WTC II/17 in A major – Prelude

This prelude is based on several motifs. Some of them spawn the

imitations and modifications typical for polyphonic material, while others
are presented with an accompaniment or with an answering motif in the
other voice. The texture is conceived in two parts both of which split
regularly to form chordal patterns.
The first harmonic progression ends with the confirmation of the home
key at m. 71. This close marks a structurally relevant point. In view of the
large-scale design of the prelude, however, one might designate it as the
end of a sub-section rather than a fully developed section since the more
convincing section ending appears, complete with a cadential formula, in
mm. 16-17. There are altogether five sections in this piece:
I mm. 1-171 tonic to dominant (A major to E major)
(1-7: tonic confirmed, 7-171: modulation)
II mm. 17-341 dominant to tonic relative (E major to F minor)
(17-231: D confirmed, 23-341: modulation)
III mm. 34-501 tonic relative to subdominant (F minor to D major)
(34-401: tonic relative confirmed; 40-50: modulation)
IV mm. 50-641 subdominant to tonic (D major to A major)
(50-632: modulation to V of A, 632-641: link)
V mm. 64-77 tonic confirmed
There are extended structural analogies in this piece.
mm. 1-4 . 17-20 and 34-37 (transposed and varied)
mm. 1-2 . 50-51 (transposed)
mm. 5-6 . 21-22 and 38-39 (transposed and varied)
mm. 7-91 . 40-421 (transposed and varied)
mm. 11-171 . 44-501 (transposed and varied)
The basic character is rather calm. The rhythmic patterns are complex,
featuring a large variety of note values from dotted and tied eighth-notes to
32nd-notes. The main rhythmic feature is the long on-beat note followed
shortly before the next beat by one, two, or three fast notes. This gives the
piece a swinging momentum, preventing too slow a tempo.
The articulation is mainly legato. Most of the interval leaps occur
either inside 16th-note patterns or between 32nd-notes and the following
longer note, i.e., in instances where detached playing is ruled out. The

495
496 WTC II/17

wedges indicated in mm. 5 and 6, which specifically demand abbreviation

of the long notes, give performers to understand that the dotted notes in
this piece are generally meant to be taken legato. As with all indications
regarding articulation or ornamentation within thematic material, these
wedges should be transferred to corresponding notes. This definitely applies
to the G in m. 393. Whether the same holds true for the A in m. 213
depends on the individual performer’s interpretation of this measure—
which may or may not be perceived as deriving from m. 5.
Conventional exceptions from the general legato touch occur above all
in cadential-bass patterns (mm. 16, 33, 49, 62, and 74). Other exceptions,
in which articulation is explicitly marked, also occur outside the thematic
material in cadential contexts. In m. 62, paired slurring indicates dynamic
shading in an active-passive pattern: these slurs reinforce the harmonic
feature of appoggiatura-resolution, rather than indicating an abbreviation
of the second note. In m. 76, the slur linking the quarter-note A to the
subsequent G serve the same function and additionally guards against any
possible interruption of the do–si–do formula.
The A-major prelude features three types of ornaments: two grace-
notes, an inverted mordent, and several mordents. Mordents occur in two
different contexts. In mm. 52-59 they function as thematically integrated
ornaments appearing before the backdrop of an accompaniment in regular
16th-notes. Each of these mordents is approached in stepwise motion, thus
beginning on the main note and containing only a single three-note shake.
The mordent in m. 76 is a cadential ornament. As it appears with no other
rhythmic features against it and marks the essential step in the final ritar-
dando, this mordent can be more elaborate. Beginning equally on the main
note, it may contain five or seven notes before it stops short. The inverted
mordent highlights the cadential-bass pattern immediately preceding the
final homophonic formula at m. 743. In accordance with the harmonic
context prevailing in this measure it must be played with C as its lower
neighbor note. The two grace-notes both appear in m. 75, as part of the
final homophonic cadence. Both are indicated as eighth-note appoggiaturas
to a main note of quarter-note value. In both cases, the graces are thus
played on the beat, together with the chordal notes in the right hand and
the bass notes, and resolve after one eighth-note into the harmonically
anticipated notes G and E respectively.
The variety of motivic material in this prelude allows for different con-
cepts, and by extension, different ways of labeling. On the one hand, a
distinction of all similar but not-quite-identical shapes would lead to so
large a number of individual shapes that performers might lose track and
A major 497

the benefit of an analysis would thus be thwarted. On the other hand,

recognizing too many different shapes as variations of a single idea might
render analytical observations meaningless. The following discussion tries
to eschew both traps by taking into consideration not only the actual
appearance of a melodic unit but also its texture and context.
M1 is presented in mm. 1-21. It consists of two manifestations of the
broken chord. In the left-hand part, a descending A-major chord appears in
a linear pattern (M1a). The skips are not bridged by the fast notes following
the dotted strong-beat notes but widened. Its homophonic companion M1b
presents the A-major chord in a pattern that focuses on off-beat block
chords preceded by auxiliary notes in the do–si–do pattern. Dynamically,
M1 describes a diminuendo, caused by the absence of any harmonic activ-
ity in this motif and enhanced by the descending direction of the melodic
part M1b. M1 recurs similarly in mm. 3, 17, 19, 34, 36, and 50. A variation
can further be found in m. 10. Moreover, M1b appears separately and in
slight variation in mm. 52, 54, 56, 58, and 59. The motif and its recur-
rences do not trigger any imitation or inversion of voices: the melodic line
always remains in the left-hand part and the accompanying chords are
restricted to the right-hand part.
M2 is an almost ubiquitous figure in this piece. Introduced in m. 2 and
appearing in a variety of pitch patterns,1 its discerning feature is the single-
voiced contour in a rhythm of regular 16th-notes beginning after a down-
beat rest and ending on the next downbeat. All variants of this motif share
the dynamic layout with an initial crescendo to beat 2 followed by a
relaxation to the downbeat. In contrast to M1, which remains confined to
its initially homophonic setting, M2 changes voices and joins a number of
other motifs.
M3, whose original version is introduced in the lower part of mm. 7-81,
consists of a three-32nd-note upbeat followed by a scalar descent in dotted
rhythm. As in M2, the dynamic climax falls on beat 2 of a measure, and
just like M2, M3 comes in many facets.2
1
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.
2
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 coun-
terpart to M2c or M2d).
498 WTC II/17

Finally, M4 differs from the three previously mentioned motifs in that

it is chordal and not linear (see mm. 24-251). It is thus closely related to the
homophonic accompaniment within M1. This motif appears not as an
active component in its own right but as a complement to M3c, which thus
grows into a two-measure figure and only drops this complement in the
liquidation process from m. 29 onward.
The following table gives an overview of the structural design in the
prelude as it results from the interplay of the three motifs and their varia-
tions, as well as showing the analogy of the first three sections.
Section I Section II Section III
M1a/M1b + M2a M1a/M1b + M2a M1a/M1b + M2a
M1a/M1b + M2a M1a/M1b + M2a M1a/M1b + M2a var
M2b/M2b var M2b/M2b var M2b/M2b var
M3a/M2c + sequ. M3d/M2d, M4/M2c + sequ. M3a/M2c + sequ.
M3b/M2f + sequ. M3d/M2c + sequ. + liquid. M3b/M2f + sequ.
cadence cadence cadence
Section IV Section V
M1a/M1b M3b/M2d, M3d/M2c + sequ.
M2a M4 var/M2c var
M3c/M1b, M3c/M2e + sequ. M3d/M2c + sequ.
expanded cadence + bridge expanded cadence

WTC II/17 in Amajor – Fugue

The subject of the A-major fugue spans exactly two measures. Begin-
ning after an eighth-note rest on the fifth scale degree, the phrase reaches
its cadential conclusion at m. 31, in a gentle ending on the third. The har-
monic background is simple.
Its most significant feature is
the progression from the sub-
I V I IV V I dominant to the dominant in
the syncopation.
The rhythmic variety in the subject comprises eighth-notes, 16th-notes,
and a syncopation of 5/16 duration. Later in the course of the fugue,
quarter-notes are added as regular features. Yet despite the presence of
these four different note values, the overall impression is one of rhythmic
simplicity. This is due primarily to the fact that while the subject and its
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 subject’s 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 augmen-
tation.
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 syncopa-
tion in m. 2 as the subject’s salient feature—a metrically highlighted note
that is also harmonically alive—will 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. mm. 1-3 A 8. mm. 22-24 S

2. mm. 3-5 S 9. mm. 24-26 A
3. mm. 6-8 T 10. mm. 32-34 T
4. mm. 8-10 B 11. mm. 35-37 S
5. mm. 13-15 B 12. mm. 37-39 B
6. mm. 16-18 A 13. mm. 41-43 T
7. mm. 18-20 T 14. mm. 42-44 (45) B
15. mm. 48-50 T

In addition to the adjustment of the initial interval in the tonal answer,

the subject undergoes several minor changes. The fourth and fifth as well
as the eleventh and twelfth statements end on the keynote instead of the
third degree. This is a modification frequently used by Bach for bass
statements, so the only surprise is that in this fugue it also appears once in
the soprano (mm. 36-37). Furthermore, two statements feature an artificial
leading-note in the first tetrachord that is later corrected (D instead of D
in B: m. 14 and T: m. 41). The final statement, while difficult to read (and
hear), is presented as an unmodified answer.
More significant modifications occur in mm. 42-44. The bass entry not
only overlaps with the preceding tenor entry for half a measure, but also
features drastic harmonic alterations that transpose its final half-measure
onto the key of D minor. The subject’s final note does not, however,
provide the expected resolution, so that the subsequent partial sequence
must be regarded as an integral part of this entry. At m. 451, this extended
subject statement concludes with an interrupted cadence (the B-major
chord is VI of D minor). Similar but not quite so drastic harmonic altera-
tions occur in the bass statement of mm. 37-39. Beginning as an implied
D-major entry, the sudden Fs toward the end suggest a turn toward D
minor. But then the final chord, expected to confirm this key, is accom-
panied by an unresolved voice and allows for subsequent modulations,
here again wrapped into a partial sequence (see at m. 394 the soprano’s
closing formula leading to a resolution in A minor).
Bach has invented two counter-subjects for this fugue. CS1, introduced
in mm. 3-5, accompanies all further statements. It consists (after a 3/16
bridge that is not part of the component) of a chromatic descent in quarter-
notes leaping back up to a do–si–do formula. Dynamically, CS1 describes
two relaxations, setting in after climaxes on the initial note and on the
closing formula’s syncopation. The two diminuendos form a striking
A major 501

contrast to the crescendos dominating the subject. Frequent variants of this

counter-subject, in which the leading-note is protracted by means of a tie,
sometimes additionally flattened or even flattened and diverted downward
(see S: m. 15 and T: mm. 23-24, A: m. 20 and S: m. 26, and A: mm. 36-37)
cause a harmonically unresolved ending. In two cases, CS1 swaps voices
(see mm. 15 and 23-24), in two others, it is reduced to its tail (see A: mm.
33-34 and B: m. 18)—an abridgment that, as the structural analysis will
show, underlines the analogous positions of these subject statements within
the design of the fugue. In three further cases we hear only the chromatic
descent (see mm. 38-39, 42-43, and 50).
CS2, introduced in A: mm. 6-8, consists of a long stream of ornamental
sixteenth-notes followed by four eighth-notes. It undergoes a modification
process that soon
leads to forms so far
from the original as
to render it mean-
ingless to retain the
labeling. As its con-
tour is designed as a
veiled CS1 parallel,
CS2 does not repre-
sent a dynamically
independent contra-
puntal component.
The A-major fugue contains eight subject-free passages.
E1 mm. 5-61 E5 mm. 26-321
E2 mm. 10-133 E6 mm. 34-351
E3 mm. 153-163 E7 mm. 39-411
E4 mm. 203-221 E8 mm. 45-483
Two of these episodes are subdivided. E5 encompasses a cadential
close in C minor that separates E5a (mm. 26-273) from E5b (mm. 273-321).
In E8, the two halves are separated conspicuously by the general pause on
m. 463. The episode material features derivations from subject and counter-
subject as well as independent motifs. The subject head in 6/8 or 4/8
extension repeatedly anticipates a complete subject statement (see A: m. 5,
A: m. 34, S: mm. 34-35, and B: mm. 203-211), and its 16th-note tail follows
some statements as a sequence or an imitation (see T: mm. 15-16, B: m. 16,
A: mm. 20-21, T: m. 21, A: mm. 21-22, 26, 26-27, 27, T: mm. 34, 34-35,
B: mm. 39, 39-40, S: m. 40, and T: mm. 40-41). The counter-subject’s
closing formula is sequenced in T: mm. 20-21, A: m. 21, and T: mm. 21-22.
502 WTC II/17

M1, introduced in S: m. 5, is developed in B: m. 10 with six sequences

(103-133) and in S: mm. 27-28 with sequences and imitations (S: mm.
28-291 and A: mm. 29-303-321). M2 (S: mm. 10-113) recurs in A: mm.
113-131, S: m. 13, and B: mm. 273-291 with imitations in S: mm. 29-303
and A: mm. 303-321. M3, the contrapuntal companion of M2, is heard in A:
mm. 102-111 with imitations in S: mm. 114-123 and in A: mm. 274-283 with
imitations in B: mm. 293-301 and S: mm. 304-313.
Two of the episodes feature cadences with explicit formulas: E3 closes
at m. 173 in F minor and E5a at m. 273 in C minor. In E7, the unresolved
D-minor cadence of the subject entry is “corrected” half a measure later
with an explicit formula in the soprano toward A minor. The subsequent
sequence of the subject’s tail leads into E minor (m. 402), which makes
way for E major, the dominant of the home key. E8 is the only episode not
to use any of the previously introduced motivic material. Its largely homo-
phonic texture sets it apart from the remainder of the fugue. In E8a, a
broken-chord figure in the bass accompanies metric block chords in a
harmonic progression from B major to E9. E8b follows with a two-part
texture in contrary-motion that reconfirms A major but is diverted with
another homophonic pattern into an interrupted cadence.
The role played by the episodes in the dynamic development can be
described as follows: E1 and E6 both bridge between consecutive entries.
Due to the prominent appearance of the subject’s head, the level of tension
falls only slightly after the subject statements and rises again for the sub-
sequent entry. E2 and E5b occupy a plane that is contrasted to that of the
subject statements and the other episodes. Therefore, these two passages
are best distinguished from their surroundings by a distinctly different tone
color and touch. E3 and E5a pick up the tension remaining at the end of the
preceding subject statements which, due to the unresolved counter-subject
ending, has not reached complete relaxation. The two episodes then
complete the dynamic decline. Similarly, the falling sequences of the
subject’s tail in E7, together with the one-by-one drop out of voices and
the harmonic resolution, also generate the effect of closure. E4 and E8 are
the only subject-free passages to create an increase in tension. In E4, the
basic intensity is generated by episode material stemming exclusively from
the subject and counter-subject, the growth by ascending sequences and
imitations. In E8a, the change of texture from dense four-part polyphony to
homophony creates a strong and sudden outbreak of almost virtuoso inten-
sity, enhanced by the contrary motion and the impact of the harmonic step
from B major to E9. By contrast, E8b, beginning after a general pause,
remains fairly soft.
A major 503

Given the overall effect of rhythmic simplicity combined with the

predominance of interval leaps and ornamental 16th-notes in the pitch
pattern, the basic character of this fugue is rather lively. The tempo should
not be too fast, both in order to avoid the effect of superficial runs without
melodic contours in the 16th-notes and to give listeners a chance to fully
appreciate the melodic closing formulas and their occasional deflection.
Owing to the greatly different character and mood of the prelude and fugue
in A major, a simple proportion can be chosen tempo relation between the
two pieces without risking dullness: an eighth-note in the prelude corre-
sponds with a quarter-note in the fugue. (Approximate metronome settings:
prelude beats = 44, fugue beats = 88.)
The articulation demands careful nuances.3 The shortest and most ener-
getically bouncing non legato should be applied to those eighth-notes in
the subject that do not form part of an ornamental tetrachord.4 A not quite
so energetic and slightly longer non legato applies in the cadential-bass
patterns and in other non-melodic eighth-notes in cadential style.5 An only
gently detached non legato is appropriate for the chromatic descent in the
counter-subject. The eighth-note leaps in M3 and the initial fourth in M2
sound most convincing if their color is lighter than the subject’s, graceful
rather than energetic. The chromatic descent in M2 can be taken either in
dense non legato or, perhaps preferable in view of the desired color contrast,
in legato. All melodic formulas are legato. This applies to the subject’s two
rising tetrachords, the closing formula in the counter-subject, and the
cadential formulas in S: mm. 16, 39, 44-45, and 50). If the counter-
subject’s closing formula is deflected, it should nevertheless retain the
legato touch to mark its melodic origins (S: m. 15, B: m. 18, A: m. 20 with
imitations in T/A/T, as well as T: m. 23, S: m. 25, and A. mm. 36-37).
While the ornamental 16th-notes in the subject are legato, those in M1 and
M2 benefit from a crisp quasi-legato touch. The cadential trill in S: m. 48
begins regularly from the upper neighbor note, shakes in 32nd-notes, and
ends without rhythmic interruption in the suffix Bach has spelled out.

3
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 contrasts—between
subject and counter-subject on the one hand and between primary material and independent
episode material on the other—are essential.
4
To avoid confusion, here is the articulation in the subject: E-C-F non legato, B up to E
legato, both Es non legato, A up to D and all remaining 16th-notes legato.
5
Cf. B: mm. 15 A-C-D-A, m. 20 G-A-D-E-A, mm. 26-27 B-E, A-D-F-G-G-C, and
mm. 47-50, all eighth-notes and quarter-notes.
504 WTC II/17

The fugue’s 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 redun-
dant 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 modula-
tion 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 begin-
ning 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 texture—virtuoso patterns in
one hand, block chords in the other, a rhythmic surprise in the weak-
beat 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 sub-
dued, 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

entry to another. E2, as mentioned above, brings with it a fundamental

change of color as well as, at the outset of the episode, a radical drop in
tension. While the ascending sequences of M1 generate a certain increase,
the redundant bass entry sounds considerably less assertive than its prede-
cessors, last but not least because, abandoned by the alto, it is left in
three-part setting.
Section II repeats the process with different means. Beginning in
unusually reduced ensemble and thus almost as softly as the first section,
the build-up of tension proceeds unhampered through three entries linked
by the intense E4. The last entry of this round resembles that of the
previous section in that it is both redundant and set in three-part texture.
The harmonization in minor completes the picture, so that section II also
ends on a softer note.
What was left out in these two otherwise almost analogous sections,
the self-contained episode, is now made up for at the beginning of section
III. Setting out very softly, this episode presents a gentle tension increase.
The three ensuing subject statements continue this development in a line
that is only transitorily suspended during E6. After the (still moderate)
climax in the bass entry, E7 provides both dynamic relaxation and release
from the minor mode.
Section IV begins with a boost of confidence in a three-part major-
mode entry. The above-mentioned harmonic digression provokes a climax
in m. 46 that serves as a culmination point not only of this section but of
the entire fugue, followed by the softer curve of E8a and the triumphant
final statement in five-part texture.