On Mon, 19 Nov 2001, Simon Wascher wrote:
> why if the beat changes with the meter, the meter (M:) isnt the field
> which defines by its content (I do *not* mean to add an extention to
> it) what Allegro (~120 beats per minute) means.
This has already been discussed. It is possible to take a reasonable
guess as to what beat is intended for a given meter. However, due to a
shortcoming in traditional notation, that guess will not always be
correct. Whether or not it's correct "often enough" is still something
that might be worth discussing.
In the traditional way of notating meter, the top number defines the
number of beats in the measure, and the bottom number gives the value of
the note that defines the beat (e.g., 4/4 means four beats per measure,
and the quarter note defines the beat; 3/2 means three beats per measure,
with half-note beats).
However, if the note value of the beat is compound (dotted), there is no
whole number that can represent it accurately, so a smaller note value has
to be used. For instance, if the composer intended two beats per measure,
with a dotted-quarter beat, the only accurate way to write it would be
something like 2/2.6666666... For obvious reasons, we write 6/8 instead.
The problem is that by doing this, traditional notation breaks its own
rules. 6/8 time seems to imply that there are actually six beats per
measure, and that the eighth note defines the beat. And to complicate the
matter further, sometimes this *is* what the composer intended.
Metronome markings are one way to resolve this ambiguity in traditional
notation -- the very fact that they are written as {note value}={beats per
minute} instead of just {beats per minute} illustrates this. In the
absence of a metronome marking, though, we sometimes have to guess at the
beat. It's not usually too hard for a human to do this, but it's quite
another thing to systematize it. So, as far as abc is concerned, either
we allow human transcribers to use their own natural intelligence to
define the beat, or we accept that whatever artificial intelligence the
standard provides (if any) for determining the beat will be wrong
sometimes.
The question at this point seems to be whether we want the standard to get
it right all of the time (define Allegro as 120, define the beat for each
piece); get it wrong only sometimes (define Allegro as 120, guess at the
beat); or get it wrong a great deal of the time (define Allegro as
1/4=120, even in 6/8, damn the beat).
And of course, there is still the option of not allowing Allegro to be
defined in abc as anything at all, leaving it to the developers to make
sense of Q:Allegro for playback.
> I belive it is not really neccessary to define the beat of allegro in
> Balkan music (like 3+3+2), I've never heard of such a definition in
> any other music notation context.
This kind of thing happens quite often in newer compositions, usually
implied by the note beaming, but sometimes written out explicitly in the
meter (e.g., 3+2 over 8 instead of 5 over 8). Some notation software
allows this... I believe Finale is one of them.
John
To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
