Gentlemen, I really want to thank you for describing this mediant/tri-tone harmonic movement thing. That's very cool. I've looked at moving major and minor triads around before, but always struggled to make it sound good. Restricting the movement to thirds and tritones seems to be the ticket. I don't know why I never realized it before.

In case anybody still isn't getting it, here's a simple experiment I just did that blew my mind. First write a simple little melody of, say 4 bars. Pick a triad that fits the first couple notes. Then change to a new triad that fits the next couple notes, making sure to move the triad root only by a major or minor third or tri-tone, up or down. Keep going until the end of the phrase. Then revoice the chords for optimal voice leading.

Say the melody is Auld Lang Syne: C F E F A G F G. We can put the first three notes over Dmin. Then the next two will go under Bbmin (pretend it's a harmonic minor scale for the A). Then we have a problem for the next two notes: G-F. We don't want to go back to Dmin. Dmaj won't work, F#maj and F#min don't work (implies a scale including F-F#-G, and no such diatonic scale exists). Or we could jump by a tritone from Bbmin to Emin, but all of the options from there are rather dull. Bbmin to Emaj works, but it's a little too dissonant for the melody, and I don't like the options it leaves for the last note, G.

I think the best option is to go to Gmaj and then end on Ebmaj. It's not great, but it does have a nice atmospheric feel to it. And it sounds even better if you start with a melody that isn't strictly diatonic. There are other solutions to Auld Lang Syne, of course, but this demonstrates the concept. Play the chords as a pad in cellos and horns and give the melody to an oboe, and you've got a film cue.

Now I'm playing with fitting odd bass lines under the whole thing. Thanks again, this is really useful stuff.

With a little more thought, I figured out why it has to be thirds and tritones.

The goal is to provide lots of harmonic motion without ever resolving. Triads are used because they have nice transparent orchestral voicings. There are other options like stacks of fourths but they have different motion rules. Diminished and augmented triads also don't work, but I haven't worked out their rules yet.

Anyway, major and minor triads can't move by fourths or fifths because it sounds too cadential, even if you mix up the majors and minors in unusual ways. You can't use minor second motion for two reasons: poor voice leading and the implication of tritone substituted cadences. Major seconds work slightly better as they don't feel cadential, but they will often feel like you're setting up a cadence as in a IV-V or vi-V type motion. They will work in a pinch, but they invariably suffer from voice leading problems.

However, movement by major or minor thirds always has good voice leading and it sounds non-diatonic as long as you avoid things like moving a major triad up by a major third to a minor triad. That and minor triad moving up by a minor third to a major triad both sound static for obvious reasons.

It's obvious when you think about it. I don't know why I never noticed before.

@kevind said:

Diminished and augmented triads also don't work, but I haven't worked out their rules yet.

Sorry, that should have been "dim and aug triads also *should* work, but I haven't worked out their rules yet."

...

...

@PaulR said:

Just a couple of points. When one uses these triads through 3rd intervals -must they be triads -or can they be inversions under the rules. Also, the root may be, let's say, C - E - G - B, but I guess you could use different bass notes?Also, I guess you could use as many intervals of a third as you wished - and then at any time you could go to the diminished 5th from the last chord whenever you liked?Paul

Paul,

The chords can absolutely be inversions. In fact, most of them probably WOULD end up being inversions to make the progression flow, and avoid that 'blocky' sound I mentioned. (Sorry, couldn't think of a better word for it.)

Different bass notes? Yep. And you can use the diminished fifth movement anywhere in the progression it feels right...not only at the end. (I just used that as an example.) Although, in fooling around I find that it does put a dramatic ending on a phrase.

I wrote a piece this week for a documentary about Matisse and Picasso which had this long expository section I was having trouble finding an idea for. I ending up writing this little harp obligatto with two notes moving in contrary motion, the bottom part playing 8th notes moving from root to 2nd of the chord, the top part moving 5th to 3rd. So the bottom part on Cmaj is C-D-C-D, etc. Top part is G-E-G-E, etc. Then I took this through a long series of mediant chord changes, added a pizzicato bass part, then tossed in the occasional woodwind line when there wasn't narration. It was very simple. I finished the scene in no time...and it just worked. The producer was trilled...commenting on how concerned he was that the scene had dragged a little, and how lively the music made it. Now THAT's a practical application!

I had been exposed to the mediant concept before, but the way Scott explained it, it just clicked. Or like I told him...'exploded' would be more accurate.

Can you tell I'm having fun with this?

Fred Story

I have worked these theories to no end, some point at which it became innate to write in suspension, and that is why I have been preaching that PolyTonality is the next step in film music. Not Atonality, but strong firm Polytonaly. Myself, I am an advocate of Bitonality and Binary forms and structures in film music as I believe it to be the purist form of emotional tug and pull. You can hear an example of over an hour of this kind of music in my best score, Hunting Humans, at:
http://www.cafepress.com/evanevans

In that score there are passages of 3 to 5 minutes without resolution, but with complete form and power.

Evan Evans

...

Hey PaulR,

There's a great book called "20th Century Harmony" by Vincent Persichetti, that you might enjoy checking out. It's basically just an overview of a whole bunch of different approaches to harmony that appeared during the past century. Polytonality is covered in there, along with many other techniques. It doesn't go into great detail, but it presents a lot of material in a very concise way, and is a brilliant way to set your harmonic imagination in motion.

Unfortunately, I gave my copy away to a musician who did me a big favor. It was a good gift, but I really miss that book sometimes!

J.

...

Quavers, yup...

I tend to feel the same way about textbooks. But because this one doesn't dwell too long on each idea, or go into extreme detail, it tends to read more like a reference. It's just enough to wet the apetite for exploration, without getting you bogged down in lenthy explanations. I (used to) just open it up at a random page, check out what it had to say, then start messing around on the piano, listening to how the ideas sounded, and experimenting with the structures they provided.

Talking about again, I think I'll try to find another copy myself!

cheers,

J.

Wonderful discussion!!!!

Many thanks.

Rob

Fred,
would it be fair to ask you to say something about that intriguing "Mediant Scale" of Scott's you mentioned? Given that, once this journey in 3rds steps begins, we've soon said a swift goodbye to any "Tonic" note?

I've been working on it but so far I'm stumped by all the possible combinations when changing chords.

Yes, quite inspiring - I like Evan's term "writing in suspension", cause that's what it's really all about. The human condition, bittersweet......

back to triads! I love it!

Nigel

Did I say that? Yeah I like that too.

[;)]

If you want orchestration knowledge and coaching from industry professionals, try http://www.thinkspaceonline.com I did their course and the tutoring was outstanding. My tutor was Robin Hoffmann who works with Russell Watson and loads of 'big names'.