Jim, I agree in general with your "clinical reality" argument, and with the idea that it might turn into some "golden moments". A failure in trying to relax could become a point of discussion about that type of thing typifying what's going on. Brian, I've got more of a problem with what you are saying. We don't know what his mind set is. Teaching progressive relaxation fosters a particular mind set, and it's not at all clear to me that it's one you would want to go for. Relaxation is relaxation, and failed relaxation is a chance to get a clearer idea of what might be causing problems. Relaxation, even Ost's applied relaxation, is not "a structured learning experience (which) promotes a modification in attitude, offers an opportunity to test client motivation" and so on. Failed relaxation can set you back. I agree with Jim that it may be worth risking, but it is a risk. It would not be unheard of for a client to drop out if it does not work. And that's not a test of client motivation, it's a test of therapist competency and ability to engage the client in a collaborative relationship even when things don't work out too well. I think that it's important to be clear as a therapist about when you are taking a risk. One way of dealing with that is to prepare the ground (this may or may not work; whatever happens, we are sure to learn something important from it). That has the advantage of being the truth, too.
Of course intervention and assessment are linked. However, without a conceptualisation/formulation, you are treading water at best, drowning at worst. Intervention informs assessment throughout therapy, but is never a substitute for it. The latter is technical eclecticism (shudder)