Avoid, just avoid

[nuffink]

In any book about harmonic theory you'll see reference to so-called "avoid" notes. What are they?

Play the 7th chord on the first degree of C major - Cmajor7 - C, E, G, B (preferably on an instrument with a bit of sustain) and then play the Cmjor scale C, D, E, F, G, A, B over it. What's going on with that F? Sounds like a big fat wrong 'un in the midle of C Major. How can that be?
Now do the same thing on the second degree - Dmin7 - D, F, A, C. Play the scale over it. No wrong 'uns. Weird.
Do it again on the third degree - Emin7 (same chord type as above) - E, G, B, D. Wrong 'uns F and C.
How can this be. You're playing entirely in key but it sounds like some of the notes are out and it depends upon the chord you're playing and the scale degree it's built on.

These are avoid notes. How can you tell where they are for the chord you're playing?

The answer is that they are a minor 2nd (semitone) or minor 9th (semitone an octave up) above one of the chord notes.
Take a look at the keyboard. The Cmaj7 has only one diatonic (white) key a semitone above a chord note - F the avoid note. Dmin7 has no white keys a semitone above a chord note. Emin7 has both C and F a semitone above a chord note.

So, diatonic chords and their avoid notes in C major

Cmaj7 - avoid F
Dmin7 - no avoid
Emin7 - avoid F and C
Fmaj7 - no avoid
G7 - avoid C (not important you can get away with murder on dominant 7ths as the inherent tritone tends to mask other dissonances)
Amin7 - avoid F
Bmin7b5 - avoid C

Curiously we seem to accept the inverse (the major 7th interval or semitone down) as relatively consonant which is why the Dmin7 has no avoid notes. This hasn't always been the case. The major 7th together with the tritone (6 semitones) have traditionally also been seen as avoid notes. A century of jazz influence has made us much more accepting of these dissonances.

So, now we know what avoid notes are what should we do with them, certainly we shouldn't avoid them. The basic guidelines are:-
Don't extend the chord with them
Don't hold them against the chord
Don't play them on a metrically strong position in the bar
Don't accent them

Jazzers swerve around the avoid note problem by playing chord scales without avoid notes over the chords. In the above example if you're playing diatonically you'll play the phrygian with it's two avoid notes over Emin7. Jazzers will happily play the Dorian, which has none, over a min7 regardless of what scale degree it's based on.

[Sascha Franck]

Nice explanations, nuffink.

But, as said in the other thread, I would really beg to differ between chords and scales used in a modal fashion as opposed to them being used in a functional way.

Regarding scales, this would mean some "target" notes (passing notes are almost always ok) and regarding chords that'd be possible tension notes.

Ok, let's first assume we have a modal vamp on a minor7 chord. We would most likely assume it's a IImin7 chord, as it would offer the least amount of avoid notes. So we can (and will) as well use the 6th. To stay in the key of C, in the case of our Dmin7, that'd be the B. Now, while not exactly being an avoid note (at least not as much as others), targeting it has a potential of sounding strange (because of the 7th, the C). And while this might often be a desired effect, one should defenitely be aware of it.
In a chord context, it becomes even more apparent. But well, there's 7th chords with 13ths, so why not use it on a min7 as well. In a modal context it's all fine with me.

But, let's now take that to a very simple chord progression, the good old II-V-I in C.
Our chords would be Dmin7, G7 and Cmaj7.
Along the movement from the Dmin7 to the G7 comes one of *the* most important note movements in such progressions. The 7th of the Dmin7 (a C) moves a halfstep down to become the third of our G7 (a B).
Now, while things will still not sound exactly "wrong", using the B as a targeted note on the Dmin7 already is certainly taking quite some of the "satisfying" effect when moving to the G7 away.

The same things are valid for a I-IV-V progression. In that case, targeting a #4 on the IV chord (in the key of C that'd be using a B over an Fmaj7) would probably too much anticipate the very same movement - the C moving down a halfstep to become the third of the G7.

Please note: I am perfectly happy with there being no "true" avoid notes on both the IImin7 and IVmaj7 chords, I am using the ones in question all the time, but still, in a functional context, they have to be used cerefully, at least once targeted or otherwise emphasized.

[Km7]

Yeah, a good topic.
The three main musical forces (melody, harmony, rhythm) can be used in such a way that they can work together or one against another for compensation. Just as in the case with dom7 chords - b9 is masked by the ever-obtrusive tritone, but it also contributes spice and tension to the dom7 - nice and not too disturbing especially if the resolution is regular. That's why in dom7 chords b9 and #9 may coexist together, if at least one of them is diatonic.

[pw]

Amin7 - avoid F

that is interesting for me. On acoustic guitar and piano, in a pop/folk context, it can work very well to do all sorts of variations between the Amin7 and the 'Fmaj7 with a sus2' (F, G, C, E - I'll leave it for you geeks to tell me what that chord is actually called).

I also find it quite plausible to play a cmaj7 in open position on guitar, using the F on the 4th string alternating with the E - another no no on your list.

I don't doubt that from a certain perspective (a very theoretical one), these are 'avoid notes' but maybe it also shows how theory can be nulled quite easily through playing style (eg- relational amplitude between the various notes being played), instrument timbre (for me a huge thing in how sound works), and conscious/unconscious learning 'outside of theory' (ie- revealing the cultural nature of the theory).

But in all seriousness, I love these explanations, and am in the process of learning western musical theory myself at the moment (after 15 years of playing guitar), so thanks!

[Toxikator]

Remember, though, that HOW you use the "avoid notes" is important. They're called "avoid" because you should generally AVOID them, not because they're forbidden.

For example, sounding an F alongside an E is generally a poor idea b/c of the m9. However, if you move from F TO E, you're doing a suspended-style motion which ENCOURAGES the dissonance as it immediately resolves it.

EDIT: Wait, NVM. When you play a CMaj chord, and you move the 4th string, you're not creating an interval of E-F (a m9), you're creating an interval of F-E (a M7). The avoid interval (at least as chordspace taught it to me ) is a m9, which is why we tend to avoid the F; it clashes if above the E. However, if the doesn't sound ABOVE an E in that context, only below it, it's not disallowed.

[Sascha Franck]

I also find it quite plausible to play a cmaj7 in open position on guitar, using the F on the 4th string alternating with the E - another no no on your list.

Well, you even said it yourself - you're alternating between the two. In that case, the C is no C anymore as soon as you play the F, but a Csus4. And a Csus4 to C movement is a really obvious and common one indeed.

[JumpingJackFlash]

I also find it quite plausible to play a cmaj7 in open position on guitar, using the F on the 4th string alternating with the E - another no no on your list.

Well, you even said it yourself - you're alternating between the two. In that case, the C is no C anymore as soon as you play the F, but a Csus4. And a Csus4 to C movement is a really obvious and common one indeed.

But he said the E was also present, so in that case technically it's not really Csus4 (certainly not in the traditional sense), more like Cadd4 or something (admittedly not a particularly common chord though). (It's pitch-class set 4-14 I believe).

[Sascha Franck]

Well, I read it as alternativ between the E and the F - and on the 4th string, you simply can't play them together.

Fwiw, on guitars there's a very common "add4" chord indeed. If you move that very C (the voicing is C-E-G-C-E) up a wholetone, it becomes D-F#-G-D-E (the G and E stay on open strings). You can even strum it and such, it will almost always sound pleasant.
But it's really an exception, and as far as theoretical explanations go, I prefer sticking to "avoid" notes as well.
Sure, we could as well all just say "anything goes". But in that case, any theory would be rather pointless.
And no, I'm not a firm believer when it comes to following theory rules, but for explanatory tasks, they do a fine job.

[Signal chain]

Thanks a lot ...

[Toxikator]

D-F#-G-D-E

What the heck is that? I have no idea what's going on with that sonority. It sounds like it needs to GO somewhere but I have no idea what. It almost works with a DFACF afterword... I just have no idea how to analyze it.

[Sascha Franck]

What the heck is that? I have no idea what's going on with that sonority. It sounds like it needs to GO somewhere but I have no idea what. It almost works with a DFACF afterword... I just have no idea how to analyze it.

It's a Dadd9add4.
Actually, I usually dampen the high E string. Sounds better to me. But the G is ringing all the way through.

[Toxikator]

Okay, but even if we call it a Dadd9add4 (which is still kind of shaky to me since it has no C or A in it at all, and w/o a 5 or 7 it's hard to establish sonority), how do we make sense of it in the context? First of all it's a major II, which in itself would serve almost NO purpose except as a V/V, which doesn't explain the success of the DFACF afterword...

Actually, here's my best analysis:
We don't consider the C to be a I, but instead a IV, and then the DF#GD (sans E, as per your note) as a DF#AD with a retardation in the G; which we can call up to A (in a DF#ACE, which is a V9) before resolving to G... which works pretty well.

[Sascha Franck]

Okay, but even if we call it a Dadd9add4 (which is still kind of shaky to me since it has no C or A in it at all, and w/o a 5 or 7 it's hard to establish sonority), how do we make sense of it in the context? First of all it's a major II, which in itself would serve almost NO purpose except as a V/V, which doesn't explain the success of the DFACF afterword...

Err, where do you see the DFACF reference?
I was just mentioning that chord as an example of a chord which is tough to explain by theory, yet absolutely common.

Besides, leaving out the A doesn't mean anything. An interval D-F#, without any further connections to whatever key, will sound like a major chord. And it's used as that very often.
And not having a 7th is, well, just common.
Then, adding a 9 is another common thing. Nothing to even remotely wonder about.
The only culprit here is the add4. Something that almost no theory book is even mentioning as a viable option. Yet, it's working very well.

And btw, that very chord is often used in a simple I-IV-V in G. I'll gladly record it for you. You won't notice any anormalities.

[Sascha Franck]

Oh, fwiw, for guitar players there's another very common chord featuring both the major 3rd and the 4th.
When you play an open E chord (E-B-E-G#-B-E) and then move up the lower 4 voices a wholetone, you'll be getting a nice sounding F#7add4. Another REALLY common chord.
Heck, I could even do a progression with this and another add4 chord in it, namely Badd4. You get it on a guitar by moving the same notes as before up another 4th. Sounds great.

[Toxikator]

Not the DFACF, but taking the E out of the DF#GDE (you said you muted it). The DFACF was a reference to my own post, as I tested the progression with a DFACF as a potential resolution and it worked well.

I agree that the 5 and 7 are not really necessary to a chord, BTW, but I would have thought that in an add9 chord you'd want to have the 5 oh well. Anyway, the reason I'd call it relevant is that to say that, much as the absence of an E in a CFG chord indicates suspension, the absence of an A in a DF#G chord would indicate that the G is a retardation.

and yeah, the I-IV-V in G was sort of where I headed with it, though I'd figure for full effect the DF#G would do well to be moved to a DF#A before resolution (at least it sounded best that way in my tests).

EDIT: No, you're absolutely right, if you hold the G all the way through, it works well. the G is consonant to the C, dissonant to the D, and then consonant to the G... I know there's a word for that but I have no idea what it is. I want to say pedal note but that's in the bass. But it's not a retardation since it doesn't RESOLVE, it holds... I know there's a word for this though.