An interesting article published in Science Magazine looks at the Science of Science (SciSci). The promise is that when Science studies itself through an iterative process, it will better itself, thus better the world.
Science as a subject for itself is fair enough, although it does raise the question of what SciSci looks at when it looks at Science. One plausible answer is demonstrably quantifiable Science artifacts — a fancy way of saying looking at wiring under the board of Science for stuff we can measure.
The article explains Science as “a complex, self-organizing, and evolving network of scholars, projects, papers, and ideas,” which is odd as the description is missing the Scientific Method. However, the framing of Science in this manner is vital as it relates to the method of examination used in the article. In this view, Science is primarily an iterative social process of knowledge production that establishes itself in complex networks of artifacts. These artifacts, like books and articles, mean that Science is intertextual, it is an idea rooted in a system of texts. However, there seem to be various other artifacts of Science to examine quantitatively too, as an example, who pays for the study, what came of the research, partnerships, scientist moving around institutions, and citations. SciSci’s examination of these all-inclusive elements delineates some of the backings of Science as a forward-moving social process.
The big idea of SciSci, seemingly, is that a greater understanding of what supports Science leads ultimately to a healthier world. That Science can give us better Science ultimately means enabling us to do better things. A rather rosy picture of Science that seems to forget all of the bad stuff but, still, nonetheless reasonably accurate.
The article goes on about problem selection, career projections, collaborations, and citation analysis, which are telltale signs of the transdisciplinary. SciSci spans scientometrics, innovation studies, econometrics, statistics, “network science approaches, machine-learning algorithms, mathematical analysis, and computer simulation, including agent-based modeling.” Combined, these disciplines form a net made of quantitative interlocking methodologies that can examine the underlying structure of Science, so they say.
Early in the article, an example of the type of SciSci findings is given, “finding that small teams tend to disrupt Science and technology with new ideas drawing on older and less prevalent ones. In contrast, large teams tend to develop modern, fashionable ideas, obtaining high, but often short-lived, impact.” So the results of the SciSci method may allow for strategies to stimulate scientific development and in turn, make the world a better place. Perhaps.