There are at least two distinct readerships for this book. The first is the early modern historian wanting to learn about power in the Tudor era. The second is the historian wanting to learn what network analysis can really do for historical research. Pleasing both with a single work is a tall order, and I suspect that despite the title, the latter group will come to cherish this book more. As a digital humanist and historian of another era, I am not the person to judge the contributions to early modern scholarship. What I can review is the methodological contribution this book makes to historical studies and digital humanities. Given the organization of the chapters around different network measures and methodological forays rather than historiographical themes, I suspect this is the real contribution the authors intended to make.
Tudor Networks of Power centers around an extensive network analysis of the 132,747 Tudor-era letters of the State Papers. This corpus of contributions by 20,560 letter writers has been brought together over centuries, through happenstance, confiscation, and curatorial effort. It is primarily a corpus of metadata—who sent to whom, topics, dates, but not transcriptions of the letters themselves. The information available is well suited to a network analysis, and the Ahnerts (one a historian, one a network scientist) have combined their talents to identify what network analysis can teach us about the collection.
Ahnert and Ahnert start with an extensive source critique, which serves as the “trust us,” written for an imagined reader who may come to the book full of skepticism about yet another digital scholar treading on their domain. Here, they show awareness of the concerns: what about varied survivability? What about the Whitehall fire of 1619? Will it not it show us what we already know? How do the decisions made by curators over centuries affect the corpus we can study? Here, they have answers for all.
From there they move through chapters that test what different network analysis measures can tell us about the people who wrote each other letters. With the exception of a chapter on “women,” all of the other chapters have names that draw on the language of networks: “betweenness,” “network profiles,” “information flow,” and “itineraries.” Each of these introduces a network analysis concept or concepts, applies it to the corpus, and then explains what that tells us about these writers and their connections to each other.
They use “degree,” “degree centrality,” “in-strength,” and “out-strength” to show that the network in question is better described as one with a “giant fringe” of activity rather than one centralized around a small number of hubs. These measures also allow them to identify people whose contributions to the conversation are most likely incomplete within the archive. They use “betweenness” measures to start to classify the types of contributors in the network, identifying the bridgers (informants, foreign merchants) and the hubs (secretaries) who collect information on behalf of the sovereign. This brings them to lesser-known figures such as ambassador Tommasso Spinelli, whose high-quality network made him an important contributor to the crown’s knowledge of the political landscape.
From there Ahnert and Ahnert start to combine measures to create unique fingerprints for each of the thousands of writers across eight different variables: in-degree, out-degree, total degree, in-strength, out-strength, total strength, betweenness, and eigenvector centrality (a measure of how many important connections a node has in the network). They show that by calculating the Euclidean (straight line) distance between any two individuals in this eight-dimensional space, you can find the most comparable people based upon their patterns of activity. They use Pietro Bizzari, a Venetian informant in the pay of William Cecil, to show how this approach draws together people who might otherwise look quite different. In this case, Bizzari was most similar to a few foreign merchants and soldiers. With their historian’s hat on, they explain why this ultimately makes sense and wouldn’t emerge without the network approach.
As women are under-represented in this network, the authors take a different approach, building “triads” to classify women into petitioners, ladies of power, and people whose fortune led them to be both at different times in their lives. This chapter is the first in the book to draw on other methods, with a simple linguistic calculation of word frequency differences between male and female writers, finding that women were more likely to use words denoting familial relations (husband, son) or the language of petitioning (beg, pray).
The final two chapters are less about network analysis and more about what other disciplinary approaches can offer. “Information flow” analyzes keyword frequency over time to identify spikes in usage, singling out terms such as “rebels” or “plague” as topics of the day, but not every day. These topics are then formed into “topic networks” where the reader can explore which types of people write about the trending issues. The last chapter turns to the geographer’s toolkit to map the networks of a number of individuals, to show the spatial range and geographic clustering of personal ties. This is a chance to demonstrate that even those with big networks did not know people from everywhere. Measures in this last chapter include average travel distance per letter and total mileage.
This book is a masterclass in the vocabulary of network analysis for the uninitiated. Anyone interested in answering any similar questions to those outlined here will find this book an invaluable starting point. Ahnert and Ahnert show how the math leads to the defensible conclusions and difficult-to-find insights. I’d hazard to call it the first true digital history monograph that centers a methodology against a historical case study, rather than the other way around. I don’t know what the early modern historians will make of it, but it will long be a wonderful teaching tool that brings network analysis into the grasp of many more historians. I’ve already assigned it to one of my Ph.D. students, and I’m sure they won’t be the last.
