Rutgers team seeks to create extensive database of African languages
A team at Rutgers is creating a database that helps develop descriptions of African languages in order to further linguistics research.
The Afranaph Project began in 2003, where only five case files of languages were developed with each containing research and information about a particular language. The project now has 33 language case files, according to the project's website, and is led by Ken Safir, a professor in the Department of Linguistics.
“There are now over 6,000, 7,000 languages in the world. About 1,500 or perhaps 2,100 of them are in Africa, and we have very few of these languages even remotely described,” he said.
The project operates by finding native speaker linguists in Africa, providing training if possible and incorporating the linguistic data they contribute into a database. Safir said in the best case, this interaction develops into the native speakers becoming collaborators.
People are usually quite proud of the languages they speak and want them to be recognized. That means individuals who are trained in linguistics tend to want to study their indigenous language. The Afranaph Project allows for individuals to participate in the international community of linguists and helps build up collaborations and careers, Safir said.
Part of the project also involves orthography, or representing the sounds of a language through written or printed symbols.
“Most of the languages of Africa are understudied or not studied at all, some of them don’t have orthography,” he said. “And so the goal of many African linguists when they first learn linguistics is to create an orthography for their language.”
Safir said Africa is an appealing continent to study because there are many languages and ethnic groups.
Lydia Newkirk, a graduate student in the Department of Linguistics and member of the project, said she enjoyed looking at the datasets and seeing how the pronouns worked in all of the different languages.
“Part of our goal with Afranaph is to collect a lot of information from a lot of different languages, especially African languages, so we can figure out how they work alike and how they work differently,” she said.
Hazel Mitchley, a graduate student in the Department of Linguistics and another member of the project, said it enables others in academia to perform more research on understudied languages.
One example is Bantu, which is a subfamily within four of the main language families in Africa. She said many of the languages in the Afranaph database are Bantu-based.
“From the Bantu language family you can say, ‘I made him walk there’ in one word,” Mitchley said.
The database also contains useful linguistic data, such as how parts of words correspond to parts of meaning. Most of the impact of the project, though, is in academia, she said.
“It will filter through over time in terms of people using it to do research and making it easier to do research, and then once research is available it starts affecting things like machine translation,” Mitchley said.
The research model for the Afranpah Project could also be applied to other fields because it trains people to provide sophisticated data that would normally take years to gather, and then makes that information accessible to anyone, Safir said.
The Afranaph database, organized by Alexis Dimitriadis, was made specifically for linguistic purposes, but it has inspired other databases with similar principles adapted for other purposes. Safir said during the course of carrying out research, the database design became innovative unintentionally.
He said people see doing fieldwork on a language as going out into the community and trying to put together all the information possible based on just seeing and speaking, so the Afranaph approach is a novel idea.
The primary goal is to scientifically explore human language. Languages spoken by humans have different characteristics than “languages” used by other species. Safir said the ultimate goal is to understand the human capacity to learn a language.
“To best encompass what’s humanly possible, we want to see what humans have actually been using,” Safir said.