Dubious invention

FROM: mclaren
TO: New Alternative Wanking List
SUBJECT: dubious invention

Jeff Scott appears to have credited me with the
invention of something involved with dynamic lattice
programs for navigating ratio space. This seems highly
unlikely for several reasons.
First, Carter Scholz wrote a routine for JICalc
which sounds pitches when the user clicks on specific
points in the ratio-space representation of a tuning.
This upgrade to JICalc appeared, if memory serves,
someime around 1993.
However, shortly thereafter Carter Scholz wrote
a much more comlex HMSL program which he used to realize
his composition LATTICE 3327. Carter's program uses real-
time calculations and multiple MIDI channels to send
out pitch-ebnd messages which retune successive notes
of a DX7II equipped witht he E! board to any desired
point in a 7-dimensional ratio space.
Moreover, Carter's HMSL program not only generates
pitches corresponding to those ratio-space coordinates in
7 dimensions, it also uses a variety of different metrics
to evaluate so-called "harmonic complexity" in 7-D ratio space.
The user can choose which metric is used by hitting keys
on the computer keyboard.
Lastly, Carter's HMSL program navigated 7-D ratio
space according to gradients which pointed in the direction
of "more complexity" (using the current measure) or "less
complexity" and also by specifying the general algorithm.
If memory serves, drunakrd's walk and Levy flight and several
others were the choices available.
-----
Doubtless others prior to Carter Scholz in late 1993
did these sorts of things. My vague suspicion is that the
League of Automated Composers did something like this on
their KIM-1s back the late 1970s, but an early computer composer
at a university in British Columbia did something similar even
earlier using real-time FM on a PDP-11.
------
The important point, however, remains that whomever
invented this particular wheel, the wheel itself has 4 sides
and is made out of marshmallow. This wheel don't work, kiddies.
My own experience with using Csound to realize various
precompositional studies in ratio space around 1994 gave rise
to the dark suspicion that ratio space metrics bore no audible
relationship to anything the human ear can hear.
Subsequent double-blind listening tests, using a Turbo Pascal
program which renamed and scrambeld audio files and wrote a record
of the new names to a text file which could be examined later,
clinched my suspicion that ratio space involves mathematical
chimerai which do not relate to anything the human ear can hear.
Consequently, manipulations of departed mathematical ghosts
in ratio space prove musically meaningless. Ratio space itself
constitutes an ignis fatuus as far as actual music is concerned --
a delusion with no discernible relation to either the roughness
of intervals, pitch height, logarithmic interval width, or any
other known property of msuical intervals or individual musical
pitches.
NOTA BENE: Representations of TUNINGS in ratio space can
exhibit some useful info -- for instance, you can see quickly whether
the JI tuning is subharmonic or harmonic or symmetrical (viz., has as
many utonal as otonal pitches), etc.
However, in terms of extracting useful musical info via
mathematical operations in ratio space, this is not possible, any
more than it is possible to turn lead into gold by chanting various
incantations, or to detect the luminferous ether with the proper
arrangement of interferometers and mirrors.
Fokker's introduction of 3-D ratio space followed the much
earlier use of 2-D ratio space by Hugo Riemann and other German
theorists even earlier in the 19th century. As far as illustrating
abstract properties of musical *SCALES*, these diagrams had some
limited use and served some very limited purporse.
When people unwisely moved on from diagramming scales to
trying to use ratio space to compose music or extract information
about actual musical properties of intervals, the whole enterprise
headed into the ditch and wound up spinning its wheels in the mud.
Some people periodically claim "Oh, no, I create music using
ratio space lattices..." Experience has shown that ratio space
diagrams and/or lattices represent a debilitating hurdle, rather than
a facilitating tool which helps people compose music. The proof is
clear and straightforward -- all the people who avoid using ratio
space compose much more music than the people who try to use ratio
space to compose music. This is no coincidence. Ratio space
represents a roadblock, a pothole on the road to composition, as much
of a conceptual barrier to composing music as the notion of
phlogiston was to creating a valid science of thermodynamics.
---------
--mclaren

> The important point, however, remains that whomever
> invented this particular wheel, the wheel itself has 4 sides
> and is made out of marshmallow. This wheel don't work, kiddies.
> My own experience with using Csound to realize various
> precompositional studies in ratio space around 1994 gave rise
> to the dark suspicion that ratio space metrics bore no audible
> relationship to anything the human ear can hear.

Absolutely correct. Distance in the lattice doesn't reflect
perception, at least for triangular lattices. Tenney complexity
is the only simple metric for JI that seems to work well enough
to mean anything, and its visualization as a lattice metric
would require... thinking about volumes in an octave-specific
rectangular lattice, I guess... or city block distance on an
octave-specific rectangular lattice whose rungs are scaled by
the log of their vector class ('zthat right, Paul?)?

> Consequently, manipulations of departed mathematical ghosts
> in ratio space prove musically meaningless. Ratio space itself
> constitutes an ignis fatuus as far as actual music is concerned --

...as far as *acoustics* is concerned, since:

> NOTA BENE: Representations of TUNINGS in ratio space can
> exhibit some useful info -- for instance, you can see quickly
>whether the JI tuning is subharmonic or harmonic or symmetrical
>(viz., has as many utonal as otonal pitches), etc.

Correctamundo. And that's what we use 'em for, over on the
big list, and tuning-math.

-Carl Lumma
instrument builder

--- In crazy_music@y..., Carl Lumma <carl@l...> wrote:

> or city block distance on an
> octave-specific rectangular lattice whose rungs are scaled by
> the log of their vector class ('zthat right, Paul?)?

That's right! It's the Tenney lattice.