here, I shall say just a few
things about the end of the work, with the help of the following illustration:

Beethovens Fifth Symphony (conclusion)

The consequent phrase of the closing subject begins in bar . This time, unlike that of the exposition (bars ff), it strives anew after the and thus gives occasion for the descent in the coda. {}. In comparison with the step motion
to , which is the only form of motion allowed in the Urlinie,17 the form of the
arpeggiation makes its influence felt even more strongly hereas Fig. a shows
as if the chordal placement of above mattered more than a linear elaboration.
For several reasons there is, namely, the effect of a caesura [Einschnitt] in bar ,
whereby at the same time the interval of a third between the e3 reached here to
the starting note c3 (bar ) becomes especially animated: [] the motivic divide
at this point (the motive from the second subject returns at exactly this point, and
is again used as it was in the development section); [] the chromaticizing of the
root G (the bass EF in bars and has already provided the impetus);
[] the construction, which verily gives the impression of the fermata that signals a concerto cadenza; [] the interruption of the conceptually expanded motive
of a fourth, which reaches only as far as e3 and is prevented from proceeding to its
end-note, d3 (see the broken square bracket in the graph of the Urlinie); [] the
concealment of the step motion, precisely in bars , by the motive of a
fourth that unfurls above it (see Figs. d and e): c 3 in bar is still part of the
upper voice, but d2 in bar is already in the inner voice. All these are the reasons
why the arpeggiation (Fig. a) irrefutably has the upper hand over the stepwise
motion (Fig. b), as is indicated by the dotted slurs in the graph of the Urlinie.
The step motion, however, is connected with the course of the harmony (see
Fig. c). The chromatic c  in bar is already pointing forcefully toward d ( ) at
II. The motive of a fourth, previously interrupted, is now resumed and in bar
reaches its end-note d2 with the effect of a suspension resolution and at the same
time a continuation from the third ce (see Fig. b)and from here onward the step
motion can also be followed clearly alongside the harmonic progression. In bars
the is inserted, as neighbor note, and finally in bar the line arrives at
the (over V), which was also the goal of the arpeggiation (see Fig. a).
Because the step motion has come to the fore, the relationship of the g3 in bar
to the e3, which had been heard a long time before (bar ), has in the meantime been obscured. The leaps of a third, eg, in bars and now
serve to refresh it (see the square brackets in the graph of the Urlinie).

Bars ff. But how far apart the three notes of the arpeggiation, bars , ,
and , still lie from each other! It is now the coda that pushes these notes together. Insofar as the master here leaves off the dotted half notes from the motive
of the modulation (bars ) and instead adds the fifth of the scale as its final
note, he leads forth the arpeggiation of bars , now expressly put together
and in a still more accelerated succession! (In this way, motives change their
meaning according to the fate that befalls them, always as a result of their setting!) Even in bars ff, broken thirds still serve the diminution, although merely
as a play in the inner voice (see the flute and also the first and second violin
parts), as if to celebrate the spirit of the arpeggiation. It is not until the augmentation of the arpeggiation of the third in bars (see the bracket in the graph
of the Urlinie) that the is reached and, with the two descending broken thirds
(for sake of contrast) in bars , finally and .
The repeat, bars ff, leads to the stretta by way of an expansion that relates
to bar .

17
At this stage in the development of Schenkers theories, the linear ascent to the highest Urlinie
tone is reckoned as part of the Urlinie itself.

18
By this notation, Schenker means that the descent from to is emphasized, by repetition (bars
, , ), before the arrival of (c2) in bar .

Bars ff. The stretta (Presto) in bars ff is also dedicated to the arpeggiation;
for this purpose it merely uses the motive of the closing subject and thus further
verifies its original meaning: c1 e1 (bars ), c2 e2 (bars ), finally g2
(bars ) as the , to which {} is attached the descent ( ).18 But it is not
until bars ff that the constant persistence of the arpeggiationin the transition to the coda, in the coda itself, and finally in the strettareveals its true and
deepest meaning: the motive of bars of the exposition returns, its first arpeggiation thus signifying the pinnacle of all previous ones, the goal of all goals. And
yet the whirl of arpeggiations is continued furthersee the imitation in the bass
and the first motive tears up to g3 (bar ). Still the arpeggiation drives on, until
the very end, in a mighty display, as if to proclaim that glory and honor belonged
to the triad alone: In the beginning was the arpeggiation; and so it is also at the
end. Does this celebration, however, not signify more than a mere bombardment
of VI cadences, for which the last group, from the end of the third [closing] subject onward, is commonly taken?

Nottebohm discusses only one sketch (Zweite Beethoveniana, p. ), which he

correctly assigns to the second transition (midway through the last movement),
corresponding roughly to bars .

tonw i l l e 6
From the corrections made by the master in the autograph manuscript, I shall
mention a few noteworthy features:

In bars , the alto and tenor trombones were higher; the same applies to the
violas in bars .

In bars , the exchange took place exactly at every quarter note.

In bars , the flutes were grouped together with the other instruments
that carried the melody.

In bars , the first bassoon and violas helped fill out the long notes with
 

their own motive, in the rhythm of .

The accompanying triplet figures of the second violins and violas in bar
each used a single note.

In bars ff, the violas originally played the same line as the cellos (as they do
later, in the development, bars ff).

In bars ff (including the upbeat), the bassoons took part right from the
fourth beat of bar . In addition, the first bassoon had a note tied over from
bar to bar ; the same thing was repeated at the fourth beat of bar .

In bar , the oboes originally played along with the flutes in sixteenth notes.

In the closing subject, the master made his way to the definitive instrumentation only with great difficulty. The changes show, above all, that he had in
mind the clarinet, and not the viola, as the actual leading voice. For the clarinet parts were marked unisono, whereas the viola (in bar !) deviated from
the tune by moving from g to d  (now in the second violin part).

In bar , the violas played a line, interrupted by rests on the first and third
beats, in contrary motion to the cellos; compare bars ff in the transition
to the coda. After he had crossed out this counterpoint, he assigned the contrary motion to the oboe and flute, thus introducing the wind counterpoint,
bars ff, in the most felicitous way. Moreover, this wind passage was originally intended for the strings.

From the fourth beat of bar , the timpani also took part, on G.

In bars the cellos originally remained on C, like the double basses.

In bars , the first violins originally followed the oboe part.

Shortly before the recapitulation, the timpani originally entered as early as
bar , though with just quarter notes. The violin eighth notes (b2) began as
early as bar .

In bar , the second violins had the same accompanying part as in bar .

And again in the reprise, we find the same intense struggle with the instrumentation as in the exposition.

The sixteenth-note figures in the first violin part in bars were originally an octave higher.
In bars ff, Beethoven originally attempted a polyphonic texture, that is,
with the first [recte: second] violins and violas also playing obbligato figures.
In bars of the coda, the horns originally played with the bassoons (as
we find later, at the repetition in bars ff), and the first violins, violas and
cellos played with the horn in bars . In addition, the motive played by
the flutes [recte: first flute] was originally assigned to the first violins.
In bars the violas played with the cellos and basses.
The horns [recte: trumpets] originally took part in the passage leading to the
stretta (bars ff).
{} Beginning in bar , the first violins kept to the high c3. But at the descending register transfer and arpeggiation, e2 g2 c3, in bars , the first
signs of edginess about ending the movement were assuaged. And again at c3,
in bar , the composer was on the verge of ending the movement: he over
came this temptation, too, and finally the third and last one, in bar .

Concerning the printed scores, the following should be noted:

. In the bassoon part in bar , the last three eighth notes should be played only
by the first bassoon. (That is what is written in the autograph score; the first
edition prescribes unisono.19)
. In the first and second violin parts in bars , the slur should cover both
bars. The same should apply to bars , and to the parallel passage, bars
ff. (The first edition actually goes against the autograph score by making
the slurring agree with that in the melodically important wind parts in bars
and ; from bar onward, it is in agreement with the autograph.)
NB. In bars the autograph shows different slurring in the otherwise identical bassoon and viola parts: slurs over the whole bar in the bassoon, but only
over the half bar (upbeat) in the violas. The slurs are brought into agreement only
in the parallel passage, bars , where the flute, oboe and bassoon play the
same notes. The first edition eliminates the difference in bars by assigning
the bassoon the same slurring as that given to the violas.

19
In the Stichvorlage, the copy of the score sent to Breitkopf & Hrtel for publication, Beethoven
actually added unis.

Beethovens Fifth Symphony (conclusion)

. In bars and , the first- and second-violin slurs extend only to the end of
the bar.
. In bars , , and , the slur in the clarinet and violas, which
lead the motive, and that in the second bassoon, which proceeds a sixth
below, should begin only from the fourth quarter [of the first of each pair of
bars]. The first bassoon should play the two half notes in bars , , and
non legato; only the first two quarter notes in bars , , and should be
legato. The first-violin slur in bars , , , , and should stop at the last
sixteenth note. Finally, in bar the viola slur should begin immediately with
the first half note.
(This is Beethovens notation throughout the exposition and the recapitulation, apart from a single slip, right in bars , where the clarinet slur actually begins with the dotted half note. And precisely this slurring was taken
up by the first edition as a model for all future instances; moreover, the edition misses out the articulation altogether in some instruments, and it provides the wrong slurring for the violin figures.
Beethovens articulation of the fourth-progression is advantageous for the
fp. And while the first dotted half note is sounding, a new half note is struck
on the off-beat, in the other instruments, which has the effectalbeit in an
artificial mannerof a kind of automatic connection of the chords.)
The first-violin sf in bar must be deleted.
. In bar , there should be a staccato on the last quarter note, B
, in the bass
parts (contrabassoon, cellos, and double basses), and likewise
. In bar , on the quarter note C. (The staccato is already missing from the
first edition.)
. In bars , the two D
s in the cellos and double basses should be tied over!
. In bar , there should be a new slur beginning at the note f in the flute, oboe,
and bassoon. (In the first editions, there are no slurs at all from bar onward.)
. In bars and , the slurs in the viola part should cover the whole
of each pair of bars; the cellos and double basses are slurred differently. (This
is the reading in the autograph manuscript; in the first edition, the cellos and
basses are slurred the same way as the violas.20)
. In bars , , and , there should be an f, not an sf, in the horn, trumpet, and timpani parts!21

NB. In the autograph score, the slurring is different in the viola and cello parts.
In spite of this, the viola articulation ought to be applied also to the cello (as is
the case in the first edition), if one considers that all the discrepancies have the
overall intention of expressing a legato. The same also applies to the repetition, in
bars ff. (Here the slurring in the first edition is also different in the two parts.)

he metronome marking  for the performance of the last movement

confirms what one probably feels: the four quarter notes of common time (not
alla breve) are most easily found by imagining four bars of the scherzo at a somewhat faster speed and, conversely, one regains the tempo of the scherzo at bars
ff, by slackening the pace of the quarter notes of the common time. If the conductor makes the effort to give emphatic accentuation to the longer notes, which
occur in the metrically weak bars, he will capture the essence of the melodic construction. A strong staccato attack on the fourth quarter in the cellos, double
basses, and contrabassoons in bars and , together with a crescendo from
the half-notes that follow until the longer C and D
in bars and , will
bring out the motive, even without any supplementary means of amplification.
The same applies to the trombones and horns in bars ff.
The climax in bars is achieved by bringing out the motivic imitations,
and by igniting the timpani strokes. It goes without saying that the C in the bass
in bar should be played louder, and with greater emphasis, than the preceding D
.
It is a feature of his art that Beethoven, for sake of more powerful climaxes,
takes care to create shadings even within a fortissimo, by dispatching the woodwind instruments first, in bars , and not adding the horns until bar .
Thus it is utterly against the sense of fortissimo for a conductor to add these arbitrarily as early as bar .
In bars , the augmentation of the motive of a third should be brought
out by a crescendo right up to the f 2 on the downbeat of bar . The same applies
to bars .
The three four-bar groups beginning in bar produce, on account of their
regular construction, a kind of restraint as the sempre pi allegro accelerates toward the presto (  ).
To conclude: All artifical means of strengthening, all recourse to instruments
that the composer himself did not specify, will prove dispensable to a conductor
who truly understands the musical text. He will triumph over all those conduc-

20

That is, the slurs in all these parts began only at the last quarter note of the first bar.
Beethoven actually wrote a thick f in the trumpet and timpani parts, canceling an earlier dynamic marking.
21

tonw i l l e 6
Lenz, on the second movement (p. ):23

tors who, for sake of a cheap brilliance, apply an instrumental makeup to the
score that runs counter to musical style.

The friendly but serious movement is based on two motives. . . . The first
theme (motive in bars ) combines quiet melancholy with a conscious
sense of grace (A
major). The triumph of the good, which awaits us, is
proclaimed in advance by the magnificent second theme (C major in bar
), supported by the timpani and trumpets.

E. T. A. Hoffmann writes of the Andante:

22

In terms of originality, it cannot be placed on the same level as the first

movement, even though the idea of introducing a stately theme in C
major, with timpani and trumpets, in a constant framework of A
major
has a striking effect. . . . All the themes of the Andante are very melodious, and the main theme is even congenial, but the very course of this
theme, proceeding through A
major, B
minor, F minor, and B
minor
before returning to A
major, together with the constant juxtaposition of
the major keys of A
and C and the chromatic modulations all express
once more the character of the entire work.

But if, according to Lenz, the second main theme does not enter until bar ,
what do bars mean for him? His eyes do not even notice these bars, not
even the error arising here; it is as if bar were followed immediately by bar .
And yet such blindness dares to venture into the furthest future, seeingfor no
other reason than the timpani and trumpets in bar of the second movementbeyond the third movement to the fourth and final movement. Oh, the
deceitful, smug, remote joys of hermeneutical blindness!
After the fermata on E
(bar ), the strings introduce a (first) episode in
which the amorous clarinet, followed by the melancholy bassoon, scatters
allusions to the [first] motive. Out of this develops an entracte (a parte)
in minor, performed by the flute, oboe and clarinet (bars ), which
resolves in the second theme, the triumphal theme (bars ff). The variations (antistrophes of the poem) are not lacking a Minore, declaimed by
flute, clarinet and bassoon in unison, embellished by the first violin, and
accompanied pizzicato by the remaining string instruments with the most
intense effect (A
minor, bars ff).

In his account of the third movement, which he calls a minuet, Hoffmann

speaks of G minor in bar (!), of G minor, D minor and C minor in bars ,
of E
major in bars , and subsequently of E
minor, B
major, B
minor, F
major, F minor, C minor, G minor, F minor, C major, and so on. The group of
bars ff appears to him as a new counterpoint in eighth notes (this note value
has not yet been used). He denies the relationship between the main theme of
the trio section and the first part, as it is revealed in the fifth-progression in bars
(); he also regards the continuation as fugal.
In between, he gives vent to his poetic heart in such terms as forebodings of
the realm of the spirits, eerie feelings, unpeaceful yearning, Terror of the
unearthlyof the fear of spirits, and many more such expressions. And yet a
more solid understanding of music, as opposed to the realm of the spirits, would
surely have brought him closer {} to the point at which he wished to extend his
spiritual antennae. One should not say that it is perhaps no more than a difference in theoretical outlook that he hears as an uninterrupted succession of keys
what is merely a collection of harmonies, or even transitional chords, within a
single key. After all, what would the poet E. T. A. Hoffmann have said if someone
had similarly torn apart the organic sentence constructions in his works and had
gone into raptures about the nouns, verbs, adjectives, and adverbs, as if they were
independent verbal entities?

Hermeneutical ignorance is easily pacified when it is permitted to stuff all that

stares it inscrutably in the face into a blinding verbal costume, for example, an
episode, entracte, antistrophes of the poem, and so on. I call this false reporting of the facts.
Lenz says the following about bars of the Andante:
Let us observe, before we take the instrumental epic further, how intimately Beethoven always connects the parts to the whole. In the accelerated pace of the Andante (bars of the pi moto), the oboe lets out a
short cry in minor (g
, f, f
), each note of which is preceded by two sixteenths (grace-notes). This figure is in anticipation of the fully outfitted
23
Wilhelm von Lenz, Beethoven: eine Kunststudie (Hamburg: Hoffmann und Campe, );
see vol. of this translation of Der Tonwille, p. , note .

22
E. T. A. Hoffmanns review of Beethovens Fifth Symphony appeared in the Allgemeine
musikalische Zeitung in ; see vol. of this translation of Der Tonwille, p. , note .

Beethovens Fifth Symphony (conclusion)

Riemann, in the Katechismus der Kompositionslehre, says the following about

bars of the Andante:24

second Allegro (scherzo), in close anticipation of the hymn of rejoicing;

see the wings that first violins, then the violas, grow using the same short
notes, when the scherzo returns after the trio section (maggiore) and in
the finale, forty bars before the final Presto.

Here we have harmony in abundance; in every individual bar [Taktmotiv]

except the first and third, there is a change of chord. In the second twobar group [Gruppe] a move towards the relative minor (half cadence on
c7) is achieved by the reinterpretation of d
6 as fVII; in the consequent the
original key returns by way of B
minor (0f is reinterpreted again as d
6).

Thus, according to Lenz, the grace-notes in bars of the second movement are related to the broken chords in bars ff of the third movement, and
with the sixteenth-note graces in bars ff of the fourth movement! This is how
the profundity of a Lenz imagines the profundity of a Beethoven: they have as
little in common as those sixteenth-note graces.

{} In the brief, quietly flowing eight-bar theme, which everyone no doubt

hears as an entity, Riemann thus hears not the single key of A
major, but a concatenation of no fewer than four keys: A
major, F minor, B
minor, A
major. (A
slightly different account is expressed by Riemann in his Grosse Kompositionslehre, vol. , p. .)25 Thus Riemann the theorist. As a melodist, Riemann hears
Beethoven the melodist as follows:

The third movement, marked Allegro, resumes the struggle in the first.
The powerful movement (not entitled Scherzo) has two themes. The
first (motive) is the mysterious ascent of the basses up the scale (C minor,
with upbeat g, bars ). The second, carved from three short notes
(quarters) and a long one (whole note) and profitably assigned to the
wind instruments, is a call-to-arms in the raw tones of the horns (bars
: later, all the winds take part). It takes time to summon the menfolk. The movement leads repeatedly to points of repose; with each step,
the designated battlefield gets nearer.

The sharply demarcated motives of subdivision [Unterteilungsmotive]

 , the frequent use of melodic motion both in the large and in the
small, should not be overlooked:
Melodic design of the two-bar groups

Hermeneutical carelessness: . . . two themes. The first (motive) . . .! Well, are

bars and themes or motives? Does a single argument help to eliminate the contradiction?

Melodic design of the individual bars:

The struggle that took place in the first movement, to occupy or not to
occupy, here addressing the question to be or not to be, heightened, inflamed, made rapturous by the basses (C major trio section, maggiore).
Only the intrinsically combative forms of a fugato are capable of representing the dualistic struggle.

If the reader now compares these with the graph of the Urlinie, he will
surelyand this is what really matterscome to understand, finally, that Beethoven could not possibly have hit upon the theme if he had taken the path that

There follows yet a gripping portrayal of the battle and all its reversals. As
everyone knows, the hermeneutistsevidently not inclined towards pacifism
take particular delight in precisely such representations, trumpeting their phrases
in the air. . . . I would remain neutral, since I consent to only one argument: is it
music, or not music? But I must ask, how was it ever possible that Lenz was ever
taken seriously as a musician by men who themselves wished to be taken seriously as musicians?
Spare me discussing the remainder.

24
Hugo Riemann, Katechismus der Kompositionslehre (Musikalische Formenlehre), vols. (Leipzig:
Max Hesse, ), vol. , p. . When this book was republished around , as the Grundri der Kompositionslehre (according to the title page; the front cover gives Handbuch der Kompositionslehre), the
chord symbols shown in the music examples (vol. , p. ) were changed from literal namesas+(A

major triad), des6(sixth-chord above D
), fVII(seventh chord reading down from F, FD
B
G), and
so onto their functional equivalents, viz. T (tonic), S (subdominant), 0S (subdominant of relative
minor), and so on.
25
Riemann, Groe Kompositionslehre, vol. (Berlin and Stuttgart: Spemann, ). In this later account of the opening eight bars, the chords are now identified by their function (T, S, 0S, etc.).

tonw i l l e 6
peated with a return from C major to A
major (half cadence on E
). Its
characteristic motive is played by the winds (music example: bars ).
This simple material is varied several times and worked out, in a lengthy
coda, in the manner of a development section.

Riemann imputes to him. That mysterious entity which, secured by the Urlinie,
guides all the individual melodic movements is so different from that which a nongenius or an uncreative person regards as melody in general and melodic traits in
particular: two different worlds, which never come into contact with one another.
On the third movement, we read, for example, the following from Riemann:26

Now, one cannot say that any of this is exactly transparent. Which are the
two thematic ideas? Is the second theme found in the concluding phrases, or
not until the episode? And isnt Riemann actually saying that the coda begins
as early as bar , well before the third variation (bars ff) appears?
The themes of the scherzo are discussed by Riemann in Thayers work exactly
as they are in the Katechismus der Kompositionslehre, with a few hermeneutical
pennants tied on, such as somewhat martially applied, conjuring the vision of
a nocturnal troop review, making sinister noises. He dwells all the more comprehensively in portraying the exterior fate of bars (see earlier).
He foregoes discussing the design the finale, and merely uses it as an opportunity to take up a position against the hermeneutical yearnings (he, Riemann!):

In the Scherzo of the Fifth Symphony, three distinct themes may be discerned: theme (music example: bars ); theme , in the main key
(music example: bars ); theme , in the parallel major [Quintwechseltonart] and treated fugally (music example: bars ). Although the
first theme starts with just two short upward runs (two four-bar half periods, both ending with half cadences, the second expanded after the second bar), its further use leaves us in no doubt about its meaning, as both
themes of the main section take turns; in the course of this, however, the
first gets longer each time it is played. The second theme steers immediately towards the parallel key of the relative minor (E
minor), the first
theme leads by way of B
minor and F minor back to C minor, from which
point the second theme returns to F minor before the principal key is finally established clearly with statements of the first and second theme.

Yet all efforts at a programmatic interpretation of the work are vague and
far-fetched. . . . What right do we have to call it a tragedy of fate? To determine the applicability of this term, we would need above all a recognizable catastrophe; this, however, is not demonstrable. {} E. T. A. Hoffmann, in his inspired review of the symphony for the Allgemeine
musikalische Zeitung (July , unsigned), steers clear of decoding the
work, but clearly runs into opposition from those who have wanted to see
in it the products of an unbridled imagination. Admittedly, he regards
Beethoven most definitely as a purely Romantic composer, but, compared to the master, ranks Haydn and Mozart distinctly beneath him in
terms of contemplative thoughtfulness [Besonnenheit] and notes, very
astutely, that only by the deepest engagement with the inner structure
of Beethovens instrumental music can one reveal the high degree of
thoughtfulness, which is inseparable from true genius and which is nourished by the continuous study of the art. If one realizes how long Beethoven struggled with this very symphony, then one will understand how
right the gifted critic is to speak of a high degree of thoughtfulness.

In order merely to determine this, it is perfectly sufficient to shut off ones

ears and simply let the eye report what it sees; this is something that a layman can
do equally well. And yet, even this comfortable account is denied by Riemann: in
bars ff, even his eyes cease to function and he entirely overlooks the descent,
the necessary reverse side of the ascent, so keen is he to pursue only the first and
second themes.
In Thayers Beethoven, Riemann writes (vol. , p. ):27
The Andante con moto in / is a piece of enchanting simplicity and
transparent construction. Its two thematic ideas comprise the eight-bar
idea (music example: bars ), which, however, secures its most intimate
charms in its concluding phrases (music example: bars ). An eightbar episode leads from the ending on A
major into C major, and is re26
Katechismus der Kompositionslehre, vol. , pp. ; Grundri der Kompositionslehre, vol. ,
pp. . In both editions, Riemann erroneously identifies the symphony as the Eroica.
27
Alexander Wheelock Thayer, Ludwig van Beethovens Leben, continued by Hermann Deiters and
completed by Hugo Riemann, vols. (Leipzig: Breitkopf & Hrtel, ). For the background to
this publication, see vol. of this translation of Der Tonwille, p. , note .

In requiring the deepest engagement with the inner structure of Beethovens

instrumental music, Riemann is thus, as one can see, in agreement with E. T. A.
Hoffmann. I, too, concur with this, except that I believe that this requirement
must be fulfilled differently.

Beethovens Fifth Symphony (conclusion)

Kretzschmar (Fhrer duch den Concertsaal, vol. [second edition, ]) at

first shares Hoffmanns interpretation of the Andante; he even shares the entire
sentence In terms of originality . . . a striking effect word for word, but without
using quotation marks (see above, under Hoffmann). But then he aligns himself (with like concealment) with Lenz in interpreting the C major episode as a
reference to the finale and its sphere of joy.
A more original interpretation of the third movement leads Kretzschmar
merely to the following hermeneutical result:

Regarding the third movement he offers the following judgment:

It is constructed in the usual form of Scherzos, with a Trio and the ordinary repeats and interchanges; and yet while adhering to these general
lines, Beethoven has departed so much from the usual proportions as to
show how far such prescribed forms can be modified without interfering
with the unity, the symmetry or the impressiveness of the whole.
That, most certainly, is not English cant but, rather, a more frightening thing:
the innately spiteful megalomania of an utter ignoramus who immodestly desecrates the modesty of the genius, which is tamed by the subject matter, because
he has lost all sense of his remoteness both from that matter and from the one
who has mastered it. At the same time, the megalomaniac illustrates a practical
skill: the one who is incapable of anything raises himself up in words to the level
of the genius, and beyond the genius, and thus spares himselfunconsciously,
and in the act of deceiving the readerthe trouble of saying what were to be said
if he knew what to say.
Bekker, whose reading of the first movement (see above) had ended with the
words: Fate has triumphed,29 continues the program in introducing the individual sections of his further account: For this world, at any rate. A new one appears in the second movement. The spirit flees from brutal reality, to the world of
more intimate visions; and so on.
It is merely a rising-up in a dream, not on solid ground; and so on.
It was promise, not fulfillment. Ones eye is again cast down to the ground,
searching, questioning, doubting; and so on.
A grotesque dance of goblins begins; and so on.
[In the last movement] no fewer than four themes follow each other, always
filled with the same, impetuous power. And yet not all hindrances have been
overcome; and so on.
Although definitely inferior to the Eroica in terms of originality, the C minor
symphony is superior to the earlier work on account of its taut construction and
the close relationship between its parts. And so on.
I do not believe that Bekker would make pronouncements about a shoemaker

The rather close kinship between the third movement and the first, as we
so often find in Beethoven, is expressed here with particular clarity. On
the surface, it may be seen in the similarity that exists between the horn
motive and the principal rhythm of the first movement; further, in the
numerous fermatas, a feature common to both movements; and still
more deeply in the primarily somber character of this Scherzo. In the
principal section, only the rhythm is of a cheerful disposition; the harmonies appear dispirited; and the melodies are questioning and melancholy, with a strangely distant quality on account of the sound of the instruments that play them in the most important places.
Things continue along similar lines, with the difficult questions concerning
form completely ignored:
Concerning the fourth movement, Kretzschmar discusses only the themes.
On these, he remarks: The themes are simple, to the point of being trivial. On
all other matters, the hermeneutist is silent; he may be excused: as a hermeneutist
he will, and must, end on a high note, bloated as he is by verbal gases.
Of Groves particular art of hearing nothingwhich only makes him fonder
of writingI shall merely transmit a few random samples. He calls bars ff of the
Andante a consequent phrase. Regarding bars ff: We now hear a further
theme (?). Bars ff: a wonderful coda (!). On bars ff: Then (?) the clarinet
and bassoon give out a motive derived from the main theme. On bars ff: After
a very short preparation, the triumphant song now begins afresh; and so on.28
28
Grove, Beethoven and his Nine Symphonies (London: Novello, ); Schenker is using the
German edition prepared by Max Hehemann, which is far from being an exact translation (see vol.
of this translation of Der Tonwille, p. , note ). Little of what Schenker quotes concerning the Andante matches Groves original wording (though Grove does, in fact, call bars ff a Coda of great
beauty). But since the opening general remarks on the third movement have been translated more
literally, I have reproduced Groves original English in the extract that follows.

29
Paul Bekker, Beethoven (Berlin: Schuster und Loeffler, ), Eng. trans. by M. M. Bozman (London: Dent, ). I have translated the original, rather than adopt Bozmans translation on pp.
; see vol. of this translation of Der Tonwille, p. , note .

tonw i l l e 6
or a tailor30 of the sort he made in the last sentence I quoted about Beethoven,
not even if he were more knowledgeable about footwear and clothing than about
music. That is, to say it yet again, the unhappy tenacity of ignorance. He concludes his hodgepodge of hermeneuticsand anyone who suspects me of taking
Bekkers sentences out of context would first have to undertake a proof of this
with the following summary judgment:

live the Emperor! is proof that this symphony must bring and say something to anyone with imagination.
Is that which the symphony has brought to Bekker truly a something? That
which, in his view, is hardly possible to fail to understand, or to misunderstand
has become possible in his very hands. And one can understand why he pursues
the so-called new music from all directions, music whichif one permits the
play on wordsis so designed that, coming from the utterly ungifted, it can
bring something only to the utterly ungifted.
One sees nothing but error in all these interpretations. The mistakes themselves
could be excused; what remains inexcusable is merely the secret pleasure of an immodesty that ingratiates itself so very cheaply in the proximity of genius. To be sure,
the work of a Beethoven, being the work of man, addresses mankind. Should that
itself be sufficient grounds to justify, conversely, mankind addressing the work?

Per aspera ad astra31 this theme, less original but surer to make an impression, is interesting on account of its being generally easier to understand, even for those with less finely tuned senses. And the four stages of
developmentstruggle, hope, doubt, victoryare illuminatingly and
convincingly marked out with such directness that it is hardly possible to
fail to understand, or to misunderstand it. The French grenadier who was
moved to cry out at the beginning of the finale, Its the Emperor: long

30
The comparison of music journalism with two intellectually less demanding lines of employment is a deliberate play on Bekkers name (Becker baker). Schenker kept some papers relating to
an earlier quarrel with the critic in a file entitled Bekkerei (bakery). See also the Appendix to this
volume (pp. ).
31
A phrase derived from Seneca, meaning that there is no easy path to the stars (i.e., to wisdom,
or truth).

True Performance
Der wahre Vortrag {Tonwille , pp. }
t r a n s l at e d b y r o b e r t s n a r r e n b e r g

There is no epoch of human history in which even one ingenious creative

The most serious hindrance at the present time lies in a craze that has been
pushed to extremes, wherein the right to a personal conception in artistic matters is derived from the same right that each person improperly claims in political and social life. Yet mankind has not even the slightest inkling that being involved in right and duty, as also in freedom and morality, is solely the prerogative
of geniuses who gain access to such superior phenomena and such extremely difficult concepts of humanity-synthesis by means of the experiences forged in the
synthesis of their own creations. Since, however, it is given to others to presume
they have the capacity for such involvement when they speak of those virtues
with mere borrowed words and yet prove the complete opposite in all that they
do and leave undone, nothing would be more natural than that, instead of tracking down self-deceptions and liberating themselves from them, they became even
more entangled in immoral misunderstanding and then carry their ostensibly
personal rights over into art without any restriction and assert those rights in the
very place where masters of synthesis, in harmony with nature as well as with the
eternal demands of art and ethics, established the rights and duties of tones in an
{} incomprehensibly high manner that far surpasses the mental powers and
abilities of others.
Thus a genius like Brahms could express himself in this way: Whenever I
play something by Beethoven, I have no individuality whatsoever, insofar as the
piece is concerned; instead I strive to reproduce the piece as Beethoven prescribed, and I then have enough to do.1 On the other hand, the many millions of
men who stand beneath Brahmshow few there are that stand above him!
emptily requite Beethoven with their personal rights, with their feeling and their
conception, and in so doing obviously have quite enough to do. And those who
might be dubbed the higher-standing among the low-standing, namely, the

achievement was understood: essence and core invariably founder upon the inadequacy of the generations. What emerges, even in the best of cases, is perhaps
an entirely external relationship, but for the most part it is only perversion and
distortion. What founders of religion and great poets, musicians, and thinkers
conceived and taught, being confined to the inner life of the idea, thus rests apart
from mankind, whose spirituality consists less in an actual result or increasing
benefit, than in a joy that derives from the impulse to imitate and pretends to be
spiritual reception.
It would constitute a challenge to nature were one to expect of her a wholesale improvement of the human race: submitting oneself in humility to the laws
of nature is the first commandment of morality and knowledge. But there have
surely been moments in the life of mankind when peopleonly, of course, under
the leadership of an individualshowed something like soul-searching and awe
before an uncomprehended greatness of spirit. Never was there such a moment
more suited than the present for calling people once again to their senses, for
today they are without doubt fully aware of their unending misery, without of
course being able to explain the causes thereof on account of their lack of depth.
The unprecedented breakdown of mankind provides an opportunity to clear
away the rest of peoples mad ambitions, until they become as humble as need be
to make them into a vessel of spirituality.

It is certain, and may well be obvious at all times, that no start whatsoever has
been made on a true assimilation of the virtues offered by the great German masters of music. In fact, the gulf has continued to widen from the time of their first
presentation down to the present day, hence the prospect of being able to start in
earnest on assimilation is diminishing.

1
[S.] Max Kalbeck, Johannes Brahms, vol. , part (Berlin: Deutsche Brahms-Gesellschaft, ),
p. .

tonw i l l e 6
much-celebrated performers and editors, even make so bold as to bring their personal conception into the work and thereby adulterate its content.2
With due deference to the feeling of someone like Brahms, who above all
subordinated himself, what do we make of the feelings of the poor in spirit who
still insist upon their own worth even in the presence of a master like Beethoven?
Is it not truly ludicrous? For would it occur to a man who is obliged and accustomed to a poor and meager diet to stuff his poor victuals into his pocket and
take them along if some rich man, out of the kindness of his heart, invited him
to dinner? Does not the hospitality offered by the rich man preclude any assistance from the poor guest? What shall we then say about the wretch who takes a
seat at the sumptuous spiritual table of a genius and suddenly gets it into his head
to display his usual poor mess for all to see!
But should we perhaps excuse ignorance merely because it is ignorance and
accept its arrogant pride with some measure of indulgence, even in such repulsive forms and dreadful consequences for art as are seen in the present day? Now,
then, just as there is no room for tolerance when it comes to implementing the
knowledge that two times two is four, so too when it comes to implementing the
knowledge of the artworks equivalent of two times two. To be sure, attending
school, the intellectual aspect of it, ultimately becomes a burden to most childrenincidentally, can one reproach the teacher who teaches reading and arithmetic by saying he pursues theory, mere theory?and they would like to leave
that place of compulsory duties in order to plunge headlong into a life that frees
them from a more complex and difficult use of the mind. If even the least of these
in like manner wanted to learn the true way of reading musical notationas if
there were a theory in art that was not also the art itself!is it then appropriate to grant them the right to express their emotions in the face of art, while at
the same time denying the millions who neither can nor want to learn reading in
school the right to make emotional judgments about intellectual matters?

is in general: a highly developed constraint chosen in freedom by a mind that

fully understands the material. Goethe had a notion of the complete uniformity
of the concept of freedom: Only law can give us freedom. {} As did Schiller:
The harsh fetters of the law bind only the servile mind that disdains it. Again,
therefore, it is the genius who, through his own practical experience in synthesis, is also led to the concept of a freedom in performance that is born from constraint. But with mankind it is different, for people prefer to understand freedom as the antithesis of any constraint. And again they bring this false
conception over into art where, under the banner of freedom, they let their
completely unfounded personal conception hold sway in a similarly unfounded
succession of moods. One sees how life and art flow in confusion and what injuries art suffers when unrefined concepts of life come into play, and, by contrast, what blessings life could receive when refined concepts of art find their
way to it.
But of course, the exigencies of performing a masterwork are so great that
even the most freedom-loving performer looks for the aid of some sort of constraint, indeed, he must look for it. What sort of constraint is this? Meter and its
markings are the first to catch his eye. He does not trouble himself much about
how they arose and what they mean in the notation, but instead struggles, with a
veritable torrent of superstition, in a frame of mind that would expect to find the
earths surface inscribed with the degrees of latitude and longitude. And he does
all that only because he does not grasp that loftier constraint, as is guaranteed
solely by the synthesis of a masterwork; for in order to detect that constraint and
freely choose it he would have to look beyond the tenth, the twentieth, or even
the hundredth bar line. But can he do that?

And if these hindrances were not enough, performance is even deprived of the

opportunity afforded by a correct text. I have descried this deplorable state of affairs in all my writings and will not tire of continuing to do so. How, I ask, shall
it be possible to achieve a truly beautiful performance when the masterwork, the
sole source of the performance, has been so adulterated in modern editions?
Compare, for instance, the following passages as they appeared in the composers
autograph or first edition (shown as a) and as they appear in modern editions
(shown as b).

The second obstacle to a correct performance of the masterworks is ignorance

of what it is that constitutes freedom of performance. What is freedom of performance? It is the same as freedom in morality or politics. It is what freedom
[S.] See [my essay Die Kunst zu hren (The Art of Listening),] Tonwille , pp. ff./I, pp. .

True Performance

{} If the performer has only these modern editions to look at, is it possible
for him to obtain even the slightest idea of Mozarts vitality and variety and learn
to employ the means for expressing such variety in performance, or to discern
Haydns ingenious pianistic inspiration, perhaps the sole example of its kind? For
do such editions in any way give impetus to the special type of performance like
that which the original commands? If it is already difficult enough to satisfy the
original when it is in front of your eyes, how far short of the mark must one fall
when one is not even aware of how the original passages appear!
Thus performance ultimately comes down to the question of printing. Having recently completed my edition of Beethovens sonatas, I will say what has to
be said about them in particular in one of the next issues of Tonwille.3 In the interim, let Brahms speak:

Mozart, Sonata in A, K. , first movement

With your permission, I will tell you just between ourselves that the accursed instructional editions of the classics generally have little or nothing
to do with art. They only matter to the purveyors purse. For this purpose,
however, the publisher is advised to come up with a quite famous and very
striking name. I, for my part, know of no instance of an edition without
such a name that is distributed beyond the city in which it is contracted or
produced. . . . We continue to phrase and mark like our classic composers.
It is not that we simply manage to make do with this, but rather that we
desire nothing else, despite the fact that we do not think as cleverly as
Mr. R. about phrases and the like.4 We even have reason now and again to
do away with the bar linebut would Mr. R. have the temerity to do that
if he wanted to be consistent? {} I need not say (nor regret) that in general, of course, there is now much interest in this sort of nonsense. Blow,
Tausig, Lebert-Stark,5 and others are admired and no one has any inkling
what a Bach edition by Czerny deserves by way of respect. My own nonsense need not trouble you (I am most likely a philistine), but I cannot
offer my congratulations, as you see.6

Mozart, Sonata in C, K. , finale

3
Although Schenker never wrote this afterword to his Beethoven sonata edition, a longer essay on
the editorial corruption of musical masterworks appeared in the first volume of Meisterwerk (),
under the title Weg mit dem Phrasierungsbogen (Abolish the Phrasing Slur).
4
Mr. R undoubtedly refers to Hugo Riemann.
5
Sigmund Lebert ( ) and Ludwig Stark () were coauthors of the Groe theoretischpraktische Klavierschule (), a popular and widely used piano method.
6
Johannes Brahms Briefwechsel XI: Johannes Brahms Briefe an Fritz Simrock, ed. Max Kalbeck,
vol. (Berlin: Deutsche Brahms-Gesellschaft, ), pp. . The letter is dated September , .

Haydn, Piano Trio in E, Hob. XV:, first movement

tonw i l l e 6
Only one thing still remains to be said here: At my instigation, some publishers have decided to issue facsimiles;7 only when a larger collection of these impressions becomes generally available will people be able to convince themselves
on their own of the devastation of the originals, provided, of course, that in the
meantime they will have made progress in their ability to read music.

7
Schenker is thinking above all of the Musikalische Seltenheiten (Musical Rarities), under the
general editorship of Otto Erich Deutsch, published by Universal Edition. The first volume of this series was Schenkers own facsimile edition of Beethovens Moonlight Sonata (); the fourth volume, by Deutsch, on five early Schubert songs, was the point of departure for Schenkers essay on
Gretchen am Spinnrade.

Miscellanea
Vermischtes {Tonwille , pp. }
t r a n s l at e d b y i a n b e n t a n d w i l l i a m d r a b k i n
On Schuberts Meeresstille. Recently I read a poem, just published, entitled Gipfelgesprch [Mountain-top conversation], which begins:
Der Abend gebietet dem Land.
Schweige.Die Nacht ist nah.

Evening bids the countryside.

Be silent.Night is nigh.

Warte nur, balde

Ruhest du auch.

In a further twenty-six lines, rivers, roads, the horizon, towns, towers, darkness,
colors, deep sounds, peace, woods, meadows, mute deer, gorges, valleys, and peak
after peak are all called upon to witness with them the evening stillness.1
Wandrers Nachtlied [Wanderers serenade] and Meeresstille [Ocean stillness]
by Goethe came unprompted to mind. Whereas the poet of Gipfelgesprch awakens a constant stream of new images in our imagination and, by so doing, deprives us of peace even more, Goethe proceeds the opposite way, withdrawing
images from our imagination so that once we have achieved inner stillness, we
imbibe through the stillness of our own mind the stillness of Nature. It is as if he
were snuffing out one realm after another:

Over every mountain top

Is peace,
In every tree top
You can detect
Hardly a breeze;

2
Albert Bielschowsky (), literary scholar, biographer of Goethe and of Friederike Brion,
author of Goethe, sein Leben und seine Werke (), which formed the basis of Goethe studies
for many years. Oskar Walzel ( ), literary critic, author of Die deutsche Dichtung seit Goethes
Tod (), and editions and studies of Heine, Chamisso, and other writers.
3
Johannes Messchaert (), pupil of Julius Stockhausen, Dutch baritone specializing in
Lieder and music of J. S. Bach. It was through his regular accompanist, Julius Rntgen, that Schenker
came into contact with him; Schenker took over as accompanist on Messchaerts concert tour of
major Habsburg cities in January and February . Messchaerts interpretations were later set down
in two volumes: F. Martienssen, ed., Auswahl von Messchaerts Schubert-Repertoire, mit seinen Vortragsanweisungen (Schott: Mainz, ). On Messchaerts relationship with Schenker, see Federhofer,
Heinrich Schenker, nach Tagebchern und Briefen (Hilesheim: Olms, ), pp. and .

the nearer world of the ear:

Die Vglein schweigen im Walde

Just wait: soon

You too will find peace.

The same process is at work, too, in Meeresstille. Here the poet prefers to
intone several times the same negation: Deep stillness reigns over the waters,
Not a breeze from any quarter, Fearful, deathly calm, The sea lies motionless,Not a wave stirs in the vast expanserather than introducing new images
and so driving our minds away from the stillness.
Only the person who can appreciate Goethes artistic concept of expressing,
even in a poem, of which the word is the very medium itself, the silence of Nature through silence of language, and not through the piling up of illustrative
words and ideasthe magic of the poem lying thus not in antithesis, nor yet in
the sonority of the word alone, as many mandarins of the art-world (Bielschowsky, Walzel,2 etc.) maintainonly that person will discover comparable mastery
in Schuberts setting of this poem. Schubert, too, lulls us into peace by taking care
not to excite the ear by piling up ever-new musical ideas.
Permit me, in this connection, to remember, alongside that great poet and
great composer, the great singer Johannes Messchaert,3 recently deceased, who in

the realm high above us,

ber allen Gipfeln
Ist Ruh,
In allen Wipfeln
Sprest du
Kaum einen Hauch;

and, with a glance down at the earth, at the soil that forms our collective
grave, finally life here below:

The small birds cease their singing in

the wood.

1
The poem, by Rudolf G. Binding, was published in the Frankfurter Zeitung on December ,
. Schenker kept a clipping of it in his personal copy of Tonwille (OC, Books and Pamphlets ).

tonw i l l e 6
his interpretation of Meeresstille achieved the corresponding effect by elongating
the notes in so utterly peaceful a way that they sounded like messages emerging
from the silence rather than notes issuing from the human breast. Only at the
word ungeheurn [immense] did any tone creep into the voice, swelling almost
imperceptibly, as far as ungeheure Weite [immense distance], but pulling back
immediately. The effect was to render the silence at the end all the more fearful
and believable.

What is more, the state today has the temerity to flout nature. From nature
which is every bit as much concerned with the waning of the generations as with
their waxing (just think what the generations would do if they were preserved
from now to the end of time like geological strata!)it dares to clutch the masses
convulsively to itself, and to protect them from dangers that they nevertheless repeatedly court through their own guilt. It thus attempts the impossible, while
merely shrugging its shoulders at the chosen spirit.
And so Schubert is proved right. The state will not be a state worthy of the
name until it at last comes to value those on the other side of the grave, the chosen spirits, for bestowing upon the millions of other people spirit as well as
money, soul and honor as well as prosperity; until it at last undertakes to impress
firmly on the beneficiaries that they are in fact beneficiaries, instead of encouraging the rapacity, arrogance, sloth, and bone-idleness of the people, of the rabble, at the expense of the spiritual ones.

State and Genius. The state should provide for me. I came into the world for one
thing only: to compose, Schubert once remarked to Anselm Httenbrenner.
Now the reverse has come true: Schubert brings money and honor to the state.
The states intellectual capacity has not progressed far enough to appreciate
how things stand: unscrupulously, without a qualm, it pockets Schuberts capital.
To be sure, its officials know all the industries, businesses, professional craftsmen,
and artists involved in marketing Schubert wares. But as to Schubert himself,
who created this inexhaustible capital investment and who is more impregnable
than all the industries of the world, of him they are oblivious as a co-producer of
national wealthin their current jargon, as a captain of industry. In the account
books of the state, the name of Schubert is nowhere to be seen.4
Moreover, the state has its mind fixed on this side of the grave, solely on those
earthly things to which the spirit already attaches a meaning beyond the grave.
Nowadays too, the Central European world is doing this, under the influence of
the West Huns (Lichtenberg5), who use their {} so-called democratic institutions merely as a pretext for spreading economic and moral servitude all around,
but without offering in return any alternative to goods of true spirit, which, seen
sub specii aeternitatis, they have in fact always lacked.
This state lives by commerce alone. It resembles more an unruly, jostling mob
than a community of people that values the things of the heart and mind. But
then, all money is short-term, whereas genius is long-term. So how could the
state, which lives by money alone, possibly live at the same time by the spirit?6

One has to grow old to appreciate all this, and have enough money to pay
for the lessons one learns. Every bon mot that I utter costs me a purse full
of gold. Half a million of my private fortune passed through my hands as
I learned what I now know, not only my fathers entire fortune but also
my salary and my sizable literary income for more than fifty years. On top
of that, I have seen a million and a half outlaid on grand ventures by
princely personages with whom I was closely associated, and in whose
progress, successes, and failures I took part.
Goethe to Eckermann,7 February ,


For myself, I must also say that I deem it true that humanity will ultimately be victorious, but I fear that by that time the world will be one big
hospital, and that we will all have to be each others human nurse.

4
[S.] Cf. Beethovens Neunte Sinfonie, pp. xvii ff./pp., . [On these pages, Schenker discusses
the economic importance of Beethoven and other Austro-German composers.]
5
Possibly Georg Christoph Lichtenberg ( ), physicist, also literary, philosophical, and
satirical writer.
6
Cf. Man does not live on bread alone, but on every word that comes from the mouth of the
Lord. Deuteronomy . and Matthew ..

Goethe, Italienische Reise, ii: Naples, May ,

7
Johann Peter Eckermann (), German writer, who published two volumes of his Conversations with Goethe in and a third in . They cover to Goethes death in .

Miscellanea

place, is not the daemon9 of a Socrates or of a Goethe, the clairvoyance of the

great composers and poets, philosophers, etc., more than all that pinching and
nipping, and airborne hovering by losers?
Loser sticks with loser. But how is he to cope with the genius?
Schuberts friend Vogl10 writes somewhere:

Occultism. Mediumism, materialization, telekinesis, cryptaesthesia, ideoplasty,

metapsychics8 Oh, what uncouth mouthfuls the fashionable new occultism has
sprouted! Only when it comes to the occultism of the powers of genius, which
ought to appeal to men more than anything, is there total silence.
Who, then, are these beings, allegedly endowed with medial powers? But
all our mediums, writes a spokesman for occultism, are from the ranks of genuine spiritualists. So they are children of the multitudeof that multitude of
which Schiller speaks:
Majestas populi
Majestt der Menschennatur! dich
soll ich beim Haufen
Suchen? Bei Wenigen nur hast du
von jeher gewohnt.
Einzelne Wenigen zhlen, die brigen
Alle sind blinde
Nieten; ihr leeres Gewhl hllet
die Treffer nur ein.

But when it comes to constructing, producing, creating, I count myself

out, especially ever since I learned from Schubert {} that there are two
sorts of composingone that, as in the case of Schubert, comes about in
a state of clairvoyance, or a trance, without any deliberation on the part of
the composer as to what he must do, through higher power and inspiration. Such a work may invite astonishment and delectation, true, but emphatically notjudgment; and a second sort that reflected, etc.

Majesty of the People

Majesty of human nature! Am I
to seek you in the crowd?
Until now, you have dwelt
only with the few.
The isolated few count, all the
rest are blind losers;
their empty bluster merely
enshrouds the winners.

As we can see, Vogl was genuinely convinced of Schuberts clairvoyance; yet he

was not above thinking he knew better than the clairvoyant Schubert on countless
occasions, making alterations to his songs which have long been recognized as
changes for the worse. What are we to make of this? Dearer still to Vogl the singer
than Schuberts clairvoyance was his own vanitynone of Vogls undoubted
virtues can alter thatsuch that he suddenly believed he could see more clearly
than Schubert and desecrated the latters precious wares: the vanity of a loser.
The same is true in greater measure when we contemplate the behavior of the
Germans toward their own clairvoyants, their own geniuses. Not a note of the German master composers do they have in their heads; they never for a moment think,
moreover, that it is only Bach and Handel, Haydn, Mozart, and Beethoven, etc.,
and not they themselves, who represents the German people. And instead of making the powers of these spirits their ownthe sum total of which offers sufficient
in fortune, prosperity, and pride for all generations of Germansthey incur the
most grievous losses in art, honor, and possessions in the process by wafting incense to the glory of spirits from other regions and peoples, which, in whatever
realm they may operate, do not begin to compare with those mighty spirits.
Were the mediums to succeed even just once, despite the limitations of their
own brand of clairvoyance, in raising a genius from the dead, what an indict-

These losers pinch and nip, hover airborne in their sances, dig up bits of the private lives from which they come and to which they belong; Rudi speaks of Minna,
Minna of Otto, and so on, and the other losers come flocking to them, mouths
and eyes agape, hooting, spouting mumbo-jumbobut when it comes to genius,
all are deaf and dumb.
Should we entrust the medial powers of nature solely to the losers? Is not
the genius, too, in her eyes, a human being? And, whether genius or non-genius,
is not man, even when he is not pinching and nipping, or hovering airborne, a far
greater wonder than the science of the occult imagines him to be? And, in the first
8
Mediumism: communication with the spirit world using a person as an intermediary; materialization: causing a spirit to appear from the dead in bodily form; telekinesis: the apparent production
of motion in objects without contact or other physical means; cryptaesthesia: supranormal perception; ideoplasty: the transmission of thought at a distance, having a physiological result such as feeling pain, or falling asleep; metapsychics: the study of psychic phenomena beyond the limits of orthodox psychology. Many of these terms are discussed in the chapter Fragliches (Questionable
things) from Thomas Manns Der Zauberberg (The Magic Mountain) of , which is contemporary with these Miscellanea. (Tonwille , though dated , was not published until April , .)

Daemon: a lesser deity.

Johann Michael Vogl ( ), baritone in the Vienna Court Opera, friend of Schubert
, and the leading interpreter of his songs.
10

tonw i l l e 6
Immediately below this Beethoven wrote:13

ment that would be of the riff-raff, which wants nothing but its own kind! But
no such raising from the dead will ever succeed. In fact, it is unnecessary: in his
work, the genius defies judgment.
In the meantime, loser clings to loserthe rabble hoots, gorges itself, and
procreates.

Should one exchange compliments with such a scoundrel who steals ones
money, or instead pull him by his asinine ears.
{} (And on the verso:)

Slovenly copyist!
Stupid fellow!
Correct the mistakes you made through ignorance, arrogance, conceit,
and stupidity. That would be more seemly than to attempt to teach me.
For that would amount to no less than the sow attempting to teach Minerva.14

BeethovenWolanek (Beethovens Smtliche Briefe, vol. , p. ).11

To Herr Ludwig van Beethoven!
Since I am unable to copy out the finale in score until Easter, and
since you will no longer need it at this time, I am sending you, together
with what I have already begun, the complete parts for your favorable
disposition.
I remain gratefully obliged for the honor rendered by the employment
you have offered me. Concerning further your otherwise inharmonious
behavior towards me, I can only smile, accepting it as a good-natured outburst. The musical world of ideas is governed by so many dissonances; why
should this not also be the case in the real world?
All that comforts me is the firm conviction that, had those celebrated
artists Mozart and Haydn been employed by you as copyists, they would
have shared the same fate as mine.
I request, merely, that you do not group me with those common copyists, who consider themselves happy to be able to maintain their existence
even by being treated as slaves.
Moreover, be assured that I will never have causenot even a grains
worth of causeto blush in your company on account of my behavior.
Respectfully yours,
Ferd. Wolanek12

It was already decided yesterday, if not even earlier, that you would no
longer be writing for me.15
You should do Mozart and Haydn the honor of not mentioning their
names.16
(And once more on the first page Beethoven crossed out Wolaneks entire letter
and wrote the following on it in large letters:)
Stupid fellow, conceited ass!

Since then, the copyist Wolanek has multiplied into millions of Wolaneks! Beethoven, however, remained the Irreproducible! Millions of Wolaneks will continue to come and, in various guises, play, conduct, write their daily pieces, interpret, and teachbut Beethoven will remain unique. If only all the Wolaneks
would take note of Beethovens words: You should do Mozart and Haydn the
honor of not mentioning their names.
13
The composers response, written on Wolaneks letter, is assigned a separate number () in
Beethoven: Briefwechsel: Gesamtausgabe and classified as a draft reply. In The Letters of Beethoven, ed.
and trans. Emily Anderson (London: Macmillan, ), which includes only correspondence from
Beethoven (not to him), this response is included, as letter .
14
An ancient Greek saying, equivalent to teaching your grandmother to suck eggs, attributed to
Demosthenes by Plutarch (Demosthenes, part ; see also his Moralia, d). Beethoven had used this
expression earlier, in a letter to Joseph Karl Bernard (see Beethoven: Briefwechsel: Gesamtausgabe, letter
, and The Letters of Beethoven, letter , where the saying is identified and explained).
15
Beethoven wrote this in the left margin of the page.
16
Beethoven wrote this in the right margin of the page, underlining the word you seven times.

11
Beethoven: Briefwechsel: Gesamtausgabe, ed. Sieghard Brandenburg (Munich: Henle, ),
letter ; an entry in Beethovens conversation books suggests that Beethoven received this note between March and March , .
12
Ferdinand Wolanek (), professional copyist, had been asked to provide performance
material for the Ninth Symphony, to be used at the Lower Rhine Music Festival in Aachen; the symphony was conducted there on May , , a year after the Viennese premiere.
Wolaneks letter is reproduced in facsimile in The Beethoven Compendium, ed. Barry Cooper
(London: Thames and Hudson, ), p. .

Tonwille
(annual volume IV, no. )

This page intentionally left blank

Beethovens Sonata in F Minor, Op.

Sonate opus {Tonwille , pp. }
t r a n s l at e d b y r o b e r t s n a r r e n b e r g
First Movement (Allegro assai)

The sonata form of the first movement proceeds as follows:

first subject:
second subject
third (closing) subject
development
recapitulation
coda
pi allegro (stretta)

antecedent phrase
consequent phrase and modulation

bars

If we disregard all the finer points of the Urlinies linear progressions shown
in the accompanying graph as well as meter in general, we are presented with the
following image of the Urlinie tones (Fig. ):1
{} What immediately strikes us as the first peculiarity of the piece is that the
tones c (c
) and a (a
) linger on for a long time, while the intervening tones b
and
g are touched upon only fleetingly at the close of the third subject and are not
elaborated until the coda and stretta. Thus the impression of the descending
arpeggiation ca
f prevails, as if the descending arpeggiation in the upbeat to
bar had become the destiny of the whole.
What strikes us in the second place is how, right at the beginning of the
movement, the circumscription of by means of the neighboring summons

forth the Urlinie motive cd
c, which henceforth fills up all parts of the movement and is also the very thing that entails the prevalence of c and a
.
Lastly, we comprehend the deepest mystery of this composition, namely, the
unique linkage of all parts founded upon the obligation of the neighbor note in
the Urlinies motive:2 the consequent phrase of the first subject and the modulation reach into the second subject; the antecedent of the second subject likewise
reaches into the consequent; and the latter ultimately into the closing subject
(note the crossed slurs in the Urlinie-Satz, shown in Fig. ).

1
The Urlinie in Fig. , like many of those Schenker described during the early s, is filled with
motivic repetitions. When he later defined the Urlinie as a single linear progression, these motivic
repetitions would be assigned to a middleground layer; compare, for example, the graph of this movement given in Der freie Satz, Fig. /.

That is, the obligation of the neighbor to resolve back to .

tonw i l l e 7
Bars ff. The antecedent phrase of the first subject extends to bar . Fig. presents the basic counterpoint [Ursatz] and its first prolongation.3

bars and . Although arpeggiation and step are oppositesthis in itself, after
all, is the difference between harmony and voice-leading, between the vertical
and the horizontalthe two here become one body, and into that body the compelling force of this voice-leading secretly infuses the unity of a soul. The second
succession substantiates this metamorphosis of a voice-leading event into a
musical soul vibrating with life. The , returning in bar in conformity with the
large-scale Urlinie motive, falls on the last (see Fig. [b] and the graph of the
Urlinie, given on p. ); the two motives are now bound together in one interval,
yet in such a way that the diminuted version of the Urlinie motive comes to the
fore. Compared to the earlier form of motivic unity [heard in bars and ],
this foregrounding of the second motive signals its detachment [from the arpeggiation]; this first detachment then makes possible a second, namely that of the
submotive D
C in the bass in bars , , and . (It is plainly a reflection of good
social etiquette when we notice how the arpeggiation introduces us to the Urlinie
motive in rhythmic diminution, and the latter in turn introduces us to the submotive.)
In the first motivic combination, the emphasis lies on the second motive, if
only because it is a copy of the genuine Urlinie motive. Therefore, we perceive the
realization of the Urlinie tones as an arpeggiation in the broader sense of the term
(see the dotted lines in the graph of the Urlinie):5 is embodied already in the
first fifth of the upbeat, yet only through the second motive in bars do we
correctly understand its manifestation; the same process is then repeated with .
Both arpeggiations move from low to high. The third arpeggiation, in contrast,
runs in the opposite direction, since the second motive takes the lead (see the
dotted line in the graph of the Urlinie6). Through this reversal of the third arpeggiation, c1 of bar is brought into agreement with c1 in the upbeat; moreover,
this makes it possible for the consequent phrase to lead the arpeggiations again
as in the antecedent, from low to high. From the standpoint of synthesis, an extremely valuable narrative resource lies in arpeggiations of this higher type: they
focus our attention on the principal tones, as in this case, and when employed
systematically under the law of parallelism, they contribute a new and particular
feature of the organic to the musical creation.

The series of intervals in the Ursatz of Fig. a is broken into two

groupings, as shown by the small brackets in Fig. b. The secret sowing4 of the descending step in the succession in the Ursatz [i.e., fe beneath c] makes it
possible to harvest the Urlinies step d
c (see the lower bracket in Fig. a) as a
submotive and as something individual and characteristic. The prolongation in
Fig. b shows a cooperation of several forces: the neighbor note links the three
tones of the Urlinie motive together tightly, yet at the same time the first step in
the lower voice effectively detaches the second and third tones as an independent
motive, while a second step answering the lower voices first step strengthens even
further the inner cohesion of the lower voice.
The implementation of these relationships now furnishes life and truth to the
diminution as well, to motives in the more restricted sense. The compelling force
of the first succession [in Fig. b] binds two motives together: {} the arpeggiation in bars and the rhythmic diminution of the Urlinie motive cdc in
3
The term Ursatz appears as a contraction of the expression Urlinie-Satz, used in the preceding
paragraph. Here the term refers to the first two segments of the Urlinie-Satz shown in Fig. . As the
labeling in the figure makes clear, the Urlinie-Satz is a two-voice contrapuntal setting consisting of
an upper voice or Urlinie counterpointed by a lower voice that rests, as it were, upon the harmonic
degrees (Stufen).
4
[S.] In Tonwille , p. /I, p. , I wrote: The great masters of German music have not merely
made the art of music: they actually are the art of music itself. By an unfathomable dispensation by
the Creator, who has sown and reaped all things, they, too, have been allowed to sow and reap in the
realm of music. But no matter how little a man contributes to Gods sowing of the seed, he may take
away from the harvest as much as he can carry; and likewise, no matter how little he contributes to
those musical masters sowing of the seed, he may take away from their musical harvest as much as
his heart desires. A seed is in the earth, and growsbut no one can say whether it has grown of its
own accord or has been raised by genius. Something whole developsbut no one can say what is attributable to its own support system and what genius has added to it. But the whole is always determined by the one seed, and thus in the small world of tones the law of Nature at large is enacted.

5
In its broader sense, arpeggiation (Brechung) signifies the transformation of a vertical situation
into a horizontal. In this instance, the coupling of the Urlinie motive in adjacent registers (c2 d
2 c2
proceeding in parallel above c1 d
1 c1) is broken apart into a single line (c1 c2 d
1 d
2 c2 c1).
6
The third octave coupling, c2 c1, is shown by a bracket, not a dotted line, in the graph. Its supporting sixths, e1 and E, are however connected by a dotted line.

Beethovens Sonata in F Minor, Op.

the Urlinie). The path from c2 (bar ) to d
1 (bar ) runs in a line that has two
bends; if one proceeds upward from some C, c in the little octave, for example:
cegb
e1 g1 b
1 d
2 c2, {} one will recognize the purpose of the arpeggiation, namely, to save d
c until the end. The third eighth note in bar is transferred
into the lower register in order to reciprocate the contra-octave of bar and at the
same time reach the two remaining tones of the diminuted form of the Urlinie
motive in a register that is appropriate for the approaching upbeat (see earlier).
The aforementioned third-progressions across bars | and | also have counterparts within this arpeggiation, one across bars |, and another, albeit on a
larger scale, in bars (ed
c). The rhythm of the submotive D
C in bar :

 

     basically continues the rhythm of the trills final segment     . The

poco ritardando in bars causes us to expect a fermata in bar ; the sudden
forte deceives us in our expectation all the more in that it also contradicts the
overall pp. And even if, after finally reaching the fermata in bar , we comprehend in retrospect the meaning of the arpeggiation as presenting the Urlinie motive for a third time in augmentation, the vivid contrast of poco ritardando and a
tempo, and of pp and f, nonetheless continues to puzzle us for the present, for it

While the background of the Ursatz (see Fig. a) presents only IVthe material does not suffice for more degreesits elaboration in the foreground produces formations whose greater weight probably permits us to speak of additional harmonic degrees (see Fig. b and the graph of the Urlinie). (Compare,
too, the succession VIV at the end of the development, as shown in Fig. , where
the existence of degrees is beyond dispute!) The lowering of II in bars [recte:
] (Phrygian II) is explained by transposition of the motive, with which the
diminished triad of diatonic II was not compatible (see Harmonielehre, pp. ff/
pp. ).
The diminuted form of the Urlinie motive in bars is decorated with two
anticipations, d2 in the sixth eighth note and e2 in the penultimate sixteenth. The
second anticipation, inserted between c2 and d2 in the suffix of the trill, drives
back to c2 (because it descends to d2) instead of upward to e2, as is seen in bars
and where the suffixes to the trills occur without the inserted anticipation.
Within the prevailing expression of an anticipation, a descending third-progression
also appears at the end of the trill, an initial sowing of the Urlinies descending
third-progressions that will come to fruition in the second and third subjects.
In bars the Urlinie motive is fashioned for a third time (see the graph of

tonw i l l e 7
All the motion within bars is only voice-leading motion functioning
within one degree, the dominant of the new key. The following figure lays bare
the voice-leading:

says definitively only that the composer will supply us with even more ruptures
and sudden twists: it is the sowing of another seed destined to bear fruit in the
consequent phrase.
Bars ff. The consequent phrase is intent upon repeating the content of the antecedent phrase, yet an onslaught of chords is again and again thrust between the
repeated barsbars bars and bar bar (). The first onslaught
starts right after the opening of the phrase in bars . And this can be explained by the fact that the two ff interruptions of bars and , which are in
the dominant, must be prepared by an interruption of the tonic triad; one need
only remove the tonic interruption to observe what force of justification inheres
in such a parallelism.
But most beautiful aspect of this harvest is the newly sown seed: for the ear,
having become well acquainted with the subject in the antecedent phrase, learns
from these interruptions to follow the subjects course despite impediments and
thus becomes capable of hearing c2 in bar (bars are, again, just an interruption) as the tone that ultimately concludes the Urlinie motive (see Fig. ).
What energy arches through the broad spans!
It should not be forgotten that the basic dynamic state of the consequent
phrase, despite the ff interruptions, is piano, increased relative to the dynamic of
the antecedent phrase only as much as the more agitated mood requires. The
brevity of the consequent phrase is compensated in the coda and stretta.

According to the main plan (see Fig. ), d
2 should be both the point of departure and the goal of the motion; accordingly, the descent of the seventh (see
Fig. a) must be avoided.7 The technique of reaching over makes it possible to retain the register in such cases (see Fig. b). As can be seen in Fig. c, reaching over
is here combined with a transfer [of the inner voice] into a higher register. In the
normal course of voice-leading (see Fig. a), the fifth b
2 would have to have
moved to the third g2 by going down across the fourth a
2; but [as seen in Fig. c],
this would have led to a series of passing chords (Kontrapunkt ii, p. /p. )
that are less fit for elaboration, so Beethoven inserted a neighboring motion
around the third (see Fig. d) that lets b
2 press upwards to c
2 [instead of descending to a
2] and thereby obtains the chords that are better suited to elaboration, though not without reinforcement in the lower octave: . The Urlinie motive makes itself felt in , b
2 c
2 b
2 (in diminution) and likewise
the submotive (bar ) in the succession f
3 e
3 (bars ), which is required
here by the reaching over as well as by the neighbor-note motion. When e
3 (e
2)
finally descends to d
2, the third-progression (f
e
d
) thereby returns, thus answering and passing along the numerous third-progressions of the antecedent
phrase. Whether the submotive f
e
is also influenced by chromatic step ee
in
the lower voice (bar ) cannot be determined.

Bars ff. The Ursatz (Fig. ) admits only a modulation through reinterpretation
(Harmonielehre ff. [pp. ]/pp. ), but the harmonic degrees in the
elaboration (see the graph of the Urlinie) necessitate the adoption of a chromatic
modulation. The approach of the modulation is heralded by an extremely unusual and unseen move: in bar the master suddenly relieves the left hand of the
tone b1, which he transfers to the right hand (compare bar ); this makes it so
much easier in bar for the left hand to become aware of the contrast between
the upward step in bar and the downward step in bar ; the hand, as it were,
physically feels its way through the modulation and so is also made capable of
communicating this feeling convincingly.
Since F minor can only modulate to A
major as the key based on III (see Harmonielehre, [pp. ]/pp. ), A
minor must be heard here as mixture within A
major proper. The fact that the minor mode here arrives in advance of the major and even {} occupies greater space does nothing to alter this.

Bars ff. The second subject is binary and projects A
major. The antecedent
phrase has a full perfect cadence but it is nevertheless bound to the consequent
phrase by the Urlinie motive (see Fig. ). Heralded by the descending third-progressions of the first subject and the transition section, third-progressions finally
appear here for the first time as Urlinie progressions: c2 b
1 a
1 and d
1 c1 b

(see the graph of the Urlinie). The second third-progression, which is transferred
[S.] See Tonwille , p. /I, p. , Fig. .

Beethovens Sonata in F Minor, Op.

into a lower octave (see Fig. ), is accompanied by the submotive (fe
); this motive lies above the third-progression and, by dividing the octave triadically, makes
the register transfer truly feasible.8 The tones of the Urlinie progression fall on
weak parts of the bar: therefore, as in the first subject, the spirit of arpeggiation
dominates; indeed, this arpeggiation recalls the arpeggiation in bars in a
stricter sense of the term.

ond (gf
). Linkage with the third subject is not the sole object of the peculiar
way in which the second subject makes its exit; it is also intended to make us
strain toward a distant future when its final word will at last be uttered decisively
and convincingly (see bar ). What far-reaching scope in the narrative design!
Bars ff. The closing subject is set apart from the second subject, to be sure, but
it is still dominated by the third-progressions of the second subject, albeit drastically reduced in duration. Indeed, even the pattern of arpeggiation is continued,
so that these compressed third-progressions fall on the weak parts of the bar. Separating
and into two different octaves (see the graph of the Urlinie and Fig. )
makes it necessary to use the reaching-over technique. The basic picture is:

Bars ff. With the consequent phrase (bar ff), both third-progressions are
transferred into a higher register; in addition, the second progression is shifted
into the minor mode, so that mixture returns after a break of only seven bars. But
now the strangest thing occurs, a mysterious force sunders the tones of the thirdprogression as though in violent jolts: bar corresponds to the first half of bar
, bar to the last quarter of bar , the whole notes c
2 and b
1 in bars
to the half notes c1 and b
in bar . Correspondingly longer note values also appear in the tones of the submotive, f
2 in bar , e
2 in bars , at first only
imagined but finally declared openly in bars . To speak of this merely as a
simple case of augmentation would, I dare say, be {} too prosaic. What best
demonstrates that this is not the conventional sort of augmentation is the fact
that no one has heretofore discerned it as such. This ingenious, dramatic feature
is firmly anchored in synthesis: without the concise form of the second thirdprogression in the antecedent phrase, Beethoven would never have been able to
devise its new shape in the consequent phrase, no more than, for example, he
could have designed the third Urlinie motive in the first subjects antecedent
phrase to flow so freely and so fluidly without the preparatory versions in bars
and . Only genius treads such paths; genius alone knows how to obtain freedom from obligation and disdains the paltry freedoms of program music and
opera, where motivic transformations are mostly dictated by an extramusical cue.
The seventh d
1 appears in the left hand in bar in order to compel a voiceleading connection between the second subjects consequent phrase and the third
subject (on their linkage through the Urlinie-motive, see earlier); accordingly,
one must also assume there is a full close in the consequent phrase. The run in
bars projects the submotive f
e
(see the first and last eighth notes in bars
and, especially, F
E
in bar ); it is a series of five diminished-seventh
chords f
g that are connected by g
for the sake of avoiding an augmented sec-

One gathers from Fig. b that the third-progressions in bars and (a
g

f
and b
a
g) belong to the inner voice and not to the Urlinie. Furthermore, the
second Urlinie progression d
c
b
is initially omitted in bars for the sake
of the cadence in bar . The necessity of avoiding consecutive fifths (see Fig. c)
no doubt led here to a exchange (in harmonic terms, to the succession I

VIII
5 V) and thus of its own accord to the submotive f
e
. As the graph of the
Urlinie shows, the same content is repeated starting in bar , until at last, in bar
, the Urlinie contributes the long awaited close
, in the most extreme
register, no less, the way to which had been paved by the passage in bar . Note
that
appears here only as passing within V. And now, from bar on, the linear progression d
c
b
a
, which has at last arrived, plunges precipitously
through three octaves {} into the low register, while the submotive f
e
meshes


with it en route. Added atop the third of the linear progression db
, the filler tone
e
sounds as though it hails from a foreign world, with its painfully lamenting sfp
always looking ahead to the next downbeat in the manner of a syncopation.

8
Coming after the first third-progression, which brings c2 down to a
1, e
1 of the submotive appears
to create a descending arpeggiation of the tonic triad that divides up the octave between c2 and c1.

tonw i l l e 7
Reached in bar , a
3 is adhered to during the motives plummet into the low register. Note how the inner voice of bars , which functions as an octave reinforcement from bar on, ultimately resolves in this register.

The chord successions functioning across the two third-progressions represent

a closed chord cycle having the effect of the harmonic progression IIIVI,
which is why it is permissible here to extract {} and retain g 3 until bar (see
the graph of the Urlinie and Freir Satz). It is important to recognize how the
first subjects motives resist an exact transfer into the major mode; compare Fig.
b with the following:

Bars . The plan of the entire development section may be represented as

follows:

Transformation of the major triad on E into a minor triad occurs in bar as

an operation of mixture.
And now a change of motives: whereas the leap from the G  (A
) of bar to
the E of bar leads across the neighbor note D , the two leaps EC and CA
,
starting in bar , are filled out with passing tones (see the graph of the Urlinie).
The arpeggiation of bars serves as a motive. The criss-crossing of rising and
falling arpeggiations is also made apparent in the graph of the Urlinie. Instead of
a major triad on C in bar (resulting from mixture), there is a minor triad,
which, with its minor third, now announces the return of the actual root A
. In
bar one must imagine that a
3 has returned (as the dotted slurs in the graph of
the Urlinie show; also see Fig. b); not until this point does it proceed to g
3 in bar
as the seventh of the chord.
And now another motivic change: the passing g
is projected by the motive
from the transition section (bars ), but it is initially inflected to major, just
as many of its secondary traits undergo a shift in direction on account of appearing in a lower register. But eventually the half step b

1 a
1 had to be worked
into bar since, as the premise of the chromatic steps of bars and , it is
indispensable. The motive in bars ff only appears to be new; close examination reveals it to be a third-progression that prepares the third-progressions of
the second subject, the content of the next segment of the development.

The first sketch (Fig. a) shows how the tonic of A
succumbs to the gravitational pull of the main key and, as III of that key, also brings further descending


fifths in its wake: VIIIVI. The third, c
(III 3), is canceled by the tonicization


III  VI, whose breadth is shown by the prolongations of voice-leading in Fig. b:
three major thirds lead from A
back again to A
,9 in the course of which the
minor third c
, enharmonically reinterpreted as b, turns upward to the major
third c when the bass makes a second leap. The leaps of a third are only passing
events, thus there can be no question of either harmonic degrees or a change of
key, especially if one is inclined for the time being to take the A
triad as dominant in the key of D
major.
Motivic elaboration of the individual leaps of a third is not uniform. The E
major triad (bars ) is traversed by two third-progressions, g  f  e (bars
) and f  ed  (bars ) (a repetition then occurs in bars ); motives
from the first subject are employed here. Distribution of the first third-progression into different octaves makes it more difficult to see the simple intervallic successions that simply exchange a sixth for a third (tenth):

Bars . The first prolongation of voice-leading in bars is shown in

Fig. a: f 2 goes to e2 with VIV and a exchange ( VIIVV) eliminates
the consecutive fifths that threatened to appear as a result of this harmonic step

[S.] Compare, for example, Chopin, Etude in A Minor, Op. , No. , bars : ECA
E;
Hugo Wolf, Das Stndchen (Eichendorff): DF  B
D; and so forth.
9

Beethovens Sonata in F Minor, Op.

progression. Transfer of the goal tone into a higher register (see Fig. b) leads
from f 2 to g2 (II) and thence via an upward arpeggiation to e3 (V). The realization
recalls the motive of the second subject; analysis of this is reflected by the layer of
voice-leading given in Fig. c. The motion of the outer voices in thirds in bars
threatens consecutive fifths;10 these are alleviated by instead of
, in the course of which the leaping passing motion in the lower voice
simultaneously projects the effect of a neighbor note, A before B
, F before G
. Yet
in reality (see the graph of the Urlinie), the lower voice reaches A and F not by
moving downward, as illustrated in Fig. c, but rather by moving upward, as required by the motive; accordingly, the second subjects motive experiences its first
alteration only toward the end of the second Urlinie progression (compare the
second half of bars and with the second half of bar ).
The root B
changes the motive of the second subject to minor. Note the alteration in the left hand from the fourth quarter of bar up to bar , compared {} to the version in bars . One need only try that version in bars
to discover the reason for the alteration: the motive requires the lowered
second scale degree in bar and that means the half step FG
in bars ,
whereby the consecutive fifths were alleviated by ; thus this procedure differs
from that in bars , where the fifths above A
B
were alleviated by means
of .11
The lower voice now moves upward in a fifth-progression to d
1 (V); g
must
therefore give way to g as the fifth of V and, in addition, a exchange
prevents consecutive fifths. Thus what the lower voice expresses, above and beyond II, is only the single bass note, D
(see Fig. a). The sheer length of the path,
the indentations made by the neighbor notes, the persistence of such significant
motivic blocks in contrary motionall fruit borne of the prolongations of
voice-leading12 stirringly project just one interval: a third (tenth)! But even that
does not suffice: the violent passion of the rising thrust drives d
1 up two more
octaves to d
3 in bar , even more impatiently and in rapid arpeggiations.

In bar the upper voice holds onto the high g3, which was reached at last in
bars ; it maintains this pitch as if obstinately opposed to the onslaught of
arpeggiations in the left hand in bar , which veritably shatter into pieces. In
bar the left hand reaches d
2 and, crowding it out, causes the right hand to give
way; the right hand takes b
3. The left hand continues to press upward, yet the
right hand successfully resists, so that nothing remains except for the left hand to
restrict its arpeggiation to two sixteenths. Thus subdued by the obstinacy of the
upper voice, the left hands upward thrust is exhausted by its excess; the tension
subsides quickly as the lower voice falls in arpeggiations from d
3 back to D
(bar
); here, at last, e appears in the upper voice, the pitch that was intimated from
the outset as a goal (see Fig. a). D
now falls back to C (bar ) as a resolution
of the suspension, and the submotive (bar ) thereby materializes. What a
summoning of voice-leading skills, figures and registers starting in bar , to
project this motive in such an exceedingly large form! Can one speak here merely
of diminutions, figural work, virtuosity?
The way to the recapitulation stood open at this point, but the register of the
first subject had yet to be prepared; therefore, the master now also retraces the
submotive in bars in the two-line octave (the sowing of the seed) and
thereby obtains c2 as of the first subject, which is now on the verge of being repeated (the harvest).
Bars ff. The overpowering excitement that brings on the beginning of the recapitulation does not allow for the original form of the first subject, for the bass
note C wants to and must ring forth, as if it were a force of nature acting for itself, following a law all its own, until a higher force again restrains it. Thus Beethoven (see the graph of the Urlinie) positions the first subject atop a tremulous
dominant pedal point in eighth notes. Naturally, on account of the overly harsh
dissonance, he not was able to hold onto the this bass note in bars [recte:
], and so resorts to neighbor notes: CD
BC.
Whereas the transition from C to D
in bar occurs precisely in the middle
of the bar, D
leaps down to B as early as the second eighth note in bar ; this is
supposed to prepare the leap that is needed in bar , where the low register {}
does not permit striking the tones CE and CF simultaneously. The half notes
in the left hand in bars ff correspond to the high bass voice in bar .
The consequent phrase breaks out suddenly in F major, but the coloring of
minor resurfaces as early as the passing events in bars ff. In bars , the

10
[S.] Compare Tonwille , p. / I, p. , Fig. ; p. / I, p. , Fig. ; see also Tonwille , p. / I, p. ,
Fig. ; Tonwille , p. / I, p. , Fig. ; and Tonwille , p. / I, p. , Fig. .
11
In both instances, the upper note of the second fifth seems to come from the seventh rather
than the fifth of the preceding sonority; but whereas the chromatic step in the bass in bar interposes a sixth between these intervals (hence ), the seventh of bar proceeds directly to the
fifth in bar .
12
[S.] Compare Tonwille , p. /I, p. .

tonw i l l e 7
trill concludes without an anticipation of the next downbeat and with only the
ordinary suffix, a figure that points out the upward passing motion; not until the
conclusion of the passing events in bar does the original version of the trill
with anticipation (as in bar ) return. The remaining course of the recapitulation corresponds to that of the exposition, even to the point of retaining the
major mode in the second subject. The change in the accompaniment in [the second half of] bar , compared to bar , is attributable to the much lower register. In the run in bars , the diminished-seventh chord has two interpolations: g
and e
.

of it is limited to the first third-progression, which is then followed not by the

second third-progression but by a repetition which introduces changes that are
required for the cadence. If one compares the versions of the second subject that
appear in the development (bars ff) and coda (bars ff) with its original
form, which harbored two third-progressions [bars ff], one notices a steady
disintegrationbehind this is concealed a balancing act that the synthesis demands from the composer. The broad arpeggiations of bar ff work in the service of the fourth-progression f 3 b
3 (see the graph of the Urlinie). The master
intuitively grasped the law of obligatory register13 when he heard the pitches g
3
(bar ), a3 (bar ) {} and, the goal of the motion, b
3 (bar ) connected
across spans of time. Again I have to ask: may such fulfillment of the law be called
virtuosity? Or is it not the case here that all the musical figuration is the very soul
of the law? The inner articulation of the linear progression b
1 b
3 in bars
into continuous steps instead of arpeggiations can be seen in the graph of the
Urlinie. In the realization of bars the Urlinie motion is superimposed by a progression that reaches back in a high arch (see the graph of the
Urlinie).
But in the lower voice, too, the bass notes are also registrally coordinated
from bar on. For how long is the contra-F left out, until contra-B
appears
(bar ), and how far is it then until C in bar , to D
in bar , and how much
further still is it to C in bar ! Beethoven also concludes the coda with the submotive D
C in the bass. The motive is extended by five short repetitions (bars
); but these would not be conceivable as such without the large-scale motivic model of bars .

Bars ff. In bar the recapitulation concludes and the coda begins. As the
graph of the Urlinie shows, f 3 stands as during bars . In the Ursatz of
these measures (see Fig. and Fig. a):

the only matter of substance is the change of interval between the outer voices,
from the octave (at the start of bar ) to a sixth (bar ), which makes possible
the passing chords that are required for the fourth-progression to in bar .
The prolongation in Fig. b shows a neighbor-note motion in the upper voice
and passing tones in the lower voice. The final realization, however, obliges us to
consider the passage in terms of harmonic degrees (see the graph of the Urlinie).
Note, moreover, that in lieu of the bass of II in bar (g1 in this case), a passing
f 1 is immediately put into play (from the perspective of the harmonic root, f 1 is a
seventh); and in bar one must likewise imagine the root c1 before the passing
b
. In bars Beethoven avails himself of the arpeggiation motive from bars
; the two figurations that, starting as early as bar ( bar ), stand opposed to one another in the sixteenth-note accompanimenta hemiola of
sixteenths in the first half of the bar and figures of sixteenths in the second
halfare employed in subsequent bars in such a way that the one accompanies
the stationary half notes in bars and while the other accompanies the
arpeggiations that proceed from them. The motive of the second subject appears
as the mediator of the cadential harmonies VIIIV in bar ff, but this version

Bars ff. Viewed from the perspective of the stretta, the coda seems almost to
acquire the appearance of a concerto cadenza (compare, for example, Beethovens
Sonata Op. , No. , first movement, bars ff). For the stretta for the first time
brings the second subject in F minor, a fulfillment of the key that had been denied it in the recapitulation. The realization refers initially to the second subject
as it appeared in the coda (bar ff), a version that uses only the first thirdprogression, but bars at last copy bars and thereby obtain the second third-progression (b
a
g) as well as the submotive d
c in the main key,
thus in one fell swoop bringing the fulfillment of closure so long withheld from
the second subject!
[S.] See Tonwille , p. ff/I, p. .

13

Beethovens Sonata in F Minor, Op.

The lower voice in Fig. a moves in step with the upper voices ; the
last two tones of its fifth-progression coincide with the upper voices , in the
sense of II and V. In Fig. b, the lower voice, as if to come closer to its goal, leaps
up a third to the midpoint of its path while is still at rest; this then impels the
Urlinie tones to make up hurriedly for lost time. In Fig. c and d, we see the ascending register transfer (a
a
1) and the inflection of the tonic triad by the
upper-fifth divider.14
In the final realization we note again, with astonishment, the Urlinie motive
from the first movement that uses the neighbor note; indeed, we even see two of
these motives linked together. The first four-bar group is held together by a cadential progression: IIIVI. The lower-fifth divider in bar supports the
neighbor note and is therefore only a neighboring chord whose root is extrapolated (see Freier Satz). The greater significance of the second b
in bar is
rooted in the harmonic degree. As early as bars and , there is a conspicuous
fifth-progression in the lower voice; it arises here in VI without any compulsion,
but it is a seed of the Urlinie progression , which does not take place until
bars ! In order to secure the effect of this fifth-progression, Beethoven must
unfold it again in its entirety in the second quarter of bar , even though he completes the first half (a
g
f) during the first beat; this is due to the appearance
of tonic on the downbeat of bar , as necessitated by the cadential progression;
since this tonic falls in the middle of the fifth-progression, it endangers the integrity of the linear progression.
Bars basically show the same linking of two neighbor-note motives (see
the slurs in the graph of the Urlinie, given on p. ); the first runs its course in the
inner voice, while the second is continued in the bass. This inverts what happened
in the preceding movement, where the erstwhile inner voice d
cd
was brought


out into the upper voice. Having placed  IV 7 in the upbeat of bar , Beethoven
writes an e instead of f
just for the sake of obtaining the visual impression of two
{} voices progressing in thirds. The second four-bar group also concludes with
the descending fifth-progression (bar ).
The second section (bar ff) is linked at the outset to the preceding events:
bars repeat the overall rising fourth of bars (a
d
1) in acutely abbreviated form. This abbreviation becomes a motive, later versions of which present
increasingly larger spans: the sixth in bars and the octave in bars . The
ascending register transfer ends with the last of these and the fifth-progression of

The combination of the two third-progressions is basically to be understood

as follows:

Accordingly, a double elision is present in bar , with a
3 omitted before g
3

in the upper voice as well as A
before A in the bass.
Beneath c2, which hovers as an echo of the submotive in bar , and reach
the decisive moment (see Fig. ). Bar twice manifests the succession a
1 g1,
while the upper voice only sounds the submotive d
2 c2 once. The same play, only
in increasingly higher octaves, is repeated in bars . Finally, in bar , the
tonic chord is filled out for the last time by the arpeggiation of bars within
the compass of f 3 and contra-F. The moment the arpeggiation touches f in the
little octave in descent, at the upbeat of bar , c2 of the submotive again emerges
above it and thus the movement concludes with the same fifth with which it
began, indeed, in the same register as the first Urlinie motive (bar ).

Second Movement (Andante con moto)

Bars ff. {} The second movement is a set of variations consisting of a theme,
three variations and a coda. The theme is binary, with repeated sections of eight
bars each. The layers of voice-leading are captured in the following figure:

[S.] See Tonwille , p. (note)/I, p. , note .

14

tonw i l l e 7

the outer voices in bars , , and , despite the downward elaboration FD
in

the lower voice.

Urlinie tones brings fulfillment of the two earlier fifth-progressions in the bass
it is the harvest. Like a shadow, the fifth-progression appears again in the bass in
bar . Constant juxtaposition of the rising tones of the arpeggiation against a

gives the impression of a choral realization15 and the impression of V in bars ,
, and ; yet the illustration in Fig. c informs us that the harmonies in these
bars function only as neighboring harmonies. The skip of a third in the lower
voice (G
E
) does not cancel the interval of a second between the outer voices;
it is only a downward elaboration that corresponds to the upward elaboration e

g
in the inner voice. Similarly, the interval of a third remains in effect between

Bars ff. The first variation is characterized by motion in eighths and delayed
bass notes struck on the offbeats. The tying of the bass notes has nothing to do
with suspension formations (Kontrapunkt ii, p. ff/p. ff ); rather, with its ponderous rhythm it means to add a certain mental weight to the downward arpeggiations, which is indispensable if the offbeat striking of the bass notes is to produce a significant effect. In the fourth bar of the variation (bar ), the delayed
attacks suddenly convert to anticipatory attacks. If one compares this with the
final bars of the variation (bars ), in which the delayed attack is maintained
throughout, one sees that the anticipatory attacks in bar were necessary, because Beethoven, in order to fashion the tonic in this bar exactly as he had in
bar , moves directly to the seventh g
in the last eighth of bar , instead of re-

15
Den Eindruck einer chorischen Ausfhrung: Schenker is probably thinking here of the malevoice choir in particular, in which the lyrical character of the first tenor partthat is, the tendency
for the highest voice to be set off from the othersis balanced against the resonance of consonant
chords sung in close position. The theme of the Andante con moto became, in fact, a popular piece
in an arrangement for male-voice choir.

Beethovens Sonata in F Minor, Op.

peating A
as in the theme. The situation is different in bar : here A
follows

g
, so that from the start nothing stands in the way of the delayed attacks. The
anticipatory attacks in bar necessitated C across bars |, since the stop of
D
D
was otherwise unavoidable. Delayed attacks continue in the rest of the
bars. The fifth-progression lacks g
in the first ending, and in the second ending
it dissolves into two leaps of a third.
In the second section of the variation, the start of the upper voices arpeggiation is delayed: in bar , where d
1 should have already appeared, a
still persists;
d
1 does not arrive until bar , so that the composer must ultimately compress f 1
and a
1 into a single bar (bar ). Though he thereby robbed of the weight it
originally possessed, he compensates in bar with the neighbor-note design,
which of course requires in turn an acceleration of the fifth-progression in bar .
All of these alterations in the second section of the theme somewhat reduce the
impression of choral performance.

The inner voice of the theme has become the upper voice (a
2), while the motive of the ascending register transfer has become the inner voice in bars
and finally, no less, the lower voice in bar . In the downbeat of bar (bar of
the theme) one must already imagine as the continuation of the lower voices
from bar ( is the descending leading tone in II; compare bars
and see Fig. ); accordingly, a
2 in the right hand follows b
2 in the left (compare
bar of the first variation). In bar , only half of the descending fifth-progression is realized and, in any case, the high a
2 is maintained during the bar for the
sake of connecting to the following section. In the repetition (bars ) the
motivic construction of the theme appears more clearly. In bars the upper
voice must be read as follows:

Bars ff. With an increase in activity (sixteenths), the second variation shifts upward an octave. It reproduces all the traits of the theme exactly. In the first section of the variation we find the peak tones of the arpeggiations on the initial sixteenth note of the downbeat and on the second sixteenth of the upbeat, with only
the working out of the suspension in bar making it necessary to reverse
this order; in the second section of the variation, it is once again the impression
of choral performance that demands greater liberties.

In the coda (bar ff), the choral aspect is developed further. Bar at last
brings , yet there is also a deceptive cadence at this point, which subsequently


leads into the final movement attacca. The effect is initially that of  IV 7, as if V
were imminent (see Fig. a):

Bars ff. {} The third variation is characterized by motion in thirty-second

notes in an even higher octave. The repeated sections are reworked and distinguished by having the thirty-second notes appear at first below and then above.
In bar , the fourth bar of the variation, the fifth-progression drops away and the
thirty-second notes turn upwards; this is due to the fact that the lower voice shifts
up to f 1 and follows the path of the inner voice in bars with fee
d
(see
bars ). In the repetition (bars ), the lower voice even moves along the
track of the next-higher inner voice, a
1 b
1 a
1 b

1(a  1)a
1.
An inversion of the voices occurs in bars (bars of the theme):

Consequently, it is the enharmonic transformation f
e (see Fig. b) that

brings about VII in F minor; and it is this harmony, then, that opens the final
movement.

Third Movement (Allegro ma non troppo)

{} The introduction presents the succession  VII7 V7, which releases the step
motion d
c in the upper voice (see Fig. b). Hence the submotive from bar
of the first movement now returns (a sowing of the seed) and in retrospect we
understand the point of this harmonic succession, which basically projects just

tonw i l l e 7
thus is represented at first by g
(
). Bars should be understood as an
augmentation of the figure that secured in the upbeats of bars and : the
augmentation arises through a fourfold repetition of that figure and at the same
time contributes the diatonic correction, g (  ). The bass in bars presents
neighbor notes with the apparent effect of VVIIIV; {} there should certainly be no question of consecutive fifths on the afterbeats in bar , since only
is relevant to the counterpoint of the outer voices (see the graph of the
Urlinie, p. ). The consequent phrase repeats the content of the antecedent, but
also openly states the Urlinie motive in the next higher octave (see the graph). In
bar , too, the fifths do not signify consecutive fifths; they belong, rather, to the
elaboration of the afterbeat thirds.

one harmonic degree, namely, V (see Harmonielehre, , pp. /pp.

). At first, in bars , the chord is played only as a vertical sonority, animated
by various rhythms; not until bar do passing tones penetrate arpeggiations and
thereby generate the motion in sixteenths. If one ignores the dots in bars of
the Allegro and puts these bars together with the last two fermatas of the Andante,
one discovers a change in rhythmic units that represents a steady acceleration:

              . Each dotted rhythm anticipates the next smaller note
value, in this way naturally introducing the subsequent sixteenth-note motion.
The manner in which the harmony is elaborated in bars shows the power of
imagination increasing in excitement and clearly bent on seeking a suitable register for the first subject. When c1 arrives in bar , VII7 turns into V7; this introduces the step c1 b
for the first time into the arpeggiation across bars |. Repeated on a larger scale (bars ), this step instigates a series of quickening
steps that complete the fifth-progression from C to F (VI).

Bars ff. It is true that the second part of the first subject group also presents
only , again accompanied by the Urlinie motive (see the graph), yet in contrast to the first part of the subject group it avails itself of reaching-over technique, employed here on behalf of decorative neighbor notes. To imitate the preceding four-bar group and its suspensions, c1 should have been introduced in bar
and led to b
(as ) with the effect of a suspension; Beethoven, however, fetches
b
from even further away, namely, from e
1; this compels him to accelerate and
later on (bars ), thereby resulting in the metric grouping of + + . In
the consequent phrase, the right hand leads the melody, while the left has the sixteenth-note figure.
Mysterious relationships hold sway between the first and second parts of the
first subject group: both the rhythm    in the consequent of the first part
and the register of c2 continue in the second part, the rhythm in the left hand and
the register in the right; c2 then functions as a bridge to the next higher register
in the consequent phrase of the second part, so that the main motive can start in
the register of c3 at the beginning of the modulation in bar .

Bars ff. The sonata form proper begins in bar :

first subject group
modulation and second subject
closing subject
development
recapitulation
Presto (stretta)

bars

The first subject group consists of two parts, bars and bars , and
each in turn is also internally binary. The Urlinie in the antecedent phrase of the
first part (bars ) presents the course , above which hovers the Urlinie
motive of the first movement, cd
c (the harvest), which was announced by the
submotive in the introduction. In bar the c1 of the submotive is implicit: the
fact that bar brings c1 was all the more reason to withhold it here.

Bars ff. The modulation section is furnished with the figuration from the first
part of the first subject group and modulates in the simplest manner to C minor
( is interpreted in bar as in C minor). The of the new key stands on two
harmonic degrees (IV and V), with the result that (bar ) falls on I. Since follows in bar within
II, we thereby enter into a cadential progression that seems
to do nothing more than confirm the modulation, but in fact it also states the second subject: a tighter connection between a transition and second subject certainly cannot be imagined. And, what is more, the motive of bars recalls

The diminution is borrowed from the introduction; as in the first movement,

transposing the motive to II necessitates a lowering of the root of that triad
even though the diminished triad would have anticipated the effect of Vand

Beethovens Sonata in F Minor, Op.

Bars ff. The following illustration provides an overview of the development:

the figure that brought about the actual conclusion to an Urlinie progression in
bars . An octave descent leads from
(bar ) to  (bar ), so that the
characteristic feature of the first subject is also mirrored in this linear progression: g
g  in the first subject, d
d  here.16 The descending register transfer engenders harmonic effects (see the graph), yet above and beyond these, the Urlinie
elects only to change
into  (i.e., Phrygian into diatonic II) without any mediation, as happened, for example, in the bass in bar of the first movement.
Bars bring a repetition of bars in a lower register: the upper voice
(originally an inner voice) at first acts as a counterpointing voice in bars ,
but then joins in as a reinforcing voice, so that (bar ) is regained in the threeline octave (c3), in agreement with its initial register in bars ff. In bars ,
notice the play of parallelism in the registral exchange of a
2 and a
3; it recalls that
of bars in the first movement.

Right at the beginning, in bar , the actual root of I is openly stated; the
ninth soars upward and does not find its resolution until bar , while in the
meantime the bass moves on (see the graph of the Urlinie). In the two g
f successions (in the bass and upper voice) we must acknowledge the submotive! The
motive of the first subject serves as diminution. IV is reached in bar , with
b
2 in the upper voice as the second tone of the descending fifth-progression (see
Fig. ).18 (The apparent key of B
minor thus turns out to be a purely diatonic
phenomenon; for moving so suddenly from C minor to B
minor would otherwise have been absurd.) The first subject is here replicated in IV, with minor alterations that suffice for the purpose of the development: the counterpoint of the
outer voices emanates from the position, the left hand accompanies with new
arpeggiations (probably stimulated by the arpeggiated third c1 a in the preceding bars), and the basss closing succession of FE
D
in bars quite obviously answers the same succession across bars | (see the slurs in the graph
of the Urlinie). In the repetition of this in bars , there suddenly appears
the decoration of a motive that exploits the lower neighbor note; this motive obviously stems from bars and is deliberately led out as a contrast to the Urlinie motive. Yet starting in bar both the Urlinie motive and this new motive
are placed in counterpoint against one another (see the graph). The arpeggiation
motive of the first subject even disappears entirely and in its place there appears
the Urlinie motive in a certain nakedness, only slightly shrouded by the rhythmic
displacement. The Urlinie motive is initially combined with a third-progression,

Bars ff. The closing subject is a consequence of the cadential tonic triad. At the
same time, however, it is also the head of the development section, resembling
the exceptionally strong linkage between the modulation and second subject in
the exposition. Once again, the first subjects motive of diminution is used, but
when the upper-fifth divider appears (bar and bar ) the motive is accelerated into eighth notes, as it were (see the graph of the Urlinie). In bars and
the scalar runs entail ba
, b on account of the upper-fifth divider (V) and
a
on account of the figures {} in bars . By contrast, the scalar runs in bars
entail ab and ba; this is due to further acceleration in the alternation of
tonic and dominant chords as well as with to the fact that the run in bar is
obliged to use ab, a fact that also requires ba in the descending motion of bar
. At the same time, the repeated alternation of D and C in the bass gives impetus to the descending fifth-progression of bars ! This fifth-progression,
combined with a raising of the third in I, leads us to expect a repetition of the exposition. Often, as in this very case, the goal chord is anticipated midway through
the progression; nevertheless, the entrance of the new chord may not be assumed
until the completion of the fifth-progression (cf. Freier Satz).17 The anticipation was all the more demanded in this passage since I of the main key is obscured [in bar ] by the raised third and the ninth (in the bass).

18
Schenker later reconsidered this account of the development section. Two sketches scribbled on
scraps of paper inserted between pages and of his personal copy of this issue of Der Tonwille
(OC, Books and Pamphlets ) show that he reinterpreted the Urlinie in the development as prolongation of g2( ) combined with reaching over to a b
2 that acts as an incomplete neighbor to a
2 at the
beginning of the recapitulation; a descending third progression connects b
2 to g2 (bars ). Neither sketch mentions c3 but instead indicate a  2 as a chromaticized neighbor preceding b
2. One sketch
is jotted on a single staff on the back of postcard announcing a Liederabend to be given by Donald
Pirnie on October , ; the other sketch, written on two printed staves and containing slightly
more detail, is undated. Emendations penciled onto Fig. also support this new reading.

16
[S.] See, for example, Chopin, Polonaise in C minor, Op. , [No. ,] in the second main part of
the ternary form: likewise a correction of d
into d in the key of C minor.
17
See Der freie Satz, , for a discussion of the closure of linear progression.

tonw i l l e 7
fe
d
(bars , ), which surmounts, as it were, the third-progression of the first subject, d
cb
(bars ). In bar , one has to imagine b
2
as the seventh of the dominant that enters at that point; clearly the pitch is not
conveyed until the third-progression cb
a
in bars and . Though it functions as but one of the means used to elaborate the fifth-progression encompassing the entire development section, this very third-progression also summons another third-progressionb
a
g (bars )en route from a
to g (bars
). To be sure, Fig. shows the head of this third-progression (b
1) as a
neighbor note between a
and g of the fifth-progression {} (bars and );
however, a fourth-progression (see the graph of the Urlinie) purposefully strives
to reach it, thereby supplying b
1 with the harmonic weight of IV.19 V is marked
by the Urlinie motive in the bass, executed on the grandest scale as a culmination
of all the Urlinie motives that are worked into the development. The realization
adds a fourth-progression to this in the upper voice, c1 d
1 e1 f: the neighbor
note B
in the lower voice appears against d
1 (bar ), while arpeggiations attend
e1 (bar ff). The waning of this arpeggiation announces the recapitulation: the
sixteenths are slowed by triplets in bars , then follow eighths in bars
, and finally half notes in bars . (The graph of the Urlinie shows the more
detailed articulation of these segments.) The right hands arpeggiation of chords
in bars replies to the preceding arpeggiation, a special type of parallelism.
The graph shows how e1 of bar (see the dotted line) moves to e2 in bar ,
then to e3 in bar , but suddenly sinks back to e, which, already taken hold of in
bar , now nestles under g of bar as its lower third (see Fig. ). We perceive
in this the conversion of a descending leading tone into an ascending one; see, for
example, ( ) in bars , , etc.

ance of c as a fifth of the F minor tonic is much less a matter of course here than
in bar , where c was the octave of the C minor tonic.
Bars ff. The movement concludes with a Presto (stretta). The organization
into repeated sections disguises a reaching-over technique that gives rise to :

According to this illustration, c2 is in principle present as early as bar ,

above g1 herein lies the reaching-over techniquebut at the moment c2 actually
enters (bar ), the root of III is also already present, followed thereafter by IV
V. The realization makes use of the third-progression a
gf, continued in imitations (see the slurs in the graph, and compare the finale in Haydns sonata, discussed in Tonwille ). Concerning the repeated sections, one would certainly have
to assume a modulation to C minor in the first, then in the second a modulation
from C minor to A
major and thence the modulation back to F minor; nevertheless, as Fig. shows, all these apparent harmonic degrees and keys, find their
raison dtre in the main diatonic system. In bars ff the Urlinie progression of
the first subject (bars ) reappears (also a third-progression). An ascending
register transfer brings the Urlinie progression into the two-line octave at the cadence (bars ) and, in bars , into the three-line octave. From bar
on, the third-progression is stripped of the Phrygian tone and is stated four times
in a row in the simplest fashion, then {} in rhythmic diminution starting in bar
. It must be observed that in the last segment (bar ff), the fifth (here, c4) is
again planted above the third-progression and holds that position until bar .
From there the fifth falls in a widely spaced descending arpeggiation across the
third a
2 in bar to f 1 in bar : the fifth is dissolved, as it were, into the triadic
arpeggiation. Following the two octave tones, f and f 2 in bars and , f 1 of
bar confirms the f 1 of bar (see the graph of the Urlinie).

Bars ff. The recapitulation proceeds without substantial modifications. It

should be observed only that Beethoven avoids a full perfect cadence within the
tonic of the closing subject (bars ff) and instead retains the fifth in the upper
voice in order to make it possible to repeat the development section. It is precisely
this that confirms my conception of the connection between the closing subject
in the exposition and the development (see earlier): the close of the recapitulation illustrates best how c is in any case absolutely indispensable as the initiating
tone of the development sections descending fifth-progression, for the appear-

German synthesissynthesis pure and simple!

20

The b
1 does not actually appear in Fig. ; but the bass b
in bar is duly marked as IV.

[S.] See Tonwille , pp. /I, pp. and p. /I, pp. ; Tonwille , pp. /I, pp. .

20

19

Beethovens Sonata in F Minor, Op.

The binding together of a tonal whole through the diatonic system in all its
forms:
Bound together by the lineage of a tone: That which is a tonal lineage here becomes an achievement of artistic truth, not of theory. That which is a main key is
here an actual birth of minor and major from the tones F and D
, F minor in the
first and third movements, D
major in the second. The contrasting keys of A

major in the first movement and C minor in the last are like contrasts within
their own families: originally they are but harmonic degrees of the main key.
Bound together by the Urlinie as a key in the horizontal dimension: A fifth-progression and a neighbor-note motive in the first movement, again a fifth-progression in the second, and a third-progression in the final movement (where of the
first movement is dreamily recalled above ). Having thus come into being through
fifth and third-progressions, the whole lives a life of holy bondedness, sheltered in
its own forces, sufficient of itself with a temporal beginning and a temporal end,
both of which, however, revolve forever in the eternal heaven of the imagination.
Bound together by Ursatz, prolongation, harmony, diminution: Again a
world full of its own destinies and history. From out of the root of only a few intervals there arises a densely branched tree full of blossoms and fruit, yet the root
is everything! The miracles of birth are not enougheverywhere the seeding of
a future harvest is at work.
Bound together by registers, by low and high: The tonal body has definite
boundaries that arise from within, from its soul, and are not given to it from
without. And harbored in some miraculous, secret way within the legitimacy of
these physical features is a legitimacy vis--vis the instrument. The keyboard outwardly bears those soul-etched boundaries of high and low, but however much
they expand, it also remains ever true to itself without assuming the false cosmetic appearance of an orchestra.21
This is German synthesis, I repeat, the proudest synthesis ever created by the
human spirit in any land.
But what can all its forms of binding together say to a human race that,
though bound to earth and body, veritably sunders limb from body, life from life,
only to die in the face of dissolution instead of living a bond! A human race that,
devoted to death and not life, prefers to cherish a tool, the common pick-ax, in
order to demolish that which geniuses built for the common weal!

The edition of Opus that I prepared as part of my collected edition of the

Beethoven piano sonatas (U.E. ) could not, unfortunately, be based on the

autograph manuscript, which is housed in the library of the Paris Conservatoire,
and I had to content myself with the first {} edition and the Urtext edition.
Prof. Carl Krebs, editor of the latter, remarks in his editorial report: Mr. Charles
Lefebvre has most kindly compared the original manuscript . . . with my printed
copy. Perhaps a facsimile of the sonata will at some point provide an opportunity to make whatever emendations might still be needed.22
Neither the title Appassionata nor any metronome markings are by Beethoven (see Tonwille , p. /I. p. ).

Performers unfamiliar with the content of the sonata as represented herein will
inevitably go astray in all aspects: in tempo, musical character [Tongebung], manner of playing, freedom, and so forth. They then defend themselves, as a rule,
with their conception, whose freedom they even direct against the composer.
But each performer means only his own conception (as is also the case, in fact, in
political and social life) and, oddly enough, attacks that of another as an offense
against the composer. If I attempt here to describe the performance of this work,
I want to appeal not to a freedom of conception that belongs to me just as it belongs to other performers, but to the well-founded reading given above.
In the upbeat to bar , the hand delays the last sixteenth, but then proceeds all
the more quickly to the downbeat: it is an incontrovertible law that, in the absence of any subdivision, the effect only of steady quarters or eighths will be produced no matter how strictly the rhythmic values are observed. The arpeggiation
of bars tolerates no delay but rather requires an acceleration. The second finger points to c2 on the downbeat of bar , legatissimo, and to a certain extent from
22
The first edition of Op. was published by the Bureau des Arts et dIndustrie in Vienna in February ; Schenker probably consulted a copy belonging to his benefactor, Antony van Hoboken
(now in the Sammlung Hoboken in the Austrian National Library). Krebss Urtext-Ausgabe is a
three-volume Sonaten fr Clavier von L. van Beethoven, published by Breitkopf & Hrtel; his prefatory remarks are dated May .
The autograph score is now in the Bibliothque Nationale, Paris. A facsimile edition was published by Edition dArt H. Piazza (Paris, ); Edition Peters in Leipzig brought out a reprint around
.

[S.] See Tonwille , pp. /I, pp. .

21

tonw i l l e 7
above, and in so doing consciously projects the meaning of the fifth across bars
|, which answers the first fifth of the anacrusis. In bar , the first anticipation
(d2) must be played legatissimo.23 The trill from below begins exactly with the
chord in the left hand; this chord must be played more strongly than the first note
of the trill. The conclusion of the trill (the last three sixteenths), which includes
the second anticipation (e2), must be played strictly in tempo (see the note in
my edition): in this manner the hand is really entrusted with the first thirdprogression. In bar , the submotive in the bass follows without any ritenuto (be
sure not to allow a legato from the last eighth to the quarter!). The poco ritardando does not really take effect until the basss final submotive in bar . To gain
mastery of the figuration in bars without the assistance of the left hand, it
is necessary to hold the right hand at a steep angle and spring from one e to the
next. In bar the original distribution of the arpeggiation (see my edition) is
preferableand, again, spring away from the two es at the start of the first and
second quarters.24 The surprising effect of the dynamic contrast fp across bars
| rules out a ritenuto.
With the fortissimo on f in the downbeat of bar , Beethoven wants to ensure
that this note lasts through the crashing chords until the next note of the arpeggiation, a
, in bar ; yet this ff ought not be overly harsh or obtrusive, since it applies only to a note in a simple arpeggiation unencumbered by chords. It is wise
to save some strength in the first chords so as to be able to increase it as the goal
approaches; a p shading must even be resumed just before the end, with a
2, and
the final intensification and acceleration must not start until that point; any
holding back of the last eighth, be it for the sake of obtaining a more forceful attack of f 3 or only out of caution, brings with it the danger that it will, in effect,
grow into a quarter and, {} contrary to the performers intention, approximate
the preceding quarter-note values. If the left hand indicates exactly the middle of
the bar, it will promote the right hands syncopations in bars , , and . The
original fingering of bars returns in the continuation of the first arpeggiation

(see my edition), as if the crashing wave of chords had not intervened at all. It is
also good to dispense with the pedal here, but this must be compensated by letting the hands lie within the span of one octave at a time. The p indicated for this
continuation accommodates the dynamic level, which has meanwhile increased
and no longer permits an unaffected return to the initial pp of the antecedent
phrase. (Remember that the actual dynamic state of the consequent phrase, despite the ff interjections, is and remains piano.) In bar the right hand should
play the first chord piano, with the fingertips, and with the hand held high; it
should make a crescendo to the end of the bar starting with the following chords:
bar then sounds as if it were brought forth, so to speak, out of the rubble heap
of the ff chords. In bar , as in bar (see earlier), the left hand must also be ushered in more strongly and distinctly than the right; in spite of being notated as a
half note, b1 of the right hand must be released and the thumb should be placed
at the disposal of the trills conclusion without further ado, as in bar . It seems
that many editors do not even know that a half note indicated by the composer
need not be held out on the keys if the effect of a half note is produced by the attack or an illusion obtained from various other conditions (see Beethovens neunte
Sinfonie, pp. xiiiff/pp. ff ). How often, in fact, does it become impossible for
singers or wind players to hold a half note for its full value, as, for example, with
marcato, non legato, etc. And since the editors of whom I speak did not know,
moreover, what Beethoven was aiming for with the detachment of b1 from the left
hand, as explained earlier, they again (see bar ) assign this note to the left
handarbitrarily and, to their way of thinking, for the performers benefit.
The suffixed trill in bar must be brought to an end in itself, and the right
hand, using the fingering , must strike the third in the next bar completely independent of the suffixed trill, facing forward, as it were: b
2, after all, is the first
note of the interpolation, which has nothing to do directly with the preceding d
2.
A certain stress on the thirds of the right hand in bars produces the portamento aimed at by their dotted values; this effect is supported by similar stress on
the thirds of the left hand (with the effect of >). The dotting disappears in bar
: how plaintively, how wearily the right hand creeps through the final half steps
on its way to the goal d
in bar (again dotted!). It is appropriate to slow down
in the last three eighths of bar .
If the right hands c1 in bar receives some degree of stressobviously the
articulation of a
1 in bar remains paramountthen the legato effect is sufficiently assured by this means and the right hand may be loosened up in what follows, thus making it easier to play the octaves and rendering the play of light and

23
Schenker explains how to do this in his unfinished treatise, Der Kunst des Vortrags: the pianist must continue holding the first note even after the d2has been played, as if c2 were notated as a
dotted half note (in The Art of Performance, ed. Heribert Esser, trans. Irene Schreier Scott, New York:
Oxford University Press, , pp. ). Schenker also uses the term portamento for this manner of
execution; see Kontrapunkt i, p. /pp. . Further helpful information on Schenkers conception of keyboard portamento can be found in the editors note on pp. of The Art of Performance.
24
Hans von Blow, for example, altered Beethovens distribution of the arpeggiation between the
two hands in his edition of the sonata.

Beethovens Sonata in F Minor, Op.

shadow more attractive. One should not suppose that the Urlinie tones in bar
have to be projected above all else; since they seem to be so hidden in the weak
parts of the bar, they should actually be left in the shadows. On the whole, one
should move ahead with bar and hold back a little starting with the first quarter of bar : by doing so the beginning and ending points of the slurred figure
[in bar ] are set off from the whole of bar , thus making it truly speaking and
singing. This is repeated in the performance of the second third progression. Beethoven joins together the antecedent and consequent phrases by means of a crescendo in bar , the particular responsibility for which must be assumed by the
left hand, while e
1 is at rest. In bar , the crescendo must last until the end of the
bar, {} thus in no way anticipating the piano of bar through a diminuendo:
only this extreme strictness prepares the contrast between f and p in bars .
The execution of bars is inconceivable without knowledge of the true content. A familiar thematic subject gets out of jointone should therefore play it
accordingly. The performer, along with the composer, should also, so to speak,
lose the reins of synthesis: it is as if he is powerless against the expansions of bar
(e
2), bar (d
2), bar (c
2), and bar (b
1) that suddenly appear (from
where?), obtruding (to what end?), as if powerless against the sudden dynamic
shifts that accompany the shaking and unloosening of synthesis, and so he
should not just rattle off the bars in as polished a way as they seem printed on the
page. He should rush into the sf of bar ; he should let his finger strike c
2 (bar
) as if stiff and cold, and should observe with complete strictness the values of
the last two notes in the bar (do not introduce b
2 too early; do not play two
eighth notes!). The trills in bars , , and must be arranged exactly in sixteenths; thus the four quarters of the bar must always be felt without being especially hammered out. The left hand probably ought to take the first f
of the trills
in bars and (see the note in my edition), and bar may even be connected
to bar by means of the pedal, thus by a pedal legatissimo that represents a type
of syncopation. The first and last eighths of the run in bars (f
3 and E
)
must be tied together by >; the rhythmic diminution (F
E
) in bar requires
the same treatment. The pedal must take part in the run, namely, by imparting
individuality to each of the descending diminished-seventh chords; when the
pedal is coordinated in this way with the content, it acts like a single pedal, but
without the disadvantages of such.
  
i
The performer should have a portamento in mind when playing the first
Urlinie-progression c
1 b
a
. In accordance with this, the f prescribed at the beginning of the bar, which merely expresses the general dynamic state of the third

subject (in contrast to the piano of the second), must be moderated at first and
the main stress must be reserved for c
; this stress renders a crescendo to c
superfluous. By contrast, the repetition in the third and fourth quarters of this bar
must be played entirely in piano and without portamento in the repeated Urlinieprogression, as if the performer were only filling up a rest during a singers performance; but the piano probably ought to be enlivened here by a slight crescendo
surge through the second and third eighths of the third quarter. The Urlinieprogressions must be kept free of pedaling; one should use the pedal only on the
first and third quarters. It can be seen from the sketches published by Nottebohm
in Zweite Beethoveniana that Beethoven originally had no intention of repeating
the Urlinie-progression in the same bar;25 the way I recommend performing the
repetition is thus validated by the composers basic intention, which uses the repetition more as a means for creating a four-bar group (hence more as a formal
expedient) than as an essential part of the content. The ff in bar must initially
be realized by the first octave of the left hand, yet even here it should be more like
a sf (no pedal!); the right hand does not appear ff until the second quarter (pedal
here) and secures an even, regular execution of the arpeggiation within an artificial piano shading by stressing and springing away from the first sixteenth. The
diminished-seventh chord must be executed in both octaves with the fingering
(see my edition). Energy and tempo should be held back in the left hands
upward arpeggiation (a type of portamento is even required); an acceleration
without crescendo should begin exactly {} in the second eighth of bar , and no
soonerthis is a rejuvenation of the octave!and the crescendo should begin
only at the last moment, leading from the sixth eighth to the sf. Do not, therefore,
attempt to force the entire arpeggiation along with the four f
s into a uniformity
of time and energy! It would be a wasted effort, an offense against both musical
logic and the laws of the hand. The octaves of the left hands upward arpeggiation in bars , in particular, urge observance of these directions. Bar presents, for the first time, the conclusion of the subject with , which is why,
despite the extremely high register, it demands more expression than it would be
accorded in normal performance: g3 of the third quarter should therefore be approached in non legato with a crescendo, which can only be achieved by holding
the hand at a steep angle during the last three sixteenths of the second quarter;
the inflections of a third (g3 b
3 and d
3 f
3) in the third and fourth quarters

>

25
Gustav Nottebohm, Zweite Beethoveniana: Nachgelassene Aufstze (Leipzig: Rieter-Biedermann,
), pp. .

tonw i l l e 7
but where, surely, the repeated sf on g
3 but also the struggle between the roots A

or B

require greater expression. In bar the approaching half step must be
heralded by the minor-mode inflection. In bar the prescribed crescendo must
be carried out only in the first and second quarters; in the third and fourth quarters, by contrast, a piano shading must be inserted artificially as the presupposition of the crescendo that reappears in the next bar: while the intensification of
the crescendo seems to move through just one tone (A
), inserting p in between
makes it possible to use appropriate expression in playing not only the third progressions that aim toward the next section {} (playing them more delicately
than one hears them played) but also, in particular, the soulful anticipation of the
second eighth in bars and (both hands using a slight portamento with
>). However, in bar the crescendo must continue to take effect in the third
and fourth quarters; here, too, the left hand must release A
over the penultimate
eighth, so that the execution of the following eighth is not hindered by the hand
stretching an octave.
What was said about bars applies to the performance of the second
subjects third-progressions that return in bars ff. The left hand observes the
strictest legatissimo throughout, particularly in the transition from the third to
the fourth quarters in bars , , and : the legato then projects in its own fashion the long path of the lower voice from D
(bar ) to d
1 (bar ) as merely
a transfer into a higher register. In the second half of bar , an acceleration
should push toward the chromatic A  and toward the forte of the next bar. The
tempo must be moderated slightly for the sake of the minor mode in bars ;
but an acceleration must take place again in the second half of bar (as in bar
). In bars there should be a > from quarter to half in the right hand,
but, conversely, there should be a < from the second quarter to the end of the
bar in the left hand while the upper voice is at rest. In the arpeggiation of bars
, the performer should embrace the disciplined approach that the master
has himself taken: he should make us see, as it were, the increasing oppression of
the right hand by the left, especially in bar ; he should underscore the goal tone
d
3 in the downbeat of bar ; and he should not overlook the eighth-note value
of the peak tones in the individual arpeggiations. Only such an artistically true
strictness of execution can truly enact the vehement passion of this moment: the
frenzied execution normally seen in this passage, which renders the masses of
tones indistinguishable in a flurry of pedaling, is just noise and not any sort of
passion. In bars it is acceptable to play the quarter-note d
s of the left
hand using two fingers at once, because of the stress required, but the preceding

should almost be tinged with legato. (Regarding Beethovens manner of seeking

expression even in the highest registers, where the likeness of the human voice,
the bearer of all musical expression, is lost, see my commentary on the first
movement of Op. , in the Erluterungsausgabe, p. /pp. .) One should
play bars [recte: ] quickly, in one fell swoop, almost as if one were actually playing just one bar and the motive just once; nevertheless, in all these bars
the e
recurring in the fourth quarter must be underscored by an unobtrusive
arpeggio, for this is the best way to ensure that the tone will continue to sound
into the following bar. The c
2s in the first three quarters of bar must be connected together as if in legato; the pedal disappears in the fourth quarter; the right
hand connects the e
3 sixteenths to one another and slows down in the last two
sixteenths.
Across bars | one simply proceeds as in the preceding bars, whereas
across bars | one should pause a little; the fact that one would thus plunge
headlong into the new E triad is what prevents one from playing this like the
anacrusis to bar (see above). A seemingly involuntary arpeggio must be recommended in bar ; this brings the short appoggiatura a2 above the left hands b1 in
such a way that the right hands g 2 and e 2 are struck at the same time as the d2
in the left hand;26 similarly in bar . In bar the sf must be approached by an
acceleration. The arpeggiations in bars ff (including the upbeats) must be
played differently than in bars : here, non legato, with accents only on the
downbeats throughout. The same is understood in bars and , where the first
sixteenth of the left hand belongs simultaneously to the upward arpeggiation of
the right hand and where, precisely for this reason, the fingering of the right hand
must also be halted in such a way that it seems as if the right hand had actually
struck that note on the downbeat (that is, with the rhythmic value of I
   ). In the
third and fourth quarters of bar the left hand must guard against any rushing
of the sixteenths: one ought not simply want to play away technical difficulties of
the sort that are demanded by synthesis (here, for example, the very long path to
C must naturally be accelerated at the end); such technical difficulties, even if
they are not exceedingly great, want to be approached and played as if they were
difficulties. One should therefore give the impression of requiring effort to play
bars , where there is unquestionably far less actual difficulty than in bar ,
26
The short appoggiatura is not present; in his edition of the sonata, Schenker expresses his belief that in this bar, and others in which no prefix is specified, it may nevertheless be appropriate to
introduce one.

Beethovens Sonata in F Minor, Op.

eighths must of course be restrained accordingly. If an actual piano in bar

were to follow directly on the heels of the rumbling in the preceding bars, it could
only produce the deplorable effect of a ppp and sotto voce; for that reason, the intensity level must be kept above piano, at least in the first half of the bar, and must
lead into p with a > starting only in the third quarter.
If the right hand is held at a steep angle as it passes above the black keys, it will
then be possible for it to obtain a legatissimo exactly as in bars , particularly if
the fingering I recommend is used. A certain stress is suitable only for each f in
which the hand rests after its flight. Pedal once in bar , twice in bar . In bar
the filler voices must be released at the fourth quarter in order to be able to play
the anticipated trill and the third progression with even greater suppleness. A >
leads from C to D
in bar . The left hands leap of a third in bar must be executed with a comfortably loose hand. The left hands half notes in bars and
must be played expressively. In bar one should at first let the right hand take the
lead in the crescendo and only let the left hand start contributing to it in the third
quarter. The pedal disappears in the second quarter of bar ; in the third and
fourth quarters of the same bar, the eighth note cs must be played rather strongly.
What was said concerning bars ff applies to the performance of the consequent phrase. The sf in bar is especially expressive, the last sixteenth, impassioned. In bars ff the pedal should again be divided (in place of a single pedal)
in conformity with the diminished-seventh chords, as in bars .
{} In bars and the shift from the hemiola to the figure of six sixteenths requires a certain caution; a scarcely perceptible ritardando before the
first sixteenth of the fourth quarter will make it possible to articulate and perceive the latter figure more clearly. Beneath the stationary half notes in bars
, the left hand should continue the crescendo. In bars ff there should not be
any clumsy or inordinate ff noise made for the sake of an alleged passion (see
what was said above concerning bars ff). The rhythm flexes, as it were, in four
quartersthe left hand takes care of thisbut the right hand has the task of
closing off each arpeggiation with a crescendo to the final peak tone (an eighth,
not a sixteenth!). Even in bars , the eighths must still be brought out; only
starting in bar does the left hand completely usurp the emphasis, until reaching d
1 in bar . The right hand finally takes over the lead with b
3. The masters
execution, which disdains giving any assistance to the right hand, must be recommended for the arpeggiations of bars ff.27 This is only made possible by the

sharpest accentuation of the quarters; moreover, a crescendo must be used to help

conclude each of the large sweeps of arpeggiation (the fourth quarter in bars ,
, , ). One should not yield immediately to the ritardando in bars ff,
since the five repetitions of the submotive stand in its way; one should perhaps
start the transition to adagio in bar with a modest reduction in tempo, yet this
adagio is not to be understood in the usual sense of the term but, rather, only as
a somewhat greater broadening of the tempo.
One should plunge into the pi allegro with a violent gesture, heedless of any
danger, and extract from this rash act the vehement passion that is indispensable
for the performance of the stretta. It is understood that the sf in bar entails
> as a retreat. The performer must be fully cognizant of the content if he wants
to do justice to the conclusion of the subject in bars and , and to the completion of and in bars ff, in particular. Bars must be connected together with a continuous crescendo, but this presupposes another suppression of
force on the downbeat of bar . The attacks of the last two chords in bar
must surpass all the preceding ones in vehemence, and for that very reason they
must followed by a moments pause, as if their deluge had ceased completely. The
piano indicated for the expansive arpeggiation of the principal motive in the following bars must still remain rather strong at the beginning, due to the ff that directly precedes it. The left hand plays beneath the right and must be responsible,
in particular, for a legatissimo effect in the arpeggiation f 1 f in bar , something
which the right hand must first make possible with an appropriate fingering (see
my edition).
The tempo marking Andante con moto is intended to direct the performers
inner eye to the whole of the ascending register transfer that binds the two sections together. In addition, there is the direction piano e dolce, which desires an
open musical inflection, ready and able to reflect an openly animated soul in an
openly animated manner. Thus there is no basis here for lapsing into an Adagio
in bars and , as is usually the case, nor any basis for a mysterious sotto voce
laden with devotional trembling.
One should move to the second neighbor note on the fourth eighth of bar
without lingering obtrusively in the second quarter of bar (the first neighbor
note); one should bring this note out with > and then ease off in bars and .
The last thirty-second in bar should float lightly; then, as if in portamento, one
should pass somewhat more heavily through the fourth sixteenth in bar . The
fifth-progression {} in the upbeat of bar (see the fingering in my edition)
must be played as if divided into two parts: pause somewhat on F in order to pro-

27

Blow had advised players to use the left hand in the lowest portion of these arpeggiations.

tonw i l l e 7
ceed all the more lightly with the last thirty-second, as if wanting to play this note
and the next octave all at once. Approach the second neighbor note, the last sixteenth in bar , in the same manner and then fall away from this note in bars
as if it were a source of light. One should not overlook the suspension, which
requires the release of >. The descending fifth-progression in bar is performed as in bar . In bar both hands must have a feel for the contrary motion
of the leaps of a third, g
e
and e
g
; the thirty-second should then lead laxly
to the downbeat of the next bar, where, to satisfy the choral design, c1 must be
provided with a slight stress. The same applies to bars . The last thirtysecond in bar expands, as it were, under the pressure of an amply intensified
crescendo; with the surge of this crescendo behind it, the hand then strikes the
chord of the downbeat with an even more pronounced rinforzando. Conclude the
theme in bars in an elevated stylebut piano nonetheless.
The bass notes on the after-beats in the first variation force an acceleration in
the overall tempo; it is as if one wished to become acquainted with the bass notes
belonging to the right hands chords the sooner the better. The bass notes must
be played poco marcato in various shades of non legato. Use a > in the fourth
eighth of the bass in bar and a diminuendo within a seemingly involuntary acceleration in bar . A > also suits the neighbor note across bars |. Bars
must be played like bars . The difference presented in the analysis between
the second section of the first variation and that of the theme necessitates
changes in performance. In bar the peak tone d
1 is still absent (see above) and
therefore the rising sixteenth notes recede in significance; even playing through
them at a steady tempo would mean underscoring a significance that is not appropriate for them. It is necessary to accelerate: and once it is used in bar , this
type of acceleration must then be applied to the analogous motives in bars
as well; in bar this acceleration finally coincides with the acceleration in the
ascending register transfer, whereby the two accelerations now gain in truth. The
intensity increases toward the end of this bar, so that the piano of bar follows
without transition; to be sure, the last sixteenth of the left hand in bar must be
allowed to run its course. In the first ending (bar ) a seemingly involuntary acceleration must be combined with >; in the second ending there must be a significant holding back in the penultimate eighth, for otherwise, given the preceding syncopation, it will sound like I
  , no matter how mechanically and exactly
it is counted out.
One must above all bear in mind the sempre ligato in performing the second
variation: accordingly, the first sixteenth in each upbeat and downbeat are sus-

tained, as are those sixteenths that convey a dissonance or a suspension (for example, the seventh sixteenth in bar or the third sixteenth in bar ). The effect
of eighth-note values produced by this artifice must be heightened in the upbeats
by using a portamento from the first to the second sixteenth; only bar has to be
played uniformly throughout, without sustaining notes and without portamento.
Otherwise, the stresses are placed as in the theme. In bar , despite the rinforzando, the first stress must be reserved for the fourth sixteenth, but g
2 must
then of course be played even more expressively. In bar the beautiful open tone
must be maintained, as in bar of the theme.
The basic dynamic of the third variation is piano; one could easily be deceived
by the sf in bars and (and in the corresponding bars of the repetition), but the forte in the other bars, due to its contrast, confirms {} the piano.
The left hand must fully support the right hands sf notes, as if the first thirtysecond in the second and fourth eighths was perhaps a sixteenth, if not an eighth.
This results in a two-voice sonority (first a fifth, then a sixth) that prepares the f
chords in both sonority and hand position, as it were. Another manner of playing, proceeding literally, would bring out the sf; the sf notes would then sound
hollow and obtrusive instead of maintaining the original expression. As in bar
of the theme, a > befits the last eighth of bar . In bar the thumb must be
used on the black keys (see the note in my edition). When played in the manner
recommended here, triads are formed on the sf in bars and . In the repetition of bars ff the melodically dominant tones in the left hand are placed on
top of triads: special stress sets them apart from the lower tones. In the second
section of the variation (bars ff), the choral aspect must be brought to the fore.
The right hands repetition of a
2 in bars necessitates a division into
groups: this division is defined by the arpeggiation motive, which stretches from
the second to the fourth eighth of the left hand in bars and and consequently suggests a slight separation of the three a
2s that lie above. However, all
six a
2s should be connected together by a large diminuendo that leads away from
the sf: they must thus be bound together dynamically; but logically, as it were,
they must be brought out as two separate groups. (My suggested fingering corresponds with this.) A slight acceleration suits the diminuendo that fades out in the
second group (the last three a
2s in bars and ). In bar the right hands g
3
should receive the first stress, and the left hand should place its first stress on the
third thirty-second. In bar the right hand consciously underscores the regular
meter in the third eighth and in so doing provides all that much more support
for the left hands syncopation. The suspension in bar slows down a little,

Beethovens Sonata in F Minor, Op.

which is why a slight acceleration has to compensate in the three eighths that follow (see bars and ). In bar the holding-up of the suspension must also be
expressed by a special fingering in the right hand (see my edition). But in bar
as well as in bars and it is even more necessary to lean forward from the first
toward the second sf in the left hand, for by this means alone is the required effect imparted to the second sf. In bars the pedal must be applied in syncopation, as if the first chord was tied over from the second to the third eighth
and the second chord from the fourth eighth to the next downbeat; yet the left
hand should not remain on the keys. In the second eighth of bar this results in
the effect of f 1 coming in legato, as it were, from the left hands g
1. The tonic of
bar ought not in any case be brought into open and sharp contrast to the preceding dominant; rather, the ff should cover over the combination of harmonic
degrees until f 1 is reached; the diminuendo should not start until this note, with
a slight slackening in the last two thirty-seconds.
The performance of the theme applies to the coda. Only the sixteenth-note
figure in bars causes difficulties. One should play the sixteenths a
and g

in bar in portamento, keep the following sixteenths flowing until the third
eighth of bar , and only then slow down with the concluding notes of the fifthprogression (F and E
), at which point a legatissimo has to connect the two sixteenths of the third eighth. The same manner of playing must then also be observed in the next sixteenth-note figure.
In bar the complementary rhythm of the sixteenths must be projected, as
if the four sixteenths lay in one hand. In bar the two quarters must be executed
in legatissimo (see my fingering) and likewise d
2 of bar , despite the arpeggio,
must be connected in a strict legato: {} only by that means will of the Urlinie
be expressed with the surprise of the deceptive cadence.
Except for the allegretto marking in the last movement of the Piano Sonata
Op. , No. , perhaps none of Beethovens tempo markings has been so seriously
misunderstood as that of the last movement of Op. , Allegro ma non troppo. For
the sake of that Appassionata title, which some fool introduced into the world,28
performers hurl themselves into a prestissimo frenzy and make a stretta long before the stretta beginsindeed, to someone who has no access to art, such a hint
as this is more welcome and more compelling than Beethovens clearest wish. In
reality, the masters tempo marking relates to the movement of the third pro-

gression in bars , which, with its diminution in sixteenths, must therefore

be lively, to be sure, but not at all passionate. Think here in four two-bar units
and let these units and their principal tonesthe very tones of the Urlinieprogressionfollow one after the other no more quickly than is needed to produce the impression of a tender sadness, as if sighs escaped the Urlinie tones and
softly drew them together.
The acceleration that constitutes the content of bars (see earlier) must be
reflected in performance. Yet the fourth eighth in bar must nevertheless be
taken somewhat more broadly, as if recalling how it derives from the quarter that
was originally imagined here in the second beat. Mounting passion then causes
the fourth eighth in bar to be an actual eighth. From the upbeat of bar it
should become increasingly agitated and accelerated. Beethoven composed this
acceleration, thus it must be determined not with a metronome but rather only
with an ear that has learned to follow the will of the tones.29 For the sake of the
eighth rests in bars and , the quarters e2 and e1 must be stressed in such a way
that a limit, as it were, is set for each note. The ebb and flow of pcrescendof
diminuendopp in bars must above all be placed in service of the step cb

across bars |, hence this step must be played with a clear crescendo from the
upbeat of bar to the downbeat of bar ; the b
that arrives here should be allowed to sound through bars , , and and only in bar should it fall with a
diminuendo to pp.
Since the arpeggiation motive that conveys the Urlinie tone begins in the upbeat of bar , and thus in a weak beat, it is appropriate to use a > that extends
to the end of the motive in bar . The Urlinie tone on the downbeat of bar is
delivered without the performers doing; the voice-leading and meter support it,
so that our ear can be sure of it even if it falls in the shadows. Both of the following two-bar units are played in the same manner. The final two-bar unit of
the group (bars ) must be played as an augmentation (see above), but for
this very reason one also has to delineate its individual building blocks with an
undulating hand; this wave-like motion of the right hand accords with the left,
which certainly does not play its eighths with the hand held stiffly in just one position. The final sixteenths of the left hand in bars , , and belong not to the
following quarters but rather to the quarters that preceded in bars , , and ,
of which they are arpeggiated tones; it is therefore a mistake to relate them, in

28
The title Sonata appassionata appeared as early as , in a version of the sonata for piano
four hands published by Cranz in Hamburg.

29
dem Willen der Tne zu folgen: a clear reference to the title of the publication in which this essay
appears; see also the essay on Beethovens Sonata Op. , No. , in Tonwille , p. /I, p. and note .

tonw i l l e 7
performance, to the following quarters, as unfortunately always happens, even if
presenting them in their true relationship is certainly a technical difficulty of the
highest order. The consequent phrase should be played in the same manner as the
antecedent phrase. The left goes over the right when the hands cross. The difficulty of the tone repetition c2 c2 in the left hands motive automatically forces the
right hand in this passage to exercise caution as well in crossing over from the last
sixteenth to the downbeat.
{} The left hand in bar traces a gentle arc from the root to the last sixteenth (as in bar ). The crescendo sign in bars applies to the four occurrences of a
in the melody, which is in that way tied together above and beyond
the rests. Play the sf in bar with warmth. In bar the left hand (see my fingering) connects the second, third, and fourth eighths in legato and matches both
the retreat and the compression of content with a slight acceleration. Far more
difficult is the task assigned to the right hand in all these bars: it has to spin out
the c2 that carries over from the consequent phrase of the first part of the subject;
indeed, the right hand has to shade it, doing so whenever it has quarters and the
left hand rests after expressing a sf, but nevertheless has to let it ring forth as if it
were only a single tone. The intervention of the neighbor note in the fourth
eighth of bar does nothing to alter this; on the contrary, this only enhances the
effect of the seemingly stationary c2. It stands to reason that the second part of the
subject, like the first, must also be tinged with the dark tone of sadness. The performance of the antecedent phrase also pertains to the consequent phrase (bars
). Only the following need be remarked: the pedal disappears by the second quarter of bar and does not return until the downbeat of bar . Bar is
played in the same manner. In bar the left hand holds onto the first sixteenth,
which presents the suspension (chord of the major seventh) within IV (see
earlier).
The first subjects arpeggiation motive employed in bars ff serves only the
inner voice (see the graph of the Urlinie); by tying the two-bar unit together with
< >, Beethoven here infuses it with a new breath. The four two-bar units
breathe as if in gentle, regular draughts and the melody, as if taking secret delight
in its own breath, is drawn forth delicately, as delicately as it appeared on the
scene in bar . All the more pressing, finally, is the crescendo from bars to
the sfp of bar .
In bars , the right hand has to project a single line arching above and
beyond the many rests (a descending register transfer, see above); nevertheless,

auxiliary accents, however difficult, are needed on the first sixteenth of the upbeat in bars , , and , and on the third sixteenth in all the other beats. The
suspensions awakened in this way reverberate through the line and make it
seem slightly ruffled. One should move more quickly with bars and somewhat hold the downbeat of bar . Play the consequent phrase (bars ) in
the same manner. Note only this besides: in bar (and ) slow down in the second and third sixteenths, and accelerate from the fourth sixteenth, where the
hand projects the descending fifth progression g2 c2.
In bars only g2 in the right hand and g1 in the left must be brought out
with a forte; even the immediately adjacent a
2 and a
1 must be played piano. The
emphasis of these two forte markings, the imitation, the high register, the pedal
(which must be withdrawn, however, in the upbeat of bar ), all of this together
leads automatically to the forte prescribed by Beethoven and renders it superfluous and harmful to make any physical exertion on behalf of individual sixteenths
or the dynamic state. The piano artificially inserted into these two bars now
comes in handy for the sf in the fourth eighth of bar , from which point another piano shadow falls on the eighths of bar , so that its repeated tones seem
to merge into a single sustained tone. And at the same time, the release of the sf
in bar benefits the left hand, for then the sixteenths in bar unfold piano
without any stress. All the sf in the following bars are sources of warmth and light
at the same time. Starting from the second quarter of bar , there should be a
scarcely noticeable holding back that announces the long-expected tone of resolution, contra-F.
{} Bars ff are to be performed in the same manner as the first part of the
subject; by their nature, the crescendo as well as the imitations in bars entail an acceleration, which then eases up in the upbeat of bar . Bars ,
too, are to be performed initially in the same manner as the first part of the subject, but then in bars one must follow the dynamics Beethoven uses to
bind the four-bar unit together. In bars ff the execution of the neighbor-note
motive is understood in such a way that only the neighbor (on the downbeat) receives a stress, and that is why special caution is required in attacking the first
note of each motive. It is this cautiousness that alleviates the technical difficulties
which the right hand has with the inner voice in the upbeat of bars and .
The high register in bars ff makes it necessary to insert an artificial p as soon
as the second sixteenth of bar , which undergoes a crescendo only in the next
bar. A p must likewise be inserted into bar in the left hand, for this alone

Beethovens Sonata in F Minor, Op.

makes it possible to play f 1 and f forcefully in the downbeat and upbeat of bar :
a forte, after all, only becomes effective when it is part of a sensible arrangement
of dynamics, for otherwise it remains a physical end in itself. The rhythmic shifts
in bars ff make an agitato necessary. The hand, held high, tosses off the eighth
of the downbeat, piano, and strikes the following quarter from above. Bars
speed up imperceptibly with >. Bars ff are to be played in the same manner as bars : forte applies only to the peak tones c2 d
2 e
2 f 2 in the right
hand and c1 d
1 e
1 f 1 in the left, whereas all the other sixteenths must be played
in piano, which also aids the unavoidably difficult fingering here. In bars
all the sixteenths following the peak tones (f 1 and f 2) must likewise remain piano.
A crescendo in the downbeat of bar leads to the chord in the last eighth. The
four-bar unit of bars is tied together dynamically with < >, but in
the next four-bar unit there must be a crescendo from the second quarter of bar
. In the four-bar units of bars and one should approach the
peak tones d
3 and e3 with a crescendo and an acceleration. The triplet in bar
must be played in the manner of a portamento, the eighth of bar must be
finely delimited. One should hurry the chords of bars and not slow down
until the upbeat of bar .
Concerning performance of the recapitulation: The arpeggiation motive in
the left hand in bar must be played as in bars . In counterpoint against
it is a figure that represents an augmentation into mere eighths, in contrast to the
augmentation into quarters (bars ). The rests following this augmentationobviously modeled after the sixteenth rests [that followed the original figure] in bars and make it necessary to accelerate toward the end, combined
with a crescendo: only then does the intended effect materialize, making it seem
as if the figure fell into a hole, so to speak. In bar an acceleration in the last
two eighths leads to the downbeat of the next bar. In bar the right hands d
3
in the downbeat must be played with beautiful, warm stress, whereas the first sixteenth of the left hand must be played in piano; in the upbeat, the reverse: the left
hand moves through the next three sixteenths in a crescendo toward the upbeat,
while the right hand plays c3 in piano in accordance with its retreat [from the rinforzando].In bars ff, the left hands chords must be played like the right
hands chords in bars ff, despite the sf on the downbeats of bars and .
Starting from the upbeat of bar , make a crescendo toward the Presto.
The sf in bar applies to a weak bar, the second of an eight-bar group, and
therefore must be approached with a crescendo and an acceleration, as if f 1 and a
1

were in legato. After the termination of the sf, the right hand strikes the keys from
above and in that way provides itself with the possibility of playing the following
chords with great elasticity in the fingertips. In bars ff, the right hand adheres
to the way it performed the arpeggiation motive in the first {} subject; the sf
apply only to the left hand. The sf on a
3 in bar is corroborated by a stress on
the left hands a
2 in bar .

More than enough has been written about Beethovens Op. theoretical,
hermeneutical, technicalI mention here, for example, only the contributions of
Marx (Anleitung zum Vortrag Beethovenscher Klavierwerke, pp. ff), Reinecke
(Die Beethovenschen Klaviersonaten, pp. ff), Lenz, Nagel, Pembaur, Bekker,
Riemann (Analyse von Beethovens Klaviersonaten, vol. , pp. ff), and so forth.30
But none of this is actually music literature: pens wrote, not ears. In none of these
writings is there is any connection with Op. , nor for that matter is there any
connection with music. What is to blame for this? False theories that cannot forge
a way to the facts of the matter, to tonal narrative and synthesis? Or feelings and
petty sentiments which, because they are addressed to themselves, are passed off
as artisticas Beethovensfacts!? All the same, Op. was not then present in
the world. Now it speaks out in tones and in words to mankindwill they be
able to understand it? Do they ultimately want to understand it?


Then tell me, Niceratos, what knowledge do to you pride yourself in possessing?
And he replied: My father, who wanted me to become a good man,
encouraged me to memorize all of Homer. And so even to this day I can
recite the whole of the Iliad and the Odyssey by heart.
30

According to the auctioneers catalog, Schenkers estate contained the following volumes: Adolf
Bernhard Marx, Ludwig van Beethoven: Leben und Schaffen, rd ed., ed. Gustav Behncke (Berlin: Otto
Janke, ); Carl Reinecke, Die Beethovenschen Klavier-Sonaten (Leipzig, ); Wilhelm v. Lenz, Beethoven. Eine Kunst-Studie. nd ed., vols. (Hamburg: Hoffmann and Campe, ) and Beethoven,
eine Kunststudie. Part , Das Leben des Meisters, new edition with supplements and commentary by
Alfred C. Kalischer, vols. (Berlin: Schuster and Loeffler, ); Wilibald Nagel, Beethoven und seine
Klaviersonaten, vols. (Langensalza, [H. Beyer and Sons], ; Josef Pembaur, Ludwig v. Beethovens
Sonaten, op. , Nr. und op. (Munich, ); Paul Bekker, Beethoven (Berlin: Schuster and Lffler,
); Hugo Riemann, Analyse von Beethovens Klaviersonaten, vols. (Max Hesse: Berlin, ).

tonw i l l e 7
But surely you must know, asked Antisthenes, that every rhapsodist
can also do this?
Of course I know this, since I get to hear one of them nearly every day.
And do you know of any people more stupid than these rhapsodists?

By Zeus, said Niceratos, that is most certainly not my opinion of

them!
All he [Antisthenes] meant by that, intervened Socrates peaceably,
was that they do not understand the sense of what they are performing.
Xenophon, The Banquet of Callias31

This extract comes near the beginning of a dialogue by Xenophon known as The Banquet (or
The Symposium), written about b.c, which recounts a dinner given some forty years earlier by a
rich man named Callias in honor of a young athlete, to which Socrates and some of his friends were
invited. The German translation that Schenker used (or made himself) differs slightly from the English versions I have consulted, especially in the last two lines:
No indeed, said Nikeratos, I most certainly do not.
And the reason is clear, intervened Socrates. For they do not understand the inner meaning of what they perform.
See, for instance, the Loeb Classical Library edition of Xenophons Anabasis, books IVVII, Symposium, and Apology, translated by O. J. Todd (London: Heinemann, ), pp. ; also The Works of
Xenophon, trans. H. G. Dakyns, vol. , pt. (London: Macmillan, ), pp. .
In Classical antiquity, the rhapsodists (or rhapsodes) were active from the sixth century b.c. as
traveling minstrels who performed epic poetry. One group, known as the Homerids, claimed to be
descendents of Homer himself; this may be the reason why Antisthenes refers to the rhapsodists as a
clan, or tribe, rather than as a group of professional singers.

The Recitative Erbarm es Gott from Bachs St. Matthew Passion

J. Seb. Bach: Matthuspassion Rezitativ: Erbarm es Gott1 {Tonwille , pp. }
t r a n s l at e d b y j o s e p h d u b i e l

The full text of this madrigalistic obbligato recitative reads:

Erbarm es Gott!
Hier steht der Heyland angebunden,
O! Geielung, o! Schlg, o! Wunden!
Ihr Henker, haltet ein!
Erweichet euch
Der Seelen Schmertz,
Der Anblick solches Jammers nicht?
Ach ja! Ihr habt ein Herz,
Das mu der Marter-Sule gleich,
Und noch viel hrter seyn.
Erbarmt euch, haltet ein!

(e)e
gb
, that the mixture of the minor mode is introduced, G minor in place
of G major (see Harmonielehre ff on chromatic alteration), so that the subsequent harmonies VI distinctly show the coloration of the G minor tonality
(which, as noted, could not have been reached directly from the key of C major).
The voice-leading prolongation in Fig. b shows the effort to avoid the immediate chromatic step CC  in the bass and skirt it through the elaboration of
thirds [Terzauskomponierungen], CBABC  that is, to dissolve it into a diatonic succession:3 after all, the chromatically raised tone, as a leading tone, entails the root lying a major third lower (see Kontrapunkt ii, pp. ff/pp. ff ). This
leads to progressions through thirds in the outer voices, combined with the
exchange which is needed to avoid consecutive fifths between the roots C and A.
{} The next transformation is shown in Fig. c. The passing b
in the upper
voice in bar actually demands that an e
follow in bar the first, quiet announcement of the key of G minoryet, on the other hand, this e
is contradicted by the following progression, to the root A, which is why an enharmonic
transformation of e
into d  must now be assumed.4 The ascent in bar is

Have mercy, God!

Here stands the Savior bound,
Oh! Flagellation, Oh! Blows, Oh! Wounds!
You executioners, cease!
Does not
The souls pain,
The sight of such misery, soften you?
Ah, yes! You do have a heart,
[But] it must be like the pillar of torture,
And even much harder.
Have mercy, cease!

The course of the music is shown in the adjacent illustration, Fig..

The recitative introduces an alto aria in G minor. Connecting to the last key
of the Evangelists preceding recitative, E minor, it begins with a C major triad,
which should immediately be understood as I of the key of C major, because of
the first motive in bars and because of the seventh b
that enters at once.2
Now it necessary to modulate from C major to G minor, which can of course be
done only with the help of a mixture of G major and G minor. The reinterpretation of I of C major as IV of G major occurs in the very first bar (see Fig. a), after
which the root of IV is raised, in bar , and then V and I follow in bars .
It is specifically through a diminished third in the chromatically altered IV, c 

3
[S.] For other examples of the avoidance of an immediate chromatic progression, see Tonwille ,
p. /I, p. , Fig. ; p. /I, p. ; Fig. c; p. /I, p. ; Fig. ; p. /I, p. ; Fig. , etc. [To be precise,
the chromatic succession is transformed into two diatonic successions that overlap; the same is true
of the other examples as well.]
4
Schenker combines two layers of elaboration in this graph, since the passing motion through b

(which, by the way, is not in the upper voice of Fig. c), but in the inner voice that will eventually
become the vocal part) presupposes the prior creation of the diminished seventh chord that includes
the motions goal, a. More than elision is at work in this case, however, for Schenker argues, quite uncharacteristically, that the pitch content of the goal chord is determined partly in response to a characteristic of the passing tone. Had his analysis created the chord first, it would presumably have been
spelled cf  ad , to lead to bg  d  ; and then the spelling e
, inferred from the passing b
, might
have been represented as merely a misinterpretation; as things stand, Schenker has it both ways, endorsing e
, at least as an initial interpretation, in the text, but parenthesizing it in favor of d  in the
graph. Curiously, he does not explicitly refer back to this discussion in connection with Fig. e, even
though an elaborate transformation there produces B
E
as a linear succession in the bass, and the

1
[S.] The introductory chorale fantasy of the St Matthew Passion will be the subject of analysis
in a later issue. [The analysis appeared in Tonwille , pp. ; see this volume, pp. .]
2
The motive in question is the descending third-progression e2 d2 c2, found in the highest voice
of the accompaniment, and presented explicitly by Schenker in his discussion of Fig. d. It is specifically against the alternative of the C triads being VI of the preceding E minor that b
speaks; the tonal
implications of this pitch in other contexts could of course be quite different.

tonw i l l e 7

for example, the outer-voice octave connected with the , in the second chord of
the succession (see Fig. a), or consecutive octaves (see Fig. b), Bach avoids in his
definitive elaboration, shown in Fig. c (cf. Fig. e). This realization of the reaching-over progressions creates effects of harmonic progression, IVV in E minor
and then in F  minor, which in this case, though, are to be understood as only the
results of the voice-leading, according to the voice-leading layers shown in Figs.
b and c. On the other hand, this realization makes it impossible to carry
through the reaching-over in the form shown at Fig. c, because d obviously
could not be presented directly above d , or e
above e .6 Hence the delay of the
reaching-over tones d in bar and e
in bar , which then has the further consequence that the root of V, D, must drop out and its place be taken by C, the second tone of the descending fifth-progression (D)CB
AG, which also introduces a progression of thirds beneath the upper voice.
Now for the final realization, shown in Fig. e. The motive of a third first appears in the accompanying instruments, in bars ; it is taken over by the voice

brought about through the technique of reaching-over5 (see Freier Satz): the
motive of the reaching-over adheres without exception to minor seconds (cb,
dc , e
d), and the tones that reach over are literally superimposed, d over b and
e
over c  (the diminished third).
Fig. d shows the working in of diminution: the motive of a third in bars .
A repetition of it is intended in bars , e
dc; the enharmonic transformation
of e
into d  is not expressed at all in Bachs notation, simply because the diminution offers no opportunity for it. In bars , the appearance of this first diminution is maintained even in the reaching-over successions: in the bass it takes the
form of neighbor-note motions, which form a counterpoint to turn-like figures in
the upper voice. Certain voice-leading disadvantages that could have arisen:

E
is again explained as D ; nor does he connect any of this explicitly to the question of e
and d  in
the diminished seventh chord near the end.
5
durch die bergreiftechnik: this technique is defined and illustrated in Der freie Satz, .

6
To be precise, what complicates the reaching-over is not the specific voice-leading transformation shown in Fig. c, but the major third in the fourth chord of the pattern, a feature common to all
the versions shown in the figure.

The Recitative Erbarm es Gott from Bachs St. Matthew Passion

with e
dc in bars . The reaching-over motives stand out clearly in the

voice; moreover, the second one also occurs in the accompaniment in bars .
Concealed in the ornament of the last eighth note of bar is a mutation of e 
through e  to e
:

CB
of the descending fifth-progression is assigned to the bass, but B
A to the

upper voice.

The vocal part follows exactly the course of voice-leading indicated here: lines

coincide with the descending third CA, lines (which end in a question)
and lines with the half-cadence-like figures of the reaching-over motives
within the ascending third.10

But then, precisely because of this ornament, the voice has to declare itself for

d in the motive at the beginning of bar , when it has already been established
that e
will occur. The notation, in voice and accompaniment alike, is thus only
what the motive demands: in the voice this is plain to see, but in the accompaniment it would not have been permissible {} to write e
immediately after e ; no
real enharmonic change can spoken of here. The vocal figure on the second and
third quarter notes of bar is to be understood as the horizontal unrolling of a
vertical succession:7

ll the preceding of course provides the only possible direction for performance: For the conductor, the sense of the bass ascending from C to C  must predominate, as well as of the breakthrough to the dominant of the new key. If every
modulation is already a motion in itself, and likewise every chromatic alteration
is an intensification, then a crescendo of course goes without saying here, where
modulation and chromatic alteration undergo such a powerful compositional
development. But how could the crescendo of the ascent AC  in bars be
sustained, unless the necessary breath had been drawn at the A? Accordingly,
however passionately the first cries of pain are given out (in forte), there must
then be some yielding within the forte over the root A in bar . I say some, since
most of the necessary effect is already provided by the descent of a third CA: the
descent of a third amounts to a decline in the forte as well, so that all that remains
for the performance is not to get in the way of this intrinsic effect of the voice
leading, i.e., to reinforce it gently.

The lower voice of the final version presents the passing tone B
on the second half note of bar , but leaves the concluding tone A to the upper voice of the
accompaniment, through voice exchange, and leaps to E
. As the sequel reveals,
the bass E
is nothing other than a D  that proceeds to E, by way of F  F, in the
sense of IV to V in A minor. Fig. d shows that E
really is derived from the upper
voice, and the rest of the succession of bass tones has its origin in a mere inner
voice. So that the root A (bar ) might descend to G (bar ) with a sense of necessity, a D 7 chord is interpolated, whose pressure (in effect that of VII in E
minor) compels the descent of A as a diminished fifth!8 There is a similar procedure in bar , where the same auxiliary chord produces the same effect.9 In bar ,

Spitta, in his study of Bach,

11 vol. , p. , writes: When the subject seems to

require it, Bach does not shrink from even the boldest means, as the striking enharmonic modulation at the end of the recitative Erbarm es Gott demonstrates.

7
[S.] Compare Tonwille , p. /II, p. , Fig. ; Beethoven, op. , no. , first movement, bars
; Brahms, Op. , bar of the theme, etc.
8
als einer vermindeten Quart: an error, as A is the fifth of the diminished seventh chord on D .
9
Here the word same requires interpretation: what is the same is that A is pressed to resolve
downward, through the superimposition of a chord that includes D  and implies an E chord to follow;
but the superimposed chord in this case is B7, simulating V of E (not D 7, simulating VII), the resolution of A is G  (not G), and the chord over the resolution includes the new dissonances E  and D  .

10
The text as Schenker presents it at the beginning of the article has eleven lines, not nine. If this
is correct, then the line numbers in this sentence should read: , with the descent of a third; ,
the question, and , with the half-cadence-like figures; and (unmentioned by Schenker) , with
the final reaching-over and cadence. Another possibility is that the two short lines Erweichet
euch/Der Seelen Schmerz should be condensed to one; but this would suppress an end rhyme between euch and gleich, and would in any event not solve the problem of the next two or three lines.
11
Phillipp Spitta, Johann Sebastian Bach, nd ed. (Lepizig: Breitkopf & Hrtel, ).

tonw i l l e 7
Spitta succumbs to a visual illusion here: notational necessity is one thing, real
enharmonic alteration another (cf. Harmonielehre, /pp. ).
{} Alfred Heuss, in his book Johann Sebastian Bachs Matthuspassion (Leipzig: Breitkopf & Hrtel, )which, despite obvious deficiencies in his consideration of voice leading and harmonic and tonal progressions, still offers plenty
of ideas that are worthy of musicians considerationwrites on pp. :

and arrangements, a p has been slipped in here, in systematic misunderstanding of Bachs intentions, and, again almost as a rule, this is how the
piece is performed. A singer may, indeed must, comprehend and sing the
piece quite realistically: these are cries being uttered. Correctly performed,
this piece, during which Jesus is whipped, exerts an elemental effect.
The wish for realism in the performance of this recitative is something I endorse; but what I want to see expressed above all is the reality of the voice leading (to use that foreign word again), whose realization is in fact the first prerequisite to any realism. First bring out the realism of the voice leading; then all
other realism will follow of itself. The art of voice leading will show itself to be
sympathetic in so far as it takes on the realism of the extremity of suffering that
it all too true to life. What Goethe wrote to Schiller on the occasion of a reworking of the dungeon scene in Faust applies here: A few tragic scenes have been
written in prose; through their naturalism and strength in relation to the rest,
they are quite unbearable. For this reason I am now seeking to present them in
verse, for then the idea shines through as through a veil, while the immediate effect of the horrendous material is muted.12

Here we are, and remain, completely on earth and look right into the
middle of the situation. In the recitative Erbarm es Gott, the famous
rhythm hammers persistently; it can never be rendered sharp and hard
enough, and the string orchestra must exert itself with exceptional artistic force if the motive is to retain its intensity to the end. Through habit
and incomprehension, conductors allow even this piece (which has a famous counterpart in Handels Messiah), almost as a rule, to be fundamentally ruined, and all realism to be taken from it. That Bach would have
wanted the accompaniment realized particularly powerfully in this case
can be seen from the absence of a p, which means at least that not only the
concertino, but the whole orchestra is to play f. In most piano reductions

12
Letter of May , , written at Weimar, no. in Der Briefwechsel zwischen Schiller und
Goethe, ed. Emil Staiger (Frankfurt am Main: Insel Verlag, ), pp. ; the quoted passage appears on p. .

Beethoven on His Quartet Op.

Beethoven zu seinem opus {Tonwille , pp. }
t r a n s l at e d b y w i l l i a m d r a b k i n

In a letter of addressed to Prince Nikolai Galitzin (Beethovens smtliche

Briefe, vol. , pp.

theme in this way. What is more, though, this version is required for the

ff:1

sake of a singing line, which must be given preference above all else.

Your Excellency!
With regard to the dispute, Zeuner2 is perfectly correct. The viola has a d

in this passage, that is

Moreover, this passage is based on

, in spite of the

in

the first violin; the latter should be understood as an ornament, or an-

. The motives3 are already written in the

anticipation, which every good singer would make, since Nature is an-

Schenkers source for Beethovens correspondence is, as always, Alfred Kalischers five-volume
edition of the letters, published in Berlin and Leipzig between and and revised (by Kalischer
and Theodor von Frimmel) between and . In its time, this edition was noted for its general
approach to the problems of transcription and providing commentary, but at the same time severely
criticized for errors of omission, duplication, and musical judgment. Although two rival editions of
Beethovens letters appeared around this time, it was not until the late s that a reliable German
edition appeared: Beethoven: Briefwechsel: Gesamtausgabe, commissioned by the Beethovenhaus,
Bonn, and published in seven volumes under the general editorship of Sieghard Brandenburg (Munich: Henle, ). The best English-language collection is The Letters of Beethoven, edited and
translated by Emily Anderson, in three volumes (London: Macmillan, ).
The discussion of a passage from Op. , which forms the basis of this essay, is taken from letter in Beethoven: Briefwechsel: Gesamtausgabe (letter in The Letters of Beethoven). It is only
a draft, beginning in French (the language in which Beethoven would have written to its addressee,
Prince Nikolai Galitzin of St. Petersburg); after two sentences Beethoven reverts to German, which
would have had to be translated. There is no record of Galitzin having received the finished letter.
The translation of the extract, and of those letters quoted and discussed in the Miscellanea of
Tonwille and , is based on the new Beethovenhaus edition; I shall note those passages in which
Schenker takes over Kalischers incorrect reading. For this essay, and for the Beethoven letters discussed in the Miscellanea, that follows, I have consulted Andersons English edition, and have
adopted her translation (with light modifications) wherever possible.
For background to the editions of Beethovens letters, see Alan Tyson, Prolegomena to a Future
Edition of Beethovens Letters, in Beethoven Studies , ed. Alan Tyson (London: Oxford University
Press, ), pp. ; the foreword to vol. of the new Beethovenhaus edition; and especially the foreword by Martin Staehelin and Sieghard Brandenburg to Ludwig van Beethoven: der Briefwechsel mit
dem Verlag Schott (Munich: Henle, ), which was a pilot project for the new edition.
2
Karl Traugott Zeuner (), a German pianist and composer active in St. Petersburg. He
had studied in Halle with Daniel Gottlob Trk, and later with Clementi; in turn he became one of
Glinkas teachers. Kalischer misread the name as Jenner.
3
les motifs. Kalischer read these words as les mot [?] ff, which Schenker (who did not have

chored in Art and, conversely, Art in Nature. Had I written

,4

however, the singing quality would have been ruined. And why? Because,
instead of the six-four chord in this passage
based on a root-position

, which is

, the chord of the sixthMMMMMMM

would have arisen, with F minor as its root

; and this would have

been contrary, foreign, to the overall course of the melody and harmony.
In short, Zeuner is perfectly correct; and it gratifies me that such an accomplished artist immediately guessed my intentions.
To this, the editor, Dr. Kalischer, remarks:
Let us begin with the examples of thoroughbass. The six-four chord a
d1 f 1, as an arrangement of dfa (root-position chord), is vehemently
rejected by Beethoven. Should the minor chord of the sixth be placed
access to the source) suggested changing to les not(es) f f, that is, referring to the Fs in the first violin and viola parts.
4
This brief example contained a printing error in Tonwille ; a correction was inserted at the end
of the next issue, Tonwille / (p.).

tonw i l l e 7
above a? Since, however, the root-position chord F minor chord is indicated precisely, the a is not correct; it must be a
, i.e., a
c1 f 1, the chord
of the sixth of F minor (the minor chord of the sixth!) . . .
. . . It is difficult to decide which of the late quartets the above-cited
music example refers to. Looking through the last quartets, it appears to
me as though the motive given above may belong to the Quartet in C 
minor (Op. ), on account of its rhythm. {} A passage from the Adagio ma non troppo, the entire passage alternates with a passage marked
Allegretto, before the Presto in C  minor,5 as follows:

correct (see Fig. c); only a musician by the name of Zeuner7 approved Beethovens voice-leading. When he was asked about it, the master explained the movement of the viola part.8 First he recalls that f f is indeed also contained in the
theme; but with f f he does not mean fortissimo, rather the two Fs ([first] violin and viola) in the previous thirty-second, and thus must have written not
mot but not(es). If one recalls the Urlinie of the theme on which the variations are based (the first two bars of the introduction are not reckoned in the bar
numbering):

etc. This might be refer to the original example in the letter. But that is
only one of the problems at hand!6

and the motive of elaboration [Auskomponierungsmotiv] in bars :

The editor has gone wrong in every respect. The passage in the letter refers to
a work dedicated to Prince Galitzin, the String Quartet in E
major, Op. , bar
of the second variation [of the second movement], and in particular to the last
thirty-second note of the third quarter-note beat of the viola part; see the starred
note in Fig. a.

then one will understand that the diminution-motive of the second variation (see
Fig. a) as a rhythmic reduction of the motive in the theme (Fig. ). And Beethoven refers precisely to these motivic origins of the passing harmony in the last sixteenth of the third quarter when he writes f f are already found in the theme.9
{} Viewed from the bass, the passing note represents a chord; see Figs. a
b. The first and second violins execute the diminution-motive in parallel
sixthsthese may in fact be understood as thirds (see Fig. b), and Beethovens
g
refers only to the first violin. Viola and cello begin simultaneously with the
first interval of a third, e
1 g
1 (first sixteenth of the third quarter), but the
middle third, d
1 f 1 (fourth sixteenth) is assigned to the viola alone, which plays
them as broken thirty-second notes.10 In addition, the first and second violins

This d
1 was felt to be in conflict with the notes a
g
2 e
3, which are sounding in
the other parts during that same thirty-second note, and c1 was regarded as more

Schenker, following Kalischer, writes Jenner.

[S.] See Tonwille , p. /I, p. . [Schenker refers to this letter in a passage concerning the inability of music historians and philologists to understand Beethovens meaning; he then goes on to
discuss a similar textual problem in a fugue by Bach.]
9
Despite not having access to Beethovens letter, which at the time was in a private American collection, Schenker invokes the concept of motive (in the form of diminution-motive) in his attempt
to make sense of Kalischers misreading of les motifs.
10
Schenker gives these notes as des1 fes1, an error.
8

Kalischer means the scherzo of the quartet, which is in E major.

Later (Schenker extracted only a third of Kalischers commentary), Kalischer admitted that even
this problem had not been solved: if the letter dates from (Beethoven: Briefwechsel: Gesamtausgabe narrows this down to around July , ), how could it refer to a piece that Beethoven did not
compose until the following year?
6

Beethoven on His Quartet Op.

play an anticipation (g
2 e
3), which collides dissonantly with the violas d
1.
Of the anticipation, Beethoven speaks explicitly; for passing tone he uses
song, melody; yet even these words, which remain for others appealing
terms of aesthetics, are used with complete theoretical correctness, since the
passing tone is actually an event in the horizontal dimension and so belongs exclusively to melody, not harmony (cf. Kontrapunkt i, pp. ff/pp. ff, and ii,
pp.ff/pp. ff ).11 He thus expresses himself very precisely when he writes
that, with c1 (as a replacement for d
1: see Fig. c), the singing quality would
have been ruined; in other words, to get from e
1 to c1 melodically, the voice
must use a passing tone, not a leap.
That is what Beethoven meant. Of examples of thoroughbass, which Dr.
Kalischer reads into those notes, there is not a trace.
Alas, when it is only a question of note-heads, the poor souls12 become
confused. If, however, one gives them dates, letters, and suchlikeit is still no
better . . . 13

Since it does not lie in things themselves, there must be something present in the disposition of men that stands in the way of their reception of
the truth, even the truth shines so brightly and its reception appears convincing in such a lively way. An old wise man sensed this, and it remains
hidden in his expression that is rich in meanings: sapere aude.14
Embolden yourselves to be wise. The energy of courage is among the
things needed to combat the obstacles that both the inertia of nature and
the cowardice of the heart place in the way of learning. It is not perchance
that, according to ancient mythology, the goddess of wisdom climbs out
of Jupiters head in full armament, for even her first act of duty is a warlike one. Even in her birth, she has a difficult battle with the senses, which
do not wish their peace and quiet to be disturbed.
Schiller, On the Aesthetic Education of Man15

11
These passages concern the status of passing dissonances, whether as single notes in secondspecies counterpoint or as polyphonic passing events arising from the combination of species.
12
den armen Menschenkindern: a biblical expression.
13
This essay was the subject of an article by Oswald Jonas, A Lesson with Beethoven by Correspondence, Musical Quarterly, (), pp. . Jonas observed that a lengthy remark in a Beethoven sketchbook from the year , the so-called De Roda Sketchbook, contained an earlier draft
of the letter, which sheds further light on Beethovens reasons for the correctness of d
1. The draft in
the sketchbook, which is now in the possession of the Beethovenhaus in Bonn (signature NE ), has
been reproduced in facsimile and transcribed in its entirety in Beethoven: Briefwechsel: Gesamtausgabe
(vol. , pp. ); the new edition of the letters also transmits further related remarks in contemporary conversation-book entries. For English versions of all this material, see William Drabkin,
Think of a Letter, Musical Times (autumn ), pp. .

14
Dare to know: Horace, Epistles. The saying gained a foothold in German intellectual writings
after Immanuel Kant quoted it in his essay Beantwortung zur Frage: Was ist Aufklrung? (An answer to the question What is Enlightenment?).
15
ber die sthetische Erziehung des Menschen, an essay in the form of twenty-seven letters published in , which became one of the most important texts arguing for a moral or ethical dimension of beauty. Schenker quotes from Schillers eighth letter.

Miscellanea
Vermischtes {Tonwille , pp. }
t r a n s l at e d b y i a n b e n t a n d w i l l i a m d r a b k i n
Florence, January ,

Dear Fiedler!
The marble statue comes to mind constantly. It is truly remarkable,
but the conception of a human body as a single entity has dawned on
me for the first time in the last few days. Up until now I had a general notion of a lot of strands of muscle moving this way and that, swelling, subsiding, forming humps here, flat planes there, and in the process knitting
the body together around the skeleton. Only now have I fully grasped the
conception of the entire musculature as a single pliable mass that clothes
the skeleton, hard in one place, soft in another, compacted in yet another;
as a mantle of muscle, its fabric capable of stretching here, contracting
there, and through which the bone occasionally protrudes.
I knew this all along, really, but just had no mental image of it. I
started with the separate parts and pieced my conception of the whole together, and consequently viewed nature as something composite. Now I
see any separate part always as just the whole viewed from one particular
angle. It sounds silly, and yet the moment in which I saw nature this way
was for me an event of fundamental import, after which under this conception my statue had acquired meaning and was totally changed. Concepts
are funny things, the way one eventually brings a whole host of separate
things under one hat! It is just a matter of bigger and bigger hats!

Like Goethe and Schiller, Beethoven also wrote xenia.

He presented them in the

form of canons to one or another of his contemporaries with whom he came into
contact, and also in the form of satirical comments in his letters. One such xenium is the romantic biography of the life of Tobias Haslinger,4 consisting of
three parts, which he commends in a letter to B. Schotts Sons of January ,
(Smtliche Briefe, V, pp. ff):5
Part One. Tobias appears as the apprentice to the Famous Experienced
Kapellmeister Fux6 and is holding the ladder to Fuxs Gradus ad Parnassum.

ing Works of Visual Art () and The Origins of Artistic Activity (). His theory of art impinged
strongly upon Hildebrands work.
3
Xenien: originally, small presents given to foreign dignitaries; later, epigrams or aphorisms, that
is, short witty sayings or poems. In Schiller and Goethe conceived a project to publish a series of
jointly authored two-line epigrams under the rubric Xenien; more than four hundred of these appeared in Schillers Musen-Almanch fr das Jahr . Many of these xenia are barbed attacks or teasing satires upon contemporary writers, institutions and cultural fashions; as such they caused a considerable stir in their time.
In later years Goethe wrote Zahme Xenien (gentle aphorisms), two- to sixteen-line rhyming
verses. A few extracts from these poems, of which about four hundred were published in six sets between and , are quoted in the Miscellanea of Tonwille , in the section German Form.
4
Tobias Haslinger (), Viennese composer and publisher. In he became the partner
of the music publisher S. A. Steiner; in , he took over the business under his own name. Haslinger
was responsible for the first collected edition of Beethovens works, which existed in manuscript in
Beethovens lifetime and was published between and .
5
Beethoven: Briefwechsel: Gesamtausgabe, letter ; The Letters of Beethoven, letter .
Schenkers source for all the Beethoven letters cited in the Miscellanea was Alfred Christlieb Kalischers Smtliche Briefe: Kritische Ausgabe (Berlin and Leipzig, ) revised by Kalischer and
Theodor von Frimmel during the three following years. Our translation of Beethovens letters here is
based on the original orthography, as transmitted by the new Gesamtausgabe.
6
Johann Joseph Fux ( ), Viennese composer and theorist, best remembered for his
counterpoint treatise Gradus ad Parnassum ().

Adolf von Hildebrand1

to Konrad Fiedler2
1
Adolf von Hildebrand (), leading German sculptor of late-nineteenth- and earlytwentieth-century Germany, theorist and writer on art, who first met Konrad Fiedler in Rome in
and became close friends with him. His The Problem of Form in Painting and Sculpture () was highly
influential. The marble statue in question was either a recent work of Hildebrands, or perhaps Verrocchios Woman Holding Flowers (ca., Florence, Bargello), which influenced his own work at that time.
2
Konrad Fiedler ( ), German art historian, theorist, and philosopher, author of On Judg-

Miscellanea

Then, as he feels inclined to indulge in practical jokes, Tobias, by rattling

and shaking the ladder, causes many a person who has already climbed
rather high suddenly to break his neck, etc. He now says good-bye to this
earth of ours and comes to light again in the age of Albrechtsberger.7
Part Two. The Fuxian nota cambiata, which has already appeared, is
now discussed with A. ( Albrechtsberger); the accented passing tones
are treated exhaustively, the art of creating musical structures8 is dealt
with exhaustively, etc. Tobias now encloses himself like a caterpillar in a
cocoon, and so undergoes a further transformation, appearing for the
third time in this world.
Part Three. The wings9 that are barely full-grown now make their way
hurriedly towards the narrow Paternostergasse,10 and he becomes Kapellmeister of the narrow Paternostergasse. Having passed through the school
of accented passing tones,11 he receives nothing from it but the bills of exchange.12 and so he creates his friend-of-youth13 and finally becomes a
member of several local learned societies,14 etc.
If you ask him, he will no doubt permit you to publish this biography.

longable musical brain that passes through the school of accented passing tones
with Fux, then Albrechtsberger (the names of the teachers could be changed),
without ever being able to grasp the general principle and at the same time understand how it might be applied to the particular situation. Thus it confronts
those who are unable to recognize strict counterpoint within free composition,
much less create prolongations that, while they appear unbound to any rule and
unrestrained in their freedom, are in fact a fulfillment of a basic principle of strict
counterpoint.
But does this angry epigram of Beethovens not also apply to the unprolongable musicians of today who, with wings that are barely full-grown, make
their way hurriedly to any narrow Paternostergasse?


{} From a letter to nephew Karl (June ):15
It was not right of those Mainz people to have done such a thing.16 But
since it has now been done, it does not matter. Our age needs stronger
minds to castigate these petty, deceitful, miserable wretches of human
soulshowever much my heart refuses to give pain to anyone. Besides, it
was merely a jest, and never my idea that something of this sort should be
printed.

The play on words, WechselnotenWechsel, comes to the fore. It is of course

harmless, and in no way the principal content of the biography. Rather, the
satirical meaning, by which alone this biography is elevated to the status of a
xenium, lies considerably deeper: it confronts the species of eternally unpro-

These words apply to the Haslinger xenium. That Beethoven really did not
want to have it published can be understood from the closing words of the letter
to Schotts Sons that was cited above, and even more clearly on a second letter to
the publishers, dated February , :17

Johann Georg Albrechtsberger (), Viennese composer and theorist, and the leading
composition teacher of his day. Beethoven himself had lessons from Albrechtsberger in .
8
die Kunst Musikal.[ischer] Gerippe zu erschaffen: Gerippe, in the plural, is usually translated as
skeletons, but it is likely that Beethoven had in mind the figurative meaning of frameworks or
ground-plans, as used by Emanuel Bach in his Versuch. Schenker may have been particularly attracted to this, as Bachs discussion of the Free Fantasy in terms of Gerippe and Ausfhrung provides
the basis for his essay Die Kunst der Improvisation, published the following year in Meisterwerk i.
(See in particular Richard Kramers note on this passage, on p. of the English translation.)
9
Flgel: probably another play on words, as Flgel can also mean pianoforte on account of its
winged shape.
10
Paternostergssl: a common diminuitive of the name of the street on which Haslingers music
shop was situated.
11
die schule der Wechselnoten Durchgegangen: probably also a play on Durchgang as a technical
term in music (passing tone).
12
die Wechsel, another play on words.
13
JugendFreund: a reference to Haslingers Musikalischer Jugendfreund fr das Pianoforte mit und
ohne Begleitung und zu vier Hnden, a twenty-five volume series of piano music for amateurs, begun
in .
14
geleerter Vereine: a favorite pun of Beethovens (gelehrt, learned; geleert, emptied).

Just for a joke, ask Tobias for the romantic biography of him which I
wrote, for that is the way to deal with people of that sort, these heartless
Viennese. He is in fact the person who advised me to have nothing to do
with you. Silentium.
15
Beethoven: Briefwechsel: Gesamtausgabe, letter ; The Letters of Beethoven, letter ; written
in August, not June.
16
Schott published Beethovens joke biography of Haslinger in their house journal Ccilia in
April , evidently misconstruing Beethovens earlier correspondence and without informing him of
their intention to go through with the joke. They added two canons that Beethoven offered to the journal (neither of which is concerned with Haslinger) and appended the composers signature. (The relevant pages from Ccilia are reproduced in Beethoven: Briefwechsel: Gesamtausgabe, vol. , pp. .)
17
Beethoven: Briefwechsel: Gesamtausgabe, letter ; The Letters of Beethoven, letter .

tonw i l l e 7
Those who derive from these words the opposite of Beethovens intentions
including Kalischer, the editor of the lettersdo him an injustice; they do not
understand his language.
Poor Beethoven!18

virtues but not my faults; but do not havesince a person must have failings
greater failings than I. Can one accuse Beethoven of making a real mistake by
writing jedoch nicht habe instead of jedoch habe nicht?
Poor Beethoven!
Neither your musical word [Ton-Wort] nor your verbal music [Wort-Ton]
speaks to the people. The true language of your love will never be comprehended
by a people who have been brought up only on the value of money, whose bodies and souls are untruthful, for whom language is merely an expedient for hiding the deceit that they practice upon themselves.

From the same letter to his nephew:

Be virtuous, be good; here you have an example how the whole world rejoices when people of that type are honoured as they should be.Be my
dear, only son; imitate my virtues, but not my faults. But since a person
must have faults, do not have worse faults than I, your true and faithful
father who embraces you.

From a letter to B. Schotts Sons in Mainz (summer ):

20

The overture you received from my brother was performed here a few
days ago. I received very high praise for it, etc. But what is that compared
with the Greatest Master of Tones on Highon Highon High, who is
rightly called the All-Mighty, while here below people only pour scorn on
it [mastery of tones]. The All-Mighty little dwarves !!! 21

The last sentence of this letter, a gripping confession by a man of the most heartfelt humility who, standing forever before the throne of God, is also willing to
bend down before a little shred of humanity, his youthful nephew. Kalischer
(p. ) comments as follows:19
Even the final sentence, which mentions his nephews failingsjedoch
nicht, da der Mensch fehlen mu, habe schlimmere Fehler als ich (?!?)
remains problematic, for Beethoven could not seriously have meant that
nephew Karl should have worse failings than those which he (Beethoven) attributes to himself!
And yet the sense is so apparent: Beethoven advises his nephew: Imitate my

This is Beethovens intoxication with God! The mighty forger of tones falls to
his knees before the forger of the All-Eternal, before the Master of Tones on
Highon Highon High, who rules over both chaos and form, over both the
thunder of the world and the ordered movement of the stars, who sets free the vibrations unleashed by the roar of the wind and brings them under control, tied
to the Tone.

18
On August , , Beethoven again wrote to Schott (Beethoven: Briefwechsel: Gesamtausgabe,
letter ; The Letters of Beethoven, letter ), expressing his outrage that the biography had been
published, protesting that it had only been written in jest, that it has always been alien to my character to hurt anyones feelings, and that the pun on geleert/gelehrt could only have been included in
a humorous sketch intended for a circle of people who are speaking in jest.
19
As can be seen from the discussion that follows, the word order in Beethovens final sentence
has given rise to misinterpretations. Andersons English translation agrees with Schenkers interpretation.

20
Beethoven: Briefwechsel: Gesamtausgabe, letter ; The Letters of Beethoven, letter . The
letter was written not in the summer of but on or about December of that year; it refers to a
performance of the Overture The Consecration of the House, Op. , that had taken place on Christmas Day.
21
The antecedent of it can only be inferred from the context. This remark, to which Schenker
had alluded earlier in the Miscellanea of Tonwille (p. /I, p. ), was added at the foot of the page.

Tonwille 89
(annual volume IV, nos. )

This page intentionally left blank

Brahmss Variations and Fugue on a Theme by Handel, Op.

Brahms: Variationen und Fuge ber ein Thema von Hndel, op. {Tonwille /, pp. }
t r a n s l at e d b y w i l l i a m r e n w i c k
But only when the two are united with each otherwhen the will freely complies with
the law of necessity, and reason enforces its rule throughout all the changes
of imaginationdoes the divine or the ideal come forth.
Schiller, On Naive and Sentimental Poetry1

third (tenth); follows, supported by the intervals , indicating the chord

sequence IIVI.
While the power of the imagination remains true to this Ursatz, all sorts of
melodic unfoldings, diminutions, motives and the like appear.

Theme

Brahms took the theme of the variations from a keyboard work in Handels
Lessons for the Harpsichord, composed for the English princesses, in the original
edition published by John Walsh (Kalbeck, p. ).2
In Handels ariathe graph of the Urlinie shows the structure of the theme
and the variations only in a single staff and in the treble clef (pp. ), the
structure of the fugue in two staves (pp. )the wonders of organic unity
originate as follows:
{} The Ursatz (Fig. a) represents the descent of the Urlinie notes . The
intervals below the , the fifth and the octave, indicate the I and V. Then appears
with V as a passing seventh that leads down to , which occurs again with I as a
1
Friedrich Schiller (), ber naive und sentimentalische Dichtung (), paragraph .
The two refers to the natural and naive, on the one hand, and to the human and artificial, on the
other.
2
The Aria, with five variations, is from the first of the Trois leons, first published in Pieces un
& deux clavecins (Amsterdam: Estienne Roger, ?), and reprinted in the second volume of his Suites
de pieces pour le clavecin (London: John Walsh, []); The princesses were Amalie () and Caroline Elizabeth (), the second and third daughters of King George II, to whom Handel was appointed as music master. See Jonathan Keates, Handel: The Man and his Music (London: Gollancz,
), p. .
The pagination varies in the several editions of Max Kalbecks Johannes Brahms, first published
between and . In the final, complete edition (), available as a reprint (Tutzing: Hans
Schneider, ), this reference is on p. .

tonw i l l e 8 9
The first voice-leading level (Fig. b) initially introduces the arpeggiation of a
third d2 f 2 in the upper voice, in place of the harmonically stated . The lower
voice immediately takes this opportunity to enliven its motion and also to accompany each third separately. Above the I, d2 and f 2 ( ) are thus set in tenths
orwhat amounts to the same thingin thirds [above the lower voice]: this is
one of the essential characteristics of the aria.
In Fig. c, d2 is at the same time retained in the inner voice, just as it actually
originated as shown in Fig. a, and it leads to c2 at the beginning of the V.
The next voice-leading level (Fig. d) now develops the melody, filling the
arpeggiations of a third with passing tones. In this manner the upper voice traverses the two spaces of a third, from the root to the third and from the third to
the fifth of the B
chord, always aiming towards f 2, of the Ursatz.
Now turning toward the , the upper voice shifts to the path of the inner
voice (see Fig. c), in order to create another unfolding of a third (c2 e
2), and by
means of this detour the appears in the melody. The bass voice falls from b
to
e
through leaps of a third (suggesting a VI chord between I and II).
The voice-leading in Fig. e presses on toward new growth, new features of
life. The Urlinie notes support embellishing neighbor notes and the increased activity of the upper voice calls for an expansion in the lower voice. The lower voice
avoids stationary bass notes, which would be too bare for the flow of the upper
voice and which would here simply yield dissonant passing tones [in the upper
voice]. Instead, it strives towards self-supporting chordstriads or seventh
chordsthat are freely based in the past of the previous voice-leading levels
(Figs. bd). Though it accompanies the ascent of the upper voice d2 e
2 f 2 with
b
cd in thirds (tenths) here (third-progression of the outer parts), it also
sounds as if it could be a canonic imitation of the ascent to (b
1 c2 d2).3 Thirds
in an inner voice accompany the unfolding c2 e
2 [in bars ].
At Fig. f, the singular and artistic unfolding of this same third is to be seen:
apparently motivated by the previous neighbor-note figure, d2 e
2, in the next
two thirds first the upper and then the lower notes are introduced.4 Only the last
third, c2 e
2, is filled in the customary manner with a passing tone.
Fig. g shows a further growth of the melody, still remaining entirely faithful

to the Ursatz: To reinforce the , a new neighbor note is inserted in the form of a
falling third-progression (e
2 d2 c2) and therewith the balance of all the neighbor notes is established for the whole. The increased unfoldings lead to new intervals, which in turn point the way to new chords. On the other hand, if the progression of chords in bars reproduces the closed cycle IIVVI (standing
for I) and if likewise the cycle VIIVV (standing for V) is completed in bars
, then in spite of all this activity in the foreground {} the last voice-leading
level apparently remains true to the fundamental meaning of the Ursatz (Fig. a)
and the first prolongation (Fig. b). In particular, the decisive feature shows
through, that is, the , f 2, remains fixed above the third of I.
The last evolution of the voice-leading thus determines the final shape of the
aria, shown in the graph of the Urlinie, except that a diminution of the motive
also enters here in the third quarter of each bar; it emphasizes
the main notes and

also carries within itself the characteristic rhythm
that permeates the whole
aria. (Moreover, the ornament in the fourth quarter of bar may also be derived
from this rhythmically diminished figure, as a twofold succession in yet further
rhythmic diminution.5)
All of these additions simultaneously promote the effectiveness of the harmony.
The form of the theme, small as it is, may be understood as a three-part song
form. The division of its contents is decisive here: the fifth-progression b
1 f 2 in
bars , the a1-section; the second fifth-progression a1 e
2 replying in bars ,
the central b-section; and a faintly diguised repetition of a1 in bars , the a2section.
The tripartite structure is further reinforced by the course of the bass, which
is particularly aimed toward this goal:

It falls through an octave in bars , then rises in bars to the fifth, and
from here goes in bars to the higher octave in bar , from whence in bar , as
5
noch rhythmisch verkleinert: in this translation, care has been taken to distinguish the two senses
of the English word diminution for music theorists: Verkleinerung, the reduction of note values in
a theme or phrase (which is always given here as rhythmic diminution), and Diminution, a term
Schenker used specifically for the creation of musical detail by the elaboration of a structure, which
is rendered simply as diminution).

The canon that Schenker suggests is realized in Handels third variation.

[S] See Elucidations, Fig. ; Tonwille , p. /II, p. , Fig. . [The illustrations in Fig. of the
Elucidations give the principal forms of unfolding (Ausfaltung). The example in Tonwille shows
the fugue subject from the Trio of Beethovens Fifth Symphony scherzo (bars ) containing a
similar unfolding.]
4

Brahms Variations and Fugue on a Theme by Handel, Op.

in bar , it again returns to the low register. There is no doubt that the two octave
descentssee the brackets in Fig. reply to one another. This is what confers
upon bars the effect of a formal section, so that, in considering bars as
transitional, the form can be no other than three-part.
The guiding outer parts lead naturally to meaningful inner voices. The d1 in
the first quarter of bar is obliged to move to c1 in the second quarter, and from
here back again to d1 in the third quarter, only in the sense of a reinforcement: it
is indeed the right of the free voices to shift between obligatory and reinforcing
characters, according to the requirements of the situation. From the last eighth
note of bar to the first quarter of bar the inner voice proceeds again in a strict
manner. In the obligatory three-part texture of the second and third quarters of
bar , where the succession
appears in open position, a reinforcement of the
upper voice appears in the lower parts, by which means -chords result in these

f
parts as well. This close position causes the harshness of the succession ea b


of the open position to be satisfactorily removed. In bar , f 1 in the second quarter (left hand) is just an incidental filler, d1 c1 in the second and third quarters a
line of reinforcement; b
in the second quarter moves to a, while the f of the bass
in the third quarter comes from the B
of bar (see Fig. , earlier). At the third
quarter f 1 can still be heard. However, its presence is strictly speaking unnecessary, since the ornament in the fourth quarter begins with f 1 as well. The keyboard style requires this kind of incompleteness in principle. In bars the
counterpoint is strict throughout.
Handel wrote five variations for this theme. Although their outward appearances are of a simple nature, in their internal behaviour they are full of a superabundance {}of the boldest voice-leading. To discuss them lies beyond the scope
of this essay; on another occasion the artistic significance of Handels variations
deserves to be discussed as well.6
Brahms wrote no fewer than twenty-five variations and a fugue for this theme.
Just as the older masters had already developed the variation set from an essentially loose form toward an organic structure with bold spirit, so the younger
master went further in enriching the synthesis in an essential way, as this discussion will demonstrate.
2

Variation . This variation introduces a simple figuration that does not alter the
voice-leading of the theme in the least. Evidently taking up the hint of a1 b
1 in
the second and third quarters of the last bar of the theme, the lower neighbor
note is used for decoration of the upper voice in this variation. (The connecting
quintuplet in the second ending of the theme is by Brahms, not Handel.) Since
the main notes thus return in every second eighth, the emphasis of the eighth
note comes to the fore and the sixteenth-note motion recedes, in spite of the fact
that the left hand imitates with two sixteenth notes in every second eighth; the
four sixteenth notes, which the right and left hand combine to produce in each
quarter-note beat, thus masks the true underlying rhythm of eighth notes.
The figure in the fourth quarter of bar (traversing an octave), acting like a
fermata,7 exceeds the normally required number of thirty-second notes ( instead
of ); likewise in bar ( instead of ). In addition, a play of imitations based on
the motive of bars , , , etc. (every fourth quarter), appears in bars and .

Variation . Again, as in the first variation, it is the lower neighbor that ornaments the main note. Herein lies the continuity between the two variations. In
contrast to the eighth-note motion of Variation , however, the second variation
exhibits instead a triplet rhythm, and consequently a faster motion. The following sketch reveals the changes with respect to the theme, and the derivation of
the diminutions:

6
Although Schenker never published a study of Handels variation technique, he used another
piece from the Leons in his essay The Art of Improvisation (Meisterwerk i). Furthermore, in his personal copy of Handels suites (Oster Collection, Printed Scores No. ), there are a considerable number of analytical annotations on the aria and variations, as opposed to text-critical markings, which
abound in the first eight suites.

im Fermaten-Sinn, that is, in the manner of a short cadenza embellishing a single harmony.

tonw i l l e 8 9

Brahms Variations and Fugue on a Theme by Handel, Op.

tonw i l l e 8 9

Brahms Variations and Fugue on a Theme by Handel, Op.

tonw i l l e 8 9
bars | imitates that across bars | (c 2 d2): f 2 in bar is thus approached exactly as d2 in bar , from above as well as from below.
The play of repetitions in the theme adapts itself naturally in this variation as
well, see Fig. c; here they emerge by a chromatic impulse from the inner voice
(see earlier), and the last of these leads to d2 on the downbeat of the third bar.
Fig. d, in which the wider compass of the voice-leading is clearly shown,
demonstrates that it is really the inner voices that carry the repetition: d2 leads to
e
2 as , accompanied by  of the inner voice.
The final realization [i.e., the score itself] also conforms to this wider compass:
The d
1 in the upper and lower voices of bar is merely a chromaticism of
mixture.
Bars follow the voice-leading of the theme; indeed in the second quarter
of bar , as if recollecting the mixture in bar , d
2 is briefly touched upon. However, the lower voice already brings back d  in the fourth quarter. In bars and
the {} chromatic activity verily takes over: that which is suggested in bar , and
which becomes continuous in bar , increases to a flood of semitone eighth notes
in the last two bars. The harmonic plan

In Fig. a the pure diatonic basis of the part-writing is still present. In bars
the lower voice unfolds a fourth-progression, which has the effect that the is
no longer the third above the {} tonic B
, but the fifth over G, the root of VI.
Nevertheless e
, the upper neighbor of , still appears here. In bar the lower
voicesee the [diagonal] dotted line in the graphlikewise contraposes g2, the
upper neighbor of , with another neighbor, that is e
1 before the d1 of bar ; but
since bar should again provide a parallel to bar , as is required for the synthesis, it introduces e
1 in the second quarter, which enables the third-progression
e
1 c1 to respond to the third-progression b
g in bar . (In any event, c1 must be
approached in contrary motion to the upper voice in order to avoid consecutive
fifths fb
g
c between the outer voices.)
Only after the completion of this parallelism do the outer voices, in bar , return to the voice-leading of the theme; see the dotted slur in the lower voice. The
attained in the third quarter of bar occurs over d1 of the bass voice. But since
this note is only a passing tone in the aforementioned third-progression e
1 c1,
the incidental third (tenth) [d1 f 2] at this point does not have the effect of a true
third within the underlying tonic chord. Indeed, the most significant difference
between the second variation and the theme lies in the fact that neither the nor
the [of the initial ascent] sounds as a third within the underlying tonic chord
(not bd
, not df ) at their first occurrence in the third quarters of bars and . Although they subsequently regain these positions in the first quarters of the third
and fourth bars, nevertheless this does not invalidate the above distinction.
The structural level given in Fig. b now intensifies, through chromaticism,
the inevitability in the movement of the outer voices. In bar the tonicization

VI 3 II requires a
(rather than a  ).8 By this means the inner voice gains the semitone pattern bc, which is followed by two further semitones: c2 c 2 and c 2 d2.
However, the neighbor note e
2 also leads to d2 from above. Once introduced into
the texture, the chromatic notes readily take every opportunity to enter the voiceleading, hence also in the third bar d
1 also appears as a passing tone in the bass
voice, replacing d2; the intended parallelism with bar thereby becomes more
convincing, and the only remaining difference is in the rhythmic displacement of
the passing tones, which extend in bar from the first to the third quarters, and
in bar from the second to the fourth quarters. Even the semitone e2 f 2 across

shows a pedal-point with cadential chords. Borne along with the warmth that the
chromatic notes introduce is also an impulse toward the higher register: the
upper voice exceeds f 2 in the fourth quarter of bar and leaps from g2 to c3 (giving an ascending register transfer of the and in bar ). Note, further, that g2 in
bars and , b
2 in bars and , and finally e
3 in bar , are merely detours occasioned by the diminution.
Variation . The final run of the second variation concludes with the first two
eighth notes of the third variation, as if it desires above all to regain the prescribed register for and . These two eighth notes then become the germ of the
diminution in the third variation, and this provides the connection with the second variation. Moreover, the aforementioned b
2 of the second variation that had
initiated the descending run is retained above the upper voice in the third variation; see the dotted line from b
2 to b
2 in the graph of the Urlinie.

[S] See Tonwille. , p. /I, p. , Fig. c. [This figure, from an analysis of Beethovens Piano Sonata, Op. , No. , shows the bass progression through the flattened seventh in bars of the first
movement.]
8

Brahms Variations and Fugue on a Theme by Handel, Op.

fourth variation. Even from a purely superficial point of view, such a manner of
linkage must at least testify to the vivacity of an artistic imagination which, because it has and is true life, immediately draws blossoming new life out of everything, even the very least idea. Moreover, this connection simultaneously becomes an intrinsic principle of the new variation, the law of its diminution, as
was the case at the join of the second variation to the third. If it is also important for the fourth variation to regain the register that the Urlinie tones of the
theme describe, then an acceleration is necessary for the diminution, which
must reintroduce what was omitted in the shift to the lower octave. In fact the
variation is governed throughout by such an acceleration. Thus in bar the
lower voice introduces the first diminution of the rising motive in the space of
two quarter notes, rather than three, so that the rushing upper voice simultaneously reaches the correct register in the second quarter at the first to third sixteenth notes. The duality of the two accelerations, in quarter notes and sixteenth
notes, is henceforth the diminution principle of the variation. In addition, the
complete octave, a summation of all the sixteenth notes in the first and second
quarters is of course also significant for the more long-range parallelisms; see
the slur in the graph.
The acceleration naturally creates space that cries out for repetition, which is
necessary in any case because of the theme.
By the shortening of the motives to sixteenth notes, the four sixteenth notes
of the second quarter naturally separate themselves from the preceding sixteenth
notes, which again establishes a special law of this variation, a rhythmic law.
If in this way the first and second quarters of the first bar of the variation have
already presented the whole first bar of the theme, then likewise the third and
fourth quarters bring the entire second bar, including the neighbor note, in its
proper place. Consequently, at the close of the first bar the variation has already
arrived at the point where the theme is at the end of the second bar. For this reason, a repetition of the last two quarters follows in bar of the variation (i.e., a
repetition of bar of the theme), which however still leaves the third and fourth
quarters free. In order to gain new content, the imagination, in other respects
wholly occupied with the adventurous acceleration of bar , throws itself toward
the neighbor note of even before this is reached. In this extension of the neighbor note g2 before the , f 2, lies the essential difference between this variation and
the theme and the first and third variations. Accordingly the voice-leading is developed in a new way: {}

However, the chief characteristic of the third variation is a considerable expansion of the neighbor note g2. This can be seen in the earliest voice-leading
levels.

In the final realization, the first neighbor note is placed right on the downbeat of the fourth bar, and precisely on this account the lower voice departs from
the course of the theme andfor the first timeapproaches V through IV,
whereby the presentation of a tenth above the bass [as in earlier variations] is
omitted. The melodic line is projected through different registers (in the right
and left hands), and is sometimes doubled.
Since, in contrast to the theme and the second variation, the unfolding of the
two thirds does not take place in bar , and since only the fourth-progression c2
f 2 can operate in place of the sixth-progression a1 f 2, a neighbor note, g2, is introduced in the third quarter of bar in order to fill the rhythmic space. Then
another neighbor note is given to g2 in bar : thus a crucial parallelism governs
g2 f 2 in the upbeat of bar and a
2 g2 in bar , suggested perhaps by G
and F of
the bass voice; if however b
2, still hovering above a
2 on the way towards a2 g2, is
finally taken out of play, and thus the third neighbor note [g2] is established, this
signifies an extreme intensification of the drive to the neighbor note, which governs this variation. This explains why no time is left for the of the Urlinie
finally provides a hasty conclusion in bar . {} The alternation of and on
the second and fourth quarters of bar is surprising: but this pattern provides a
smoother connection to the on the down beat of the sixth bar. In the course of
the lower voice, G
, the bass of the second chord (first quarter of bar ), is of
course to be understood as a neighbor note.
Variation . The third variation closes, in accordance with the requirements of its
diminution technique, with the repeated of the left hand, d1, in the seventh
eighth note of the last bar. In the same register, the right hand now begins the

tonw i l l e 8 9
melody again makes an ascent to f 2, by means of only a fourth-progression, as
in variation (not by a sixth-progression a1 f 2, as in the theme and in variations
). Therefore, in bar , again in order to fill temporal space, g
2 is placed before
f 2 as a neighbor to the neighbor note; filled thus, the upper voice can now keep
pace with the unfolding of the sixth AF in the lower voice.
This arrangement of the voice-leading now governs the final realization of
the diminutions. The f 2 appearing in the fourth quarter of bar is consequently
not yet the , but a passing tone within the fourth d2 g2; see Fig. c and the graph
of the Urlinie. With this fourth-progression, the lower voice moves basically the
same way as the upper voice, so that the consecutive octaves must be removed by
the interpolation of tenths; see the graph of the Urlinie. In bar , the neighbor note
g2 appears directly on the {} downbeat, and at this point is heard the imitation
of bar of the variation, i.e., the imitation of the basic motive in the rhythmic
diminutions of quarter notes and sixteenth notes. Notwithstanding the G chord
( VI), however, the imitation should immediately be perceived as a true parallelism to bar , corresponding to the parallelism between bars and in the
theme; this parallelism should by no means be sacrificed to the various abridgments of the motive, for it is only slightly veiled by the G chord; the returns
strongly in the third quarter of bar , marked by the B
chord, so that the variation in any event traces the course of the theme once more. In bar the descending fifth-progression from f 3 (with a stationary ) is repeated in the lower octave,
the original register of the theme, according to the inner law of the variation. In
the same bar the last sixteenth note of the second quarter [in the bass] occurs an
octave lower, in order to avoid consecutive octaves with the upper voice.
In bars the fourth-progression c2 f 2 is unfolded in a different way (see earlier), accomplishing an inversion of voices compared with bars of the theme
(and also variations and ): the fourth-progression b
e
is in the lower voice,
with the illusory effect of IIV, while the upper voice proceeds in thirds; here the
consecutive fifths are removed by means of . Nevertheless, the cycle of
chords still remains complete here, as VIIVV, and, as in the theme and in the
second variation, again basically signifies simply the V. Besides, the lower voice explicitly adheres to F1 at the beginning and the end of the progression.
Bar of the variation maintains the parallelism with bar of the theme, although here, unlike in the theme, d2 does not represent of the Urlinie, but has
above it the , which moves to and only in bar (see earlier). The two rhythmic diminutions in the first and second quarters of bar conform to those in bars
and . The movement of the figure beginning in the third quarter of bar is es-

In Fig. a, the neighbor note is still shown in its proper position: it follows the
, which appears right at the beginning; the lower voice likewise goes its customary way: B
DF. En route to the , a neighbor note is again inserted, above the
E
in the lower voice (a reference to IV; see variation ), which causes the Urlinie
to be delayed and only to reach the at the end.
In the voice-leading level shown in Fig. b, the (d2) appears at the beginning
according to the principle of unfolding (cf., Figs. bc). In this new arrangement, the neighbor note g2, which was understood earlier as a prolongation, now
of course appears temporally before the ; a leap leads from d2 to g2. The lower
voice proceeds from B
to E
(the neighbor to D
) by leaps of a third, so that the
midpoint G simulates a VI chord. However, the underlying thematic movement,
B
DF (see Fig. a), is not altered in the least by this change. From c2, two unfoldings of a third, c2 e
2 in bars and d2 f 2 in bar , lead upwards to e
2 as the
in bar . The two thirds also bear neighbor notes: f 2 in the fourth quarter of bar
and g2 in the first quarter of bar . (However, inasmuch as the , f 2, is still retained above these unfoldingssee the broken slur and compare Fig. aa secondary effect of two f 2 e
2 d2 third-progressions is achieved at the same time, of
which the first progression in bars is only a motion from the upper to the
inner voice, but the second in bars signifies the conclusive descent .)9
Finally, in Fig. c one sees the ascent to and the insertion of the supporting

D 3 chord which tonicizes the root G; from this note the upper voice derives the
fourth-progression that now binds d2 and g2 together melodically. In bars the
9
[S] See Tonwille , p. /I, p. . [This reference, to Schenkers analysis of Haydns Sonata in E
,
discusses bars of the first movement, in which the first descent from f 2to b
1 (bars ) is
only a motion to an inner voice, whereas the second and more fundamental descent
completes the first part of the second subject.]

Brahms Variations and Fugue on a Theme by Handel, Op.

In Fig. b, then, can be seen the neighbor note e
2, which is introduced above
the
II (Phrygian) chord, although the resulting
is immediately corrected diatonically to  (c2) at the V chord.
In the final realization, the diminution clearly replicates the ascent to , but
necessarily accelerated because of the octave unfolding, as in the fourth variation.
Another imitation of the ascent is inserted between d
2 and e
2 (neighbor note) in
bar of the right hand, in the form of a leap of a third b
1 d
2, which represents
the rhythmic diminution in the theme. In the lower octave, the left hand actually
leads the way with the same motion of a third.11 The previous inverted figure,
d
2 b
1 (via c2) is closely related to that leap of a third which, arising from the
fourth variation, had initiated the fifth; see above. The resultant figure is repeated
in the first and second quarters of the [following] bar. Only now is the expanded
neighbor note introduced, in the form of the third-progression e
2 c1, in contrast
to the theme; see the graph of the Urlinie. Then in the fourth quarter of bar it
joins in the ascent to , further accelerated to sixteenth notes since the octave unfolding of the parallelism must be squeezed into a quarter-note beat. In accordance with the special law of the variation, this repetition appears also in bar ,
in the third and fourth quarter-note beats.
In bars and the motivic parallelism to bar requires that, in spite of the Ursatz (Fig. a), the elaboration [Auskomponierung] should at first follow the minor
V, which has the purely diatonic third (see Harmonielehre, pp. ff/pp. ff ),12 thus
a
1 and a
2 in bar , after which the first instance of the leading note a1 follows in
the last eighth of bar . Because of the sixth-progression a
2 f 3 introducing the
in the third quarter of bar this variation again relates more closely to the
theme and the second variation, where the sixth-progression, although in different circumstances, nevertheless goes from a1 to f 2.
The appears in bar . The special law of the variation now requires a repetition in bar as well; but both the and the cadence that accompanies it, in contrast to bars and , here require the repetition to be arranged with different har
monies, II and V 3. By this circumstance the upper voice is pushed upwards from
d
2 by way of e
2 here with the effect of a neighbor note to the to f 2; then the
correction of
to  , as well as the descent (see earlier), occurs beneath the f 2.
Precisely this arrangement of outer parts, f 2 above b
1, bridges across to the next
variation.

sentially continuous until the third quarter of bar (to ), but according to the
inner structural law of the variation it, too, is divided into two parts by the operation of the repetition; see the graph of the Urlinie. After all, in bars and the
first part deviates a little from the structure that has by this time been established,
which is adhered to exactly only in the second part. This, too, contributes to the
apparent effect of repetitions; at the same time, however, the repetition signifies
a sort of ritardando of the whole progression. The two forte strokes at the end of
the variation, which are meant to emphasize the leap of a third from the neighbor note (f 2) to d2 as the , are replications of the two forte beats at the conclusion of the first part.10
Variation . The last beats of the fourth variation call forth the fifth variation
there is merely a change to minor and to pianoalso, the arpeggiation of the octave d
d
2 from the first to the third quarters of bar of the fifth variation is related to the octave span in bar of the fourth variation. What beautiful evidence,
once more, of the zeal of an inspired creative power! The Ursatz of the new variation is shown here.

{} Although the appears again as a tenth here (see Fig. a), it no longer belongs to the tonic chord, as in the theme or in the fourth variation, butand we
encounter this change for the first timeto III, here a major chord. At any rate,
this chord signifies the midpoint on the way to V more than a true III (with its
own fifth-relationships); this is confirmed moreover by the fact that the tonic
chord softly steals in just before the entrance of the V. In this sense B
D
F in
the bass motion of the variation, notwithstanding the very clearly expressed III
here, still bears a resemblance to the bass motion of the theme B
DF, again signifying merely IV. The voice-leading for at level a) is almost completely
identical to that in the theme.

11

Terzzug: but Schenker means the leap in the last two sixteenth notes of the third quarter.
The point is repeated in Der freie Satz, .

10

12

Those at the end of the first part are not specifically marked forte.

tonw i l l e 8 9
Variation . This variation is also in B
minor. Its Ursatz

left and b
2 in the right; see the graph) the upper voice finally attains b
3 and,
most importantly, does so in accordance with the mysterious arpeggiation b
2
b
3 in the last eighth of bar .
Variation . Here are the voice-leading levels of the seventh variation, which is
again in major:

reveals complete identity with that of the fifth variation; see Fig. a.
{} Nevertheless, the final legacy of the fifth variation, f 2 over b
1, leads the
diminution to completely different paths here. Since b
1 is again possible at the
beginning of the new variation, because of the conclusion of the fifth variation,
the motive of the ascent to can begin in the first quarter as well, again as in the
theme and variations . In contrast, then, to the fifth variation the first unfolding of the octave can fill out the space b
1 b
2 here, with b
1 c2 d
2 being
taken as the Urlinie.
The rhythmic diminutions appear within the rise and fall of the arpeggiations
again, but the neighbor note of the theme has no place. This is accounted for by
the use of the technique of strict canon (imitation at the distance of a quarter; in
bars the upper voice begins) for the presentation of the diminution here,
and, besides this strict obligation, all the octave progressions are controlled by a
second bond, the hidden arpeggiation b
2 d
3 f 3 b
3; see the broken slurs in the
graph. Precisely this second requirement, which we encounter for the first time
in this variation, presupposes above all the arpeggiation notes, and on this account the neighbor note is now omitted.
Like the fifth variation, the sixth also progresses to the tenth in bar (at the
) and restates the B
chord immediately before the V in bar .
In bars the diminution introduces the in an extremely artful manner.
The canon proceeds in contrary motionthe lower voice leadsand into it is
woven the unfolding of a third, as in the theme and the first and second variations; see the graph.
Bar brings in the . The canon, however, with the lower voice still in front
as in bars , returns to direct motion and thereby the leaps of an octave again
occur also, as in bars . From the height attained in the third and fourth quarters of bar , there is stepwise movement down an octave; since, however, in each
case an ascending register-transfer takes place after the fourth eighth note of the
fifth-progression (thus after F in the left hand and f 2 in the right, after B
in the

{} The Ursatz, Fig. a, is similar to that of the fifth and sixth variations (see
Figs. and ), in so far as the , f 2, occurring as a tenth, is part of the III chord
that serves to mark the midpoint along the path from I to V; on the other hand,
in the manner of its descent , the seventh variation resembles the fourth
(see Fig. a).
In the prolongation shown in Fig. b, however, the unfoldings and
arefor the first timemissing: the and are given only chordally. In order to
emphasize the III chord, the chromatic passing tones c  and e come into play, in
place of the diatonic c and e
.13 As compensation for the ascent to , the upper
voice subsequently expands beyond f 2 to a2 introducing this form of the
diminution, too, for the first time.
13
The c  is an implied note; it appears in three of Schenkers voice-leading levels (Figs. bd), but
not in Brahmss score.

Brahms Variations and Fugue on a Theme by Handel, Op.

eighth of each of bars and and thus fortunately avoids the threatened simultaneity of e
1 and d2 in bar . {}

But that is not all: the is introduced an octave lower (see Fig. c), in the
inner voice. Therefore the fifth of the chord must appear as a filler note in the
upper voice. The continuation in direct motion then necessitates the exchange.
Fig. d shows the downward step of a fourth voice an octave above the lower
voice. This upper voice has the same course as the lower voice: the sixths by which
the consecutive fifths are avoided serve also the avoidance of consecutive octaves.
The diminution in Fig. e shows a previously illustrated position: the and
acquire their neighbor notes, as in the theme. By this circumstance the vertical
position of the upper voices is presented horizontally (through leaps); see the

d f
fourths bf a c in Fig. d. In the use of filler notes as a cover, the connection with
the preceding variation becomes clear: there b
2 d
3 f 3 in bars , here b
1 d2
f 2, and so on.
By way of leaps, the highest voice reaches the high f 2 precisely when the inner
voice reaches a1: at the point where the true Urlinie goes above the , the highest
filler voice confirms the with its f 2, by which means the overstepping of the
inner voice is rendered ineffectual at the same time. When the upper voice immediately falls back to f 1, it does so only in order to create space for the further
rising leaps in bars , which extend to e
3.
The necessity to connect the unfolding of a fourth in bars with the added
leaps above it presents a great difficulty for the composer, especially in bar : e
1
should and must go to its neighbor note, f 1, b
1 should and must go to f 2; there a
step of a second, here a leap of a fifth that can only be bisected. (Indeed only b
1
d2 f 2 is admissable, though b
1 e
2 f 2 or b
1 c2 f 2 would have better fitted the
underlying F harmony.)
The final realization of the variation adds only a little to the voice-leading in
Fig. e, and yet makes changes in all the difficult situations. It replaces the missing ascending series b
1 c2 d2 and d2 e
2 f 2 with neighbor-note motions (d1 e
1
d1 and f 2 g2 f 2: see the graph); however, these neighbor notes, which appear in
the second quarter of each bar, must not be confused with the neighbor note in
the fourth quarter of bar in the theme and in this variation. The final realization overcomes the difficulties of bars in the following way: it refers back to
the unfolding of a third in the themesee Figs. fgand allows a1 b
1 in the
third and fourth quarters of bar to follow from c1 d1 in the first and second
quarters (thus, in inversion, sixths instead of the original thirds); by maintaining
the flowing rhythm of bar , it uses a1 and d2 of course in the fifth [recte: sixth]
1

Variation . The rhythm of the previous variation, and also the arpeggiations
piled up in bars , want to assert themselves further; they strive toward the
continuation of their species in a new variation: the former is continued by the
drumming root tones, the latter in the arpeggio motive of sixteenth notes, which
counterpoints against the Urlinie motive:

Fig. shows how the crowding of arpeggiations demands the ascending register-transfer; for it is obvious that g2 would go better with e
2 in the fourth quarter of bar than the unison e
2.
The voice-leading transformations are as follows:

tonw i l l e 8 9
Fig. b). One can clearly see from this example how the survival instinct of a phenomenon that has already blossomed into life, as well as new difficulties in the
voice-leading, prove to be forces that produce a new configuration. How blessed
appears the imagination of the artist here: it is open to all stirrings of musical life
and thus resembles creative nature, for which the force of reproduction and the
necessity of new circumstances are preconditions for creation.
And, as the counterpoint of sixteenth notes to be adapted to all the unfoldings shows in bars : g
1 already appears as the first sixteenth note in the third
quarter, and likewise b
1 in the third quarter of bar , in order to lead smoothly
to the following chord (see the fifth-progressions, b
1 g
1 e
1 in bar and d
2
b
1 g
1 in bar , indicated by slurs in Fig. e).
But now also to maintain in these very delicate voice-leadings the special
rhythm of the variation: what a sense of consistency, what sure-handed mastery
is required! If in bar , as in bar , things begin on the second eighth note, the first
eighth note must somehow be filled; but since this can only be governed by the
last harmony in bar , the master shifts the imitation of bar in bar in such
away that it is completed on the first eighth note of bar . Therefore e
2 f 2 in bar
takes a different place than c2 d
2 in bar : for the sake of a higher plan, whose
defining characteristic is the eighth rest on the downbeat of bar , an irregularity
as perceptible as this is introduced. {}
The repetition of bars brings about the inversion of the voices and alterations in the dynamic markings.

The Ursatz (Fig. a) matches that of the theme, from the retention of the root
of I to the in bars .
The first prolongation (Fig. b) betrays the intention to gain in bars by
the inversion of the two sixths that serve the in bars , i.e., to turn the sixths
into two thirds. The of bar bears a neighbor note that occurs on the downbeat
of bar ; it is followed quickly by .
The prolongation shown in Fig. c shows the diminution in progress: a uniform series of sixths in bars , which yields, through inversion in bars , a
likewise uniform series {} of thirds. Chains of sixths and thirds now express
clearlyin each of the first three quarters of bars and the ascents to and
; an ascending register transfer leads from d2 to c3 (instead of c2) and likewise
from f 2 to e
3 (instead of e
2). The organic quality of the inversion increases the
emphasis on and . Then, in bars and , the fourth-progression c2 f 2, accompanying a sixth-progression of the lower voice, gives the . The similarity of this
voice-leading to that of the fourth variation, Fig. c, is thus clarified, except that,
instead of the neighbor note g
2, a descent f 2 e
2 is used (from the neighbor note
back to the ). There now follows a renewed ascent to which, in the manner of
the first ascent from the beginning, is founded on an inversion of its intervals.
Fig. d shows the next voice-leading level for bars . In bar a substitution
appears in the sixth-progression of the lower voice: e
1 d
1 for ab
, which shifts
to the inner voice; for the sake of parallelism, g
1 f 1 replaces e
1 f 1 in bar as well.
A further development of the diminution in bars is shown in Fig. e: vertical thirds are replaced by horizontal leaps of a third (upward and downward).
What is more, the graph introduces inversions in bars , like those in bars
(cf. Fig. c).
All that remains for the final realization is to iron out all the unevenness coming to light through the diminution, and to provide for the synthesis. One sees in
Fig. , where the rhythm of the bass follows that of the counterpoint in each first
and third quarter, what a lameness this rhythm must cause: in order to remove
this defect and stagnation, the composer completes the arpeggiation in three sixteenth notes and adds a note repetition in each fourth sixteenth note in the third
quarter of bar and in the first quarter of bar . On the other hand, the master
removed the danger of emptiness that the octaves would have produced in the
first and third quarters of bar (or, in the inversion, the unison in the first and
third quarters of bar ) through the displacement of b
1 and d1 to the position of
the second, whereby sixths arise; and in precisely this way he underscores the
sixth as the one interval that is fundamental for the voice-leading here (see

Variation . The eighth variation bequeaths to the ninth the long, sustained notes
of the bass, as well as the falling sixth-progression of the upper voice in the last
two quarters of bar (the inversion could not be shown in the graph of the Urlinie.) The fundamental notes, still subdivided into drumming reiterations of the
bass in the eighth variation, are reduced to held notes in the ninth variation, and
the broad reverberations of single solemn strokes appear in the place of a forever
surging presence. In this transformation, the new variation is made manifest.
b
d f b

Four sound-pillars, four octaves highB
D F B
form the tonal framework
into which the diminution is built.
The voice that guides the tones of the Urlinie is in the left hand, an octave
lower than in the theme; the counterpointing sixth-progression is in the right
hand. The theme is first altered in the third quarter of bar , where c1 returns to
b
. This is possible here since the counterpointing sixth-progression finishes pre2

Brahms Variations and Fugue on a Theme by Handel, Op.

likewise in bar : during these new triplets {} the bass returns through arpeggiation to the original note, which becomes all the more necessary as bars
soar ever upward.
The ninth variation represents an example of the outer limits of keyboard texture. It dares to be imagined (the four sound-pillars), but at once confirms itself
by the most honest means to be a true piano piece that lies well under the fingers.

cisely with d2 in the third quarter and is thus suited to continue with d2 e2 in the
original register; see the graph of the Urlinie.
Also new is the raising of the third in the III chord, where the is reached. It
at once stimulates the imagination of the artist to maintain the Urlinie motive exactly in the major form of the III chord as well, which is a minor chord in the
major system. In the last quarter of bar , f 2 is of course changed back to f 2 on
account of the V chord, so that the achieves its diatonic fulfilment as well.
The imagination remains consistent also in bars : if it had allowed the
motive initially to triumph over the diatonic system in bars , so it allows the
first form to rule in bars , as if the progress of the Urlinie tones did not concern the diatonic system at all. In spite of this, the is introduced in the third
quarter of bar .
Evidently on account of the power of the non-diatonic f  in bars , the repetition of bars takes up the Urlinie-motive in the chord of F ; it is understood, however, that this apparent eavesdropping F  is in reality a G
(
VI),
since the B
(I) follows.14 (In the graph of the Urlinie, the repeat [of bars ] had
to be omitted.)
Linear progressions that are related only to the requirements of the sustained notes are also woven into the diminution. Thus triplets are formed by the
counterpoint of the right hand in two places, by which one can clearly make out
the individual notes belonging to the sixth-progression, which touch only the
chain of notes [Tonstrang] b
2 d3 f 3 b
3, stirring up new waves, so to speak.15
Likewise, and only in order to carry forward the waves, the roots are lightly restruck in the left hand. This light touching of what are essentially stationary
notes proceeds, to be sure, with a determined regularity (in the low and high
registers), though it is certainly not derived from the note repetitions of the bass
in the eighth variation; it is a regularity of its own kind, born from the circumstances of the new variation.
The bass G in the fourth quarter of bar is to be understood as the divider at
the lower fifth of the next root, III, above which e2 leads on already as a passing
tone: G(5 )6. The falling third B
G is answered by the rising third DF of the bass
in bar . (Apart from the violation of the parallelism, e2 as the fifth of the upperfifth divider A would have produced an unnecessarily harsh effect.) After the
triplet of the second quarter in bar , another follows in the third quarter, and

Variation . The new form of the motive, as the ninth variation has produced
itthe motive returns to its starting noteis also determinative for the tenth
variation. In relation to the b
2 in the first quarter of bar , this variation is set an
octave higher than the theme. Nevertheless, the flight of the bass toward the low
register in bars , which is characteristic of the previous variation, is still felt
here and expresses itself in a change of register from quarter to quarter, always in
the downward direction. (This can be only partially indicated in the graph of the
Urlinie.)
The b
2 ( ) is approached from a sixteenth-note figure that stems from the
eighth-note ascent in the last bar of the previous variation. It is also this figure
that then leads to the high d3 and c3 in bars and . Traits of modal mixture are
encountered in bars and in the fourth quarters, and in bars and in the first
quarters. The voice-leading in bars is illustrated by the following levels:

The lower voice, with its neighbor note (see Fig. b), reflects that in bars
and of Variation , and yet the same neighbor note has a different function here:
it stands in the service of another horizontal unfolding (b
cd
). The notes of
the Urlinie are covered by higher filler voices.
Variation . The eleventh variation is linked with the tenth, as that is with the
ninth, by the descent of the bass.
The voice-leading levels of the new variation are as follows:

The enharmonic link A  B
provides the direct connection between the two chords.

14
15

That is, the last note of each descending sixth becomes the next sustained note.

tonw i l l e 8 9
its fifth, the same play of neighbor notes occurs, as does the opposition between
b
a of the lower voice at the end of bar and b1 a1 in the inner voice.
The paths of the diminution in bars are predetermined by the requirement that b
2 should be attained in bar , for the sake of the parallelism to bar .
Clearly, f 2 is placed in the first quarter of bar as a reference to the of the first
section. In the second and third quarters the inner-voice c2 b
1 acts as an imitation of d2 (c 2 )c  2 of bar ; the transfer of the sixth, b
1, to a higher register follows in the fourth quarter, which introduces the exchange (to avoid consecutive fifths). According to this, the third-progression f 3 e
3 d3 should then have
appeared, and the been introduced in the second and third quarters of bar , in
imitation of the earlier third-progression motive: however the cue-note of the b
2
[recte: b
1] in the first quarter of this bar leads away from here and back to the
course of bar , while the is surrendered to an inner voice. What an impulse to
the synthesis! From the third quarter of bar the lower voice is filled out, accompanying the last two neighbor notes with passing tones in whole tones and
semitones (in contrast to the lower voice in bars )a prelude to the next variation. The b
1 of the inner voice in the second quarter of bar no longer leads upward: the impetus for repetition is extinguished.

Fig. a shows again the usual motion of the upper and lower voices in tenths
in bars , whereby the occurs at the minor III chord. In bar we encounter
the inversion of the dominant chord for the first time; it suggests an exchange
of voices,16 {} which introduces the sequence e
2 d2 in bar , as in the
upper voice. Two neighbor notes then tremble in the wake of the in bars and .
The elaboration shown in Fig. b proceeds in the manner of a reaching over:
thus the upper fifths of I and III arise. In bars the voice exchange, shown in
Fig. a, is worked out through exchanges.
The diminution structure of the final realization is colorfully animated. In
bar an eighth-note figure arising from b
2 is added above the grace-note b
1, the
beginning of the actual ascent b
1 c2 d2; it runs directly through the neighbornote harmony in the second quarter, advanced by the lower voice, whichfor the
first timeoriginates from the inversion of the tonic chord, and first reaches
the root in the third quarter. By this means a slight undulation arises, which is
also passed on to the following variation. But within this the repetitions, which
continually burst forth from the inner voice, can be clearly discerned; see the
graph of the Urlinie. These recollect the second variation (see Fig. cd), while
the arrangement of the third-progression e
2 d2 c2 from the second to the fourth
quarter of bar calls to mind Variation , where the expanded neighbor note follows the shorter neighbor note in the same way.
The lower voice arpeggiates the B
chord through the fifth f in the first quarter of bar , but without the fifth gaining the status of a divider. In the fourth
quarter of bar , g precedes this fifth f; and likewise the root B
at the third quartet of bar is preceded by e
: the two notes g and e
have neighbor-note functions,
like the upper voice e
2. The succession g
f 1 in the inner voice in the second and
third quarters of bar enters in opposition to the gf of the lower voice across
bars |. Likewise across bars |, at the similar division of the D chord through

Variation . The charmingly disguised third-progression f 2 e
2 d2 in the last

two bars of the previous variation, as well as the repeated reminder of the neighbor note e
2, still continue in the twelfth variation: in the g2 f 2 neighbor-note
play, and in the falling figure of bar . In order to remove the significance of
from the primary tone of the figure, f 2, an inner voice enters and gives the ascent
to . Under the chime of delicate neighbor notes, the motive thus enters in the
inner voice, but then skips up to the upper voice in order to continue here with
the course of {} the neighbor note and the returning d2 c2 b
1, see variations
and . The same pattern is repeated in bars , , and .
The parallelism between f 2 and b
2 at the high points of bars and is significant, as is that of the same notes in bars and ; see the graph of the Urlinie.
The motivic gesture in bar pushes upward above the , as in bar of the second variation. It serves the same function as an ascending transfer of : there
through a leap to c3, here through a leap to c2, which is indeed approached by the
upper voice in accordance with the unique law of this variation (it comes from
b
2). In all other respects the conduct of the variation is like that of the theme.

[S] See Elucidations, Fig. . [Schenker is thinking of the third part of Fig. b.]

16

Brahms Variations and Fugue on a Theme by Handel, Op.

In bars the extrapolated roots give the misleading impression that the
counterpoint, including that of the passing D
chord, is initially a movement in
parallel thirds, as shown in Figs. c and c. From the concealed octaves on the
downbeat of bar , the outer-voice counterpoint in bar proceeds in tenths; {}


why the tenth eC is not likewise introduced in bar has been explained earlier.
The diminution is replete with ornaments of Hungarian-gypsy origin, as in the
fourth quarter of bar , in the first quarter of bar , in the last eighth of bars
and , etc.

Variation . The Ursatz of the variation

shows the same course as the fifth and sixth variations in bars (see Fig. ). In
bars , in contrast, we see the V chord return by way of III to I in the manner
of a fifth-progression. However, since in bars the is retained longer here
than in the theme (see also the third, fourth and tenth variations), the danger of
consecutive octaves looms in bar , where the fifth-progression of the Urlinie
tones should lead toward its conclusion, if at the same time the bass also continues its path with D
CB
, in order to complete its own fifth-progression. Therefore, the bass suddenly takes the opposite direction and meets the (c1) with the
root of V at the interval of the fifth.
The diminution of the realization still stands entirely within the influence of
the motivic form newly established in the ninth variation (ascent to and return). Even the return d
1 c1 b
appears in the foreground here, formed by two
large third-progressions (see the graph).
An arpeggiation of a third leads to the , here d
1 the variation lies an octave lower than the themewhich one can speak of simply as a bare chordal skip
to , since the passing c1 is omitted. This manner of formation of the is sufficient for the descent to appear to be only a returning motion, despite the breadth
of its realization. One cannot be deceived about this, even by the ascent to the ,
whichstarting already in the third quarter of bar brings not the expected
d
1 e
1 f 1, but links over to the final note of the return, b
c1 d
1, and thereby
makes possible the chordal setting of the as well.
The neighbor note of , e
1, is only hinted at tenderly, in the ornament of the
fourth quarter in bar ; it does not participate in any way in the voice-leading,
even though the ornamentation of c1 e
1 apparently strives after a parallelism to
the arpeggiation of a third, b
d
1, in the first and second quarters. In bar , the
organization is at first the same as in bar ; on the other hand, the fourth quarter
clearly brings the neighbor note to , g
1, powerfully supported by the divider at
the lower fifth.

Variation . The two falling fifth-progressions in bars and of the thirteenth

variation, g
1 c1 and f 1 b
, suggest the falling diminution of the new variation,
and perhaps the sharp contrast of largamente there and sciolto here also stems
from the Hungarian influence of the previous variation.
The falling sixth-progression of the diminution in bar recollects that in bar
of the eleventh variation, but is not as obvious: does it replace the ascent b
1 c2
d2 (see the graph) or is it only a chordal placement of the , here of course ex

pressed in inversion: bd ( )? In any case, the diminution and the neighbor note do
appear.
Bar in the theme a broad elaboration of the neighbor notedenotes only
a bare repetition here: in this respect the fourteenth variation is like the seventh.
That the occurs with the III chord establishes a further connection with the seventh variation.
In bar the master substitutes the lower-fifth divider for the upper. This is
connected with the necessity to avoid a doubling of the leading note, c . This c 
2

replaces c as the proper seventh note of a minor scale in the elaboration of the
minor III chord, thus signifying a greater divergence from the basic diatonic content, which is even further emphasized in the sixteenth-note run by the steps of
an augmented second. Now the bass must still adhere to the divider at the lower
fifth in bar , all the more because the run in the upper voice begins only from f 2
and is more compatible with the lower-fifth divider in the second quarter with c1
than with c 1.
The voice-leading in bars proceeds as in the theme and in the first and
second (and fifth) variations. In bar the inner voice is especially striking; as it
enlists the motive of the diminution in the preparation of bar (see the graph).

tonw i l l e 8 9
Variation . This variation, too, resembles the previous one with its characteristic sixth, b
2 d2, (see the first and second eighths in bar ), and also in the chordal
setting of the and , which is even more pronounced here; in addition, the sixteenth-note figure refers to that in bars of the variation .17
Here, the ascent to is replaced by a neighbor-note figure, as in variation
(see the second and third eighths in bar ), and the rhythmic diminution and
neighbor note follow, just as in that variation; see the graph.
In the service of is a twice stated descending third-progression, fe
d,
first in the one-line octave (bars ), then in the two-line octave which at the
same time expresses the Urlinie. In these third-progressions, there is no shortage
of neighbor notes; see in bar , last eighth, f 1 as neighbor to e
1, and in bar , first
eighth, g2 as neighbor to the , and third quarter, f 3 (register transfer of f 2) as
neighbor to the . Moreover, the first third-progression lays claim to a hitherto
unfamiliar expansion, explained thus: The arpeggiation b
2 d3 f 3 b
3 is spread
over the whole variation (see the graph); in addition, the first space of a third is
filled with c3 (see the last eighth-note in bar ) and the fourth f 3 b
3 is filled with
all the chromatic passing-notes; although b
3 is reached only in the following {}
variationa special bond of the two variations through an arpeggiationthe
full passage through the interval of a fourth is nevertheless the cause of that extension. Also connected with this is that, in bars , G
appears before F in the
lower voice.
Finally, observe the syncopated effect of the Urlinie notes in bars (see
the graph): f 2 ( ) in the fourth quarter of bar is joined to the neighbor note g2
in the first quarter of bar , and e
3 ( ) in the second quarter of bar is joined
to the neighbor note f 3 in the third quarter; consequently the and , when considered without the two neighbor notes, each has the value of a half note, in the
form of I .

Already the Ursatz (Fig. a) shows for the first time the inversion of V with
the . If in the earlier variations the bass went upward from B
through D to F,
regardless of whether D figured in a I6 or a III chord, it now falls from B
through
A to F. Bars arefor the first timefilled out by the alone, so that the concluding falls directly on the first bar of variation .
The first unfoldings are at work in the voice-leading shown in Fig. b: first
the downward arpeggiation of a third in bars , then in bar the upward


arpeggiation of a fourth. In bars the single third, ec , is twice unfolded, downward in bars and upward in bar .
Fig. c shows a filling-in of these unfoldings, with a neighbor note (bar )
and two passing tones (bar ). Moreover, both unfoldings are linked by a chord

built upon II 3, which confirms the step of a second in the lower voice (B
A) as
a necessity. This very step of a second gains in motivic significance; see G
F in
the lower voice in bars (is this an echo from bars of the previous variation, or a substitute for the neighbor note omitted in bar ?) and a
g1 in the
inner voice in bars .
The full realization (see the graph of the Urlinie) shows imitations in the service of all these unfoldings. Again, however, as also in the earlier variation, {} a
multitude of filler notes within the interval of a fourth f 3 b
3 float over the real
upper voice: twice b
3 leaps back down to f 3 in bars and , while in contrast f 3 (bar ) leaps back up to b
3 (locally in bar , but more properly in bar of
the next variation). If the first leap in bar goes to a high register, this links up
with the upward direction of the transition through the interval of a fourth in
bars of the fifteenth variation. Only from bar on do the leaps turn downward; the first downward leap, f 3 f 2, already accentuates the , even before the
unfolding of a fourth arrives at f 2 (f 1 in the realization). In bar g
3 sounds as well
2

Variation . The new variation is linked with the previous one in the following
ways: b
3, which begins the sixteenth variation and at the same time completes the
fifteenth; the leap in the last bar of the fifteenth variation, which leads directly to
the new one; and finally the sixteenth-note figure.
On the other hand, the course of the voice-leading is entirely different:
17
In the autograph score currently housed in the Library of Congress, Washington, DC (a manuscript apparently unknown to Schenker), variations and appear in reverse order.

Brahms Variations and Fugue on a Theme by Handel, Op.

cessitates the same rhythmic setting in bars . In bars an ascending register

transfer (e
2 for e
1) takes place through a reaching-over, which clearly resembles
a stretto.
In this variation, too, a filler voice also appears in leaps, above the real path
(see the graph); these leaps give expression to the falling fourth b
3f 3 in bars
and the ascending fourth f 3b
3 in bars . Indeed, in bars the leaps actually divide {} into a polyphonic structure, whose voices partly strengthen the
given voices, and partly also follow new obligatory paths: g
2f 2, c
3b
2 and f 3e
3.

(neighbor note to f 3 in bar ): it gives a unison in the motion of the neighbor

notes in the outer voices. Considered in relation to each other (see the dotted
slurs in Fig. c), the highest filler voice and the upper voice in bars give the
appearance of
|| in the form of downward unfoldings: g
2 e
2, f 2 d2,
e
2 c2.
In bar , c2 follows the neighbor note e
2 as early as the second quarter; this is
a rhythmic shift which, insofar as c2 should be apportioned the value of a half
note (see in Fig. c: e
2 in bar and d2 in bar ), has as a consequence an acceleration of the notes d2 e
2.

Variation . The special features of variation , namely the quarter rest before
the motivic elements and the chaining together of the leaps, work again in distinctive new ways in variation : an eighth rest arises from the quarter rest, and
from the leaps comes a continuous sixteenth-note figure, which at the same time
takes part in the syncopation of the main voice.
The motivic segments are developed alternately by the two hands, in the left
an octave lower than the theme, in the right at the original register.
Bars show only slight differences in comparison with the theme; to be
sure, the most striking is the articulation of the motivic segments: the first two
notes are always marked portamento, the third and fourth notes legato. Its charm
lies in its artful contradiction of the original meaning.
In bars the voice-leading exhibits new characteristics:

Variation . The of the previous variation falls on the first quarter of the new
variation: a stronger linkage is scarcely imaginable. In addition, the upward unfolding c2 d2 e
2 in the last bar of the earlier variation becomes the seed of the
new one. And, finally, the two variations are additionally bound by their leaps: in
variation this always occurred once per bar (from the second to the third
eighth), except twice in bars and , and the doubling in the last bar plainly promotes a compression in variation .
The motive of the ascent in the left hand, an octave lower than in the theme,
brings c1 again in the second quarterhere imitating bar of the sixteenth variationbut smoothes the eighths to quarter notes. The first note, b
1, is absent,
partly since the conclusion of the previous variation makes use of the first quarter, and partly since the imitation of the last bars requires it to be so: yet through
this the motive gains clarity in comparison to the realizations in variations .
It is striking that the appears as early as the second bar and that the neighbor
note g1 claims all of bar for itself.
The following figure shows the voice-leading of bars :

Here the ascent c1 e
1 (e
2) proceeds by way of the semitones d
d, which the
lower voice accompanies with a division of a falling fifth-progression FD
B

(VI). In the realization this progression takes up bars , so that the does not
appear until the last bar and follows it in rapid quarter notes.
The shifting of individual motivic elements in bars by a quarter note ne-

Fig. a, which gives the fourth-progression c2 f 2, is indeed similar to that of

variation : one thinks particularly of the neighbor note G
in the lower voice. But
Fig. b already shows another unfolding: the inner voice a1 rises to the neighbor
note b
1 and falls from here to f 1, including that g
1 as well. From here to Fig. c,

tonw i l l e 8 9
Variation . The ascending fourth-progression in bars of variation may
have provided the initial stimulus for variation .
The repetitions in bars and recall the second variation; on the other
hand, the ascent to , which occurs as early as the third quarter of bar , reminds
one of variation . Most ingenious is the progression of the lower voice. The
fourth-progression B
F, along with the new elaboration from B
in the third
quarter of bar and the dwelling on A
in inversion in bar , hinders insight into
the true facts of the case, as the graph of the Urlinie shows them; but if one puts
the true bass notes in bars and , one obtains with the greatest surprise B
D

F, the most usual progression that stems from the theme.
The voice-leading in bars is even bolder. The lower voice falls from the
dominant through D as the halfway point (III) back to I in bar , while the upper
voice traverses the progression c1 c 1 d1 during which is retained (see the
graph). The does not appear until the fourth quarter of bar , and follows hastily in bar . One also gathers from the graph why, in the final realization,
the upper voice cannot bring in the in the first quarter of bar : the lower voice,
still proceeding from the A (second eighth-note of bar ) in chromatic steps, goes
to D precisely at this point.

where the lower voice assumes the path of the inner voice, only one step remains.
For the diminution of the fourth-progression, a motive related to that in bars
of the previous variation is used. The , however, remains in control in the first two
bars, and even in bar , where it is lead up to b
2. Against this fourth-progression
in the main voice, the upper voice (see the graph of the Urlinie: Fig. c shows an
inversion of the two voices)presents the summative third-progression f 2 e2
e
2 d2, which connects up with f 2 e
2 and f 2 e  2 in bars . The progression
V
I here is also new.
Not until the final bar do and follow. A voice exchange in the third and
fourth quarters brings b
2 into the upper voice. The arpeggiation that soars
above the whole variation also finishes on this note: b
2 d3 f 3 b
2 (for b
3) in bars
.
Variation . The idea of motivic segments alternating between registers is also a
basis of variation : The ascent to in bar appears in the lower voice in the left
hand, but its repetition occurs an octave higher, in the inner voice in the right
{} hand, etc.; see the graph. The rhythm is that of a siciliano; the leading inner
voice is ornamented with an inverted mordent [Pralltriller].
As in variation , an arpeggiation spans bars : f 1 a1 c2 f 2; furthermore,
the f 1 g1 a1 expansion above the is common to both variations, as is the bass
motion B
DF. On the other hand, the neighbor note is, for the first time, assigned to the voice that carries out the arpeggiation: f 1 g1 and a1 b
1 in the third
and fourth quarters of bars and .
The repetition of bars occurs in a higher octave, and the inverted mordent appears in the highest voice.
Bars retain the fourth-progression c1 f 1, which introduces the as well.
The upper voice accompanies it with the third-progression f 1 g1 a1, evidently
imitating the third-progression in bar , while the lower voice adds the neighbor
note g. In this way a succession of chords arises which recalls the older manner
of the a cappella style, a succession in which every chord is in root position and
not even a inversion is admissable.
Bar repeats the arrangement of bar and contains the together with its
neighbor note in the highest voice (f 2 g2). The impetus of this neighbor note, and
also the third-progression f 1 a1 in bars , now drives the upper voice up to b
2,
while the inner voice takes .
These bars are repeated in the same way as bars .

Variation . This variation, completely separated from the earlier one, begins a
new group of variations founded on a more or less stationary bass. Further, it is
the only one to give up B
as the tonic; it is in G minor. But to be sure, the leading voice of the Urlinie behaves here as before, as if it remained in B
. If one considered this voice alone, one would not for a moment fancy that the tonality of
the theme had been abandoned. (In the graph I have accordingly deemed it appropriate to indicate the Urlinie tones in relation to B
major.) I assume that {}
Brahms used as the model of this variation the sixth variation in Beethovens
Eroica Variations, Op. , where the true tonality of the upper voice (E
major)
is given the lie in a like manner through an artificial harmonic progression (C
minor). One might think that this artifice is out of context here; and yet, one can
understandably be lured into reflecting a familiar construction in two keys simultaneously, as it were. The point of the matter therefore is not that G minor
should be denied altogether; it is, and remains, the key of this variation. The
artists task was only to adapt the pure B
major note-series to this tonality for the
sake of a delightfully concealed contradiction.
In other respects, the course of the variation is simple (see the graph), apart

Brahms Variations and Fugue on a Theme by Handel, Op.

graph). In bar the neighbor note e
1 appears, now likewise a wave among waves,
but in bar the neighbor note g
1 is replaced by a higher melodic arch.
{} Bars astonish through their extraordinary art of voice-leading, in
spite of its similarity with that in variations , , and . The fourth-progression
c1 f 1 in bars is surely unmistakable; nevertheless, the upper voice is forced
into leaps of a third (in lieu of steps of a second) by the lower voice, which continues in half steps in accordance with the wave-motive. Then, from the fourth
quarter of bar , sevenths instead of sixths follow, which are effective at the same
time as a means of avoiding consecutive fifths (see the graph). More than that: at
the third quarter of bar , the motivic parallelism requires the A major chord, for
which reason e1 must necessarily replace e
1 here. In addition, the lower voice
must again continue, as before, by a semitone from the third to the fourth quarter. The difficulty is great, yet the master knows how to conquer it; he submits
himself to the inevitable and shows undaunted consistency, and in so doing acknowledges the highest law of synthesis: he makes the last quarter of the sixth bar
the bearer of I! He has such faith in the force of this very rhythmic shift that he
confidently leaves to the listener the task of restoring e1 to the diatonic and thematic e
1, and in bar immediately ventures both the leap of a fourth downward,
as a chordal arpeggiation (standing for a leap of a fifth upward), and the very
harsh dissonant clashes in the second and fourth quarters. Then in bar he introduces a voice exchange (the ascending register transfer parallels that of bar ),
and by this means obtains the (see the graph). The root E
in the first quarter
of bar is merely a divider at the lower fifth, its importance intensified by the leap
of a fifth which contrasts the earlier leap of a fourth. The b
1 above it in the inner
voice is not its fifth but a figuration, more precisely a passing tone in the thirdprogression c2 b
1 a1; and the final note of the third-progression, a2 (standing for
a1), completes the neighbor-note chord a1 c2 e
2 g
2, whose functionVII
7
Vis actually in opposition to that of the divider at the lower fifth. (More concerning this kind of phenomenon in Freier Satz.18)

from the diminution, which uses inversions of the intervals throughout. The
main voice proceeds in short grace-notes, and that which juts out from them
must proceed beneath them: that is to say, the peak tones of the triplets belong to
inner voices. The true register in which the inversion is to be understood is at any
rate well understood at crucial points. There is a subtle charm in the alternation
between the true intervals and those created by inversion, especially when the
change occurs in every quarter, as in bars and . When the variation has ended
with b
1 as the in B
majorbut here as third of the tonic in G minorone has
real difficulty in crediting the note with the meaning of , so magical is its course,
swaying between two worlds.
Variation . This variation is connected with the previous one by the pedal
point. The bass voice is anchoredwe have returned to B
majorin the higher
register of b
; additionally, an inner voice maintains, with only slight deviation,
the fifth f 1, thus giving the character of a musette.
Like the theme, this variation places the ascent to and the neighbor note in
bar . By contrast, bar resembles variations , , and by bringing a repetition
and, following on, the ascent to beginning as early as the third quarter. Bars
and prolong by repetitions; only the neighbor note g3 is materially important.
Bars give the fourth-progression c3 f 3 as in the theme, and also with the
same yield, the ; also as in the theme, a succession of chords actually develops,
which is only partly concealed here by the stationary voices. In particular, the
succession f 3 g3 a3, which leads above the fourth-progression (although only a
reinforcement of the inner voice), could easily deceive us about the real voiceleading. In the fourth quarter of bar , the neighbor note supported by the G
chord is striking, being the first of its kind. In the progression of this chord to the
dominant, consecutive fifths appear between the upper and inner voices; these,
however, are mollified because the sustained root of VI is understood merely as
the bass of a chord (and not a root-position triad), and because the diminution
slips a third in between.

Variation . This variation appears at first glance to be simply a figuration of the

previous one.

Variation . The stationary bass as well as the stepwise undulation of the inner
voices in the previous variation prompt the new one. Moreover, there is a contrast
of high versus low at work between them, and of sixteenth notes versus triplets.
The and the in the first quarters of bars and are gained by an ascending register transfer, linked with the wave-like motive in the inner voice (see the

18
The final bars of variation were a favorite passage of Schenkers, as he had previously discussed them in Kontrapunkt i, p. /p. (examples and ), noting that only by similar derivations can the true relationship between strict and free counterpoint be found. The concept of divider at the lower fifth (Unterquintteiler) is not specifically mentioned in Der freie Satz.

tonw i l l e 8 9
It is possible to view, in the low register, the fifth, too, as stationary (see the
graph) and thus to see an intensification compared to the earlier variation.
It is on account of the sixteenth-note diminution that, after completing the
ascent to which follows the neighbor note, in bar the upper voice circles
around f 1 (with f 1 g
1 f 1 g
1) instead of around d1. This height could be dared
here in view of the higher placement of the ascent to in bar : d2 e
2 f 2 for d1
e
1 f 1, and also in view of the final intensification in bar , which also absorbs
completely the high register of bar .
The diminution also resumes the play of the neighbor notes, as in variation
, in the inner voices, in the upper voice, and finally also in the bass (see the last
sixteenth note, G
). However, it was impossible to include all of these motions in
the graph of the Urlinie.
Bars follow bars of variation with complete fidelity.

in the fugues of Handel and Bach, fully absorbs all this formal freedom, speaking
through every register, high and low, but without forsaking the paths of the Urlinie, and also without in the least overstepping its limitations by miming an orchestra. Anyone who can perceive the course of this fugue will be amazed by the
sublime faculty that inspires the intellect. How extensively these paths lead, too,
through realms of two and three thirds, of fourths, fifths, sixths, octaves, in ascending register transfer, arpeggiation, passages, reaching over, etc.the voiceleading always obeys the master lightly and quickly, and all the expansions are
clearly illuminated by the fire and splendor of a divinely blessed vision.
[Bars ff.] The fugue is in three parts: bars , bars , bars . Its subject is as follows:

Variation . The final variation, motivated by the arpeggiation in the last bar of
the previous one, returns to the simplest form.
The lower voice is again founded on the progression B
DF, and the and
also appear again as tenths.
{} As in variation , the and are approached from above.
The unfolding in bars is like that in the theme, except that the sixth-progression af is displaced from the upper to the inner voice and, conversely, the
thirds of the chords are shifted from the inner to the upper voice.
In bars a voice-exchange occurs, carrying the leading note upward into
the upper voice, instead of the . (The path of the exchange is clarified in the
graph.) All else is registral play (ascending and descending register transfers) and
figuration.

It reproduces the ascent of the theme using the shortest path, and leaves
to the remainder of the fugue the task of completing the necessary descent in
its own way.
Despite the brevity of two bars, the subject maintains the essence of the ascent: thus the four quarters in bar stand for with the neighbor note e
1,
and in bar the arrives amid the diminutions. The has a double function here:
it appears to complete the ascent , as though e
1 were not the neighbor note
but, rather, the en route to , and at the same time seems to arise from another
voice, which reaches over, giving the sense of a suspension and thus also necessitating the descent e
1 d1.20 If all these events contain harmonic meaning,
then one must understand the first three quarters in bar as I, the fourth quarter
as IV, and the second bar as VI. It becomes clear that the subject, equipped with
these features, was destined to have the final say in the work. Specifically, it is important for the progress of the piece that, whether or not one speaks in terms of
harmonic degrees, the subject remains in the tonic and does not end in the dominant.

Fugue. Among the fugues for the piano, this may well be reckoned as the boldest. Without aiming at alcune licenze, as in Beethovens Op. ,19 it hardly
stands in the shadow of that gigantic fugue. While it is certainly monothematic,
in accordance with fugal principles, it moves as naturally and freely as any of the
freest fugues of Handel or Bach. Life-basis and life-force in every sixteenth note,
everywhere urgency in the voice-leading and synthesis, but at the same time a
veritable ecstasy of indomitable happy freedoms. And the piano, even more than

20
[S.] In Bachs first Chorale Fantasy of the St. Matthew Passion, the same reaching-over technique occurs for the same purpose; see Tonwille . [Schenker is referring to bars in the upper
parts, and then the bass at the entry of the choir in bar ; this is illustrated in the graph of the Urlinie.]

19
The last movement of the Hammerklavier Sonata, Op. , is marked Fuga a tre voci, con alcune licenze.

Brahms Variations and Fugue on a Theme by Handel, Op.

{} The answer follows in bars and, in accordance with fugal theory, begins in the key of the dominant. The answer is real, since the fifth, f 1, in bar is
not answered by the tonic, b
1, in bar . A tonal answer would have immediately
obscured, indeed destroyed, the meaning of the subject. But the stormy mood
and tension was still destined towards the high register, which promotes an immediate connection to the third entry, and here it was entirely sufficient that only
the last sixteenth note in bar provides for the modulation back to B
major. According to conventional theory,21 the answer is in the key of the dominant, thus
in F major; however the tonal sense of the great masters, guided by synthesis and
tonality in the large, hears therein only the expression of the dominant chord. In
the graph of the Urlinie, the chords are explained from the standpoint of the
tonic key; the customary manner of consideration is indicated within small
parentheses.
As a counterpoint to the answer, sixteenth-note runs based on complementary rhythm appear in bar for the first part of the entry; the second part, in bar
, uses a staccato motive which is also thematic, as a step of a second, but which
in its staccato setting relates more to the final variation.
The third entry is in the bass in bar . The first countersubject remains, and
the second immediately takes up the staccato motive, so that both features of the
first countersubject are already in counterpoint with the first part of the entry in
bar , whereby the staccato motive, which had been set in a different rhythm, now
actually adapts itself rhythmically to the run. At the second part of the entry, in
bar , where both countersubjects execute the same staccato motive with the same
rhythm, this unity promotes the clearness and flexibility of the polyphony, which
is so vital for this fugue. A voice-exchange in bar removes the consecutive octaves between the upper voice and the bass; see the straight dotted lines in the
graph.
Immediately after the third entry the fourth follows in the tenor in bars ,
which, according to traditional theory, should be in F major but in fact remains
in the main key of B
major. In the first part of the entry, bar , the sixteenth-note
figure provides counterpoint in thirds and in contrary motion. This now lowers
the upper voice to c2 (see the graph): one could imagine the Urlinie descent
here already, except that the events which follow show that this is only a mo-

tion to the inner voice, indeed the greater part of a falling fifth-progression, f 2
b
1, in which the is still by no means disturbed;22 nevertheless in bars
an ascending register transfer occurs, leading c2 up to c3, which then continues to
the b
2 (instead of b
1) in the first quarter of bar as the conclusion of that fifthprogression. The voice-leading in bars shows consecutive fifths, which are
removed by the interpolation of seventh chords.
In the first quarter of bar , the upper voice reaches f 2 and the lower voice
again b
, the root of I; while the former still continues the path f 2 c3 (the second
part of the ascending register transfer), the latter establishes the root, which brings
about a rhythmic shift. Strictly speaking, b
2 a2 was bound to follow in the third
and fourth quarters of the upper voice, according to the series (see the
graph); however, through voice-exchange the composer brings in b
2 c3 in the
upper voice, whereby the upward direction of the next entry above is most opportunely introduced, and since it thereby retains d2 e
2 in the inner voice, at the same
time it introduces the sixths that characterize the voice-leading of the next entry.
The entry in bars , being a fifth entry, is redundant; it is present only in
order to secure the high register. If one looks back from the highest point, f 3 in
bar , to the earlier occurrences, then one can understand the logic of the
arrangement of the entries from low to high {} and also the direct connection:
through all five entries blows a wind, a storm! The sixteenth note a
2 in the first
and third quarters of bar makes no claim to a tonicizing role, and is thus no
seventh (not I
7), but merely one of a number of sixths, which only involuntarily
substitutes for b
2: it weaves its own mystery about this single fugal atom, which
belongs however to the vast world of Brahmsian spirit.
Next, the upper voice falls from f 3 in bar to f 2 in bar , and then to d1 in
bar ; at the same time, with the help of a divider at the upper fifth, the lower
voice descends an octave, in bars . In order to avoid consecutive octaves between the outer parts, f 2 must certainly enter first, in the last eighth of bar . In
terms of this divider, f 2 e
2 in the upper voice signifies the simple motion
(see the graph).
In bar a
1 reaches over d1, in order to lead to g1 as . The second Urlinieprogression begins with this note on the downbeat of bar and develops as far
as the
.

21
Der Theorie nach: here, and in his later essay on fugue (Meisterwerk ii), Schenker uses Theorie to mean the conventional approaches to the study of fugue, to which his own approach is opposed.

22
Bar is discussed further in Der freie Satz, and Fig. /a, as an example of linear progressions through an augmented fourth in the upper voice, e  b
and d  a
on the second and fourth
beats.

tonw i l l e 8 9
The , bars , is situated within a passing motion. Of course, in endeavoring to avoid an unfruitfully dissonant interval for the elaboration, the masters
of counterpoint frequently prefer to elaborate the passing motion in a consonant
manner,23 as in Fig. c:

the upper voice traverses the third-progression c  ba, but only in the sense of a
motion to the inner voice, so that, at most, c 1 only drops to c1. The movement in
the bass from A to F in bars is inverted from a descending third to an ascending sixth, while the upper voice traverses a fourth upward. Finally, in bars
(), the last chord experiences an unfolding of a sixth in the upper voice.
Fig. e) illustrates the ascending register transfer that is definitive for the
final realization, Fig. f.
Now for the particulars of the realization:

where there appears to be a I, although a true I is inconceivable within the movement from IV to V. A fifth-progression closely related to the fugue subject leads
from the bass note E
to B
in bars . A second fifth-progression leads from
the bass note B
(bar ) to F (bar ); the consecutive fifths in bars are corrected by interpolations of a seventh (see the graph). Also in bars , the bass
alternates in a playful manner between octaves, deciding in bar in favor of the
higher register, where it leads the second fifth-progression quickly to the end; not
until bar is the original register restored, by two steep octave-descents. From
f 1, the in bar , to e
2, the in bar , the upper voice goes through a seventhchord arpeggiation; see the dotted slur in the graph.

Bars . The path from the B
chord in bar to the F chord in bars ff is
carved by two fugal entries, which are in B
minor and F minor according to the
general theory. Chromatic notes are prominent in both entries, in the third quarter of bars and : d before e
and a2 before b
2; they are not introduced through
mixture here, but much more signify chromatic notes of the tonicization in the

form of I 3 IV, which nevertheless in its special way simulates major within the
minor in the fugue subject. Octave leaps, too, continue to surprise us in the lower
voice in bars , as a continuation of the octave plunges of bar : in voiceleading terms they are understood only as reinforcements, hence octaves also
occur in the same way in the upper voice of bar . By this means a master can
embrace the full range of the piano for fugal polyphony, without destroying the
system of obligatory voices! In other respects the counterpoint remains almost
the same as in the earlier entries, except that in bar the rhythm is altered for
the first time in comparison to bar . But before the upper voiceas shown in
Fig. ebegins its crucial ascending register transfer, it has to be lowered again
to d
2, since in the meantime it has been driven up to c3 by the entry in bars
; at the same time, the lower voice keeps strictly to the indicated path.

Bars ff. Fig. clarifies the paths in the second part of the fugue.
The Ursatz (Fig. a) shows the in the descent as a simple passing
tone of the second (or ninth).
The first prolongation (Fig. b) makes the passing tone consonant: the lower
voice goes from D
to F, giving the I(
III)V motion familiar to us from the
theme and most of the variations.
Fig. c shows the enharmonic transformation of the D
chord: this enables
a new path to progress downwards through two major thirds to the root of V, F,
rather than by the leap of a third upward, as given in Fig. b.
In Fig. d the unfolding moves more strongly: The first succession of chords,
B
D
in bars , is enlivened with a passing motion in which bass notes are
extrapolated, again in order to provide consonant support for passing tones that
would otherwise have been simple and unfruitful.24 Then, in bars , the lower
voice falls from D
(actually C ) to the lower third A through {} four fifths, while

Bars .25 A new technique characterizes the new section, which begins with
the entry in D
, bars , and introduces the descent of a third D
(C )A; see
Fig. c). Although it is again fugal entries that create the diminution, now for the
first time they are in the form of inversions, whose purpose it is to prepare and
develop, through contrast, the original form of the subject. The entry in bars
and the inversions in bars and all remain within the same D

chord; but to the second inversion in bars is added the lowered seventh, as

[S.] See Elucidations, Fig. .

Bars are also discussed in Der freie Satz, and Fig. d/.

23

Schenker writes T. here, a result of miscounting.

24

25

Brahms Variations and Fugue on a Theme by Handel, Op.

a chromatic note in the tonicization (D
7 V 7), by which the path for the next
fifth-progressions is opened.
In bar the bass figure, in the second beat of the entry, showsfor the first
timethe shift to the dominant chord: it is understood, however, that this signifies merely a divider here, opening a path to the lower sixth, F, in bar , which is
enharmonically changed to E  in bar . The bass continues this figure in the following bars: though appearing to be no more than pianistic arpeggiation, it still
carries obligatory voices throughout, including the stationary d
. The ascending
register transfer in bars is shown in Fig. f and in the graph of the Urlinie.

minor and in B minor) appear in the service of the diminution. Since as inversions they have a falling direction, the counterpoint can now present even bar
of the subject in contrary motion: through the ascending direction and the
fourth-progression it prepares for the next entry in the original form, which begins the third part of the fugue. Moreover, the rhythm of the augmentation, {}
the progression of the fugue theme in eighth notes instead of sixteenth notes,
was to be vitalized in advance; it is for such a preparation of the eighth notes
that the alterations are meant in the subject as well as in the counterpoint: they
are worked out in rhythmic opposition to each other (see the graph), whereby
the emphatic accentuation of the eighth notes asserts itself; this includes the
first eighth-note rest, which (in the place of the sixteenth-note rest) is organically a part of the subject in augmentation. In addition, the entry in inversion
in bars shows the further alteration, namely that in bar , as the first
part of the entry, e3 does not fall to b2 (understood as part of a fourth-progression)
as yet, but only to c3. The reason for this is to be seen in the requirements of the
voice-leading:

Bars . Yet another new technique again characterizes the section comprising bars , the section in which a transition from the A chord to the F
chord is sought (see Fig. d).26 Two complete fugal entries (apparently in A
26
Schenker writes T. here and in the heading for this paragraph, a result of the same miscounting as noted earlier.

tonw i l l e 8 9

According to this, a faulty doubling of the leading note was unavoidable unless
the obligatory parts were strictly distinguished by the help of that alteration as
well as by the complementary rhythm, so that G  of the bass, following at the
distance of a quarter note, is associated with d3, and likewise A in the first quarter of bar is associated with the fourth quarter c3 in bar . In bar the form
of the fugue subject already offers occasion to reintroduce b2 and thus to restore
its meaning again. In bar the imitative rhythm is accelerated: the bass follows
at the distance of an eighth note instead of a quarter note, through which it is
finally led, in the middle of bar , to the rhythm in which the upper voice coincides exactly with the lower voice and continues symmetrically with it; by this
means, the parallelism of the two third-progressions f 3 e
3 d3 performs an essential service.

{} Here it is striking that the bass, beyond merely providing the entry in
bars , proceeds through a large-scale fifth-progression which indeed also
has a thematic relationship to bars , while at the same time the upper voice
of bars keeps to the and moves only in far-ranging arpeggiations and
passing motions.
The realization, as before, of course uses additional entries for the purpose of
diminution.
In bars the bass introduces the augmentation in B
minor, with the already familiar chromatic D from the second part of the fugue. Also, the counterpoint is augmented (something which the sixteenth-note figure easily disguises)
and, moreover, guided by two voices in two octaves, for which the very augmentation of the subject offers occasion (see the graph). Even the previous counterpoint returns in bars (compare with bars and ), except that here the
leaps of a fourth are filled melodically with passing tones.
Brahms then uses the Handelian techniquesee for example in the fugue of
Suite No. in F minor, bars by returning to the original, faster form of
the subject in the tenors entry in bars , and at the fifth (F minor, according
to the general theory).27 However, the true answer to the earlier entry in augmentation does not occur until bars , in the upper voice, an entry at the
fifth and in augmentation as well.
In bars the bass falls from F through D
as a midpoint of the fifth, in
order to reach B
. Bars are an augmentation of that motive which arises
from bars and if, by way of abstraction, the final eighth-notes are placed
next to one another, f 2 g2 a
2 g2 f 2 (see the smaller brackets in the graph of the
Urlinie)the diminution makes use of the eighth-note motive in the third and

Bars . In principle, the B
chord (I) has already been reached in the third
quarter of bar , but it is still necessary to introduce the unfolding of a sixth in
the upper voice (see Fig. d) in order to arrive at a note no higher than b
2
(standing for b
1) as the , and at the same time to give upward direction for the
counterpoint, in contrary motion to the falling progressions of the inversion. The
one- and two-line octaves take part in this sixth-progression, see d1 e
1 f 1 g1 in
the inner voice in bars (left hand) and g
2 a2 b
2 in the upper voice in bars
. To begin with, the lower voice accompanies it with the following passage:

which clearly expresses the fugue subject with the ascent B
CD and the neighbor note E
; then however, in bar , it takes the root, in order to prepare the pitch
and register of the coming entry.

27
The Handel passage in question, from the second movement (Allegro), bars , is based on
a stretto at the fifth above, in three voices. The second entry, divided in two, completes itself after the
third entry, and the first entry, similarly divided, completes itself after the second entry, by which
means the rising fifths are counterbalanced by falling fifths.

[Bars ff.] Fig. summarizes the events of the third part:

Brahms Variations and Fugue on a Theme by Handel, Op.

fourth quarters of bar . A second augmentation, with the voices inverted, follows in bars : through both augmentations, however, the upper voice gains
only a half-step, expressed in two octaves: f 3 g
3 and f 2 g
2 (see Fig. ).
In bar the bass has arrived at the root of I, while in the meantime the upper
voice has already advanced from the fifth of the chord to the sixth with g
2.
Thereby a chord results as the initial member of a chain of chords, which continues to c
3 in bar . In bars the inner voices, beneath a stationary uppervoice g
2, work from g
1 up to b
1 likewise by means of the eighth-note motives
of the previous bar, but differently rhythmicizedand in this way the position of
bar is regained in a higher register on the downbeat of bar . The effort of the
inner voices is, however, not just a question of confirming the first register, but
through its acceleration of motives can suggest a still greater acceleration, which
is indispensable for the passage through the chromatic notes in bars (see
the graph). It is also these notes that initially suggest b
2 c
3 in bars , rather
than b
2 c3 (see bar ). In bar the bass finally reaches the fifth of the progression, here functioning as a dominant, while the upper voice enters again with b
1,
that is, as a suspended fourth, so that the resolution of the suspension can at the
same time unite the broad progress through the octaves to f 3 in bar . Fugal entries, which, with a few alterations (see bars ), accommodate the new objective, are again used for the purpose of diminution. The bass provides counterpoint using inversions of the subject, so that the entries proceed in both directions
simultaneously. In bar f 2 is attained, in bar b
2, and in bar f 3. In bars
a fourth-progression of the upper voice is thus united with a third-progression
of the lower voice (see the graph), and the looming {} consecutive fifths in this
passage, like those in the following bars, are removed through interpolated sevenths or sixths. However, all this voice-leading does not yet signify a pedal-point,
but only an elaboration of the V chord by a suspension and its resolution.
The V is now prolonged until bar , while the upper voice falls through a
fourth-progression from f 3 to c3. This downward motion, which exceeds that in
the first part of the fugue (bars ) by one note, can be explained here not just
in the sense of the Urlinie, but also as an inversion of bar ; this is confirmed here
by the progressions in contrary motion in bars ff. The descending fourthprogression is also used in quarter notes, of course, in bars as the main
diminution motive, in accordance with the quarter-note progression of the motive in bars ff.; it is even developed in imitations, which follow at the space
of two quarters (see the graph). We see in bars a series of such fourth-

progressions, which cover the space of the octave f 2 f 1 (accelerated in bar from
c2 onwards, as the divider of the octave). At the inversion in bars , the motive appears in a rising motion, again accelerated in bar from c1 as the divider
of the octave. The falling fourth-progression returns in bars , in a still lower
octave; this time, however, unlike bars , the acceleration tears away
almost violentlya bar earlier, and at this point, in bars (see the graph),
draws in an expansion of the falling fourth-progression in the lower voice, which
thus falls through two octaves. Consequently in bar the endings of the two
falling fourth-progressions in the upper and lower voices converge. By appropriate distributiond2 appears as early as bar and arpeggiation in contrary
motion in bars , the consecutive octaves are avoided.
Not until now does I enter, in bar , and the bass at first takes the lead. Along
with the motive of the fourth-progression, which corresponds to bar of the subject, Brahms begins in bars simultaneously the last augmentation of the
same with B
CD; e
2 in bar continues this progression and then introduces
f 3 in bar as a neighbor note! In bar d3 finally appears, coming from e
2
(which stands for e
3). In the following bars this d3 is topped by b
3, as in inversion,
signifying the (bar ) that finally follows from the (bar ). This manner of
superposition of b
3 above d3 recalls the unfolding of a sixth at the close of the second part of the fugue, where d2 b
2 is likewise expanded within I (see Figs. d
and f). What force in the diminutions, and what certainty of Urlinie direction!

Thus the young Brahms writes the variations and the fugue.
He himself, being blessed with a love that is able to create life, and having also
an ear for the creative love of a Handel, hears in the aria the sounds that Handel
has woven out of sounds, the life in their rise and fall, the connections throughout, the diminutions, neighbor notes and unfoldings, in short everthing that
moves within them purposefully, and therefore, like a second father, can beget
likenesses of the aria.
But their ordering, too, comes to him from creative love. For the variation
images do not simply glide by like scenes in a panorama, from one bell signal to
the next;28 nor are they based only on the linking effect of opposites or of in28
Among the many types of panorama popular in the nineteenth century was the diorama, invented in by Louis Jacques Mand Daguerre (), in which an audience, seated on a

tonw i l l e 8 9
creasingly intense movement: to him the variations become literally a sequence
of generations, as one could say in the genealogies of the Old Testament: and the
first begat the second, the second the third, the third the fourth, and so on. For
him, nothing in this process is too small, too insignificant, that it might not be
called upon for a new act of procreation.
{} The fugue is for him not a piece of artistic homework, to be accomplished according to a textbook model. The last, highly passionate variation
wishes to bemust besurpassed: Brahms now hearkens to its compulsion toward even greater desire, and, with almost carnal force,29 he follows it by a fugue
that is jubilant to the heavens, as if the fugue itself would become an equally vast
sequence of ever new variations once again, but here united into one great
whole. The laws of the fugue willingly confront this untamed delight: the answers at the fifth, transpositions to other modes, augmentations, inversions, and
all kinds of other imitations; all of this, tamed through tonality and Urlinie,
preaches the soul of the Aria with a thousand tongues, in testimony of its ever
ascending curve of life.
For Brahms the exposition is not merely a textbook obligation to present the
four voicesrather, he storms through the voices because he wants to gather
their octaves as if in one ringing sound, the joy of B
major.
To the first part he adds the middle part with of the Urlinie, and
through the voice-leading he forces into the main tonality all the chords resulting from mixture, chromaticization and passing motions as apparent keys (D

major, G
major, etc.): in what a bold arch of voice-leading does Brahms bring
about a resounding of a tone of the Urlinie as the primordial breath of its expansive movements!
And life becomes ever bolder and more high-spirited in the third part: the
counterpoint sounds as if coming from a lung full of octaves, spreading strength
of force, significance and jubilation in all directionsfinally there are passionately elevated expansions, unfoldings and augmentations, full of desire, which
now add breadth to the height and depth.
He came fresh from heaven and did not need to strive first toward purifying

himself on earth. God knew him, he knew God, but those who know not God
have no access to him.
He was naive, chaste and pure, like every genius, and his proudest virtue was
modesty. He was, in truth, endowed by God and thus could be modest; for only
those who are endowed by men become immodest, defiant, bitter, and ungrateful.
The sun-storm of synthesis which raged through German art for two centuries bore him aloft on its mighty wings; it was as if the synthesis wanted to
make sure of its own self again, of its power and its blessing. He came into the
world in a breath of necessity: his birth was a historic necessity, birth and death
called him into the ranks, even into the physical proximity of the greatest.30
Like them, he created from the background of the tonal space, from the
simple melody of the seven-note Urlinie,31 and forged ahead with his own original voice, by means of highly secret transformations, into the foreground full of
the most ravishing motives and diminutions.
He created what was perfecthe realized dreams of human lifeand took
part in artistic immortality as a God-given substitute for corporeal immortality,
a synthesis of art in place of a synthesis of humanity. Therefore he, like the greatest, will force humanity to stand alongside him.
Even though humanity, because it is ever coming and going, will always demand what is new, it will nevertheless have to remain standing alongside Brahms.
In the long run, mens ingratitude, their lack of understanding, will prove to be
too weak; for they will be more strongly attracted by the secret of his powers of
synthesis and melody.
Idle questions bounce off him: is he a Romantic? a Classicist? an Epigone?
a Modernist? a North-German? a Viennese? All these questions are posed by
phrasemongers, by people who belong to no time, not to yesterday, today, or tomorrow, who also have no {} spirit, and therefore most zealously pursue the
spirit of the age [Zeitgeist]. He will not be turned into a monument, nor will he
be overcome by todays progress machines and their slaves. He survives the rebirth of all melodies which, with rather less art or nature, refresh the moment but
cannot expand the moment of existence. He does not succumb to the new German freedom and self-determination, which the peoples of the world have given

revolving platform, viewed various remarkably lifelike scenes. The bell signaled a man to operate the
mechanism which would rotate the platform and audience from one scene to another (usually contrasting) scene. See Helmut and Alison Gernsheim, L. J. M. Daguerre: The History of the Diorama and
the Daguerreotype, nd ed. (New York: Dover, ), pp. .
29
Schenkers wording in this sentence (hochentflammt . . . Lust . . . mit geradezu animalischer
Kraft) has strong sexual connotations.

30
By rtliche Nhe der Grten, Schenker is linking geographic proximity to artistic distinction,
Vienna being the natural or adopted home of Brahms and a great many other German and Austrian
mastersand of Schenker himself.
31
In Schenkers reading of the aria, the Urlinie comprises the ascent from to plus the descending fifth-progression to .

Brahms Variations and Fugue on a Theme by Handel, Op.

us out of wisdom and generosity, and for which Otto Ludwigs words are so appropriate:

First of all it should be observed that Brahms has altered all Handels
to
tr, except those in bars and , where he put
; there is certainly nothing objectionable in this alteration.33
Dynamic markings are lacking in the original score of the aria; Brahms has
not provided any, either. The dynamic should be understood as something like
pf,34 an open tone which does not contradict the melodic character of the Aria
and permits dynamic fluctuations up to f and down to p. This sketch of the dynamics follows the form of the Aria:

He who seeks true freedom, should first work towards making himself
free, i.e. making his life into the fullest expression of obedience to the commandments. Dear God, if freedom, which we are to obtain, bears a resemblance to those who purport to provide it for us, then I would add an
eighth request to my Lords Prayer, thus: and deliver us from freedom.32
He is not shaken by the delight in servility of the German intellectuals who
have deserted genius, and who, out of a lack of any talent, of any refinements,
even of personal character, because they have abandoned genius, embrace all
Russian, French, Italian, and English crudeness and dilettantism with grateful
artistic enthusiasm, even though these are already well surpassed in synthesis and
culture by, for example, Negro melodies, which have grown in the natural wild.
He does not give way to the national ruses of certain people, who, in order to aggrandize the smallest of their own, wish to diminish him in particular. He, finally,
also defeats the worldliness that misrepresents the materialism and cunning of
today, particularly for personal gain, as the goal of life and as the source of a new
world art. All the wealth of worldliness is not sufficient to capture one little tonespirit [Tonseelchen] of Brahms: he is not made of stocks and bonds, nor of
newsprint; he is all soul, all genius, therefore it is only through the soul, in fervor
and humility, that one can reach him.
He remains standing. Henceforth, like the greatest, he shall be part of humanitys calling: for just as the individual human being, man or woman, is for a
long time mistaken and afraid until he takes root in his vocation and his task,
thus also humanity as a whole is ever searching for its vocation, and at some time
humanity will find that only geniuses are its true calling, in which the ultimate
purpose of its existence is fulfilled.
Brahms was one of the greatest, a geniushe created from the background
and therefore remains in the background of eternity!

T. 1
























5














7




 
















In bar the thumb is to be used on the first quarter note b
1, and bars and
should also be played with the hand in this position. The trill in the second beat
of bar should be performed in thirty-second notes (in the first eighth two trill
pairs; in the second, a triplet and the written-out ending). In the third beat the
first two sixteenth notes are joined in a legato fashion and {} should certainly
be detached from the third, with which the diminution begins; this however is to
 . The neighbor note in the last eighth of the bar should be played
be played as i
expressively; the same applies, to a lesser degree, to the neighbor note in the last
eighth of bar , according to the dynamic plan given above. The quarter note f 2
at the beginning of bar must be stronger than the previous g2. This bar is legato
from f 2 to d2. The trill in the first half of the second beat should be played rather
richly and freely, but nevertheless keeping the neighbor note in the last sixteenth
note of the beat strictly to its sixteenth-note value. In the fourth quarter the
thirty-second note figure leading back to the repetition should be articulated
motivically: f 1 drives toward a1 in the second eighth and g1 toward b
1 in the first

33
Brahms himself mentions this alteration in a letter to Breitkopf & Hrtel (April , ): The
theme is found in the second volume of Handels works, page , and according to that the notation
could be improved only in that I have occasionally placed tr instead of
on purpose. See Johannes
Brahms Briefwechsel, vol. , ed. Wilhelm Altmann (Berlin: Deutsche Brahms-Gesellschaft, ), p. .
See also Kalbeck, Johannes Brahms ( edition), vol. , p. .
34
Schenker probably means poco f, which is also what Brahms uses at the start of variation ; pf is
not a generally recognized dynamic marking.

Now, some remarks on performance.

32
Otto Ludwig (), German novelist, poet, playwright, and critic, studied music with Mendelssohn in Leipzig, and wrote a comic opera Die Kohlerin (The Coal Burner) in . His masterpiece
is the novel Zwischen Himmel und Erde (). The source of the quotation has not been found.

tonw i l l e 8 9
quarter of bar . Also in bar the diminution (from f 1 onward) is to be separated
from the previous sixteenth notes, which should be played legato. The anticipation in the first beat of bar is legatissimo: c2 may even continue for a full quarter note here in the sense of the longest legatissimo (in addition to the d2). At the
crescendo beginning in the third beat of bar , the left hand takes a particularly
lively part, in that it must also play the notes in bar like a horn part. On the
other hand, at the p the neighbor note e
2 in the last eighth of bar should be soft,
as in the dynamic plan. Across bars |, notwithstanding the leap of a third, the
fourth finger should be used twice in succession, on e
2 and c2.35

one adheres to the crescendo of the basic plan (especially in the left hand); this is
the pattern for the thirty-second note figure in the fourth beat of bar , which is
then no more of a hindrance for the realization of the dynamic plan: hence the
figure is to be played p and the final sixteenth notes in the left hand are to be performed without haste.
Variation . The animato direction in this variation refers to the increased motion of the tripletsthey automatically prevent any respite, even if the master
had not expressly indicated itaccordingly, one succeeds in this acceleration of
the tempo only when one has found the right tempo for the theme and the first
variation as well. It is not correct to play the two g2s in bars and and the two
b
2s in bars and with the same weight; each time, the repetition is to be played
softer, despite Brahmss indications < >, since it is free from the burden of
the ascending motion. However, an emphasis is surely required for the e
1 in the
left hand in each third beat of bars and . In bar the fourth-beat b
2 should be
supported by a strong c1 in the left hand. In bar one should brush d
2 only
lightly. The quarter notes that have been notated as passing tones from the fourth
beat of bar in the inner voice are understood only as the expression of a {} pianistic legato, nothing more. The crescendo indicated in bar clearly agrees with
that recommended by me in the basic dynamic plan. It would be wrong however,
if the player thereby allowed himself to be led to f in bar ; since Brahms notates
a crescendo across bars | on account of the ascending register transfer, he obviously desires a p in the first beat of bar as a preparation for the crescendo. The
master always places his marks only with the intention of synthesis; they are not
to be carried out to the letter, except when one has also considered the continuation; when therefore, as here, after the crescendo another follows very quickly, it
should be understood that a p shadow lies in between.

Variation . In the first variation the dynamic plan remains the same as in the
theme. The direction of the composer pf confirms the dynamic quality which is
brought about by the compositional method of the variation itself; thus it should
not have a further pf or indeed f as a result of this indication. All the sixteenth
notes in the right and left hands should be played p and detached, while a pressure should be given to the following eighth notes (demanded by Brahms with >).
From the eighth notes to the sixteenth notes, however, the hand must slide in an
easy legato, especially in each of the third and fourth beats of bars , , and , and
in the first and second beats of bars and , where leaps of a third or fourth are
used instead of steps of a second: here the right hand may even emphasize the
melodic detours of the upper voice with a light turning-out of the forearm toward the right, as if painting, onomatopoetically.36 However, the chief requisite
of the performance remains the continuous participation of the low bass notes
(Grundtne) in all dynamic undertakings according to the basic plan given above.
In the ornamental figure on the last beat in bars and , c3 and b
2 are to be
expressed as octave reinforcements (which Brahms aims at with the marking
< ). In bar , notwithstanding the thirty-second note figure in the third beat,

Variation . Because of the suspension effect, greater weight should be given to

the first eighth note of each beat throughout, even though it is always unaccompanied, than to the chord that follows in the second eighth! Equally, a portamento
play is to be continued from eighth note to eighth note throughout. In bar , b
2
is above all to be played delicately: the hand lingers in the high register and makes
the tone ring on (see the graph). The crescendo at bar remains within the basic
plan, although this leads here not to the but only as far as the neighbor note.
Before Brahmss < , introduced at the last eighth of bar and extending to the
third beat of bar , a p is presumed in bars , with which the indispensable

35

This fingering direction illustrates Schenkers view that even fingering should contribute to the
voice-leading conception. Using the fourth finger twice reinforces the linear discontinuity of e
2 c2,
which helps to project the deeper connection d2 c2. For more on Schenkers view of fingering see
Charles Burkhart, Schenkers Theory of Levels and Musical Performance, Aspects of Schenkerian
Analysis, ed. David Beach (New Haven, CT: Yale University Press, ), pp. .
36
Schenker rarely appeals to the visual aspect of performance practice for the sake of musical expression. However, Carl Philipp Emanuel Bach, whose writings Schenker greatly admired, commends
such means of expression; see the Versuch ber die wahre Art das Clavier zu spielen (), part I, chapter , paragraphs ff, in English as Essay on the True Art of Playing Keyboard Instruments (), trans.
and ed. by William Mitchell (New York: Norton, ), pp. ff. See also the remarks on the performance of variation .

Brahms Variations and Fugue on a Theme by Handel, Op.

light touch on G
in the bass in the first beat of bar agrees very well. The complete sonority in bars and as far as g2 can not be expressed fully enough. In bar
one must not overlook that the V chord is suspended in the second eighth note
of the second beat; allow the peak of the arpeggio, f 2, to ring on freely, and pause
a little before proceeding to the third beat (I). At the end of the variation, the
player should form the left hand in a somewhat stretched position, as in presentiment of the coming one: the left hand waits until the right hand can move to its
position precisely, in order that the loosening of the hands will also be expressed
distinctly in the touch.

sion (use the fingering , not , in all these places): if the third [of the chord]
in the right hand now follows the third in the left hand at the distance of only one
sixteenth note, then {} both thirds will flow characteristically into each other,
and the sum gives a truly wonderful portamento from b
to d
and from d
to f, as
if to make at any given time only a single third there. The first sixteenth note, b
1,
in the left hand in bar must be sustained, in order that the b
1 on the second
beat in the right hand can follow within the sphere of the echo: in this way the
pianist can fulfil the law of the obligatory register on its own account there, too,
where the notation imposes limits on the composer. The full weight of the V
chord is achieved in the fourth sixteenth note of the second beat in the left hand
in bar : one links this sixteenth note to the identical second sixteenth note of the
third beat, and this artificial connection provides functional strength, thus compensating for the brevity of the chord.

Variation . The octaves are to be grouped around the sf (in bar Brahms has
only indicated >): a stronger and at the same time warmer touch at the sf chords,
which almost lifts the forearm from the keys, easily brings the octaves that follow
the sf as a result of the rebounds. In particular the octaves in the left hand in bars
should be brought into action in this way; they have the ability to skip regularly in each single salvo like a pebble spun across the water. After the last sf in bar
, a crescendo of accumulated sound, there follows an artificial rest in which the
sound disperses, so that the downbeat of the following bar takes an artificial p
shadow37 of its own accord; then the next sf at the last sixteenth note of the second beat restores the connection with the previous sf at the f 3. The concluding
passages of both the first and second parts are stern, pithy, and to be played detached. This manner of playing in the variation is actually suggested by the appearance of the beams, which the master has drawn in accordance with the content: the chords provided with sf stand alone, the beamed chords are heard in a
contrasting piano touch; the last two chords are both marked forte.

Variation . For the masters, the instruction p sempre always indicates a sort of
motionlessness, stillness, as if the imagination of the composer held its breath; it
is so in this variation as well, despite the moving eighth notes and sixteenth notes.
The piano keys are to be articulated only lightlylegato is achieved partly
through the fingering, partly through the touchonly the high points in the
right and left hands should be emphasized: b
2, d
3, and so on. Even in the repetition of the same high notes at precisely regulated intervals, the restraint of the
sempre p can be found. Do not in any event play slowly on account of the minor
key and the octaves; rather, keep in mind the overall effect of the high notes. At
these points, if the player strokes the hand (right and left) gently on the keys, they
will end almost the way in which a string player ends a note with a longer bow
stroke in spite of its durational value. In bars and the sixteenth notes are to be
grouped as follows: the first two belong together, the last two prepare an easy
glide over the keys (lower and upper keys), so that the second sixteenth note is
not in any way carried along in the gesture as well.

Variation . With full awareness the player should convert the last leap of a third
in the previous variation to the first leap of a third in the fifth variation. The indication espressivo refers to the interior swelling < > in the second half of
bar and the first half bar , and so on. The performance of these signs must not
be exaggerated, but should rather conform to the flow of eighth notes and sixteenth notes. In the left hand, the d
1s in bars and , the f 1s in bar and, likewise
at the end, the d
1s and e
1s in bars and are to be tied over with warm expres-

Variation . Brahmss markings > here are indications for the technical realization; the touch must explode, and the release then ushers in the sixteenth notes
and facilitates the mechanics of the repetition. This applies in particular to the
groups of six sixteenth notes beginning on the second eighth of the second beat
of bars , and ; if the rebound operates properly, the hand will shake out all the
sixteenth notes almost as a single action. The sign > above the last eighth of bar
affects the neighbor note. The last eighth note in each of bars and is marked

37
ein knstlicher p-Schatten, that is, a p that is not actually marked in the score but is to be supplied by the performer; likewise Kunstpause (artificial rest) immediately above and, in the performance notes to variation , ein knstliches legato.

tonw i l l e 8 9
Variation . A stronger stroke on the first triplet eighth note will gain the following two eighth notes automatically. But anyone who, distrusting the natural
law of pressure and release, fails to provide the necessary elasticity at the first
eighth note, must reap the ill, that the third triplet eighth note will move to the
next in legato. The player must be especially prepared for the change of register,
particularly in bars and where the direction suddenly reverses.

sf within a legato; the more calm one preserves in the previous note repetitions,
the easier it is to play the sf. From the crescendo marked in bar , one can infer
that the first beat of the bar should be played piano.
Variation . In the left hand, not only is the finger exchange possible, but
also the use of the third finger alone (with the thumb remaining close by). The
sixteenth notes will achieve pressure and release by themselves; the pressure will
be easier to achieve if, at the same time, a pressure is applied in the right hand
(see the masters indication < on the beat in question). A non legato is selfevident, except in bars , where legato is also indicated several times. (In these
very same bars, the player, if he has previously used only the third finger in the left
hand, can change to a fingering; thus refreshed, he can return to the exclusive use of the third finger in bars .) In bars of the repetition, that is, in the
last two bars of the variation, the player should strive to attain the complete sonority from b
to b
2 as a pillar of sound; in this way he will prepare the next variation.

Variation . The tempo is to be maintained fluidly here on account of the increased diminutions and the wide-ranging passing motion in the left hand. At the
fourth beat of bars and the tempo yields a little, which however presumes a
slight acceleration on each first beat here. In bars and the play of gf
against g
f, as well as b
a against ba, must not be overlooked (see the analysis). In bars and the imitation in the upper and lower voices should be emphasized, that is to say, the legato is to be performed in the strictest manner in the
i
left hand, hence a sliding fingering from d
1 to c1 and from e
1 to d1 can be applied without further ado. The indication p dolce at the beginning of bar refers
back to bar and reveals the synthesis of the variation! At the third beat of this
bar, the imagination leaps to the octave b
B
in advance, and the left hand will
then easily traverse it as a unit. In the second beat of bar , the two sixteenth-note
b
1s are to be played with the second and third fingers.

Variation . The direction poco sostenuto must not, as we have come to understand it, be interpreted as an indication of adagio; apart from the poco which limits it in any case, {} the tempo of the variation is determined above all by its
connection with the previous variation, and the sostenuto refers thus to the pillars of sound in bars , , , and . Since the octaves must be played here in such
a way that their reverberation extends across two full bars, it indeed demands a
moving tempo. In the repeat of bars , the succession F (G
)B
should consciously be contrasted to the earlier FB
; the effect can not succeed, either, if the
tempo is broadened. The left hand takes precedence, owing to the motive; here
one leads to the highest point (fourth beat) with a < ; in the next bar one returns to the diminuendo with a relaxed hand, and as well with a sort of acceleration to the starting point. In the right hand in bars , and thus also in bars
, and so on, one should brush the middle eighth note of each triplet only lightly;
the hand always remains in position for the descending motion of the sixthprogression, and therefore the legato is not actually executed through the triplets
but is played as an artificial legato, which, without endangering the effect of legato
at all, thus preserves the sixth-progression as well as enabling the octaves to be introduced. The >, marked from the second to the third beat of bar , confirms
the manner of performance recommended here. The p that follows directly after
this sign requires a further softening, something that is usually rejected by pianists who play the variation too broadly and are apt to use a < at this point.

Variation . The left hand is non legato throughout, thus even in places where it
has motivic elements. The indication soave refers primarily to the right hand. In
fact, the g2 f 2 neighbor-note play in the first and second beats of bar (and similarly in bars and ) cannot be presented tenderly enough: g2 is scarcely touched:
it supports the eighth-note f 2 with only the slightest pressure. By contrast, g2 in
the third beat receives a stronger touch and f 2 becomes softer; move quickly from
this note to the penultimate sixteenth note, b
1, and strike the neighbor note on
the last sixteenth softly, and consciously imitate with a slight turning-out of the
forearm. In bar observe the strictest legatissimo in the chromatic descent in the
right and left hands; here, too, the leap of a fourth c2 f 2 (in the third beat) should
be imitated with a slight turning out of the arm. In bar the ornamental play of
the small slurs in the right hand is to be introduced cautiously, but accelerated
from the third to the fourth beats. {} The right hand participates expressively
with the motion of the motive (left hand) in bar , especially in the third and
fourth beats; in the last eighth of the bar there should be pressure in the right and

Brahms Variations and Fugue on a Theme by Handel, Op.

left hands. The chromatic progression in the first and second beats of bar
should fall with the effect of a glissando.

until the last eighth-note of the bar if the player does not, through fear, change
his manner of playing in between. If the principle of attack and rebound is not
successful here, then it will be necessary to proceed much too cautiously, indeed
fearfully, sixth by sixth, whereby either the tempo will be delayed, contrary to the
indication sciolto, or, for security against unavoidable technical weakness, a much
too lengthy pedal will be showered over the whole bar.39 The two forte markings
at the last eighth note of bar and the first eighth note of bar apply also to the
bass, all the more so as the execution of the leaps in the whole of the second bar
depends on the first forte in the bass of bar (see earlier). In spite of the crescendo
marking at the end of bar , strength must be reserved through the course of bar
for a crescendo from the fourth beat of this bar to the forte in bar .

Variation . In spite of the indications largamento ma non pi and f espressivo, in

spite of the rolled chords in the left hand and the figures in the right hand, which
clearly point to a gipsy fantasy to be performed with a swing and caprice, in spite
of all this, the variation should be played as a funeral march.
The tops of the rolled chords strike the melody notes. Make a crescendo from
the first beat to the second, a diminuendo at the falling third-progression in the
second and third beats, then < > again at the figure in the fourth beat. The
quintuplet in the second eighth of bar should be played freely; in particular, the
first two thirty-second notes should be delayed a little. At the third and fourth
beats in bar , continue with crescendo. From the fourth beat of bar to the
downbeat of the following bar, emphasize the suspension by a crescendo. The
thirds in the right hand in bar to follow the style of the variationare all to
be fingered . The quintuplet in the last eighth should break loose wildly, in the
gipsy style. The thirty-second-note figure in bar of the repeat is to be articulated
precisely (in the original edition an eighth-note rest appears by mistake instead
of a sixteenth-note rest),38 so that the second f falls exactly over the arpeggio.
Across bars | a b
sixteenth note actually rises up: a stronger pressure on this
note, together with a lingering, reducing the final thirty-second-note b
1 almost
to a sixty-fourth note, will give the proper sound to this passage. The following
chord in the right hand at the first beat of bar should be struck passionately and
warmly; at the same time the bass note, if likewise struck warmly, will radiate the
chord broadly. Bars want to be played freely, but still within the sense of the
structure: the end of the diminuendo should coincide with that of the first fifthprogression (third beat). The fourth beat should be given a new impulse of
warmth, and not until the following bar should one recede with a diminuendo.
Again, the quintuplet should be played freely, as in bar . (Should not the thirtysecond notes in bar according to the first editionbe sixty-fourth notes?)

Variation . The start of the sixteenth-note figure in bar requires an artificial p

and a careful realization; a crescendo (with acceleration) occurs at the last sixteenth notes, which introduce the diminution. The first pedal of this bar extends
only to the second beat, the third and fourth beats are repedaled. In bar , where
the figure is extended, the artificial p continues to the third beat (the pedal also
continues to this point), and only from the {} third to the fourth beats should
there be a crescendo (without pedal). In bar the descent from d2 to c2 at the third
beat is to be expressed through a > (no pedal here). In bars one achieves
a full sound in the high register by means of the crescendo.
Variation . The tempo is the same as in the previous variation. The direction
ma marcato refers to the sharp staccato eighth notes in the high and low registers;
a stronger emphasis on each eighth note is thereby understood. The repetition of
the marking p in bar presupposes a crescendo in bar (on account of the dissonant chord). In bars ff, every first eighth note should be emphasized; one
should not overlook that the eighth notes leap in the opposite direction here [in
the right hand]. The repetition of the p in bar signifies a prior crescendo in bar
(on account of the chromatic a
). In addition, emphasize the third beat constantly; for were one to yield completely to the law stipulating that the introduction of a motive on a weak beat requires its own emphasis, one would thereby
lose the fundamental meter altogether.

Variation . The difficulty of playing the sixths in this variation is easily removed
if one simply trusts nature and applies the requisite pressure in the first beat in
the right and left hands; the impetus of the rebound will then certainly suffice

39
Clara Schumanns detailed fingering of this difficult passage can be seen in the Library of Congress autograph.

38
In the Library of Congress autograph, the sixteenth-note rest is indeed crossed out and replaced
by an eighth-note rest in bar but is not altered at the repeat.

tonw i l l e 8 9
it (despite the < >). Brahms accompanies the ascent d1 f 1 with a crescendo
mark. In bar the motion increases a little, as in bar , and slackens from the second beat of bar . {} Bars and are to be played as a single idea, which
Brahms expressly shows with a broad < (at the same time a sign of synthesis).
In bars , p espressivo refers to the lower register, which requires a warmer tone.
However, it should not be overstated, but sounded as in sotto voce.

Variation . The indication pi mosso expresses the accumulation of leaps,

which as such do require an acceleration of the tempo. The motivic elements in
the left hand in bars should be played (with a certain amount of acceleration)
almost as if under a continuous slur, with a hesitation only at the two eighth
notes across bars |. In general, the height of the right hand must be continually
varied in the execution of the leaps, always in accordance with the motion and
expression of the motives in the left hand; especially in the final bar of the variation, the colours of such a manner of playing help to clarify the ending.

Variation . The continually pressing desire to invert the original intervals here
(see the analysis) will be appropriately expressed in a moving tempo as well. In
the midst of this continual need, however, the player should seek a pause in each
first beat of bars and and each third beat of bars and ; one accelerates toward these beats from the third beats of bars and . The extremely beautiful effect of the further heightened interplay at the rearrangements in bars and will
be promoted by a crescendo that continues to bar . In bar there should be a
diminuendo leading away from the first beat. The triplet in the second beat of bar
, like that in the second beat of bar , is to be performed portamento: a softer
touch from tone to tone by gently undulating hands.

Variation . The indication grazioso expresses itself here in the gracefully arranged alternation of portamento and legato in the motivic elements in the left
and right handsthis is, as a rule, overlooked in public performanceas well as
in the elegantly playful sixteenth-note figures in both hands. A warmer touch in
the first beat of the legato group separates it slightly from the portamento group
and additionally gives a welcome pause for the player in the charming flow of
syncopations. Brahmss particular indication < > in bar serves the synthesis, and at the same time requires continuation in a single progression to this
point, in spite of all the internal fluctuation. All of this applies also to the execution of bars .

Variation . The musette can not be perfectly expressed unless the f 2 in the right
hand, precisely according to the indication, receives its proper touch, from one
eighth note to the next, so that it actually overpowers the upper voice. The
melody will nevertheless assert itself if it merely maintains its inherent brightness
at a few crucial places: at the third beat of bar , the third beat of bar , and at the
quarter notes and eighth notes marked marcato by Brahms in bars and . In addition, observe the marcato signs also placed by Brahms in the last eighth of bars
and ; these, however, are meant not for the eighth-notes but for the first thirtysecond note in the inner voice, so that, thereby provided with a profusion of colors, the musette really appears in the garb of an orchestral piece. At the last eighth
of bar , the tempo should be held back a little.

Variation . By leggiero e vivace, Brahms requires the chords to be played staccato. The inverted mordent [Pralltriller], wherever it occurs, whether in the inner
or the upper voice, is to be dashed off as quickly as possible; thus there should be
no legato from the ornament to the next sixteenth note. In bars and , one
should not overlook the sign >, which demands an emphasis in the appropriate
chords.
Variation . The tempo is quick. The motion should be directed at the quarter
notes marked by Brahms with < >: the fourth beat of bar , the second beat
of bar , and so on. The warm breath of the chromatically moving legato bass
promotes this movement. The variation expresses a restrained, manly passion;
contrary to the sense of the variation however, and contrary to the composers instructions, most virtuosi play it as a solemn, lofty adagio, to which they are obviously seduced by the visual appearance of the chords.
The fact that the fourth-progression of the bass does not finish until the second beat of bar gives the player an indication to hold back a little here; nevertheless, a slight acceleration already from the third beat of bar must prepare for

Variation . As pointed as the staccato is here, and as dark its mood, it must
nevertheless be strangely warm at the apex points, which Brahms indicates with
crescendo and marcato signs. The stationary bass note, the altogether low register,
and the aforementioned accents give to the variation the movement and expression of a ballade. This is particularly expressed in the three marcato signs of bar
and in the climax b
1 in bar . When in bars the variation soars ever higher,

Brahms Variations and Fugue on a Theme by Handel, Op.

one should not forget that here the high register is only a respite from the darker
low register, but by no means a bright, friendly light.

continuing fortethough here, since indeed the composition of a fugue is not

developed chordally like a homophonic composition, the forte is to be understood as only that of a simple, open tone.
In the performance of the subject, the sixteenth notes should always be played
to some extent as an upbeat leading to the quarter note. Continued throughout,
this manner produces a commendable quarter-note emphasis, which makes the
composition more directed and quicker than would be accomplished by utterly
uniform accentuation of the sixteenth notes.
Also, at the staccato counterpoint in bar , the first sixteenth note is a dynamic
upbeat to the staccato accented eighth-note. This principle of performance, too,
remains valid for the entire fugue.
In bars the strange a
2 and a
1 should also be played with appropriately
strange expression. The a
1 is to be played darkly in bars . In bars , play
the left hand with a sensitivity for descending octaves. Bars should be understood, for sake of performance, only as a succession of two harmonic degrees
[Stufen] with an interpolated passage; therefore the uniform dynamic markings
pcrescendof also have meaning here, as an image of synthesis.
In bars [recte: ], the subject should be played beautifully with regard to the legato, the chromaticism, and the restoration of diatonicism. One
should retain the same beautiful expression in the answer at the fifth, in bars
. The pianistic euphony of the entry on the D
chord, bars ff, will result from
the conduct of the lower voice (left hand), which verges on the homophonic style
(see the analysis). The expressiveness increases in bars through the tendency of the
at work there: the chromatic quality of the tonicization (in the
general way of speaking) requires greater emphasis here. In bar the octaves in
each first sixteenth note of the second and fourth beats are endings to the previous groups of octaves. In bars one should strive, like the composer, to
achieve a metric balance, that is to say, to overcome the previous emphasis, and
in the end give happy expression to the joy of success, so to speak, with an f in
bars .
Even in the augmentation of the fugue subject in bars ff, the eighth notes
preceding the half note should be restrained in an artificial shadow of p (like the
earlier sixteenth notes preceding the quarter notes). The same in bars ff; here
Brahms expressly notates > on every third beat. The < >, added in bar
and also valid for bar as well, are only movements within p; they are understood, by the way, also in bars . The manner of performance already recommended for the fugue subject should in particular be strictly observed from

Variation . This variation paints an even sweeter, even more ballade-like effect,
through the tumble of sixteenth-note figures in darkest piano. Individual expression appears here also through the marcato signs in bar and through b
2 and a2
in bar . The difficulty of managing a < in the figures of the left hand in bars
and as opposed to the marking of > on each beat in the right hand will easily
be overcome if one first practices the left hand alone, in order to introduce the
swell; it supports itself on the bass note until the fifth has been reached, and then
gives a small jerk up to G
and returns, and so throughout.
Variation . Although this variation presents the player with no particular technical difficulties, I must express my astonishment and disappointment that the
dotted eighth note on the third beat of each bar, marked marcato by the composer, is never performed according to these instructions but is always released in
the same manner as the preceding staccato eighth notes. Players fail to give the
most basic care here, which they surely owe, all the more since they do not compensate through spirit.
In bars one should play ascending broken octaves, not descending broken sevenths. In bar an artificial p is to be inserted at the fourth beat, so that a
swell (with slowing down) can then be achieved in the third and fourth beats of
bar .
Fugue. {} The tempo of the fugue, although quick, should still be as broad as is
necessary to allow the original tempo of the aria somehow to be recalled by the
subject. Above all, the rising motion permits the essential reference to the Aria to
be continued in the subject. However, it can only achieve this if it also retains
something of the tempo that is appropriate for the Aria. Nowadays this fugue,
like fugues in general, is dashed off much too quickly, without consideration
for the form in the large and the small, not to mention the details. The dynamic
markings of Brahmss fugue, by the way, can be taken as an example of how a
great master conceives the performance of fugues in general: on no account
continuously forte, but in changing colors that always follow the course of the
fugue.
The basic dynamic of the first part is fortethe p and crescendo in bars and
, like the similar directions between bars and , are only shadings within the

tonw i l l e 8 9
the third beat of bar onward, particularly in the left hand, where the motive is
found in a rhythmically oblique position. Thus the sixteenth notes are always to
be played as upbeats to the accentuated quarter notes. From bar onward, one
still seizes upon the third beats as a rhythmic halt and thus accomplishes the intensification through {} each first and third beat, as in an alla breve (consequently, all the second and fourth beats lie in the shade). In bars ff, return to
the original manner of performing the subject, with the stress only on the second
and fourth beats. The left hand should go lightly over the keys here, since it is
used by turns also in the low register, where it actually plays thirds. The direction
col Ped. does not signify anything like continuous pedaling, but only sustaining
as the occasion merits: it is thus at the discretion of the player. In bar the animation led by the left hand across several registers is to be played as such, with a
crescendo added from the fourth beat to the downbeat of the next bar. But here,
in bar , at the second sixteenth note of the second beat, an artificial p should be
inserted, to release the tension; this will set up a crescendo to the marcato of the
fourth beatall notwithstanding that bars are to be played in a torrent.
The emphasis on the quarter-note beat still applies to the motive in the left hand
at the ff of bars ff. The realization of the marcato signs in the right hand in bars
presents no technical problems; nevertheless, the performance must still
give the illusion of a certain level of difficulty: appearance stands for reality here
and serves the spiritual expression. On the other hand, in bars , , and
only the chords on the downbeat are to be played with real emphasis. The left
hand in bar is like the right hand in bars ff; in bar however, Brahms
marks only the root of the dominant with >, thus a > is to be played from the
second to the third beat here, likewise in all the descending fifths that follow as
well. In the last two bars one should also attempt to express the last falling fifth,
and with it also the falling third f 1 d1 in the left hand.

better than my earlier ones, much more practical and also more easily to be disseminated), but also for the public. At an evening at Wagners, on February ,
, in Penzing near Vienna, he played this work on the express desire of his
host. (If this variation set was no more successful in building a bridge to Wagner
than any of his other works, the fault certainly did not lie with Brahms; Wagner
had no ear for the wonder of synthesis.)41
Disagreeably, new editions of the work have appeared. Brahmss publisher himself was guilty of the first crime (see the Erluterungsausgabe of Op. , pp.
), namely, the failure to distinguish Reisenauers fingerings more clearly from
the composers.42 Since then, editors have taken pleasure in corrupting the articulation, etc., which grossly attests to the decline of the ear. In my remarks on performance, I have addressed many mistakes and infelicities on the part of the virtuosos; the enormous gulf separating them from the work is expressed there, too.

41
This, the only meeting between Brahms and Wagner, is widely reported in the literature; see, for
example, Malcolm MacDonald, Brahms (London: Dent, ), p. . Wagner is recorded as having
remarked, One sees what can still be done with the old forms in the hands of one who knows how
to deal with them. One cannot deduce from his words whether he liked or disliked the piece, still less
whether he lacked the aural faculties that Schenker is referring to in his slighting remarks.
42
The preliminary remarks to the Erluterungsausgabe of Op. includes a lengthy diatribe
about contemporary editorial practice, including the following remarks that refer to a new issue of
the original publishers edition of Brahmss Op. , with additional fingering by Alfred Reisenauer
(), a concert pianist and professor at the Leipzig Conservatory:
In front of me is a communication from the publishing house Simrock, dated December ,
, in which it is openly admitted that Reisenauers fingerings for Brahmss Op. were
used on the original plates and [that] earlier prints are no longer available. Thus, before our
very eyes, the fingering of a pianist whose close relationship with Brahms was never once attested has actually been included on the engraving plates; and the player will receive an edition in which he has no choice but to assume that the fingering, too, is by Brahms himself.
In the Erluterungsausgabe, and in his later editions of the Beethoven sonatas, Schenker used different type fonts to distinguish his own fingerings from those of the composer, in the hopes of winning respect of musicians for alcohol-free, i.e. editor-free, editions, as he describes them a little further on in these remarks.

The literature for this work, and for the piano music of the master in general, remains scanty and insignificant; even the phrasemongers, at all times eager to
help, for once have not yet ventured forth.
Brahms always held his Op. in high regard (see Kalbeck, I, ff),40 not
only for the publisher (he wrote to Breitkopf & Hrtel: I consider the work much
40
In the final complete edition of Kalbecks Johannes Brahms (see note ), the relevant passage appears in vol. , p. .

Genuine and Sham Effects

Wirkung und Effekt {Tonwille /, pp. }
t r a n s l at e d b y r o b e r t s n a r r e n b e r g

Body, time, unfolding in space and time, goal setting, the pleasure of the jour-

must go astray. He changes and distorts it, brings things in and leaves things out,
and high-handedly destroys the prerequisites of synthesis and thus converts the
effect of the artwork into its complete opposite. Unaware that the listener is included from the outset in the work of genius, in its content and compositional
style, he instead imagines the listener to be detached from the work, pitted
against it, entirely on his own, and, because he himself has sunk, he imagines the
listener is at a level so low that nothing is any longer possible: he underestimates
the person in the auditor. Then behind the geniuss back, the conductor or performer places himself in a direct connection with this listener that he has so
deeply subdued, and negotiates with him, as it were, for applause at the expense
of the work: how much applause, for example, will he give for this exquisite oboe,
this dreamy horn, or that sweet flute. He deals in sham effects [Effekt].
And then what must happen will happen. The listener is cut off from the
genius.
The genius ceases to speak to the listener: it becomes impossible for the listener to find himself again in the genius. The genius ceases to be a tongue for a
mankind eternally mute: mankind struggles for expression and culmination in
the genius.
When Beethoven once said to Schuppanzigh, Do you think I care for your
wretched fiddle when the spirit speaks to me?3 he wanted to say that what synthesis offers should be a law, to himself, to all violins, and to all Schuppanzighs.
All of themthe composer, the instrument, the virtuososhould be included
in this synthesis, must come from it, and are obliged to return to it. {}

ney, protraction of that pleasure, remembrance, repetition, contrast, expectation,

striving, fulfilling, deception, and similarly fundamental experiences of the
human soul, to name only the most essential, are also repeated in music as a
human art. Nothing human is foreign to it, only geniusand genius alone
adds synthesis. The non-genius is not familiar with the geniuss experience and
does not attain the geniuss success, not in art, not in life.
When the genius releases a work into the world of appearances, he speaks not
just to himself but also to each individual listener,1 whereas each listener, by contrast, recognizes himself in the genius and feels expressed and culminated only
through the genius. All the necessary preconditions are already given in the work,
so that nothing else remains for the conductor or performer to do in regard to
this effect. Anyone who is privy to the inner sense of the geniuss compositional
stylethe superior way in which it serves synthesis, expounds the heroic direct
line of a keys seven tones, and divulges the compositional structure right down
to its tiniest details2 will feel himself fortunate to unearth, without vanity or
self-interest, this effect of sublime, deeply ethical, and artistic reciprocity between
genius and listener; indeed, he may even flatter himself to educate mankind
about synthesis.
By contrast, anyone who misapprehends or mistrusts the compositional style
1
[S.] Cf. Beethovens neunte Sinfonie, p. vii/p. ; Tonwille , p. / I, p. ; Tonwille , p. /I,
p. .
2
[S.] Schiller (ber naive und sentimentalische Dichtung): This kind of expression, where the
term completely vanishes in what is being referred to and where speech leaves the thought which it
expresses as it were naked, while the other type can never represent it without at the same time concealing it, this is what in style [Schreibart] one calls above all inspired and the work of genius.
[Schillers Werke, ed. Ludwig Bellermann, nd ed. (Leipzig: Bibliographisches Institut, n.d.), vol. ,
p. ; On the Naive and Sentimental in Literature, trans. Helen Watanabe-OKelly (Manchester: Carcanet New Press, ), p. .]

3
Ignaz Schuppanzigh (), violinist, the foremost string quartet leader in the first quarter
of the nineteenth century. The anecdote relates to the first of the Razumovsky Quartets, Op. ,
No. , in F major, a work of exceptional difficulty for its time (). Schenker may have taken it from
Adolf Bernhard Marxs Ludwig van Beethoven: Leben und Schaffen, first published in by Otto Janke in
Berlin in . (In the second edition of , it appears in volume , on p. .)

tonw i l l e 8 9
Hence, one should play the work of a master as one finds it written. That is
the best effect of all: that is a genuine effect [Wirkung].
One should create no antithesis between genius and listener. That is the worst
effect of all: that is a sham effect [Effekt].
A genius has never written against the listener, or against conductors or performers; so let these musicians not work to the contrary!

decision and definiteness, which means they are clear and free from obscurity. A truly capable mind always knows definitely and clearly what it
is that it wants to express, whether its medium is prose, verse, or music.
Other minds are not decisive and not definite; and by this they may be
known for what they are.4
Schopenhauer

When one sees the number and variety of institutions that exist for the
purposes of education, and the vast throng of scholars and masters, one
might fancy the human race to be very much concerned about truth and
wisdom. But here, too, appearances are deceptive. The masters teach in
order to gain money, and strive, not after wisdom, but after the outward
show and reputation of it; and the scholars learn, not for the sake of
knowledge and insight, but to be able to chatter and give themselves airs.
Every thirty years a new generation comes into the worlda youngster
that knows nothing about anything, and after summarily devouring in all
haste the results of human knowledge as they have been accumulated for
thousands of years, aspires to be thought cleverer than the whole of the
past. For this purpose he goes to the University, and takes to reading
booksnew books, as being of his own age and standing. Everything he
reads must be briefly put, must be new! he is new himself. Then he falls
to and criticizes. And here I am not taking the slightest account of studies pursued for the sole object of making a living.5
Schopenhauer

4
Arthur Schopenhauer, Parerga und Paralipomena, chap. , Selbstdenken, , in Smtliche
Werke (Wiesbaden: Eberhard Brockhaus, ), vol. , p. ; translation taken from Complete Essays
of Schopenhauer, trans. T. Bailey Saunders (New York: Willey, ), vol. , pp. .

5
Parerga und Paralipomena, chap. , Ueber Gelehrsamkeit und Gelehrte, , in Smtliche
Werke, vol. . p. ; translation taken from Complete Essays of Schopenhauer, vol. , p. .

The works of all truly capable minds are distinguished by a character of

Elucidations
Erluterungen {Tonwille /, pp. }
t r a n s l at e d b y i a n b e n t

The fundamental chord in nature (Der Klang in der Natur) is a triad:

because only consonance, with its tonal spaces (as shown earlier) can, by contrast
with dissonance, promote new passing-tone progressions and freshly burgeoning
melodies. This comes about through prolongations in ever-renewing layers of
voice-leading, through diminution, through motive, through melody in the narrower sense; but all of these hark back to the initial tonal space, and to the initial
passing-tone progressions of the Urlinie. As the outcome of all these transformations and unfoldings, there emerge the harmonic degrees [Stufen]:

Because of the inherently narrow compass of the human voice, arta human
activitycan avail itself only of the shortened form of the chord of nature, which
when sounded successively defines tonal space:

The Urlinie measures out the tonal spaces within the chord, and thereby articulates the chord for the first time, bringing it to consciousness.

The Urlinie is the first passing-tone progression [erster Durchgang]. As such

it constitutes the first melody, and at the same time provides the diatonic content
[Diatonie].
There are no other tonal spaces than those of , , and . There is no
other origin for passing-tone progressions, or for melody.
The first passing-tone progression comprised by the Urlinie generates dissonance (second, fourth, seventh). Dissonance is transformed into a consonance

{} Despite the notes being sounded successively, the arpeggiation of a chord

remains a harmonic phenomenon:

Passing-tone progression, by contrast, is a melodic phenomenon. It for ever

generates dissonance, even though it can also be articulated in consonant form
by means of transformation. The same is true of the neighbor note, which derives
from passing-tone progression:

[S.] From Neue musikalische Theorien und Phantasien, vol. II Freier Satz; see also Kontrapunkt
i, the Erluterungsausgabe of Op. , and Tonwille . [These Elucidations were first published here,
reprinted in Tonwille , and reproduced with a few minor changes in Meisterwerk i and ii.]

tonw i l l e 8 9
To compose creatively from this background is to open up to oneself an
unlimited world of foreground and melody ready to be articulated through an
unlimited world of genius. In all this infinitude of genius and melody there is but
one boundary. This is the boundary drawn by Nature itself with its primary
chord, and by man with his tonal space and Urlinie. Genius is grateful for this
boundary, for it offers a necessary protection and control of freedom.
Deficiency in music is deficiency in that sense of tonal space within the individual; it can apply to the totality of society through the sum of its individuals.
In the works of its great composers, German music commands the broadest
spans of tension, the most powerful transformations within its layers of voiceleading, the most unrestrained processes of dissemination and unfolding [Auflockerungen und Ausfaltungen] in its harmonic degrees and its passing-tone progressions. These it exploits, in a compulsive quest for unity, as the articulation
of a tonal space replete with passing-tone progressions in the background:
that is to say, its synthesis, governed by necessity yet at the same time free. German melody, the true melody of music, holds the monopoly of the melody of
synthesis. The other nations, with precious few exceptions, lack the musical
power and stamina with which to engender similar long-term relationships
and tensions-spans. Their melody is an end in itself, good only for the passing
moment; no matter how sweet that moment, the melody is sterile, unripe for
synthesis.
Musicians may be divided into those who create from the background, hence
from the tonal space and the Urliniethe geniuses; and those who operate only
in the foregroundthe non-geniuses, who must therefore compose entirely in
terms of the succession of surface events, just as they hear and read in terms of
successive events. Between the two groups lies an unbridgeable chasm.
Where art is concerned, only geniuses pertain, for they bring to bear the utmost economy in feeling and creative activity. It is much simpler to generate from
a single nexus, such as , the boldest and most expansive foregrounds, than
it is to churn out foregrounds without such a nexus, foregrounds which must forever be consigned to chaos.
The art of geniuses is as simple as the simplest passing-tone progression; but
for precisely that reason it is forever inimitable and unattainable to non-geniuses.
The latter lack a bond that is for them unfathomably God-derived, the bond that
connects back to ultimate simplicity; consequently they go in a thousand different directions, always anxious because they have no sense of origin.

Neighbor-note motion on the third and its derivatives:

Other types of unfolding include conversion of a chromatic step into a diatonic

progression (Fig. a), or transformation of a vertical situation into a horizontal
one (Fig. b):1

Only genius is imbued with a sense of tonal space. It is its innate awareness,
just as the concepts of physical space (as extension of the human body) and time
(as growth and development of the body) are inborn, innate in every human as
part of the sense of their own body.
Genius alone creates from the background of tonal space, from the first
passing-tone progressions comprised by the Urlinie. While non-geniuses, whether
they work creatively {} or just for self-enjoyment, invariably run aground on
the succession of surface events in the music, the genius harnesses freedom in the
successive events at the foreground to the compulsion of the passing-note progressions in the background.
1
In one of his personal copies of Tonwille (OC Books and Pamphlets No. , Schenker penciled
a further illustration of how bf to ce may be unfolded, namely, as the succession b1 f 1 plus e1 c2.
1

Miscellanea
Vermischtes {Tonwille /, pp. }
t r a n s l at e d b y i a n b e n t a n d w i l l i a m d r a b k i n
An die Vollendung

To Perfection1

Vollendung! Vollendung!

Perfection! Perfection!

O du, der Geister heiliges Ziel!

Wann werd ich siegestrunken,
Dich umfahn2 und ewig ruhn?

O thou, sacred objective of the gods!

When shall I, flushed with victory,

embrace you and rest eternally,

Und frei und gro

Entgegenlcheln der Heerschar,
Die zahllos aus den Welten
In den Scho dir strmt?

and, freely and generously,

return the smile of the countless legions
who stream from the worlds
into thy bosom?

Ach ferne, ferne von dir!

Mein gttlichster schnster Gedanke
War, wie der Welten
Fernstes Ende, ferne von dir!

Ah, far, far short of thee!

My most divine, most beautiful thought
fell, like the farthest reaches
of the worlds, far short of thee!

Und fleugt3 auf des Sturmes Flgeln

Aeonen lang die Liebe dir zu,
Noch schmachtet sie ferne von dir,
Ach, ferne, ferne von dir!

And though on the wings of the storm

love, aeons long, flies toward thee,
still it languishes far short of thee,
ah, far, far short of thee!

Doch khner, gewaltiger,

Unaufhaltbarer immer
Fleugt durch Myriaden Aeonen
Dir zu die glhende Liebe.

Yet bolder, more powerful,

still more inexorable,
doth glowing love ever fly toward thee
through myriad aeons.

Voll hoher Einfalt,

Einfltig still und gro,
Rangen des Sieges gewi,
Rangen dir zu die Vter.

Imbued with great innocence,

innocently silent and great,
our forefathers strove, confident of victory,
strove toward thee.

Ihre Hlle verschlang die Zeit,

Verwest, zerstreut ist der Staub,
Doch rang des Sieges gewi,
Der Funke Gottes, ihr Geist dir zu.

Time devoured and rotted their mortal frame,

its dust is dispersed;
yet their spirit strove, confident
of victory, of the spark of God, to thee.

Sind sie eingegangen zu dir,

Die da lebten im Anbeginn?
Ruhen, ruhen sie nun,
Die frommen Vter?

Are they at one with thee,

those who lived there from time immemorial?
Are they now at rest, truly at rest,
our devout forefathers?

Vollendung! Vollendung!
Der Geister heiliges Ziel!
Wann werd ich siegestrunken
Dich umfahen und ewig ruhn?

Perfection! Perfection!
O thou, sacred objective of the gods!
When shall I, flushed with victory,
embrace you and rest eternally?
Hlderlin4

Schenkers quotation of Hlderlins An die Vollendung in its entirety reflects his equation of mastery with perfection and, ultimately, with immortality. In the Brahms essay in this same issue of Tonwille, he writes: Er schuf Vollendungen, that is, Brahms created perfect things; and he goes on to say
that he contributed to the immortality of art (Tonwille /, p. /II, p. . In the following year he
wrote in similar vein in the foreword to Meisterwerk: Idea, Perfection, masterwork are one concept:
by achieving perfection, the masterwork partakes of the eternal life of the idea. . . . Perfection is true
life, a true eternity (Meisterwerk i, p. /p. ).

1
Vollendung is perfection in the sense of completion, accomplishment, not in the abstract or
moral sense.
2
umfahn: an old form of umfangen.
3
fleugt: an old form of fliegt.

tonw i l l e 8 9

has {} anyone yet noticed the place specified where this passage sounds?
Has no one heeded the strict logic governing it, making it absolutely nothing otherit can be nothing otherthan the accompaniment (understood in the broadest sense) to a song:

A Letter from Dr. K. E. Neumann

Vienna XVIII, Gentzgasse , April ,

Froh, wie seine Sonnen fliegen

Durch des Himmels prchtgen Plan,

My very dear Sir,

Forgive me, if an unknown person and a mere musical amateur does not
wish to deny himself the opportunity to express his most heartfelt thanks
to you for your splendid and seminal work toward the understanding of
the Ninth. After a long, long epoch of the most shameless barbarism, you
have succeeded in enabling those who have been led so contemptibly
astray to understand and appreciate again the strict and serious art. And
in your thoroughly painstaking and equally valuable work, as destroyer
and creator at the same time, you have succeeded magnificently. Even if
the consequences and success may not follow immediately but can only
take place in the course of time, perhaps as we move into a somewhat
nobler age, you will come to enjoy the respect that you deserve as the first
to have mightily promoted the sort of renaissance we have hoped for.
As an expression of my thanks, might I nevertheless be permitted to
ask you to examine more closely a passage from your thoroughly comprehensive work? You always have presented and explained everything so
extraordinarily beautifully, carefully and precisely: but one of the most
truly splendid Beethovenian wonders, the Allegro alla marcia in B
major,
is dispatched on p. [/p. ] of your book in just a few words. And,
moreover, these words, taken from Nottebohm, are nothing more than an
unbelievably superficial misunderstanding: Turkish musicfirst pianissimoa few notes ppa few reststhen at full strength.5 Is that really
all that one can offer toward a deeper understanding of that unbelievably
magnificent, indeed magically overpowering passage? Turkish music:
that is pure nonsense, even if, formally speaking, a remotevery, very
remoteanalogy may be made. Even Richard Wagner, who himself races
over this uniquely stupendous passage, has expressed himself somewhat
better when he speaks of war-like sounds, etc. Another commentator
has thereupon detected something like a call to arms ( la Carmen). But:

Cheerfully, as its suns fly

Over Heavens magnificent plan,

That is, Beethoven, here with eternally mysterious ingenuity (or, rather,
clairvoyance), makes, so to speak, the suns dance the quadrille, in eternal
play, in conformity with clear laws, along the paths of octaves; and thus
he portrays, as it were, the law of gravity by rhythmic means: everything,
of course, held together in a single mold, in single, ingenious experience
and conception.
Therefore: this passage has nothing to do with Turkish music, in
spite of the triangle, nor with the apprentice warriors or the bugle corps.
For the chorus begins, in perfect clarity, as I have already said, with the
words: Froh, wie seine Sonnen fliegen.
Sapienti sat.6
Most respectfully yours
Dr. K. E. Neumann
To introduce the author of this letter today would be pointless: his name and
accomplishments are on everyones lips.7 If I publish the letter for the first time
he wrote it just a year after my monograph on the Ninth Symphony appeared
I do so because I am convinced that the public have a right to know what occupied and interested this much-admired and much-appreciated man outside his
main lifes work. If I remember correctly, I referred in my reply once again (as I
discussed in my book) to the necessity of a contrasting key, B
major versus D
major, that is, to the synthesis of the variations. It is not the task of man to fathom
the inspiration of a genius; this should be accepted as the genius himself intended, as a gift.
6
Sapienti sat (sufficient for him who has a clear understanding): a quotation from the Latin
comic dramatist Plautuss Persa (The Persian Woman), line ; also in Terences Phormio, line .
7
Karl Eugen Neumann (), translator of the collected sayings of Buddha, including Die
Reden des Gotamo Buddhos aus der mittleren Sammlung ( vols., ) and Die Reden des Gotamo Buddhos aus der lngeren Sammlung ( vols., ). Schenker kept a clipping about Neumann, Die Weisheit des Buddho: zum Gedchtnis Karl Eugen Neumanns, by R. M. Behrensdorf,
printed in the Deutsche allgemeine Zeitung of November , .

5
Schenker extracted this remark from Gustav Nottebohms Zweite Beethoveniana (Leipzig: RieterBiedermann, ), p. . Beethoven entered it on folio r of Autograph / (currently housed in the
Biblioteka Jagiellonska, Krakow), a late pocket sketchbook for the Ninth Symphony (winter ).

Miscellanea

in such an age there will be no time truly to come to understand a work like the
Ninth Symphony. It suffices that the work is bought and soldthis is the order
of the day in our materialistic worldand, being an antiquity, like a valuable old
piece of furniture in ones living room, is stuffed into the recesses of ones skull.
This sterile tradition has suddenly been broken by a felicitous idea of the
Wiener Konzerthaus-Gesellschaft10 which, as a way of celebrating the works centennial, reconstructed its first performance in all respects. The number of players
in the orchestra was supposed to be the same as then; the score and tempos were
to be resurrected exactly according to the masters wishes and intentions. {} In
the program booklet for the concert, the conductor, Paul von Klenau, referred to
these intentions himself.11

One Hundred Years of the Ninth Symphony

On the jubilee performance of the Wiener Konzerthaus-Gesellschaft, May ,

At one point in his Farbenlehre,

Goethe writes:

The Bible itselfand this is something we have not thought about sufficientlyhad practically no effect in olden days. Hardly had the books of
the Old Testament been gathered together than the Nation from which
they originated was completely dispersed; it was only the letter9 which
those who were dispersed gathered themselves around, and continue to
gather around. Hardly had the books of the New Testament been collected than Christendom broke apart into countless sects. And so we discover that the people did not occupy themselves so much with the work
as around the work. And they parted company over the various means of
interpretation that one could apply to the text, or foist upon the text, or
use to conceal it.

10
Wiener Konzerthaus-Gesellschaft: founded as the Wiener Konzertverein, moved to the Konzerthaus and changed its name when that building was completed in . It held a series of regular
concerts with fairly adventurous programming, and a popular Sunday afternoon series.
11
Schenker kept a copy of the concert program (OC, File B, item ), for which Klenau wrote the
following preface:
For historical reasons, Beethovens Ninth Symphony will be performed today approximately
as it may have been heard a hundred years ago. The same number of orchestral instruments
will be used as were then used (according to surviving documents). Beethovens original
scoring will be used, without any additions or modifications.
The tempos, as Beethoven himself indicated them by metronome markings, have been
determined after a careful checking of the original score. (Thus, for example, the tempo
marking Presto  on p. of the Peters edition of the full score is wrong, and results
from a printers error: it should read  . Certain performance and phrasing signs, as
they appear in the original manuscript, will be followed exactly.
If, as a result, todays performance perhaps sounds in many respects different from
what we have become accustomed to as a result of changes that have accrued over many
years, we nevertheless feel justified on this occasion in excluding as much personal interpretation as possible in order to arrive at Beethovens intentions as closely as the surviving
historical material permits us.
Vienna, May ,
Paul Klenau
The concert also included the other works that were performed at the premiere of the Ninth: the
Consecration of the House Overture, Op. , and the Kyrie, Credo and Agnus Dei from the Missa
solemnis. It was also scheduled to begin at the same time of day ( p.m.).
Paul August von Klenau ( ), a Danish conductor and composer, was a close friend of
Alban Berg and, briefly, a pupil of Arnold Schoenberg. He made his reputation mainly in Germany
and Austria, and during the s served as choral conductor of the Vienna Konzerthaus-Gesellschaft.
Like Furtwngler before him, Klenau sought out Schenkers advice on the interpretation of orchestral
and choral repertory pieces, and visited him several times between and . Letters from Klenau
during this period invariably express gratitude for Schenkers generous and perceptive musical advice.
Schenker, for his part, was initially skeptical about Klenaus capacity to follow his argument, let alone
put it into practice in the concert hall; by the time of the Ninth Symphony jubilee, however, he was
more than satisfied with the conductors results. (He also seemed happy to have been associated with

Things have gone no differently with Beethovens Ninth Symphony in the hundred years since its synthesis first resounded through humanity. In my monograph I showed the transformations it has undergone, and additionally the judgments drawn from theory and practice that have been put forward by artists,
theorists, and conductors. They all confirm that, in the past hundred years,
people have been occupiedto repeat Goethes phrasenot so much with the
work as around the work.
In recent times the Ninth, along with everything else connected with religion,
art and knowledge, has maneuvered its way into the market-place; it belongs to
the commonplace symbols of a world of film-makers, globetrotters, and international broadcasters who, though their existence affords them pleasure in bits
and pieces, have utterly forgotten that true progress lies behind them, not before
them. In an age in which intellect is given trivial assignments, even if it wants to
work as hard as the stomach, in an age in which the gramophone robs the people
of their song, the newspaper robs humanity of its Bible, its Odyssey and its Iliad:
8
Johann Wolfgang von Goethe, Zur Farbenlehre: nebst einem Hefte mit sechzehn Kupfertafeln
(Tbingen: Cotta, ).
9
Nur der Buchstabe war es, that is, only the literal meaning was important.

tonw i l l e 8 9
To determine the original make-up of the orchestra was not difficult, as the
precise details have been passed down.12 In spite of all the so-called retouchings
of Wagner, Mahler, and others,13 the first edition of the score still survives
of course, it does not reproduce the composers manuscript in all its details
without error; one can prove this to oneself by comparing it to the facsimile,
which has now been published14 nevertheless, a return to the tempos prescribed by Beethoven must have created the greatest difficulties, even insurmountable ones.
Should we perhaps be annoyed with those who heard the first performance
for not also telling us something about the tempos? Could they have thought
about that matter at all, when such a wealth of human ideas struck them for the
first time? They took in the performance as it was given to them, without asking
later whether Beethoven had succeeded in realizing his intentions. They did what
the contemporaries of geniuses had always done, and will always do: their words
mirrored above all their hearts, their feelings. In this respect the conductors job,
so far as questions of tempo arose, was an entirely new one: he knew only that the
tempo indications were the masters own,15 and was obviously determined to find
the path to him. But in this respect, help could be provided only by ascertaining
the synthesis and the meaning of the notationas understood at the deepest
levelas these bear a most intimate relationship to the tempos. The jubilee performanceI was present at it16 achieved an effect that exceeded the usual standards, as synthesis is always superior to a cobbling together. It confirmed Bee-

thovens creation in its unity of synthesis, notation, and all detailed instructions;
it succeeding in refuting the opinion of those who have moved away from Beethovens practice, believing it to be in need of their own assistance (as if Beethoven were not the practitioner of all practitioners), and who, because synthesis is
inaccessible to them and because they are after the cheapest effects, imagine that
the master inhabits the desert of a theoretically dry, non-viable performance
while they themselves are the true artists of good effect. In short, the conductors
intentions exceeded all expectations. What he still owes the work, beyond what
he succeeded in accomplishing the first time, is something that he probably
knows best himself.17 The audience not only expressed their satisfaction, they
also responded enthusiastically, perhaps without being conscious of the difference between it and other performances. What remains beyond dispute is that
the conductors efforts were not in vain: one day he will receive our thanks, when
we are on the track of Beethovens synthesis and have begun to have faith in his
metronome markings.
The stubborn weeds of the day18 deceived themselves, as everoh, they have
their traditionthey either kept silent or repeated their old errors. But there is
also the tradition of the great synthesis, to which Beethovens Ninth belongs,
which will survive their ruminations; it will triumph, even if in the course of the
centuries hundreds more Ninth Symphonies accrue.
Following on from the successful jubilee performance of the Ninth Symphony, I suggest that the good, dear city of Vienna consider whether, instead of
packing in so many festivals, one after the other, which confer neither prestige on
the city nor profit on the art, it might wish to organize solemn festivals of the Viennese classics,19 using Bayreuth or Oberammergau as a model and adhering to
the intentions of the masters as they have been precisely transmitted; the established example of performance could, in every respect, be handed down quasicanonically, from generation to generation, by an orchestra expressly comprising
and led by the very best musicians of the land. I can imagine that such worldclass art ought also to become world-class commercial enterprise, to the advantage of both art and commerce.

the event, noting in his diary that the concert program mentioned his monograph.) For further details on Klenaus relationship to Schenker, see Hellmut Federhofer, Heinrich Schenker, nach Tagebchern und Briefen (Hildesheim: Olms, ), pp. .
12
The orchestra is reported to have included twelve first and twelve second violins, ten violas, a
total of twelve cellos and basses, and doubled wind parts; the chorus was to have comprised twenty
to twenty-four singers on each of the four parts. See Alexander Wheelock Thayer, Thayers Life of Beethoven, ed. Elliot Forbes (Princeton, NJ: Princeton University Press, ), pp. .
13
Schenker had written on this subject himself, in an article entitled Beethoven-Retouchen for
the Wiener Abendpost (a supplement of the Wiener Zeitung) of January . This article is reprinted
in Heinrich Schenker als Essayist und Kritiker, ed. Hellmut Federhofer (Hildesheim: Olms, ),
pp. .
14
A facsimile edition of the Ninth Symphony autograph score was published in Leipzig in by
the firm F. Ristner and C. F. W. Siegel.
15
[S.] See Tonwille , pp. ff/I, pp. . [This section of the Miscellanea is entitled Beethovens Metronome Markings.]
16
Schenker recorded the event in his diary; see Federhofer, Heinrich Schenker, nach Tagebchern
und Briefen, p. .

17
[S.] Herr von Klenau has just recorded his initial experiences in vol. of Pult und Taktstock
(published by Universal Edition). [Die klassische Neunte appeared in July . Schenker kept a
copy of this article in his scrapbook, as it refers to his monograph on the Ninth Symphony.]
18
die Wegeriche des Tages: probably a reference to critics writing for daily newspapers.
19
Weihfestspiele der Wiener Klassiker: perhaps a dig at the pretentious terminology Wagner used
for his later operas, Bhnenfestspiel (the Ring cycle) and especially Bhnenweihfestpiel (Parsifal).

Miscellanea

conquer their original desire for vengeance and renounced their right to
retaliation, seemed to me often in so doing to gain greater advantage over
their enemy than was normally compatible with the notion of pure selfrenunciation, because the profound wisdom and intelligence jointly at
work in forgiveness led to the opponent wearing himself out and destroying himself in fruitless raging. {} It is this granting of pardon, too,
that in great historic contests augments and authenticates the superiority
of the victor after he has manfully fought out a quarrel, and attests also to
his moral attainment of manhood. Thus, sparing and raising up the defeated opponent is more a matter for secular philosophy. Genuine love of
an enemy in full possession of his powers and still wishing us harm is, on
the other hand, something that I never have seen.

No man can cultivate kind-heartedness and love without himself becoming ennobledmost spectacularly so when the recipient is identified
as an enemy or adversary. This most distinctive fundamental teaching of
Christianity struck a sympathetic chord in me since I, easily wounded and
provoked, was all the quicker to forgive and forget. And later in life, when
my mind began to back away from the doctrine of divine revelation, I became intensely curious to ascertain the extent to which this law was
merely the expression of a necessity already inherent, and recognizedly so,
in mankind. For I saw that it was obeyed in a pure and disinterested fashion by mankind only in one particular quarter, namely by those whose
own natural dispositions led them to do so. The rest, who struggled to

Gottfried Keller, Der grne Heinrich20

20
Kellers autobiographical Der grne Heinrich (Green Henry), written in and expanded in , is a Bildungsroman, a novel concerning the education and character formation of
its protagonist. Concluding part , chapter (Childhood, Elementary Theology, the School Bench),
which describes an incident in which Henrys schoolmaster chastises him by a violent physical attack,
and then accidentally puts his head through a pane of glass in front of Henry, this passage is a reflection on the line from the Lords Prayer: And forgive us our trespasses as we forgive those who trespass against us.

This page intentionally left blank

Tonwille 10
(annual volume IV, no. )

This page intentionally left blank

The Opening Chorus (First Chorale Fantasy) from Bachs St. Matthew Passion
J. S. Bach: Matthuspassion, Einleitungschor (Erste Choral-Fantasie) {Tonwille , pp. }
t r a n s l at e d b y j o s e p h d u b i e l

That Bach was guided by the image of a particular musical form, among other

The opening chorus is generally called a chorale fantasy. Unable either to per-

things, in composing the opening chorus of the St. Matthew Passionthe graph
of the Urlinie is given on p. cannot be doubted. But with respect to this
form, which of course is at the same time the content of the piece, alas, every performance is a failure, as, no less, is all the literature (discussed later). The St. Matthew Passions opening chorus itself is incontrovertible evidence that the art of
our geniuses stands too high to be the collective property of humanity in any serious sense. The common understanding of music is by its nature too feeble to
raise itself to such heights, and no illusion, no yearning, will build a bridge to
them. And to overstep the limits of nature, to underestimate Gods grace, election, vocation and occupation, gathering and deepening in the highest dedication
of an entire life, because one hopes for more blessings, more fruits, from a wider
dissemination, from the artificial gathering together of humanity and its compulsory cultivationthis is assuredly not the way that leads to the high art of genius. Those who pass off the people as the best guides to art, and who because
they confuse art with our daily bread, do not do enough themselves to prepare an
artistic meal for the people, who even flatter themselves that they are struggling
for progress, although they admit to knowing nothing in particular (and, consequently, cannot actually know whether they are really struggling for progress or
only for the rudiments of art), all the spokesmen and entrepreneurs, loudmouths
and leaders, only dilute awareness of art, reduce general responsibility, destroy
high art as well as lowin short, they damage art through the people, the people
through art. The danger threatens that an art so profoundly human and eternally
vital as that of our geniuses will become a sphinx, stared at by humanity, deaf to
its solicitations and changing interpretations. Endeavoring in all my work to
demonstrate the self-sufficient perfection, sovereign for all time, of the work of
genius, I will now snatch the opening chorus of the St. Matthew Passion away
from the sphinx-gaze to which it has been subjected for two centuries.

ceive the chorale or to understand its role, people have had recourse in their embarrassment to the broad and indulgent concept of a fantasy, buthow remarkablewithout having any idea that the concept applies here even in the strictest
sense, because the chorale really does permeate the content of the piece from the
first tone to the last.
Between the so-called chorale prelude and the chorale fantasy, as both find
their highest fulfillment in the art of J. S. Bach, a specific distinction can be
drawn. In the chorale prelude, the chorale melody is presented clearly as the main
substance of the material, in complete adherence to a single key, which the apparent departures at the fermatas {} do not contradict; the individual lines, if
not merely ornamented, are given preludes and postludes, usually with figuration.
The chorale melody is still generative of the material in the chorale fantasy, of
course, but its relationship to the other material takes a contrastingly freer form.
In order to make just one thing clearnamely how, in this opening chorus,
as in a genuine chorale fantasy, the motives in the large and the small are everywhere connected to the chorale melodyI shall in this instance omit all observations relating to voice-leading layers, harmonic progression, and so on.
At the outset I present an illustration of the chorale O Lamm Gottes1 in a
1

The text of the chorale:

O Lamm Gottes unschuldig,
am Stamm des Kreuzes geschlachtet,
allzeit gefunden duldig
wie wohl du wurdest verachtet:
All Snd hast du getragen,
sonst mten wir verzagen
Erbarm dich unser, o Jesu!

Oh innocent lamb of God,

slaughtered on the tree of the cross,
found to be meek at all times,
how much you were despised:
You have borne all sin,
otherwise we would have had to despair.
Have mercy on us, oh Jesus!

tonw i l l e 1 0

The Opening Chorus (First Chorale Fantasy) from Bachs St. Matthew Passion

The modulation to G major occurs only during the first notes of the chorale
in bar , no sooner, and the return modulation to E minor follows during the
last notes of the chorale in bars , no laterwhich is to say, specifically not
after a conclusion of the chorale in its own key of G major. The manner of modulation and return modulation is now the distinctive characteristic of the piece,
and has its own poetic meaning beyond the purely musical intention to provide a
contrasting key: in the midst of the reality of the main key, the chorale is a supernatural apparition, in effect a message from above. It is just for this reason that G
major is introduced and departed from so quietly: only a vision comes and goes
so quietly.
To recognize the motivic likeness between chorale and diminution, it is necessary merely to transpose the major-mode chorale melody into the minor.
Accordingly, the two orchestras render the melody of the first two lines of the
chorale in bars ; in bars , chorus and orchestra unite in a repetition of
the prelude with variously altered diminution.
In bars , the chorale melody is reinforced (with figuration) by the two
flutes of the first orchestra.
In bars , the two flutes of the first orchestra play a prelude to the melody
of the second line, assisted at one point by oboe and flute [recte: violin ].
In bars , various groups and instruments, different from those in bars
, take part in the reinforcement of the chorale melody.
Bars present a boldly compressed postlude to the melody of these two
lines. If this presents only the leap of a third, b1 d2, in bar (see the flutes of the
first orchestra), instead of the chorales initial leap of a fifth (see g1 d2 in bars
), and closes with the note b1 in bar , instead of with the octave g1 as the chorale
does in bar , it absolutely is not meant to do anything like contradict the chorale;
rather, the full cadence imposed by the chorales g1 in bar was to be thwarted for
the moment, which Bach accomplished by setting b1 and b2 above g1 so as to impede closure. Thus, the chorales initial fifth appears bisected in this variation.
This very return to the third makes possible yet another, summarizing,
postlude in the sequel. Only this postlude, in bars , begins at once with e3
(see the chorales e2 in bar ), as though what preceded it had as a whole expressed just d3, and finally descends to the octave g2 in bar , where it is connected to the repetition for the third and fourth lines of the chorale melody of
bars . Through its citation of the content of bars in bars , and
through the full cadence, the second postlude of course also has the effect of a
transition to the repetition.

transposition of the setting in F major found in the collection of Bachs chorales

edited by [Ludwig Christian] Erk (Edition Peters, vol. II, no. ):

To this I append an illustration of the Urlinie, without going into the justification of its specifics, only to establish that even a chorale melody cannot do
without one, if it wants to follow a secure course in the diminutions that are part
of it, as of every melody:

The basic key of the chorus is E minor, but the actual chorale in bars
is in G major. (Figure shows it at this transposition, to make it easier to recognize the correspondences and contrasts between the two realizations.)
Thus Bach has overcome the most difficult aspect of synthesis, the question
of tonalityunknown to the chorale preludeby addressing the need {} for a
contrasting key, imposed by the fantasy form, precisely with the chorale, which
he sets in G major: thus the main key receives its contrast and the chorale is presented at the same time. But this was possible only because the chorale was already contained within the main key, at least as transformed in its diminutions,
in bars !

tonw i l l e 1 0

The repetition of the chorale melody for the third and fourth lines takes up
bars .
In bars , the postlude is again intensified (compressed), as in bars
, also with the same excursion to the third, b2.
{} But in bars , instead of a second postlude, there is a presentation of
the chorales first move of a fifth, transposed to G major, carried out in the manner of bars , which by itself already produces the effect of a prelude, indeed
even the deceptive effect of dividing the entire piece into two parts. The descending series of tones that is attached to the move of a fifth (as in the first prelude) reaches only as far as b2 in this instance.
Next come preludes to the fifth and sixth line, in bars .2
The melody of the fifth line of the chorale is traced by flutes and and the
first violins of both orchestras.
In bar , flutes and of the first orchestra present a postlude to the last
notes of the line, an augmentation of which takes up all of bars .3
Something similar recurs with the sixth line.
In bar the prelude to the seventh verse begins. Here again, a presentation
of the chorale melodys initial move of a fifth: after the first one, in E minor, and
the second, in G major, now the third in the harmony of II in G major. The descending line from e3 in bars then draws the melody of the seventh line irresistibly along with it.
In bar the last note of the chorale is set down against the tonic of the main
key, not that of G major (see above), whereupon the motive of bars makes
a resolution of the entire chorus.

A comparison of this realization with those on the same chorale among the

chorale preludes for organ (see the Gesamtausgabe, vol. , or the Peters edition
of Bachs organ works, vol. V, no. , and vol. VII, no. ) will clarify the difference between the two forms.4 It will be recognized that the boldness and breadth
of the motivic extensions is much greater in the introductory chorus of the Passion than in the chorale preludes for organ, even in the extremely dramatic third
stanza of the A major prelude. Precisely this depth of variation, and the constant
play of instruments in the service of diminution, would be impediments to a
grasp of the content; no one was prepared for such broad arches, such figurations. Now is the time to learn to understand the opening chorale fantasy of the
Passion in the sense offered here. Then one might hope, further, to find the path
to Bachs art of melody and variation, and to learn to distinguish the melody of
genius, as artistic melody, as melody at all, from up-to-date proletarian melody,
which is all that the mob comprehends as melody.

Just a word about compositional elaboration [Auskomponierung]: The chorales

first leap of a fifth is elaborated in bars by the technique of reaching-over:

2
By preludes [Vorspiele] Schenker means melodic shapes that anticipate the chorale melody. It
is easy to see how the fifth line is foreshadowed by the descent from b2 to d2 in bars ; to find a
hint of the sixth line in bars , one would have to hear a continuing descent from d2 through c2 3,
extending across the immediate descent b1-a1 2-g1 2, to b2 overlapping with the actual fifth line. This
leaves questions about the music between the fifth and sixth lines, to be taken up below.
3
Schenker reads e3 d3 c 3 b2 in bar , and e2 d2 c2 b1 in bars , as transposed repetitions
of the end of the fifth line, g1 f 1 e1 d1, rather than as anticipations of the end of the sixth line, e2
d2 c2 b1; apparently the latter possibility is ruled out by the absence of an initial d3 or d2, as the case
may be. But the omission of the beginning is no obstacle to hearing a partial repetition, untransposed,
of the sixth line in bars , d2 c2 b
1 a1 which must represent the similar treatment of this
line referred to in the next paragraph. Presumably it is then the reentry of the first chorus in bar ,
with its opening material, that ends the sixth lines postlude and situates Schenkers next few fourthprogressionsfrom a1 2 to e1 2 in bars and from e2 to b1 in bars in a prelude to the seventh line, discussed in the paragraph after next.

Already, therefore, just because of the reaching-over, the ascending motive of the
upper voice must be granted thematic precedence over the descending motive;
the same holds in the sequel, despite a more or less elaborated rearrangement of
the voices in bars and as well as in bars and . But it should
also not be overlooked that, in {} bars , the entire first line of Picanders text
Bachs chorale preludes for organ on O Lamm Gottes unschuldig, BWV (from the Orgelbchlein) and BWV , were published not in volume but in volume / of the old Gesamtausgabe. (They appear in volumes IV/ and IV/ in the Neue Bach-Ausgabe.)
4

The Opening Chorus (First Chorale Fantasy) from Bachs St. Matthew Passion

is assigned to the choral basses in a significantly decisive way, and so not to the
descending motive of the sopranos.5

ing out includes two choruses with their own text and two orchestras,
while a third, monophonic chorus sings the cantus firmus. (p. )

Then what could Spitta have been hearing in bars , , , ,

and so on?

I shall refrain from devoting even a few words to the performance of the piece,
on the grounds that I have already given; but may I expect that conductors, now
initiated into the mysteries of the content, will find their own way to express the
chorale melody in all variations, preludes, and postludes.

But it is clear that the text of the opening chorus of the St. Matthew Passion could only be the product of a viewpoint that has transferred everything essential to the Passion story into the bearing of the cross and its
further course. What Bach has shaped musically is a magnificent image of
a seething crowd advancing, filled with people captivated by grief. But at
the same time he has held to the custom of opening a passion setting with
a chorale. A third, monophonic choir works the chorale O Lamm Gottes
unschuldig into the tangled fabric of the voices of the two other choirs
and the divided orchestra. Given the preference that obtained in Leipzig
for the combined effect of several choirs set in different places in the
church, it is highly probable that in this case the chorale was sung by a
treble choir stationed next to the main organ. It must be credited to Picander that he set the individual segments of the madrigalistic text in very
close conformity to the successive lines of the chorale. Thus the chorale
appears externally and internally as the power dominating the whole, and
with this the powerful musical picture was comprehensibly confined
within the boundaries of the Protestant church style. (pp. )

Spitta writes, in his study of Bach:

But twice, in the choruses at the beginning and end of the first part, he
extends the simple form of the chorale fantasy, freely admitting personal
sensibility. These two pieces are nonetheless still differentiated from one
another by the fact that the subjective element in the final chorus of the
first part is expressed more in a purely instrumental image, as opposed to
that in the introductory piece, through the two principal choral bodies,
which have their own texts. (p. )
The expression chorale fantasy occurs here, but by it Spitta refers only to the
text of the principal choral bodies, not to the preludes and postludes, to the
uniquely decisive likeness of the motives.
For even the opening chorus is, from the standpoint of its musical form,
nothing other than an arrangement of the chorale O Lamm Gottes unschuldig. Admittedly, an arrangement in which the contrapuntal work-

That Picander plays on the successive lines of the choral text in his madrigalistic textin my sketch, see the words wie ein Lamm in bars , die
Geduld in bars , and so forththis completely external trait, Spitta observes; but that, contrastingly, more vitally and specifically, the motives are connected with the chorale, flesh of its flesh, breath of its breath, this he has not noticed, and he must forfeit not only the joy of following Bach, but even the
possibility of fully appreciating this artistic offering.
{} Alfred Heuss writes, in his book on the St. Matthew Passion:7

Picanders text:
Kommt ihr Tchter, helft mir klagen, Come, ye daughters, help me mourn,
SehetWen?den Brutigam,
BeholdWhom?the bridegroom,
Seht ihnWie?als wie ein Lamm,
See himHow?like a lamb,
SehetWas?seht die Geduld,
ObserveWhat?observe his meekness,
SehtWohin?auf unsre Schuld.
LookWhere?at our guilt.
Sehet ihn aus Lieb und Huld,
Watch him, out of love and grace,
Holz zum Kreuze selber tragen.
Carry the wood for the cross himself.
6
Phillipp Spitta, Johann Sebastian Bach, nd ed. (Lepizig: Breitkopf & Hrtel, ). Page numbers
are given at the end of each quotation.

Technically speaking, this first chorus is also a chorale fantasy, like the closing chorus of the first part. But one must be wary of attributing decisive
7
Alfred Heuss, Johann Sebastian Bachs Matthuspassion (Leipzig: Breitkopf & Hrtel, ). Page
numbers are given at the end of each quotation.

tonw i l l e 1 0
significance to this fact. In this chorus the chorale is not the decisive germ
at every point, but an accessory, an addition splendid beyond all measure,
called forth by a valuable suggestion derived from the preceding Passion
and fused with the rest. The grandiose image developed by Bach in the
first choir could not have been created from the chorale text; what can be
created from that, Bach has done magnificently in his chorale fantasies for
organ. (p. )

an elemental suffering; they weigh all the more heavily on the whole. Only
gradually do they engage in the motivic fabric. ()
Heusss musical standpoint is the choir of mourners; from here he hears the
themes of the orchestral prelude, and finds them to be like those in the choral
section, but does not find the themes in both places to be like those of the lines
of the chorale. Therefore he goes wrong in his assessment of the themes and his
judgment of the chorale passage of bars as the outbreak of his powerful
lament.

The content of the opening chorus can scarcely be more strongly contradicted than by an assessment of the chorale as merely an accessory, an addition, such as Heuss offers here. That Bach has worked out the chorale much
more profoundly in the first choral fantasy of the St. Matthew Passion than in the
organ preludes, I hope he might now be able to recognize. Jesus way of the cross,
motion, agitation, the cries of the masses, choral prayer: to represent all this just
through a melody and its transformationsis such fulfillment of the whole by
one light not a much more astonishing phenomenon than the grandiose image
that Heuss abstracts from the mass effects alone, through his removal of the
chorale, the thematic sun?

In the second line of verse, Seht den Brutigam, the lament contains
something wondrously moving, the tone becomes more animated, more
open, richer in outbreaks of violent pain. The theme with the leap of a
fifth:

{} and particularly the flute figure that sounds with it:

The cry is expressed in the higher voices. The themes of the orchestral
prelude are the same as in the choral section. The first one in the soprano
is the specific theme of the cross, and must be understood as it is presented
in the orchestra in a few places, namely as chromatically descending:

which never occurs except with the word sehet, is taken directly from
the language of gesture: the daughters of Zion point to Jesus. And now the
second choir joins in with its questions. . . . Here, too, begins the chorale
O Lamm Gottes, which has already been discussed. (pp. )

The second principal theme, which appears first in the bass voices, is a
contrasting theme to the first, and works its way upward in a manner similar to that in which the upper theme sinks into the depths:

But all these individual traits are only traits of the composing-out, which flow
into the descending Urlinie series b2 e1 (like the descending series d2 g1 of the
chorale melody). About the relationship of the G major tonality of the chorale to
the main key of E minor, Heuss says nothing.
The new, brief questioning motive, divided antiphonally between the two
orchestras:

From these themes Bach develops the outbreak of his powerful lament in
the first chorus, using the words of the first line of verse. Then flutes and
oboes present only the bare outline of the theme, as though representing

The Opening Chorus (First Chorale Fantasy) from Bachs St. Matthew Passion

and therefore cannot evolve any further. It is self-explanatory that the

crescendo that arises on the basis of quite natural musical laws at the beginning therefore falls away in the repetition. (p. )

belongs to the second choirs lively Wohin; but the first choir has become quite subdued and still. Its orchestra no longer laments with it, the
chorus itself stammers its Seht auf unsere Schuld in isolated phrases.
The reason lies in the personal turn taken by the text: The consciousness
of sin depresses and subdues. (pp. )

Contrary to this, I am of the opinion that conductors would do well to follow

the preludes and postludes and to listen to what these in themselves tell clearly
enough by their scope, by the circumstances of the cadences, of the text, etc.
Albert Schweitzer writes in his book on Bach:8

Here the thematic relation of the motives in orchestra and chorus to the
melody of the fifth and sixth lines, i.e., the prelude-like aspect of the diminution,
is not recognized.

Picander imagined the introduction as an aria with chorus, exactly like

the first piece of the second part. According to his conception, the words
Kommt ihr Tchter, helft mir klagen! and Sehet . . . den Brutigam!
Seht ihn! . . . als wie ein Lamm! were to be sung by the daughters of Zion,
while the interjections Wen? . . . Wie? . . . Was? fell to the chorus. But
Bach the musician felt otherwise. He saw how Jesus was led through the
city to his crucifixion; his eye glimpsed the crowds of people who
swarmed through the streets; he heard how they called and responded to
one another. From this vision he created the introduction to his Passion
in the form of a powerful double chorus. The singular Kommt ihr
Tchter, helft mir klagen is allowed to remain, as though to bear witness
to this large-scale violation of the text. (p. )

It is quite wonderful when, in this section, the spiritual high point of the
chorus, the chorale enters with the line All Snd hast du getragen; the following passionate orchestral entry, pouring out quite suddenly in the high
register in the midst of these words, makes the most moving effect in the
entire chorus:

Most decidedly, the chorale has brought about this figure, which connects
the thought of our own sin to the one who is to atone for it. Just here, also,
the passionate cries in the orchestra join together with the sharp and
painful motive used in the manner of echo in the two orchestras, which
finds its most intensive application at this point. (p.)

Bach may very well have seen all this and heard all this, but his eye and his ear
were the chorale. Therefore Bach the musician did indeed feel otherwise, and his
vision was still more powerful, because it drew everything that was to be seen
and heard into the musically cosmic unity of the chorale melody.
Therefore it is probably not right to understand this double chorus as a
composition that represents an idealized suffering, and accordingly to
perform it toned down and in a slower tempo. It is conceived realistically,
and represents swaying, pressing, howling, and calling. What Bach writes
in the vocal parts here is no coloratura, but the presentation of the unfixable, long drawn-out rise and fall of the voices of a great crowd of people
sounding through each other. This is revealed at once in the manner of
the first soprano entrance. (Musical example.)

Here, again, the postluding is not recognized.

One might almost assume as a certainty that the orchestral introduction
begins piano and then gradually rises to a powerful forte. Because Bach
strikes up in a quite low register, it is exactly as though the mighty
mourning procession is only approaching from a distance {} and then
only gradually coming nearer. In this connection, note well how Bach
does not begin again in the low register for his repetition in the relative
major of the orchestral introduction . . . , but proceeds by maintaining
the register: The procession is now underway, moving before our eyes,

8
Albert Schweitzer, J. S. Bach (Leipzig, Breitkopf & Hrtel, ). Page numbers are given at the
end of each quotation. This is a much expanded translation of a French original, J. S. Bach: le musicien-pote (also published by Breitkopf, in ), in which shorter versions of the quoted passages
occur on pp. . Ernest Newmans English translation (London: A. & C. Black, ) is of the
German version, with some further expansion; this discussion is on pp. of the second volume.

tonw i l l e 1 0
If this analysis is right, then even the orchestral prelude is to be performed with heavily weighted accents and a certain inner unrest, so that
the bass, waiting on one pitch, and the terrible inexorability of the harmonic progression elicit an effect of increasing anxiety. This much might
hold true, in any case: that the tempo of this piece is usually taken much
too ploddingly; and that, through repeated study of this chorus, one retreats from all ingenious dynamic shadings and seeks its effect more and
more in lifelike declamation. (pp. )

In fact a livelier tempo does suit the pieceSchweitzer is right about this
but less because of the representation of swaying, pressing, howling, and calling
than, much more, because of the great arches of diminution in the preludes and
postludes, which should not be extended entirely beyond the range of hearing.
The conductor should situate himself on these arches: within them, according to
Bachs intention, all realism is concealed.

Haydn: Austrian National Anthem

Haydn: Oesterreichische Volkshymne {Tonwille , pp. }
t r a n s l at e d b y j o s e p h l u b b e n

The Austrian national anthem, which confesses to political synthesis in a deep

(see Fig. b) into the whole-step g1 a1 and the fourth-progression a1 d2, contrary
to the more common practice of coupling two third-progressions: g1 b1 d2. We
also see here how he transforms a1, which at Fig. a is still dissonant, into a fifth
(see Elucidations, Fig. ), and further, how he unifies the fourth-progression
and sets it apart by leaving the bass-note D stationary and by chromatically altering c2 to c 2.
Now comes the influx of even more numerous harmonic degrees, shown in
Fig. c: at , IIV, and with the fourth-progression, again transforming passing

tones into consonances, the circle of harmonic degrees VIVIII 3 V, expressing the background unity shown in Fig. b) by beginning and ending with V.
The diminution at Fig. d works with superelevations2 above the actual path of
the Urlinie: g1 climbs to b1, which falls again to g1; imitation yields the third-progression c2 a1, from which we can conclude that the first ascent to b merely provides an opportunity for the true falling motive. {} This is also confirmed during
the diminution of the fourth-progression by the ascent from a1 to c2 and the two
successive falling third-progressions that follow it immediately. One notices, however, that the last two notes, c 2 and d2, are stripped of such falling progressions.
Further superelevations are at work at Fig. e: in bar c2 is flipped above b1; in
bars we get even e2 d2 above the actual third-progression. The constant pressure manifested in these superelevations pushes upward to g2:

and sincere manner, is paid homage by Haydn with an incomparable artistic synthesis. Never again will such a song be sung to this nationtherefore they should
at least know what a treasure they possessed in their proud past, so that they may
forever be edified by it in a more troubled future.
Though apparently in three parts, the song is in reality two-part, as the following illustration shows:1

The ascent constitutes the first part, the descent the second part (see
Fig. a). In accordance with this Ursatz, remains diatonic, since at this point a
chromatic note is not necessary.
However, under the constraint of the text Haydn divides the fifth-progression
2
berhhungen. In this essay, Schenker uses this term frequently, and in a similar manner to the
recently coined bergreifen, to denote a raising of the register by the application of linear progressions to higher notes of the triad. (This seems especially clear in Fig. , which is specifically labeled
arpeggiation of the G triad.)

Der freie Satz contains three graphs of this piece, Figs. /, /, and /. In his personal copy
of Tonwille (OC, Books and Pamphlets ), Schenker sketched some changes onto Fig. , principally
concerning the lower voice, bringing Fig. a more closely in line with Figure /.
1

tonw i l l e 1 0
On folio recto, we see an earlier version of bars ff:

from which a compensation in the falling direction is sought. Bit by bit the superelevation is given up, so that in bar the lower sixth, b1, is reached. Now the
diminution returns once more to the fourth-progression a1 d2; it condenses the
earlier fourth-progression of bars , but inserts the diatonic c2, as opposed to
the earlier chromatic c 2, according to the laws of nature. Meanwhile, rested; it
continued to sound through the falling sixth-progression and the rising fourthprogression (see the dotted slur in Fig. e), and still asserted itself in bar as the
head of the falling Urlinie.
Now we grasp fully the purpose of the synthesis: d2 in bar is not yet the ,
rather it belongs to the motive of the superelevation. Not until the c 2 in bar
does Haydn point to the in bar here is the completion of the ascent, and
thus the fermata! The sum total, as it were, of everything that was lost by the
omission of the third-progressions to c 2 and d2 in bars (see above) is expressed in the expansive descent from g2 (compare Fig. e).
This synthesis also explains why Haydn begins the song on the third beat;
only in this way was it possible to place the harmonic degrees critical to delineation of the form on the strong beats of bars , (), , and . If one tried to
begin the song on a downbeat, half cadences and full cadences alike would fall on
upbeats. In this regard, the acceleration of in bars was decisive;
made possible by the omission of the third-progressions, it placed d2 on the correct beat of bar . This is the point about which Haydn may well have organized
both the beginning and ending of the song.
The thrust of the ascent, which is twice restrained at , succeeds all the more
powerfully in the following fourth-progression, . This force crowns itself with
the peak-tone g2, and also with the justness of the descent, so that in honor of the
great singer one may transform the proud words of the Emperor, justitia regnorum fundamentum, into the equally proud words of the artist, justitia artis fundamentum.3

{} but the final version is already visible on the penultimate line of the same
page; its superior value abides in the following:
The (a1), which in bars and is expressed far less fully than , is clearly expressed in bar by the significant expansion that is lacking at that point in
the sketch; moreover, c 2 occurs in place of c2, and with it the stronger statement
of the fourth-progression a1 d2; finally, the division of the falling sixth-progression g2-b1.
On folios the composition is written down in score for voice and orchestra. In this version the song is preceded by a G major chord, to which
Haydns NB refers: The very first note is played only at the beginning, in order
to give the pitch to the people. The distribution of the orchestra is: trumpets
in C, timpani, horns in G, oboe , oboe , flute, bassoon, violin , violin , viola,
voice, basses. The following are significant as striking passages of instrumentation: the placement of the horns on the downbeat of bar with the octave d1 d2
that they hold until the third beat of bar , in order to fuse together the groups
formed by bars and ; and similarly, the placement of the timpani on the
downbeat of bar (in sixteenth notes), so that the last group, bars , is
strongly bound to its repetition. In addition, the score exhibits the most careful
dynamic markings.4
On folios are written the variations with which we are familiar from the
C major string quartet with the Emperor Hymn (Op. , No. ).
To the volume is also appended the first edition, in which Haydns phrase
markings have already been misinterpreted.
The imprimatur is from January , , and the first edition bears the remark: Sung for the first time on February , .5

n the Austrian National Library in Vienna (the former Court Library), several
autograph leaves relevant to the Emperor Hymn are preserved in a single volume.
4
A modern edition of this score is published in H. C. Robbins Landon, Haydn: Chronicle and
Works, vol. (London: Thames & Hudson, ), pp. ; a facsimile of the first page of Haydns
autograph appears on p. .
5
The imprimatur, or authority to print the song in its final form, came from Count Joseph Franz
Saurau, president of Lower Austria. February was the date of the emperors birthday.

3
The Latin phrases may be translated as justness is the fundamental principle of (imperial) rule
and justness is the fundamental principle of art.

Schuberts Impromptu, D. (Op. ), No.

Quatre Impromptus, op. , Nr. {Tonwille , pp. }
t r a n s l at e d b y j o s e p h l u b b e n

The piece is in three-part song form:

a1
b
a2

The peculiar form of the first section can only be grasped by means of the Urlinie and its voice-leading prolongations (Fig. ).
The Urlinie descends in the space of a third, bg, in bars (see Fig. a).
The path through the third b1 a1 g1, which concludes in bar , is merely an elaboration [Auskomponierung] of . Not until bars , and their repetition in bars
, does arrive. Thus the a1 section concludes not with the cadence in
bar , but with the one in bar .
Of course, the third-progression in bars itself breaks off initially at a1 in
bar , in the manner of an antecedent (IV), and only the new linear progression
that strikes out from b1 leads conclusively to g1 in bar , with the effect of a consequent (IIVI). One observes clearly in this example what the Urlinie means
to a composer. If Schubert had sensed actual closure as early as bar , it would

bars
bars
bars

[S.] Schuberts op. has become known by the title Quatre Impromptus. His autograph does not
bear this title; further, it has No. in G
major, not in G major, and has a time signature of four half
notes to the bar (or four-two time: , not ) or . Only the collected edition by Breitkopf & Hrtel
[Franz Schuberts Werke: kritisch durchgesehene Gesamtausgabe (Leipzig: )] reproduces the text
of the autograph; all other editions have accepted the arbitrary alterations of the publisher [of the first
edition (Vienna, )], Tobias Haslinger. In view of this deplorable situation, I am compelled to base
my analysis on the accessible version of the piece. Compare the critical report to the collected edition,
series XI (edited by Julius Epstein and Eusebius Mandyczewski), p. .

tonw i l l e 1 0

have been impossible for him to set the subsequent group (bars ) as if it still
belonged to the first section.
What is also special about the path of the Urlinie in the a1 section is that it
lacks the neighbor note that usually appears in such cases. But the masters intuition is also aware of this, for the middle section (bars ) sets out precisely to achieve the neighbor note c2. Since (c2) did not come about through
, it should at least appear as a neighboring elevation [berhhung] of . Harmonic degrees of the principal key serve the neighbor-note structure in the b
section: VIIIVI (see bars , , and ).
In Fig b we see at the very beginning, in bars , a falling third-progression
as the first elaboration [Ausdeutung] of . It prepares the next third-progression,
bars , which likewise only serves the (not to diminish the importance of

the more numerous harmonic degrees that arise from the increased activity of
the lower voice).
The unfolding of , beginning in bar , is even bolder. This Urlinie-tone is
placed above V, which, according to the Ursatz in Fig. a, descends to I. However,
the fifth-progression is divided in two: D to B, the halfway point, in bars ,
and B to G in bars . The passing bass note C in the first space of a third may
well have incited the upper voice to push upward to e2, whereby a tenth above C
is achieved in the outer-voice setting in bar , {} likewise as a passing tone. A
second tenth, d2, follows in bar . Now if the upper voice had fallen from d2 to g1
at the same time as the lower voice fell from B to G, it would have yielded consecutive octaves, had not the lower voice, deferring to the upper, moved in contrary motion. All of these unfoldings create the impression that there are two ca-

Schuberts Impromptu, D. (Op. ), No.

dences, VIIIVI, while according to Fig a there is only one falling fifth. (If
one wanted to assign the status of an Urlinie to the upper voice throughout bars
, it would have to ascend from to , and then fall to ; to be sure, d2 could
be considered, by abbreviation, to be right at the outset in bar , whereby it
would follow that is a neighbor notesee the dotted slurs in Fig. b.)
In the b-section, the neighboring inflection (see Fig. a) is divided in two:
bars and . The fact that each tone is supported by its own chord in
both groups of bars necessitates the admission even of an autonomous key, E
minor (instead of VI in G major), and a return modulation to G major.
The precise path which the lower voice had to take in bars , in order to
provide the upper voice with lower thirds, is shown in Fig. c: given the upper
voice d2 e2 in bars , the inner voice b1 c2 is prescribed by the diatonic
systema half step, which as an applied leading-tone motion requires the appropriate falling fifth GC (see Kontrapunkt ii, pp. ff/pp. ). For the sake
of parallelism, the half-step g 1 a1 instead of the diatonic succession g1 a1
and the corresponding falling fifth EA had to be installed previously in bars
. Because of this, however, the danger of consecutive fifths would have arisen,
insofar as a fifth in the outer-voice setting of bar was indispensable, and would
have to be imitated in bar .
In Fig. d, one can observe the elimination of these consecutive fifths through
contrary motion in the lower voice. The unfolding of the middle section here becomes more diversified: it is as if the first falling second, b1 a1, in bars had
stimulated the composers fantasy toward imitation. Similar seconds are immediately placed in the inner voice in bars , twice over the divider at the upper
fifth, and the third time over an ascending third, EG, or, what amounts to the

same thing, a descending sixth (the G major chord expresses III 3, which serves
the tonicization of VI in bar see Fig. c). The seconds in the upper voice in
bars are counterpointed by the dividers at the upper fifth in the lower
voice, the seconds in bars by other seconds.
Now to the unfolding in Fig. e. The richer path of the lower voice in bars
leads to the effect of harmonic degrees, but the more background voice-leading
level, Fig. b, clearly demonstrates that the circle of harmonic degrees VI
IVVI is nothing more than VI.
The falling fifth-progression of the upper voice in bars and
presages the sixth-progression of bars ff in the middle section. Everything that
in Fig. d had entered in the form of a second is unfolded in sixth-progressions,
once again providing an opportunity for harmonic degrees. The weight of the

sixth-progressions, additionally fortified with harmonic degrees, requires the assumption of a modulation to C major in bars ff, instead of the progression to
VI in E minor shown in Fig. b. The elaboration of the dominant of the new key
is especially bold: first, in bars , the upper voice presents a fourth-progression; the lower voice overlaps this in bar with a fifth-progression, in the middle
of which the upper voice steps in again in bar with a fifth-progression (see the
graph of the Urlinie, p. ). The significance of these linear progressions is
matched by the harmonic degrees of these two cadential cycles. The new key is
expressly confirmed in bars , and specifically by a cadence {} in bars


, I IV 3 VI, which is repeated in the following four bars (except that this
3
time IV is used). The leaps of a sixth in the lower voice of bars and
, which are inversions of the preceding leaps of a third, point the lower voice
toward the leap of a third to A in bar what power there is in the parallelisms!
In bars the play of linear progressions relates to bars ff: there, sixthprogressions, here, diminished-seventh progressions. The richer motion demands
recognition of harmonic degrees. From bar to bar , a cadence unfolds in the
key of E major (the result of mixture), and here again, just as was the case with the
cadence in C major (bars ), it is achieved by means of interlocking fifthprogressions. The two cadences, bars and bars , are related by parallelismyet another reason to accept a modulation to C major in the former case.
In bar , a reinterpretation of I in E minor as VI in G major brings about
the return modulation, which also completes the Ursatz of Fig. a.
Regarding the final realization, the following points should be noted. In
bars , the lower voice carries a leap of a third (IVI), and this event in the low
register immediately elicits a similar one in the high register, the leap of a third
b1 g1 in bars . Thus is the kingdom of breathtaking synthesis established.
In the upper voice of bars , there is a second leap of a third, a1 f 1, this time
decorated by an accented passing tone; since the lower voice has meanwhile completed the fifth-progression with a second leap of a third in bars , the upper
voice, too, completes the fifth-progression, a1 d1, with the third-progression f 1
d1 in bar . Now it is the turn of the lower voice once again, with the fifth-progression DBG in bars . From the nature of the fifth-progression, it follows
that I should be assumed only after the root appears in the second half of bar ,
yet by this point passing tones are already slipping away above it (see the graph
of the Urlinie). In bars , there are fifth-progressions in the upper and lower
voices, the outer voices are separated by the interval of a third (tenth), and there
are chromatic notes in the inner voice (see bars and ); moreover, in bar

tonw i l l e 1 0
progressions in bars and and thus resolves the leading tone b1 to c1 instead
of c2. The emphasis on g1 offers greater advantages for expressing the dominant
and the modulation than would the sustaining of b1, all the more since the descending register transfer, b1 c1, by itself implies the ascending step b1 c2 (see the
graph of the Urlinie). The new path is so secure that Schubert, in his realization
of the final fifth-progression in bars , can even risk a grand melodic arch
in the upper voice, bars , without having to fear that the fifth-progression
will be subject to misunderstanding. The grand arch then demands, to be sure,
imitation by a smaller one, bar . The upper voice reaches the actual register of
the Urline (c2) through an ascending arpeggiation in bar .
In bars , the path through the diminished seventh (upper and lower
voices in contrary motion) is richly decorated with accented and unaccented
passing tones. In bars the accented passing tones are transformed into suspension formations. Also in bar , b1 appears in the last eighth and picks up
where c2 left off in bar (see the graph of the Urlinie), so that the unfolding of
the seventh c2 d 1 in bars , just like the sixth-progressions in bars
and , once again proves merely to be a linear progression into an inner
voice. In bar , however, b1 is missing; this note is omitted here, because in any
event in bar it appears at the head of the diminution.
The upper voice descends in bar all the way to the upper octave of the
chord, e1, but it swings back up to c2 right away in bars , from which it follows that the fifth, b1, must be supplied by the Urlinie in bar .
The repetition of the first section begins in bar , exhibiting some modifications. Bars state an exact repetition of bars . By contrast, bars
correspond not to bars , rather to bars . Bars correspond to
bars . After bars stand in for bars , the composition finally returns in bars to the path of bars . Thus simply for the sake of variety, bars are thoroughly scrambled in the repetition. It is important here,
however, that Schubert in bars initially returns to the intensified version,
bars : this follows from the precept of synthesis whereby the repetition of
a parallel structure should commence with the second, already developed version.1 Only the parallelism of bars and {} bars remains in the second section as it was in the first (see bars and bars ).
The coda begins in bar : the Urlinie circles the leading tone four times (g1
1

f ), before finally coming to rest on , beginning in bar .

a neighbor note substitutes for the passing tone in the lower voice. The fifthprogression in the upper voice, however, should be conceived only as motion into
the inner voice: b1 moves to a1 (bar ), with a detour to e1 in the inner voice.
Again in bars , the upper voice leaps (from b1 to g1), only after the lower
voice has led the way with a third progression (compare bars ). The in bars
is decorated by a suspension.
This suspension becomes an ornament in the following bars (see d2 c2 in bar
, f 2 e2 in bar ). In bars , the originally unaccented passing tone (see the
graph of the Urlinie) is transformed into an accented one, clouding the effect of

I 7. Thus this linear progression demonstrates that in the composers fantasy the
initial voice-leading in Fig a and b played a greater role than did the harmonic
degrees that maintain the foreground realization. The diminution in bars ,
the fifth-progression e2 a1, is conceived and deployed as a prototype of the next
one, the fifth-progression d2 g1, that is unfurled broadly with stepwise motion
in bars ! In bars , there is an unfolding of thirds (see Elucidations,
Fig. ): the lower notes g1 f 1 lead, the upper notes b1 a1 follow in the second half
of bar . The repetition in bars adds significant decorative suspensions
and trills, a filling-in of passing tones in bars , colorful mixture in bar
(e2 e
2), chromatic notes in the inner and lower voices in bar , and so on.
In the bsection, the melodic detour of bar and the series of suspensions
are diminutions of the sixth-progression. The composers consciousness of this
expresses itself clearly in the retaking {} of b1 in the last eighth of bar , which,
in addition to its effect as an anticipation, may very well signify the expression of
a b1 that remains in the background of bars (see Figs. d and e, and the
graph of the Urlinie), beneath which the inner voice descends to d 1. The leap f 1
a1 in the upbeat of bar expressly prepares the leap of a sixth d 1 b1 in the upbeat of bar . The repetition in bars ff. exhibits minor alterations: a neighbornote inflection in bars (contrast this with bars ) gives rise to another
neighbor-note inflection b1 c2 b1 in bar (contrast this with bar ). The leap
f 1 c2 in the upbeat of bar (compare bar ) motivates the leap d 1 b1 in the
upbeat of bar , again confirming that the register of b1 remains in force in this
group of bars as well. After bar in the final realization, b1 gradually relinquishes its role as the primary tone of the sixth-progression b1 d1 (see Fig. d and
the graph of the Urlinie). Another leap to the leading tone, b1, as in bars and
, would have placed the upper voice on c2 in bar ; Schubert, however, indulges himself in the boldness of a registral exchange: he moves from b1 to g1 as
the upper octave of the dominantit is also the fulcrum of the fourth- and fifth-

[S.] See, for example, Chopins Etude in C minor, Op. , No. , bars and .

Schuberts Impromptu, D. (Op. ), No.

The diminution in the coda avails itself of a fifth-progression in the lower

voice, which indulges in mixture with an E
in bar . The upper voice leaps
down to c 1 in bar , a note that belongs to the inner voice (here as part of IV;
see the graph of the Urlinie). The upper voice does not arrive at f 1, the Urlinies
prescribed leading tone, until the last quarter of bar after a detour through
the inner voice (c 1 d1) and a descending arpeggiation from c2. On the one hand,
the upper voice keeps the chromatic c 1 (bar ) and the diatonic seventh c2 (bar
) separate from each other, each in its own register; on the other hand it binds
them together, which is exactly what the charm of the diminution consists in. In
bars ff, the repetition of the fifth-progression is reinforced with new linear
progressions. The neighbor note a
1 in bar is exceptionally expressiveit derives from the neighbor note a
1 in bar , which in turn has precursors in bars
and bars !for its sake, Schubert leads the lower voice in bars
through E
D
C
, as if it were tonicizing A
with a falling fifth E
A
. In bar
, a diatonic C would also have been possible in the lower voice, but the master
must have felt that, on account of the unusual leap a
1 c 1 in the upper voice of
bars (contrast this with the leap of a diminished fifth in bars ), the
lower voice too could and should risk something unusual. He therefore uses C
instead of C. It is as if the outer-voice setting were overflowing its banks: a
1 beyond
g1 in the upper voice, and C
beyond C in the lower voice! The powerful intensification in bars (compared to bars ) can only be heard correctly if the
ear perceives it at the same time against the norms of the diatonic system. In bar
, the leading tone is again squeezed into the last quarter, as in bar . In bar ,
the inner voices d1 e1 d1 imitates the neighbor-note motion of bar .

melodic line should remain at the prescribed pp, but without lapsing into a sotto
voce, which would be inadmissible here.
The law of performance governing note repetitions in general demands that
the two half notes of bar be directed toward the whole note of bar , with the
sort of acceleration that the accentuation in bar makes obvious. The changes of
pulse in bars in this way hold the key to the performance of the entire piece.
All is lost if b1 in bar is more strongly emphasized than b1 in bar , and if bar
is not directed towards bar in the manner of an upbeat. Precision with respect
to the first eighth in the third quarter of bars , , , , etc. is helpful to the alla
breve, as is also a smooth, unhurried presentation of the [triplet] eighths. The
legato from b1 to g1 in bars should be {} supported by a legato of the pedal
as well, not to be released until the second quarter of bar . Beginning in the third
quarter, the pedal should be depressed again. Schuberts marking < > in bar
demands a movement in bars and whose goal is b1 in bar thus one should
not get lost in the quarters of bar . On the last quarter of bar , one should hesitate a little. In the consequent phrase, b1 in bar should be accented, as it was in
bar , and then likewise a1 in bar . By letting the quarter notes hover in bar ,
one appropriately prepares the following whole note, a1.
Beginning in bar , the bass participates more, and it should become more
colorful. In bar , it is sufficient to consider applying a touch for d2 that would
even allow the finger to leave the key, and to play a1 on the second quarter with
the fourth finger in such a way that the eighth note would continue to sound, like
a quarter. Yet it is also important that, following the immediate release of pressure, the first two accompanying eighth notes, d1 and e1, should be played with
slight hesitation; the difference should be made up by an acceleration in the third
quarter. At the anticipation on the fourth quarter, the pedal should be lifted, and
not depressed again until bar likewise in bars . In bar , the chord in
the left hand requires a marcato attack, in spite of the pp. The right hands e2,
however, should be presented as if through a change of voicingveiled, to be
sure, but brought out as if by a >. In bar , the eighth notes in the first quarter
should hesitate, and therefore the remaining quarters should accelerate towards
the whole note in bar , which once again bears a >. Bar should be played in
the same way as bar . In bar , a > should be applied on the third and fourth
quarters. The free manner of playing that I demanded in bar is also required
in bar , all the more on account of the grand arch in the right hand; the fifth
finger should thus leave the key, provided that the pressure is sufficient, and migrate to b1. In any event, the first two eighth notes must be delayed, and the eighth

Schuberts time signature of   points the way to the correct performance. Accordingly, one should not simply announce one note after the other; rather, one
should lead toward and retreat from significant notes. This results in spoken
melody, or sung speech. The player must first find the courage to confront a bar
comprising four half-noteshe will then be able to reproduce correctly the melodic arches.
The melodic line must be prominent, penetrating in tone, floating clearly
above the dark accompaniment of the right and left hands. If the marcato chords
and octaves of the left hand are kept entirely in the background, completely colorless and indifferent, the song will rise above it all the more beautifully. The

tonw i l l e 1 0
notes in the third quarter accelerated accordingly. Across bars |, the passing
G in the bass should slip by almost unnoticed. Although it could have been a
quarter note, it was kept as an eighth note; this abbreviation will not at all tolerate the clarity and brightness of conscious intent, but should instead be performed as if in a trance. The left hand in the third quarter should exchange the
fifth for the third finger, in order to express the third span AGF with ;
the hand should remain angled steeply over the keys, while the fourth finger
lightly floats across G. The left hand in bars also has to take care to use the
appropriate fingering in order to express the fifth-progression: thus in bar , on
i
F, , so that the fifth finger prepares for C in bar . On the way, to be sure, it
would be advisable to exchange fingers during the trill: ; the grace notes
i
maintain , on C in bar .
The dotted half notes that are marked sf in bars , , , , , , etc.,
should be played only as halves, if they are provided with sufficient pressure; the
hand, freed thus on the third quarter, should be lifted up, so that falling lightly,
piano, on the triplets of the fourth quarter, it performs them as an upbeat to the
following downbeat. Because of the leap in the right hand in bar , the first
eighth in the left hands triplet should hesitate slightly, while the others hurry forward in equal measure. In bars and , the final eighth notes, f 1 and f 1, should
be played with the fourth finger, not with the second or third.2 In bar , while
the melody rests on g1, the passing D in the bass should be underscored slightly,
despite the pp; when the pressure [on this note] is released, the broadly arching
melody in the right hand will descend all the more calmly and comfortably. The
leap of a sixth g1 e2 across bars | wants to be rendered as if with a singers or
string players portamento: the hand is lifted, {} it strokes the air in an arc (mimicking a glissando), and finally plays e2 as tenderly as possible. In bar , the first
quarter note hesitates, the remaining ones move forward more rapidly. The same
holds true in bar , despite the closing-off of the section, so that the last quarter
note, d1, moves without pause toward c1 in bar . The sectional division is not to
be expressed by the usual ritardando at the closing cadence; rather, an artistic
means must be used to link the sections togetherhere it is the task of the left
hand, which strikes the opening note of the new section underneath the concluding fifth-progression in the right hand. In bars , the left hand must be
aware of the opposition of CA
1 and CA1; in spite of the four bars separating

them, the modal mixture can be expressed quite well through parallel movements and a corresponding touch. The left hand should remain on C for the entire duration of bar . This not only facilitates the sixteenth-note run, it also correctly interprets the sixteenths as nothing more than the means of executing a
glissando in the leap of a sixth. In bars , the accented passing tones should
be strong, while the dotted half notes that follow them should be maintained at
p. Such a performance can only be achieved though constant practice and a precise conception of the effect; it is consistent with the performance practice of the
best singing artists, and entirely appropriate for Schubert, the singer of all singers.
In bars , the masters indication < > serves as the most effective
means of synthetically binding the accumulated appoggiaturas together, thereby
avoiding the danger that the numerous figures might appear too disparate. In bars
, the bass actually remains on B1, and the eighth notes in the third and fourth
quarters of bar should not counteract this impression. Therefore, the left hand
should be held erect on the first eighth, and the figure, beginning with the second
eighth, should be played lightly, almost as if one did not intend to produce actual
sounds. The fingering suggested for the eighth-note group is , and for
i
B1 in the next bar, . In bar , the passing half note A1 (because of the stationary d 1) should, as in bar , be slightly underscored. The performance of bars
should be like that of bars . In bar , the left hand should linger a
little on the first eighth of the triplet, then accelerate through the remaining two.
In bar , the second eighth note in the third quarter should stand out slightly,
and then recede in the acceleration of the last triplet in the fourth quarter. In bar
, one must be careful to hold a1 as well as c2 under the fermata.
In bar , the right hand should strike c 1 with the third finger, because of the
sf, but then exchange it for the fifth. In the last quarter of this bar there should be
a slight hesitation, providing the opportunity for c 1 to ring out. In bar , there
should be a hesitation in the first quarter, and a resumption of motion in the remaining quarter notes (compare bar , etc.). Bar should be played the same
as bar . In bars and , Schuberts prescribed > demands a slight push
towards the third quarter. In bar , there should be a pause, like a fermata, at e2:
the breath should freely indulge itself thereinin the last two quarters, the original tempo should be regained. The final bars conclude in a comfortably animated way. There should be no pedal in the last quarter of bar .

2
[S.] Compare my fingerings for Beethovens Sonata in C  minor [Op. , No. ], first movement,
bar .

Schuberts Moment musical in F Minor, D. (Op. ), No.

Schubert: Impromptu, F-moll, op. , Nr. { }
t r a n s l at e d b y j o s e p h l u b b e n

The piece has a ternary form: a , bars ; b, the middle section in A
major,
1

more elegant when the upper voice now leaps down the same path even more
quickly (see the dotted slurs in bars of the graph of the Urlinie, p. ). In the
second eighth of bar , a
1 ( ) has already been retaken, and in the last eighth, g1
( ). Now the law of obligatory register serves notice of its demands, and f 2 ( ) is
brought in by an ascending arpeggio in bar . At the same time, the inner voice
in bars repeats the basic motive, for the second time and in augmentation.
In bar , the two neighbors, b
1 and b
2, are struck simultaneously; in bar the
inner voice leads, in bar the upper. It is understood, however, that in bars
both voices want to take the same path.
In the second quarter of bar the F minor chord is merely passing; it is not
yet I. Had the root of IV actually been sustained through the second quarter, a
1
would have created a seventh in the inner voice (IV8-7 V6 5); now in order to
convert this dissonance into a consonance (see Elucidations, Fig. ), the root F
is inserted, which indeed makes a
1 into a third but nonetheless shares the passing character of the upper voices f 2.

bars ; a2, bars ; coda: bars .

Bars ff. The first section occupies the tonal space f 1 a
1, through which the Urlinie descends. A neighbor note, , is inserted in bar .
The is set chordally and circumscribed by a motive of diminution (bars
) that appears similar to a turn.

The beautiful blossoms of all the diminutions are contained, as in a bud, in this
very turn figure; it even provides impetus for the Urlinie ( )truly
a secret of becoming that only the organically creating fantasy of a genius has the
power to bring forth, to express.
Now to the particulars of the unfolding. It proceeds from a
1 to f 2 by means
of an ascending register transfer with reaching-over technique in bars : the
upper voice reaches over with c2 in bar , then with f 2 in bar ; the inner voice repeats the first motive in bars . The leap of a fourth c2 f 2, together with the division in bar of the falling fifth-progression to IV by the third, D
, prompts a
second leap of a fourth, through which b
2 is finally reached in bar .
In this sort of ascending register transfer there is a great deal of elegance and
humorthe persistence of the first motive as a microcosm set against the macrocosm of the ever-increasing range of the upper voiceand the effect is even

Bars ff. The middle part also devotes itself to the same a
that in the first part
had borne ; here, however, this note is not merely a III, rather, nothing less than
the key of A
major {} is granted autonomy by the elaboration of an octave line.
The diminution draws its magic from a deeply hidden background. There is at
first an augmentation of the basic motive in bars (see the graph of the Urlinie)who can hear in this motive in A
major, speaking with all the tongues of
note repetition, its origins in the first F minor one?then two leaps of a third
and fourth in bars (e
2 as a short grace note) up to a
2 ( ), imitating the
leaps in the first part. How uniquely beautiful is the effect, when in bar the
upper voice now recedes as an inner voice and repeats its motive beneath the Urlinie in bars , just as it did in the first section (see the graph of the Urlinie);
thus in bars the inner voice simulates an upper voice! The Urlinies fourthprogression fashions for itself in the middle section an independent, repeated

1
The piece belongs to a set of six Momens musicals (now orthographically regularized to Moments
musicaux), published by M. J. Leidesdorf in Vienna in , the year of Schuberts death. But it first
appeared in print in December , with the title Air russe, in the first of a two-volume Album musicale brought out by the same publisher (under the name Sauer & Leidesdorf). The label impromptu in the title of Schenkers essay is a mistake.

tonw i l l e 1 0

subsection. Because it occupies the two-line octave in the middle section, a
2 g2

was suppressed in bar in favor of a
1 g1: the register should feel new. How fortunate this play of registers turns out to be for synthesis! Bars provide the
completion, . In bars , the neighbor note f 2 ( ) is inserted, approached

from the III 3. The half-step e2 f 2 in bars may very well be related to c2 d
2
in bars and . Be that as it may, the third-progression a
2 g2 f 2 takes off
above f 2 (which has already been reached by the downbeat of bar ). It alludes
to the just-concluded fourth-progression of the Urlinie. Howeverand this is
what is crucialthe altered harmonic interpretation confirms that this series of
notes no longer has anything to do with . Bars repeat bars .
The root of VI, F, is reserved however for the upbeat of bar (thus the position

in bar , and on the downbeat of bar ). This provides C, the root of III , with
a lot of momentum, which then sweeps along with it the falling fifth-progression
of the cadence. Beginning in bar , the most beautiful unfoldings of a third appear in the upper voice (see Elucidations, Fig. ): ascending in bar , descending in bar , and both ascending and descending in bars . Moreover, the
way that the Urlinie, seeking closure, unites with the inner voice that carries the
first motive (compare bars and ) is truly astounding, exhibiting complete harmony of synthesis. Across bars | Schubert inverts the leap of a third
in the bass into a leap of a sixth. He does this once more in bar , whereby he
reaches the root of the dominant in the great octave in bar . This depth is
sought out in order to provide the A
chord in the sum of bars with the requisite chordal completion.

Bars ff. Bars repeat bars . The last bar of the group is placed in the
service of the coda, as its first bar (see the graph of the Urlinie). The Urlinie traverses the space of a third ascending and descending, in two groups, bars
: , and bars : ; it is as if it wants, after the fact, to regain by
means of the ascending that which it forfeited at the beginning because of
the initial chordal placement of . In bar , the notes in the upper voices appear
already as passing tones above the root of IV. The g2 of the upper voice ( of the
Urlinie) is thus an accented passing tone, arising from the abbreviation of .
In bar , the minor mode is left behind, and mixture yields the major mode; a2
enters in place of a
2, and, within a phrase expansion (bars ), the cadence
concludes in F major.
In bar , is decorated by the leading tone, in the manner of a neighbor
note: . The following sketches provide information concerning the path of
the upper voice, which implies the unfolding of a two-voice setting (see Elucidations, Fig. ):

{} The sixth af , reached by the upper voices on the downbeat of bar , inaugurates a short two-voice passage (see Fig. a). Given the consistently descending
arpeggios, the unfolding (Fig. b) is unmelodic, on account of the leaps to e2 and
1

Schuberts Moment musical in F Minor, D. (Op. ), No.

f 2, In Fig. c this drawback is overcome by means of the melodic steps a1 b
1 and

e2 f 2 (seen in the light of Fig a, this sketch shows that the upper voice descends
to the inner voice and then returns to the upper voice). In Fig. d, a descending
register transfer comes into play (see the dotted slurs and the graph of the Urlinie).
The arpeggiation of a sixth in bars is thus stretched across the frame of the
downbeat sixth in bar ; it is this very diminution that leads down to a1 and (according to Fig b) makes the stepwise motion across bars necessary. The
diminished fifth in bars imitates the earlier arpeggiation of a sixth.

thereby signaling the transformation with respect to bar : the short grace note
becomes an eighth. In bar , there should be an acceleration in the last eighth, expressing surprise at the ascending register transfer. The quarters in bar should
be performed as in bars and . Beginning in bar , the tone should be fuller and
more open; but the performance of the quarters should remain as before. In bar
, the first sixteenth should be given a light emphasis, on account of the suspension; for the same reason, the second quarter of bar should be played quietly.
The suspensions in each first eighth of bars and should also have a quieter
release. In bar , one should take into account that the first and third eighths in
the bass belong to the lower voice, while the second and fourth bear only accompanying inner voices. Thus the former are to be played more forcefully (i.e., fp
fp). In bars and the suspensions should be brought out. In bar , the
raised note in the bass and the upper voices b
2 should be emphasized (the diminished octave of figured-bass theory); Schubert expressly indicates this with
< >. This bar should conclude unhurriedly, even hesitantly. Its motion
should not spill over into the following bars, because the surprising inversion of
voices in bars demands an independent, accelerated movement. Beginning
in bar , the tempo of bars and should be resumed. Specifically, bars ,
which represent an exact parallel to bars and , must proceed without pause;
and only just before the closing chord should there be a short breath, to prepare
a marcato attack.

he tempo marking, Allegro moderato, is determined by the marcato quarters

in bars , , etc., which are indeed determinative for the entire performance of
the piece. One should always preserve a comfortable distance between the two
quarters (moderato). While the appropriate touch leaves the hand on the keys
for only an eighth, it nonetheless gives the effect of a quarter. Unlike violinists,
wind players, or singers, pianists obey too slavishly the prescribed note values,
because it doesnt occur to them that their fulfillment is granted to the performance that recognizes that the score only calls for effects, not the means by
which to achieve them.2
In bar , one should hesitate a little between the first and second eighths,

2
[S.] See the Preface to Beethovens Neunte Sinfonie, pp. ff/pp. ff. [Schenkers views on textual
criticism and musical orthography are summed up there in similar words (p. /p. ): Orthography
announces and seeks effects, but says nothing at all about the means of producing them!]

Mendelssohns Venetian Gondola Song, Op. , No.

Mendelssohn: Venetianisches Gondellied, op. , Nr. { }
t r a n s l at e d b y j o s e p h l u b b e n

A piano piece by the master, one which all the world has played and yet plays,

master reifies all of thatwater, boatman, and songin a single voice-leading

progression, in the ascending register transfer of that first ag  in bars :

but which no one ever really knew and no one yet knows. Mendelssohn himself
was the source of the title, and the musical public has stuck with it.
Nobody really believes that gondoliers in Veniceor anywhere elsesing
like this. In their songs, nature remains distant from art. On the other hand, there
are indeed gondola songs that are counted as artistic; but not until when one
hears Mendelssohns gondola song does one begin to realize how little art those
others possess. They are too corrupted for nature, too raw to be art, like unhappy
mongrels that belong neither to one category nor the other. A German master
sings a Venetian gondola song, and look, even Italian nature pales by comparison.
Arts higher wisdom takes on the traits of profound simplicity, so that by tonepainting and imitation of reality it even appears to be a product of nature.
The masters art is not any less great here than it is in his proudest and most
extensive compositions; this fifty-five-bar synthesis demands no less respect than
those others. Progression after progression, we witness the presence of a brilliant
tonal spirit, who listens to the slightest rustlings of the tones, to their needs and
wants, and brings his soul into accord with theirs.
The piece has a three-part song form. The first section, a1, bars , is confined to the tonal space f  a, wherein the Urlinie executes a falling third-progression. The middle section, b, bars , sets out from , moving directlyin the
moment that appears (bar )into the third section, a2, bars . A brief
coda closes the piece, bars .

The register transfer proceeds (see Fig. a) by means of the reaching-over technique that the master is able to join ingeniously to an octave exchange (see
Fig. b). {} If we now compare the last sketch with the realization,1 we are able
again to recognize one-by-one, in the interplay of reaching- over voices and registers, the events that formed part of the composers conception. Thus the different registers are divided, as if between the water and the singer: a in bar belongs
to the water, g 1 in bar belongs to the singer, d1 in bar is again the waters, and
c 2 in bar is the singers. The reaching over across bars | takes a bold tack: the
development of the octave a1 a2 in bars begins with a1 in bar , but the c 2
that comes from d1 arrives on the downbeat of bar . Thus the last two eighths of
bar stand in a strange twilight: according to their relation to g 1 in bar and g 
in bar , they belong to the dominant chordthis is contrary to the reachingover technique, which is already inclined toward the tonic.

Bars ff. But, right from the start, what mastery sets that in motion! Out of
the waves depths, ag  rushes up (bars ), we hear a secret, but we dont suspect
that it signifies the first step of the Urlinie, . Then suddenly a call rings out,
as if out of the mouth of a gondolier who is just turning the corner. The call rings
out, the waters rush, the song begins: a call-and-response of waves and men. The

1
Vergleichen wir das letzte Bild mit der Ausfhrung. This expression would normally denote the
relationship of the graph of the Urlinie to the actual score, but here Schenker is probably referring to
Fig. b, whose realization can be seen in the graph of the Urlinie itself, on p. .

Mendelssohns Venetian Gondola Song, Op. , No.

The octave a1 a2 in bars is laid out in a broad arch. The unfolding of the
octave f 1 f 2 follows in bars . The descending arpeggiation of a third is expressed by both of the octaves, a1 f 1 and a2 f 2 (see the slurs in the graph of the
Urlinie, above). The arpeggiation of the next third responds by taking an ascending course in bar , bearing upward, and responding in rhythmic diminution to
the call of bars . is placed above V, which is approached from below by way
of the third, A (arpeggiation). The realization uses a sixth as an inversion of the
third, granting at least the weight of the lower register to the barely established V.
The arpeggiation of both thirds (ascending and descending), the development of both octaves, the beautiful arch of the [boatmans] song, the exuberant
note repetition in bars , the diminution in eighths in bars that scurries
down from a2, the way in which a1 b1 c 2 in bars responds to the eighths of
bar , retrospectively confirming that those eighths belong to the tonic chord
all of this is the profound serenity of the path, and at the same time a dreamlike
cohesiveness of the Urlinie, in short, an intertwining of relationships that is at
once both strict necessity and a breathing-forth of the most outward freedom.
The dominant supporting in bars has the effect of a half cadence, [the
end of] an antecedent phrase. However, , insofar as it recalls the situation in bar
, now becomes a cue for the reaching over that will also be repeated. Still in bar
, d1 enters in the inner voice (contrast this with bar ), and soon the consequent
phrase comes into play with the octave unfolding a1 a2. The contraction of the
ascending register transfer is self-evident, since it would not do to introduce the
singer again, as in bars .

It occurs to the composer, who traces over the human song with pianistic coloratura, to place and therefore in the lower octave. The arch from a2 to f 1 in
bars is priceless, expanded compared to the one in bars , and perfectly
formed. The endpoints of the third, a2 and f 1, are now in the same descending
arpeggio, resulting in an even warmer one, on account of the immediately expressive relationship achieved in spite of their distance. The arpeggio is delayed
by syncopations (see the graph of the Urlinie); but if a1 in bar , between c 2 and
f 1, is for its part only an eighth, this gives the impression of an acceleration at the
end of the arpeggiation.
The next third, in bar , is transformed in the same way with respect to the
one in bar . At the same time, how artfully is the suspension placed into the
service of the diminution: g 1 appears not on the downbeat, but only on the third
eighth, thus corresponding to the register of a1 on the weak beat of the preceding
bar. {}
Bars ff. The middle section begins in bar : a broad passing motion rises from
f 1 (standing for f 2) to e 2 in bar an Urlinie descent is thus expressed by
means of ascending passing motion (see Elucidations, Fig. )now at this
point one comprehends the meaning of the descending register transfer at the
close of the consequent phrase: the composer already has the ascending register
transfer in mind. The leaps of a fifth, with a pattern, are in the service of the
passing motion. Of course, in bar there is an F  instead of G  in the bass, so
that d 1 in the upper voice functions as an accented passing tone, i.e., as a passing

tonw i l l e 1 0
tone that comes from c 1 (). However, the rising fifth F  C  (in place of the
falling one) serves as an anticipated divider at the lower fifth. (While it is true
that, as concerns the passing motion, the dividers at the upper and lower fifths
are entirely equalsee Kontrapunkt ii, p. /pp. , and Elucidations, Fig.
in the case of ascending passing motion in the upper voice the divider at the
lower fifth nonetheless frequently has the advantage of a more engaging effect: it
transforms the ascending passing tone into an accented one that, despite its origins as a simple passing tone, is somehow more artistic.) F  here provides moreover a useful service by reviving and preserving the remote actual root; G  C 

tave (with e2 b1 in bar moreover delivering a repetition), thereby paving the

way for the same octave unfolding that was the hallmark of the a1 section. The
a2 section does in fact begin in bar , with the incipit unfolding a1 a2, and develops in the same manner as the consequent phrase of the first section.
c 3 enters above a2 in bar : the firmament of this note is stretched from the trill
of bars and overarches even the coda (see the dotted slurs {} in the graph
of the Urlinie). It is from musical justice, the law of obligatory register, that such
sweetness gushes through such an expansive space.
Hardly has the octave progression of the Urlinie reached its conclusion in bar
, than it is taken up anew by the inner voices; it is propagated in the waves,
while the singer continues to cadence with the leading tone. Here once again one
should notice the last eighths of bars and , a whiff of naturalism in the premature entrances. In f 2 of the last two measures we find not only the conclusion
of the Urlinie, but also c 3 falling to f 2; the waves subside, a final falling fifth, and
the waters surface comes to rest in the bass.

would needlessly anticipate the dominant effect, which in any event is brought to
bear strongly enough in bars . One observes also how the passing diminution makes use of the opening motive, again with shades of twilight (the thirdprogressions e 1 g 1 and g 1 b1 across bars | and | are both anticipations). The repetition of c  in bar pursues a special goal: to highlight the
rhythmic augmentation that extends the eighth-note motive to complete bars.
In bars we again see the ascending arpeggiation of a third, as in bars
and bar . And once again, as in those bars, the inner voices are stirred up
here it happens not in the service of an ascending register transfer but, rather, in
order to fill in space, so that the Urlinies call, which moves from  (e ) to diatonic  (e), is made stronger.  is introduced by means of an especially artful ascending arpeggiation of a third: c 3 appears in bar , an imposition that may be
understood as nothing more than an artistic imitation of the naturalistic Italian
singing style,2 and it is maintained in bars , until finally, on the first eighth
of bar , e3 appears and resolves the tension. The extended trill on c 3 increases
the tension on the way to e3. (In the graph of the Urlinie, e3 is placed where it already belongs, on the downbeat of bar .) The descending arpeggiation d3 b2
appears in bar in response to the ascending arpeggiation c 3 e3, and here as
part of over IV. How logically is the influence of one third on another expressed! Here everything flows forth from the nature of the Urlinie, which must
follow to its end the path through the tonal space of the chords!
Since between  and  there was an interval of an octave, the ascending register transfer had to be given back: in bars the diminution descends one oc-

Only the Gesamtausgabe and first edition

guarantee the correct performance.

They show the high art of phrase articulation that Mendelssohn, like the older
masters, possessedcreated out of the depths of synthesis; these slurs have since
been replaced by the unartistic, indeed completely unmusical phrasing slurs4 in
the newer editions, so that precisely where it is especially relevant, in a piece such
as this, the breathing is blurred, even annihilated.
Ignorance of the content misleads the player into a sluggish tempo. The Allegretto tranquillo refers here not to the eighth, but to the half bar. Only in this way
does the song preserve the fluidity that it demands on account of its particular
representational character.5
The e 1 in bar should be brought in before the root, as if anticipatedthe
player will be reminded of the premature eighths in bars and but the
value of the anticipation should not be increased to an actual eighth. Through
such liberty, the gondoliers call, an imitation of real life, maintains its own distinct outline. It has the effect of being disjunct from voice-leading in general, and

2
[S.] In Schuberts Nacht und Trume, we find an extremely expressive use of a similar sospiro
across bars : one has to have heard it sung by the incomparable J. Messchaert, in order to comprehend the gripping effect of what is basically a naturalistic compositional trait. [Johannes Messchaert was a Dutch baritone, for whom Schenker was the accompanist during a concert tour in .
For an appreciation of Messchaert, see Tonwille , p. /II, pp. .]

3
Felix Mendelssohn Bartholdys Werke: kritisch durchgesehene Ausgabe, ed. J. Rietz (Leipzig: Breitkopf and Hrtel, ). All the sets of Songs without Words for piano solo were published in vol. xi/.
4
Schenkers antipathy toward editorial slurs in particular, and modern editions generally, forms
the basis of his essay Weg mit dem Phrasierungsbogen, published in the first volume of Meisterwerk.
5
The tempo marking Allegro tranquillo, given in the graph of the Urlinie, is incorrect.

Mendelssohns Venetian Gondola Song, Op. , No.

In the left hand of bar , one should use the thumb for the second eighth,
and the second finger for c 1. In bar the thumb is allowed even on the first
eighth, the root. One should not overlook the sf in bar , within a ff, and should
prepare it. The trill in bar should be introduced with four even thirty-seconds
in the last eighth, and the thirty-seconds should continue in the next two bars as
well. The sf in bar is intended for the suspension; with its beautiful warmth the
note makes up for its late appearance. After that, one should proceed from d3 to
b2 with expressive portamento, from b2 to b1 with a steady hand, dividing the motion lightly only once, at e2. Not until the upbeat of bar should pressure be applied in order to emphasize the bass as well as b1; the release of this pressure, beginning with a2, ultimately gives flight to the next octave unfolding. The c 3 in bar
should be treated as if played by a flute.6
One should hurry the fifth-progression f 1 b in bars , since it is an
inner-voice progression; one imagines seeing and hearing the oar-strokes. The
trill in bars is to be played, like the earlier one, in even thirty-seconds. The
forte refers only to its beginning, and for that reason one should practice the trill
slowly at first. The last two bars can only be done justice by using an array of bold
pedal techniques: in the first eighth of bar the pedal is still engaged from bar
; in the second eighth, no pedal; the pedal again in the following two eighths,
but it is lifted just before the last eighth, and not depressed again until the third
eighth of bar , so that the floating f 2 remains illuminated by the final light.

precisely for this reason it is sudden and surprising. Mendelssohn marks a forte
in bar this too is a masterly stroke. In bar , a warmer expression for d1 is
called for. The p cantabile (bar ) indicates an open tone, but this does not mean
every note, rather, only that certain notes should sing out, overshadowing the
others. The right-hand eighths in bar hover delicately above the surface, gliding
toward a distant goal. The repeated notes in bars and press ahead, as if driven
toward the < > in bar . This symbol clarifies the synthesis: to play all of the
c s in bars and broadlyon account of the p cantabileand thereby drag the
tempo, would be incorrect. The hand should somehow know that the first unfolding ends with c 2 in bar , and that the second begins with f 1. Mendelssohn
separates the two unfoldings with an eighth resthow ingenious!therefore
the hand should be lifted lightly off of c 2 and allowed to fall in a gracious curve
to f 1. The second unfolding also demands movement, and the master specifically
notates < in bar , to the sf in bar . The g 2 in bar has to ring out, not only
in the prescribed value expressed by the syncopation, but beyond into a1: only by
means of such a linking of g 2 and a1 does one do justice here to the synthesis (the
notation is incapable of expressing this sort of maneuver). In bar one should
hesitate on the first two eighths, and accelerate from the third eighth to the next
quarterthereby expressing the acceleration (see above), and at the same time
serving the repeated notes (two c 2s) in an elegant, almost naturalistic manner.
The dotted quarter f 1 in this bar should be supplied with sufficient pressure, so
that the notes of bar seem to be {} swept away by the last bit of breath that
the broad eighth-note coloratura has left.

6
[S.] Compare the c 3 of the second violin in Mendelssohns wonderful String Quintet Op. , first
movement, twenty-two bars before the repeat sign [i.e., bar ].

Mendelssohns Song without Words, Op. , No.

Mendelssohn: Lieder ohne Worte, op. , Nr. {Tonwille , pp. }
t r a n s l at e d b y j o s e p h l u b b e n

The background of the piece can be seen here:

b1 g 2, as in bars . As significant as this detour from f 2 b1 and from b1 back to

g 2 is, it also confirms the Ursatz of Fig. a, where f  moves directly to g .
The middle section lingers in the register of b1. In the lower voice, a passing
tone can be seen between B and G . Because the passing seventh a1, in bar ,
would have been dissonant against the root of the divider G , the neighbor note
D  appears before the E. The upper voice works out a motive, borrowed from the
first sections (see the small slurs under the large dotted one). In the third
section, the third-progression f 2 e2 d 2, ( ), undergoes an expansion,
through the repetition of e2 d 2 ( ); not until bar does the falling fifth to I
usher in .
{} The last voice-leading layer, Fig. c, brings to fruition that which the preceding ones only hinted at in their outlines. A third-progression leads from g 1 to
b1, the take-off point. It is imitated by b1 d 2 in the middle section, and thereby
attains the status of a motive. The antecedent is enriched by an unfolding that inserts VIIIV. The unfolding in the consequent is even more bountiful: the pressure of its increased weight even expresses the key of the dominant, and the fifthprogression. grows in significance [almost to attain the status of] an Urlinie.
The return modulation then becomes the task of the middle section. It is precisely here that the composers extraordinary talent for narrative indulges in a

very artful prolongation of the ordinary V 7. He immediately attaches an imitation of the introductions opening third-progression (see above) to the descent of
the lower voice, BAG . By then making another ascending leap of a sixth from
d 2 (like the one in bars ), he reaches b2, confirming the two-line octave as the
home of the Urlinie progression. One should understand the master correctly:
he had the immensely difficult task of, on the one hand, emphasizing the register
of b1, on account of the coming leap of a sixth, as the point of departure in the
middle section (see Fig. b); on the other hand, the law of obligatory register demanded that, if possible, the two-line octavesince it is the register of the Urlinieshould also be incorporated into the middle section. He therefore con-

Fig. a unveils the first space of a third in the E major chord, eg , as the tonal
space of the whole.
Since is placed directly above the tonic chord, this tonal space is filled in by
the descending third-progression of the Urlinie, (g f e).
In the breadth of this narrative, turns twice toward before setting out once
more and finally continuing to . In the language of form, the first two statements
of , bars and bars , signify antecedent and consequent of a first section. We then recognize in the fifth-progression divided at the third, B(G  )E,
the outline of a middle section, bars , which is followed by the return appearance of at the beginning of the third section.
The voice-leading layer of Fig. b shows the deployment across a broader registral span. This is what provides the opportunity for the falling fifth-progression
f 2 b1 in bars , giving greater weight to the consequent than to the antecedent. One does not grasp the true meaning of that fifth-progression, however, until
one acknowledges it as a path from the upper to the inner voice. The inner voice
finds its way back to the upper voice in bars , using the same leap of a sixth,

Mendelssohns Song without Words, Op. , No.

The ascending leap of a sixth to e2 in bars aims for a parallelism to g 1

e1 in bars . The leap of a sixth to in bars , a 1 f 2, is interwoven with the
descending arpeggio in bars . A similar interweaving (descending fifth and
ascending sixth) also takes place in bars , and the last leap reaches up to .
At the same time, all of these unfoldings have the value of harmonic degrees (see
the graph of the Urlinie). The periodicity is regular, in four- and eight-bar
groups, but the reaching over in bar , seen in terms of the upper voice, creates a
sort of upbeat; this constantly flies in the face of the regularities, creating a contradiction that is, to be sure, overcome by the norma fascinating game of regularity and contradiction.

stantly seeks a balance between the two registers, as indicated with arrows and
slurs in Fig c and the graph of the Urlinie. The version in Fig b wins out: the
two-line octave disappears, b1 holds firm in the upper voice and can, in the end,
leap to g 2.
In the a2 section, the of bars and bars is decorated with a
double neighbor note, joined by the secondary effect of a third-progression (see
f 2 e2 d 2 in bars and ).
Third-progressions appear in the coda as well, in bars ff (see the inner voice).
Now for the final realization:
Bars ff. In the introductory bars , the content presents itself as nothing more
than a simple accompaniment, waiting for the imminent melody; but we know
from the preceding remarks that the third-progression has the effect here that its
end-note leaps to g 2. The chordal placement of thus passes through the fifth of
the chord (compare Fig. b and c). The third-progression also contains another
motivesee the arpeggiations g 1 e1and a1 f 1 in the graph of the Urlinie (above),
in bars and bars . It follows that b1 in bar does not actually come from
a1, rather from an [inner] voice that reaches over.
The leap of a sixth in bars , which establishes the structural range of the
upper voice (see Kontrapunkt ii, pp. ff/pp. ), finds a corresponding descent from g 2 back to b1 in bars . This ascending and descending motion now
becomes raw material for diminution, gracefully disguised by the motives ( in the
strictest sense of the term) and the artful articulation.

Bars ff. {} The beginning of the consequent phrase is concurrent with the
modulation (by harmonic reinterpretation: see Harmonielehre, pp. /pp.
). The voice-leading is the same, yet the meaning of the harmonic degrees has
changed. The C  minor chord, II in B major here, is extended in bars , compared to its original appearance in bar of the antecedent. In order to understand
this, one must append the c 2 in bar to the g 2 and e2 in bars and (compare bars and ), and thus hear the descending arpeggiation in bars
(which in any event is required by the diminution technique) as a summarizing
triadic result of all of bars ; one then also understands why the extension
brings g 2 to the surface once again in bar it is necessary precisely because of
the summation. (It is another matter that c 2 in bar is immediately captured by
the V.) Refreshing g 2 in bar is also at the same time beneficial to the progress

tonw i l l e 1 0
of , which by the proximity of g 2 and f 2 in bars is made clearer here
than in the case of the corresponding notes in the antecedent. The descent in bars
is followedas has always been the case up to this pointby an ascent in
bars , and this in turn precipitates the subsequent descent of the long fifthprogression that carries the Urlinie in bars . What synthesis! How the rising and falling serve the modulation and the Urlinie at the same time!

bars lingers in an extension of V, bars . In terms of voice-leading,

the extension is achieved by retaking b1 in the inner voice in bar in spite of
the e2 that juts out above itand then proceeding from there through a fourthprogression back up to e2 in bar . It is as if in bar , before the fourth-progression
had ascended, a registral inversion had taken place (b1 as upper voice); thus, one
should understand an elision of phrases in bar and hear c 2 in bar as coming equally from b1 and from e2. This type of voice-leading serves a valuable purpose, particularly in large symphonic movements, where it often aids in the most
significant extensions. {} The fifth-progression in bars responds to the
one in bars ; here d 2 is even more clearly expressed, without a suspension,
making possible the leap of a fourth b1 e2 across bars |, in all its purity and
without the concealment found in bar .
The coda begins in bar . It contains, at the beginning, a confirming cadence
in bars . The upper voices g 1 f 1 in bars is actually an inner
voice (see the graph of the Urlinie): two sixths that follow one after the other are
broken into a descending and an ascending leap, respectively (see also Elucidations, Fig. ). In bars , a leap of a fourth is propagated, new life from the
living leap of a fourth that had governed the third section overall (replacing the
leap of a sixth in the first section). One should not fail to hear, however, the ancillary effect of g 1 f 1 e2, as a kind of newly engaged , the third-progression
of the Urlinie; it recedes in the face of the primary effect. In bars there is
the same treatment of the upper voice as in bars ; once again the leap of a
fourth b1 e2 is emphasized as the actual goal. This not only increases the significance of the goal, but additionally prepares the final group, bars , which at
the same time, leap by leap, carries out the arpeggiation of the chord: ascending
from e2, through g 2 (bar ) to b2 (bar ), and descending from b2 through g 2
(bars ) to e1 in the last bar.
A word concerning performance:
This piece has come to be known as Lullaby in recent editions, but in truth
it approximates a waltz type. Two more contradictory conceptions are hardly
imaginableeven if the infant were a Hercules in the cradle, he would not dance
a waltz!yet truly, the lack of understanding surrounding this wonderful piece
of music, this piece that lives in every blood vessel, is even greater than the distance between those two conceptions.1

Bars ff. The last fifth-progression, filled with passing tones, now stimulates the
growth of further fifth-progressions in the lower voice, likewise filled with passing tones. The meaning of the rising third-progression in bars has already
been stated; within it, in bars and , little replicas are embedded, whose own
upward striving rejuvenates the ascent of the large-scale third-progression. The
descent in bars corresponds to the ascending leap of a sixth to b2 in bars
; likewise, the descent in bars corresponds to the ascending leap of
an octave to d 2 in bars . (Regarding the descents to a 1 and c 1, see Fig. c.)
The connective linear progression in bar is also, like those in bars and , a
replica of the third-progression.
The note b1 in bar , which, according to all of the voice-leading layers and
the final realization, is the goal of the middle section is preceded by a neighbor
note a 1. This prepares the sixteenth-note figure, which itself is a diminution of
the leap of a sixth (compare bars and ff). This half-step in the upper voice in
turn now demands one in the lower voice as well; here, however, C  D  ultimately prepares the way for D  E (see the brackets in the graph of the Urlinie),
the half-step that had already been planned long ago (see Fig. b). How wonderfully one force leads to another, each step becomes an event that embeds itself in
the ordered succession, everywhere there is cause and effect, pattern and copy.
Bars ff. The lower voice in bars is laid out as a rising fifth-progression
that orders the harmonic degrees (compare Fig. a); the first-inversion tonic
chords in bars and are already pressing upward toward V. The diminution
in bars signifies a stationary e2, which was approached by leap from b1 in
bar , in the manner of the previous leapsthus the same law of diminution
governs the third section of the piece as well. Its rising and falling becomes even
more effective in the leaps that follow: c 2 f 2 in bar , f 2 b1 in bars , b1 to
e2 in bars , etc. (but it is self-evident that the passage in bars , according to the Urlinie, signifies f 2 e2d 2: see Fig. b and c). The repetition in

1
Schenkers working copy was Theodor Kullaks edition of the Lieder ohne Worte (C. F. Peters,
Edition No. a), OC Scores, No. . The piece does not have a formal title there, but Kullak, in common with many other editors, notes that it was familiarly known as Wiegenlied (Lullaby).

Mendelssohns Song without Words, Op. , No.

Undoubtedly the publisher did not recognize the upbeat, and did not comprehend the conflict between norm and upbeat. Thus the false phrasinghow
they think only of performance!which engages and then destroys the four- and
eight-bar groups.2
Bar presents a pianistically difficult task: the first eighth, b1, should under
no circumstances be intoned forcefully, since the entire bar signifies an upbeat
(see earlier), and this would convert the upbeat into a strong beat. Also, the entire sixteenth-note figure must be played extremely lightly, in spite of a slight emphasis on a 1, with an acceleration to g 2, and even with a bending of the arm to
the right that physically reproduces the figure. Finally, bar signifies only a stationary b1 all the more reason to use the figure only as a springboard. By contrast, the point of emphasis should be postponed until bar , when b1 returns; this

note should be the only goal of all previous motion (this is why Mendelssohn notated a > in this bar). The release of tension that follows b1 in bar technically
makes bar into an upbeat to the next four-bar group; one thus comprehends the
importance of the point of emphasis, which can be found each time only in the
third bar of the four-bar group.
One must really see the diminution in bars ff in the original notation
there is nothing more artful than this colorful, extremely delicate arrangement of
eighths according to number, legato, portamento, staccato; a veritable singing in
syllables and words, an enchanting feast in gradations of color, with the most
elastic breath and the broadest aural horizon! How high stands this German
singing, in comparison with that of the Latin countries!

2
Schenker probably has in mind Kullaks continuation of the portamento slurs to the b1 in bar ,
the f 2 in bar , and so on, which would imply the grouping of bars , , , and so on. In
the Mendelssohn Gesamtausgabe, edited by Julius Rietz (Leipzig: Breitkopf & Hrtel, ), these
slurs stop at the last eighth note of bars , , and , etc.

Schumanns Scenes of Childhood, Op. , No. Of Foreign Lands and Peoples

Schumann: Kinderszenen Nr. , Von fremden Lndern und Menschen { }
t r a n s l at e d b y j o s e p h l u b b e n

The foreground of the piece has arisen from this voice-leading past:

tion of the seconds occurs, representing a thematic contraction. The upper voice
falls from b1 to e1 (motion to the inner voice), and thereby deprives itself of the
neighbor note c2 in bar . c2 is thus all the more effective in the ascent ,
and the whole of bars () gives the impression of an enlargement of the first
arpeggiation of the third.
The outer-voice setting in bars shows motion through a third in the upper
voice, and through a fourth in the lower voice. Here is the origin of the phrase:

A harmonically supported is lowered to in bars (see Fig. a): this is the

first section of the little three-part song form. lingers in bars , which represent the middle section. The neighbor note ( ) aids the expansion here, and by
its upward stepwise motion also signals the ascent (compare Kontrapunkt
i, p. /p. , and ii, p. [recte: ]/pp. ). The third section, bars , is
a repetition of the first.
The piece thus avails itself only of the tonal space between the third and the
fifth of the G major chord, and closes with instead of ; the matter of this ending will be discussed later.
The first unfolding1 is visible in the graph at b). It begins as an arpeggiation
of the third in the upper voice, causing to fall on weak bars, on the second and
fourth bar of the first four-bar group. Subsequently, unfoldings in seconds follow,
causing and to fall similarly on weak bars. In the middle section, an accelera-

In Fig. a, the lower voice moves by a simple passing tone to the root of IV;
{} the passing tone forms a chord with the stationary middle voices. In Fig. b,
instead of the passing tone, a neighbor note to the root of IV is triedbut here
the unfruitful chord threatens again; only through the conjunct motion of the
inner voice, shown in Fig. c, is the appropriate chord achieved.
The graph of the Urlinie (p. ) illustrates further stages of diminution. The
first third-arpeggiation is enlivened by a melodic detour that leads to the from
above. The same succession of pitches could belong just as well to an inner voice;
and thus the example clearly illustrates what little effect a melodic superstructure
can have on the course of the Urlinie.
Leaps of a fifth are inserted in the lower voice in bars . In the service of
the motive, they simulate harmonic degrees, though they are only passing. In bars
the progression IVVI is presented melodically by a fourth-progression,
replying in a parallel manner to the preceding one in bars ; this is an extraordinary piece of synthesis. In bar the inner voice has sixths beneath the
upper voice (ga), and, incited by this, in bars it carries a third-progression

1
die erste Ausfaltung: coming at the head of the paragraph, Schenker appears to be using this term
here in the general sense of elaboration (Auskomponierung). The components of this elaboration are
then described as arpeggiations of a third (Terzbrechung[en]) and unfoldings of a second (Ausfaltungen in Sekunden).

Schumanns Scenes of Childhood, Op. , No. Of Foreign Lands and Peoples

childlike curiosity so focused on the world outside that we do not fully experience the life of our inner world.
Regarding performance, one should note the following: in bar the chromatic
note C , the root, requires a delicate touch; the same touch is required for the first
quarter note of bar in the right hand, in order to restore the balance of the twobar unit. The symbol >, which Schumann notates in bars , clarifies the synthesis, i.e., it is intended to express the closing off of the first section. The two
obligatory ritardandi in bars and must absolutely be performed in relationship to the basic rhythm, i.e., not too broadly. In particular, the passing motion
in the inner voice in bar must reach its goal despite the fermata, which therefore must not be overextended.

in quarter notes. This bold passing motion of the inner voice binds the middle
and third sections across the fermata.
In the final realization, the fundamental arpeggiation of the diminution (see
Fig. b and the graph of the Urlinie) is even more firmly established and emphasized by the upward arpeggiations in eighth-note triplets. The play of the arpeggiations in quarter notes, from bar to bar, indeed even from section to section (as
in the connecting of the middle and third parts), gives to the whole a sense that
somehow refers to the world outside. Waves carry us further and further away,
and the journey has no end; when the piece concludes, we are still, as it were, underway, only surmising that the goal is far, very far away. The composer entitled
the piece Of Foreign Lands and Peoples and thereby expresses its epic character. Borne and buffeted by the constant motion of the element, we remain with

Schumanns Scenes of Childhood, Op. , No. Trumerei

Schumann: Kinderszenen op. Nr. , Trumerei {}
t r a n s l at e d b y j o s e p h l u b b e n

The course from background to foreground is as follows:

At the same time, the Ursatz at Fig. a indicates the harmonic progressions. In
the middle section, outer-voice octaves are conspicuous at and (I and IV) in
bars and .
Fig. b shows the raw material of the diminution, and the richer harmonic
progressions that it generates. The particular hallmark of this diminution is the
melodic construct above the actual notes of the Urlinie. Through ascending
arpeggiation and descending passing motion it describes a high-arching bow at
each succession of two notes of the Urlinie; deviations from this are evident in
the middle section at and of the Urlinie.
Just as the path of the Urlinie in itself stretches calmly and patiently upward
first to , then to , and finally to againso this effect is made even stronger by
the diminution that, constructing bows above the Urlinie, similarly appears to be
reaching for something. Is it dreaming in general, or the content of a particular
dream?
Without considering metric issues, the sketch given as Fig. b illustrates the
following characteristic features of the piece. At the first ascent, , the diminution also summons IV. If the harmonic progression IVV had been combined
with a motion, without diminution, from to of the Urlinie, the fifth-succesf g
sion b
c would have occurred; thus one sees how through contrary motion the
diminution, as well as serving an inherent purpose, also helps to avoid errors in
counterpoint.
The number of harmonic degrees also increases in bars . Moreover, here
the diminution falls short of the Urlinies , which is then only implied by the

harmonic progression II 3 V. (Since the Urlinie moves only by step, it follows that
it traverses as passing motion just spaces of a third or fourthsee Elucidations). Hence it is impossible to grant to , a3, a connection to the last note of the
descending progression, c2, in bar , though this was possible in the similar case
in bar .)
In bars , basically the same ascending fourth progression as that exhib-

Fig. a shows how the Urlinie rises from to , only to fall back again to ; it
climbs there a second time, and indeed all the way to , only to sink back again
to . Finally, it repeats its first motion, ascending a third time to and falling back
to . Thus the Urlinie also clearly illustrates the form, the small three-part song
form, whose middle section is highlighted by .
In Schenkers annotated personal copy of Schumanns piano music (Edition Peters , edited
by Alfred Drffel), as in all other editions of the Scenes of Childhood, Trumerei is the seventh piece
of the collection, not the ninth.
The word Trumerei suggests the act of day-dreaming; the closest equivalent in English is perhaps reverie. Schumanns piano piece is, however, universally known by its German title, which has
been retained here.
1
An undated analytical sketch of Trumerei forms part of the material in the Oster Oollection
intended for Der freie Satz, but not included in the final version (File , item ). It indicates an interrupted Urlinie in which the second is embellished by an upper neighbor. The exact sequence of
notes that comprises the Urlinie of Fig. a appears in the sketch as a later middleground.
In a personal copy of Tonwille (OC, Books and Pamphlets ), Schenker superimposed this
very reading, in pencil, over the printed graph of the Urlinie.

Schumanns Scenes of Childhood, Op. , No. Trumerei

eighth-note value becomes a precedent for the eighth-note upbeat in bar , so

that at least a new parallelism takes place, subsequently justifying the eighth note
c1 in bar . In order to conclude this issue, let bar be considered: since here
(g1) actually occurs in the fourth beat, any possibility of an upbeat in general is
lacking; in this most dire plight the composer can only preserve it by pushing it
through forcefully as a short grace note. (The great masters of synthesis would,
on the other hand, have fully developed the upbeat in any event, and so would
have led the diminution from the beginning up until this point in such a way that
the upbeat quarter note could have been preserved. As an example, I refer to the
manner in which Beethoven, in the first movement of his Sonata Op. , preserves the construction of the first theme group at the beginning of the development section, see the Erluterungsausgabe of Op. , p. /pp. .) In connection with the motive of a fourth of the upbeat, in bar the f 1 replaces a1 ( ); to
be sure, the bass, completing the large-scale fourth progression, brings in the A
(before B
) at this place. From this example, one can plainly see what difficulties
synthesis can present.3
Regarding the diminution of the final section, one should note the following:
The first two eighth notes of bar stand for a contraction of the upbeat; in
addition, the second eighth note acts as an anticipation. These two eighth notes
become a specific diminution motiveas shown in bar by the first and third
quarters, in bar by the first quarterwithout, however, raising the question of
whether the peak tones should be taken for part of the outer-voice setting (see the
graph of the Urlinie). Through the combination of each pair of such eighth-note
progressions, a new motive arises (see the motive of the soprano in bar at the
first and second beat); in the third and fourth beats, the alto imitates this new
motive, but changes the leap in the fourth beat to a second (on account of the
bass). Similarly the tenor in bar uses a leap in the first beat, but in the second
beat a step instead of a leap of a third (gb
). The original accenting of the weak
beats (see above) is disrupted by none other than this new motive.
It should also be mentioned that across bars | the consecutive fifths between
the outer voices (see below Fig. b) are eliminated by the f  in the inner voice at
the last eighth note; specifically, e2 moves to d2 through e
2. Thus, the outer-voice

ited by the Urlinie is expected of the lower voice; the threat of consecutive octaves
is avoided by inserted thirds (tenths). The semitone of the Urlinie, , a1 b
1,
practically demands the semitone f  g (given the falling fifth of the roots, DG,
f  1 g1 is better than f 1 g1), and therefore also requires that the chords be arranged
for an apparent cadence in G minor. In bars the essentially simple root
succession, B
C, IVV, is animated by a fourthprogression that parallels the
previous one.
IIVI takes part in the final statement of . is elaborated by the typical
third-progression falling to . However, decorated by a neighbor note (d2), it
assumes a form similar to that of a nota cambiata (see Kontrapunkt, i, pp. ff/
pp. , and ii, p. /p. ).
Beneath Fig. b are shown two contrapuntal problems2 in both cases, consecutive fifthswhose circumvention must be the concern of the next level of
diminution.
Now to the graph of the Urlinie (p. ). It shows a leap of a fourth as upbeat.
This leap of a fourth motivates the arpeggiation that follows in bar .
The quarter note of the upbeat and the first quarter note of bar clarify each
other: had Schumann placed the fully voiced chord in bar on the third beat instead of the second, the quarter-note value of the upbeat would thereby have
been destabilized. The chords appearance on the second beat, though thus motivated by the upbeat, also proves to be fruitful for synthesis, since from this point
on the rhythm of the diminution motive is directed toward weak beats.
In bar the quarter-note value of the upbeat is still maintained; in contrast,
in bar it shrinks to an eighth note. This is an indication of an imbalance in the
synthesis: diminution and upbeat fall into opposition. The final note of the ascending progression, c 2, must be accorded the value of a half note, on account of
its status as an ending note, as well as in consequence of the motion in half notes
that is established in the preceding bars. Since on the one hand this c2 would not
do as an upbeat, but on the other hand the requisite c1 would not occur naturally,
given the voice-leading, the upbeat must somehow be brought in arbitrarily, as if
behind the back of g2, the implied note of the Urlinie (or g1 as ), and of c2 (as the
final note of the falling motive). So the composer smuggles c1 in stealthily, moreover in the value of an eighth note (which on account of the arpeggiation is already completely out of the question as an upbeat to bar ). Admittedly, the new

3
The discrepant forms of upbeat in Trumereiquarter note at the beginning of the piece,
eighth note at the end of bar , grace-note at the end of bar are the subject of a handwritten leaf
in OC (file , item ) in which Schenker contrasts Schumanns inconsistency with the practice of
Classical composers (Beethoven, Mozart) in similar situations.

2
Bei c) werden zwei Satzschwierigkeiten aufgezeigt: these short illustrations are not actually
marked c) in the music example.

tonw i l l e 1 0

setting in no way allows two perfect fifths to follow one another, rather, a perfect
fifth follows a diminished one.
In bars and the falling fifth-progression of the upper voice, d2 g1, becomes an imitation motive: at the inner-voice d1 g in bars , and the fragment d1 b
in bar (see the graph of the Urlinie). A similar play of imitations is
also developed in bars , and it is in fact the last fragment in the upbeat half
of bar that brings in and .
The graph of the Urlinie shows how the diminution in the last two bars,
, is to be interpreted. The bass presents a neighbor note in the first beat of bar
(the motive, admittedly, begins on the third beat of bar compare the motive of diminution in bars and ). The neighbor note gives rise, in the middle of
the resolution of the suspension, to a sixth-chord that appears to drop out of the
piece; its elaboration [Auskomponierung] feigns an inherent purpose, almost
causing us to fear that the voice-leading has somehow gone to pieces, until the
second beat of bar creates clarity.

sound as though one wanted to conceal something, the sense of which is not
understood. One should accelerate through the eighth notes in the fourth beat,
hesitating however at the first two eighth notes of the first beat of bar , enough
so that between the two f 2s the new root of IV, the grace-note sixteenth B
, can be
inserted without haste. The pedal should be applied just to this root, and then the
half note f 2 should coincide with the half note d in the left hand. Beginning with
the second beat of bar the pace should be quickened slightly, and the tempo
given back only at the fourth beat of bar . In bar , the second eighth note in the
left hand, g1, is to be played fully, and allowed to linger somewhat beyond the prescribed duration; from it a shadow falls on the remaining eighth notes, with the
effect of a diminuendo, which essentially takes the pressure off of the anticipation
in the right hand; one can aim for the most tender touch for the little note of anticipation, provided that the shadow is clearly established.
Bar should proceed as in bar , with no hesitation until the second eighth
note in the third beat of bar . The last eighth notes in bar should still have a
hesitant step, looking back to the preceding fermata; in bar the second and
third beats should proceed in tempo, the fourth beat of bar and the first beat
of bar being presented with their own shading, as if for their own sake; and
only beginning with the second beat of bar should the ritenuto really take over
completely.

he first concern of performance is that, in bar , the first beat exactly maintain
the rhythmic value of the upbeat. A definite, properly rigid touch belongs to the
chord on the second beat; a tender beginning would just be puzzling; it would

Appendix

This page intentionally left blank

Music Criticism
Musikkritik (unpublished: Oster Collection File , items )
t r a n s l at e d b y w i l l i a m d r a b k i n

For the people, only the best is good enough: so goes the well-known rallying

in anything that exists in the deepest sense. It is a gift from heaven, like a
beautiful voice, a special feeling of being driven toward art: an ability,
which one either has or has not, and which is indeed among the rarest. To
be sure, it can be disciplined, and there are matters on which differences
of opinion are inadmissible; but the boundaries for these are drawn very
tightly, much tighter than one would generally suppose. Whether a violinist plays a note or passage in tune or out of tune {} is something that
is not open to question; on this point observers must always agree, otherwise one would be justified in suspecting that the critical listener himself
had no capacity for discrimination with respect to intonation. But beyond this and similar empirical statements of the most common sort,
there are generally no absolute facts.

cry of those who champion the people. It is not made clear whom they mean by
people; I shall only show briefly how the application of that saying to music
criticism is outlined in the head of a critic whose love for the people is genuine
and to whom he ascribes an influential role even in art. Recently this criticfor
the moment, his name is of no importance herewrote the following about his
profession:1
There is one thing that must be established as a fundamental truth: criticsm is not a matter of understanding, or of knowledge. It is not something that can be learned, and for this reason cannot be grasped logically
This unpublished essay amplifies a long polemical passage from the Erluterungsausgabe of
Op. , deleted at proof stage, which is severely disparaging of Paul Bekker, the chief music critic for
the Frankfurter Zeitung and the author of a Beethoven book () that Schenker had pummeled in
the literature section of numerous Beethoven essays. In its final form, it is marked TW , as it was
intended to form part of the Miscellanea of Tonwille , which Schenker submitted to his publisher in
April (OC /). Because of its excessive length and a shortage of small type at the printers,
Universal Edition had to postpone publication of the entire Miscellanea until Tonwille ; Schenker
was also required to make substantial cuts in it. Deleting the whole of Music Criticism from the
Miscellanea is something that Emil Hertzka personally insisted on; in a letter of May , , he points
out that Der Tonwille, a journal with a small print run, is an inappropriate place to defend oneself
against an attack by a critic for a major metropolitan newspaper (OC /); in a long letter, written
about two weeks later, he goes over several points made in Schenkers essay to show the weakness of
some of his own arguments in the counterattack upon Bekker (OC /).
The entire text of Music Criticism is in the neat hand of Jeanette Schenker. The first eleven
pages (i.e., items ) have been copied almost word for word from four columns of text deleted
from the page proofs of the Erluterungsausgabe of Op. (OC / ).
1
All the extracts discussed in this part of Musikkritik are taken from a newspaper article entitled Kritik und Persnlichkeit (Criticism and Personality), by Paul Bekker. It is preserved in the
Oster Collection (File , item ) amid a collection of newspaper clippings by the same author, kept
in a folder marked Bekkerei. According to Jeanette Schenkers identification, the article was published in the Frankfurter Zeitung on November , ; thus it is contemporary with work on the Erluterungsausgabe of Op. .

That someone who is ignorant of the facts of musical life can simply deny
their existence is understandable; but since when is ignorance reckoned as a gift
from heaven, as the rarest ability? From heaven, says our critic, in an age that
has come to hate the word, let alone admit its existence. But he feels free to use it
with confidence, since he means only his notion of heaven, whose boundary
with ignorance has, for sure, been drawn very low. He writes:
It is the same with virtually all things that are the subject of criticism:
there is no objective yardstick, no reckoning of any factual inventory.
Their real value depends entirely upon the person making the valuation.
To such an ardent enemy of musical facts, {} it is easy to attribute statements such as the following:
But what actually is this understanding, and how should I explain it
more narrowly? How does one achieve it? Experience often teaches us
that two specialists (in the serious sense of the word) contradict one another sharply; and it is not seldom that we find that neither judgment

appendix
who understands nothing about compositional technique. (If, for example, I may
speak about myself: I would really not advise that critic to compete with me at
playing or conducting, or to express critical opinions.)
With regard to knowledge of ones subject, about which our man likewise
knows nothing, he additionally wrote:

proves lasting in the course of time. Knowledge of ones subject, then, offers no security.
I agree that knowledge of ones subject, as it has hitherto been amassed, offers no security; but I am convinced that there isthere must besuch knowledge that does offer security. If, as I see it, the tragedy of the present state of music
lies precisely in the fact that the immense ingenuity of our great masters has not
found the slightest reverberation in music theory, does that automatically exclude
the possibility of a true knowledge of ones subject? But it would be a mistake to
argue with a hermeneutist who is so miserly in measuring out his alms from
heaven that he is surely incapable of reckoning a better future for knowledge of
ones subject, although his pen is able to glide easily over idle progress and future.
{} One reads further:

Knowledge of ones subject can at best provide belated enlightenment

about the material origins of an effect. It can lead in this way to valuable
insights, but these will always remain locked in the material. {} They
lack real explanatory power unless they serve a spontaneously reacting
artistic feeling. Thus artistic feeling alone should be the basis of judgment! But how do things stand with respect to this artistic feeling? Can it
be in any way recognized, conceptually understood, even if only in relation to our selves? Or are we, when we take a particular stance, exposed
entirely to the subjective arbitrariness, the uncontrollable driving force of
an instinct about which we do not know whether it is based on genuine
feeling or is leading us astray?
Now, then, this risk is something we must take on. There is no absolute security, and all decisions, even in artistic judgment, are ultimately
dependent on feelings. That is not to say that, in criticism, the only correct state of affairs is absolute anarchy, with every driveller being allowed
to rely on his feelings alone, regarding all understanding of the material
as superfluous or, worse still, harmful. Even the most sensitive feeling
must be founded on an evaluative process in order to secure and cement
its compelling judgments, {} and this always reaches into the substance
of the artwork. The special art of criticism lies in making judgments
based on feeling understandable and believable from the material penetration of the artistic substance.

I can analyze Beethovens Ninth Symphony harmonically and thematically, right down to the last details, and yet inwardly may stand further
from the work than some listener who knows not the first thing about
compositional technique. Knowledge of ones subject alone is worthless,
because the work of art is never the product of such knowledge; on the
contrary, the latter is only a speculative derivation and specialization of
artistic creativity, achieved after the event.
As little as I would equate our critic with some listener who knows not the
first thing about compositional technique (despite being such a sworn enemy
of compositional facts), so little would I wish to be reckoned among those who
analyze Beethovens Ninth Symphony harmonically and thematically, right
down to the last details, and yet inwardly may stand further from the work,
although I have myself written a monograph on the work. Even in spite of his
stingy alms from heaven, he has, as a good German cosmopolitan of the postWilhelmine, highly blessed social and democratic epoch, no real need to remain
so cowardly; {} to his credit, I thus assume that he would probably have referred
to me by name in that passage if he had actually meant me. So it is not as the author of that monograph, but really only as a critic of the critic, that I say that anyone who is capable of what our man has attributed to the commentatorand it
must indeed be assumed that he also accepts the factual correctness of the commentary, if he is not to be accused of having spoken merely the crudest nonsensewill definitely and in every respect be far superior both in terms of inner
feeling and in power of expression, competence in criticism, and so on, to the one

Dishonest, or stupid? Or both? Suddenly our critic wants to be a musician

again. It does not suit him to be regarded merely as a driveller, and all at once he
is able to demand an evaluative foundation that reaches into the material substance of the artwork. But how is this possible if knowledge of ones subject offers no security? Of what use is any substantive penetration, on what basis can an
evaluation be made, if knowledge of ones subject is worthless? How wretched
must the gift from heaven be in this pitiable man, how empty his inner feeling and compelling judgment, if in the end he is still reaching for something
which even for him belongs to a higher plane, though he cannot admit this him-

Music Criticism

self, still less betray it to his readers. For, in spite of personality, feeling, and
gift from heaven, he wishes above all {} to be a connoisseur and a musician!
Of course he is no different from any of the cheap moral and progressive braggarts
of our age who surround him, poking about the difficult-to-reach and difficultto-accomplish, hoping to present something great to the world with very cheapest of means. (This is the cardinal principle of every democracy, every huff over
progress.) Thus our critic, as indeed all who are gathered around him, initially
degrades what is good, only to revalorize it so that he might count for something
himself. But the fraud has not yet come to an end:

personality values that shape life and movement. It would result in a sort
of intellectual indigestion, a severe blockage in the development and progress of intellectual values. {} Above all, it would actually turn criticism
into what it always has been for the lazy thinkers among the reading public: the inviolable expression of the way it was. In truth, however, every
piece of criticism is invisibly headed by the words: the way I see it.
God save us thenalthough today it is really more urgent than everfrom ultimately discovering who Bach is, who Beethoven is, how Beethoven differs from
Wagner, and how these two are in turn utterly different from the symphonic illiteracy of, for instance, a Berlioz or Mahler. May God preserve us from this, for
it is of still greater importance for our critic that, above all, no sterility enter the
profession. On the other hand, I think it would be better for humanity to remain
stationary with a well-understood Beethoven for centuries, millennia, even if that
meant no Wagneriads or Straussiads,4 to say nothing of todays rubbish. We
would be more blessed indeed to sleep in peaceful reveries for millennia with a
Beethoven than to awaken in darkness each day with the likes of our critic. For
sure, then, even human mediocrity would make quicker progress than were it,
fearing sterility, to be constantly renewed by lies that are yet unable to help it {}
shed the skin of its mediocrity. The way I see it is our critics ultimate answer
with respect to music; yet he would never dare give full vent to his personality, his
I, in such a way on another subject if he knew so little about it as he knows
about the it of music. And now for the best of all:

Such an art of criticism can be disciplined up to a certain point, but must

be understood as individually determined. It is therefore incorrect to compare two judgments that contradict one another in order to demonstrate
the worthlessness of criticism. What do such contradictions prove, other
than that they derive from different persons and from different psychic
limitations? Should criticism do morecan it do morethan transmit
perceptions?
What can one say in response? A few sentences earlier, our man concluded {}
from the situation in which two specialists (in the serious sense of the word)
contradict one another sharply that knowledge of ones subject offers no security, by which he intended to condemn such knowledge once and for all. On the
other hand, two contradictory critical judgments fail to make a case against the
value of criticism! How can these statements be reconciled? The answer is simple
enough: even a critic must earn a living, and he would rather write music criticism than go down a coal mine or suchlike place, despite having every sympathy
for the working classes. Should however the danger ever arise that he, as is the
case now, is led to the conclusion that criticism is worthlesshey, presto!he
turns worthlessness into a worth (following the principle of democracy and the
wage-church2), which he proclaims all the more loudly. He says explicitly:3

Beyond this, however, one should be aware that criticism is not a matter
of reason and knowledge, but of talent and temperament. To restrict or
suppress it would be as foolish and fruitless as any struggle against spiritual forces; with criticism, it would be probably the most foolish and
fruitless.
The matter is, however, even simpler: someone who throws such statements
about himself is lacking in all these qualities: reason and knowledge, talent and
temperament. And against such poverty, of course, no struggle can be of any use.

It can therefore be in no way desirable that critical statements agree with

one another. Such a result would be, on the contrary, regrettable for criticism, signifying sterility of thought and feeling, the renunciation of the

4
Wagneriaden und Straussiaden: the first of these terms, used in a derogatory sense, refers to the
Bayreuth Festival, which had been running for nearly fifty years. There was no comparable festival
celebrating the works of Richard Strauss, but Schenker may be alluding to the composers five-year
appointment as joint artistic director of the Vienna State Opera in , a position in which he soon
incurred the wrath of the Operas management for his financial extravagance.

Lohn-Kirche: a term used disparagingly in The Mission of German Genius to epitomize Marxist ideology; see Tonwille , p. /I, p. and note .
3
The following appears near the beginning of Kritik und Persnlichkeit; it is the only extract
that Schenker quotes and discusses out of sequence.

appendix
That our critic reckons the cultivation of all this poverty as ethos of personality
could only have become possible in our age of an unspeakably depraved democracy, whose distinguishing feature is to speak about ethos and to do the opposite.

It is not the purpose of this notice to recount in detail the content of

Kurths book, in order to add critical glosses.
And then:

It is a basic feature of Kurths work that aesthetics and music theory are
brought into close relationship with one another, thus giving aesthetics a
professional, scientific basis and theory an intrinsically organic connection with the theory of perception and, ultimately, with the view of the
world in general. Of late, this has been attempted on several occasions,
sometimes in a doctrinaire manner as, for example, by Heinrich Schenker,
whose method of measuring and counting {} has been prized by many
as a learned approach. August Halm has done something similar, and incomparably better and more seriously; he would probably have been able
to reach significant conclusions were it not for the fact that his vision was
restricted by a one-sided view of Bruckner. Kurth is without prejudice,
and is thus intrinsically free; in addition, he is more thoroughly educated
as a theorist. He has not fixed himself to a single phenomenon;7 his thinking is determined primarily by scientific reasoning and keeps an objectifying distance from the material. By this means he is able to balance theory and aesthetics in a meaningful way and needs neither to be influenced
by aesthetic prejudices to the detriment of theory, like Halm, nor indeed
to let aesthetic perception become confused by theoretical a priori statements, like Schenker.

{} I first mentioned our critic by name in the discussion of the secondary literature in my Erluterungsausgabe of Op. (), the first volume to appear in
a series of explanatory editions of the late piano sonatas of Beethoven. The contradiction in the work of men, who at the very least protect the most wretched
paintings from the rubbish issued by nonentities, to say nothing of the most accomplished ones, but nevertheless regrettably fail to protect in like manner works
of poetry and music, seemed to me to have reached such a pitch that I had to demand justice for musical works of art. Since then, I have continued to come to
terms with our critics book on Beethoven, most recently in the Erluterungsausgabe of Op. (). Anyone can see for himself that I confronted our critics
words with Beethovens content simply for the purpose of faulting their singular
lack of objectivity, and their hermeneutical misguidedness. I confess that my zeal
to further develop my own theories and work out the content of our masterpieces
allowed me no time to speculate about whether the author of this Beethoven
book would mount an objective counterattack. As far as I knew, however, he had
not in all the years between and mentioned me by either by {} name
or by work. Perhaps he felt personally wounded, for it is a characteristic of those
lacking in objective judgment to take it as a personal attack when someone
dares to oppose them on a matter that they would gladly talk or write about, even
when they do not understand it properly. But suddenly, in , the critic mentioned me by name for the first time. Mr. Paul BekkerI shall now, finally, mention his name here5 does this in the Frankfurter Zeitung on the occasion of referring to a new work by Professor Ernst Kurth.6 He writes there:

Being myself an artist and not an aesthetician, still less a learned man, I confess not to have a clue about what Bekker may mean by all of this. But I cannot
help being a bit suspicious of his habemus papam-joy.8 Perhaps readers with a
fine sense of hearing also treat me in the same way; but Bekkers readers {} may
be left to work out for themselves what he is saying, even in those placesand
this really is worth demonstratingwhere he cannot understand himself. Moreover, one finds idle wordplay in Bekkers brief notice: living science, stylistic
psychology on a grand scale (applied to Kurth), an entirely naturalistic way of

5
[S.] It may also be noted that the first part of this essay is a sort of subordinate clause taken
from the Erluterungsausgabe of Op. . Sapienti sat [may the wise know it].
[The subordinate clause comprises about four columns of text from the page proofs of the Erluterungsausgabe of Op. , which were deleted from the final version.]
6
The work under review was Kurths Romantische Harmonik und ihre Krise in Wagners Tristan; a
copy of the review, which appeared in the Frankfurter Zeitung on March , , is pasted into
Schenkers scrapbook (OC , p. ).
The publication of this review coincides with the start of a friendship between Bekker and Kurth;
their surviving correspondence is transcribed by Luitgard Schader in Ernst Kurths Grundlagen des
linearen Kontrapunkts (Stuttgart and Weimar: Metzler, ), pp. .

7
Einzelordnung: Bekker is referring to Schenkers Urlinie, a term first used in the Erluterungsausgabe of Op. .
8
That is, the joy of having elected a new pope (Kurth) of music theory. Schenker mounted an
extended attack on Kurths Grundlagen des linearen Kontrapunkts in , at the end of a pair of essays
on Bachs unaccompanied violin music (Meisterwerk i, pp. /pp. ).

Music Criticism

conceiving music (applied to Kretzschmar), and so on, culminating in an expression of the favourite notion of all democratic Sunday drivers in the arts: In
this respect [Kurth] expresses at the same time the limitations placed on the
methods of recognition or, to be more precise, their respective partial value.
Soon afterward Bekker came back at me, this time concerning a facsimile edition of Beethovens C sharp minor Sonata for which Universal Edition invited me
to write a foreword.9 I believe I can say quite categorically that I was the first to
undertake such a task objectively: I disregarded everything that did not strictly
pertain to the manuscript and determined only which notes the master changed,
here and there, and why he replaced what he cancelled with something new. In
short, I went through the manuscript line by line and spoke in few words about
things that any musician {} could verify with his own eyes. Concerning my
notes, Bekker comments on:

such a thing. His readers may, however, be satisfied with the following sentences
{} as a sufficiently objective statement:
Beyond this, it is desirable to know the original version of a musical work,
to set against the editions that have been marred by instructive remarks
and other comments, and to know that they are protected by facsimile reproduction from the danger of being destroyed.

I certainly need not assure readers of my Erluterungsausgabe that, with the

words mysterious expressive charm, Bekker does not mean the charm in the
notation of some musical idea, for that would presuppose that he recognized

But he has taken this idea from my work, which today is readily acknowledged even by my opponents. (The democrats joy in expropriation does not even
recognize the obligation to thank those whose ideas they have stolen.) Can one
not see clearly how Bekker begrudges me my objectivity, only to denigrate it as a
personal expectoration just for the sake of denigrating me? Pure vengeance,
which only proves that he finds personal expectoration in matters of art
exactly as I doinsipid and iniquitous. Moreover, he appears to rely for support
on the widely read Frankfurter Zeitung, in whose services he is engaged.10 But
thatat least insofar as it concerns meis ill-contrived arrogance since, in the
first place, his paper is only a daily newspaper and, secondly, only a democratic
paper. And so I can say boldly that I am stronger than Bekker and his paperhis
shieldbecause I am engaged in the services of the aristocracy of genius. Or, to
express this in laymans terms: {} I give good milk and good butter and not, like
Bekker, merely rancid phrases that have no nutritional value. And so he should
learn to be modest; otherwise he should say wherein I have been factually incorrect in my criticism of him. And he should do so in the straightforward language
of a musician who is not satisfied merely with the capacity for discrimination
with respect to intonation. Meanwhile, I shall have ever greater faith in reincarnation: the sixteenth Bekker will finally have to learn to read music, and indeed
in my school and in no other. And he will not understand why the first Bekker
kicked about with his short democratic legs and struggled so wildly against learning to read music, yet found the courage to set up a corner-shop for democratic
phrases and celebrate the future before his people.

9
Schenkers facsimile edition of Beethovens Moonlight Sonata () was the first volume in
Universal Editions Musikalische Seltenheiten series, of which Otto Erich Deutsch was the general editor. Bekkers unfavorable comments appeared in an article entitled Musiker-Faksimiles in the
Frankfurter Zeitung of April , , in which Schenkers edition was one of a number of publications
discussed. A copy of this review was pasted into Schenkers scrapbook (OC , p. ).

10
In a letter of May , (OC /), Hertzka admits that Bekker had been spiteful in his review, but cautions Schenker about the futility of using Tonwille to mount an attack upon the critic
(see note ). This may have only further inflamed Schenkers anger over Bekkers comments on the
Moonlight facsimile.

Schenkers insipid remarks about the autograph, together with his Urlinie. If one must publicize such personal expectorations, then at the very
least they are out of place here, and ruin the impression made by this sort
of publication.
But then he seeks to enlighten his readers, as follows:
There lies in the arrangement of the musical notation, as fixed by the
composer, a mysterious expressive charm, which exceeds the magic of a
literary manuscript. It is as if the secret lapidary notation, set down in the
musicians personal handwriting, mirrored the mystery of the creative
process with remarkable clarity and, moreover, the shorthand of feeling
brought the artists individuality to the fore in an especially characteristic, almost pictorial way.

This page intentionally left blank

Index

cussed as theoretical concepts, though they appear frequently: *arpeggiation (Brechung); *ascent, linear (Anstieg); *divider (Teiler, [Ober-]Quintteiler); *mixture, modal (Mischung);
*outer voices, contrapuntal setting of the (Aussensatz); *register transfer, ascending and descending (Hherlegung, Tieferlegung); *synthesis (Synthese).)

Preliminary Note: Apart from Urlinie and Ursatz, which have been left untranslated, technical terms used in Tonwille are given in English with the original German (if different)
in parentheses. Terms that appear very frequently in the text are marked by an asterisk (*);
page-number references are givenin italicswhen these terms are explained or discussed
as theoretical concepts. (Several such terms have no page references because they are not disAdler, Guido, ix n.
Albrechtsberger, Johann Georg,
Altmann, Wilhelm, xiv
Amalie, Princess, n.
Anderson, Emily, n., n.
Antisthenes,
*Auskomponierung. See Elaboration.
Austria, xiii, n.
Bach, Carl Philipp Emanuel, xv
Versuch ber die wahre Art das Clavier zu
spielen, n., n.
Bach, Johann Sebastian, ,
O Lamm Gottes unschuldig
chorale, ,
organ chorale preludes, BWV and
,
St. Matthew Passion, xiii
No. Kommt, Ihr Tchter, chorale
fantasy, n., n.,
No. Erbarm es Gott, recitative,

Twelve Little Preludes, xv
Well-Tempered Clavier
Book I, Prelude and Fugue in C minor,
xv n.
Bartk, Bla, ix n.

See also under Schenker, Heinrich, editions of music and Erluterungsausgaben

String Quartets
Op. , No. , in F, n.
Op. in EI, xiiixiv, xvi,
Op. in CT minor,
Symphony No. in C minor, viii, xv
analysis, xiixiii, ,
autograph corrections and sketches,
,
literature,
performance, ,
textual matters, , ,
Symphony No. in D minor, xiv, n.,
,
Variations in EI (Eroica Variations),
Op. ,
Bekker, Paul, viii n., xivxv, n., ,
,
Berg, Alban, n.
Berlin, Prussian State Library, n.
Berlioz, Hector,
Bible, quotations from and references to,
xiii, , n., , ,
Bielschowsky, Albert,
Binding, Rudolf, n.

Baumgarten, Theodor, ixxii

Beethoven, Karl van,
Beethoven, Ludwig van, xiv, , , ,
,
Fidelio, n.
letters,
metronome markings, , , n.
Mass in D (Missa solemnis), Op. ,
n.
Overture in C (Consecration of the House),
Op. , n., n.
Piano sonatas
Op. , No. , in F minor, xiii, , n.
Op. , No. , in C,
Op. , No. , in F, n.
Op. , No. , in CT minor (Moonlight), n.
Op. , No. , in D minor (Tempest),
Op. , in F minor, viii, xii; analysis,
xiii, ; Appassionata title,
, ; arrangement of second-movement theme for male-voice choir,
n.; artistic assessment, ;
sketches, ; literature, xiv; performance, xiii,
Op. in A,
Op. in BI (Hammerklavier),

Brahms, Johannes, , n.
Variations on a Theme of Handel, Op. ,
vivii, xii, n.
analysis, xiii,
artistic assessment, ,
autograph score, n., nn.
literature, xiv,
performance, xiii,
textual matters, ,
Brandenburg, Sieghard, xvi, n.
Breitkopf & Hrtel, music publisher, vi n.,
n., n.,
Broesicke-Schoen, Max
Bruckner, Anton,
Blow, Hans von, , n., n.
Bureau des Arts et dIndustrie, Viennese
publisher, n.
Carmen, opera by Bizet,
Chopin, Frdric
Etude in C minor, Op. , No. (Revolutionary), n.
Etude in A minor, Op. , No. , n.
Polonaise in C minor, Op. , No. ,
n.
chorale prelude, chorale fantasy, ,
Christianity, ,

index
Clementi, Muzio, n.
Cologne, University of, xv
commerce, in relation to state,
Cotta, J. G., publisher, vii
Cranz, music publisher, n.
Cube, Felix-Eberhard von, xii n.
Czerny, Carl,
Daguerre, Louis Jacques Mand, n.
Dahms, Walter, vi n.,
degree, scale-step (Stufe), xiv,
Demblin, August, x
democracy, democrats, , ,
Der Heimgarten, magazine, xv
Deutsch, Otto Erich, , n., n.
diatonic content (Diatonie),
Die Musikantengilde, journal, xv
*diminution, n.
Drffel, Alfred,
Dumba, Nikolaus, ,
Drr, Walther, , n.
Eckermann, Johann Peter,
effects, genuine and sham,
*elaboration (Auskomponierung),
epigones,
Epstein, Julius,
Erk, Ludwig,
exchange, (Quint- und Sext-Auswechslung),
, , , ,
exchange of voices (Stimmentausch),
nn. , ,
extrapolation (addition) of a root (Auswerfen eines Grundtones), ,
Fiedler, Konrad,
form, musical, , , n., , , , , ,
,
Frankfurter Zeitung, newspaper, n., ,

Franz Joseph, emperor of Austria,
Friedlnder, Max, n.
Frimmel, Theodor von, n.

Furtwngler, Wilhelm, vi n., n.

Fux, Johann Joseph,

Jenner. See Zeuner, Karl Traugott

Jews, Jewishness, ix
Jonas, Oswald, xxii, n., n.

Galitzin, Prince Nikolai, xiii,

genius, artistic, xiv, , , , ,
, , , ,
George II, king of England, n.
Germany, the Germans, xiv, , , , ,
, ,
Goethe, Johann Wolfgang von, xii, n., ,
n., n., , , , , ,
Grove, George,
Gutmann, Albert J., bookshop in Vienna,
viviii

Kahl, Willi, xv
Kalbeck, Max, ,
Kalischer, Alfred, xvi, ,
Kalmus, Alfred, xii
Kant, Immanuel, n.
Keller, Gottfried,
Kiel, Free Student Brotherhood of, xii
Klenau, Paul von,
Klopstock, Friedrich, n.
Krakow, Bilioteka Jagiellonska, n.
Kramer, Richard, n.
Krebs, Carl,
Kretzschmar, Hermann, viii n.,
Kullak, Theodor, nn.
Kurth, Ernst,

Halbmonatsschrift fr Schulmusikpflege,
journal, xv
Halm, August,
Handel, George Frideric, ,
Lessons for the Harpsichord, HWV :
Aria and Variations in BI,
Messiah,
Suite in F minor, HWV ,
Haslinger, Tobias, ,
Haydn, Joseph
Austrian National Anthem (Emperor
Hymn), xiii,
Piano Sonata in EI, Hob. XVI:, , n.
Piano Trio in E, Hob. XV:,
String Quartet in C, Op. , No. (Emperor),
Hehemann, Max, n.
hermeneutics, ,
Hertzka, Emil, vxii, , n.
Heuss, Alfred, ,
Hildebrand, Adolf von,
Hoboken, Antony van, n.
Hoffmann, E. T. A., xiv, ,
Hlderlin, Friedrich,
Homer, , n.,
Horace, n.
Httenbrenner, Anselm,

*layer (level), voice-leading (Schicht),

Lebert, Sigmund,
Lefebvre, Charles,
Lenz, Wilhelm von, , ,
Leidesdorf, M. J., publisher, n.
Lichtenberg, Georg Christoph, n.
*linear progression (Zug), n.
Ludwig, Otto,
Mahler, Gustav, ,
Mandyczewski, Eusebius,
Mann, Thomas, n.
Marx, Adolf Bernhard,
Marxism, n.
Mendelssohn, Felix, , n.
Song without Words, Op. , No.
(Venetian Gondoliers Song), xiii,

Song without Words, Op. , No. , xiii,

String Quintet in A, Op. , n.
Messchaert, Johannes,
Moormann, Ludwig, xv

Mozart, Wolfgang Amadeus, ,

Piano Sonatas
K in C,
K. in A minor, xv
K. in A,
Symphony in G minor, K. ,
Mozart-Jahrbuch, ix n.
Musikbltter des Anbruch, journal, viiiix
Nagel, Willibald,
Negro melodies,
*neighbor note (Nebennote),
Neumann, Karl Eugen,
New York, xii n.
New York Public Library. See Oster Collection
Niceratus,
Nottebohm, Gustav, , , , ,
Oberammergau, a
obligatory register (obligate Fhrung der
Tonlage),
occultism,
Oppel, Reinhard, xii n.
Oster Collection (New York Public Library),
xvxvi, xxii, n., n.,
n.,
Schenkers scrapbook xiv, n.
sketches for works analysed in Der Tonwille, n., n.
annotations in Schenkers personal copies
of Der Tonwille, n., n.,
n., n.
Paris
Bibliothque Nationale, n.
Conservatoire,
*passing, passing tone (Durchgang),
Paumgartner, Bernhard, xv
Pembaur, Josef,
perfection, artistic (Vollendung), ,
performance, musical (general matters),

Peters, C. F., publisher, vi n., ix n.

index
Plautus, n.
prolongation, xiv, , ,
Pult und Taktstock, journal, n.
reaching over (bergreiftechnik, bergreifen,
berhhung), ,
Reisenauer, Alfred,
rhapsodists (rhapsodes), n.
Riemann, Hugo, , , , n.,
Rietz, Julius, n.
Rntgen, Julius, n.
Sammelbnde fr vergleichende Musikwissenschaft, ix n.
Sauer & Leidesdorf, music publisher, n.
Saurau, Count Joseph Franz, n.
Schader, Luitgard, n.
Schenker, Heinrich
as accompanist, n.
criticized as analyst, viii n., xv,
dealings with publishers, vxii,
editorial principles, xiiixiv,
editions of music
Beethoven, Moonlight Sonata, autograph facsimile, viii n., n.,
n.,
Beethoven, piano sonatas, xixii, ,
, n.
Erluterungsausgaben (editions with critical commentary) of Beethoven
sonatas, xi, xiv
Op. , xvi, ,
Op. , , , ,
Op. , ,
polemical writing, vi, xiv, xv,
scrapbook. See Oster Collection

Schott, B., & Sons, music publisher,

Schubert, Franz, xiv,
autograph scores in facsimile, , n.
Gretchen am Spinnrade, D. , xiii, xv,
Ihr Bild (from Schwanengesang, D. ),
Impromptu in GI, D. (Op.), No. ,
xiii,
Meeresstille, D. ,
Moment musical in F minor, D.
(Op. ), No. , xiii,
text setting,
Wandrers Nachtlied, D. ,
Schumann, Clara, n.
Schumann, Robert
Of Foreign Lands and Peoples (Scenes
of Childhood, Op. , No. ), xiii,

Trumerei (Scenes of Childhood, Op. ,
No. ), xiii,
Schuppanzigh, Ignaz,
Schweitzer, Albert,
sexuality, n.
Simon, Josef, vii
Socrates, ,
sowing-harvest (Saat-Ernte) metaphor, ,
, ,
space, tonal (Tonraum), ,
Spitta, Phillipp, ,
Stark, Ludwig,
Staehelin, Martin, n.
Steiner, Sigmund Anton, n.
Stockhausen, Julius, n.
Strauss, Richard, xiv,

writings, published
Beethovens fnfte Sinfonie, in book
format, xiii
Beethovens neunte Sinfonie, n., ,
n., , n., n.,
Fnf Urlinie-Tafeln, xii
Das Meisterwerk in der Musik, x, xii, xv
n., , n., n., , n.
Der Tonwille: childrens issue, xiv; contents, xiixiv; Elucidations, xiv,
n., n., n., , ,
, ; materials deleted, xvi;
publication dates, v; publication history, v-xii; reception, xivxv;
reprintings, xi; The Mission of German Genius, vi, xii, xiv; title of, v
Neue musikalische Theorien und Phantasien, xixii, xiv: I Harmonielehre,
vii, , , , , , ; II1
Kontrapunkt, vol., , n., ,
; II2 Kontrapunkt, vol., , , ,
, , , , ; III Der freie Satz,
xii, n., n., n., n.,
n., n., n., n.;
writings, unpublished or posthumous
Die Kunst des Vortrags, n.
Freier Satz (early drafts of Der freie
Satz), xiv, xxii, , , ,
Musik-Kritik, xvi,
Schenker, Jeanette, ix, xiii,
Schenker Moritz, ix, xi
Scheu, Gustav, xi
Schiller, Friedrich, n., , , , ,
, n.
Schindler, Anton,
Schoenberg, Arnold, n.
Schopenhauer, Arthur,

Tausig, Carl,
Terence, n.
Thayer, Alexander Wheelock,

theory, as conventionally understood, ,

,
tonicization (Tonikalisierung),
Treitschke, Georg Friedrich,
Trk, Daniel Gottlob, n.
*unfolding (Ausfaltung, Auflockerung),
Universal Edition, music publisher, vxiii,
xv, n., ,
*Urlinie, n.
Urlinie-Satz, xiii, , n.
*Ursatz, n.
Verrocchio, Andrea del, n.
Vienna, n., ,
Austrian National Library, n.,
Gesellschaft der Musikfreunde, , n.
Krtnertor Theater, n.
Konzerthaus-Gesellschaft,
Philharmonic Orchestra, xiv
Stadt- und Landesbibliothek (Municipal
Library), n.
State Opera, n.
voice exchange. See exchange of voices.
Vogl, Johann Michael,
Vrieslander, Otto, viii n., ix, xv
Wagner, Richard, xiv, , , ,
Walsh, John,
Walzel, Oskar,
Winter, Hugo, ix, xi
Wolanek, Ferdinand,
Wolf, Hugo, n.
Xenophon,
Zeitgeist,
Zeuner, Karl Traugott,