iim7 options for diminished substitution

[Ioscommuter]

I am familiar with substituting a diminished 7 chord a half step higher for a Dom7. For example replacing G7 with G#, B, D, F and that each degree could be considered the root. I am wondering if this means that there are 4 different m7 chords that could function as the ii preceding the V that we are substituting for? For example: dm7-G#dim7, fm7-Bdim7, abm7-Ddim7, bmin7-Fdim7? I am assuming here that all 4 would have the flat9 in place of what would be the root if it were a dominant chord? If this is a thing, are there examples in songs someone could point out? Also it should be possible to resolve to 4 different tonic notes. Can the "ii" be inconsistant with the "I"? For example bmin7-G#dim7- Cmaj7?

[KirkP]

I may be over my head here, but after experimenting with the idea I concluded that the answer depends on what key (or chord or scale) you want this turnaround to take you to.
For example, if the third chord will be C, Dmi7 obviously works as the first chord, since it's the ii. Fmi7 also sounds good as the first chord, since it's the iv.

(edit) Abmin7 and Bmin7 don't sound great to me as the first chord if the target is C. But as I experiment it seems like one could get away with them if each chord is voiced such that the dissonant tones walk to the key of C.

Bottom line, if your target is the key of C this diminished chord is functioning as a G7b9, no matter how you name it.

[JonR]

I am familiar with substituting a diminished 7 chord a half step higher for a Dom7. For example replacing G7 with G#, B, D, F and that each degree could be considered the root. I am wondering if this means that there are 4 different m7 chords that could function as the ii preceding the V that we are substituting for? For example: dm7-G#dim7, fm7-Bdim7, abm7-Ddim7, bmin7-Fdim7? I am assuming here that all 4 would have the flat9 in place of what would be the root if it were a dominant chord? If this is a thing, are there examples in songs someone could point out? Also it should be possible to resolve to 4 different tonic notes. Can the "ii" be inconsistant with the "I"? For example bmin7-G#dim7- Cmaj7?

I agree this ought to be possible, but what you're really doing is invoking other dom7s for which the dim7 could be standing for - regardless of the eventual tonic.

Remember the dim7 is actually the leading tone chord (vii) of the key - borrowed from minor, but common enough in major. Eg, Dm7-G#dim7-Cmaj7 should strictly be called Dm7-Bdim7-Cmaj7, regardless of what the bass note on the dim7 is. (Think of it as built from the 3rd of G7 if you like, but really the viio7 is a chord in its own right.)

So, to choose Fm7 means the following Bdim7 will suggest Ddim7, the vii chord of Eb (standing for Bb7). So a following Cmaj7 - although it may work - will sound like a deceptive cadence.
In fact, in jazz you do often see Fm7-Bb7-Cmaj7 - the so-called "backdoor" progression - but in that case the Bb7 is a lydian dominant chord - for which a dim7 is not a suitable sub (it might have 9th and 13th extensions, not compatible with Ddim7). IOW, the Bb7 is an extension of Fm7, forming something more like a minor plagal cadence (iv-I) than a perfect cadence (V-I).

The other options work as follows:
Abm7-Ddim7 = Abm7-Fdim7 (rootless Db7b9), so you expect Gb or Gbm (F#m) to follow.
Bm7-Fdim7 = Abm7-G#dim7 (rootless E7b9), so you expect A or Am (F#m) to follow.

However, I do think all three work reasonably well! - they sound fine to me going to C - as long as the chords are voiced so the voice-leading works naturally. Deceptive cadences are, after all, what a lot of jazz changes are all about!

[czardas]

I have followed this topic with interest, but decided to wait and see what others had to say about it before I responded. I am no expert in chord substitutions: which seems something of a rather large open book to me.

From my perspective, all these chord progressions should work but they may not be suitable as substitutions. I think of such chord sequences as mini-modulations with the dominant as the pivot chord. This is great for creating interesting and unusual voice movement when writing original melodies. If the melody is already set, then options for accompaniment will be more limited.

In some cases there will be a degree of commonality between the two keys: allowing a different interpretation of the subdominant. Under these conditions, chances are that such a chord substitution might work quite well: I haven't spent a great deal of time analysing this. Consider that the diminished chord (vii dim7) resolves in eight different directions: that is when you include the relative minor key resolutions. That's a few more sequences you can try out. You can also try backcycling - substituting the dominant with several mini-modulations before resolving to the actual key.

[Ioscommuter]

Thanks Czardas,
for an example of relative minor resolutions would that be
bmin7-G#dim7- Amin7 in place of Cmaj7?

[czardas]

Yes and similar:

Bmin7b5 ... G#dim7 ... Amin6 (sounds Latin )

[JonR]

Thanks Czardas,
for an example of relative minor resolutions would that be
bmin7-G#dim7- Amin7 in place of Cmaj7?

Right, because G#dim7 is vii (leading tone chord) in key of A minor.
Call it Bdim7 and it's vii in C (C minor originally, but often borrowed for C major).
So the same chord could also go to Cm and A major .

Two other target keys would be common - respelling the dim7 as the other two possible leading tone chords:
Ddim7 > Ebmaj or Ebm
Fdim7 (E#dim7) > Gb(F#)maj) or F#m

Less common resolutions would be the common-tone diminished move - i.e., to any tonic whose root is one of the notes in the dim7 - so 4 options there, more likely major than minor. (Not sure if this is how czardas counts his 8 moves?)
This move is more common in the middle of a sequence (like an embellishment of, or chromatic approach to the tonic) than as a final cadence.

[czardas]

I was thinking purely of vii-i in both major and minor as producing 8 possible resolutions - for any diminished chord. The other sequence you mention is interesting: I'm not sure how to interpret it. Consider this sequence:

Bdim7 ... C7 ... F

If you omit C7, you are left with exactly the sequence you describe (Fdim7 ... F). Including the C7 doesn't appear to significantly change the overall movement or impression. To me there seems to be a relationship between these two ideas, albeit a slightly tenuous one (perhaps). The diminished chord could plausibly be considered a secondary dominant with the resolution having been omitted (like taking a shortcut to the final F chord). This is probably not the only way to think of it and I would be very interested to hear anyone else's interpretation.

[JonR]

I was thinking purely of vii-i in both major and minor as producing 8 possible resolutions - for any diminished chord. The other sequence you mention is interesting: I'm not sure how to interpret it. Consider this sequence:

Bdim7 ... C7 ... F

If you omit C7, you are left with exactly the sequence you describe (Fdim7 ... F). Including the C7 doesn't appear to significantly change the overall movement or impression.

I think it does. It makes Bdim7 into the vii of C. Rootless G7b9 if you like. (vii/V in relation to F).
Omitting the C7 obviously changes it, and (IMO) forces us to hear the Bdim as a cto.

[czardas]

What's a cto?

The dominant C7 can be squeezed between in any case. It should work under most normal circumstances. That being said, the sequence fits this description. Chords can be omitted from cadences: IV - V - I ==> IV - I.

This way of thinking has always served me well when I compose music: add a few chords and remove a few.

Edit
I have a little more time now to elaborate on my thought processes. We start with a simple cadence V7 - I.

C7 ... F

Next we add a secondary dominant to introduce the dominant (VIIdim7) - V7 - I.

Bdim7 ... C7 ... F

This is a well known sequence. Any of these chords can be omitted. If you bypass C7; you are still going to (and coming from) the same place. In essence the overall movement does not change. What actually changes is the path you take. C7 is nothing more than a slight detour which adds momentum (figuratively speaking). I see this treatment as more of a theoretical device capable of creating much variety. We can insert, substitute or remove chords from standard sequences in a variety of situations. The resulting sequences will always be theoretically connected to the original sequence.

For me this approach often validates chord sequences which I find hard to justify otherwise. When provided with additional justification, it is always a bonus. BTW, theoretical justification is never a prerequisite for a chord sequence to work in actual practice.

I often marvel at some of the ideas used in jazz, however I stopped thinking purely in terms of modes a while ago: partly because it seems a bit like using tricks (although I probably ought to revisit them). Don't get me wrong, I think they create some wonderful effects, but I tend to feel a little constrained and always want to break out of the box. Thinking melodically is where it's at.

[JonR]

What's a cto?

Sorry, "common tone diminished".

The dominant C7 can be squeezed between in any case. It should work under most normal circumstances. That being said, the sequence fits this description. Chords can be omitted from cadences: IV - V - I ==> IV - I. [/quote]Well, exactly, it's a different cadence!

IOW, its not about what we can and can't do, but about what we call it when we do it. E.g., the difference between a functional substitute, and a replacement that does a different job.

Edit
I have a little more time now to elaborate on my thought processes. We start with a simple cadence V7 - I.

C7 ... F

Next we add a secondary dominant to introduce the dominant (VIIdim7) - V7 - I.

Bdim7 ... C7 ... F

This is a well known sequence. Any of these chords can be omitted. If you bypass C7; you are still going to (and coming from) the same place. In essence the overall movement does not change. What actually changes is the path you take. C7 is nothing more than a slight detour which adds momentum (figuratively speaking). I see this treatment as more of a theoretical device capable of creating much variety. We can insert, substitute or remove chords from standard sequences in a variety of situations. The resulting sequences will always be theoretically connected to the original sequence.

I see what you're saying, but I still think we need different terminology here.
In particular, I think the C7 is more than a "slight detour". Bdim7-F sounds different enough from Bdim7-C7-F to be described differently. In particular, the function of the Bdim7 is different. In the first case, it doesn't work as vii/V, because (a) the V isn't there, and (b) Bdim7-F works as a common-tone diminished, direct to F, quite a different thing from how it moves to C7.
None of this is about limitations on what we can do! Just defining terms in the most useful and consistent ways we can.

For me this approach often validates chord sequences which I find hard to justify otherwise. When provided with additional justification, it is always a bonus. BTW, theoretical justification is never a prerequisite for a chord sequence to work in actual practice.

Sure. Theory is not "justification" for music in any sense. It's vice versa. The sounds justify themselves, and theory is only the names we give them. Where we can distinguish different sounds, or different effects, it makes sense to have different terms, even where those sounds are closely related.

I often marvel at some of the ideas used in jazz, however I stopped thinking purely in terms of modes a while ago: partly because it seems a bit like using tricks (although I probably ought to revisit them). Don't get me wrong, I think they create some wonderful effects, but I tend to feel a little constrained and always want to break out of the box. Thinking melodically is where it's at.

Right! I was lucky not to have discovered modes or chord-scale theory until I'd been playing music (including some kinds of jazz) for 25 years. I always played melodically (as much as I could), and discovering jazz theory didn't stop me!

[czardas]

Well, exactly, it's a different cadence!

Okay I'll grant you that: the second Amen is ecclesiastical and the first one's kind of more Gospel.

its not about what we can and can't do, but about what we call it when we do it.

A sweeping statement.

I see what you're saying, but I still think we need different terminology here.

I have the tendancy to try and fit pieces of the puzzle together. I see my cadential argument (thank you for reading it) as merely adding weight to support the viability of the sequence. I vaguely remember seeing the term 'common tone diminished' before. The fact that the sequence has a special name suggests there ought to be some theory as to why it is considered important.

(a) the V isn't there,

I don't see this as a problem: since it can be included (or perhaps implied) most of the time - I think.

Where we can distinguish different sounds, or different effects, it makes sense to have different terms, even where those sounds are closely related.

Well I thank you once again for updating me on the terminology. There are some pretty big holes in my theoretical knowledge.

Diminished chords seem to have so many intriguing properties. They also connect to each other chromatically, so you can go practically anywhere you want using them. Minor seventh chords share a similar trait: with parallel movement being possible. I wish I could explain it. Still, what works works. I have just been having some fun with Ary Baroso's Brazil. His music is full of unusual progressions.

I appreciate your responces JonR. You put forward some intelligent counter-arguments to these musings of mine.

[TLerch]

I love Diminished harmony and feel it is an area that can lead to some great sounds. Here is my two cents worth. G7 is related to Bb7, Db7 and E7 via the diminished cycle. All four chords relate to Bdim7 which is one of the 3 dim7 chords related to the key of C. (the three are vii dim i dim and bii dim)
Each Dom7 then can have a "companion Minor7" and depending on voice leading each pair can resolve to C major.
So it's
ii V I
iv bVII I (so called "back door" ii V)
bvi bII I ( so called "Wes" ii V)
and vii III I. (least common and almost always resolves to vi)
you can also mix and match (again with good voice leading) for instance ii bII I or even ii iv bVII bII I if you really want to stretch it out. Of course you can also resolve to vi via the same pathways.

there are two other dim7 chords related to the key of C, bii dim7 which functions similarly as above but resolves to ii and IV and the other is i dim7 which resolves to I (CTD), iii, and very often ii especially if its in first inversion.
example Ebdim7- D-7
There are other secondary resolutions (for instance i dim7 can also act like V of V) or (vii dim7 can act like CTD of IV)
its fun to go thru the "Real Book" or other collection and try to identify any dim7 as one of the above types.
all the best
Tim

[czardas]

There's some good information there Tim. Thanks for sharing.

[TLerch]

You are welcome, its my pleasure. Such a fun subject, there is a lot more too it but i'm sure the good minds in this group will turn over many stones.
Tim

[Irez87]

Barry said that all horn players should know how to play in and out of Bb7, Db7, E7, and G7. I guess if we wanna cop some horn licks... Try this:

1. Go up a Bb dominant scale to the b7

2. The b7 becomes the "5" of the Db7, go down to the b7 of Db dominant scale

3. The b7 of the Db7 becomes the "5" of the E7, go up to the b7 of E dominant scale

4. The b7 of the E7 becomes the "3rd" of the Bb7

[JonR]

Barry said that all horn players should know how to play in and out of Bb7, Db7, E7, and G7. I guess if we wanna cop some horn licks... Try this:

1. Go up a Bb dominant scale to the b7

2. The b7 becomes the "5" of the Db7, go down to the b7 of Db dominant scale

3. The b7 of the Db7 becomes the "5" of the E7, go up to the b7 of E dominant scale

4. The b7 of the E7 becomes the "3rd" of the Bb7

Yes, that's the dim7 link. The same dim7 chord will substitute for any of those four dom7s (it's clearer when you add the b9).

To summarise:

There are THREE possible ways a dim7 can move - and only three (because of the symmetry of the chord).

1. When one note of the dim7 moves up a half-step to the root of the following chord.
This is the most common use, known as "vii" or leading tone chord. The natural derivation is from the vii degree of harmonic minor, but it can be used to resolve to a major chord, and in a secondary function too.
E.g., Bdim7 can go to C or Cm, whatever key the C or Cm is found in. (And the C can have a b7 or maj7 too)
This also means that if you see Fdim7, Ddim7 or G#dim7 going to C, they should really be called Bdim7, regardless of what the bass note is.*
This is also commonly seen as a sub for the V7 of the following chord, but vii is a chord in its own right.

2. When one note of the dim7 is the same as the root of the following chord.
This is the "common-tone diminished" (cto), and usually resolves to a major chord.

3. When one note of the dim7 moves down a half-step to the root of the following chord.
I'm not sure if there is one accepted term for this, but I call it a "chromatic dim7" (not totally satisfactory, because most dim7s are chromatic!) or passing dim7 (also a bit vague).
The most common occasion is between two min7s a whole step apart. You see this in Jobim quite a lot, and there are two in "Night and Day".
You could see it, in those instances, as the vii of the preceding chord.
Eg Fm7 - Edim7 - Ebm7. Edim7 sounds a bit like vii of Fm, or a rootless C7b9 (function #1 above). But in relation to Ebm7, it's function #3.

* Although any dim7 can have four names, looking for the voice-leading (nearest note to the root of the following chord) tells you which of these 3 moves is in play, and therefore which is the most logical name for the chord. (Even so, it can be handy in chord symbols to name the dim7 after the desired bass note, regardless of its function.)

[czardas]

I made a mistake above, thinking that Jobim wrote this, but it was written by Ary Baroso. The error has been rectified.