Some thoughts on “Aproximación cuantitativa al sistema de personajes de La Celestina”

Twitter has proved an excellent place to meet other people interested in Celestina and it was precisely through Twitter that I came to know Jeniffer Isasi’s blog and to read her entry “Aproximación cuantitativa al sistema de personajes de La Celestina“.

First of all, it must be said that her blog entry is far from being definitive, but it is merely a tentative approach to the possibilities offered by Digital Humanities tools for the study of Celestina. More specifically, the author decided to comment on the results she obtained after applying techniques of data mining to the text and use that to try to ascertain the nature and the importance of the relationships between characters in Celestina. On account of these relations, the author aims to discover if the central role given to the go-between by the title Celestina is justified or not.

This said, we can ignore certain flaws in her methodology that, obviously, relate to this being a simple experiment. For example, the decision of using the digital edition of Celestina at Project Gutenberg is based exclusively on the convenience of having a digital plain text available, and not on philological criteria. Something similar happens with the decision of using act XII as a reference (despite being completely justified by the lack of technical knowledge and time). Act XII is highly convinient because of its central role in the plot, but it is also an act in which the lovers and the servants are given much more preeminence as usual and in which the otherwise static action of Celestina becomes quite dynamic. It is also the act in which Celestina dies, a fact that the author does not emphasise as much as I would have liked her to do, as I am not sure if she is taking this detail into account when calculating the percentage presence of each character in the work.

This last is a major problem and poses relevant questions to her results. For example, having a look at her first table, Celestina speaks 21,85% of the time. This is certainly impressive if we take into account that she dies in act XII, that is nine acts before the end. In addition, this necessarily changes the approach to the results on how many times each character is mentioned by the others, as it is not the same being mentioned after death than while being alive. In fact, the other characters keeping the dead ones in mind is revealing, as we should not forget that the deaths of Celestina and the servants trigger Elicia’s and Areúsa’s vengeance. Moreover, if we take into account the Voyant’s graph (in which I miss an explanation of the sections in which she has divided the text and a lot of characters, the author not having explained why they are not included in her analysis or why she is not using the traditional system of acts), the fact that Calisto’s and Melibea’s lines increase from section 9, that is, the deaths of Celestina and the servants, could make a difference in the number of times she and he are mentioned, as both live only for themselves and each other and it is only logical that their names is the thing they will be saying most.

Leaving aside the problems posed by the nature of act XII, which determine very much the conclusions of the author, there is something that she only mentions marginally but in which I find the greatest contribution of her analysis: being able to determine who quantitatively leads the conversation. The (lack of) correspondence between the statistical results and the practical efficacy of the dialogue could reveal interesting aspects on the dialectic ability of characters. However, what interests the author most is to show that Calisto is the central figure in act XII and which bounds together all characters. This will not surprise any reader of Celestina, as it is obviously Calisto’s falling in love which triggers the action and, without him, nothing would have happened. Therefore, it is only obvious that he acts as the main link between characters, but it is always nice to have a visual confirmation of this.

With regard to the conclusions, they are not really different to what textual criticism has concluded in the last fifty years. And this brings me to something that tortures me everytime I have to deal with Digital Humanities: and now, what? Please, do not misinterpret me. I am great fan of applying new technologies to research, despite lacking the technical knowledge. However, everytime I am confronted with all these visualisation tools and data mining statistic results I cannot help feeling that they are completely void. That they are research interrupted directly after the data gathering stage. Consequently, I fear that developing these tools and generating research data will become an end in itself and I fell that many current Digital Humanities projects lack a strong interpretation component. This is not bad in itself, I mean, for sure someone will be able to make use of data and come up with a clever interpretation someday, if you give them all the facilities. But I fear this attitude will end up turning Digital Humanities into a self-centered discipline exclusively based on studying the possibilities of new technologies applied to the analysis of the traditional objects of study of the Humanities, not the ancillary tool of research I would like it to be. Not in the case of the above mentioned blog-post, of course, as it is obvious from the beginning that the author is exclusively messing around with data to see what happens and she would really be able to come up with an interpretation someday. However, it is still a good example of how raw data leave us “vacíos y secos”.

---

---