Toolkit: Tritone substitutions

The orthodox meaning of tritone substitution is replacing a V7 chord with the dominant seventh chord half an octave away.  This works because the two notes that imply a dominant seventh chord–third and seventh–match between these two chords.  If you play B and F together, it’s a jarring interval and they want to resolve to C and E.  In other words, those two notes imply a G7 chord, especially in a C major context.  However, you could call them F and Cb and resolve them in the opposite direction, to Gb and Bb, implying a Db7 chord.  Tritone substitution is using one in place of the other in a V7-I cadence.  A variation is to use the V7 and go to the substitution for an intermediate measure before resolving to the I, especially on turnarounds.  To help establish the tonality, I put a ii chord in these example cadences: Dmin7-G7-C; Dmin7-Db7-C; Dmin7-G7-Db7-C.

What’s interesting to me is that a typical chord resolves to something a half step away from a tritone.  G resolves to C or D, either side of the tritone Db.  Put the other way, if you want to get from Db to C, substituting a chord a tritone away gets you there.  Of course, if you just pound three chords in a row, Db-G-C, it’s an awfully high contrast of outside and consonance.  However, if you play Db and C chords but play a melody of G scale to C, it’s still outside but it flows.  If you play a melody that starts in a Db scale and drifts through a G scale into C, you get a nice mix of savory chords and sweet resolution.

I do both in chromatic progressions.  In something like “A Night in Tunisia” that goes back and forth between e minor and F, I’m willing to do a Bb arpeggio or scale over the e minor on one measure somewhere in the late part of the phrase, resolving to the F and into the cadence.  However, in a modal piece like “So What” or “Little Sunflower”, that’s entirely too angular.  Instead I’ll work in the long measures of e minor more and more flatted fifths and major sevenths, and coming up on the F chord, start resolving those to the four and seven instead of to the tonic chord.  Over e minor, play patterns of notes like Bb-A-G-E or D#-E-F#-E; then put the D# and D together in some sort of E-D#-D descending line to start suggesting an Eb-D resolution; then alter the motifs to include an F natural, then start leaving out the G in the first one to highlight the Bb-A resolution; and I’m already comfortably in F before the chord changes.

It’s either that or play up the sudden chord change after sixteen bars of pure e minor.  I figure the head has defined that well enough, so I’m trying to draw smooth lines that blur that boundary into one continuous musical thought.  At any rate, that’s what I’m going for.

Toolkit: Sight-transposing

I learned this exceptionally useful skill early on, and sometimes I take it for granted.  From ages 13-18, I’d roll out of bed every Sunday morning and play the church service cold, not even looking to see what the hymns were until the prelude was underway.  Practice makes second nature, and that’s served me well at many a gig.

Several different people tried to teach me transposition by several different methods.  For those not in the know, a standard Bb trumpet (or clarinet, tenor/soprano sax, treble clef baritone, etc) has to play a note one whole step above the note you want to hear; it’s just the way music is written (a topic for another day).  Other instruments transpose different amounts or not at all.

One suggestion: read the treble clef parts as though they were tenor clef and adjust the key signature.  That’s clever, except that I learned to read tenor clef by pretending it was treble and reverse-transposing.  It might work for someone else who learned tenor clef first, but I can go years without seeing any kind of K clef so it really doesn’t help me.  (Eb saxophone/bari sax players can pretend they’re reading bass clef, with the same assumption.)

Another suggestion: count two half steps before playing each note.  Variation 1: count a major second before playing each note, which means constantly shifting your key center.  No thanks.

Third suggestion, which is how I started: adjust the key signature, then play the note right above the written note (space above the line, line above the space).  It takes surprisingly little practice before it just clicks, and sometimes I habitually do this when the part is already transposed for me.  How embarrassing.

Nobody suggested this, but it’s what I eventually started doing: Translate everything into scale degrees (or solfeggio), then translate that into whatever key you need.  This sounds like doing it the hard way because it’s a two-step process, but again, after a while it clicks.  (Of course, you have to know your scales and keys very well.)  It doesn’t work for everything; anything in an unusual scale (basically not major, modal, or harmonic minor), anything with a lot of accidentals, or anything atonal pushes me back to reading the note above the written, at least for a measure or two.  However, this approach lets me transpose to any key, which means I can read off a sax part or allow the accompanist to choose a key (after negotiation, because let’s face it, some keys are better for the instrument).  It also means I can adjust if someone tells me the wrong key, an accompanist forgets to remove a guitar capo or MIDI transpose, or a singer/choir drifts out of pitch.  Perhaps best of all, it means I can use the extension on the piccolo to put me in a key that’s friendlier to the fingers and has better intonation.

What I didn’t mention is that this technique also requires you to be able to put music into and out of your memory on the fly.  Remind me to tell you some time about sightreading from memory.

Posted in Toolkit. 1 Comment »

Toolkit: dim to 7b9

A progression that comes up often is a diminished chord followed by a dominant 7b9 a fourth up.  It’s a form of ii-V, sometimes in an ending cadence or turnaround.

The easy way to get through it is to treat them as the same chord.  They’re the same notes, excepting the bass.  Technically, they imply different scales (though the pairing defines the scale of the 7b9), and it matters a good bit what comes before and after; but there’s nothing wrong with keeping it simple and playing with small arpeggios from time to time.

“Stella by Starlight” is full of these, as is “Dear Old Stockholm”.  My weakness here is a temptation to play the same lick each time they come around, essentially doing a turn of root, b9, root, 7 of the second chord.  Sometimes I’ll do an arpeggio of the diminished chord with a half-step trill on each note.  Chromatic slides along the diminished arpeggio work too.  (I’m using a fully diminished chord, four notes each a minor third apart.)

If I have enough of these gestures ready to hand, I can pick one that fits in whatever larger phrase I’m constructing.  This is a very fluid chord change, and I’ve been able to internalize it pretty well, so I’m fairly graceful with it.  (And therefore I like “Stella” and “Stockholm”.)  It helps that there are only three fully diminished chords (spelling can cause you to see a lot of different 7b9s).

Taking it farther, Mark Levine (The Jazz Theory Book) points out that Fmin-maj9, Gsusb9, Abmaj7#5, Bb7#11, Dm7b5, and E7Alt are in fact all the same chord–unless you’re the bass player!

Toolkit: Playing the Words

“I can’t improvise on the melody without knowing the lyrics first.” — Ben Webster.

“When [Billie Holiday] sang it was like she was telling you things from her life, not remembering words to songs a songwriter wrote. Amazing. I learned a lot about the stories of songs from Ben Webster. He once told me, “I can’t improvise on the melody without knowing the lyrics first.””  — John Levy, in an interview on JazzWax

Here’s the thing: most of these standards we play have lyrics, and that’s an integral part of the song that you don’t get from a horn.  That means you have the opportunity to say something different, perhaps something beyond vocal language, but it also means that a melody from a good songwriting team is crafted to the lyrics.  Maybe the music says the same thing as the words, but you can’t count on everybody getting the full message that way.  Maybe it just provides a context, but that means you have to communicate the message.  Maybe the audience knows the words, and you just have to play the head in a way that doesn’t interfere with their recollection of the lyrics.

But when you’re ready to do your thing, and you’ve got a bunch of tools on your belt, you need to start talking to the audience.  Sometimes I practice improv by singing spontaneous beat poetry.  When I know a song’s lyric well, part of the time I’m mentally singing words I’ve made up as I play solo phrases.  So, I might start a chorus of “Misty” with the first three notes of the melody, in time, and go from there: “Look at me, on a branch, up a tree, my paws are grabbin’ at a robin,” pause for a beat, “It’s a long way down and I’d land on my head.  I see birds, I see stars, in your eyes…”

I’m staying on the theme, and I’m thinking in terms of imagery and sentence structure but not rhyme.  Melodic rhyme is not quite the same as language rhyme.  It’s actually more like rhyming in sign language, which is based on visual symmetry.

As with many other techniques, most people aren’t going to be getting every literal thought.  Not everything in my head gets communicated with perfect clarity, and frankly I’m grateful for that.  Still, you know whether you’ve heard someone recite a poem or just ramble at the microphone, and that’s the difference I’m cultivating here.

There are practical applications as well, especially for a directional instrument like trumpet.  I’ve managed, in the middle of a solo, to communicate to a waiter that he should leave that drink where it is because it’s mine and I’m not done with it.  Hey!  You!  Walk away, fella!  ‘s cool.

Toolkit: Beyond the Melody

Why not just play the melody?

Well, your audience has heard it.  Sure, they like to hear it again, but that’s what the head is for.  You can develop a solo off the melody.  However, there are a lot of jazz players in the world; other musicians and the audience as well have heard that too.

If you follow what people in the examples did by starting with the melody and developing from there, when you go to an open jam you probably want to go straight to the fourth or fifth chorus, skipping the first few on the assumption that everyone there has already covered that ground.

However, after years of this, you start to think in terms of larger and larger musical sentences.  The truth is, there are only so many classes of motifs, and eventually everything is going to remind you of some combination of things you’ve heard before.  Just as a child progresses from syllables to words to clichés to forms, a longtime jazz player is interested more in when you say something than what you say, how you construct the sentence more than what words you use.  OK, you played a fast scale; it might or might not be particularly impressive from a technical perspective, and it’s neither good nor bad that we’ve all heard it before; the question is, what does it mean at that point in the song, over that chord, after what you just played–and what are you going to play next that might expand its meaning?

Try it first by quoting other songs during your solos.  It’s cute, but what’s the difference between a random phrase and something that makes the audience laugh?

Toolkit: Motif

Usually I have the first solo chorus in the quintet.  We try to put some overall structure into the song as a group, with a form incorporating all our solos together.  That sort of thing doesn’t have a detailed schematic, but we do think about it.  I often stay close to the melody or do something simple; that leaves room for the others to stretch their legs, and it provides a transition for the audience.

One of my tricks is to take a three- or four-note motif from the first bar of the melody and work it.  It’s relatively easy to do, and it preserves some sense of it being the same song.  For example, the song “Four” starts with five three-note phrases: three notes ascending three times (the same notes), three notes descending, and three notes ascending into a chord change.  Each is three eighth notes followed by an eighth rest.  So I’ll start with three notes ascending and go from there.  I can find three notes of a scale on any chord, so I’m free to think more in terms of the solo’s broader shape.  I’ll do several ascending, then instead of a descending set do a big jump to a low note and do more ascending.  That has the same effect of breaking the pattern.  I’ll leave out the eighth rest between two sets, then put longer and longer rests between the next few.  Then I’ll extend it and do sets of five instead of three (five makes a lot more rhythmic opportunities than four).  By then it’s pretty much time to wrap it up.  (Using a pattern this way is called a sequence.)

Now, I don’t know how much I buy into the idea that good composition techniques have a big effect on the quality perceived by the listener.  What does matter is that I’m confident in what I’m playing and can spare attention for structuring the chorus, looking at the audience, hitting some accents with the drums, or otherwise focus on the music beyond the technique.

When I don’t have the first chorus, I usually take something from the solo before mine.  That’s particularly effective when we trade off in shorter increments.  Listen to how these guys play off each other: Printup, Gransden and Gunn playing “Birdlike” at around 8:00 on this YouTube.

Posted in Toolkit. Tags: . Leave a Comment »

Toolkit: What’s Wrong With the Melody?

A guitarist I knew used to wonder why players would whirl off into rainstorms of eighth notes whenever it came time for a solo.  “What’s wrong with the melody?” he would ask.

Actually, in a vocal band I usually play the melody straight if I have a solo.  Sure, I’ll decorate a bit, but the audience is either dancing or talking, and in both cases they want things to stay in line.  In a jam band, however, they already heard me play the melody on the head.  There’s an unwritten rule that you don’t repeat the whole form, but rather offer something new to say.

Still, it’s a good starting point.  If you’re new to improv, or if you just can’t get a handle on a song, play bits of the melody.  Take some liberties, insert something if you get an idea, build on it.  Do sexy music theory stuff like retrograde inversions.  Or just stay pretty close; at least you know it’ll fit, and we’re trying to play music, not impress anybody.

I was having trouble with St. Thomas, and this guy laid it out: GelZero on YouTube playing St. Thomas.  I don’t know who he is, but the dude knows how it’s done.

But to the original point, I want to hear a player put something of themselves into it, but I want to hear that from a player that is able to do the song as it is.

Posted in Toolkit. Tags: , . 1 Comment »