Changing Planet

Language Diversity Index Tracks Global Loss of Mother Tongues

 “For the past several years, we had been hearing anecdotal reports about endangered languages–how we’re losing languages by the day, how we may lose 50-90 percent of languages before the end of the century. But nobody had any reliable quantitative data to corroborate these claims,” says Luisa Maffi, co-founder and director of Terralingua, an international NGO devoted to sustaining the biocultural diversity of life through research, education, policy, and on-the-ground work.

ILD paper.jpg“But now a new Index of Linguistic Diversity (ILD), the first of its kind, shows quantitatively, for the first time, what’s really happening with the world’s languages,” Maffi adds. “The ILD shows in quantitatively rigorous ways what the trends have been over the past 30 years in the numbers of mother-tongue speakers of the world’s languages–and the news is not good: an overall decline of more than 20 percent in that period alone.”

David Braun of National Geographic News Watch interviewed Luisa Maffi, David Harmon, and Jonathan Loh about the new Index of Linguistic Diversity.

Harmon, of the George Wright Society/Terralingua, and Loh, of the Zoological Society of London/Terralingua, are the co-authors of a paper, Index of Linguistic Diversity, published in the journal Language Documentation & Conservation. Volume 4 of 2010. Their work was underwritten by The Christensen Fund, a nonprofit which supports National Geographic News coverage of biocultural diversity issues.

Braun: What is language diversity, and why are we potentially on the brink of a mass extinction of languages?

Harmon: There are 7,000 languages, but there’s more to diversity than just separate languages. There’s diversity within languages and structures of languages, and all that.

The reason why we’re coming up to the brink of a mass extinction of languages is simply that there are a lot of pressures in the world that are enticing or even forcing people to switch from generally smaller, more geographically restricted languages to larger languages, especially global languages like Mandarin Chinese, English, or Spanish, or even languages more regionally dominant than smaller languages.

So we have 7,000 languages, which is the consensus number of discrete languages that are out there. But most of the people who study endangerment of languages are predicting that there is a potential for a mass extinction of these languages within the 21st Century. By extinction they mean that the languages are no longer going to be spoken by people as mother tongues, their principal languages.

“There is a strong possibility that we’ll lose languages that people are using as their main vehicle of expression, which they may regard as one of the linchpins of their self-identity.”

Some of these languages might still be spoken after they are lost as mother tongues, in a restricted way, in ceremonies or in special usages like that. But in essence there is a strong possibility that we’ll lose languages that people are using as their main vehicle of expression, which they may regard as one of the linchpins of their self-identity.

So all the pressures that are out there in terms of globalization, government policies that may favor certain official languages and actively or at least tacitly suppress smaller languages, economic pressures, all these things come together to put pressure on smaller languages. Therefore the diversity of languages is going to be compressed, from 7,000 separate languages to something much smaller than that.

But it is even more nuanced than that. There is also the factor of distribution of languages and how even that distribution is, and that is part of our conception of linguistic diversity. Most people talk about separate languages and they talk about extinctions. But one of the things that we are doing in this ILD is trying to move the conversation beyond those two factors, to try to get to a richer view of linguistic diversity.

Loh: I want to emphasize one thing here. When a language goes extinct, it does not necessarily mean that the ethnic group that speaks that language has gone extinct.

Jonathan Loh.JPG

What often happens is that when a linguistic group is small, when the number of speakers may only be a few hundred or thousand, there is enormous pressure to shift from their native mother tongue to a more dominant language spoken in the country where they live, which could be Spanish or Chinese or Russian or Portuguese or Arabic, or whatever, because of the social and economic advantages of speaking that bigger language.

Photo: Jonathan Loh

When that process happens, perhaps the younger generation starts to become bilingual, and then the next generation has a weaker grasp of its mother tongue, and by the third generation they can no longer speak to or understand their grandparents and great-grandparents. So it can happen quite rapidly, the shift from one language to a larger, dominant language.

Maffi: I would add that in some cases the shift to a dominant language can very much be part of government policy, and it can happen from one generation to the next. That’s what happened to native communities in North America, both Canada and the U.S., where the system of residential schools was put in place, and children were consciously taken from their families and communities and put in residential schools far away, where they were forbidden to speak their own language.

When they came home they were not communicating with their parents and grandparents in their own language, they spoke English. For some of them the pain and the shame were such that they didn’t want to speak their own language anymore, because they were told it was primitive, and anyway it was associated with all their suffering.

So they didn’t transmit their language to their children and grandchildren, and now we are faced with a situation in which many of these communities are beginning major efforts to re-acquire and re-affirm their languages as part of their identity.

So when we focus just on the idea of language extinction, that we are losing individual languages, to some extent we lose sight of the process of how this happens. Extinction is the end point of a long process, and the ILD helps us see that because it helps track the shift from the smaller languages to the larger languages, so it gives us a sense of that process.

Braun: Why should we be concerned about language diversity? Isn’t it the view of some people that it would be a lot easier if we all speak the same language?

Harmon: Diversity is really the underpinning of all life on Earth. We’ve just come through the International Year of Biodiversity. There was a lot of focus on that concept during this year, but here at Terralingua we have a biocultural diversity approach. Our view is to emphasize those parts of biological, cultural and linguistic diversity that inter-penetrate one another.

You can make the argument that as part of that larger biocultural diversity language diversity is absolutely essential for both individuals and human groups to establish identity, to understand who we are, what our place is in both the larger civilization of humans and also the natural world.


At that deep level is why you really need to be concerned about language diversity, because ultimately you are diminishing our own humanness. That’s an argument that I have made in previous work that I’ve published.

Photo: David Harmon

As for the second question, isn’t it a lot easier if we all speak the same language, I have to say that that’s a very easily retailed myth, that people who have other agendas like to put out there when they try to pooh-pooh the enormity of the language diversity crisis. You will see this in published responses, when people will say, ‘Well let’s all speak the same language, it will be a lot easier.’

What I always say to this is, OK, fine, let’s all speak the same language — but we’re going to put 7,000 ping pong balls in a basket with a language name on each one, and we’re all going to agree to speak whatever language we pick out of the basket. I think a lot of people who make that argument would quickly retreat, because what they are really saying is wouldn’t it be a lot easier if everybody in the world spoke not just any old language, but MY language.

So it’s a highly loaded political question, which in fact is also incorrect, because things aren’t necessarily easier when people speak the same language. We can point to situations like in Northern Ireland, where there’s strife among people who speak the same language. There are hundreds of examples like that.

Loh: Voltaire said he heard someone at Versailles exclaim, “What a dreadful pity that the bother at the Tower of Babel should have got language all mixed up: but for that, everyone would always have spoken French”.

Maffi: Similarly to what Dave was saying, when people come up with that kind of question when I give a talk, I always tell them to imagine if the history of the world had gone a little bit different and the dominant language was Russian or Chinese or Spanish, and you were a speaker of English, and someone said you have to stop speaking your language and start speaking Chinese.

You may do it because you have no choice, but how would you feel and what kind of difficulties would you encounter, and what kind of frustration in your ability to express yourself, especially when you are put in emotional or very personal situations. You would probably not be too happy, and somehow you would discover that so many of the things that you take for granted in your own culture, embedded in and conveyed by your own language, all of a sudden they wouldn’t be there.

ILD Key Findings

  • Globally, linguistic diversity declined 20% over the period 1970-2005.
  • The diversity of the world’s indigenous languages declined 21%.
  • Regionally, indigenous linguistic diversity declined over 60% in the Americas, 30% in the Pacific (including Australia), and almost 20% in Africa.
ILD Global Chart.jpg
Indigenous ILD Americas.jpg
Indigenous ILD Pacific.jpg
Indigenous ILD Africa.jpg

Between 1970 and 2005, global linguistic diversity declined by 20% while the diversity of indigenous languages declined by 60% in the Americas, 30% in the Pacific, and 20% in Africa.

Graphs courtesy of David Harmon and Jonathan Loh. © Terralingua 2010



Braun: Anyone who speaks more than one language knows that there are concepts and ways of thinking in one language that aren’t present in another language, so you do lose more than language when you lose a language, right?


Maffi: That’s right, and it’s an important point. The majority of people who make the kind of argument [that it’s easier if everyone speaks the same language] are monolingual, speakers of some majority language, who just don’t have the perception of the diversity and variation that there is even among languages that are closely related, let alone languages that are so different and have absolutely nothing in common with one another.

It should almost be prescribed for everybody to be at least bilingual so that we would begin to get a sense not only of what is lost when you can’t speak your own language, but what is gained when you can speak more than one.

Braun: Why do we need a linguistic diversity index? What does it measure?

Loh: The idea of creating the index was the attempt to quantify the loss of linguistic diversity, which hasn’t been until now expressed in quantitative terms. But there was a sense that the world is rapidly losing linguistic diversity, that we’re heading to a mass extinction of languages.

We wanted to put this in quantitative terms. So we thought about measuring it by looking at data for the numbers of speakers of different languages around the world, and the index is based on trends in speaker numbers across a large sample of about 1,000 of the 7,000 languages. We then looked at these trends, and we said overall what’s the average trend across these 1,000 or so languages?

But it’s a slightly more sophisticated measure than that, because we had to take into account the fact that the world’s population is growing, and over the period which the index of linguistic diversity covers, which is 1970 to 2005, the world’s population roughly doubled. This meant that many languages, if not most languages, are increasing in numbers of speakers, just by dint of population growth.

What we really wanted to measure was the concept Dave mentioned–evenness. How are different languages shared among the world’s population? What the index actually measures is how even is the distribution of speakers among the world’s languages. What we see is that that distribution is becoming more and more skewed.

In other words, some languages are growing faster than others, and for most languages the share of the world’s population speaking those languages is declining [against] a few dominant global languages which are increasing their share of the world’s population. So the distribution is becoming more and more uneven and speakers are becoming concentrated into an ever smaller set of languages.

The charts below can be enlarged by clicking on them.


World Population and Languages Chart.jpg
Largest World Languages Chart.jpg

Charts courtesy of David Harmon and Jonathan Loh. © Terralingua 2010 

Harmon: It’s an index of the concentration of people across languages. It’s a landscape of languages of the world and people are shifting to different places in that landscape. Some languages are colonizing more of that landscape than others. Even if the numbers of people who speak a language are growing numerically, their portion of the overall landscape of languages that their language occupies is being compressed by the larger languages growing even faster than they are.

In some regions of the world, particularly in the Americas and Australia, our index goes down much more sharply than it did for the world as a whole. Those are places where populations of the speakers of the smaller languages, which are largely indigenous languages, are not growing any faster than the rest of the population of their region, so they are actually showing numerical declines in the numbers of people speaking those languages.

In those areas, ILD plunges over the period 1970 to 2005.

The ILD doesn’t predict the future, but if current trends of concentrations continue to shift, at some point population growth hopefully will max out at the middle of the century, and then you will start to see numerical declines in speakers of even those languages that are growing right now.

It’s the idea of trying to account for population growth that’s the extra nuance the ILD addresses in its methodology.

Maffi: Jonathan has a nice metaphor of the jelly bean index.

Loh: Imagine languages are represented by different color jelly beans. You have a tall jar half filled by ten different color jelly beans. You add more jelly beans to the jar, but only red ones, until it is nearly full. Then you put only a couple more of each color on top, put the lid on and shake it up.

What happens is that the overall color of the jar has become redder and the diversity has reduced, because now the red jelly beans are dominating and the other color jelly beans occupy a smaller proportion of the total jar. Although the overall number of all colors has increased, the diversity is reduced. The evenness is reduced–and that’s exactly how the index of linguistic diversity works. It measures what is that evenness of jelly beans in the jar.

Maffi: Here’s another metaphorical way of explaining what’s going on with the world’s languages. Let’s imagine (and it’s not far from reality!) that each language community represents a “choir”, singing/speaking in their own language. Well, the choirs of those 25 or so big, dominant languages are getting bigger and bigger (that is, the languages are adding disproportionately more members to their choirs). When the choirs “sing” in those languages, the sounds they make are more and more drowning the sounds that the 6,975 others can make, even if those others still manage to add a few members to their choirs here and there.

In many cases, especially in the Americas and Australia, the numbers of members of the smaller language “choirs” have actually been radically diminished — by actual elimination through war and disease, especially in the early days of colonization, or by later forced assimilation that caused complete or near complete language shift. In those cases, we see the sharp declines of the regional linguistic diversity indexes.


But in all cases, when you have such “big bullies” of languages all around you, chances are that you’ll be forced, whether by social or economic circumstances, or even by government political will, to shift to a majority language. And if you, as an individual, shift to another language (instead of keeping yours and adding one or more others, thus becoming multilingual, as is entirely possible and actually beneficial), chances are you’re not going to teach your language to your children, so intergenerational transmission is broken.

Photo: Luisa Maffi

That’s how, even in cases in which there hasn’t been colonization, over time the numbers of mother-tongue speakers of a given small language can actually diminish. When this happens, it’s no longer just the case that the global distribution of speakers of the world’s languages is getting more skewed, less even: rather, you begin to witness language loss, the end point of which is language extinction.

To explain how language shift occurs: the members of the smaller choirs may valiantly continue to sing, and they may even add members to the choirs–but if their voices are being drowned more and more by those of the bigger choirs, some of them may end up saying, to hell with it, if I can’t have my rightful place in the polyphonic chorus of humanity, I’m joining the bigger choir! Add to this the active recruitment by the bigger choirs (read government policies and other assimilation mechanisms), and you get the idea of how we’re getting more and more linguistically (and therefore culturally) homogenized.

Needless to say, it’s also (and some would say foremost) a matter of human rights for the speakers of the smaller languages to be able to continue to speak their languages. The UN Declaration of Human Rights states (Art. 2) that everyone is entitled to all the rights and freedoms set forth therein, without distinction of any kind, including based on language.

“Discrimination based on what language one speaks is against human rights.”

From this one can infer that discrimination based on what language one speaks is against human rights. And the UN Declaration on the Rights of Indigenous Peoples touches extensively on rights to own language–and, as Dave and Jonathan point out in the ILD paper, 80%-85% of the world’s languages are indigenous! So to apply linguistic human rights to indigenous languages really means protecting an overwhelming proportion of the linguistic diversity that we have on this planet. (Of course, the speakers of the “big” languages also have the same rights–only, they shouldn’t exert them at the expense of the other languages!)

The many language revitalization efforts around the world are obviously valiant (and often successful) attempts to counter all this. But we still have to deal with government attitudes and policies, which still tend to see linguistic (and cultural) diversity as a threat, or at the very least make all kinds of excuses for not providing enough support for mother-tongue education.

Certainly Canada, the US, Australia, and other economically powerful countries don’t nearly do what they could and should about this, although attitudes have changed somewhat.

And some countries are going in the opposite direction: I just recently read an article (Globe and Mail, Nov. 10, 2010) titled “China’s dying languages ‘not that important’ “, which reported that China (which previously prided itself, at least in official statements, with recognizing its minorities) has now taken the question about mother tongue out of its national census. What this spells for China’s 88 endangered languages is everybody’s guess.

Essentially, what you see in the process is the increasing dominance of just a few languages out of the 7,000, and they are taking up more and more of the total “language space” that we have on the globe.

If you imagine that there is a linguasphere, comparable to the biosphere, and all these languages are taking some of that space, the growth in number of speakers of just a few dominant languages is overwhelming any growth that may occur in the other languages.

So the majority of the 7,000 languages is getting squashed, squeezed into a smaller and smaller portion of the linguasphere or language space. In some cases, as happened in the Americas and Australia, actual losses, and often catastrophic losses, of speakers have occurred. The end point of that process is language extinction.

This way is how you begin to see how it happens. The big bullies of languages put increasing pressure on the smaller languages.

Loh: Continuing the jelly bean analogy, languages come in many different sizes. Some have tiny numbers of speakers and they occupy a very small proportion of the jar. Now when that proportion reaches the lowest threshold, the jelly beans change colors, and either through their own volition or through coercion they switch to the dominant color, and that accelerates the process until the original color disappears.

Harmon: The ILD tells us that more and more of the world’s people are becoming concentrated into just a few larger languages at the expense of smaller ones. That represents the loss of linguistic diversity short of languages actually going extinct, because up until now most of the language endangerment discussions centered around extinction.

You will often see quotes to the effect that one language is going extinct every two weeks. That may be true, on average, but that’s actually derived from a sort of back-of-a-napkin calculation from the mid-1990s by the linguist Michael Krauss, and now it’s gotten the status of one of these factoids that’s out there, floating free from any actual calculation.

It is important to know how many languages are going extinct, because that’s probably the easiest way to explain this to any layperson. But it’s really important to know that there is a lot of language diversity that’s getting lost before languages go extinct. I think that the ILD tells us quite a bit about that.

Maffi: The ILD is telling you what the process is and how a language gets to the point of extinction, and that there is a lot that is happening before that. It’s a process of greater and greater homogenization, basically–one of the many ways in which we are getting homogenized.

Languages are carriers of cultural values, knowledge, worldviews, and therefore through the reduction of linguistic diversity we are also experiencing a reduction of cultural diversity and greater cultural homogenization.

Braun: The ILD shows little change in linguistic diversity in Eurasia, but as you have pointed out, it charts a precipitous fall in linguistic diversity in the Americas and Australia. Does the index show which specific languages are coming out on top, gaining at the expense of others?

Loh: The index doesn’t get down to that level of detail, but we can see from the underlying data which are the dominant languages. Globally, about 25 languages are spoken by 50 percent of the world’s population. So half the world population speaks one of those 25 languages, the other half of the world’s population speaks one of the remaining 6,975 languages. So there’s an extremely skewed distribution.

The languages on top are what you would guess, Mandarin, English, Spanish, Hindi, Arabic, Russian, Portuguese, French, and so on.

Top 25 World Languages


Top 25 Languages Chart 2.jpg

Chart courtesy of David Harmon and Jonathan Loh. © Terralingua 2010

Braun: Your paper mentions that the language extinction metaphor does not necessarily imply absolute irreversibility. How can the ILD assist with irreversibility?

Harmon: The way we use the term language extinction, we are talking about extinction as a mother tongue. In other words, a first language people are using, not necessarily exclusively, because a lot of people are bilingual or multilingual, but are using it in their daily lives and would consider their allegiance to that language as their first language.

When you talk about extinction as mother tongues, you are talking about people who no longer use their ancestral language for the whole wide range of life-navigating purposes. They may still retain a language that was their mother tongue, but maybe they speak it only during religious ceremonies or special occasions. Or more likely, they will know just a few words of that language. Their competency in being able to produce sentences and express themselves in that language will decline.

What happens is that these languages are lost as mother tongues and they are assumed to have gone extinct. There are no more native speakers of whatever the language might be.

That doesn’t mean that those languages couldn’t be revived as mother tongues, as was the case, most famously I think, with Hebrew, which was dormant as a mother tongue for a long time, and then through some passionate advocacy in the early part of the 20th Century was revived and now is the mother tongue of a lot of people.

Languages can also be revived as languages of heritage, as I call them, which means they might not come back as full mother tongues of many people, but they might be a language retained by speakers and taught to children, and so remain vital in at least some domain. There are a lot of language revitalization programs going on around the world.

We can lose lots and lots of languages in the coming decades as mother tongues, and we should fight to avoid that as much as possible. But even if that happens, the languages might still be around as languages of heritage. There’s a lot of value to having languages like that.

Language revitalization programs need to be supported, even if there isn’t a realistic prospect of ever reviving such languages a mother tongues. They can still have a place in the world of linguistic diversity. They might be diminished in terms of richness, but languages of heritage can still be extremely valuable. Extinction in that sense is not the end of the story.

Maffi: There are extreme cases in which indigenous communities have, after going past the brink of extinction, made heroic efforts to recover as much as possible of their languages from existing data.

Work like this done by linguists and communities has been extremely valuable. I know for instance, of some Native Californian languages where the only extant documentation was really in pages of transcriptions by linguists done in the early 1900s or even earlier, and recordings on wax rolls. Members of those native communities have gone out to museums and archives to retrieve those materials, put them on computers and try to reconstruct as much as possible of the language, then train themselves by talking to a computer.

There are still elders who are fluent speakers, and those elders have been put together with children in so-called language nests so that the children begin to learn their language the way they would normally, just by talking to adults and being spoken to by adults. Or even putting together elders with teenagers in so-called master-apprentice programs where the youth are essentially in full-immersion mother-tongue medium, with the elders speaking to the youth and doing activities together so that the language becomes alive.

In a large number of cases, native-language communities around the world have established school programs. Just the other day I was visiting a school near where I am, on Salt Spring Island in British Columbia, on the west coast of Canada, run by the Saanich First Nation–British Columbia having been a hotbed of linguistic diversity, with 32 different languages, most of which are endangered.

In some cases like this one there are still elders around who speak the languages and who became language advocates and contributed to documenting their language and creating a script for the language, and on that basis language programs have been established in the schools. They are not getting all the financial support that they need in order to establish full-fledged language programs or teaching in the native language, but that’s the direction in which things are going. People express a lot of pride in this. It motivates them to get their language back.

This post is part of a series of features focusing on biocultural diversity, produced with the support of The Christensen Fund.

12418031_10153900711084116_8462971761216697621_nDavid Braun is director of outreach with the digital and social media team illuminating the National Geographic Society’s explorer, science, and education programs.

He edits National Geographic Voices, hosting a global discussion on issues resonating with the Society’s mission and major initiatives. Contributors include grantees and Society partners, as well as universities, foundations, interest groups, and individuals dedicated to a sustainable world. More than 50,000 readers have participated in 10,000 conversations.

Braun also directs the Society side of the Fulbright-National Geographic Digital Storytelling Fellowship

Follow David on Facebook  Twitter  LinkedIn

Forty years in U.S., UK, and South African media gives David Braun global perspective and experience across multiple storytelling platforms. His coverage of science, nature, politics, and technology has been published/broadcast by the BBC, CNN, NPR, AP, UPI, National Geographic, TechWeb, De Telegraaf, Travel World, and Argus South African Newspapers. He has published two books and won several journalism awards. He has 120,000 followers on social media. David Braun edits the National Geographic Society blog, hosting a global discussion on issues resonating with the Society's mission and initiatives. He also directs the Society side of the Fulbright-National Geographic Digital Storytelling Fellowship, awarded to Americans seeking the opportunity to spend nine months abroad, engaging local communities and sharing stories from the field with a global audience. Follow David on Facebook  Twitter  LinkedIn
  • Khalid Mahmood

    It is an interesting article. I am afraid in the table of languages the raking of Punjabi or Western Punjabi is shown low. From the statistics are taken?

  • Christopher.phipps

    This is an important topic and you’ve done a nice job of detailing these researchers’ responses, but you’ve included a set of graphs that are not possible to understand without more explanation. The ILD Key Findings graphs show a line deviating form a 1 on the y-axis with no explanation of what that means. I had to track down the original article ( large pdf) and I’m still a little confused.

    I believe a 1 on the y-axis = no change in each language’s share of world’s population; rising = more diversity, falling = less. I’d like this data explained more thoroughly.

  • Johanna Horn

    I always find language death so increadibly sad and beyond any and all description personally.
    Also understanding different dialects can be just as hard as learning a new language.

About the Blog

Researchers, conservationists, and others share stories, insights and ideas about Our Changing Planet, Wildlife & Wild Spaces, and The Human Journey. More than 50,000 comments have been added to 10,000 posts. Explore the list alongside to dive deeper into some of the most popular categories of the National Geographic Society’s conversation platform Voices.

Opinions are those of the blogger and/or the blogger’s organization, and not necessarily those of the National Geographic Society. Posters of blogs and comments are required to observe National Geographic’s community rules and other terms of service.

Voices director: David Braun (

Social Media