Original data file, sporting DOS/Windows line-ending characters and inconsistent quoting behavior that breaks github's display of delimited files, was originally found here:
However, Many Eyes and the website above no longer exist (https://www.computerworld.com/article/2930326/ibm-to-shutter-dataviz-pioneer-many-eyes.html).
The file downloaded above was very lightly processed to yield the more machine-readable file:
lotr_wordsSpoken.tsv
Actually all I did was read it into R and write it back out to file.
Context:
When creating simple examples of automated data analysis pipelines for STAT 545A, I didn't want to use simulated data and I was sick of Gapminder. So I did this. Of course.