Académique Documents
Professionnel Documents
Culture Documents
Bachs
Well-Tempered Clavier
In-depth Analysis and Interpretation
Siglind Bruhn
EDITION
Fachverlag fr
ORZ
eisteswissenschaften
Bruhn, Siglind.
J. S. Bachs Well-Tempered Clavier: In-depth Analysis and
Interpretation
Waldkirch: Edition Gorz, 2014.
kontakt@edition-gorz.de http://edition-gorz.de
Second, completely revised edition
(First edition Hong Kong: Mainer International Ltd., 1993)
ISBN 978-3-938095-19-5
Siglind Bruhn 2014. All rights reserved
No part of this book may be reproduced or transmitted in any form
or by any means, electronic or mechanical, including photocopying,
recording, or by any information storage and retrieval system without
the written permission of the author, except where permitted by law.
Bibliographic information, including detailed bibliographic data,
are obtainable from the Deutsche Bibliothek under http://dnb.ddb.de
Printed in Germany by rombach digitale manufaktur, Freiburg
Table of Contents
How to use this book
Introduction
a) The Well-Tempered Scale
b) The Clavier
c) The Preludes in Bachs WTC
d) The Fugues in Bachs WTC
Questions Regarding a Prelude
Questions Regarding a Fugue
Additional Information
7
9
9
11
13
15
17
19
21
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
47
52
66
77
91
105
116
129
139
150
161
169
180
192
204
214
226
235
246
256
270
283
292
303
315
6
Considerations and Model Answers for
WTC II/1 in C major
Prelude
WTC II/2 in C minor
Prelude
WTC II/3 in C major
Prelude
WTC II/4 in C minor
Prelude
WTC II/5 in D major
Prelude
WTC II/6 in D minor
Prelude
WTC II/7 in E major
Prelude
WTC II/8 in D minor
Prelude
WTC II/9 in E major
Prelude
WTC II/10 in E minor
Prelude
WTC II/11 in F major
Prelude
WTC II/12 in F minor
Prelude
WTC II/13 in F major
Prelude
WTC II/14 in F minor
Prelude
WTC II/15 in G major
Prelude
WTC II/16 in G minor
Prelude
WTC II/17 in A major
Prelude
WTC II/18 in G minor
Prelude
WTC II/19 in A major
Prelude
WTC II/20 in A minor
Prelude
WTC II/21 in B major
Prelude
WTC II/22 in B minor
Prelude
WTC II/23 in B major
Prelude
WTC II/24 in B minor
Prelude
Further Reading
Table of Contents
327
337
347
357
371
381
393
403
413
425
437
445
453
465
477
485
495
507
519
527
537
547
557
569
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
Fugue
325
330
340
351
363
375
385
397
407
417
429
440
448
458
468
479
488
498
510
521
530
540
549
560
573
579
questions to be pondered
additional information
discussions of possible answers to the questions
with regard to each of the 48 preludes and fugues
Abbreviations used:
SATB
UML
v1, v2, etc.
S
CS
dux
comes
M
CM
rh
var
inv
E
m. 51
mm. 13-31
TSD
T relative
t relative
I IV V
i iv V
r.h., l.h.
17
18
1.4a
1.4b
1.4c
1.4d
1.4e
20
Additional Information
1.1 What features in your piece help you
determine the prelude type?
Music is often defined as a sequence of sounds arranged chiefly on
three levels: the horizontal, the up and down in a tune, called pitch pattern;
the vertical, where several pitches coincide thus forming harmonies; and
the temporal, with longer or shorter, regular or irregular time spans from
the beginning of one tone to that of the next, called rhythm.
Each of these levels is usually carefully organized. The horizontal level
contains melodic units, phrases, and sections; this is what we are after
when analyzing structure. The vertical level distinguishes the ways in which
simultaneous tones relate to one another, defining harmony and texture.
Finally, on the level of time, the multitude of rhythmic values is packed
into measures of usually equal length in an orderly way, with a definite
hierarchy between their beats; this is called metric organization.
Although any piece of music will feature most if not all of these components, composers usually seek particular expression of their artistic idea
by molding these components so that a certain design is recognizable. For
such a design there is one basic choice to make: does he want to compose
along the lines of a standard model that can be recognized as such, or does
he prefer his piece to sound more like a spontaneous improvisation?
In the case of established structural models there are certain rules by
which composers must abide: dances are defined by certain metric and
rhythmic features, canons by a specific texture, sonata form by a particular
sequence of material, etc. In the case of a composition put together more
like an improvisation, they may choose one or two of the six components
mentioned above and shape their piece primarily according to these. In a
prelude, i.e., in a composition carrying a title that does not reveal anything
about the content of the piece, discovering this determining force is
therefore of vital importance for both listeners and interpreters.
a) Preludes determined mainly by harmonic processes
Whenever melody and rhythm are negligible or neutral, harmony is the
ruling factor. This rule of the thumb sounds simple enough. But are
melody and rhythm not always present?
21
22
Additional Information
Additional Information
23
24
Additional Information
Additional Information
25
c) The sections
The number of structural sections in a prelude is theoretically not
restricted, although it is rare to find more than six. Their boundaries are
always delineated as described above: a conclusion in the harmonic progression coinciding with a conclusion in the melodic line if there is one.
d) What exactly constitutes a structural analogy?
Structural analogies recall the simpler binary and ternary designs and
are therefore a very popular building principle in preludes. Analogies may
consist of as little as two or three measures but may extend to the length of
a complete section that is in some way recapitulated.
The simplest, but also least frequent, manner of creating an analogy is
by literally repeating several measures. More ingenious and more widely
used are variation, transposition, and correspondence.
In variations, several measures recur with modified features. Various
details may be altered: One or several of the voices may be written
in a more elaborate pitch and/or rhythmic pattern, the leading line
may be given to another voice, or an accompanying voice may
change its pattern. The features that will always remain the same in
literal variations are the key, the overall harmonic progression, and
the length of the segments.
Transpositions in their simplest form keep all the original details
intact but merely transfer them to a new tonal environment. If there
are, in addition, more than arbitrary changes in the details, we would
speak of a varied transposition.
In correspondences, the structure recurs in principle rather than
literally. The surface changes may be so considerable as to make it
difficult to spot the analogy at first sight. For a good understanding
of the structure, however, correspondences are just as important as
other analogies, not least because they are much more frequent.
Imagine the following structural pattern in an early section: two-bar
model with sequence + modulation to a new key + cadential close with twobar closing formula. Let us assume that the segment recurs analogously at
a later stage of the piece, initiated by a different two-bar model with its
sequence, that the modulation takes an extra measure to get to where it is
heading, and that the closing formula, although of equal length, features a
different melodic design in the treble. This is a subtle way of creating correspondenceand very rewarding to detect. Understandably, such analogies
only make sense if the surface designpitch pattern, rhythm, and texture
is neutral enough to draw sufficient attention to structural processes.
26
Additional Information
Additional Information
27
for the longer values. The main exception in the longer values: an
appoggiatura is always inseparably linked to its resolution, and the
keynote / leading note / keynote (dosido) formula is always legato.
Longer notes in a rather lively piece need further specification
whenever we are dealing with notes prolonged by a dot or tie. While
these are certainly longer, they are detached only where the prolongation itself (i.e., the value of the dot or tie) is of longer
duration. If the time value of the prolongation equals one of the
shorter notes in this composition, the note will be linked to the
following one in quasi legato. Thus in a composition where non
legato quarter-notes and quasi legato eighth-notes are a rule, a
half-note prolonged by a dot or tied into a quarter-note would be
detached from the following note; but a note of any value prolonged
by an eighth-note tie would not be detached.
b) How do you make decisions regarding ornament realization?
Ornaments in polyphonic compositions of the Baroque era are, more
often than not, decisive features in the material they embellish. Their
convincing execution depends on a smooth beginning, a speed that relates
to the rest of the piece, and an appropriate ending.
The most common ornament in Bachs fugues is the long trill, indicated
by one of several symbols: the abbreviations or t, the mordent symbol,
or one of the signs for compound trills. Here is how the long trill is played:
In keeping with the basic rule for ornaments in this epoch, it begins on the
upper neighboring note. However, there is a very frequent exception:
Whenever the trilled note is approached stepwise, i.e., whenever the main
note is preceded by the interval of a second from above or below, the trill
begins on the main note. Another exception, albeit much less frequent,
applies whenever an ornament decorates the first note of a phrase (see e.g.,
the initial note of the G-minor prelude from vol. I). In this case, too, the
ornament begins on the main note. The speed of the shakes is related to the
general motion in the composition in such a way that the value of each
ornamental note is preferably twice as fast as that of the fastest note values
that do not qualify as written-out suffixes.
While trills beginning on the upper neighboring note enjoy regular
motion throughout, those launched from the main note and therefore comprising an uneven number of notes hold the first trill note for the duration
of each of the following two-note shakes, in such a way that all further
upper auxiliaries fall on the stronger pulses within the trill, thus evoking
the effect of a long appoggiatura.
28
Additional Information
Whenever the note following the trill is its harmonic resolution and
falls on a strong beat, the end of the trill should be prepared (announced)
with a suffix, i.e., a turn to the lower neighboring note and back to the
main note. Conversely, no suffix is desirable if the resolution appears too
late, too early, or not at all. The first case happens if the trill is prolonged
with a tie; in this case the shakes must stop short on the last main note
before the bar line, so that the tie can be fully appreciated. The second case
occurs in dotted note patterns; here the shake ceases preferably shortly
before the dot. Finally, if the trill is succeeded by a rest or a leap not
providing harmonic resolution, the shake should continue up to the very
end of the ornamented note value, ending again on the last main note.
Any ornament appearing in the subject or one of the counter-subjects
must be regarded as integral to this component of the thematic material.
Regardless of whether or not the composer repeats the ornament symbol in
each recurrence of this component, the performer should retain such a trill
throughout the work (provided the subject or counter-subject does not
appear in drastically modified shape).
Additional Information
29
Secondary features
Beyond these basic considerations of chordal relationships there are
two secondary features that must be given special attention since they will
regularly repeal the rule mentioned above: the sequence and the pedal note.
The term sequence describes the process by which a modelthis can
be a melodic unit or, in the case of a harmonically-determined work, a
group of chordsis repeated once or several times on different pitch
levels. In such a case, the primary dynamic process is always established in
the model. The sequential groups must follow the same pattern, or else the
relationship would not be comprehensible. The model thus sets an example
that will be observed in each sequence, regardless of the actual harmonic
relationships there.
The pedal note also creates a law of its own. In the majority of cases,
it comes as a bass note, usually entering on the dominant of the home key,
which is sustained (or, more common on instruments other than the organ,
reiterated) for several measures while the harmonies made up by the other
voices sound reluctant to surrender to the truth that the piece is soon to
come to an end. Pedal notes have their roots in Baroque organ music. Yet
while a sustained or reiterated note on the organ will objectively maintain
a constant dynamic level, subjectively or psychologically it will work its
way ever deeper into the listeners consciousness. This is the reason why
a basically simple eventthat of a sustained or reiterated noteinsinuates
a gradual, smooth but persistent increase of tension.
Finally, any truly melodic event overrules the processes of chordal
tension in pieces determined by harmonic relationships. If a melodic unit,
however small, has enough character of its own, it may create motions that
are momentarily independent from the underlying harmonic process.
30
Additional Information
Additional Information
31
32
Additional Information
Additional Information
33
34
Additional Information
Additional Information
35
36
Additional Information
Additional Information
37
Although this basic truth is valid in all fugues, the degree of impact
exercised by the subject on its surroundings may vary slightly in each case.
Some of the common constellations are: the subject can be spread regularly
across the fugue and always be accompanied by other prominent musical
ideas. Thus, although leading, it will appear as one in a group. As the
subject retreats momentarily, other characteristic motifs may develop.
Thus, the subject may appear as the leader of one team that is contrasted
with another (admittedly less important) team representing a different
color. Last but not least, the subject may not have the same importance at
all times but, e.g., demand more and more attention as the fugue develops.
It may do so by presenting its statements in more powerful variations, or
by appearing in several voices almost at once.
a) When you discuss the subject statements, what are you looking for?
A fugue may contain any number of subject statements. The twentyfour entries in the opening fugue of the Well-Tempered Clavier represent
by no means the largest number. These statements are usually named after
the voice in which they sound. The following conventions are practical:
In a 4-part fugue, the established names for the voices are those of
the vocal ensemble: soprano, alto, tenor, bass (abbreviated S A T B).
In a 3-part fugue, using a selection from this 4-part combination
would seem arbitrary. A good solution is to refer to the voices as
upper voice, middle voice, and lower voice (abbreviated U M L).
In the few 5-part fugues, the least problematic solution is to count
them from v1 to v5 (i.e., from first or uppermost voice to the fifth or
lowest voice. The attempt made by some analysts to retain the vocal
terms by introducing alto I and alto II or tenor I and tenor II has led
to lengthy arguments among scholars. Such a quarrel seems somehow quite beside the point, and the v1-to-v5 option may prevent
further disputes.
b) What kind of changes can you expect to occur in the subject?
In the course of the fugue, the subject may appear in various guises.
This is important to remember when trying to locate all the statements.
One of its intervals may be modified to adjust to a different
harmonic background. In many fugues, this is the case in the second
and fourth entries, i.e., in the answer, the entries beginning on the
dominant. An entry with such an interval adjustment is called a
tonal answeras opposed to the real answer in which all
intervals remain unchanged.
38
Additional Information
The end or the beginning of the subject (or both) may feature
variations in pitch and rhythm. The most frequent modifications are
prolonged or shortened first notes and delayed final resolutions.
The subject may appear partly or entirely in rhythmic variation (e.g.,
showing dotted-note figures where the original was in regular motion)
or in metric variation (e.g., beginning on a weak instead of a strong
beator vice versaand continuing slightly off beat).
The subject may appear upside down. This is called an inversion.
Finally, the speed may appear changed in such a way that all note
values are doubled and the subject is twice as long. This modification is called the augmentation. Also possible, although less
frequent, are entries in diminution where the time value of all
notes is decreased, usually by 50%, and the subject therefore takes
only half the time it did originally.
Additional Information
39
soprano:
S
alto:
S
CS1
tenor:
S
CS1
CS2
bass:
S
CS1
CS2
CS3
There are some, though not many, fugues with no counter-subject at all.
This happens if the subject is so dominant that it does not accommodate
any rivaling ideas. In other cases, a counter-subject may accompany only
a few subject statements and then disappear, never to be heard again.
When trying to find out whether a line qualifies as a counter-subject,
you should ask yourself the following questions:
Does it recur? A melodic detail that only sounds once will not be
given a formal name.
Is it independent? A line that duplicates part of the subject in
either rhythm or pitch pattern is not regarded as independent.
Is it melodic? A neutral scale segment or a melodically meaningless group of notes would not be recognized even if it recurs.
a) How can you determine the length of a counter-subject?
The length of a counter-subject will always roughly correspond to that
of the subject. Where exactly it begins and ends can best be decided when
comparing it with later appearances.
40
Additional Information
Additional Information
41
42
Additional Information
Additional Information
43
c) How do you decide what relation to choose between the tempo of the
prelude and that of the fugue?
All through the Baroque and Classical era, pieces of music that jointly
form a larger work (i.e., the movements of a sonata or suite, or prelude and
fugue, prelude and toccata, etc.) were conceived as being related with
regard to their pulse. This concept is comparable to the much more familiar
demand of a rhythmic relationship in architecture. Imagine a group of
buildings constructed in the same style but to the order of different patrons
and therefore with unrelated measures (height and width, angle of roof,
pattern of windows and doors, etc.). Such buildings will appeal to the eye
as individual houses, each of them a unit in its own right. A complex of
related buildings, howevera castle, a temple district, or a similarly
structured ensemblewill display a subtle yet very definite relationship
between all its measurements.
As music unfolds in time, which is measured in pulsations, the
equivalent factor of a rhythmic relationship is created by proportions of
the pulses. For this purpose, any of the metric values in a piece of music
can be used as a point of reference: the actual metric value indicated in the
time signature, the larger unit of a half or whole measure, or the smallest
note value appearing in the piece.
A tempo proportion is then created in such a way that one note value of
the first piece relates to any value of the subsequent piece by being either
equally, half, or twice as fast (1:1 /1:2 / 2:1), or three times as fast or slow.
Even an implied but not actually sounding value, like the triplet fraction in
a piece not featuring triplets, can be related to a pulse in another movement. This kind of pulse translation may be preferable particularly in the
case of successive pieces of the same time signature and rhythmic organization, which might otherwise present too little distinction. To give a few
examples (out of many possible ones):
The quarter-note beat of a fugue may equal
a quarter-note in the prelude
an eighth-note in the prelude
a half-note in the prelude, etc.
an entire measure in the prelude
three eighth-notes / three quarter-notes / three sixteenth-notes in the
prelude
a virtual (imagined, not materializing) triplet eighth-note or triplet
quarter-note in the prelude.
44
Additional Information
This rule of thumb will almost always help to determine the length of
the first section (or exposition). It often even enlightens the entire plan
of the composition, with all its consecutive rounds. Concerning the number
of rounds, you can expect at least two. As for the maximum, there is no
established rule. (There was a time when people, influenced by theories of
the late nineteenth century, tried to detect a three-section structure in every
fugue. Despite the fact that this has long since been unmasked as a
misconception originating in sonata and ternary forms, these theories still
seem to linger in some minds.)
Additional Information
45
As soon as each voice in your fugue has entered with the subject, the
full ensemble is reached. However, while the musical debate relaxes (as
in an episode) or embarks on a fresh start (as at the beginning of a new
section), one of the voices often takes a rest. A subject statement that
sounds accompanied by less than the full ensemble may therefore indicate
the beginning of a new section.
Analogies in a fugue are not necessarily obvious at first sight, although
some are. What is required is that a sequence of events (like a number of
subject entries, accompanied by certain material, perhaps including an
episode) recurs later in the piece in the same or a very similar order. This
similarity does not refer to the position of the material in the ensemble; so
the order may be the same with different voices being used to present the
materialwhich is why some analogies are at first hard to find.
b) What do the harmonic features tell us about the design?
While a larger part of the piece will always appear closely related to
the main key, there are usually passages that show a harmonic digression
to either a different tonal area and/or to the opposite mode (to the minormode area in a major-mode fugue or vice versa).
To determine the harmonic background of a subject entry, it is best to
look at both its beginning (i.e., the first two or three notes, to avoid
misinterpretations in tonal answers) and its end. Subject entries within one
section often appear in tonic / dominant / tonic / dominant progression. A
change in this pattern is worth noting. Significant cadences are usually
those that appear as obvious formulas or cadential-bass patterns outside the
main material of the fugue.
c) How can you sketch the design of a fugue?
Such a design is usually sketched using a horizontal column for the
measure numbers, corresponding horizontal lines showing the material
sounded in each of the voices, and vertical columns for the juxtapositions.
In the colored graphs included in this book, bricks of different shade and
intensity represent subject and counter-subject entries, roughly like this:
mm. 1
S
A
T
B
S
S
CS1
5
S
CS1
CS2
7
8
S
CS1
CS2
CS3
10
episode
(rest)
etc.
46
Additional Information
50
WTC I/1
C major
51
In keeping with the law defining dynamics in sequences, the same relationship must apply to the following two chords, on a softer level because the
sequence descends. From the last chord of the sequential pattern onward
there follows a further gradual release of tension. The dynamic analog to
the process in this portion of the prelude may be expressed in these terms:
mf + - mp + - mf - mp - mp - p+ - p.
A similar development occurs in the third section. Like the preceding
one it starts with a sequence, and again, the relationship between the first
and the second chords in the model is one of relaxation. The release of
tension is even stronger here than in the earlier sequence as mm. 12 and 14
each consist of a diminished seventh chord resolving onto an inversion of
the supertonic (m. 13) and of the tonic itself (m. 15). At the end of this
sequence the harmonic tension therefore appears already abated to something very close to the softest shade used in this piece. The following four
measures, a transposition of mm. 8-11, should portray a dynamic outline
exactly like the one in the corresponding measures so as to assist listeners in
grasping this analogy. The concept of mm. 12-19 is thus approximately:
poco f - mp + - mf + - mp - mp - mp - p + - p.
The fourth section is almost as long as the first three sections together.
The emergence of the dominant pedal in m. 24 serves to divide it into two
subsections. The first of these subsections, from m. 20 to the downbeat of
m. 24, ends in an imperfect cadence. The harmonic development sets out
from the C-major seventh chord (m. 20) but then leads away from the tonic
in bold steps. The process includes two diminished-seventh chords; in
addition, there is a hint of an independent bass line, which enhances the
sense of urgency. Upon closer inspection, the end of this bass line reveals
a circular movement preparing the beginning of the pedal note by sounding
both the natural leading-note (from the semitone below) and the artificial
one (from the semitone above) to the dominant keynote G. The audacity of
these harmonic steps can best be conveyed in the following dynamic plan:
mp - mp + - mf + - poco f +.
After this forceful increase, the next subsection begins with a sudden
hush to a level hardly above piano. From here the pedal note takes effect. Its
typical gradual and smooth dynamic growth continues not only while the
bass remains on G but all the way to the end of the piece, thus closing the
prelude on something like a triumphant forte chord. The dynamic balance
in this section may therefore be represented quite convincingly as follows:
p + - mp - mp - mp + - mf - mf - mf +- poco f - poco f - poco f + - f - f.
52
WTC I/1
The graph attempts to show the processes determining this prelude.1
Carl Czerny added an extra measure. He argued that first, it seemed highly improbable, for
a composer as conscious of subtle numeric balancing as Bach was, to have written a piece
consisting of the uneven number of 35 measures; second, a bass line with a diminished third
such as the one in mm. 22-23 seemed melodically incorrect and would therefore require a
chromatic link. Czerny therefore inserted a tonic six-four chord to correct Bachs error,
and by doing so destroyed the forceful tension built up here with so much ingenuity.
C major
53
54
WTC I/1
the harmonic progression underlying the subject statements in mm. 2-4 and
5-7). As far as the subtler harmonic progressions from one eighth-note to
the next are concerned, there are so many variations throughout this fugue
that it is not possible to state one unequivocal solution.
Ludwig Czaczkes in his analytical
book on Bachs WTC made his own
choice when he reconstructed in fourpart texture what he regarded as this
subjects harmonic basis.
When deciding on subtle dynamic
processes in the subject, the aspects to
be taken into particular consideration
are the two rhythmic phenomena and their respective harmonic functions:
the syncopation is rhythmically stronger, but it falls on the harmonically
rather weak tonic-to-dominant movement; the dotted eighth-note is rhythmically weaker but supported by the stronger harmony, the subdominant.
The preference of one climax over the other therefore rests on whether
interpreters feel the harmonic process more strongly than the rhythmical
one, or vice versa.
If a performer chooses option (a) for the phrase structure and reads the
subject as a unit that allows for no further subdivision, only one climax has
to be determined, which will come smoothly prepared by an increase in
tension and be followed by a relaxation. Performers perceiving the subject
as made up of two halves (b) have to decide which of them is stronger and,
consequently, which of the two shorter tension build-ups is more powerful.
In the latter case the subject would contain two releases, with the E at the
end of the first subphrase as a transitory solution, i.e., not quite as soft as
the final E.
The design of the C-major fugue has often been called programmatic:
as the first of a collection of twenty-four fugues it contains twenty-four
subject statements. These appear as follows:
1.
2.
3.
4.
5.
6.
7.
8.
mm. 1- 2
mm. 2- 4
mm. 4- 5
mm. 5- 7
mm. 7- 8
mm. 7- 8
mm. 9-10
mm. 10-12
A
S
T
B
S
T
A
B
9. mm. 10-12
10. mm. 12-13
11. mm. 14-15
12. mm. 14-15
13. mm. 15-16
14. mm. 15-16
15. mm. 16-17
16. mm. 16-18
A
T
A
T
B
S
S
A
17.
18.
19.
20.
21.
22.
23.
24.
mm. 17-18
mm. 17-18
mm. 19-20
mm. 19-20
mm. 20-22
mm. 21-23
mm. 24-25
mm. 24-26
T
B
T
A
S
T
T
A
C major
55
56
WTC I/1
Only twice is the density of the material in this fugue briefly interrupted. As the subject is absent in these measures, they qualify as episodes:
E 1 = m. 13 (last three eighth-notes) to m. 14 (first eighth-note),
E 2 = m. 23 (after the first 16th-note) to m. 24 (first eighth-note).
Furthermore, the fugue ends with two subject-free measures: the last
statement concludes on the first 16th-note of m. 26, thus: E3 mm. 26-27.
None of these subject-free measures is even remotely related to the subject;
neither are there any episode motifs. The two subject-free passages within
the fugue are both clearly identifiable as cadential closes. The second half
of m. 13 features typical closing formulas in all three sounding voices,
leaving no doubt that something is about to come to an end here. In mm.
23-24, soprano and bass once again present those typical patterns while the
inner voices join in neutral, quasi-chordal style.
Surprisingly, the ending of the composition avoids all these established
formulas. Two measures earlier Bach had already started a tonic pedal,
thus making it clear that the piece is approaching its end. Now he allows
C major
57
the other three voices to continue freely, with harmonic and melodic developments independent of both material bounds and pattern restrictions. One
after the other the voices then come to a halt: the tenor withdraws first,
upon which the alto becomes less eloquent. The soprano is the last to
remain active by running up to the high C.
The final note in the soprano, the high C, deserves a mention. This note
has not been previously reached in this composition. As is widely known,
it marks the upper limit of the keyboard instruments in Bachs time, just as
the preludes final low C marks the confines at the other extreme. These
two tiny details are another hint that Bach, in writing his twenty-four
preludes and fugues, aimed to demonstrate in full all that was possible on
this instrument.
The dynamic role each of the three subject-free passages plays in the
development of this fugue will have become apparent from what has been
said above. The two inner episodes serve as cadential closes and thus have
a relaxing tendency. The two final measures, on the contrary, succeed in
creating somethinghowever smallof their own, thus contributing to a
build-up toward a final climax.
Both the overall stepwise motion and the rhythmic structure with its
four different note values and frequent syncopations suggest that the basic
character of this composition is rather calm. Within the overall frame of
this character, the chain of three leaps in the subject adds an element of
contrast that should be feltand playedas such.
The most appropriate tempo is a calmly flowing one: flowing enough
so that the quarter-notes, rather than the eighth-notes, are felt as a pulse;
calm enough so that the 16th-notes still sound serene. Regarding the
relative tempo of the fugue to its prelude, retaining the beat would lead to
a somewhat dull result since both the time signature and the 16th-note
motion are the same in both pieces. Therefore, a proportion of 3:2 or 2:3
seems better suited. [Approximate metronome setting: prelude beat = 80 or
90, fugue beat = 60.]
The articulation in the fugue is mainly legato. One conspicuous (and
regular) exception occurs with the three leaps in the subject, which, being
consecutive jumps, must be detached. Similarly, leaps and cadential-bass
patterns in non-subject portions of any voice are also exempt from legato
articulation.
The fugue contains three ornaments, namely in mm. 13, 18, and 19.
The one in m. 18 is indicated in the Urtext as deriving from Bachs own
manuscript. It is a mordent that begins on the upper auxiliary note and
touches down twice on the main note, thus resulting in four regular notes
58
WTC I/1
A-G-A-G. The two other ornaments both embellish typical soprano closing
formulas. As the Urtext printing in brackets indicates they do not appear in
the autograph but were added in various early copies. Ornamenting this
characteristic dotted note in a cadential formula was common practice, and
performers should include some kind of embellishment in any case. The
simple mordent (as suggested in m. 19 by the symbol in brackets) is a more
likely solution in both cases than the compound ornament recommended
by the small-print version in m. 13 since an anticipation of the final note in
the soprano formula (see A-A mm. 13-14 and D-D m. 19) normally replaces
and thus precludesa trill with a suffix ending.
The design of the fugue is clearly laid out. Bach orders the voices
presenting the subject, and the sequence of single and stretto statements, in
a consistent pattern:
mm. 1-7
entries in all four voices, no stretto
mm. 7-14
entries in all four voices, two with stretto partners
mm. 14-19
entries in all four voices, all with stretto partners
mm. 19-27
two stretto pairs including three of the voices, followed
by one repeated (redundant) stretto pair in the coda.
Within these four groups, the coupling of voices in stretto also seems
to follow a plan: the second round joins the higher and the lower voices
in each register (S + T, A + B), the third round groups the adjacent and the
outer voices (S + A, A + T, T + B, B + S), the fourth round leaves out the
bass completely but uses the next lowest voice, the tenor, in all three
combinations.
The complete ensemble is built up four times: The first four entries
produce the regular assembling of all voices involved in the fugue (see
mm. 1-7: from one to four voices). The ensemble is consequently reduced
to three voices in the first stretto statement. In both the following alto entry
and the next stretto, one of the voices is resting, so that the full ensemble is
again reached at the fourth statement of this group (see mm. 7-13, from
three to four voices). After the cadential close in A minor, the ensemble is
momentarily reduced to only two voices. This marks the return to C major
as a genuine new beginning. The following three strettos then sound in full
ensemble (see mm. 14-19, two to four voices). Finally, the stretto in mm.
19-20 that overlaps with the closing formula has the soprano pausing. The
full ensemble is restored with the ensuing statement (see mm. 19-27, from
three to four voices).
The harmonic progression within this fugue leads first from C major
to its relative A minor, confirmed by the cadential close of mm. 13-14. A
new start in C major is diverted after four entry pairs to D minor. This key
C major
59
is again confirmed by a closing formula (m. 19), although the two outer
voices alone hold on to this tonal anchor while the two inner parts continue
with stretto statements of the subject. As can be seen from both the keys of
the stretto group-leaders and the prolonged bass notes (see mm. 19-20: D,
21-22: G, and 24-27: C), the fugue then reverts to the home key, which is
confirmed by a cadential close in C major in mm. 23-24. In terms of the
harmonic development, the last four measures thus appear as a coda, since
the final cadence has already taken place and the tonic bass note sounds as
an extended pedal note.
There are several interesting analogies within the four sections. Both
the first and the second sections feature a harmonic progression launched
from the tonic and followed by entries on the unusual repeated dominant.
mm. 1-5: A (I), S (V), A (V) mm. 7-12: S + T (I), A (V), B + A (V)
The first sections two initial entries relate to their counterparts in the
second section (counting again the strettos group leader as the relevant
voice) in inversion, as do the third and fourth entries of these two rounds.
mm. 1-7: A, S T, B
mm. 7-13: S, A B, T
The first stretto of the third section (mm. 14-15) is analogous to the
first stretto of the second section (mm. 7-8) insofar as both the pitch level
and the distance between the second entry and its leader (two eighth-notes)
are identical. The second stretto in the third round (mm. 15-16) appears as
an intensified variation of the corresponding second entry in the second
round (mm. 9-10), again with the same pitch level. Hence:
mm. 7-10:
mm. 14-16:
stretto (S + T, on C + G)
stretto (T + A, on C + G)
distance 2 eighth-notes
distance 2 eighth-notes
single entry (A) on G
stretto (B + S) leader on G
60
WTC I/1
Finally, the fourth stretto of the third section (mm. 17-19) brings a
modulation into a new key, as does the fourth entry of the second section
(mm. 12-13); in addition, both the group leader here and the single entry
there sound in the dominant of the key of their destination.
mm. 12-13
mm. 17-19
fourth entry of section II
fourth stretto of section III
on V of A minor
leader on V of D minor
The first section shows a gradual, constant build-up of the ensemble
without any interruption by episodes. The tension rises steadily but, in the
absence of additional intensifying factors, reaches only medium level. The
second section, consisting of a twofold stretto-plus-single-entry sequence,
indicates a slight twofold relaxation. The first is supported at its end by the
closing formula (see soprano mm. 9-10: GF-G), while the second is
enhanced by the modulation to the minor key and the ensuing resolution
within the cadential close. The third section builds up tension from two to
four voices in the densest imaginable stretto setting. The dynamic processes are similar to those in the first section but considerably intensified.
Tension reaches its height at the end of this round; here the modulation
with its final turn to D major, the Picardy-third version of the expected D
minor, with the elaborate cadential formulas and the impatiently overlapping first stretto of the ensuing section create a supreme climax. The
fourth section resembles the second insofar as it appears made up of two
halves. After two entries that modulate back from D through G to C, thus
initiating a harmonic relaxation, the cadential measure 23 represents an
obvious caesura. The following coda replaces the expected stretto + single
entry with a stretto over an extended pedal note, but complements this with
two measures of subject-free development.
There are thus striking relationships between the two halves of the
fugue: The first section builds up tension. This build-up extends into the
beginning of the second section but is then not developed further, owing
both to the divided layout of this section and to the regressing harmonic
development. The third section builds up tension toward the overall climax
of the fugue. This elevated tension is continued into the overlapping beginning of the fourth section, after which it subsides gradually, both because
of the divided layout of this section and its softening modulation and
cadential close.
A comparison of the shorter original version of this prelude with the C-major prelude
reveals many parallels: As in the C-major prelude, the first 24 measures of the C-minor
prelude all show a half-measure model followed by its repetition. The extension of the first
section, the chain of descending sequences in the modulating second section, and a
protracted dominant pedal in the final section are all identical in both pieces.
61
62
WTC I/2
C minor
63
The practice described above was quite well-known in Baroque times. Performers were
expected to understand, from their immediate grasp of the musical sense, whether or not
an arpeggio included the upper note, and then play accordingly. The fact that there was
obviously no need to specify this in writing may indicate that musicians usually did
understandor, perhaps just as often, that they were keyboard players studying with the
composer and thus could be taught during the lesson.
64
WTC I/2
In this matter it seems difficult to decide which of the two versions reflects the composers
true intention. The text of the initial Little Prelude shows m. 18 with two equal halves and
a repeated bass note, while the version preferred in todays Urtext editions of the WellTempered Clavier has a bass note B on the third beat of this measure. The latter version
certainly sounds harmonically more convincing, but it does so at the expense of continuity:
this measure diverges significantly from the pattern observed throughout the remainder of
this part of the piece.
C minor
65
66
WTC I/2
mm. 21 22
mp + mf
mm. 28
f
30
pf
32
mf
23
mf
24
25
mf + pf
26 27
pf + f
28
f
34a 34b 35 36 37 38
pf mf + mf mp + mp p
The graph sums up the tension development throughout the entire prelude:
C minor
67
iv
i6
V9
68
WTC I/2
There are altogether eight subject statements in this fugue:
1. mm. 1-3
M
5. mm. 15-17
M
2. mm. 3-5
U
6. mm. 20-22
U
3. mm. 7-9
L
7. mm. 26-28
L
4. mm. 11-13 U
8. mm. 29-31
U
Apart from the interval adjustment in the tonal answer, the subject
appears unchanged throughout the fugue. Both the interval variation in the
major-mode statements and the Picardy third at the end of the last entry are
regular features. The fugue does not contain any strettos or parallel entries.
It features two counter-subjects, only the first of which is truly independent.
CS1 makes its first appearance immediately after the initial subject
statement; i.e., it stretches from the second 16th-note of m. 3 to
the downbeat of m. 5. The conspicuous change in the rhythmic
pattern (only 16th-notes in m. 3, only eighth-notes thereafter) and
the pitch level (see the tenth interval C1-E 2 in m. 3) reveal this
counter-subject as structurally conceived with two contrasting
subphrases. The first segment presents a descending scale that
describes a decrease of tension. The longer second segment also
moves essentially in falling direction, with its most likely interpretation being an even more distinct diminuendo. (Climaxes on
either the lower C, the target of the descending scale, or the
admittedly interesting F are inadvisable. As both notes coincide
with tension peaks in the subject, stressing them would counteract polyphonic clarity. The rhythmic evenness in the countersubjects second half already weakens its impact in the context
of simultaneous lines; an additional dynamic parallel would
offset the requirement of independence.)
CS2 is introduced against the next subject entry. It starts belatedly but
concludes together with the other components on m. 91. Rhythmically, CS2 is almost entirely attuned to CS1. Nor is its pitch
pattern, which undergoes several changes in the course of the
fugue, very significant. Thus the subjects second companion
stages little contrast of its own. The overall impression in the
fugue is of a dialogue between the subject and one countersubject with a homophonic accompaniment.
The following sketch shows the phrase structure and dynamic design
created in the combination of the subject and its two counter-subjects.
C minor
69
Note that in the stretto imitation used in these episodes, the [decreasing] chromatic ascent
of this extension coincides with the [increasing] regular eighth-notes that open the head motif.
This takes many performers unawares: they lose sight of which voice is leading and when.
70
WTC I/2
C minor
71
subject in M
E1
E4
subject in L
subject in U
E2
E5
subject in U
subject in L
E3
E6
subject in U
The harmonic progression in this fugue leads from C minor to the
relative major key, E major (in the fourth subject entry), but returns to the
home key very soon thereafter and never leaves it again.
72
WTC I/2
The dynamic tension curve in the first section shows a gradual increase
through the first three entries, caused by the buildup of the ensemble and
the increasing gesture of the first episode. There follows a slight decline
during the second episode and in the fourth entry, which is not only redundant but also sounds in the major mode, thus appearing more relaxed than
the original minor-mode version. As this fugue is not dramatic but rather
cheerful and dance-like, the rising and falling dynamic developments in
this first section are not particularly strong. The second section repeats this
dynamic pattern: the first three subject statements, including the episodes
linking them, increase very smoothly. (The second of these episodes, E5,
begins by decreasing tension before it realizes that this is premature and
that there is one more entry to come in the buildup.) The cadential close
after the second sections third entry brings the tension release, so that the
redundant entry, coinciding with the coda, sounds like a softer afterthought.
The relationship between the two sections is one of enhanced repetition.
The second section sets out in a more involved way with its first subject
statement accompanied by both counter-subjects. The episodes are longer,
thus creating a higher degree of anticipation for the next subject entry. The
final entry, in its detachment from the fugues polyphonic texture, may
depict more of a retreat than the last entry of the first section, which
appears lighter mainly because of its changed mode.
The first cadence ends at m. 71. This cadential close lies embedded in
a melodic flow that continues uninterrupted. A change of surface pattern
occurs in m. 8, where two melodic voices lead into the inverted-voice
texture of mm. 9-15. In this sense, this initial cadence is only an indirect
indication of a subordinate structural ending within a larger context. There
is, therefore, no caesura, and no cut after the reappearance of the tonic.
Exactly the same holds true for the following harmonic progression which,
now in the tonal realm of G major, draws to a cadential close at m. 151.
Here again, the melodic pattern continues through another measure before
giving way to a continuation in inverted voices that marks the beginning of
a new harmonic development.
Owing to the 3/8 time, the structural units determined by these cadences
appear short and, with their eight-measure extension, supremely regular.
The phrase in mm. 25-31 is the first not to be followed by such a voice73
74
WTC I/3
swapping link. Instead, the second half of the cadential close serves as the
beginning of a new development. The following diagram shows the
phrases and their tonal areas in the entire prelude. The bridging measures
that link consecutive phrases by prolonging the tonic of a cadential close
before a renewed change of voices do not harmonically belong to either of
the closed progressions; they are thus deliberately omitted here. The
graphic arrangement tries to visualize the harmonic progressions.
1. mm. 1-7 C major
2. mm. 9-15 G major
3. mm. 17-23 D minor
4. mm. 25-31 A minor
5. mm. 31-35 A minor/D minor
6. mm. 35-39 D minor/G major
7. mm. 39-43 G major/C major
8. mm. 43-47 C major/F major
9. mm. 47-53 F major
10. mm. 55-61 C major
11. mm. 63-73 G pedal
12. mm. 75-83 C major
13. mm. 87-104 G pedal, resolving to C only in the final measure
C major
75
76
WTC I/3
C major
77
6
4
Baroque polyphony, in contrast both to the contemporary dance types and also to the
music of the ensuing period, knew frequent metric shifts of its thematic material. This is
especially true in quadruple time where a subject or motif first introduced in the middle of
the measure could be placed at the beginning of a measure, and vice versa, in later
statements within the same piece.
78
WTC I/3
There are only two rhythmic values: eighth-notes and 16th-notes. The
pitch pattern is characterized by leaps rather than steps; note the melodic
broken chord at the beginning (E-C-G) and the alternating sixth and
seventh intervals at the end. There are few steps; all can be identified as
written-out ornaments: a turn in m. 1 and an inverted mordent in m. 2. In
the subjects harmonic progression, the active step to the subdominant falls
on the downbeat of m. 2. An analysis of the underlying chord progressions
that Bach uses later in the fugue reveals the G in m. 2 as an appoggiatura
to the following F. These two notes, G and F, thus form a pair that may
under no circumstances be separated by either phrasing or articulation.
The climax in the subject occurs unmistakably on the downbeat of the
second measure. Here, two powerful tension-enhancing features coincide:
the appoggiatura and the active harmonic movement from the tonic to the
subdominant (or, more often in this piece, its relative minor on ii). The
peak note E (which may tempt all those who connect strong feelings with
high pitches) is in reality only part of a broken-chord pattern on the tonic
and therefore melodically and harmonically insignificant. The dynamic
curve in the subject thus begins with an energetic crescendo through the
first segment up to the downbeat G. This crescendo should develop evenly
and not burst out too early, so as to give the E and C enough impetus
toward the appoggiatura G. In the fairly abrupt tension decay that follows
from this appoggiatura to its resolution, approximately half of the tension
is lost. The remainder is then released gradually throughout the series of
leaps.
There are twelve subject statements in this fugue:
1. mm. 1-3 U
5. mm. 14-16 L
9. mm. 42-44 U
2. mm. 3-5 M
6. mm. 19-21 M
10. mm. 44-46 M
3. mm. 5-7 L
7. mm. 24-26 U
11. mm. 46-48 L
4. mm. 10-12 U
8. mm. 26-28 M
12. mm. 51-53 U
In three of the subject entries, nos. 4, 7, and 12, the upbeat eighth-note
is replaced by three 16th-notes, while in statement 10 it appears split into
two 16th-notes. Metric displacementsa beginning on the last eighth-note
of a baroccur in statements 7, 8, and 12. The subject does not appear in
either stretto or parallel but takes three counter-subjects. CS1 is introduced
against the subjects second entry (see mm. 3-5: U) and remains a faithful
companion ever thereafter. It is exactly two measures long, a little longer
than the subject, as it begins slightly earlier in the bar. In its full scope it
C major
79
While the mind may easily accept this truth, the fingers seem more reluctant and often find
it difficult to resist stressing the longer note.
80
WTC I/3
upbeat followed by a descent in longer note values. These features and the
long release of tension resulting from this melodic shape reveal its
relationship with CS2. This impression, however, is weakened both by the
harmonic progression and by the concluding cadential-bass steps.
The sketch shows the phrase structure and dynamic design in the
primary thematic material of this fugue:
C major
81
in tension be also increased. Also in E1, the lower voice presents a motif
that can be traced back to the first counter-subject. It sets out with the same
inverted-mordent figure, followed by ornamental waves recalling the final
groups of CS1 in inversion. This motif is also frequently used within the
fugue and will be called M2. In terms of tension, M2 contains very little
active power. The short upbeat-like impulse in the inverted-mordent figure
is followed by a long, subdued drop in tension.
E2 is related to E1, but both M1 and M2 recur in considerable transformation. There are some significant changes that create an entirely different
character: In M1, the leader of the imitative pattern is now the middle
voice and the tie prolongations in the motif are replaced by rests, creating
a definite interruption of the tension. Moreover, the imitating voice does not
follow its leader but restates the second, relaxing half of the motif. Toward
the end of the episode, both voices abandon the motivic context altogether
and join in a cadential figure. M2 sounds in the upper voice where it no
longer appears as a sequencing one-bar figure, but is extended to a two-bar
curve. In free inversion, it is pushed upward to a slide-decorated peak; the
effect of this climax is enhanced by the fact that it sounds in a diminishedseventh interval to the C in the lower voice. The ensuing release ends in
a so-called female extension, a melodic tail after the harmonically resolved strong beat, on the fifth 16th-note of m. 14.
None of the subject-free passages serves exclusively as a cadential close.
But the final 2 measures (E7) and the first 1 measures of E4 both
present non-motivic material that leads to perfect cadences with distinct
closing formulas (see the cadential-bass patterns in L: mm. 21-22 and 55 as
well as the two typical melodic formulas in U: mm. 22 and 55). In the case
of E4, the cadential close divides the episode into two segments (E4a: mm.
21-223, E4b: mm. 223 -244). Another episode, E5, is subdivided even
further (E5a: mm. 283-303, E5b: mm. 303-344, E5c: mm. 344-422).
Finally, several episodes or their segments are varied repetitions of
earlier models. E1 recurs in three varied repetitions: E3 uses the tensionextended version of M1 from m. 9, E6 is closest to the model but begins
with a half-bar extension, and E5b appears most remote in its use of the
motivic material: the voices are exchanged, with the lower voice in the
lead, the middle voice reduced to a broken-chord figure, and the upper
voice recalling M2.3 E2 recurs once: E5a recalls it in inverted voices, with
3
Despite the varied beginning, the statements of M1 in the lower voice should nevertheless
retain the tension curve characteristic for this motif (with, e.g., a crescendo in mm. 304-311
followed by a diminuendo up to m. 31m), while the figure in the middle voice is too
removed from the original to take part in any subtle dynamic shaping.
82
WTC I/3
the lower voice now featuring the dramatic ascent (a slide should be added
on the peak note in m. 29) while the upper voice, partly crossing over the
middle voice, recalls the developed version of M1 and the middle voice
just fills the texture. E5c, the last segment of the longest episode, features
an internal correspondence: U: mm. 35-38 (with 3/16 upbeat) are taken up,
voices inverted, in L: mm. 39-42 (with 3/16 upbeat). One may get a clearer
picture of what is happening in these episodes by completing the earlier
diagram as follows:
E1
E2
E3 (. E1)
E4a
E4b
E5 (. E2 + E1 + E4b)
E6 (. E1)
E7
The role these episodes and their segments play in the dynamic development that shapes the fugue as a whole is both relevant in each case and
significant with regard to the understanding of the overall structure. E1, as
it is determined by the introduction of new material, demands a change of
register or color (the same holds true for its three variations). Its sequences
progress downward, thus causing a relaxation and suggesting that a section
is drawing to its close. (In fact, all three voices have already stated the
subject.) The last sequence, however, extends the rise within M1 and reestablishes a higher level of tension, thus preparing the listener for more to
come: a redundant entry. E3, the first variation of E1, shows even more of
this extended rising. The ascending trend within the melodic units counterbalances the relaxation implied in the descending sequences and thus
defines this episode as one linking adjacent subject entries. By contrast, E5b,
the second variation of E1, stresses the decline. Its role in the overall
tension is one of announcing the forthcoming end of a section. E6, the third
variation of E1, returns to the pattern of the original: the smooth tension
decay in the descending sequential pattern is arrested at the last moment,
thus granting the following (redundant) subject entry to be perceived as
still being part of the section.
Both the original E2 and its variation in E5a are self-contained units. In
a color distinctly different from that of the subject-determined passages,
their dynamic outline runs in curves; within each of them, a rise to the
climax and a subsequent relaxation are concluded by a cadential close.
Compared to the relaxing E1 and the self-contained E2, E4b represents the
type of episode that conveys a preparation for a subsequent entry. The
incomplete subject statements serve to suspend the tension before the
ensuing full entry. This impression is further enhanced by three facts: this
episode segment sets off after a cadential close, it is presented in reduced
ensemble, and the secondary voice is confined to non-motivic material.
C major
83
84
WTC I/3
The fugue features three ornaments: the slide in E2 (and, correspondingly, in E5a), the cadential mordent in U: m. 22 (not included in the fair
copy but deriving from a copy, presumably because Bach regarded this
conventional ornament as self-evident), and the compound ornament toward
the end of E5 (see U: m. 38). The slide often causes confusion, probably
because it is conventionally printed slightly to the left of the note head it
ornaments. Yet just like other Baroque embellishments, it also begins on
the beat. Thus in m. 13, the right-hand downbeat G falls on the left-hand
C. It is followed, in 32nd-notes or faster, by A and a B that is sustained for
the remainder of the note value. Correspondingly, in m. 29 the left hand
plays C-D-E with the C (not the E) coinciding with the middle-voice F.
The cadential ornament in m. 22 is a simple mordent, beginning on the
upper neighbor note and consisting of a double shake: E-D-E-D. The
symbol for the complex ornament in m. 38 asks for a turn progressing into
a trill. Because of its tied ending and delayed resolution, this trill should
conclude without a suffix. The result is, in 16th-note motion, an initial
A-G-F-G.4 This is followed by six A-G groups. (In order to further
enhance the suspension it is possible to play only five A-G groups and
stop a little earlier before the bar line.)
The most prominent feature of this fugue, in terms of design, is the
striking analogy of mm. 1-12 and mm. 42-53. Here are the details: The
order, position, and keys of the three initial entries recur identically. The
surrounding counter-subjects also correspond (although mm. 42-46 now
feature an additional voice, owing to the fact that the ensemble in a fugue
never drops back to a single voice). As has been shown above, the
subsequent episode, E6 in mm. 48-51, is a variation of E1, with its initial
half-measure extension serving to modulate. The subject entry that follows
also corresponds with the redundant entry in the first section although, as
a result of the modulation in the episode, mm. 51-53 now sound on the
tonic and not on the dominant as did mm. 10-12. However, the order and
position of the voices are the same. The analogy of these two entries is
additionally enhanced by the fact that they are the only ones in the entire
fugue to be accompanied by CS3. This prominent analogy defines the
major structural traits of the fugue. In addition, the episodes play an
important role in determining the design.
4
The pitch of the lower auxiliary needs a comment. The harmony underlying these
measures of retransition is normally interpreted as the dominant, represented by an
alternation of dominant-six-four (C major with G in the bass) and dominant-seventh
chords. If this interpretation is adopted, i.e., if G major acts as a dominant, its seventh is F
(heard repeatedly in these bars), whereas its leading note in ornaments should be F.
C major
85
There are three instances in the C-major fugue where the concluding
force of an episode sheds light on Bachs intention of partitioning the work
into sections. The first is the cadential close that ends E2 at the beginning
of m. 14. The second is the explicit cadence in the middle of m. 22, already
mentioned repeatedly. The third is more complex. In E5a, the variation of
E2 seems to conclude something in the middle of m. 30. However, this
cadential close is followed by the variation of E1 that, as has been shown,
makes no attempt to launch any kind of new development. Neither does the
ensuing segment in which the tension is suspended. To see these three
episode segments follow one another is already unusual enough. To see
them trying to surpass each other in tensionlessness is even stranger.
However, it is this very strangeness, this long retreat from the active
striving in the piece, which gives this portion its particular effect: as a
protracted buffer before the last section of the fugue it succeeds in
highlighting the symmetrical design.
Finally, looking for features that might indicate section beginnings we
find that two subject entries in this fugue appear in reduced ensemble.
They are the first minor mode statement in mm. 14-16 (which is thus triply
justified in being regarded as a section beginning) and the first entry in the
recapitulating final section.
86
WTC I/3
The harmonic outline confirms the other findings: The first four subject
statements remain in the home key of C major. The following two are in
minor modethe relative minor keys of the tonic and the dominant
respectively. E minor, the relative of the dominant G major, is also the
key in which Bach concludes this section with a cadential formula. The
episode segment E4b modulates back to C major, and all remaining
subject entries are presented in the home key, in the conventional
alternation of tonic and dominant.
The analogy of the first and last sections requires correspondences also
on the level of dynamics. Both times, the first three entries sound gaily
bouncing, with a slight increase caused by the growing number of voices.
The redundant fourth subject statements regain this mood after the very
timely pickup of tension at the end of the respective episodes. The second
section also shows a slight tension increase between its two subject statements, mainly because of the growth from two to three voices.
In the third section, however, the first of the two entries most probably
contains more tension than its successor. The main reasons are that it
appears at a point of heightened expectancy (after the two incomplete
subject statements) and that, in unmodified ensemble strength, the subject
appears in the upper voice, making it appear more powerful than the
following statements middle-voice position. This decreasing tendency is
then continued through the long string of episode segments that, as has
been shown above, become ever lighter.
Among the four sections of this fugue, the analogous outer ones take
the lead. The second section sounds softened, due both to its minor mode
and its shorter extension. In the third section, the four measures containing
the two subject statements return to the home (major) key but contain no
special features that would emphasize them in any way. While they may
sound more self-assured than the preceding minor-mode entries, this mood
is overshadowed by the nearly sixteen measures of surrounding episodic
material.
C E G + A C E. Note that this combination of pitches is exactly the same as that in the
final interrupted cadence of the C minor fuguecompare fugue m. 112 with prelude m. 5.
87
88
WTC I/4
the cadence in E major to one in C minor. This return to the tonic passes
almost unnoticed because of Bachs decision to continue and develop the
same melodic idea throughout the next measures. Moreover, mm. 8-10
recur in mm. 23-25, transposed and with the middle and lower voices
varied. (See also the remote resemblances in the following measures.)
The character of the piece is calm, yet the time signature might present
a pitfall in terms of tempo: the calm character expresses itself by means of
a half-measure pulse. (The harmonic tempo, proceeding exclusively in
half-measure steps, supports this pace.) While these pulses should convey
a generously swinging motion, the actual tempo of the quarter-notes is not
really slow. The appropriate articulation is an overall legato not even
interrupted for cadential-bass formulas since Bach writes all of them
explicitly with tied notes (see e.g. mm. 13, 34, and 38).
There are several ornaments. The opening phrase features arpeggiated
chords, grace-note groups, single grace notes, and a mordent. The arpeggios
and the grace-note pairs represent basically the same musical feature; both
are played on the beat and fast. The notes are sustained whenever they are
essential to fill the chord (as in mm. 1, 3, and 12); they sound equally fast
but unsustained whenever they contain pitches also provided for by other
voices (as in mm. 2, 4, and 8). The single grace-notes (see mm. 2 and 4)
are appoggiaturas since they represent harmonies different from the main
notes, and must be played with due weight. Appearing in the context of
imitating voices, in which case the rules of polyphony require the retreating voice to give way to the imitating one, these appoggiaturas are best
resolved after one quarter-note each. The mordent, approached stepwise by
the appoggiatura, begins on the main note and consists of a single
three-note shake. Both the grace-note (appoggiatura) and the mordent on
the main note (resolution) should be transferred to all recurrences of the
same motif, wherever harmonic and structural progressions allow this.2
The grace-notes in the second phrase are somewhat more problematic.
Harmonically, they represent appoggiaturas and thus require time. Melodically, they interrupt the smooth scalar motion and create note repetition.
Rhythmically, they would have to be played in 16th-notes. Perhaps most
importantly in terms of the impact on the character of the prelude: as this
splitting of an eighth-note falls on the final note of a motif, an impression
of congestion arises together with a blur of the phrasing. These reasons
combined suggest that it may be advisable to do without these ornaments.
2
Left hand m. 3 add grace-note A on beat 1, mordent on G beat 2, right hand m. 4 add
mordent on D, and left hand m. 9 add grace-note C, mordent on B.
C minor
89
90
WTC I/4
The only other complete quotation of M1 occurs in mm. 8-9. There are,
however, quite a few developments, deriving mainly from the first half bar.
In mm. 9-10, the right-hands threefold use of this partial motif brings
about a rise in tension. The middle voice answers with a single quotation
that leaves out the octave leap (m. 10-11), and the lower voice continues
with its own little build-up in a twofold motion whose final steps point
downward (m. 11-12). A similar development of M1 can be witnessed in
mm. 24-28: the threefold rise in the right-hand line corresponds with that
in mm. 9-10, and the lower voice also answers with a twofold quotation of
the partial motif, which this time takes up the original octave leap. It is
then extended through two additional sequences with upturned endings,
resulting in a powerful increase toward the peak note F (m. 28m). All the
while the right hand uses the half-measure figure for a gradual descent. A
third development of this kind involves only the right hand: mm. 30-31
feature a fourfold rising motion that prepares a climax on the G at m. 321.
These developments aside, there are short quotations of thematic components that mainly serve to create overall unity. The partial motif frames
the end of the third phrase by appearing in m. 12 (twice) and in m. 14, it
wanders through the voices in the fourth phrase (see the almost continuous
motion from L: m. 15 to M: m. 18), and concludes this section (M: m. 20).
Two further quotations appear toward the end of the piece (L: m. 34 and
M: m. 37). The rhythmic figure from the center of M1 also occurs independently (mm. 9-11, 15, 17, 21-27, 30-31, 35, 36, and 38). Even the cadential
formulas in mm. 13-14 and 34-35 rely in their leading voices on this
rhythmic figure.
M2 is exactly one measure long. First presented in mm. 5-6, where it
begins after the downbeat and ends on the first eighth-note of the following
measure, it consists of an ascending broken chord that brings with it a
natural though soft increase in tension, and a falling scale that resolves this
tension. This motif only appears twice in the prelude: in mm. 5-7 where the
model is followed by two descending sequences, and in mm. 20-23 where
the same progressionmodel plus two descending sequencesis extended
through a fourth sequence whose end melts into the M1 development of
mm. 23-28.
The preludes overall dynamic level remains gentle. As the tension
rises slightly through the first phrase and falls through the descending
sequences of the second, there is an initial furtive climax in the middle of
m. 5. The second overall climax occurs in the third phrase. While it is
somewhat stronger than the first, it is weakened by the fact that the treble
reaches it in the middle of 10, the bass only on the downbeat of m. 12.
C minor
91
The fourth phrase features three dynamic curves in its three subphrases.
They gradually rise in intensity, yet none of them is important enough to
serve as an overall climax. The more significant peak follows at the beginning of the long section where, corresponding with the two initial phrases
of the piece, it falls on the first peak note of M2, at the end of m. 20.
The three most powerful climaxes in the prelude are thus:
right hand,
middle of m. 25,
left hand,
middle of m. 28,
both hands together, downbeat of m. 32.
Thereafter, the tension recedes (no accent on A in m. 33!). The final
passage, after the interrupted cadence in mm. 34-35, features an extremely
soft, subdued build-up through mm. 35-36 to the downbeat of m. 37.
92
WTC I/4
C minor
93
m. 1-4
m. 4-7
m. 7-10
m. 12-15
m. 14-17
m. 19-22
m. 22-26
m. 25-29
m. 29-33
m. 32-35
v5
v4
v3
v2
v1
v4
v4
v3
v5
v3
11. m. 35-39 v4
12. m. 38-41 v3
13. m. 44-48 v2
14. m. 48-51 v1
15. m. 51-54 v4
16. m. 54-57 v2
17. m. 59-62 v1
18. m. 66-69 v1
19. m. 73-76 v5
20. m. 76-80 v1
21.
22.
23.
24.
25.
26.
27.
28.
29.
m. 81-84
m. 89-92
m. 94-96
m. 95-97
m. 96-98
m. 97-100
m. 100-102
m. 107-109
m. 112-115
v4
v1
v1
v2
v1
v5
v4
v1
v2
The note group D C F E in v1 mm. 54-57, however, is not a subject entry. The combination of an initial major second with a perfect fourth is too unlikely, and assuming a
stretto at this early stage in the fugue would mean ignoring Bachs design, which seems so
consciously to reserve the stretto intensification for later.
94
WTC I/4
C minor
95
96
WTC I/4
C minor
97
strong introverted urge, the second subject is generally lighter in its mood,
while the third subject shows a jolly, extroverted vivacity.
The three subjects appear in all possible combinations. Determining
what kind of mood or spirit results from each of these juxtapositions lays
a good foundation for an understanding of the entire composition.
The coupling of S1 + S2 brings about a quality that is more serene,
less urgent than that which determines the main subject when heard
alone (S1 introverted/intense + S2 introverted/relaxed).
In the S2 + S3 combination, all the darker drive has gone. What is
left is a momentarily intense quality, soon giving way to one of
relaxation (S2 introverted/relaxed + S3 extroverted/active).
Conversely, when S1 is paired with S3, two strong forces join,
expanding both inwardly and outwardly (S1 introverted/intense
+ S3 extroverted/active)
The range of tempo in this composition is confined on the one hand by
the basic character, which is rather calm, on the other hand by the time signature, which requires an alla breve pulse. Ideal is an eighth-note motion
that is unhurried but can nonetheless be clearly felt in half-measure groups.
The corresponding articulation requires legato for almost all melodic notes.
An exception can be made within the third subject where the first interval,
the ascending fourth, and the ensuing note repetitions may sound in a
bouncing non legato in order to give credit to the special outgoing mood
of this component of the material. The Urtext of this fugue does not
indicate any ornaments.
The relative tempo of the prelude to the fugue that does justice to their
characters sets larger metric units in proportion. Owing to the strong
half-measure feeling in both pieces, a good effect is achieved like this:
half a bar
corresponds with
one bar
in the prelude
in the fugue
(Approximate metronome settings: prelude beat = 100, fugue beat = 66.)
The fugues design is complex. Within the first large part, the full
ensemble of five voices is attained only transitorily in mm. 19-20; the fifth
entry and most of the redundant subject statement sound in still reduced
ensemble. After the ensuing cadence in mm. 21-22 the number of voices
drops to only three, a fact suggesting the beginning of a new round. This
second section ends in m. 35 in full ensemble and with a pronounced
cadence, followed again by a reduction of the ensemble to three voices.
The third section introduces the second subject, which sounds both
against the main subject and independently, both in its original shape and
in inversion (see mm. 41-44). In m. 49, the third subject enters for the first
98
WTC I/4
time. Its immediate combination with both S1 and S2 marks the beginning
of something new: the fourth section. From here to the end of the large
middle part, there are two structurally analogous progressions:
mm. 48-72
correspond with
mm. 73-93
5 three-subject juxtapositions : 4 three-subject juxtapositions
grouped 3 + 1 + 1
:
grouped 2 + 1 + 1
(mm. 48-57 / 59-62 / 66-69
:
mm. 73-79 / 81-84 / 89-92)
interspersed with two passages featuring only S2/S3
(mm. 57-58, mm. 62-64 : mm. 79-80, mm. 84-88)
rounded off by passages with distinct cadential-bass steps
(mm. 69-71
:
mm. 92-94)
After the end of these two sections, the fugues fourth and fifth, the second
subject is dropped. Instead of three-subject juxtapositions, the sixth section
begins with S1 and S3 in simultaneous stretto. All this together provides a
picture of six sections, with the following statements of the main subject:
S1:
I
mm. 1-22
v5, v4, v3, v2, v1, v4;
II
mm. 22-35
v4, v3, v5, v3;
III
mm. 35-48
v4, v3, v2;
IV
mm. 48-72
v1, v4, v2, v1-v1;
V
mm. 73-93
v5, v1, v4, v1;
VI
mm. 94-115
v1v2, v1v5, v4, v1, v2.
As both the second and the third subject enter considerably later, it is
to be expected that their own rounds do not entirely coincide with those
of the main subject. For the fifteen statements of the second subject, four
groups can be made out:
S2:
I
mm. 35-48
v1, v4, v5, v1;
II
mm. 49-65
v2, v3, v1, v2;
III
mm. 66-71
v5, v1;
IV
mm. 72-94
v3, v5, v2, v1, v4.
The third subject enters last but seems the strongest among its rivals. It
surpasses both the other subjects in the number of its statements. It catches
up with the second subject by already finishing its fourth round in m. 92;
and it equals the main subject insofar as it also builds a total of six rounds.
The 37 entries of the third subject are as follows:
S3:
I
mm. 49-64
v3, v1, v5, v2, v4, v1;
II
mm. 64-73
v3v5, v2, v3, v2;
III
mm. 74-84
v4, v2, v5, v3;
IV
mm. 84-92
v4v2, v5, v5;
V
mm. 92-101
v1v2, v4v3-v4v3, v2v4,v1, v5;
VI
mm. 102-115 v4v3, v2v4, v3+v2v4, v3.
C minor
99
The harmonic outline will here be described with reference to the main
subject. Its first six statements, all on the tonic or dominant of C minor,
are concluded with a cadential close in G minor in m. 21-22. The next
four entries effect a modulation to the relative major key, through F minor,
C minor, and B major to E major, with a cadential close in m. 34-35. The
third section finds S1 returned to the C minor field. It closes, interlocked
with the beginning of the next round, in C major at m. 491. The fourth
section begins firmly anchored in the subdominant, with two entries in F
minor and one in the relative key of A major. However, the last two S1
statements revert to C minor, with the repeated entry in the first voice
ending in a floating position on the dominant-seventh chord. The fifth
section confirms the return to the tonic by placing all its four S1 statements
on C itself and concluding this part with a cadential close in C minor
(mm. 93-94). The final section basically remains on the tonic. Stretto
entries occur on various pitch levels but all relating to an implied key of C
minor. From m. 105 onward the G pedal, giving way only momentarily to
its leading-note F, prepares the end of the composition, which is reached
with a final subject statement over a tonic pedal.
100
WTC I/4
102
WTC I/5
D major
103
104
WTC I/5
+
p
mf
mp mf
mf
mp mp p
p
A process roughly similar to that described above occurs in the subsequent section. The first measure is yet another version of the active part of
the modulation. After a renewed climax in the first half of m. 15, the
tension begins to fall but is suspended in its descent by a pedal of 2
measures on B (mm. 16-18). Here, the hidden two-part structure that so far
characterizes the bass pattern is replaced briefly by a broken chord pattern,
thus giving these two measures the even larger frame of whole-measure
pulses. At the conclusion of the pedal, m. 18 seems to lead to a new key
(A minor). Yet again this key is not confirmed but passed through on the
way to G major. This time, however, there are no repetitions of the two
target harmonies. As in the corresponding measures discussed above, this
final small-scale modulation sounds like an appendix to what precedes it.
mm. 14
15
16
17
18
19
20
p
mp + mp mp + mf mp p
The recapitulation of the first five measures should, of course, retrace
the dynamic outline from the beginning of the prelude; in other words:
mm. 20
21
22
23
24
25
p
mp p
mp + mp p
The preludes last section begins with two measures that prepare the
final pedal on the dominant. Unlike similar final pedals in other preludes,
this dominant bass note is not preceded by its leading-tone. Here, instead,
this leading-tone is made up for in the middle of the pedal (m. 302-31)
where it combines with virtuoso peak notes in the right hand and thus
creates a strong emotional climax. After this interruption, the dominant
pedal continuespartly as a sustained note, partly implied under the
cadenza-like setting of the right hand in mm. 33-34and only resolves
onto the tonic in the very last measure.
For the dynamic design this means a steady increase, with only slight
inflections on the surface in those measures where the harmony seems to
come to a halt, i.e., mm. 30-31 and m. 33. The overall dynamic development in this section may be described as follows:
mm. 25 26 27 28 29 30 31 32 33 34 35
p
mp mp mp + mf mf mf + pf pf
pf + f
D major
105
106
WTC I/5
This subject undergoes none of the modifications otherwise encountered in Bachs fugues: it is never inverted, augmented, or diminished, and
its pitch and rhythm patterns remain untouched. The only exception occurs
in the final statement (B: m. 24-25), which some scholars do not consider
equal to the other statements as it is extended in both directions. The
bustling 32nd-note group from m. 244 is preceded by three similar groups
that gradually ascend in fourths toward the tonic level (see the groups from
B, E, and A). Symmetrical to these anticipating figures, the subjects
second half is extended with sequences in m. 26: the original notes of this
half, G F E D, beginning here with an octave displacement, recur in a first
transposition a fifth down (C B A G), while a second sequence is modified
to accommodate the cadential-bass steps (F E D A). Moreover, while
there are no real strettos in this fugue, the final entry is partially set in
parallelsand as the parallel begins one beat (i.e., one whole figure) later,
the deceptive impression of a stretto does arise for a moment. To complete
this unusual entry, in its second half and throughout its sequences the three
upper voices move in rhythmic alignment but in contrary motion (see the
right-hand chords in mm. 25-26). The strong homophonic element in this
final, extended subject statement enhances the outsider position it occupies
in the design of the fugue.
D major
107
The counter-subjects are unusual. They confirm that the French overture,
an essentially homophonic genre, is indeed one of this fugues god-parents.
The contour appearing in continuation of the first subject statement and
against the second entry consists of the notes G-F-E-D-E-A. It meets the
three basic requirements of a counter-subject: to be easily recognizable,
independent, and taken up again later in the fugue. Nonetheless the term
counter-subject seems a little too big for this simple unit, not least because
it sounds very much like a cadential-bass pattern and does not wander from
voice to voice as a polyphonic unit should, but remains essentially at the
base of the texture. A close look at the two other counter-subjects comes to
similar conclusions: Although they are formally independent of the subject
and taken up repeatedly in the course of the fugue, the fact that they
resemble closing formulas prevents one from taking them too seriously as
counter-subjects.
Dynamic shaping in the contrapuntal components allows for but little
independence. The cadential-bass pattern of CS1 should logically approach
the note that represents the subdominant as its climaxa climax thus
coinciding with that of the subject. Similarly, CS3 with its quarter-note/
half-note/quarter-note rhythm will accent the half-note, which falls again on
the same beat as the other climaxes.
Only the syncopated CS2 could sound
in diminuendoif it were left alone;
but more often than not its first note
falls together with that of the subject
beginning (see mm. 4, 5, 8, and 13).
The dynamics of these four parts thus
give a distinctly homophonic result.
Five times are the subject entries
interspersed with episodes: E1 = mm. 3-4, E2 = mm. 6-7, E3 = mm. 9-11,
and E4 = mm. 16-23. Among the three motifs that make up the episodes,
one, appearing in four of the five episodes, uses the 32nd- note figure from
the subjects beginning. Moreover, E4b (mm. 17-23) is dominated by the
dotted rhythm of the subjects second half. Hence, most of the episodes in
this fugue are closely related to its subject.
The close relationship with the primary thematic material explains
some of the dynamic developments in the episodes. M1 is an ornamented
version of the unresolved second counter-subject (mm. 32-41 . A: 132-141),
a descending step in decreasing intensity. In analogy to the subjects
second half, M3 also proceeds in decrescendo. Conversely, M2 with its
relationship to the subject head seems at first to build up tension. Yet there
108
WTC I/5
are three reasons that speak against this rendition: on occasion of its first
entrance in mm. 6-7 the motif ends with the neutral interval of a fourth. In
all other episodes it is launched on a strong beat and ends on a weak one.
And above all: the fugues structure would become incomprehensible were
all M2 entries to sound at the energy level as the subject head.
The role played by the episodes in the course of the fugue is also
puzzling. E1 with its underlying lines of parallel syncopations links two
pairs of subject entries within the same section. Z2 has the same effect,
although it marks the opening of a new section. By creating this relaxing
effect in front of what should be a new impulse, Bach seems to thread the
first two sections closely together, so that the feeling of a structural ending
only arises when a new tonal sphere is reached, i.e., with the downbeat of
m. 9 that establishes B minor. E3 is the first episode to be self-contained.
It seems intended as a buffer between two larger portions of the piece.
The pattern of the subjects statements in mm. 9-16 (S, A, S, T, B) with
its twofold soprano lead suggests a subdivision into two rounds. The
surprising deceptive cadence at the end of the alto entry and the M2 variant
accompanying the concluding tone underline a caesura that is otherwise
hardly tangible. By contrast, the extensive final episode is subdivided in
itself: the brief first segment (E4a) is a cadential close and thus brings a
feeling of relaxation (mm. 16-171), while the longer second segment (E4b)
develops the material of E3 (mm. 17-19 + 21 . 9-11). After descending
sequences in mm. 17-20 it uses a dense web of imitations to rebuild tension
within a single measure. The subsequent cadential close with its soprano
ascent to the third does not allow a complete resolution. After another
measure with M3 in three-part homophonic texture Bach repeats the
cadential close, now with conventional formulas in the soprano and bass.
The character of this fugue is determined more by the mood of a French
overture than by the conventions it shares with other polyphonic compositions. It is stately, lively but not light. Articulation in French overtures is
an intricate affair: The 32nd-note groups are legatovirtuoso figures that
should never sound melodious, which gives an indication for a minimum
tempo. The dotted-note groups reflect the stately character of this piece
best if played in a rather heavy non legato. The longer note values in the
subjects companions require three slightly different types of articulations:
neutral non legato in the cadential figure of CS1, a more melodious non
legato in CS3, and true legato in the dosido formula of CS2.
Of the 16th-notes in the upper part of M3, the first threea written-out
inverted mordentshould be played legato, while the fourth 16th-note
may sound as non legato as the accompanying voices.
D major
109
As the prevalent features in this fugue and its prelude are so different,
a risk of monotony does not arise. The tempo proportion may therefore be
simple: a half measure (two quarter-notes) in the prelude corresponds with
one quarter-note in the fugue. (Approximate metronome settings: prelude
beat = 120, fugue beat = 60.)
The fugue features three ornaments that, based on their print size in the
Urtext edition, stem from Bachs own hand: the cadential ornaments in
mm. 20 and 22 and the grace-note in m. 10. The first, a mordent, should
consist of four notes (D-C-D-C), with a triplet of 64th-notes against the
first two 32nd-notes in the left hand. The symbol for the compound trill in
m. 22 indicates a turn + mordent figure: D-C-B-C-D-C-D-C must be
fitted into less than a dotted eighth-note as the anticipation should come
after a short stop. A brilliant 64th-note execution, with the final C on or
before the dot, is the solution. The grace-note in m. 10 represents an appoggiatura, held for only a 16th-note to avoid the D-C parallel with the bass.
If played, it must be transferred to S: mm. 11 and B: mm. 18, 19, 20
Other ornaments stem from copies. They may or may not go back to
the masters suggestions but are worth considering. The subject could be
played with ornaments on two of its dotted notes. The first is an inverted
mordent (to be played with a whole-tone neighbor in all cases apart from
mm. 8 and 15). The second is a mordent with three or, better still in overture style, five notes. Both ornaments must be either omitted or included in
each subject statement. The mordent in M1 (T: m. 3) seems inconsequential
but sounds pleasant. The inverted mordent in m. 15 may serve to enhance
the fact that this subject statement, unlike all others, ends unresolved. The
two cadential ornaments in mm. 16 and 26 are very typical. Both begin
with the main note, ascend twice to the upper neighbor note, and end in a
suffix. Their rhythmic realization depends on a decision that, unique to this
particular fugue, has an impact on the rendition in general: Performers who
wish to underline the French-overture character of the fugue will render
these ornaments in the corresponding style, i.e., as fast as possible, concluding them in the time span of a dotted eighth-note or less and leaving
ample room for a point darrt. Performers preferring to stress the fugue
over the genre piece will interpret the compound symbol as an indication
for a note-filling trill; they will therefore sustain the initial main note for a
16th-note, followed by three regular 32nd-note-pairs.
This brings us to a rhythmic feature to be mentioned in this context.
French overtures are known to feature over-dotted rhythms: some of the
notes written with single dots were traditionally played as if they were
double-dotted. Again: whether one wishes to play this composition in the
110
WTC I/5
D major
111
112
WTC I/5
voice. The next entry sounds in full ensemble and in the expected G major;
but again, its resolution comes with a seventh (this time the major seventh).
The third entry in this group reaches E minor, the subdominant relative and
the key of the explicit cadence at the end of this part. The ensuing sevenbar episode is longer than any of the preceding sections; it is even longer
than the entire six-measure second part.
Within the first part, the four subject statements of the first section
bring forth an increase in texture. The increase in loudness, however, is
comparably small because already the first statement should have fully
declared the stately character of the piece. Throughout sections I and II, the
tension remains almost unchanged. The two similar episodesone within
the first section, the other between the two sectionscause slight but inconsequential relaxations. It is only the third episode, at the end of the first
part, which succeeds in conveying a more noticeable tension-decrease.
Within the second part, the buildup is again caused by the growing
number of voices. While among the two statements in three-part texture,
the alto statement in mm. 12-13 with its exposed pitch position sounds
more outgoing than the soprano statement in mm. 13-14, the ensuing tenor
statement in four-part texture and the last entry in the bass, accompanied
by a parallel of CS3 in the two middle voices, share in the climax. As the
end of this statement does not bring about the expected harmonic resolution but keeps the middle voices in suspension, the tension is allowed to
subside only very gradually in the subsequent cadential measure.
The long episode opening the third part begins with the material from
the tension-releasing third episode and thus creates a sense of continued
relaxation until m. 201. Immediately following this, the fourfold imitative
interplay of M2 propels the tension so high that even the cadential pattern
with its typical closing-formula mordent cannot dissipate it fast enough.
Another measure with M3 and the varied repetition of the cadential close,
this time without the stormy 32nd-note stretto, is required to achieve relaxation. The following measure with its descending tendency continues this
direction. As a result the final, unusually extended subject statement with
its powerful parallel motion begins from a relatively soft level, only to
unfold its tremendous buildup all the more forcefully.
The relationship between the three parts is obvious from all that has
been said above: the first two parts are similarly built, with rises both times
from about mf to a good f. The third part seems to delay the outbreak, and
when it finally allows its only subject statement to make its declaration,
this by far outranks that of all the previous entries.
114
WTC I/6
D minor
115
The second section is determined by its two motifs. The longer first
motif (M1), introduced as an independent melodic unit in the lower-voice
of mm. 6-8, is actually not completely new. Its stepwise descent recalls the
bass line in m. 4, and its ending is reminiscent of the cadential-bass pattern
in mm. 5-6. Some details, however, are new, and they are decisive: M1
begins on a pitch that is the seventh of its chord. This creates high melodic
tension right from the start of the motif. In its second half the line returns
once more to the same note, which is now redefined as the root of the
deceptive chord in the subsequent cadence. The tension created here is of
a harmonic nature. In the right hand, the uppermost chordal notes present
several short parallels to the left-hand line. Particularly the ascent in the
middle of m. 7 serves to enhance the impact of the deceptive chord. The
much shorter second motif (M2), introduced in m. 10, depicts a melodic
descent combined with a harmonic process of relaxation and thus counterbalances the gesture of M1. The dynamic processes in this section may be
described as follows:
M1
mm. 62-82
mf mp mf + p
sequence
mm. 82-101
mf + mp + pf p +
M2 + sequences mm. 101-122 mf mp + mp p +
M1 + extension mm. 122-151 pf mf pf mp mf + p
The third and concluding section combines various secondary melodic
processes of a different nature. The lower part sets out from the newlyfound keynote D. After a broken chord that propels it almost two octaves
upward, it splits into a hidden two-part structure with D as a pedal in the
lower part and a melodic descent in the upper notes, accompanied by a
peak-note line in the right-hand part. The anticipated release of all tension
is delayed when the peak-note line takes on a virtuoso form in mm. 18-19.
But, as this virtuosity seems to bear little effect against the decreasing
power of the bass line, the previous peak-note pattern is reinstated.
Soon after the bass descent has been concluded, the concealed two-part
structure of the left-hand part turns into a real one: an independent middle
voice evolves. As middle and lower voices proceed, two distinct tritone
intervals mark points of high tension at mm. 213 and 223. In the right-hand
part, the peak note line is taken over alternately by the top and bottom
notes of the broken chords. The overall impression is one of descent, so
that the downbeat of m. 23 is reached in a softness almost comparable to
that of its corresponding measure, m. 15.
From here, both hands are once again propelled up almost two octaves,
with the now firmly established middle voice adding extra drive. In an
116
WTC I/6
D minor
117
The preparation and the resolution of this climax are slightly irregular.
The process of tension-growth leading up to the climax is called into
question by the three-note slur in m. 2, and after the climax there are only
two notes entrusted with resolving the rather powerful tension. Yet instead
of ending on one of the two melodically relaxed notes of the tonic chord,
F or D, the subject comes to a melodically incomplete-sounding halt on A,
the fifththe one scale degree that regularly fails to convey a satisfactory
feeling of resolution. The slur in the subject becomes more comprehensible
if we remember that articulation symbols in Baroque polyphonic music
derive from markings for string and wind players. Violinists approaching
an unmarked line would play each note with a separate bow movement.
They would thus be able to increase the tension through each note. But,
seeing the slur in the subject, they would combine the group of notes
comprised under it in one single bow movement. The dynamic effect is
that the initial note under the slur is active while the notes following it
sound passive and do not continue the tension. The slur thus creates a
two-leveled structure: the dynamic gesture leads through each of the notes
in the first measure to the first note
under the slur and from there implicitly on to the climactic B, after
which it breaks off and then
resolves through the trill into the
subjects ending note.
There are seventeen complete and seven incomplete but structurally
relevant statements (the latter are here marked with an asterisk):
1. mm. 1- 3 O
9. mm. 17-19 U
17. mm. 28-30 M
2. mm. 3- 5 M
10. mm. 18-20 M
18. mm. 29-31 U
3. mm. 6- 8 U
11. mm. 21-23 U
19. mm. 33-34 M*
4. mm. 8-10 O
12. mm. 22-24 O
20. mm. 34-36 U
5. mm. 12-13 M* 13. mm. 23-25 U
21. mm. 35-36 O*
6. mm. 13-15 O
14. mm. 25-26 M* 22. mm. 35-36 M*
7. mm. 14-16 M
15. mm. 26-27 U* 23. mm. 39-41 U
8. mm. 14-15 U*
16. mm. 27-29 O
24. mm. 40-42 M
118
WTC I/6
in mm. 35-36), or it may appear inverted (see entries nos. 5, 7, 12, 13, 14,
15, 16, 18, and 22). Moreover, there are two conspicuous strettos, both
involving complete entries appearing in the original shape (mm. 17-20 and
39-42). These are surpassed by a stretto sounding in a chain of three entries
two of which are inverted (mm. 21-25). Finally, there are several strettos
made up of a mixture of complete (2/2) and incomplete (1/2) statements, using both the original shape (O) and the inversion (I):
mm. 12-16: 1/2 I (M) + 2/2 O (U) + 1/2 O (L) + 2/2 I (M)
mm. 25-31: 1/2 I (M) + 1/2 I (L) + 2/2 I (U) + 2/2 O (M) + 2/2 I (L)
mm. 33-36: 1/2 O (M) + 2/2 O (L var) + 1/2 O (U) + 1/2 I (M)
Parallel statements do not occur. There are, however, two instances
where the (inverted) second entry of a stretto coincides with an incomplete
entry in the original shape: see mm. 14-15 M + L and mm. 35-36 U + M.
Bach invents only one counter-subject for this fugue. CS grows out of
the final note of the preceding subject statement. The fact that the subject
seems to ends in a tied note thus brings up the question of the note value
with which the counter-subject begins. To what proportion the A on the
downbeat of m. 3 is divided between the subject and its counter-subject
can be understood when comparing later counter-subject statements: the
few times when CS does not follow immediately after a subject statement,
it opens with a first note of dotted eighth-note duration (see, e.g., U: m. 6).
The counter-subject thus begins, like the subject, on the second eighth-note
of a measure and ends after two measures on the downbeat. It consists of
two subphrases. These are interrelated through their rhythmic structure.
Both set out with a syncopated dotted eighth-note (U: mm. 3 . 4) and
progress in regular 16th-note motion until the subsequent downbeat. Both
are structurally analogous: in the first subphrase, four descending 16thnotes are sequenced; in the second subphrase, the same happens to a turn
figure. The two subphrases are most clearly distinguished in mm. 6-8
where the first is allocated to the upper voice while the second follows in
the middle voice. Considering these features, each subphrase will find its
climax on the initial syncopation, after which the falling 16th-notes effect
a relaxation. These dynamic developments are very gentle. With regard to
the relationship between the
two subphrases, the second
subphrase expresses a higher
tension level than the first
both because of the shift upward and, more importantly,
D minor
119
because of the greater emphasis the real syncopation creates in comparison to the merely implied syncopation at the beginning of CS.
Determining the number of episodes in this fugue depends on how one
evaluates the incomplete subject statements. If one assumes that, as in
many other fugues, considerably shortened subject statements function as
episode material, then the number of episodes would amount to eight.
However, since the incomplete statements in this prelude often engage in
primary-level strettos with complete entries, the label episode seems illfitted for measures characterized by these shortened entries. The following
table lists all the episodes in this fugue, but places measures with
incomplete subject statements in brackets:
E1 mm. 5-61
[E5 mm. 25-271]
E2 mm. 10-121 [-131]
E6 mm. 31-331 [-341]
E3 mm. 16-171
E7 mm. 36-391
E4 mm. 20-211
E8 mm. 42-44
Almost all episode material in this fugue seems closely related to the
subject. This is particularly true for the first three episodes and their
analogs later in the piece: In E1, the upper voice imitates the subjects
second half (see m. 5) while the middle voice recalls the beginning of
CSincluding the way in which CS emerged out of the subjects ending
note (M: m. 5 . U: m. 3). In E2, both the upper and lower voices extend
the preceding entry in sequences (mm. 9-101 are sequenced in mm. 10-111
and 11-121). Only the middle voice gains some independence by creating
a motivic figure of one-bar length. In mm. 12-131, the first incomplete
entry in the middle voice is combined with the subjects second half in the
lower voice (as it appeared in E1) and a long note in the upper voice. E3 is
a varied sequence of the measure preceding it (mm. 16-171 . 15-161). By
contrast, E4 is a typical cadential close: the upper voice provides an
elaborately ornamented dosido formula, the middle voice contributes the
syncopated trill with resolution on the next downbeat that often appears in
closing formulas, and the lower voice plays a cadential-bass pattern. In E5,
two incomplete subject entries in the middle and lower voices combine
with a figure derived from the end of the counter-subjectextended and in
inversion. The three final episodes are related to three episodes in the first
half of the piece: For E6 see E2, although the voices are inverted and considerably varied (mm. 31-331 or 341 . 10-121 or 131). For E7 see E3, the
latter being an extended version of the former, with three bars similarly
continuing as a sequence of the preceding bar. For E8 see E4, as the first
measure of E8 is an exact transposition of E4 (mm. 42-431 . 20-211). This
120
WTC I/6
D minor
121
For the ornament in the subject there are two possible solutions; they
depend on how one interprets the wedge on the climax. For performers
who feel that the wedge creates a sudden, rather dramatic halt in the
melodic flow and that the subsequent G comes in after something like a
phrasing, the trill should begin according to the rules for ornaments at
phrase beginnings, i.e., on the main note. For performers who feel that the
wedge creates articulation rather than phrasing and that the tension is
suspended throughout this interruption and picked up at an almost undiminished level in the G, the trill should begin according to the rules for
ornaments within a phrase, i.e., on the upper auxiliary note. In each case
the trill shakes in 32nd-notes and ends in the whole-tone suffix F-G.
There are several indicators that help to determine the design of this
fugue: A subject entry in reduced ensemble can be found at the beginning
of the four-measure passage with multiple strettos (mm. 212-251). This
reduction of voices occurs after the episode identified as a definite cadence
with strongly conclusive powers. The fugue is thus laid out in extended
structural correspondence. Working backward from the analogous cadences
one finds: mm. 20-211 . 42-431, mm. 17-191 . 39-411, mm. 12-(161) .
33-(381), mm. 9-111 . 30-321. Moreover, the subject statement following
the first distinctly cadencing episode, launching the four-bar phrase with
dual stretto (mm. 212-251), is set in reduced ensemble. If we accept that the
first four single entries in the fugues initial half (mm. 1-10) correspond to
the first two strettos in the latter half, the rest of the analogy falls into
place. The two passages form the first and third sections, concluding in
mm. 131 and 341 respectively. The two passages just ascertained as
analogous, mm. 9-21 and 30-43, constitute the second and fourth sections.
122
WTC I/6
124
WTC I/7
Try playing m. 15 as the penultimate measure of an E-major cadence. The soprano would
resolve onto E, the alto descend to G, the tenor to E, and the bass leap down to E. As m.
161 shows, this is not what happens: all voices except for the alto take unexpected turns.
E major
125
It need not, however, exceed the tempo of the spelled-out turn figure immediately preceding the trill, as closer inspection reveals that this figure is in fact part of a longer compound
ornament beginning on the tied eighth-note E.
126
WTC I/7
A missing accidental in the Urtext, found also in several other editions, should be carefully
considered: in m. 20, the tenor features B-A-B despite the A in both soprano (m. 19) and
alto (m. 20). As the soprano statement of M2 suggests B major for these measures, adding
a natural to the leading note seems essential.
4
The variations indicated by the asterisks above are: * The third note may be prolonged to
twice its value, delaying the fourth note to a weak beat where it follows usually as yet
another syncopation. ** The motif may sound with its first note lengthened to a syncopation. This may then be followed either by only the third note or by the two remaining notes.
*** The fourth note may bend back instead of leading downward.
E major
127
The dynamic design of the two subjects is, at least at the beginning,
determined by the motifs from which they are derived: S1 begins with a
tension-increase toward what was the final note of M1; the second, longer
subphrase begins anew and carries an even stronger crescendo through the
motifs varied sequence. The climax falls preferably on the quarter-note E
and is followed by the resolution through the subsequent do-si-do figure.
The dynamic shape of S2 reflects its evolution from the fugato motif: a
climax on the syncopation E
followed, after slight relaxation serving as a new start, by
a stronger second climax on
m. 21. The descent to the keynote of the target chord brings
the relaxation.
S1 appears twelve times in the course of the fugue and S2 twentytwo times. The following chart lists the statements of both subjects:
mm. 25-27
S1
A + S2
B
mm. 27-29
S1
S + S2
T
mm. 30-32
S1
B + S2
A
mm. 35-37
S2
A + S2 S
mm. 38-40
S2
S + S2 T
mm. 41-44
S2
A + S2 T + S1 S
mm. 46-49
S2
B + S2 T + S1 A
mm. 49-51
S1
A + S2
B
mm. 53-56
S1
S/B + S2
T + S2 A
mm. 56-58
S1
B + S2
A + S1 S
mm. 58-60
S1
A + S2
S
mm. 60-63
S1
S v S2
A + S2 B
mm. 64-67
S2
A + S2 S + S2 A
mm. 68-70
S1
A + S2
T
Both subjects appear with several irregularities and variations: S1 may
come without its final note (as in mm. 27-29) or even without the last two
notes (as in mm. 60 and 63). It may be deprived of its entire first subphrase
128
WTC I/7
(as in mm. 30-32), shortened in the middle so that the first subphrase leads
directly into the note repetition of the ending (as in B: mm. 56-57), or it
may, while shortened in the middle as described above, find its concluding
note repetition replaced by a single tied note (as in S: mm. 56-57). S2 may
come without its final note (as in mm. 27-29), it may have lost both its
final notes (as in mm. 30-32, 35-37, 38-40, etc.), or its beginning may
sound ornamented (as in m. 60).
S2 builds several strettos (see particularly mm. 35-49 and 64-67), the
last of which sounds briefly like a parallelalthough it is not. In addition,
there are a few interesting voice crossings: in m. 28 the tenor does not
descend directly after the two rising fourths of the S2 beginning. Instead it
M1
M1
M1
M1
M1
M1
M1
M2
M1
M1
M1
M1
M1
M2
M2 expanded
M2 expanded
M2
M2
M2
M2
M2
M1
M1
M1
M1
M1
M1
M1
M1
M1 S1
S1
S2
S2
M1
S2
M1
M1
S2
M1
S2
M1
M1
M1
M1
M1
S1
M1
S2
M1
S2
S2
S1
S2
S2
S1
M1
S2
S1
M1
S2
M1
S2
M1
M1
M1
S1
S2
S2
M1
M1
S2
M1
M1
M1
M1
M1
S1
M1
S2
M1
M1
S1
M1
S1
S2
M1
M1
S1
S2
S1
M1
S2
M1
M1
M2
S2
S1
S2
M1
M1
M1
E major
129
130
WTC I/7
There are two subphrases within this subject. This can be deduced not
only from the rest in its middle but much more from the varied partial
sequence with which the second subphrase begins: the two prominent
eighth-notes in the subjects second half sound like an elevated sequence
of the two eighth-notes before the rest, while the preceding groups of
16th-notes is shortened but made up for by the final strong-beat note.
The pitch pattern in the subject features mainly broken chords:
m. 11-2
= E-major chord (with auxiliary note)
m. 12-3
= A-major chord (with auxiliary note)
mm.14-22 = F7 chord (with passing note5)
A brief look at the remainder of the fugue shows that these and other
broken-chord patterns are prevalent throughout the piece. The rhythm in
the subject consists mainly of 16th-notes and eighth-notes, with the
exception of the quarter-note carrying the trill. The same predominance of
these note values can be detected in the entire piece.
The subjects harmonic background is determined by the modulation
within its confines. The active step (tonic to subdominant) occurs melodically immediately after the second beat, but is metrically confirmed only
on beat 3. This subdominant harmony is sustained even when the second
eighth-note of beat 3 returns to B (a note that, in its unaccompanied
melodic version, is necessarily heard
as a return to the tonic). The pivot
chord (V7/V) beginning melodically
with the A can be felt metrically
either in the rest at m. 14 or at m.21.
The E at m. 21 seems the obvious choice for a climax: it falls on a
downbeat, it represents the metrical place of the pivot chord that triggers
the modulation, and it is reached in the high-tension interval leap of a
minor seventh. Moreover, as mentioned above, it is conceived in varied
sequence to the process in the first subphrase that finds its climax on the C
at m. 13, the note that represents the subdominant harmony and sounds
rhythmically as the first halt after the opening 16th-notes.
The tension in the subject thus moves in two curves. The first begins
with a moderate crescendo and is complemented with a one-note relaxation
(C-B). A second crescendo then surpasses the first and is complemented
with a longer diminuendo for a more perfect resolution of the tension.
The subject appears altogether nine times in this fugue:
5
The use of the second eighth-note in m. 2 as a passing note is confirmed in Bachs harmonization of later subject statements; see, e.g., mm. 4, 7, and 18.
E major
1. mm. 1- 2
2. mm. 3- 4
3. mm. 6- 7
131
U
M
L
4. mm. 10-12 U
5. mm. 17-19 M
6. mm. 20-22 L
7. mm. 25-27 L
8. mm. 28-30 U
9. mm. 33-35 M
The subject always sounds in its complete length. Its last note appears
most often as a 16th-note but may be extended to an eighth-note (as in mm.
27 and 30) or even to a quarter-note (as in m. 35). Moreover, several
statements begin with a syncopated anticipation of the first note (see mm.
10-11, 25-26, 28-29, and 33-34). In addition to these small changes in the
appearance of the subject, interval adjustmentsbetween the first two
notes and across the restoccur in all tonal answers (see m. 3 etc.). No
stretto or parallel are used.
Bach has invented only one counter-subject for this fugue. CS is introduced in mm. 3-4 against the second subject statement, where it begins
slightly later than the subject with the eighth-note A and ends, together
with the subject, on the G at m. 43. This counter-subject appears as a
faithful companion to the subject in all but the initial and final entries,
featuring only one slight variation of its beginning in m. 20. It fulfills its
task of counter-balancing the subject in several ways: Against the brokenchord patterns dominating the first two-thirds of the subject, CS sets
stepwise motion, and against the stepwise motion at the end of the subject
it sets a broken dominant-seventh chord (see m. 42). While the subject is
made up of two subphrases, the counter-subject is conceived as one
indivisible line. And while the subject has two climaxesa softer one on
the third beat and a stronger one on the fifth beat after its beginningthe
counter-subject features a single
unbroken tension-curve. Its
climax falls either on the first
eighth-note, followed in this case
132
WTC I/7
E6 is the only episode to show a relationship to the subject: its first two
measures quote a variation of the first subphrase in their upper voice. All
other episodes are entirely independent from the primary material. They
present a number of characteristic motifs that are used with great consistency: E1 introduces the half-measure motif M1 featuring two broken
chords, each with a subsequent step downward to the next beat (mm. 2-3:
F-D-B-A, F-D-A-G). This motif plays a leading role in all episodes of
this fugue, occasionally in an extended version with a final unaccented
jump upward as in E3 and E4. E2 combines a prolonged M1 in the middle
voice with M2, a figure in the upper voice that also consists of broken
chords, this time in ascending direction. Significant features are the long
syncopations. M2 also recurs, sometimes unchanged, at other times represented only by the accented upward leap in eighth-note rhythm. E3, the
first episode in three-part texture, introduces M3 (M: mm. 7-8: F-D-B-E ),
a third motif that is subtly related to the other two: like M1 it begins with
a descending broken chord, and like M2 it ends with an accented upward
leap. Yet as its rhythm is built exclusively from eighth-notes, it is at the
same time quite distinct. These three motifs appear in a variety of combinations (see particularly in E3, E4, E5, and E8). The second half of E6
then introduces a new pattern that recurs shortly afterward in E7. In it,
sequences of ascending eighth-notes in the bass and a 16th-note dialogue
with varied segments of M1 in the two higher voices create a pattern that
we shall call M4.
Several relationships exist among the episodes of this fugue: the shape
of E5 and E2 is similar although the hands are inverted (mm. 4-5 . 19-20),
E3, E4a, and E8 are analogous except for the inversion of voices (mm.
7-10 . 12-15 . 30-33), and the second half of E6 is taken up, as was
mentioned above, in E7. The only episode segments to remain without any
correspondence are the cadential close of E4b with its preparation in mm.
15-173), the first half of E6 (E6a = mm. 22-241), and the final episode E9.
As all episodes apart from E6a are conceived as independent in
material and character from the subject and counter-subject, they should
sound like self-contained units, in a color and intensity distinctly different
from those characterizing the subject and its counter-subject. The dynamic
gestures within this contrasting and much lighter color are as follows: In
E1 the descending sequences create a natural decline of tension. Similarly,
the final episode E9 sounds in manifold falling lines. E2 with its analog in
E5 and E3 with its analogs in E4 and E8 all show relaxing tendency in
descending sequences. The first half of E6 also features a pitch direction
that points downward; here again the dynamic tendency is decreasing. The
E major
133
Note that the trills in mm. 7, 12, 21, and 27 require a double note to be played in the
middle of the trill motion. This becomes much easier in m. 30 if the twofold middle-voice
E is taken by the right handwhich is more comfortable than one might think. In m. 35,
the trills main portion can be played with the right hand while the suffix sounds smoother if
taken by the left.
134
WTC I/7
136
WTC I/8
This is very important, particularly since the somewhat complex rhythmic structure in the
piece may tempt performers into a transitional stage of counting in quarter-notes. While
this may be essential during the process of preparation, it is equally vital that this stage be
ultimately overcome in favor of the larger swinging pulse with only three beats to a
measure.
2
The use of the pedal is a very delicate matter and may appear somewhat complicated
where 16th-note motion in one of the lines requires a clean and unblurred rendition. The
only solution is thus to pedal on or slightly after the last rhythmic value in a measure. As
every performer will aim at evenness in tone color throughout each measure and each
phrase, this means that the pedal cannot be depressed before the last 16th-note fraction of
each half-note. The only stretches that might do without any pedal are the beginning of the
second half (mm. 20-21) and the cadenza-like measure near the end (m. 35).
E minor/D minor
137
138
WTC I/8
E minor/D minor
139
heard particularly well after the completion of the first perfect cadence: the
first subphrase sounds in the lower melodic voice, leading harmonically to
a dominant-seventh chord, while its answer, given in the upper voice,
supplies the respective resolution. In mm. 12-14 the motif is freely developed in the upper voice, followed by a variant in what appears as a middle
voice (mm. 14-16). A last statement, presented in the lower voice in m. 16,
is extended and merges into the cadential-bass pattern that ends this
section. Yet another variant emerges in mm. 22-25. Finally, toward the end
of the prelude, M1 recurs stripped of its question + answer design in the
form of a chain and in different harmonic guise.
M2 is introduced in mm. 20-21, in the very center of the 40-measure
prelude. Spanning only one measure, it shares its rhythmic and metric
features with M1. After only one beat, the lower voice sets in, creating a
stretto. The whole imitative pattern is then sequenced. The tension curve is
quite different from that in M1, mainly because the entire figure expresses
a single harmony. The climax falls on the beginning, the character is
stately. M2 is developed once, in mm. 26-28, where it appears in a singlevoiced version without imitation. Owing to the harmonic change at the end
of the first of these statements and the inverted pitch pattern in the second,
the dynamic curve is also inverted, with a crescendo to the final note of the
motif.
The interpretation of this prelude is determined by its broadly swaying
basic character and its texture: a melodic dialogue before the backdrop of
softly arpeggiated chords. Large-scale dramatic buildups are not the aim in
this meditative composition.
140
WTC I/8
E minor/D minor
1. mm. 1-3
2. mm. 3-6
3. mm. 8-10
4. mm. 12-14
5. mm. 19-22
6. mm. 20-22
7. mm. 24-26
8. mm. 24-26
9. mm. 26-29
10. mm. 27-30
11. mm. 30-32
12. mm. 36-38
M
U
L
L
M
U
U
M rh
U
M
U inv
M inv
141
13. mm. 39-41
14. mm. 44-47
15. mm. 45-47
16. mm. 47-50
17. mm. 47-50
18. mm. 52-53
19. mm. 52-53
20. mm. 52-53
21. mm. 54-55
22. mm. 54-55
23. mm. 54-55
24. mm. 57-60
L inv
L inv
U inv
M inv
U inv, rh,
L
M
U
L inv
M inv
U inv
U
Apart from the usual adjustments in the tonal answer, the most frequent
guise of the subject is the inversion (inv) in which all intervals appear
upside down. Augmentations (augm), i.e., statements in which each note
value is doubled, and rhythmic variations (rh) each occur three times. In
these variations, the dotted-quarter-note value of the subjects second note
also affects the fourth and sixth notes, thus stretching the first subphrase
in such a way that the second subphrase has to be shortened.4 Furthermore,
two-part and three-part strettos occur frequently.
Since this fugue does not feature any counter-subject, the intriguing
juxtapositions are those three-part strettos in which one of the voices
appears in augmentation. The example shows mm. 77-83:
Moreover, the first note is abridged in mm. 12 and 61, lengthened by anticipation in m. 26,
and ornamented in m. 39. The final note is omitted in m. 50, delayed in mm. 14 and 26, and
reached after an escape note in mm. 29 and 79 or after a chromatic passing note in mm. 6364. Finally, in the very dense three-part strettos of mm. 52-53 and 54-55, only the subjects
main thought can be tracked, but even this fragment contains irregularities at its end.
142
WTC I/8
They should be mentioned not because of their thematic importance but because a
performer would wish to pay attention to shaping them consistently.
In mm. 14-16, the middle voice line C-F-E-D-C is imitated an octave higher (see
U: mm. 16-17 from C).
The same line may also be recognized, with a variation in the first interval and a cut
before the last, in mm. 18-19 (see M: C-B-A-G).
Also in mm. 17-18, the upper-voice line C-D-E-F-E-D-C is followed by a varied
sequence (see U: mm. 18-19 from D).
The same curve recurs in inversion twice in mm. 33-34 (see U: from E; M: from G varied), and six more times, in several slightly different variations, in mm. 41-44 (see U:
m. 41 from G, m. 42 from F, m. 43 from C; M: m. 42 from C, m. 43 from F, m. 44
from F).
In mm. 82-85, overlapping with the end of the preceding subject entry, the same curve
recurs three times in inversion (see M: m. 82 from A, m. 83 from A, m. 84 from F).
Also in mm. 82-85, the syncopated final note of the subject entry, together with the
subsequent falling broken chord, sets the model for a little curve that recurs twice (see U:
m. 83 from A, m. 84 from F.
Finally, all three voices are involved in stating a four-note scale segment as an upbeat
figure in mm. 85-87.
E minor/D minor
143
144
WTC I/8
III
IV
VI
E minor/D minor
145
There are few explicit cadential closes in this fugue. The work features
a twofold harmonic development, with a return to the tonic and a new
progression occurring roughly in the middle of the fugue. The first four
separate statements are in D minor, followed by a modulation to the
dominant (A minor; cadence in m. 19). The first stretto then modulates to
F major, which is reached in closing formulas at m. 301. After this, the
three separate inverted statements return to the home key of D minor. The
second stretto group sets out from the tonic and leads to the dominant on
m. 521, thus repeating the harmonic process observed in the first section.
Subsequently, the fifth section begins and ends in A, with a short detour
to F in the second three-part stretto. Similarly, the augmented-entry strettos
begin and end in A while the middle group touches several other transitory keys. It is left to the final episode to modulate back to the home key.
Owing to this harmonic device, the impressive fugue ends without a coda.
146
WTC I/8
II
E
separate
entries
(original)
p-mp
III
E
two-part
strettos
(original)
dim
mf
separate
entries
(inverted)
dim
IV
E
V
E
two-part
strettos
(inverted)
p-mp dim
mf
VI
E
three-part
strettos
(orig/inv)
three-part
strettos
(w/augm.)
148
WTC I/9
These are: U: none; M: mm. 1: G, 5: E, 6: A, 10: all, 13: F, 15: C, 19: A, 20: D;
L: mm. 6: F, 7: G, 9: E, 10: all, 14: B, 20: B, 21: C.
E major
149
preceding the symbol designates. For performers with good finger dexterity,
a rendition with eight ornamental 32nd-notes sounds most convincing. A
longer ornament might blur the lower-voice entry of the motif.
m. 1
m. 7
m. 4
Note that in these measures, the motifs head carries on into another broken chord. For a
meaningful performance it is vital to distinguish between the original, active broken-chord
rise, found here on the first beat of each measure, and the passive one that serves as a
rhythmic extension to the melodic target note.
150
WTC I/9
E major
151
Only two of the subject statements are varied. The upper-voice entry in
mm. 20-21 includes an interesting octave displacement: a 16th-note-group
sets out one beat early from middle B. The extra time created by this
metric shift is filled with a figure leading to the climax C in the higher
octave. Similarly, the final entry also begins on the wrong beat. This time,
however, no correction is made. Instead, this statement concludes in its
metrically anticipated position and thus makes room, as it were, for an
accented cadential-bass leap to the tonic at the end of the piece.
The fugue features only one counter-subject. Introduced in the middle
voice where it is pitted against the second subject statement (from the
second 16th-note in m. 2 up to m. 31), it is almost omnipresent. This
companion is thus considerably longer than the subject. As it begins immediately after the downbeat, it bridges the three-eighth-notes rest that
precedes the first subject note. The counter-subjects pitch pattern and
perfectly even rhythm confirm the ornamental nature and, by extension,
the basic character of this fugue. A spiraling motion around a central note
(mm. 2-3: around E) and an obvious lack of harmonic or melodic features
that might enhance tension discourage any explicit dynamic shaping other
than a very soft rise at the beginning and a relaxation during the second
152
WTC I/9
half. For the same reasons and because of the very even rhythmic pattern,
subphrasing is also unlikely. Nevertheless, one can clearly distinguish two
segments. They are of equal length (each spans eight 16th-notes and ends
on a strong beat), which allows them to appear in various combinations.
Separate or joined, they play an important role in the fugue. The first
segment, CS-a, consists of a scale section from D to B followed by a
broken chord B-G-E. The second segment, CS-b, features the inverted turn
D-E-F-E and its varied repetition D-E-F-D. These segments appear
both as companions to subject entries and as episode motifs. They are
therefore worth being traced in detail.
As a companion to the subject, the complete version of CS-a + CS-b
appears only against the subject entries in mm. 2-3, 7-8, 20-21 and, with a
slight variation, in mm. 21-22. Other subject entries are accompanied by
1. CS-b + CS-b in sequential pattern (see mm. 3-4, 25-26)
2. CS-b + CS-b in a two-part imitative setting (see mm. 9-10)
3. a free figuration followed by CS-b (see mm. 6-7)
4. CS-b dissolving into a free figuration (see mm. 28-29)
5. no counter-subject material at all (see mm. 16-17, 19-20).
Within the episodes, the segments may appear in the counter-subject
order of CS-a + CS-b (mm. 4-5 L: D-D; mm. 8-9 M: A-A). In this order,
they may form descending sequences (mm. 11-12 and 12-13 L: G-G and
E-E; mm. 13-14 and 14-15 U: B-B and A-A; mm. 17-18 and 18-19 M:
B-B and G-G; mm. 22-23, 23-24 and 24-25 M: D-D, C-C and B-B).
They may even incorporate small variations that change the harmonic
outline (mm. 26-27 and 27-28 U: D-G and F-B). Or the episodes may
recall only one of the counter-subjects segments, either separately (m. 3
first half U: A-B) or in ascending sequences of CS-a (mm. 5-6 L: C-C,
mm. 15-16 U: G-G). In other words, there is not a single measure in the
E-major fugue without some quotation of the counter-subject segments.
Because of its unusual nature without any explicit dynamic shaping,
the counter-subject does not challenge the subjects supremacy. Yet one
must distinguish different levels of intensity in its 16th-notes and shape the
beginning carefully: while the
four ascending 16th-notes that
S
end the subject constitute a distinct diminuendo, the ensuing
group of ascending 16th-notes
represents a tension increase
(though a mild one owing to the
ornamental structure).
E major
153
154
WTC I/9
M2 is more prominent than M1 both in length and in further impact. It is set more clearly
apart from the ornamental character of this fugue owing to the melodic legato found in its
appoggiatura-resolution figure, and its descending sequences effect an overall relaxation.
E major
155
The tempo proportion of prelude and fugue also allows for two
choices, depending on just how fluent or how gently flowing the prelude is
taken. If the prelude is played with a pulse of metronome 80 or above, the
proportion can be simple: a beat (dotted quarter-note) in the prelude
becomes a beat (quarter-note) in the fugue. If the prelude is played with a
considerably slower pulse, a translation of eighth-notes into eighth-notes
works best. (Approximate metronome settings: [a] 96 for the dotted quarternotes in the prelude and the quarter-notes in the fugue; [b] 72 for the dotted
quarter-notes in the prelude; 108 for the quarter-notes in the fugue.)
The paramount force determining the design is this fugue lies in the
two explicit cadential formulas as well as in the structural analogies. Some
of these correspondences were already mentioned above, some should be
added here. Summing up, these are the factors determining the layout of
the fugue: E2 and E4 both consist of two segments, the first ending with a
cadential close emphasized by a closing formula, the second creating a
color contrast by introducing a motif that is unrelated to the primary
thematic material. Interpreting these two analogous cadences as the ends of
the first and second sections respectively, and looking backward from
these ends, one can observe that both are preceded by the same pattern of
two subject statements (on tonic and dominant), followed by an episode
that serves as a bridge, and complemented by another statement in the third
voice (mm. 1-43 . 64-113). The second section differs from the first owing
to the extensions of roughly one measures length each at both sides, i.e.,
additional episode measures both before the above-mentioned pattern.
Another analogy occurs between the second segment of E4 and E6.
Both present almost exactly the same material, albeit in different keys and
with the upper and middle voices inverted. If the second segment of E4 is
regarded as the beginning of the third section in this fugue, which follows
from what was said above, it would be logical to accept E6 as the beginning of the fourth section. Looking back again from the respective section
endings one can detect the following analogy: The three subject statements
preceding the end of the third section recall those of the first section. Like
them they are presented on the tonic, the dominant, and the tonic (mm. 1-43
. 193-22). The episode that, in both the first and the second sections,
serves as a bridge inside this pattern, now connects the entire pattern with
the additional subject statement at the beginning of the third section. While
we found that the second section was extended in comparison to the first,
we can now confirm that the third section continues this process of
extension at least at the beginning: the episode-segment opening the
section is longer (mm. 13-163 . 53-6), the linking episode inside the
156
WTC I/9
section is also longer (mm. 17-193 . 83-9 . m. 3), and there is an additional subject statement.
While the third section reveals itself as related to the first by the key of
its three subject statements, the fourth section similarly recalls the second
one. The two statements in the fourth section take up not only the keys of
the first and last statement in the second section, they even appear in the
same voices (mm. 25-26 . 6-7: statement in U on the tonic, and mm.
28-29 . 9-10: statement in L on the dominant).
The Wilhelm Friedemann version differs little from the Well-Tempered Clavier version:
The G in m. 91 avoids the seventh jump downward by appearing one octave higher; the line
is thus smoother, but the bass-note descent from m. 6 to m. 9 is less satisfactory without its
expected target in the lower octave. In m. 13, the re-sharpening of the F already occurs in
the second half of the measure and not, as in the later version, only in m. 14. In the second
half of m. 15, the original version features G-B-C-D-C-D-C-B, the harmonically logical
figure that, in the adaptation, is modified to G-A-B-C-B-C-B-A in favor of an assimilation
with the ensuing measures. Finally, mm. 21-22 were originally erected on a repeated tonic
pedal. As the reworked prelude for the Well-Tempered Clavier is destined to continue with
a weighty second half, these measures are changed to support the intended modulation.
157
158
WTC I/10
E minor
159
shortest appearing values (i.e., with four notes to a 16th-note). For pianists
who find this technically too demanding or musically too congested, there
are, in this particular piece, indicators why a trill speed in slower values
might be not only acceptable but, some believe, even advisable: The fastest
occurring note values, the 32nd-notes, all appear in a strictly ornamental
context. These groups could thus be interpreted as spelled-out unaccented
ornaments and embellishments (see e.g., m. 1: an upbeat turn, m. 3: an
embellishment of the step A-G, etc.). It would therefore make sense to play
the ornaments appearing as symbols in the same motion. Moreover, four of
the five trills end in suffixes that are spelled out. This is either because an
interrupted trill is normally not expected to conclude with a suffix (as in
mm. 10 and 12), or because Bach desires a suffix differing from the
common shape (as in m. 14). Note, however, that the very regular suffix in
m. 1 is also written out. These suffixes appear in 32nd-note notes, a fact
that supports the rendition of the entire trills in 32nd-notes.
The four trills that are approached in note repetition (mm. 1, 10, 12,
and 20) start on the upper neighbor note. The fifth (m. 14), preceded by
stepwise motion, begins on the main (16th-) note. Regarding their ending,
only the trills in mm. 1 and 14 feature regular motion up to the written-out
suffix. The others are either interrupted by a sudden rest (cutting out one
32nd-note pair but leaving the remainder intact; see mm. 10 and 12), or
they end in an anticipated resolution, requiring a point darrt stop before
the anticipation (see m. 20). The example shows the complete rendition of
these trills in the slower of the two possible speeds.
160
WTC I/10
E minor
161
This leaves only the final measures of each half of the prelude to be
discussed. In the first half, mm. 21-22 modulate actively toward A minor,
thereby suggesting an increase in tension. In mm. 33-41, a rather abrupt
initial increase is followed by the considerably softer beginning of the
pedal note B. For the duration of this pedal (mm. 343-39), the dynamic
growth is smooth and gradual. However, it manifests itself here not only in
the obstinate repetition of the bass note but also in an intensification of the
texture from three to four voices (compare m. 34 with m. 38. Note that this
increase of texture does not change the essential figures in the least). The
logical target for this crescendo is the interrupted cadence in m. 40, after
which the tonic of the home-key is approached in an overall diminuendo.
The simplified outline of the prelude in E minor given in the following
music example tries to capture these processes.
162
WTC I/10
regarded as the final note, thus creating an unusual overlapping of the end
of the first subject statement with the beginning of the subsequent entry,
lies in the subjects melodic structure. The pitch pattern is unusual in that
it is composed in hidden two-part structure. In an attempt to determine
which notes belong to the melodic strand and which to the harmonic background, the following layers can be revealed: The melodic strand comprises an initial E (represented by either or both of the key notes in the first
chord), followed by a chromatic descent to the B on the downbeat of m. 2.
The end of the line is then provided by the A on the second beat in m. 2
and the two upbeat groups
G-F-G and F-E-D. The harmonic background is represented first by the third and
fifth of the tonic and then by
a keynote pedal in an offbeat
repeated-note pattern. This
segment of the backdrop is concluded by the ornamented keynote pedal
(E-D-E) after the downbeat of m. 2. The modulation to B minor manifests
in this strand with the offbeat notes C and A in m. 2. And these find their
resolution onto the target key on the equally offbeat B in m. 3. The example
above spells out this hidden
two-part structure. Consecutive notes hardly ever belong
to the same strand, thereby
transcending the concept of
intervals.
The rhythm in the subject comprises only 16th-notes and eighth-notes.
These two basic values remain predominant throughout the composition,
and even the third rhythmic unit, the tied-over quarter-note in the countersubject, does not change the fundamentally simple rhythmic pattern. The
subjects harmonic progression, too, is very simple. Aside from the initial
tonic and the two chords representing the modulation it only contains a
rudimentary subdominant in m. 13.
The swift motion permits no emotional involvement in melodic details.
At the same time, neither the simplicity of the harmonic design nor the
rhythmic pattern in uninterrupted 16th-notes offers any climax. Therefore,
the only feature capable of influencing the development of tension in this
subject is the descent in the melodic part of the structure. This leaves the
subject with a very straightforward dynamic gesture: an energetic beginning followed by a gradual decrease in tension.
E minor
163
164
WTC I/10
The eight subject statements in this fugue are very neatly grouped in
four pairs, each of which is followed by a subject-free passage.
E1 mm. 5-10
E3
mm. 24-29
E2 mm. 15-19
E4
mm. 34-42
The second half of E4 recalls subject segments (compare U: m. 39 and
L: m. 40 with the first subject measure, as well as U: mm. 41-42 with the
second half of the subject). E1 and E3 quote the counter-subject, with the
expected final note replaced by a large leap downward that is followed by
the rising sixth with which the subject concludes. One might describe this
hybrid figure as counter-subject body with subject tail. Its companion is
the first genuine, i.e., independent episode motif (see U: mm. 5-7). M1
consists of two rising broken chords followed by a four-note ornament, a
descending scale, and a long closing note. Two dynamic designs are
possible in this episode motif. The climax can fall either on the peak note
of the broken-chord sequence (e.g., the G in m. 5) or on the rhythmically
exposed tied note (e.g., the D in m. 6). The first rendition underscores the
motifs independence while the second, owing to the analogous length of
the climax note with that in the counter-subject, makes it sound like a free
imitation of the other voice. E2 and E4 provide two more motifs. The
lower voice in mm. 15-16 introduces M2, which is characterized by a
simple broken chord in eighth-note motion complemented by the repetition
of the chords third and octave. The upper voice contrasts this with M3,
which consists of two descending scale segments in an uninterrupted
16th-note pattern. With these two motifs Bach achieves a very balanced
effect by subtle means: the two ascents in M2 are contrasted in M3 by two
descents, and while in M2 the four-note broken chord was followed by the
shorter two-note leap, M3 begins with the shorter scale segment that is
then followed by the longer one.
The combination of M1 and the hybrid motif described above as a
counter-subject body with subject tail is followed by a descending
sequence and rounded off by two measures of almost completely parallel
structure in the two voices (see mm. 9-11 and 28-30). Similarly, the combination of M2 and M3, covering only a single measure, is followed by a
double imitation and a descending sequence before this episode-type, too,
is rounded off by a parallel motion of the two voices (see mm. 19-20 and
38-39). The relationship between the episodes is thus very straightforward:
E1 recurs in E3 in inverted voices. Similarly, E2 is recapitulated in
inversion in the first half of E4 (mm. 15-19 . 34-38).
The role the episodes play in the development of the composition is
only slightly more complex. E2 and the first half of E4, consisting entirely
E minor
165
166
WTC I/10
The dynamic outline follows the large-scale analogy with two gradual
build-ups. The tension continues to increase through the bridging episodes
E1 and E3 and is released in the concluding episodes of each half.
168
WTC I/11
F major
169
170
WTC I/11
F major
171
172
WTC I/11
M: mm. 56-57 is imitated through all voices in mm. 57-61 and 63-64. By
contrast, E4 consists of nothing but a cadential close, as does the first
segment of E5, which repeats this cadence in varied transposition. The
final episode, too, is only a slightly more elaborate cadential formula. This
structural equivalence of the cadential closes aside, the episodes in this
fugue show neither analogies nor other relevant patterns.
The role each episode plays in the development of this composition can
be easily deduced from what has been observed so far:
E1, anticipating the ensuing counter-subject appearance, links two
statements, as does E2, quoting the entire counter-subject against
rising sequences.
E3 with its descending sequences brings a slight release in tension.
E4, acting as a cadential close, represents a definite relaxation.
E5a creates the same effect of relaxation, whereas E5b, determined
by its ascending eighth-note scales, creates the impression of an
increase in tension.
E6 sets out with the same ascending eighth-notes but then relaxes in
the final cadence.
The simple rhythmic pattern in conjunction with the scalar or ornamental structure of the 16th-notes indicates a rather lively basic character.
The tempo should be fast enough to reflect the bouncing mood of the
eighth-notes and the ornamental nature of the 16th-notes. The appropriate
articulation is quasi legato for the 16th-notes and non legato for all eighthnotes except for those in appoggiaturas and closing formulas.2
The relative tempo of the prelude to the fugue poses a problem because
of the inherent triple meter in both movements. The simple proportion, in
which an eighth-note in the prelude corresponds with an eighth-note in the
fugue, is possible. It strings the two pieces so firmly together as to almost
blur the confines of each. Moreover, there is the danger of monotony in
two consecutive pieces both based on mainly two note values in exactly
the same speed (dotted quarter-notes = 63). I propose another proportion,
which may seem more complicated but has the advantage of giving a
livelier, more interesting result. In it, an artifical group of three 16th-notes
in the prelude corresponds with an eighth-note in the fugue. (Approximate
metronome settings: dotted quarter-notes in the prelude = 72, eighth-notes
in the fugue = 144.) The different speed helps to differentiate touch and
color in the fugues heavier eighth-notes from the preludes lighter ones.
2
In mm. 45, 55, and 71, the ornamented notes in the upper voice are not detached from the
notes that follow. Likewise in the middle voice, the do-si-do figure in mm. 55-56 is legato.
F major
173
The fugue features mainly two kinds of ornaments: a long trill on the
penultimate note of the counter-subject and mordents in the typical closing
formulas. The trill in the counter-subject abides by the same rules that
would apply to a similar trill in the subject: it is approached stepwise and
thus begins on the main note, its speed is in 32nd-notes as these are twice
as fast as the shorter written-out note values, and it ends in a suffix before
resolving stepwise. The ornaments in the closing formulas (mm. 45, 55,
and 71) are short since their resolutions are in each case anticipated and
fall before the next strong beat, thus demanding a point darrt. They all
begin on the upper neighbor note and move in 32nd-notes or slightly
faster. In the two cases with dotted-note rhythmic figures, the ornament
stops short before the third eighth-note beat. The ornament in m. 28 is a
variation of the long trill mentioned above since the upper voice in mm.
26-29 features a variation of CS. Although shorter here, its beginning,
speed, and end should match the long trill. Finally, the mordent at m. 481
begins on the upper neighbor note and fills the note value with four swift
sixty-fourth-notes.3
Several features in this fugue provide indications for its design. The
subject statements beginning in mm. 21 and 46 both sound in reduced
ensemble and should be earmarked for possible section beginnings. In a
third instance, the stretto launched in m. 36 is accompanied by a sustained
pedal and thus also creates the effect of not actively involving all three
voices. The episodes between the second and third statements (E1) and
between the third and fourth ones (E2) are both related to the primary
material insofar as they quote the counter-subject. They thus suggest a link
between entries belonging to a single structural section. Furthermore, both
episodes have an ascending pitch outline. E3, by contrast, serves as a kind
of extension to the preceding subject statement and shows a descending
pitch line. The triple stretto + closing formula in mm. 36-46 is structurally
analogous to the triple stretto + closing formula in mm. 46-56. The final
cadential close is built along the same pattern as the two preceding ones
(compare particularly mm. 71-72 with mm. 45-46).
Harmonically, the fugue remains rooted in F major until m. 31. The
subsequent measures undertake a shift to D minor (see m. 32, reinforced
by its dominant, the A-major chord in mm. 34 and 36), which becomes the
basis of the triple stretto from m. 36 onward. A cadence in this relativeminor key closes this segment in m. 46. The following measures pick up
3
As this ornament seems somewhat out of keeping with the remainder of the piece, it may
be worth considering whether one might ignore the indication.
174
WTC I/11
the triple stretto in inverse order of entering voices and in G minor, the
relative minor to the subdominant. This key is confirmed with an explicit
cadence in m. 56. Two measures after this cadence the fugue is back in the
home key of F major, in which the final, varied subject statement appears.
This is again rounded off with a cadential formula similar to the two
previous ones.
The design that becomes apparent from these observations is unusual
insofar as it reveals a fugue consisting of parts rather than sections.
The first part, encompassing two sections, is united by the home key; its
two sections are glued together by the bridging power of E2 (which does
not allow for phrasing before the fourth statement, although the entering
order would suggest this) and by the density of material immediately afterward (which does not allow for phrasing after the fourth statement either,
although the reduced texture would suggest this). This part is rounded
off by E3, the relaxing episode in descending sequences that introduces the
modulation. Although the second part does not literally encompass
several sections, the explicit closing formulas in E4 and E5a give the
impression of three segments.
The first half of the fugue, welded together of two sections, describes
a gradual increase of tension. In the second half, each of the three segments
begins with a climax in mighty intensification followed by an episode
concluding in complete relaxation.
176
WTC I/12
F minor
177
178
WTC I/12
crescendo. In mm. 5-63, the tension decreases slightly as all melodic
figures point downward (see in the soprano the central notes C-B,
in the alto the descending sequence, and in the bass the target notes
B, preceded by its leading-note A, and A, preceded by cadential
steps). The extended parallel ascent in mm. 63-73, in which all four
voices participate, then provides an even more intense increase of
tension, which peaks on the syncopated high F, relaxing afterward
toward the cadential close on m. 91.
The second section is distinguished from the first by a much higher
portion of highly emotional intervals. The recurrence of the hidden
parallels (mm. 10-11) prepares the ascent to the climax in m. 113,
after which the successive pairs of leading-note + resolution in
m. 12 bring about a prompt relaxation (see alto: B-C, F-G; tenor:
B-C; alto: F-E; tenor: F-E). The ensuing 3 measures of the third
section contain two tension-curves: a gentle one in mm. 13-14 and
a slightly steeper one beginning with the last 16th-notes in m. 14,
climaxing in m. 152, and ending in the interrupted cadence. The
fourth section begins with a dynamic curve identical to that at the
beginning of the piece. Its descent, however, is extended all the
way to the end of m. 18. A final curve (short increase and longer
decrease) concludes the prelude on a soft note.
In the overall design, the moments of highest tension are at the end
of the first section (where the drive is greatest in the extended
parallel) and at the beginning of the second section (where the
interval intensity is highest). As this interpretation renders the 16thnote motion as an uninterrupted line to which all voices contribute
in complementing manner, no melodic phrasing occurs.
b) If the prelude is regarded as a metrically oriented piece, the features
determining the performance will be sought in the quarter-notes.
The focus in the texture now tends toward the lower voices, and all
dynamic inflections are of a delicate nature. Within the initial
section, a first gentle curve with a climax in the middle of m. 1 is
followed by sequencing one-measure groups. As the bass in these
groups falls ever further, a gradual diminuendo results. This leads
to a state of floating weightlessness in the measure where the
quarter-note motion is temporarily suspended (see mm. 53-63). The
subsequent extended ascent leads to a slightly more pronounced
climax on the downbeat of m. 8, followed by a relaxation through
the cadential close.
F minor
179
The two sections that build the middle of the composition are less
regular in their pulse, featuring several passages in which quarternotes are interspersed with half-note beats (see the first halves of
mm. 9, 10, and 12 as well as the second half of m. 14). This impairs
the effect of meditative calm that prevails in the outer sections.
The soothing continuity of the quarter-note beats is restored with
the interrupted cadence and not broken again before the prelude
ends. Dynamic increases and decreases are even more subdued in
this section than in the preceding ones, concluding the composition
in a state of complete calm.
c) If the prelude is regarded as a structurally oriented piece with the
outline of an invention, the focus will be on the components of the
thematic material. These components are introduced within the
initial two measures. The example shows the components, in the
order of their importance in the composition, with their inherent
dynamic shape:
the main motif
The main motif appears eleven times, more often than not with
a variation of its end. These are its statements: (ext = extended, var
= varied; inv = inverted)
m.
1 S
mm. 6-7 ext T
mm. 13-14 var T
mm. 2-3 T
mm. 9-10 var T
mm. 15-16 inv B
mm. 3-4 T
mm. 10-11
T
mm. 16-17
S
mm. 4-5 T
m. 13 var
S
In four of these eleven statements (mm. 2-3, 3-4, 4-5, and 10-11),
the motif is accompanied by its regular companion. On two other
occasions (mm. 6-7 and, in inversion, mm. 15-16), it comes with a
varied version of the stepwise ascent characterizing the companion.
The syncopated figure (mm. 2, 4, 5, 6, 11, and 17) and the relaxing
link (mm. 2, 3, 5, 5-6, 12, 15, 17, 18 [2x]), 19 [2x]), and 21 [2x])
account for much of the remainder of the piece.
180
WTC I/12
The first section is thematically dense, with main-motif entrances in
almost every measure and, after a short relaxation in mm. 5-6, an
extended quotation of the motif in the left-hand part. The two
sections that form the preludes center feature thematic measures
interspersed with cadential or imperfect closes. The final section
provides a liquidation of the thematic material. In this structurally
oriented interpretation, this liquidation together with the extended
pedal bass determines this last section as an overall relaxation in
which the pitch curve in mm. 19-20 plays only a subordinate role.
F minor
181
The dynamic outline should mirror the melodic and harmonic particularities. Beginning on the exposed fifth degree requires a slightly elevated
tension in the first note. The ensuing steps toward the natural leading-note
and on to the artificial leading-note create a powerful increase. Next, the
extra-scalar fourth interval that reaches the third leading-note represents
the apex of this dynamic curve. (For any wind or string instrument, the
climax would be between the two notes B and Ea solution that is not,
unfortunately, open to keyboard players.) The resolution onto the keynote
F brings such a relief after all these daring harmonic twists that it involves
an almost complete decrease of tension. What little is left will abate
gradually through the chromatic descent.
The 58-measure fugue comprises only ten subject statements:
1. mm. 1-4
T
6. mm. 27-30
B
2. mm. 4-7
A
7. mm. 34-37
A
3. mm. 7-10
B
8. mm. 40-43
T
4. mm. 13-16
S
9. mm. 47-50
S
5. mm. 19-22
T
10. mm. 53-56
B
182
WTC I/12
S
CS3
CS2
CS1
F minor
183
The further development of the counter-subjects includes several irregularities that may make a convincing rendition of their phrase structure and
dynamic design difficult. The two most confusing details are the crossover
of voices1 and the swapping of sub-phrases.2
The F-minor fugue contains eight subject-free passages.
E1 mm. 10-13
E5 mm. 37-40
E2 mm. 16-19
E6 mm. 43-47
E3 mm. 22-27
E7 mm. 50-53
E4 mm. 30-34
E8 mm. 56-58
No episode is related to the subject. Instead, all seem to derive, in one
way or another, from the first counter-subject. It is helpful to distinguish
two main types of episode in this fugue: E1 is dominated by a motif that
uses the first subphrase of CS1 and continues with a segment from CS2 (A
mm. 103-11 . T mm. 83-91: a quarter-note followed after a 16th-note tie or
rest by a descent in 16th-notes). The CS1-segment is imitated in the bass
with a new ending, partly doubled by the tenor. The ensuing measures
sound like descending sequences.3 This episode recurs in exchanged voices
and with some variation in E4. Here, the bass is the leading voice, imitated
by the tenor, while the soprano adds an extended parallel to the descent.
1
After the middle beat in m. 7, the tenor interval A-F crosses over the descending alto leap
F-E; they regain their normal position in the middle of m. 8. In m. 13 immediately after the
middle beat, the same tenor interval moves the voice into what seems like alto position;
when the alto comes in with the upbeat to m. 14, it does so considerably lower than the
tenor; they recoup their ordinary places with the fourth beat of m. 14. In mm. 292-303,
soprano and alto cross. In mm. 47-48, there is such a gap between soprano and alto that one
easily mistakes the alto for a lower voice. When the tenor comes in an octave above the
alto, the confusion is complete. However, the voices disentangle on the last beat of m. 48.
2
In m. 19, the alto begins with the first subphrase of CS1 but carries on with the remainder
of CS2 (which, on top of everything else, involves an octave displacement on m. 212). At
the same time, the soprano (after having set off, in m. 19, with notes not belonging to the
primary material) takes over CS1, which the alto had abandoned, and concludes it properly.
In mm. 27-30, the tenor quotes CS1 but reduces its ninth leap to a simple second. The alto
presents the first subphrase of CS2 (see G-C-B-A), continues with notes aping the large
leap missed out by the tenor, and from the middle beat onward finds itself in CS3. The
soprano, after three non-consequential notes, completes the interrupted CS2. In mm. 34-37,
the first subphrase of CS1 is heard in inversion in the tenor, while the remainder sounds in
the soprano. The final trill is prevented by a tied note, the resolution avoided altogether. In
mm. 47-50, CS2 appears in the (alto-positioned) tenor, deprived of its first subphrase and
with a varied ending. Similarly, the CS1 statement beginning in the alto is allowed neither
its ninth leap nor its final trill, which is substituted by a written-out figure-work.
3
There is no way to make the actual notation heard, which has alto and tenor inverted; this
fact would therefore appear inconsequential for the interpretation.
184
WTC I/12
E6 also relies on the same material; this time, the alto takes the lead. (Here
Bach strings the motif to its sequence by filling the phrasing space with an
additional 16th-note; see the D after the middle beat of m. 44.) The imitation is presented in the bass, with soprano and tenor accompanying.
A second episode-type is established in E2. Here the initial impulse is
given in the soprano by a three-note figure that quotes only half of the first
CS1 subphrase. A counterpart enters with the bass, which derives its figure
from the inversion of the same first CS1 subphrase (the extension bends
back upward). The texture is completed by the imitation of the bass figure,
sounding in the alto, with the extension of the inverted CS1 subphrase
continuing downward. As in the first episode-type, this one-measure model
is also sequenced twice (mm. 16-17 . 17-18 . 18-19). This episode recurs
in the first half of E3 (mm. 16-19 . 22-25). Here, the leading voice is the
alto (which is not at all easy to convey under the parallel soprano). Its main
counterpart sounds in the tenor, with the imitation in the soprano. The
second half of this episode is based on the same material but uses it freely.
E7 is another episode to follow the model of this second episode-type. The
texture is very similar to that found in E2; the soprano is in the lead, while
the lowest voice, here the tenor, sounds the counterpart and its imitation
appears in the alto (mm. 16-19 ~ 50-53).
The two remaining episodes, E5 and E8, do not follow either of these
models. E5 features a figure derived from the first subphrase of CS1 (see
bass m. 37, sequenced in m. 38, and m. 39, sequenced in 40). The other
two voices involved in this episode recall the tied-quarter-note-plus-3/16note figure from CS2, matched here with free extensions. E8 features a
measure presenting nothing but this CS2 segment in four-part texture,
followed by two cadential measures.
The role played by each episode in the dynamic outline of this fugue is
determined mainly by their ascending or descending sequence patterns.
Descending sequences are found in E1, E4, and E8. Episodes that embark
on a relaxation before suddenly turning into a final crescendo are E3, E5,
and E6, while in both E2 and E7, the ascending motion dominates.
Both the complexity of the rhythmic pattern in this fugue and the high
degree of intensity expressed in the interval structure indicate a rather calm
basic character. The tempo is confined by features inherent in the composition. On the one hand, the 16th-notes must be calm enough to allow for
the expression of melodic intensity; on the other hand, the quarter-notes
must convey the impression of stringent movement in order to be felt as
the relevant pulse (instead of surrendering this task, as often happens, to
the eighth-notes).
F minor
185
186
WTC I/12
The two cadential ornaments are also note-filling trills. The one on the
sopranos E begins on the main note, while the compound ornament in the
tenor begins as indicated from below. Both then move in 32nd-notes and
end in parallel suffixes.
The episodes in this fugue end in unusual ways. All of them close on
an imperfect cadence, and several even present these as unresolved chords
(see E1 in m. 13: F is suspended into a chord on C; similarly E2 in m. 19
and E3 in m. 27). This makes the episodes appear ill-suited as section
endings. By contrast, a cadential pattern that conveys structural closure
occurs in the context of a subject statement, at the end of the tenor entry in
mm. 40-43. These observations suggest that in this composition, episodes
are not conceived to round off preceding statements but rather to prepare
ensuing ones. A further factor that may give hints regarding the structure
of this fugue is Bachs use of the ensemble. Two subject statements appear
in reduced ensemble (T: mm. 19-22 and S: mm. 47-50) and are therefore
likely to mark section beginnings.
The harmonic progress through the subject statements is straightforward. The first six statements sound in the tonic area; four of them begin
like the first entry on C, only two in the dominant position on F (tonal
answer, m. 4) or G (real answer, m. 19). The seventh and eighth statements
relate harmonically to the area of the relative major. The former (mm. 34-37)
represents the tonic position of A major but is harmonized with reference
to F minor; the latter reads like the dominant of A and ends accordingly.
In the detailed steps of their harmonic progression, however, both are not
in keeping with the subjects original harmonization. The two final subject
statements return to the F-minor key, representing the dominant and tonic
positions respectively.
The design of the fugue thus presents itself as follows: The two statements in reduced ensemble must be regarded as first statements in their
respective sections. These two sections therefore begin with E2 (in m. 16)
and with E6 (in m. 43). The latter section-beginning is confirmed by the
explicit cadential-bass pattern at the end of the preceding statement. The
two statements harmonically referring to the relative major belong together
and form a section of their own. This section begins with E4 and ends with
the cadential-bass steps in m. 43. Its two entries are further united by their
harmonization of the subject, which deviates from the pattern established
earlier in the fugue, and by the fact that the subject appears abandoned by
its counter-subjects: CS2 and CS3 are entirely missing in both statements
while CS1 occurs only against the first of these entries and is varied at both
ends (see S: mm. 34-37).
F minor
187
Within the first section, the tension rises steadily from one subject
entry to the next. An increasing density of the material, up to a texture that
confronts the subject with three counter-subjects, supports an equally
growing intensity. The episode preceding the fourth statement grants a
transitory change of color along with a short relaxation, which is cut short
by the full force of four thematic voices. The end of the section seems
somewhat unresolved since a suspended G mars the final F-minor chord in
m. 16. This creates an effect as if this section was strung, if only by a thin
thread, to the subsequent one.
The second section begins with an episode of rising tension, causing its
first statement to enter already on a somewhat elevated level. The ensuing
episode generates a decrease of tension in its first half (up to the transitory
188
WTC I/12
cadence to A major in m. 25) but then turns around and prepares for the
next subject statement with a dynamic increase in mm. 26-27. This statement equals the final statement of the first section in the density of its
material and, consequently, in its intensity. Its ending furnishes the longexpected perfect cadence in F minor which, this time, sounds unimpaired
by any suspended voice.
The episode opening section III provides a more drastic change of
color owing to the position of its material: the two motivic parts are in the
bass and tenor, the alto pauses, and the soprano contributes only a parallel,
without independent force of its own. This darker color is enhanced in the
descending sequences. The episode linking this sections two entries is
built similarly to that linking the two statements in the second section. Its
first measures feature descending sequences, followed after a transitory
cadence in E major (m. 39) by a last-minute preparation for the ensuing
subject entry. The two statements themselves create much less tension than
those in preceding sections: with regard to texture, because of the much
lower density in thematic material; with regard to melody, because of the
major sixth that is more neutral in its tendency than the strong minor sixth;
and with regard to harmony, because their dissimilar harmonization omits
all the most striking steps. The second entryand with it the section
closes, as was mentioned before, in a definite cadence.
The fourth section begins with a four-part episode presenting three
measures of high material intensity. This comes as a surprise after the
preceding subject statement, which generated so little thematic motion
around it. The sequences, and with them the tension, descend toward the
middle of m. 46 but are followed by a most powerful, partly chromatic
ascent in the three lower voices. To fulfill the expectations raised in this
sudden crescendo, the soprano with the ensuing subject entry and the tenor
with CS2 begin almost two octaves higher than they ended a little earlier.
With the bass pausing, this creates a most striking shift and propels the
tension up immediately. The linking episode in the middle of this section
combines rising and falling tendencies, leaving the final subject entry as
unprepared as was the last statement in the first section. This statement
returns to the register in the center of the keyboard and to four-part texture
but replaces two of its counter-subjects with syncopated appoggiaturas (see
T mm. 53-54: C-B, A-G and A mm. 54-55: D-C, G-F). At the closure of
this fugue, Bach thus substitutes thematic density with emotional urgency
and creates a new kind of climax. The final episode can do nothing but
gently resolve this cumulated tension.
190
WTC I/13
From these two requirements results the following articulation in the bass line:
legato
mm. 1-3 (E); phrasing before C; legato mm. 3-5 (C-C)
non legato
in the cadential pattern mm. 5-6
legato
mm. 7-10 (E); phrasing before D; legato mm. 10-11 (D-D)
non legato
in the cadential pattern mm. 11-12
legato
mm. 12-14 (A); non legato mm. 14 (A)-15
legato
mm. 16 (first half); non legato mm. 16 (C)-17
legato
mm. 19-20 (F); phrasing before B
legato
mm. 20 (B)-21 (C), phrasing
legato
mm. 21 (D)-22 (G); legato mm. 22 (F)-23 (F) (non legato also possible)
non legato
in the cadential pattern mm. 23 (A)-24 (F)
legato
mm. 24 (F)-27 (tied C); non legato up to the end.
2
F major
191
beat but is always both delayed and approached indirectly. This thematic
ornament can be rendered as a mordent (as indicated in m. 1), in which
case it includes four orbettersix notes whenever it begins on the upper
neighbor note, as in all cases of note repetition, and five notes when it
begins on the main note as in m. 121, or as a compound ornament (as indicated in m. 7), in which case it comprises eight notes, beginning always on
the upper note, describing a turn to the lower neighbor and back to the
main note, and ending with two trill pairs. The appropriate rhythmic
position of this interrupted ornament is such that the point of interruption
avoids any beat. The example shows two possible solutions for each version:
The initial section features all three thematic components. The brokenchord figure begins in the upper voice and is imitated in the lower voice.
The complementary-rhythm figure consists of two segments with overlapping phrase endings (phrasing in U after m. 41, in L already after the third
dotted eighth-note in m. 3). The closing formula begins in the right-hand
part on the middle beat of m. 5, in the left-hand part on the second dotted
eighth-note of the same measure. Harmonically this section describes a
modulation from F major to the dominant C major.
In the second section, the three components appear with different emphasis. The broken-chord figure is merely transposed; the complementaryrhythm figure much extended. Two descending subphrases in the treble
(see mm. 78-87, mm. 88-101) are accompanied in the lower voice by a single
unphrased descent followed by an additional curve. The closing formula,
however, is only represented by its bass part while the upper voice
continues in the complementary-rhythm pattern. The harmonic progression
leads from C major to D minor, the relative minor key of the tonic.
The third section is considerably shorter but features all components in
compressed versions. It begins at m. 121 with the broken-chord figure in
the bass, imitated in the treble before returning once more to the bass. The
bass then gives way almost directly to an extended cadential pattern. In the
treble, a little link precedes a very condensed complementary-rhythm figure
followed by the complete closing formula. The structure of the fourth
section is similar but even more compressed.
192
WTC I/13
F major
193
The only modifications in the subject, both very common ones, occur
at its beginning and end. The initial interval is adjusted in the first answer
but never again thereafter, and the final interval is on two occasions (mm.
6-7 and 21-22) enlarged to a fifth to include a drop to the keynote. No
stretto or parallel statements of the subject appear.
The two counter-subjects are of very different character and importance.
CS1 is introduced against the second subject statement (see U: mm. 3-5).
The first of its two subphrases begins with an ascending octave followed
by a circling figure and a halt on a syncopation. The second subphrase is
an extended transposition of the first, beginning with a longer preparation
and closing with an additional dosido formula.3 The climaxes in the two
subphrases could fall either on the analogous syncopations (this results in
a somewhat saucy expression) or on the bottom of the curve before (on E
and A respectively), thereby underpinning the tender mood of the countersubject. The second subphrase, lower than the first, is softer overall.
3
194
WTC I/13
Although CS1 is an almost constant companion to the subject, it undergoes significant modifications on its way through the fugue. Its initial
interval, originally an assertive octave leap, becomes gradually smoother:
reduced first to a seventh (m. 5: C-B), then substituted by a three-note
figure beginning with a fifth (m. 11: G-C-B), it is finally leveled to a
stepwise figure (m. 15: A-G-F; similarly in mm. 20 and 31). Moreover,
CS1 also loses momentum at its end. In mm. 11-13, the last syncopated
figure is replaced by a quarter-note and an appoggiatura that resolves
belatedly after a tie suspension. In mm. 15-17 and 31-33, the second subphrase turns into a cadential-bass pattern, in mm. 20-21, CS1 is shortened
to its first subphrase, and in mm. 28-29, it is entirely omitted.
The second counter-subject is normally expected as a companion to the
third subject statement. However, the line appearing in the upper voice of
mm. 5-7 turns into a rhythmic parallel of the subject. It is thus necessary to
look further, to the ensuing subject entry in mm. 12-13, in order to find the
true second counter-subject. CS2 begins belatedly half a measure after the
subject. It is characterized by a rising scale interspersed with pairs of the
repeated keynote. In keeping with this straight gesture in the ascent, the
development of tension also depicts only a single direction: no subphrasing
interrupts the continuous crescendo to the upper F. This counter-subject is
heard only three times altogether: after its introduction it recurs in U: mm.
20-22 and in U: mm. 28-30.
Another line that accompanies the subject twice and might, at a first
glance, appear as a counter-subject occurs in U: mm. 15-17 and in M: mm.
31-33. Upon closer inspection one detects that its end is composed as a
complement to the CS1 turned cadential bass, is actually making up, in
free style, for the omitted syncopated figure at the end of the first countersubject. This note group should therefore not be regarded as a primary
component in its own right. The following sketch shows the phrase structure
and dynamic design in the thematic material of this fugue.
F major
195
196
WTC I/13
Interestingly, Mcs2 actually appears earlier than the counter-subject from which it stems.
L: m. 28 belongs to this episode in view of its material, although there is an overlap with
the beginning of the next subject entry.
F major
197
This type serves as a bridge. Another type, represented by E1, E3, and E6,
gains limited independence by using a distinct episode motif. In both E1
and E6, a cadential close is present, whereas this more definite ending is
missing at the tail of E3.6 This episode type provides a color contrast to the
primary material and conveys the impression of structural closure. The
third type is represented exclusively by the five measures of E4b. Utilizing
only segments from subject and CS2 and carrying these through various
harmonic steps, this episode seems purposefully to postpone the ensuing
subject statement.7 The message of this episode type is therefore neither
bridging nor closing but delaying.
The material of this fugue does not express a single character. In the
subject, the first counter-subject, the episode figures derived from these,
and the independent episode motif M1, the complex rhythmic pattern and
the overall stepwise motion clearly indicate a rather calm basic character.
By contrast, the second counter-subject and the episode figure derived
from it (Mcs2) feature a simple rhythmic structure and a pitch pattern consisting exclusively of skips, thus suggesting a rather lively basic character.
The conclusion to be drawn from these observations should accommodate
both facts. The tempo is moderately flowing; swift enough to convey
liveliness in the 16th-notes of CS2, but moderate enough to convey calm in
those of CS1. The articulation should take into account the differing
characters represented by the material. In the subject and the motifs
derived from it as well as in CS1 and M1, all notes are legato, whereas in
CS2 and Mcs2, all notes are quasi legato.8 Non legato articulation is
appropriate in the cadential-bass notes of mm. 11, 16, 22, 33, and 34 as
well as in the octave leap of m. 25.
The relative tempo of the prelude to the fugue is ideally represented by
the compound relationship of 3:2. This translates as follows: three dotted
eighth-notes in the prelude equal two quarter-notes (half a measure) in the
fugue. Approximate metronome settings: prelude beats (dotted eighthnotes) = 96, fugue beats = 63.)
6
It may not seem too farfetched to speculate that this missing close of E3 is made up for by
E4a which, as was shown, presents an explicit cadential formula.
7
It reminds one faintly of that episode portion in the C-major fugue that serves as something like the development section in a sonata form fugue. The reader may wish to
refer back to the chapter on WTC I/3.
8
For the sound balance within these components it is of greatest importance that the
frequent note repetitions do not sound with a greater degree of detachment than the other
intervals; any effect of slurring should be carefully avoided.
198
WTC I/13
F major
199
return to the home key, the third episode modulates from C major to the
tonic relative minor that, after a subject entry in D minor in mm. 20-22, is
confirmed in the cadential close of m. 22-23. E4b undertakes the return to
the home key before the next subject entry. Thus all but one of the subject
statements appear in F major, either on its tonic or on its dominant.
The following conclusions may be drawn from what has been observed:
The first section encompasses three subject statements and one episode,
with the cadential formula at the end of E1 concluding this section. The
second episode comprises two statements that appear strung together by
the bridging E2, followed by the episode type that, elsewhere in this fugue,
concludes a segment but here lacks the final cadential close. The ensuing
statement appears thus almost as redundant, an impression enhanced by the
fact that it is the only one in the minor mode. The cadential measure of E4a
concludes this second section in m. 231. The third section begins with an
opening (delaying) episode followed by the passage that was recognized
as structurally analogous to the second section without its quasi redundant
final entry. In this section, the third subject statement is now completely
stripped off.
Within the first section, the tension grows from the first to the second
subject entry along with both the growth of the ensemble and the emergence
of the contrasting first counter-subject. However, from here to the third
200
WTC I/13
entry, there is almost no increase as the upper voice does not contribute
independent material but runs a rhythmic parallel to the subject. The
ensuing episode begins with a distinct color contrast. It then allows for a
slight increase of tension as M1 comes in and is sequenced upward, only to
give way to the relaxing cadential close that concludes both the episode
and the first section.
The beginning of the second section can be regarded as the climax not
only of this section but of the entire fugue because the juxtaposition of the
subject with its two counter-subjects creates the greatest degree of polyphonic contrast. The bridging episode sounds softer than the surrounding
subject statements but not necessarily as different in color as did E1. It is
followed by an entry that makes no use of the newly introduced CS2 and
thus sounds less assertive than the preceding one. The ensuing episode
begins again with a color contrast and a slight increase of tension in connection with M1. Its relaxation is interrupted by the quasi redundant entry
that is the most subdued both in this section and in the fugue.
The third section sets off in an almost floating atmosphere. The remaining two subject entries and two episodes repeat the development of
the second section, omitting, however, both the climax and the anticlimax.
The first subject statement of this section is accompanied by CS2 but not
CS1 and thus sounds less polyphonically exciting, and the quasi redundant
final entry is dropped entirely.
202
WTC I/14
C1
C2
Whether the A on the downbeat of m. 2 or the third 16th-note F constitute the intended
melodic ending is almost impossible to decide.
F minor
203
1
S
C
2
3
C1
S
4
5
C2
S
mm.
U
L
12 13 14 15 16
S
C2 C3 S inv
C S
S inv C3
6
7
S
C2
C2 S
9
10
S
C1
11
12
17
19 20 21
C3 S
S
S par
22
S
C
23 24
C2
S
18
The few subject-free passages in this fugue are episodes in the usual
sense. Some take their material directly from the primary features; see,
e.g., the episodes in mm. 3-4 and 5-6, which display a motif (in imitation)
composed of the subjects turn figure and a leap that recalls C1. The lower
voice in mm. 8-9, the upper voice in mm. 10-11, and the lower voice in
mm. 16-17 use the same motif, while mm. 11-12 and 17-19 constitute
2
See in m. 19 the upper notes of the right-hand part. The immediate relationship between
these chords and the chords in mm. 14-15 becomes apparent when one compares the lower
notes here with the uppermost pitches there.
204
WTC I/14
F minor
205
The subjects harmonic layout is simple in its large steps but fairly
complex in the small ones. This is caused both by the artificial leading
notes in the subject and by the high degree of chromaticism throughout the
fugue. The main steps of the simple progression are represented by the
three long notes, with A (m. 1) for the tonic, B (m. 2) for the subdominant,
both C and the ornamented G (m. 3) for the dominant and the final F
(m. 4) for the return to the tonic. The more complex harmonic progression,
which Bach uses, e.g., in mm. 15-18, is shown in the example below. The
Roman numerals reveal how Bach in his harmonization of the subject
actually passes through the keys of B minor and C majorthe subdominant and dominant keys of F minor:
206
WTC I/14
The two notes in a true sigh are usually neighbor notes; they always represent different
harmonic steps. Leaps, especially if they occur within a single harmony, are therefore not
sighs and should not be played as pairs. Notes that appear in the context of sigh figures
but do not belong are: the octave leap in m. 15 where the lower F is the end of one subphrase, the higher F the beginning of the actual CS1; similarly in mm. 19, 22, 29, and 32.
F minor
207
208
WTC I/14
The fourth episode stands out as different from the others because it
emerges from the preceding subject statement without any clear beginning.
The soprano in mm. 23-24 continues the descent in syncopations begun
toward the end of the subject entry (see mm. 22-23 F E, continued in mm.
23-25: D-C-B). At the same time, the bass presents sequences of the
variation of the CS1 beginning that accompanied the end of the inverted
subject (mm. 22-23: D A . 23-24: B-F and, shorter, . 24-25: G-E).
Similarly, the alto also extends the end of the subject statement in varied
sequences (compare m. 23: G-A-B-C with the ensuing groups F-B and
E-F-G-A). This entire episode can thus best be regarded as an extension
of the preceding subject statement.
The structure of the episodes and their interrelationship is interesting.
E1 recurs faithfully (except for the octave displacement) in E5 and thus
creates a distinct symmetry. E1 also reappears, slightly varied and with an
accompanying voice, as the first segment of E2 (see T + B: mm. 11-12) and
of E3 (see S + B: mm. 18-19). The second segment of E3 features a parallel
presentation of M2 in the outer voices (soprano and bass) followed by
imitation, also in parallel (in what seems like the inner voices but is actually
the alto and the bass after an octave displacement, since the tenor is
resting). This pattern is taken up in E6: see m. 35 with a sequence in m. 36.
This creates a second structural analogy.
Finally, there are two cadential formulas apart from the one in the final
measure. One appears in m. 20 where the key of C minor is confirmed in
a perfect cadence. The other formula, with a dangerously similar-looking
bass line, marks the end of the above-mentioned subject extension (see
m. 25). This one, however, is not only an imperfect cadence, but it also
overlaps with the beginning of a new subject statement in the soprano. It
thus does not qualify as a structural caesura.
The features described above also determine the role played by each
episode in the dynamic development of the composition. E1 and E5 both
begin slightly softer than the end of the preceding subject entry, after
which they prepare the subsequent entry in a twofold crescendo. E2 sets
out with a similar buildup. This is followed by a gradual release generated
by the overall descending direction of the M3 appearances. A complete
relaxation is held at bay by the frequent quotations of the outgoing M1 in
the bass of mm. 13-14. E3 begins with the same active gesture followed by
a gradual diminuendo in m. 19 and a complete relaxation in the perfect
cadence. E6 picks up the gradual release. It is a bridge to the subsequent
entry. Finally, E4 continues the diminuendo in the subject ending and thus,
at a moment of particularly low tension, creates an anticlimax.
F minor
209
Both the fugues pitch pattern with its prevalence of stepwise motion
and the rhythmic organization with its variety of note values are unequivocal in determining the basic character of this fugue as rather calm. The
frame for the tempo is set on the one hand by the required tranquility of
character, and on the other hand by the rather long note values, which still
need to be perceived as alive. In other words, the eighth-notes must be
slow enough to convey serenity, but not so slow as to impede the listener
from taking the entire subject on one breath, and thus cause the subject to
fall into pieces. The articulation is generally legato. Only the first half of
the first counter-subject as well as the episode motif derived from it feature
slurring in pairs, following the pattern of the sigh figures. The only notes
to be played non legato are the consecutive leaps in the second half of the
counter-subject (e.g., m. 6: D-G-C) and the cadential-bass patterns in
mm. 20, 25, and 39-40. For the relative tempo of the prelude to the fugue,
a good and feasible solution is: one measure in the prelude corresponds
with half a measure in the fugue. (Approximate metronome settings:
prelude beats = 112, fugue beats = 84.)
The only ornament in this composition is the trill in the subject. As it
is complemented by a resolution, it is note-filling. Approached in stepwise
motion it begins on the main note sustained for the duration of an eighthnote and then shakes in 16th-notes, ending with a suffix. This trill, being
an integral part of the subject, must be transferred to the D in m. 6 and to
the Gs in mm. 10 (where it is indicated in brackets), 17, and 31, i.e., to the
second-to-last notes of all original-shape entries,4 as well as to the B in m.
23 and the E in m. 34, i.e., to the second-to-last note of inverted entries.5
The entering order of the voices and their respective surrounding
texture, in conjunction with the explicit cadence at the end of E3, define
the fugues binary structure: The full ensemble of four voices is reached in
the fourth statement. In the middle of m. 20, a cadential formula closes this
section in C minor, the minor dominant of the home key. The subsequent
entries manifest a protracted suspension before the full ensemble is reached
once again:
4
In the subjects final statement, the Picardy third of the ending impedes execution of the
trill. The ordinary trill on G would shake with A as its upper note. This A, however,
clashes with the A in the middle voice. A trill with A is obviously also out of the question
because of the preceding A. This subject ending must therefore remain unornamented.
5
Students sometimes ask whether trills are also upside down in inverted statements. The
answer is: no, ornaments are not inverted.
210
WTC I/14
212
WTC I/15
the second in a pair of measures in hidden two-part structure and represents a harmonic resolution, while m. 9 is the harmonically active first in
a pair.
The basic character is easy to determine as rather lively: This is
unambiguously suggested both by the simple rhythmic pattern with only
two note values and by the frequent broken chords (in the triplets) and
leaps (in the eighth-notes). Like the fugue it precedes, the prelude is downright virtuoso and should be played accordingly, in a very brisk pace. It
does not feature any ornaments. With regard to articulation one should
differentiate between the coarse level of simple contrastnon legato for
the eighth-notes and legato for the 16th-notesand the level of subtler
shades. These include: non legato in a light, neutral touch for the accompanying eighth-note leaps, a more expressive non legato for eighth-notes in
hidden two-part structure (where they move predominantly stepwise), quasi
legato with a crisp touch for the 16th-notes in the broken chords, legato in
a moderately expressive touch for the 16th-notes in hidden two-part
structure, legato with some melodic expression in the 16th-notes forming
motivic figures, and legato with melodic intensity for the eighth-notes in
U: mm. 11-13 representing appoggiatura-resolution pairs.
The G-major prelude can be regarded as consisting of two halves. Each
embodies a shorter introduction followed by a longer portion that develops
and complements the material established before. The first half spans mm.
1-111. It is based on a harmonic idea presented in the right hand with
virtuoso broken-chord patterns, accompanied by a rhythmicized pedal note
with octave displacement. In the first simple cadence (mm. 1-2), the
descending pitch direction of the broken-chord pattern is so dominant that
it overruns all considerations for shaping along harmonic lines. In character with this virtuoso figure, the prelude thus begins with a rather energetic
tone color followed by a diminuendo through two measures. The third
measure is composed as a harmonic sequence; it takes up the V7-I pair
from m. 2 and relocates it to D major. The equivalent to this feature in
performance is a sequence also in the dynamic process, i.e., a repetition of
the second half of the previous diminuendo. The question whether the
dynamic level in m. 3 is generally higher or lower than that in m. 2 can be
solved with the help of three observations: the pitch level is slightly lower
in the second half of m. 3 than in the second half of m. 2, the second half
of m. 3 features rests in the right hand part, the thinner texture adding to
the effect of decreasing intensity, and m. 3 was recognized as the end of
the first short section. A definite relaxation serves to underscore this fact in
performance.
G major
213
f -mf mp -p mp p mp mp
pf
mf f
mp mp + mp
The second half of the piece, covering the remaining nine measures, is
based on motivic treble figures that introduce a new expressive quality.
Beginning with the first little motif (see m. 112-3, with three sequences up
to the E-D on m. 131), the melodically conceived 16th-notes followed by
the appoggiatura-resolution pair in the eighth-notes add charm and grace to
the prelude. (Note the dynamic shaping with U: crescendo to the appoggiatura, L: diminuendo from each strong beat onward. Dynamic imitations
would be ill-advised here.) The second motif (mm. 132-142) presents a
development of the first. It is twice as long, more elaborate in its triplet
figures, and the appoggiatura is now indirect (D-E embedded in D-E-F-E).
This motif also spawns sequences (mm. 142-152 and 152-162.) The accompaniment, which had contained old material in the first motif, combining
the broken-chord figure and the octave leap from mm. 1-4, is now much
more intricate as it sets out with the varied broken-chord figure from m. 6
and continues it freely, albeit in virtuoso fashion.
214
WTC I/15
The section ends with a figure of only one quarter-note length (see U:
m. 163), complemented in the left hand by a unit that, because of its
rhythm, has to be regarded as overlapping (m. 163-4 C-A-F-D etc.). Both
figures are sequenced many times, the treble right through to the end,
whereas the bass figure gives way to a cadential-bass pattern in m. 18.
The development of tension in this second half is determined entirely
by the direction of the sequences. Thus the first motif, which begins in
something like mf, is followed in its descending sequences by a diminuendo.
The second motif, which sets out with more vigor anyway, undergoes a
further crescendo in its ascending development. As if to compensate this
outburst, the third little figure, launched from a climax of less melodic
intensity than its predecessors, describes a continuous relaxation up to the
end of the prelude.
G major
215
a varied partial sequence of the first subphrase: the eighth-note descent and
the syncopated seventh leap recur, but they are linked by a smaller interval
than before. The third subphrase [c] then consists of the downbeat
eighth-note in m. 4 and the twofold scalar descent (with or without the note
in m. 51). If one follows the third of the above-mentioned concepts and
assumes the closure of the subject already on the downbeat of m. 4, there
would only be two subphrases, [a] and [b], with [b] complemented by a
resolution.
The pitch pattern consists primarily of stepwise motion, with only three
leaps: two in m. 2, one in m. 3. Two of the leaps appear as consecutive
intervals (m. 2: G-D-C). Among the seconds, the longer note values, i.e.,
the eighth-notes in mm. 2-3, have melodic value while the shorter are
either ornamental, as is the case in the two turn figures at the beginning, or
constitute scalar passages, as happens in the two five-note groups at the
end of the subject.
In terms of its rhythm, the subject features three values: 16th-notes,
eighth-notes and (syncopated) quarter-notes. The same three note values
also constitute the predominant rhythmic pattern throughout the entire
composition. A particular attribute of the rhythmic pattern in this subject is
that it displays perfect symmetry in mm. 1-4. Were one to give a name tag
to the rhythmic patterns of each measure, the result would read x-y-y-x.
When analyzing the subjects harmonic background it seems worth
anticipating an important trait of this fugue, i.e., that the subject frequently
appears in inversion. The harmonic outline of this inversion is considerably
different from that underlying the original shape. The following examples
demonstrate this with harmonizations taken from mm. 11-15 and 20-24.
half cadence
perfect cadence
In our search for the intended climax we may thus not be able to rely
on harmonic features. However, both the rhythmic pattern and the pitch
outline provide congruous guidelines. Concerning the rhythm, the two
216
WTC I/15
G major
217
218
WTC I/15
G major
219
material we can distinguish three episode types: E1, E4, E6, and E9 are all
built on M1 and M2, while E3, E5, E7, and E8 neglect M2 in favor of the
other two motifs. E2 is related to all of the others. It sets out with M1 in
non-motivic surrounding; then combines an M1 inversion with an M2
parallel, and finally presents sequences of the M3/M1 pair, complemented
by a neutral middle voice.
The relationship between the episodes is evident from the table above.
E1 is related to E4 and also, though less closely, to E6. E5 and E7 are
similar, and E8 shows analogy to the second segment of E3. E2 can be
divided to reveal relationships to E1 and E4 in its first two measures, to E5
and E7 in its remainder. None of the episodes exhausts its function as a
cadential close. In one instance, however, a closing formula determines the
material of a final segment: see the do-si-do figure in the upper voice as
well as the typical bass pattern in the first half of m. 69.
The role each of the episodes plays in the development of the composition also stems directly from its material. Those of the first type, based
only on M1/M2, act as bridges, while those ending with descending
M3/M1 sequences have concluding character. The two episodes featuring
the M1 + scale combinations lie in-between; they provide more contrast to
the primary material than the first type but lack any features of conclusive
force. One might thus say that they insinuate a gap that they nevertheless
span.
The basic character of this fugue is best interpreted as rather lively.
This decision is supported mainly by the pitch pattern with its leaps and
ornamental figures. The rhythmic pattern contains four melodically relevant note values, three already in the subject, and additional 32nd-notes in
the variation of M1, but all these values fit smoothly into a generally simple
rhythmic structure. The tempo is swift, playful if not outright virtuoso,
allowing for an energetic and non-superficial touch. The articulation
demands non legato for the eighth-notes and quarter-notes and different
kinds of legato for the 16th- and the few 32nd-notes. Among these, truly
melodic quality seems appropriate for M3 while an equally dense sound,
though for different reasons, suits the hidden two-part structure of M1.
Ornamental 16th-notes like the turn figures and scale portions in the
subject and the longer scales in E3 and E8 are best rendered in quasi
legato. An ideal proportion of tempo between the prelude and the fugue is
founded on the larger pulses of each: one quarter-note in the prelude corresponds with half a measure in the fugue. Approximate metronome settings:
80 for the beats (quarter-notes) in the prelude and the compound beats
(dotted quarter-notes) in the fugue.
220
WTC I/15
m. 64
mm. 69-70
G major
221
222
WTC I/15
In each section, the tension rises from the first to the last entry, due
either to an increasing number of participating voices (I, III, and IV), the
enhanced presentation of the subject (V: from single entry to stretto), or
the analogy with the first section (II). Section II is dynamically more
intense than the first since its three entries are launched in full ensemble
and follow one another without interruption. Section III falls back because
of both its reduced range and its minor mode, appearing almost as a softer
appendix. Section IV develops to an even higher dynamic level. Finally,
section V inverts the situation in the pair to which it is analogous (II/III . V)
by setting out from the softer level of the single entry and evolving in a
powerful increase toward the stretto with parallel tail.
224
WTC I/16
Furthermore there are instances where two voices complement each other
to form one of the motivic figures, thus creating extra melodic intensity. In
view of this complexity, the preludes basic character should be interpreted
as rather calm, and played in a tempo restrained enough to allow for full
melodious substance in the 32nd-notes. The corresponding articulation
requires that all notes except cadential-bass patterns and consecutive leaps
be played legato. Cadential-bass patterns can be found in mm. 10-11, 1718, and 18-19. Consecutive leaps that need to be detached occur in the
lowest voice in mm. 5-6, whereas the seeming leaps in m. 14 form part of
a hidden two-part structure. They require legato between the two layers so
that each of the two voices sounds disconnected.
A tricky question arises in the area of articulation with regard to the
figure introduced in the upper voice of m. 2. The Gs clearly do not belong
to the main melodic line, but they do not form a secondary line of their
own either. Thus they can be interpreted, and played, in two ways. One can
regard them as a pedal background in a hidden two-part structure, in which
case they are not separated from their surroundings by articulation, but only
set apart by a different tone color. Or one can perceive them as charm
notes, the playful escape notes that were to become so common in the
music of the later 18th century. In this case they are treated as integrated
parts of the melodic line that sound just lighter and, because of their
interval gaps at both sides, non legato.
There are five ornaments, all of them marked as trills and all of them
appearing at a moment of harmonic affirmation (i.e., on the tonic chords of
the keys reached by way of modulation in the course of the piece). None of
these ornaments needs, or pretends to aim for, any kind of resolution. This
is most obvious in connection with the trill on the final note, but equally
true in the case of the four other ornaments, which all end in tie prolongation followed by the beginning of a new phrase. Having said this, it
follows that none of the trills concludes with a suffix. As to the beginning,
a launch from the main note is in order in the case of the trills that are
approached stepwise, as in mm. 3, 7, and 19, or begin without any preparation, as in m. 1. And although the context of the trill in m. 11 would allow
a different solution, it should probably match the corresponding ornaments
and also begin on the main note. The speed of the trill motion (twice as fast
as the shorter note values) is in 64th-notes, or four notes to each 16th-note.
In the four trills of one measures duration, the shake comes to a halt on the
last main note before the bar line, i.e., on the final 16th-note that is then
tied over. The duration of the shake in m. 19 is not fixed in any way; it
could be anything from an eighth-note upward.
G minor
225
The first of the three rhythmic figures that determine the piece is introduced in m. 1. Its texture consists of three parts, two of which, the upper
and the lower, are unmoving pitches: a trill in the treble and a repeated
pedal in the lower voice). The rhythmic pattern is even, with three different
values sounding simultaneously (eighth-notes, 16th-notes and ornamental
64th-notes). This figure recurs three times. In m. 3, the key and the order
of the voices are the same, with an octave transposition of the treble. The
eighth-note pedal, however, has been abandoned. A complementary pattern
of the middle and lower voices now provides both the eighth-note and the
16th-note pulses. In mm. 7 and 11, we find the figure transposed and in
inverted voices; the pedal has been redeemed and now appears as an
offbeat eighth-note pulse as part of the upper voice.
The second rhythmic figure defines the playful contour in the upper
voice of m. 2 that has already been mentioned. It takes various accompaniments as it recurs, varied not only with regard to its pitches but also in its
length and in details of its rhythmic design (see mm. 4-6, 8, and 15-16).
The third rhythmic figure first appears in m. 9. Its basic element spans only
a quarter-note. While it starts out in only two melodic guises, later recurrences (see mm. 12-154 and 163-19) show it varied not only in pitch but
also in texture. Thus its only constant is its rhythm.
With regard to the preludes layout one may state: Section I presents
the first two figures and subjects them to a first development. Section II
briefly takes them up in transposition before concentrating on the presentation of the third figure. Section III begins once again with the first figure
but then develops the third, only once interrupted by a reminiscence of the
second figure (mm. 154-16). In the vivid three-part texture of the concluding
segment, which sounds above a newly added tonic pedal, the third figure
reigns supreme.
The dynamic range within this rhythmically determined prelude is not
large. Small-scale dynamics depend exclusively on the overall pitch
direction defining each rhythmic figure. The melodically static first figure
suggests only the slightest increase in tension, while the third figure seems
always composed in superimposed descending pitch lines and thus appears
determined by dynamic relaxation. Only the second rhythmic figure
changes melodic orientation and, with it, dynamic direction, allowing for
diminuendo (as in mm. 2, 5, and 15-16), crescendo (as in mm. 6 and 8) or
a dynamic curve (as in m. 4).
With regard to the overall development of tension, the prelude
describes a soft, largely spaced dynamic curve. The second section implies
226
WTC I/16
more tension than the first but is surpassed by the third, while the coda
returns to the level of the beginning. Two climaxes can be determined,
falling on the downbeat of m. 9 and on the middle beat of m. 12. Both are
motivically connected with the third rhythmic figure but derive their
expressive power above all from their harmonic underpinning.
The G on m. 24 does not belong to the subject. Three reasons support this: Melodically,
this note recurs only in one of the subsequent statements, in m. 6. Harmonically, the tonic
is already reached with B. Metrically, a subject that begins after a strong beat can usually
be expected to conclude on a strong beat.
G minor
227
Although many performers emotionally prefer interrupting rests, the harmonically more
consistent rendition is the one that resolves the tension of the main climax in a single
diminuendo. As the fugue progresses, more evidence arises to support the concept of an
undivided entity.
228
WTC I/16
In the course of the fugue, the subject suffers only minor modifications,
most of them inconsequential. Its answer is tonal, with its first interval enlarged from a semitone to a minor third (e.g., m. 2: G-B). The subjects
beginning is varied in m. 23 where the initial eighth-note is replaced by
two ascending 16th-notes. The final note, being the third scale degree,
appears as a Picardy third in m. 34, a common feature in a minor-mode
fugue. A modification that is more consequential can be observed when, in
the major-mode section from m. 12 onward, the two initial intervals are
changed: the semitone with its particular urging quality is substituted by a
major second and the leap appears as a major sixth that, unlike its minor
brother, is not a high-tension interval. These two changes have considerable influence on the buildup of tension.
In two instances does the subject appear in stretto: mm. 17-18 feature
a combination of bass and alto statements at a distance of four eighthnotes, and mm. 28-29 present a corresponding group of two complete
entries in soprano and tenor, fortified by a third entry at the same distance
in the bass that deviates immediately after the climax.
The fugue features a single counter-subject. CS acts as the subjects
faithful companion throughout the fugue and is only omitted in the final
statement. It begins in a metrical position equivalent to that of the subject,
but half a measure late, after the subjects climax. (This belated beginning
is the reason for the additional G on the fourth beat of mm. 2 and 6, a note
that does not form part of any material but whose function is to support the
beginning of the subject entry when all other voices are resting.) In pitch
pattern and rhythm, the counter-subject is strikingly related to the subject,
particularly to the its answer, which it almost seems to read upside down
and with its halves exchanged.
G minor
229
S
or
CS
M1 has a forerunner that, consisting only of its ascending portion, appears in E2 (see the
soprano in mm. 8 and 9) and in E3 (see the tenor in m. 16-17).
230
WTC I/16
slightly varied, in the upper voice. This imitation should retain the dynamic
curve of the model.4
As can be seen from this overview, none of the episodes serves
exclusively as a cadential close. Only one episode segment fulfills this
purpose. In E2, the home key of the fugue is confirmed (motivically still in
the context of the subject-related figure) with a perfect cadence in G minor
on the middle beat of m. 10. The remaining 1 measures are neither
needed for the closure of the episode, nor do they present any (new or
continued) episode material. Instead this extension functions as a modulation to the related major key.
There is no structural relationship between the episodes of this fugue.
The role played by each episode in the development of the composition is
determined by the direction in which the motif derived from the subject
ending it sequenced. The following details can be observed: In E1 and E3,
the subjects final half-measure is taken up in ascending sequence. Both
episodes thus serve as bridges heightening the tension toward the ensuing
subject statements. E2 begins similarly with an ascending sequence of this
figure but continues thereafter in a generally descending direction (mm. 910: soprano B-A-G, bass E-B-D-G). This descent and the cadential close
it encompasses represent a relaxation. In E4 and E6, the final half-measure
of the subject is not sequenced but imitated. These imitations, though
ascending in pitch direction, appear as a more indirect continuation than
the sequences in the previous examples. Yet while they do not convey the
feeling of anticipation, they succeed in slowing down the decay of tension.
In both cases, however, the episode concludes with an inversion of the
figure and with a plunging final interval that generates a strong decrease.
Thus a gradual diminuendo ends in a steeper drop of tension.
The overall stepwise motion interrupted only for high-tension intervals
suggests that the basic character is rather calm. The rhythmic pattern seems
neutral. On the one hand, three different note values are regularly used
and, in the episodes, supplemented by syncopations; on the other hand, the
rhythmic structure does not seem complex enough to require real tranquility.
The result is a placidly flowing tempo.
4
The frequently heard mistake that renders the descent in the imitation in diminuendo and
the ascents in crescendo actually cuts the motif in two, and each half will automatically cry
for a new partner. The result is usually that middle and upper voice sound as if in a single
track, so that even the original motif in the middle voice is then destroyed. A good way to
avoid this pitfall is to play the leading voice (M) on a slightly higher level of intensity than
the imitating upper voice.
G minor
231
232
WTC I/16
The relationship between the sections is complex. Owing to the modification of the interval structure in the major-mode subject, section II begins
in reduced emotional vigor. Its concluding stretto may surpass the final
entry of section I in loudness but probably does not reach its passionate
quality. Section III, while returning to the melodic intensity of the minor
mode, is dynamically static and thereby falls back even behind section II.
After the additional color contrast in very soft shades brought about by the
material change in E5, the beginning of section IV then presents the
sudden climax. The fugue ends in considerable intensity.
234
WTC I/17
closing formulas, must be legato. Within the quasi legato, various shades
of touch are required in order to bring out the different textures. Passages
in hidden two-part structure with a melodic line moving before a backdrop
of a repeated pedal (as in U: mm. 15-16, L: m. 32, and U + L: mm. 39-40
where weightier and lighter notes alternate regularly) should be distinguished from those with an ornamented pedal (mm. 13-14 and 30-31 where
the structural notes are interspersed with feathery inverted mordents) and
from the remainder of the 16th-note passages in which all notes take part
equally in the melodic line.
The prelude contains two kinds of ornaments, indicated by the same
mordent symbol. The two closing formulas in mm. 17-18 and 34-35 each
feature the typical interrupted (point darrt) trill with anticipation of the
resolution note. Both begin on the upper neighbor note and shake in
32nd-notes. The other four mordent signs, appearing on the quarter-notes
in mm. 36 and 38 and on the 16th-notes in mm. 41 and 42 respectively, can
be played as just that: three-note mordents that begin on the main note
either because they are approached stepwise, as in mm. 36 and 38, or
because they embellish the initial note of a phrase, as in mm. 41 and 42.
or
mm. 17-18
mm. 34-36
A major
235
236
WTC I/17
A major
237
Only once, in an episode where the combination of interval and rhythm pattern suggests
false entries, does a brief parallel occur; see mm. 214-221 alto/soprano.
2
Introduced in m. 22-4 (T), M1 recurs, with the opening note displaced to form an ascending
scale, in m. 4 (B). In rhythmic variation with intersected syncopation it is also found in m.
27 (A, imitated in S). The inversion, M1a, appears in mm. 24-32 (T), 44-52 (B), 153-161 (A),
223-231 (A), and 251-3 (T, with two sequences). M1b, an extended version combining the
ascending scale with the M1 inversion, is found in mm. 53-62 (B), 72-81 (A), 82-91 (B),
92-101 (B), 232-241 (T), and 242-251 (T). An even more extended version materializes in
mm. 291-302 (T). M1c, the (free) inversion of the previous motif, occurs twice, in mm.
16-17 (S) and 18 (B). As a further development from the same root, the scalar ascent is
complemented, after a syncopation, by a new tail in eighth-notes: for M1d see mm. 7-8 (B),
8-9 (S), 9-10 (A; here it gives way to a melodic closing formula), and in mm. 16-17 (B).
238
WTC I/17
M1a
M1b
M1c
M1d
M1e
M2
or
M3
Introduced as a two-fold partial sequence of the subject, this motif is first heard in B: mm.
3-4 above an E pedal (see E-B-G-A, E-A-F-G). It recurs as a separate motif, with
occasional ornamental variation, in mm. 11-13 (B), mm. 14-15 (A) and mm. 19-21 (S).
Remote relatives can be heard in mm. 25-26 and 31-32. A final remote relative combines
the scalar ascent with a subsequent descent. This figure appears several times, with a
varying number of free sequences; see mm. 11-13 (S), 14-15 (B), 19-20 (A) and 31-32 (B).
4
First presented in the context of M2, this motif could, in fact, be read as a parallel to the M2
peak notes. The consecutive syncopations recur in mm. 11-13 (T), 14-16 (S) and 19-21 (T).
A major
239
descending sequences (as in E3, E4, E5, and E6), or in hidden descents (as
in E2 where the soprano features the falling A-major scale).
With the exception of the syncopated half-notes in M3, the rhythmic
pattern in this fugue is simple. Moreover, the pitch pattern is made up of
broken chords alternating with ornamental figures. It is thus easy to
determine the basic character of this fugue as rather lively (or even very
lively). The tempo is confined by the character of its primary material: the
broken-chord leaps in the subject should sound vigorous and thus not too
slow, while the various 16th-notes figures should by no means appear as
mere virtuoso patterns, particularly in view of the manifold motifs Bach
invented and developed, and thus not too fast. The corresponding articulation consists of bouncing non legato in the subject and M2, a fairly
intense quasi legato in all motifs deriving from M1, and a melodious non
legato in M3. The only longer note values that must be played in strict
legato are the cadential formulas in S: mm. 15-16, 22-23, and 34-35, and in
A: mm. 9-10. The fugue does not feature any ornaments.
The relative tempo of the swifter and lighter prelude in triple time to
the more vigorous fugue in quadruple time is best established by equaling
metric, and not rhythmic, values: an ideal proportion is one in which one
measure in the prelude equals half a measure in the fugue. (Approximate
metronome settings: prelude beats = 120, fugue beats = 80.)
The most obvious indicators that help determine the design of this
fugue are the cadential formulas in mm. 9-10, 15-16, and 22-23. As the
subject entries preceding the first of these closures constitute a perfect
round of all four parts (T B S A), the first section seems thus ascertained.
Between the cadential closes in mm. 9-10 and 22-23, the correspondence
of E3 with the first segment of E4 and E5 creates a larger unit from m. 11
to m. 23. The reduction of the ensemble from four to merely two voices in
mm. 23-24 further enhances the structural importance of this harmonic
closure. Inside this superimposed frame of twelve measures, the closing
formula in the middle of E4 generates a smaller caesura, confirmed in its
structural value by the fact that the ensuing subject statement sounds in
reduced ensemble. As no explicit cadential close occurs between m. 23 and
the end of the fugue, one has to look for other indicators of structural
layout. One hint can be found in the entering order of the voices. The five
final statements (B T A S S) seem to form a group: they follow one another
not only very closely and in a logical arrangement of gradual ascent, but
also lead, after the tension-increasing E7, to the conclusion of the fugue in
a redundant soprano statement. The return to the home key with the bass
entry in mm. 27-28 further ascertains the confines of the final section.
240
WTC I/17
A major
241
242
WTC I/17
Although mollified transitorily during the episode, this urge continues even more powerfully in the four consecutive statements that
mark the beginning of the fifth section. The ensuing episode E7,
interestingly enough, is the only tension-sustaining one apart from
the very first episode. After its deceptive cadence (m. 332-3), the
final statement sounds in an almost homophonic setting, crowning
the fugue with a triumphant closure.
244
WTC I/18
I
II
III
IV
or:
I mm. 1-91 (1-51, 5-91)
II mm. 9-171 (9-131, 13-171)
III mm. 17-29
G minor
245
presents the same figure (L: mm. 22-23) along with the inversion of the
original 16th-note group (M: m. 25, U: m. 26) and a version of the group
in its first shape but with a broken triad at the end (L: mm. 19-21). In
addition to all these transformations of the motifs head, this episode also
recalls the motifs tail (mm. 19-21, in imitation between U and M).
In order to determine the dynamic processes in this prelude, one may
wish to distinguish a varying intensity among the statements of the motif.
Intensity is created here not only, as is generally true for all polyphonic
compositions, by the number of voices surrounding the statement or the
mode in which it is set, but also by the nature of the accompanying material.
In this prelude, Bach seems to use parallels as a characteristic means. The
following hierarchy can be deduced from this observation:
The lowest level of intensity is expressed in statements quoting
the motif without any doubling (mm. 9 and 10, 15 and 17, 27
and 28).
Slightly more emphasis is created by the parallel of the motifs
tail (mm. 1, 2, and 7); less, where the doubling is set in contrary
motion (as in mm. 8 and 16).
The fortification of the 16th-notes alone generates even greater
density (mm. 5, 6, 14 and, somewhat more indirectly, m. 13).
This last level of intensity is exceeded only once where both
halves of the motif are doubled. In m. 18, the 16th-notes appear
in parallel and the eighth-notes are matched in contrary motion.
Relating these observations to the structural features of the prelude one
detects that
the endings of all sections appear in lessened intensity;
in the first and second sections, the climax occurs immediately
after the diminuendo resulting from the episodes descending
sequences and the mid-section cadence;
the first section begins in moderate intensity, with two statements, while the second section, which introduces the motifs
inversion, sets out with only one statement of low intensity;
the third section contains only one complete statement of the
motif that, appearing at the beginning of the section, exhibits
the highest level of intensity in the entire piece;
the coda returns to a very relaxed state.
Here is a graphic representation of the invention, in the second of the
two readings offered above, highlighting the appearance of structurally
analogous sections.
246
WTC I/18
G minor
247
to the artificial leading-note of the fifth scale degree. The resolution of this
leading-note marks the end of the first subphrase. The ensuing major-sixth
leap is thus not a melodic interval but rather a split between the two subphrases. The second subphrase comprises whole-tone steps and a perfect
fifth. Conspicuous note repetitions on the fourth and fifths scale degrees of
D minor enhance the impression of a cadential-bass pattern. Having
observed this one understands why these notes seem to convey so little
melodic message. They are harmonic notes (as representatives of the chords
they imply), and their significance is vertical rather than horizontal.
The subjects harmonic background is most intriguing in a segment
where it might be least expected: the eighth-note-descent B-A-G. The
melodic return to the keynote at m. 21 is not matched by a similar return in
the harmonic progression. On the contrary, after an initial alternation
between tonic and dominant (G minor and D major, with or without their
seventh) in m. 1, this downbeat represents the harmonically most active
step in this phrase, i.e., that to the secondary dominant triggering the
modulation. The melodic resolution of the artificial leading-note C thus
coincides with the harmonic resolution of the diminished chord into the
new tonic. The freshly established key is then confirmed in a cadential
progression.
The subjects main climax occurs in the first subphrase. The melodic
and harmonic developments both favor the first beat of the second
measure, where G-C constitutes a high-tension interval while at the same
time representing the pivot chord of the modulation. The end of the first
subphrase on the measures third eighth-note provides the resolution to
both the harmonic tension and the melodic leading-note, thus generating a
steep decrease after the gradual increase during the first measure. By
contrast, the dynamic outline of the second subphrase is very gentle. The
first note (F) serves as an upbeat to the subdominant representative G,
which is followed by an even relaxation. Although the subjects two subphrases could thus hardly be more contrasting, they nonetheless complement one another symmetrically with regard to tension: the gradual rise
through most of the first subphrase is answered by a gradual decay through
most of the second, the sudden release after the expressive first climax
finds its counterpart in the concise upbeat to the secondary climax.
248
WTC I/18
The subject appears altogether twelve times.
1.
mm. 1-3 T
7. mm. 17-19
2.
mm. 3-5 A
8. mm. 19-21
3.
mm. 5-7 S
9. mm. 24-26
4.
mm. 7-9 B
10. mm. 26-28
5.
mm. 11-13 T
11. mm. 32-34
6.
mm. 15-17 B
12. mm. 37-39
T
A
S
B
T
S
Apart from the alteration of the first interval in the answer, no modifications of shape or length occur in the subject throughout the fugue.
Inversions, parallel statements, or strettos are not used.
Bach invented two companions to the subject that appear repeatedly.
However, just as the subject itself displays a somewhat unusual melodic
line in its second subphrase, the counter-subjects also do not quite conform
to ordinary expectations of polyphonic counter-parts. CS1 is introduced
against the second subject statement in mm. 3-5. It begins with an upbeat
(which is later occasionally dropped or varied). The following long note
and particularly the ascending groups on the sixth and seventh eighth-notes
of the measure sound so much like a parallel to the beginning of the
subject that they may hardly pass as contrapuntal. Only the middle segment with its sequences of the ascending figure leads a polyphonically
independent life (T: m. 3-4 F to B), while what follows thereafter qualifies
as a very traditional closing formula (G-A-F-G). Notwithstanding this
limited independence, CS1 accompanies almost all subject statements, with
exceptions occurring only in the entries in mm. 24-26 and 26-28, where
CS1 is omitted altogether, as well as in the final entry, where its beginning
and end are considerably varied. CS2 is also first presented in its expected
place against the third subject statement (T: mm. 5-7). Its characteristic
features are the syncopated fourth leap and the descent in quarter-notes
concluding in a tie. The beginning as stated in m. 5 with upbeat eighthnote and quarter-note (this value again in rhythmic parallel to the subjects
initial note) is later dropped or varied, just like that of the first countersubject. The suspension created by the tie at the end of CS2 redefines the
harmonic surroundings of the subjects final note as still awaiting resolution. Statements accompanied by this counter-subject are thus unlikely at
section endings. CS2 recurs four times, in mm. 11-13 (A), 15-17 (S), 19-21
(B), and 32-34 (S).
G minor
249
250
WTC I/18
the first half of E6, while E3 is taken up in E4a. The role played by each
episode in the development of the composition also stems immediately
from the material employed. The episodes that are based on the formula
give a concluding impression due to the salient cadential pattern, while
those displaying the imitative motif are either accompanied by a similar
cadential-bass line (as in mm. 28-30) or followed by an explicit cadential
close (see the perfect cadence in A major in m. 24).
Triggered by the two-faced subject, the character of the entire fugue is
ambivalent. It changes constantly between calm, melodious lines of high
expression and cadential patterns of almost neutral tone quality. As a
result, a definition of the basic character must be attempted separately for
the two subject segments and the material resulting from them. In the first
subphrase, the pitch pattern with its poignant leading-notes and tritone
interval as well as the corresponding rhythmic variety suggests a rather
calm basic character. In the second subphrase, by contrast, all melodic
expression seems aborted in favor of a non-committal formula.
The tempo is moderate in order to accommodate the leading-notes
(F-G) and high-tension intervals (G-C) appearing in eighth-note (m. 2)
or even 16th-note rhythm (mm. 21-23, 28-29). The articulation should
reflect the ambiguous character depicted above. In the subjects first
subphrase, the appropriate articulation demands legato throughout. In the
second subphrase, however, non legato articulation is needed to convey the
cadential character. Both the pitch pattern with its note repetitions and
perfect-fifth interval and the regular rhythmic pattern support this interpretation. Correspondingly, the remaining material also divides into two
fields. All cadential formulas maintain the detached style, while the motif
and the counter-subjects tend toward a melodious character. CS2 is
basically legato; only the fourth leap may be detached. The episode motif
with its complex rhythm and semitone as well as high-tension intervals is
entirely legato. Only CS1 shares the ambiguity of the subject: its first
segment is legato while in the second the non legato intention is enhanced
by the written rests.
The most straightforward tempo proportion between the prelude and
the fugue is probably also the best. It is achieved by equating half a
measure in the prelude with a quarter-note in the fugue. (Approximate
metronome settings: 60 for the compound beats in the prelude and the
quarter-notes in the fugue.) No ornaments need to be considered in this
fugue.
When trying to determine the structure of the G-minor fugue, only the
first section is unambiguous as it is distinguished quite clearly by the
G minor
251
entering order of the voices. Four subject statements, presented in uninterrupted succession by the four parts involved in this fugue, are followed by
the concluding first episode. The next statement appears in reduced ensemble and thus confirms the beginning of a new section at m. 111. In the
absence of any assistance from explicit cadential formulas and structural
analogies in the further course of the fugue, the reduced number of voices
involved in a subject statement as an indicator for the beginning of a
section must be complemented by observations concerning the material.
Reductions of the ensemble appear two more times. In mm. 19-21, the
soprano is resting during the entry of the alto, and in mm. 24-26, the bass
is temporarily suspended during much of the soprano statement, although
the first two beats of the entry sound in complete four-part texture. The
recurrence of the tenor statement in m. 17 distinguishes this subject entry
as a redundant one, thus signifying the imminent closure of a section.
As the final statement of the first section (mm. 7-9) and the final
statement of the fugue (mm. 37-39) both appear polyphonically less
intense than the statements preceding them, it may make sense to look into
the appearance of counter-subjects throughout the fugue. If one assumes,
as a hypothesis, that Bach may have composed lessening density of
contrapuntal material towards the end of each section, the following
groups can be established:
mm.
S
CS1
CS2
1
T
3
A
*
5
S
*
*
7
B
*
11
T
*
*
15
B
*
*
17 19
T A
*
*
24
S
26
B
32 37
T
S
* (*)
*
The harmonic outline of this fugue describes a very simple curve. The
minor dominant serving as the secondary key is reached in m. 11. However, as the subject itself in its original version also modulates to this key,
this D-minor cadence does not seem to establish a truly new tonal center.
Moreover, the ensuing statements still remain very closely linked to the
original tonic: already the tenor entry in mm. 11-13 modulates back to G
minor, the bass entry follows with a subdominant-tonic version, and the
redundant tenor statement closes once again on the minor dominant. Only
the third section leads into new harmonic fields. Its first episode modulates
from G minor to A major, the bass statement begins in F major (the
dominant relative) and concludes in B major (the tonic relative), and the
second episode (E4) modulates back to the tonic G minor.
252
WTC I/18
This fugue seems singularly static with regard to tension. With the
exception of the three initial subject statements, all increases in texture in
the course of a section are annihilated by a simultaneous decrease in
polyphonic intensity. The dynamics of this fugue thus live entirely from
the contrastparticularly in the third sectionbetween subject-dominated
passages and episodes.
254
WTC I/19
The ideal tempo is moderate with gently paced quarter-notes and rather
swift 16th-notes. The articulation is legato for the 16th-notes and non
legato for the eighth-notes and quarter-notes. Among the quarter-notes,
those with distinct melodic quality (like the four chromatic steps in L: mm.
1-2) are only softly detached, while notes of cadential character (see, e.g.,
L: mm. 2-3, where notes proceed in sequencing descending fifths) sound
more clearly separated. Among the eighth-notes are several appoggiaturas
(U: mm. 7-8 and 15-17), which must be perfectly linked to their resolutions.
The only other longer notes that require legato appear in the trebles
closing formulas in mm. 11-12 (F-E -F) and 24 (A-G-A).
The A-major prelude features only one indication for ornamentation,
the cadential mordent in m. 3. Since it is approached stepwise, this mordent
begins on the main note and can do with only three notes. In m. 14, the
upper voice features a similar figure. It also appears in the context of a
cadential close and equally consists of a dotted-note group on the dominant
before its resolution into the tonic. Although Bach did not indicate ornamentation here, it may safely be assumed that such a typical closing formula
would have been decorated with the same cadential mordent (in this case
beginning on the upper neighbor note and featuring four notes).
The subject of this fugue consists of two subphrases. The first ends
with a 16th-note rest at m.14; the second, consisting exclusively of regular
16th-notes that descend in half-measure sequences, concludes with the C
at m. 33. The fact that the rest in the middle of the subject is of structural
importance and not tension-sustaining follows from the relationship among
the preceding notes. The F at m. 13 is composed as an appoggiatura that
resolves indirectly (i.e., through an artificial leading-note) into the eighthnote E. This resolution concludes a dynamic curve and thus marks the end
of the first subphrase. The scalar ascent after the rest picks up the tension
and prepares a second climax at m. 21. The 16th-note figures in the second
subphrase contain an interesting feature. Stripped of what may be recognized as ornamental splitting, the underlying quarter-note pattern performs
a descent in consecutive fifth that parallels the ones heard in the middle and
lower voices in simple intervals. The example below shows this simplified
melodic line.
The subject recurs five times. The only modification it undergoes
occurs at the beginning of the second section where the four initial 16thnotes sound an octave higher than the remainder of the phrase. (This is
certainly due to the confines of the keyboard in Bachs time: continuing in
the high register for even the first subphrase alone would not have been
possible).
A major
255
256
WTC I/19
mm. 1
2
3
U subject
M
CS2
L
CS1
mm. 12 13 14
U subject
M
CS2
L
CS1
5
6
CS2
CS1
subject
7
8
9
10 11
12
episode
CS1
cadent.
with
subject
for9 motif A
CS2 9 mula A
15 16 17 18 19
episode
CS2
with
CS1
9 motif A subject
20 21 22
subject
CS1
CS2
9
23 24
cadent.
formula A
A major
257
See m. 16: the initial B is harmonized in B minor but the statement continues as if in
F minor, later modulating to E. See also m. 42: the initial F is harmonized in F minor but
the statement continues as if in A major, with only the final resolution returning to F minor.
258
WTC I/19
A major
259
from F minor to B minor in this final extension of the subject. The fugues
last subject statement (see mm. 44-46) features a freely varied ending.
Oddly, the only true stretto in this fugue is the one between the initial
statement and its answer. Further in the course of the composition, subject
entries appear neatly separated from one another. There is one instance,
however, where the rhythmic pattern of the voices accompanying a subject
entry contains a hint of stretto. In mm. 25-26 the middle and lower voices,
accompanying the upper-voice entry, feature a single eighth-note on the
second of the compound beats, followed by a 3/8 rest and a sequence of
eighth-notes in leaps of fourths. Although neither the intervallic connection
between the two subphrases nor the end conforms to the subjects original
shape, the entry of this rhythmic group gives the impression of a stretto.
Owing to the many sequences within this subject, this impression soon
turns into one of parallel motion or homorhythmic contrary motion
respectively (see m. 26).
In the main body of the fugue, the voices accompanying the subject
statements display hardly any characteristic or even recurring features, and
true contrapuntal lines are non-existent. The small figures that do appear
repeatedly are all derived from the subject in a more or less direct way.
Only with the introduction of regular 16th-note motion from m. 23 onward
can one distinguish voices to such an extent as to follow more than one
longer melodic idea at a time. These 16th-notes create a subject companion
that recurs once. In its original appearance, CS spans M: mm. 231-248.
When taken up in mm. 27-29, the counter-subject is extended along with
the subject; its end now falls on m. 291. This counter-subject does not seem
to comprise any features that would distinguish its dynamic outline. The
absence of any rhythmic hallmarks within the constantly flowing
16th-notes and of particular intervals or otherwise highlighted melodic
features classify it as a passively jingling accompaniment rather than a
competing partner. However, in order to distinguish CS from the many
other 16th-note figures yet to come, it is possible to stress the hidden line
inside the runs.
In the passages of the fugue that are devoid of 16th-note motion one
can discern three little motifs that appear repeatedly. M1 first emerges in
U: mm. 3-4. With its fourth leap, tied note, and weak-beat resolution it
260
WTC I/19
sounds like a free sequence of the subjects final five notes. This motif
recurs at the beginning of mm. 5, 6, and 7, each time in the middle voice.
Moreover, variations of this figure can be found in U: m. 43 and M: m. 44.
M2 is introduced in U: mm. 58-65. Here an initial tied note and weak-beat
resolution are complemented by two notes in zigzag motion (a third or,
later, a fourth leap up / a fifth leap down). This figure imitates the end of
the subjects answer plus the two notes following it (see M: mm. 36-41).
M2 recurs in mm. 7-8 (U/M in parallel), 8-9 (M), 10-11 (M, imitated in U),
mm. 11-12 (L, imitated in U), 17 (M, imitated 17-18 in U, then sequenced
in M), 18-19 (U), 22 (M), 43-44 (U), 46 (M), 47-48 (U), 51 (U), 52-53 (M).
If one allowed for interval modification, many more recurrences could be
observed. M3 appears as an emotionally even more intense combination,
consisting of two tied-note-plus-resolution pairs in ascending sequence.
This figure occurs three times: in mm. 12-13 (M), 13-14 (U), and 16 (U).
A major
261
E4b (mm. 20-22) exposes in its upper voice a little 6/8-note figure that is
imitated in stretto (see U/M: mm. 20-21) before moving on to a rhythmic
variation of the subject, also with imitation (see mm. 21-22; owing to the
pattern of consecutive eighth-notes, both imitations soon form parallels).
E5 features the first half of CS (U: m. 29, imitated in L) as well as freely
leaping eighth-notes (L, imitated in U). E7b transposes this combination
(mm. 49-50), while E6c recalls it in free variation (mm. 39-411).
The only episode portion to feature a longer motif is E6b, the central
segment of E6 (mm. 36-394). This is particularly noticeable since all three
voices establish patterns of 12/8 length, thus momentarily weakening the
metric order of the 9/8 time.2 Both motifs are then launched again in their
original position, where they break off (see mm. 388-394 along with the 16thnote chain of equal extension in L: F-C, sequenced as G-D).
As the above-listed material reveals, several of the episodes are related:
E6a
.
E3
E7a
.
E4a
E7b
.
E5
In a wider sense, E6 and E4 are conceived in structural analogy: both
begin with a segment providing a cadential close, after which they proceed
to present some genuine episode material that is duly imitated. Both then
continue with a variation of primary material (a variation of the subject in
E4, a variation of the counter-subject episode in E6), and both conclude
in another cadential formula.
The role these episodes play in the development of tension within this
fugue is determined, on the one hand, by the frequent use of extended
cadential formulas and, on the other hand, by the relative independence of
the material presented. There are episodes or segments thereof that complement the preceding subject statement with a straightforward relaxation;
this is the case in E1, E2, E3, as well as in E4a and E6a. Other episodes
serve as bridges. This is true of E5, but less of the passages corresponding
with it. Then there are those episode segments that, after a cadential close
but before the beginning of the next subject entry, attract attention for their
own sake. They are the ones requiring the most radical change of color in
order to be fully appreciated. E4b and E6b are obvious examples. The only
subject-free passage not to fall into any of these patterns is the final
episode (E7). Its three segments, which all allude to different models heard
before, seem primarily to express a reluctance to end the fugue.
2
In mm. 362-374 see U: D-B, imitated in M: E-C, and M: F-E, imitated in U: G-F.
262
WTC I/19
The pitch pattern with its predominance of fourth leaps, and the
rhythmic pattern with its regular eighth-notes accompanied in the center
sections of the fugue by equally regular 16th-notes, both suggest a rather
lively basic character. The tempo may be fairly swift. Bachs 9/8 time
should be interpreted as a choice of notation rather than of pulse, to be
rendered with the idea of a compound triple time.
Articulation in this piece needs careful planning if it is not to destroy
essential musical details. The basic attitude corresponding to the character
encompasses distinct non legato for the eighth-notes (in a crispness that
comes close to the staccato touch of later musical eras), non legato of more
extended duration for the quarter-notes, and a quasi legato of almost
classical leggiero quality for the 16th-notes. These categories, however,
include important exceptions: Within the non legato, all appoggiaturas
must be linked to their resolutions. This applies to the appoggiaturas in the
subject as well as to those in M1, M2, and (twice) in M3. Moreover,
several notes in the cadential formulas demand legato (see mm. 15-16 M:
D-C-D; mm. 19-20 M: E-D-E, mm. 22-23 U: A-G-A; mm. 48-49 M:
A-G-A; mm. 53-54 U: A-G-A). Within the quasi legato of the 16th-notes,
a distinction is desirable between the hidden melodic notes in the
counter-subject and the remaining notes of more directly virtuoso quality.
The tempo proportion between prelude and fugue can be rendered in
two slightly different ways. Performers wishing to emphasize the contrast
of characters, with a more lyrical prelude preceding a more virtuoso fugue,
will opt for a 16th-note in the prelude that corresponds to an eighth-note in
the fugue. Conversely, performers seeking to diminish the contrast and to
present a prelude in Allegretto motion followed by a fugue in Allegro ma
non troppo tempo will choose a proportion in which a quarter-note in the
prelude corresponds with a dotted quarter-note in the fugue.
The fugue comprises two ornaments. One of them, that in m. 8, is a
typical attribute of cadential formulas. As such it should be transferred also
to the corresponding formula in m. 41 (on the dotted G). The second ornament is designated in m. 26. All ornaments begin on the upper neighbor
note and move in 32nd values. The one in m. 8 comes with a written-out
suffix, while its structural counterpart in m. 41 is followed by a weak-beat
resolution. The latter thus represents an interrupted trill that stops short on
the fourth note of the shake, which is then tied over and sounds as a
syncopation to the measures final eighth-note. The trill in m. 26, like that
in m. 8, is a note-filling trill comprising eight 32nd-notes including the
suffix notes.
A major
263
mm. 8-9
41-42
26-27
The fugue contains several features that indicate its structure. These
appear, in the first instance, in the change of rhythmic pattern, the choice
of the episode material, and the texture, and in the second instance, in the
order of subject entries and the harmonic design.
The emergence and disappearance of the 16th-note accompaniment
results in three differently colored passages. The first ends at m. 231, the
second at m. 421. The two structural caesuras are preceded by episodes
that, as was shown above, are conceived in structural analogy: E6 and E4
both consist of a segment with a cadential close followed by genuine
episode material and a variation of primary material. The fact that the
subject statement in mm. 23-24 appears in reduced ensemble further corroborates the first caesura.
Within the first larger passage, the redundant lower-voice entry in mm.
6-7 announces the imminent end of the first section, confirmed by E1 with
its cadential formula. The second section, beginning in m. 9, also features
a redundant lower-voice statement (a repetition of the preceding entry).
Within the second larger passage, the only internal episode, E5, concludes
the third section after a complete set of three statements, one in each of the
voices. The fourth section then begins in m. 31 and contains two subject
entries as well as the closing episode. The relationship between the statements of these sections is also confirmed by the harmony: while the three
statements of the third section appear in the keys of tonic and dominant,
the two entries of the fourth section belong to the subdominant (D major
and its relative B minor). The third passage comprises only two statements
followed by a long episode. No further subdivision into sections is possible
here. However, three facts demand a slight modification of this notion: the
final episode consists of several structurally distinct segments, the first of
which ends with a cadence in the home key (mm. 48-49). Above all, the
16th-note motion that is characteristic for the fugues central passage but
abandoned in the fifth section is taken up again from m. 49 onward. All
this indicates that one should regard the fifth section as completed at m.
491 and followed by a coda in the final six measures.
264
WTC I/19
A major
265
The harmonic outline confirms the structure described above. The four
initial subject statements are in A major, the key in which the first episode
concludes the first section. After another entry on the tonic, E2 modulates
to the relative minor key. The following two statements each modulate
(F minor to B minor and B minor to E major), but E4 closes the section
again in the home key. The third section resembles the first one in that its
statements all relate to the tonic. The fourth section, just like the second,
leaves the home key area, this time for the subdominant and its relative
minor. The episode closing this section re-establishes not the home key but
its tonic relative F minor, leaving the final return to the second, modulating statement in the fifth section.
As a result of their descending order, the four initial entries of the
fugue create the deceptive impression of a four-part fugue. The tension
grows throughout and is only released in the concluding short episode. The
second section features a reverse of this process. Its first entry is the only
harmonically stable one in the original major mode, while its second statement begins in minor and ends in a modulating extension. The third entry,
conceived again as a redundant statement, is varied to such an extent that
it appears much weaker than the preceding one. This section comes to a
transitory close in very soft shading on the downbeat of m. 20, after which
the remaining segments of E4 create an independent little tension curve.
As the third section sets out in reduced ensemble and, what is more, as its
three entries in L, U, and M are arranged in such a way that their actual
pitch position (particularly that of the initial notes) sounds in ascending
order, a gradual increase of tension similar to that in the first section is
created. The fourth section recalls the second one insofar as it also
proceeds from the major to the minor mode in decreasing tension, and also
comes to a transitory close in soft color after the first segment of its
concluding episode, after which the remaining segments of E6 attract fresh
attention for their motivic material. The fifth section begins in the minor
mode. Its second subject statement, although returning to the major mode
and even, in its extension, to the tonic, remains comparably inconspicuous.
In the coda, however, the prolonged ascent of the leaping eighth-notes in
mm. 49-51 engenders a final climaxone that is virtuoso rather than
thematic.
The relationship between the sections is an important feature of this
fugue and should by all means be conveyed in performance. The increase
of tension in the first section and the complementing decrease in the
second section, followed by the small tension curve in E4, all find a
faithful correspondence in the increase of tension in the third section, the
266
WTC I/19
The harmonic conclusion of section II is very strange. We expect a return to the tonic, and
the chords in mm. 20-21, D minor and G diminished-seventh, seem to announce this with
their ivvii7. At the last moment, however, the lower voice sidetracks with a leap to C instead
of a step from the leading note to A. The tonic chord thus appears in inversion and fails to
satisfy the listeners demand for a resolution. Even more strangely, the process is repeated
similarly three measures later, where the left-hand progression G-A-B (A) diverts to C and
an unusual major-seventh leap adds to the harmonic irregularity. It is only after yet another
cadential measure that the downbeat of m. 26 finally presents the true tonic bass.
267
268
WTC I/20
tempo should be fairly swift, with the 9/8 time signature read as a convenience for 3/4 time with triplets. The appropriate articulation includes non
legato for the eighth-notes and quasi legato for the 16th-notes. The score
does not contain any ornament symbols.
The one-measure principal motif M1 consists of an A-minor chord
broken in zigzag, changing direction with each of the three compound
beats: up, down, up. Only the first interval is filled with a passing note.
This motif is sequenced twice, one step up in m. 2 and another three steps
up in m. 3. Thereafter, the gradual ascent is complemented with the upper
keynote on m. 41. The subsequent passive descent consists of a run and a
falling A-minor chord in a complementary pattern between both hands.
This descent appears more like a link than an integral part of the fourmeasure phrase, and later modification can easily be anticipated. Only the
very last eighth-note in m. 4 announces the shift from the tonic to the
minor dominant with a D, representative of the chord B-D-F-A, the
dominant-seventh of E minor. In this key the four-measure phrase is then
imitated by the lower voice.
The link following the imitation in m. 8 features the run and broken
chord in ascending direction. The final eighth-note of the measure, A,
seems once again about to initiate a harmonic shift, this time to the key of B.
However, the motifs ensuing entry in L: m. 9 ignores the modulation and
begins once more on E. The shape of the four-measure phrase is modified
here. The two sequences of M1 are descending in steps and the linking
measure, instead of displaying a complementary pattern, splits the original
ingredients. With simultaneous runs in the upper voice and broken chords
in the lower, it ends the first section of the piece.
Throughout the entire initial section, the contrapuntal companion is
characterized by a long note on the first compound beat in each measure
followed by a trill figure on the second beat. Within the initial phrase, both
the long note and the trill represent a tonic pedal, while in the imitation and
the developing phrase, block chords and moving trills accommodate the
steps of the harmonic progression. The final 3/8-note group in each
measure serves as a small-scale link, consisting of broken chords in the
initial phrase and of varying runs in the second and third phrases.
The second section sets out in m. 13 with M1 and its original fourmeasure phrase back in the upper voice, the sequences once more moving
upward. There are, however, two new modifications. The counter-motif
omits the trill figure and instead complements the long notes at the outset
of each measure with runs. Moreover, the link at the end of the phrase
A minor
269
combines the idea of a run with that of the broken chord by retaining the
chord notes from within the run in a split-voice texture. This phrase ends
in C major, without any sign of a modulation even in the final eighth-note
of m. 16. The next measure thus presents a surprise both in pattern and in
harmony. Over a repeated pedal note on the pitch a very mindful listener
may have anticipated for the pivotal eighth-note (F, of D-F-A-C), the
upper voice fits the remaining chord notes into a new complementary
pattern. The resolution of this modulating chord follows immediately with
the newly established pattern, in the final measure of the C major phrase
(compare m. 18 with m. 16). A sequence of the pivot-chord measure
then leads to D minor in m. 20. The last two measures of the second
section come full circle with the principal motif, now in the lower voice,
accompanied by a free variation of the modified counter-motif (compare
U: mm. 20-21 with L: mm. 13-15). The unusual harmonic ending of the
section was already mentioned.
The beginning of the third section gives the impression of a recapitulation: the upper-voice motif begins as did the original in m. 1, and the
counter-motif also takes up the trill-figure familiar from the first section
(compare m. 22 with m. 1). The third beat of the measure, however, features
a variation in the motif, the second measure is harmonically diverted
(compare m. 23 with m. 2), and the expected third segment of the
sequential pattern is substituted by a closing link (similar to the one that
concluded the initial section in m. 12). This leads to the reiteration of the
cadential-bass deviation (see mm. 24-251). The remainder of m. 25 continues the scalar descent and, more importantly, corrects the cadential close
by reiterating it with the proper bass steps and melodic leading note.
Finally, mm. 26-28 are built upon a tonic pedal. The lower part with its
voice splitting is reminiscent of the beginning of the second section (mm.
13-15), while the upper part displays a free variation of the principal motif.
The trill figure, added at the very last moment in a split-half of the upper
voice, recalls once more the characteristic feature of the original companion and thus ends this invention with a five-part A major chord.
The dynamic presentation of the composition should be designed to
reflect the structural outline as well as the varying thematic density. The
climax of the first section falls on m. 9 where, after two four-measure
phrases in continuous crescendo, the developing phrase sets in with a sudden major mode, both voices sounding in a fairly high register. The second
section reaches its dynamic peak at the end of its initial phrase (m. 16).
Thereafter, the tension recedes in gradual waves. (Careful: The G at m.
270
WTC I/20
211 often blurts out because of the preceding large leap but should actually
be very soft, in keeping with the pending cadential close.) In the third
section, the dynamic curves are very gentle as long as tonal stability is still
awaited. Then, however, in the final three measures with their multiplying
parts, a powerful crescendo leads to a luscious completion.
A minor
271
In the course of the fugue, Bach harmonizes this subject with slight
variations. In mm. 8-9, e.g., the dominant already appears on the downbeat
of the second measure, while in mm. 4-5 the tonic is retained much longer
and the dominant is only reached on the middle beat. In the second half of
the same two statements, Bach harmonizes m. 101 as a subdominant
representative, while the analogous m. 31 returns to the dominant after only
an eighth-note resolution to the tonic. Beyond these differences, the basic
harmonic outline affirms a progression in which the two relevant steps are
the short-lived resolution of the dominant to the tonic on the last eighthnote of the second measure and the perfect cadence that underlies beats 3
and 4 of the subjects third measure and resolves on the final downbeat:
Note that this fugue, while basically in four parts, introduces a fifth voice after the general
pause in m. 80. This new voice establishes a tonic pedal in mm. 83-87. It seemed practical,
however, to retain the counting of the original four voices as soprano, alto, tenor, and bass
during these measures of five-part texture and use bass 2 for the additional voice that
does not participate in the polyphonic play.
272
WTC I/20
A minor
273
Given that the imitation distance in the strettos is usually four eighth-notes,
i.e., the same length as the sequence at the beginning of the subject, this
results in a short parallel. By contrast, parallels of any larger segments of
the subject do not occur.
In a fugue that is so essentially based on stretto work, it is certainly not
surprising that we will look in vain for a characteristic and independent
counter-subject. There are two short accompanying figures that deserve to
be mentioned since they recur. Their common feature is the fact that both
establish a short parallel to one fragment of the subject. They are thus not
quite polyphonically independent. In mm. 4-5 the alto, having just introduced the subject, builds a counterpoint to the beginning of the subject
answer. It begins with two descending tetrachords (see the eighth-notes
A-G-F-E and C-B-A-G), continues with a parallel of the first half of the
second subject measure, and ends, at the latest, with the B on the middle
beat of m. 5. This counterpoint is taken up in mm. 8-9 where it is shared
between soprano and alto, and in mm. 11-12 in the alto (where the
tetrachords are reinforced by a parallel in the soprano). Later in the fugue,
the sequencing tetrachords alone, without the ensuing parallel turn-figure,
recur frequently as a companion to a subject entry; see particularly mm.
14-16 (B, T, A), 28-29 (B), 53-54 (T), 57-58 (T+B), and 73-74 (T). In mm.
18-19, the soprano creates a short parallel to the second segment of the
subjects first subphrase. Similar brief parallels occur in mm. 21-22 (A)
and mm. 24-25 (T). As these parallels can also be read as rudimentary
imitations of the subject beginning, an interesting interpretation is that they
are embryonic forerunners of the stretto entries to come.
This fairly long fugue encompasses altogether sixteen subject-free
passages; most of them, however, are only very short.
E1
mm. 7-8
E9
mm. 52-53
E2
m. 14
E10 mm. 56-57
E3
mm. 17-18
E11 mm. 61-62
E4
m. 27
E12 mm. 63-64
E5
m. 31
E13 mm. 71-73
E6
mm. 35-36
E14 mm. 79-80
E7
mm. 40-43
E15 mm. 82-83
E8
mm. 46-48
E16 mm. 86-87
In the material that makes up these subject-free passages, two facts
stand out: many episodes are simple cadential formulas of one-measure or
even only half-measure length, and several others feature the subjects
head motif. E2, E3, E4, E15, and E16 present variants of the cadential bass
steps, E2 and E4 additionally feature the dosido formula, E3 and E4 the
274
WTC I/20
typical melodic turn with dotted leading note and anticipated resolution. In
three other episodes, the same components make up only a portion of the
subject-free passage; see the second half of E8 as well as the conclusion of
E12 and E13. In the latter case, the sopranos cadential formula overlaps
with the beginning of the subsequent statement in the bass. The subjects
head motif is found in seven of the episodes, often accompanied by a
scalar passage (see E1, E3, E6, E7, E10, E12, and E16). Finally in E5, E8,
and E9 the scalar segment appears without the subjects head motif
Only three of the episodes feature figures that are immediately
sequenced and thereby attain a minimum of independence: bass and alto in
E7 mm. 40-41, the trill motif in alto and tenor of E9 as well as in the
soprano of E10, and the imitations of soprano and tenor including their
sequences in the alto of E13. E8 is the only episode presenting an imitation
of the subjects tail (see T: mm. 45-47). The 16th-notes in all episode types
form ornamental figuration or broken chords.
There are few structural analogies among the episodes. One can detect
a correspondence between two cadential and two non-cadential passages
(E2 and E4, E1 and E5). Moreover, E9 and E10 share the same prominent
motif and scalar descent. The role each episode plays in the development
of tension within the fugue is determined by motifs and closing formulas.
E1 bridges between two consecutive statements but does not create
a color contrast since the Ms figure appears as a sequence of the
previous measure. Conversely, E2 gives the impression of a close,
with a definite relaxation. E3 is ambiguous: the bass pattern speaks
of conclusion, the false entry in the alto seems to point forward. E4
follows with an unequivocal closure and a definite tension release.
Like E1, E5 connects two entries. E6 with its ascending sequences
leads to the subsequent stretto. E7 introduces the first color contrast
owing to its novel material and, thereafter, a decrease of intensity.
E8 presents the long-awaited closure with its extended cadential
formula. Surprisingly, it is succeeded by what must be interpreted as
a preparation for the ensuing subject statement.
E9 and E10 serve as dynamic bridges. E11 like E7 presents genuine
episode material with a concomitant color contrast but no conclusion (the stretto in mm. 62-63 is structurally analogous to that in
mm. 43-46), and E12 with its cadential close corresponds with E8.
E13 also suggests a color contrast because of its genuine little motif.
Its ending with the overlap of a cadential close with the beginning of
an ensuing subject statement recalls E2 and distinguishes this as
another bridging episode. E14 differs from all its predecessors by its
A minor
275
For appoggiatura-resolution see mm. 6-7 [alto: C-B, E-D, D-C], m. 11 [soprano: D-C], m.
26 [soprano: B-A], mm. 40-41 [tenor: G-F, F-E, E-D], m. 62 [alto: F-E]; m. 64 [alto: G-F];
m. 70 [tenor: A-G]. For do-si-do formulas see mm. 13-14 [alto: E-D-E], mm. 16-17 [bass:
G-F-G], m. 23 [soprano: C-B-C], mm. 26-27 [soprano: A-G-A], m. 32 [tenor: E-D-E], m.
34 [alto: E-D-E], mm. 47-48 [soprano: C-B-C], m. 60 [bass: A-G-A], m. 64 [tenor:
D-C-D], m. 73 [alto: F-E-F], mm. 75-76 [bass: G-F-G]. There is only one closing formula
in the bass that requires partially legato articulation; see mm. 47-48 [D-E-F-G = legato].
276
WTC I/20
figures in the soprano carry the mordent that is so typical for this formula.
The brackets indicate that the composer himself did not find it necessary to
specify this obvious ornamentation, but one of his students took it down.
The same brisk mordent should be added in m. 17 on the dotted F and in
m. 73 on the dotted G. Among these, only the mordent in m. 17 begins on
the main note and therefore makes do with only three notes; the remaining
ones are launched from the upper auxiliary and comprise the regular four
notes.4 The trill in the motif appearing in E9 and E10 represents a
note-filling ornament. Starting each time on the upper neighbor note and
shaking in 32nd-notes, it comprises eight notes including the suffix that is
spelled out in all cases.
Much headache has been caused by the apparent impossibility to play
the fugues final five measures with their sustained bass pedal and the
texture with up to seven voices. The following excerpt attempts to help by
suggesting one possible execution, fingering, and use of the middle pedal.
In m. 27, the figure appears in the alto and thus makes ornamentation much more difficult
to execute. However, for performers with a good technique it is possible to play the
mordent in the left hand, then take over its sustained final note in the right thumb, the
ensuing 16th-note A again in the leftquite without interrupting the legato required for the
upper voice.
A minor
277
278
WTC I/20
more striking is the fact that the first stretto in section V is at first entirely
unaccompanied. By contrast, the beginning of section VI with three active
voices gives the impression of reduced ensemble but actually encompasses
four parts; the indirect pedal C in the soprano suddenly comes to life again
in m. 75. This substantiates the interpretation that sections V and VI are
more closely connected than any other consecutive sections. The fugues
overall design thus contains two pairs of corresponding sections (sections
I/II and III/IV), followed by the only minimally structured larger body
containing sections V/VI and a short coda.
The harmonic outline supports the structural correspondences.
The first section comprises entries on the tonic and dominant of A
minor before ending on an E-major cadence. Section II begins on
the dominant and returns to the tonic with an A-minor cadence.
I/II = tonic - dominant - tonic
During its three initial strettos section III is still firmly rooted in the
home key. After the longer interruption of E7, however, the redundant entry-pair presents the relative major key, which is confirmed,
at the end of E8, with a cadential close in C major. Section IV
returns to the home key but ends, both in its second stretto and the
final abridged pair, in the subdominant region, which is confirmed
in E12 with a cadential close in D minor.
III/IV = tonic - its relative major - tonic; end on subdominant
The two final sections, earlier ascertained as forming a larger unit,
further underscore the subdominant field. Section V begins with a
stretto of entries in A minor and D minor, and were its final cadence
not placed across the beginning of the sixth section, it would conclude in F major, the subdominant relative. Section VI, after several
harmonic adventures,5 ends with a perfect cadence in A major, the
Picardy-third version of the tonic (m. 83). In the coda, however, the
final four-part stretto is led by a soprano statement in D minor, a last
reminder of the subdominant key set against an overwhelming
assertion of the tonic (three A-minor entries and the A pedal). The
final cadence, as expected, confirms the A major conclusion.
V/VI = subdominant - (its relative major) - major tonic
coda = subdominant/(minor) tonic - major tonic
5
A striking number of pedal notes mark section VI as concluding. See mm. 73-75: C
(soprano), mm. 76-77: G (bass; beginning at m. 753 but escaping momentarily to its leading
note), mm. 79-80: D (bass, interrupted), and mm. 83-87: A (additional second bass voice),
as well as the two inverted dominant-seventh chords that, enhanced by the voice-splitting
into six parts, precede the reiterated general pause in mm. 80 and 82.
A minor
279
280
WTC I/20
282
WTC I/21
B major
283
The chief processes in the prelude rest on the facts described above.
The first phrase is determined entirely by figure 1. It appears in a pattern of
descending half-measure sequences and is rounded off by a half cadence.
The descending sequences are counter-balanced in the following phrase by
two ascending sequences built from combinations of figures 1 and 2. The
third phrase picks up the ascending motion, from B through an octave to
A, beginning in straightforward chromatic steps and continuing in a diatonic passage with octave displacements. While the ascent presents itself in
figure-1 pattern, figure 2 in m. 8 returns in a double curve to the low E, the
leading-note of the long-anticipated dominant. A one-measure cadential
pattern with figure 1 confirms this key and with it the preludes first half.
In m. 10 Bach subtly creates the impression of both rounding off and,
at the same time, announcing something new. While the first half of the
prelude is made up entirely of the two figures with open-position broken
chords and scalar runs, the tail of its final cadence introduces a combination of the two figures: a pattern that, like figure 2, uses one-track texture
but, like figure 1, moves in broken chords. This combination (here referred
to as figure 1+2), turns out to frame the second half of the piece.
The first statement of the dotted-note figure 3, presented in m. 11, is
linked to its two descending sequences by extensive statements of figure 2
whose dive and soar respectively span almost three octaves. Despite the
inverted direction of the run in the second statement of the figure, the
listener can still perceive the continuing sequential pattern. After the second
sequence, the recurrence of figure 1+2 seems to announce the impending
close of this section. Yet once again the addition of a seventh redefines the
tonic triad and thus diverts the cadence. Another statement of figure 1+2
leads to the compound ornament before one measure each of figures 2 and 1
complete the preludes larger frame in symmetrical fashion.
284
WTC I/21
Phrasing in the subject therefore occurs after the downbeats of mm. 2 and 3,
with exactly one measure length for each of the two subphrases, while the
subsequent two measures represent a single unit.
The interval organization in this subject includes sixths, fourths, and
seconds in the eighth-notes and predominantly small intervals in the essentially ornamental 16th-notes. As the sixths occur in weak metrical positions
(i.e., after the beat, not toward it), these large leaps do not convey emotional tension but an energetically bouncing temperament. Corresponding
with this lively interval pattern, the rhythmic pattern throughout the fugue
is very simple, consisting almost exclusively of eighth-notes and 16thnotes. Even the regular syncopations in the main contrapuntal voice (see
U: mm. 5-7 etc.) cannot diminish this effect.
The subjects harmonic background is characterized by two active
steps within the first two subphrases: both times a simpler chord underlying
the 5/8 upbeat is followed by an incomplete minor-seventh chord. The
third measure concludes the cadence with a straightforward V7-I. The
harmonic progression thus ends already at m. 41. Yet for the dance-like
character Bach intended for this fugue, he prefers phrases with an even
number of measures. The additional eight eighth-notes, consisting of an
exact repetition of the ones just heard, thus appear as a metric complement
with neither melodic nor harmonic information.
I ii7
vii7 iii7
V7
V7
The subjects dynamic design should convey the phrase structure, the
metric organization, and the harmonic features. A distinctly virtuoso aspect
enters with the ascending sequences and their growing amount of embellishment. Within the first two subphrases, the climax falls on the respective
downbeatsboth because this is the point of harmonic emphasis and
because any other accent would blur the metric structure. (If, as happens
easily for lack of attention, the peak notes B and C are accented, this will
necessarily give listeners the mistaken impression that the subject begins
with three upbeat eighth-notes followed by the first downbeat on B.) By
the time the third subphrase is approached, the meter is firmly established.
Therefore, the fact that this incomplete sequence comes without what was
previously the climax and instead just relaxes from the E (V7, m. 33) to the
final D (I, m. 51), suggests that in this subphrase, highest pitch and
dynamic peak coincide. With regard to the relation between the three
B major
285
Except for the interval adjustment in the tonal answer, the subject
remains untouched throughout the fugue in detail as well as in shape. No
stretto or parallel statement ever materializes. But ever after its first entry,
it comes escorted by companions that remain faithful to the end. They are
not, however, quite as independent from their leader as true contrapuntal
technique would require, particularly in the case of the second companion,
which runs largely in parallel to the subject. Although the term countersubject might thus not seem appropriate in its strictest sense, the usual
abbreviations are employed here for easier reference. CS1 is introduced in
U: mm. 5-9. Its three subphrases coincide exactly with those of the subject.
As in the subject, the first and second subphrases are related by ascending
sequence (although the first subphrase appears more elaborate than the
second here). Also as in the subject, the third subphrase consists of a onemeasure model followed by its repetition. (This subphrase encompasses a
four-note parallel to the subject; see m. 81.) In terms of the dynamic
outline, the three climaxes fall slightly earlier than those in the subject. The
prominent features within the first two subphrases are the final syncopations, which capture the peak of tension. In the third subphrase, the short
upbeat-like ascent preceding the protracted note repetition indicates a soft
crescendo, which is followed by a long diminuendo. CS2 seems more like
a filler than a polyphonic partner. Introduced in U: mm. 9-13, it consists of
four short gestures interrupted by rests. The first two, again conceived as
ascending sequences, complement the two 16th-notes missing in the
rhythmic patterns established by the subject and CS1. Their second note
sounds in unison with the other companion, and their final eighth-note,
286
WTC I/21
B major
287
each sequence determines the dynamic tendency and, with it, the role each
episode plays in the course of the fugue. Thus E1 momentarily surpasses
the tension expressed in the fourth subject entry but subsequently recedes
with each measure. E2 is conceived, despite its two-fold structure, as a
single decreasing line, and the first segment of E3 resembles an echo from
which the cadential close resurfaces.
One can thus state that this fugue is only rudimentarily contrapuntal in
texture and entirely determined by its subject. At the same time, the subject
remains virtually unmodified throughout the composition. This may
suggest that what matters in this fugue it is not the material, and that the
chief task fulfilled here by the subject is to establish the vigor and mood
upon which the entire piece depends.
The basic character is lively. This is supported both by the simple
rhythmic structure and by the pitch pattern with its frequent leaps and its
ornamental 16th-notes. The tempo may be very swift, particularly since the
demands of polyphonic playingand hearingin this fugue are almost
negligible. The articulation combines an energetically bouncing non legato
for the eighth-notes with a crisp quasi legato for the 16th-notes. Ornaments
do not occur in this fugue. The relative tempo of the prelude to the fugue
uses the larger pulse for a proportion: a half-note in the prelude translates
into a dotted half-note in the fugue. Approximate metronome settings:
prelude beats = 72, fugue beats = 108.
In the absence of changes in texture or conspicuous cadential patterns,
an analysis of the structure must rely on the episodes and the harmonic
development. As has been shown, the episodes are related with one another
in that two patterns introduced in E1 return in reverse order in E3 and E2.
This may be interpreted as suggesting an axis symmetry (a+b b, a) and
can thus hint at a design in which the first section would find its correspondence in the joined second and third sections. The tonal organization
supports this view. The four initial subject entries remain in the sphere of
B major. The first episode then modulates via the relative key G minor to
its dominant D Major. The fifth and sixth subject statements represent the
sphere of the relative-minor key (with G minor in mm. 22-26 and C minor,
the subdominant relative, in mm. 26-30). The ensuing episode confirms the
key of C minor. Thereafter, the incomplete statement leads back to the major
mode. Thus the next complete statement can enter in the subdominant E
major, followed by the return to the tonic in the final statement.
A big dramatic development is clearly not the purpose of this virtuoso
fugue. Once the subject and its two companions have been established,
there is little distinction between successive entries. The interpretation of
288
WTC I/21
290
WTC I/22
The tempo in which this character can best be expressed is fairly slow. The
quarter-notes of the time signature can be imagined as a solemnly swaying
movement, and the eighth-notes perceived as the continuous pulse should
be completely without haste. The articulation required in such a piece is a
dense legato.1 The score does not contain any ornaments.
The pulse in this metrically determined prelude is thus two-fold: the
main swinging motion is in quarter-notes, whereas the consistent pounding
is experienced in the eighth-notes. This ambivalence is established in m. 1
where the unremittingly throbbing pedal note contrasts not only with the
rhythmic motif but also with the mid-texture chordsand these, in their
iambic pattern, underpin the larger quarter-note pulse.
rhythmic model:
chordal background:
repeated pedal:
As the example shows, the preludes initial phrase displays a constantly changing number of voices (from four to seven, though mostly
five). Thereafter, the texture from m. 3 up to the downbeat of m. 20 is in
consistent four-part writing. The recapitulation of the initial phrase then
resumes the five-part texture, with extra voice splitting only in m. 22,
where beat 2 features seven and beat 3, nine voices.
While the initial phrase might easily pass for the introductory line of a
homophonic composition, there are several passages that may be described
asmodestly yet very consistentlycontrapuntal. A plausible interpretation of these confusing data is the following: the prelude can be regarded
as a work in four-part writing whose opening phrase and its recapitulation
1
As this prelude contains such a large number of repeated notes, the help of the damper
pedal will probably be inevitable to produce the desired smooth result. This needs careful
pondering. Using the pedal only where there are repeated notes would create an uneven
tone color. Thus regular pedaling throughout is the only alternative to not using the pedal
at all. The cleanest solution is to depress the pedal after the last 16th-note fraction of each
eighth-note beat, and to release it with the onset of the following eighth-note. The nature of
this meditative piece, however, admits also a slightly less transparent execution in which
the pedal is simply changed on each eighth-note, thus giving the second 16th-note in each
rhythmic figure a somewhat murky coloring. Eighth-notes, however, must be clear.
B minor
291
In m. 22, bass 2 splits first to the third C/E, then to the triad A/C/E, only to resume its
former single-voiced sounding with the low F on the fourth eighth-note of m. 23.
292
WTC I/22
the two leading voices (mm. 15-17). The section tail, which like those of
the other sections is freer in style, sets out in m. 18 with a joint four-part
realization of Rh2 and ends by combining features from the first and
second sections (compare the sustained soprano and the descending bass in
the first half of m. 19 with m. 6, and the bass repetitions leading into an
imperfect cadence in mm. 19-20 with mm. 9-10). In the fourth section, the
varied recapitulation of the initial phrase accompanies the leading soprano
motif with parallels first in the tenor, then in the alto, and finally in both. It
gives up Rh1 for Rh2 at the end of m. 21 and ends in a many-voiced
fermata on a vii7 chord. The ensuing cadential formula once again alternates
the two rhythmic patterns while returning, both in the sustained soprano
and in the repeated bass 2, to the tonic pedal and to the double-third texture.
The dynamic rise and fall in such a meditative prelude should be much
smoother than in other pieces. It develops as a secondary feature and must
never lure listeners away from the all-pervading pulse. A rendering that is
both natural and unobtrusive follows the course of the ascents and descents
of pitchboth those in the overall design and in sudden leapswith very
gentle increases and decreases. As a result, the first section undergoes a
gradual tension increase (approximately from p to mp) up to m. 5, followed
by a short release, while the second section presents a gradual release
(approximately from p + to p ), followed by a mild rebuild of the original
piano color in the cadential formula. The third section begins with very
delicate touch in the sparse texture of its first phrase, but in its second
phrase engenders the strongest and steepest build-up of tension in this
piece. This leads to the first overall climax (preferably no stronger than mf)
on m. 162, from which the remainder of the section descends in gradual
diminuendo. The fourth section contains the second overall climax. Despite
its compact chord, it should sound introverted and intense rather than extraverted and loud. The prelude closes on a soft note, giving the impression
that it returns to the level of the beginning.
B minor
293
metric endings in Greek poetry, these different shapes of the same trunk
are often referred to as male and female. The male version appears
condensed, concluding with the last essential harmonic step without any
softening addition. The female ending adds, in poetry, an unaccented
syllable after the final accent. In music, it extends the tonic reached in the
perfect cadence with notes passively remaining in the same harmony.
This regular extension is conceived as a smooth continuation, both in
rhythm and pitch pattern, of the second half of the main body: mm. 2 and
3 of the subject contain exclusively quarter-notes in stepwise motion. This
gentle curve appears in sharp contrast to the subjects beginning where two
half-notes followed by a downbeat rest create rhythmic tension. At the
same time, the falling perfect fourth, in itself a relaxed interval, is followed
(across the rest) by a minor ninth. This interval, rarely used for horizontal
progressions in Bachs time, expresses an emotional gesture of the highest
intensity. The question thus arises whether or not this exceptional melodic
jump is really meant to be perceivedand playedas an interval, i.e.,
whether the consecutive notes belong to a single structural unit.
Thus not only the ending but also the phrase structure of this subject
can be interpreted in two ways. In the first option the rest, together with the
unusually large distance between the two notes enveloping it, is regarded
as an indication of structural interruption, comparable to a comma in
verbal language. As a result, the subject consists of two subphrases that
differ considerably in all features: the first subphrase begins on a strong
beat and contains only two long note values in a falling fourth, while the
second subphrase sets out on a weak beat and features only shorter note
values in stepwise motion. In the second option, the rest is interpreted as
tension-sustaining, as a silent continuation of the melodic flow, and the
interesting interval is thus fully exploited. This minor ninth, whose tension
is enhanced by the fact that it spans across a suspended downbeat, then
becomes the subjects primary melodic focus.
The subjects harmonic background
can be depicted as shown in the example.
Modifications are frequent. In several cases,
i V i ivV i - - - - the male ending is raised a semitone and
harmonized not as a tonic but as a major
chord with seventh on the tonic root. This chord is heard as a dominantseventh and resolves correspondingly to the subdominant. The melodic
line supports this modulation by lowering the entire female ending one
whole-tone step, thus leading to the root of the target harmony. The fairly
frequent occurrence of this harmonization in the fugue provides additional
294
WTC I/22
B minor
295
The subject appears both in stretto and in parallel. The stretto settings
involve two essentially different cases of premature succession. There are
entries in which only the female ending of the leading statement overlaps
with the subsequent beginning (see mm. 3, 12, 27, 55). In other cases, the
distance is as close as possible, i.e., the second subject note in one voice
becomes the initial note in the ensuing statement (see once in m. 50, twice
in m. 68 and twice in m. 69). In the light of the latter, the parallel entry in
mm. 55-58 might appear as an even closer stretto.
In addition to the modifications caused by modulation, the subject
undergoes changes of melodic shape in connection with its tonal answer,
where the original falling fourth is augmented to a fifth and the ensuing
large interval spans a minor tenth (see, e.g., mm. 3-6). While this tonally
determined lowering of the subjects second note remains without effect on
the close on the tonic, some cross-combinations of these intervals occur
that do in fact alter the harmonic background of the statement. In mm.
25-28, the combination of an initial falling fourth with an ascending minor
tenth places the subjects tail one note up and causes a modulation from D
major to E major; conversely, in mm. 53-56 the falling fifth is combined
with the ascending minor ninth, thus engineering a modulation from B to
E. A last-minute tonal adjustment can also happen by way of a variation
in the female ending (see the final ascent to E in mm. 12-13).
The B-minor fugue features no counter-subject. This may be partly due
to the unusual female ending frequently serving as a contrapuntal accompaniment to the beginning of the second or later subject entries in a group.
The fugue comprises six subject-free passages.
E1 mm. 6-9
E4 mm. 40-46
E2 mm. 18-24
E5 mm. 58-67
E3 mm. 35-36
E6 mm. 72-75
Two of the episodes, E2 and E5, develop from the preceding statement
by sequencing the subjects second and third measures (see v5: mm. 18-20
and, in a more modified version, v2/v3: mm. 58-60). E1 and E4 contain a
motif that consists of an inverted-mordent figure ascending to a dotted
half-note on the downbeat. A genuine episode motif, M1 is sequenced and
imitated. It not only fills the entire subject-free passage but even extends
into the beginning of the next statement (see v1/v2: mm. 11; v5/v2: mm.
42-47; note that v2 is the uppermost of the three voices involved in these
measures). The two remaining episodes, E3 and E6, are essentially nothing
but extended cadential closes. E3 features a short cadential-bass pattern and
a longer dosido formula in v1, while in E6 these figures are of opposite
length (see v5 in whole-notes: mm. 73-75, v1 in half-notes mm. 74-75).
296
WTC I/22
Moreover, the same combination of cadential traits can also be found at the
end of E2 (mm. 23-25). The structurally analogous E5 does not show this
cadential close. Instead, this close appears in the context of a full-fledged
subject statement, in mm. 53-55, where it is easily overlooked.
As these details reveal, the episodes in this fugue are arranged in a
symmetrical pattern of structural correspondence: E1 relates to E4, E2 to
E5, and E3 to E6. With regard to the dynamic outline in each, E1 and E4
with their ascending sequences of M1 describe an increase in tension and
thus lead from the preceding subject statement onward into the subsequent
one. E3 and E6, on the other hand, conclude a structural section. In the two
intermediate episodes the overall pitch tendency is very pronounced. In E2,
the initial sequence from the subject tail begins lowerand therefore
probably softerthan the end of the preceding entry. At the same time, the
first voice completes a descending line begun immediately after the disappearance of M1 and continued through two entire entries, in ever larger
note values (v1, mm. 11-20: C-B-A-G-F-E, D-C-B, A). This long
descent is countered by sudden ascents in all five voices (see particularly
mm. 20, 21) that discontinue their build-up only near the cadential formula
at the end of the episode. In the structurally corresponding E5, these lines
seem reversed. The initial sequences from the subject tail begin in an
elevated position (mm. 58-59) but are then followed by descending lines.
The most prominent of these is the one in the uppermost voice which, as in
E2, concludes in whole-notes (v1 mm. 62-67: G-F-E-D-C-B-A-G-F).
This episode thus places its climax at the beginning and subsequently
conveys an almost complete release, so extended that the anticipation for
the events to come is heightened all the more.
The basic character of this fugue is rather calm, not only because of the
complex rhythmic pattern with its many syncopations and tied notes, but
even more so because of the combination of stepwise motion with high
melodic tension. Yet the tempo should not be slow since Bachs time
signature indicates a pace in half-note beats. The appropriate articulation
demands legato in all note values, with the exception of the cadential-bass
steps mentioned above. There is one detail, however, that depends once
again on the interpretation of the subjects phrase structure. For performers
who conceive the subject as consisting of two subphrases, the falling fourth
is the only larger interval; it may therefore be played legato, particularly
since it comes in strongly ebbing dynamics. By contrast, performers who
regard the subject as indivisible are faced with two consecutive large leaps
both of which express an active increase in tension. These should then be
detachedthe first by articulation, the second by the written rest.
B minor
297
Owing to the fugues alla breve indication, the relative tempo of the
prelude to the fugue is most convincing if rendered in the complex
proportion of 3:2, where an assumed eighth-note triplet in the prelude
corresponds with a quarter-note in the fugue. (Approximate metronome
settings: prelude beats = 40, fugue beats = 60.) The fugue contains only one
ornament, the trill in bar 49. It begins on the upper neighbor note, shakes
in three pairs of sixteenth-notes and ends with a two-sixteenth-note suffix
leading into its resolution on the ensuing downbeat.
Many of the observations made above in the discussion of the episodes
encourage an exploration of a possible structural symmetry in the fugue.
This symmetry does in fact exist; it is complex but striking. Here are the
relevant details:
first half (mm. 1-36)
second half (mm. 37-75)
S
in v1, v2
S
in v2
E1 with ascending M1
E4 with ascending M1
S
in v3, v4, v5
S
in v3, v5, v1/v2, v4
E2 with gradual ascent
last entry with cadential close
ending in cadential close
S in v2+v3
E5 with gradual descent
S
in v1, v2, v4, v5
S
in v1/v2/v3/v4/v5
E3 with cadential close
E6 with cadential close
Within each of the two structurally corresponding halves, the entering
order of the voices, together with the sense of closure conveyed by some of
the episodes, allows two sections to be distinguished. In the first half, the
layout is straightforward: section I comprises five subject statements linked
by a bridging episode (E1) and rounded off by the cadential formula at the
end of E2. Section II consists of four entries and the short E3 with its
cadential close. As to the structural boundaries in the second half of the
fugue, two interpretations are conceivable.
In the first concept (which accords with the rules derived from
other Bach fugues), section III, like section I, contains five subject
statements, one of which is a stretto. They are linked by a bridging
episode (E4) and closed by the cadential formula accompanying the
last entry. Section IV then consists of a parallel statement, an episode
(E5), the impressive five-part stretto and a short final episode (E6)
with cadential close. The textural density supports these findings:
sections II and III both begin with only three of the five voices
taking part (see mm. 25-28: v4 and v5 resting; mm. 37-45: v1 and
v3 resting), and at the beginning of section IV, after a full five-part
cadential close in mm. 53-55, v1 pauses for twelve measures.
298
WTC I/22
The second concept builds on the admittedly strong tendency of
relaxation in E5, on the structural analogy of this episode with E2
in the fugues first half (which serves as the final segment of the
first section). Moreover, the fact that the v4 statement in mm. 53-56
overlaps, in its female ending, with the ensuing parallel entry makes
it more difficult for both performer and listener to perceive a
section ending and new beginning here. In terms of the dynamic
development, this concept is much easier to realize. In structural
terms, however, it not only leaves the fourth section with a single
group entry as opposed to two (one parallel, one stretto), but also
results in a somewhat unbalanced overall picture.
B minor
Concept I: sections
I
mm.
1-24
no. of bars 24
299
II
25-36
12
III
37-54
18
IV
55-75
21
300
WTC I/22
to notice that Bach postponed the full five-part texture in this section until
this entry, and that, by contrast, the subsequent parallel statement sounds
in four-part setting only, as does the long E5, with the upper voice resting.
(This gives rise to a very sensitive question for performers: how does one
convey the absence of the highest voice when one is in fact playing high
pitches?)
The fourth section is thus, under several aspects, the weakest one. It
comprises only two (group) entries, its build-up of texture is smallest (only
from four to five voices), and its episode is the only one in the entire fugue
to constitute a protracted decrease of tension. In this regard, the fourth
section continues the gradual decline of dynamic power from one section
to the other. When the final five-part stretto counter-balances this overall
decline by erecting a powerful super-climax, this should come as a true
surprise to listeners.
302
WTC I/23
B major
303
304
WTC I/23
B major
The fugue comprises twelve subject statements:
1. mm. 1-3
T
7. mm. 18-20
2. mm. 3-5
A
8. mm. 20-22
3. mm. 5-7
S
9. mm. 21-23
4. mm. 7-9
B
10. mm. 24-26
5. mm. 11-13
T
11. mm. 29-31
6. mm. 16-18
A
12. mm. 31-33
305
Sinv
Ainv
B
T
A
S
In the course of the fugue, the subject undergoes two kinds of modifications, one at the beginning and the other at the end. The initial intervals
of both subphrases are enlarged from seconds to thirds in all tonal answers
(mm. 3, 7, 31). In the inverted answer, only the opening step of the first
subphrase is enlarged (m. 20). The final resolution is delayed (mm. 7, 31),
diverted (mm. 18, 20), or omitted (m. 22). In one instance, the subject
ending is varied but arrives in time at the proper resolution (m. 26). In
other words, only four entries are quite original: nos. 1, 5, 9, and 12.
Parallel statements do not occur; nor do true strettos in which a substantial segment of one subject entry overlaps with another. The only
instance where an entry begins at less than two bars distance from the
beginning of the previous one occurs in m. 21, i.e., in connection with the
one entry in the fugue that omits the final resolution. Thus the process one
is hearsas opposed to what one sees in the scorecan be said to be
concluded by the time the subsequent statement enters.
Bach invents one counter-subject for this fugue. It is introduced against
the subject answer in mm. 3-5. Beginning two eighth-notes later than the
subject, the counter-subject also displays two subphrases separated by a
change in pitch level (see m. 34). The phrasing occurs one eighth-note after
that in the subject. Moreover, in its first appearance the counter-subject
even ends with a metrically delayed resolution, i.e., one eighth-note after
the subject (m. 51 E-D). Besides this similarity in phrase structure, the
counter-subject is also
related to the subject
in its pitch pattern as it
shows an overwhelming participation of
seconds and a scalar
ascent in the second
subphrase. Its dynamic
306
WTC I/23
M1
M1a
M1b
B major
307
This is certainly the impression we get when we play the voice alone. It is also what any
string or wind player, performing this fugue in a quartet, would feel. Only pianists, busy
with figures in other parts but still in their own ten fingers, occasionally ignore the singlevoice demands, with the excuse that there is already enough happening at this point.
308
WTC I/23
often neglected, with dynamic shading that reflects the decrease of tension
in the subject at this moment. The problems that arise with the trill in this
fugue are caused by several irregular endingsthose that were earlier
mentioned as statements with delayed resolution. In all of these cases,
the trill begins in the same manner but ends prematurely and without a
suffix, stopping short on the last main note before the bar line.2 To provide
performers who now decide against playing this fugue (just because of the
trills) with an incentive for trying, the following examples give the most
prominent occurrences spelled out, with some suggestions for appropriate
fingering in the trickier cases:
mm. 2-3 (similarly in mm. 4-5, 8-9)
mm. 30-31
mm. 12-13
mm. 32-33
mm. 6-7
mm. 19-20
The only statements that forgo the trill altogether are those with omitted resolutions or
varied endings, i.e., those in mm. 17, 21, and 25.
B major
309
When trying to determine how this fugue is structured, one cannot rely
on any of the data that normally guide such an analysis. Except for the
final measure, there are no explicit cadential formulas outside the confines
of the subject statements. The only obvious cadential-bass patterns mark
the endings of the fifth and sixth statements respectively. They thus follow
one another too closely to indicate section endings. The texture is unusual
insofar as only two among the twelve subject statements sound in full
ensemble. All other entries either feature one resting voice or maintain the
full four-part texture only for a short span. The harmonic design, too, is
atypical insofar as no noteworthy modulation takes. Except for one entry in
the subdominant, the statements alternate regularly between the tonic and
the dominant positions of the home key.
The only clues to determine the structural layout of the composition are
thus the analogies observed earlier in the design of the episodes. These
analogies can now be expanded:
4 consecutive subject
4 consecutive subject
statements (mm. 1-9)
.
statements (mm. 18-26)
(E3 regarded as an extension)
E1 (mm. 9-11)
.
E4 (mm. 26-29)
2 additional statements .
2 additional statements
interrupted by E2
.
rounded off by E5
As both the fourth statement in m. 9 and its counterpart in m. 26 close
with a perfect cadence while the ensuing episodes feature open endings
leading into the entry following next, we may assume that these episodes
open sections rather than close them, as in so many other fugues. The
result of these comparisons leaves us with four sections, of which
the first and third each consist of four consecutive subject statements and end in a perfect cadence (m. 9: F major, m. 26: C major),
the second and fourth each show an opening episode of analogous
material and design, two subject statements, and another episode
(incorporated into the second section but closing the fourth).
Concerning the dynamic design valid in the sections of this fugue, two
basically different patterns can be distinguished. The first and third sections
consist exclusively of subject statements. Within the first section, the
tension rises gradually along with the usual increase in the number of voices.
Owing to the fact, already mentioned above, that this section does not truly
establish the expected four-part texture, the growth should be restrained in
such a way as to avoid the sensation of a powerful climax. In the third
section, the two initial statements appear in inversion. Given the particular
shape of this subject with its two rising motions now converted to falling
310
WTC I/23
lines, the inversions sound much less cogent and self-confident than the
original. As a result, the four entries of this section also give the impression of curbed tension.
The second and fourth sections feature alternations of episodes and
subject statements. The decisive musical message in these sections is that
of color contrast. A very light and delicate touch in the episodes, conveying both a playful character and melodic openness, is set against a much
more assertive touch in the subject statements. In contrast to the episodes,
the statements are clearly directed toward their goal and assuredly closing.
Performers who wish to weigh the two entries in each section against each
other will find that in the fourth section, the final statement surpasses the
preceding one, both because of the rise in these consecutive entries from
the alto to the soprano and because of the temporary four-part texture. In
the second section, however, where the statements are separated by a
substantial episode, the question of any dynamic relation between the
statements seems beside the point.
312
WTC I/24
B minor
313
The preludes first phrase comprises two group statements: the initial
middle-voice statements of M1 and M2 and their stretto imitations, starting
at half a measures distance in the upper voice, in which the final note of
M1 melts with the beginning of M2, and an M1 entry in the upper voice of
mm. 4-5 imitated now in the middle voice (ending rhythmically varied). A
further M1 statement, this time with a rhythmically varied beginning, can
be recognized in M: mm. 61-71. Its imitation in the upper voice bends its
final note back upward and thus creates a do-si-do formula. The second
phrase features a string of ascending M1 sequences in the upper voice that
follow one another in such a way that the final note of one simultaneously
serves as the beginning of the next. All are imitated after half a bar. The
same process can be observed, with the middle voice in the lead, in the
third phrase. After the perfect cadence in m. 16, the above-mentioned
extension to the imperfect cadence completes the section.
In the preludes second repeated section Bach introduces seemingly
new material. All of it is, however, related to the two earlier motifs. M3 is
a rhythmic variation and diminution of M1. Introduced in the middle voice
like the two earlier motifs, it consists of three eighth-notes leading to a
strong beat. The pitch pattern contains an ascending fourth followed by
descending steps. This motif reigns, extended through a few linking notes
in each voice, in the short fifth phrase of the prelude. M4 enters after the
D-major close in m. 21. It is rhythmically close to M3 with three eighthnotes leading to a strong beat, but its pitch pattern points in a single
direction, thus revealing its relationship to M2. The motif is introduced in
ascending direction in m. 21, but later recurs also in descent as, e.g., in
mm. 23 and 26. After a closely knit pattern with alternations of M4, M3,
and M2, the middle segment of the fifth phrase displays a pattern in
complementary rhythm with half-notes and syncopated half-notes (see
mm. 24-26) that closes with M2. The closing formula of this phrase, as
well as that of its modulating extension, presents yet another motif, made
up this time of the three ascending eighth-notes from M4 and a diminution
of the first motifs variation that sounds like the do-si-do formula. The new
motif, here called M5, also serves to indicate a cadential close (see M5 in
mm. 26-27 and 28-29, middle voice).
The sixth phrase, beginning after m. 293, sets out with motifs that are
by now familiar: in the upper voice, M2 gives way to a variation of M5
and partial sequences thereof, followed by a string of M4 sequences. The
middle voice eventually presents a sixth motif (M6 see mm. 324-333) that
is to play an important role in this phrase. Related to M2, it begins with a
syncopation followed by three eighth-notes leading to a strong beat. The
314
WTC I/24
B minor
315
a conspicuous ascent not only in the two melodic voices but even in the
lower voice, which for once participates in the sequential patterns and even
describes a melodic contour in its metrically enhanced notes (see in mm.
32-38, U: G1 to B2, M: E1 to F2). This passage presents the climax of the
whole prelude. Its position in the structural layout corresponds with the
(lighter) first climax: one occurs in the middle of phrase III, the other in the
middle of phrase VI, the third phrase of the second section.
The final phrase is not restricted to concluding function but builds up
its own little climax at m. 453. In the metrically unusual cadential close, the
tonic is reached on the middle beat, while the two upper voices display a
distinct reluctance to give in to a final release of all tension. This generates
a close that can be interpreted either as hesitant or as fairly powerfuland
be played accordingly, with either a return to complete piano or a rich,
mezzoforte ending.
316
WTC I/24
unpaired note, the B at m. 31. The second subphrase, much more concise,
comprises after the broken-chord upbeat only its target, the trill, and the
ensuing harmonic resolution.
When examining the subjects harmonic background it is crucial first
to determine the nature of the note-pairs. Slurred by Bach himself, each
pair acts out the relationship of appoggiatura-resolution. Having established
this, we know that only the resolutions are essential for the harmonic outline while the appoggiaturas create local harmonic relationships of secondary order. The example visualizes the two layers of harmonic events and
analyzes Bachs harmonization as found, e.g., in mm. 21-24 of the fugue.
B minor: i
F minor:
i V V/i7v
V/v7
V
7
V/ 7 V i iv 5 V i
V
In our search for the subjects dynamic outline, we want to look for
tension-enhancing features in each of the two subphrases before pondering
the relationship between the two climaxes. The highest degree of harmonic
tension within the initial two measures is reached in the chord marking the
modulation, i.e., in the C7 that determines the second half of m. 2. Within
this half measure, the D particularly captures attention. This note is exceptional in two respects. In terms of structure, it appears as the peak of the
ascending sequences (B-A, C-B, D-C). In terms of its scale degree, it
represents the sixth in F minor, thus serving as a secondary leading-note.
The climax of the second subphrase is the long G. This note is not only
much longer than all other values in the subject, but also incorporates the
two most essential steps of the target-key cadence: the subdominant (as a
six-five chord) and the dominant-seventh chord of F minor. Balancing the
two climaxes against each other, one discovers that the second represents
natural tension that is resolved immediately afterward, while the first
expresses artificial tension that, owing to the structural cut between B and
C, is released only indirectly. On a higher structural level one could thus
claim that the first subphrase creates a tension that is resolved in the
second subphrase.
This fugue contains thirteen full subject entries. Another seven, marked
with asterisks in the following table, are incomplete:
B minor
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
317
mm. 1-4
mm. 4-7
mm. 9-12
mm. 13-16
mm. 21-24
mm. 30-33
mm. 34-35
mm. 35-36
mm. 38-41
mm. 41-42
A
T
B
S
A
T
A*
S*
B
S*
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
mm. 42-43
mm. 43-44
mm. 44-47
mm. 47-50
mm. 53-56
mm. 57-60
mm. 60-63
mm. 69-70
mm. 70-73
mm. 74-75
A*
B*
T
B
T
B
T
T*
B
A*
See, e.g., mm. 9-12 where an inversion of CSa appears in the currently highest voice
(which is the alto). The tenor follows with CSb + CSc. Similarly in mm. 13-15: an inversion of CSa is heard in the tenor, CSb + CSc in bass.
318
WTC I/24
The incomplete subject entries have here been counted among the essential subject statements, both because of their material and because they appear accompanied by countersubject segments. This is, of course, a matter of interpretation. It is theoretically equally
possible to regard these incomplete statements as subject-related episode material. This
would lead to slightly different results in the counting of the episodes:
E4
mm. 24-30
E6
mm. 41-44
E11
mm. 73-74
E5
mm. 33-38
E10
mm. 63-70
E12
mm. 73-76
B minor
319
320
WTC I/24
B minor
321
The only ornament in this fugue is the trill on the subjects penultimate
note. This trill always begins on the main note as it is approached stepwise.
After an initial note of 16th-note duration it moves in 32nd-notes (i.e.
twice as fast as the faster note values in the piece). Wherever the subject
ends with the resolution falling on the subsequent strong beat, the trill is a
note-filling one and ends with a suffix. The trill constitutes an integral part
of the subject. In this fugue as in many others, Bach does not add the
ornament symbol in later subject statements, just as he eventually stops
indicating the slurs. Contemporary performers could be trusted to retain all
the subjects characteristic features, particularly articulation and ornaments,
in all further statements without renewed reminders from the composer.
Regarding long trills, the decisive word is: where musically applicable.
It is necessary to study the melodic and harmonic surroundings of a
subject-ending (and not just the technical feasibility) to find out where a
trill is needed, where it may have to be modified, and where it should be
omitted. The three-part rule of the thumb is: (a) if the originally ornamented note is resolved on time, play the original trill; (b) if the originally
ornamented note is resolved early or belatedly, play an interrupted trill
without suffix, stopping short immediately before the bar line (in case of a
delayed resolution) or before the dot (in the case of an anticipated resolution); (c) if the originally ornamented note remains unresolved, no trill is
warranted. Moreover, this fugue features two instances where a complete
execution of the ornament is thwarted when the subjects contour is crossed
by a line from another voice. (In these cases, the trill may be pianistically
impossible but would be played in, say, a string-quartet rendition.) These
cases are marked with an asterisk in the table below. The conclusions of
the trill in subject statements are thus as follows:
m. 6 (a),
m. 11 (a), m. 15 (c), m. 23 (a)*,
m. 32 (a), m. 40 (a), m. 46 (b)*, m. 50 (a),
m. 56 (c), m. 60 (a), m. 63 (c), m. 72 (a).
The structural layout of this fugue does not reveal itself as easily as
that of others. For one thing, the texture is quite peculiar: although the
fugue is written in four voices, only two of the thirteen complete subject
entries appear in four-part setting: entry no. 4 in mm. 13-16 and the final
complete entry in mm. 70-73. As the latter only confirms the obvious, i.e.,
that the fugue is about to end, only the earlier four-part entry assumes a
crucial position. Moreover, as there are no episodes ending with earcatching cadential closes, structural understanding must rely on the three
cadential closes occurring within episodes and on the usage, so striking in
this piece, of incomplete subject statements.
322
WTC I/24
This in itself would not be remarkable at all. What does appear both unusual and ingenious
is the fact that E2 displays a clearly relaxing tendencyattentive listeners will get the
impression that the exposition of the ensemble is completed after the third entry!
Correlating this observation to the fact that four-part entries are in fact an exception in this
four-part fugue, it seems as if Bach consciously created the impression of a three-part
composition with only occasionally enhanced density in chosen moments.
B minor
323
Looked at from the same angle that earlier revealed three expositions,
one now finds that sections IV and V develop what has been presented
earlier. Section IV returns to the uncluttered structure with no incomplete
entries (thus leaning on sections I and II), but imitates the entry pattern of
section III (B T B becomes T B T). By contrast, section V takes up the
abridged entries from section III as well as the beginning with M2 / M3
from section II and the four-part statement from section I.
324
WTC I/24
tonal answer). They are harmonically arranged in such a way that a return
to the tonic is nevertheless granted: the first modulates from i to v, the
second reciprocates with iv to i. The third section begins once again in the
i-to-v environment but then modulates for good, so that the remaining two
entries move from the tonic relative to the dominant relative and back
again, shading and concluding this section in the major mode. The fourth
section begins and ends on v, leaving the crucial return to the home key in
a iv-to-i modulation to the fifth section.
The dynamic outline is quite different in each of the five sections.
Section I is characterized by a build-up of tension that is gradual though
interrupted. The superimposed increase of tension occurs from the singlevoiced entry to the four-part statement. The first episode, after an initial
drop in intensity, contributes to the impression of mounting tension in its
ascending sequences. Only the second episode decreases and thus suggests
that we are dealing with a three-part fugue and a further entry would be
redundant. Yet the full four-part texture of the fourth subject statement
defies this. The section ends on a high level of intensity that, owing to the
fake entry and the fully maintained ensemble of four voices in the concluding episode segment, hardly abates before the cadence. In section II,
the protracted episodes with their descending motion precede two relatively soft statements, so that the entire section remains somewhat subdued. In
section III the apparent density of materialwith eight occurrences of the
subject headraises the overall tension level. At the same time, the
incompleteness of five of the entries and the fact that they appear accompanied by the relaxing CSd restrict the dynamic development, so that only
the final consecutive entries in T and B gain momentum.
Sections IV and V both begin with longer spans of diminishing tension.
The final entry of section IV is the only one to do without any segment of
the counter-subject. Instead it is accompanied by sequentially ascending
patterns in two voices that create intensity. Section V, having recalled
section II with its opening episode, then leaps directly to the final statement of section I. Prepared only by the incomplete tenor entry, the bass
entry in mm. 70-73 soon gains full four-part texture and thus provides the
fugue with a glorious ending. The full texture is maintained throughout the
ensuing E11 (which builds up tension in ascending sequences) as well as
the incomplete entry, and surpassed in the five-part setting of mm. 75-76.
When relating the climaxes to one another, it seems irrefutable that the
endings of sections I and V represent the overall highlights. In their midst,
section II is probably the softest, while the climax of section III is surpassed
by that of section IV.
328
WTC II/1
These eight sections thus represent, on a large scale, the steps of the
simple progression: tonic (mm. 1-5), subdominant (mm. 6-20), dominant
(mm. 20-25) and return to the tonic.
The prelude comprises two extended passages that are built in strict
correspondence to one another: mm. 54-142 . 202-284. Another analogy
established through the use of pedal notes rather than melodic resemblance
exists between the opening measures and the closing measures: compare
mm. 1-31 with mm. 32-34. We can thus condense the structural overview
into the following schema:
opening section
mm. 1-3/3-5
main section
mm. 5-14
middle section
mm. 14-20
main-section recapitulation mm. 20-28
closing section
mm. 28-32/32-34
In this texture of voices complementing one another in a seemingly
endless stream of sound, no articulation or obvious phrasing is desirable.
Instead, the sound flow should meander from one voice to the other in an
unbroken legato. An appropriate tempo has to take into consideration both
the details of the surface pattern and the underlying metric pulse. It must
be calm enough to allow for clarity in the 16th-notes and occasional 32ndnotes in the weaving lines, but not so slow as to render the quarter-note
beats imperceptible.
When describing details of tension growth and decrease, we must keep
in mind that these appear only as shadings before the backdrop of the
predominant harmonic developments. The first phrase (mm. 1-31) displays
an ornamental line falling over an entire octave. The dynamic development
follows the shape, creating a single diminuendo.
The line reads like a hidden two-part structure in
double sixths. It anticipates what later frequently
materializes as a complementary pattern of two
adjacent voices.
The second phrase (mm. 3-53) appears as a more complex reiteration of
the same process. The octave descent occurs within the first half measure
in the form of a scale that falls through the alto and bass registers. Thereafter it is extended, first imitating the more elaborate ornamental pattern of
the first phrase (see mm. 33-4) and then continuing in a cadential bass
pattern (mm. 4-53). The falling line spans more than two octaves.
Only from m. 6 onward do listeners get the impression of true four-part
texture. Secondary features emerge with timid tendencies to counteract the
overwhelming impression of descent. The tenor emancipates itself in m. 4
C major
329
with a four-note ascent (a little crescendo), the soprano, which had been
motionless for more than a measure, follows with an ascending fourth (m.
4: D-G; a smaller crescendo), and the alto adds an even weaker ascending
step (m. 4: B-C). After this, all voices blend once again into the overall
relaxation. This is completed with the weak-beat ending in the bass. The
four ascending 16th-notes are passive here, and the C at m. 54 is best
played as a pianissimo note.
The first phrase of the main section introduces a figure that recurs
several times as an active gesture: the zigzag broken-chord descent in
16th-notes that, after reaching an artificial leading-note (F), resolves
indirectly onto G (B: mm. 54-62). This short figure will be referred to as
M1. Its little tension-curve with the dynamically active beginning inspires
two similar curves in the soprano (mm. 6-7) and bass (mm. 6-7). A last,
softer curve in the soprano (mm. 7-8) leads to the final relaxation of this
phrase with a weak-beat ending at m. 82. Here, too, the ascending notes in
the tenor are passive and end in pianissimo.
The second phrase of the main section also begins with M1 in the bass,
followed by a one-measure dynamic curve in the soprano (mm. 8-9). The
end of this phrase corresponds with that of the previous phrase (compare
S + A: m. 7 with B: m. 10, and the passive ascent in T: m. 8 with the passive
descent in S + T: m. 11). The performers shaping should be geared toward
underscoring such resemblances.
The next phrase consists of two consecutive larger curves. One begins
in m. 11 with three active gestures (S, B, A) followed by a protracted
relaxation through m. 132. The other is launched primarily by the prominent
bass line that builds up some tension in mm. 13-14. The phrase ends once
more with passive gestures (here in A + T).
In the middle section, overriding large-scale pitch motions determine
the tension design. The first peak line occurs in the bass. After two active
gestures describing ascending fourths (see mm. 14-15: C to F and F to B),
it descends gradually in an ornamented line that extends through the
remainder of this middle section, i.e., through six full measures. This bass
line is joined, in the second phrase within this section, by a similarly
embellished descent in the soprano. The most appropriate interpretation,
which admittedly asks for performers with a long breath, is one that
renders the entire descent as a single gradual diminuendo.
From m. 20 onward, the recapitulation of the main section follows,
developing along the same lines as the motions described earlier. It is
complemented in mm. 28-30 and 30-32 by two short phrases featuring
fragments of M1 in the bass (see mm. 28-29 and m. 30). These are
330
WTC II/1
C major
331
ornamental rather than melodious. The two large intervals in mm. 1-2 form
consecutive leaps, thus corroborating the assumption of a lively character.
The target of these consecutive leaps, the A at m. 21, is supported in many
ways: melodically the highest pitch, rhythmically the first of two longer
notes, and harmonically the representative of the subdominant, it is further
emphasized by an ornament proper (see the inverted-mordent symbol)
before it relaxes slightly with the ensuing step downward.
If one regards mm. 3-5 as a varied sequence of mm. 1-2, seeing that
(GF)G-C-AG could become an ornamented (FE)F-(E)-DE, the subject
appears as consisting of two fairly balanced subphrases. The rest in m. 3
must then be played as an interruption before a new beginning, the
written-out inverted mordent following it (m. 3: F-E-F) being more active
than the fairly relaxed G in m. 2, and the D at m. 41 a second, softer
climax. Another view is equally possible and perhaps more conducive to
the transmission of overall unity. The entire string of 16th-notes in mm.
3-5 can be regarded as ornamental. One would then define the main
melodic steps in the subject as (GF)G-(C)-AGFE. In this case,
the gradual descent from A to E requires an uninterrupted line. The rest in
m. 3 is now perceived as tension-sustaining, after a G that sounds only
minimally softer than the preceding climax. The step A-G and the tensionsustaining rest will thus allow for a further release through the following F
and all its gradually retreating ornamental surroundings to the final E. This
interpretation thus renders the subject as an uninterrupted unit.
The subjects harmonic background does not reveal anything that might
decide the matter. The active step to the subdominant appears at m. 21
but then there never were any doubts about the overall climax anyway. The
subdominant relative, another harmonically active chord, is represented by
m. 4, followed by the dominant and the final tonic.
IV I
ii7
V7
332
WTC II/1
Apart from the interval modification in the answer (where the falling
fifth in the first measure becomes a fourth), the subject does not undergo
any changes. Neither does it appear in any stretto or parallel setting.
Against the answer, i.e., at a point where one expects a counter-subject,
the middle voice continues with a melodic line. Later in the piece, this line
recurs twice in its unabridged version (M: mm. 25-29 and 51-55); in two
other cases, only the initial measure reappears (U: mm. 9 and 39). While
independent in structure, this counter-subject is not entirely independent in
material. Its first two measures (see from D, the second 16th-note in m. 5)
are closely related to the subjects final two measures. Only its second half
contributes new components with an ascending scale, syncopation, and
closing formula. Another shortcoming of this counter-subject is the fact
that its unabridged version remains restricted to the middle voice, appearing exclusively in accompaniment to an upper-voice subject statement, and
thus lacks true polyphonic versatility.
The internal structure of this phrase poses a question that is crucial for
the fugues contrapuntal setting. Following the pitch pattern of descending
sequences in the first two measures, most performers will choose to
interpret these as expressing gradually lessening tension. After the lowest
note A, the ascending scale would then support an increase toward the
syncopation, after which the closing formula provides a relaxation.
With this dynamic outline in mind, let us return to the two options for
the interpretation of the subjects phrase structure. We shall find that the
concept described above for the counter-subject is ideally suited to balance
the concept of the subject as an indivisible phrase. In a reading of the subject assuming two dynamic curves, however,
this interpretation of the
counter-subject is unlikely as it would create
simultaneous phrasing
in both voices and thus
undercut the polyphonic
texture. Performers who
prefer a divided rendering of the subject should
thus play the countersubject as an unbroken
phrase, beginning with
an extended crescendo.
C major
333
334
WTC II/1
C major
335
the last F over to the next downbeat. No suffix is possible in this case since
the ornamented note lacks a resolution.
At this point it may be interesting to learn that Bachs first version of
this fugue (according to the Kellner manuscript) ended on m. 681. The
composer later rewrote the perfect cadence in mm. 67-68 as an interrupted
cadence and added 16 measures on a pedal note C. This extended coda
concludes the piece much more convincingly. At the same time, the fact
that it was not intended at the time of the earlier version is of great help for
a true understanding of the fugues architectonic design. This is determined by cadential formulas together with the key sequence of the subject
entries and the dynamic buildup in some of the episodes. Only the end of
the initial section may cause some doubt. In mm. 21-22, Bach presents the
first closing formula with typical features in both the upper and the lower
voices, thus creating a strong feeling of closure. The beginning of a subject
statement in the middle voice overlaps with this closing formula for an
entire measure. In this manner Bach strings the first and second sections
closely together.
Given that the ensemble is reduced to two voices after the cadence in
mm. 24-25, which would normally advocate the beginning of the second
section, one needs good grounds for a differing view. Four reasons support
the assumption that section I ends already in m. 22: As all three voices have
already presented the subject, the first section can only close here or after
an additional (redundant) entry. Yet the overlapping subject statement in
the middle voice ends with a cadential close that is not quite satisfactory
owing to a sudden break-off in the upper voice (m. 251), which creates a
strong link between the upper-voice line in m. 24 and the new beginning in
m. 25. The overlapping subject statement in mm. 21-25, while beginning
in the harmonic surroundings of G major, soon reveals its loyalty to the
key of D minor (see the Cs and Bs from m. 22 onward), thus forming a
pair with the following upper-voice in A minorthe minor dominant of
the preceding statement. A closer look at the beginning of both statements
further strengthens this reading as they display the interval structure of
subject + answer. (Compare mm. 21/25 with mm. 1/5.) If the subject entry
in mm. 25-29 were the first of a new section, then the episode that follows
would have to be regarded as linking two consecutive statements. The long
and definite tension decrease in E2, however, makes such an interpretation
very unlikely. Lastly, the dynamic design of E1 is distinctly related to that
of the original E4 (until m. 68): both describe a full curve, using similar
material.
336
WTC II/1
Having said all this, a conclusion for the remaining sections follows
without problems. The harmonically related entries in mm. 21-25 and
25-29 constitute, together with the decreasing E2, the second section. The
return to C major and to three-part texture in m. 39 marks the beginning of
the third section, which comprises three statements and a closing episode
(like the first section, the only difference being the inserted E3). The
exceptionally long coda must be regarded in the light of its relative in the
Classical period: it rounds off the entire piece, not just the final section.
In the first and the third sections, the tension rises from one entry to the
next. In the first section, this is bolstered by the gradual increase in texture;
in the third section, Bach uses for the same effect an ascent from the very
low keyboard register (mm. 39-43) and a mid-section episode with a
strongly increasing tendency. The tension diminishes in the second section
owing to the reduced number of voices in all but the first four measures,
the minor mode, and the strongly decreasing direction of the episode.
The coda begins softly but then builds up strength as the pedal moves
to the lower C and the lower-voice figures suggest hidden two-part structure.
When the other voices also split, the fugue ends in full resonance.
338
WTC II/2
The initial motif (M1) begins in a metrically weak position. The bass
note C on the downbeat serves as an anchor: it takes part in neither the
melodic lines nor the regular background features. Its tone quality is
therefore sonorous but neutral. M1, as shown in the example above, consists of a melodic descent in quarter-notes doubled in compound thirds. On
the first level of background, a 16th-note-figure in the shape of an inverted
mordent serves both as a melodic link between the notes of M1 and as an
octave ornament of its parallel. Finally, the off-beat eighth-notes in the
accompanying voice represent a repeated C that, although it gives way to
the harmonically required B at the end of the measure, conveys a strong
sense of an indirect pedal note. All features recur in inverted voices in m. 2.
Moreover, they turn out to be the constituents of the entire piece.
M2, introduced in mm. 3-4, appears as if enlarged on various levels.
The steps of the melodic line, now ascending, move in strong-beat halfnotes so that we hear pairs of repeated notes in active/passive grouping:
D-D, E-E, F-F, G-(E). The parallel in the lower voice also begins, in m. 3,
with an ornamented half-note G. Only in m. 4 do we hear chromatic steps
in quarter-note pulse (A-A-B-B-C). The indirect pedal is abandoned in
this motif. M3 in mm. 5-73 drops an additional feature: the accompanying
eighth-notes. Instead, the parallel melodic steps, falling on beats 1, 2, and
3 of each of these measures, are surrounded by the ornamental figure that
is now intensified in parallels, and complemented by a scale that overruns
the obvious two-part texture by creating the illusion of a link between the
C minor
339
upper and the lower voices. In mm. 73-83, the lower part continues the
previous pattern while the upper part emancipates itself briefly with a
repeated ornamental figure. The harmonic progression is concluded in
mm. 83-93 with a closing formula in the bass and in the melodic layer of
the upper part (E-D-E). The first section ends with a transposition of M1
(compare mm. 93-113 with mm. 1-2) followed by another cadential close in
mm. 113-12.
The second section begins with a slightly more complex pattern. In
mm. 13-14, we hear a hidden-three part texture consisting of an upper
layer with melodic quarter-notes followed by a straightforward melodic
figure and its descending sequence, a lower layer with neutrally-colored
accompanying eighth-notes followed by quarter-notes, and an ornamental
background that, here, does not move together with the melodic lines but
presents an indirect pedal on B (m. 13, right hand), followed by an ornamented G and F (m. 14, left hand). The entire pattern is then repeated in
transposition (upper voice a fifth down, lower voice a fourth up), with only
minimal adjustments at the beginning of the pattern.
In the substantial center of this second section, Bach establishes no
motifs at all. Variations of earlier features can be made out, e.g., a modified version of M3 in mm. 19-20, a remote resemblance of M2 in m. 22,
and ornamented indirect pedal-notes in mm. 17 (r.h.: C) and 21 (l.h.: F).
Yet although a minimal sense of melodic recognition is encouraged by
sequences, the predominant features are not found in the details but rather
in the large-scale lines.
The following sketch, showing an excerpt of this prelude in its skeletal
version, concentrates on these features and thus gives guidelines for
intensity shading and dynamic shaping:
After the harmonic return to C minor, the preludes last three measures
recapitulate fragments of the various features. There is a variation of M2
beginning in the left hand with an upbeat of three 16th-notes to m. 26.
They are paralleled in the middle voice in m. 26 and picked up by the
340
WTC II/2
C minor
341
342
WTC II/2
C minor
343
344
WTC II/2
In either case there are a few leaps outside the subject that require nonlegato rendering: the octave leap in m. 3 (G), the cadential leaps in the
lower voice of mm. 8-10 and 13-14, the alto on the middle beat of m. 19,
and the bass in m. 23 as well as in the three final notes in mm. 27-28.
The tempo of this fugue is moderately flowing: slow enough to allow
for full appreciation of the texture in all measures with a complementaryrhythm structure, but not so slow as to stretch the augmented statements
beyond recognition. These should still move in such a way that they could
be sung on one breath. The relative tempo of the prelude to the fugue may
be chosen in simple proportion since the two pieces are conceived in so
different a character: a half-note in the prelude corresponds with a quarternote in the fugue. (Approximate metronome settings: prelude beats = 108,
fugue beats = 54)
C minor
345
346
WTC II/2
348
WTC II/3
m. 411, and the final ten measures confirm the tonic with voice-splitting to
a four-part chord in m. 50. Here is an overview of the layout:
Prelude
I
mm. 1-3
I
(C major)
II
mm. 4-6
I-V
(modulation to G major)
III
mm. 7-13
V-ii-vi-I-IV (modulation to F major)
IV
mm. 14-20
IV - I
(return to C major)
V
mm. 21-251 I - V
(modulation to G major)
Fugato
VI
mm. 25-341 V - I
(return to C major)
VII mm. 34-411 I - V
(modulation to G major)
VIII mm. 41-50
V-I
(final return to C major)
The challenge posed by this piece is to create the greatest possible
difference between the two halveswhile at the same time conveying the
message that they belong integrally together.
Difference in character is achieved mainly by variations in touch and
intensity: In the prelude, no note, whether of 16th-, eighth-, or quarternote duration, should stand out from the chordal texture of which it is but
a part. Dynamically this means that any small-scale increase, any accented
or agogically delayed stroke automatically sounds like a pretense of melodic
independence, which is not what is wanted here. (In order to create this
non-melodic effect, performers might even strive to counteract natural
tendencies and play the soprano-ascent at the beginning of each half
measure with an imperceptible decrease.) In addition to such dynamic
treatment of note-groups, the intensity in each single note should be kept
very low. This is obtained by a combination of neutrally colored touch and
very even articulation. While the connection between the 16th-notes is
definitely legato and should not pose a problem, an even articulation in the
eighth-notes is less easy to achieve and needs more attention than pianists
are often willing to pay in the case of such secondary features. The choice
is between, on the one hand, legato (which means finger-legato in note
repetitions) and, on the other hand, a very gentle non legato (in which
skips are no more separated than steps).
In the fugato, the opposite holds true: the texture consists of nothing
but melodic features. Every note is alive in color and touch, and every line
is dynamically shaped. Articulation, too, is different: the legato in the
16th-notes is lighter than that in the prelude; unmarked eighth-notes are
well detached, and the wedge-carrying eighth-notes in the fugato-motif
(as well as in all further entries where this is not specifically indicated)
sound staccato. Unity between the two unequal halves is achieved by
C major
349
350
WTC II/3
C major
351
mm. 1
14
18
21
The subject of the C-minor fugue in vol. I has only five notes, but these expand over more
than three measures in fairly calm tempo. The subject of the E-major fugue from vol. I is
slightly shorter than this one and also fast-paced but rhythmically far more exciting.
352
WTC II/3
and simplicity seem noteworthy, even more surprising is that this brief unit
appears in stretto from its very first appearance. Listeners will necessarily
assume an even shorter subject, one comprising only the four notes that
sound unaccompanied before the second entry. Although the harmonic
layout proves that this segment is insufficient for a subject, Bach seems to
take the foreseeable misunderstanding into account when he uses this
four-note fragment several times later in the fugue to replace the entire
subject (see particularly in mm. 25-35). If the early entry of the second
subject statement takes listeners by surprise, they must be all the more
astonished to hear that the third entry (M: mm. 2-3) already uses the
inversion. Moreover, almost every subject statement is followed by a motif
that, serving as a regular extension to the entry in the particular section of
the fugue, is imitated in all voices. These motifs, which often sound against
a subsequent subject entry but are actually designed as strung behind the
preceding one, confound listeners expectations for a counter-subject.
Let us now look into details. The pitch pattern comprises a broken
chord followed by a short stepwise descent (C E C G F E). The steps of
this descent are interrupted by rests; hence the overall effect of a melodic
line in separated notes. The rhythm is simple, comprising only eighth-note
values. Rhythmic variety is provided in the course of the fugue by the
regular motifs, which introduce 16th-notes, 32nd-notes, and tied notes. The
subjects harmonic background is as simple as its melodic line. Owing to
the stretto overlap, only the minimal i-V-i materializes: the three initial
eighth-notes represent the tonic; the remaining three notes stand for the
dominant, the dominant-seventh, and the final return to the tonic. The
dynamic follows this simple design. The climax falls on G as it represents
at the same time the middle of the subject, the onset of the only non-tonic
harmony, and the target of the largest leap.
When listing the subject entries of this fugue, one can adopt different
views: Counting only the twelve entries that quote the subject in its full
lengththese will be listed below in bold faceis not good enough for a
structural explanation as it leaves the fugues second half devoid of its
subject. Statements that, although reduced to their incomplete four-note
scope, retain the original rhythm (or its augmentation or diminution) must
therefore be included; these are given below in Roman characters printed
in normal, widened, or narrowed typeface. Inversions are indicated by an
asterisk. Not included among subject statements are figures that resemble
the abridged subject but are falsified in meter or interval pattern, either
because their climax falls on a weak beat or because their contour is not
launched with the characteristic third. This still leaves 34 statements:
C major
1
2
3
4
5
6
7
8
9
10
353
mm.
mm.
mm.
m.
mm.
mm.
mm.
m.
mm.
m.
1-2
1-2
2-3
3
4-5
4-5
5-6
5
5-6
6
L
U
M*
L
U
M
L
U
M
L*
11
12
13
14
15
16
17
18
19
20
mm. 7-8
mm. 7-8
mm. 8-9
m.
9
mm. 9-10
mm. 9-10
mm. 10-11
m. 11
m. 11
mm. 11-12
L
M
U
L
M
U
L*
U*
M
L
21
22
23
24
25
26
27
28
29
30
31
32
33
34
mm.
mm.
mm.
m.
mm.
m.
mm.
m.
m.
mm.
mm.
mm.
mm.
mm.
14-15
15-16
15-16
17
18-19
19
19-20
24
25
25-26
25-26
27-28
28-29
31-32
L
M
U*
L
M*
L*
L
L
U*
M
L
L
U
L
354
WTC II/3
C major
355
356
WTC II/3
The rise and fall of tension in this fugue describes a regular pattern.
Each section begins with fairly high intensity. The intensity is slightly
higher in the second and fourth sections than in the third because of the
additional false entry with Ms, and higher even in section V because of the
particular density and the augmented statement. In each section, the tension
then decreases. In section I tension declines sharply after the stretto, in
section II it subsides gradually through the diminished entries, and in
section III it abates even more gently through two strettos with incomplete
entries. In sections IV and V the decrease in tension seems to follow that in
section I, but is then prolonged by an extended stretch in moderate
intensity. Sections II and III end in fairly soft tones, while sections I and
IV present material up to the very end and thus do not create so much
relaxation. Only the extended cadential close in section V provides a
complete release.
The harmonic layout shows a small but significant difference. In mm. 33-37 the material
sounds on the subdominant. This is a function traditionally used at the beginning of Baroque
recapitulations as it guarantees closure in the tonic without the need of adjustments in the
modulation: I-V becomes IV-I. It comes therefore as a surprise that the imperfect cadence
at the end of the main theme (mm. 4-5: G-C-G) is changed (mm. 36-37: C-F-G7). The
transposition of the melodic lines continues (U: mm. 56-74 . M: mm. 376-394), but the
changed harmonic circumstances allow the second statement of the main theme to enter in
the original key (M: mm. 76-116 . U: mm. 396-433). The end of the phrase is once again
harmonized differently: m. 111 represents V/V while m. 431 is a simple imperfect cadence.
Adding the observations regarding the differently harmonized phrase endings to the fact
that the voices carrying the main theme in exposition and recapitulation are exchanged,
we discern a cross relationship in the symmetrical portions: mm. 1-5 corresponds with mm.
39-43 insofar as in both, the main theme is sounded in the upper voice and harmonized as
V-I-V. Conversely, mm. 7-11 correspond with mm. 33-37 insofar as in both, the main
theme appears in the middle voice and the phrase ending is harmonized as V-I-V/V7.
357
358
WTC II/4
C minor
359
360
WTC II/4
C minor
361
Finally, there are several long trills. Three of them occur in a chain
and, uncomfortably, in the left hand (mm. 14-15, 19-20, and 59-60). In all
three instances, the second trill is approached stepwise and thus begins on
the (prolonged) main note. The same holds true for the first trill in m. 19.
The other two trills embellish chromatic progressions, thus also suggesting
emphasis on the main notes. As a result, all trills may begin on the main
(16th-) note, then shake four times in 32nd-notes with the upper auxiliary,
and end in a suffix as indicated. The three remaining trills, occurring in U:
m. 31, U: m. 54, and L: m. 50, have exactly the same shapeexcept that
the latter, very softly, shakes twice as long.
The complex question of ornamentation in this aria-style prelude duly
addressed, let us turn to the works layout. The first section (mm. 1-171)
comprises three phrases and a free continuation (codetta). The leading
features are the main theme with its imitation, the contrapuntal accompaniment with its later free variation, and an interlocked motif (M1) with
its own contrapuntal line. The main theme includes three subphrases.
The first is one measure long, with a climax on the appoggiatura on m. 21.
The second subphrase consists of a 3/16-note ascent preparing the climax,
and a subsequent double-note repetition providing some relaxation. The
third subphrase begins like a varied sequence: it reaches its climax in an
immediate leap, its double-note repetition is rhythmically extended, and a
one-measure tail complements the phrase. Of the three climaxes, either the
first or the second can be played as prevalent in the theme. Their metric
placement is intriguing as it creates the impression of hemiolas: they fall
consecutively on beats 1, 3, and 2. After the main climax, the impression
of a gradual decrease should prevail, and the tail is best kept free of any
further accent. The contrapuntal accompaniment to the main theme
contains two subphrases: one ends on the long B in m. 2 after climaxing on
the tied C, while the second is characterized by a protracted decrease after
a climax on the appoggiatura (m. 31). A third voice fills in the texture in
mm. 1-31 but displays more independence thereafter. It creates a tensionincrease toward the appoggiatura (m. 41) before it resolves together with
the other two voices.
On the occasion of the second main theme statement, the contrapuntal accompaniment is strongly varied. With only its second climax
retained (m. 91) it describes a simple curve. The third voice keeps a low
profile during the statement but then adds a shortened imitation of the third
subphrase from the main theme (see U: m. 10-12). The first independent
motif (M1) is introduced in mm. 5-7. Its two subphrases both peak on the
downbeat-appoggiaturas (U: mm. 61 and 71). The accompaniment in the
362
WTC II/4
lower voice (M1a) retains many features of the main theme, particularly
the division into two subphrases and the similarities at the beginning of
each subphrase. The climaxes thus fall on mm. 52 and 62 respectively.
To distinguish the free development in the codetta from the preceding
tight-knit phrases, the dynamic lines should be kept as simple as possible
there. A convincing solution is a long and gradual build-up (from L: m. 11
and U: m. 12 respectively) that peaks in m. 15 and is followed by a
relaxation up to the close of the section at m. 171.
The second section begins, over an accompaniment launched once
more from an ascending broken chord, with a second motif (M2). This
motif, again, contains two subphrases. The first is closely related to that of
M1 (U: mm. 5-6 .17-18), and even in the second we can discover, with
some imagination, traces of the second half of M1 (mm. 6-7 .18-19). Not
surprisingly, climaxes fall on the respective downbeats. As this section is
characterized by the immediate imitation of M2, the performers attention
will be captured by one voice at a time. From m. 23 onward, the lower
voice dominates with partial sequences of the last-heard motif statement.
Climaxes of gradually lessening intensity fall on mm. 233 and 243. The
section is rounded off with a two-measure cadential close (mm. 255-271).
The gradual decline that begins with the third imitation of M2 and
continues through to the end of the section is underpinned by a descending
peak-note line in the upper voice (see U, mm. 21: F, 22: E, 23: D, 24: C,
25: B, 26: A-G-F, 27: E).
The third section begins similarly with a motif in three-part imitation
(M3, see U: mm. 27-28). While M2 recalls M1, M3 appears as a condensation of M1a. Like the second subphrase in M1a, M3 reaches its climax
after the broken-chord ascent on beat 2 and continues in a single unbroken
decrease. As in the previous section, the third motif statement is taken up
in two partial sequences (L: mm. 30-321) and rounded off by a short
cadential close.
It has been mentioned earlier that a large portion of section IV, mm.
33-451, is conceived as a (somewhat irregular) recapitulation of the
thematic passage of section I, and that another passage (mm. 50-55) very
faithfully takes up section III. Between the two, the lower voice retrieves
the habit achieved in the two preceding sections and continues with partial
sequences (see L: mm. 45-481). The other two voices recall fragments of
earlier motifs in variation (e.g., M: mm. 45-46 from M1, partly sequenced
but differently varied in mm. 47-48). A descending peak-note line in the
upper voice creates once more the effect of gradual decline (U mm. 44: A,
45: G, 46: F-F, 47: E-E, 48: D-C-B-B, 49: B-A-G-F-E, 50: D).
C minor
363
After the weak-beat cadential close in m. 56, the seven final measures
represent the coda. As was mentioned earlier, Bach had already used this
coda in the conclusion of the first section.
The internal design of this prelude is thus far from simple. It contains
enough repetitions to serve the intended mood of calm expressive power,
and enough variation to provide for constant surprises.
section
I
II
III
IV
Coda
mm.
1-17
17-27
27-33
33-56
56-62
material
m.th.
M2
M3
m.th.
M3 codetta
M1
M2
M3
M1
M3
m.th.
M2
M3
m.th.
M3
codetta sequences sequences M1
sequences
close
close
sequences
364
WTC II/4
Four of the subject statements are inverted, but none appears in stretto
or parallel. The subject receives a real answer without interval adjustments.
Other modifications, too, occur only rarely. They never affect shape but
only tonality and materialize only toward the end of the fugue. In m. 53,
the thematic fifth interval is diminished to a fourth, and the two notes after
m. 541 are shifted one tone up. In m. 55, Bach introduces an additional
accidental on the note before the climax, and in m. 62 he raises the final
note to the major third. The most dramatic effect among these small alterations occurs in mm. 67-68 where the beginning of the subjects answer
sounds for once not in G major but on the fifth degree of C minor and the
perfect-fifth leap is significantly altered to a diminished fifth. As this
subject entry also picks up alterations heard separately before (the raising
of the note before the climax and of the final note; see B m. 68-69), the
final entry of the fugue sounds harmonically somewhat eccentric.
C minor
365
The only counter-subject Bach invents for this fugue comes with two
surprises. It develops as it goes, passing through several entries and
various attempts before reaching what can later be recognized as its final
shape. Moreover, this component does not feel confined to the passages
reserved for the primary material. Instead, three counter-subject statements
invade the long subject-free passage in the middle of the fugue. They behave like entries in a section and have thus led some analysts to assume a
fugue with two subjects. While we can appreciate that the three countersubject statements in mm. 35-39 create this effect, we must acknowledge
that this thematic phrase was heard before. It is not presented in an independent exposition here (or anywhere else, for that matter).
In its final shape (L: mm. 30-31, U: mm. 35-36, L: mm. 48-49, U:
mm. 55-56, L: mm. 61-62, U: mm. 66-67, and M: mm. 68-69), the countersubject describes a continuous diminuendo. The first four notes are melodic,
whereas the final leaps give the impression of a cadential-bass pattern.
Comparing the other statements with this version, we find the following
variations. The first statement of CS (L: mm. 2-4) enters late, shortening
the initial note. This note is tied and thus launches a diatonic descent
instead of the chromatic one characteristic of the final version. The high
pitch Fneeded here to complement the harmonyreplaces the expected
D. The second and third statements of CS (M: mm. 17-19, U: mm. 20-21)
come close to the final version. The former still omits the chromaticism by
suspending the initial note, while the latter introduces the chromatic descent
but features a variation (a three-note group instead of a leap) toward the
end. The sixth and seventh statements (M: mm. 36-37, L: mm. 37-39)
begin a quarter-note late; in addition, the former entry omits the final note.
Owing to the harmonic modifications in the final subject entry, the partnering CS (L: mm. 68-69) begins and ends half a measure late.
The contrapuntal dynamics of subject and counter-subject are simple:
366
WTC II/4
entry and the beginning of the ensuing statement on the next downbeat.
These filler-episodes, linking what should be conceived as consecutive
entries, will be indicated below with an asterisk.
E1 m. 4
E5 m. 25*
E9 mm. 498-52
E2 mm. 68-15
E6 m. 27*
E10 m. 54*
E3 m. 19
E7 m. 29*
E11 mm. 568-60
E4 mm. 218-23
E8 mm. 318-47
E12 mm. 628-65
E13 mm. 69-71
Within these episodes, sequences of the subjects ending play a major
role; they will be referred to as Ms (motif derived from the subject).
Several other components recur frequently and therefore deserve to be
pointed out. One is the conventional melodic closing formula consisting of
a syncopated keynote followed by the leading-note and a downbeat keynote (see, e.g., L: mm. 4-5). In this fugue, this do-si-do group is frequently
used outside cadential conclusions. For convenience, it will be referred to
below as the close. Furthermore, Bach enriches the episodes with four
independent motifs. M1 is first heard in U: mm. 107-126, M2 in U: mm.
117-1212, M3 in U: mm. 13-14 (from B to B, a convex curve ending in a
syncopation), and M4 in U: mm. 448-451. The very short episodes and segments of the two very long ones are built in symmetry: E1 . E3, E5 . E6,
and E7 . E10. Moreover, within E2 and E8, mm. 6-8 . 31-33, mm. 8-13
. 39-44, and mm. 13-15 . 44-47.
The role each of the short episodes plays in the dynamic design of the
fugue follows from the direction of their sequences. Thus E1 and E3 are
decreasing, but E5, E6, E7, and E10 are increasing. All of them link consecutive subject statements. The dynamic layout of the two long episodes
is more complex. Each features a conspicuous peak-note descent marked at
its end with a close and a cadence (E2 mm. 6-9: A-G-F-E-D in U, close
in M, cadence in C minor; mm. 10-13: F-E-D-C-B-A-G in U, close in
M, cadence in E major). In these segments, the tension is diminishing. In
the final segments, however, the direction of the sequences is ascending
and, with rising tension, prepares the ensuing subject entry.
Of the remaining episodes, E4 is divided: mm. 217-227 diminish with
descending sequences in all voices, while mm. 227-23 increase again. E9 is
simple in structure, with only decreasing tendency. E11 sets off with a
cadential close in C minor (mm. 56-57), overlapping with the beginning
of an Ms stretto and a dynamic curve formed by ascending M2 sequences
followed by a cadential resolution. All this occurs over a dominant pedal.
E12 is once again divided in its dynamic shape, featuring a decrease (up to
m. 647) followed by an increase that prepares the next pair of entries.
C minor
367
Some of these melodic closing formulas are easily overlooked, so here is a list: L: mm. 4-5
and 5-6; M: 8-9, 12-13, 39-40, 43-44, and 48-49; U: mm. 19-20, 32-33, 53-54, 56-57, and
70-71.
368
WTC II/4
C minor
369
370
WTC II/4
372
WTC II/5
mm. 1-51
4 mm. opening
D majorD major
mm. 5-131 8 mm. modulation (1 fifth up) D majorA major
mm. 13-16 4 mm. confirmation
A majorA major
mm. 17-211 4 mm. opening
A majorA major
mm. 21-331 12 mm. modulation
A majorB minor
mm. 33-411 8 mm. return to the tonic
B minorD major
mm. 41-451 4 mm. modul. (1 fifth down) D majorG major
mm. 45-531 8 mm. modulation (1 fifth up) G majorD major
mm. 53-56 4 mm. confirmation
D majorD major
The character of this prelude is rather lively. This is indicated in the
broken chords and ornamental 16th-notes as well as in the way Bach
expresses his idea of meter. The dual time signature is revealing: while the
12/8 time reflects typical Baroque reluctance to accept triplets as basic note
values, the alla breve not only hints at the occasional suspension of the
compound metric pattern in favor of a simpler order, but also determines
that the main beat is to be the half-measure pulse. The dual time signature
also serves to eliminate doubt about the value of the eighth-notes: they take
up a third of a quarter-note in all patterns like that of m. 1, but half of a
quarter-note in all patterns like that in the second half of m. 2. Bach seems
to conceive all notes outside the groups of three in the alla breve pulse.
Thus wherever he uses accompanying rhythms in gigue-style, he writes
these in dotted rhythm (see, e.g., mm. 5-7, 12, 16, etc., and mm. 23-26).
These dotted-note groups have the rhythmic value that matches the triplets,
i.e., they stand for a quarter-note-plus-eighth-note combination. (Bach may
have avoided writing this way because he is using the quarter-note as a
larger unithalf of the half-note beatelsewhere and would thus have
created more confusion with two kinds of quarter-notes.)
The preludes tempo should combine two requirements: swiftness (in
the 16th-notes) and stateliness (in the quarter-notes and non-triplet eighthnotes). The triplet eighth-notes act as a link between the two. A good way
to approach them is to make it a point not to think in groups of three but in
groups of six, in accordance with the alla breve indication.
Ornaments occur in various forms. The cadential mordents in mm. 12
and 32 begin on the main note in an interpretation that stresses melodic
steps, but may begin on the upper note if a performer chooses to focus on
the virtuoso element. More melodically defined mordents like the one in
m. 23, however, must begin on the main note. The inverted mordents in
mm. 13 and 33 must fit into the local tonality. They thus touch use the
leading-notes G and A respectively. The inverted mordents in mm. 13,
14, and 15 should be transferred to the corresponding mm. 53, 54, and 55.
D major
373
374
WTC II/5
two original components of the fanfare, run and zigzagging broken chord,
are split between the upper and lower voices in m. 8, while only runs are
left in m. 9. A cadential formula closes the development in mm. 12-13.
The four measures preceding the repeat sign present a descending
sequence (compare the second to ninth eighth-note in m. 13 with the same
segment in mm. 14 and 15). The material consists of varied fragments of
the fanfare (descending runs and a three-note broken chord) along with
the drum-beathere with an ornament replacing the dotted-note group.
The fourth measure adds a renewed cadential close in the dominant.
With regard to this preludes material and form, two observations are
crucial:
All components of the thematic material, even when expressing
different colors, are derived from the initial two measures. The
fanfare may appear split into fragments of various lengths, the
homophonic cadential formula may assume different shapes, and
the assertive opening octave beat may turn into different drumbeat patterns. Yet all can be traced back to a single two-bar cell.
The overall design reveals a ternary form, with some specific attributes: The initial section is not only repeated but also recurs in
corresponding design and material. The significant difference lies
not in the inversion of voices (U/M: mm. 1-9 . M/U: mm. 41-49)
or the few melodic variations (compare mm. 13-16 with 53-56), but
in the harmonic development. The combination of a repeated
exposition that modulates to the dominant, is then followed by a
middle section, and is finally complemented by a recapitulation that
begins and ends on the tonic, suggests sonata form.
We are thus dealing here with a prelude in sonata form based on a single
thematic cellin other words: a monothematic sonata movement.
The middle section fits well into the notion of development section.
Its beginning in mm. 17-20 is based on the main motif, i.e., the
combination of octave-beat, fanfare, and cadential close. For the first
time in the prelude, the entire two-bar unit appears in inversion and at the
same time contrasted with a counterpoint (U: mm. 18-20). In mm. 21-331,
Bach develops the chamber-music version of the thematic components
as it appeared in mm. 5-141. Familiar passages with only little modification
(see e.g. U/M: mm. 21-231 and L: mm. 23-251) are set against new material
(see L: mm. 21-231 and U/M mm. 223-251). The combination is then
further developed (mm. 25-26), gives way momentarily to disguised
parallels and virtuoso figures (mm. 27-27), and ends by returning to
familiar grounds (m. 30 . m. 11, m. 31 . m. 10, m. 32 . m. 12).
D major
375
376
WTC II/5
1
2
3
4
5
6
7
8
mm. 1-2
mm. 2-4
mm. 5-6
mm. 6-7
mm. 10-11
mm. 11-13
mm. 14-15
mm. 14-16
T
A
S
B
A
S
A
S
9
10
11
12
13
14
15
16
mm. 21-22
mm. 22-23
mm. 22-24
mm. 25-26
mm. 27-29
mm. 27-29
mm. 28-29
mm. 33-34
T
S
A
B
B
S
A
T
17
18
19
20
21
22
23
24
mm. 33-34
mm. 33-35
mm. 40-41
mm. 43-44
mm. 44-46
mm. 44-46
mm. 45-46
mm. 45-46
A
S
T
B
S
A
T
B
Only one of these entries is varied: the alto statement in mm. 28-29
substitutes the syncopation with a rest and transposes the four final notes
a fourth up. Modifications of the beginning or the end do not occur. There
are no interval adjustments in the answer, and the subject is never inverted.
Strettos, however, are frequent and involve from two entries (introduced
already in the first section; see mm. 5-7) to all four parts (see mm. 44-46).
Bach does not invent a counter-subject. The fugues secondary material
derives entirely from the subject itself, thus giving this piece a singularly
dense and unified atmosphere. The subjects four-note ending (Ms for
simpler reference) appears both as a counterpart to the subject and as an
episode motif. When accompanying the subject it can appear in three
positions: as a rhythmic complement, beginning after the fourth subject
note (see mm. 3, 5, 6, 10, 21, 25, and 40), as a rhythmic parallel to the
beginning eighth-note motion (see mm. 5, 11, 14, 21, 25, and 33), or as a
stretto imitation to the subjects ending (see mm. 7, 29, 34, and 41).
D major
377
E4: ending of subject stretto (A/S) imitated by stretto 1 with A/S/A, stretto 2 with T/S/T,
stretto 3 with B/S/B/A/T, non-stretto imitation in B. E8: ending of subject stretto (T/A/S)
imitated by stretto 1 with T/A/S, stretto 2 with T/A/S/T/B/A/B/S/A+B/S/A, non-stretto
imitations in A and in S. E10: ending of subject stretto (S/A/T/B) imitated by stretto 1 with
T/S/B/A, stretto 2 with T/B/A, stretto 3 with T/A/S, non-stretto imitation in T.
378
WTC II/5
any obvious closing features, but it distinctly marks the return to the tonic
(see particularly the two-octave descending scale in the bass of mm. 38-40).
Structured in themselves, these three episodes are also the most selfcontained in their dynamic shape: E4 describes a curve, beginning with a
buildup (see the special treatment of texture in mm. 16-18) and ends with
a definite relaxation; E8, despite its descending D-major scale at the end,
strikes listeners with steady ascents (see the chromatic ascent in the bass of
mm. 35-36, followed by ascending sequences, and the rising peak-notes in
the soprano). The conclusion of this section is thus prepared as a triumph.
Conversely, E10 features only descending lines and thus completes the
piece on a soft note. A fourth group serves only to prepare or close a subject
statement.5
The subjects pitch pattern with note repetition, broken chord, and consecutive fifth and fourth intervals indicates a rather lively character, as
does the rhythm that, as mentioned earlier, is simple throughout the piece.
The alla breve time signature confirms this reading. The tempo, however,
finds its upper limit in the desired clarity of the material: the density of the
Ms strettos can be neither properly expressed nor fully perceived if the
pace is too fast. The two beats in each alla breve measure can therefore
best be imagined as generously swinging. In view of the existing contrast
in prevailing note values, the tempo proportion between the prelude and
the fugue may be simple: half a measure in the prelude corresponds with
half a measure in the fugue. (Approximate metronome settings: 92 for the
dotted quarter-notes in the prelude, 46 for the half-notes in the fugue.) The
articulation includes non legato, quasi legato, and legato. Within the non
legato we should distinguish between more definite cuts after the quarternotes (fourth subject note) and smoother interruptions after the syncopations
smooth enough to guarantee that the syncopation can truly be heard as
swallowing up the strong beat. The quasi legato, too, may sound denser
in the ending of the subject (Ms) than in the note repetitions. True legato
applies to the few 16th-notes but also, more importantly, to the closing
formulas (see U: mm. 20, 27, and 44; M: m. 50). The score does not
feature any ornaments.
The fugue consists of five sections. The development of tension inside
each section as well as in the entire fugue is determined by the occurrence
of strettos and by the tendencies expressed in the episodes.
5
E4a: a single statement of Ms serves, together with the two bass notes, to modulate from
A major to B major. E6: cadential formula in F minor, for once no quotation of Ms.
D major
379
Section I comprises the initial round of statements, the linking E1, and
E2, which concludes the section in the dominant key. The tension rises
through the subject entries but declines in E2. Section II presents two single
entries followed after the bridging E3 by a stretto of the same voices. E4
closes this section, once more in the dominant key. By unexpectedly
weakening the cadential resolution in m. 10 with a resting tenor, Bach
combines the two sections. Their union is also fortified harmonically: the
weakened A-major cadence of m. 10 is taken up in a fully-voiced close at
the end of section II. Beginning in four-part texture, this section picks up
from the raised level of the previous development. The linking episode
with its ascending sequences of Ms aids the rising motion. The growth is
furthered by the advent of the first stretto and is doubled shortly in the first
portion of the ensuing episode. Only then does the tension abate. The first
and second sections together thus build a pair not only on the harmonic
level but also with regard to the development of tension.
Section III consists of the half-bar modulation in E4b, four partly overlapping entries, the linking episode E5, and the cadential close in E6. It
relates harmonically to B minor/major and F minor/major. The beginning
launches a new tension development, enhanced by the change of harmony
and the thinning of texture. Although the sequence of the three opening
entries dramatically raises the tension level, the mood is more restrained
here than in the preceding section. The subsequent single statement and the
simple closing episode underpin the lessening of drive in this short section.
Section IV encompasses two tightly interwoven three-part strettos and
two longer episodes with multiple Ms strettos. It begins harmonically in
the dominant (stretto at the octave), progresses through the subdominant
(m. 331, also m. 351), and ends with a return to the tonic. Section V begins
with two single subject statements linked by E9. It is then crowned by the
triumphant quadruple stretto, after which its tension subsides gradually
throughout the descending lines of the final episode E10.
Sections IV + V together form a large pair, corresponding to that built
by sections I + II: Just as the A-major cadence in m. 20 completes the transitory A-major cadence in m. 10, so does the D-major cadence in the final
measures complete the return to the tonic that at the end of section IV lacks
the support of any cadential formula. Just as the four single entries of
section I were surpassed not by the two single entries at the beginning of
section II but by its stretto, so the two three-part strettos in section IV are
outdone not by the two single entries at the beginning of section V but by
its four-part stretto. The consistent decline of the pitch line in the final
episode is understandable after such a powerful buildup.
380
WTC II/5
382
WTC II/6
reached, G and not G is the proper auxiliary. In mm. 43 and 45, however,
the inverted mordents use the whole-tone step.
The falling octave in the lower voice at the beginning of the prelude
never recurs and should therefore be regarded as apart from the thematic
material. Moreover, on two strategic occasions (see mm. 5 and 26) the
descending scale is conceived as an upbeat feature beginning in a metrically weak position on the second 16th-note of a measure. One might thus
contend that the prelude begins with a kind of assertive signalthe
triple Dand only launches its motivic material from the next weak beat.
This material encompasses four distinct motifs. One of them appears with
a specific accompaniment that it retains, a second is escorted by varying
non-motivic note-groups, and the other two are introduced in stretto
imitation, thus not leaving room for any contrapuntal figure.
M1 consists of a melodic strand in eighth-notes (L: mm. 2-51) and an
accompaniment (M1a) in which a descending scale triggers a broken-chord
motion. While M1a is harmonically determined and reaches its climax on
the subdominant chord (m. 2) followed by a diminuendo until the end of
the motif, the leading part of M1 follows melodic features. The rests
together with the sequential layout at the beginning indicate sub-phrasing.
The climax of the entire motif falls on the B in m. 4, a note that is
preceded and followed by high-tension intervals (a minor sixth and a
diminished fifth respectively). In this motif m. 3, a kind of harmonic
variation of m. 2, takes part in the gradual buildup to the climax and should
therefore increase in intensity.
D minor
383
384
WTC II/6
D minor
385
mm. 38-39. In the transitional measures, the lower voice continues the
decreasing tendency while the upper voice opposes this with a stringent
increase in (mostly chromatically) rising eighth-notes.
Section III differs from the two preceding sections in almost every
respect. No motif is quoted in its entirety. Instead, the initial measure with
fragments of M1 is immediately followed by a development that combines
faint remembrances of M1 with the extended pedal concluding section I
(see A in L: mm. 43-46 and in U: mm. 47-49). Two measures of hidden
two-part structure with parallels recall the end of section II (mm. 51-52 .
38-39) and lead to four measures in toccata-style (mm. 53-56). After the
home key is reestablished in its major mode (Picardy third from m. 57
onward), the prelude ends with four cadential measures on a bass pedal D
and a final chord that, with its four voices, confirms the four-part structure
indirectly evoked in all the preludes non-polyphonic passages. Dynamically, the third section describes a prolonged decrease of tension. Caused
primarily by the absence of thematic substance, this decrease is further
supported by a descending peak-note line (see treble E: m. 43, D: m. 44,
C: m. 45, B: m. 46, A: mm. 47-49, G: m. 50, F-E-D: m. 51, E-D-C: m.
52, B: m. 53). While the toccata-like measures describe superficial waves
rather than a consequential increase in tension, the final development over
the tonic pedal musters the force to counteract the decreasing tendency and
conclude the prelude on an assertive note.
386
WTC II/6
D minor
387
One occasionally hears the very subtle version that increases tension in the first subphrase
only up to the B before resolving into the tonic on A. This is beautifulas long as it does
not lead to a similar up-down on the first beat, which would with one stroke ruin the
subphrase by making it reminiscent of the emotional waves of 19th-century music.
388
WTC II/6
The D minor fugue comprises nine complete subject statements.
1
mm. 1-3 M
5
mm. 14-15
M
2
mm. 3-5 U
6
mm. 14-17
U
3
mm. 6-8 L
7
mm. 17-18
M
4
mm. 10-13 U
8
mm. 17-18
L
9
mm. 25-27
U
None of the further statements sounds exactly like the initial one. The
second and third entries set out without major modifications, but the final
note concludes the melodic cadence even more definitely than at the
beginning of the fugue by falling back to the keynote. The fourth subject
statement breaks off immediately before the expected syncopation and,
after an octave displacement, sequences the eighth-note descent but alters
the ending and breaks off unresolved. Statements 5 and 6, as well as 7 and
8, are positioned in stretto, with the latter pair using the subject inversion.
None of these entries is complete. The fifth entry anticipates the syncopation (M, m. 15: on A instead of G) and then breaks off suddenly. Its
partner corrects the position of the syncopation (U: m. 15) but, instead of
resolving afterward and thus bringing the subject to a close, continues and
expands the chromatic descent in sequences of the rhythmic figure. In the
following stretto with inverted entries, the leader behaves like the leader of
the preceding stretto: it anticipates the syncopation, lengthens it and then
breaks off (M: m. 18). Its partner is even less complete, as it does not reach
any syncopation but gives way to another motif halfway through the
second segment. Only the statement that concludes the fugue recaptures
the original version. Yet while the syncopation is in place, the cadential tail
is substituted by a melodic closing formula.
This fugue features only one proper counter-subject. It is introduced
from m. 3 onward where it begins immediately after the middle beat. Its
end is unclear. One feels that the counter-subject was meant to end with a
closing formula AG-A. But Bach obviously decided to deviate from this
expected ending, to write G instead of G and delay the final note, which
now resolves late and into another key. (In M: m. 5, the resolution comes
after a suspension on the second 16th-note and suggests not a chord on the
dominant A but one on the tonic D. The ending on the 16th-note F is thus
the logical conclusion, although it is quite impossible to convey this in
performance as suspension and resolution overlap with the beginning of
the next motif.)
D minor
389
390
WTC II/6
D minor
391
392
WTC II/6
394
WTC II/07
of the beats appear as gentle embellishments that lack not only rhythmic
variety but also melodic intensity.
As a result, articulation is not an issue here. At any given moment, the
sound flows without interruption in at least one of the voices. As Bach
takes the trouble to write many of the triple-meter beats in the left-hand
part with rests (see mm. 3, 11, 20-23 etc.), we may assume he took it for
granted that all notes, whatever their value, would be sustained for the
entire duration. The performers concern is therefore merely to distinguish
between passages of true complementary motion, in which neither interruption nor overlapping may occur (see e.g. in mm. 245-324, 565-609), notes
to be sustained explicitly in order to transform broken chords into vertical
chords (mm. 18, 34, 37, 9-10, and 26-28), and measures in which a
subordinated rhythmic pattern needs to be observed (particularly those
with a pedal note: mm. 5-8, 13-16, etc.).
In the interest of the atmosphere the composer tries to convey, the
dynamic level throughout the prelude should remain even. Fine shades in
intensity create light and shadow or foreground and background, but strong
increases and decreases in tension are not called for. Details that may guide
performers in this delicate task can be discovered in answer to questions
like these: Does the melodic flow proceed from one eighth-note to another?
(In this case, each note carries active melodic weight.) Are subsequent
eighth-notes conceived as spread-out chords with additional rhythmic
fills or tail? (In this case, a larger number of notes may represent a
single musical unit, and may therefore have to appear under the impression
of a single gesture, with not more than one weighted note.) Do subsequent
notes represent two different melodic layers? (In such cases of hidden
two-part structure, different coloring of the parts is required.)
Several ornaments appear in this piece. The grace-notes all represent
eighth-note appoggiaturas resolving into quarter-note resolutions.1 The
inverted mordent in the final measure is straightforward while the one in
m. 50 raises second thoughts regarding the pitch of the lower neighbor
note. Despite the A in the key signature, this ornament requires an A as it
occurs in the harmonic context of G minor. The cadential mordent in m. 68
begins on the main note and contains three or, better still, five notes.
1
The question whether the dotted note should be split into one third appoggiatura and two
thirds resolution (as applies most often in J.S. Bach) or into two thirds appoggiatura and one
third resolution (as demanded by his son C.P.E. Bach in the famous treatise On the True Art
of Playing the Keyboard) does not arise here, for in all cases has the lower voice abandoned
the harmony on the third beat and a resolution at that moment would thus come too late.
E major
395
396
WTC II/07
The following simplified version of the seven phrases aims at visualizing the main features:
cadence a
motif 1
cadence a
II
rhythmic pedal
motif 1
rh. pedal
- - - - - cadence b - - - - -
III
rhythmic pedal
motif 2
IV
- - - - cadence b - - - -
motif 3
4-bar
insert
motif 2
sequences
VI
dominant pedal
VII
E major
397
Dynamic shading in this prelude is very restrained so as not to jeopardize the meditative calm of the composition. At the outset of phrases II-VI,
the three motifs describe melodic curves of one, two, and three measures
respectively, while the endings of these phrases and the framing cadences
are defined by very slight harmonically induced tension developments.
398
WTC II/07
I IV V7
I IV- - ii7
IV V V7 I
The phrase structure behind all these details encompasses three subphrases. The first coincides with the first cadential progression, thus
closing at m. 31. The second subphrase begins with the syncopated C and
closes before the sequencing syncopated B, i.e., at m. 51, and the third
subphrase consists of the complete or shortened sequence of the second. In
the light of this phrase structure, our earlier assessment of the interval
structure calls for a small revision. The perfect fourths are not actually
conceived as intervals (in the sense of steps within a melodic context) but
mark cuts between adjacent subphrases. The pitch pattern thus balances the
initial fifth and the broken chord at the end of the second and third
subphrases with stepwise motion in-between.
Determining the overall climax is not difficult. The C in m. 3 not only
stands out rhythmically and melodically, it also represents the definite turn
to the subdominant harmony. Furthermore, the gradual relaxation in the
second subphrase is echoed in the third subphrase, with a smaller climax
on the syncopated B and a more complete relaxation to the tonic. In the
first subphrase, the initial note should be felt as an implied increase of
tension (which, obviously, the keyboard is unable to produce but any
string or wind player would provide very convincingly). The subphrase
then peaks fairly gently on the B in m. 2 before relaxing toward its end.
The fugue comprises twelve subject statements. Among them, one
(marked with an asterisk in the table below) is significantly changed at
both the beginning and the end. All other subject statements are given here
in their full melodic length, including the contested tail.
1 mm. 1-7
B
7
mm. 37-43
A
2 mm. 7-13
T
8
mm. 38-44
S
3 mm. 14-20
A
9
mm. 53-59
T
4 mm. 21-27
S
10
mm. 55-59
A*
5 mm. 30-36
T
11
mm. 59-65
S
6 mm. 31-37
B
12
mm. 60-66
B
E major
399
Apart from the adjustment of the initial interval in the answer and the
overall modification in the tenth entry, the subject remains basically
unchanged throughout the fugue. Slight changes in pitch appear in mm. 31,
38, and 60 where the second interval in the tenor, alto, and soprano entries
respectively is altered to a note repetition instead of the original descending
second. The only rhythmic variation is inconsequential. It occurs in the
statements beginning in mm. 30, 31, and 53 where the initial whole-note is
shortened to a half-note.
Inversion and augmentation are not used, but strettos appear frequently.
Their structural position and the kinds of combination chosen are interesting. The four stretto settings in this fugue include pairing of the two
lower voices (T+B, mm. 30-37), the two upper voices (A+S, mm. 37-44),
the two inner voices (T+A, mm. 53-59) and the two outer voices (S+B,
mm. 59-66). In each of these pairs one entry is modified. An entry with an
interval adjustment or any other significant variation carries less weight
than one that sounds exactly like the original. The sequence of the leading
voices of the four sets is thus B S T B (from TB, AS, TA, SB).
Proceeding in the traditional way, one can expect a counter-subject
surfacing against the second subject statement (i.e., against the answer).
Within the melodic progression introduced in the bass line in mm. 7-13,
two segments can be distinguished. Their different recurrence in the course
of the fugue will tell whether or not any of the segments, each one of them,
or only the whole qualifies as a self-assured companion to the subject.
The first segment begins with a descending sequence of the subjects
tail (mm. 7-81), which it extends into a cadential close (up to m. 91). This
gesture recurs, with small variations at the end, three timesalbeit never
against the subject. Instead, it appears as an extension of subject statements
(see S: mm. 27-30, A: mm. 43-45, S: mm. 44-46, B: mm. 66-68; shortened
also in T: mm. 13-14). We must thus conclude that this figure is a horizontal extension of the subjects tail rather than a vertical counterpart in the
manner of a counter-subject. The second segment introduced against the
answer (mm. 92-121) begins and ends with the keynote (here, on the
dominant, this is B ). After an initial syncopation, the curve between the
two Bs resembles a familiar closing formula, particularly from the dotted
note in m. 11 onward. This melodic unit recurs twice, against the third and
the fourth subject statements (see T: mm. 16-19 and A: mm. 23-26; in the
latter, see also the parallel in the tenor). We can safely thus speak of a
counter-subject. The fact that it does not reappear later may be excused by
the grouping of all further subject statements in strettos.
400
WTC II/07
S
CS
The number of subject-free passages in this fugue depends on the interpretation of the subjects length, as the assessment of the subjects tail
makes all the difference. The table below is based on the following
distinction: Whenever a subject statement reaches its harmonic conclusion
only one measure before the beginning of the next statementif, in other
words, the subjects tail ends on the initial note of the following entrythe
tail and its vertical surroundings are not considered a subject-free measure,
although it must be recognized that it is harmonically defensible to hold
this view. Wherever the tail is accompanied by a motif that plays a role
in the further development of the episode, the measure is considered part of
the episode. This leaves five genuine episodes:
E1 mm. 12-13
E4 mm. 43-53
E2 mm. 19-20
E5 mm. 65-70
E3 mm. 26-30
Four of these five episodes grow out of the end of a preceding subject
statement by extending the tail in sequences: In E1, the tail and its
sequence are accompanied by a four-note motif in the bass (B-B-A-G)
that is also sequenced a third lower. In E3, the same bass motif appears,
albeit with a smaller initial interval, against the soprano entrys tail and its
sequence (see F-B-A-G in mm. 26-27, sequenced again a third lower).
After this variation of the first episode, E3 concludes with a cadential
formula in B major. In E2, the subjects tail and its descending sequence
are accompanied by partial parallels in the tenor and (descending) cadential steps in the bass. In E5, the descending sequence of the subjects tail is
matched again by descending sequences in all other voices (mm. 65-66,
66-67). The conclusion follows here with an extended cadence.
The only longer episode is E4. Its bass motif, though not immediately
related to the earlier ones, is one measure long and sequenced in descending
thirds, once in identical and once in varied shape (mm. 43-46). Thereafter
the two voices of the preceding stretto introduce a counterpoint of two
substantial motifs (M1: alto m. 46-48 G to B, M2: soprano m. 46-48 E to
E). Exchanged in cross imitation these form a threefold chain. The overall
motion is falling, accompanied by a two-measure bass figure.
E major
401
The role played by these episodes in the development of the fugue can
easily be deduced from the observations made above. E1 and E2 link
consecutive entries with a slight decline in tension each time. E3 and E5
conclude harmonic as well as structural progressions with full cadences.
E4 also creates a gradual decrease of tension but does not spell out an
explicit close, thus maintaining some connection with what follows.
The fugues basic character is determined by the variety in its rhythmic
pattern that, particularly with its manifold syncopations, needs much calm
to unfold. Meanwhile, the alla breve time signature prevents too slow a
motion. The resulting calm but flowing style is also reflected in the interval pattern, which demands an alternation of tight legato in all steps and
single skips and non legato for the characteristic broken chords in the tail
of the second and third subphrases of the subject.
As the contemplative mood of the prelude with its triplet motion differs
considerably from the duple meter in the fugue, the tempo proportion
between the two pieces can be simple: a beat (= a dotted quarter-note) in
the prelude becomes a beat (= a half-note) in the fugue. Approximate
metronome settings: 72 for the dotted quarter-notes in the prelude and the
half-notes in the fugue.
The subject with its distinct material pervades much of the fugue. Its
articulation should therefore be carefully studied. In the first subphrase,
E-B and A-G are legato. The interruption by the tension-sustaining rest
should be very smooth, giving the impression that the fully sustained
half-note reaches silently across the rest where the A picks up in almost
equal intensity. (A drop in tension after the resta mistake that occurs all
too easilycompletely destroys the expressive significance of this rest.)
Thereafter, GC and FB two measures later are separated by phrasing.
The descents C-B-A and B-A-G are legato, followed by the gentlest of
breaks for the note repetitions. The inverted-mordent figures resume the
legato, before a softly detaching non legato distinguishes the broken
chords A-C-F and G-E-E.
The obvious grouping of subject statements into three blocks makes it
seemingly easy to determine the scope of the sections, but only the first is
simple. It contains the initial statements in each of the four voices, connected by the short episodes E1 and E2 and concluded with a perfect cadence
on the dominant at m. 301. Thereafter, the introduction of strettos marks a
new section. The first stretto leads without interruption into the second,
which is followed by E4, the episode that stands out both for the significance of its motivic material and for its length. Two more strettos, equally
tightly linked, are rounded off by the final episode that closes the fugue.
402
WTC II/07
The sequence of two strettos + episode, two strettos + episode may lead
one to assume a division of these measures into two sections. The fact that
each stretto-pair features a complete round of voices would support this
view were it not for the lack of a harmonic close between the two blocks.
In m. 53 where the tenor sets in, the other voices are in the midst of their
motifs. Moreover, the B-major harmony is inflected by the D on the middle
beat and grows into an E7 chord on the next downbeat. The expected
resolution into the subdominant A major materializes only in a metrically
weak position after the second quarter-note in m. 54.
Another approach to this fugues structure is therefore to assume a long
second section, which develops from the first through overall expansion:
I = entries (B, T), E1 (link), entry (A)/E2 (link)/entry (S), E3 (close)
. II = strettos (TB, AS), E4 (link), strettos (TA, SB),
E5 (close).
In the first section, the tension rises very gradually with the increasing
number and rising order of entering voices. As both the subjects tail and
the links of E1 and E2 with their descending sequences and retroactive
harmonic steps have an appeasing quality, the buildup does not reach a
high level, and the cadential close in mm. 29-30 concludes almost as softly
as the piece began. The dynamic processes in the fugues second section
repeat those of the first section on a heightened level of intensity. The rise
of tension between the first and the second strettos is stronger because of
the increase in voices and the more drastic rise in pitch level. E4 with its
descending sequences causes a depletion of tension, but it is counteracted
by tight polyphonic interplay. The third stretto enters with slightly reduced
intensity, a fact that is supported by the texture (resting soprano) and the
choice of subject-carrying voices (T+A form the least exposed combination).
The final stretto brings the climax, with a brilliant combination of S+B
subject statements and a parallel of the two inner voices.
404
WTC II/08
D minor
1.
2.
3.
4.
5.
6.
405
mm. 1-2
mm. 2-3
mm. 9-10
mm. 10-11
mm. 11-12
mm. 12-13
U (d)
L (d)
L (F)
U (F)
U (a)
L (a)
7. mm. 17-18
L
8. mm. 19-20
U
9. mm. 28-29
L (d)
10. mm. 29-30
U (d)
11. mm. 323-333
U
12. mm. 333-343
L
13. mm. 343-353
U
M1 undergoes a number of small changes in the course of the piece. In
mm. 181, 201, 333, and 343, the descent in its second half begins not from
the octave but from the seventh, with the effect that it reaches the keynote
on the final 16th-note of the measure. Consequently, the interval to the
ensuing downbeat is modified in all cases. In mm. 18 and 20 it is raised to
a fourth, whereas in mm. 33 and 34 the transition to the counter-motif is
such that the strong-beat note is not perceived as belonging to the motif.
Bach creates two companions for M1. In keeping with the terminology
used in polyphonic compositions they will be referred to as countermotifs. CM1, introduced in mm. 2-3, plays around the keynote and fifth.
It begins with three eighth-notes (the leading-note followed by the keynote
and its octave), centers in an ornamented descent, and concludes, after a
rest, with another leading-note/keynote pair. The single climax of this
counter-motif falls on the middle beat and thus coincides (almost or
exactly) with that of the main motif. CM1 recurs several times, albeit
heavily disguised. In m. 9, the 16th-note figure begins with F-E (instead
of E-F) and ends on the octave. The ornamented descent is the same, but
the resolving final pair is omitted. In m. 12, the initial eighth-notes octave
leap is diminished to a sixth. The descent, twice as fast as before, occurs an
octave lower than expected and is followed by another leap, and the ending
(G-A) appears faster and a little earlier than it should, with unexpected
notes following. Finally in mm. 28 and 29, only the descending figure,
now even on a different degree of the scale, can be identified.
The second counter-motif (CM2) is introduced in the second half of the
prelude (U: mm. 17-18). It is characterized by a symmetrical structure in
which the second half acts as an inversion of the first. There are two broken
seventh chords in the same rhythmic pattern on beats 1 and 3 respectively
and two turn-figures on beats 2-3 and 4-1. Regarding its dynamic design,
the rhythmic details, surprising in a composition with an otherwise so
simple rhythmic pattern, demand that the climaxes fall on the two syncopations. This counter-motif thus exhibits a greater polyphonic independence
than the first. It recurs four times, in L: mm. 19-20, L: mm. 323-333, U:
mm. 333-343 and, considerably modified, in L: mm. 343-353.
406
WTC II/08
The invention features two other motifs besides M1. Naming them
creates a little problem as the first obviously derives from the main motif
and the second, though not related to the main motif, derives very
definitely from the previous one. To avoid confusion, they will appear here
with consecutive numbering regardless of their relationships. The initial
two statements of M1 are followed by M2 (see U: mm. 3-4). It begins on
the second eighth-note of the measure with a fragment of the zigzag known
from M1 and continues with a short descent also reminiscent of the main
motif. The two segments are then complemented by an eighth-note leaping
up a fourth to an ornamented peak note that is tied over. This pattern distinguishes this motif from the preceding main material. M2 is imitated in
stretto (L: mm. 33-43). The sequence of both the original and its imitation
(U: mm. 4-5, L: mm. 43-53) features a strongly varied first half and thus
severs all ties with M1. A partial sequence of the fourth leap follows (L:
second half of m. 5) and can be recognized, without its ornament, at the
beginning of the accompaniment figures in the three following measures.
With these broken chords in the lower voice Bach creates a gradual
descent that contributes to the shaping of this portion of the prelude:
E-G-C (m. 4), D-F-B, C-E-A (m. 5), B-D-G (m. 6), A-C-F (m. 7),
G-B-E (m. 8), and F (m. 9). M2 recurs, complete with imitation and
varied sequence but in inverted voices, in mm. 21-23. In addition, its first
half, the segment that is related to M1, reappears three times in mm. 13-14
in almost complete parallel motion and a further four timestwice with
neutral accompaniment, twice with a disguised parallelin mm. 30-32.
M3 also comes in two versions: as a trunk, and with an extension. The
trunk begins like M1 and M2 after the strong beat and ends on the next
strong beat (U: m. 5 E-E). Among the eight regular 16th-notes, the initial
four and the last are melodic, describing a turn-figure and, after interrupting escape notes, its resolution. (The melodic idea is thus conceived as
E-D-C-DE.) In this format M3 is sequenced (see mm. 5-61: D-D).
In three further sequences, the non-melodic inner segment is extended and
thus lengthens the motif to full-measure scope (M3a see mm. 6-71: C-C,
7-81: B-B, 8-91: A-A). The dynamic shaping of this motif takes the play
with melodic and non-melodic components into account. The first four
16th-notes sound in a dense legato of considerable intensity, the brokenchord insertion is lighter both in touch (quasi legato) and in dynamics
(diminuendo), and the final note, while soft at the end of the decrease,
picks up the more intense tone color of the melodic notes.
The prelude comprises four sections. The first introduces all motifs
with the upper voice in the lead (mm. 1-91). After the initial M1 statements,
D minor
407
408
WTC II/08
D minor
409
410
WTC II/08
where its second half is sequenced. A four-note figure that accompanies the
motifs first appearance (m. 5: B-A-A-D) also recurs (A: mm. 37-38).
M2 appears in a preliminary shape already in E4 (T: m. 27, from B to C)
but gains its full size only in E6 where it is followed by a sequence (S:
mm. 36-37). Its extension features a descending 16th-note group (S: mm.
38-39) that is imitated and sequenced throughout the remainder of E6.
Four of the subject-free passages suggest a harmonic conclusion, but in
only one of them does this conclusion coincide with the end of the episode:
in E8 the cadence concludes not only the episode but the entire piece. In
E3, the soprano sounds a traditional do-si-do figure, but this formula with
its cadential return to D major at m. 241 overlaps with the subject entry in
the alto. Similarly, the F-major cadential formula in mm. 29-30 ends on
the fourth eighth-note of the next subject entry. In E6, by contrast, the
cadential close occurs after the first 12 measures, in m. 353, while the
episode continues for another five measures, dividing this episode into two
segments (E6a and E6b).
The role each episode plays in the dynamic design is determined by its
motivic and cadential content. Thus E1, E2, and E6b all create a contrasting
color that is used as a kind of negative preparation for the following subject
statement. E3, E5, E6a, E7, and E8 wrap up a preceding entry. They bring
forth a slight diminuendo, but not with a change of color.
The rhythmic variety, together with the high content of leading notes
and other alterations, indicates a rather calm basic character. Yet as the
eighth-notes and longer note values carry the melodic line while the 16thnotes serve as embellishments, the pace should not be too slow. The tempo
proportion between prelude and fugue must be complex: Given the identical time signature in both pieces, a simple translation of the pulse would
give a dull result. One good way of relating the two pieces is to use an
imagined eighth-note triplet (instead of the actual four 16th-notes in each
beat) for the pivot: a triplet eighth-note in the prelude corresponds with a
16th-note in the fugue. (Approximate metronome settings: prelude beat =
80, fugue beat = 60.)1
The articulation in the fugue is predominantly legato. Detached playing
is only required in connection with the leaps in the episode motifs (in M1
after the D and Fin A: m. 6, before and after the higher A in T: m. 5,
1
The translation is achieved by mentally continuing the preludes pulse while dividing
each quarter-note into a triplet. These triplet-beats are then regrouped in pairs and, as soon
as they are perceived as independent pulses, each pair is converted into one of the opening
eighth-notes of the fugue.
D minor
411
412
WTC II/08
Bach designs the fugues first half as a graded tension increase. The
first two subject statements are contrasted with one, the following two with
two quotations of the episode motif. The ensuing four subject entries in
uninterrupted succession heighten the tension further before it is rounded
off by a concise cadential formula contributing the expected relaxation.
The second half begins with an overlap and a stretto, thus immediately
propelling the tension upward. Three further subject entries in four-part
setting, interspersed only by half-measure episodes, retain the tension at a
high level that is only abandoned at the cadential close of this section. The
episode opening section IV provides the longest span of contrasting color
in the fugue. The section ends, after the cadential return to the home key,
with the climax in the parallel subject statements.
414
WTC II/09
Note that there is no immediate modulation. In mm. 5-6, B major is still the dominant of
E major. B major as a new tonic is only established after the F7 chord in m. 7.
2
E major
415
416
WTC II/09
E major
417
E7
B
M1 (2 + 2 mm.)
theme (4 mm.)
M5 + cadence (1 + 1 + 1 mm.)
M6
(1 + 1 + 1 mm.)
B-C-G
C
M2 (1 + 1 + 1 mm.)
M7
(1 + 1 mm.)
E7
A7
cadence (1 + 1 mm.) B
M3 (1 + 1 + 1 mm.)
M4 (b+b+b mm.)
close (1 mm.)
= 24 measures
B
F
M8
(1 + 1 + 1 mm.)
end of theme (1 m.)
A7
M6
(1 + 1 m.)
B
M4
(b+b+b mm.) E
close ( mm.)
E
M3
(1 + 1 + 1 mm.)
B7
= 24 measures
+ cadence + M9 (3 + 3 mm.)
E
418
WTC II/09
various tied notes. The time signature is somewhat puzzling since we are
used to reading the as 2/2 time. Here Bachs indication refers to the
half-note as the relevant beat, without specifying that he is fitting four halfnotes, and not just two, into each measure. The subjects harmonic
background describes the simple progression IiiI64 V7I. The dynamic
development follows harmony (rising from the tonic to the subdominant,
falling on the way back to the tonic) as well as melody (increasing from
the keynote through ascending step and skip, and diminishing with the
descending line).
The fugue comprises 31 subject statements in a great variety of shapes
and sizes, but nevertheless in strikingly regular order.
1. mm. 1-2
B
17. mm. 26-27
S
2. mm. 2-4
T
18. mm. 27-28
A
3. mm. 4-5
A
19. mm. 28-29
T
4. mm. 5-7
S
20. mm. 28-29
B
5. mm. 9-10
A
21. mm. 30-31
B
6. mm. 9-11
T
22. mm. 30-32
A
7. mm. 10-12
B
23. mm. 30-31
T var 2
8. mm. 11-12
S
24. mm. 31-32
S var 2
9. mm. 16-17
A
25. mm. 31-32
T var 2
10. mm. 17-18
S
26. mm. 35-36
A
11. mm. 19-20
B
27. mm. 35-36
S
12. mm. 20-21
T
28. mm. 35-37
T
13. mm. 23-24
S var
29. mm. 36-38
B
14. mm. 23-25
A var
30. mm. 37-39
S
15. mm. 25-27
B var
31. mm. 40-41
B
16. mm. 25-27
T var
E major
419
420
WTC II/09
CS3 appears closely linked to CS2; they only ever materialize together.
The third counter-subject is introduced in the bass of mm. 16-17. It
consists of four ascending semitones in a short-long, short-long rhythm
pattern, and a concluding do-si-do figure that recalls the ending of the first
counter-subject. Its recurrences coincide with those of CS2: mm. 17-19 (T)
and mm. 19-20 (A).
E major
421
active passage, in E5, the gradual buildup through the ascending sequences
of the CS1-variation begins already under the umbrella of the preceding
subject statement, and E3 and E6 create nothing beyond a cadential close.
Only in E4 is the harmonic close not only diverted into an interrupted
cadence (see m. 303: C minor instead of E major) but also overlaps significantly with the beginning of the following bass statement.
The basic character of this fugue is calm. The four half-note beats3 can
indeed be understood as fairly flowing, while on the other hand, fast
tempo for the half-measures would be incompatible both with the eighthnotes and, more importantly, with the spirit of this fugue. The relative
tempo of the prelude to the fugue is most convincing when larger units, not
smaller beats, are chosen to establish proportion. A good solution is to
make one prelude measure correspond with half a measure in the fugue.
(Approximate metronome settings: prelude beats = 100, fugue beats = 66.)
The articulation is primarily legato. Non legato applies only to quarter- or
half-notes that form cadential-bass patterns or consecutive leaps (B mm.
2-3: D-B-E; T + B mm. 7-8; etc.). Ornaments do not appear in this fugue.
The structure of this composition is determined by the introduction of
new transformations of the primary material. The concluding quality of all
episodes except for E4 supports this concept. There are five sections in the
E-major fugue.
Section I comprises four statements in ascending order followed by
E1. It concludes at m. 91 with a harmonic resolution into B major,
after a one-measure dominant pedal. Except for the unaccompanied
first statement, all subject entries are accompanied by CS1.
3
The meter and, by extension, the tempo require some pondering. This is due to the time
signature. As has already been mentioned, the (alla breve) usually stands for 2/2 time.
Thus performers may wonder whether in measures of two whole-notes each the information should be read to mean only two beats in a measure or rather count half-notes.
A helpful piece of evidence is found in Bachs Partita No.6 in E minor. In the final
version that performers use today, its gigue appears written with measures of double
duration, with the time marked by a slashed circle ( ). That this unusual metrical indication must be read as a 2/1 (= count two whole-note beats in each measure) is supported
by the corresponding time signature found in the earlier version of this partita, included in
Bachs Notebook for Anna Magdalena, where all notes of the gigue are written in double
speed (eighth-notes originated as 16th-notes etc.) so that each measure contains four
quarter-notes. The matching time signature is (alla breve), which unquestionably means
count two beats in each measure. As it is most unlikely that the final version of the gigue
with its augmented note values was not meant to retain this metric order, one may be
permitted to deduce that the symbol stands for two beats in a measure of eight
quarter-notes. By extension, the in the equally long measures of the E-major fugue from
WTC Book II should be read as count half-notes, i.e., four beats to the measure.
422
WTC II/09
Section II also encompasses one subject statement in each of the
four voices followed by a concluding episode (E2). The statements
are grouped into two strettos here, with the follower entering in
both cases on the second note of the statement presented by the
leader. The episode is longer and much more substantial in
material than the one in the first section. One could almost read this
as an attempt to make up for the shrinking of the section as a result
of the stretto overlap. The section ends at m. 161 in C minor, the
tonic relative.
Section III, too, comprises four subject statements and an episode.
As before, the entries are grouped in two strettos, with the distance
between leader and follower now an entire measure. New contrapuntal material is presented in the form of the paired components
CS2 and CS3. The section closes at m. 231 in F minor, the
subdominant relative.
Section IV introduces the variation of the subject, presented in all
four voices, which are once again grouped in two strettos. The distance between the entries of each stretto is smaller than ever before,
and both strettos together cover only four measures. Without an
episode to serve as a buffer, another transformation of the subject is
introduced: its diminution. The distribution (one diminished statement in each of the four voices) and the structure (two strettos) are
not surprising, even less so since these strettos are built by the
voices that were paired just before in the strettos with the varied
subject (mm. 23-24 S + A and mm. 25-26 B + T: subject variation;
mm. 26-28 S + A and mm. 28-29 T + B: subject diminution). The
distance in the stretto resembles that of section III, albeit in half size.
After E4, which serves as an interruption without closing, the fifth
stretto combines one voice from each of the two groups in a combination of two different versions of the subject (mm. 30-31 B:
diminution, A: original values). One could thus describe the entries
as presenting two intertwining rounds. There are 4 + 1 entries of the
varied subject and 4 + 1 entries of the diminution. The redundant
stretto combining the two different shapes is further distinguished
by the surrounding material: it is accompanied by three entries of a
diminished and inverted subject variation (mm. 30-32: T, S, T) as
well as a variation of CS1 that extends, with its sequences, into the
subsequent episode. The fourth section thus combines features from
all three preceding sections before concluding at m. 351 in G minor,
the dominant relative.
E major
423
424
WTC II/09
The first ending continues with a link to the repetition in m. 1. As a representative of the
dominant chord, this link uses the major third and the leading-note D.
2
This section-ending cadence is followed, as was the end of section III, by a one-measure
link that modulates to D major, the (major-mode) relative of the (minor) dominant.
425
426
WTC II/10
Equally analogous are the first and second sections of the preludes
first half, but here we are dealing not with literal transpositions but with
structural correspondences:
mm. 1-111
.
mm. 11-231
motif, imitation
motif, imitation + sequence
motif + 2 sequences
motif + 2 sequences
To sum up: Both halves of the prelude feature inner correspondences
between their first and second sections. These are not literal but structural;
they concern primarily the leading voice in each measure. The latter parts
of the two halves display a very close analogy. The correspondence mostly
involves both voices and uses literal transposition.
The prelude is based on a simple rhythmic pattern with a predominance
of 16th- and eighth-notes. Exceptional 32nd-notes occur only in mm. 3, 4,
12, and 22 where they represent written-out ornaments (inverted slides)
and in mm. 77 and 97 where they spell out the pitches of the trill suffixes.
The basic character of this invention is thus rather lively. The tempo may
be fairly swift; sensing whole-measure beats is preferable to counting in
eighth-notes. The corresponding articulation consists of a light non legato
for the eighth-notes, a crisp quasi legato for the 16th-notes, and legato only
for the ornamental 32nd-notes. One exception to the non legato in eighthnotes is indicated by a slur (U: m. 51). It is possible to interpret the
following m. 52 as a varied sequence, in which case one would play it with
a similar two-note slur.
There are a host of ornament symbols to be considered: mordents,
inverted mordents, turns, compound ornaments and trills. Mordents occur
in mm. 18, 20, 51, 52, and 71. In mm. 18 and 20, they form an integral part
of a melodic figure. In both cases they begin on the upper auxiliary and
comprise four notes. The mordent in m. 51 is approached in stepwise
motion and thus begins on the main note; a three-note shake is sufficient in
the swift tempo. The remaining two mordents are printed in brackets. As
the circumstances differ, the two cases must be decided independently. In
m. 52, the 16th-note figure C-B-C is itself a written-out ornament, so that
an additional mordent might cause congestion.3 Conversely, the mordent in
m. 71 appears in a typical cadential formula that, in Bachs time and style,
would always have been played with an ornament, whether or not this is
indicated. Inverted mordents appear in mm. 37-41, 92, 95, 96, 102, and 107.
In all cases, the lower neighbor notes are taken from the E-minor scale. As
3
A further consideration for omitting this mordent is that this would be the only case in the
entire prelude where a 16th-note is ornamented. All other ornaments decorate longer notes.
E minor
427
this ornament is used twice in closing formulas (see mm. 102 and 107), it
may be a good idea to play the corresponding note in m. 47 (C) with the
same inverted mordent.
Turns are found in mm. 47, 57-59, 78, 102, and 107. The first begins
on the upper neighbor B and uses the sharpened lower neighbor note G.
Structurally identical turns occur in mm. 102 and 107. Both also begin on
the upper neighbor, and the first also requires a sharpened lower note (see
m. 102: D-turn with C). The turns in mm. 57-59 all appear in the context
of scalar motion and therefore begin on the main note. Within their fivenote figures, the lower neighbor repeats the pitch of the preceding eighthnote (e.g., in m. 57: G). As this ornament is integrated into a motivic
figure that continues in sequences, it should be transferred to the second
eighth-notes in mm. 60 (L), 61 (U), 62 (L), and 63 (U). The turn in m. 78
repeats the same five-note ornament once again.
Trills are required in mm. 29-32 and, correspondingly, in mm. 33-37,
86-88, and 89-92, as well as in m. 97. The four earlier trills serve to
prolong the sound in a sustained pedal note rather than taking part in a
melodic line. They had therefore best begin on the main note and launch
their shakes in 32nd-notes right away, so that the main note and not the
auxiliary meets the melodic notes in the other voice.4 These extended trills
all end either with a tie or before a rest. None of them requires, or even
allows for, a suffix. Instead, they all come to a halt on the last main note
before the bar line. The remaining trill in m. 97 is most straightforward as
it is an ornament with melodic content: it begins on the upper auxiliary E,
shakes in regular 32nd-notes and ends with the suffix as marked by the
composer.
Compound ornaments appear in mm. 43 and 77. The former begins (as
indicated by the vertical stroke at the symbols left) with an appoggiatura
E that resolves after an eighth-note, i.e., against the B in the lower voice.
The shake that follows is that of an interrupted trill: it consists of four fast
notes and stops short in a metrically weak position. This ornament may be
transferred to the corresponding note G in m. 98. The second compound
4
18th-century style distinguished melodic and non-melodic trills: The purpose of melodic
trills is to heighten the interest of the ornamented note by repeated appoggiaturas; thus an
approach from the upper neighbor note is generally the rule. Wherever the melodic surroundings bring about a beginning on the main note, the initial note is lengthened so that all
other beats in the measure are met by the upper neighbor note, giving the impression of a
repeated appoggiatura. The purpose of non-melodic trills is usually, as in this piece, to
sustain the sound of a note that, on a keyboard instrument, would otherwise fade far too
early. In this case the sound itself, i.e., the main note, is the issue. It thus falls on the beats.
428
WTC II/10
ornament begins, as the convex curve preceding the symbol indicates, from
the lower neighbor C. As it is followed by a written-out suffix, it should
be interpreted as a note-filling trill. It therefore shakes in regular 32ndnotes (C-D-E-D-E-D-C-D).
The main motif of this invention spans two measures. It begins after
a downbeat rest and ends at m. 31. The motif can be regarded as consisting
of two symmetrical halves: the first, from G to B, contains a four-note
ascent, a skip in the opposite direction, and a final ascending step; the
second half, beginning again on G, correspondingly encompasses a
four-note descent followed by a skip in the opposite direction and a final
descending step. In terms of tension, the motifs first half presents an
increase with the climax at m. 21. The second half adds the complementary
decrease. In its complete and unvaried shape (but including inversions), the
main motif recurs fifteen times:
1.
mm. 1-3 U
9. mm. 17-19 L
2.
mm. 3-5 L
10. mm. 19-21 L
3.
mm. 5-7 U
11. mm. 21-23 L
4.
mm. 7-9 U
12. mm. 49-51 Linv5
5.
mm. 9-11 U
13. mm. 53-55 Uinv
6.
mm. 11-13 L
14. mm. 73-75 Uinv
7.
mm. 13-15 U
15. mm. 77-79 Linv
8.
mm. 15-17 U
The initial statement of the main motif is accompanied by the octave
leap E. This does not form part of the polyphonic pattern but only serves as
a harmonic support. It should therefore be played in neutral touch. Later
statements of the main motif are accompanied by various counter-motifs
(CM). The unifying feature in all of them appears in the second measure:
CM1 accompanies the climax of the main motif with an inverted slide
followed by an ascending seventh leap. Preceded by varying figures, it
occurs exclusively in the upper voice: see mm. (3)-4, (11)-12 and (21)-22.
CM2 features three ascending eighth-notes followed by a cadential step
(fifth down or fourth up) to the final downbeat. Equally preceded by
varying upbeats, it occurs exclusively in the lower voice; see mm. (5)-7,
(7)-9, (9)-11, (13)-15; also, inverted and with varied ending, in mm.
(53)-55, (73)-75. CM3, the only melodically distinct companion of the
principal motif, appears only twice, in U: mm. 17-19 and 19-21.
Each of the four inversions is followed by two partial sequences; see L: mm. 49-53,
U: mm. 53-57, L: mm. 73-77, U: mm. 77-81.
E minor
429
M1a in mm. 23-251 (U), 24-26 (L), 25-27 (U), 26-28 (L), 25-27 (U), 81-83 (L), 82-84
(U) as well as divided between mm. 28-29 + 23 and mm. 84/85 + 81;
M1b in mm. 29-32, 32-33 (U), 33-35, 35-37 (L); 85-87, 87-89 (U), 89-91, 91-93 (L);
M1c in mm. 37-39 (L), 38-40 (U), 39-41 (L), 40-42 (U), 94-96 (L), 95-97 (U) as well as
modified in mm. 93-95 (U);
M1d in mm. 57-59 (L), 58-60 (U), 59-61 (L), 60-62 (U), 61-63 (L), 62-64 (U); and
M1e in mm. 63-671 (L) and 67-711 (L) as well as inverted in mm. 67-701 (U).
430
WTC II/10
This line constitutes an important guide for the performer who might
otherwise lose sight of the whole in the face of so many enchanting parts.
It is also helpful when we set out to determine the subjects dynamic
shape. The local climaxes in the three sequencing subphrases of the beginning form a crescendo while the three syncopated climaxes initiating the
remaining subphrases describe an overarching diminuendo. The most expressive of these climaxes is the C in mm. 2-3. As the first syncopation it
brings a rhythmic surprise, as the peak of the simplified line it marks the
natural turning point of the tension, and as the bearer of the subdominant
chord it also highlights the active harmonic step. The following example
shows the combination of harmony and dynamic outline:
The scope of this final subphrase is achieved by an indirect extension. It is indirect because
it cannot be omitted without leaving the subject incomplete. However, one might wish to
try playing the final subphrase (from F in m. 4) with the eighth-notes in the first half of m. 5
transposed one note lower and the middle beat bending back and resolving onto E: the
subject might convincingly have ended here. It thus does not come as a surprise that Bach
later uses the curved triplet run (from C down to E) as a motif independently of the subject.
E minor
431
L
M
U
L
432
WTC II/10
The subjects final measure is taken up in E1a (L: 2x), E2 (M: 1x mm. 35-36), E3 (L: 2x),
E4 (U: 3x and M: 1x), as well as in E5a (M: 1x and L: 1x). The five-note descent alone
recurs separately in E1b (M: 3x), E2 (M: mm. 36-37, all voices alternating mm. 37-41),
E5a (M: m. 68), E5b (L: 2x), and E6c (U: m. 84). Another derivation, also featuring six
eighth-notes in this metric position and in scalar motion, is the ascent with varying endings
appearing in E1b (U: 3x), E5a (U/L 3x), E6a (M: mm. 78-79), and E6b (U: mm. 84-85).
E minor
433
434
WTC II/10
never once coincide, thus the problem of polyrhythm does not arise. The
dotted-eighth-note figures, however, regularly sound against triplets. As
Bachs manuscript displays his writing of note head against note head very
clearly (see, e.g., mm. 18ff), these dotted-note figures are intended to be
read in gigue rhythm, i.e., in a proportion of 2 + 1 instead of 3 + 1. While
most performers intuitively choose this reading in m. 12 and from m. 18
onward, many are unaware of having played a different, strictly dotted
rhythm in the three initial statements of the subject. Consistency requires
that a chosen rhythmic rendering of a recurring segment should be retained
throughout a composition. This means that the dotted-note figure in m. 3
should already be given the rhythmic shape it is going to have throughout
the piece. (A slightly odd though basically corresponding case occurs on
the final beat of m. 83: the 16th-note upbeat in the middle and lower voices
also makes more sense if read in triplet rhythmwith the dotted note here
substituted by rests.)
The tempo does not permit much individual variation. It must be fast
enough to convey the alla breve pace but moderate enough to allow for
transparency in the ornaments 32nd-notes. Thus unhurriedly swinging
half-notes give a good measure. The tempo proportion between the prelude
and the fugue gives best results if one measure in the preludes 3/8 time
corresponds with half a measure (a half-note) in the fugue. (Approximate
metronome settings: prelude beats = 180, fugue beats = 60.)
The score contains both thematic and cadential ornaments. The first
thematic ornament appears in the subject at m. 11. It is an inverted mordent
printed in parentheses on three occasions (mm. 1, 7, 13). The question is
not so much which notes to play but rather whether or not to play it at all.
The three consecutive markings indicate that the ornament is to be treated
as an integral part of the subject; omitting the symbol in further subject
statements was usual practice. The parentheses mark it as a later addition,
the full print as stemming from Bachs own hand. (Additions found in
copies by other hands are distinguished in the Urtext by small print.).
Performers who can technically master the required speed and middlevoice acrobatics (in mm. 30 and 50) should not forgo this embellishment.9
The second thematic ornament decorates the counter-subject. It is printed
only once, though without brackets (m. 10). Placed against a half-note, a
note that must be sustained and linked, it poses a much bigger problem.
The option is therefore to play it only onceor not at all.
9
The statement in mm. 41-47 features the mordent on the first syncopation. This seems
somewhat arbitrary and could probably be silently adjusted.
E minor
435
436
WTC II/10
All four sections describe an increase in tension from their initial to the
last statements. In sections I and II, these increases are not interrupted by
episodes, and E3 in section III with its rising tendency also maintains the
drive from one entry to the other. Only section IV reverses these facts. The
initial (subdominant) entry appears more intense than the second (tonic)
entry because of the harmonic release from subdominant to dominant and
tonic, a prevailingly relaxing attitude in E5, and the three-measure thinning
of the ensemble in mm. 702-732, which makes this subject statement the
only one in the fugue to appear in reduced surroundings after the full threepart texture has been established. The final episode then describes a full
dynamic curve analogous to those in the other three section-concluding
episodes.
Among the four sections, the second one in the major mode may be
regarded as most outgoing in character. The third section is still on an
elevated level, and only the final section returns to the minor-mode level
and intensity of the fugues beginning. The differences, however, are not
dramatic. This is another playful fugue in which the thematic material as
such captures the attention more than its manifold modifications.
438
WTC II/11
F major
439
mm.
Section I
1-4
5-71
7-91
9-111
11-143
143-161
16-171
pattern
clue notes
structure
P1
P2
P2
P1
P1
P3
P1
F-D-B-G-E-C-A-F-C + G-E
D-F-(C)-B-G-E
C-E-(B)-A-F-D
D-B-G-C-A-D
F-D-B-G-E-C-A-F-D-B + E
C/AC/C
G-E-C
complete phrase
two-measure model
two-measure model sequenced
short version, phrase 1
variation of phrase 1
closing formula
link
Section II
17-20
21-231
23-251
25-271
27-293
29-322
32-331
P1
P2
P2
P1
P1
P1
P1
C-A-F-D-B-G-E-C-G + D-B
A-C-(G)-F-D-B
G-B-(F)-E-C-A
A-F-D-G-E-A
D-B-G (parall. F-D-G) + C-A-D
G-E-C-A-F-D-B-G
A-F
complete phrase
two-measure model
two-measure model sequenced
short version, phrase 1
development of phrase 1
variation of phrase 1
link
Section III
33-37
37-381
38-401
40-411
41-431
43-451
45-471
47-491
49-511
51-541
54-56
P1
P2
P3
P1
P3a
P3a
P3a
P1
P1
P1
P1
D-B-G-E-C-A-F-D-B + D-G
E-B-C
A/EE/D
F-D-B
B / G A-F-D
D / B C-A-F
F / D E-C-A
F-D-B-G-E-C
E-C-A-F, F-D
A-F-D, F-D-B, D-B-G
E-C-A-E, B-A-E-C
development of phrase 1
development of pattern
development of pattern
link
two-measure model
two-measure model sequenced
two-measure model sequenced
development of pattern
free imitation
partial sequences
cadential close + link
Section IV
57-60
61-623
623-642
643-671
67-702
703-721
P1
P2
P3
P1
P1
P3
F-D-B-G-E-C-A-F-C + G-E
D-F-(C)-B-G
B / E D / B
E-E, G-E-C, F-D-G
B-G-E-C-A-F-D-B-G-E
F / D F / F + A-F
440
WTC II/11
the third section. The pattern establishes a model in which the initial ascent
and its imitation are complemented by a triple turn (see mm. 41-431). This
variant is referred to in the table above as P3a. It is followed by two
sequences. In both cases this pattern, like that of P2, forms a curve. The rise
and decline in P3 are slightly steeper than their counterparts in P2.
F major
441
Any decision regarding the dynamic shaping in this subject should take
the allusion to the gigue into account. In light of this virtuoso dance and
the exemplary simplicity of phrase structure that is typical of it, the three
consecutive upward thrusts (F-C, A-D, and C-F) take the lead over any
considerations of harmonic tension. The subjects outline thus consists of
three short crescendos, each surpassing the preceding one, followed by an
unbroken decrease through the octave from F to F.
This fugue features eight statements of the subject.
1
mm. 1-5
U
5
mm. 52-56
M
2
mm. 5-9
M
6
mm. 66-70
L
3
mm. 14-18
L
7
mm. 85-89
U
4
mm. 21-25
L
8
mm. 89-95
L
The subject undergoes few changes. In the tonal answer, the first leap
is modified from a fifth to a fourth. In three instances, the female ending of
the initial statement (i.e., the melodic ending on a weak beat) is substituted
by a male ending (see entries 4 and 8, mm. 25 and 89 respectively, and m.
95 where the harmonic resolution occurs on the downbeat and is followed
by a new active step in the 16th-notes.) No inversions, strettos, or parallels
of the subject are used in this fugue. In addition to the modification of the
ending, the two final statements feature further changes. The penultimate
entry fluctuates between the major and minor modes: its beginning is
surrounded by the F-minor chord but the subjects fifth note is A. This is
followed by D, which in turn gives way to D (see mm. 85-87). The same
entry also surprises with its thickened texture: mm. 86-87 contain two
chords with five and one with six parts. The final entry is expanded not
vertically but horizontally, by way of additional sequences of its head
motif (see mm. 89-93). It is further set apart from the remainder of the
composition by an accompaniment in 32nd-note runs that does not occur
anywhere else. Bach does not give this subject any regular companion or
counter-subject.
As the list of the subject statements reveals, by far the largest portion
of this fugue is taken up by passages in which the subject is absent. Only
34 of the 99 measures feature the subject, while the remaining two thirds
fall into six episodes, some of them surprisingly long.
E1 mm. 9-14
E4 mm. 56-66
E2 mm. 18-21
E5 mm. 70-85
E3 mm. 25-52
E6 mm. 95-99
442
WTC II/11
As regards this distinction: it is common to call a melodic unit a motif if it recurs in more
than one place and in more than one voice. Sequence models, by contrast, are short-lived,
usually involve more than one voice, and remain unchanged in their texture.
F major
443
444
WTC II/11
446
WTC II/12
Neutral non legato: mm. 1-4 (L), 4-8 (M), 8-16, 20-24, 28-32 (L), 52-56 (U+M), 56-58 (M).
Appoggiaturas requiring intense legato may come in parallels (U+M: mm. 1-4, 9-16, 28,
29-32, 40, and 57), grow single-voiced out of a suspension (M: m. 34 and U: mm. 36),
appear single-voiced without a marking (U: mm. 37, 38, 45, and 46, L: 49, 51, and 52, and
U+M: m. 70) or slurred (U: mm. 42 and 44). For a closing formula in legato see U: m. 39.
F minor
447
448
WTC II/12
The following table lists the preludes material in order to show what
correspondences hide behind the many variants. Strongly diverging
variants are marked with an asterisk, developments with the plus sign.
mm. 02-122 M1, M2, M1..............
inserted:
mm. 122-202 M1, M2
..............
inserted:
mm. 202-282 M3, M3/M2 ..............
mm. 282-402
mm. 402-482
mm. 482-562
mm. 562-622
mm. 622-70
f/p/f
mf
f/p
f
mf
F minor
449
450
WTC II/12
The motif is first heard with sequences in the lower voice of E4b and taken
up once at the outset of E5a, with an imitation. The sequence model, introduced in mm. 17-18, quotes the subjects note repetition in double thirds
with an ornamental extension in the upper voice against quasi-melodic
curves in 16th- and eighth-notes in the lower voice. E2b continues with
two descending sequences, one faithful, the other varied and extended. In
the structurally corresponding E3b the upper and middle voices are inverted
and the extension of the second sequence is slightly modified. Another
analogous episode is E6, while E5b is shortened in its second sequence.
The relationships between the episodes are manifold and revealing for
the structural layout. Note particularly the four larger passages that correspond with one another while being distinguished from their surroundings
by texture owing to the three-part model: E2b . E3b . E5b .E6. All have
harmonically concluding endings. These are thematically open in mm. 24
and 71 but characterized by closing formulas in mm. 39-40 and 84-85.
Performers may wish to consider setting the distinct sequence models in
the four corresponding blocks apart from the remainder of the fugue by
choosing a different color.3
E2b and E3b are prepared by analogous episode segments that extend
a preceding entry, whereas E5a consists of two corresponding halves.
Dynamically, E5a is the only episode that represents truly active increases
in the ascending sequences of the subject head. (One might have a point in
arguing that these make up for the missing third entry that the balance with
section I would require). Only E1 and E4, which are not so distinct in their
material and structure, serve as links between two subject statements
within a section. Both begin with a slight increase before giving way to
relaxation, while all other episodes represent a decreasing tendency.
The thematic leaps and broken chords as well as the simplicity of the
rhythm indicate a rather lively basic character. The tempo should be fast
enough to create the effect of a true 2/4 meter, with only one strong beat in
each measure, and to guarantee the light character in the 16th-notes. The
relative tempo of the prelude to the fugue needs careful planning. A simple
proportion of quarter-note = quarter-note (and thus: measure = measure)
may sound dull. A complex proportion, while more difficult to imagine,
gives more satisfactory results for each of the two pieces. It implies using
3
Having the second manual of a harpsichord in mind may help. On modern pianos, one can
shade the fingers soft touch in these segments by depressing the left pedal. This gives an
effect that is both convincing in terms of the structure of this fugue and in keeping with
Baroque performance practice. Exact timing of the left-foot pedal is, of course, essential.
F minor
451
a note value that does not literally occur, the triplet eighth-note, as a base:
one triplet eighth-note in prelude tempo corresponds with one eighth-note
in the fugue. (Approximate metronome settings: prelude beats = 66, fugue
beats = 100.) The appropriate articulation includes non legato for the
eighth-notes and a crisp legato for the 16th-notes. The non legato may
sound bouncing in the subject itself, slightly less assertive in the secondary
material, and particularly gentle in the sequence models.
The only ornament suggested in the score appears, in parentheses, on
the first down-beat note of the first subject entry. If played, it must be
added to all other complete entries. An ornament that is not indicated even
in brackets but which almost all performers will intuitively add is the
(four-note) cadential mordent on the dotted G in m. 84.
The structure of this fugue is very clear, as can already be seen from
the outline and function of the episodes. The four corresponding episode
segments each conclude a section. As they end with different degrees of
melodic conclusiveness, the impression arises that sections I+II and
sections III+IV form the two larger blocks of a binary design. This concept
is supported by the observation that the first and third sections contain
bridging episodes between their entries, while the second and fourth
sections do not. The following table visualizes the correspondences.
section III:
section I:
subject in L + E4 (linking)
subject in U, M, + E1 (linking)
subject in L + E2a (short opening)
subject in M + E5a (long opening)
E2b (sequence model;
E5b (sequence model;
no closing formula)
no closing formula)
section II:
section IV:
subject in U, M
subject in U, M
E3a (short opening)
The harmonic plan supports the binary layout in that both the first and
third sections set out from the tonic. The developments inside each half of
the fugue, however, are different: Section I contains three statements in the
field of the tonic, after which E2 modulates to E major (the dominant of
the tonic relative A major). The second section is held in the major mode,
with its two entries in A and E major respectively. E3a then modulates to
C minor, and E3b closes in C major (the dominant). The two statements of
the third section are both on the tonic, after which E5 ends on an F7 chord
(the dominant of the subdominant B minor). The fourth section features
entries on the subdominant and tonic and thus closes the circle.
452
WTC II/12
The dynamic development from one entry to the next is not the foremost intent in this fugue. In section I, the tension rises throughout the three
entries (the dynamic curve in the bridging episodes constituting no interruption but just a suspension) and then diminishes gradually through the
concluding episode E2. Correspondingly in section III, the tension
increases through two entries (suspended during E4) and again in the
sequences of the subject head, before it subsides gradually throughout the
concluding E5. In both cases, the harmonic ending on the dominant of the
key in which the subsequent section will set out creates further release and
thus mollifies the beginning of sections II and IV. The reduced texture
supports this, as the entry in mm. 24-28 appears manifestly (and that in
mm. 71-75 practically) in two-part setting. In both cases, the final subject
statement is placed in the middle voice, the most subdued of the three
possible positions. Any growth in tension that might be brought forward
by the increase in texture is thus annihilated, and sections II and IV remain
much less outgoing than sections I and III. The concluding episodes E3
and E6 end both halves of the fugue on a soft note.
mm.
1-16
17-44
45-56
57-67
68-75
phrase structure
phrase length
3+3+5+5
16
3 + 3 + 3 + 3 + 5 + 4 + 4 + 3 28
4+4+4
12
3+5+3
11
5+3
8
453
454
WTC II/13
Bachs dotted notes sometimes confuse players. In the early 18th century, the dot after the
note did not yet have the rigid meaning it was to acquire later, i.e., to lengthen the note by
exactly half of its value. In this era it meant a little longer, the following notes determining
in each case how much extension is implied. Whenever the actual value of the dot is only
a 32nd-note (as in mm. 1 and 12-16), the same legato applies between the prolonged note
and its complement as would be used between other melodic 32nd-notes in this piece.
F major
455
piece. The grace-notes in mm. 15, 41, 43, and 441 should therefore be
played as 16th-notes followed by eighth-note resolutions. In m. 12 the
appoggiatura precedes a dotted quarter-note. Most performers would argue
that the melodically relevant portion of this note, here as elsewhere in the
piece, is the rhythmically predominant dotted eighth-note, and therefore
render the grace-note as a 16th-note. Reading the almost identical closing
formulas of mm. 44 and 67, by contrast, is unambiguous since a specific
rhythmic rendition is established for this variant of the dosido figure:
| ).
Inverted mordents appear, as so often in Bachs music, on notes that
are thereby specifically enhanced; see, e.g., m. 12. As the figure presented
in this measure is sequenced twice in mm. 13 and 14, it is recommended to
transfer the inverted mordent to the corresponding notes. By the same
token, the ornament may also be added on the syncopations in the analogous phrases, in mm. 65 and 66. In m. 22, by contrast, the printed inverted
mordent seems less appropriate: the note it embellishes is not melodically
significant and has remained unornamented in identical circumstances just
before (compare U: m. 22 with L: m. 19). The two mordents in mm. 28 and
74 adorn typical closing formulas and begin on the main note. The first
contains only one three-note shake while the ornament in the final cadence,
appearing in ritardando, may be played with five notes.
Trills occur in mm. 26-27, 29-32, 38, 44, 52, 56, and 67. Most of them
begin regularly on the upper note; only the three trills that are preceded by
appoggiaturas (mm. 44 and 67) or approached stepwise (m. 52) begin on
the (prolonged) main note. The speed of the shake depends on ones
interpretation of the fast note values in the piece. Performers who feel that
all 32nd-notes represent written-out ornaments may decide to shake in the
same values. There is, however, good reason to regard the 32nd-notes as
fully weighted melodic notes; this is supported particularly in figures like
the one in mm. 1-2 etc. Another reason for reconsidering is the fact that the
final cadential formula features written-out 64th-notes, which should not
sound faster than the trill notes. The suffixes must in any case be rendered as
spelled out by Bach and thus as slower than the 64th-note shakes, a
common practice particularly in dances and other non-polyphonic music of
the era. The examples
show possible executions
for the trills in m. 26
(with m. 27 accordingly),
m. 32, and m. 44 (with
m. 67 accordingly).
m. 26
456
WTC II/13
F major
457
The second melodic phrase answers in the left-hand part of mm. 4-71.
It is conceived in hidden two-part structure, with a background featuring
a repeated indirect pedal on C that briefly gives way to the leading-note at
the very end of the motif, and a foreground manifested in the downbeat
notes consisting of a descending line (F-E-D-C) that is paralleled in
thirds on beat 3 of each measure. The dynamic equivalent is a diminuendo
throughout the melodic notes, while the background is best kept very soft
and evenly colored throughout, imitating as much as possible the effect of
a sustained pedal note. The right hand, in this first statement of the phrase
as always in this prelude, realizes the standard accompaniment pattern.
The third melodic phrase emerges in the left-hand part of mm. 7-8. Its
prominent feature is the combination of a descending broken V7 chord with
the harmonic resolution, represented by steps 8-5-1 also in a descending
pattern. Both the harmonic and the melodic shape point to the seventh (see,
e.g., at m. 72) as the climax of this two-measure motif. The model is
sequenced in mm. 9-10 and partially again in mm. 11-12. Together with its
two sequences it thus triggers a stepwise modulation (m. 8: F major;
m. 10: G major; m. 12: A minor).
The preludes first section begins with the original statements of the
three phrases followed by a free development of the characteristic rhythm
in which dotted-note groups with one 16th-note and three 32nd-note complements are set against each other in a quasi-contrapuntal texture (mm.
12-16). The standard accompaniment pattern is suspended during this
development. The dynamic shape is determined by ascending peak note
lines in both voices (mm. 121, 131, 141, L: A, B, C; mm. 122, 132, 142,
U: F, G, A). A reminiscence of the accompaniment pattern in m. 14,
descending lines in m. 15, and a cadential close in mm. 16-171 complete
the tension curve.
In sections II and III Bach develops the same material with multiple
variants and new combination. In section II he concludes with a cadential
formula in D minor (see mm. 44-451). Section III is rounded off with four
measures of gentle transition in which a parallel version of the standard
accompaniment pattern is interspersed with reminiscences of the very first
development in mm. 12-16.
Section IV shows traces of a recapitulation: mm. 57-601 review the
principal phrase with its original accompaniment and an almost identical
melodic line, mm. 60-651 are a transposition of the third phrase, and mm.
65-66 recall mm. 12-13, equally transposed. The section concludes with a
closing formula similar to the one heard at the end of section III.
458
WTC II/13
Similar harmonic fillers can be found in several other fugues where a delayed entry of
the counter-subject would otherwise leave the beginning of the answer unaccompanied.
F major
459
The subject features five different note values: 16th-notes, eighthnotes, quarter-notes, dotted quarter-notes, and a half-note (this is the actual
value of the upbeat, hidden behind the trill and its written-out suffix). The
pitch pattern displays mainly stepwise motion, with only a few skips that
are confined to m. 3. The phrase structure, however, is anything but simple.
Whether or not all the skips in m. 3 are in fact interval leaps in a melodic
context, or whether one of them marks the line between two subphrases, is
a question of eminent importance for the performance as well as for the
understanding of the piece. As so often, Bachs harmonization delivers the
clue. It consists of four essentially analogous harmonic gestures in a metric
organization of perfect regularity. Each weak beat represents a V7 chord
that resolves on the following downbeat. In the initial upbeat as well as in
m. 4, this V7 is instituted on the middle beat, while in mm. 1 and 2 it only
materializes on the final quarter-note and eighth-note respectively.
7
7
7
Guided by the harmonic pattern one detects that the melodic line contains an equivalent structure: a triple sequence followed by a sigh motif,
all of it wrapped in linking and embellishing notes:
460
WTC II/13
D in m. 31), after which the rising phrase ending relaxes gradually toward
a relatively soft G. Finally, the dynamic shaping of the sigh motif
follows the established pattern of upbeat / heavylight.
The fugue comprises eleven entries of the subject.
1 mm. 0-4 M
5 mm. 32-36 L
9 mm. 64-68 L
2 mm. 4-8 U
6 mm. 36-40 M
10 mm. 70-74 M
3 mm. 8-12 L
7 mm. 40-44 U
11 mm. 76-80 U
4 mm. 20-24 U
8 mm. 52-56 M
The subject undergoes few modifications. The answer does not need
any interval adjustment, and inversion, stretto, or parallel are not used. The
only variation occurs where the trill on the first subject note is replaced by
a written-out ornament in a similar pitch pattern but a different rhythm (see
m. 20). The same variation is even more disconcerting in mm. 70-71 where
the original trilled ascent appears as a parallel in the lower voice.
Bach invented two counter-subjects for this fugue. CS1, introduced
against the answer (M: mm. 4-8), has particularly close ties to the subject
since its initial sigh motif with upbeat appears like an echo of the subjects
final subphrase. This figure is then sequenced in chromatic descent, complemented by a third subphrase that culminates on a long trill before
concluding with its resolution on the following downbeat. Once introduced
the first counter-subject remains a faithful companion of the subject.
CS2 enters against the third subject statement (M: mm. 8-12). Like
CS1 it begins with a one-measure unit that is sequenced a whole-tone
lower. The twofold figure leads to a climax on the syncopation, after which
it resolves downward onto the tonic. The second counter-subject can be
found several times with small modifications (in mm. 32-36 and 64-68 it
swaps endings with CS1). In other instances variations are so significant
(see M: mm. 20-24 and U: mm. 37-40) that they make recognition quite
difficult, particularly for listeners.
F major
461
The episodes marked with one asterisk begin as follows: in the voice carrying CS1: after
the downbeat; in the voice carrying the subject: after the resolution on the third eighth-note.
At the double asterisk, the episode begins after the downbeat in the middle voice but after
the middle beat in the upper and lower voices.
462
WTC II/13
The role played by the episodes in the fugues dynamic design follows
from their different content. The color contrast in the two episodes characterized by independent motifs was already mentioned. The descending
sequences in the two longer episodes determined by sequences from the
subjects tail as well as in the shorter E6 imply a gradual diminuendo. By
contrast, their two other relatives, E5 and E7, quote the sequences in
ascending direction and thus build up tension. The cadential close of the
final measures features the rhythmic pattern and melodic formula typical
for gavottes, traits one associates not with a release but with a triumphant
close, thus concluding the fugue on a confident note.
The highly emotional quality of the subject in conjunction with the
complex rhythmic pattern determines the basic character as rather calm. In
the alla breve meter, the half-notes are gently flowing but far from hasty.
The tempo proportion between prelude and fugue can be simple as the
preludes rhythmic features guard against monotony: thus a quarter-note in
the prelude becomes a half-note in the fugue. (Approximate metronome
settings: all beats = 56.)
The articulation that corresponds with this character demands legato
for all notes in melodic context. Non legato occurs only where quarternotes form cadential-bass patterns (L: mm. 55-56, 83-84) or consecutive
leaps (L: mm. 13-14, 19-20, and 45-46, in M2). There are, however,
passages in this fugue that require a different approach. The five episodes
that develop the subjects tail display features that would normally indicate
a rather lively character: simple rhythmic patterns with only eighth- and
quarter-notes shaping lines with frequent large intervals. The leaps in
quarter-notes result from Bachs decision to use in these episodes not just
the subjects last subphrase but also the sixth that precedes ita leap that,
in the subject, divides two subphrases and thus is not really conceived as
an interval but becomes one here. The accompanying eighth-notes feature
regular written-out inverted mordent figures, another indication of lively
character. While the tempo in these episodes may certainly not divert from
that chosen for the subject-dominated passages, their material invites a
different touch quality. In the quarter-notes, lightly bouncing non legato is
interrupted only for the appoggiatura-resolution pairs; in the eighth-notes,
quasi legato alternates with tight legato in the ornamental figures.
To sum up this point: the F-major fugue thus encompasses three fairly
different colors: one is reserved for the subject-dominated passages, a
second color applies in the motivically determined episodes E1 and E3,
while a third distinguishes the other five episodes.
F major
463
One of the exceptions to the rule for trills in 18th-century music generally beginning on the
upper note concerns notes opening a new phrase. This applies in all cases where the voice that
presents the subject sets in newly (at the beginning of the piece or after a rest). Even in mm.
36 and 40, the considerable leap preceding the subject entry supports the interpretation of
phrase end and new phrase beginning, while the lower-voice figure found in mm. 31-32 and
63-64 seems to lead into the subject entry, so that phrasing here would sound artificial.
6
While Bach, apparently out of consideration for his contemporaries, omits this ornament in
the lower-voice entries and in the context of large intervals (see L: mm. 12, 36, 68, and M: m.
56), it is preferable to retain the mordent in each statement. The same holds true for the five
episodes built on the subjects tail. Bach signals ornamentation only in comfortable positions but not in double sixths and in the lower voice. Given todays standard of pianistic
technique, such inhibition is no longer valid. What is more, this mordent can contribute
tremendously to the effect of imitation. It should therefore be added on the downbeat notes
in M: mm. 27, 28, 31, 32; 57, 58, 61, and 62; U: mm. 59, 60, 63, and 64; and L: mm. 69
and 70. Note that in all cases, the mordents upper neighbor note must match the harmony
realized by the other voices, which may be different from the home key and its key
signature. Thus the mordents on D in M: mm. 271 and 571 must both begin on E, that on
A in U: m. 301 on B, that on G in U: m. 591 an A, and that on A in U: m. 631 on B.
464
WTC II/13
a redundant entry and an episode derived from the subjects tail. The third
section (mm. 64 84) is shorter and contains only three entries linked by the
latter type of episode. Harmonically, the first section is in the key of the
tonic, the entries in the second section are on the dominant, tonic, tonic
relative, and subdominant respectively, and the statements of the third
section return to the realm of the tonic.
466
WTC II/14
F minor
467
The score features three ornaments in mm. 9, 23, and 25. All appear as
mordents in brackets, suggesting that they do not stem from the main
source todays scholars regard as authoritative for Bachs intentions. If
played, all begin on the main note and comprise a simple three-note shake.
It might, however, be a good idea to consider abstaining from additional
ornamentation in a piece so rich in written-out embellishments.
The musical events in this prelude can be viewed from two entirely
different viewpoints. One approach is to trace a hidden large-scale framework that, while only the skeleton of the prelude, might provide important
orientation for the integration of small-scale events into a larger design.
The large-scale processes, together with the dynamic outlines in the main
components, are better shown visually than explained verbally.
468
WTC II/14
The other, more conventional approach traces the melodic patterns and
sequence models. The principal component, together with is contrapuntal lower voice, the filling broken chord, closing formula, and subsequent
melodic imitation in the middle voice, recurs in mm. 12-13 (imitation
shortened) and in mm. 30-31. The syncopation figure introduced in U: m.
3 reappears (always in the treble) in a large variety of pitch patterns
throughout the prelude (see mm. 5, 22, 24, 28, 32, 33, 34, 36, and 41). In
mm. 132-152, a three-part model is presented that is sequenced one note
lower in mm. 152-172 (with small variations in the lower voice) and then
followed by two contracted sequences (see mm. 172-182 and 182-192) and
a cadential close. A second three-part model is introduced in mm. 21-22
(U: from m. 212 E; M+L from m. 211). This model is sequenced one note
higher, also followed by three contracted sequences (mm. 25, 26, and 27)
and a harmonic conclusion in an imperfect cadence with fermata and
general pause (m. 29). The preludes final section, having begun once
more with the main idea, does not establish any new models but freely
alternates the syncopation figure with patterns formed from 16th-note
triplets. An interrupted cadence at m. 391 diverts a formula that seemed
headed for C minor in favor of an arrival, in one voice after another, at the
tonic with Picardy third .
F minor
469
as consisting of two similar halves, with the second slightly extended: see
upbeat ( ) / syncopation / inverted mordent / long note varied to become
upbeat ( ) / syncopation / inverted mordent / short note + tail (F-G-F).
In this light, the subject presents itself as divided into two subphrases. The
pitch pattern presents a different message. It can be described as combining a broken-chord upbeat followed by an ornamented descending scale.
This scale features three syncopated half-notes leading, after a quarter-note
inserted to correct the meter (so to speak) to a strong-beat half-note that
resolves onto the keynote: upbeat-DCBAGF. In this view,
subphrasing is inconceivable and the subject appears as one indivisible
unit.1 The subjects harmonic layout (which can be observed most easily in
mm. 16-19) emphasizes the syncopations by allotting each of them a shift
from a primary function to a secondary dominant:
There is no right and wrong. At this early point in material appreciation, the preference
depends entirely on the view adopted by each individual interpreter. Later in the fugue we
may find that other components work better with one of these concepts than with the other,
but so far, the choice is one of taste only.
470
WTC II/14
The complete subject appears ten times in the course of the fugue.
1
mm. 1-4
M
6
mm. 34-37
L
2
mm. 4-7
U
7
mm. 51-54
M
3
mm. 8-11
L
8
mm. 54-57
U
4
mm. 16-19
U
9
mm. 60-63
L
5
mm. 28-31
M
10
mm. 66-69
U
Inversion, stretto, and parallel do not occur, and the changes in the
subjects shape are minor. The characteristic adjustment of the initial interval in the answer occurs only in m. 4 as the two further subject statements
appearing in the harmonic position of an answer (mm. 54 and 60) both
feature a varied beginning. The subjects ending is modified in mm. 36-37,
where the ornamented step is replaced by a cadential-bass pattern, and in
m. 69, where the resolution of the trill is delayed.
Bach invents no regular counter-subject for this fugue. Instead, components of the subject itself appear in different combinations as contrapuntal
material.2 Nonetheless, this is by no means all there is to be said about the
primary material, for as one discovers only later, this F-minor fugue is a
triple fugue: a fugue with three subjects. The second and third subjects are
introduced separately before they are combined with the first. Moreover, as
with the first subject, several of their components appear as contrapuntal
material, eventually accompanying any of the three subjects.
The second subject makes its first appearance in L: mm. 20-22 after an
A-major cadence. Beginning thus in the major mode, its initial descent
with the characteristic dotted-note group leads to a raised fifth that implies
a modulation back to the original tonic: E, the leading note to F, leaves A
major. The ensuing fourth leap to the syncopated F seems to introduce a
traditional closing formula, and indeed this dosido figure constitutes the
ending of the second subject in several of its later statements (see e.g. M:
mm. 56-57 and mm. 62-63; L: mm. 68-69; with slight variation also L:
mm. 24-25). In the first few presentations of this second subject, however,
the resolution is omitted, replaced by a chromatic descent to E. This is tied
and resolves belatedly (outside the subjects actual scope) onto a note that
initiates a cadential-bass pattern in B major. The rhythmic pattern of this
2
See mm. 4-7. What begins like an independent voice (diatonic ascents interrupted by a
tritone leap downward and various rests) ends with a partial imitation of the subjects end.
In mm. 8-11 the secondary voices both imitate the subjects initial broken chord.
F minor
471
See M: m. 42 = fifth instead of second; L: mm. 44-45, U: mm. 45-46 and L: m. 55 = three
16th-notes instead of one eighth-note.
472
WTC II/14
F minor
473
474
WTC II/14
F minor
I
II
III
IV
475
consists of the three initial entries of S1 (the third concluding in
a closing formula), the short linking E1, the more substantial E2,
a redundant entry, and the concluding E3 whose cadential-bass
pattern confirms the modulation to the tonic relative (A major)
on the middle beat of m. 20 (S1: M U L, U).
encompasses two segments. IIa presents three initial statements
of S2 plus a redundant entry. This fourth entry introduces the
final version of the second subject with its resolving ending and
thereby concludes this segment at m. 251, still in A major
(S2: L U M, L). IIb begins with an episode featuring three
entries of the S2-derived motif. It is followed by a S2 stretto, a
S1/S2 juxtaposition, a longer episode, and an S1 entry accompanied by the S2-derived motif. The section concludes with a
cadential formula in C minor, the minor dominant, at m. 371
(S2: U M L, S1: M L).
encompasses three segments. IIIa is launched prematurely in a
one-measure overlap with the previous section. It presents the
three initial statements of S3 plus a redundant entry. This
redundant entry is accompanied by a melodic closing formula
that marks the modulation to G major, the secondary dominant
V/V (S3: M U L, M). IIIb is very short, containing only two S3
entries and a reiteration of the cadence in G major (S3: L M).
IIIc begins, similarly to the last segment of section II, with an
episode featuring the S3-derived motif. It is followed by two
further entries of S3 and a longer episode concluding in m. 52
with a melodic closing formula in the subdominant B major5
(S3: L U).
begins almost like section III with an overlap (here of eighthnotes only). Its opening S1 statement represents the subdominant
key, but the second half of m. 54 modulates back to the tonic.
There follow the three three-subject juxtapositions, linked by
two episodes and rounded off by a short cadential conclusion.
For the four S1 entries of this section Bach has chosen the same
voices as in the four statements of the first section, thus closing
a large bracket (S1: M U L, U; S2: M M L; S3: L U M).
Note that there is a misprint in the score here: in mm. 49-50, all voices cancel the G to G
except for the lower voice on the middle beat of m. 50 where the natural is almost certainly
omitted by mistake.
476
WTC II/14
As is obvious from the layout, the issue in this fugue is the separate
presentation and eventual juxtaposition of contrasting subjects. Superimposed dynamic developments are therefore not likely to be the composers
intention. While the exposition of S1 constitutes a gradual build-up of the
ensemble from single-voiced to three-part texture, the lack of enhancing
contrapuntal features limits the dynamic development. The expositions of
S2 and S3 present a similar picture. Section IV is the most intense in the
fugue. This applies to the entire section rather than to any single entry. The
fugue thus presents itself as a composition with four sections in four
different shades of emotional involvement. The only color contrasts within
each section are provided by the episodes.
See, e.g., the written-out inverted mordent on the first beats of mm. 7-11. If played in the
same touch as the surrounding 16th-notes, the middle notes of these ornaments might appear
as belonging to the melodic line in the hidden two-part structure, while the melodically
relevant tone is actually the main note of this mordent, the measures second 16th-note.
2
477
478
WTC II/15
four notes and must be played fast particularly in front of the octave
displacement of its resolution (e.g., as a 32nd-note triplet plus 16th-note).
In two cases, the same cadential formula is additionally embellished. In m.
45, there is a mordent on the third 16th-note, enhancing the virtuoso
character of the piece. It can easily be integrated, again as a 32nd-note
triplet. Performers may consider transferring this mordent to the corresponding note in m. 13. The additional turn in m. 26, on the other hand,
occurs in a surrounding of melodic eighth-notes not adorned elsewhere.
Playing it underscores the preludes playful side but obscures the parallels
of the two hands. If included, this turn begins on the main pitch and
encompasses five regular notes. The cadence concluding the first section
also features two ornaments. Both the accented bass note of the V7 chord
and the final I carry inverted mordents. In the corresponding cadential close
in mm. 47-48, the first of the two ornaments is omitted. If played in m. 15,
it should also be added here. As to the pitches of the lower neighboring
notes, only the D in m. 16 requires an artificially raised leading note since
the modulation has brought us to the key of D major. All other inverted
mordents use notes from the G-major scale.
Finally, there are four purely melodic ornaments. Three are taken up by
the inverted mordent that enhances the E in m. 32 and its two sequences.
All are meaningful ingredients both with regard to the character and the
structural clarity of the piece. The fourth, a mordent in m. 20, is also very
listener-friendly: it emphasizes a new beginning within the otherwise
almost confusingly long chain of eighth-notes unfolding before the background pedal in the hidden second part.
Thematically the prelude is based on several homophonically conceived
units with a distinct texture. Three of these models define the first two of
the preludes three sections. Model 1 consists of two pedals: a sustained
pedal on the first scale degree and a repeated indirect pedal D on the weak
beats of the hidden two-part structure in the other voice. In between, there
are two lines moving in eighth-notes: one is written in eighth-note values,
while the other forms the melodic strand of the hidden two-part structure.
These lines move in parallel sixths. In the ensuing measures, the model
recurs in a mixture of sequence and imitation: The sustained pedal and the
repeated indirect pedal have exchanged hands, as have the same two lines in
eighth-notes, which now move in compound thirds, the inversion of parallel
sixths. When Model 1 recurs in mm. 17-22, the two pedal notes sound in
octaves. The third recurrence of M1 inverts the perfect-fifth interval of the
pedal notes so that they form compound fourths. As if slightly uneasy with
this interval, the pedals move from measure to measure (see mm. 37-40:
G major
479
480
WTC II/15
While mm. 1 and 2 play within the tonic chord, m. 3 along with the first
harmonic shift establishes a figurative pattern that is sequenced in mm. 4 and
5. No subphrasing is needed or even desirable in this virtuoso play of
broken chords. The harmonic outline describes an active step mm. 2 to 3,
thus enhancing the outset
of the above-mentioned
sequence. The dynamic
7
line can only follow the
simple layout expressed
in harmony and pitch. The result is a two-measure increase up to m. 31,
followed by a gradual decrease to the final note.
The fugue comprises only six subject statements:
1 mm. 1-61
U
4 mm. 33-381
L
2 mm. 8-131
M
5 mm. 40-451
U
3 mm. 15-201
L
6 mm. 65-701
M
Apart from the adjustment of the first interval in the answer, the
subject remains unchanged throughout the fugue. It never appears in either
inversion or stretto.
Bach invents two counter-subjects for this fugue and uses them very
regularly albeit with some variation. CS1 is introduced against the subjects
answer (U: mm. 8-13). In its original shape it consists of a five-16th-note
upbeat followed by slower note values in an overall descent including two
syncopations. It ends in m. 13, in the weak position on the third 16th-note
where the suspended E resolves indirectly into F. CS1 accompanies all
further subject statements. It recurs in M: mm. 15-20 in a simplified
pattern: the descent seems as if cleaned, i.e., freed of all ornaments and
escape notes, and the final resolution occurs at m. 201. In M: mm. 33-38
the initial 16th-note run is replaced by a dotted-note group, and in mm. 4041 the upbeat is moved to another voice (L instead of M). Finally in U:
mm. 65-70 the main rhythmic pattern is substituted by a chain of suspended
notes, with the last broken into a chord. Only the descending direction may
still remind listeners of the original CS1.
CS2 is introduced in mm. 16-20, i.e., against the subjects third statement. Entering one measure late, it is characterized by a group consisting
of a syncopation followed by a fifth leap down to a pair of rising 16thnotes. The little group is sequenced twice in descending direction. The
final syncopation resolves belatedly on the weak beat after the completion
G major
481
of the subject (just as the final note of CS1 had done 7 measures earlier).
CS2 accompanies the following two subject statements but is absent in the
final entry. In the course of its two further entries it, too, is slightly varied:
the one-measure rest before the belated beginning is filled in with 16thnotes (see U: mm. 33-34 and L: mm. 40-41), and the ending is diverted
(see U: mm. 37-38 and L: mm. 44-45).
In view of so
consistent a use of
the contrapuntal
material, it may
come as a surprise
that, upon closer
inspection, these
counter-subjects
are found lacking
a crucial quality
required in true polyphony: they are not
independent of the subjects pitch pattern
but actually ornamented parallels! The
examples show the subject with its two
companions as found in mm. 15-20, first in
Bachs setting and then in skeletal simplification revealing the dependency, which
extends to the dynamic design.
In this fugue, each subject entry is separated from the subsequent one.
There are thus as many episodes as there are subject statements.
E1 mm. 6-81
E4 mm. 38-401
E2 mm. 13-151
E5 mm. 45-651
E3 mm. 20-331
E6 mm. 70-72
Bach has invented three episode motifs. The first is introduced between
the dux and the comes. It is related to the beginning of CS1, which it thus
anticipates (U: mm. 6-71 . mm. 8-91) and recurs in similar function in E2
and E4 as well as varied in E3a. The two other motifs are derived from the
subject. Ms1 (L: mm. 22-251 . mm. 1-31) dominates the ten-measure E3b,
Ms2 (L: mm. 45-471 . mm. 3, 4, 5) the even longer E5a. The two episode
motifs are accompanied by remotely analogous companions.
The two long episodes E3 and E5 are subdivided by D-major cadences
whose melodic closing formulas in mm. 22-23 and 61-62 suggest section
endings. The harmonic development on the inner sides of these cadences is
482
WTC II/15
G major
483
484
WTC II/15
In the second section, the first two entries are linked once more by the
dynamically increasing motif derived from the counter-subject. At the
same time, the choice of voices (L, U) and the extension of the pitch range
to the highest octave available on Bachs keyboard instruments support
another growth in tension. After this climactic moment, the long E5a with
its descending sequences brings about a gradual relaxation. Although the
brief E5b presents a powerful rush upward, the final subject statement
enters on a much softer note, owing mainly to its relatively weak position
in the middle of the texture and its reduced polyphonic density, as it is
surrounded by a simplified CS1 and an accompanying bass instead of CS2.
The concluding cadence and particularly the appoggiatura in the final
measure confirm the soft ending.
This playful fugue thus presents itself dynamically in a design that is as
simple as it is capturing:
The others are the B-minor fugue I/24 (Largo) and the B-minor prelude II/24 (Allegro).
485
486
WTC II/16
mm.
II
III
IV
11
13
VI
15
17
19
21
G minor
487
tail. (Several of these segments recur separately, which is why they are
named here.) The head M1a aims for E , the sixth scale degree that, in
the minor mode, represents high tension. The complement M1b ends with
a descending tritone, one of the high-tension intervals that in Bachs time
often characterized highly emotional pieces in slow tempo. The climax is
marked by the largest interval leap, C-A, which is particularly conspicuous
as it emerges from a descending line. The tail M1c provides the expected
relaxation with a gentle stepwise figure that, were it not for the dotted
rhythm, could be read as a written-out turn. The texture of this model can
be described as two-part polyphony with a pedal in the lowest voice. The
dynamic outline of M1 results from all the details stated. Tension grows in
M1a, is enhanced by the tritone in M1b, peaks on the high A, and resolves
in the gentle curve of M1c. The model is imitated a fifth up in mm. 2-31.2
It also recurs in mm. 5-61, 6-71 (with additional filling voices and the
complement now in the lowest part), and in mm. 9-101. M1a, the head
alone, is quoted several times; see mm. 7-81 (4x), mm. 8-9 (as a link at the
end of the cadence), mm. 11-121 (3x), m. 16 (2x), mm. 173-183 (4x, a varied
transposition of mm. 7-81). M1b recurs in two crucial positions: in mm. 1516 it marks the beginning of the peak note line and in m. 19 its end. M1c
exerts its soothing power as one of the complements to M2 (see below).
Moreover it also appears in connection with the beginning and end of the
peak-note line (twice in m. 15, once in m. 19).
M2 is introduced in mm. 3-41. It is related to M1 in two ways: its final
figure (see mm. 3-4: D-A-C-B) quotes M1a and its second complement
(see m. 3, alto: G-F-E-F) takes up M1c, both transposed into a majormode context. Easily overlooked upon first reading is the rhythm of the
first complement (tenor: m. 31-2), which replaces, for the first time in the
piece, the second 32nd-note of the rhythmic unit with a 16th-note. The two
complements in different voices give the impression that this model is set
in three-part polyphonic texture over the pedal D in the lowest part. The
dynamic design of M2 is once again a simple curve, with the climax here
on beat 3 where C appears as the apex of the leading line. Owing to a lack
in high-tension intervals, this curve is much gentler than that in M1. M2 is
sequenced in mm. 4-51 a fourth up and recurs in mm. 10-111. In these two
measures, the principal voice and its main complement are note-identical
but the former pedal in the bass now takes a more active part.
2
As the notation in m. 2 reveals, the uppermost voice in m. 1 is not the highest to appear in
this piece. A neat distinction of voices is, however, not very meaningful since Bach has not
conceived this prelude in consistent part writing.
488
WTC II/16
G minor
489
490
WTC II/16
entries (see mm. 55-56: S and m. 63: B), the most drastic modifications
occur in mm. 67-75. The soprano statement, after beginning regularly in
mm. 69-72, features a delayed resolution which, enveloped in the metric
pattern of a hemiola, finds its satisfactory conclusion only at the end of a
cadential formula at m. 751. The preceding tenor statement can either be
read as shortened (i.e. without the entire final subphrase, in which case it
would break off on the second beat of m. 69), or, more likely, as extended
in its middle by two extra sequences (see mm. 67-68 sequenced in 68-69,
69-70, and 70-71) and then complemented by a free and again very much
extended version of the final subphrase.
Bach invents only one true counter-subject for this fugue. Introduced
in mm. 5-9 in the tenor, it displays a phrase structure strikingly similar to
that of the subject without its initial note: a one-measure subphrase is
followed by two descending sequences the second of which is extended to
two-measure length. The brackets in the example mark this similarity:
G minor
491
The bass line in E1 begins after phrasing in m. 17 with the 16th-note D. There follow a
climax at m. 173, phrasing in m. 182 after E, another climax in m. 183, and another phrasing
in m. 191, etc. In E2 the climaxes fall on T: m. 243, S: m. 253, T: m. 263, and S: m. 273; in
E7 on B: m. 752, S: m. 753, T: m. 761, A: m. 762, S+T: m. 763, and S+B: m. 772.
492
WTC II/16
tions, and ornamental figures. The articulation requires quasi legato for all
melodic 16th-notes except for written-out ornaments, which should be
legato, and non legato for all eighth- and quarter-notes. It is a good idea to
distinguish the quality of touch in the cadential versus the melodic eighthnotes (e.g., in B: mm. 9 and 24-28). Note-pairs consisting of appoggiatura
and resolution are always legato; this applies also to the tied prolongations
in mm. 14 and 15. Phrasing (e.g., T: in m. 10 after A, m. 11 after G, m. 14
after A, and m. 15 after G) is best expressed through dynamic means.
The tempo of this fugue is moderately fast. The 16th-notes provide the
measure for an appropriate pace: they should be swift enough to convey
their ornamental character but not so brisk as to jeopardize their melodic
quality. Transparency must be guaranteed as well as the possibility for
unhurried phrasing between 16th-notes (as is demanded, e.g., at the
juncture of subject ending and counter-subject beginning). The tempo
proportion between the prelude and the fugue uses larger metric units for
translation: a quarter-note in the prelude corresponds with a whole measure
in the fugue. Approximate metronome settings: prelude beats = 28 (eighthnotes = 56), fugue beats = 84.
The only ornament symbol in the score is found in m. 21. The mordent
is approached stepwise. It thus begins on the main note and consists of a
single three-note shake. Beginning this shake with two 32nd-notes, so that
the return to the main note falls on the bass note C, represents a perfectly
acceptable and clear solution. The rendition of the ornament as a triplet
concluding before the bass motif begins is correct but risks blurring.
In terms of its layout, the G-minor fugue falls into two almost equal
halves (44 + 40 measures). While this major dividing line is beyond doubt,
the distinction between sections in each half appears blurred by contrary
information from within the musical text. Here are the details. The first
half contains eight subject statements in T A S B T A S B. The episode
following the last of these entries describes a self-contained dynamic curve
and concludes with a perfect cadence in C minor. The second half presents
parallel entries in the three neighboring voice pairs T+A, A+S, and B+T as
well as the stretto TS with a varied ending. The entering order of the
leading voice in each pair, T A B S, is almost identical with the order
observed twice in the first half of the fugue.4 The final entry in the bass
4
The leader in a parallel statement is traditionally the one which is placed on the scale
degree identical with that of the original subject. Thus in the first pair which is
harmonically in F major, the tenor begins correctly on the fifth and carries on accordingly,
while the alto, doubling in thirds, does not maintain the original interval structure.
G minor
493
seems the odd one out in several ways: it is the only statement in the fugue
that is considerably varied, the only one to contain a harmonically effective
modification of the pitch pattern (B), and it follows after two conspicuous
homophonic formulas in hemiola pattern, thus appearing more like a coda.
To sum up: Bach introduces a new contrapuntal technique in the second
half of the fugue, which he then uses consistently through four entries and
four different voice combinations. The single section comprising four
parallel statements is then followed by a redundant entry.5
In the first half, the repeated entering order suggests two complete
rounds of statements, but the texture seems to impart a different message.
In mm. 20-24, the tenor statement retains the full four-part ensemble
established in the preceding bass entry throughout the subjects first and
second subphrases. The ensuing subject statements in alto and soprano,
however, appear in reduced ensemble as the bass is resting in mm. 28-36.
Yet since neither the harmonic development nor a distinct cadential close
support one or the other solution, we can only deduce that Bach chose to
leave this boundary ambiguous. What he creates amounts to a double
section in which the dividing line can be determined by careful analysis
alone but does not constitute, particularly for listeners, a structural caesura.
In a way, the composer is thus matching the long section in the fugues
second half with a combined unit in the first.
The development of tension in the fugues first half thus consists of
two increases. Interpreters seeking to emphasize the identical entering
order will tone down the second tenor statement despite its initial four-part
ensemble, in contrast to those who perceive a beginning of the second
section only at the point of textural reduction. The fugues second half
begins with a kind of explosion: in full four-part ensemble and with the
5
494
WTC II/16
496
WTC II/17
A major
497
A list of the relevant appearances could look like this: M2a see mm.2, 4, 18, 20, 35, 51;
M2b see L: mm. 5/6, U: mm. 21/22, L: mm. 38/39; M2c see U: mm. 7-9, 29-31, 40-43, 68,
70/-73 and L: mm. 24, 26, 28, 32, 65, 67, 69; M2d see U: mm. 23, 25, 64, 66; M2e see U:
mm. 53, 55, 60; M2f see U: mm. 112-121, 122-151, 442-481.
M3a see L: mm. 7-101, 40-441 (always as a counterpart to M2c); M3b see L: mm. 12-151,
45-481, 64-671 (the latter against M2d); M3c see L: mm. 52-601 (alternatingly against a M1b
variant and against M2e); M3d see L: mm. 23-241, 25-261, 27-281, and as a partial quotation
in L: mm. 29-31 and U: mm 32-33, then again in U: mm. 65-661, 67-691, 70-741 (as a counterpart to M2c or M2d).
498
WTC II/17
Section II
M1a/M1b + M2a
M1a/M1b + M2a
M2b/M2b var
M3d/M2d, M4/M2c + sequ.
M3d/M2c + sequ. + liquid.
cadence
Section IV
M1a/M1b
M2a
M3c/M1b, M3c/M2e + sequ.
expanded cadence + bridge
Section III
M1a/M1b + M2a
M1a/M1b + M2a var
M2b/M2b var
M3a/M2c + sequ.
M3b/M2f + sequ.
cadence
Section V
M3b/M2d, M3d/M2c + sequ.
M4 var/M2c var
M3d/M2c + sequ.
expanded cadence
A major
499
answer feature beats that are only once or not at all subdivided, the fugal
voices thereafter complement one another in such a way that they create a
continuous motion up to the general pause in m. 46. In fact, all through
these inner 42 measures the 16th-note motion only misses four beats:
one each in mm. 13, 14, 16, and 19.
The subjects pitch contour features a number of consecutive leaps:
three of them appear at the beginning in increasing interval spans (third,
fourth, and fifth), two more are added before m. 21 (octave and fourth). In
this rhythmic shape, the two ascending tetrachords (m. 1: B-C-D-E,
m. 2: A-B-C-D) constitute popular formulas. Finally, the 16th-note figure
filling the latter part of m. 2 has ornamental character, embellishing the
suspended resolution D-C. It may be interesting to visualize this subject
stripped of all its ornaments (including, for this purpose, the octave split).
The emerging skeleton is striking with regard to both its interval pattern
and its rhythm: the subject turns out to be based on a progression of rising
fourths and falling fifths,
evolving in a pattern of
gradual rhythmic augmentation.
The phrase structure in the subject can be interpreted in two ways,
depending on whether one seeks to emphasize the structural details or the
harmonic and rhythmic features. Performers choosing to stress the fact that
the tetrachord in the first half of m. 2 is a sequence of that in the second
half of m. 1 will render the subject as consisting of two subphrases that are
divided by an octave leap. Its phrase structure may then be described as a
three-note upbeat + ascent followed by a single-note upbeat + ascent + tail
(ornamented resolution). Conversely, performers regarding the syncopation in m. 2 as the subjects salient featurea metrically highlighted note
that is also harmonically alivewill interpret the subject as an undivided
unit encompassing two consecutive upward thrusts that peak in the
syncopation. The dynamic design differs accordingly. If the emphasis is on
the sequence of two tetrachords, the three-eighth-note upbeat leads to a
first climax on B, followed by a decrease on the way up to E. The lower
E then provides the active upbeat to the second climax on the A, which
launches a gradual relaxation throughout m. 2 to the final C. Conversely,
if the emphasis is on the syncopation, there is only one climax, reached in
an unbroken crescendo and followed by a diminuendo of almost equal
extension. Especially for a lively interpretation of this fugue, the second
option for phrasing and dynamic shaping is probably preferable.
The fugue comprises fifteen subject statements.
500
WTC II/17
1.
2.
3.
4.
5.
6.
7.
mm. 1-3
mm. 3-5
mm. 6-8
mm. 8-10
mm. 13-15
mm. 16-18
mm. 18-20
A
S
T
B
B
A
T
8.
9.
10.
11.
12.
13.
14.
15.
mm. 22-24
S
mm. 24-26
A
mm. 32-34
T
mm. 35-37
S
mm. 37-39
B
mm. 41-43
T
mm. 42-44 (45) B
mm. 48-50
T
A major
501
502
WTC II/17
A major
503
Distinctions may seem small and, particularly to impatient performers, hardly worth the
effort. Yet they do make all the difference in this fugue where color contrastsbetween
subject and counter-subject on the one hand and between primary material and independent
episode material on the otherare essential.
To avoid confusion, here is the articulation in the subject: E-C-F non legato, B up to E
legato, both Es non legato, A up to D and all remaining 16th-notes legato.
5
504
WTC II/17
The fugues structural layout is easy to grasp. The entering order of the
subject statements, combined with cadential endings followed invariably
by a reduction of the ensemble, creates a particularly unambiguous design.
The only irregularity consists in the position of an episode at the beginning
of a section. These are the details:
Section I comprises the initial round of subject statements (A S T B)
connected by E1. It is followed by the (self-contained) E2, a redundant bass entry, and the closing E3. While all five entries are in
A major, in regular alternation between dux and comes (tonic and
dominant), the closing episode, which concludes the section on the
middle beat of m. 16, modulates to the relative minor key.
Section II features four subject statements, two pairs bridged by E4,
the episode closest to the thematic material. The subject statements
are launched once more from the home key, repeating the modulation to F minor, this time in the final entry. This alto statement is
conceived as redundant in the round (A T S A), thereby announcing
the end of the section, which is confirmed in E5a with the cadential
close in C minor.
Section III begins with a five-bar subject-free passage (E5b). It is
followed by three subject statements (T S B) that are linked by E6
and closed by E7. All three entries are in the minor mode (E minor,
B minor, and D minor), representing the minor dominant, the
subdominant relative, and the minor subdominant respectively).
The return to the home key, prepared in the modulatory process in
E7 and confirmed with the tenor entry in m. 41, marks the beginning of the fourth section. This section encompasses three subject
statements, the last of which is again conceived as a redundant
entry (T B T). The harmonic digression at the end of the bass entry
and in the subsequent E8 enhances the impression of redundancy,
giving the fugue a particularly well-rounded ending. This section is
distinguished not only by its use of texturevirtuoso patterns in
one hand, block chords in the other, a rhythmic surprise in the weakbeat general pause, and a four-part cadence ending deceptively
but also by particularities of its internal structure. The four opening
measures feature the only overlap of subject statements in the
fugue. The counter-subject accompanying the two entries involved
in the stretto is varied and merged into a single prolonged chromatic
descent (see m. 41: A to m. 44: B), and the bass entry is extended
up to m. 451. The impression arising here is that of a superimposed
four-measure phrase.
A major
505
The first and second sections are moderate and fairly similar with
regard to their overall intensity level. Thereafter, the third section with its
consistent harmonization in the minor mode appears comparatively subdued, while the fourth section is even more outgoing than the opening.
Within section I there is a gradual increase of tension due to the growth
in the number of voices. E1 constitutes only a short lessening of tension
without any serious interruption in the dynamic build-up from one subject
506
WTC II/17
The first interpretation entails a register plan throughout the piece in order to convey what
might have been Bachs idea beyond the very scarce indication of mm. 3 and 5. This requires
specialized knowledge about early keyboard instruments.
507
508
WTC II/18
Since the rhythm is simple and the pitch pattern includes many large
intervals, the basic character of this prelude is lively. The articulation
requires 16th-notes in light legato touch and non-legato eighth-notes. The
fact that Bach slurs the eighth-note pairs in mm. 9 and 10 as well as the
similarly built pairs in mm. 37 and 38 indicates something beyond nonseparation. These are characteristic appoggiatura-resolution pairs, a quality
that, without the slurs, might have been easy to overlook in the complex
harmonic setting of these measures. Not only should such pairs under all
circumstances be rendered legato, the slurs add a sense of weightylight.
The similar slurs in mm. 44-45 serve to distinguish metrically analogous
but harmonically different groups: on beat 2, the appoggiatura is to be
linked to its resolution, while beat 4 features two sixths belonging to the
same harmony which must be detached.2
The tempo of the prelude is flowing: fast enough to allow listeners to
hear swiftly moving quarter-notes, but not so rushed as to convert the 16thnotes into virtuoso runs. The appoggiaturas that are distinguishing features
in the prelude must retain some of their sighing quality. Ornaments include the grace-notes and two trills. All grace-notes appear as eighth-notes
preceding quarter-notes. The value of the main note is thus split into equal
halves, with the appoggiatura on the beat (mm. 2, 4, 17, 31, and 42). The
trills embellish 16th-notes that form part of written-out ornamental patterns
(U: m. 18, M: m. 19). They are therefore nothing but suffixless mordents,
beginning on the upper neighbor note and comprising four 64th-notes.
Both the structure of the prelude and its dynamic layout are primarily
influenced by the succession of motifs. Meanwhile, orientation for a largescale design is provided by several indirect lines. Thus the first three of the
preludes four sections are strung together by consistent overall descents in
the outer voices. In section I, the bass moves gingerly downward over 1
octaves, slowly at first, then picking up speed in diatonic steps followed by
more closely spaced chromatic steps, only to rise again in consecutive
fourths. Similar though even more pronounced processes can be detected
in section III (l.h. from m. 25, r.h. from m. 36) where extended chromatic
descents even create rhythmic patterns. Section II is more static with merely
a few bass-note lines. In the fourth section are such lines of no consequence.
The example shows this with an excerpt from the first three sections:
2
G minor
II
509
mm. 1 2 3 4 5 6 7
mm. 16 17 18 19
III mm. 25 26
20
27
10
11
12
13 14 15
21 22 23 24
28
mm. 36
29 30 31 32 33 34 35
37
38
39
40
This brings us to the motifs characterizing this prelude. M1 is a twobar unit with three elements. In m. 1, a 16th-note line accompanied by two
metrically placed chords describes a twofold curve in the treble. The lines
long descent into the bass register splits into a hidden two-part structure
consisting of a typical bass pattern below a repeated pedal. In m. 2, new
treble voices present two sigh motifs in double thirds. The resulting
impression of a three-strand texture should be underlined: there are the
block chords, the 16th-notes, and the little melodic gestures. M1 is then
repeated (according to Bach: as an echo), with the sigh motifs in inverted
voices as parallel sixths. Variations of M1 can also be found in mm. 16-17,
41-42, and 43-44. In all three cases, the accompanying chords are replaced
by melodic figures based on segments of the other two M1 elements.
M2 is introduced in mm. 5-7. In its accompaniment, it is related to M1.
The melodic double notes are also taken up, together with their particular
metric organization. The double notes describe what appears like an incomplete one-bar curve. Beginning with a written-out inverted mordent
followed by a sudden leap, they embark on a descent that seems to prepare
a conclusion on the next downbeat. Yet the expectation is deceived twice
before the sequence finally reaches its goal (see m. 81). Dynamically the
three measures describe a gentle descent in which the escaping sixth leaps
should sound as charms, i.e., softer than their surroundings. M2 recurs
once, without its second sequence and, more importantly, without the final
resolution (see mm. 21-22).
M3 consists of two contrapuntally conceived voices. M3a, the lower
part, spans a single measure (m. 8). M3b, the right-hand part, begins after
beat 2 and ends in the following measure with the above-mentioned slurred
510
WTC II/18
3-4
M1
p
5-7
8-113
M2
M3
f di mi nu en do
Section II
mm. 16-17
M1 var.
18-20
developmt.
21-22
M2
Section III
mm. 25-33
developmt.
dim.
34-35
x
cresc.
393-40
36-393
M3
expansion
di mi nu en do
Section IV
mm. 41-43
M1 varied + expanded
113-15
x + close
crescendo
23-24
expansion + close
46-49
development + close
G minor
511
512
WTC II/18
G minor
513
Two further S1 counter-subjects recur only once each in considerably altered shape; they
thus exert no real impact on the polyphonic design of the fugue. Both combine dosido
formulas with eighth-note runs, thus mixing elements of both primary components.
514
WTC II/18
The final 8 eighth-notes of S1 are quoted in U: mm. 37-391 and imitated in M: mm. 39-411.
Both quotes are accompanied by the second subphrase of CS1 (see M: mm. 37-391; U: mm.
39-411). This material builds E4a, which ends with a cadential close in B major at m. 411. In
E14, only the final 6 eighth-notes of S1 appear, but all the more frequently (see L: mm. 128129, M: mm. 129-130, U: 130-131, M: mm. 132-133, and L: 134-135). The ending of CS1
is considerably shortened here and dropped after only one occurrence (see U: mm. 129-130).
S2 material is heard in two versions: as a chromatic descent (taken from the initial six notes
of S2, see M: mm. 115-116 and L: mm. 117-118) and as an ascent of one whole and two half
steps (derived from the five rising notes of S2, see U: mm. 83-84, M: mm. 84-85, and L: mm.
85-86, 86-87, 87-88, 88-89). CS2 appears twice as an extending sequence (see in E7: L mm.
65-66, and in E8 varied: U mm. 70-71). Later on, the entire second half of CS2 (i.e., its final
twelve eighth-notes) is quoted in E13 (see U: mm. 123-125). This recurrence is accompanied
by the shared ending of CS1 and S2 (see L: third eighth-note m. 123 to 1251).
5
For M1 see U: mm. 9-101, for M2 see M: mm. 9-101, for M3 see M: mm. 17-181, and for
M4 see L: mm. 17-181. M5 is introduced in U: mm. 17-191 and continued with a chromatic
variant from U: mm. 107-1091 onward.
G minor
515
Three of the episodes are distinguished from the remaining ones by the
fact that they feature motifs that appear nowhere else in the fugue. As these
three episodes are at the same time the longest subject-free passages, this
observation certainly deserves attention. The components are: in E3, a syncopation followed by five eighth-notes (L: mm. 26-27, M: mm. 27-28, and
L: mm. 28-29), in E10, two consecutive syncopations with upbeat and
subsequent mordent (U: mm. 85-871, M: mm. 86-881, U: mm. 87-891, and
M: mm. 88-90) and the figure characterized by a slide in 16th-notes
(U: mm. 90-91, M: mm. 91-92, and L: mm. 92-93), and in E13, a three-note
ascent with artificial leading-notes (U: mm. 119-120, M: mm. 119-120,
U: mm. 120-121, M: mm. 120-121, and U: mm. 121-122).
The role each of the fifteen episodes plays in the dynamic development
of the fugue can be described as follows: E6 and E15 are nothing but
elaborate cadential closes, extending the preceding subject statements
without any color contrast and bringing forth an immediate and complete
relaxation. E4a, E7, E8, and E14 evolve from the subject entries preceding
them by way of imitation or sequence. They, too, do not bring forth a color
contrast but just a gradual relaxation. The three long episodes (E3, E10,
and E13) describe self-contained curves. They all begin with an increase of
tension followed by an extended decrease and a final relaxation underpinned by a cadential close. The remaining episodes (E1, E2, E4b, E5, E9,
E11, and E12) are short and display independent material. E1, E2, E5, and
E9 are determined by descending sequences, while the others, particularly
E11 and E12, display rising lines. Besides their individual increases or
decreases, all these episodes establish a color contrast to the surrounding
subject statements.
In this fugue, the main subject with its simple rhythm and restricted
melodic expression can be dangerously deceiving with regard to the works
mood, misleading performers into too swift and light-paced a mood. The
complexity of the overall rhythmic pattern and the high content of
chromatic lines and altered notes do in fact indicate a rather calm basic
character. The tempo should express the compound meter in the manner of
a very moderate half-measure pulse.
The relative tempo of the prelude to the fugue sets the larger beats into
proportion, thus providing enough variation on the surface with 16th-notes
in one, triplet eighth-notes in the other piece: a quarter-note (a beat) in the
prelude corresponds with a dotted quarter-note (half a measure) in the fugue.
(Approximate metronome settings: 76 for the quarter-notes in the prelude
and the dotted quarter-notes in the fugue.)
516
WTC II/18
G minor
517
E1: M1 + M2
+ extension
E2: M3 + M4 + M5
E3: M3b + M4a + M5
+ extension with E3 motif
E7: M2a + CS2 sequence
E8: M2a + CS2 sequence
E9: M2a + M3a + M5
E10: Ms2
+ extension with E10 motif
E5: M1 + M2
+ extension
E4b: M3 + M4b + M5
E4a: Ms1 + Mcs1
E6: cadence
E11: M3a
E12: M3a + M4/4a + M5a
E13: Ms2 + M3a + M4a
+ extension with E13 motif
518
WTC II/18
520
WTC II/19
These pre-beat ornaments are eventually complemented by a few onbeat ornaments. These are also spelled out and involve, in contrast to the
ones mentioned above, the only 16th-notes in the prelude. In m. 92, the
upper voice can be read as a dotted quarter-note F precede by a grace-note
appoggiatura and ornamented with an inverted mordent. In m. 21, the
16th-notes hide a downbeat G adorned with a slide. In view of the
ornamental structure of the lines it would be misleading to describe the
pitch pattern as consisting of alternating steps and skips, since these unfold
on different levels of the melodic process: the skips in the background
pattern, the steps on the ornamental surface.
The tempo of this prelude is calm but flowing: calm enough to avoid
any hurried impression (which might be caused particularly owing to the
silent finger shifting necessary for perfect legato) and flowing enough to
convey the feeling of gently swinging compound 4/4 time rather than
creeping eighth-notes. The appropriate articulation is legato in all notes
appearing in the upper and middle voices as well as most notes in the
lower voice, with the exception of octave leaps (see mm. 2, 5, 15, 23, 26,
and 32) and a cadential-bass pattern (m. 30: A-F-D-E-A) which should be
gently detached.
Although it is certainly possible to describe the initial melodic figure
of the prelude (explained above as an ornamented broken chord) and to
state its recurrences throughout the prelude, such a listing does not truly
facilitate an approach to the piece.1 What Bach presents here are general
pitch lines, elaborately decorated but basically simple. They can be
described as follows:
1
27 28
31 32/33
32
32
27 28
31
30
A major
521
Section I
mm. 1-41 rising
4-61 falling
U: C-D-E-F U: F-E-D-C-B
6-81 rising
M/L
Section II
mm. 97-131
falling
13-151
rising
15-161
falling
Section III
mm. 16-201
falling
20-211
rising
21-221
falling
Section IV
mm. 22-251
rising
25-271
falling
27-291
rising
8-97 falling
U/M/L
29-311
falling
Section V
mm. 31-33
rising
Material and layout make it easy to identify section IV as a recapitulation of section I and section V as the coda. Yet whether it is meaningful,
or even necessary, to declare sections II and III as development, remains
arguable. Structure, just like thematic material, is but a secondary issue in
this piece. The dynamic realization can follow the essential pitch lines
traced above. Thus rising lines would be expressed in gentle crescendo,
falling lines in diminuendo. The basic tone color, however, remains the
same throughout the prelude. There are no contrasts and no moments of
dramatic tension, as implied by the calm and soothing impression this
piece conveys.
522
WTC II/19
The rhythmic pattern, both in the subject itself and in the fugue as a
whole, is simple. The basic features are chains of 16th-notes and dotted-note
groups. These two frequently combine in such a way that the 16th-note
pulse appears as the result two parts in complementary dotted-note groups
(see mm. 5, 7-8, 10, 12, 20, 24, and 28 where the effect of uninterrupted
16th-notes is achieved by the interplay of the accompanying voices with
the subject).
The subjects harmonic background is basically simple, although the
ornamental curves and their syncopations add spots of flavor and tension.
The music example shows the
subject as Bach presents it at the
opening of the fugue and below it
the underlying large-scale curve;
both marked with an interpretation of the main harmonic steps.
The dynamic design follows
the simple phrase structure with
an increase throughout the first measure, a climax on the second syncopation, and a subsequent relaxation. The syncopation on the last eighthnote of m. 1 captures the highest amount of tension: it stands harmonically
for the subdominant, i.e., the most active step in the simple cadence, and
metrically for the only downbeat in this subject, an important fact in a
composition where meter, owing to the characteristics of the rhythmic
pattern, plays a decisive role.
There are ten subject statements in this fugue.
1. mm. 1-23
L
6. mm. 12-133
M
2. mm. 23-41
M
7. mm. 16-173
L
3. mm. 5-63
U
8. mm. 20-213
U
4. mm. 7 -83
L
9. mm. 233-251
M
5. mm. 93-111
U
10. mm. 273-291
U
A major
523
mm. 12-13) involve the third and the fourth scale degrees (mm. 20-21:
G/G/G/G, mm. 23-24: G/G/G, and m. 28: D/D/D/D). There are no
inversions, strettos, or parallel statements.
Neither does Bach invent a regular companion to the subject that
would display a minimal degree of independence. Instead, the subject
comes accompanied by several kinds of rhythmically varied parallels. The
most frequent pattern moves in thirds below the simplified curve (M: mm.
5-6, 9-10, 20-21, and 27-28) or in sixths above it (U: mm. 23-24). The
quarter-note values are broken into dotted-note groups that split the pitch
into note repetitions (mm. 5-6, 9-10, and 27-28) or add chromatic
semitones (mm. 20-21 and 23-24). Interestingly, such parallels only ever
occur between the upper and middle voices. There is only one lower-voice
entry that incites a parallel. Yet this is even more explicit than the
previously mentioned ones as it involves more fractions of the ornamental
pattern in the subject (see m. 7: L + M).
In four of the previously identified cases (mm. 5-6, 9-10, 20-21, and
27-28), the lower voice adds a further double-third parallel to the subjects
quarter-notes on beats 3, 4, and 1. As a result, this stretch sounds in parallel
triadsan extremely rare phenomenon in a genre renowned for complex
polyphony. This further parallel also comes in a dotted-note rhythm, with
the splitting appearing as a note repetition, an octave displacement, or a
chromatic semitone. The following comparison of one of the excerpts with
its simplified version shows these parallels:
mm. 5-6
524
WTC II/19
A major
525
The fact that this ornament appears in brackets should not induce us to think it is optional.
This kind of cadential trill was so commonplace in Bachs time that performers would
automatically play it even where it was not explicitly stated. Its appearance against running
16th-notes, and the alleged technical problem arising hereof, is no reason for omitting the
ornament either. Anyone who can play this fugue in all its required rhythmical accuracy will
certainly be able to tackle this trill, particularly if two of the final 32nd-notes in the middle
voice (D-C) are taken over by the left hand.
526
WTC II/19
In the absence of tension-enhancing developments in the subject statements, no dramatic climaxes are built up. By the same token, the close
thematic relationship of the episodes with the subject discourages explicit
color contrasts between the primary and the secondary material. Instead,
the only large-scale shading occurs in the minor-mode middle section,
which should sound less brilliant than the surrounding major-mode sections.
This is a playful, virtuoso fugue in which joyfulness rather than dramatic
developments are the goal.
527
528
WTC II/20
ornamentation (for details see below). The basic character of the piece can
therefore be interpreted as rather calm. The tempo may be fairly slow; the
chromatic 16th-notes should be given time to unfold their full emotional
content, and the 32nd-notes ought not to appear virtuoso.
The corresponding articulation is a continuous legato. This is only
interrupted by phrasing and in the few cases of cadential octave leaps (L:
mm. 16, 28, and 32). Phrasing is a matter that needs to be pondered with
great care. At first glance one would think this to be straightforward, as
one-measure motifs and the imitative texture are obvious. Closer analysis
reveals that there is frequent overlap of a tied notes harmonically required
resolution at the end of one motif with the initial note of the subsequent
motif. Such overlap, in which an interruption of the sound flow would
therefore be wrong, occurs in almost every measure. In all cases, the
progression from the end of one phrase to the beginning of a new one can
only be expressed by exceedingly eloquent dynamics and occasionally by
a change of tone color. The following table gives an overview of the
phrasing at the point after the downbeats in each measure. The caesura ()
suggests a slight interruption of sound flow, the slur ({) indicates that an
interruption is unwarranted.
mm. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
U
{ { { { { { { {
L { { { { { { { {
mm. 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
U { { { { { { { { {
L
{ {
In addition, there is a breath after the middle beats in U: mm. 10, 14, 15,
20, 23, and 24 as well as in L: mm. 3, 6, 7, 12, 19, and 27.
The prelude feature three ornaments, occurring in the final measures
of the two halves respectivelythe two measures that, through the use of
scalar passages and leaping octaves, appear as virtuoso and thus distinctly
different from the measures preceding them. In m. 16, the lower voice is
embellished with what appears in the score as a mordent and a trill. The
mordent, however, is followed by a suffix-like group and thus given away
as another note-filling ornament. Both trills shake in 64th-notes, i.e., twice
as fast as the fastest note values in the piece, the regularly occurring 32ndnotes. The one at m. 161 is approached stepwise and therefore begins on a
slower main note and also ends in slower values, as indicated by Bach. The
other trill begins and ends regularly, launched from the upper note and
concluding with a suffix in the speed of the shakes. As both ornaments
A minor
529
occur in the left hand and against fast notes in the right hand, they intensify
the virtuoso impression that is created in the written-out notes and further
enhanced by the sudden appearance of the additional middle voice. The
inverted mordent in m. 32, by contrast, is simple: it consists of the usual
three notes (A-G-A), which should be played fast enough for the trill to
end before the B that initiates the run in the lower voice.
The preludes main motif (M1) is introduced in mm. 1-21. More
precisely, the lower-voice part of the motif begins on the second eighthnote, after a downbeat note that (like the one in the upper voice of m. 17)
serves as harmonic and metric support but does not belong to the thematic
material. These downbeats should therefore be played in a neutral tone
color and not at all assertively. The three lines of M1 are conceived in
parallel motion, presenting a single gesture in what may be described as
ornamented homophony:
The corresponding dynamic gesture is an overall diminuendo, triggered by a short but fairly strong crescendo in the initial four-note ascent.
While the decrease in tension is gradual and straight in M1b, the leading
M1a should be subtly shaded in such a way that the predominant chromatic
descent stands out against the backdrop of a softer secondary line in the
higher register.
M1 recurs without modifications in mm. 2, 4, 5, 11, 13, 25, andwith
small variationsin mm. 8, 9, 26, 30, and 31. Inversions occur in mm. 17,
18, and 22, modified also in m. 21 and as a partial sequence in m. 20. Very
free adaptations can even be detected in mm. 19 and 23-24. In all this, the
dynamic shaping of the inversion raises a question. Should one retain the
original design with its diminuendo or follow the natural temptation and
ascend in crescendo? This determines whether the B section, which is
where most of the inverted statements occur, emerges as the dramatic
climax of a prelude in ternary form (which results if the rising lines are
rendered in rising tension) or rather as an integrated section in a prelude in
binary form (as happens if the original dynamic shaping is retained also for
the motifs inversion).
530
WTC II/20
A minor
531
and second-to-last intervals are perfect fourths, and while the third leap
represents a diminished seventh, the third-to-last uses the complementing
minor third. The bridged connection between the two halves of the phrase
is marked by a gap both rhythmicallythe duration of the rest is
longer than any of the note values
major fourth dimin.
min. fourth major
surrounding itand in terms of
tritone
third
seventh
third
third
pitch, with the interval that expresses the highest degree of tension:
the tritone. Owing to this symmetrical correspondence of the two
halves, the phrase must be regarded as an indivisible unit in which the rest,
far from indicating phrasing, represents the moment of highest tension.
This view is supported by the harmonic progression, which displays a
simple cadence. The dominant (in Bachs harmonization often appearing in
the form of a ninth chord) falls on m. 21
and resolves, through the rest and the
leaping eighth-notes, only on the final
note at m. 31. The dynamic design correi iv V9 - - - - - i
sponding with these features consists of
a powerful increase of tension toward the G, followed by a tensionsustaining rest and a gradual release through the eighth-notes.
The subject appears eight times.
1 mm. 1-3
L
5 mm. 13-15
U
2 mm. 3-5
M
6 mm. 17-19
M
3 mm. 6-8
U
7 mm. 21-23
U
4 mm. 94-113
L
8 mm. 254-273
L
Apart from the adjustment of the initial interval in the tonal answer
and the enlarged leap in the major-mode statement (see m. 10: A-B = a
minor seventh instead of a diminished seventh), only small rhythmic modifications occur. In mm. 9, 17, and 25, the subjects initial note is shortened
to an eighth-note.1 Inversions, strettos, or parallels are not used.
1
At first glance, the fourth and the final entries seem to leave it open whether we are dealing
with the original of the subject beginning, with a third in which the gap is filled by a passing
note, or with the answer entering an eighth-note late. Upon closer inspection one discovers
that the metric organization of the melodic lines preceding the entries, together with the
harmonic logic, determine the eighth-note on beat 4 as the end of the preceding development.
532
WTC II/20
Bach invents two companions to the subject. These appear with great
regularity, accompanying all but the initial and final entries. CS1 is
launched after an intermittent partial sequence of the subjects tail from the
sixth eighth-note of m. 3 onward. Its conspicuous feature is a five-note
plunge occurring four times in descending sequences. The first two of
these scale segments are separated by an interval leap and a rhythmic gap.
This interruption is subsequently softened when the gaps in both pitch and
rhythm are fleshed out. After a fourth plunge, an ornamented ascent
concludes CS1 on a relaxed note. This counter-subject is thus independent
from the subject in every possible respect. The dynamic equivalent to the
compositional details consists in a four-fold increase of tension, with the
climax falling each time on the lowest note, followed by a decrease in the
gentle upward motion. CS2 enters almost a whole measure after the subject
and ends undetermined (compare L: mm. 7-8 with M: m. 11, L: mm. 14-15,
U: mm. 18-19, and M: mm. 22-23). Its most characteristic segment contains the first four eighth-notes. These complement the subjects rhythm by
filling the mid-phrase rest in a way that ingeniously combines the three
outstanding features from the subject: The interval pattern in the CS2
segment sounds like a free imitation of the subjects first half, the rhythm
is an anticipation of the subjects second half, and the distinctive interval,
used twice in the CS2 segment, is the tritonethe interval that connects
the two palindromic halves of the subject. At the same time, the CS2
segment forms a parallel in sixths with the central portion of CS1. Rather
than completely independent, CS2 is thus a supporting and enhancing
complement to both the subject and its primary companion. Dynamically,
a crescendo
throughout
the opening
four eighthnotes is answered by a
diminuendo
that parallels
the one in
subject and
CS1.
The fugue features eight subject-free passages. As the two countersubjects enter later than the subject, overlaps of episode material with
subject statements occur regularly. The table below lists only the episode
segments that play a role in the interpretation of the fugue.
A minor
533
E1 m. 3
M: second to fifth eighth-notes
E2 mm. 5-6
M + L: up to the fifth eighth-note
E3 mm. 8-9
M: up to the first or fourth eighth-note in m. 10
E4 mm. 113-13 U: up to m. 131, L: up to the fifth eighth-note
E5 mm. 15-173 L: up to the fifth eighth-note
E6 mm. 19-21 M + L: up to the fifth eighth-note
E7 mm. 23-253 M: up to m. 261
E8 mm. 273-28
The episodes combine motifs derived from subject and counter-subject
with independent motifs. Ms is a free sequence of the subjects second
half. It is first introduced in m. 3, the minute link between the initial
subject entry and the answer. It is later accompanied by a motif derived
from the end of the counter-subject (for Mcs see L: mm. 11-12, M: m. 12,
U: mm. 12-13), which soon spawns developments of its own (see mm.
11-13, all three voices) and ultimately even acquires an active attitude (see
L: mm. 19-213).
M1 is the first independent motif. Neither its pitch nor its rhythm
derives from the primary material. This motif is introduced in E2 (see L:
m. 51-53). In its symmetrical design of ascent and descent, the number of
notes (4 + 4), and the tension-sustaining rest, it shows a certain structural
relationship to the subject.2 M2 appears exclusively in E6 where it is heard
in a pattern with imitation and varied sequence (see U/M: mm. 19-20).
The use of episode material establishes a number of relationships
among the subject-free passages. (Subdivisions in some episodes appear on
the basis of material, not as a result of harmonic closes.)
E1 . E4b
E2 . E3a, E5a, E7b
E3b . E5b
E4a . E8
material
Ms only
Ms + M1
M1a imit.
Ms + Mcs
compare mm.
13 (L)
8-91 ,15-161,24-251
16-17
27-28
with mm.
2 (L)
5-61
9-10
11-131
As was already mentioned, E6 is in a category apart owing to its combination of M2 and the ascending sequences of Mcs. Outstanding for other
reasons is E7, which consists of three segments. E7a is an extension of the
preceding subject entry, E7b follows with a transposition of E2, and E7c
rounds this episode off with scales that are not even abandoned at the
advent of the subsequent subject entry, but replace the counter-subjects.
2
Variants of the motif are the compact version without rest in E3 (see. M, O, M: mm. 9-10)
as well as different extensions that increasingly approach the shape of simple runs.
534
WTC II/20
A minor
535
note. Its shakes are usually taken in 32nd-notes. This is acceptable, though
choosing twice the speed might be preferable for anybody who can manage.
The faster trills not only add drama and brilliance to the work, but also
avoid the problem of fusing ornamental notes with melodically essential
ones (e.g., the 32nd-notes in CS1, M1, and M2). The trill ends with the
suffix in the rhythm notated by Bach.
There are two exceptions in the execution of the trill. In M: m. 27, the
ornament begins on the main note and thus with a note of twofold duration
before launching its shakes.3 In m. 28, a compound ornament beginning
with an inverted mordent is indicated. The trill thus begins, in full speed,
with the lower neighbor note.
As has been shown above, Bachs design of the episodes divides the
fugue into two halves. The first half ends with a cadence in the home key
at m. 131. The second of the slightly uneven halves literally consists of two
sections: section II (mm. 13-211) features two subject statements in the
upper and middle voices, followed by section III (mm. 212-28) with two
statements in the upper and lower voices. The relationship among the
subject entries in this fugue, particularly between U: mm. 13-15 and M:
mm. 3-5 as well as L: mm. 25-27 and L: mm. 9-11, stresses the design in
two halves over that in three sections. The second half is only slightly
more extended, as a table of the corresponding episodes displays:
E1
E4b
E2
E3
E5 (E6, E7a) E7b (E7c)
E4a
E8
The first section begins with a majestic increase from the first to the
second subject statement. The episode that follows, conceived in descending
sequences, reduces the level of tension, which is then picked up all the
more forcefully by the third subject entry in the upper voice. E3 brings
another gradual decrease. Interpreters who regard the redundant lowervoice statement as a false fourth-voice entrya possible viewshould
aim at slightly surpassing the dynamic level of the previous entry. Performers who do not underwrite this concept should render the entry in slightly
lesser intensity, as is fitting for a redundant statement. The final episode of
section I, E4, provides the dynamic closure.
3
Note that in Henles Urtext edition, the counting of the measures is wrong at this spot since
the half measure at the end of the second-last line and its complement at the beg inning of the
last line have accidentally been separated by a bar line and are consequently counted as two
measures. The fugue totals 28 measures.
536
WTC II/20
In Book II of the WTC, Bach thus includes a whole suite among his preludes: an allemande
in D minor, a courante in E major, a sarabande in F major, and a gigue in Bmajor.
537
538
WTC II/21
V mm. 33-411
F major to F minor (perfect cadence)
VI mm. 41-4812 to G minor (perfect cadence, tonic relative)
VII mm. 49-5212 B major (imperfect cadence)
VIII mm. 53-5612 to C minor (perfect cadence, subdom. relative)
IX mm. 57-6412 C minor to B major (imperfect cadence)
X mm. 65-761
modulation to F-major7
XI mm. 76-831
continuation of F-major7
XII mm. 83-8712 return to B major (tonic)
These sections encompass three analogous passages:
mm. 812-12 . mm. 5212-56 transposed
mm. 13-171 . mm. 37-411
transposed and varied
mm. 28-32 . mm. 83-87
transposed
Another feature that is vital for a first overview is the sequence. Three
categories can be distinguished: sequencing in all three voices, sequencing
in the leading voice with free contrapuntal work in the other voice(s), and
sequencing involving the inversion of voices.
model
mm. 9-10
mm. 13-14
mm. 19
mm. 221-227
mm. 237-241
mm. 25
mm. 65-66
mm. 72
sequence
mm. 11-12
mm. 15-16
mm. 20
mm. 227-231, 231-237
mm. 241-247, 247-251
mm. 26
mm. 67-68
mm. 73
model
mm. 33-34
mm. 37-38
mm. 41
mm. 457-467
mm. 57
mm. 62
mm. 70
mm. 767-787
sequence
mm. 35-36
mm. 39-40
mm. 42
mm. 467-477
mm. 58, 59, 60
mm. 63
mm. 71
mm. 787-807
m. 7
m. 26
B major
539
In addition to these trills, there are three more ornaments. The mordent
in m. 2 begins on the upper auxiliary and contains four notes, while the
one in m. 8 is approached stepwise, begins on the main note, and needs
only a single three-note shake. The inverted mordent in the principal motif
appears in brackets, which points to its origin not from Bachs manuscript
but from a copy. While the choice whether to include or disregard it is thus
left to each performer, adopting it is, in fact, a good idea as it can be of great
help to listeners in the manifold versions of the main motif. If the inverted
mordent is incorporated in m. 1, it should be transferred to all measures
featuring the version of the motif in which the 16th-note scale segment is
complemented by three dotted eighth-notes in reverse motion, i.e., to mm.
3 (L), 5 (U), 8 (L), 49 (U), 65 (U), and 67 (U). What remains to be decided
is whether an inverted mordent is preferable in all cases or whether the
inverted motif (e.g., L: mm. 3-4) might not benefit from an inversion of the
ornament (i.e., a simple mordent).
The thematic material of this prelude is determined by four motifs. M1
comes in numerous guises, which all share the beginning after a strong
beat and the scale section in 16th-note leading to the next strong beat. With
five regularly appearing variants, this motif dominates a large portion of
the prelude (M1a = U: mm. 1-21, M1b = M: mm. 18-27, M1c = L: mm. 2-31,
M1d = U: mm. 222-8, M1e = L: mm. 31-321). M2, first heard in mm. 9-10,
consists of two contrapuntally dependent parts. It can be heard as a pedal
note with above it an embellished descent in the alternating strong beats of
the two voices (see the octave parallel B-A-G-F, the first three notes with
written-out inverted mordents, interspersed with broken chords). In pitch
pattern, length, and texture M2 is thus distinctly different from M1. The
dynamic shape follows the line formed by the peak notes and thus describes
a two-measure diminuendo. M3 consists of a one-measure model in m. 13
that is sequenced three times in mm. 14-16. The texture is reminiscent of
toccata-style. The right-hand pattern can be described as hidden two-part
structure (see, e.g., m. 13: A-G-G before the harmonic background of a
G-minor broken chord). The left-hand part accompanies with another
two-part pattern consisting of cadential notes in the bass interspersed with
upwards leaping fourths in the high register. M4, a two-measure unit
restricted to the upper voice (mm. 28-29), is similarly designed in hidden
two-part structure: as an ascent above a pedal C ornamented with inverted
mordents. M5 (mm. 58-63, with imitation and inversion) is purely local.
The following table shows these motifs, albeit without specifying
inversions. Cases of voice swapping are marked with *, developments or
extensions with +, modifications with '.
540
WTC II/21
mm.
1-31
3-51
5-77
77-8
9-10
11-12
13-16
17-18
19-20
21-227
22-237
237-281
motifs
M1a, M1b, M1c
M1a, M1b, M1c
M1a, M1b, M1b
M1a, M1b var
M2
M2*
M3
M1b, M1b, M1b, M1b
M1d, M1c, M1d, M1c
M1b, M1b, M1b
M1d, M1c, M1d, M1c
M1+
28-301
30-32
M4, M1b
M1e, M1e, cadence
mm. 33-34
mm. 35-36
mm. 37-40
mm. 41-48
mm.
49-521
52-531
motifs
M1a, M1b, M1b, M1b
M1a, M1c
53-54
55-56
M2
M2*
57-581
58-647
M1d, M1c
M5, M1+
motifs
M1, M2, M3
M1+, M4
M4+, M3, M1+
M1, M2, M5
M1+, M4
harmonic function
tonic subdominant
dominant
modulating
tonic
tonic
modulating
B major
541
that set off from a note outside the home-key chord: The fugue in F major
(WTC II/13) begins on the leading note, thus approaching the key note
from below, while this fugue in B major begins on the second scale step,
preceding the tonic from above.
Phrasing within this subject is regular. Four segments, each of onemeasure extension, are grouped in the pattern of figure/sequence, figure/
sequence, with the effect that, despite a lack of outstanding rhythmic or
melodic features, the subject is easy to remember even for listeners. The
rhythmic pattern within the subject consists exclusively of eighth-notes.
Throughout the fugue, these are joined by quarter-notes and occasional
longer note values, but the overall effect remains one of simplicity. An
analysis of the pitch pattern reveals that the first subphrase and its varied
sequence are based on broken-chord patterns (see, e.g., B-F-D in mm. 1-2,
with B ornamented by a written-out turn). The third subphrase and its
sequence, by contrast, are composed as three-note descents of iambic note
repetitions. There is thus an
underlying structural skeleton with a distinct rhythm
behind the apparent uniformity of eighth-notes. Highlighting this may present a
challenge for performers:
The harmonic design of the subject is also simple. Although slight
differences occur in the course of the piece, the progression consists
mainly of tonic (mm. 1-31), subdominant (mm. 32-41), dominant-six-four
(m. 42), dominant-seven (m. 43), and tonic (m. 51). The dynamic outline
follows the dance-like design: the first two subphrases increase slightly,
while the next two provide the corresponding relaxation.
The B major fugue encompasses ten subject entries.
4. mm. 21-251 L
7. mm. 47-511 L
1. mm. 1-51 M
5. mm. 32-361 M
8. mm. 54-581 M
2. mm. 5-9 1 U
6. mm. 40-441 U
9. mm. 63-671 U
3. mm. 13-171 L
10. mm. 78-821 U
Apart from an interval adjustment in the answer at m. 53, the subject remains unchanged. Only the final statement features harmonic variation: it
opens and closes in F major but diverts in between to the minor mode.
542
WTC II/21
Bach gives the subject several companions. In the course of the second
to fourth entry one recognizes two recurring patterns, replace by two others
from the fifth subject statement onward. The first two do not meet the
requirements of true counter-subjects, i.e., to be melodically characteristic
and independent from the subject. Their recurrence, however, requires
them to be mentioned. They will be referred to in lower-case letters, as a
reminder of the above reservations: In its most basic form, cs1 consists of
nothing but a protracted dosido formula. As such it appears in M: mm.
14-171 and, with a diverted resolution, in U: mm. 22-25, while mm. 62-91
function as a precursor. Introduced in U: mm. 14-171, cs2 doubles the
subjects second to fourth subphrase in parallels of differing distance.
CS3, the first full-fledged counter-subject, is first heard in U: mm.
33-36. Consisting of three sequential ascents whose syncopations avoid the
subjects phrase-cuts, its dynamic equivalent is a protracted increase of
tension up to the final tie (see, e.g., m. 353), followed by a relaxation that
is either explicit (as in m. 36: G-F) or implicit (as in m. 44). CS3 recurs
faithfully in every further subject statement, although with several modifications including an abridgement at the beginning, a belated resolution,
and a transposition a fifth down. CS4 is paired with CS3 in a manner
similar to cs1/cs2. Introduced simultaneously in the lower voice, it moves
in half-notes and dotted half-notes which sound like a widely spaced
cadential-bass pattern. A dynamic rendering would have to consider the
inherent tension between harmonic steps, with the result that the climax of
CS4 coincides with that of the subject, once again restricting polyphonic
independence. Occasionally shortened like CS3, CS4 also recurs in each of
the remaining entries. Surprisingly given its similarity with a cadential
pattern, it not only wanders through all three voices but also changes its
pitch position with reference to the subject: while in mm. 33-36 it confirms
the subject statements F-major key with a pattern in F major, it ends
rooted in D in entries of quite different harmonic definition, like those in
mm. 40-44 (B major) and 47-51 (G minor).
All entries except for the first are followed by a subject-free passage:
E4 mm. 36-401
E7 mm. 58-631
E1 mm. 9-131
E2 mm. 17-211
E5 mm. 44-471
E8 mm. 67-781
E3 mm. 25-321
E6 mm. 51-541
E9 mm. 82-93
With regard to their material and harmonic formulas we can ascertain
the following subdivisions:
= mm. 58-601-631
E2a/E2b = mm. 17-191-211 E7a/E7b
E3a/E3b = mm. 26-291-321 E9a/E9b/E9c = mm. 82-861-901-93
B major
543
This fugue challenges interpreters more than many others with the task
of distinguishing episodes from thematic phrases since the secondary
passages contain so many fragments of the subject and its counter-subjects.
Thus E2 begins like an extension of the preceding entry, and E5 takes up
the subjects first three measures including the early quasi-contrapuntal
companions. E8, quoting the two distinct subject segments immediately
after the return modulation to the home key, gives the impression of a false
entry. Other episodes establish analogies among themselves. Thus E6 and
E7 are remotely related to E2a (mm. 17-191), E9b (mm. 86-901) is reminiscent of E3a (mm. 26-291), and E9c (mm. 90-93) is an exact transposition of
E3b (mm. 29-321). The role each episode plays in the dynamic design of
the fugue is determined by the rising or falling direction of the sequences
and secondary lines.
The simple rhythmic pattern, the ornamental concept of the subject,
and the comparatively low degree of contrapuntal activity all indicate a
playful fugue of rather lively character. The tempo may be fairly swift.
Feeling a whole-measure pace brings a good result, but care should be
taken that the slurred pairs retain their traditional active-passive or
heavy-light patterns and do not sound hammering. The relative tempo of
the prelude to the fugue may establish a proportion between the larger
beats: half a measure in the prelude corresponds with one measure in the
fugue. (Approximate metronome settings: dotted quarter-notes in the
prelude = 84, quarter-notes in the fugue = 126.)
The articulation includes non legato for the quarter-notes and other
longer note values, a crisp quasi legato for the eighth-notes outside the
slurred patterns, and true legato for the slurred note-pairs in the second half
of the subject. Note, however, that Bach indicates these slurs only once.
According to the conventions of Baroque performance practice, it was the
interpreters responsibility to act accordingly in all corresponding cases. In
the case of this fugue this entails that analogous note pairs must be
articulated analogously. This regards the second half of all subject entries,
their recurrences in episodes, and the parallel segments from cs1.2
2
544
WTC II/21
Phrasing is easily overlooked between the first two eighth-notes in L: m. 17, M: m. 18,
L: m. 36, L: m. 39, M: m. 52, U: m. 58, L: m. 61, L: mm. 70, 72, 74, 75, and M: mm. 83-84.
Conversely, the first two eighth-notes must be slurred owing to their harmonic relation in
U: m. 19, L: m. 20, M: m. 37, M: m. 39, M: m. 51, and L: m. 59.
B major
Section I
M
U
L
L
545
B major
F major
B major
F major
tonic
dominant
tonic
dominant
dominant
tonic
tonic parallel
Section III
M
E major
subdominant
U
c minor
subd. parallel
(L
false entry in episode
U
F major
dominant
Section II
M
F major
U
B major
L
g minor
redundant entry
546
WTC II/21
548
WTC II/22
Its first subphrase [a] begins with a gentle curve in eighth-notes before it
turns to quarter-notes. Reading the figure D-C-B-A-B as D + turn figure
on B, one recognizes that the whole subphrase consists of broken chords
representing a complete harmonic progression: D-B-F, G-E-C, F-D-(B).
The harmonic progression ends at m. 31. That the D ends the phrase
despite marking the center of a broken triad will only become apparent in
two later statements, in which the two subphrases are entrusted to different
voices (see from m. 42-52: L/M, M/U). The overall descent in subphrase [a]
is dynamically best represented in a continuous diminuendo. The second
subphrase [b] begins with the rising octave leap followed by what could be
described as an ornamentation of the simple group B-C-B. This segment
is once again harmonized as a complete progression. The tension grows
toward the last downbeat, to be released in the final ornamented resolution.
(On the level of minute shaping, the crescendo evolves in two staggered
increases, rising twice through the sequenced tripartite note repetitions and
climaxing on the next downbeats appoggiatura before one brief and one
longer relaxation).
ii
VI
iv
B minor
549
550
WTC II/22
Of particular impact are the rests. All three are equal in duration and metric
place: a quarter-note each falling on the first half of one of the weaker
beats in a measure (see mm. 13, 23, and 42). These interruptions in the
melodic flow raise a crucial question regarding the structure of the subject:
Do all or only some of the rests mark the end of a subphrase? In the case of
the first rest, this seems easy to answer. This silence cuts into an ascending
line (A: mm. 1-2: B-C-D-E). Complemented by a falling tritone and its
resolution, which returns to the keynote B, the ascent presents the first
half of a perfect curve that exerts a strong unifying force. The rest in m. 1
thus has tension-sustaining function. Conversely, the silence following the
return to the keynote in m. 23 marks a breathing point after the end of the
first subphrase. Finally, the third case is ambiguous. On the one hand, the
figure built by the three notes at the beginning of m. 4 depicts a releasing
gesture and thus allows the consideration of another breath. On the other
hand, mm. 3-4 can be identified as a varied sequence of the first subphrase,
preceded by a quarter-note upbeat and augmented so as to extend into the
quarter-note on the next downbeat:
Bachs subject
its skeleton
tritone + resolution
tritone + resolution
ii
i vii7 I
vii7 i
ii
vii7 i vii7 V7 i
B minor
551
The subjects real answer is either identical with the original (as in T:
mm. 17-21) or differs only in its major-mode ending (as in S: mm. 5-9).
Other modifications concern the shape of the phrase ending, extended into
552
WTC II/22
B minor
553
554
WTC II/22
B minor
555
556
WTC II/22
The development of tension in this fugue presents itself as a mighty increase in consecutive sweeps. After two overall crescendos, I < II, III < IV,
section V begins in greater intensity than any of the preceding sections and
gains even more dynamic momentum in its episodes rising sequences.
Only then does the tension abate somewhat toward the second mixed
stretto. The deceptive cadence should surprise listeners, and the coda may
outshine everything.
558
WTC II/23
In the lively tempo one is likely to choose for this piece, the number of
ornaments required may pose a problem. It helps considerably to imagine
all trills, slides, and appoggiaturas as fully-written note values. Not only
does that take away some of the fear, it also clarifies their metric position.
This is important since trills in the left hand should have the same speed as
those in the right and slides like the one in m. 23, which is part of a
melodic figure, should be transferred to all recurrences of the same figure,
in this case to the two sequences. (A similar slide in m. 263 is spelled out,
probably because of the required accidental.)
mm. 1-2
B major
559
hand) and the middle beat (right hand) respectively. The following longer
development is strung together by a peak-note curve in the right hand (see
the ascent/descent in mm. 3-6/7-121: C-D-E-F / A-G-F-E-D-C-B-A).
The left hand sets off with a hint of independence but then resigns itself to
mere accompaniment. In this homophonic pattern, the climax thus falls on
m. 71. In section VI Bach uses the same components for a somewhat different pattern, as seen in the right hands descending line in mm. 38-451.
The best dynamic renderinganything but easy given the long breath
neededwould be a continuous diminuendo from m. 38 to the end.
Section II begins with four measures in which the two uppermost
voices of a texture that is momentarily enriched to three parts play melodic
eighth-note groups forming a superimposed line. Section IV, the shortest
of the prelude with only five measures, is the melodically most intense. It
stands out even more than it would elsewhere because the sections surrounding it, III and IV, are composed in toccata style and thus form no
melodic lines, neither directly nor indirectly. By contrast, section IV is
conceived in consistent three-part texture and even presents a small motif.
In its upper voice (mm. 24-25), a richly embellished falling triad leads into
an octave leap before being complemented by relaxing eighth-notes in the
middle voice. The combination is sequenced, at first exactly, then with an
extension. After more sigh motifs and a quasi-cadential trill, an imperfect cadence concludes this section.
The following music example shows the lines in the sections discussed
abovethe preludes skeleton.
mm.
1 1
mm. 12
13
14
15
16
17
10
11
12
560
WTC II/23
mm. 17
18
mm. 23
24
mm. 28
29
mm. 33 34
19
20
21
25
30
35
26
31
36
22
37
32
38
39
23
27
28
41
42 43
33
40
44
45
46
B major
561
562
WTC II/23
The lack of modification in the subject itself leaves ample room for
regular contrapuntal work. Bach has invented three counter-subjects for
this fugue. Two of them appear exclusively in the first round of entries,
while the third accompanies the subject for the remainder of the fugue and
also pervades some of the episodes. CS1 is introduced in B: mm. 5-81 with
a female ending extending until m. 91. Its trunk consists of three measures
with similar patterns, laid out in sequences. The first subphrase (mm. 5-61)
begins with an upbeat to a syncopated half-note and concludes with three
eighth-notes in stepwise descent, thus describing a perfect curve. The
following two subphrases represent ascending sequences, each of them a
fourth higher than the previous one. The entire group should therefore be
rendered as an increase in intensity. Phrasing between the subphrases may
or may not be enhanced by actual cuts in the sound flow. Its expression by
dynamic means (relaxation followed by a new active rise) is, however,
crucial. The tripartite body is finally complemented by a female ending
that prolongs both the final harmony and the melodic descent at the end of
the third subphrase. CS1 recurs three times, once in each of the four voices
in ascending order following the textural rise of the subjects statements
(T: from m. 10, A: from m. 14, S: from m. 19). CS2 is first heard in B:
mm. 10-141. It also begins with an upbeat to a syncopation. But both this
note and its resolution are much more extended than in CS1, so that the
first subphrase spans two measures. A second subphrase follows with an
increase to m. 131 and a decrease to the final note.
These two early counter-subjects are designed to draw the listeners
attention to the subjects female ending. This is particularly obvious in the
case of CS2, which modifies the harmonic resolution on the final note of
the subjects trunk in favor of an interrupted cadence: see, e.g., m. 131
where the expected conclusion in B major is thwarted by the G in the
bass, which converts the chord into a G-minor triad (vi of B major) and
postpones the resolution to the final note of the female ending in m. 141.
or
B major
563
--
---
--
---
--
--
An identical process occurs in mm. 17-18. Interpreters who have opted for
shaping the subject along harmonic lines, i.e., with the climax on the
subdominant and the final tonic as a dynamic resolution, will find Bachs
diversion to the deceptive chord particularly rewarding to play.
CS3 is introduced in S: mm. 28-301 and accompanies almost all of the
subjects further entries. It consists exclusively of eighth-notes winding
downward in a pattern that can be read as an ornamented imitation of the
subjects central segment. This imitative relationship determines the dynamic design in this counter-subject, allowing for the same two solutions
that are possible in the subject:
or
564
WTC II/23
B major
565
566
WTC II/23
B major
567
568
WTC II/23
When comparing the level of intensity in the first half of the fugue, the
second section appears somewhat lessened in comparison to the first. This
is due, on the one hand, to the fact that the entering order B T A S B
creates a stronger impact than that of its structural sequence T A S B T. On
the other hand and perhaps more importantly, section I features its five
subject statements surrounded by two independent counter-subjects, while
section II only contains a single regular accompaniment of the subject.
What is more, CS3 is entirely dependent on the subject in terms of both
pitch pattern and dynamic design. Last but perhaps not least, the change of
mode toward the end of section II also softens the increase in this section.
Within the second half of the fugue, the converse process can be
observed: section III contains longer episodes in contrasting color than
section IV, thus suspending the rise of tension between consecutive entries
more. Furthermore, the bass entry in section III is not accompanied by any
counter-subject, thus forfeiting some of its climaxing power. By contrast,
the corresponding soprano statement in section IV comes supported by
CS3 in the tenor and a varied version of CS3 in the alto.
The dynamic rendering of the B-major fugue should thus aim at
depicting the two-fold symmetry as eloquently as possible: the large-scale
structural analogies between sections I/II and III/IV and the harmonic axis
symmetry of the fugue with the correspondences of sections I/IV and II/III.
570
WTC II/24
On the basis of these analogies, the prelude can be described as comprising four sections:
I
mm. 1-16 M1 + imitation on the tonic (B minor)
II mm. 17-40 M1 + imitation on the tonic relative (D major)
and M1 + imitation on the subdominant (E minor)
III mm. 41-58 M1 on the dominant (F minor)
IV mm. 59-66 M1 on the tonic (B minor)
Given that the composer has prescribed the character he desired for
this piece, the first questions performers usually face when dealing with
Bachs music are already answered: the character is lively. Almost all
16th-notes are written-out ornamental figures. There are turns in different
metric positions (U: mm. 2 and 21-23, L: mm. 1-3, U: mm. 10 and 12),
inverted turns (mm. 9 and 11), and off-beat trills with diverting suffixes
(mm. 7 and 13-16). Regarding the articulation of the three note values
determining this piecequarter-notes, eighth-notes, and 16th-notesthe
distinction characteristic for compositions in lively character between the
faster and slower note values thus concerns the eighth-notes and quarternotes. The eighth-notes should be played in a light, transparent legato
touch, while the quarter-notes are mostly non legato. There are, however,
a number of exceptions: the light legato of the eighth-notes should be
interrupted where escape notes separate slurred note-pairs (thus in U: m.
26, the F does not form part of the main melodic line). And quasi legato
must give way to dense legato where this is explicitly indicated, as in U:
m. 58. Particularly easy to overlook are interruptions of the eighth-note
flow owing to phrasing. This is particularly crucial in U: mm. 13-16 after
the downbeats and in mm. 23 and 61 before the final eighth-note where the
last of three almost identical sub-phrases must be separated from the
typical upbeat + appoggiatura-resolution group. In the quarter-notes, the
usual non legato touch must be suspended for all note-pairs constituting
appoggiatura-resolution.1 Other quarter-notes to be played legato are those
in the traditional dosido formula (U: mm. 65-66). The quarter-notes that
are marked with wedges (U: mm. 21-22 and 59-60) are not necessarily
shorter than those without but heavier, pointing toward an overall climax
that, after having been twice avoided (in the downbeats after the wedged
notes), is finally reached in the appoggiatura-resolution group after the
third sub-phrase.
1
B minor
571
572
WTC II/24
B minor
573
574
WTC II/24
The fugue comprises ten subject statements, one of them incomplete:
1. mm. 0-61
M
6. mm. 44-501
L
2. mm. 6-121
U
7. mm. 54-601
M
3. mm. 15-211
L
8. mm. 70-761
L
4. mm. 26-321
M
9. mm. 81-871
U
5. mm. 35-411
U
10. mm. 96-981
M incomplete
Apart from the usual adjustments of the first interval in the answer, the
only modification the subject undergoes is the varied and extended upbeat
in m. 70. Neither inversions nor grouping into strettos or parallels occur.
Moreover, the polyphonic density never surpasses that of two thematically
shaped voices. While there are two counter-subjects, they never coincide.
CS1 only accompanies the subject in mm. 6-12, 15-21 and, partially, in
mm. 26-32. As soon as CS2 emerges, CS1 breaks off and disappears. (This
retreat of CS1 at the arrival of CS2 is repeated in m. 73). CS1 begins with
a varied imitation of the subjects first subphrase, continues in a trill with
anticipated main note and ascending resolution, and ends after a linking
scalar descent with the sequence of the same trill. Dynamically, the
imitating segment follows the design in the subjects first subphrase and
the trill prolongs the relaxation. The descending scale then serves as a new
upbeat, complemented by the second trill with a second decrease. CS2 is
first presented in L: mm. 29-321. Its features are most unusual for a contrapuntal figure in Bachs polyphonic style. Moving in even 16th-notes that
zigzag in large leaps, it resembles accompaniment patterns in homophonic
textures. Its hidden two-part design shows a bass line BAGC-F
below an ornamented descent A-G-F. The dynamic layout with a single
decrease presents no polyphonic contrast whatsoever to the subject.
B minor
575
E2
E3
E4
E6
E7
Ms/Mcs1
Mcs1a
M1/M1/Mcs2
M1/M1a/M1b M2/M2a
contrapuntal imitative imitative +
contrapuntal
imitative
accompaniment
E5
link + M1
E8
Msb/Mcs2a
contrapuntal
E9
E4 + link + E6 + S/CS1 incomplete + cadence
imitative
contrapuntal
accompaniment
576
WTC II/24
77 79 87 88 89 90 92
93 94 95 96
U
U
M
O
M
O
M
M
M
M
M
G-A F-E C-B F-E B-A E-D G-F F-E A-G G-F B-A
B minor
577
As mentioned above, the first three subject entries are accompanied exclusively by CS1, the fourth statement and all further full entries by CS2. This
change defines the boundary between the first and the second sections. The
entry in L: mm. 70-76 is singular since its beginning is not only varied but
anticipated in another voice (see M: mm. 69-711). The sequence of this
anticipation makes one believe to hear CS1 revived after a long absence.
But the remainder of the subject is accompanied by CS2, the countersubject on duty in this portion of the fugue. This exchange is reminiscent
of that which in M: mm. 26-32 opens the second section.
578
WTC II/24
relationship of the episode material in E2 and E3 allows for no real
color contrast. The tension thus rises gradually, subdued only
briefly in E1 and in the dynamically decreasing E2 with its
descending sequences.
Section II begins, overlapping with the end of section II and
growing out of the ascending sequences of E3, on a fairly elevated
level of tension. As soon as CS2 emerges and CS1 is abandoned
(which means: at the moment when a polyphonic companion is substituted by a virtuoso accompaniment pattern), the earlier intensity
gives way to extroverted playfulness. This is supported by the
episodes, which invite contrasts in color owing to their unrelated
material, and contrasts in mood, owing to their graceful suspensionresolution figures. Moreover, the tonal definition of the subject
statements enhances this extroverted appeal insofar as two of them
are in the major mode (see U: mm. 35-41 and L: mm. 44-50).
Section III brings a return to the more intense character of the first
section. Beginning in a setting of reduced voices and without any
overlap with the previous material, it is launched as a genuine new
beginning. In the further course of this section, the return of the
counter-subject posing as a homophonic accompaniment pattern
(CS2) and of the episodes in contrasting colorboth characteristics
of section IIsuggests a return to the more outgoing mood. E9
combines a diminuendo in the descending sequences of the original
E4 with a crescendo in the ascending lines of the original E6. This
inverted dynamic curve is then topped by the incomplete subject
statement that retrieves, for the last time, the initial phrase of CS1.
The fugue thus ends in a very confident mood.
Further Reading
Benary, P., J. S. Bachs Wohltemperiertes Klavier: Text, Analyse,
Wiedergabe, Aarau 2005.
Bergner, C., Studien zur Form des Wohltemperierten Klaviers von
J. S. Bach, Neuhausen-Stuttgart 1986.
Bodky, E., The Interpretation of Bachs Keyboard Works, Cambridge,
MA. 1960.
Borris, S., Das Bearbeitungsverfahren bei den 11 kleinen Prludien im
Friedemann-Bach-Buch, in Die Musikforschung 5 (1952), pp. 50-52.
Brandt Buys, H., Het Wohltemperirte Clavier van J. S. Bach, Arnhem
1955.
Buchmayer, R., Cembalo oder Pianoforte, in Bach-Jahrbuch 5 (1908),
pp. 64-93.
Constantini, F.-P., Zur Typusgeschichte von J. S. Bachs Wohltemperiertem Klavier, in Bach-Jahrbuch 55 (1969), pp. 31-45.
Czaczkes, L., Analyse des Wohltemperierten Klaviers, Wien 1956/1965.
Dadelsen, G., Zur Situation der Auffhrungspraxis Bachscher Werke,
in R. Brinkmann (ed.), Bachforschung und Bachinterpretation heute.
Wissenschaftler und Praktiker im Dialog, Kassel 1981.
Dahlhaus, C., Bemerkungen zu einigen Fugen des Wohltemperierten
Klaviers, in Bach-Jahrbuch 41 (1954), pp. 40-45.
Dannreuther, E., Die Verzierungen in den Werken von J. S. Bach,
in Bach-Jahrbuch 6 (1909), pp. 41-101.
David, J. N., Das Wohltemperierte Klavier: Der Versuch einer Synopsis,
Gttingen 1962.
Dreyfus, L., Matters of Kind: Genre and Subgenre in Bachs Welltempered Clavier, Book I, in P. Brainard u. R. Robinson (ed.),
Festschrift fr W.H. Scheide, Kassel/Chapel Hill 1993, pp. 101-119.
Drr, A., Zur Frhgeschichte des Wohltemperierten Klaviers I von J. S.
Bach, Gttingen 1984.
Drr, A., J. S. Bach Das Wohltemperierte Klavier, Kassel 1998.
Emery, W., Bachs Ornaments, London 1953.
Engels, M. W., Bachs Well-tempered Clavier: An Exploration of the 48
Preludes and Fugues, Jefferson, N.C. 2006.
579
580
Further Reading