A briefs relating to Application Composition is included
below.
‘Application
composition’ in the set of Preludes presented on the CD
“Atmosphere’s Edge.”
‘Application
composition’ is a heading by which I intend to respond to the
question of
interactivity. It is a term whose over-use
requires that one make a special effort to define it. The
question is no longer ‘what is it,’ but
rather ‘what does one mean when one employs the term.’ To
begin, an application composition is a
computer program written to do
three things. First, respond to a control source. In
my works, the control source means either
pitch tracking an acoustic instrument, or a controller invented for the
task. Second, manipulate the data so that
it can be
implemented in a specific control scenario, and set up higher level
control
mechanisms by which that control can be dynamically routed and
networked. Third, provide sound generating
routines that
are capable of receiving the processed data in ways that are
efficacious to a
performer on the control instrument, and a listener.
Generation of control data, processing and
mediation of that data, and implementation of the processed data in
sound
generating routines; interaction refers to the relation of these three
stages. It is another story, but I would
like to
suggest that this is actually a flow diagram of the performance
practice on any
acoustic instrument. It speaks to the
genesis of my ideas as I am first an acoustic performer, and the
acoustic
instrument is my main design paradigm.
In the
first instance of control source, this CD presents 9 different
applications
which respond idiosyncratically to the Japanese shakuhachi, the Chinese
xun, or
the western silver flute adapted with a glissando head-joint.[1]
Depending on the overall
application
design concept, the kind of derivations from the raw control data
provided at
the beginning of the pitch tracking process can be more or less simple
or
complex. Some applications require of
the pitch tracking only that it report on a standard midi resolution
of 0 -127,
especially for example when they are tracking general registration, or
when
they are being used to track interval sequence triggers which rely on
the
players familiarity with standard 12 tone equal temperament. Others applications provide arbitrarily set
microtonal equal tempered grids through which the ‘frequency
tracking’ is
routed, for example taking a bandwidth of three octaves and dividing it
into
150 equal steps. This kind of frequency
tracking reporting is quite useful in routines which reconstruct the
acoustic
source sound using synthesis routines. This
kind of hi-resolution data facilitates
’re-synthesis’ in a
non-technical sense of the term. In
addition, some applications track partials, and for example implement
their
amplitude values in controlling some timbre parameter such as an
intensity
index in a waveform synthesis routine. [2]
In the
second instance of control source, I am co-inventor of one such
invented
controller, a prototype instrument called the Cilia[3]. Three tracks on the CD present this
instrument. The Cilia is an
electronic
‘flute controller’ with raw control processed and
implemented in a software
application (MaxMSP). Its performance gestures are modeled after the
Shakuhachi,
the traditional Japanese flute. Like any
flute, the Cilia splits an air column produced with an open
lip technique. This is
nothing like the pressure sensors used on commercial wind instruments. The mouthpiece splits the air column and
derives control data from an analysis of the split air column’s
dynamics. In place of finger holes, there
are 5
high-resolution (values from 0 to 6000) three-dimensional (X,Y and Z)
track-pads
which enable the numerous finger techniques that can be used on
non-keyed
flutes like the shakuhachi. It is first a complex ‘event’ controller (sound
produced is simultaneous with the control gestures).
It is second
a ‘process’ controller (sequences, algorhythms,
sonifications, pre-recordings). This
priority comes out of an aesthetic
choice, but also out from an historical design problematic.
All electronic
controllers are by process controllers by default; it is not difficult
to
design triggers. The Cilia confronts the
more difficult task of making an instrument capable of the nuance of
control
that matches and surpasses the ‘event controller’ standard
set by good acoustic
concert instruments. The
Cilia’s control routing
concepts derive from an analysis of three categories of control found
in
acoustic concert performance; direct discrete control (direct
implementation of
triggers (whether simple switches or sequence tracking routines) and
direct
continuous control (data streams (e.g., continuous values from the
track-pads)), dynamic control
(controllers directly controlling controllers), and networked control
(controllers indirectly controlling controllers by way of networks of
controller).
With
this last articulation of categories of control, I cover the essentials
of my answer
to the second application requirement pertaining to the manipulation or
processing (“massaging, wrangling”) of raw control data. This discussion is too technical for a brief
such as this. Suffice it to indicate a
problematic and stimulate a polemic. Most
electronic controllers barely if at all design
dynamic
control. It is not easy to find examples
of complex networked control even at the margins of invention.[4] What
could be called networked control is discernable in acoustic
performance and I
would contend is what makes for ‘nuanced’ performance. As long as instrument designers are
specifically
interested in the human controller, acoustic instruments are still the
best
paradigm for controller design, and thus the standard to meet and
surpass. I believe that this is largely
acknowledged,
but one could still complain that the analyses of this paradigm have
not relied
enough on first hand performer experience.
The sound generating routines which
implement the control data are… personal. In
the world of film one encounters the expression
‘personal cinema,’
referring to film makers whose technique, vocabulary and interests are
so
idiosyncratic as to defy public classification. Well,
this has always been the promise of electronic
music. It is a place where the bricoleur
finds himself most at home. One purpose of
the Cilia is to enable a
composer to improvise with unheard of
‘orchestral’
resources. This idea of orchestra is
that of a single instrument capable of producing sound with an
indefinitely
complex inner horizon. Inner horizon of
sound refers to the complexity demonstrated within the simultaneity of
a single
musical gesture. Currently, timbre and
layering are the concepts by which I organize the Cilia’s sound
worlds. There are four types of timbre;
instrument
timbre, harmonic timbre, timbre density, and texture.
Of layering, there are three; monophony,
heterophony and polyphony. I have
customized complex and idiosyncratic FM, waveform, and additive
synthesis
techniques, sound file manipulation and spectral re-synthesis
procedures using
maxMSP. Aesthetically and sonically, the
sound ranges from the pure to the noisy; e.g., the sine wave to crafted
non-periodicity, the familiar semantic reference of a sound file to the
completely unfamiliar musings attainable through convolution techniques.
Finally, why is the application composition
a composition? It is a composition by
virtue of being rule bound. All pieces
on this CD are predetermined precisely by how interaction is written
down. Like a traditional score which
provides a set
of instructions, so too does the program articulate the conditions by
which
musical events are generated. The
program could be represented as a kind of consequence logic. It is structurally predetermined; most
generally, given ‘x,’ not just anything will follow! If structure is predetermined, then form and
specific content are encouraged. The
player can always say no to the consequence logic of the particular
application,
and introduce independent, unrelated and arbitrary
counter-considerations. In
short, this is structured improvisation.
These application compositions are
compositions by virtue of not being rule bound. But
a good performer will always say that one cannot say
‘no’ before one
has understood what it is to say ‘yes.’
Here lies a divide between what the critic could call
self-indulgent
performance, and on the other hand, structurally informed performance
which
through its peculiar ‘no’ is then capable of emotional
narrative. In the
end, music is always narrative, even when it attempts to be
non-narrative and …
interesting. This is the entry way into
another
topic, what does music mean? I only
knock on the door. More pertinent to
the immediate topic of application composition is the distinction, just
below
the surface, between composition and
improvisation.
Too much
is made of this distinction. Experienced
score-bound composers and script-less improvisers both know this. Composition and improvisation are two
sides
of a coin. The composer premeditates,
bringing a wide array technical skills to bear on a musical problem. It is not an automated process; there is
spontaneity, surprise, accident, trial and error. Similarly,
no improviser goes into a
performance without having practiced, without a set of techniques and
musical
agendas, ready to be used as the moment inspires. Even
improvisers who endeavor to be
anti-technicians have inversely honed their skills so as not to sound
like they
are in control. It is nevertheless,
control.
This last
item is one of those obvious things to say that nevertheless needs to
be said.
Bruce
Gremo
January
15, 2009