Re: chord labeling

[Juergen Reuter]

On Tue, 24 Feb 2004, Edward Sanford Sutton, III wrote:

>   I know a chord can be labeled as Cm, but is there a way to label them for 
> harmonic analysis? I have only figured out how to get so far with stuff like:
> 
> <c e g>_\markup "i"
> 
> The last time I attempted such markings was on a cadential I 6/4 chord. I 
> found that using \column in a markup statement was about as far as I was 
> getting (which would still need work).

AFAIK (if I am wrong, please tell me!), functional harmonic analysis 
always assumes the existence of a tonal basis as reference point.  For 
example, in c major, <c e g> is annotated as "I", <f a c> is "IV" and <g b 
d f> is "V7", since c, f and g are the 1st, 4th and 5th in c major.  Some 
people also write "T" (tonic), "S" (sub dominant), "D" (dominant) for "I", 
IV", "V", respectively; thus <g b d f> is "D7" in c major.

Unfortunately, the key signature of a piece seldomly is useful as a tonal 
basis reference point, since (especially for music after 1800) the tonal 
basis changes quite often within a peace, but the key signature of the 
printed music does not follow these fine-grained tonal changes all the 
time.  Finding the tonal basis automatically may be feasible for pieces 
with very simple harmonics; but for such pieces, harmonic analysis is 
probably not very interesting to apply.  Figured bass also does not give 
any hints for finding the tonal basis, since it uses the bass note as 
reference point rather than the tonal basis.

Hence, I currently do not see any way for automating annotation of 
harmonic analysis unless you explicitly provide the tonal basis (maybe one 
could add kind of fine-grained key signature change feature to lily that 
does not affect the key signature engraver, but is used for harmonic 
analysis only).  So, I guess, using markup is currently the only sensible 
way for annotating harmonic analysis.

Greetings,
Juergen