In the world as we understand it, based on Einstein, nothing can go faster than light. This prediction based on the general theory of relativity has proven itself countless times in empirical research. And now, lo and behold, a group at CERN has observed neutrino’s racing through earth from France/Switzerland to Italy at the World-record breaking speed of slightly above light-speed! This finding has received a lot of coverage in the media. Why is this so interesting?

Of course this research finding has given rise to wild speculations, including the possibility of time travel (would Richard Branson be willing to start a company on this one?). Various wild theories have already been formulated that possibly explain how both Einstein and the CERN-researchers can be right. It could be the case that further replications will show that in fact neutrino’s cannot exceed light speed. And finally it could eventually turn out that Einstein was wrong.

The observation of these speeding neutrino’s contradicts our best understanding of physical reality. We simply cannot explain these observations. What I find the most interesting about this all, are the huge efforts of scholars really trying to understand and especially to replicate these highly counterintuitive observations. The group at CERN has done its own replications, accounting for some of the criticism on the original approach, and calls for other groups to do the same. This response to conflicts between theory and empirical reality is to me a sign of a healthy scientific practice.

The social sciences do not deal with neutrino’s, but can learn a lot from this scientific practice. Recently the Dutch science community was stirred up by a very serious case of fraud in social-psychology research, with a prime investigator faking experimental data. Enough has been said about this, and there is no need to discuss it all over again here. A comment by one of his non-fraudulent co-authors, however, was very interesting. This co-author was given a table with the results of analyses on a new set of data (that eventually turned out to be made up completely, but she did not know that at the time), and used that information to test their hypotheses. The results supported the hypotheses, as a result of which, she argued, she didn’t question the validity of the data as much as she should have.

Now we know how unfortunate that turned out. But there is a point in this: people, including scientists, tend to be less critical when what we observe aligns with what we know. In the social sciences this all too often results in a tendency to find support for theories. And while any theory needs some level of support, the consequence of this practice is that all too often many theories are formulated on a specific phenomenon that are intrinsically different, while no tests are performed which of the theories actually performs best. Factors such as this confirmation bias, but also other factors such as the theory-ladenness of observations, the unmeasured variable problem, our tendency to downgrade the perceived importance of exceptions, and our mind that is trained to immediately formulate ad-hoc explanations for observations, dictate that we cannot simply base a scientific discipline on confirmation and illustration.

So, in line with the response to the speeding neutrino’s, social science could benefit from more replications, more attempts to reject theories, fewer attempts to ‘prove’ theories by finding supportive evidence, and especially more attention to observations that do not fit a theory. Instead of yet another confirmation, solving puzzles, contradictions, and paradoxes should be the principal task of the social sciences.

The OPERA Collaboraton: T. Adam et al. (2011). Measurement of the neutrino velocity with the OPERA detector in the CNGS beam Arxiv arXiv: 1109.4897v2