Paradox of Hoaxes: How Errors Persist, Even When Corrected

4 October 2012

Here is an interesting challenge that think tanks face on a regular basis -a challenge often created by other think tanks and linked to the fact that think tanks CAN get it wrong. In Paradox of Hoaxes: How Errors Persist, Even When CorrectedSamuel Arbesman argues that:

Despite our unprecedented ability to rapidly learn new things and crowdfix mistakes, Knowledge and its sinister twin Error continue to propagate in complex and intriguing ways.

Even after an error has been corrected (false information has been updated, a flawed theory has been refuted, or  lie has been caught and shamed) it has a high chance of making a comeback. Like one of those joke candles of our youth.

What caught my attention in this article is Arbesman’s excitement. The world of think tanks is full of talk about evidence based policy and grand programmes based on single studies, a few months’ worth of research, and one or two pilots at most. Donors often fund think tanks in developing countries avoiding overlap: one focusing on health, another on growth, another on education, etc. But as Arbesman points out:

It would be so convenient and predictable if all knowledge stood the test of time. But if that were the measure of being a scientist, then no one would be a scientist. No one would explore or write or even be willing to read about our latest (even if recapitulated or inaccurate) findings. Of course, we still have to be scrupulous; but the good news is that while knowledge is fickle and changing, the way it changes does obey some rules and regularities. There is a method to the madness.

So we should all keep in mind what a former professor of mine said after lecturing his classes on a certain scientific topic on a Tuesday. On Wednesday, he read a paper that was published and that invalidated the lecture. On Thursday, he went into class and told his students, “Remember what I told you on Tuesday? It’s wrong. And if that worries you, you need to get out of science.”

Here then is a role for think tanks. As they pursue certain policy decisions they must also ensure that their ideas are sound (technically, politically ethically, etc.). The often heard comment that ‘all the research has been done’ should be grounds for concern about the intellectual robustness of the organisation. Specially in the field of social sciences where things can’t ever be 100% certain.