can an experiment make us learn less?

Mary tends to do useful things like post poems, prose selections and quotes on her blog. Today, she put up a list of R. Buckminster Fuller quotes. I love Bucky as much as the next guy, but I had to take exception to one of his opinions. Here is the comment I posted on her site, which serves as a useful independent post, methinks (slightly modified):

I wish I could agree with this one:

"Every time man makes a new experiment he always learns more. He cannot learn less."

But I just don't think it's true. I think there are probably entire fields of studies whose assumptions generate experiments that shackle our imaginations. The first one that comes to mind is economics -- by assuming that humans are rational beings attempting to increase efficiency, there are probably many economic studies that are causing us to ignore data about how people actually behave. And I'd say there are analogues in the physical sciences too. Of course, there are also experiments that are just poorly designed. I think you could argue that we learn from these failures in the long run, as a species, but they limit us in the short term. So I'd like to say that I might agree with the intent of the statement, but not its content. I'd say that while science represents a fulfillment of human nature, our inherent experimental impulse is still primary. We shouldn't claim that the scientific process itself is inherently good. Instead, we should say that human curiosity is the primary good, and science is good to the extent that it furthers that curiosity (rather than stifling it).