A couple of years ago, I went to a museum exhibition of DaVinci's notebooks. I was completely unprepared for what was my primary reaction: that DaVinci just do

A couple of years ago, I went to a museum exhibition of DaVinci's notebooks. I was completely unprepared for what was my primary reaction: that DaVinci just documented every idea in his head, and that some of them just happened to pan out. But for some reason, we credit him for the small percentage of ideas that just happened to work, and ignore all the wrong ideas. To me, he didn't seem like a genius, he seemed very much like a broken clock that just happened to be right twice a day. There's a certain survivor bias in these sorts of discussions, where we identify great contributors as taking huge risks, and so forth and so on. This is true, but we also maybe tend to give them too much credit because we ignore all their failures. We tell ourselves that's why perseverence is so important, but then turn around and make fun of people or even become angry at them for wasting our time when so many things don't turn out. The difference between perseverence and shotgun approaches to pursuing knowledge seems arbitrary to me lots of times. If you followed every one of DaVinci's ideas, for example, you'd be screwed, because he was so wrong about so many things. Today, he'd probably be seen as a crackpot because it would be so difficult to know when his idea was right and when it was wrong. The value of his ideas lies as much in his audience, the people who realized what of his ideas was correct, as much as, if not more, than in DaVinci himself. I guess I've grown really disillusioned with advancements in science, because so much of it involves a sort of mythmaking that seems fairy-tale to me. To me, the reality seems more distributed, involving small contributions, gradually changing over time, with a lot of randomness involved. I don't mean to come across as saying we shouldn't encourage experimentation and tinkering, or that it shouldn't be valued. In fact, I'd argue the opposite. But at the same time something seems off about our incentive structures--the way we give credit seems off. Think of it this way: we do science because at some level, what's correct and what's not is an empirical idea. You could have 10 very reasonable explanations and 9 of them will be wrong "just because." In many cases, being intelligent has nothing to do with why the 9 are wrong, otherwise all we'd need would be philosophy and math. So why do we pretend that the 1 person who was correct was correct because they are a genius? It seems like you can't have geniuses and a need for science at the same time.