Re: Roman numeral chord notation in lilypond

[Miguel Lopes Santos Ramos]

> From: "Rick Hansen (aka RickH)" <address@hidden>
>
> Patrick Taylor Ramsey-2 wrote:
> > 
> > Hello, all.
> > 
> > I'm a bit flummoxed; I've been searching for the last hour and a half for
> > a way
> > to typeset roman numeral chord notation (ii, V65/V, N6) under a staff in
> > lilypond.... and found nothing, except for a lone message dating from 2003
> > stating that said notation wasn't yet supported.  Is this still the case? 
> > And,
> > if so, why?  It does seem like there are far more arcane notations which
> > *are*
> > supported... and this one is used by every beginning music theory course
> > in the
> > us (and most likely elsewhere)

Of course, analysing chords in terms of their relation to the main key is
fundamental to tonal harmony, but there are some many different notations...

The particular version you use it's unknown to me... You use the 6 5 and 6
which come from the figured bass, lower case for minor, and I can't guess the
N. Around here we've thrown away the figured bass and use a, b, c, d for
inversions, slashes forward and backward for diminished or augmented,
zeros for missing fundamentals, there are different oppinions regarding
missing fundamentals...
The most standard perhaps is the one from Schoenberg's books, which suffers
from figured bass reminiscence and has no graphical distinction between
minor/major/diminished/etc...

Then there's graphical problems... lots of glytches...
If you just want to print roman numerals, which is the only thing that's
standard, then...

> You could enter it as a lyrics staff.  The big issue with roman is that when
> a piece changes key for example what used to be a "III" chord is really not
> a "III" chord anymore, unless you have some way to depict to the reader to
> increment/decrement the key center by some other roman numeric value. 
> Because the roman numerals are based on "current" key center.

Of course that would be interesting...
All it would take would be to have one variable, the current key, which
needn't be a roman numeral...
Perhaps two variables, the current key, or tonal region, and the main key,
so you could also print the relation of the current key to the main key
(sd, S/T, Np, as Schoenberg does in Structural function of harmony).

The analysis part is easy, and it could be independent of notation, or
the notation configurable. What seems hardest to me is the graphical part,
the superimposition of symbols to the roman numerals, the little numbers, etc.

All it would take would be someone with experience on Lilypond, which I don't
have. I would be able to do the analysis part in Common LISP though...

Miguel
Lisboa