A first look at Mary Hamilton’s social networks

Description

Abstract
Social Network Analysis is a vital tool for historians of all kinds, including (socio)historical linguists looking for possible correlations between network ties and the spread of linguistic features (Milroy 1980, Milroy & Milroy 1985, Marshall 2004). This paper showcases an innovative method for mapping the social networks of hundreds of participants in a large and fascinating archive of letters, diaries and manuscript books dated c.1740-c.1850, The Mary Hamilton Papers.

Network-informed work on the history of the English language typically assigns a tie strength to every individual pair of persons, whether ‘strong’ vs. ‘weak’ or a numerical score (e.g. Bax 2000, Tieken-Boon van Ostade 2000, Bergs 2005, Henstra 2008, Sairio 2009). I argue that such methods are impractical or inappropriate for the number of participants in The Mary Hamilton Papers and the shape of a family archive centred, unsurprisingly, on one person. In our approach, by contrast, networks are mapped by means of data taken from the existing mark-up of the digital edition of the Papers, combining metadata from all relevant material in the archive, whether or not transcribed, with information derived from people mentioned (‘mentionees’) in transcribed letters and diaries. The strength of ties between individuals is derived from the frequencies of co-occurrence of author-recipient, author-mentionee, recipient-mentionee and mentionee-mentionee pairings, counted once each per letter or day's entry. The contribution of each type of frequency is weighted by careful sampling. By going beyond the metadata and utilising textual mark-up already present in the transcription files, we can begin to show not only relationships involving Mary Hamilton or her husband (who between them are author or recipient in the bulk of the material) but relationships between others – family members, friends and acquaintances, courtiers, members of the Bluestocking circle, and so on. The reconstructed networks can even display important relationships involving persons with little or no correspondence in the archive, such as children and servants.
The paper reviews some of the challenges encountered in developing a data-driven methodology (one which has some analogues in cell biology, scientometrics, etc.). I discuss the solutions we have arrived at and ask whether the proposed method is objective and replicable, how well it works for The Mary Hamilton Papers, and whether it could be applied to other collections.

Period	4 Jan 2023 → 6 Jan 2023
Event title	British Society for Eighteenth-Century Studies
Event type	Conference
Conference number	52
Location	Oxford, United KingdomShow on map
Degree of Recognition	National