John Walsh writes:
| John Chambers writes:
| >As you may recall, the approach that I used for my Klez/Balkan tunes
| >is the rather trivial extension of K: to allow:
| > K:<tonic><mode><accidentals>
...
|
| But John, I thought this was so obviously desirable that it needed
| no further comment.
Well, maybe I should write up a description and see if there's any
reaction this time. And if I casually ignore the existence of clefs
and other orthogonal concepts, maybe the discussion won't wander off
into discussions of how transposition should work. ;-)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
The K: header line specifies a key signature. At least one K: line
must be present in each abc tune, and more may be used to change the
key signature. The first K: line marks the end of the abc headers and
the start of the music section. The format of a key signature is:
K:<tonic><mode><accidentals>
All of the fields are optional. If <mode> is present, the <tonic>
must also be present, since a mode by itself isn't sufficient to
specify a key signature. If all the fields are omitted, there is no
key signature. This is similar to (but subtly different from) K:C.
The <tonic> is a capital letter which may be followed by a # or b for
sharp or flat respectively. The case isn't significant.
The <mode> may be any of the classical modes, and may be abbreviated
to the first three letters. The letter 'm' alone may be used for
minor. If only a <tonic> is given, a <mode> of major is assumed. The
mode names are:
lyd lydian
ion ionian (same as major)
mix mixolydian
dor dorian
ael aelolian (same as minor)
phr phrygian
loc locrian
Case isn't significant in these mode names. There are two additional
keys specifically for notating highland bagpipe tunes:
K:Hp % key signature of ^f ^c =g
K:HP % no key signature at all
Both cause note stems to drawn downwards, while stems on ornament
notes go upwards.
You may also use an explicit list of accidentals for a key signature.
The tonic note isn't required in this case, though you should include
it if you know it. If you give tonic, mode and accidentals, the
accidentals are added to the key signature of the tonic+mode. Here
are some examples:
K:^f % same as K:G and K:Gmajor, but no tonic specified
K:C^f % same as K:CLydian
K:A^f % same as K:ADorian
K:E^g % any of several Balkan or Middle-Eastern scales
K:_B_e % same as K:Bb or K:Gm or K:Cdor or K:DPhrygian
K:^f_B_e % a Middle-Eastern scale, D freygish/hejaz or C misheberach
K:Dphr^F % D freygish/hejaz with ^F rather than ^f
K:Cdor^F % C misheberach with ^F rather than ^f
K:Amix=g % A mixolydian with explitic g natural in the signature
K:^f^c=g % the same, but as explicit list of accidentals.
K:D=C^c % an Indian scale with different 7ths in the two octaves
K:D_B_e=C^F^c % D zengule (the Turkish scale)
Note that the case of accidentals is significant, since there are
musical styles like the last two examples that use different
intonations on some notes in different octaves.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
That might be all we need to say for now.
One question is about spaces. I'd prefer to say that the fields of a
K: line may be separated by spaces, for readability. The only problem
with this is the abc 1.6 description of "global accidentals", which
use the same syntax but with the accidentals separated from the
<tonic> and <mode>. However, I've been unable to find any abc that
uses global accidentals, or any software that implements it. So
perhaps we should decree "global accidentals" no longer part of abc's
syntax, and permit spaces between the K: fields.
The primary motivation for including explicit accidentals in the K:
line is to allow abc to represent non-Western scales. A secondary use
is in transcriptions for which the tonic is difficult to determine,
due to either the lack of musical knowledge of the transcriber, or to
the non-tonal nature of the music. Use of the tonic and mode should
be encouraged when they are known, since this information is useful
in computerized searches of music.
There's gotta be a typo up there somewhere ...
To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
