Also known as “why transcription isn’t just for solos,” and “why you should always learn the tune from the record.” I’ve recently been digging into Jobim’s music (particularly the incredible “Elis & Tom” record) and trying to really play his music correctly. Why? 1) Because his tunes have beautiful, simple-sounding melodies that often disguise complex harmony underneath. 2) Because he’s the master of both chromatic harmony and diminished chords (which are related). 3) Because he’s written so many enduring standards. And 4) Because we so often play his tunes WRONG. And “Triste” is a great example of all these aspects. I’ve been playing this tune wrong for years and wanted to share my findings from comparing 3 Jobim recordings of “Triste” – from Wave (1967), Elis & Tom (1974), and Sinatra & Company (1971). Even though Elis Regina sings it in G, I’ve put all the changes in A Major for ease of comparison, as the other two recordings are in A. The first page of the document shows the changes from “Elis & Tom,” the second page shows the changes from “Wave” and the Sinatra record, as there is only one substantial difference, and the third page shows the standard “bad” changes (view the changes in the embedded Scribd document below).
There are a few things that immediately jump out – firstly, I’m not sure why the “standard” key is Bb Major, as I couldn’t find a Jobim recording in Bb. Second, and this is a big one, in all three recordings, the form is 32 bars – I couldn’t find a single instance of the 34 bar form with the 4-bar “vamp” at the end. It’s yet another example of why I wish the original Real Book authors would have listed their source recordings.
The next big thing to notice is in bars 27-28 of the form (the second to last phrase of the tune). The “bad” real book changes have the phrase resolving from iii-7 to VI7 to II7. Dominant II sounds strange here, and is a strange resolution point for Jobim. Jobim’s progression actually goes from iii-7 straight to biiidim7 which is a much more typical Jobim sound, and it leads really nicely into the ii-7 chord at the beginning of the last phrase.
Lastly, check out the chord on the downbeat of measure 15. People tend to play this chord as a iii-7, as if it is part of a iii-7 VI7 ii-7 V7 leading into the second half of the tune. This way, you have two bars (measures 13-14) in IIIMaj, and then two bars leading you back to IMaj. But it’s actually a IIIMaj7 – I think of it as still belonging to the previous two bars in IIIMaj, so you end up with three bars in IIIMaj with a quick turnaround back to I. It’s a small difference, but it’s important because that’s how Jobim wrote the tune and it’s indicative of a problem I was discussing with another pianist earlier today. It seems like there is a gradual dumbing down of harmony to make tunes easier to play over. The typical improviser has more “stuff” to play over a iii-7 VI7 ii-7 V7, so it’s “easier” to play over, but that’s not how the tune was written. We as improvising musicians need to strive to be as true to the intent of the composer as possible, regardless of the difficulty of the harmonic progression. After all, I would think very few of us got into music, and jazz especially, because it’s “easy.”
Although the original Real Book is known for having all sorts of inaccuracies, I must give credit to Sher Music’s “New Real Book Vol 1”, as their lead sheet for this tune address all the inaccuracies from the original Real Book (which are also present in the iReal Pro changes).
One quick last thing – the biggest difference between the Sinatra version and the version from Wave is in bar 23 of the form. Jobim (or the arranger?) added a quick little chord substitution to III7alt which is a refreshing sound. It sounds close to the iii-7 played in bar 7 of the form (the same spot but half a chorus earlier), but just different enough to make you listen twice. Check it out at about 1:02 in: