Earlier this week, an article in the Financial Times (1) caught my attention and eventually motivated me to write this post. While I find myself in stark disagreement with the views presented in this piece on research in the physical sciences, this read gave me the chance to ponder and organise my thoughts on the interplay between science and the communication of scientific results.
As a physics graduate, the title of the FT article is difficult to forget: “Physicists brought low by thinking big”. Further, the standfirst claims rather openly that scientists hunt breakthroughs in order to “seal their reputations”. I think that these statements need clarification. If researchers did not “think big”, many important scientific discoveries may have never seen the light. Once researchers set out for their investigation, I doubt that the main drive is to “seal their reputations” with their ground-breaking results. Surely some scientists enjoy the spotlight and actively search for it; as long as they play by the rules in order to gain visibility, what is wrong with this? Even more – in all fairness, is it not the case that the media happily takes advantage of the ‘spotlight attraction’, and sometimes even operates to create a sensation out of what should remain a purely scientific debate within the community? I shall come back to this aspect later on.
The FT piece was partly inspired by an earlier article published in Nature at the beginning of July (2). The author of the latter – Jan Conrad, a professor of astroparticle physics in Stockholm, Sweden – argues in favour of stricter policies when it comes to the announcement of (alleged) physics breakthroughs to the scientific community as well as to the media and, consequently, to the general public. Conrad is genuinely concerned, and therefore urges physicists to implement effective changes to their codes of conduct in the light of how science is now done and reported – the open release of data and free online archives for preprints being two examples of novel trends. In his opinion, physicists have possibly become too relaxed about the way in which they communicate the outcome of their research, both to peers and to the taxpayer. To support his view, Conrad refers to results in particle physics and astrophysics which turned from discoveries into, well, instances of how not to do science. One such example dates from 2012, when data from the OPERA experiment supposedly flagged the existence of neutrinos travelling faster than the speed of light (3). In the search for evidence of the expansion of the early universe, the BICEP2/Planck collaboration was recently at the centre of a similar deflation on earlier claims made by the BICEP2 team (4).
In his piece, Conrad comes to the conclusion that the physics community, along with academic publishers and funding bodies, needs to acknowledge these breaches to scientific best practice as serious matters. Quality checks and collaboration should prevail over a somewhat sloppier attitude, even when demanding requirements for grant applications and internal competition almost seem to suggest otherwise. Interestingly, the article in the Financial Times draws on the example of the OPERA experiment to oppose Conrad’s view: science is “messy”, and it would be untruthful to deliver a ‘purified version’ of the research process to the public. It strikes me that the author of the FT piece did not notice that Conrad only marginally mentions the role of the general public; one could criticise him for overlooking this point, but this is an entirely different matter. Most importantly, I would like to stress that the OPERA and BICEP2 examples do not, and should not, lead to the conclusion that science is a “messy affair”.
In both cases, I agree with Prof. Conrad when he writes that researchers rushed to make their results public, and in doing so postponed cross-checks on their instrumentation as well as on their data analysis which, in fact, they then performed anyway. They should have reversed the order – not because in this way physicists would have looked flawless in the eyes of the general public. This is not what is at stake here. Physics relies on a methodology crafted over centuries of scientific progress, and every researcher should keep in mind that a theory holds until it is proved wrong. In this sense, it is good practice to perform all known checks before publication and not following a press release. I doubt that Conrad’s suggestion is to hide the winding paths that sometimes lead to scientific advances; rather, I believe that his piece acts as a reminder to the members of the physics community – learn from your mistakes and do not forget the basics of your discipline.
If I look at the two articles published in Nature and the Financial Times, I would be tempted to conclude that the dialogue between science and the media is almost as relevant as effective codes of conduct to modern research. Scientists should not fall into the trap of declaring a breakthrough ahead of time. Yet they are humans and may become impatient – at which point it would be up to the media to play the wiser one: if a press release mentions an ‘indication’ or an ‘anomaly’ in some experimental data, it may be worth tuning down the enthusiasm by a fraction while waiting for further developments. Both parties have their share of responsibilities in the dissemination of scientific advances. The good news is – as with research, there is always room for improvement.
Note: The title of this post was inspired by a quote attributed to physicist W. Pauli, who apparently used the phrase “It is not even wrong!” to describe a result which is fundamentally flawed (for example, because it cannot be falsified by experiment).
(1) http://www.ft.com/cms/s/0/88d22dea-23e6-11e5-bd83-71cb60e8f08c.html#axzz3fHyKF2se — Unfortunately, the article does not seem to be freely accessible at the time of writing.
(3) http://profmattstrassler.com/articles-and-posts/particle-physics-basics/neutrinos/neutrinos-faster-than-light/opera-what-went-wrong/ — I decided to cite this link because I found this online post very exhaustive; it also clarifies well a few aspects which are incorrectly reported in the FT piece (1).