>From reading the list, it looks like I'm not the only person using harmonic axes. However, I haven't seen any mention of the full algebraic treatment I apply to them. I will outline some of my ideas here, and someone can tell me whether they are original and, if not, how they are notated.
First a column matrix needs to be defined to correspond to harmonic axes:
oct (log(2)) H ----- (log(3)) log(2) (log(5))
H can be generalised to have as many dimensions as the matrix you are multiplying it by. Using this, a fifth is (-1 1)H, a major third (-2 0 1)H and a subminor seventh (-2 0 0 1)H.
H is useful for defining harmonies. For melodies, though, it is more convenient to use the smallest intervals that occur in a scale. For a diatonic scale with comma shifts, these will be:
(t) ( 1 -2 1) (s) ( 4 -1 -1) H (p) (-4 4 -1)
Where t is a minor tone, s a diatonic semitone and p a syntonic comma. These intervals completely define a 3-D JI, and so the matrix can be inverted:
( 5 2 3) (t) H ( 8 3 5) (s) (12 4 7) (p)
This will be true of any three intervals whose defining matrix has a determinant of +/- 1.
You can add commas for as many harmonic dimensions as you like. Equally tempered scales can then be defined according to the number of steps in these intervals:
This is the clearest representation I know of. From the defining matrix, nTET 12*r5 + 7*q5 + 3*p5.
That'll do for the time being.
Graham
Received: from ns.ezh.nl [137.174.112.59] by vbv40.ezh.nl with SMTP-OpenVMS via TCP/IP; Sun, 16 Mar 1997 19:52 +0100 Received: by ns.ezh.nl; (5.65v3.2/1.3/10May95) id AA30779; Sun, 16 Mar 1997 19:52:24 +0100 Received: from ella.mills.edu by ns (smtpxd); id XA30778 Received: from by ella.mills.edu via SMTP (940816.SGI.8.6.9/930416.SGI) id KAA27175; Sun, 16 Mar 1997 10:50:07 -0800 Date: Sun, 16 Mar 1997 10:50:07 -0800 Message-Id: Errors-To: madole@mills.edu Reply-To: tuning@ella.mills.edu Originator: tuning@eartha.mills.edu Sender: tuning@ella.mills.edu