One of the most painfully overused phrases in science is ‘paradigm shifting’. The roots of the term as used in a scientific context come from the philosopher Thomas Kuhn, who utilised it in his model of scientific advancement. While researchers have interpreted Kuhn’s work in different ways, a general sense of the model is as follows:
Science proceeds under a paradigm of knowledge, methods, and techniques, which together define a kind of overarching global perspective. As scientists continue to accumulate knowledge, anomalous results begin to build up, until they are no longer explicable as merely errors under the existing paradigm; once the community of scientists accepts that these anomalies require a new global perspective to fit these anomalous findings in, then science undergoes ‘a paradigm shift’ to a new framework of knowledge, approaches and methods.
In my brief editorial experience, it seems like many researchers are big fans of this term – many use it in cover letters to suggest that their work is valuable and significant. I don’t intend here to either question the model of scientific advancement suggested by Kuhn, or to debate the various merits of the supposedly paradigm-shifting work submitted by different authors. No, here I’d like to contend that the modern relationship between science and publishing makes a genuine paradigm shift in the context described by Kuhn rather more difficult; much more difficult, in fact, than one might believe based on the frequency at which the term is used in the media and in cover letters.
Two factors are at play here, I think. First, the critical process of peer review means that anomalous results are potentially more likely to receive intense scrutiny, making it ever harder to publish work that might significantly undermine the existing core perspective. Secondly, since number of publications tends to be an important metric by which academics are judged, there is an incentive to break down radically anomalous findings into smaller publishable pieces. While individual small publications can still add to the body of anomalous results, for researchers with a grand, game-changing idea, the potential lure of multiple papers might outweigh the hard work in building a large case for a new mode of thinking that cannot be supported under the existing framework.
Peer review is considered a vital part of modern science, but when Kuhn published The Structure of Scientific Revolutions in 1962 (in which he proposed the debated model of scientific advancement), peer review was only beginning to be formalised. As part of the post-war scientific boom in the west, peer review was becoming increasingly important to secure funding1, but it is notable that many of the paradigm shifts considered by Kuhn, such as the shift from a Ptolemaic view of planetary motion to a solar-centric one, was based on science prior to the advent of review. Galileo and his contemporaries were able to publish without first getting their results past their peers, who may have had personal bias against such radical ideas. I’d suggest that it’s worth asking how easy it is today to persuade referees of a novel idea when they are working in an existing paradigm.
The second point is arguably more subtle. I think it’s fair to say that even a short contribution can dramatically change the way we think about the world, but making a case for a dramatic shift in scientific frameworks can require a large body of evidence. The Origin of Species is not a short book; Charles Darwin used a vast range of examples and data to build his case, and in combination they provide a new framework for understanding life on earth. The modern publishing incentives seem unlikely to encourage such large compilations, however. Judgement of researchers based on the number and citation count of their publications encourages splitting of projects into smaller parts (or even, derogatorily, ‘Least Publishable Units’), while simultaneously discouraging scientists from putting out anomalous results without context, which would be unlikely to achieve high impact (and as suggested above, may have trouble getting through review). I’d suggest that The Origin of Species as a series of small papers would be unremarkable until the final short-format piece that linked it all together; I’m unsure whether this would be a successful way to build a career in the modern academic environment.
Younger researchers are encouraged to publish more (and thus potentially split their work up more), and may therefore be more prone to this kind of effect. Older, more experienced scientists may have more intellectual and emotional capital invested in the framework within which they have spent their careers working; the tendency to promote game-changing suggestions may thus be more limited amongst more established researchers; I’d love to hear counter-examples, though.
A positive suggestion to address these aspects might be to emphasise the importance of conferences! There, unrefereed work can be judged by a broader community, and anomalous results presented concurrently, by researchers of all ages and backgrounds. With plenty of discussion and open-mindedness, these should serve as highly productive ground for giant leaps in our understanding of the world around us.
Perhaps these suggestions do not, in reality, limit the progression of science under Thomas Kuhn’s model. I do think, however, it’s worth questioning the modern understanding of the term, especially as science has changed so much in the past decades. When some studies suggest that global scientific output is doubling roughly every nine years2, it’s worth considering whether our models to describe its advancement are still valid.
1: Csiszar, A. Peer review: Troubled from the start. (2016), http://www.nature.com/news/peer-review-troubled-from-the-start-1.19763
2: Bornmann, L., & Mutz, R. Growth rates of modern science: A bibliometric analysis based on the number of publications and cited references (2014). Uploaded to arxiv.com: https://arxiv.org/abs/1402.4578