Sunday, January 29, 2012

Tracking Down Stella .... By Starlight

   The 1944 Movie Uninvited introduced Victor Young's classic tune  Stella By Starlight.  Sixty years later the tune is still considered a fixture in the Jazz Musicians' repertoire and, as is customary in the Jazz World, it has had 60 years of speeding up and 60 years of natural selection, slowly morphing into a totally different animal.  The video below, with Ray Milland and Ruth Hussey at the helm,  is the movie feature of the tune, and you will notice right off, that the song is a ballad, and if you're a musician, you will notice that it starts of with a Diminished 7 Chord and not a min7(b5) Chord as is now customary.  Those are 2 pretty different qualities of harmony, even though they differ by only a single note.    But I'm not too concerned about that, since the replacement chord comfortably fits the up-tempo style that people have applied to this tune.  And, I must say, many people have applied themselves to this tune - about 800 versions pop up in Rhapsody!

STELLA BY STARLIGHT    from UNINVITED   1944

 
  But,  I really wanted to do some detective work to figure out the confusion behind the Real Book I published version of this tune and the awkward chord progression that has become defacto Standard on this tune.  I remember this tune from the 50's, as a kid, and how I loved the chord progression ( some things I guess you're born with!).  What I remember hearing are the several spots in the tune where there's a Suspension in the melody against a Major Chord  and, simultaneously a 5th in the bass.  Yes, most of my friends we outside playing tackle football while I was doing this, and staying safe.    At the time I didn't know what that sound really was, but I just remember the sound being absolutely captivating.  That wasn't something you heard every day -- neither was hearing  a song that STARTED with a Diminished 7th chord.  (The other song I remember starting that way was Gershwin's "Our Love is Hear To Stay", but we've also blown off THAT diminished chord  in recent times.)
  In time, though, that unique "HOOK" that was built into the tune ( major 3 AND sus4 AND 5th in the bass), gradually disappeared.  I say "gradually" because I did hear a number of early recordings where the HOOK was still there.. case in point, the Anita O'Day version at the very bottom of this post).  In the abbreviated Chord Chart below, those 2 "hook" spots are marked in PINK. The original chords, taken from the movie score, clearly show this  tension-resolution moment,  built into the song.  Fake books like "Real Book 1" tend to simplify-down the harmony by leaving out important aspects of original compositions, important aspects such as  a Fifth or Third or a Seventh in the bass, and that could explain why one never hears this tune played with much adherence to the original chords - musicians tend to get their chord knowledge from fake books and NOT from old recordings.  And can you blame them?... the song is from a totally different era, after all.  Someday there should be a discussion about this philosophy of using old movie songs as jazz "standards" and "blowing over changes" and the merits of doing so.  But I digress.  

   The SECOND  element of confusion on this song happens in bars13-16 (YELLOW).  The original chords have a really nice descending bass line with a SURPRISE Diminished 7 chord along the way.  It's exactly what I remember listening to the radio back then when the tune was popular with the mainstream culture.
  Here's where some detective-work has paid off.  Miles Davis recorded the song in 1958 and 1964, about 15 years after the original movie,  and supplanted the original chords with the ones you see below in GREEN.  They are actually very effective, as you might expect from Miles,  as long as you blow off the Fifth in the Bass on F chord that leads into them.  They work really well --- and the Bass progression weaves a nice logical and elegant path.   The Real Book version ( in BLACK), which came AFTER Miles recorded his version,  is a bit of a bastard child of the Miles Davis chord changes and the original chord changes, taking one from one and one from another.   It just doesn't work. It goes to the Em7(b5) chord earlier than did the original chords but then it just comes out of it in a really dumb way.  The Bass line suffers and the progress of tension and release  is ambiguous.   I guess that's about as well as I can describe it.  Many versions on YouTube are worth hearing if you can sift through the fluff.   Check out Anita O'Day at the bottom of the post.  Despite all my rantings, Stella By Starlight is still a great tune.

  MILES DAVIS  GREEN        REAL BOOK 1   BLACK       ORIGINAL SCORE   RED


(DOUBLE CLICK TO ENLARGE)
Stella By Starlight FIRST 16 BARS

 

5 comments:

  1. I would not say the real book changes are at all simplified, taking a diminished chord and replacing it with a dominant chord or ii-V progression is a very common bebop practice and is the opposite of simplification. As for calling the real book changes a bastardized version of the original and Miles is wrong; both Bill Evans and Stan Getz play the "real book" changes far before the real book is out. To call the real book changes bad and say the other changes "work" just seems silly. The changes presented by the real book are pretty hip and by no means incorrect or bad at all. Trying to trace how changes over standards have evolved is a cool thing to do but do your homework first.

    ReplyDelete
    Replies
    1. Hey Doc, I am in total disagreement with you that replacing the diminished chord with the dominant or ii-V progression is not simplification. Quite the opposite: It is. The reason I say this is that jazz players have a habit of thinking certain ways, and that is one of them. Jazzers reduce everything to stock structures so it will fit their practiced vocabulary, and they don't have to think in new or unfamiliar terms when they improvise on chord changes. And because one person used a particular set of chord changes, and then that set of chord changes happened to be published in the "real" book, it does not mean they are "the" changes.

      Delete
  2. Hey Doc, I am in total disagreement with you that replacing the diminished chord with the dominant or ii-V progression is not simplification. Quite the opposite: It is. The reason I say this is that jazz players have a habit of thinking certain ways, and that is one of them. Jazzers reduce everything to stock structures so it will fit their practiced vocabulary, and they don't have to think in new or unfamiliar terms when they improvise on chord changes. And because one person used a particular set of chord changes, and then that set of chord changes happened to be published in the "real" book, it does not mean they are "the" changes.

    ReplyDelete
    Replies
    1. Taking something original and replacing it with the same chords used in every other song - I'd call that dumbing it down. Jazz musicians can handle ii-V7 in their sleep with their hands tied behind their backs.

      Delete
  3. Hi Steve,
    Loved the post on Stella.


    RE: MEASURE 14
    Try this on for size

    G7 C7 C/Bb
    / / / /
    of which
    Bm7b5 Bbºma7
    / / / /
    is a variant (or vice-versa)

    One can see how closely it relates to the original if one merely precedes the ORIGINAL Bbº with the Bø7 (on beat 1-2 with the ORIGINAL Bbº following on beats 3-4.
    We both agree that the Em7b5 is just plain 'bad'.

    The
    Bm7b5 Bbºma7

    has the benefit of voice leading well from the preceding F bar (especially if one were to play the typical descending bass line in quarter notes F-E-D-C) AND the Bbºma7 is much better getting to the Aø7 in the following measure.

    ReplyDelete

Please leave comments here!