SongTrellis
Music and Musical Know-how For You

Author:  David Luebbert Posted:  9/24/06; 2:07:28 PM Topic:  II V7 I Practice in F Msg#:  4965 (in response to 4961) Prev/Next:  4964/4966 Reads:  780

Bernie,

This is a useful comment for me. In that exercise, Steve Breitbach is responsible for choosing which kinds of chords should be played and for how long. The actual voicings were calculated by a copy of the Macintosh SongTrellis Editor which handles requests to edit or play workscores. That software uses a set of rules that are applied as it calculates a succession of chord voicings. So I'm actually responsible for the voicings that it produces, since I selected the set of rules used and decided how they would be invoked and in what situations.

If you can tell me about arranging ideas, and tell me when they are appropriate to use and how often they may be used without wearing out their welcome, and what kinds of ideas are related to one another, I can perhaps boil those recommendations down to rules that I can use in a new version of the Editor.

The general idea you are presenting here is to locate linear ideas that can be used to organize the topmost voice in adjacent chords, so that the top pitches ascend or descend in a line.

I guess I could use the really specific rule that over a I-ii-V-I sequence, I can pick out an ascending four note major scale fragment which ends on the root of the final I chord, with the root note doubled or inverted to be the top pitch of the last chord voicing.

It seems like you would not want to apply this idea in every I-ii-V-I situation, since that's a very common harmonic idea and it might lead to tedium when over applied. Where is this idea best applied? Would you do this every time the situation presents? I would guess that there might be other formulas that you might use in the same situation for variety's sake. It seems like it would be most interesting to hear it towards the end of a chorus or of an eight-bar section where the progression is doing its final march to the I chord, or else as a progression starts out.

If I'm applying the more general rule of identifying linear patterns to use in the top voice, it seems like the idea would be to find top voice motion that moves either up or down, which follows some identfiable scalar path (chromatic scale, major scale, one of the minor scales, diminished scale) that moves as a succession of major or minor seconds. Do you have patterns like this in your repretoire when you voice progressions? How many chords do you try to tie together with these patterns? Is the idea best used for groupings of three or four chords or is it too tedious to carry this over a section of eight or more chords?

Following a different idea, the voicings rules I use in the SongTrellis Editor don't attempt to often invert chords so that non-root tones are used in the bass. At the time I wrote that code, I didn't feel like I knew enough to do this kind of inversion safely. Could you give any guidance about how to do these kinds of inversions and not screw the harmony up?

Please join our community at SongTrellis. Our contributors welcome your comments, suggestions and requests. As soon as you join the site (or login if you are a member) a response form will appear here.