Class uses historic Cherokee archive, Skype to hone skills in linguistic fieldwork

A new collaboration between the Yale Department of Linguistics and the Beinecke Rare Book and Manuscript Library has resulted in an innovative approach to teaching students how to master the field methods used in linguistic research.

Typically, linguistic fieldwork is conducted verbally between a researcher and a native speaker of a language in a natural setting, such as the speaker’s community. Linguists ask questions about a language to understand aspects of it that cannot be learned by simply reading books about the language. It is an especially essential skill when studying languages that are not widely known or well documented.

For the first time this spring, Claire Bowern, associate professor and director of graduate studies in the Department of Linguistics, made use of the Beinecke’s Kilpatrick Collection of Cherokee manuscripts to teach students the skills needed in fieldwork to elicit linguistic data from a native Cherokee speaker. The class both took the unusual step of using historical texts as the basis for the interviews, and made use of Skype instead of face-to-face interviews.

“The class is a unique experiment in teaching field methods,” said Lisa Conathan, head of digital services at the Beinecke. “I’m not aware of any other time that this specific topic in linguistics has been taught this way.”

“Normally we work face-to-face, asking questions about the language that you don’t get the answer to by looking at the grammar,” said Bowern.

Bowern’s class — two undergraduates and four graduate students — studied the Cherokee language with native speaker Tom Belt, the Cherokee language program coordinator at Western Carolina University, using the Beinecke’s historical texts as source materials.

“We get excited about languages that are complicated and challenge us,” says Parker Brody, a first-year graduate student in the Department of Linguistics, who took Bowern’s class. “Working with these original materials is something new. They showed us interesting things, like word play, that you wouldn’t necessarily find just by asking a native speaker.”

In addition to Belt, students worked with Hartwell Francis, director of the Western Carolina University Cherokee Language Program, who was among the researchers to examine the Kilpatrick Collection in 2013 with Conathan.

An unparalleled collection 

The Kilpatrick Collection includes nearly 2,000 documents from the 1880s through the 1960s, written in the Cherokee syllabary by native speakers. Material ranges from shopping lists and personal correspondence to family histories and committee notes to foundational documents related to the political restructuring of the Cherokee people after the 1934 Wheeler-Howard Act, also known as the Indian Reorganization Act.

Bowern and Conathan, who is also a linguist, had discussed ways to use the material in coursework. Bowern was drawn to the idea of trying a different, more text-based approach to field methods, while still being able to work with native speakers.

“It’s a beautiful set of materials to work with,” said Bowern. “I don’t know of any collection that is as rich. We can do this for English and major European languages, but it’s rare to have this much material for a Native American language, or other indigenous languages.”

While some of the material dates back to the late 19th century, Belt said he felt a familiarity with the historic written materials.

“Even the committee reports are exciting. It may not be that exciting to know they voted to bake cakes for some event, but it’s absolutely exciting to know how they wrote this,” he said. “It’s like listening to somebody speak.”

Researching a new language

Students in Bowern’s class paired off at the beginning of the semester, with each team choosing a short text from the Cherokee materials to begin their research.

“My class partner, Faruk Akkuş, and I picked something that looked like it might be a shopping list,” said Brody. “It had a list of nouns, which seemed like a good starting point.” Brody and Akkuş converted the words in the Cherokee syllabary — a set of written characters representing syllables — to the Roman alphabet and used a Cherokee dictionary to try to figure out as many of the words as possible.

“We got about half of the list done that way. Then we worked with Tom Belt and Hartwell Francis on data collection and cultural information,” said Brody. “We didn’t know, for example, that in Cherokee you can take a verb, add a suffix, and turn it into a noun. It describes what you used that item for — like ‘baking soda.’ It wasn’t in the dictionary because it isn’t really the base noun of the word.”

The students met as a group once a week in class and scheduled additional sessions with Belt or Francis. The scholars in North Carolina worked with the students each week in New Haven via Skype and email, answering questions about morphology and sentence structure.

“I wanted to get the experience of Skype interviewing,” said Brody. “I didn’t know going in what it would be like. It was surprisingly very successful.”

He noted that the first time they used Skype, New Haven had just had its first major snowstorm of 2015. Classes had been cancelled, but the research moved forward. “It was 9 a.m. and I was in my apartment, Faruk was in his, and Tom and Hartwell were in North Carolina. We were working from three different places, but it was great.”

Brody said he was excited by the complexity of the Cherokee language.

“One verb can be 15 syllables long, tell you about who is speaking, the addressee, who or what you’re talking about, and if it’s past, present, or future,” he explained. “For someone interested in morphology, it’s a goldmine.”

The Beinecke texts were helpful in seeing the actual usage of a language and the topics that native speakers chose to write about, noted Brody. “Instead of saying, ‘I want to know how to say these 10 sentences,’ we had texts that would bring up stories and cultural information that you wouldn’t get otherwise.”

After working on the shopping list, Brody selected a letter from the collection and found several aspects of the language that are not well described or understood, and could be researched in the future. He noted that the field methods class taught him how to gather data in a way that is flexible and open-minded.

“You’re coached to scientifically get at the information you want — just like in a chemistry experiment: You want to make sure to control for every possibility,” he said.

Bowern said these are exactly the kinds of skills she hopes her students learn in her field methods class.

“I want them to be able to ask questions about a language to get the information they need. Learning how to phrase questions to someone who’s not a linguist is a very important skill,” she said, adding that students doing this for the first time can find the class challenging.  

“It’s the unknown. They’re used to looking at certain types of problems and if they don’t know the answer they know where to get it. Here, they can’t look up an answer because it isn’t in any books. They have to figure out how to ask a speaker to get the information they need,” she said.

Finding aid

In addition to Bowern’s class, Brody also spent much of last semester working on a finding aid — a guide to the collection — for the Cherokee materials at the Beinecke, which are now available to the public. The project is being managed by Conathan and is the latest step in identifying all the material in the collection in order to make it available to researchers.

“This is a long-term project at the Beinecke,” said Conathan. “We want to make sure we do this right, treat the material correctly, and understand what it is.”

In addition to the finding aid, Conathan has launched the website Transcribe Cherokee, which has a small selection of the Kilpatrick materials online that any student or scholar can transcribe.

“By transcribing them, it makes them easier to read, and it also means that you can search across the text for somebody’s name or a specific word. We’re hoping that with increased community engagement more people may be able to use these in a classroom context,” she said.                                                

Conathan added that she hopes to have an ongoing conversation with Cherokee teachers, students, and scholars to find out how they would like to use the material, what they find of interest, and what is most useful.

Next steps

Bowern and Francis plan to continue working with the Kilpatrick Collection. In particular, Francis would like to see the next generation of linguistics students and scholars visit Yale to study the Beinecke’s collection. For his part, Belt is excited about the possibilities for revitalizing the Cherokee language.

“We have to cultivate our own scholars, so this doesn't go away,” he said. “The language is complex and hard, and it opens up all kinds of possibilities. Not just for scholarly work, but for the culture itself.”

According to Brody, linguists are finding a lot of opportunities in diverse fields, from academia to technology to data analysis.

“Right now a lot of technology companies are hiring linguists for things like voice recognition and voice processing software. We’re also being hired for data analysis because of the way that we look at massive amounts of data from a language we’ve never seen before, and pick it apart, logically and mathematically,” he noted. “There is a lot of math and logic in linguistic theory.”

He added that while the field of modern linguistics is young, it is moving in interesting, cross-linguistic directions.

“It’s going to take a long time to get at the idea of how human language works,” said Brody. “Right now we’re in the stage of collecting data, comparing things, and branching out into new language families.”