Text Analysis for History Classes with Voyant

The interdisciplinarity of digital humanities is one of the field’s greatest strengths – that’s a truism, but of course it’s also a truth. DH tools help us capitalize on that interdisciplinarity even in our own field-specific, core courses. It’s nothing new to incorporate cultural products like poetry into a history class. Poems, songs, paintings – all of these make for engaging and perceptive primary source discussions and help students think not just about events but about ideologies, values, uses of the past and visions of the future. Digital tools can help us change our gaze slightly and read those primary sources through a slightly different lens.

One of my favorite and most successful deployments of poetry as a primary source was in a modern world history course during a discussion of immigration, race and racism, and nationalisms in the early 20th century. We paired Langston Hughes’ “Let America Be America Again,” published in 1936, with Kahlil Gibran’s 1926 “Message to Young Americans of Syrian Origin.” Both engage with questions of citizenship, the immigrant experience, the construction and weaponization of racial and ethnic categories, and other ideas that remain as pertinent today as they were in the 1920s and 30s.

Last time I taught these poems, I did it in the “traditional” way: I assigned them as readings; had students annotate them and come with thoughts, questions, and observations; and placed them at the center of our primary source discussion. It was incredibly productive, and students found the poems both helpful for contextualizing the historical moment and beautiful in their own right. But what if we could do more?

Enter Voyant, an online tool for text analysis. The visualizations and statistics it produces help us consider the text in a different way. When I read a poem (and anecdotally, when my students do), we’re looking at phrases and stanzas, and we’re following a narrative or line of argument. But the computer takes the poem as a corpus, as a collection of words (and the ideas attached to them). So it can help us to view the text at slightly more of a distance.

Langston Hughes, “Let America Be America Again,” 1936

Kahlil Gibran, “Message to Young Americans of Syrian Origin,” 1926

Even just these word clouds help tease out some similarities and differences in the tone and message of each poem: the prominence of “dream” in Hughes and “believe” in Gibran; the juxtaposition of Hughes’ “land” and Gibran’s “labor”; that despite the prominence of Gibran’s “freedom,” “America” appears only once in the entire poem, while it is the most frequent word in Hughes.



The Links graph helps to further contextualize those words and gives us some sense of their role within the text, the values assigned to them. These observations tell us very little about history in isolation, but they offer exciting avenues for conversation and investigation as students seek to connect text with context.

One of the dangers of DH’s interdisciplinarity is that we sometimes borrow too uncritically from other fields’ vocabularies and methods without considering the theory and discourses that have produced them. So to be clear, I’m not attempting to reinvent the field of text analysis. I’m not claiming to have discovered it for historians either – it’s been around for a very long time and in a lot of contexts. In fact my “corpus” of just two short poems might make scholars of text analysis scream. So as with any DH tool or method, I strongly encourage you to dig into the field of text analysis a little bit and try to peg down some of its assumptions and goals – not least because that too can be a fruitful discussion with students about interdisciplinarity, different frames for reading, and the role of technology in mediating information! There are tons of resources out there, but I found this workbook by Brandon Walsh and Sarah Horowitz to be an excellent starting point.

If you use Voyant in a history class, I hope you share your experiences and any insights you have to offer! I’d love to hear about your successes, your failures, and your ideas for improvement.

Leave a Reply

Your email address will not be published. Required fields are marked *