This post is half tutorial and half discussion, but the main focus is on discussion. It is about problems in expressing music effectively using music notation. Basically, I want to start an open discussion on these issues and hear what views people have.
As a piano composer and enthusiast of music notation, I am often faced with a struggle between these two extremes:
- Expressing music in the cleanest way possible.
- Showing as much detail as possible.
The first extreme involves the ease of reading of the notation. To express this in the form of a question: Which decisions can I make so that the notation is uncluttered and easy to read for the performer? Meaningful elements such as slurs, articulations, and pedal markings are often included in a score to guide the reader to more closely understand the composer's intentions, but in some cases this can possibly hinder the reading of the music. On the other hand, showing only notes may not be enough information for the performer.
In the second extreme, I refer to details such as slurs, articulations, and pedal markings, but more importantly the delineation of voices on a piano score. We usually denote different voices by stem direction. In the most general sense, we can usually show four voices on a grand staff without being too cluttered.
Unfortunately, music notation can only communicate so much with the performer. There are constraints such as horizontal and vertical spacing on multiple levels, page turning, and economy of production, but in this post, I will focus on a smaller-scale scenario involving making decisions within a four-measure passage that fits on one line. To illustrate these issues, I will use the case of a composer arranging a piece of popular music for solo piano. I will be using a four-bar passage from the song Scarlet from the anime Ayashi no Ceres. (For those of you know the anime and/or song, this will be an extra treat.) Here is a MIDI excerpt that goes along with the article. If you want, you can probably find the full song on YouTube. Just search "Ayashi No Ceres Scarlet".
The first step I take in creating solo-piano arrangements of non-solo-piano music is to listen to the recording and to choose the best notes to include in the piano arrangement. Things that I need to keep in mind are:
- Maximum stretch of the hands
- Distribution of notes between hands
- Important notes that determine the harmonic progression
So here is what I came up with:
Example 1
The music doesn't show a lot of detail yet, but it serves as a basis as to how I can split up the notes into voices or parts. The directive "con pedale" allows the performer to use the pedal to his liking, but doesn't give details such as where to pedal and where to not pedal in order to evoke the correct harmonies and prevent any unwanted dissonances.
Here is my first attempt:
Example 2
In this version, the melodic line has been singled out and further indicated by slurs to guide the performer with voicing (as the singer would sing it). In the third measure, notice how the upper staff has only one melodic line, yet the first chord is implied as belonging to an accompanying part.
Q: In such cases, would you separate the voices, and use rests to show the silences, or keep it the way it is for sake of clarity?
In the next attempt, the accompaniment is split up into multiple voices.
Example 3
A few changes have been made. The accompaniment is now delineated to show the bass line and the middle harmonizing part, which is shared between the two hands. At some points, the middle voice shares a similar rhythm with the bottom voice.
Q: In such cases, would you use two voices throughout to show the delineation of voices, or use a second voice only where the rhythm is very different?
Q: Is it necessary to mark phrasing for accompaniment voices?
In the end, this is what I came up with:
Example 4
The only major change here is the addition of a bass line that emphasizes the harmonic progression. You will notice that the notes don't always extend to the next chord change: sometimes it's a half note; sometimes it's a whole note. The reason for this related to the pedaling and the abstract concept of where a note ends.
In the first bar, the left hand can only sustain the low G until the A, and then it has to change positions to play the rest of the measure. The harmony implied here is a G-minor chord with an added second. I lifted the pedal at the B-flat because the A would create too much dissonance if it is sustained till the end of the measure.
In the second bar, I thought it would make sense to use a whole note, since the pedal is pretty much held for the entire bar, since all of the notes in the measure belong to the D-minor chord. So in an abstract sense, the F does sustain till the end of the measure, except the performer doesn't physically hold it down with his pinkie.
In the fourth bar, the notes played by the left hand span no more than an octave, so the performer can hold down the D with his pinkie and keep a clear harmony and use the pedal to produce a legato effect in the right hand.
Q: So my question here is: Where do you stand in terms of expressing the abstract idea of an implied harmony versus expressing what the performer can physically do at the piano?
Q: Also, which of the above solutions would you favour and why?
In addition to these questions, I would like to hear what people have to say about music notation in general.
