This website is the online home for Visualizing Tolkien with R. It serves two functions: (1) it is an interactive repository of basic natural language processing (NLP) techniques used on Tolkien's many fictional works, and (2) it is designed so that readers with no coding knowledge can follow along and write their own code.  Attached files include the R script (code), visualizations, and analyses. This project places special focus on these texts in particular:

• Leaf by Niggle

• The Hobbit

• The Lord of the Rings

• The Silmarillion

• Farmer Giles of Ham

​

This project uses these methods of NLP in particular: 

• token distribution analysis

• sentiment analysis, including relative sentiment

• topic modelling

​

You can find my GitHub repository for this project, too.

​

______________________________________________________________________________________________

​

This project is partially funded by the Critical Digital Humanities Initiative at the University of Toronto, under the

DH Training Bursary program. Further information about the program can be found here.

​

______________________________________________________________________________________________

​

A note about my methodology

A literature review on Tolkien studies and Computational Literary Analysis

______________________________________________________________________________________________

​

Why not use other NLP tools, like Voyant, to process my texts?

 

I choose to use R for exploration and flexibility needs.

​

First, I need to see how different tools can give me different types of results; there are many tools used for natural language processing in the humanities. We can use pre-made applications like Voyant or the Gale Digital Scholar Lab, which will give us a very interactive, easy-to-understand layout with many options for distant reading. With these pre-made applications, data is clearly organized, and running distant reading tests are quick and straightforward. It is my intent to run the same files I use here in Visualizing Tolkien with R through Voyant and Gale, just to see the variety of distant reading tests I can use and add to my toolbox of computational literary analysis. However, with make-it-yourself projects like using R or Python, humanities scholars are afforded a different, more intimate view of the project.

​

I also have a desire for flexibility on the project. With Voyant, we can upload many different file types, like RSS feeds, XML, and text files, without needing to learn how to write code that will make the application accept these file types. However, the Visualizing Tolkien with R project only needs a program to accept text (.txt) files, and I want to explore the greater degrees of freedom given to me when designing my own commands to run the tests I want. In a sense, the reason I choose R instead of Voyant or Gale was to explore distant reading on my own terms. In the future, when designing rigorous research questions about Tolkien studies, I need to decide if pre-made applications are enough, or if a thorough make-it-yourself program through R might be needed for accurate results. For this project, though, R is the tool that I need.

​