Wednesday, January 14, 2009

The End of the "Nightmare"

Sometimes it helps me, as a composer, to write the ending first. My rationale is, that way I know where I’m going. Maybe not always first but early in the creative process, one way or another. That doesn’t mean it can’t change by the time I get there: it just helps put everything else in some kind of focus.

So I’d been working out the different segments of this little two-minute scherzo, the one where four contrasting elements are “blenderized” into an interlude before the final movement. It’s not pulverized into a scherzo slurpie, just the idea of some kind of seemingly random juxtaposition of these fragmentary sections, a little more scrambled than your standard rondo with some of them overlapping and no real connection between them.

Because of its brevity and odd form, this is the one piece of the set of five (which has now become, somehow, a Violin Sonata) that probably wouldn’t stand on its own. Since the Nocturne (the first piece to be composed but the last of the five pieces as it’s set out) is also fairly short, these two could make a pair. The fact that the opening of the Nocturne is essentially the resolution of the 4th piece helps connect them as a unit which at this point could be called “Nightmare and Nocturne.”

I’m not sure how nightmarish this interlude really is: the recurring scorrevole section evokes that kind of mood, unsettling and a bit scary, followed by the kind of “suspended animation” where one might stop suddenly as if thinking “what was that!” The two other elements are too normal, perhaps, but then the more frightening aspects of a dream might be more frightening when compared to something fairly normal.

The only problem is the “suspended animation” bits may not be anticipatory enough. Each segment is built on different groups of six notes (hexachords), each with different harmonic content and what I’d chosen for this second element is basically a whole-tone scale. That makes it harmonically ambiguous but it also doesn’t really have an edge to it, lacking the tension implicit in the others’ dissonance. While Debussy’s whole-tone impressionism can be vague and, depending on how it’s handled, unsettling, I’m not sure yet how well that’s going to work, here. We’ll see. I’d already sketched out a few of these segments, but they, too, might change.

The past week or so was focused primarily on Mahler for my pre-concert talk with the Harrisburg Symphony and now I’m looking ahead to next week’s event, a school presentation at the Capital Area School for the Arts for an up-coming recital with Market Square Concerts, where the major work to be presented is Bach’s Goldberg Variations as well as several somewhat newer pieces by Ligeti, Carter (his “90+” written to celebrate a friend’s 90th Birthday) and Harrisburg’s own Jeremy Gill (his “Eliot Fragments” written to celebrate Elliott Carter’s 100th Birthday). In the midst of all that – and doing a little bit of blogging – I’ve been working on the last five measures of this piece, the end of the “Nightmare.”

Now, at this tempo, it literally flies by in about 13-14 seconds, so spending 7 of the last 11 days on it may seem “inefficient” but that’s the way it goes. Over the years, I’ve become a slow composer and I’ve found that writing fast music takes longer than writing slow music (actually, Berlioz complained about that, too, so it’s nothing new).

Scorrevole means just what it sounds like – scurrying. The original idea was that the whole brief piece would be barely audible, but then I thought I would make each of these fragments a bit louder till we’d reach the climax and then it would fade back into near-inaudibility. But then I started thinking a gradual crescendo from beginning to end might be more compelling. Chopping it off in mid-air at the very end would probably be even more effective, dramatically, like dreaming you’ve fallen over the cliff but you wake up in a cold sweat before hitting the ground. That kind of thing.

(There’s a cool piece by English composer Alastair King he called “Hit the Ground (Running, Running Running)” but I wanted to avoid thinking of this as “Hit the Ground (Splat)”...)

The other problem was how to write something like this that could end loudly.

It would be difficult for the piano to control the kind of rushing figures I had in mind at that dynamic level, so they should go from a single scurrying legato strand to playing them in both hands, first an octave apart and then two octaves apart. As the violin moves from a couple of beats per bow to fewer and fewer notes per bow until the final measure where it’s one bow per note, the piano then breaks into pounding octaves alternating between the right and left hands (a Lisztian approach I’d already used somewhat differently in the Chaconne). This helps the crescendo but also helps propel the momentum forward: it feels faster even though the tempo remains the same.

So here are the last five measures of the 4th movement of my nearly complete Violin Sonata.

Now, the freebie software I’m using is very limited (it wasn’t designed for the kind of stuff I want to do), so take these grains of salt as you check it out: the original is in 4/4 with 16th note sextuplets at 1/4-note = 90 – but since Finale’s Notebook doesn’t permit me to write it that way, I’ve transcribed it as ten measures of 12/8 with 8th notes at dotted 1/4-note = 180. Same thing, audibly, but I have to admit seeing all those 16th notes just makes it “look” faster!

Now, I haven’t figured out how to record the play-back with its geeky synthesized sound-approximations, so you’ll just have to look at it. At least it’ll give you an idea.

And now to write the opening and finish the second half...

-- Dr. Dick

No comments:

Post a Comment