chord voicings
Okay, this should do everyone some good, but it may be difficult for me to convey to guitarists. Apologies in advance for my lack of guitar skills. If you are habitually playing voicings with the root in them (especially in the middle or at the bottom), this article is for you.
Generally you learn theory by examining harmonic intervals (two simultaneous notes) first, and then progressing toward triads (three simultaneous notes. Once you learn triads, you just keep building upwards: seventh chords and then extensions of these.
Triads and seventh chords are built of the intervals known as thirds. A major triad is a minor third stacked on top of a major third. A minor triad is a major third stacked on a minor third. Adding a seventh involves stacking another third atop the ones you have. chords built of thirds are said to be tertian voicings.
This is not the only way to build chords, and it’s not the most pleasing (although I guess I can allow for preference somewhat). It has advantages–the systematic way of stacking upward from the root is easy to understand at first compared to other voicings.
In jazz, it’s idiomatic for the chording instruments (guitar and piano/keyboard) to omit the root from their chord voicings, since the bassist is usually covering that function. Once free from that obligation, a new world of harmonic structures opens up in front of us.
A completely different sound emerges when you build chords using the interval of a fourth. These are known as quartal voicings. Adding these sounds to your harmonic vocabulary will expand your playing, literally and figuratively.
We’ll go through a few quartal structures and look at how they can be used in place of the tertian structures you learned early on.
The very first thing to note is that the pentatonic scale can be expressed as a stack of perfect fourths. So for A major, you can take the pent scale A B C# E F# and arrange it as a stack, building upward from the third of the chord: C# F# B E A. Play this quartal structure over the root A and you have achieved an A major sound, with considerably more openness than the conventional tertian voicings. The doubled root on top is optional. Pianists probably need two hands to cover this compass, so it’s important that you not try to double the root on the bottom–your hands will be needed up higher.
The relative minor is F# minor. Note that the same five notes are F# minor pentatonic. But, you get a doubled root fairly low, which is best to avoid. The top three notes, B E A, form a nice quartal voicing for F#mi11.
There’s another structure that is more useful for minor. This is sometimes called the “So What chord,” beause it appears prominently in the Miles Davis composition of that name. It is a stack of fourths built upward from the root, but the topmost note is lowered to give a major third as the topmost interval. So it’s not purely quartal, but still has a lot of the favorable characteristics.
The So What chord for F# minor would be built as: F# B E A C#. I will note for the guitarists, this is the same interval structure formed by your five lowest open strings: E A D G B.
So now we have two entirely new shapes–shapes that are not stacks of thirds or inversions of those. The wider spacing between notes gives substantially different character to these. Play them to get the sound in your ears.
These voicings span a larger compass too, which adds to a full sound. You can collpase them into a single octave when you prefer a closer spacing. The resulting voicings will be full of seconds, which will add a lot of tension compared wth tertian voicings. Varying the spacing between open and closed is great for creating tension and release–now you have more devices available thanks to these new voicing formulas.
What about the dominant quality? Let’s lower the bottom most note in e So What chord by one semitone. We get a structure that built upward is a tritone, two perfect fourths and a major third: F B E A C#.
This formula is very well-used. It’s the quartal way to spell G7#11, and also Db7#5#9 (also notated Db7alt). Since it is rootless, it works as either chord. The tritone forms the third and seventh of the dominant seventh chord in either case. This can be thought of as an A triad over G7. Or as A triad over Db7.
I think the alt chord and its various uses deserves an entire article. So I won’t really go into it here, except to note that the superlocrian mode (seventh mode of ascending melodic minor) is the scale associated. So you could play the D ascending melodic minor scale (D E F G A B C#) over the chord we just built (F B E A C#).
Okay, you now have three entirely new structures to add variety to your chord voicings.
For major, you have the stack of four perfect fourths, built up from the third of the major chord. For example, F major would be built as
A D G C F,
with the bass playing the root. Note that the major seventh does not appear in this, but the 6 and 9 do.
For minor, you have three fourths and a third, built up from the root. So F minor would be built as
F Bb Eb Ab C,
and we are doubling the root. This same structure can be Dbmaj7, or GbMaj7#11 by moving the root accordingly.
For dominant you have a tritone, two fourths, and a third, built up from the third or seventh of the dominant chord. This gives you two possibilities. F13#11 would be built up from the seventh as
Eb A D G B,
and F7alt would be built up from the third as
A Eb Ab Db F.
These are duals–F13#11 is B7alt, and F7alt is B13#11.
These three shapes are the “big three” you’ll use a lot in playing jazz. They can add a lot of interest to your comping. I’m not advocating the total abandonment of tertian harmony–just know about these quartal shapes and use them to get out of the sonic rut that triads put you into.
I’ll stop having outlined the basics of this idea. There’s more theory that applies here, but I want some feedback before I launch that discussion.
(programming note–I’ll add audio files soon so you can hear this theory put to work)

Nice post! I was just talking to my piano teacher about these. I haven’t seen those #11 or alt voicings before though (from flattening the root of a So What chord), they sound good. I was thinking that for the dominant chords, I’d just take the major stacked fourths voicing and just raise the 6th in the left hand to a flat7. Or build stacked major fourths voicing off the b7 of the dominant. Any comments on that ?
I’ve been working on smooth voice leading with these kinds chords, has been a little tricky, and a little tedious b/c without the root, the harmony tends to sound very spacey and I don’t feel a very strong cadence; or at least I’m not used to it yet. Thanks for the fresh perspective and good luck with your 100 Standards.
Comment by whitecraneboxing — April 15, 2007 @ 8:14 pm
I would definitely say it takes getting used to rootless.
On your other question, I find the best sounds come when the voice spacing is closer up high and wider down low, as a general rule. My readings in psychoacoustics indicate that’s becuase the harmonic series is spaced wide low and close high–seems reasonable. But in any case, I like to keep the tritone interval on the bottom. I won’t say that your solution doesn’t work, it just doesn’t usually occur to me as I build chords.
Comment by Rob — April 15, 2007 @ 8:36 pm
Hi, my question is this: Let’s say I choose a jazz standard i.e. Autum Leaves or All of me, and want to harmonize some parts of the melody by using quartal voicing but starting from the melody note downwards, as tis is the way I was asked to. Would this work out well without loosing the feeling of the original melody? The hint here is that I MUST start from melody note down. How can this be done?
Comment by pepearm — April 26, 2007 @ 6:41 am
The real key there is that my formulas will work out best if the root or fifth is in the melody, and least well for the third or seventh, since I place those on the bottom mainly.
So the melody notes in the first phrase of Autumn Leaves are the thirds of the chords, which isn’t a great start. So I wouldn’t pick that example.
Comment by Rob — April 27, 2007 @ 12:22 am