But recently, I’ve been reading a little of Henry-Louis de La Grange’s first volume (the 1973 edition) of his epic Mahler biography – managing some 200 pages and skimming another 100 or so before the book was due back to the library (being an interlibrary loan, the renewal policy is fairly limited).
In addition to that, I’ve been slowly working on the revisions for my music appreciation thriller, The Doomsday Symphony which I finished back in February but have been reluctant to follow through the process of slicing and dicing my way to the final product. I'm actually trying to avoid working on the complete rewrite of The Lost Chord because I know this will take more time than I have, now, but that doesn't keep ideas from bubbling up in the creative stew...
And since around May 1st, I’ve been busy sketching my new Piano Trio which I thought was going well till the other day when I finished the first two segments of the piece (less than a third of the work’s total length) I calculated that in 55 days I’ve spent over 240 hours producing some 101 pages of sketches (this does not include a few that pertain to the original idea for a piano sonata scribbled down in late-December last year) which have so far translated to 105 measures or 7 minutes of music…
The trio is basically a four-movement work in one movement, except in this case, the movements are cut up and spliced into the continuous fabric in various segments so that before one movement is finished, the next movement has begun and it may be a while till we get back to that point of departure. Consequently, this has involved a good deal of planning to balance the symmetries and proportions of the form this creates.
The major problem this past month has been working out the second movement which is a chaconne, similar to the one that formed the central arch of a five-movement violin sonata.
I’ve never been a big fan of the “sectional” variation form – thirty-two variations mostly in the same key and all, basically, the same form (say, “rounded binary”) one after the other, regardless of the amount of variety the composer can squeeze out of often very insipid material. While I love, say, Brahms’ “Variations and Fugue on a Theme by Handel” (something I actually could play, once upon a time), there have been few performances or recordings of it that didn’t strike me as the equivalent of Chinese water torture on the macro-structural level.
And chaconnes are basically the same kind of thing – a chord progression that repeats over and over while something noodles around above it. (The passacaglia, close cousin to the chaconne, can have similar issues.) There are very few that can hold my interest after a while.
So my idea (which I’m sure is not original) was to come up with a chord progression that can modulate which means that, rather than having 32 variations (or 512 measures) all in the key of, say, D Minor, it can actually have a continuously varying tonal palette.
By the way, I should point out two exceptions to this problem (at least for me) and both are by Bach: the Chaconne in D Minor for solo violin from the Partita No. 2 (which I used to play in Brahms’ transcription for piano, left hand which I originally took on when my tendonitis was acting up) and the Goldberg Variations which are not only sectional variations but also based on a repeating harmonic progression. The difference is, you’re never hit over the head with the idea "this is the same thing, over and over again."
So you might call my response to this, “Chaconne Awe.”
Anyway, I set up a series of chords that have a logical harmonic direction but which can also evolve in different ways. By carefully crafting the tension between dissonant three-note chords and standard (but not standardly used) major and minor triads, I created a pattern of chords that point to certain resolutions, thereby moving from one “tonal level” to another as the piece unfolds.
And a lot of this can be varied simply by using different inversions of these chords: in certain instances, a three-note (non-major/minor) chord could go off in a different direction in a different inversion; a close-position chord could have more tension than the same one in an open-position. A second inversion major triad will have a different sense of resolution than a first inversion or root position triad. By using these chords in a consistent manner, you can create your own harmonic context of dissonance and resolution.
To avoid the monotony of sameness in the rhythmic structure – the often pedantic pounding out of 4+4+4+4 measure units – I based the length of each variation on the structural proportions by dividing the time-line according to the Golden Section, something I’ve been doing for years, anyway. This means some variations are shorter than others and that, as the harmonic motion drives you to a particular climactic point, the variations becomes shorter until the rhythmic motion is driving you to that climactic point just as the harmonic motion is as well. (That’s nothing new: Beethoven did it all the time, writing shorter and shorter phrases as he approached a significant cadence.)
Above this – as in a traditional chaconne – would be a melodic layer, something that rises out of the chord progression and changes continuously or may, in itself, become the source of variations.
Only in my case, this layer becomes more independent until it seems to have no relationship to the harmonic layer.
In fact, the sense of line cadences at different points from the harmonic layer, only merging at certain significant points.
This creates a kind of temporal counterpoint that still fits “logically” within the harmonic and melodic expectations. Whether a listener senses this “logic” is not the point but it helps underline a hopefully emotional response to the idea of what a cadence – whether it’s by Bach, Beethoven or Schoenberg – can be (or should be, if the performer is at all aware of the emotional nature of what’s happening in the music).
Getting these two lines to work together was not a matter of just slapping notes down on a page (usually too often the way “modern composers” in any era are accused of working). There was still a context that needed to work harmonically as well as linearly, just like it did in all those counterpoint exercises I should have done when I was a student but usually didn’t because counterpoint in general was something generally overlooked).
Now, in a piano trio, there are so many ways you can subdivide the instruments. If my harmony is based on three-note chords, two string instruments cannot always be playing three notes, so they must be carefully worked out in such a way that this is possible. Also, having the melody in the piano meant it was either doubled in both hands or I had to work out some kind of “accompanimental line” so the texture wasn’t so spare but then this became another layer of complexity to work (contrapuntally) with the harmony and the melody.
There were days I just stared at blank pieces of paper, scratching out potentialities, only to sit back and think, “ya know, this would be a lot easier if it were for orchestra” – as if having more instrumental options made the instrumental challenges less challenging. Or “maybe it would be easier if I just started over and did something else.” Or “perhaps tomorrow will be better,” and I’d put it aside.
Then one day, without so much as an “aha!,” a solution presented itself without needing any significant changes, no need to “start from scratch” or any reason to doubt my sanity. Go figure…
The thing is, in order to make the chaconne work for the segment I was composing now, I needed to know what the whole chaconne was doing. On the other hand, sketching out the entire chaconne means that, when I finally do get around to writing that part of the Trio, it’s already done. Voilà…
So now I’m ready to start the first segment of the third movement, a scherzo, and probably the same thing will happen – I need to sketch out the whole movement, not just the portion of I need at the moment. Besides, it overlaps with the other segments as if fading in and out of our perception, so all of that has to be worked out in advance.
It’s not whether these are first, second or third movements, because they will appear in various orders at various times. In fact, the Trio ends not with the fourth movement but with the final segment of the first movement which is, essentially, the recapitulation of the opening, however affected it becomes by everything that’s happened in between.
But each movement is definable (easily recognizable) by its mood or tempo – or, in the case of the chaconne, its “procedure” – not by its location in the time-line. The scherzo is fast, the last movement to be introduced (if not the finale) is actually a slow (by comparison almost suspended) nocturne-like movement. Yet the tempo throughout is the same – the metronome set at a consistent “quarter note = 60” (the silent common denominator of a ticking clock) while the perception of the tempo frequently shifts by the number of notes we hear in a given pulse.
This creates something not nearly as complicated looking as Elliott Carter's 7-against-13 passages or metronome markings like 163.3 or Leon Kirchner's Piano Trio II which has six metronome changes in the first ten measures.
Well, anyway, time to get back to work.
- Dick Strawser