I've been trying to brainstorm ways of representing the "canon" of literature that Amazon.com produces when it suggests "related" books to users. While there are many problems in doing this research (i.e. Amazon personalizes everything, so my Amazon recommendations are different from someone else's…), I just wanted to do a trial experiment with social network analysis. Here were my rules in studying Amazon.com:

1. I would start with Zora Neale Hurston's Their Eyes Were Watching God, and this would be in my "source" column.

2. I would then scroll down to the most privileged placement of a suggested book that Amazon.com members had also bought alongside Their Eyes. In this case, it was F. Scott Fitzgerald's The Great Gatsby. This text would then be put in my "target" column. 

3. I would then click on Fitzgerald's The Great Gatsby (now this was my next entry in my "source" column, and I would see the book associated with the Fitzgerald text in, again, the most privileged spot on the "other customers purchased this" section. This was J.D. Salinger's The Catcher in the Rye.

…Then I repeated that several times over, except in the case where the suggested book in the most privileged place was one that was already in my "source" column twice over, in which case I would choose the next over (these differences in privileged position on Amazon became my "type" differences). 

It turned out that it was really hard to get off Fitzgerald. It was like a circle, always going back to all these writers I read in high school, whereas nothing ever led back to Zora Neale Hurston (or even a female or African American writer). While this sample size is WAY TOO SMALL to draw any conclusions, it would be interesting to visualize a pattern on Amazon.com that circles back to a small canon of white writers. Despite being the largest bookseller in the U.S., if not internationally, Amazon.com might produce a relatively small circle of suggestions. Here is my graph in Gephi:

You can see, from this graph, that Amazon kept suggesting a relatively small circle of authors that always ended up going back to each other. Hurston, however, is an outlier--a search never goes back to her works. Hurston is associated with other canonical modernist texts, but those texts aren't associated with her. I wonder what would happen with lesser known modernist (or other writers) like, say, Jamaica Kincaid, Sterling Brown, etc...