>>From Alan Young
> Reed's C7 in m. 1 is most enlightening! But I still don't much like any of
> the solutions to the change in m. 4-5. Bm7 works better than G after the
F#
> chord, though, to my ear. And I think, Reed, that you meant to write F#
> there, not F?
No, I'm pretty sure Reed meant F7 ! This is not a ii-V-i pattern to the Bm7,
but rather a #IVm7b5->IVm6->I with the I being Gmaj7, or a plagal cadence
with a preparation chord. If you place a F#7 there, then it would go nicely
to the Bm7, but imo that's not what is intended
Not to beat on a dead chord, but the F7 is the backdoor dominant, that could
also be notated Cm6/F.
Pedro