How to record every language in the world

To say that Frederico “Freddie” Andrade is ambitious is an understatement. Through his non-profit Wikitongues, he is building the first public archive of every language in the world. Wikitongues relies on volunteers who create videos and contribute to word lists, phrases, and dictionaries, all catalogued on Wikitongues’ open source platform.

We spoke with Freddie about the project’s genesis, the stories the organization has collected, and how they’ve been able to activate volunteers from around the world.

How was the idea for Wikitongues born?

I was born and raised in Brazil and moved to New York for college in 2009. I met my buddy Daniel Bogre Udell, who came up with this idea for his thesis. When we graduated, we wanted to move forward with the basic premise of asking people what they spoke, recording them, and putting it on YouTube.

After you decided to get started, what was one of your first recordings?

We recorded Jerry Wolf, a village elder guy at the Eastern Band of the Cherokee Nation, outside of Asheville, North Carolina. In casual conversation, he told us that he spoke French and learned it when he was at D-Day. We did this while being in school and working full-time and at the end of 2014, we decided to go all in. We took the following six months to organize and prepared ourselves to do a Kickstarter campaign.

What was the main motivation behind the project?

Academically half of the languages are expected to die in the next 80 years. People just don’t know or talk about it but language and cultural heritage disappear. You can’t rewind. What I can do is build a narrative around the value of language, culture, and identity.

How do you source content?

We are the stewards of the mission and the context comes from us, but first and foremost, we want people to come in and make their own. We couldn’t pay people since we weren’t equipped to make that happen. We resolved to start building an open source community of software developers to contribute and now we have fifteen software contributors. We have an audience of 30,000 people and it’s growing.

How do you activate volunteers?

Anyone with access to the Internet can go on it to document their language or any other language that exists. It can be dictionaries, phrases, or books. Ultimately, the goal is to have lessons and anything that is language be documented on the platform. Partnerships have been a big area for us.

How have partnerships helped you?

The reason partnerships became a vehicle for us is that there are a lot of communities that are adverse to outsiders. Our approach is about partnering with them and supporting them through technology, tools, etc. One of the big difficulties is that there isn’t really a clear distinction between what is a language and what is a dialect. There’s a real challenge of how to ask the question that gives you the answer you need. So that’s where partnerships become really helpful.

Sarah Doyle and Afro Amado dancing at an event in Vanuatu.

Do you have advice for how to best activate volunteers?

My advice is to communicate your story. How are you going to tell it and who is going to hear it? Our volunteer community started with the YouTube channel where people would show up and want to contribute. Our first fifteen volunteers were from all over the world. We would start publishing unfamiliar languages and that was enough to start perpetuating this conversation on a public platform. It was entirely organic and word of mouth.

Are there certain languages you’re looking for?

It’s important to address that we’re not looking for specific languages. Some people think we shouldn’t include sign language, but there are are almost 300 in the world so it’s super important to include them! We also include constructed languages like Esperanto.

What are some of the biggest lessons you’ve learned in building Wikitongues?

That things take time and take money. I arrived with the perspective of startups and all of a sudden we’re a nonprofit. We can’t give you equity and that was a significant learning curve. Being nonprofit and being able to say we haven’t taken money from X, Y, and Z gives us a lot more freedom to make decisions to our mission.

How do you make money?

We have three prongs: foundations and grants, low ticket fundraising from individuals, and now we are working with a couple of professional fundraising. There is definitely room for business models to show up like producing original content such as documentaries.

Are there any stories that particularly stand out?

The first story that comes to mind is when Dan and I were just working on this idea. All of a sudden we get a video in a language called Nafasana. We did our due diligence and nothing came up. Nothing on Wikipedia, no academic data. We realized this language had never been studied before.

[The other story is that] people don’t usually know that in Japan, there is another language called Ainu. It is spoken in the very north of the country and they are a community that has been persecuted and discriminated against with forced assimilation policies and language bans. There are now just ten to thirteen living native speakers. If there was one thing that I needed to give priority to, was finding a speaker of this language. We bent over backwards and squeezed our volunteer community in that direction. After a year and a half, Lindie, a South African and our first volunteer, found a woman who owned a restaurant [there] and recorded a video.

This process was more serendipitous than it was one of us scouring the Earth for someone; rather, Lindie came to us on her own accord and she did the real work. We just pointed her in the direction.

Wikitongues is about so much more than language.

The point of the project is that we’re not talking about alien monkeys on Mars. It’s about humans. The stories are all the same.

Related project
Aprendices Visuales.org (Visual Learners)
Aprendices Visuales.org (Visual Learners)

Books with pictograms for children with autism

Log in to follow