About construction in my trio for clarinet, violin, and cello, Absinthe, Spirit of Green Dream
Matt Fields
A harpsichordist asks:
"I wanted to ask you about the process of writing down one of these
(current tense, of course). Does it go from mind to Finale, or is there
paper inbetween? I know for some, they need to "scribble around" for
awhile, but then others, like a friend of mine at Caltech, puts his
combinatorics stuff straight into AMSTeX. It just seems with music, some
of the idea/editing/inspiration comes from auditory input, so....I wonder
if your ears work too, during the process."
I respond:
My personal method is pretty messy. I'm likely to take a fugue
subject, or a tone row, or a pseudo-sequence in invertable
counterpoint and mull it over in my head for a few days, scribble
stuff on paper, try it out at the keyboard, when suddenly a light-bulb
will go off and I'll discover a way of writing it that's ten times
simpler and thus clearer and more effective than anything I'd tried up
to that point.
Sometimes, though, I can work backwards... when writing some of the
recapitulatory musics in Absinthe, I knew that I'd carefully
constructed the expositions of them so that simply retrograding those
materials would yeild recognizable variants with the opposite dramatic
curve and amusing rhythms with an emphasis on the same harmonies, so I
worked on the recap with the expo in hand, manually adjusting things
on the fly to feel and sound nice for the instruments (and transposing
the entire mess to follow naturally from the preceding music and lead
naturally back to the pitch level that the piece started at). I also
then extended some of the music and trimmed some of it to yield
different timing. At every step of the way, I pause and work through
the music in my mind to see how it flows, both the local section and
all the music from the beginning of the piece onward. If there's
anything less than groovy about it, I try to catch that early on, for
were I to re-use some music in a later part of the piece, I'd want to
have fixed the grooviness ahead of time and not find myself fixing it
in multiple places later (this is sort of like code reuse in computer
programming---solve an engineering problem correctly once and re-use
the solution like mad).
For me, serial materials are an approach to counterpoint: I'm making
audible patterns with melodies and chords, while building phrases that
more often than not contain 12 pitch classes. I tend to build longer
phrases by overlapping forms of the row, e.g. using the last note of
one row as the first of the next, or using the last 2 notes of the
first row as the first two notes of an RI form...
Let me give an example of harmony and melody from my current piece:
P E Eb Ab D B A F C# F# A# G C
R C G Bb Gb Db F A B D G# D# E
RI9 A D B D# G# E C A# G C# F# F
I9 F F# C# G Bb C E G# D# B D A
x y z y' z' x' x" z" y" z` y` x`
All the "melodies" suggested by this chorale are forms of the row;
the "harmonies" are limited to revoicings of the three chords x, y, and z.
Phrases containing 12 pitch classes can be gotten by simply presenting
3 chords at a time, or you could cut the thing up other ways, like:
P E Eb Ab D B A F C# F# A# G C
R C G Bb Gb Db F A B D G# D# E
RI9 A D B D# G# E C Bb G C# F# F
I9 F F# C# G Bb C E G# D# B D A
x y z y' z' y z y' z' x' x" x' x" z" y" z` y` z" y" z` y` x`
and thus make phrases with different numbers of rhythmic events in them,
still with all 12 pitch classes, with every melody part of a row form,
and all the harmonies subsets of the chords x, y and z...
This doesn't work for every row in the world, and realizing it gets
you no closer to a composition than realizing that the key of the dominant
is just one sharp away from the key of the tonic. But it's a fun toy.
Let me slowly introduce another toy found in the same composition:
With any row in the world, you can make an aggregate (phrase of 12
pitch classes--see, by controling the rate of pitch classes you get a
consistent-sounding harmonic rhythm) by stacking the first 4 notes,
next 4 notes, last 4 notes into 4 3-note chords, thus:
P E Eb Ab D B A F C# F# A# G C
P E Eb Ab D B A F C# F# A# G C
P E Eb Ab D B A F C# F# A# G C
----------
(aggregate)
And as each P is exhausted, you could start another P in the same
voice, and thus continue the canon with every 4 chords being an
agregate. I like the open-fifths chord E B F# that starts each such
group---it's very dreamy, to me---so I would emphasize it. Again,
every melody is the row, so we have harmony and melody emphasized
together.
But this row happens to have its middle 4 notes invertable onto
themselves, e.g. an inversion of B A F C# is B C# F A. Thus:
P E Eb Ab D B A F C# F# A# G C (f# g d g# b c# f# a e...
It f# g d g# b c# f a e c eb bb(E Eb Ab D B...
P E Eb Ab D B A F C# F# A# G C (f#..
---------- ---------- ----------
(aggregates)
This mode of construction produces another never-ending inversion
canon: when each voice finishes its P, it can start an It just
afterwards, and vice-versa. Every melody will thus be from P or It,
and again, the harmonies are restricted to just a few chords, with
every 4-chord aggregate starting on E-B-F#, which I might (and did)
emphasize to bring out the dreamy quality of the music. (for
comparision, in Bartok's strings-percussion-celeste, at the moment
when quartal harmony of this kind first takes over, the celeste enters
with a serenely dreamy sound...).
But all of this I have to play with, sometimes for weeks, before I
spot a way of putting it to use in service of the dramatic images I
have in mind. I actually e-mailed my tone row to composer Frank
Brickle, who lives in Princeton NJ but is no longer affiliated with
the U there, and he mailed back to me a bunch of transforms and
patterns it suggested to him, all of which were potentially
groovey---for another piece---but it turned out I couldn't wedge them
nor about half of the patterns I'd found into the dramatic curve of
the piece. So it's kind of like building a 3-part invertable
counterpoint, then *not* using all 12 configurations in which it
works, only those that you need for your piece. I have limited
patience for those who use a composition merely to exhaustively
catalogue a pattern, not to generate beauty.