In April, I wrote about a little debate some coworkers and I were having about whether politics could be studied scientifically, and whether that approach was valid (I said it could be, and that it was one among many valid approaches). One argument tendered in opposition was that we simply don’t have enough information to even make a successful politico-scientific model, let alone to test whether it works or not. Any attempt to construct such a model, or to delineate variables and units of measurement, would be in vain.
I think that is incorrect, because in science, the cool thing is that fail = win. What I mean is that science is all about testing something, getting results, and developing a conclusion based on those results. If your model sucks and your experiment fails as a result, you’ve learned something. You rebuild, you try something else, and you chalk your first try up on the list of things that don’t work. Launch and iterate, my friends. It’s the Google way, and we’ve done alright so far.
Stories are just as much about what’s written as what is not written. Pictures are just as much about what is captured as what is not captured. Science is just as much about what works as it is about what does not work. There is another debate in the research community about whether it is worth publishing failed experiments in academic journals. Of course it’s worth it! People learn most from failure! Of course maybe more space should be reserved for the epic wins, but there is just as much to be gleaned from the details of an epic fail.
That is why I am so stoked about the Energy Department’s announcement that they are going to fund a bunch of cutting-edge research experiments, some of which currently exist only in conceptual form. Says the leader of the new initative, Dr. Arum Majumdar:
“We don’t know which ones are going to work, but we’ll try them,” he said, “and if many of them fail but one works, that’s great, we’ve solved the problem.”
Exactly. And for the projects that are funded but ultimately do not prove feasible or scalable, we’ll still have a win on our hands. We will have explored and ruled out approaches based on data instead of instinct, and we will also have encouraged researchers and entrepreneurs to take risks and to think outside to box. That’s America, baby.
Image used and modified under a Creative Commons license courtesy of willandbeyond.
