To do so, Erikson develops very clear argumentation in six independent chapters. She starts with a characterization of the evolution of the themes economic works focused on between 1550 and 1720, using compelling methods of topic modeling (the Latent Dirichlet Allocation method allows her to get a sense of what was being discussed in a huge set of documents). She also describes the increase in economic publications in chapter 1. Chapter 2 is where she adopts a more qualitative approach, allowing close-reading of early modern economic texts the reader might not be familiar with, and leading to the conclusion that debates (on issues such as bullionism or the importation of Indian cotton) drove the increase of publication. In a theoretical chapter 3, the author discusses how the narrative she defends fits with the existing literature. She quantitatively tests her hypotheses in the next chapters with three sets of methods: social network analysis (chapter 4), time-series analyses with models of vector autoregression (chapter 5), and a comparative analysis between the British case and the Dutch (chapter 6). The synthesis of the results she reaches allows her to quantitatively ground the idea that “contentiousness, organizational structure, and resources of chartered companies led individuals [mainly merchants] to publish economic arguments in order to challenge or support company privileges and practices in a context in which government influence was constricted.” (p. 177)
One of the most remarkable aspects of Erikson’s book consists in the efforts she puts in detailing and explaining data collection and treatment. Indeed, most of her argument is built on the exploitation of previously existing digital collections (such as the
The same care can be found in the interpretation of data. Erikson constantly defines the quantitative tools she is using. Very different in their elaboration, all the models developed in her work are clearly presented, giving the readers a clear idea of what they can be used for (and then potentially adapted to others’ research). The description and interpretation of results are quite satisfactory because they are rich and detailed. Network analysis appears in two chapters of the book. In chapter 1, the author uses clustering methods to organize the network of topics that she draws from topic modeling of 2,353 texts that she found and are fit for computational analysis. Clustering allows her to establish the five topics in which 17th century British economic thought can be divided: religious and moral issues, politics, husbandry, travel, and finally trade, finance, and national prosperity (p. 41). In chapter 4, Erikson builds bipartite networks to assess the centrality of authors of texts -especially merchants- dealing with economic matters in chartered companies (as investors) and spheres of power (especially their participation in committees of trade). She uses metrics like degree centrality, topological overlap, and clustering coefficients to do so.
Even if Erikson acknowledges the fact that economic ideas crossed borders in Early modern Europe (p. 216), the reader could have hoped to learn more about this phenomenon. Indeed, as works about the Republic of Letters (among others) have shown, the 17th century was characterized by a very intense circulation of ideas that could have affected the evolution of economic writing in Great Britain. Many of the dynamics described regarding English-written texts seem to have also happened in France if one considers the works of Montchrestien or Boisguilbert. For instance, the very rich database of texts exposed in chapter 1 could have allowed an exploration of quotations of other works by English authors. Far from contradicting Erikson’s conclusions, such an approach could have helped consolidate them.
Overall, this book is a pleasant and very solid read, showing how a wide range of statistical methods can be very elegantly articulated. Such an approach based on transparency regarding data-collection, exploitation, and results interpretation should be praised. In the end, a work of this kind shows how exciting can be the use of sociological concepts and quantitative methods to highlight dynamics that occurred centuries ago and affected the construction of modern day’s economics.