Uncategorized

Blog 2 – Voyant

In beginning my tinkering with on Voyant, I entered the full text of Herland in the search, as entering the URL of the e-text did not at all give accurate results regarding the contents of the text. Once being navigated to the Corpus, it became clear that the novel at least has something to do with women and men named Terry and Jeff, as these were among the most frequently used words. The word “country” also popped up quite a bit (137 times, compared to “Jeff” being mentioned 152 times); this caused me to assume the women and these men had something to do with a country. Looking back after reading the first few chapters, it’s safe to say that is correct.

After exploring the tools Voyant has to offer, Correlation, Bubblelines, and WordTree are among my favorite features to use. To further investigate the significance of the most popular words in the book, I decided to narrow my search to “Terry” and “women”, as the Cirrus and Summary tools revealed those to be the two most frequently used words at 247 and 209 words, respectively. Beginning with Correlation, I found that “women” was most highly correlated with “atmosphere”, “horses”, “speaking”, “tutor”, “cattle”, and “walled”. With “Terry”, “can’t”, “kill”, “advanced”, “climbed”, and “colored” were among the most highly correlated. Interestingly enough, the highly correlated words for both “Terry” and “women” seemed to perfectly match up with the frequency of use per section seen in the Trends tool. Moving on to the Bubblelines tool, which appears to be a similar tool to Trends, it is apparent that “Terry” appears consistently up until around ¾ of the way through the novel, in which it then suddenly drops off and is barely used. This begs the question, “What happened to Terry?” Around this same time, “women” is used less, but is still used more consistently. The last tool I found quite useful is WordTree, which is exactly what the name suggests – a word tree. I typed in “women”, and the results were quite interesting and different from what I saw in previous tools used. The words connected with women were “careful”, “multitude”, and “parthenogenic”. This, to me, suggests a somewhat sinister, perplexing twist to the novel.

Although Voyant is a helpful tool in trying to figure out what are the crucial terms and ties between those terms in a novel, for a novice at the program such as me, that is all it is – trying. I suppose, in a way, it is similar to asking someone for a quick synopsis of the book in passing. Another important point, though, is that this program seems to be more statistical and technical than it is focused on the true literary analysis of the text, something I am more acquainted with. With that said, it would be more user-friendly if Voyant had an easier tutorial or had an “About” tab when you put your mouse overtop a tool, such as Bubblelines; that way, the user knows for sure exactly what they are analyzing as well as how to make better use of it. Overall, though, Voyant is an exceptional tool for quickly analyzing a text’s most important terms, aiding in deciphering a text before even reading it.

5 Comments

  • molwils

    I am so glad I am not the only one who just tried to enter the url! I too, narrowed my search to “Terry” at one point. It was interesting to see the ups and downs of his role in the novel. “Although Voyant is a helpful tool in trying to figure out what are the crucial terms and ties between those terms in a novel, for a novice at the program such as me, that is all it is – trying.” I completely agree with you on this statement. I definitely felt like I was trying and failing a lot more than I was succeeding!

  • wschmid

    Making sure you get just the proper text loaded took me a few tries as well. The copyright and disclaimers really skew the data. It sounds like you have a pretty good grasp on using Voyant. Comparing the correlations of Terry with the correlations of Women, I think is just what we’re supposed to do. Play around until the tools beg questions of us. That gives us something to look for when we’re close-reading. I agree with you on their support documents. They can be a bit unclear.

  • laujack

    I found the word tree tool to be helpful as well. It was interesting to use the other tools (cirrus and the summary) to find initial information, but word tree helped me to make connections between these words. It seems like you have a good grasp of the tools on Voyant. And your description of how you proceeded was really well written and helped me to have a better understanding of some of the things that I was struggling with. I have always had more emphasis on analysis of the text’s meaning then analysis of the text so this was an interesting experiment for me as well.

  • kcookso

    You mention adding an “about” tab in the website, this would be very beneficial to beginners (like us). Hopefully they are still tinkering with their own website and are adding new things all the time to make it easier and more user friendly.

  • Prof. DeSpain

    You’ve done some excellent work here, Colleen. I can imagine doing a full corpus study for the final project as an experiment and seeing what broader trends are revealed.