Saturday, May 24, 2014

Are scientific press conferences bad?

I fear that may be the case.
Previous cases of premature announcements include cold fusion, "life on mars" [really dead germs on meteorites from mars], neutrinos travelling faster than the speed of light, a Caltech theoretical chemist claiming he had solved high-Tc superconductivity,.....

In march BICEP2 scientists called a press conference to announce they had discovered evidence for cosmic inflation. This coincided with them placing a paper on the arXiv and Stanford releasing a Youtube video, that subsequently went viral, showing Andrei Linde being presented with the exciting news.

However, now questions are being asked. The chronology is described by Peter Woit on Not Even Wrong and there is a nice discussion of the science by Matt Strassler. The key issue seems to be the method used for subtracting the background signal due to galactic dust. It seems that BICEP2 scientists estimated this background signal by "scraping data" off the powerpoint slide from a talk given by their Planck competitors! But was this a robust estimate?

The issue has received coverage in the press including this Washington Post article.

I think there is a broader issue here of the role of rumours in the social media age. I am skeptical that one can have a forthright, robust, constructive, and thoughtful scientific discussion via tweets and blog rumours, when not all parties have access to the relevant information and there are a bunch of journalists watching. The problem is accentuated if people have already make strong public claims that have been further hyped up by the media and institutional press offices.

I thought that this issue of science via the media was a relatively new one. However, I learned this week that even Einstein was not immune from it! There is an interesting article in APS News, A Unified Theory of Journalistic Caution by science journalist Calla Cofield. She points out how Einstein went to the press to publicise his [now discredited] theory of distant parallelism. The New York Times covered it uncritically, since he was Einstein, after all.


  1. Is it not more ludicrous that BICEP would have to go to such measures to get the foreground data in the first place? Why wouldn't Planck just share it with them? What is the point of having "competitors" in science as opposed to just other research groups that studying the same thing?

  2. Thanks for the comment. I agree that this incident highlights that this kind of "competition" is problematic for good science.