> Hello! Harald Mueller speaking of 20-note arrangements, and Ben Jackson searching for a musical paper weight playing "Over The Waves" -- the first theme of this famous waltz must pose the worst-ever challenge to the arranger of lesser-keyed instruments!
I do not think so. In the MIDI file attachment, I have placed the first five minutes of work that I would do to arrange this waltz (I hope we are talking about the same theme). If you listen to the music (or read it), you will probably say, "Well, but he did it wrong! Exactly the 'interesting' semitones are missing, etc." Here are some explanations about arranging that explain my choices:
There are two problems with "limited-scale" organs like the 20er: one is that the spread of the melody range can be smaller than the range required by your piece of music; the other are the missing semitones. For both, there are solutions:
Reducing the spread:
Move things up or down an octave. If you can do this at a motif boundary, it's quite easy. If not (e.g. a sequence such as C1-E1-G1-C2-E2-G2-C3...), then you can use the fact that chords with many harmonics cannot be genuinely placed in a certain octave by the ear. Therefore, you can e.g. harmonize the chord sequence as follows:
- on a reed organ (which has many overtones), just add the standard triad chords; they will already blur the octave placement.
- with flutes, add jazz chords, mostly seconds in the middle range. This will add the required harmonics
- Replace the semitone with the next natural note. This is often acceptable in ornaments. Sometimes it helps to have a neighboring note play _as if_ the semitone was there. Then, the natural note is understood to be the semitone. A concrete example:
Original: trill G - F# - G
Arrangement: G+E - F+D#(!) - G+E (the D# would of course usually accompany a F#)
- Replace the semitone with some other note from the current harmonic chord.
- Move the whole piece to another key. Example: The first D# in Scott Joplin's Entertainer. This note alone requires that (on a B-flat-based 20er) to arrange the whole piece in the key of C-major instead of the much better supported B-flat.
How do you decide whether such a trick is acceptable? There are two rules:
(1) It must sound "correct". It is hard to define this, but in our western musical world there is whole range of acceptable, somewhat-acceptable, hardly-acceptable, and many not-acceptable combinations of chords and sequences. You have to "understand" these sequences.
Good training: Go to a concert of piano playing pupils and identify the errors, without knowing the pieces. Or, maybe easier, download many classical MIDI files from the Internet and find the errors by just listening. (You should _not_ know the pieces; the classical MIDIs on the Internet are on average horribly error-ridden.) If you can do this reliably (i.e., you can spot more than 50% of the errors, especially those in the tenor and alto inner voices, or the accompaniment and counter-melody), then you have this sort of "musical understanding" and "musical logic".
(2) You must know what the original composer wanted. If a trill is just an unimportant ornament, it is much more okay to replace it with a different ornament, or even leave it out, than if a note defines a theme or even a whole piece.
This is the reason why you must, in my opinion, always start arranging from the original score, not a piano arrangement or the like. Even better, transcribe the piece from a CD, because then you will (with a good conductor) learn much more about the _intent_ of the notes, chords, sequences and ornaments than by just transcribing the notes.
Essentially, when you arrange for mechanical instruments, you must be composer+arranger+conductor in one person -- then you will create acceptable results.
Finally, of course, a 20er organ is just a limited instrument after all, so some of your compromises will be horribly audible -- maybe you should then decide to not do the arrangement, after all.
(There are quite a few more rules and tricks of the trade. I'd really like, one day, to do a "workshop for crank organ arrangers", maybe in Waldkirch or wherever they would accept my proposal...)
Harald M. Mueller
Grafing bei Muenchen, Germany
14 Nov 2002 13:53:34 +0100 (MET)