Under the direction of computer science professor Eric Alexander, Liz Nichols ’20 (Aurora, Colo.) spent the summer analyzing speech patterns in Shakespeare’s plays.
Project: “We used language processing software to break the text from Shakespeare’s plays into phonemes, which are the individual sounds that make up speech. We looked at the percentage of each character’s lines that were made up of a given phoneme, then plotted them as 42-dimensional vectors.”
Goal: “We wanted to see if you could distinguish one character from another based on how they sound—not just the words they use, but the quality of their speech. Do fat, rich men tend to use round vowels? Do anxious characters have clipped, higher-sounding speech? We quantified and compared the phonemes in each character’s dialogue, and built tools that scholars can use to explore these questions.
“My co-researcher, Estelle Bayer ’19 (Seaside, Calif.), compared the language sounds of protagonists, antagonists, and fools. For my project I made a tool to graph each character’s speech, visually comparing it to the average among all Shakespeare characters.”
Relevance: “We want to add the strength of computing to the field of literary analysis. Digital tools can give Shakespeare scholars more questions to ask, such as, ‘Why does this character have some of the darkest-sounding speech in all of Shakespeare?’ or ‘Why does this protagonist sound so different from all the other protagonists?’ Shakespeare’s characters tend to have very distinct voices, and we can quantify that.”
Fun fact: “Linguistics research has established a light to dark scale for vowels. Light or ‘front’ vowels—like ee—are made with the tongue in the front of the mouth, whereas dark or ‘back’ vowels—like oo—are made with the tongue at the back of the throat. This research is used often in brand marketing. For example, cleaning product brand names tend to sound bright. Swiffer somehow sounds cleaner than swaffer or swoeffer. People have broad associations with certain sounds that are universal across all languages—if I show you a small horse and a large horse and ask which one’s Kit and which one’s Kat, you’re likely to say the little one’s Kit and the big one’s Kat.”