Globally Speaking
Podcast 044
The Logic of Localizing African Languages
R I am Renato Beninatto.
M And I am Michael Stevens.
R Today in our podcast we’re going to talk about African languages. This fascinating world, the future of business, where the growth of technology and trade is expected to accelerate in the next 10-20 years. A new generation of users of content.
M Yes, and it’s great, because we have taken a look at some of the health crises that have gone on and what language has to do with that in Africa. Today we get to hear from someone who for some of their career has spent time developing products and making sure products work in Africa. There’s a whole world and as long as I’ve been in the industry, it seems like it’s a future state, but we’re getting closer to it.
R To the reality. So, let’s listen to it.
MN I’m Manuela Noske, I’m a Communications Manager at GALA, the Globalization and Localization Association, and I am, by background, an African linguist.
R But you’re not African.
MN No, and I don’t have to be.
M No. Did GALA hire you because of the large presence they have in Africa?
MN No, they didn’t. So they hired me because they wanted somebody who comes from the client side of the localization industry, bringing that to GALA there. It’s an industry association that is value to them, because that is part of the audience that they’re trying to serve. And Communications Manager is a new role for me. I haven’t done that in the past, I find it fascinating, very interesting. Lots to learn, and I’m very happy with what I do.
M And GALA’s a great organization. We know we have lots of listeners who have found us through them, so that’s wonderful. You mentioned being on the client side and African languages. Is that when you were first exposed to working with African languages, or did you have an interest before that?
MN No, my interest goes way back. I have a master’s degree in African Languages from the University of Cologne, decades ago. Done research in Africa, done some business in Africa, so there’s been an ongoing interest in the continent.
R But African languages seems like an awfully big amount of languages.
MN It is a big amount of languages, yes.
R So do you specialize in all the African languages or do you focus on certain parts?
MN Depending on how you count them, there’s between 1,000 and 2,000 languages spoken in Africa, and the languages that I focused on for my academic interest were languages in East Africa. So the region of Kenya, Uganda, Tanzania was the area that I was interested in, but professionally I’ve worked with a wider variety of languages in Southern Africa as well as in Western Africa. So the largest languages on the continent are the ones that have gotten localized into, and so those are the ones that I’ve worked with.
R So these would be how many?
MN At Microsoft we localized into 12 African languages. There are three that we’re doing in Nigeria, Hausa... or we were doing in Nigeria. Hausa, Igbo and Yoruba, Wolof in Senegal, in Ethiopia, Amharic and Tigrinya; Swahili for all of East Africa, and then in South Africa, we do a total of five, including Afrikaans. I forgot Rwanda. Kinyarwanda in Rwanda.
R Okay. That’s fascinating. Because for localization you need also to have a significant consumer base, you need to have computers to be done there, but it’s a little bit the chicken and the egg problem. It’s a classic problem that we have in localization. Do you get people to use computers because the software is localized, or do you localize because people have computers? What was the strategy, the commercial strategy for localizing into African languages, if there was one?
MN I think it’s a hard problem. I think the thinking was exactly that we wanted to... there is a market there, and that the understanding slowly evolved that Africans don’t speak as much English or French or Spanish as people commonly believe. So the common people in Africa tend to speak their local vernaculars, and on top of it they’re polyglot. They speak a lot of the lingua francas, the vernaculars in the region, and so I think the thinking was simply that we needed to have and build a presence in those countries, in those languages, to attract more customers.
R So what is the difference between localizing into a European or Asian language and an African language?
MN Well, there’s lots of difference actually, and I think it starts with the problem that many of these languages are spoken languages; they’re hardly written, and there aren’t any real standards yet. So, the languages that I’ve listed, the 12, they’re on a spectrum essentially. Swahili has a fairly long history of writing already; so does Amharic, but languages like Zulu, for instance, their written heritage is younger, and so there are lots of problems that come with that in terms of terminology, writing standards. How do you actually spell these different words? How do you call something and how do you actually use the language for written communication?
Those are huge topics in African languages. Plus, there’s the perception by many people that these languages are really languages of the heart, languages that are used in the home to communicate with your loved ones, they’re not languages that you use in business. So how can you develop terminology for businesses in these different languages? People sometimes reject the notion, like, ‘I speak Zulu at home, but I wouldn’t speak it in the workplace, so why would I read Zulu in the workplace?’ So, trying to bridge that and basically try to assist in the development of these languages was part of the goal as well.
M So just an average speaker would be hesitant to use their native language in the workplace.
MN Yes, And sometimes they don’t have much of a choice, because these are all multi... with the exception of Kinyarwanda, maybe, or Lesoto, all of these countries are multi-lingual countries, and so...
R Because they are artificial boundaries created by the colonial powers, essentially.
MN Yes, and there are lots and lots of different ethnic entities. So, when you come into the city, any number of languages are spoken, and so in East Africa, fortunately, Swahili has been developed as the lingua franca, the language of commerce and business; it’s the commercial language—sorry, it’s the national language as well—and so people use Swahili then to communicate with people from other areas. So, when you’re in the workplace, you can’t be sure that anybody really speaks your language, so then the first choice is Swahili, the second choice would then be English, or in some other areas of Africa, French.
R Have you monitored how behavior towards technology has changed after you localized the products into those languages?
MN We’ve tried to understand what’s the adoption rate? How receptive are people to it? These are difficult questions, and I don’t think we’ve really done a stellar job in trying to figure that out and really unpacking what’s happening there. What we’re finding when we sit down, we’ve done it only... I’ve only been involved in one informal study for Swahili, is when you first show the interface in Swahili there is true bewilderment. People look and go, ‘What’s this? What are you saying?’ In worst cases, they would start laughing about what you’ve presented to them and then, you know, ‘What? Is there something wrong? Is there something funny?’ and it’s just, ‘No, it’s unusual, it’s uncommon.’
For many of the languages, especially the larger ones that we localized into, there’s also a state called diglossia, which is there’s an accepted written standard, but what people speak is quite different from that standard. So, they will look at what you’ve produced and say, ‘Oh, that’s the good Swahili or the right Swahili, but it’s not really the language that we speak.’ So, the central problem is that we need to really—the nut that we need to crack—is how can we produce content or a software interface that actually reflects the language that is spoken and not the language that the academy has decreed is the right kind of Swahili?
M Are people more apt to embrace a new technology when it’s in their spoken language and what they’re more familiar with, rather than the... or is it a feeling thing? Why would that be the standard rather than the academic standard?
MN You know, it’s a good question, and I’ve seen both reactions to it, so let me give you an example. Last time I was in Kenya, for kicks we went around and asked people, ‘When you use an ATM...’ and we asked just about anybody, like receptionists and waiters and whatnot. ‘When you use an ATM, what language do you use to withdraw money?’ because the ATM displays in both English and in Swahili, and about half of them said, ‘Well, when it comes to money, I want to make sure, so I use English,’ and the other half said the exact opposite, ‘When it comes to money, I want to make sure; I use only Swahili.’
And so you’re like, ‘Well, what is it? Is it more trustworthy in your language, or is it just the opposite? Do you feel this is the field or the area of business and commerce and so English is the more appropriate tool to use?’ So, I think the word is still out if it really helps people to understand technology or participate better in technology when it’s in their local language, rather than a foreign language. The English and French still carry a lot of prestige and a lot of status, and it matters a whole lot.
R It’s interesting, because this new generation, the digital natives, in those countries, it’s one thing when you localize it for the first time a few years ago, but now there is a whole generation that is growing up with mobile devices, mobile is very advanced in Africa, banking, mobile banking is more advanced than in many countries, and there is a whole generation that is growing up already with content in Swahili.
I remember from my times working in the Czech Republic that during the period of Czechoslovakia, everybody understood Czech and Slovak, and they say that today the Slovak kids have a hard time understanding Czech and vice-versa, even though the languages are very similar. I imagine that something like this might start to happen because of Snapchat in Swahili or in Kinyarwanda or whatever other language, and Facebook. I
I remember a story from Iris Oris from Facebook, that she said that they did a survey, and they asked if people wanted to use Facebook in Swahili or in English, and the majority said English. When they launched the product in Swahili, 90% or 70%, a very high percentage, I don’t know exactly what percentage, but a very high percentage of the users immediately switched to Swahili, because that’s the language they speak at home.
MN Yes, and honestly Swahili is a slam dunk case. If you want to have a presence in East Africa, you have to do it in Swahili. It’s the language people really identify with, but you have to look at the different regions. In other regions it’s a different case and some of those other... like West Africa, they may truly prefer French. But in East Africa Swahili is a must-have.
R But you have a favorite language, and that’s not Swahili.
MN Well, I have a number of favorite languages, yes.
R And it’s not German either.
MN It’s not German, no. No, I mean, when it comes to African languages I am partial to the Nilotic languages, like Maasai, Turkana, Karamojong, Teso. Those are the ones I’m most fascinated by. So those are the classic cattle herders, the nomads of East Africa. They’ve won my heart, and those are the languages I studied academically, but those are not languages that I think anybody would ever localize into.
R What is so attractive about these languages?
MN Oh, it’s not the languages. It’s the cultures—you know, the way the people interact, the way they communicate, their world view—that has really attracted me. They’re almost the opposite of all that we hold dear in the Western world, and so I’m just fascinated by that contrast, to be quite honest.
R But these Nilotic languages, what do they have in common? Sounds? Grammar? What is the element that unifies them?
MN It’s all (of it). They’re related languages. There are three different branches, and so they’re more or less related, but within those branches they’re genetically related. Essentially, they go back to one ancestor language, and then they developed, just like German and Dutch are related languages, so all of these languages are related. Yes, so they share some grammatical elements, they share some lexicon, all of that. Some of the phonology’s shared as well, so yes, they’re related.
M And talk to us a little bit about the culture that’s so attractive. My grandfather was a cattle herder, so I completely relate that cattle people are great.
MN Yes, well, it’s about survival, and it’s about survival all the time, and so the decisions that you make on a daily basis to survive and to get on. They’re very hostile, and it’s funny to be talking about this and finding this attractive, but they’re the only people I ever lived among who really will not smile at you, regardless of what happens. They stare at you and they stare at you fiercely, it’s very intimidating. There’s just no culture of friendliness and openness, and they’re also known as very avid cattle raiders, so they will go after one another. They used to go after one another with spears and now, unfortunately, they would with assault rifles, and the cattle raids are deadly. Sometimes 10, 20, 30 people will die, and it’s all about the cows.
So, I don’t know if you know the story about the Maasai and the cows, but their thinking is that any cow that exists in this world must at some point have been a Maasai cow, and if it’s with another tribe, then it must have gotten stolen from them. So that is their attitude. There’s true pride and ownership, and that leads to these cattle raids, because those cows over there, they look attractive…they must have been ours and somebody took them.