Author Archives: sjt

Why Transcribe So Much?

Why do you transcribe so much music? Isn’t that just stealing? Can’t you write your own stuff?

Think of this way… a jazz player will memorize solos. All musicians practice. Orchestral players memorize famous solo passages and repertoire. So what do composers and arrangers do?

That’s right, we transcribe music by ear. Put it down on paper. If you, don’t you should. Let me explain.

Why Not Just Read Scores?

The goal of composing for large ensembles like this, is to get the sound that’s inside your head down on paper so it can be performed.

You can go a long way with studying scores, and you need to do it, but it’s not enough.

First reason is availability. Most modern scores are simply not published, or if they are, it’s probably something that everyone already knows anyway. What if you want to learn how an unpublished composer puts things together?

Second is you need to really win the knowledge for yourself. It’s sort of like trying to play a jazz solo from written music. Music is not the notation, the music is the sound. If you just snatch out lines, harmonies, and voicings from other scores, it won’t sound right when you go to use it. The voice leading won’t be right, the orchestration won’t be consistent.

The best way is to transcribe and keep doing it, all the time. It’s practicing. Want to get even better? Transcribe something by ear, and then check the score afterward to see how close you got.

The more you do it, the easier writing and arranging gets, because now all these sounds are in your head. They’re part of you.

OK, Strings Are Fun Too

Here’s a little preview of a new project for Lauren.


Writing for strings is pretty damn cool too, given their extreme range.

I’m doing two orchestrations for this. One for the studio, where we can overdub in woodwind parts and not have to deal with instrument changes. A second for live performance, where physical reality constrains the orchestration.

The second one is a lot harder given that I need to put the strings somewhere. Dropping them entirely would destroy the charm and grandeur of the piece, so this is a fun puzzle to solve.

For the live version, I’m adapting these soaring lines into the saxophones and rhythm. I’m undecided though, maybe it will be better replacing the saxophone with a piccolo, 2 flutes, and two clarinets to capture all these high notes.

You might notice I have about four or five projects and post threads going on at the same time. That’s pretty much how I work… inspiration can’t be scheduled!

Reversing the band adaptation (1)

Since we have a “young band” score to start with, the next step to restoring the full brilliance of the orchestral score, is finding the differences between the two and seeing if there is any pattern to them. We can do this with visual inspection and a cursory listening, and then a more detailed listen.

Instrumentation differences

Rather obvious is the orchestration. No strings, but we do have the saxophone family. I’m also reasonably sure there is no baritone (a “small tuba”), alto clarinet (seriously, how many bands really have these?), and probably the bass clarinet. When writing the band score, John had to find out where to put the string parts… either omit them, or mimic them in the winds, saxophones, percussion, or combinations thereof. He also had find “something to do” for the other instruments even if the color is not needed.

Don’t forget that this was conceived as a vehicle for Whitney, who has enough vocal firepower to project over an entire orchestra. This adaptation doubles the melody everywhere, so, it looks like it’s intended for use without vocalist. Or, a student vocalist who needs the melody doubling to counteract any limitations in technique.

Most, if not all, of this melody doubling is unnecessary. The places where John does so in the original are then color choices. The excess doubling then detracts from the orchestral colors. The jazz-influenced middle section (11-18), is particularly apt here. The soft string, horn. and wind colors are just harmony and countermelody, which lets Whitney use a softer, more intimate tone. This also sets up the contrast for the dramatic “B” section to follow.

Second, concert bands usually have multiple people on a single part, which is rarely true for wind ensemble, jazz bands, and orchestras. So we have to be mindful that the band version is written for that, and adjust backwards accordingly. A unison line on 8 flutes sounds very different than two flutes in octaves.

Ability differences

The main thing with student editions is instrument ranges, as the extreme ends of each instrument are more difficult. We put up with all that difficultly (French horn players can relate here) since they also have the most personality and interesting colors.

Examples: the flute never goes higher than a Db, which is clearly to avoid the complicated altissimo fingerings. But the flute has another usable octave on top of that. Similarly, the top fifth of the oboe, bottom fifth of the bassoon, are missing. The clarinet is missing both the top and bottom octaves due to its wide range.

Lines are either moved around to other instruments to cover the range where it matters, changed octaves, reharmonized into an easier range. That’s just the woodwinds; it goes on for every other instrument. The main point is this is lot of color we need to restore.

Fourth, rhythmic changes and simplifications. The main snare pattern is offset by a sixteenth-note to make it easier to play and for the band to come in correctly. Trumpets replace 32nd notes with sixteenth to avoid double-tonguing. Flurries of 32nd notes on the winds are reduced to glissandos, which can be faked. We can restore all of these those and let the players figure it out.


Writing for woodwinds is always my favorite part.

This excerpt is 2:05 in the video below – the “rockets read glare” part, highlighting the top vocal note.

As I said earlier, simplicity. Here’s a quickie analysis of what’s going on:

  • Bars 1 & 3 are closed-voiced tremolos of the chords: A♭/E♭ A♭sus⁴/E♭, then D♭m⁶. This gives the shimmer and energy behind the brass chords.
  • Bars 2 & 4 are written-out diatonic glissandos, voiced in thirds, over the chords D♭6/9#11 and C⁷sus♭9. This gives an epic sweeping feel, again over the brass.
  • Bar 5-6 starts with the most important chord of the entire piece (“proof”): Fmin(maj⁷), followed by a unison statement of the melody (“through the night”).
  • Bar 7 is drop-two harmony (“that our flag was still”) since the range is so high, for the B♭9sus and Bb13♭9 chords. In this range, gives a “sweeter” sound than close harmony would. And, it’s more idiomatic for the string parts being mimicked.
  • Bar 8 is my favorite, the flurry of 3-part diatonic chords, with the lead doubled an octave down. This propels the entire bar of otherwise nothing but a held chord, right into measure 9 (“D”, which is the “say” in “oh say does that…”).

All of this is really hard to hear in the recording, and it’s worth digging out. It’s what puts all the energy behind the fairly simple melody we all know, love, and maybe hate sometimes. Without it, you just have a blarrrrr of loud held notes, as gorgeous as those rich chords are.

Project: John Clayton, Jr.’s “Star Spangled Banner”

One of the most important pieces of music in my life is John’s arrangement of SSB. I’ll spare you my story of why this is so — but suffice to say is that this is the only good rendition of SSB ever arranged. Really, brings tear to my eyes.

Whitney Houston dials in a performance that’s equal parts technical brilliance, personality, and reverence. By reverence, I mean it is not this:

How not to sing the national anthem. Ahem, Mariah.

A few months ago I was thrilled to discover that it was for sale. Legally. Arranged by John himself for concert band. I quickly ordered the entire set, but when it arrived, I was disappointed to learning it was a “Young Band” stripped down arrangement. (There is also a marching band arrangement, which is helpfully “not available”. Come on, take the damn page down, stop being jerks.)

I could write a few articles of analysis on this piece alone – all 35 bars of it – and what makes it so majestic. But let’s get to the arranging part now.

Selecting the Orchestration

Score or not, I’d have to reduce it down to a smaller band orchestration. Just starting with a real score, however, gives me quite a leg up, and saves a lot of time with the initial pass. This particular piece is extra challenging for accuracy, as the only recording is the open-air Super Bowl performance. In 1991. No studio edition. No alternate takes. (Well, almost none.)

A quick listen to the original sounds to be a typical orchestra – strings (violin, viola, cello, bass; brass: 3 trumpets, 3 trombones, 4 horns, tuba; winds: 2 flute (doubling piccolo), 2 clarinet, oboe, bassoon; percussion: snare, bass drum, timpani, crash cymbal, harp (yes!!), and maybe just maybe – is that a marimba at the end?

Our band is 5 reeds, 5 trumpets, 4 trombones, keyboard, bass, guitar, drums. We can work with that, given competent doublers on the reed parts. Ain’t no saxophones in the original.

Reading the Score

A cursory reading of the score while listening is a little disappointing, as it has been heavily modified for beginner players. Still, having the score is a gold-mine, as I can listen to what does and doesn’t match, note them down, and work from there.

Transcribing large-scale music by ear is a like solving a puzzle. Nobody can hear absolutely every part in isolation, but you can hear the lead parts, the bass, selected lines here and there, the tessitura, the rough type of voicing (octaves? close parallel block chords? drop-2? 4-part vocal?), and so on.

But key is the the snippets that you can hear – oh, that’s a clarinet playing and F and Eb over here, that’s the French horns playing this phrase over there, there’s the trumpets playing 32nd notes… you get the picture. Those are like the “easy” puzzle pieces that can only exist in one spot on the board, unlike the vast piles of ambiguous gradient background. I mark every single one of these “keystones” down.

This, by the way, is a continuous process – every time you listen to it with different headphones, EQ settings, and so on, you’ll hear different bits. So you have to be ready to move those puzzles pieces around and solve the problem. This is learning, you’ll get better each time.

Or you can skip all that, and your arrangement will sound… meh. It will “work”, if you’re competent enough an arranger. But it won’t sparkle or resonate. And you won’t get any better.

The rest is just… when you get right down to it, filling the blanks.

I forget where I read this, but if the music you are working on seems chaotic and complicated, it’s because you haven’t discovered the simplicity of it yet. There always is some underlying framework, you just have to find it.

Coming next: the strategy for this particular score.

Still here

It’s been a busy summer actually playing music instead of writing about it. As we ease back into winter, I’ll be doing a lot more posting.

Also, I finally have a business name for all my music services: No Rehab Music.

Why the name? Dunno. I’m messed up in many ways, and probably would fail out rehab.

A month or two after coming up with the name, it dawned on me that’s it’s also a tribute to one of my favorite singers. Totally unconscious decision, until I spent a semi-lucid evening listening to the amazing breakout Frank.

RIP Amy. You still are influencing generations of singers and writers.

Transcribing a Big Band Chart

(ed: originally January 2009. Comments for 2021 follow.

At the time of writing this, John’s chart wasn’t published, as pretty much nothing was back then then. Happily, over the years John has published many of his excellent arrangement. Go buy them and support music, directly to the creator if you can.

If you compare scores, you’ll see I came pretty damn close. But the reality is without access to unmixed recordings, you’ll never get every single note in every single inner part. There are things you simply can’t hear, and too many ways to voice things that sound identical. You have to settle with something that gives the sound and effect of voicing. If a tree falls in the woods, and all that. I’ve learned over the years the point is the sound, not the notes.

That said, I’ve gotten a lot better and faster in the 10+ years of doing this. Wait till you see my Thad Jones and Kenny Wheeler score and analysis!

I’m tempted to rewrite all of this, but I’ll restrict it to minor clarifications.)

Part I: Introduction

Musician friends! I thought some of you might be interested in seeing a big band chart put together in real-time for a working band. Though I do my own arrangements, I also like to transcribe tunes from other arrangers. Not only do I get a lesson in their writing style that you just can’t get from looking at scores, I also get a good chart that nobody else has. Good vocal tunes are hard to find. None are published.

I find transcribing large ensemble tunes by ear is a totally different ballgame than doing small group charts. You really have to know all your voicings, and be able to identify them by ear. 4-part block, 5-part drop-2 and 4, basic ensemble, triads with melody 8vb, clusters, etc. There’s just no way to pick out every single instrument’s line in the mix.

OK, the chart. It is I Love Being Here With You arranged by incredible John Clayton, Jr., and sung by Queen Latifah. I actually started work on this in October, but I only spent an evening or two on it, and it’s been idle ever since. You are witnessing me transcribe it in real time.

The first step, obviously, is to listen to it a bunch of times and love it. So that’s what we’ll do. Go ahead, put it on your iPod (ed: wow, so 2009) and listen!

I chose this tune, thanks to our trumpet player Peter, who is a great scout for new tunes. It’s a good fit for our band. It’s also a style that our singer Britt will nail to the wall. Finally, I like it. It’s John Clayton, and has a ton of interesting stuff in it that I want to see how it works. So, rule #1: write for the band you have now. Rule #2: you gotta like it.

Next up: the tools I use, and the rough sketch.

I’m back?

Apologies for the delay, to all my zero readers. Just rummaged through my defunct Facebook page for some old content which I’ll be releasing soon.

LFCC Progress

It’s been about a month since I started LFCC, and I’m still only part way through phase 1 – which is interval-based diatonic scales with directional changes. I’m nearing completion of just the first page, and in major only. That’s 14 exercises x 12 major keys, going from q=80 up to at least q=160, often up to q=200. It would be a bit faster if I stopped at 160.

The rest of just phase one is the above in minor (1 more month) and then both in 4ths, 5ths, 6ths, 7ths, and octaves. Each interval group is 4 exercises, so this roughly 5 months at the current pace.

Adam suggests 6 keys per week (3 in major and minor) and says “in one direction”… which is unclear. Is that one exercise (up, down, up/down, down/up) or all four?

In order of difficulty, I would say they are:

  • #1 – diatonic scale (12345678)
  • #14 – 4 note segments, up/down (1234 5432 3…)
  • #18 – 4 note segments, down/up (4321 2345 6…)
  • #8 – 4 note segments, up (1234 2345…)
  • #10 – 4 note segments, down (4321 5432…)
  • #4, #22 (dup) – thirds up (1324 3546..)
  • #26 – thirds up/down (1342 3564…)
  • #26 – thirds down/up (3124 5346…)
  • #24 – thirds down (3142 5364…)

Note that due to the direction changes, some of the “thirds” exercises are mixtures of thirds and seconds (22, 26, 28), while others are thirds and fourths (24).

I’m not sticking to 6 keys per week at the moment, but rather going as quickly as possible. For the easy keys, I can often knock off 6 keys (e.g., FCGDAE) for one or two exercises in a single session. For the harder keys, 2 exercises (e.g, Db/F#) in one key is more like it.

Some Critique

I have to guess exactly what phase each exercise is in, rather it being explicit. Each exercise should be marked with a phase number. I believe Phase 1 is pages 1-3, measures 1-67.

The log sheets are organized one page per exercise, iterating over 12 keys. I think it’s better to have one key per key, with only 2-4 exercises on it. After doing an exercise in one key, it’s easier to stay in that key for a while than jump around.

Similarly, I think it’s easier to do all the majors first, and then the minors; or, pair the majors with their relative minors (C major with A minor) as they share much of same notes. The former is more challenging since it gets into F# and B much faster, but it’s helping break bad habits faster.