Annals of Ambiguity

Evidence Isn’t Proof and Proof Isn’t Everything

Evidence is important. We’ll start with that. We want to do things that we have reason to believe will work. And evidence gives us a reason to believe. So it’s no surprise that we hear a lot about evidence-based approaches to tough problems. In health care and education, in social work and even criminal justice, “evidence-based” is something like a seal of approval.

And no one wants to reinvent the wheel. People working on projects say it a lot: “Let’s not reinvent the wheel.” What does it mean? Do we even know, or is it just something we say – an empty phrase that nonetheless expresses the unexamined privilege given to evidence-based approaches?

Consider what it meant to invent the wheel. We don’t call it the accident of the wheel, so in a fundamental way we think the wheel was the result of human agency. Accordingly, there are two possibilities why “Let’s not reinvent the wheel” is said.

When we say it, we might be surrendering to a nearly inexpressible doubt: “We don’t really know what we’re trying to do or even why.” We’re overturning what until that moment had passed for understanding the problem or condition we’re charged with revolving, an understanding that vanishes on closer consideration. And by implication we’re suggesting the wheel might have been invented because the problems it would solve were not yet understood. That is, inventing the wheel was an act of opening up a new world.

The other possibility rests on the assumption that problems arise independently of us – and our understanding of them objective – affording us the detachment required for solving them. “Let’s not reinvent the wheel” means the group’s project is not something that hasn’t been done before. It’s this way of saying “Let’s not reinvent the wheel” that’s decisive. It’s the one given authoritative thrust by its implicit appeal to evidence, to what’s been done, and worked, before.


Both these possibilities manifest a particular disposition. In the first case, the group regards an assignment (a problem) as merely a starting point from which the way forward is deliberately held in obscurity. In this way, in the first case, “Let’s not reinvent the wheel” is synonymous with the exhortation: “Let’s explore!”

The second case is a disposition that holds the problem as a matter already fundamentally settled. That is, it’s been defined, an outcome determined, and an approach determined. All that’s left is execution.

Here’s a radical question: when we work to solve a problem, what if our efforts stem from the same kind of thinking that gave rise to the problem? And does this kind of thinking seem like the only kind of thinking there is. (This might be what Einstein meant when he said: “We cannot solve our problems with the same thinking we used when we created them.”)

Only the first case obligates us to overturn our understanding of a problem and thereby initiate a different kind of thinking than the kind that, Einstein tells us, created it. And this necessitates de-privileging evidenced-based approaches, because evidence only has currency when we assume that we already understand a problem.

Can we imagine project work that is, if not “wheel reinventing,” creative in the way inventing the wheel must have been? This kind of work would begin by admitting that the prevailing understanding of the problem will not inspire the kind of work that will solve it.

Maybe we do want to re-invent the wheel – and open up the new worlds that accompany new kinds of thinking.