Many disciplines in the scope of humanities has doctrines and/or techniques which have developed and reached to today by passing from person to person. Moreover, some other disciplines are directly involved with the observation of human relations. Therefore, studies involving applications of social networks are applicable almost all branches of humanities. Three previous examples of such applications are chosen among the abstracts of Digital Humanities 2013 to be discussed in this article.
Extraction and Analysis of Character Interaction Networks From Plays and Movies
The first abstract is from literature. In their work, Suen, Kuenzel, and Gil have selected 173 plays and 580 movie scripts to be examined. Four different algorithms for extracting interaction networks was applied to determine to most useful one. Then the social networks appearing in the texts were converted into weighted graphs using the best-performing algorithm for each document. Obtained weighted graphs – one for each play / movie – were analyzed for their statistical values, coefficients. Among these are clustering coefficient, single character / relationship centrality, their variances, number of characters, etc. On the other hand, properties of the texts from which the weighted graphs are extracted are used for the classification. Whether it is a play or a movie, its date, its ratings, its genre, and the author are such properties.
Results of this study showed that plays and movies are easily distinguishable. Plays are more likely to have a main character with sharp distinction and every other character is placed around this main character. It was also seen that plays tended to have more disjoint groups of characters before 1800s. The study also justified some genre stereotypes and assumptions. Horror movies have specific statistics regarding their interaction networks while those of romance and comedy movies can be said to be almost the same. The work has olsa confirmed one of the defining characteristics of Shakespeare, which is the drastical changes in the importance of the mains characters among his plays. Lastly, as expected, no correlation is found between the statistical values of the interactions networks and the rankings that the play/movie got.
A social network analysis of Rousseau’s autobiography “Les Confessions”
The second work which is to be presented here is a similar application of social networks in search of answers to different questions. Rochet, Bornet, and Kaplan analyzed Jean–Jacques Rousseau’s autobiography “Les Confessions” which comprises 50 years of his life. The network is again represented as a weighted graph using an algorithm which reasonable adjusts the weights of the edges and prevents erroneous edges to appear. Construction resulted in an undirected connected graph which is composed of 435 vertices and 3572 edges.
In the analysis part of this work, authors of the paper found out that the diameter of the constructed graph was 10 – away from 4, diameter of a random graph – which is explained by the networkal distance between the persons in his earlier years and those in later years. It was also revealed that there was a high ratio of closeness of triplets of nodes and a high global clustering coefficient which indicates a pretty sharp grouping of people in his memories / thoughts. Interestingly, the degrees of vertices and the weights of edges were realized to be distributed as power-law shape, and distribution of edges has mode 6 which points out that J. J. Rousseau mentioned the characters either at least a few times or with many other characters.
Documentary Social Networks: Collective Biographies of Women
The third paper – last, but not least – is about a more general study on biographies of women conducted by Booth and Martin. The authors introduce a new term “documentary social network” for the network of social relations which can be traced by biographical documents and the term “documentary siblings” represents two persons who share a link on this documentary social network. With the idea of creating a documentary social network in mind, more than 13000 biographies of more than 8000 women with different nationalities and occupations from diffirent centuries are scanned.
These prosopographical documentary social networks reveal unexpected connections between different types of women. Paper examines the network by centralizing a devoted nurse Sister Dora and analyzing her connections to a celebrity Lola Montez, a quenn Anne Boleyn, and a Gothic novelist Ann Radcliffe. At the end of their analysis, the authors of the paper claim that “digital exploration of the persons, narratives, collections, and documentary social networks in the CBW genre opens a prolific field of ramifying versions of female biography impossible to decipher through an approach to individuals or single full-length biography, and invisible through the customary lenses of historiography, time and period, place and nationality”.
To sum up, analyzing the social networks appear in any field of humanities opens a gate through a new understanding in those fields. Different possible kinds of analysis in different areas offer a huge package of new information which were not visible with traditional methods.
1. Suen, Kuenzel, Gil. (2013, July). Extraction and Analysis of Character Interaction Networks From Plays and Movies. Retrieved from : http://dh2013.unl.edu/abstracts/ab-251.html
2. Rochat, Bornet, Kaplan. (2013, July). A social network analysis of Rousseau’s autobiography “Les Confessions”. Retrieved from : http://dh2013.unl.edu/abstracts/ab-405.html
3. Booth, Martin. (2013, July). Documentary Social Networks: Collective Biographies of Women. Retrieved from : http://dh2013.unl.edu/abstracts/ab-289.html