George,
the paradigm shift Kuhn describes is what I mean by radical change: New observations cummulating to a point where the whole structure tumbles over, giving birth to something completely new. But I think between this radical changes there's a lot of finetuning going on.
And, of course, can a radical new theory not be understood from the old point of view. That's why it's radical new in the first place.
As for tendig to view the history of sciences as linear: Same problem with evolution. Looking back, there seems to be a direct line of progression from the first amoebas right to us humans (Well, ok, some dead branches included.) That's the equivalent to Kuhns "view prior work in the context of the current paradigm."
Changing the perspective to looking from the past gives a whole different result: At a specific point in the past no one could have judged which species would take the price and pass the torch on to its descendants.
So, as I see it, Kuhn, you and me agree on the point of radical change/paragdigm change as crucial for scientific development.
BTW, George, you never really said if YOU agreed with Kuhns view 
Bye,
Jens
