Research team is revitalizing Indigenous language with a first-of-its-kind blueprint

With 75 percent of Canada’s Indigenous languages facing accelerated language loss, an Indigenous-led research team from Port Hardy is setting a precedent for language revitalization by developing a unique approach that will teach people how to ‘live’ their language.

“Language revitalization isn’t just about learning to speak forgotten words. It’s about learning how to look at the entire world through an Indigenous lens,” said Sara Child via a press release announcing the research team’s findings.

Child is a professor in Indigenous Education at North Island College who is leading the research effort to recover Kwak’wala, one of the most complex Indigenous languages in the world. Her team’s breakthrough work — supported through the Sanyakola Foundation, a not-for-profit society Child established in 2017 to focus on Indigenous language revitalization in consultation with Kwakwaka’wakw elders — is one of dozens of Canadian research projects highlighted in a new Innovation Trends video series, called Their World, Our Future, by national innovation organization Mitacs.

“Indigenous languages are medicine for our people and their revitalization is vital to our individual and collective wellness and the wellness of the earth,” added Child, a Kwagu’ł band member whose interest in Indigenous language was piqued as a young child, when she saw how proud it made her grandparents to hear her family speak it.

“Suicide and traumatic death is no stranger to our people, and to our youth in particular, and I believe deep in my soul that relearning our languages will change that, as long as our language learning is done in a way that respects our traditional values and beliefs and our connection to the land,” she said. “I am passionate about this work, which I dedicate to my nephew, Eugene Kenneth Child, who left us far too soon in life.”

A multi-faceted, collaborative effort that involves elders and knowledge keepers, and engages youth, the Sanyakola project to recover Kwak’wala launched in 2017, and in 2020 was one of the first projects to be approved under Mitacs’ new Indigenous research stream. Though well documented, Kwak’wala is “dangling by a very slender thread and needs to be “woken up,” said Child. “This can be supported by tapping into the memories of elders and transcribing hours upon hours of translated tapes, videos and books from archives into the official writing system adopted for Kwak’wala,” she added.

What sets the research team’s approach apart is their understanding that the language is intrinsically tied to the land and wellness, meaning it requires a different pedagogy than conventional Western teaching methods. For example, Kwak’wala has concepts that can only be fully understood by performing land-based and cultural activities that go along with them, such as “respecting the ocean” by expressing gratitude while performing a cleanse in the ocean.

“Our people believe that if you’re physically harmed, you’re also emotionally and spiritually harmed, so we have to conduct ceremony to brush away the harm of your spirit and your emotion so that physical healing can begin,” Child explained. “That’s part of our work, to ensure people gain understanding of this essential aspect of our language along with the translation.”

The four Mitacs-funded interns involved in the project are currently working in consultation with elders to build a framework for what a new pedagogy grounded in a Kwakwaka’wakw lens will look like. When the pandemic derailed their plans to meet with elders in person, they went online, working to build their own fluency in the language as they continued to build capacity for immersive language summer camps that will hopefully be held outdoors once again on Kwakwaka’wakw homelands later this year. Their strategy also includes developing leading-edge voice-to-text technology, such as a mobile app capable of identifying everyday items from pictures snapped on a smartphone, with Kwak’wala words pronounced in the voice of an elder.

While the team found that online learning during COVID-19 has a silver lining in that it is helping to build momentum for language revitalization, Child explained that the primary challenge for Indigenous languages in technology sectors is that the vast majority of voice-to-text technologies are developed for English, yet Indigenous languages are verb-based and therefore function very differently from English. That means the researchers are forced to develop their own recognition system from scratch, carefully authenticating transcribed words from archived sources before feeding them to an advanced machine learning system that is slowly but surely starting to learn Kwak’wala. “This work is arduous and time consuming, adding complexity and barriers to Indigenous language revitalization,” Child said.

“One of those barriers is that English, French and other large languages of the world have billions of dollars of support, whereas Indigenous languages receive miniscule amounts of funding,” Child explained. “If we break the code for building voice-to-text technology for Kwak’wala — and I’m sure we will — the impact will be absolutely phenomenal across the globe.”