Can we talk Jazz chord substitutions?

[Free]

I've studied Jazz chord substitutions from various internet sources. I plan to buy a book on this, because there seems to be some variances from player to player on what subsitution methods are best. That said, I'd really appreciate if some theoretically advanced players would elaborate a bit on chord substitution theory - a general overview from fundamentals to advanced? Thanks much!

-Mike

 

[JonR]

First consideration is chord function: what job is the chord doing? Can you find another chord that will do the same job (or near enough)?
This means basically (a) how it harmonises the tune, and (b) how it leads to and from the chords either side.
The melody is fixed - that's a basic rule. However, almost any chord can be made to harmonise a given note, so that's not a great limitation.

So most chord substitution in jazz is focussed on (b).
There are three basic chord functions in tonal music:
1. TONIC. I, vi or iii chords.
2. DOMINANT. V or vii chords.
3. SUBDOMINANT. IV or ii chords.
Substitutions don't always work both ways, however. With dominant and subdominant they more or less do, but while you can replace a I chord with the iii or vi, you probably shouldn't replace a iii or vi with the I.

It's with the dominant chord where jazz has most fun with substitutions.
The two most common are the tritone sub and the dim7.

The tritone sub means using another dom7 a tritone away from the first. This works because both chords share the inner tritone between 3rd and 7th (roles being reversed). And the 3rd and 7th are the key guide tones of the chord, which determine its function (root and 5th being less important).
A tritone sub has the additional attraction of providing a chromatic bass descent. (Remember these are used for V chords in cycle-of-5th moves.)
Eg, Dm7-Db7-C instead of Dm7-G7-C.

The dim7 sub means building a dim7 chord on the 3rd, 5th or 7th of the dom7 - or a half-step above the root.
This derives from the minor key, where the V chord is (typically) a 7b9, and the vii is a dim7 which is a rootless version of the V7b9. (Eg, E7b9 and G#dim7 in the key of A minor.)
Minor key cadences are often used in major keys, so dim7s are widespread subs in sequences in minor or major.

(This will do for a start. More later - and not just from me, I'll bet...)

 

[KRosser]

First - for many tunes there are commonly played substitutions (for instance, the 1st two bars of Girl From Ipanema - FMaj7 Bbmaj7/Am7b5 D7alt/ G13 /). Many of these aren't written anywhere - you have to kinda pick them up 'off the street'. Play with lots of experienced players, keep your ears open, and don't be afraid to ask the piano player, "Hey, what was that sub you used on such-and-such tune?". It always surprises me how many players don't want to ask questions of the people they play with, for fear of looking insecure or whatever.

As far as 'methods' go, one way of looking at chord subs that I like is somewhat reductive - take the melody note(s) and make a list of every chord that fits

For instance, let's try reharmonizing the second bar of "All The Things You Are"

First bar - melody note Ab, first chord Fm7

Second chord - melody note Db - the given chord is Bbm7. Db (C#) also belongs to:

Bb7#9
Bm9
B13
C7b9
Dbm7
Dbmaj7
Db7 or alt
Dbdim
Dmaj7
DmMaj7
Eb7
Ebm7
Ebm7b5
E13
Em13
F7aug
Fmaj7#5
F#m7
F#maj7
F#7
Gmaj7#11
G7b5
Gm7b5
Ab7sus
Abm11
A7
Amaj7

...and that's not everything!

Next thing to do is figure out the bass movement - bass movement tends to be the most ear-friendly if it's approaching via a descending half-step, moving in an interval pattern (i.e., successive ascending or descending half steps, whole steps, m3rds, maj3rds, etc), follows the circle of 4ths/5ths, or doesn't move at all (pedal tone). Important: bass movement does not mean 'root movement', necessarily - inversions can be your friend here....

The next step is to find how any of those chords (including their inversions) can be used to bridge the first Fm7 and the following Eb7 using these ideas...

I'm gonna stop now...

The lesson here, if I have one, is that you can make something work often by reconciling only the bass movement and melody - the inner voices (that will actually make up the bulk of the chord) can be up for grabs if one's ears are willing to be stretched...

 

[diego]

Pay attention to where you put them. Very important and it's rarely mentioned. The even-numbered measures create tension (weak mesures) which resolve to odd-numbered measures which are stable (strong measures). Since chord substitutions typically create more tension, they work better in the odd-numbered measures. Of course this is a general rule, but a good one. Listen to Coltranes Mr. P.C. and count the form using the name of the measure on the first beat... 1234 2 (second measure) 234 3234 4234, etc. for the twelve-bar blues, and listen to where the lines based on substitutions fall (you'll hear they sound "outside" the harmony and create more tension. Or take a blues and simply anticipate the next chord by a half step. Put the sub in the weak measures, but try the strong measures too, so you get a feel how all this works.

Ken: You're playing in San Francisco on the night I have tix to see Bill Frisell! It's not fair!

 

[Free]

As far as 'methods' go, one way of looking at chord subs that I like is somewhat reductive - take the melody note(s) and make a list of every chord that fits.

Thank you very much Jon, KRosser and Diego. I particular like the fundamental concept of simply being cognizant of the melody notes and considering any chords that fit those notes. Of course, a melody is extremely dependent on it's effect in the context of a harmony - a melodies effectiveness depends a lot on the harmonic context, so ultimately it's a process using one's ear to determine what sounds best. Maybe that is more important than substition methods? I'm starting to think chord substitutions are best employed in solo choruses and not over a song's composed melody? I really don't know. I'm still trying to crystallize all this... Thanks again!

-Mike

 

[KRosser]

Thank you very much Jon, KRosser and Diego. I particular like the fundamental concept of simply being cognizant of the melody notes and considering any chords that fit those notes. Of course, a melody is extremely dependent on it's effect in the context of a harmony,

To be certain: this is a method of reharmonization that definitely deals with a post-Ornette concept - melodic pull trumps harmonic function. And it assumes a melodic pull from both above and below. A few of the masters of this kind of approach are Andy Laverne or Richie Beirach, etc...

so maybe it's a process using one's ear to determine what sounds best.

Well, this should always be your final arbiter, recognizing that ideally this is always subject to flexibility/change/evolution/de-evolution/whatever....

What I've suggested is merely a method for 'opening up one's ears'. If you find something is too far, an important parameter has been established, too,for your own musical awareness...I think....

 

[Free]

Also, I think this question was kind of missed. I've wondered about this a while: Does the term substitution also apply to when one plays different root scale modes over the a given chord when soloing? Like when a player like Pat Martino solos with an Em7, Bbm7, etc scale over A7 #5? I mean is substituting scales considered chord substitution too?

 

[rosscoep]

The concept of tri-tone subs flipped my mind when I first learned it.

 

[KRosser]

Pay attention to where you put them. Very important and it's rarely mentioned. The even-numbered measures create tension (weak mesures) which resolve to odd-numbered measures which are stable (strong measures). Since chord substitutions typically create more tension, they work better in the odd-numbered measures. Of course this is a general rule, but a good one. Listen to Coltranes Mr. P.C. and count the form using the name of the measure on the first beat... 1234 2 (second measure) 234 3234 4234, etc. for the twelve-bar blues, and listen to where the lines based on substitutions fall (you'll hear they sound "outside" the harmony and create more tension. Or take a blues and simply anticipate the next chord by a half step. Put the sub in the weak measures, but try the strong measures too, so you get a feel how all this works.

In my own experience, the bar numbers don't matter much when dealing with reharmonization. Even in light of tension/resolution, a resolution can still be a sub.

Ken: You're playing in San Francisco on the night I have tix to see Bill Frisell! It's not fair!

When faced with the above quandary, the smart person would opt for Frisell...

Not to worry - I had to bail out on that gig due to logistical concerns

Enjoy Frisell quandary-free!

 

[Free]

To be certain: this is a method of reharmonization that definitely deals with a post-Ornette concept - melodic pull trumps harmonic function. And it assumes a melodic pull from both above and below. A few of the masters of this kind of approach are Andy Laverne or Richie Beirach, etc...

Thanks for the response. Could you clarify "melodic pull from both above and below"? What is above and what is below exactly? Thanks much.

-Mike

 

[JonR]

Also, I think this question was kind of missed. I've wondered about this a while: Does the term substitution also apply to when one plays different root scale modes over the a given chord when soloing? Like when a player like Pat Martino solos with an Em7, Bbm7, etc scale over A7 #5? I mean is substituting scales considered chord substitution too?

IMO, no.
It comes back again to function. A "substitution" (IMO) implies something that takes over the same function as what it's replacing.

This has been discussed elsewhere, and my feeling is that Martino's "substitutions" are more like "outside" tensions on the chord. (IOW, nothing wrong with them, it's just a question of definition of terms.)
Em7, Bbm7, Gm7 and Dbm7 ALL have at least one note that works against the function of A7#5. Therefore they are not exact substitutions - they don't (can't) do the same job. In conventional terms, the outside notes would need to resolve back to a chord tone before the chord moved on. (Eg, the D in Em or Gm7 would need to be heard to go to C# before the A7#5 went where it was going.)

But then the rule mentioned by krosser above - melodic pull trumps harmonic function - will take care of a lot of those issues. As long as a melodic line is strong and "comes out all right in the end", any disagreements with underlying chords can often be ignored.

(Mind you, that begs the question - what are those chords doing there in the first place? If we're going to ignore them, why not just play a different song? )

 

[Free]

IMO, no.
It comes back again to function. A "substitution" (IMO) implies something that takes over the same function as what it's replacing.

This has been discussed elsewhere, and my feeling is that Martino's "substitutions" are more like "outside" tensions on the chord. (IOW, nothing wrong with them, it's just a question of definition of terms.)
Em7, Bbm7, Gm7 and Dbm7 ALL have at least one note that works against the function of A7#5. Therefore they are not exact substitutions - they don't (can't) do the same job. In conventional terms, the outside notes would need to resolve back to a chord tone before the chord moved on. (Eg, the D in Em or Gm7 would need to be heard to go to C# before the A7#5 went where it was going.)

But then the rule mentioned by krosser above - melodic pull trumps harmonic function - will take care of a lot of those issues. As long as a melodic line is strong and "comes out all right in the end", any disagreements with underlying chords can often be ignored.

(Mind you, that begs the question - what are those chords doing there in the first place? If we're going to ignore them, why not just play a different song? )

That's very helpful information, Jon. Thanks much! This thread has already crystallized a lot of insights for me. Very cool.

-Mike

 

[KRosser]

But then the rule mentioned by krosser above - melodic pull trumps harmonic function - will take care of a lot of those issues. As long as a melodic line is strong and "comes out all right in the end", any disagreements with underlying chords can often be ignored.

Semantics, maybe, but I wouldn't say ignored, since that "disagreement" becomes an essential piece of the fabric.

Even still, just to get all post-modern on ya - I wouldn't put all my eggs into Ornette's basket as much as I wouldn't put them all in Bird's either. I get to cherry-pick from all the methods used throughout music at will, regardless of history/genre, ultimately...

 

[diego]

Certainly with experience, almost anything can work... I've heard Richie Beirach play major 7ths over dominant chords -- but I still think that it's good for people learning substitution concepts to work with an awareness of form, because it does have an impact on how the substitutions sound. You might have a beginner putting some "out" chord on the first beat of a tune, you know...

Ken: thanks for gig update, hopefully you'll get back up to SF.

 

[KRosser]

But then the rule mentioned by krosser above - melodic pull trumps harmonic function - will take care of a lot of those issues. As long as a melodic line is strong and "comes out all right in the end", any disagreements with underlying chords can often be ignored.

(Mind you, that begs the question - what are those chords doing there in the first place? If we're going to ignore them, why not just play a different song? )

Because a song is not a chord progression.

If I take a familiar chord progression and write a different melody, it's a different song.

If I take a familiar melody and reharmonize it, it's the same song.

 

[KRosser]

Thanks for the response. Could you clarify "melodic pull from both above and below"? What is above and what is below exactly? Thanks much.

-Mike

Above - melody

Below - bass movement

 

[Dickie Fredericks]

The lesson here, if I have one, is that you can make something work often by reconciling only the bass movement and melody - the inner voices (that will actually make up the bulk of the chord) can be up for grabs if one's ears are willing to be stretched...

This is something that plagues me. My ears dont wanna stretch.

Im working on it though.....

 

[KRosser]

Certainly with experience, almost anything can work... I've heard Richie Beirach play major 7ths over dominant chords -- but I still think that it's good for people learning substitution concepts to work with an awareness of form, because it does have an impact on how the substitutions sound. You might have a beginner putting some "out" chord on the first beat of a tune, you know...

Like Frisell, you mean?

Ken: thanks for gig update, hopefully you'll get back up to SF.

Any/every chance I get...one of my favoritest cities in these here united of states....

City Lights and numerous Italian restaurants in North Beach are depending on it!

 

[JonR]

Because a song is not a chord progression.

If I take a familiar chord progression and write a different melody, it's a different song.

If I take a familiar melody and reharmonize it, it's the same song.

Absolutely!
But I was talking about the situation where you have a given chord sequence and are improvising over it - not reharmonising it.
Of course, you can reharmonise and then improvise over the new chords...

Or (and I think I can guess the answer... ) are you saying that two kinds of harmonisation can be happening simultaneously? One in the chords, the other implied by the improv?

 

[KRosser]

Absolutely!
But I was talking about the situation where you have a given chord sequence and are improvising over it - not reharmonising it.
Of course, you can reharmonise and then improvise over the new chords...

Or (and I think I can guess the answer... ) are you saying that two kinds of harmonisation can be happening simultaneously? One in the chords, the other implied by the improv?

I come from a school of jazz where that exploratory, anything-can-happen, who's-got-the-ball thing was the thing...

So yeah, take a guess...