For the final project, Matt, Iris and I together have conducted a sentimental analysis based on Linn’s journal, and after combining it with the battles’ dates, we have some very interesting results. Before we started, we have envisioned several plans. For example, I was interested in the relationship between weather and memory/emotion and longed to explore that if weather functioned as an indicator for his emotional states or as a factor contributing to the change in sentiments. Another direction we could go is to analyze the sentiments in the journals as well as those in the letters, and when presenting both emotions from sources in the same visualization, we might find the difference (such as the letter has more positive sentiments and deliberately hides negative sentiments), which is valuable when we think about the audience and the objective of Linn’s writing. I still think that these could be the potential research questions for future plans regarding Linn’s journal. However, due to the limited time available, we finally chose to investigate the change in sentiments in Linn’s letter and its relationship with the battles Linn experienced.
After we determine our research question, we divided the journal into three parts so that everyone can mark up the emotions and share their impressions in the second meeting. I read the first part of the journal which is from the beginning to Feb 18th. However, we didn’t specify how to mark up and have not standardize our markups until the second meeting. For example, I marked up by words instead of ideas; considering that different readers might have different readings towards the same text, I was very cautious in only marking up the vocabularies that are definitely positive or negative. In addition, even though we made sure that we all used <state> and <note>, I marked up the word like “<state type= “emotion”>pleasant</state> <note>positive</note>.” Later, when we talked to Iris, she told us that it was more convenient for her to extract data if we marked up the text like: “pleasant<state type= “emotion”><note>positive</note></state>.” Therefore, I then cleared up my markup and made sure that it consistent with the rest of the text. Below the is screenshot of my markup (on GitHub):
Thanks to Iris for making the visualization to illustrate Linn’s changes in sentiments throughout the days. Below is the screenshot of the final visualization:
Here are the dates for the battles:
Battle of Roanoke Island: February 7-8, 1862 (Union victory)
Battle of New Bern: March 14, 1862 (Union victory)
Battle of South Mills: April 19, 1862 (Confederate victory)
After the first battle, Linn was deeply frustrated even though Union claimed victory. He had recorded in his journal of all the wounded soldiers and dead corpses, and his negative sentiments were also reflected in our graph. Before that, he was usually happy about the beautiful views of nature, and it seems like that he didn’t realize what war would be like. Therefore, he was extremely disillusioned after the first battle in his life. Linn did not write any journals during the second battle. While he seemed to be peaceful, the text suggested that he paid more attention to the details in life, and probably he did not know how to react to the battle that easily and absurdly destroyed the beauty of life. Finally, Linn seemed to be gave due to the loss of the Battle of South Mills. However, based on the journals, we don’t have many details about what happened after April 19th.
For the future projects, it would be interesting to see if there is any difference in what Linn valued before and after Linn’s experience of battles. Nonetheless, it is clear that he was most disillusioned after participating the first battle, and he did not really know how to cope with the latter battles and the violence he witnessed.
This is an interesting project and it is only a start. I am so happy that I learnt about TEI in this class, which I think is a powerful way to enhance “close reading” as well as to provide data for “distant reading.” I also learnt that standardization is especially important in group projects, and in the end I want to thank my wonderful teammates, Matt and Iris! I hope to see further developments of Linn’s project!