Italian Trulli

"'How the World Jogges': Interconnectedness, Modularity and Virality in Seventeenth Century News"

Yann Ciaran Ryan (y.c.y.ryan@qmul.ac.uk), Queen Mary University of London, United Kingdom

'Viral news' is often associated with twenty-first century social media, but early modern news behaved in a similar way. News spread faster than would be expected in either a linear or diffuse model, much like the spread of information on Twitter or the global spread of contagious disease. Recent scholarship has shown that early modern news operated as a network (Raymond, 2016: 115). Crucially, it is theorised that to operate in this way, the network could be said to adhere to a 'Barabási-Albert' model: a model for reconstructing networks using preferential attachment, which explains why in many real-world networks the vast majority of connections go to a small number of entities (Barabási and Albert, 1999: 509). Network analysis has been used by early modern scholars to examine correspondence networks between individuals, with recent work showing how network analysis can uncover both broad and specific patterns of communication. (Ahnert and Ahnert, 2019: 28). This poster shows that the same methods can be used to examine an early modern network where the nodes are cities rather than individuals.

This poster uses data from seventeenth news sources to examine one aspect of this network: modularity. Modularity is the tendency for a network to divide into individual components: it is a measure of the extent to which a network is a series of cliques or, on the other hand, a closely connected whole (Blondel et. al., 2008: 2). Modularity measures and 'community detection' algorithms can score any network on its degree of heterogeneity and then suggest the mostly likely divisions in that network. Understanding these divisions is key to understanding the impact and spread of early modern information.

Between 1641 and 1649 London saw the development of a relatively free printed news industry. This included regular news from abroad: news mostly from Europe but occasionally further afield. It has been shown that this news moved in individual 'chunks', organised on the level of the paragraph rather than issue or even article (Slauter, 2012: 253). The regular, paragraphed structure of these early 'newsbooks' means that structured data can be collected from the texts, either manually or extracted automatically. Utilising a manually-collected dataset of paragraphs of news, it is possible to analyse the underlying network of cities behind the information found in the newsbooks.

The poster will demonstrate how the use of such a dataset with network analysis has led to the discovery of communities of news and information, specifically using network modularity and community detection to suggest the extent to which Europe could be divided into individual clusters of cities closely linked by the sharing of information, and how this can be used to understand the viral nature of early modern news.

The network analysis shows that, at least from the perspective of London, Europe's news network was highly modular. It can be divided into eight distinct, closely-connected communities, with twenty-eight key cities holding together the entire networked system. This modularity explains how Europe's news system was highly viral and efficient. The communities found share different properties which are explored through this poster: sometimes regional and in close proximity to each other, but always joined by lines of communication rather than politics, language or religious confession. They come together to make a sketch of Europe's communities of information.

The poster presents maps, network diagrams and sample data to illustrate the suggested communities of information in Europe, as found from London newsbook data. It suggests reasons for the individual communities, and outlines the extent to which the communities can be seen to follow linguistic, confessional or political lines, or whether they deviate from these traditional divisions. The poster explores the extent to which understanding the network as a series of closely-knit communities can explain the reasons for the viral nature of news transmission, even in an early modern setting.

Appendix A

Bibliography
  1. Ahnert, R. and Ahnert, S. (2019). Metadata, Surveillance and the Tudor State. History Workshop Journal, 87: 27-51.
  2. Barabási, A. and Albert, R. (1999). Emergence of Scaling in Random Networks. Science, 286 (5439): pp. 509-512.
  3. Blondel, V., Guillaume, J., Lambiotte, R. and Etienne Lefebvre. (2008). Fast unfolding of communities in large networks. Journal of Statistical Mechanics: Theory and Experiment, 10: 1-12.
  4. Raymond, Joad. (2016). News Networks: Putting the ‘News’ and ‘Networks’ back in. In Moxham, N. and Raymond, J. (eds), News Networks in Early Modern Europe. Brill, pp. 102 – 129.
  5. Slauter, Will (2012). The Paragraph as Information Technology: How News Traveled in the Eighteenth Century World. Annales. Histoire, Sciences Sociales, 67(2): 253 – 278.