FAR-23 defines the stall as the point at which the aircraft departs controlled flight or the pitch control is on the stop. Therefore by definition, the aircraft can stall at any weight and CG condition - it's just that the characteristics of the stall may change.
2D - you missed my argument a bit. I was not arguing that all pilots should be taking an aircraft to a fully developed power-on stall (although I'm as-yet unconvinced it actually does any harm in a sorted aircraft). No, my argument was that a power-on stall can be achieved inadvertently by a very experienced pilot (and thus probably an inexperienced pilot too), and that it's clues and warnings are not identical to those of of an idle stall. Therefore a pilot should be made fully aware of what it looks like and how to recover safely from it. I wouldn't have a problem with that involving just taking the aircraft to incipient if the stall characteristics are particularly nasty.
In that context, I think that the FAA instructor was quite right to refuse to sign off our original poster if he couldn't deal with the power-on stall - although it might have been more polite to turn the sortie into a bit of instruction in the subject and then do a second checkride a little later. I'd particularly bear in mind that this was a 152, which takes some pretty gross mishandling to spin - even from a power-on stall.
G