the same thing has happened 
  with jazz.  gotta hand it to bach: he pretty much wrote the book on 
  "music theory" without meaning to do so.
    
   ** 
  don't you think that bach picked up a lot from other folks and then just 
  made it his? he seemed to have at least borrowed from people like buxtehude 
  and vivaldi, no? 
 
   
  >most people 
  seem to start with a melody that would lead the harmony as it 
  goes<
  
this is not always the case with traditional jazz. often, 
  composer x will start with harmonically interesting chord changes (think giant 
  steps, countdown, etc.) and the melody comes next.  it makes the most 
  sense this way.  since the harmonic progression (changes) will be the 
  basis on which composer/musician x will be improvising, the chords therefore 
  have more significance than the melody and thus come first.  charlie 
  parker wrote new melodies over old tunes, keeping the chord progression 
  intact...although this has just as much to do with not wanting to pay 
  royalties on tunes he played/recorded.
  
** well, i did say 
  *most* - - not *all*. in my opinion, for parker the tunes were more a means to 
  blow, not necessarily for being themselves; it was nice that they were good 
  tunes, but not really necessary. a guy like ornette, on the other hand, 
  doesn't really use chords - - neither did miles in some of his more modal 
  moments. i think they're thinking more in *tonal areas* or some such. art 
  pepper didn't know anything about chord theory for quite a while, it was all 
  by ear. people who really play well through "rhythm" changes are probably not 
  really thinking about the chords at all. but, i guess my original point is 
  that the melody better really sing or the harmonic hipness may be of no real 
  import. a real synergy is to be hoped for. 
   
  stig