I have mentioned before that Arabic traditional music does not encompass the musical parameter of harmony. Nonetheless it is an important part of my research, because I want to incorporate aspects of Arabic music into my jazz repertoire, and harmony is a very big part of jazz music. To discover some ideas on harmony in Arabic music, I have decided to arrange a traditional piece, and analyse a traditional piece played with a jazz band. For the first piece (Ya Fajar) I will look at chordal possibilities that have worked out for me.
Secondly I will look at a piece by Lebanese singer Rima Khcheich, and her band, which is formed of jazz-educated players. To look at the possibilities for me as a bass player, I have decided to do an analysis of the bass notes in comparison to the melody and the tonal centre of the piece. Also I will compare it to the traditional oud accompaniment that can be heard at the beginning of the track.
Harmony From Chords
To research and play harmony within the Arabic tradition, arranging is a necessary tool, since the original classical music from the Arab world does not have the parameter of harmony. To research the possibilities I decided to write an arrangement to the Arabic traditional song Ya Fajar, a beautiful song that compares love at first sight to the touch of the first sunlight at dawn. For the non-microtonal passages in maqam-based music it is fairly easy to write chords. For microtonal material it is somewhat trickier, because the 3rd has to be avoided. A very simple trick is to use sus4 or sus2 chords, since they don’t show harmonic colour. A leadsheet of the final arrangement is depicted on the right, and can be found as well in appendix 3.
The song is written in maqam bayati on D. Therefore it only seemed natural to start on a D-minor chord. Straight away I know I have to be weary of the E, also in my voicings. For the first phrase I was looking for sounds that complement the melody and add a slight change in colour, but nothing drastically. Because of that I stayed within the tonality of D-minor, firstly moving to F major. Because the harmonic rhythm of the first 3 bars is: 2 bars of D-minor, 1 bar of F-major, I wanted to keep that for bars 4-6. Furthermore I wanted to make a turn, away from F major, so I moved to Bb, which sounds like a maj7 chord with the A in the melody. This has a somewhat stronger sense of dissonance, making it want to go somewhere. I tried to increase this by writing a Bm7b5 chord in bar 6, adding more pressure, before releasing it back to D-minor, which has a similar 3rd relationship to Bm7b5 as F major. The A7+ in the second house is there to signal that it’s a different ending.
In the next part I used another method. This seemed like a good place to have the bass notes of the chords move in the opposite direction that the melody notes move. So the melody moves down: Bb, A, G, F, and with some ornamentations ultimately back to D. The bass has an upward motion: Gm, Am, Bb, Bm7b5 and then moves down again to Bb, A7 and Dm, for resolving purposes.
In the B part I used a third method, namely one of a drone. I tried to keep the harmonic rhythm quite static, because this is the part of the song where the vocals have their most ecstatic lines, and there is room for much ornamentation. In order to give the vocalist space, not only rhythmically but also harmonically, I deployed the tactic of using sus chords here. That way the melody can move freely over the quartertone B without clashing, but the sound of the chord is still full. I decided to harmonize the strong rhythmical accent in every second bar, for dramatic effect.
The C part is the point where the melody starts moving from its highest point back to the root of the tune. Because of the absence of quartertones in the first bars of this section, I could use simple chords to support the melody. The last line of the tune is the same as we saw before with the bass line, making a counter movement to the melody, before ending together on a D-minor chord. The biggest conclusion I can draw from this analysis is that it is in fact possible to use chords for a traditional song, and make it sound good within a jazz context. One thing I notice is that I have mostly avoided using the note E in the chords (apart from one Am and an A7 chord). This is because the piece is written in maqam bayati on D, which has an E.
Here is a recording of the melody as played by NAJWA.
Harmony From Bass Notes
In order to have a clearer picture on the possibilities of accompanying a traditional melody, I decided to transcribe and analyse a song by Rima Khcheich, Futina Al-Lathi, which was released on the record Falak from 2008 on Jazz in Motion. This song is written in two maqamat, Rast and Nahawand (the equivalent of minor). This particular record is interesting to me, because the song starts with an old recording of when Khcheich was young, which is accompanied by an oud. After this the tune is played with a jazz quartet (guitar, bass, drums and saxophone). I decided to compare both accompaniments, looking at the bass notes chosen in relation to the tonal center, and the melody. I will draw some conclusions based on my findings. You can find a PDF of the score with the analysis on the right, as well as in appendix 4.
An important thing to keep in mind is that not all ajnas have a ghammaz on the 5th note. We bass players like to switch between the root and the 5th of a chord. In Arabic music the 5th is not always the note to switch between. The ghammaz is usually a safe note to use as a note to alternate between, as an accompanist. In the example below however, the ghammaz is the 5th (G to C), so there is no confusion between the terms dominant and ghammaz.
The song starts with the bass introducing the tonal center, which is C. In bar 3, an Ab is played to add colour, and to anticipate the Ab in the melody. The line continues in bar 4, moving in the opposite direction of the melody, to land on a G, creating a sense of a G7-chord. In the resolution the bass follows the melody in unison (bar 6). In bars 7-9, the bass keeps stabilizing the tonal center, while landing on the strong melodic note G in bar 9. This also creates a pull back towards the C.
In bar 10 an important change in method occurs. These four bars display a distinctive harmonic movement. I find bar 11 very interesting, because the melody lands on an A, while the bass moves to F, setting the scene for the F that the voice will sing later. This creates a very strong feeling of F major, a common chordal change in jazz music. In the bars following the bass still follows the melody, creating a sensation of a I-IV-II-V7 chord progression. The melody helps with this sensation, even adding a Ab to the II chord, making it sound like a Dm7b5, with a stronger gravitational pull towards the root note C, to which it resolves in the next bar.
In bar 14 and further, the bass plays a more melodic role, following the contour of the melody, but a 3rd lower, to create a diatonic harmonic movement. In bar 17 the bass slows down in terms of melodic density, to wrap up the phrase. After that, bar 18, the tonal centre is reconfirmed, and the bass follows distinctive patterns of the vocal melody, doubling many of its notes. Bars 22-25 have quite a clear division into two segments of two bars. It sounds like a call and response, because of the melody. The bass uses this nicely, playing an Eb, making a harmonic change.
In bars 26-29 the melody itself creates a sense of harmonic motion, using intervals to create a Cm and a Dm7b5 arpeggio, as well as using chromatic colouring notes to help solidify the feeling of a (temporary) change in tonal center, moving to a subdominant. It could be argued that in bar 27 an F would be more accurate as a bass note, following the melody. This actually happens when the oud plays this part, which we will see later on. In jazz harmony theory, a 2nd degree is far more common to be the subdominant than the 4th degree.
From bar 30 we can compare the modern jazz quartet version to the older, more traditional version where the oud plays the accompaniment. It is interesting to see the different choices both players make. While the upright bass creates the sensation of G7, Eb, and Fm going to Dm7b5, the oud chooses to go from Cm, via F to Bb, then to Fm, and back to Cm. It seems like the oud stays more loyal to the tonal center, and to the melody notes, while the upright bass creates a more colourful passage here.
In bars 34 and 35 the oud and bass use similar methods of accompaniment, switching between the root and the dominant (which is also the ghammaz in this case). In bars 36 and 37, the bass repeats itself from the bars before. The oud chooses to follow the melody instead, using the Eb as a bass note.
The upright bass, in bars 38-41, uses a pedal on C to give the vocalist a stable ground to sing the melody, and space to move freely. The oud takes another interesting route here, especially in the second repeat. The oud player follows the melody, as it does in the first repetition, but it harmonizes itself using parallel 3rds, below the melody note. For this it uses the Eb, instead of the E, creating a sense of Cm dorian (alternating between Cm and F).
In bars 42-45 both the bass and the oud follow the melody notes to create a bassline. At the end of 45 the bass walks down towards the root note C, to create a bigger sense of conclusion. The four bars after, we have seen already in bars 22-25. We have information here about the oud now though. Much more than the bass, it follows the first notes of the melody (dropping the G later in bar 48). On the second repetition the oud lands on C, creating a sense of conclusion for this section.
Bar 50 until the end is also quite interesting, because both the oud and the bass follow the notes of the melody to create a bassline, but in a different placement in time. The bass sticks more to the time signature, landing on the first beat of the bar. The oud chooses to emphasize the beats where the melody lands instead, creating almost the illusion that the time signature changed here. Both instruments end with playing the last phrase in unison with the voice. A nice detail is that the oud plays a major third at the end. This is something that would happen a lot in western classical music, but apparently is adapted to Arabic music as well.
From this analysis I can conclude that there is a number of options for the bass to accompany a melody in traditional Arabic music. To categorize the options we can make a distinction between bass lines that either:
- Confirm the tonal center of the music at that moment;
- Follow the starting note, or the most important note of the melody;
- Form a colouring note with the melody or tonal center, so a harmonic motion is created;
- Follows the melody in unison;
- Follows the contour of the melody, but not in unison (i.e. in 3rds or 6ths);
- Plays a melodic countermovement to the melody;
- Form a drone or pedal tone, to give the melody space to move freely.
These options are not necessarily mutually exclusive, and can be combined within a composition, as we’ve seen in the example.