Rewriting Annoying Music: Gloria — S 280, Gloria in excelsis, by Robert Powell

March 13, 2017

Today I return to some Episcopal liturgical organ music I worked on in late 2016. Let's first review the general problems found in annoying music.

THE PROBLEMS

1. Overwriting
2. Unnecessary difficulty
3. Haphazard texture
4. Poor or inconsistent harmony
5. Readability Issues

When to Rewrite Music

Playing music should be not only worthwhile, but ideally enriching, ennobling, character-building. Playing poorly-written music is counter-productive. Music that is well written isn't always easy to play, but the difficulty of playing good music is worth the effort. For me personally, the only time and effort worth spending on poorly-written music is the time and effort needed to improve it — to rewrite it. When I find some combination of the above problems in liturgical music that I am supposed to play for a service, I have no choice but to rewrite it, because playing poorly written music is not how I want to spend my time.

The Gloria in excelsis S280, written by Robert Powell, is a likeable and memorable tune containing triplet rhythms and some chromaticism that can be challenging and fun for a congregation to sing. Most groups I've heard do reasonably well with some practice. I found a typical rendition from an average small congregation on YouTube. The video editing is comical, and please note that the organist does not play the accompaniment flawlessly here, but this is the best version I could find online. [The video was removed from YouTube]

With no offense to Mr. Powell, the organ accompaniment in the service book is full of problems. It is mainly overdone and unnecessarily difficult to play. Nobody needs to play 2 against 3 in the tenor line during the Gloria. The harmony is consistently four-part, but the voiceleading has problems. One aspect of the accompaniment that is particularly bothersome is a B-natural in the melody which is unprepared by the harmony. Congregations consistently sing this note wrong. This happens in a couple of spots. I'll focus on the second one here. The original passage is below.

In the video above around the 56-57 second mark, you can hear the congregation (and choir) sing this note particularly flat. The harmony preceding this note needs to prepare it so that the B-natural sounds like the correct note to sing, not like an unexpected chromaticism. Solving this problem of the B natural is not exactly easy to do, but the situation can be improved. This is not the only problem with this passage. Here is an analysis with problems marked as I see them.

The reason the B-natural is problematic is that the harmony clearly emphasises B-flat in the measures leading up to it. There are also other problems in the preceding measures. In bar 19, the bass and tenor begin a standard hunting-horn arpeggiation which almost works, but not quite. The middle harmony ends up being a suspended chord, and the tenor leaps up for an appogiatura on the G for yet another chord without a third (this time the fourth is the appogiatura) and there is no good reason for that other than an attempt to imitate the triplet in the melody, but it doesn't quite work. The triplet in the tenor is approached by leap whereas in the melody it's approached by step. The following C Major chord lacks a fifth and instead has both the third (E) and the fourth (F) in it, which isn't the best choice. The 3-against-2 rhythm in the tenor is more complicated than it needs to be. This is a load of trouble in one measure. Here is a list of all the problems in this passage, bar by bar.

• Bar 19
  1. second chord of arpeggiation lacks a third
  2. third chord of arpeggiation also lacks a third via forced appogiatura
  3. The C Major chord lacks a 5th, and includes a simultaneous 4th and 3rd
  4. 3-against-2 in the tenor is unnecessarily difficult to play
• Bar 20
  1. retrogression V to IV from previous measure
  2. doubled third in the B-Flat downbeat then gets tripled (!) by the melody
  3. beat 3: the D minor 7 in third inversion is not a strong harmony
  4. harmony is particularly weak on the last beat, again no third there
• Bar 21
  1. B-natural in the melody sounds unexpected because of preceding emphasised B-flats
  2. B-natural is harmonised as F Major, against an A in the tenor, so the sound is distinctly lydian
  3. the tenor leaps up to an A on the downbeat, in similar motion with the melody, doubling the third of F
  4. the melody steps up from A and the returns to it, while the tenor stays on A, twice doubling the third
  5. beat 3 the tenor imitation results in virtually parallell motion with the melody in beats 3-4
  6. the tenor 2-against-3 is unnecessarily difficult to play
• Bar 22
  1. beat 3, doubled third again in the melody with the tenor, in similar motion continuing from the previous bar
• Bar 23
  1. parallell octave in outer voices, G to A
  2. implied similar motion on strong beats between soprano and tenor, B to A
  3. cross relation between alto and soprano C-Sharp to C (natural), with intervening chord having no third

The first problem to solve is the B-natural in the melody. Ideally, the congregation should have this note in their ears already from preceding harmony. Powell's melody makes that difficult to do, because the melody focuses on F (harmonic antipode to B), and the melodic contour is deliberately lydian (ascending F G A B) in this measure. The challenge then is to minimize awkwardness and at least make the B natural part of some logical harmony rather than an exotic anomaly. My strategy was to move from F to D Minor and from there to A Minor. Moving to D Minor is not a problem. I could have used a dominant in bar 19 to make this progression even stronger, but I decided on using E diminished in first inversion, for a plagal effect. Moving to A Minor is more difficult, because the melody steps up from F G A B, and there is no opportunity to reasonably include a G-Sharp in the harmony until the B appears. So I decided instead to use use another diminished supertonic in inversion, a B half-diminished in 3rd inversion, which sounds similar to G-Sharp diminished 7 over tonic (a baroque idiom), without the G-Sharp, so it gives a leading tone cadential effect.

Needless to say, in my arrangement, the voices interact in a more balanced way, and there are no more awkward 2-against-3 rhythms and no voiceleading errors. Whenever a triplet appears, the other voices simply follow suit or continue obliquely. The C natural in the melody in bar 23 is prepared by a shift in the alto away from A Major to A Minor — this is a bit gloomy for a Gloria, I admit, but it's passable as an imitation of the tenor moving from F-Sharp to F-natural a few beats prior.

Here are the two passages together for side-by-side comparison.

While I wouldn't claim that my version is completely fail-safe, it at least improves the chances for the congregation to sing the B-natural correctly.

A similar exegesis could be done for the entire piece, but it would take a long time. By now you know how this works: There are problems in the original, I find them and fix them, and the result is music that makes more sense and is more satisfying to play.

The complete arrangement is below, with a link to the .pdf file for you to download, print, and use in your services if you like. I beg your pardon for my playing on this recording. You'll notice a slip in the tenor line near the end.

Thanks for your attention. If you enjoyed this post, you may want to browse previous entries on this topic using the search field at the top of the page.

Regards,
Aaron