While this Lab was not very hard, at least for me, it was definitely one of the most tedious labs we have done. To my surprise, I got the hang of using Terminal and MALLET pretty easily and quickly. Hands down the most annoying part of all this was the attention to spelling. I found myself having to go back 3-4 times to rewrite commands. To be honest, when we first started discussing and working on this lab, I did not see the usefulness in learning or understanding any of this. However, after completing the lab, I now see the value in understanding both MALLET and Terminal, in terms of literary analysis and overall knowledge of computers.
In my experiment I used the State of the Union addresses Corpus from 1951-2000. Due to the large number of texts in my corpus, I decided to focus on just 5 topics. I figured by keeping the number smaller, I would get a more refined or narrowed set of data to go off, just as Jockers suggests in chapter 8 of Macroanalysis. Whether or not that is actually the case, I’m still a little unsure of, but we’ll go with it. Going into it, I thought using the State of the Union addresses would be a little bit of a challenge just based on number of documents, but it really didn’t make too much of a difference. The much bigger challenge was making sense of the data that came out of using MALLET. There are some portions of the data that make sense, but for the most part, I was pretty confused with what all of this meant in terms of textual significance. Obviously, we are given the most prevalent topics and when/where they are used, as well as their frequency, but as we have discussed a number of times in class, this is only a minor piece in the puzzle in making sense of it all. After completing this Lab it is becoming more and more apparent as to why close reading plays such a vital role in this class, as well as in life.
The topic that received the highest frequency out of the five I decided to examine was: people congress year american world government years time nation america make peace budget americans great tax federal union president today. This seemed to make sense given the corpus I was working with. I just used the topic with the highest number (2.74), not sure if that’s actually correct, but that’s what we’re going with.
Overall, I feel as though this Lab has further opened my eyes to the power of computers and their value in literary analysis. Topic Modeling tools, such as MALLET, can help us understand the vast relationships and differences between various texts in the most efficient way possible. As I stated above, close reading is one of the most valuable skills any literary scholar can acquire. However, the relationship between close reading and distant reading is one that is very much a two-way street. Furthermore, tools like MALLET, allow us as readers the ability to develop and implement our close reading skills in a way that could not even be imagined if not for the power of computers. While the usage and understanding of these various technologies are by no means easily attained, after some practice, as with anything else in life, value is brought to the forefront and purpose is made much more clear.