I finally completed my moderate-length fugue last night in the wee hours of the morning (ironically with the help of my synthesizer, set on organ, and not with my violin, which I used to compose parts of the previous sections). The piece in its entirety is 98 bars long, placing it on the same scale as the g minor fugue in the first Sonata, but a good deal shorter than the C major fugue, which I’ve been told that, at 354 bars, is Bach’s longest fugue.
Obviously, I’m not Bach; my contrapuntal skills are meager and, perhaps more importantly, my aims are not the same. Although the purpose of the fugue, I think, can be nothing other than the glory of God and the Heavens (even as I am a secular person, I cannot help but feel this in the form), the musical aims can vary greatly. For Bach, the interplay between the voices, the toying and molding of the theme, that was key. While I am interested in these ideas, I am, as a product of contemporary culture and of the Romantic Era, also interested in melody. In that sense, I place myself in a difficult position, where the upper line in particular must be melodic enough to *stand alone as a line* without requiring the “explanation” of the other voices. The soprano line, played alone, should sound like the violin part to a violin-piano sonata (in fugue style, if such a thing has ever been written).
Because “melody” itself comes most often from the shape of notes and intervals, this means that I have the challenge of adapting melody to this context, where stepwise is almost the only option. Consider that there are 3 or 4 voices (I do not follow a strict number, as the violinist would die from 98 bars of 4 part texture), and 4 strings, this allots something like the range of a fifth or sixth to each voice. A leap of a perfect fifth would land the voice smack dab in the middle of another voice.
This is the primary challenge of writing a fugue for the violin, and I do admit that I had to reduce the “countersubject” to a “countermotif” of particular ideas and interplays without being able to preserve the entirety of the line. At this point, I do want to point out that, to my knowledge, only two other such fugues exist, the aforementioned two by Bach. In the g minor fugue, the countersubject does not appear in every one of the expositional entrances, although it does appear later, and always maintains its rhythm. The C major fugue, however, has an undying loyalty to its subject-countersubject pairing, although I think that Bach knew very well that he could make that much easier on himself by having the countersubject be nothing other than the famous lament bass (descending chromatic scale), which works brilliantly as a recurring countersubject (but is terrible as a subject).
Here is a summary of my piece in great detail:
The opening features the tenor playing the theme in the home key of d minor at a soft dynamic. The first two measures and a quarter note into the third comprise a basic ascending scale, ornamented by various eighth notes of little consequence. However, the tendency to crescendo must be restrained, as the tritone drop in the third bar lands on an accentuated E that must be the height of the phrase, despite its position in the scale. The subject formally ends on the first beat of the fourth measure, and there are three beats of bridge that vary based on what key the next entrance will use.
The subject itself is very simple, very stepwise, and contrained to the range of a sixth. It does not take a genius to realize why Bach’s g minor and C major fugue subjects are, although varying in length, placed squarely within the range of a fifth, creating the interesting scenario in the C major where that fifth goes from a D to an A, that is, there is not a single C in the theme, despite its key. There is no practical way of handling a wild subject on the violin.
I treat the subject as consisting of four main motifs, if the bridge is included. The first is the eighth note sequence of the first measure, D-D-C#-D. The second is the second measure’s last three beats, the falling and rising palindrome. Next is the leap in the third measure. Lastly, the cascade of eighths and quarters in the bridge constitute something of a motif.
The structure of the piece consists of four bodies of “heavy” counterpoint in which I explore the different voicings and inversions of the subject in many voices. Between these bodies are three free sections of different proportions that feature faster notes and single-motif sequences. Only one or two motifs are used at a time, often as accompanying gestures. There are generally just two voices in these passages to allow for more agile playing.
Now, back to the exposition: the alto comes in a fifth higher, taking the key of A. There is not much to say except that the second measure of the countersubject is the most preserved portion throughout the piece because it fits perfectly with the subject. There is rarely occasion to reinvent it.
Accidentals in the a minor line bend it back towards d minor, but by measure 7, it is apparent that F major has subverted d, and the soprano comes in on the E string with the major version of the theme, which, among other things, has a fifth of a leap and not a tritone. After cadencing in measure twelve, the soprano climbs higher and begins a three-measure passage that is simultaneously a tribute to the g minor fugue and to pop ballads, if such a gesture is possible. The subsequent sequence is then a tribute to the C major fugue, and after this, I’m done with referring to precedent. I think that you need to do a bit of searching to find what I’m referring to, as my idea of “reference” in music involves a good deal of inventing and very little copying. Only the shape or texture remains, not the harmonies or voices or any of that.
The bass comes in very late, so late in fact that he must begin on the third beat of the subject, having completely missed his entrance some 6 bars earlier. However, all the other voices suddenly freeze and allow the bass to belt out the lowest possible rendition of this theme on the violin – starting on the A on the G string. The other voices return again and close off the first contrapuntal section with a cadence on a Picardy third in A.
The first exploratory section is arguably the hardest to play. It involves just two active voices that play with a diminution by half of the subject, brought through a circle of fifths from d minor to g minor to C major and finally landing in f minor, where only the first subject motif is developed in a downwards sequence. Three measures of tonicization prepare the piece for the second heavy-voice section in Eb major. The return of the theme in normal form is unmistakable, but it lasts only very briefly. Then, there is a winding, almost arhythmic segment of wailing suspensions and chromaticisms, marked “dolcissimio e legato.” Eb major has a dual identity: as a pompous, royal key, and also as an effeminate, melancholy one. I try to juxtapose the two identities and accentuate their differences. A flurry of notes brings Eb major to a full cadence – the first of only three in this piece. By full cadence, I mean the whole nine yards: pre-pre-dominant, pre-dominant, dominant, and tonic with a whole lot of ornamentation.
Following is the three-bar third interlude, which explores the tritone leap. It is cut short, however, when it lands in g minor. The g minor fugal section was arguably the hardest to write, because the subject is banned from entering in the top voice. It appears below a continuing line in g minor, and then slips in between the soprano and bass in d minor. It is barely audible when sandwiched between two other dominant lines, and it is not meant to be heard, just felt.
Although I could have ended the piece here, I opted to cadence lightly, and in an anachronistically romantic style, into C major, which is a brief foray into the mirror fugue territory. The voices enter in reverse order, with the themes inverted, and with the countersubjects also inverted. This was surprisingly easy to do compared to writing the g minor section, and the glorious mood is allowed to end normally with a decisive C major cadence … except not. The final chord is pushed away by the vii chord of G, and the third interlude begins, developing the bridge motif, which, if you recall, is the only one that has still not been developed.
The section culminates in a D major chord (on purpose), with a space for improvised cadenza (intended to be very short, but this is not a requirement. I always play a very long cadenza here when I play the fugue. By long I mean minutes, haha!).
The recapitulation begins with the opening transposed into g minor. This sets up the second entrance to begin in d, and the remainder of the piece is nothing but repeated material, extended cadences, and spontaneous fluorishes, reaching full four voice texture in the last two measures to close out the piece in an appropriately dramatic way. On a D major chord, of course.
Although I wrote this largely in a Baroque style, following most rules of good voice leading (I do have unresolved leading tones in the inner voices, but this is generally “excusable”), I do not intend for it to be played in the manner that many people play Bach. For me, in my pieces, tempo is an effect of the heart. The notes must come out at the speed of their feeling, and vibrato along with dramatization are very welcome. In essence, it should be played in the same fashion in which one approaches Rachmaninoff.
Anyhow, I’m still in the process of typing it into the computer, but anyone is free to borrow my neatly handwritten copy of it if he or she is interested.