compositional methods

--- In crazy_music@y..., xed@e... wrote:

/crazy_music/topicId_unknown.html#716
> Joe, do you hear a specific microtonal melody in your head,
> and then try to find the ptiches in ratio space? Or is the process
> interactive? In otherwords, do you hear a kind of template, and
then refine it (sort of like twisting a telescope into focus) as you
move through ratio space?
> Reason I ask is that the latter tends to happen to me,
> particularly when I compose in JI.

mclaren poses here an *extremely* interesting compositional
question.... i.e. how does one actually go about it, and what is the
process. The suggestion that the ideas are first "in the head" is a
significant one. Basically, this is the way *I* work all the time...
I virtually *always* HEAR something before I write it down or try to
find the pitches to match it.

So, in that sense, I don't "improvise" if, "improvise" means just
trying out different things and settling on something acceptable.
However, I doubt very seriously that fine improvisers work that way,
either. Most probably they *also* are striving to find something
that is ALREADY in their heads... and use their muscular-auditory
coordination and senses to realize that aim...

So how does this process differ in microtonality from 12-tET? Well,
of course, I have had VASTLY more experience in 12-tET, having
composed *well* over 10 hours of music in that tuning.

So how is this process working in microtonality? Well, in this case,
I believe I absorb the scale as a kind of "template" and STILL try to
find pitches I am initially hearing *in my head* matching these
pitches to the pitches of the scale.

For this reason, I believe my basic "style" of which I seriously
believe I have one, after many years of composing, is for all
practical purposes THE SAME in 12-tET AND in microtonality.

It's just that I have a new, "refreshing" (for *me* at least)
template of pitches that remains in my mind after working with a
scale a bit.

Then, I try to "find" the pitches I need for my piece from this range
of possibilities. Since the pitches are xenharmonic and,
therefore, "fresh" to me, the compositional experience at the moment
is more interesting and enjoyable that it would be in 12-tET.

And, generally speaking, when *I* am getting more out of my
composing, an audience, conceivably, will as well!

__________ _______ ________
Joseph Pehrson

Joseph responded :

>
> mclaren poses here an *extremely* interesting compositional
> question.... i.e. how does one actually go about it, and what is the
> process. The suggestion that the ideas are first "in the head" is a
> significant one. Basically, this is the way *I* work all the time...
> I virtually *always* HEAR something before I write it down or try to
> find the pitches to match it.
>

Yeah. I believe my problem is that, even though I spend most of
my time improvising on an instrument, when I do write, it is
nearly always "from my head". Now, my head doesn't naturally produce
things that have never been heard before, just like Tarzans ability
to learn English language speach by finding English language books
is highly suspect. However, if I pour enough of certain sounds
in, eventually they will settle, metamorphise and come back out
as original music.

I have had very few ideas which were distinctively microtonal and
they were basically little splashes in an otherwise 12 et piece
of a 7/4 seventh or a neutral third.

> So, in that sense, I don't "improvise" if, "improvise" means just
> trying out different things and settling on something acceptable.
> However, I doubt very seriously that fine improvisers work that way,
> either. Most probably they *also* are striving to find something
> that is ALREADY in their heads... and use their muscular-auditory
> coordination and senses to realize that aim...
>

I improvise all the time, some of the time its real-time-creation,
more of the time its real-time-rearranging of known chunks, like
building something new with Legos, and some of the time it stinks.

As I said before, improvisation is usually not the way I will
get myself generating a piece. (Thouh some improvisations were
actually the realisation of a piece "hey, thats been in my head
for six months now, I should write it down or something".)

In improvisations, (electric guitar) I freely use 'gestural'
microtonality with very little idea of specific pitches desired.
Its more like "I want this note to have a little more emphasis so
make it this much sharp". "Do I want this dyad to climb out a little
bit more..." Notice that these are non-verbal thoughts, but
real-time feedback across your ears/heart/fingers.

>
> So how is this process working in microtonality? Well, in this case,
> I believe I absorb the scale as a kind of "template" and STILL try to
> find pitches I am initially hearing *in my head* matching these
> pitches to the pitches of the scale.
>

I think this is what will happen with me, but it sees to take a long
time for me to make headway. I DO have to wrap myself around some sort
of "understanding", something I didn't have to do with Western Harmony
since I learned it almost like a baby learns to talk. But a new
language... I would both have to practice it and do the 'grammer'
sort of stuff.

If I could learn microtonality without ear-work I WOULD be Tarzan!

Bob Valentine

>
> Joseph Pehrson

--- In crazy_music@y..., "D.Stearns" <STEARNS@C...> wrote:

/crazy_music/topicId_unknown.html#757

> <<And, generally speaking, when *I* am getting more out of my
> composing, an audience, conceivably, will as well!>>
>
> Yeah, even if it means you and your audience are now one and the
> same... you!
>
> --Dan Stearns

Hi Dan...

Thanks for the commentary. Well, fortunately, with the xenharmonic
stuff, this *isn't* presently the case. In addition to the Internet
things which *many* people have heard, including yourself, I have had
*multiple* public performers of both the live and electronic
xenharmonic music of late, so, fortunately for me, it hasn't been
so "hermetic..."

best,

Joe

________ _______ _____
Joseph Pehrson

FROM: mclaren
TO: The new Alternative Wanking List
SUBJECT: compositional methods

Joseph Pehrson has written about
his methods for composing xenharmonic
music, also mary Ackerley. And since one
of the words in the title of this list
is "MUSIC," discussion of various methods
which prove useful for composing microtonal
music seem far more appropriate for this
forum than numerological mumbo-jumbo such
as "Tenney dissonance" and "harmonic
entropy," whose relevance to real microtonal
music in the real world remains so minimal
as to require a scanning tunneling electron
microscope to detect.
Experience shows that my own compositional
methods fall into 5 broad categories:

[1] Write down the notes using standard
European common practice notation.
This involves a form of tablature. The
notes on the 5-line staff represent the MIDI
note numbers, not the pitches usually indicated.
Thus, in (say) 49 equal tones per octave, a
C above middle C might indicate an interval
smaller than a minor third above middle C,
whereas in 5-note Pythagorean the same written
note indicates a pitch more than 2 octaves
above middle C.
Sometimes when using this compositional
method it proves useful to write the notes
down first on conventional score paper,
and only transfer them to the computer for
MIDI sequencing when the paper composition
has reached a stage of final polish. In other
cases -- particularly when using complex
embedded n-tuplet rhythms -- it proves
easier and more musically effective to enter
the notes directly into a computer program
such as Finale.
ADVANTAGES: This compositional methods
allows extraordinarily complex rhythmic
relationships to be realized with great
precision. It also allows extremely precise
control of polyphony -- for instance, the
composer can insure that polyphonic lines
never cross, or that no vertical note-complex
has more than a given number of pitches, or
that metric modulation can proceed with
great smoothness. Other advantages include
the fact that such a composition can be
brought to state of high perfection, nudging
and tweaking individual notes in a feedback
process that involves listening, making
changes in the score, and resequencing the
MIDI file.
DISADVANTAGES: By far the biggest
disadvantage of this method is the fact that
enormous amounts of expression must be added
*after* the MIDI file has been created. This
not only involves a great deal of work (nudging
individual grupetti back and forth in time,
adding a specific MIDI velocity for each note,
ramping the tempo map up and down for particular
sections of the score), it also creates more
serious problems. Specifically, this method
of compositional tends to disconnect the dynamics
and expressive nuances of the composition from
the larger-scale comopsitional structures (viz.,
phrases, periods, chord progressions, recapitulations,
etc.) Experience has shown that an extremely
important part of making a composition work
involves the close connection twixt expressive
nuance and the compositional structure. For
instance, using rubato combined with a slight
crescendo in a particular phrase can make or
break not only that musical phrase, but the
entire section in which it occurs. This means
that entire section of the music (at least in
my experience) must sometimes get re-composed
because when the pitches and vertical note-complexes
come out, they just don't work due to their
relation with previous sets of expressive
dynamics and articulations used in a previous
musical section. During live performance or
live improvisation, the performer or improviser
has the luxury of thinking ahead and preparing
for the next musical phrase with proper phrasing
and articulation and dynamics and rubato...but
when composing entirely on paper (at least in
my experience) this becomes much more difficult.
Accordingly composing using this method of
writing the notes down beforehand often creates
difficulties and produces unforeseen problems,
although it also offers some remarkable musical
advantages not available otherwise.
EXAMPLES: 5 tone equal piano piece #5 from
"240 Piano Pieces," the 47 and 48 and 49 and 50
tone equal compositions from "Mclaren - Microtonal
Music: Volume 5."

[2] The opposite pole from writing it all
down beforehand involves improvising a section
live and practicing it until it works, then
simply switching on the MIDI sequencer and
recording it as MIDI data. This method has
the advantage that it allows the utmost nuance
and emotinal affect...at the price, however,
of limiting the complexity and subtlety of
the music to the composer's technical proficiency.
With MIDI sequencers, it's possible to get around
this problem to a certain extent. Often I will
slow down the MIDI sequencer when recording and
"think in slow-time" to create rapid chromatic
runs or complex virtuosic sequences which sound
convincing when speeded up by a factor of 150%
or even 200%. "Thinking in slow-time" involves
putting your brain into a kind of place in which
you try to visualize what the relatively leisurely-
paced melodies and chord progressions you're entering
will sound like when sped up. This takes a lot
of practice, and some days it just doesn't work.
However, with enough experience, "thinking in
slow-time" can prove an effective technique for
getting virtuosic orchestral or instrumental lines
without losing the human articulations and nuances
that go with a live performance.
However, there's a sharp limit to the amount of
formal sophistication you can create by doing it all
live. Inevitably, my experience shows that creating
a composition using other methods invariably produces
a richer and more complex musical structure. Now,
if musical structure be the main issue, pure
improvisation won't work. But the human imagination
can prove surprisingly complex (a number of Chopin's
and Debussy's piano pieces appear, from the autograph
evidence, to be improvisations written down), and
complexity of musical structure remains a subsidiary
value when all is said and done. The overall emotional
impact of a piece of music counts for far more than
any arbitrary intellectual measure such as "formal
complexity." Thus, this method has it uses...particularly
when combined with other compositional methods mentioned
here. For example, sometimes I start out by improving
a section, then continue using other compositional
methods, then if I hit a tough patch, move back to
improvising a section, and so on.
ADVANTAGES: Pure improvisation creates the most
"natural" sounding and the most musically nuanced
possible performance -- important with MIDI!
DISADVANTAGES: Pure improvisation tends to favor
virtuosic effects at the expense of musical depth,
and also tends to limit the composer to hi/r technical
proficiency level, placing a bar of musical complexity
AND overall musical coherence beyond which the composition
can't go in real time. In the end, compositional methods
which work outside of real time tend to give the composer
an unfair advantage over pure improvisation, since
outside of real time the composer has the luxury of
going back and changing earlier sections to fit with
later ones, develop transitory themes and harmonic
progressions more fully, etc. A pure improvisation is
like a sledgehammer tube of toothpaste -- what you get
is what you get, and you can't change it. MIDI allows
this limitation to be overcome to a certain extent,
but only (realistically speaking) insofar as the other
compositional methods mentioned here get applied to
an improvisation *after* it has been recorded and
sequenced.
EXAMPLES: The theme for the 19-tone equal variations
on "McLaren - Microtonal Music: Volume 2," the piece
in the 14th root of 3 on "McLaren - Microtonal Music: Volume
1."

[3] One big idea.
This remains the sole and exclusive compositional
technique for most so-called "new" music in serious
music circles nowadays. The "idea" can involve anything,
but its distinctive characteristic remains its tendency
to overwhelm the music itself. For example, Gyorgy
Ligeti wrote a composition which calls for 100 metronomes
to get wound upa nd then slowly wind down and slide out
of synchrony -- the composition ends when the last metronome
stops ticking. Other "idea" compositions involve, for
instance, Nancaroow's "Canon X," in which one polyphonic
line starts out at a slow tempo and progressively speeds
up while the other polyphonic line starts at a very rapid
tempo and progressively slows down. Iannis Xenakis' best-
knwon compositions involve the "big idea" of treating
individual musical notes as atoms in a gas and lettting
many (50 or even 70) instruments use independent polyphonic
lines to create a statistical overall effect. John
Chwoning composition "Stria" is built around the "big idea"
of using the modulation index of FM synthesis to control
timbral morphing, while Harry Partch's "And on the Seventh
Day Petals Fell In Petaluma..." is based on the "big idea"
of creating a series of one-minute-long duets for various
combinations of his microtonal instruments.
ADVANTAGES: This advantage of this compositional technique
is that you can create a striking piece of music using a
very simple basic idea. The disadvantages include the
fact that once having created such a composition, other
compositions using the same or similar "big idea" tend
to sound trivially similar -- also, this kind of compositional
procedure creates the very real trap of spending all your
time running around hunting for different new "big ideas"
to use for compositions. Once again, in the end the
main point involves the dramatic gut-punch and emotional
effect of the overall composition. If a "big idea" helps
create that, great...but all too many contemporary composers
spend their careers in a frantic hunt for ever more exotic
and bizarre "big ideas" (burn a piano, flip a coin, roll
a naked girl in paint across a piece of manuscript paper,
play a 'cello in the nude, ad nauseum) until the music
itself gradually recedes into unimportance and only the
desperate quest for the shock value of "big ideas" remains.
In the end, music ain't an idea -- it's sound, and this
compositional technique can tend to submerge that basic
fact.
EXAMPLES: 26-tone equal piano piece number 3 from
"240 Piano Pieces," which uses a Maxwell-Boltzmann
distribution of note density and tempi; metal bar scale
Csound piece from "Mclaren - Microtonal Music: Volume
4," which uses the free-free inharmonic vibrational
modes of a theoretical metal bar for both melodies
and harmonies: and 28-tone equal piano pieces number 2
from "240 Piano Pieces" which uses 2 identical melodies
an octave apart starting at the smae time, but differing
in tempo by 1%.

[4] Composition using layers, typically proceeding
one thread or strand at a time, with slow continual
modification of each melodic line and/or harmony to
relfect the growing structure.
This method seems favored by the most composers
as a general approach. By allowing the composer to
work outside of real time, it allows much greater
complexity and subtlety, and also give the composer
the opportunity to unify the composition to a much
greater degree than with most other methods. For
instance, by modifying ealier or later musical phrases
to foreshadow or echo later or earlier material, the
composer can create a much greater impression of
musical unity. Alternatively, if on repeated
listenings a section proves too "dead" sounding
or too repetitive, the composer can simply slice
it out and insert new material to add variety. Since
all music involves a continual dynamic balance twixt
unity and variety, this compositional method allows
the best overall balance twixt unity and variety.
ADVANTAGES: This compositional method, in
my experience, serves as the msot flexible vehicle
for the widest possible variety of musical material.
Whether in composing a virtuoso solo study or a complex
dense polyphonic structure, this method of composition
seems to distort the musical material least of all.
By contrast, using a "big idea" can often twist the
music into the service of an obtrusive abstract
concept...while composing using notes on paper
tends to overemphasive the pitch and rhythm at
the expense of less visible (but equally musically
important) qualities like drama and excitement and
emotional impact.
DISADVANTAGES: Composing by laying down layers
can and in my experience sometimes does lead the
composer into dead ends. There is a tendency to
produce one or two melodic lines, or a combo of
melody and harmony, which sound great by themselves
but less impressive when combined. This problem
couldbe called "the war of the parts against the
whole." When laying down a composition in layers,
it's hard to keep in mind that the overall goal
remains creating a gestalt that works musically...
It can be frustratingly easy to create two
individual melodic lines which sound impressive
and very effective each by itself, but which
lose their musical impact when combined. Often,
this compositional method can create frustration
for the composoer because it sometimes becomes
necessary to throw away wonderful harmonic
progressions or melodic passages because however
impressive and effective they sound by themselves,
they just don't fit with the overall composition
as it grows and lengthens throughout time.
EXAMPLES: 43-limit just intonation piece
from "Beyond All Limits -- Volume 1," 28-tone
equal piano piece #4 from "240 Piano Pieces,"
Csound infinite continued fraction scale
compositon #1.

[5] Pick an overall structure and fill it in.
This compositional method, taught as "the" method
in most undergrad college compositional courses
(if the textboosk provide any guide), remains
extremely useful. Abandoned for a number of
decades beginning around 1905 or 1910, this
compositional method has re-emerged as composers
realized by around 1975 that there exist only
a finite number of comprehensible overall methods
of structuring compositions.
Twixt WW II and 1975, the emphasis in "serious"
music (so-called) involved the attempt by each
composer to create hi/r own private code, what
Stockhausen called "a separate genetic code"
embedded in eachcomposition which serves as its
structural blueprint. All too soon, however,
composers found that while they could create private
compositional codes of unlimited variety and
complexity, listeners could not perceive an
unlimited number of limitlessly complex private
compositional structures.
By around 1975, composers began to realize that
many traditional forms owed their musical power
not to tradition but to basic aspects of the
human ear/brain system and the human nervous
system. For example, long slow accretions of
pitch tend (whatever the composer intends or
mathematically tries to produce) to create a
sense of tension. Likewise, sudden highly
chromatic fortissimo staccato note-clusters
tend to generate (regardless of what the composer
intends or calculates) to create emotions of
distress and anxiety. And so on.
As a result, age-old compositional structures
like the A-B-A arch form and the rondo and the
variation form, however they may get stretched
by contemporary composers, have retained their
musical effectiveness.
ADVANTAGES: Classic methods of structuring
music, like the binary form or the rondo form,
produce musical effects which seem more reliably
audible than new private compositional codes
invented by the composer for that piece of
music alone. These relatively simple classic
methods of structuring music offer the advantage
of relatively clear audibility, at the expense
of formal complexity. In the Romantic period,
composers added to formal complexity by lengthening
the overall classic structure (viz., adding more
recapitulations to a sonata-allegro form, or
tacking on more extended digressions when
the countersubject of a fugue entered), but these
options are generally speaking not available today.
Romantic composers had not only the advantage of
an audience more steeped in traditional forms
than today's audience, but the Romantic composer
also dealt with an audience used to a more leisurely
musical pace (which allowed much greater formal
complexity courtesy of extensions and digressions
from classic forms like the rondo). Today's
audience demands faster-paced symphonies -- to
my knowledge, for instance, no contemporary young
composer has written a 90-minute-long Mahler-length
new symphony. New symphonies today written by
contemporary composers tend to be much shorter
than traditional Romantic-era or even classical-era
symphonies...often as short as 10 minutes or 15
minutes, compared to the broad expansive musical
forms of a Brahms or a Wagner, to see nothing of
a Mahler. However, another great advatnage of
using a classic form like the rondo or the
variation form is that it allows the composer to
produce a very comprehensibly organized piece of
music without spending unnecessary effort to
dream up some exotic new method of musical
organization (which might not prove audible to
the audience, after all's said and done). For
instance, Lou Harrison uses the clever method
of sometimes simply writing "pivot" notes at
key points of a phrase and then drawing in broad
arcs to indicate the general length of the phrase--
he then goes back and fills in the phrases in
detail later. This allows maximum control over
the general pacing and large-scale audible structure
of a composition.
DISADVANTAGES: The flip side of that coin
involves the tendency to focus on the macro-level
and ignore the individual melodic or harmonic details
which can often add much of the life and verve to
a piece of music. By tending to force the composer
into a panoramic view of the overall piece of music,
this compositional method can create an overall
sense of "blahness"...the sense that something
musically comprehensible and coherent is going on
at a large formal scale, but about which the listener
doesn't much care--since the specific vivid surprises
and quirks and unexpected melodic and harmonic turns
are lacking. Since all music involves a dyanmic
balance twixt unity and variety, this compositional
method can create a danger by overemphasizing
predictability and large-scale form at the expense
of those crucial unpredictable exotica that often make
or break a composition...for instance, the abrupt
breakdown of the two-melody pattern at the very
end of Bach's second prelude (c minor) in WTK I,
or the unexpected juxtaposition of different musical
styles that occurs in Stravinksy's "Rite of Spring"
(third section).
EXAMPLES: Second movement from 15 tone equal
piano concerto (an ABA' binary form where A' uses
the inversion of the primary melody in A) from
"McLaren - Microtonal Music: Volume 3"; 19 tone
equal variations from "McLaren - Microtonal Music
Volume 2"; 9 tone equal canon by inversion from
"Mclaren - Microtonal Music: Volume 1."
---------
How do these general compositional methods
relate specifically to microtonal music?
In my experience, very often the tuning
provides either the "big idea" or part of the
binary form or rondo form or variation form...
and the tuning certain exerts a powerful
influence on an improvisation. For example:
improvisations in 17 equal will tend to
emphasize 17's melodic brilliance, while
improvisations in (say) 19 equal typically
gravitate more toward the conventional-sounding
major and minor harmonies available in 19.
In a non-just non-equal-tempered tunings
using an inharmonic series, improvisations
tend to sound more expansive than in, say,
harmonic series 1-60 or subharmonic series
1-60. In a relatively higher JI limit such as
61, composing using layers tends (in my
experience) to take advantage of the increased
melodic subtleties available in such high
JI limits, while composing using layers in
(say) 17-note or 19-note Pythagorean tends
to generate more active dyanmic composition
to make up for the more limited melodic
resources.
In my experience, the harmonies available
in a particular xenharmonic scale don't exert
that much influence. In a JI composition, for
instance, if while composing a piece it turns
out that a required chord lacks a smooth
fifth I will typically merely use an octave
or a third instead and the listener will typically
"fill in" that dyad as a functional triad if
it occurs in the context of other smooth
functional JI triads nearby. Likewise, the
lack of smooth triads in equal temperaments
like 11 or 13, or in JI non-octave tunings which
may use a perfect fifth but no recognizable thirds,
or in exotic NJ NET tunings, has little overall
impact on the overall structure or emotional
effect of the composition (in my experience).
The reason for this is that melody proves
far more important in music than harmony,
and even in some JI compositions in which I
have deliberately made use of triads with p5ths
which are rougher than just by a comma, the listener
often gets carried along by the melodic momentum
to such an extent that the roughness or smoothness
of specific harmonies just don't seem to matter.
Much more important that the acoustic smoothness
of the intervals available in a given xenharmonic
tuning are issues like the presence of absence
of overall musical drama...the degree of excitement
created by a melodic arc...and the degree to which
a large-scale formal buildup (which Nadia Boulanger
called "the long line") pays off with a musically
satisfying conclusion. Compared to these issues,
acoustic smoothness of intervals in the tuniing
(as measured by pseudoscientific gibberish like
"harmonic entropy") plays a trivial role of
the utmost inconsequentiality.
---------
Very often, certain tunings will inherently
suggest certain large-scale musical forms. For
example, 35 tone equal immediately suggests a
rondo form -- 7 equal followed by 5 equal followed
by 7 equal followed by 7 + 5 equal simultaneously
followed by 7 equal followed by 35 equal as a
chromatic field (and so on).
Or a Wilson CPS, which contains other smaller
CPSs within it, may suggest a rondo form or a
binary form -- A B A, with A (say) a dekany and
B a hexany inside it, etc.
Certain non-just non-equal-tempered tunings
may immediately suggest specific overall musical
forms -- for example, if an inharmonic series
is self-similar it may offer the xenharmonic
composer a particular musical form like a canon,
in which the leader might use on section of an
"octave" as the leader of the canon (in Erv's usage
-- my term is "cyclic multiplier," any repeating
frequency multiplier) and a self-similar section
of the same "octave" with smaller NJ NET intervals
as the follower.
Or, for example, a sonata form might use
a harmonic series for the statement of the theme
and a subharmonic series where the classical
composer would ordnarily modulate to a secondary
key center. In an NJ NET tuning, these would
take the form of an inharmonic series versus
the subinharmonic series.
---------
Clearly one of the great advantages of composing
outside 12, aside from the sheer increase in the
composer's emotional vocabulary, is the tremendous
potential for new and exciting variants of audibly
comprehensible compositional methods. One of the
great problems facing late 20th century composers
is the relative exhaustion of traditional compositional
methods -- whenever a tonal composer modulated to
the relative minor in 12, for instance, everyone
knew what it was going to sound like. Efforts to
break out of traditional compositional procedures
and invent entirely new ones enjoyed only limited
success, since as Fred Lerdahl and Leonard Meyer
have pointed out comprehensible large-scale musical
structures and audible methods of musical organization
tend to take much of their power for structuring
music from the close linkage between sensory consonance
and cultural musical syntax. This linkage cannot be
broken arbitrarily...80 years of experience by
audiences who find atonal or statistically or
purely mathematically-organized music audibly
unstructured assure us that basic aspects of the
human nervous system and the human ear/brain system
constrain the range of possible audible compositional
methods on many levels.
One solution involved the arbitrary creation of
large-scale statistical methods of organizing
music -- Ligeti's and Penderecki's and Xenakis'
"statistical" structuring of music. This tends
to lead to a dead end, however, since eventually
every "statistical" piece of textural music tends
to sound alike after a while -- large-scale jittery
sounding "fuzzy" stuff going on, lots of simultaneous
glissandi, and so on. Lacking are the kinds of eloquent
melodies capable of the "long line" which proves so
important in creating effective musical drama.
Another solution invovles greatly increasing
rhythmic complexity while arbitrarily reducing the
tonal language to a relatively less chromatic
(often modal) subset of pitches. This solution was
pioneered by Conlon Nancarrow can continues in use
today by many younger composers such as Julia Wolfe
and Michael Gordon, with considerable success.
However, a third solution involves refracting
traditional compositional methods (such as the
5 mentioned above) through the lens of a microtonal
scale, creating a rainbow of new musical possibilities
based on tried and true methods of musical organization.
As J. J. Nattiez has pointed out, the central pathology
of so-called "serious" modern music involves "neopathy" --
the obsessive compulsion to do something new at all
costs...even at the cost of destroying audible
comprehensibility and perceptible musical organization.
Xenharmonic tunings offer such a solution.
---------
While Joseph Pehrson and Mary Ackerley have written
in some detail about the compositional methods they use
to create microtonal music, other practicing microtonal
composer like Jacky Ligon and Jeff Scott and Kraig Grady
have not.
Have about it, folks?
I know something about the overall notation you use
for some of your music, Kraig, from your article in
Xenharmonikon... But what about giving us some details?
---------
--mclaren

--- In crazy_music@y..., xed@e... wrote:
> FROM: mclaren
> TO: The new Alternative Wanking List
> SUBJECT: compositional methods
<snip>
> --mclaren

Excellent article. I wish I had more to say about my compositional
method. As some of you know, my pieces are, apart from the
improvisations, mostly pop type music running towards comedy. How I
compose: something pops into my head, usually a vocal or instrumental
line, and it grows longer and develops in my mind's ear until it is a
whole piece, at which point I write it down as a lyric sheet with
chord changes and any instrumental riffs that are important. About
half the time the whole idea does not develop, so I write down the
fragment and put it in my files. Sometimes I can come back and look at
the fragment and finish it, sometimes not.

John Starrett

--- In crazy_music@y..., "John Starrett" <jstarret@c...> wrote:

/crazy_music/topicId_726.html#864

> --- In crazy_music@y..., xed@e... wrote:
> > FROM: mclaren
> > TO: The new Alternative Wanking List
> > SUBJECT: compositional methods
> <snip>
> > --mclaren
>
> Excellent article. I wish I had more to say about my compositional
> method. As some of you know, my pieces are, apart from the
> improvisations, mostly pop type music running towards comedy. How I
> compose: something pops into my head, usually a vocal or
instrumental
> line, and it grows longer and develops in my mind's ear until it is
a
> whole piece, at which point I write it down as a lyric sheet with
> chord changes and any instrumental riffs that are important. About
> half the time the whole idea does not develop, so I write down the
> fragment and put it in my files. Sometimes I can come back and look
at
> the fragment and finish it, sometimes not.
>
> John Starrett

John, perhaps you could fill us in more about how you do modulations,
particularly in 19-tET. Do you develop the ideas first in 12-tET (I
rather doubt that), or sketch 19-tET chord changes right off the
bat...

And what kind of notation do you use for 19-equal...
the "traditional" enharmonic kind??

You actually do *more* strictly-defined xenharmonic modulation that
anybody I know, with possibly the exception of Herman Miller...
(well, and Neil Haverstick, of course...)

________ ________ ______
Joseph Pehrson

Thanks, John, Brian, and Monz, for sharing.

As some of you know, I'm not actively composing in microtonal
tunings right now. The reason for this is simple -- I know how
I'd like to compose microtonally, and the tools I want don't
exist yet. So, I've chosen to use my time to create those tools
first. It shouldn't take more than 10 years to do this, but if
it does, so be it. In the meantime, I'm composing in good old
12-tET, for which there are excellent tools, and I'm learning to
sight-read classical music at the piano.

It became apparent to me early on that, unless you're a guitarist,
tools do not exist to make world-class microtonal music. So, the
first step was to learn enough theory to know what kind of tools
would do the job. That step is done. Quite contrary to what
Brian says, theory does a great job here. I also checked out the
two established ways to make microtonal music:

1. I built a 15-limit guitar. I'm not a guitarist, but this
instrument is very valuable to me for ear training. John
Starrett (hi, John!) also showed me that in a guitarist's hands,
the instrument was capable of making world-class music. I
almost gave it to him on the spot.

2. I did a lot of something I like to do anyway, and that's
sing. To date, Barbershop is still the only established musical
form that both claims, and delivers, anything higher than the
5-limit, as far as polyphonic, consonant music goes.

Then, I experimented with the following methods of composing
microtonal music:

--1. I took the manuscript-writing skills I learned with 12-tET,
the microtonal ear training I did with my guitar, and the knowledge
of meantone accidentals (non 'enharmonic equivalent') I got from
studying microtonal theory, and I wrote some music. I put the
music in my computer, and got out (12-tET) MIDI files.

I used JdL's adaptune software to retune some of them, but the
software doesn't pay attention to the accidentals (actually, my
score entry package doesn't even preserve them in the MIDI files
it makes), so those retunings aren't anything like I had in mind
(also, I usually compose in the 11-limit, while the version of
adaptune I had only went to the 7-limit). I took this experience
and wrote an algorithm to do adaptive tuning the way I think it
should be done. That's new tool number 1. I'm currently learning
C at work, and I imagine it'll be up and running within 2-3 years.

I also had a brass quartet (two trumets, french horn, and tuba)
perform one of the works at an AFMM concert in New York. Sadly,
it proved impossible to get the boys together enough to have them
learn 11-limit JI. But they got the 5-limit fairly well. Remember
everybody, nobody in New York has more than 5 minutes for anything,
no matter how much you pay them (just one of the many reasons
that city is high on my list of suggested nuclear test sites).

--2. I looked at various 12-tone scales, and created a few I liked.
I learned how to tune and regulate pianos, and retuned two pianos.
On the tuning list, I learned that one of the scales was used by
Wilson/Grady/Poole, and the other appeared (rotated in the lattice)
in a paper issued by Wilson in 1969. I recorded about an hour of
improvised music in the first, and put some mpeg excerpts on mp3.com.

I then bought a Kawai K5000s keyboard (for $450 at Sam Ash, I got
one of the last new K5000's sold in the states). This is one of
the best keyboards ever brought to market. It can only do quarter
tones, but I used Graham Breed's MIDIRelay to retune it from my PC.
Unlike Brian, I don't believe in using more than 12-tone scales on
the Halberstadt keyboard. I recorded a bunch of music with this
setup that I haven't released, and will never release. It's for me.

Back in 1997, I had a couple of righteous jams with excellent
musician Pat Pagano, on a Baldwin organ retuned to 7+ limit scale
by Denny Genovese (still don't know exactly what it was). A CD
of this material is available to anyone who asks nice, except for
Brian McLaren. The rest of it was stolen from my car in Oakland
in 1998, but not before several people had heard it.

...and those are the two methods I've used. Now let's talk about
new methods. We've already seen new method 1 (my adaptive tuning
algorithm, which was described in an article on the tuning list,
and on my web site).

New Method 2.

I want a general MIDI engine that takes note-ons from one MIDI
channel, and uses them to re-root a tuning on another channel(s).
Then, I'll use it with: (a) A two-manual organ with full pedal,
so that root-control may be assigned to either manual or the
pedal, or turned off, and may be sounding or silent. and (b) a
score entry package where a staff may be assigned to root control,
sounding or not...

New Method 3.

A keyboard with more than 12 tones per octave. I've read and
traveled extensively to learn about what I want... I've got every
extant paper by Bosanquet on microfilm/xerox, I've met with Erv
Wilson, Michael Zarkey (and played his 19-of-31 harpsichord), and
Scott Hackleman (and played his 19-tET clavichord), Harvey Starr
(and fingered his prototype MicroZone), and Paul Vandervoort
(and played his Janko-keyboard piano). I even lived with Norman
Henry for three months, playing his 11-limit harpsichord and
helping him develop his 15-limit fortepiano. My conclusion is
that Wilson's hexagonal keys are not desirable. Vandervoort has
the best bet, and his MIDI controller (with polyphonic aftertouch,
optical action, for about $2000!) will be along shortly.

--
Both these new methods use a MIDI synth. I've learned from contacts
on the tuning list (special thanks to John Loffink) that only three
synths are really up to the job... 1. Kurzweil K2500-2600 series,
2. Emu Proteus 2000 family, and 3. Kyma. These also make up some of
the best sounding synths on the market (also notable for good sound
is the Korg Triton series). Of these, the Kurzweil has the root
choosing engine already built in (thanks to Wendy Carlos), but the
price/perform. isn't that good, seeing as how Kyma costs less and does
way more (you just have to write the stuff yourself, in SmallTalk).
The Proteus can be had for $850 or so, and I may buy one this week.
Kyma is the best choice for everything, but it is expensive, and you
have to make everything yourself -- it's not so much an out-of-the-box
musical instrument as it is a sound computation engine, though Marcus
Hobbs has a library of tools he's willing to share, and Breed will
no doubt be out with some stuff too.

That's it! Thanks for listening.

-Carl

Hi Carl,

> I used JdL's adaptune software to retune some of them, but the
> software doesn't pay attention to the accidentals (actually, my
> score entry package doesn't even preserve them in the MIDI files
> it makes), so those retunings aren't anything like I had in mind
> (also, I usually compose in the 11-limit, while the version of
> adaptune I had only went to the 7-limit). I took this experience
> and wrote an algorithm to do adaptive tuning the way I think it
> should be done. That's new tool number 1. I'm currently learning
> C at work, and I imagine it'll be up and running within 2-3 years.

If you are using windows c, then I have a c code snippets page
that might be of interest.

Just shows how to midi relay in C from Midi In, but may be useful
if you haven't done that before. Assumes one already is familiar
with the basic ideas of Windows C as in Petzold's book. Doesn't use
C++; I'm a pure C type programmer by inclination and what I've learnt.

http://www.robertwalker.f9.co.uk/code_snippets.htm

(there are other ways of doing it, and if one wants to send
the desired note times along with the midi messages, instead
of just time stamping them with the current time as they are
played, one will need to re-write the code - that would be desirable
if one wanted to quantise the midi out. I plan to do experiment with
that in fact some time and maybe do second version of the teeny
tuned sequencer, as preliminary to adding option to do same thing in FTS.)

I plan to add more to it in time.

> I want a general MIDI engine that takes note-ons from one MIDI
> channel, and uses them to re-root a tuning on another channel(s).
> Then, I'll use it with: (a) A two-manual organ with full pedal,
> so that root-control may be assigned to either manual or the
> pedal, or turned off, and may be sounding or silent. and (b) a
> score entry package where a staff may be assigned to root control,
> sounding or not...

Interesting!

Some of it sounds easy enough to do in software. One could
easily program to look at one of the midi channels and use it
just for the purpose of changing the tunings used in the others.
E.g. that the tuning is in just intonatino 12 tone (or whatever
scale has all the chords one wants in it), and then using the
root control channel, you change the point at which the scale
starts. E.g. play an A in the root control panel and it will
shift the scale so that the 1/1 is A. Might not necessarily be
the root of the chord you want I suppose, but one could get
used to which notes to press where for which chords.

Then, you have question of how to dovetail that to the notes
already played, and I suppose you soon then get into
comma shifts, or methods to correct pitch drift, but doing
it this way with your idea of a root control channel,
seems to me there mightn't be that much to do, just note how far you
have got from the original pitch of the note and if it is further
than a certain amount, let the pitch of the piece drift back home
at some fixed maximum drift rate (or maybe let the rate of
drift back depend on how fast the passages are, or on how far
away one has got).

If though the root control is for the chord, and prog. needs
to recognise that, then that I gather from John it is far harder
to do as not all notes have been played yet and it isn't easy for
a program to anticipate what notes are yet to be played in
a chord as it is played. That is, unless one were to put in a
delay there, so that the notes sound, say, a half second
after they are played, which one could prob. get used to
using as the delayed action would be a bit like a (non electically
controlled) acoustic organ.

Another idea, maybe one could maybe devise a kind of figured
base type idea where you play several notes at once in the
root control manual to tell the software the exact chord to
expect, with one of them as the default if you play
just a single note, and maybe do it so most common chords
only need two notes (e.g. play two notes a tone apart for
dom7th with root on upper note)... Might work if one is
using one hand for root control, and one for playing a solo
line, perhaps.

John and I have discussed on TL, I think while you were away,
idea of leisure adaptive tuning of a piece in his program,
then another program, such as maybe FTS has job of going
through a second playing of the piece, the actual performance,
and retuning it according to the information in the leisure
adaptive tuning. So that is something else one can explore.

Issue there, especially for practicing using this method,
is how to quickly find ones place in the leisure time retuned
file if the performer skips to another part of the piece,
or stops and plays a section again. Bearing in mine here that
the chords will prob. not be played in the exact same order
of notes at microsecond level, which complicates matters
if one is trying to recognise them. We explored a few ideas,
enough to see that it might be a feasible thing to do.

> New Method 3.
>
> A keyboard with more than 12 tones per octave. I've read and
> traveled extensively to learn about what I want... I've got every
> extant paper by Bosanquet on microfilm/xerox, I've met with Erv
> Wilson, Michael Zarkey (and played his 19-of-31 harpsichord), and
> Scott Hackleman (and played his 19-tET clavichord), Harvey Starr
> (and fingered his prototype MicroZone), and Paul Vandervoort
> (and played his Janko-keyboard piano). I even lived with Norman
> Henry for three months, playing his 11-limit harpsichord and
> helping him develop his 15-limit fortepiano. My conclusion is
> that Wilson's hexagonal keys are not desirable. Vandervoort has
> the best bet, and his MIDI controller (with polyphonic aftertouch,
> optical action, for about $2000!) will be along shortly.

What about Margo's idea of playing two simultaneous ordinary
keyboards each with 12 notes? I'm hoping to try that idea out
in near future to use with scales of more than 12 notes per
octave, and add programming for it to FTS.

Robert

> Just shows how to midi relay in C from Midi In, but may be useful
> if you haven't done that before. Assumes one already is familiar
> with the basic ideas of Windows C as in Petzold's book. Doesn't use
> C++; I'm a pure C type programmer by inclination and what I've
> learnt.

Me too.

> http://www.robertwalker.f9.co.uk/code_snippets.htm

Thanks! I've put it in the queue (yes, I'm using windows c).

>> I want a general MIDI engine that takes note-ons from one MIDI
>> channel, and uses them to re-root a tuning on another channel(s).
>> Then, I'll use it with: (a) A two-manual organ with full pedal,
>> so that root-control may be assigned to either manual or the
>> pedal, or turned off, and may be sounding or silent. and (b) a
>> score entry package where a staff may be assigned to root control,
>> sounding or not...
>
> Interesting!

Another one of those ideas that many people have independently
discovered.

> Some of it sounds easy enough to do in software. One could
> easily program to look at one of the midi channels and use it
> just for the purpose of changing the tunings used in the others.
> E.g. that the tuning is in just intonatino 12 tone (or whatever
> scale has all the chords one wants in it), and then using the
> root control channel, you change the point at which the scale
> starts. E.g. play an A in the root control panel and it will
> shift the scale so that the 1/1 is A. Might not necessarily be
> the root of the chord you want I suppose, but one could get
> used to which notes to press where for which chords.

That's the general idea, yes. It's worth noting that I don't
only want to re-root to members of the current scale (I believe
this is the behavior the K2500/K2600's support, and was the
behavior Justonic's software was supposed to support), but also
to members of a different scale entirely. You can get a
tonality diamond by modulating a harmonic scale by a subharmonic
scale. Or, you can use meantone roots for a JI scale, and
get around many comma problems (a suggestion due to Paul Erlich).

> Then, you have question of how to dovetail that to the notes
> already played, and I suppose you soon then get into
> comma shifts, or methods to correct pitch drift, but doing
> it this way with your idea of a root control channel,
> seems to me there mightn't be that much to do, just note how far
> you have got from the original pitch of the note and if it is
> further than a certain amount, let the pitch of the piece drift
> back home at some fixed maximum drift rate (or maybe let the
> rate of drift back depend on how fast the passages are, or on
> how far away one has got).

I don't see how drift could be an issue, since the scales are
rooted to one master scale. Comma shifts will occur, but will
be a feature, and not a bug, of music composed for the setup.
For playing existing music where the syntonic comma should
vanish, Paul Erlich's suggestion (above), should work fairly
well in most cases.

> If though the root control is for the chord, and prog. needs
> to recognise that, then that I gather from John it is far harder
> to do as not all notes have been played yet and it isn't easy for
> a program to anticipate what notes are yet to be played in
> a chord as it is played. That is, unless one were to put in a
> delay there, so that the notes sound, say, a half second
> after they are played, which one could prob. get used to
> using as the delayed action would be a bit like a (non electically
> controlled) acoustic organ.

Not quite with you. The root control is for the every note not
on the control channel. I don't envision any 'smarts' at all.
As a musician, the idea of alternate tunings is one of: more
stuff I can control. Software that does stuff for me may be good
for making the kind of music we're already hearing _sound_ better,
and this is a good idea, but I also want new music that isn't
possible in meantone!

>> New Method 3.
>>
>> A keyboard with more than 12 tones per octave. I've read and
>> traveled extensively to learn about what I want... I've got every
>> extant paper by Bosanquet on microfilm/xerox, I've met with Erv
>> Wilson, Michael Zarkey (and played his 19-of-31 harpsichord), and
>> Scott Hackleman (and played his 19-tET clavichord), Harvey Starr
>> (and fingered his prototype MicroZone), and Paul Vandervoort
>> (and played his Janko-keyboard piano). I even lived with Norman
>> Henry for three months, playing his 11-limit harpsichord and
>> helping him develop his 15-limit fortepiano. My conclusion is
>> that Wilson's hexagonal keys are not desirable. Vandervoort has
>> the best bet, and his MIDI controller (with polyphonic aftertouch,
>> optical action, for about $2000!) will be along shortly.
>
> What about Margo's idea of playing two simultaneous ordinary
> keyboards each with 12 notes? I'm hoping to try that idea out
> in near future to use with scales of more than 12 notes per
> octave, and add programming for it to FTS.

Go for it!

In stark contrast to this encouragement, I will repost my initial
reaction to this scheme, from September 6th, 2000:

>... completely limiting. Organs and harpsichords have had multiple
>manuals for hundreds of years, yet I know of no serious microtonal
>work ever done in this manner. Can you play chords across manuals
>1-handed, for instance? What if I split your favorite piano
>keyboard into seperate 7- and 5-tone manuals. How would you like
>that?

:)

-Carl

Hi Carl,

> Just shows how to midi relay in C from Midi In, but may be useful
> if you haven't done that before. Assumes one already is familiar
> with the basic ideas of Windows C as in Petzold's book. Doesn't use
> C++; I'm a pure C type programmer by inclination and what I've
> learnt.

Me too.

> http://www.robertwalker.f9.co.uk/code_snippets.htm

Thanks! I've put it in the queue (yes, I'm using windows c).

>> I want a general MIDI engine that takes note-ons from one MIDI
>> channel, and uses them to re-root a tuning on another channel(s).
>> Then, I'll use it with: (a) A two-manual organ with full pedal,
>> so that root-control may be assigned to either manual or the
>> pedal, or turned off, and may be sounding or silent. and (b) a
>> score entry package where a staff may be assigned to root control,
>> sounding or not...
>
> Interesting!

> > Another one of those ideas that many people have independently
> > discovered.

Well, that's true of many good ideas. When the time is ripe you find maybe three
people have come up with it independently or something, maybe more.

I thought I'd invented the idea of a fractal tune when I first started
on FTS actually, then discovered this whole active field of algorithimic
composition.

> > Some of it sounds easy enough to do in software. One could
> > easily program to look at one of the midi channels and use it
> > just for the purpose of changing the tunings used in the others.
> > E.g. that the tuning is in just intonatino 12 tone (or whatever
> > scale has all the chords one wants in it), and then using the
> > root control channel, you change the point at which the scale
> > starts. E.g. play an A in the root control panel and it will
> > shift the scale so that the 1/1 is A. Might not necessarily be
> > the root of the chord you want I suppose, but one could get
> > used to which notes to press where for which chords.

> That's the general idea, yes. It's worth noting that I don't
> only want to re-root to members of the current scale (I believe
> this is the behavior the K2500/K2600's support, and was the
> behavior Justonic's software was supposed to support), but also
> to members of a different scale entirely. You can get a
> tonality diamond by modulating a harmonic scale by a subharmonic
> scale. Or, you can use meantone roots for a JI scale, and
> get around many comma problems (a suggestion due to Paul Erlich).

Right. This is far easier than my first thoughts about what
you were sayiang. Just a matter of assigning a scale / mode
to one of the channels as the root channel, working out the
position relative to 1/1 in usual way, then multiply all the
notes played by the resulting value. That would do the
tonality diamond just as it is - play the control note
and note on the main keyboard, and you'll hear the product.

Then for e.g. mean-tone modulation of j.i. scale, I suppose
one will want e.g. the A of the main keyboard to still sound
an A whatever the root of the j.i. scale is, e.g. if root
key pressed is the F, you play the A as a 5/4 above the meantone F.
That would require all of another couple of lines of code or something!

The whole thing would prob. only need maybe two or three dozen lines of
programming if I were to add it to FTS, what with the way everything is set up
already for this type of thing. Most of the work involved in fact
would be thinking where to put it in the program, the user interface,
and how to relate it to everything else. I could do it at same
time as I do the generalisation of Graham's scale morphing idea
to arbitrary modes / scales, as both will require one to have a
second scale or mode to hand. FTS actually already has a multiple scales
in play at once option for the fractal tunes, and I could adapt that
in some way, or may decide it is clearer to have a new section.

Does the idea of doing it in FTS appeal? If so I'll add it to
my to do list for FTS. Could mean you can get to play around with
it a bit before you reach point of programming it yourself.

> Then, you have question of how to dovetail that to the notes
> already played, and I suppose you soon then get into
> comma shifts, or methods to correct pitch drift, but doing
> it this way with your idea of a root control channel,
> seems to me there mightn't be that much to do, just note how far
> you have got from the original pitch of the note and if it is
> further than a certain amount, let the pitch of the piece drift
> back home at some fixed maximum drift rate (or maybe let the
> rate of drift back depend on how fast the passages are, or on
> how far away one has got).

> I don't see how drift could be an issue, since the scales are
> rooted to one master scale. Comma shifts will occur, but will
> be a feature, and not a bug, of music composed for the setup.
> For playing existing music where the syntonic comma should
> vanish, Paul Erlich's suggestion (above), should work fairly
> well in most cases.

Yes, I was still thinking in terms of a kind of adaptive tuning of the
type John does, where one needs to recognise the chord, and retune
it's notes according to the notes in play. But with your idea, seems
commas won't be an issue, and as you say, will be a feature when they
occur, rather than a bug.

> If though the root control is for the chord, and prog. needs
> to recognise that, then that I gather from John it is far harder
> to do as not all notes have been played yet and it isn't easy for
> a program to anticipate what notes are yet to be played in
> a chord as it is played. That is, unless one were to put in a
> delay there, so that the notes sound, say, a half second
> after they are played, which one could prob. get used to
> using as the delayed action would be a bit like a (non electically
> controlled) acoustic organ.

> Not quite with you. The root control is for the every note not
> on the control channel. I don't envision any 'smarts' at all.
> As a musician, the idea of alternate tunings is one of: more
> stuff I can control. Software that does stuff for me may be good
> for making the kind of music we're already hearing _sound_ better,
> and this is a good idea, but I also want new music that isn't
> possible in meantone!

Yes, sounds an interesting line to follow. See whre it goes.

>> New Method 3.
>>
>> A keyboard with more than 12 tones per octave. I've read and
>> traveled extensively to learn about what I want... I've got every
>> extant paper by Bosanquet on microfilm/xerox, I've met with Erv
>> Wilson, Michael Zarkey (and played his 19-of-31 harpsichord), and
>> Scott Hackleman (and played his 19-tET clavichord), Harvey Starr
>> (and fingered his prototype MicroZone), and Paul Vandervoort
>> (and played his Janko-keyboard piano). I even lived with Norman
>> Henry for three months, playing his 11-limit harpsichord and
>> helping him develop his 15-limit fortepiano. My conclusion is
>> that Wilson's hexagonal keys are not desirable. Vandervoort has
>> the best bet, and his MIDI controller (with polyphonic aftertouch,
>> optical action, for about $2000!) will be along shortly.
>
> > What about Margo's idea of playing two simultaneous ordinary
> > keyboards each with 12 notes? I'm hoping to try that idea out
> > in near future to use with scales of more than 12 notes per
> > octave, and add programming for it to FTS.

> Go for it!

Okay, thanks for the encouragement!

> In stark contrast to this encouragement, I will repost my initial
> reaction to this scheme, from September 6th, 2000:

> >... completely limiting. Organs and harpsichords have had multiple
> >manuals for hundreds of years, yet I know of no serious microtonal
> >work ever done in this manner. Can you play chords across manuals
> >1-handed, for instance? What if I split your favorite piano
> >keyboard into seperate 7- and 5-tone manuals. How would you like
> >that?

> :)

Yes, I've thought about that too, it may work best if the manuals are
very shallow; sometimes you see ones that are. Would be nice if one
could play chords one handed. Only the top manual needs to be shallow
of course.

Robert

> Well, that's true of many good ideas. When the time is ripe you
> find maybe three people have come up with it independently or
> something, maybe more.

Def.

> I thought I'd invented the idea of a fractal tune when I first
> started on FTS actually, then discovered this whole active field
> of algorithimic composition.

Wow!

> Right. This is far easier than my first thoughts about what
> you were sayiang. Just a matter of assigning a scale / mode
> to one of the channels as the root channel, working out the
> position relative to 1/1 in usual way, then multiply all the
> notes played by the resulting value. That would do the
> tonality diamond just as it is - play the control note
> and note on the main keyboard, and you'll hear the product.

Right.

> Then for e.g. mean-tone modulation of j.i. scale, I suppose
> one will want e.g. the A of the main keyboard to still sound
> an A whatever the root of the j.i. scale is, e.g. if root
> key pressed is the F, you play the A as a 5/4 above the meantone F.
> That would require all of another couple of lines of code or
> something!

I don't think so -- it's still just mulitplication.

> The whole thing would prob. only need maybe two or three dozen
> lines of programming if I were to add it to FTS, what with the
> way everything is set up already for this type of thing. Most of
> the work involved in fact would be thinking where to put it in
> the program, the user interface, and how to relate it to
> everything else. I could do it at same time as I do the
> generalisation of Graham's scale morphing idea to arbitrary
> modes / scales, as both will require one to have a second scale
> or mode to hand. FTS actually already has a multiple scales
> in play at once option for the fractal tunes, and I could adapt
> section.

Boss!

> Does the idea of doing it in FTS appeal? If so I'll add it to
> my to do list for FTS. Could mean you can get to play around with
> it a bit before you reach point of programming it yourself.

There's more than too many good ideas out there to code...
anyone beats me to it, and I'll send him flowers.

I realized that it may be time to write up a precise spec
of what I have in mind. Would you like me to send you a
copy (1-2 weeks)?

> Yes, I've thought about that too, it may work best if the manuals
> are very shallow; sometimes you see ones that are. Would be nice
> if one could play chords one handed. Only the top manual needs to
> be shallow of course.

Carry this one step further, and you have the stuff of a
generalized keyboard.

-Carl

> John, perhaps you could fill us in more about how you do
> modulations,
> particularly in 19-tET. Do you develop the ideas first in 12-tET (I
> rather doubt that), or sketch 19-tET chord changes right off the
> bat...

Most of my music is basically diatonic, so for most of my pieces my
mind's ear hears diatonically. I have been playing in 19 long enough
so that I hear 19 as I compose. Some of my music sticks to the
diatonic 19 until I mess with modulations after writing it down, and
some I hear with the unique 19 modulations built in. I will use a
couple of examples that are available on mp3.com at the Tuning Punks
page.

Limp Off to School has two modulations based on the progression 6 5 1
4 (where I am using the convention that arabic numerals denote the
usual triads built on that diatonic major scale degrees), but using
the small minor third available in 19 to move to the 6 so that the 4
is now one 19tet chromatic step higher than it would have been with
the usual large minor third down. When I composed it I had originally
wanted to do that trick 3 times (the section contains 3 of the 6 5 1 4
progressions) but I found when it came time to write it down that I
couldn't gracefully get back in the key and had to use two small minor
thirds and one large.

Citified Notions on the other hand was completely diatonic at
conception and didn't get it's funny modulations until I was
rehearsing it.

> And what kind of notation do you use for 19-equal...
> the "traditional" enharmonic kind??

Traditional... A A# Bb B B#/Cb C C# Db D D# Eb E E#/Fb F F# Gb G G# Ab

> You actually do *more* strictly-defined xenharmonic modulation that
> anybody I know, with possibly the exception of Herman Miller...
> (well, and Neil Haverstick, of course...)
>
> ________ ________ ______
> Joseph Pehrson

John Starrett

--- In crazy_music@y..., "John Starrett" <jstarret@c...> wrote:

/crazy_music/topicId_726.html#926

> > John, perhaps you could fill us in more about how you do
> > modulations,
> > particularly in 19-tET. Do you develop the ideas first in 12-tET
(I
> > rather doubt that), or sketch 19-tET chord changes right off the
> > bat...
>
> Most of my music is basically diatonic, so for most of my pieces my
> mind's ear hears diatonically. I have been playing in 19 long
enough
> so that I hear 19 as I compose. Some of my music sticks to the
> diatonic 19 until I mess with modulations after writing it down,
and
> some I hear with the unique 19 modulations built in. I will use a
> couple of examples that are available on mp3.com at the Tuning
Punks
> page.
>
> Limp Off to School has two modulations based on the progression 6 5
1
> 4 (where I am using the convention that arabic numerals denote the
> usual triads built on that diatonic major scale degrees), but using
> the small minor third available in 19 to move to the 6 so that the
4
> is now one 19tet chromatic step higher than it would have been with
> the usual large minor third down. When I composed it I had
originally
> wanted to do that trick 3 times (the section contains 3 of the 6 5
1 4
> progressions) but I found when it came time to write it down that I
> couldn't gracefully get back in the key and had to use two small
minor
> thirds and one large.
>
> Citified Notions on the other hand was completely diatonic at
> conception and didn't get it's funny modulations until I was
> rehearsing it.
>
> > And what kind of notation do you use for 19-equal...
> > the "traditional" enharmonic kind??
>
> Traditional... A A# Bb B B#/Cb C C# Db D D# Eb E E#/Fb F F# Gb G G#
Ab
>
> > You actually do *more* strictly-defined xenharmonic modulation
that
> > anybody I know, with possibly the exception of Herman Miller...
> > (well, and Neil Haverstick, of course...)
> >
> > ________ ________ ______
> > Joseph Pehrson
>
> John Starrett

Thanks so much, John, for describing this process. You know, it
seems that in "Limp Off to School" there are quite extended
modulations.... Do you feel that the increased modulatory
possibilities in 19-tET led you to thinking in that direction?? It
would seem to me that it might. I really dig those modulations...

__________ _________ _______
Joseph Pehrson

> Thanks so much, John, for describing this process. You know, it
> seems that in "Limp Off to School" there are quite extended
> modulations.... Do you feel that the increased modulatory
> possibilities in 19-tET led you to thinking in that direction?? It
> would seem to me that it might. I really dig those modulations...
>
> __________ _________ _______
> Joseph Pehrson

Yes, the modulation from the +[6 5 1 4] (where +[ denotes the usual
diatonic progression but started one chromatic step higher) done twice
brings it up 2 chromatic steps, then a 4 -[5 4 -[5]] brings it down 2
chromatic steps. The progression for the whole piece is written below,
and the notation is as above with the additional conventions that all
chords are major or minor depending on their root unless otherwise
noted, a chord with a "d" following it denotes a dominant of the new
key (and thus major), and the new key is indicated by */* where the
first * is the root note of the chord in the old key, and the second *
is the same chord related to the new key. |: :| is a repeat. "b"
denotes a diatonically flatted scale degree, as in b7 for the minor
seventh degree from the root rather than the major seventh.

|1 |4 |5 |5 | Intro 3 chromatic steps below target
|1 |4 |5 |6d | key. Now on to target key, which is a
perfect fourth up from the 6d by virtue
of the "d"
|:1 |5 4 |5 |5 :| Ordinary diatonic
+6 |5 |1 |4 | New key one chromatic step higher
|2 |3d |6/1 |1 | New key tonic on 6th degree of old
|+6 |5 |1 |1 | New key one chromatic step higher
Now we are 3 chromatic steps below the original key.. a 19tet whole
step. This is where the line "No gym class no recreation..." starts.
Note the flatted 6th degree of the old key now functions as the fourth
degree of the new key by the notation -6/4
|-6/4 |4 |-5 |5 |
|4 |4 -5|5 |
Now we are back at the 5th degree of the target key. Here comes the
chorus, "Limp off, limp off to school"
|b7/1 |4 5 |5 |
|1 |4 5 |5 |6d |
|2/1..... etc.

Yes, there are all kinds of cool tricks you can use in 19tet
modulation. The major second is three chromatic steps and the minor
second is two chromatic steps, but since there are pitches in
between these can be used to fool the ear when they are substituted
for the one expected.The two sizes of minor third can be used to great
effect by using a step wise walk down or up, for instance, and every
once in a while sneaking in one of the chromatic steps. I have
uploaded two samples of guitar moving between a functional 4 and 5
chord. See if you can tell what I am doing here, and what the
difference between the two is.

John Starrett

--- In crazy_music@y..., "John Starrett" <jstarret@c...> wrote:

/crazy_music/topicId_726.html#957

> Yes, there are all kinds of cool tricks you can use in 19tet
> modulation. The major second is three chromatic steps and the minor
> second is two chromatic steps, but since there are pitches in
> between these can be used to fool the ear when they are substituted
> for the one expected.The two sizes of minor third can be used to
great
> effect by using a step wise walk down or up, for instance, and
every
> once in a while sneaking in one of the chromatic steps. I have
> uploaded two samples of guitar moving between a functional 4 and 5
> chord. See if you can tell what I am doing here, and what the
> difference between the two is.
>
> John Starrett

Well, I'll have to study this is more detail, but right "off the top"
it sounds, in the second example, like you are substituting
a "sharper" chromatic step between the 4 and 5 for the normal one on
the ascent... (??)

_______ ________ _________
Joseph Pehrson

I've really been enjoying this thread, it's great to see other composers'
perspectives on this highly elusive process of creating music. Here's my two
cents for a 3/2 on the subject.

Although I can compose with paper, pencil and eraser, and have used this
process for some film scoring, the guitar is a highly idiomatic instrument
and I prefer to compose hands on the instrument. Often I start with an open
tuning, and improvise. If I come up with an idea I like in the improv, I
keep coming back to it to develop it further. Sometimes the process takes
months or even years, having little incomplete bits hanging around, waiting
for the right perspective for completion. Every now and then I get lucky and
a piece just "shows up", fully formed and ready to go.

For a long time, I've liked the idea of composing using instruments I didn't
play, so as to not be boxed in by my abilities on guitar. I'd compose a
melody on an instrument other than guitar, then transfer it to guitar. A
fringe benefit of this approach is that I've ended up learning to play many
of these instruments with proficiency.

I tend to be lazy about notation, I find it extremely arduous to try and
notate exactly things that I play slightly differently every time. Using
Finale doesn't make things any easier. I usually record my pieces so I don't
forget them. Of course, recording can be pretty stressful as well. These
days I'm less interested in touring around the country, sleeping on floors,
playing the same dozen pieces over and over again so as to sell a few CDs,
so I'm trying to get more disciplined about writing down my music so other
performers can play it, and send me those fabulous royalty checks :-)

As far as composing microtonal pieces go, it has been a slow process. I've
used intentional deviances from 12-tET in my slide guitar and fiddle playing
for years, without saying that a particular note was a five or seven limit
tone, just playing what felt right. Some of my current learning of JI has
been a way to further describe and hone what I was doing naturally. Like
planting a seed, watering it, and letting the seasons and elements do their
part, I am taking this approach with microtonality, over time it is becoming
a natural part of my musical language. I've been aquiring various
instruments, such as fretless guitar and banjo, didjs, overtone flutes, and
previously to that, 31 and 19 tET guitars, and learning to play and sing in
JI accurately. I've done alot of improvising, and a fair amount of reading
and studying. I know that at some point in time, in the not too distant
future, I will better internalize this language, and more music will
happen/be composed.

Seth

--
Seth Austen

http://www.sethausten.com
emails: seth@...
klezmusic@...