I recently read <a href="http://arstechnica.com/science/2015/08/mit-claims-to-have-found-a-language-universal-that-ties-all-languages-together/">this</a> article about the supposed discovery of a language universal by researchers at MIT. I was only able to read this news article and the original research paper's abstract, but knowing a little bit about linguistics and a lot about probability I was underwhelmed. I don't want to say that it was a dud of a paper, but their conclusion seemed pretty obvious a priori, and it has almost nothing to do with the Chomsky-style language universals it was being touted as. The paper is a classic example of work that is 20% incremental progress, 80% phds students trying desperately to publish something.
But I can only see that because I DO know about linguistics, probability, and the inner workings of academia; to outsiders this work would look like an actual discovery. This misconception is encouraged by the journalists, who have every incentive to sell incremental growths in evidence as if they are breaking news.
This is a very innocuous case. Do I need to mention the one, solitary research paper that found a link between vaccines and autism, and that was later completely discredited? Or how about the economics paper about the dangers of high public debt, that was used as the basis for policy decisions and was <a href="
http://www.bloomberg.com/bw/articles/2013-04-18/faq-reinhart-rogoff-and-the-excel-error-that-changed-history">recently</a> found to contain an Excel error. The list goes on. And it is all the worse because these things discredit the scientific endeavor as a whole, giving fuel to those who reject established scientific truths.
So I propose that, in general, journalists and politicians should steer clear from delving into scientific papers. They should instead focus on scientific *consensuses*, the invaluable knowledge we acquire when the scientific sausage is done being made. There is a ton of room for debate on this point, but I've been thinking about it lately and would love some discussion. Thoughts?