Traversing the Graph: the Coded Text of the Prelims Project

Hello and welcome back! Two weeks have passed and it is time go back to the Preliminaries Project graph and take a fresh look at what  lies hidden within its multicolored mini-universe of nodes and edges and digitally encoded information. Sound daunting? It is. Coming from a primarily humanities background, I still retain a slight fear of the digital world; it intrigues me and at the same time  it frightens me. Doesn’t look like a book…Doesn’t smell like a book…Hey that’s not a book! Humanities, welcome to the 21st century.

Although my Gephi generated friend and  constant companion the Preliminaries graph is most definitely NOT a book, it does have some similarities with something like–I don’t know–maybe a historical text written in Ancient Greek: it contains a lot of information if you know how to read it. For a guy like me, reading the Preliminaries graph is a ground-up process, I have to decode as I go. Obviously there are a variety of excellent texts and websites to learn about social network analysis; however, it is important to recognize that each data set is unique and to approach it as such. This blog post will share the experience of my first attempt to crack the code of the Preliminaries graph.

A quick note about functions:

From now on I will include any code that I write as a hyperlink to a Gist in which the code is stored. Feel free to download them, play with them, change them, or leave comments. Remember I am a Python (programming) rookie so please be kind!

Ever since I started looking at these kinds of graph I have been particularly interested in the measure Betweenness Centrality because I feel that it is an effective way of measuring a node’s importance in a graph. It is a measure of the number of Shortest Paths that pass through the node, and is an excellent measure of the load placed upon the node in terms of mainting close conectivity between the various communities of a network. I decided to use this measure as a starting point (fig.1) and everything else just started falling in place.

Figure 1 Prelims Graph in Black and White

The above figure is the Preliminaries graph rendered with the OpenOrd algorithm with node size adjusted for betweenness centrality and a size range from 5-40. Obviously, Lope de Vega dominates the graph, but other nodes also stand out: the Count of Lemos, the Duke of Lerma, Felipe Bernardo de Castillo, Gonzalo de Céspedes y Meneses, and Miguel de Cervantes. While each of these other important nodes will recieve further  investigation, today I will focus on the results of a line of inquiry that starts, but does not finish, with Lope de Vega.

First I decided to take a look at Lope’s Publication Network(fig.2). In the previous blog I introduced the concept of the publication network as an important way to understand the social circles associated with literary production. In the Prelims data set a publication network is an ego or neighbors network that extends to four degrees of seperation. This range includes the editors, censors, and other individuals important in the formal process of publication. The publication network also includes any other individuals that are more directly connected to the author: personal realtionships, patronage, dedications in the form of poetry or letters etc. Using the new find_neighbors function I wrote to identify neighbors networks at n degrees of seperation, I determined that Lope’s publication network consists of 1083 nodes or 67% of the nodes in the graph. Ok, he was well connected, that is something we can assume based on the extremely prolific nature of his literary production…so what can the Prelims graph tell us? Well for starters, it can tell who wasn’t in his publication network.

Fig. 2 Lope de Vega’s publication network

Lope de Vega.color = red

In order to do this, I removed the subset of nodes representing Lope’s publication network from the other nodes that make up the graph, then using another new function filter_by_type combined with return_label I was able to return a list of the names of all of the people who are not in Lope’s publication network. A quick review of this list produced some interesting results: I found several authors based in Spain including Gonzalo de Céspedes y Meneses and El Inca Garcilaso, and two authors active in Peru, Diego Dávalos y Figueroa and Pedro de Oña. While a quick look at both Céspedes and El Inca produced interesting results, the two Peruvian authors–actually neither was really Peruvian, Oña was born in Chile and Dávalos in Spain, but they both published books in Lima–attracted my attention. In this period social circles were indeed highly influenced by geography, and it is logical that these authors find themselves at the periphary of a network centered geographically in Madrid; however, I was curious about how far removed they actually are from Lope’s network and furthermore,  how their situation compares to authors working in New Spain (modern Mexico). Using set_intersect to indentify intersecting portions of the various neighbors networks, I found that both Pedro de Oña and Diego Dávalos y Figueroa are connected to Lope’s network at 3 degrees of seperation through their dedications to the Viceroy of Peru Luis de Velasco y Castilla, and at four degrees through Juan de Zúniga, Diego de Ojeda, and the Order of Santiago (fig. 3 and fig. 4)

Fig. 3 Diego Dávalos y Figueroa and Lope de Vega

Lope de Vega.color = red

Diego Dávalos y Figueroa.color = yellow

set_intersect.color = blue

 Fig. 4 Pedro de Oña’s and Lope de Vega

Lope de Vega.color = red

Pedro de Oña.color = yellow

set_intersect.color = blue

As previously noted, the relative lack of connections with the Madrid circle could be inferred based on geographic concerns, so in order to put this in perspective, I decided to take a look at several writers active in New Spain, which although located in the Americas was considerably closer to the center of Spanish literary production. I chose Bernardo de Balbuena an author and politition, who published in Mexico City and Spain, and Juan de Torquemada, a Franciscan chronicler who published in Madrid, both of whom were quite well connected in Spain as shown by the following images(fig.5 and fig. 6).  It is important to note here that there is considerable variation between the networks of the two authors associated with Mexico, an avenue that I hope to explore in the future.

 Figure 5 Juan de Torquemada y Lope de Vega

Lope de Vega.color = red

Juan de Torquemada.color = yellow

set_intersect.color = blue

Figure 6 Bernardo de Balbuena and Lope de Vega

Lope de Vega.color = red

Juan de Torquemada.color = yellow

set_intersect.color = blue

In order to compare the “Mexican” authors with the “Peruvian” authors, I combined the four social networks based on geographic constraints: however, I found that at 4 degrees of seperation there was no direct overlap, so I upped the parameter in find_neighbors to 5 degrees of seperation and produced the following image in which the “Mexican” network is red, the “Peruvian” is yellow, and the overlap is blues(fig.7).

Figure 7 American Connections

“Mexican” Authors.color = red

“Peruvian” Authors.color = yellow

set_intersect.color = blue

As we can see here, even at five degrees of seperation there are few overlaps between the networks. In a way this is a good thing, because it allows us to focus in easily on important points. Coming from the Peruvian side, we can see that the Order of Santiango generates an important connection originating from the Viceroy of Peru, Luis de Velasco and connecting with several individuals who are shared by both social circles(fig. 8).

Figure 8 The Order of Santiago

Also, we can see a connection through Juan de Zúñiga that originates with the Erratas of Miscelánea Austral by Dávalos y Figueroa(fig.9).

Figure 9 Erratas of Miscelánea Austral

Following this connection a bit further, we arrive at what I consider to be very interesting territory: the importance of the House of Zúñiga in the Americas(fig.10).

Figure 10 The House of Zúñiga:Peru

It is well known that the house of Zúniga was powerful in both Spain and the Americas, and also that certain members of this house were important patrons of the arts and literature; however, I don’t know if their role in transatlantic literary production has been adequetly explored–my next mission will be to search for studies that focus on this noble family. Let’s follow this path through the graph a bit more and see where it goes (fig.11).

Figure 11 House of Zúñiga:Mexico

Here we can see the obvious importance of the Zúñiga’s in Mexico during this period (an Archbishop and a Viceroy). This illustrates not only the political importance of the house of Zúñiga in New Spain, but also the importance of  political figures/nobility in publication and how the members of one house can spread their cultural influence throughout geographic space. To take this concept one step further, let’s follow the Zúñigas back the Spain(fig. 12).

Figure 12 House of Zúñiga: Spain

Here we find Alonso López de Zúñiga y Pérez de Guzmán, the Duke of Béjar, and–not surpisingly for all the quixotephiles out there–the first part of Don Quixote. It turns out that American authors were not the only ones soliciting support from the House of Zúñiga: Miguel de Cervantes dedicated part 1 of Don Quixote to the famous Duque de Béjar…hmmm. Interesting? Could be, at the very least worthy of further inquiry.

Here I have conducted a very brief first look at the Prelims graph. Although all of the above connections deserve more attention and some might even merit some real-deal academic concern, it is too early to say with any certainty. For the time being I will continue trying to crack the code of the Prelims graph and faithfully provide any results I can find. Two steps: 1- Hit the books, and then 2- do some old-fashioned number crunching using the Python software package NetworkX and produces some custom .gexf files to use with Gephi…wait did I say old fashioned? HeHe…

Until next time: Saludos


P.S. Thanks versae for all the constructive criticism!


Filed under Uncategorized

4 Responses to Traversing the Graph: the Coded Text of the Prelims Project

  1. Awesome, David!

    That remembers me to my very initial research of the Zúñigas, Análisis preliminar de Preliminares: De Jesuítas y Zúñigas.

    What I found, briefly, is Casa de Zúñiga did control what arrived to Spain from America, and Jesuits did the same respect to what arrived to Spain from India.

  2. Yes Javi I am pretty much trying to do what you proposed in your ‘future work’ section. This project is communal CulturePlex all the way!

  3. BTW, take a look on this conference: SocialCom.

  4. Pingback: Math is the Path: Degree Distribution of the Prelims Network and Other Random Graphs | Preliminaries Project

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>