At a recent languages-themed discussion dinner hosted by the Provost at Queen’s, one idea kept resurfacing: the story we tell ourselves about languages in the UK may be wrong.
Duolingo UK and Europe Director Michael Lynas brought to the dinner a perspective grounded in the behaviour of millions of learners. Duolingo data suggests that interest in languages is not declining, particularly among under-22s. Instead, it is thriving outside the classroom but too rarely carried through into formal education.
What followed was a conversation about missed connections: between apps and classrooms, cultural interests and curricula, home languages and qualifications. Here, Michael reflects on where those connections break down and how they might be rebuilt.
You’ve said that the UK isn’t lacking an interest in language-learning so much as lacking good pathways. What do you mean by this and where do you think the pathway breaks down most sharply?
One of the things that I’ve been frankly quite surprised by is the Duolingo data on language learning. The story we tell ourselves in the UK is that we speak English and we don’t care about languages, and we’re a bit ashamed about that. I just assumed this would show up in the data, but the UK is second in the world in terms of time spent learning on the app. It’s also frequently one of the top countries in terms of people learning multiple languages. We’re seeing what people are doing, not what people are saying and that reveals that people in the UK are actually very keen on languages. Even more interesting is that people under the age of 22 are the keenest. So, when people say that young people don’t care about languages anymore because they’ve got translation apps, that doesn’t seem to be borne out by the data either.
We’re seeing what people are doing, not what people are saying and that reveals that people in the UK are actually very keen on languages.
If the interest is there, then the question I’m focussed on is how you create the pathways between informal learning and that which is classroom- or university-based. How do you bridge the divide a little bit? At the key moment when you’re 13 or 14 and selecting your GCSE choices, we need to help people to connect their interest in languages and things they’re doing outside school, with what they’re learning at school. I’ve got some ideas, but I don’t think there’s an easy answer to this because you need to take a general interest and make it compelling enough for people to actively choose something. It’s also vital to recognise the challenge of this early enough because without a GCSE, you won’t do the A level, and without that you probably won’t choose languages at university. This filters right into teaching and then becomes a big problem.
You’ve previously noted that the appetite for languages doesn’t always show up in the places we’re used to looking. Where should educators be looking instead?
Obviously, there are places like Duolingo and other new technologies where people are using technology to learn and explore languages. There are also more cultural drivers so things like Korean pop bands or Japanese TV shows can spark an interest in the language. It could also be books or sports so it’s good to look at what people are interested in in their personal lives and see how we might draw a line between those interests and the study of language at school. That’s not always happening and it’s a missed opportunity. We can also consider home and heritage languages: languages, other than English, that are spoken at home. I know this is something that the Queen’s Translation Exchange (QTE) does in its work with schools. What we’ve uncovered is that sometimes people don’t think these languages are an academic thing that they can take a qualification in; it’s just a thing they have. So, reframing this is another good idea.
Duolingo started with a mission to make education universally available. How did this develop into a realisation that how people learn matters as much as what they learn?
We’re part of the app generation of technology. The challenge we embraced was what and how could people learn on a phone? Clearly, there are some constraints, but it also opens up some different opportunities. One constraint is that you’re not in a classroom setting so no one is making you turn up to learn. Duo tends to be what people are doing in their free time and that means you need to work a bit harder to compete with other calls on attention. A big part of Duolingo’s mission is making it fun to get people’s engagement. We need both the quality of learning, but also it needs to be fun because if it’s not fun, you could have the best education in the world, but people won’t commit to it and make any progress.
I think where we stand out is the fact that we’ve made it fun and memorable. People keep coming back and you hear a lot of people talking about their streak, because there’s a source of pride there. We’re very open about gamifying the learning and we’re not ashamed of that, but there is also a very serious side. We have lots of people with language PhDs involved behind the scenes, along with learning scientists who have helped to develop a lot of techniques. That’s enabled us to help people learn other things as well like music, maths, and chess.
We’re very open about gamifying the learning and we’re not ashamed of that, but there is also a very serious side. We have lots of people with language PhDs involved behind the scenes, along with learning scientists who have helped to develop a lot of techniques.
Universities have traditionally prized depth and slow thinking, while platforms like Duolingo optimise more for speed and bite-sized learning. How do you see those models learning from each other?
We think that what we’re doing and what happens in a classroom or university setting complements each other. When I speak to people at the University Council for Modern Languages, it struck me that lots of them and their students are using Duolingo, either to learn additional languages or to practice their vocabulary and speaking skills. Teachers across the UK are saying, hey, maybe in your spare time, you could practice your French a bit on Duolingo. Students may only be spending a few hours a week in their French or German class, but if they can take a bit of the time that they might otherwise spend on other apps, and spend it learning a language, that’s going to accelerate their learning. I also think it’s fantastic that someone could pick up a language for fun on Duolingo and then decide to take it forward at school and university and that’s something we see happening.
In terms of deeper learning, we’ve recently introduced the Duolingo Score to encourage people to spend a little bit longer learning, maybe 15 instead of five minutes a day. The score is basically a measure of how far you’ve made it through the course and we’re encouraging people to capitalise on their habit of learning and push themselves a bit further.
The HEPI report A Languages Crisis highlights a widening participation gap in language-learning. How can technology help?
The thing we’re particularly excited about at Duolingo is the fact that technology is making a great education much more accessible. We’ve got an unusual business model whereby 90% of people learning on the platform are learning completely for free. As a company, we’re passionate about keeping that education free and accessible for millions of people, particularly people who might be learning English to get a better job or get into university. The mobile phone is becoming ubiquitous around the world, and more people in the developing world now have a phone than access to a computer. This provides opportunities for people which it would be more expensive to get access to in other formats.
The thing we’re particularly excited about at Duolingo is the fact that technology is making a great education much more accessible.
AI could be changing what it means to learn a language. What impact do you think this is going to have?
We’re not seeing an impact in our numbers: in fact, the number of people engaging with what we do is going up each year. There are now 135 million monthly active learners on Duolingo. Fundamentally, I think it’s because the reason people want to learn languages is not because they want to get a perfect translation of what people are saying.
The reason people want to learn languages is not because they want to get a perfect translation of what people are saying.
You don’t need to be word perfect in another language to be able to understand someone or say a few things, to show a bit of respect, and build some rapport. It doesn’t seem likely to me that that rapport will be the same if you take out your phone and use a translation app on it to communicate with others. That’s not really the point. There are deeper reasons for learning a language; ultimately, language is the way we understand the world around us, and I think to really understand another culture, you need a knowledge of language that goes beyond a literal translation.
Ultimately, language is the way we understand the world around us, and I think to really understand another culture, you need a knowledge of language that goes beyond a literal translation.
One thing I’ll say about AI is that it’s creating new opportunities for how you can learn. One of the things we traditionally struggled with on Duolingo is practising speaking and what AI has done is enabled us to have some great practice conversations with some of our characters. Previously, we tried to set this up with other people, but we found that learners were far more self-conscious and reluctant to engage with this.
As you know, Queen’s is home to the Translation Exchange, which treats translation as a creative act, not just a technical one and Duolingo has become famous for making language-learning playful, even funny. How can we bring back the joy to formal language education in the UK?
In the conversations I have with Charlotte (QTE Director), we often talk about the need to regain that sense of fun and exploration. Duolingo is all about making language learning fun and engaging. Many of the reasons that young people are connecting to languages are related to art and music and those are all really engaging, exciting things that can open up a whole world of opportunities. Language learning is inherently creative and exciting and if that hasn’t always been people’s perception of it in formal learning, then that’s a shame. Of course, language learning can take time, and it can be hard but that shouldn’t mean we have to lose sight of all the joyful aspects. We need to make the habit fun.
We need to make the habit fun.
Data shows under-22s choosing Japanese, Korean, and Chinese in growing numbers. What does that tell us about how cultural interest drives language-learning today?
There’s a lot of stability in terms of the top languages being learnt on the app, for example in the UK it’s French, German, and Spanish. But we also see that with younger age groups, there are different languages emerging – with Korean, for example, 80% of people learning Korean are under 30. This can be driven by popular culture but it’s also due to world events so, for example, we saw a big interest in learning Ukrainian in the UK after the invasion of Ukraine. People want to engage with geopolitics linguistically so they can say a few words to refugees in their own language and make a connection.
Can you recommend a book?
I’m going to recommend three books, which are all part of the same series. It’s called The Alexander Trilogy by Mary Renault. They’re about Alexander the Great and Ancient Greece and what’s special about them is they evoke a completely different world. You have conflict and love, like we do today, but it’s a very different world that you can access through the lens of fiction. I think there are some similarities to how languages also open us up to different cultures.
What’s your streak?
760 (in French).
