Sunday, April 30, 2017

Waltz in C Major for Solo Guitar

Inspired by Dr. Gjerdingen's Music in the Galant Style, I have written a charming little waltz in C for the guitar. Early on in the book, he quotes a Mozart horn concerto that had a cliche I recognized from several classical pieces I'd heard. It's over a schema he calls a Jupiter, after the finale of Mozart's Jupiter symphony, which has the famous do, re, fa, mi underpinning (My favorite single movement in all of Mozart's numerous symphonies). The harmonic continuity is I, V, V, I, so there are several possibilities: do, ti, re, do; do, re, ti, do; do, ti, sol, do &c. You chose the form of the underpinning depending on what you want to do with the melody. In the Mozart horn concerto example, the gist of the melody was do, di, re, ri, mi. Like I said, I'd heard this chromatic lick many times before. Mozart's piece was in 6/8, but I recognized the simplest way to present it would be in 3/4, and so I was writing a waltz (More modern than the ancient minuet).

Mozart's continuation went on to modulate, which was a pro forma aspect of galant writing, but I recognized immediately that if you didn't modulate, you could continue the chromatic lick through fa and fi to sol, and in so doing get an augmented sixth under fi to target the dominant seventh chord on the last beat of measure six. This screamed out for a fermata, which really makes the a section pluperfect. The conclusion after the (written-out) fermata is also the irreducible essence of simplicity. You can make a nursery rhyme profound if you have something profound to say with it. The section doesn't modulate, but it does have all twelve pitch classes in just eight measures!

The difficult thing for this piece was the B section. It took a couple of days of working on it as I went to sleep before I got the idea to go from sol back down to ti for the fermata on the V(m7) sonority. This allowed me to use some of my favorite secondary dominant harmonies: V(d5m7)/IV, V(d5m7)/iii (The harmonic point of origin for what the old fashioned theorists called the French Augmented Sixth chord), V(d5m7)/ii, and finally the traditional French harmony, V(d5m7)/V in second inversion (The so-called German Augmented sixth is a V(d5m7m9)/V with the root missing). The B section really is the pluperfect answer to A, and, as a result, this piece sounds like something that has existed forever, I just happened to be the one to discover it.

Here's the audio (AIFF file, so you'll need QuickTime active in your browser): Waltz in C Major

And here's the score:

As you can see, the piece begins with a quarter note pickup on the dominant note G, which was part of Mozart's formula. That's the open G string, of course, which is why this has to be in C for the guitar. I laid out the melodic trajectory in three-note directional units: C, C#, D in measures one and two; D, D#, E in measures three and four, and then the E, F, F# in five and six that must resolve to G. In between are reiterations of the open G. Nothing could be simpler. For the underpinning, I used the do, ti, sol, do option of the Jupiter schema. The inner parts are in the traditional waltz accompaniment pattern, and the conclusion has a super-cool contrapuntal lick, followed by a soft cadence with a 9-8 attacked suspension to set up the repeat.

For the B section, then, I went from the open G to the G an octave higher and worked my way back down (The piece is so simple, everything is in open position!). The secondary dominants presented are: V(d5m7)/IV on the last beat of measure nine, V(d5m7)/iii on the last beat of measure eleven, V(d5m7)/ii on the final beat of measure thirteen, which goes into the V(d5m7)/V in second inversion at the beginning of fourteen. Then the lower voicing of the dominant seventh chord at the written-out fermata. All twelve pitch classes are in this section too. Pow!

Take a look at the voice leading from the V(d5m7)ii into the traditional French sonority: You have A to A-flat in the bass, and E-flat to E in the lead. Traditional composers - or theorists - would not allow this, as it seems to be a diminished fifth into an augmented fourth; a forbidden parallelism. Luckily for me, I studied Schillinger, and so I know this isn't a parallelism at all, as the voices transform in a crosswise manner: Root becomes diminished fifth/diminished fifth becomes root, and major third becomes minor seventh/minor seventh becomes major third. It's only a parallelism if the voices don't change function. Parallel perfect fifths are fine if the voices transform as well.

To make this into the dimensions of a traditional waltz, which has six sections usually, I had to make the form A, A, B, B, A, B, A'. This will doubtless be in my next sonata for solo guitar, so I will most likely make it into a scherzo by composing another piece to go on the interior. That would make it go to something like an A, A, B, A, C, D, C, A, B, A' form. Time will tell!