The structure and evolution of scholarly interests from antiquity to the eighteenth century

Human societies reap tremendous benefits from acquiring scientific knowledge, which provides an unparalleled understanding of our universe, along with economic growth and increased life expectancy. Science is fueled by scientists’ curiosity. To better understand this curiosity and its structure, we rely on a database, Cultura 1.0, which lists all scholars found in online catalogs, in order to look for patterns in their interests. To limit the effects of Western influence, we circumscribe our study to scholars born before 1700. Since many scholars have more than one interest (e.g. mathematician and astronomer), we can construct a network of co-occurrences of interests with the 2317 ‘polymaths,’ defined here as scholars who have at least two interests (out of the 13,556 scholars of the database). While attempting to account for the overrepresentation of European scholars, our results suggest some above chance commonalities in the co-occurrences of interests, suggesting an underlying structure in scholars’ interests and curiosity. For instance, in Europe and in the rest of the world, philosophy and mathematics co-occur relatively frequently. Clustering the network of interests, three broad domains emerge: Human (e.g. history), Natural (e.g. zoology), and Abstract (e.g. mathematics). We find that every world region in the database (e.g. ancient Mediterranean) contains a similar portion of scholars devoted to each of these domains, and that the Human and Natural domains appear substitutable for each other at the margin. We believe that our findings shed light on long term historical trends in scholarly interests, and also help us better understand the structure of human curiosity.