Wow, there's a little thought-grenade for the "that's not falsifiable" crowd...I had to read his source article to really (well, and very basically) appreciate what he was getting at.
I do agree with this idea; once you are that insistent on testable stuff, once you see it as _the_ hammer for every nai...err, problem, you are basically the stubborn hermetic hold-out as far as the broader conceptual space is concerned.
Alder seems to aim more at the social space, but I think it's a bigger problem than his dinner-party-style example would indicate.
It may even be important for people like that to define for themselves how and when the imagination is useful...
(There is also a broader question about sets of psychological tooling which would be pretty fascinating to reconcile with this kind of thought model. I can think of at least a few groups of people who would have a pretty good counter to an argument like this, especially one with a winner-takes-universe feel to it)
Briefly, I can't tell if your question is really getting at the point I'm thinking about. I also don't know your background and I'm not even sure you understand where I'm coming from, so there would be a lot of ground to cover before any convincing of someone to care would be on the table for me.
