My question is about playing upper neighbors to embellish basic chord tone motion.

Lower neighbors like this...

enter image description here

...seem to pose no problems. Just use half steps below and the neighbor sounds like a leading tone and it never outlines a tritone

But the upper neighbor seem more complicated. All the diatonic upper neighbors seem fine except on the ii chord...

enter image description here

enter image description here

...where a tritone is outlined. In minor this doesn't seem a concern, because that tritone is just outlining all chord tones on both iim7♭5 and imin6.

Only A B A F on the ii chord in major seem like a potential problem.

If the lower chromatic neighbor acts like a leading tone, solfege TI, then the upper neighbor seems analogous to solfege FA. If you play all the diatonic thirds with an upper neighbor, they outline perfect fourths with the upper neighbor like FA, except the one that outlines a tritone.

For a jazz player, would the "basic" approach put a flat on it? Like, A Bb A F over ii in C major.

I understand that both diatonic and chromatic lower/upper neighbors are all part of the jazz vocabulary and can be used depending on context and style. My question is just about a "basic" diatonic approach.


In this particular example you might consider that maintaining the sequence over-rides the incongruous tritone. Or you might, indeed, flatten the upper note. Or you might consider the leading note to be an 'avoid' in a ii7 chord. Or you might like the vaguely Lydian sound of the B♮ (a Lydian ♯4th is becoming as much a jazz staple as a 'blue' ♭3rd used to be). Your call. The main point is that you're aware of the issue and listening to the effect of what you play.


My take on this is the B upper neighbor is not a great choice, not so much because of the tritone with the F but because it telegraphs the B in the G7 in the next bar and the C to B is the main difference between those two chords in terms of voice leading. If the B wasn’t a pivot note it wouldn’t have that effect. The A is the only note on a Dm7 ii chord that doesn’t work well with an upper neighbor. I would suggest slightly breaking the pattern and using the neighbor two houses down instead, A-C-A-F.

I’d say the Bb is out of play since you are looking for a diatonic approach but I’d pick it over the B sonically speaking if I had to choose.


In major, the B-natural over the ii chord I like a lot and would keep, but I would definitely avoid the Bb.

I tested these by first improvising over the ii-V-I, establishing C major in my ear. Then I moved to the melody patterns proposed. The B gives a very appealing "out" sound to the ii chord, especially in first inversion, but is clearly within the key of C and functioning as an upper neighbor.

Bb however, sounds quite bad to my ear. It's obviously out of key and clashes with the F in particular. My sense is that the ii chord operates to the ear a bit like a IV6 chord -- that is, as though we've shifted temporarily to F major. (This remains true for me regardless the inversion). Put another way, I agree with the interpretation that the Bb serves as FA against the F in the chord.

Jazz players learn very early on to avoid (read: be very, very careful with) FA in a major key.

  • Am I reading you correctly? "...Bb serves as FA against the F in the chord..." and that is uncharacteristic of jazz? That "out" sound, would that also be the lydian/whole tone sound some like with a #4 over a major chord? Like Thelonious Monk. – Michael Curtis Dec 23 '20 at 5:26
  • @MichaelCurtis Generally speaking, the perfect fourth above the root of a major chord (including 7th chords built on a major triad) sounds pretty bad. Jazz players learn to use it carefully. (It's an excruciatingly common rookie error to sit on, say, an F over a C7 in a C blues. Not that I never did that, of course.) But the #4, yes, offers a sort of Lydian or whole-tone sound, depending on how it's used. In your exercise, I'd put it more in the Lydian category. – Aaron Dec 23 '20 at 5:33
  • Part of the reason for my question is the teaching of "avoid note" and the surely related "rookie error" to sit on a P4 and not hearing it isn't a chord tone. "Avoid" as in "avoid because it's one that can't be justified as a chord tone" versus "carefully exploit non chord tone dissonances for effect" IMO is the difference between good and bad teaching. In other words, lower/upper & diatonic/chromatic neighbors are all valid in jazz, but being clueless about what your chord tones are is not. Right? – Michael Curtis Dec 23 '20 at 14:54
  • @MichaelCurtis - I think the good v bad teaching issue has a lot more to do with 'use your ears' rather than get bogged down in the theoretical use of a particular note. Either work - it depends a lot on where in the bar the 'suspect' note is played. There is a great danger in carrying theory - any theory - too far into playing.More important maybe to make sure a student is aware that actually either note is fair game, given circumstances. I certainly wouldn't want to send a student off believing one works and the other doesn't - cos that's not always the case. – Tim Dec 23 '20 at 16:25
  • @Tim, but that is my point. "Avoid note" just send you off thinking it's one that doesn't work with no explanation of how it can work in various contexts and why. The theoretical question here is whether an outlined tritone, where the tritone is not chord tones, should be avoided? How can you be aware if that's fair game with some music theory to explain why? – Michael Curtis Dec 23 '20 at 16:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.