Incidentally, one of the arguments usually brought up against direct entry midwifery is that these women "just" shadow other midwives to learn (which is not true as stated; there is academic learning involved as well). But isn't that model the same as the one used to train doctors? Years of book learning followed by years of clinical supervision?

Perhaps the difference is in the very nature of how we store information. Objective data is easy to put on paper, and load into spreadsheets. But the thousands of little anecdotes that we collect over our clinical careers is hard to quantify in this way… until you bring in the human brain. We have an ability to synthesize all of this data into a remarkable level of coherence that a computer and statistics seem to lack. This is why certain areas of medicine demand hands on learning rather than reading to become proficient. Almost everything I know about surgery came from operating with surgeons better than I, not from reading books. I might get ideas from reading books, but until I can put it into practice it isn't really knowledge.

There's a place judgement and there's a place for data. Both are necessary but neither is sufficient. I can look at any well reputed journal and find you several studies that are so poorly designed that their results are absolutely meaningless. It takes judgement to know that. You have to know what data is real and what is nonsense, and when good data still doesn't apply to your patient.

At the same time, sometimes experience can lead us down the wrong track…. "Experience is what allows us to make the same mistakes with greater confidence".

Now, is that science?