As any one in here knows from troubleshooting complex systems, the first seemingly sensible idea is usually wrong. To understand, you need many ideas, and to systematically update your model and disprove each hypothesis, using the evidence, and the build up in your intuition, to tune your next hypothesis. Science is more about hacking than it is about getting a well-written paper past a committee. The feedback loop shouldn't be slowed down with a political step.