Scientific knowledge is ever-changing and ever-expanding, and according to “The Half-Life of Facts” by Samuel Arbesman, it is estimated that scientific knowledge expands by a factor of ten every fifty years, leading to a knowledge ‘half-life’ of 45 years. It is the idea of pessimistic induction- theories that seem to be successful within the scope of today’s understanding may well be disproved or made obsolete by new experiments and different theories in the future. This has happened repeatedly in the past, like for example, the numerous advances in models of atomic structure since Dalton in 1808. Conversely, Einstein came up with the Theory of Special Relativity in 1905, and it hasn’t been disproven since.
As the life time of what is perceived to be true can vary so much, how soon should scientists make new results and theories publically available? In order to have any weight in the scientific community, any new papers are required to go under peer review, and, if accepted, go on to be published in a journal from where they can be referenced as a ‘reputable’ source. However, this can take months, and if initially rejected, years to finally complete.
This can sometimes lead to researchers getting impatient, and creates the problem of how soon to let the media, who can have a habit of sensationalism, know when a major new discovery is made. In 2011, the OPERA group were quick to publically announce (although with caution) that they had apparently measured neutrinos to travel faster than light, and in doing so apparently disproved Einstein’s aforementioned 1905 theory. The results, if true, could have had a huge impact in the way we understand physics, but after further research by OPERA and other groups, and extensive peer review, it was found that the findings were incorrect. This ultimately led the resignation of the OPERA group’s lead scientist.
Although it is important to get new findings out quickly for the benefit of the scientific community, researchers have to be absolutely certain that what they have found is reproducible, especially if the findings have huge implications. Peer review is, for the most part, effective at separating the poorly performed experiments and badly interpreted results from those that are credible, and perhaps only after passing though this stage should scientists and the public class any findings as ‘true’.