WTC I/12 in F minor – Prelude 

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

I/12.1.1 The prelude type


This prelude is truly intriguing as there are, or so performing pianists convince us, three quite different approaches to its interpretation.


The melodic intensity in the sixteenth-note lines compels concentration on these as the primary feature of the composition (ex. 63a):


By contrast, the almost uninterrupted string of pulse-giving quarter-notes elicits interpretation of the composition as a metrically determined piece conveying a meditative character (ex. 63b):


Finally, the frequent recurrence of the figure exposed in the soprano of the first bar, in the context of an almost consistent four-part writing, allow taking the prelude for an "invention" (ex. 63c):

The examples give hints as to what is exactly at stake. In view of these choices, performers obviously face a host of interesting decisions; melodic intensity, texture coloring and tempo, to name just a few, will be different in each of the three cases.


I/12.1.2 The overall design of the prelude

The first harmonic progression concludes in bar 2m. (The subdominant falls on bar 1 beat 2, the dominant, as a ninth chord, on bar 2d.) In terms of an "invention", this cadence coincides with the second entry of the motive (see T: bar 2m) and should therefore not be regarded as a structural break.

The larger harmonic progression ends on the downbeat of bar 9 in Ab major, the relative major key. An explicit closing-formula (see bars 8/9) reveals this cadence as an obvious structural caesura.

Altogether there are four sections in this prelude. Three of them constitute closed harmonic progressions, while the fourth one ends in an interrupted cadence which can only be ascertained as structurally relevant because of its obvious correspondence: the beginning of the prelude recurs as a seeming recapitulation which then turns somewhere else; compare bars 1-2m with bars 16m-17. This is the only structural analogy.


bars 1-9

i - III

F minor to Ab major


bars 9-12 end


Ab major to C major


bars 13-16m

V - i

C major to F minor,
ending with interrupted
cadence on Db bass


bars 16-22


confirmation of F minor


I/12.1.3 Practical considerations for performers

In this prelude, the appropriate choice of tempo, dynamics and articulation depends largely on the performer's interpretation of the prelude-type (refer to paragraph 1).


Performers who feel that the message of the composition is conveyed primarily by the sixteenth-notes, will have to choose a tempo slow enough to allow these note values to develop full melodic expressivity.


In this case, the main concern is with the arabesque-like lines winding through all voices. They will require careful and expressive shaping provided with all shades of emotional quality, while great care is taken to ensure the constant flow of a segment in one voice into the next.


As for the articulation, the calm character requires that notes contributing on any level to the melodic fabric be played legato, and only cadential-bass patterns and consecutive leaps be non legato. However, since the first note in several of the leaps is at the same time a quarter-note in another voice and therefore cannot be shortened (see e.g. left hand: bars 4, 5), the few eighth-notes which actually sound detached appear above all in the context of cadential formulas (see e.g. left hand: bars 6, 8/9, 11/12) or in similar bass lines (see e.g. lowest voice bars 9/10, 13).


Performers who wish to express a meditative character by means of focussing on the continuous quarter-note pulse, will have to choose a faster tempo so as to gear the listener's attention to the metric feature rather than the surface structure.


In this case, all emphasis will be on the four beats in each bar which will then be perceived as a regular rocking motion with a very low-key melodic intensity. The sixteenth-notes (and to a lesser degree the eighth-notes) will be treated as purely ornamental and played with extremely light touch, like delicate lace-work behind more solid lattice.


As for the articulation, all quarter-notes and eighth-notes will be gently detached.


Performers who decide to draw on the inherent structural features and interpret the prelude as an invention, are free to choose any tempo as long as this allows for the thematic figures to be perceived as units.


In this case, the interpretation will emphasize a kind of touch and dynamic shading which distinguishes clearly between the different components of material, giving most intensity to the main motive and its possible companion, less to any other recurring figures and least to all other passages.


As for the articulation, either of the two approaches mentioned under (a) and (b) is possible, depending on whether the sixteenth-notes are regarded, and played, as essentially melodic (a) or as ornamental figurations (b).

There are four kinds of ornament in the piece, all appearing within the first ten bars.


The symbol which decorates the counter-motive (i.e. the regular contrapuntal companion to the main motive, as introduced in U: bars 2/3; for more details see below) denotes a long trill which begins on the upper neighboring note, moves in thirty-second-notes and ends in the suffix as indicated by Bach. This ornament recurs three times in the course of the prelude (see the fourth beats in bars 3, 4 and 10).


The symbol in bar 5 also designates a note-filling ornament which commences here on the main note and accommodates five notes, including a suffix.


The trill in bar 8 is composed in such a way that it lacks a resolution – both rhythmically (since it does not fill the time up to the next strong beat) and with regard to pitch (since the resolution note A follows only after three intermittent notes). This is thus a short ornament – a mordent which, launched from the upper neighbor note, contains only one double shake.


Similarly, the trill on the middle beat of bar 10 is also a "point d'arrêt trill". It includes five notes (as it is introduced stepwise) and stops short with noticeable separation from the following sixteenth-note.


I/12.1.4 What is happening in this prelude?

As any description of a composition sets out from a premise as to what the distinguishing features are, the F minor prelude allows for three fairly different descriptions, depending on the three choices expounded above.

a) If the prelude is regarded as a melodically oriented piece,
the features which determine the performance will all be sought in the sixteenth-note lines.

Within the first section (bars 1-9d) there is an initial curve in which the focus is entirely in the upper pitches and the climax falls on the prolonged Db on the downbeat of bar 2, with a subsequent diminuendo until the middle beat of the same bar. This curve is followed by the first of many embellished parallels (see bars 2m-3, soprano Ab-Bb-C and tenor F-G-Ab). These parallel ascents, different in detail but joined in their general outline, are sequenced twice (see bars 3m-4 and 4m-5) and create a gradual crescendo. From bar 5 to the middle of bar 6, the tension decreases slightly as all melodic figures point downwards (see in the soprano the central notes C-Bb, in the alto the descending sequence, and in the bass the target notes Bb, preceded by its leading-note A natural, and Ab, preceded by cadential steps). The extended parallel ascent in bars 6m-7m, in which all four voices participate, then provides an even more intense increase of tension which peaks on the syncopated high F, relaxing afterwards towards the cadential close on bar 9d.

The second section is distinguished from the first by a much higher degree of those intervals engendering melodic tension (see particularly in the soprano the first half of bar 9 and the first half of bar 10). The recurrence of the hidden parallels (bars 10/11) prepares the ascent to the climax in bar 11m, after which the successive leading-note / resolution pairs in bar 12 bring about a prompt relaxation (see alto: B-C, F#-G; tenor: B-C; alto: F-E; tenor: F-E). The ensuing three and a half bars of the third section contain two tension-curves: a gentle first one in bars 13/14 and a slightly steeper one beginning with the last sixteenth-notes in bar 14, climaxing on the second beat of bar 15 and ending in the interrupted cadence of bar 16m. The fourth section commences with a dynamic curve identical to that at the beginning of the piece; its descent, however, is extended here until the end of bar 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). It is further worth noting that in this interpretation which renders the sixteenth-note motion as an uninterrupted line catered for by complementing voices, no melodic phrasing occurs.

b) If the prelude is regarded as a metrically oriented piece,
the features which determine the performance will be sought in the quarter-notes. The focus in the texture tends towards the lower voices, and all dynamic inflections are of a delicate nature.

Within the initial section, a first gentle curve with a climax in bar 1m is followed, from bar 2m onwards, by sequencing one-bar groups. As the bass in these groups moves ever further down, a gradual diminuendo results. This leads to a state of floating weightlessness in that bar (see bars 5m-6m) where the quarter-note motion is temporarily suspended. The following extended ascent leads to a slightly more pronounced climax on the downbeat of bar 8, followed by a relaxation through the cadential close.

The two sections which build the middle of the composition are less regular in their pulse, featuring several passages where the quarter-notes are interspersed with half-note beats (see the first halves of bars 9, 10 and 12 as well as the second half of bar 14). This impairs the effect of meditative calm which 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.

All these components, which occur regularly in the prelude, are introduced within the initial two bars. Ex. 64 shows these components, in the order of importance in the composition, with their inherent dynamic shape (ex. 64):

the main motive

its regular companion

the relaxing link

the syncopated figure

The main motive appears eleven times, more often than not with a variation of its end. These are its statements: (ext = extended, var = varied; inv = inverted)

bar 1


bars 6/7 ext


bars 13/14 var


bars 2/3


bars 9/10 var


bars 15/16 inv


bars 3/4


bars 10/11


bars 16/17


bars 4/5


bar 13 var


In four of these eleven statements, the motive is accompanied by its regular companion (see bars 2/3, 3/4, 4/5, 10/11); on two other occasions it comes with a varied version of the stepwise ascent which characterizes the companion (see bars 6/7 and, in inversion, bars 15/16). The syncopated figure (see bars 2, 4, 5, 6, 11, 17) and the relaxing link (see bars 2, 3, 5, 5/6 (l.h., var), 12, 15, 17, 18 (2x), 19 (2x), 21 (2x)) account for much of the remainder of the piece.

The first section is thematically dense, with main-motive entrances in almost every bar and, after a short relaxation in bars 5/6, an extended quotation of the motive in the left-hand part. The two sections which form the middle of the prelude feature thematic bars interspersed with cadential closures or imperfect closures. The last section brings a liquidation of thematic material. In this structurally oriented interpretation, this fact together with the extended pedal bass determines this last section as one overall relaxation in which the pitch-curve in bars 19/20 plays only a subordinate role.


WTC I/12 in F minor – Fugue


I/12.2.1 The subject

The subject of this fugue is three bars long. It commences on beat 2 of a 4/4 bar, thus giving the impression of a long upbeat. Its ending is defined by the return to the keynote after the half-note G which represents the dominant.

With regard to the phrase structure, two views are theoretically possible: The subject can be regarded as a single phrase, or as consisting of a main subphrase (through the first six quarter-notes) followed by an "afterthought" (see the descent Bb to F). One might, however, feel that this "afterthought" is not quite substantial enough to represent a subphrase of its own. The most convincing concept, then, is a combination of the two ideas; in this case the paramount buildup and decline of tension takes place within the first six notes, followed by a mere extension of the previous dynamic decrease in the remaining descent.

The pitch outline of this subject is most intriguing. Commencing from the elevated position of the fifth scale degree C, the stepwise motion touches first the natural leading-note to C (i.e. the minor sixth Db), then the artificial leading-note below C (i.e. the raised fourth B natural). This triggers an unusual step: not only is the expected resolution to the leading-note omitted, but what is more, the interval which follows, while looking like nothing more special than a perfect fourth, is in fact located entirely outside the F minor scale! To top it all, its target is yet another leading-note (the raised seventh leading upwards to the keynote) followed by its resolution F. The ensuing descent seems to pick up from the recently heard B natural and continues through three chromatic steps before falling onto the lower F.

Within the subject itself, the rhythm consists only of quarter-notes and one half-note. However, a cursory glance beyond the boundaries of the first subject statement reveals that throughout the fugue, eighth-notes and sixteenth-notes are values to be reckoned with on a regular basis.

The harmonic background of the subject is unusual in that it sets off from the dominant, reaches the tonic for the first time in the middle of bar 2 and only then continues in an ordinary subdominant /dominant /tonic progression. Another extraordinary feature in the harmonic outline is the fact that the first steps (which melodically approached the two leading-notes to C) are both composed as interrupted (or "deceptive") cadences:


the dominant (i.e. the implicit harmony under the initial C) proceeds not to the tonic F minor but to Db major;


the ensuing F7 chord (i.e. the implicit harmony under the next C) which one expects to move into Bb minor leads instead into a G7 chord.


Ludwig Czaczkes, in his very thorough two-volume work Analyse des Wohltemperierten Klaviers takes into account all transitory harmonies created by the counter-subjects. The full harmonization of the subject as Czaczkes conceived it is shown here first (ex. 65a);


A Roman-numeral analysis follows
(ex. 65b).

The dynamic outline should certainly mirror the melodic and harmonic particularities mentioned above. The beginning on the exposed fifth degree already requires a slightly elevated dynamic level for the first note. The ensuing step towards the natural leading-note, followed by one to the artificial leading-note creates a powerful increase in tension. After this, the extra-scalar fourth interval which 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 natural and E natural – a solution unfortunately not open to keyboard players!)

The resolution onto the keynote F brings such a relief after all those daring harmonic twists that it involves an almost complete decrease of the tension built up until then. What little tension is left will then abate gradually through the chromatic descent.


I/12.2.2 The statements of the subject

There are only ten subject statements in this fugue of 58 bars.


bars 1 - 4



bars 27 - 30



bars 4 - 7



bars 34 - 37



bars 7 - 10



bars 40 - 43



bars 13 - 16



bars 47 - 50



bars 19 - 22



bars 53 - 56


(ex. 66)

The variations in the appearance of the subject throughout the fugue are minimal. The adjustment of the initial interval in the tonal answer occurs only once (bar 4), and the final note is only once delayed by a tie-prolongation of the half-note before it (see S: bar 50). No parallel or stretto of the subject are used.


I/12.2.3 The counter subjects

Bach invents three counter-subjects of very distinct character.


is introduced against the second subject entry in bars 4-7. It consists of five subphrases, thus creating the greatest possible structural contrast to the subject. The predominant note value is the sixteenth-note, another major contrast to the subject and its prevailing quarter-notes. The first subphrase of CS1 (see T bar 4: F-G-Ab--Ab-Bb-C) presents a slight increase in tension. This is separated from the remainder of the counter-subject by a leap of a minor ninth, and thus launches a completely new rise in tension. The second to fifth subphrases are then built as sequences which ascend towards a peak (and dynamic climax) on the syncopated C.


is first heard as a companion to the third subject statement in bars 7-10. It encompasses three subphrases. A first short one contains four eighth-notes, with a gentle climax on C. The second subphrase begins almost like a partial sequence (F-Eb-D in bars 7/8 can be heard as a sequence of C-Bb-Ab in bar 7), with the stronger climax on its first note F and an extended descent to the lower F (the rest is here a tension-sustaining one and does not indicate phrasing, as in CS1). A third small subphrase ascends to the dotted Bb as its mild climax, and relaxes throughout the last three notes.


only appears twice in the fugue. It is introduced in its expected place, i.e. against the fourth subject statement. Beginning at a later point of the subject than any of the other counter-subjects (see from the upbeat C-D-Eb at the end of bar 13) it ascends in a gradual sweep which, though interspersed with eighth-note rests, represents only one single phrase, with a climax on Db.

The further use of the counter-subjects includes several irregularities which may make their detection (or, at least, the convincing rendition of their phrase structure and dynamic design) difficult. These are the two most confusing details:


Voices cross over each other in several places.


In bar 7 immediately after the middle beat, the tenor interval Ab-F crosses over the descending alto leap F-E; they regain their normal position in the middle of bar 8.


In bar 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 bar 14, it does so considerably lower than the tenor; they recoup their ordinary places with the fourth beat of bar 14.


From bar 29 beat 2 to bar 30 beat 3, soprano and alto cross.


In bars 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 bar 48.


Counter-subjects are adventurous enough to swap subphrases!


In bar 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 beat 2 of bar 21). At the same time, the soprano (after having set off, in bar 19, with notes not belonging to the primary material) takes over CS1 which the alto had abandoned and concludes it properly.


In bars 27m-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-Bb-Ab), continues with notes which ape the large leap missed out by the tenor, and from the middle beat onwards finds itself in CS3. The soprano, after three non-consequential notes, completes the interrupted CS2.


In bars 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 and a completely avoided resolution.


Finally in bars 47-50, CS2 appears in the (alto-positioned) tenor, deprived of its first subphrase and with a drastically varied ending. Similarly, the CS1 statement which begins in the alto is allowed neither its ninth leap nor its final trill, which is substituted by a written-out figure-work.

The following sketch shows the phrase structure and dynamic design in the primary material of the F minor fugue as found in bars 13-16 (ex. 67):



I/12.2.4 The episodes

The F minor fugue contains eight subject-free passages.


bars 10 - 13


bars 37 - 40


bars 16 - 19


bars 43 - 47


bars 22 - 27


bars 50 - 53


bars 30 - 34


bars 56 - 58

None of the episodes 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 motive which uses the first subphrase of CS1 and continues with a segment from CS2 (compare A bars 10m-11d with T bars 8m-9d: a quarter-note which, after a sixteenth-note-tie or rest, is followed by a descent in sixteenth-notes). The CS1-segment is imitated in the bass (with a new ending) which in turn is partly doubled by the tenor. The ensuing bars sound like descending sequences. (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.)

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. (Performers should be careful with this soprano segment: it is the least important part of this episode.)

E6 also relies on the same material; this time, the alto takes the lead. (Here Bach strings the motive to its sequence by filling the phrasing space with an additional sixteenth-note; see the D natural after the middle beat of bar 44.) The imitation is presented in the bass, and both soprano and tenor are accompanying voices.


A second episode-type is established in E2. Here the initial impulse is given in the soprano by a three-note figure which 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 upwards). The texture is completed by the imitation of the bass figure, sounding in the alto (with the extension of the inverted CS1 subphrase continuing downwards). As in the first episode-type, this one-bar model is also sequenced twice (compare bars 16/17 with bars 17/18 and 18/19).

This episode recurs in the first half of E3 (compare bars 16-19 with bars 22-25). Here, the leading voice is the alto (a fact which is not at all easy to convey under the parallel soprano!); its main counterpart lies 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 (compare bars 16-19 with bars 50-53).

The two remaining episodes, E5 and E8, do not follow either of these models. E5 contains a figure derived from the first CS1 subphrase (see bass bar 37, sequenced in bar 38, and bar 39, sequenced in 40). The other two voices involved in this episode recall the tied-quarter-note-plus-three-sixteenth-note figure from CS2, matched here with free extensions. E8 finally features one bar presenting nothing but this CS2 segment in four-part texture, followed by two cadential bars.

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 which 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 (though not without drawbacks towards the end of each of these episodes).


I/12.2.5 Character, tempo, articulation, ornament realization

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 sixteenth-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).

The articulation which corresponds with this character is legato in all melodic parts. With regard to the components of the material in the fugue this means:


All notes in the subject and its three counter-subjects are legato; exceptions occur only where phrasing separates a note from the beginning of the next subphrase (see particularly in CS1 and CS2 after the first subphrases respectively).


Non-thematic parts may feature consecutive skips or cadential-bass patterns, as in bars 34/35, bars 40/41 and bar 43 (bass); these must be taken non legato.


In the episodes, octave leaps and cadential steps in the lower part must be detached (see B in E1: C,C, Db etc.; similarly in E6; also in E3).

The relative tempo of the prelude to the fugue varies, of course, depending on which view a performer adopts regarding the prelude-type (refer back to the distinction made between melodically determined, metrically determined and "invention"-style prelude under paragraph 1 of this chapter).


In the interpretation which finds the melodic intensity in the sixteenth-note lines of the prelude, the appropriate proportion is a simple 1:1; i.e.

one quarter-note

corresponds to

one quarter-note

in the prelude

in the fugue.


In the interpretation of the prelude as a metrically determined piece, the ideal proportion is the more complex 3:2; i.e.

three quarter-note beats

correspond to

one half-note

in the prelude

in the fugue,


six sixteenth-notes


one quarter-note

in the prelude

in the fugue.


In the interpretation of the prelude as an invention, either of the two proportions is possible, depending again on how one perceives the prelude sixteenth-notes.

A last comment on this matter of tempo choice: One should not neglect that the tempo of the fugue, too, may vary depending on the character of the preceding prelude. Thus after a metrically oriented prelude in which the shortest note values served as mere lace-work, the fugue may give extra weight on the melodic quality of its sixteenth-notes and sound slightly slower than after a rendition of the prelude in which the smaller values were granted full expressive power. (Approximate metronome settings: (a) all beats = 66; (b) prelude beats = 84, fugue beats = 56.)

The fugue contains two kinds of ornaments. One is part of the thematic material where it marks the end of the first counter-subject; the others are cadential ornaments decorating the final closing-formula in bars 57/58. The trill at the end of the counter-subject abides by the same rules which apply to a trill in a subject. As the note it ornaments resolves properly onto the following downbeat, this is a note-filling ornament; it commences on the main note (having been approached in stepwise motion), shakes in thirty-second-notes and ends with a suffix. As an integral part of the thematic material, this trill must be transferred without changes to the following bars:

bar 9


bar 30


bar 15


bar 55


bar 21


The two cadential ornaments are also note-filling trills. The one on the soprano-E begins on the main note, while the compound ornament in the tenor commences as indicated from below. Both then move in thirty-second-notes and end in parallel suffixes. A written-out version of this final closing-formula is given below (ex. 68).



I/12.2.6 The design of the fugue

One salient feature in this fugue is that the episodes show unusual endings. All of them close on an imperfect cadence, and several, in addition, present these as unresolved chords (see E1 bar 13: F suspended into chord on C, also E2 bar 19 and E3 bar 27). Both features make the episodes of this fugue appear ill-suited to provide section endings. By contrast, the only distinct cadential pattern (before that in the final bars) to convey the impression of a structural closure occurs in the context of a subject statement, at the end of the tenor entry in bars 40-43. These observations invite the conclusion that in this composition, episodes do not round off previous statements but prepare the ensuing ones; they thus open sections and do not close them.

A factor which might provide further enlightenment regarding the structure of this fugue is Bach's use of the ensemble. There are two subject statements which appear in reduced number of voices (see T: bars 19-22 and S: bars 47-50); both are therefore likely to mark section beginnings.

The harmonic background to the subject statements seems very straightforward. The first six statements appear in the tonic area; four of them commence like the first entry on C, only two in the dominant position on F (tonal answer, bar 4) or G (real answer, bar 19). The seventh and eighth statements belong harmonically to the area of the relative major; the former (see bars 34-37) seems to represent the tonic position of Ab major but is harmonized in such a way that it refers back to F minor; the latter reads like the dominant of Ab and ends accordingly. However, it may be worth noting that in the detailed steps of their harmonic progression, both are not in keeping with the original harmonization of the subject. The final two subject statements are back in the F minor key, representing the dominant and tonic positions respectively.

The conclusions to be drawn from these findings for the design of the fugue are as follows:


The two statements in reduced ensemble (tenor entry in bars 19-22 and soprano entry in bars 47-50) must be regarded as first statements in their respective sections. These two sections therefore commence with E2 (in bar 16) and with E6 (in bar 43) respectively. The latter section-beginning is confirmed by the explicit cadential-bass pattern at the end of the preceding statement.


The two statements which harmonically refer to the relative major belong together and form a section of their own. They 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 completely missing in both statements, CS1 occurs only against the first of these entries and is varied at both ends (see S: bars 34-37). This section thus commences with E4 and ends with the cadential-bass steps in bar 43.

For a sketch showing the design of the fugue in F minor, see ex. 69.



I/12.2.7 The development of tension

Within the first section, the tension rises steadily from one subject entry to the next. An increasing density of the material, up to the texture which confronts the subject with three counter-subjects, supports an equally growing intensity. The episode which precedes the fourth statement grants a transitory change of color together with a short relaxation, but it is then hit by the full force of four thematic voices. The end of this section seems somewhat unresolved since a suspended G mars the final F minor chord (see bar 16); this creates the effect of stringing this section together, if only by a thin thread, with the second one.

The second section begins with an episode of rising tension, so that its first statement appears already on a somewhat elevated level. The ensuing episode generates a decrease of tension in its first half (until the transitory cadence to Ab major in bar 25m) but then seems to turn on its heels and prepare for the ensuing subject statement by a dynamic increase (see bars 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 long-expected perfect cadence in F minor which, this time, sounds unimpaired by any suspending voice.

The episode opening section III provides a more drastic change of color because of the position of its material: the two motivic parts are in the bass and tenor, the alto is resting 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 the two entries of this section is built similarly to that linking the two statements in the second section. Its first bars feature descending sequences, followed (after a transitory cadence in Eb major in bar 39m) by a last-minute preparation for the approaching subject entry. The two statements in this section, too, create much less tension than those of the preceding sections, for three reasons: with regard to texture, because of the much lower density in thematic material; with regard to melody, because of the major sixth which is much more neutral in its tendency than the strong minor sixth; and with regard to harmony, because of their dissimilar harmonization which omits all the most striking steps. The second entry – and with it the section – closes, as mentioned repeatedly, in a definite cadence.

The fourth section commences with a four-part episode which presents three bars of rather 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 until the middle of bar 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 previously. With the bass resting, 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 equally unprepared as was the last statement in the first section. This statement returns to the lower pitch level in the center of the keyboard and to the full four-part texture, but replaces two of its counter-subjects by syncopated appoggiaturas (see T bars 53/54: C-Bb, Ab-G; A bars 54/55: Db-C, Gb-F), thus substituting thematic density with emotional urgency and creating a new kind of climax. The final episode can do nothing but gently resolve all this cumulated tension.