Networks, boundaries, communities in early modern letters

Letters were one of the main means of communication in early modern times and many correspondences have survived in private and public archives. Historians have long recognized the value of letters for history, so many of these correspondences have been edited in source publications since the nineteenth century. There are exchanges of letters in many fields of life including political correspondences between officials, letters between scientists and religious exchanges between clergy (catholic and protestant) en cultural correspondence between various types of artists. Recently, the promise of the digital collection of correspondences for the study of the exchange of ideas has been explored in several international projects about the republic of letters and the circulation of knowledge.¹
The republic of letters as an 'imagined community' of scientist was supposedly in existence in the seventeenth and eighteenth century and there has been much debate as to how much of a real borderless community it actually constituted. For political correspondences the case was clearly very different, as these involved realms with real borders and relations of authority and political friendships and enmities, but also family power networks.² Apart from the obvious cases of science and politics there are many more networks and communities in the cultural realm, in religion but also in that of trade and of course family business.
In the digital compilations of the last decades, the emphasis in the communications within especially the scientific community has been on the reconstruction of social networks. This is difficult because of two inherent properties of correspondences of one person:

• - one person is usually part of different realms and may write about different subjects in a single letter; this may complicate the analysis of the extent of his correspondence
• - the main character of the correspondence is (logically) at its center. Even a cumulative reconstruction of correspondence networks tends to consist of a number of more or less overlapping circles.
• These two properties are illustrated by the visualization in figure 1 that gives an impression of the political and literary parts of the network of Constantijn Huygens, a seventeenth century Dutch civil servant, literary author and scholar. There are less literary correspondents than political, but the networks are mostly apart; the people whose names are in maroon appear in both networks are only a minority and they tend to be the ones with whom Huygens wrote most anyway (and about all subjects), among whom his employers and his family.³

fig 1. Overlapping political (blue) and literary (orange) networks of

Constantijn Huygens (1596-1687). 

Names of people appearing in both networks in maroon

In the context of communities, boundaries and identifications, these properties are much less of a limitation. Here, the 'reach' of a correspondence of a single person and the entanglement of different realms can throw light on the nature and boundaries of different communities and their mutual interactions. By mapping correspondences on maps where (changing) boundaries and terrain are already shown it becomes possible to assess and visualize the influence of political and physical boundaries for the exchange of information in different realms of life and, by using correspondences from different periods, the changes over time in the boundaries of the communities and the extent of the connections.
For an example of such a change, William I of Orange started out as a prince in the sixteenth century Habsburg empire under emperor Charles V, but later turned the leader of the Dutch Revolt against Philip II. In his correspondence this is evident by a change in his correspondents who were predominantly German before severing the ties with Philip II and an orientation to a number of different partners after 1572, when the revolt had really taken off. Only the correspondence with members of his family showed continuity throughout the period. Apart from communications about his political affairs and patronage, the correspondence also contains letters about the management of his family and economic affairs.⁴ Like the correspondence of William of Orange, the exchange of letters between different parts of the Habsburg empire is also an indication of the strength of association between its constituent states and polities.
Methodologically the sending and reception places of the letters can be plotted onto maps. The number of letters that is exchanged is an indication of the strength of association and is used for drawing network graphs. This is not new in itself, but it is when combined with different 'layers' of communities and the shifts over time. This can be accomplished by using the metadata of the letters, that are in many cases available or can be made available by curating data from the digitized correspondences, using existing indexes. In addition, the letters may be used to assess several other characteristics of the communities in which they were used for communication:

• the language in which the letters were written shows the language of communication in a realm. For science, this was Latin, but the vernacular was used for other international contacts in the early modern period. For example, the correspondence of William of Orange contains letters in Dutch, French, German, Italian, Latin and Spanish and apparently this is the case for most of the political correspondences of the time. In the eighteenth century, however, international diplomatic communications were mainly in French.
• Letters give an indication for the means and the speed of communication in a certain realm. While there have been some studies of the postal services, it is not clear if there were differences between different realms of life - whether political correspondences was exchanged faster than scientific letters for example and neither do we know if there were shifts over time from the 16^th to the 18^th centuries. This may be derived from the time lapse between letters from an exchanges of letters between two individual correspondents (and their differences and shifts over time). In addition the letters contains some indications, that may be harvested by using linguistic tools and text mining techniques
• There have been experiments with the exchange of ideas and the circulation of knowledge, using text mining.⁵ This has proven to be quite a challenge, not in the least because this is multilingual material but as tools improve the results will likely improve as well.

1 See http://republicofletters.stanford.edu/; http://ckcc.huygens.knaw.nl ; http://dalembert.academie-sciences.fr/ ; for a much broader 'union catalogue of early modern letters' see Early Modern Letters Online at http://emlo.bodleian.ox.ac.uk;a list of 'Some Sources for Early Modern Letters' from humanists is available at http://warburg.sas.ac.uk/research/projects/scaliger/sources-for-early-modern-letters/

2 See for this last for example Luc Duerloo, Dynasty and Piety: Archduke Albert (1598-1621) and Habsburg Political Culture in an Age of Religious Wars, (London, 2013); Francesco Benigno, “Der Adel in den Italienischen Provinzen der Spanischen Monarchie im 17. Jahrhundert”, in: Ronald G. Asch Der europäische Adel im Ancien Regime. Von der Krise der ständischen Monarchien bis zur Revolution (ca.1600-1789, (Cologne, Vienna, 2007), 385-406; Anna Maria Rao and Stainar Supphelen “Power Elites and Dependent Territories” in: Wolfgang Reinhard (ed.), Power Elites and State Building, (Oxford, 1996), 79-99

3 Data on bases of the Constantijn Huygens correspondence as included in the epistolarium at ckcc.huygens.knaw.nl.

4 Rik Hoekstra, “Correspondentie totaal. Patronen en trends in de briefwisseling van Willem van Oranje”, Eef Dijkhof, Michel van Gent (red.), Uit diverse bronnen gelicht. Opstellen aangeboden aan Hans Smit ter gelegenheid van zijn vijfenzestigste verjaardag, (The Hague, 2007), 117-131, pp. 127-129, also available at https://www.academia.edu/1340538/Correspondentie_totaal_patronen_en_trends_in_de_briefwisseling_van_Willem_van_Oranje

5 See for example http://ckcc.huygens.knaw.nl