This blog is a sum up of three articles about the construction and the use of a social network analysis.

The first article is titled “converting medieval documents into a searchable database”. It describes the work achieved by the collaboration of five universities. They implemented numerical techniques to automatically store texts pertaining to a German city from the year 601 to 1545. In few words, it is system, which extract all metadata of a book/text, using typographical means. Then it cleans information in order to create a sort of index: the typical output is a year, a place of issue, often names, signatories and archival information. Also it provides an extensive index with more details on each character like place of living and relationship.

The goal of this article is to show how relatively simple techniques can be applied to historical collection and thereby make it much more accessible and useful.

As it can be now very rapid to find the right document for a specific research, it would be relevant to know to what extent this research can be interesting. This is well illustrated in the second article from Cornell-Alexander Jackson, entitled “using social network analysis to reveal unseen relationships in medieval Scotland”. The title is pretty clear about the topic. Though the author prove, how social network analysis can be trivial to understand some circumstances. The first point of this article explains that relationships can shape human’s behavior above and beyond the influence of his proper attributes. Because of this, it would be relevant to study an historical situation, through a network analysis of its main characters. The study of Mr. Jackson is a full analysis of all the connections between the governors of Scotland under the reign of William I (XI century). During his researches he noticed that one of the governors, Duncan II, had more private connections in politics than the others. Historians have no idea how prominent was Duncan II during his time; they have no idea whether or not he played a role in running the Scottish government. There is also a second research solution for this kind of study, which is to create a location network, but in this Scottish case there wasn’t enough data.
As if this research is not achieved yet, the author has already discovered unknown things. At least, this article shows that a network analysis can bring a new perspective for studying history.

There are other projects that have recently started to use this network analysis method. One special example is a network, which connects all the Saints in medieval Italy. This network is rather unique because all the connections are not based on texts, but on paintings. This work led by Thomas Lombardi is very rigorous. This is the subject of the third article. Thomas Lombardi built each connection link with a confidence coefficient. He also added a weight for each link. The result is a huge network hierarchized with a thousand of connections. The goal of this work is to gauge the relative prestige of each Saints before and after the Black Death (1346–1353). In other words, the issue is to determine if the perception of these saints have changed in the eyes of painters and their clients.

It is very instructive to discover a new approach for analyzing historical situation. Perhaps this method should be applied to understand some mysterious stories happening today. I would be very curious to see huge network linking our politicians together and with their relations!

References

Dharchive.org, (2014). DHArchive. Available at:http://dharchive.org/paper/DH2014/Poster-720.xml [Accessed 19 Oct. 2014].

Jackson, C. (2014). DHArchive. Dharchive.org. Available at:http://dharchive.org/paper/DH2014/Paper-881.xml [Accessed 21 Oct. 2014].

Lombardi, T. (2014). DHArchive. Dharchive.org. Available at:http://dharchive.org/paper/DH2014/Paper-643.xml [Accessed 21 Oct. 2014].

Advertisements