Quantitative analysis - Amauta
Hanno and Jörg: thank you for this fascinating presentation. I would like to ask you, first, about the quantitative discourse analysis you outline. I've not encountered this before, and I would be interested to hear more about your rationale, and what insights this approach enables. If I understood it correctly, the diagram you showed represents the frequency of certain key words and phrases (?) in Amauta - is that right? I can see that that might reveal patterns of influence, and the emergence of particular terms and rhetorical devices. But I didn't grasp how the diagram represents this information - there doesn't appear to be a temporal dimension, or any hierarchy in the numbered references. Did I completely misunderstand it? I look forward to hearing. Many thanks.
thank you for your question; we are happy to explain. We were interested in distinguishing distinct focuses of the discourse on Mexico in Amauta in the 35 texts and providing quantitative evidence for the results. We used a methodology which becomes more and more common in the digital humanities, namely Vectors Space Models or VSMs. Based on word frequencies, this approach calculates the thematic closeness (or distance) between texts, in this case: journal articles. It is well suited for a quick overview of a large text corpus, and we were ourselves astonished by the clear results we gained from it, which provided a starting point for further qualitative explorations. The methodology is described in more detail in Andrew Piper, Enumerations. Data and Literary Study. Chicago: The University of Chicago Press 2018: 45–49.
Our diagram uses the most frequent terms (after the removal of stopwords) in all of the 35 articles from Amauta, but does not make use of phrases or rhetorical devices or takes temporality into view. Also, it does not provide a hierarchy.
On a related note, Simon Hengchen et al have used this methodology with regard to the 'vocabulary of the nation' in several newspapers with a broad temporal range: Simon Hengchen, Ruben Ros, Jani Marjanen, A data-driven approach to the changing vocabulary of the ‘nation’ in English, Dutch, Swedish and Finnish newspapers, 1750-1950, see https://dev.clariah.nl/files/dh2019/boa/0791.html
Does this answer your questions in a satisfactory way?
With best regards, Jörg