Thanks for the replies, I see what you are saying about the perils of thinking about infinite span aerofoils. Maybe I should re phrase the question:
Is the entire back end of the power curve completely a function of induced drag, ie, backwards tilting of the lift vector due to downwash, or is it also due to the increase of pressure and parasite drag with increasing angle of attack?
as bookworm said:
If you have a look at the drag polars of aerofoils (rather than finite wings), you still see a roughly quadratic dependence on lift coefficient (i.e. on AoA). For cambered wings, the minimum drag is not at zero lift coefficient. But that's not induced drag.
J