No. 20 in A minor, BWV 889: Prelude

A friend once told me that there is nothing more difficult than playing this prelude musically. Unlike previous preludes, it appears elusive, and one falls into the illusion that the same thing continues endlessly. When encountering such a piece, analysis is paramount. Analysis often reveals what was previously unseen.

As this prelude features repeat signs characteristic of Book II, it is already divided into two halves: the first and second. Let's first examine the first half. In measure 1, the subject descends chromatically (right hand). Simultaneously, the left-hand accompaniment also descends chromatically. Measure 2 is merely an inversion of the upper and lower voices. Between these melodic phrases of the subject, numerous sequences appear. These can be two beats long or span an entire measure. The next appearance of the same thematic pattern as in m. 1 is in m. 4, where it modulates to E minor. And m. 5, like m. 2, is simply an inversion of the hands. In this way, we will trace only the tonality of the subject, disregarding the sequences.

Then, in m. 8, it modulates to C major (though it may not sound like C major, it becomes clearer if you swap the hands and play the right-hand figure in the bass clef), to G major in m. 9, to D minor in m. 11, and returns to A minor in m. 13.

Let's summarize the keys that appear again:

• A minor
• E minor
• C major
• G major
• D minor
• A minor

If we rearrange them as C G D A E, we can see that each key is separated by a perfect fifth.

Now, let's focus on the pitch. In the first half of this prelude, the C in the right hand on the first beat of m. 8 is at the highest point. This means that the highest pitch occurs at the point of modulation to C major.

Next, let's consider the spatial relationship between the subject and accompaniment. Just as in measures 1-2, there are some subjects where the subject and accompaniment are separated by more than an octave, while others, like in m. 11, are quite close.

And looking at the final measure of the first half of this prelude, although it ends in A minor, it clearly ends on the dominant (V). This means the first half concludes with high tension.

Considering these many facts comprehensively, and also leaving room for the performer's subjectivity, we determine the dynamics for the first half of this prelude. Rather than simply deciding on forte because the pitch is high, we make a comprehensive judgment based on many factors. I imagine this prelude might have been performed on an organ. Perhaps the type of pipes changed with each key. For instance, m. 11 might suggest a relatively thick pipe, and I feel that the subjects in A minor and E minor might also have used different types of pipes.

I believe this approach is effective for determining the dynamics.

Regarding the second half, since the first half ended on the dominant of A minor, it naturally begins on the dominant of the same key. In m. 17, the subject appears in the left hand. At first glance, it might seem similar to the subject in m. 1, but in fact, this is the first time the subject ascends in the second half. The first beat descends, then it ascends chromatically from the second beat, and descends on the fourth beat. This is the exact opposite pattern from before. Considering the prelude as a whole, I personally feel that m. 27 has the highest tension and the greatest volume.

And as an example, the subject appears simultaneously in two voices on the first beat of m. 31. Compared to the first half, the second half contains more sections with rising tension. Performers should carefully consider where to place the climax.

Fugue

This is a very rare fugue. There are actually two types of fugal subjects: real answers and tonal answers. It's quicker to demonstrate this visually than to explain it in words, so let's use m. 1 as an example. Let's assume the subject begins on the second beat of m. 1 and ends on the first beat of m. 2 (of course, some may consider the subject to be longer than this). In that case, the intervals for these four notes are: E to C is a major third, C to F is a perfect fourth, and F to G# is a diminished seventh. Keep this in mind. The second subject appears in the alto voice in m. 3. This subject is A G C D#, and the respective intervals are a major second, a perfect fourth, and a diminished seventh, meaning the first interval differs.

Originally, B G C D# would be correct; this is called a real answer, meaning the true answer. The A G C D# actually written are notes chosen with tonality in mind, and this is called a tonal answer.

If one searches for real answers in this fugue, only three appear throughout the piece. Therefore, the subject includes tonal answers, but it doesn't stop there. For example, the three quarter notes written in the bass voice from beats 1-3 of m. 10 inevitably sound like the subject. In this case, the final interval is a minor seventh, not a diminished seventh, and furthermore, the first note is missing. While it is possible to consider the first note as the eighth note from the previous measure, many subjects appear under such borderline conditions for what constitutes a subject. Those with a strict interpretation might not consider them subjects, resulting in performances where these are not clearly brought out. The judgment in this regard is left to the performer.

Another important point is the issue of articulation. While it is uncertain whether Bach himself wrote this, unusually, the eighth notes in m. 2 have staccato markings. If I were to offer advice, I would suggest not playing the first four quarter notes legato, but rather detaching them slightly just before moving to the next note. This makes the performance easier later on and conveys that the mood of this subject is by no means smooth. Furthermore, for example, the B natural in the bass on the third beat of m. 10 is a quarter note; even if the other two voices are quite high and distant, by sustaining it for its full quarter-note value and strictly adhering to the note values, one can aim for voice independence.

I feel the peak point is between measures 21-22, but what do you think? It is a very firm and resolute fugue. It would be wise to be careful not to let it flow too horizontally.