FUGUE 25 (BOOK II, 1)

C MAJOR (THREE VOICES)

- Exposition: the subject is announced in the alto.

The answer is tonal, and is given to the treble. It will be noticed that
as Bach treated the first note of the subject as the dominant of C,
not the tonic of G, the auxiliary note, F, lay a whole tone below it,
while in the answer the corresponding interval is a semitone. There
is no regular countersubject, and no codetta. The exposition ends on
the first semiquaver of bar 13, at the completion of the subject in the
bass.
Middle Section. In the first episode (bars 13 to 21) the second
section of the subject is worked as a bass to a free canon founded
on its first section. In bar 21 the subject enters in the alto,
modulating into D minor, and followed by the answer in the treble,
modulating to A minor. The second episode (bars 29 to 39) is for
treble and alto alone, beginning with a new treatment of the material
of episode I, and developing (from bar 33) into a canon in the fifth.
Final Section. The entries, which are technically the final group,
appear rather early in this fugue. The answer in the bass (bars 39 to
43) is separated from the two succeeding entries by a third episode,
reaching to the beginning of bar 47, and furnishing new
counterpoints to the semiquaver figure. The subject in the alto (bars
47 to 51) and answer in the treble (bars 51 to 55) complete this
group. The fourth and longest episode (bars 55 to 68) separates this
from the coda; it begins with a free transposition of episode I to the
fifth below, with a new continuation from bar 62, still built on the
semiquaver bass. The coda contains four partial entries of the
subject, in bars 68, 75, 76 and 79; a fourth voice is added from bar
80.
Great unity of character is given to this fugue by the derivation of
nearly the entire material from the subject itself.

WTC II/1 in C major Fugue

II/1.2.1 The subject
With its length of four bars, divided by a rest exactly in its middle, the subject of the C
major fugue reveals a very regular phrase structure. The beginning after an eighth-note
rest converts the first bar into an upbeat to bar 2; similarly the third bar, with its
downbeat rest, also serves metrically as an upbeat to bar 4, after which the phrase
concludes on the first sixteenth-note of bar 5.
The question whether this phrase consists of two subphrases or one indivisible unit with
a tension-sustaining rest in its middle, allows for two answers. These have to be in
keeping with the interpretation of several other features in the subject of this fugue. Let
us look at these features first.
The pitch outline develops in a fairly restricted range, between the middle C as the
lowest and the sixth (A) as the highest note. The beginning on the fifth gives the subject
a hint of being already in the middle of things, while the conclusion on the third has a
gently releasing quality. The pitch pattern contains two accented inverted-mordent
figures (see bars 1 and 3) and an unaccented one (see bar 4). These written-out
ornaments suggest that the remaining sixteenth-notes (in the subject, and in the fugue as
a whole) should also be interpreted as ornamental rather than melodious in their
intent. The two larger intervals within the subject appear as consecutive jumps (see bars
1/2), thus confirming the lively character. The target of these consecutive jumps, the A
on bar 2d, is supported in many ways: melodically the highest pitch, rhythmically the
first of two longer notes, and harmonically representing the subdominant, it is further
emphasized by an ornament in full speed (see the inverted-mordent symbol) before it
relaxes slightly in the following step downward.
One could now perceive bars 3-5 as a varied sequence of bars 1/2: (GF)G-C-A-G would
have become an ornamented (FE)F-(E)-D---E. In this view, the subject appears as two
(fairly balanced) subphrases. The rest in bar 3 must then played as an interruption
before a new beginning, the written-out inverted mordent following it (see bar 3 F-E-F)
would be more active than the fairly relaxed G in bar 2, and the D on bar 4d would
come out as a second (though softer) climax.
Another view is equally possible - and perhaps more conducive to the transmission of
overall unity in the subject. The entire string of sixteenth-notes in bars 3-5 can be
regarded as ornamental. One would then define the main melodic steps in the subject as
(GF)G-C-A-G-F---E. In this case, the gradual descent from A to E requires an
uninterrupted line. The rest in bar 3 is then perceived as tension-sustaining, after a G
which sounds only minimally softer than the preceding climax. The A-G and the
tension-sustaining rest would thus allow for a further release through the following F
(and all its gradually retreating ornamental surroundings) to the final E. In other words,
this interpretation renders the subject as an uninterrupted unit.
The harmonic background of the subject does not do reveal anything that would decide
the matter. The active step to the subdominant appears on bar 2d - but then we never had

any doubts about the overall climax anyway. Bar 4 represents the relative minor to the
subdominant, which is then followed by the dominant and the final tonic (ex. 10).

II/1.2.2 The statements of the subject

The three voices of the C major fugue present altogether eight subject entries:
1.

bars 1-5

5.

bars 25-29

2.

bars 5-9

6.

bars 39-43

3.

bars 9-13

7.

bars 47-51

4.

bars 21-25

8.

bars 51-55

(ex. 11)

Apart from the interval modification in the answer (where the falling fifth in the first bar
becomes a fourth) the subject does not experience any changes; neither does it appear in
any stretto or parallel setting.

II/1.2.3 The counter-subject

After concluding the subject, the middle voice presents a melodic line against the
answer, i.e. at a point where one expects a counter-subject to be introduced. In the
further course of the piece, this line recurs twice in its unabridged version (see middle
voice bars 25-29 and 51-55); in two other cases, only the initial bar reappears (see upper
voice bars 9 and 39).
While independent from the subject in structure, this counter-subject is not completely
independent in material. The first two bars (see from the second sixteenth-note D in bar

5) are closely related to the subjects final two bars, and only the counter-subjects
second half with its ascending scale, syncopation and closing-formula contributes new
components. Another shortcoming of this counter-subject is the fact that its unabridged
version remains restricted to the middle voice, appearing always in accompaniment to
an upper-voice subject statement, and thus lacks true polyphonic versatility.
The inner structure of this phrase poses an important question for the contrapuntal
setting in this fugue. Following the pitch pattern of descending sequences in the first
two bars of the counter-subject, 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 towards the syncopation, after which the closingformula brings a relaxation.
With this dynamic outline in mind, let us go back to our two options for the
interpretation of the phrase structure in the subject. 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 the case of a subject consisting of two dynamic
curves, however, this interpretation of the counter-subject is most unlikely as it would
create simultaneous phrase cuts in both voices and thus interrupt the polyphonic texture
in quite an awkward way. (Performers who definitely prefer a divided interpretation of
the subject must therefore render the counter-subject as an unbroken phrase, beginning
with an extended crescendo.)
The two sketches display the options available for the phrase structure and dynamic
design in subject and counter-subject.
(ex. 12)

II/1.2.4 The episodes

The C major fugue contains only four subject-free passages.

E1

bars 13-21

E3

bars 43-47

E2

bars 29-39

E4

bars 55-83

At this point it might be interesting to learn that Bachs first version of this fugue
(according to the Kellner manuscript) ended on bar 68d. The composer later re-wrote
the perfect cadence in bars 67/68 as an interrupted cadence and added sixteen bars on a
pedal note C. This extended coda certainly concludes the piece much more
convincingly. At the same time, the fact that it was not yet there in the earlier version is
of great help for a true understanding of the fugues architectonic design.
The episodes make ample use of material from the subject.
-

E1 contains an imitative pattern based on the subjects first half, complemented by

an eighth-note instead of the rest (see bars 13-19, upper and middle voices; the
final imitation in the middle voice is inverted and varied). This pattern recurs
faithfully in bars 55-61. In both instances, the lower voice adds a figure which is
derived from the last bar of the subject and/or from the first bar of the countersubject. The lower-voice sixteenth-notes then continue, with a little more liberty,
until the end of the episode (bar 21 downbeat) and until the final cadence of the
earlier version (bar 67 downbeat) respectively.

The head of the subject occurs further in the coda where it is imitated through all
three voices (see bars 68-72: L, bars 72-76: M, bars 76-80: U). The first eight bars
are again accompanied by the sixteenth-note figures from subject ending and/or
counter-subject beginning.

In E2, the first segment contains a two-bar motive which is only in its beginning
rhythmically related to the subject head (see bars 29-31: U, sequenced in bars 3133); this is again accompanied by the string of thematic sixteenth-notes. The
second segment of this episode brings an imitation built exclusively on the
sixteenth-note pattern (see bars 33-37d: U, bars 34-39d: M).

E3 displays a lower voice which is even more closely related to the end of the
subject as it sequences the final bar (see bars 43-46: L).

Apart from the final bars of the coda, the only components of the episodes that are not
related to the primary material appear in the upper and middle voices of bars 43-47 and

61-67. They do, however, recall the other episodes insofar as they also contain an
imitative pattern in sequences.
The role each episode plays in the dynamic design of the fugue is easily determined in
accordance with the sequential patterns. E1 builds up tension through bars 13-19 but
then brings a slight release in bars 19-21. E2 keeps a very low profile in its constantly
falling lines. E3, on the contrary, ascends continuously in all three voices and thus
produces a considerable increase. E4, like E1, begins with rising lines and a crescendo
(bars 55-61) followed by a gradual descent (bars 61-65) and a cadential release - which
now, in its revised version as an interrupted cadence, expresses much more intensity
than did the earlier perfect cadence. The coda begins in a fairly leveled softer shade,
followed only in bars 80-83 (see the split voices) by a final strengthening.
The uniformity and simple relationship throughout both the primary and secondary
material adds to the playful character of the fugue. This character is further enhanced by
rhythmic continuity: before the final cadence, the sixteenth-note pulsation is almost
constant, with minimal interruptions only on four occasions (see immediately after the
downbeat in bars 11, 21, 22 and 41).

II/1.2.5 Character, tempo, articulation, ornament realization

After all that has already been observed, the basic character of this fugue can no longer
be a mystery. The simplicity of the rhythmic pattern, the ornamental structure of the
pitch line and the jumps that occur both in the subject and in the episode material clearly
express a rather lively character. The tempo, too, should be fairly swift. The articulation
that follows requires non legato for the eighth-notes and quarter-notes, and legato for
the sixteenth-notes.
The relative tempo of the prelude to the fugue should be chosen in complex proportion.
The reason lies in the prevailing patterns of almost continuous sixteenth-notes in both
pieces which would sound dull if the respective pulses were in simple proportion. A
good and feasible solution is to translate each of the four quarter-notes in the final bar of
the prelude into triplet eighth-notes (instead of the previously felt four sixteenth-notes),
and to turn these imagined triplet eighth-notes - after having freed them of their
grouping into lots of three - into the eighth-notes of the fugue (Approximate metronome
settings: prelude beats = 72, fugue beats = 108). In other words:
corresponds with
an (assumed) triplet eighth-note

an eighth-note

in the prelude

in the fugue

The prominent ornament (now talking about those notated in symbols) is the inverted
mordent in the subject. Its pitch does not pose a problem as it always uses the natural
(white key) for its lower neighbor note. As is the case with all ornaments which form a

characteristic feature of a fugue subject, this inverted mordent must be played even
where the composer did not write it again (i.e. in bars 22 and 26). By contrast, the
episode motives deriving from the subject head need not be ornamented as Bach does
not indicate this in a single case. (The presence or absence of the ornament has the
further advantage that it tells listeners already after one bar whether what they are
hearing is going to be a subject entry or a subject-related motive - a distinction which,
due to the uniformity of material in this fugue, would otherwise not be so easy to make.)
Another ornament appears in two of the three statements of the counter-subject. In bar 8
it is printed as an inverted mordent, while bar 28 features a mordent symbol without the
slash. The first reading is slightly confusing in a typical closing-formula where the
experienced performer might have added a full trill even without any ornament symbol
from the composers hand. Thus the second reading appears more convincing and can
safely be chosen in both cases. As a leading-note trill approached stepwise it begins
with a sixteenth-note on the main note, followed by six thirty-second-notes including
the suffix, and resolving smoothly and without interruption on the following downbeat.
In bar 54 the closing-formula is varied and needs no ornament.
Finally, the long trill with a tie-suspension in bars 37/38 begins regularly, i.e. from the
upper note; it then shakes in thirty-second-notes until the very end of the bar where it
ties the last F over to the next downbeat. No suffix is possible in this case since the
ornamented note lacks a resolution.

II/1.2.6 The design of the fugue

Cadential formulas together with the key of the subject entries and the dynamic build-up
in some of the episodes determine the structural layout in this fugue. Only the end of the
initial section might cause some doubt. In bars 21/22, Bach presents the first closingformula with typical features in both the upper and the lower voices, thus creating a
strong feeling of closure. The fact, however, that the beginning of a subject statement in
the middle voice overlaps with this closing-formula for an entire bar strings the first and
second sections closely together.
As the number of voices is reduced to two after the cadence in bars 24/25 - and thus
would normally advocate the beginning of the second section only here, one obviously
needs good grounds for a differing view. The reasons which lead to assume the closing
of the first section in bar 22 are as follows:
(a) As all three voices have already presented the subject, the first section can only
end here or after one more - a redundant - statement. Yet the overlapping subject
statement in the middle voice ends with a cadential close which is much less
satisfactory as a close, due to a sudden breaking off in the upper voice (see bar
25d) which creates a strong link between the upper-voice line in bar 24 and the
new beginning in bar 25.

(b) The overlapping subject statement in bars 21-25, while commencing in the

harmonic surroundings of G major, soon reveals its key as D minor (see the C#s
and Bbs from bar 22 onwards). This statement thus forms a pair with the following
upper-voice entry which, from its second bar onwards, is in A minor - the minor
dominant of the preceding statement. A closer look at the beginning of both
statements further strengthens this view as they display the interval structure of
subject + answer. (Compare bars 21/25 with bars 1/5.)

(c) If the subject entry in bars 25-29 was the first of a new section, then the episode
which 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.

(d) The dynamic design of E1 is very much related to that of the original E4 (until bar
68): both describe a full curve, using similar material.
Having said all this, a conclusion for the remaining sections follows without problems.
The harmonically related entries in bars 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 bar
39 marks the beginning of the third section which encompasses 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.
For a sketch showing the design of the fugue in C major, see the following graph (ex.
13).

II/1.2.7 The overall dynamic outline of the fugue

Within both 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, whereas in the
third section, Bach uses two particular devices to create a similar effect: the beginning
in a very low keyboard register (see bars 39-43 in which even the upper voice sounds
mainly below middle C) and the inserted episode with its strongly increasing direction.
The second section is distinguished by three factors which decrease tension: the reduced
number of voices which is valid for all but the first four bars, the minor mode, and the
strongly decreasing direction of the episode.
The coda begins somewhat subdued for the first eight bars but builds up gradually as the
pedal moves to the lower C and the lower-voice figures suggest hidden two-part
structure (from bar 76 onwards). When the other voices finally also split, a fully
resonant close is reached.