Melodic minor chord families?

[jster]

I suspect this is an easy question, but I'm not sure. For the harmonized major scale we divide the chords into (typically) three categories: tonic, subdominant, and dominant. How come we don't do the same for Melodic Minor? Or do we? And I suppose the same could be asked about Harmonic Minor.

I could develop the question, but I'm more interested in just the short answer. So I'll keep the question short.

Thanks in advance.

[Jeff Brent]

Some consider the jazz minor to be one big chord: R b3 5 7 9 (relative to parent).

Otherwise, if you consider the ascending melodic minor and harmonic minor as members of the composite minor family, the composite minor, of course, has chords functioning as dominant and subdominant built off the fourth and fifth degrees.

[jster]

Thanks Jeff,
That's basically the answer I hoped to hear, not in so far as I have an opinion of my own, but in so far as it least confirms that I am paying attention! One minor matter (no pun intended): why only five notes in your big chord?

[michaelsorg]

Originally Posted by jster

Thanks Jeff,
That's basically the answer I hoped to hear, not in so far as I have an opinion of my own, but in so far as it least confirms that I am paying attention! One minor matter (no pun intended): why only five notes in your big chord?

Melodic minor does present some problems in creating 7-tone extended chords. It's usually only done for the 4th and 7th mode. So for C melodic minor, you can have an F13#11 or a B7 #11#9b9#5 (B7 altered). In other chords, you run into problems of dissonance with the B-F tritone or the B-C minor 9th interval. For a half-diminished chord, you can effectively use 6 tones (Am9b5 add11).

As far as the functions, yes that's less important here because the "jazz" 7-tone melodic minor is seldom used to set up chains of progressions the way harmonic minor is. It's used more modally - each of the modes has its unique sound because of the reactions of the bass with the upper tones.

For harmonic minor, you can check out the chain of chords yourself - it's self-explanatory.

[jster]

Thanks Michael. That's amazing. I just asked about the half diminished with an add 11 on the guitar forum from an unrelated direction. This is great. So let me then ask two follow up questions:

1) I was motivated to ask my original question because I was reading an improvising book where the author, once he argued for using the altered mode (over a V chord) and the locrian #2 mode (over a iim7b5), went on to argue that he could just use any four note arpeggio from MM. And I didn't see why that made any particular sense. He seemed to favor min(maj7) and maj7#5 both in place of the altered mode and in place of the locrian #2 mode. (So over Balt he likes Cmin(maj7) arpeggio, etc, etc.) So I guess here we have two questions: 1a) Why would one limit oneself to arpeggios when one has already argued for the whole 7 notes, especially absent any chord family story? 1b) What's so great about min(maj7) and maj7#5?

2) When we consider the 6 note half diminished chord, there are two minor 2 intervals. How come we don't run into problems?

Thanks!

[Jeff Brent]

Originally Posted by jster

why only five notes in your big chord?

It's not *my* big chord

Levine states that those are the defining notes of the jazz minor scale.


I have a different method of assigning the defining notes of the jazz minor. It partly revolves around whether or not the absence of any one note from the scale could imply another different scale.

The b3 defines that the scale is not some kind of major. It's absence would result in ambiguity.

If the nat7 were absent, it could be mistaken for dorian.

If the nat6 were absent, it could be mistaken for harmonic minor.

The P5 is non-defining for the same reason that it is non-defining in the majority of chord structures.

A P4 does not help to define western scales either.


This method of analysis is also partially based on the avoidance of chromatic clusters in common occidental scales.

The M2 can't be lowered because that would create a chromatic cluster between the nat7 and root, and it can't be raised because #2=b3. As it is the only option possible, it must be non-defining.

Using this same criteria, it is interesting to note that the root itself can be considered non-defining as it is the only option possible as well:

A #1 creates a chromatic cluster between it and the M2 and b3, and b1=nat7.

[michaelsorg]

Originally Posted by jster

Thanks Michael. That's amazing. I just asked about the half diminished with an add 11 on the guitar forum from an unrelated direction. This is great. So let me then ask two follow up questions:

1) I was motivated to ask my original question because I was reading an improvising book where the author, once he argued for using the altered mode (over a V chord) and the locrian #2 mode (over a iim7b5), went on to argue that he could just use any four note arpeggio from MM. And I didn't see why that made any particular sense. He seemed to favor min(maj7) and maj7#5 both in place of the altered mode and in place of the locrian #2 mode. (So over Balt he likes Cmin(maj7) arpeggio, etc, etc.) So I guess here we have two questions: 1a) Why would one limit oneself to arpeggios when one has already argued for the whole 7 notes, especially absent any chord family story? 1b) What's so great about min(maj7) and maj7#5?

2) When we consider the 6 note half diminished chord, there are two minor 2 intervals. How come we don't run into problems?

Thanks!

Well, the 5 notes Jeff and Mark Levine and your other source mention do allow graceful motion within this tonality. They can be considered the first place to start with this sound. If you have an accompaniment source such as Band-in-a-Box, plug in an F13#11 chord and improvise using these 5 tones: C Eb G B D. You'll notice it's hard to make a mistake; it's in a way like using the major or minor pentatonic scales. Now throw an F into the mix. You'll notice that the potential for not-so-graceful leaps in your melody line has significantly increased. But if you use the F in the context of the minor pentachord, it usually comes out fine (C D Eb F G). Or using the A within the context of a G major pentachord (G A B C D)

Anyway, back to the 5 notes: Whether you're playing Cm(maj7), Ebmaj7#5, F13#11, Am11b5, or B7#9#5, those same tones will serve you well. That being said, it's a worthwhile goal to learn to use the entire scale to create viable, melodic, and unique lines. One chord pair that sounds nice, especially over the B7 alt would be G major triad alternating with F major. Try running all the seventh chords from the scale. You'll see that some sound very smooth, some less so, but you can use all these in a solo - you just have to find a way to de-emphasize the less effective and lay into the more sonorous chords.

My best advice for learning these sounds: Use them over ballads!

[michaelsorg]

Hi jster, here's a little exercise using melodic minor over a minor ii-V-i. The first 4 bars use some parallels between patterns in F mel mi, Ab mel mi, and C mel mi. The second 4 bars bars use directly parallel patterns, which is a very effective way to use mel mi, even though it's not a novel idea.

[kenlars99]

Originally Posted by jster

I suspect this is an easy question, but I'm not sure. For the harmonized major scale we divide the chords into (typically) three categories: tonic, subdominant, and dominant. How come we don't do the same for Melodic Minor? Or do we? And I suppose the same could be asked about Harmonic Minor.


Thanks in advance.

It seems to me that jazz minor is often used in a very different way than the major scale... not so often that there are multiple chords in a row all diatonic to the same jazz minor scale.

jazz minor can function as all 3, in particular these very common usages:
locrian major 9 as a subdominant
altered scale as a dominant
minor with a major 7 or major 6 as a tonic

however all 3 of these have different "root" of the jazz minor.

[michaelsorg]

Originally Posted by jster

2) When we consider the 6 note half diminished chord, there are two minor 2 intervals. How come we don't run into problems?

I just realized that I hadn't addressed part 2 of the question. In an improvised melody, a minor ninth can create problems, but usually only if the player didn't hear it in his/her head before playing it. In chords, it's fairly difficult to make a minor 9th interval sound good anywhere but between bass note and b9. Another sort of min 9th could be used ocasionally for unusual tension and color.

That's why the half-step structures within a set of tones are normally voiced in chords with half-steps or as major 7ths - that's usually not a problem, and can create some nice voicings. We don't usually leave a half-step on the top - that gets you back into the region of unusual tension.

[jazz oud]

Originally Posted by michaelsorg

We don't usually leave a half-step on the top - that gets you back into the region of unusual tension.

I think it's perfectly fine to voice a half-step on top, as long as it isn't a melody note. It is tense, but a usable tension in modern jazz IMO. I guess you're right that it isn't the "usual" voicing, but it's not really problematic either.

I think the problem would be less the half-step tension than the obscuring of the melody. I would tend to avoid a whole step below the melody as well.

I.e., you can voice the chord with the maj7 on top, but if the singer/melody instrument is singing/playing the root in the same register, it won't sound good.

[michaelsorg]

Originally Posted by jazz oud

I think it's perfectly fine to voice a half-step on top, as long as it isn't a melody note. It is tense, but a usable tension in modern jazz IMO. I guess you're right that it isn't the "usual" voicing, but it's not really problematic either.

I think the problem would be less the half-step tension than the obscuring of the melody. I would tend to avoid a whole step below the melody as well.

I.e., you can voice the chord with the maj7 on top, but if the singer/melody instrument is singing/playing the root in the same register, it won't sound good.

Yeah, I like to do left hand voicings under a solo with a half step on top, but the improvisation is happening above that. If the melody or solo instrument is in the middle or low register, I'm very careful about doing that. Also in arranging, you have to get the balance and intonation perfect to make those voicings work. Register also plays a role - the higher that half-step is, the more it bites.

In terms of the maj 7, I'm speaking of the interval between 2 given chord tones, most likely with a tone in between, not necessarily the maj7 of the chord.

You run into the minor ninth problem in voicings just dealing with the diatonic scale. Say you've got an Fmaj7 to voice for piano, guitar, or band. You can start on the E below middle C, build it up in 4ths: E A D G C, beautiful. But put an F on top of that? Hmmm...not for me. You can double the E, and that works real well for melody harmonization; then you have the melody on top and doubled two octaves lower. Or you can content yourself with 5-note quartal chords; they work great and never incur that minor ninth interval.

[dogbite]

Originally Posted by jster

I suspect this is an easy question, but I'm not sure. For the harmonized major scale we divide the chords into (typically) three categories: tonic, subdominant, and dominant. How come we don't do the same for Melodic Minor? Or do we? And I suppose the same could be asked about Harmonic Minor.

I could develop the question, but I'm more interested in just the short answer. So I'll keep the question short.

Thanks in advance.

the (well, mine anyway) short answer is this:

there are two usages of melodic minor in music. 1) the classical use as a minor tonality during mostly cadences, where the ascending and descending forms are different and the functions of tonic and dominant remain identical to the major mode, and 2) the jazz melodic minor (ascending form) modes where modes are substituted for the functions found in classical harmomy...

for example, F melodic minor for E7(alt) resolving to Am is functioning as a dominant, even though the vii of melodic minor is not a dominant scale step. same for the Bb7(9/#11/13), where the Bb is the subdominant of F melodic minor; however, may function as a dominant (V7) of Eb major. in other words, the subdominant (IV) of F melodic minor is functioning as a dominant (V7) of Eb major in the following progression:

Bb7(#11) Ebmaj9

E        F
D        D
Ab       G
Bb       Eb

in which case the pitch set F G Ab Bb C D E (F melodic minor, ascending form) may be used as a mode of the Bb lydian dominant scale, functioning as the dominant of Eb.

sorry if "short" wasn't so short after all...

[jster]

Originally Posted by Jeff Brent

It's not *my* big chord

Levine states that those are the defining notes of the jazz minor scale.


I have a different method of assigning the defining notes of the jazz minor. It partly revolves around whether or not the absence of any one note from the scale could imply another different scale.

The b3 defines that the scale is not some kind of major. It's absence would result in ambiguity.

If the nat7 were absent, it could be mistaken for dorian.

If the nat6 were absent, it could be mistaken for harmonic minor.

The P5 is non-defining for the same reason that it is non-defining in the majority of chord structures.

A P4 does not help to define western scales either.


This method of analysis is also partially based on the avoidance of chromatic clusters in common occidental scales.

The M2 can't be lowered because that would create a chromatic cluster between the nat7 and root, and it can't be raised because #2=b3. As it is the only option possible, it must be non-defining.

Using this same criteria, it is interesting to note that the root itself can be considered non-defining as it is the only option possible as well:

A #1 creates a chromatic cluster between it and the M2 and b3, and b1=nat7.

So as you see it, the b3, 6 and 7 define the jazz minor? Am I reading you correctly?

[jster]

Originally Posted by michaelsorg

Hi jster, here's a little exercise using melodic minor over a minor ii-V-i.

Thank you Michael. So your lesson here was mostly to see how smooth things sound when we leave out the 4th and 6th of the parent?