WTC II/21 in Bb major - Prelude

from Siglind Bruhn
J.S. Bach's Well-Tempered Clavier
In-depth Analysis and Interpretation

II/21.1.1 The prelude-type

This prelude is composed in polyphonic style. Passages in three-part texture alternate with phrases in two-part texture where the middle voice keeps extended silences. Within each section, part-writing is kept very consistent.

Most of the techniques characteristic of Baroque polyphonic style are used: (strict or free) imitation, (full and partial) sequences of motives and contrapuntal models, inversion of shape, and inversion (exchange) of voices.

The structural design is in two repeated halves, the second of which is considerably longer. The time signature is twelve-sixteenth, a compound meter reminiscent of the gigue. The uninterrupted flow of sixteenth-note and frequent "gigue-rhythm patterns" (eighth-note + sixteenth-note, see e.g. in the first halves of bars 2, 3 and 4 respectively) enhance the impression that this prelude is composed along the lines of the Baroque dance which traditionally closed a suite.

 

II/21.1.2 The overall design of the prelude

The first harmonic progression concludes on the downbeat of bar 3 where the tonic is confirmed in a perfect cadence. This cadence should not be considered as structurally relevant, both because it marks the ending of the very first three-part combination of the main motive, and because the tied notes in the upper and lower parts essentially weaken the conclusion on the downbeat.

Harmonic closes which coincide with structural caesuras occur on the downbeat of bar 17, where the subdominant key Eb major is established, and in bar 32 which confirms the dominant key F major; these two perfect cadences thus divide the first half of the prelude into two segments of almost identical length.

An analysis of the material presented in this prelude reveals several smaller sections which are utterly important for a correct structural assessment of the piece in spite of not being harmonically concluded. It is therefore advisable to describe the design of this prelude on the basis of combined observations of harmonic processes and material presentation.

There are twelve (four + eight) sections in this prelude:

I bars 1- 8end Bb major to F major (imperfect cadence)
II bars 9-17d F major to Eb major (perfect cadence, subdominant)
III bars 17-28d Eb major to F major (imperfect cadence)
IV bars 28-32 F major (perfect cadence, dominant)
V bars 33-41d F major to F minor (perfect cadence)
VI bars 41-48end to G minor (perfect cadence, relative minor key)
VII bars 49-52end Bb  major to Bb  major (imperfect cadence)
VIII bars 53-56end to C minor (perfect cadence, relative of subdominant)
IX bars 57-64end C minor to Bb  major (imperfect cadence)
X bars 65-76d modulating to F major7
XI bars 76-83d remaining rooted in F major7
XII bars 83-87 returning to Bb major (tonic)

Several structural correspondences can be detected:

bars 8end-12 recur in bars 52end-56 transposed

bars 13-17d

recur in bars 37-41d transposed, varied
bars 28-32

recur in

bars 83-87

transposed

Another feature of supreme importance for a quick structural overview is found in the sequential patterns. These appear in such great numbers here that it is worth listing them. The following table includes

(a)   sequences that involve all voices

(b)

sequences of the leading voice
with free contrapuntal work in the other voice(s)

(c)

sequences involving an inversion of voices

model
sequence model sequence

bars 9/10

bars 10-12 bars 33/34 bars 35/36
bars 13/14 bar 15/16

bars 37/38

bars 39/40
bar 19 bar 20 bar 41 bar 42
bar 22d-22m bars 22m-23d, 23d-23m bars 45m-46m bars 46m-47m
bar 23m-24d bars 24d-24m, 24m-25d bar 57 bars 58, 59, 60
bar 25  bar 26 bar 62 bar 63
    bars 65/66 bars 67/68
    bar 70 bar 71
    bar 72 bar 73
    bars 76m-78m bars 78m-80m

II/21.1.3 Practical considerations for performers

The basic character of this prelude is rather lively. This character is revealed above all in the rhythmic pattern which is very simple and, as was already mentioned, contains the typical features of a gigue. The tempo should be chosen accordingly; it should be fast enough to depict the jerkiness of the gregarious dance, but not so rushed as to render the intricate patterns of imitations as a simple technical show.

The prelude contains several kinds of ornaments. As the two compound symbols in bars 7 and 26 indicate, Bach expects four trill notes against each of the sixteenth-note - a good measure for the limit of maximum speed (see ex. 204 below). Besides these, three simple ornaments occur. Two of them are mordents; the one in bar 2 begins on the upper auxiliary and contains four notes, while the one in bar 8 is approached stepwise and therefore commences on the main note, encompassing only a single three-note shake (ex. 32):

The inverted mordent poses no problem with regard to its pitch since it is immediately preceded by D which also serves as its lower neighbor note. The fact that this ornament appears in brackets points to its origin not from Bach's manuscript but from a copy. While it is thus left to the performer to include or disregard the ornament, adopting it is, in fact, a very good idea as it can be of great help for the listener in distinguishing the manifold versions of the main motive, all of which begin with scale portions and are thus difficult to discern. If the inverted mordent is incorporated in bar 1, then a similar ornament should be transferred to all bars which contain the version of the motive in which the sixteenth-note scale segment is complemented by three dotted eighth-notes in reverse motion. Such transferal is then advisable for bar 3 (L), bar 5 (U), bar 8 (L), bar 49 (U), possibly bar 52 (U, i.e. on A), bar 65 (U), and bar 67 (U). What remains for the performer to decide is whether an inverted mordent is preferable in all cases or whether the inverted shape of the motive (see e.g. L: bars 3/4) might not be more convincing if the ornament is also inverted (an "inverted" inverted mordent being, of course, a simple mordent).

 

II/21.1.4 What is happening in this prelude

The thematic material of this prelude is determined by four motives.

M1

comes in numerous guises; what they all have in common is the beginning after a strong beat and the scale section in sixteenth-note which leads to the next strong beat. Five versions appear regularly.
M1a is introduced in the upper voice (see bars 1-2d). It extends over a full bar and complements a scalar descent with an ascent in dotted eighth-notes which may be distinguished by an ornament (see the paragraph above).
M1b, presented as the immediate imitation in the middle voice, contains only the scale in straight motion, whereas M1c, the next imitation (see lower voice: bar 2) features the scalar motion ending in an interval - perfect fifth or fourth - in the opposite direction.
In M1d (see e.g. bars 19, 22), the final strong beat repeats (or retains) the previous pitch, and in M1e (see e.g. U: bars 30/31, L: bars 31/32), the sixteenth-note motion is extended to form a one-bar curve.
With regard to texture, M1 is a polyphonic component which wanders through all voices, is regularly inverted and set in a continually varying contrapuntal environment.
The dynamic shape of M1 is determined by the common feature, the initial scale. Its metrical position as an extended upbeat clearly indicates an increase in tension; in the case of a complement (as in M1a) or an extension (as in M1e), this increase is followed by a decrease and thus forms a complete curve.


M2 is first heard in bars 9/10. It consists of two contrapuntally dependent parts. M2a commences in the lower voice with the two final sixteenth-note of bar 8 and ends with a tied note F two bars later. Its characteristic feature is an ornamented descent (see the strong beats Bb-A-G-F, the first three with written-out inverted mordents). The weak beats in-between are filled by broken chords which, since they are all rooted in a kind of pedal note (C), are gradually narrowed down until they form a scale (see bar 10). M2b in the upper voice complements the lower part with a harmonically adjusted imitation; the ornamented descent Bb-A-G-F falls here on the metrically weak positions. M2 is repeated in inverted voices immediately afterwards (see bars 10end-12); here the upper voice is leading.
Pitch pattern, length and texture of M2 are thus as different as could be from those in M1. The dynamic shape follows the line formed by the peak notes with their written-out ornaments and thus describes a two-bar diminuendo triggered by a two-sixteenth-note upbeat.

M3 consists of a one-bar model which is sequenced three times (see bars 13, 14, 15, 16). The texture is reminiscent of toccata-style. The right-hand pattern can be described as hidden two-part structure (see e.g. bar 13: the melodic descent D-A-G-G before the harmonic "background" of the broken G-minor chord), and the left-hand part accompanies with another two-part pattern, consisting of cadential notes in the bass register interspersed with fourth jumps in the high register. The visual representation of the score seems to indicate imitation after two bars, but this impression is deceptive: no change of voice occurs - the reorganization is one of technique only. (Taken literally, this means that the "upper voice" moves to the bass register and stays there until bar 39! It may be best to assume that consistent part-writing is suspended here.)
The development of tension in this motive follows the four-bar descent in the sequential pattern with a long diminuendo.

M4 is a small motive which first occurs towards the end of the first half, in U: bars 28/29. It consists of a repeated pedal note ornamented by inverted mordents (see the threefold C-B-C), alternating with three-note ascents in rising succession (D-Eb-F, Eb-F-G, F-G-Ab). In its scope of six sixteenth-note and its one-track design, M4 recalls M1. It does not, however, take part in polyphonic development as it is never imitated but restricted to the upper voice. The dynamic shape of this motive corresponds with the direction of its ascending note groups and thus represents a crescendo.

M5 is restricted to the middle of the prelude's second half. Introduced in U: bar 58, with sequence in M, it appears like a compound ornament preceding a strong beat. Correspondingly, its dynamic shape is a slight increase.

The design of the prelude in Bb major can be traced very convincingly from the occurrence and interplay of these five motives. For the sake of clarity, the table below does not specify inversions of motives and eventual irregularities. (An asterisk designates inversion of voices, an apostrophe marks occasional modifications of a motive. For details of sequences, refer back to the beginning of this chapter.)

bars motives   bars motives
1-3d M1a, M1b, M1c
---
49-52d M1a, M1b, M1b, M1b
3-5d M1a, M1b, M1c
---
52-53d M1a, M1c
5-7m M1a, M1b, M1b      
7m-8

M1a, M1bvar

     
9/10 M2
---
53/54 M2
11/12 M2*
---
55/56 M2*
13-16

M3
     
17/18 M1b, M1b, M1b, M1b      
19/20 M1d, M1c, M1d, M1c
---
57-58d M1d, M1c
21-22m M1b, M1b, M1b   58-64m M5, M1dev
22-23m M1d, M1c, M1d, M1c      
23m-28d M1dev   64m-67d M1a, M1a, M1b, M1b', M1b'
      67-70d M1a, M1b, M1b', M1b', M1b
      71-76m M1b'dev
      76m-83d M1dev
28-30d M4, M1b
---
83-85d M4, M1b
30-32

M1e, M1e, close

---

85-87

M1e, M1e, close
 
bars 33/34
    M4 in indirect parallels
 
bars 35/36
    M4 in indirect parallels
 
bars 37-40
    M3
 
bars 41-48
    M1dev

The design of this "prelude in the character of a gigue" thus reveals itself as a remote relative of sonata form. This becomes particularly obvious when one combines structural observations derived from the distribution of the motives with those based on the main harmonic steps in the piece.

  bars motives harmonic basis
exposition section I 1-16 M1, M2, M3 tonic - subdominant
exposition section II 17-32 M1dev, M4 dominant
development section 33-48 M4dev, M3, M1dev modulating
recapitulation section I 49-64 M1, M2, M5 rooted in tonic
recapitulation section II

65-74/83-87

M1dev, M4

rooted in tonic
insertion 74-82  

modulating

(in lieu of coda)

     

 

WTC II/21 in Bb major -- Fugue

II/21.2.1 The subject

Commencing after a eighth-note rest and concluding on the downbeat of bar 5, this subject is exactly four bars long. The beginning is particularly noteworthy because the first note does not form part of the tonic chord. (Among the forty-eight fugues of the Well-Tempered Clavier, there are only two which set off from a note outside the home-key chord: The fugue in F# major (volume II) commences on the leading note, thus approaching the key note from below, while this fugue in Bb major begins on the second step, preceding the tonic from above.)

Phrasing within this subject is perfectly regular. One can detect four segments, each of one-bar length. They are grouped in the pattern of model/sequence, model/sequence, with the effect that, despite a lack of outstanding rhythmic or melodic features, the subject is not difficult for a listener to remember.

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. Bb-F-D in bars 1/2, with Bb ornamented by a written-out turn). The third subphrase and its sequence, on the other hand, are composed as three-note descents linked by anticipations on the preceding weak eighth-notes. This observation is particularly interesting as it reveals an underlying structural rhythm behind the obvious uniformity of eighth-notes; to bring this out might be a special challenge for the performer (ex. 33):

The harmonic design of the subject is also simple; although slight differences occur in the course of the piece, the outline consists mainly of

  tonic subdominant dominant-six-four / dominant-seven tonic
bars 1/2 32-4d 42 43 5d

The dynamic outline follows the dance-like design: the first two subphrases increase slightly, while the next two provide the corresponding relaxation.

 

II/21.2.2 The statements of the subject

The Bb major fugue encompasses ten entries of the subject.

1. bars 1- 5d M 6. bars 40-44d U
2. bars 5- 9d U 7. bars 47-51d L
3. bars 13-17d L 8. bars 54-58d M
4. bars 21-25d L 9. bars 63-67d U
5. bars 32-36d M 10. bars 78-82d U

(ex. 34)

Apart from an interval adjustment in the answer (which occurs only once in the piece; see bar 5 beat 3), the subject remains unchanged and uninverted. Only the final statement contains harmonic variation in that it commences and closes in F major (see bars 78/79 and 81/82) but touches the minor mode (see the Ab) in-between.

 

II/21.2.3 The counter-subjects

There are several companions to the subject. In the course of the second to fourth entry one can recognize, albeit behind strong variation, two recurring patterns; two others substitute them from the fifth subject statement onward. The first two, as will be seen, do not meet the requirements or true counter-subjects, i.e. to be melodically characteristic and independent from the subject. Their consistent recurrence, however, requires them to be mentioned, and they will therefore be referred to in small letters, as a reminder of the reservations.

cs1

consists, in its most basic form, of nothing but a protracted keynote / leading-note / keynote formula. As such it appears in M: bars 14-17d and, with a diverted resolution, in U: bars 22-25. In connection with the first accompanied subject statement in bars 5-9d, the same formula is very much concealed in a variation: the middle voice contains the note F first in bar 6 (second eighth-note) as the beginning of a parallel to the subject's second subphrase. The same F is regained in bar 7, beat 3 where its tie-prolongation recalls the simpler shape of the formula. E, the leading-note to F, appears ornamented with a written-out inverted mordent (see bar 8) before it resolves timely into F.
Although one can state that this element recurs with regularity, it is so obviously a simple closing formula that it cannot be recognized as a true counter-subject, last but not least because it does nothing to "counter" the subject but only supports it. A similar restriction applies to the dynamic shape: while a closing formula traditionally expresses a single gesture of relaxation, effective shaping is not possible owing to the very long note value at the beginning of the formula.


cs2 is introduced in bars U: 14-17d. The first four and the last six eighth-notes form parallels in compound sixths to the subject's second and fourth subphrases respectively, and the subject's third subphrase is doubled in compound thirds. No contrapuntal independence whatever can thus be attributed to this companion. Nevertheless, the same figure recurs in M: bars 22-25d. Interestingly, its first bar alone is also heard in U: bars 54/55 and M: bars 63/64.

CS3

is first encountered in U: bars 33-36. Consisting of three sequential ascents which, by their tied notes, conscientiously seem to avoid the phrase-cuts in the subject, this companion is phonically independent. Its dynamic equivalent is a protracted increase of tension until the final tie (see e.g. bar 35, beat 3), followed by a relaxation which is either explicit (see bar 36: G-F) or implicit (see e.g. bar 44).
CS3 recurs faithfully in every further subject statement, although with several modifications including abridgement at the beginning, belated resolution, and transposition a fifth lower.


CS4

is paired with CS3. Introduced in L: bars 33-36, it moves in half-notes and dotted half-notes which sound like a spaced-out cadential-bass pattern. A dynamic representation would therefore have to consider the inherent tension between harmonic steps and place the climax on the subdominant note (see e.g. the Bb in bar 34) - with the effect that the climax of this counter-subject coincides with that of the subject.
Occasionally shortened like CS3, this counter-subject also recurs in each of the remaining entries. Interestingly, the cadential pattern not only wanders through all three voices but also changes its pitch position with reference to the subject: while in bars 33-36 it confirms the F major key of the subject statement with a pattern in F major, it ends rooted in D in entries of quite different harmonic definition, like those in bars 40-44 (Bb major) and 47-51 (G minor).

As the analysis reveals, this Bb major fugue resembles its counterpart in the first volume of Bach's Well-Tempered Clavier, in that it features very faithful counter-subjects with, however, a stunning lack of melodic eloquence.

The sketch shows the subject in its manifold quasi-polyphonic surroundings. Although the two groups of counter-subjects never occur simultaneously as presented here, it is interesting to view them as parts of a single, almost homophonically designed pattern. Dynamic independence is only gained by CS3 and - theoretically though not perceptibly - by cs1 (ex. 35):

 

II/21.2.4 The episodes

This fugue encompasses subject-free passages following each but the first entry.

E1 bars 9-13d E6 bars 51-54d
E2 bars 17-21d E7 bars 58-63d
E3 bars 25-32d E8 bars 67-78d
E4 bars 36-40d E9 bars 82-93
E5 bars 44-47d    

Taking into consideration the material exposed in these episodes as well as the harmonic closes incorporated in them, one can further distinguish a number of sub-divisions:


E2 consist of E2a (bars 17-19d, with a cadence in Bb major) and E2b (bars 19-21d); similarly, E7 is also divided by a cadential formula.


E3 includes E3a (bars 26-29d) and E3b (bars 29-32d, characterized by an extended cadential pattern with chromatic bass line); similarly, E9 can be partitioned into E9a (bars 82-86d), E9b (bars 86-90d) and E9c (bars 90-93, featuring the same extended cadential pattern with chromatic bass line).

Material from the subject and the counter-subjects is prominently present in the episodes. If, for easier reference, one calls the subject's first subphrase with its ornamented broken chord "a" and the third one with the slurred pairs "b", then the following design can be made evident:

E1 contains a two-bar motive based on a variation of a with a one-bar extension (see M: bars 9-11d, sequenced in 11-13d). Its counterpart is also based on a which is preceded here by a long syncopation. The direction of the sequence in the middle voice is falling while that in the upper voice is rising; the overall effect is thus one of suspended tension.

E2a combines the entire second half of the subject (compare M: bars 17-19d with bars 3-5d) with a parallel in the lower voice which, beginning as an extension of the subject, after one bar gives way to a cadential-bass pattern. As the upper voice is kept in long note values, this episode segment gives the impression of extending, in changed voices, the S + cs1 + cs2 combination heard just before, which it rounds off with a cadential close. The tension in this segment is thus clearly releasing.
E2b employs basically the same material for a new build-up. The increase of tension is enhanced not only by the rising double thirds but, more importantly, by the modulation to the dominant key F major.

E3a is based on a variation of a which, together with its descending sequence and its lower-voice counterpart in long syncopations, recurs immediately in inverted voices. (Only the middle voice remains aloof of the thematic processes.) Tension is only slightly diminishing here.
E3b quotes b and a before it concludes in a cadential pattern. The chromatic bass line adds to the strong release in tension.

E4 is dominated by the declining gesture of b which, occurring first in imitation and then in parallels, creates a strong effect of declining tension.

E5 recalls the first three bars of the subject and its early companions: a in M: bars 44/45 is not sequenced but imitated (U), while the first two bars of cs2 are shared between L: bars 45/46 and M: bars 46/47. The tension is rising, enhanced in bars 46/47 by the ascending lines in U and L.

E6 seems remotely related to E2a. Two bars with b in the middle voice, long syncopations in the upper voice and the characteristic cadential-bass pattern in the lower voice create a strong bond, although the scope of the episode is extended here and consequently includes modifications. The dynamic representation is a diminuendo.

E7a is yet another remote variation of E2a, ending after decreasing tension with a cadential close in C minor on the downbeat of bar 61.
E7b contains both b and a, together with a rising line in the upper voice. It renews the cadential close in C minor but leaves the upper voice unresolved, thus creating an open, forward-directed gesture.

E8

is longer than all preceding episodes and the most intriguing of all. Its first two bars modulate from C minor back to the home key Bb major, thus creating a strong expectancy for a subject entry on the tonic. When the lower voice (where one expects the entry, after recent statements in M and L) begins, in bars 69/70, with a variation of a, the listener must feel reassured - prematurely, as the immediate combination with b to a two-bar entity shows. This two-bar unit, complete with its own accompanying patterns (see M: the quarter-notes in bars 69-71 and U: the inverted mordent and syncopation), is then sequenced twice in descending direction. An extension quoting both a and b and ending with a cadential close in F major completes this episode. The corresponding dynamic development is a long and very gradual diminuendo.


E9a provides yet another false entry. The combination of a, b, b and a one-bar complement to a four-bar phrase appears all the more convincing as the surrounding voices represent CS3 (U) and CS4 (L) respectively, with only very small modifications. Dynamically, this episode segment thus represents the same curve as built in the subject entries.
E9b contains, in the lower voice, a two-bar quotation of a, followed by a descending sequence. Tension in this segment abides slightly. (These bars can be read as vaguely reminiscent of E3a.)
E9c

is an exact transposition of E3b and thus completes the fugue with a continuation of the diminuendo and a very persuasive cadential close.

The structural correspondence of E3b and E9c, with or without that of E3a and E9b, is an important factor in the design of the fugue.

 

II/21.2.5 Character, tempo, articulation, ornament realization

The very simple rhythmic pattern, the ornamental concept of the subject and the comparatively low degree of contrapuntal activity all indicate a playful fugue of basically rather lively character. The tempo may be fairly swift; feeling a whole-bar 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 should establish a proportion between the larger beats:

half a bar
corresponds with
one bar
in the prelude
 
in the fugue
(Approximate metronome settings:
dotted quarter-notes in the prelude = 84, quarter-notes in the fugue = 126.)

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 marked 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 interpreter's responsibility to act accordingly in all corresponding cases. Such cases include analogous note-pairs in all further subject statements, in cs2 and in all quotations of b within the episodes. (Special care should be taken in note pairs which are not slurred; mix-ups happen easily and result in confusion for both listeners and performers.) In the subject, the fourth jumps at the beginning of the third and fourth bars respectively are separated by phrasing and must therefore be distinctly detached. Similarly, quotations of b in episodes must be separated with the same phrase cuts.

(Here are some of the instances outside the subject statements themselves where phrasing is easily overlooked; all indications refer to the first two eighth-notes of a bar:
  bar 17 [L] bar 18 [M] bar 36 [L] bar 39 [L]
  bar 52 [M] bar 58 [U] bar 61 [L] bar 70, 72, 74, 75 [L]
  bars 83/84 [M]      

In the following cases, however, the first two eighth-notes in the bar must be linked, due to their harmonic connection as suspension / resolution:

  bar 19 [U] bar 20 [L] bar 37 [M]  
  bar 39 [M] bar 51 [M] bar 59 [L]  

The score contains only one ornament, a mordent in the lower voice of bar 26 which is notated in brackets. If played, it begins on the main note and contains a single three-note shake. As it bears no melodic or structural relevance and is not even imitated, it can easily be omitted without loss for the composition.

 

II/21.2.6 The design of the fugue

The entering order of the subject statements, the cadential closes within the episodes and the introduction of two new contrapuntal companions from bar 32 onwards make the design of this fugue very obvious. There are three sections.

Section I encompasses the two initial statements, a bridging episode (E1) and the third entry followed by a cadential close in the home key (E2a). The second half of the same episode then represents a "change of mind" in that it undertakes a new start with a modulation and thus prepares for a redundant statement. Thereafter, a slightly longer episode with two dynamically decreasing segments (E3) completes this section with a strong cadential close in the dominant key F major. With regard to material, this section is unified by the exclusive appearance of cs1 and cs2 as companions to the subject.

The beginning of section II is marked by the simultaneous introduction of the two counter-subjects CS3 and CS4 which stay with the subject thereafter. The section contains three subject statements, linked by short episodes and completed by a passage which takes up the first, aborted cadential close of the fugue (compare E6 with E2a), concluding here in Eb major, the subdominant key of the fugue. The fact that this cadence remains unresolved in the middle voice ties the second and third sections together. This impression is reinforced by the use of the same contrapuntal material for both sections, in contrast to the first section.

The third section consists of three statements separated by long episodes. The observation that the third statement is redundant (it repeats an upper-voice entry), together with the recognition of a "false" lower-voice entry in the preceding episode (E8), leads to the assumption that section III is built along the same lines as section I, i.e. as a disguised recapitulation.

The table below shows the entries and their keys. For a sketch showing the entire design of the fugue in Bb major see ex. 36.

 

II/21.2.7 The overall dynamic outline of the fugue

This is a playful fugue in which powerful build-ups of tension are not the issue. The alternation of subject statements and slightly lighter episodes determine most dynamic processes. Additional developments are created by the dynamic gesture of each episode and by the change of mode in the subject statements.

Within the first section, the tension rises from a graceful but not too subdued beginning through three subject statements and a bridging episode. It subsides during the cadential close of E2a, only to be picked up immediately and brought to a first climax in the redundant lower-voice entry which, owing to its pitch position a fourth below the previous one, gives the (deceiving) impression of a fourth-voice entry. The seven-bar long E3 provides a gradual relaxation.

Section II begins, despite its one-bar long rests in the upper and lower voices, in a more assertive color as the subject is now enhanced by the slightly more independent new counter-subjects. Despite a dynamically diminishing episode, the second entry maintains the elevated tension, mainly because of its very exposed pitch position in the highest register of the Baroque keyboard. The next episode with its increase then provides a powerful preparation to the first minor-mode statement in this fugue which, due to the particular character of the subject, is for once much heavier and more powerful than the original in the major mode. The tension then subsides in the final episode of this section.

Section I        
M
U
L

Bb major
F major
Bb major

= tonic
= dominant
= tonic
} main round of entries
L   F major = dominant   redundant entry

 

Section II

       

M
U
L

F major
Bb major
G minor

= dominant
= tonic main round of entries
= tonic relative

} main round of entries

Section III

       

M
U
(L)

Eb major
C minor

= subdominant
= subdominant relative
(fake entry in episode)
} main round of entries
U F major = dominant   redundant entry

The first entry of section III combines the parallel motion of cs2 (extended here to two bars and even joined in bar 55 by a third parallel in the lower voice) with the two counter-subjects. This statement can thus be interpreted as representing the climax of the fugue. The inverted curve of E7 leads to another minor-mode statement which, also reinforced by a cs2-parallel in its first bar, picks up the high tension of the previous statement with slightly different means. The long episode E8 with its false subject entry then provides a very gradual relaxation. It is into this considerably lowered tension that the redundant statement makes its entry - much softer than its predecessors, last but not least because its first two bars are conceived in reduced ensemble. The first segment of E9 thwarts an immediate decline of the tension, but the final eight bars of the fugue lead to a completely relaxed ending.

 

">