How often do you use spellchecker, electronic dictionaries and other language tools on your computer? For most “minor” languages (numbering in the several thousands but spoken by only tiny percentages of the world’s population), those tools simply do not exist. And many of these minority languages belong to indigenous people.
Meet Antti Arppe, professor of quantitative linguistics at UAlberta, and his research project, “21st century tools for indigenous languages.” Together with his fellow faculty members and students at the U of A, and with collaborators in Canada and abroad, Arppe aims to create software and computer technologies to help people learn and use minority languages. The pilot language is Plains Cree, the most widely used Canadian indigenous tongue. “Crucially, our research team has two native speakers and scholars of Plains Cree, Dorothy Thunder and Jean Okimâsis, and we have started reaching out to Cree communities as well,” says Arppe.
Arppe explains that many indigenous languages have a very complex morphology compared to English. Morphology concerns the structure of the many forms of words, such as “walk,” “walked” and “walking.” Generating all forms of the 100,000 most common English words in your average learner’s dictionary would result in 400,000 to 500,000 word forms. In Plains Cree that figure would be tens, if not hundreds, of millions — far too many to list.
But the computer models Arppe constructs can solve this problem. These models can then be used as the basis for spellcheckers and electronic dictionaries. The end goal is to create similar computer tools to those available for majority languages such as English and French.
Language plays a major role in the identity of a community.
However, the project is not simply about providing tools to assist in using these languages in all spheres of modern life; it also addresses some deeper philosophical issues. “One could argue that language plays a major role in the identity of a community,” says Arppe. “[Supporting] the continued use of a language also helps enable the survival of that community’s culture, traditions and values.”
Participants of the project kick-off meeting (LR): Conor Snoek, Sjur Moshagen, Jordan Lachler, Arok Wolvengrey, Jean Okimâsis, Trond Trosterud, Dorothy Thunder & Lene Antonsen (missing from photo: David Beck, Ahmad Jawad, Juhani Järvikivi, Tim Mills, Arden Ogg, Sally Rice and Earle Waugh.)
Arppe has followed a long and winding road from his native Finland to his work on Plains Cree in the Faculty of Arts. He was led here in part by the strength and depth of knowledge in indigenous languages at the Department of Linguistics. The work has been a departure from his previous research and from his own language abilities. Arppe speaks Finnish, English and Swedish; can converse in Norwegian; possesses “survival” skills in French and a few other European languages; and can even read a bit of Akkadian and Sumerian. And Plains Cree? “I’m still only a beginner,” he says with a smile.