Searching for the Dutch Corner in the Republic of Letters: Network Science and Wittgenstein’s Family Resemblance

This article has a twofold contribution. First, it offers a new way to consider communities that are conceptually and analytically difficult to define. This method has been based on the philosopher Ludwig Wittgenstein’s notion of ‘family resemblance’, applied here through a novel quantitative process, using an online knowledge base – Wikidata – and bespoke visualisations to identify communities in large epistolary datasets. A bipartite analysis based on shared characteristics, rather than evidenced personal connections, makes it possible to map a large community of individuals, revealing clusters of similarity and shared features. In a break from most existing applications of historical community detection, a different approach is used here, using a network object known as a multigraph, which allows characteristics to be viewed both as individual attributes and as parts of overlapping sets. Second, this article describes an application of this method to uncover a community that is often discussed but notoriously difficult to define and analyse: the Dutch ‘Republic of Letters’: this loose early modern sub-community included members from different denominations, professions, origins, and affiliations, but has been recognised qualitatively as a body of prolific and intellectually bonded individuals, with links within and beyond the United Provinces. In applying this method, we create a truer picture of how communities such as these operated and cohered, while avoiding reductive criteria for inclusion of such individuals. The uncovering of this particular Dutch community is a clear demonstration of the method in action and therefore offers a successful alternative to other community detection methods.