Knowing what scale to use

[guitarjazz]

I had a counterpoint teacher, a German man, who said he practiced Bach first thing in the morning to 'clear his head'. I like that!

[engelbach]

Originally Posted by guitarjazz

I had a counterpoint teacher, a German man, who said he practiced Bach first thing in the morning to 'clear his head'. I like that!

I'm not much of a classical player, but I work through Bach chorales for the same purpose.

[Music997]

If you're in C Major and you see a diatonic chord such as G7, obviously you will use a G mixolydian scale which is the same thing as a C Major scale starting on G and ending on G.

If you're in C Major and you come across a chord such as A-flat Major 7th, then you know you're dealing with a chord outside of the key of C Major and will use an A-flat Major scale which is the same thing as the A-flat Ionian Scale.

[engelbach]

Originally Posted by Music997

If you're in C Major and you see a diatonic chord such as G7, obviously you will use a G mixolydian scale which is the same thing as a C Major scale starting on G and ending on G.

If you're in C Major and you come across a chord such as A-flat Major 7th, then you know you're dealing with a chord outside of the key of C Major and will use an A-flat Major scale which is the same thing as the A-flat Ionian Scale.

Forgive me if I offer a slightly different approach.

These examples are only true if you follow a narrow interpretation of Chord-Scale Theory, according to which every chord has its own scale and only that scale, which I don't recommend.

A G7 in C major only gives you four notes. You have eight more notes to choose from, some of which will sound more consonant than others. In fact, it's normal in jazz to add at least one altered note over a dominant for more interest, such as a b9; i.e., in this case Ab instead of A natural.

The same goes for an AbΔ in C major. The only notes you're given are Ab C Eb G, only two of which are outside of C major. You could play in C major except for E and A, which you would flat to accommodate the chord.

An isolated chord that contains non-key notes does not have to denote a key change. And even if it does, there are other possibilities; for example, an AbΔ may not denote the key of Ab, but the key of Eb (same as Ab Lydian), which would take an Eb scale, with a D natural instead of a Db. Or C harmonic minor, with both D natural and B natural.

The important thing is to create melody, not scales, and thinking of chords in relation to the key will produce more interesting melody than thinking of them in isolation.

[mwtzzz]

Originally Posted by Music997

If you're in C Major and you see a diatonic chord such as G7, obviously you will use a G mixolydian scale

No, what you use depends on what kind of sound you need.

1. Straight G7. You may improvise using tones G, A, B, D, F, (and E as passing). this provides a diminished sound

2. G7sus4. You may improvise using tones G, A, C, D, F (and E as passing). This is a suspended chord. It is sometimes written as a Dm7 with 11 (or G on top), but its function is a suspended G7. You may follow it with a G7, G13, G7b9, or follow it directly with C6 or Cmaj7.

3. G13. You may improvise using tones G, A, B, D, E (and F as passing). this avoids the diminished sound of the straight G7 chord.

4. G7#5 You may improvise using tones G, A, B, D#. This provides a C major blues sound due to the inclusion of the blue note.

5. G7b9. You may improvise using tones E, F, Ab, B, D. The G13 often precedes this chord, which is then followed by C6.



The chords given above are different chords with different sounds which are appropriate in different situations depending on which chord comes before and which one comes after.

Any of the above chords can be followed by a Cmaj variant (Cmaj, C6, Cmaj7). By the way, C6 is the same thing as Am7 and if you notice the tones on (5.) you will see that they resemble an E7, which is the secondary dominant of Am7.

[guitarjazz]

Originally Posted by engelbach

Forgive me if I offer a slightly different approach.

These examples are only true if you follow a narrow interpretation of Chord-Scale Theory, according to which every chord has its own scale and only that scale, which I don't recommend.

A G7 in C major only gives you four notes. You have eight more notes to choose from, some of which will sound more consonant than others. In fact, it's normal in jazz to add at least one altered note over a dominant for more interest, such as a b9; i.e., in this case Ab instead of A natural.

The same goes for an AbΔ in C major. The only notes you're given are Ab C Eb G, only two of which are outside of C major. You could play in C major except for E and A, which you would flat to accommodate the chord.

An isolated chord that contains non-key notes does not have to denote a key change. And even if it does, there are other possibilities; for example, an AbΔ may not denote the key of Ab, but the key of Eb (same as Ab Lydian), which would take an Eb scale, with a D natural instead of a Db. Or C harmonic minor, with both D natural and B natural.

The important thing is to create melody, not scales, and thinking of chords in relation to the key will produce more interesting melody than thinking of them in isolation.

Not to open a can of worms but...George Russell, who might be accused of starting the whole chord-scale approach, was equally in favor of looking at the relationship with the key, a horizontal approach as he called it, as having access to scales that acknowledge a vertical relationship with the chord of the moment. Most players are sensitive to both approaches. Focusing entirely on one or the other would give a greyscale view of a colorful tonal world.

[engelbach]

Originally Posted by guitarjazz

Not to open a can of worms but...George Russell, who might be accused of starting the whole chord-scale approach, was equally in favor of looking at the relationship with the key, a horizontal approach as he called it, as having access to scales that acknowledge a vertical relationship with the chord of the moment. Most players are sensitive to both approaches. Focusing entirely on one or the other would give a greyscale view of a colorful tonal world.

My post was in response to the suggestions by Music997, which I found too narrow.

My point is that one can choose from any number of scales at any time — including implying a key change on every chord, if that's what one wants.

[Ken Valentino]

The Chord Scale way of thinking turned into a dead end for me as well. Certain styles can work fine with the chord scale way of thinking, but for someone who wants a greater number of note choices it can be almost a bad habit to think that way. Any note can potentially work at the same time as any chord.

It's possible to have just the notes in a C major chord sound unintended or all 12 notes(even microtones in-between) sound completely intentional. The main goal is understanding how our ear makes sense of it all, which actually leads to knowing what something will sound like. For me it started clicking when I looked at the combinations of tendencies.

[mwtzzz]

[QUOTE=Ken Valentino;609462] Any note can potentially work at the same time as any chord.

Welll ... "potentially" is the key word, right. Whether a tone will work depends totally on harmony, which involves the harmony you come from and which you're going to (because harmony by its nature is backward and forward looking).

Saying "any tone can work" really is meaningless unless you qualify it by saying under which conditions a particular tone works.

It's possible to have just the notes in a C major chord sound unintended

Depends on which chord comes before and which comes after. In other words, harmony.

[guitarjazz]

[QUOTEAny note can potentially work at the same time as any chord.[/I]
.[/QUOTE]
George Russell's work also articulated this approach.

[Ken Valentino]

Originally Posted by mwtzzz

Welll ... "potentially" is the key word, right. Whether a tone will work depends totally on harmony, which involves the harmony you come from and which you're going to (because harmony by its nature is backward and forward looking).

Saying "any tone can work" really is meaningless unless you qualify it by saying under which conditions a particular tone works.



Depends on which chord comes before and which comes after. In other words, harmony.

I think we are almost saying the same thing, at least in that a note only works because of a relative relationship. The main difference is that I don't believe harmonic connections are the only reason ever for a note to sound intended.
We can define Harmony as a more general word to include several tendencies. but then at some point we are back where we started. For instance if we have just a C major chord and a G major chord, I need to know what the other tendencies are to know what I'm really measuring from, which in turn will give me a more consistent label for that sound.

[guitarjazz]

[QUOTE=Ken Valentino;609496]

Originally Posted by mwtzzz

I think we are almost saying the same thing, at least in that a note only works because of a relative relationship. The main difference is that I don't believe harmonic connections are the only reason ever for a note to sound intended.
We can define Harmony as a more general word to include several tendencies. but then at some point we are back where we started. For instance if we have just a C major chord and a G major chord, I need to know what the other tendencies are to know what I'm really measuring from, which in turn will give me a more consistent label for that sound.

Russell called 'tendencies',tonal gravity. He's not the only one postulate such things and may well have gotten the idea from reading Hindemith.

[mwtzzz]

[QUOTE=Ken Valentino;609496]

Originally Posted by mwtzzz

I think we are almost saying the same thing, at least in that a note only works because of a relative relationship.

Absolutely. It always depends on where you're coming from and where you're going to. There's two kinds of "relative" relationships: the relationship between two or more tones played at the same time, and the relationship of each of these to the tone(s) that came before, and then the tone(s) that will come after. This is where people start getting confused. They only think of a single chord and ask "how do I improvise on this chord" when in fact this is the wrong question because a chord by itself has little to say and the improvisation - just like the chord itself - must take into account the harmony that comes before and that comes after. Obviously this is complicated, because you're dealing with one or more tones in the present moment (played simultaneously) and also dealing with a set of tones before and a different set of tones after. Many-to-many relationships, not the 1-1 relationships that people like to deal with in their daily routine.

[guitarjazz]

[QUOTE=mwtzzz;609506]

Originally Posted by Ken Valentino

Absolutely. It always depends on where you're coming from and where you're going to. There's two kinds of "relative" relationships: the relationship between two or more tones played at the same time, and the relationship of each of these to the tone(s) that came before, and then the tone(s) that will come after. This is where people start getting confused. They only think of a single chord and ask "how do I improvise on this chord" when in fact this is the wrong question because a chord by itself has little to say and the improvisation - just like the chord itself - must take into account the harmony that comes before and that comes after. Obviously this is complicated, because you're dealing with one or more tones in the present moment (played simultaneously) and also dealing with a set of tones before and a different set of tones after. Many-to-many relationships, not the 1-1 relationships that people like to deal with in their daily routine.

The chord has plenty to say about how to play over it. Any moderate amount of transcribing will show you that sometimes the 'greats' approach parts of their solos like this.
It's surprising that a more holistic approach isn't encouraged more often.

[Ken Valentino]

The Quotes got a little messed up, as far as who wrote what. I just fixed my previous post. Sorry about that.

As far as whether a chord matters or not. It could matter a great deal, maybe almost a key change, or it could be where nothing in it makes an obvious root.

I want to know what measurements my ear is making. If I get stuck in thinking that a chord (simultaneous notes) will always create an obvious root, then my labeling of sounds becomes more random. If instead I only think about the Key then still my labeling of sounds becomes more random.

The more I've become aware of exactly what measurements are taking place, the more I've been able to predict what something would sound like.