I know this is an old thread. But, I have only one thing to add that might fix everything. This will explain Jazz, Classical, Rock ‘n’ Roll, and all other types of music…
Read Frank Zappa’s autobiography “The Real Frank Zappa Book.”
After you have read it all things will be clear.
Since that book isn’t in my personal or nearby public library (although I have been a Zappa fan), can you condense or review it for us?
Nitpick: The “70% of all music sold in the US was once jazz” statistic is pretty bogus. It assumes that anything played by a big band or with a swing beat in the 30s and 40s was jazz, but that just ain’t so. It’s like calling every pop record today that uses any electronic instruments “electronica.”
Wumpus,
Just out of curiosity, how would you classify something with a swing beat played by a big band then? That sounds like jazz to me. Sure, they might be playing anything from a bossa nova to a Steely Dan arrangement but those wouldn’t have the “swing beat”. To me, if it’s swingin’ and there are horns, it is probably jazz. I wouldn’t call every pop record with an electronic instrument “electronica”, but I would call it pop. Could you clarify?
musicguy, I’d classify it as swing music or big band music. But not necessarily jazz. Swing is actually a broader category than jazz.
Suppose you were look at people who described themselves as hardcore jazz fans in the late 30s and early 40s–the sort of people who wrote newletters and made discographies. If you were to ask these people “Does Glenn Miller make swing records?” they would have answered, “Sure.” But if you were to ask, “Does Glenn Miller make jazz records?” they’d say, “Ack, no, no, no, that’s not really jazz.”
Today we tend to think of swing as a subcategory of jazz, but when swing was at its peak, it was the other way around. People considered jazz as a subdivision of the massive pop music sensation known as swing. There were swing records, which included Glenn Miller, and the Andrew Sisters, and Kay Kaiser. Then as a subcategory of that there were jazz records, which tended to appeal to a more select audience of rabid fans who looked down on Glenn Miller as mere pop.
Of course, some artists like Benny Goodman and Duke Ellington appealed to both the mass swing audience and the jazz snobs. That’s the legitimate basis for saying jazz was more popular in the late 30s/early 40s that it’s ever been since. But not all swing records were jazz records.
(Incidentally, you may need to rethink your definition of jazz. There’s lots of music that swings and involves horns that isn’t jazz–like, for example, much of the Rhythm and Blues recorded between the end of WW2 and the beginning of the rock era. For that matter, a whole lot of early pre-Ringo Starr/Keith Moon/John Bonham rock’n’roll swung and involved horns. Think “Rock Around the Clock.” On the flip side, there’s lots of jazz that doesn’t swing–like all jazz made before the swing era for starters, not to mention a hefty percentage recorded since the free jazz era began in the 60s.)
Wumpus
Thanks for your response. The Glenn Miller example defintely alters my perspective and is causing me to rethink a few things. Must ponder this for a bit while I listen to some Coltrane 
What factors would you say are unique to jazz, yet not swing as a whole? I’m thinking that improvisation is probably the main one.
Of course, by todays radio stations playing “smooth jazz”, it muddies up peoples idea of what jazz is even more. It would appear that anything that is instrumental is considered jazz now. Thats a bit disheartening to me.