Curt Siffert

A few years ago, back around the time that Kanye West interrupted Taylor Swift, there was another bit of news in the music media. Coldplay had come out with a song, Joe Satriani heard it, and felt pretty devastated by what he heard.

The suit was eventually settled and dismissed. The rumors I've seen were that Satriani got a cash payment from Coldplay, but that's hard to confirm. Coldplay didn't admit wrongdoing.

But what was interesting about it was that the argument mostly centered on the chord progression! The chord progression in question was VI-VII-III-i , or IV-VII-III-i. (VI and IV are pretty similar and serve the same function due to common tones.)

That chord progression seems rather distinctive - a far cry from the cliché four chord songs everyone likes to joke about. Right?

Right?

Well... let's take another look. Both of those chords end up on the minor "one". And minor keys have relative major keys. What would those chord progressions be in the relative major? IV... V... I... vi. And, ii... V... I... vi. That's pretty much the "four chords" right there. The latter one is basically every jazz turnaround ever at the end of a standard. The only twist here is that they chose to label the chords in terms of the relative minor. And that's arguably wrong of them, anyway. Coldplay's song is in a major key the entire way through (not a V/vi to be found), whereas you could argue that Satriani is minor in the verse, and modulating to the major in the chorus before it goes back to the relative minor.

To me, the similarities are more convincing because of tempo, groove, and the key parts of the melody - the melodic rhythm in particular. I think that's where the strength of Satriani's claim was, supported by the chord progression - not in the chord progression itself.

However, I suspect that in trials such as those, they don't look at it in terms of each particular ingredient - they just play it for the jury and ask them if it sounds "too" similar. I'm also not surprised this lawsuit turned into a settlement, because if Coldplay had been found liable based off of a combination of factors, it would have opened them up to a ton more lawsuits from everyone else who happened to write four-chord songs that they believed sounded similar. (Which almost happened with Cat Stevens anyway, due to his ii-V-I-IV progression...)

I found this plugin at Noteflight today and am as pleased as punch. I might find myself using it quite a bit on this site.

Here's a favorite chord of mine that seems to sneak into a lot of my music. One good example: Before A Kiss.

(Note, if my music is playing in the background, hit pause in the player in my right sidebar. :) )

As good as I am at recognizing patterns, it's humbling when you miss some! I went into my most recent lesson feeling like I knew my octatonic/half-dim scale. Tony calls it a symmetric diminished scale.

So, he asked me to play it in C. Then he asked me to play it in F. Then he asked me to play it in G. I figured he was just testing me in some random keys.

It became clear later that the point of asking for those three are that those are the only scales I really need to know. All the rest of them are just the same scales as the same as those three because of the symmetry of the scale. Eb, F#, and A are the same as C. Ab, B, and D are the same as F. Etc.

He showed me some sequences then too - any pattern of notes in that scale can become a sequence of playing the same intervals lower in the scale. I probably just need to see some more relationships in the scales and then I'll be able to start putting those sequences together in real time.

One of the interesting things about learning jazz piano, and jazz theory in general, is how all the theory starts to feel inter-related after a while.

Most jazz piano books, and Mark Levine's is no exception, seem to take the tack of throwing a bunch of data at you and hoping it sticks. I've had a little trouble with that so far. For instance, I put a lot of pressure on myself to start trying to memorize jazz standards about eight months ago, and I got discouraged really fast. I just felt like I was trying to drill myself and none of it was sticking. I was missing glue.

I've learned a lot about learning in my days of memorizing classical piano pieces. The main thing is that you can't memorize something in only one way. If you memorize simply by muscle memory, then it requires a huge amount of focus to stay in the brainspace to rely on only that when you're performing. It's the same with memorizing by ear - what if you have a brain lapse and forget how something is supposed to "go" next? But if you can memorize it in five different ways, you've got redudancy - you've always got a fallback. I would always try to memorize in as many ways as possible - muscle memory by playing the piece on a table. Ear, by playing through the whole piece in my mind when falling asleep in bed. Visually by trying to photographically memorize the score, and also by visually memorizing what the piano keys looked like - and finally, analytically with harmonic relations, how the chords changed from a theory perspective.

Basically, what matters is not really the data, but how linked it is. And that is what makes jazz theory so difficult for beginners, and then so much more weirdly easy once you really get going. What matters is how much glue you have.

I had a few brainbending moments yesterday while practicing. For a while I've been stuck in the phase of improvising diatonically, with a few blues scales mixed in. But as soon as it would get into things like the more interesting scales one could play for the more interesting chords, I wouldn't know how to link that together in my head.

Two good examples are the octactonic scale and the alt scale.

I've known the octatonic scale (aka the half-dim scale) for a long time, from a theoretical perspective. Half-step, whole-step, half-step, whole-step. But that bit of knowledge is basically meaningless to me. It's like all I could do was either struggle through playing the mental pattern (while the visual pattern is so different from key to key), or drill myself to learn it with muscle memory. I didn't see any other way to look at it other than it being two of the three dim-7 chords clustered together - and I don't make a practice of looking at scales as clustered chords while I'm trying to play scales, so that wasn't helpful. And some people define the half-dim scale as starting with the whole step, which just confuses the issue.

So this week my teacher (Tony Pacini) got me working on quartal chord voicings, and one of the most common ones is the dom7 voicing that starts on the 7 and goes up by tritones and fourths. In F: Eb, A, D, G, C, F. And we were also talking about tritone substitution, and how basically any voicing of dom7 chord works over the root or the note a tritone away from the root. And he also made the point that you really only have to think about the octatonic scale that starts with the half-step.

So here were the brainbending moments:

• While I don't know the half-dim/octatonic scale, I do know my modes. The two most common are Lydian (raised fourth; film scoring, Grieg, John Williams), and Mixolydian (lowered seventh; it's all over the place). If you combine them, you have a lydian-mixolydian scale, with both a raised 4 and a lowered 7. I've always loved that sound. While playing with the half-dim scale that starts with the half-step, it started to look and sound familiar. It turns out that the half-dim scale is just a lydian-mixolydian scale, with a crunched up beginning (instead of a 2, you play the notes on either side of the 2). And that made it click. All of a sudden, I can now play it in every key.
• That lydian-mixolydian scale shows up elsewhere, too. Take the above dom7 voicing and crunch it down to within one octave, over the tritone root (B natural). B, C, D, Eb, F, G, A, B. What's that? That's the alt scale. What is that over its original root? F, G, A, B, C, D, Eb, F. The B isn't actually in there, but you have to use a #11. That's the lydian-mixolydian scale. Yes, in other words, the alt scale is always the lydian-mixolydian scale, starting on the tritone.
• I liked looking at the half-dim scale as being like the lydian-mixolydian, but with a crunched-up beginning. So what's another way of looking at the alt scale? It turns out it is like a whole-tone scale, except with a crunched-up beginning. Again, instead of the 2, it's the notes on either side. It's like it's half a half-dim scale, and then half a whole-tone scale.

I guess that was a productive few minutes when that all clicked together for me.

One of those things that is cool about youtube. They've got people out there that are rigging video cameras up above their pianos, and demonstrating how to play jazz piano. This guy took it to the point of actually annotating his performance.

It's interesting to note how much of purist different performers are about jazz. People that don't play jazz don't always understand what it means to improvise. A friend was recently dumbfounded to realize that I play my piano solos differently every time, perhaps thinking that all jazz performances were pre-written. Others see it on the other end of the spectrum - that improvisation is purely inspired, akin to being psychic or touched by God. The truth is it's somewhere in between. (Okay... maybe not as close to the "touched by God" end...)

But here it's interesting to note that the performer is very definitely planning certain licks for certain points. The prepared bass part has a riff at about 1:57 that the piano mimics - he's clearly planning to play that part then. And when combos rehearse tunes for performance, it's common for instrumentalists to work out favorite licks to the point where a solo section becomes a bit more structured and prepared; very different than what would come out at a fakebook jam. It's more prepared than I'd want to be for a gig.

I still get stuck in wanting to know what the "right" way to do it is - if I'm improvising "too much" or relying "too much" on licks I already know. I think the truth is there just isn't a right (or wrong) answer.