Science is about failure
You know it’s true. Academic science in particular is a near-constant stream of not-quite-succeeding that is sometimes broken by a rare statistical anomaly — something works.
This is true at all steps and for almost all aspects of our daily lives. Positions are competitive. Grants are competitive. Reviewers are jerks. Scholarships are competitive. Journal space is limited. Experiments come back negative. Equipment breaks. The signal is buried under the noise. Administrative rules are Byzantines.
And yet everything we show of our research is what worked. The grants we got. The statistically significant results. The accepted papers. The long moments of frustration, struggle, hopelessness are almost never publicly talked about. We know it, of course. And we know that our colleagues know, and we know that they struggle too; but there is almost never a paper that will spend 90% of the time listing all of the dumb mistakes, the dead-ends, the failed attempts. Even though 90% of the project was spent failing.
Last week, my grad students suggested that I watch a TED talk by Uri Alon, on why innovative science demands a leap into the unknown. This is an important think to watch, and an important discussion to have. The feeling of being lost is never going away. But it is a good sign. It means that something will happen. But we ought to be explicit about this, especially with students. Feeling lost is normal. Trying a hundred things to have on (barely) working is routine.
Failing less does not seem very possible. Being better at judging your own chances of getting the grant, publishing the paper, obtaining the results helps, of course. But science is about exploring the unknown, and most of the unknown, it turns out, is boring, unremarkable, and non-significant. The feeling of discovering something new is worth it — imagine this, for a moment, you are the only person in the history of humanity to know something new about the world; this is equally humbling and thrilling, but it comes that the cost of an awful lot of failure.
What we can do is fail better. I am not failing much less than when I was a graduate students. But I am failing more efficiently, maybe more creatively, and sometimes more spectacularly. Research is fundamentally about not knowing what is going on, and so research training should develop the ability to be comfortable in this state. Or as I say (and did I stole this from someone?), I have no idea where I’m going, but I know how to get there.
Accepting that there will be failures takes time. There are still moments when I grow frustrated with a project, when a grant rejection hits me particularly hard, or when I wish the data would speak a bit louder. But hey, this is the price to pay to do science.
Worth it.
