When we read the news related to dying languages, normally this is painted as bad news and it's really important to preserve the language, see Language at risk of dying out (Guardian) or Digital tools 'to save languages' (BBC), for example.

Is this view shared by most linguists, or is it mainly the view of news reporter or anthropologist? If we think about it, maybe human communication would be better off if we have fewer languages. The diversity of languages cause communication difficulty, and in some cases can even cause conflicts, war, etc.

So what are the scientific reason for preserving dying languages, if any?

  • 4
    As for whether this view is shared by most linguists, it isn't. I have come across some who care about all languages, others who care about ancient languages from just their own countries, and still others who don't care about any of them. I'm not exactly a linguist, but I care for very few languages and none of them is even remotely "endangered". I can't even give you percentages for each of the three because I haven't asked each linguist in my circle.
    – prash
    Commented Mar 5, 2012 at 20:00
  • 2
    "If we think about it, maybe human communication would be better off if we have fewer languages. The diversity of languages cause communication difficulty, and in some cases can even cause conflicts, war" --- It seems like you were using these sentences as a devil's advocate for a scientific reason to want fewer languages, but something being 'better off' in this context is not really a scientific question, but an ethical one. So I am not sure how that would be any sort of scientific reason. Of course, scientists tend to be humans and thus have ethical beliefs :P. Commented Mar 6, 2012 at 4:32
  • 2
    Yeah, I'd also like to question the assumption that fewer languages would lead to less war. It's true that ethnic conflict often does pit speakers of different languages against each other. But in multiethnic regions where everyone speaks the same language, ethnic conflict still arises -- it's just rationalized in terms of race or religion or access to resources instead of in terms of language. (Still, this is a very good question!) Commented Mar 6, 2012 at 16:08
  • @Dan I didn't say that having fewer languages will eliminate wars. Off the top of my head, there have been fewer wars in Latin America than Africa, despite similar history, although I don't have any data to show that the language homogeneity is related to this in any extent
    – Louis Rhys
    Commented Mar 6, 2012 at 17:05
  • 1
    language is culture, the more languages, the more culturally rich the world is.
    – CptEric
    Commented Jun 29, 2016 at 6:44

9 Answers 9


Newmeyer and Emonds (1971: 301) mention that a

"...result of the reliance on outside funding agencies is the occasional deliberate falsification of the nature of linguistic work."

Probably a bit too cynical, but it comes close to the present situation. The endangered languages cause has captured the imagination of the public, and it is the best shot the field has had at making its work recognized as worthwhile since the idea of automatic translation. Even though language preservation falls more within the bounds of applied linguistics, it is something that professional linguists have found themselves involved in, partly because it is a good selling point for government funding agencies, and partly because they would like to find a way of giving back to the communities that they work in (in the case of field linguists).

For those interested in how language endangerment has been approached by traditional linguists, a "classic" publication is a collection of essays published in 1992 in Language (Hale et al 1992). The previous year the French journal Diogène (no.153) dedicated an issue to the topic (though with much less exposure). A good and very recent report on language vitality by family can be found in Language 88(1):155--173 (similar open access material here). The contributors to the 1992 article stressed the following points:

  1. Linguists should be concerned about the rapid loss of their data sources.
  2. Language extinction is progressing at a very rapid rate.
  3. Speaker communities must be empowered to save their languages.
  4. Loss of linguistic diversity is as serious a threat to the world's natural resources as is the loss of biological diversity.

The first two points are generally uncontroversial, though linguists tend to take different stances as to the last two. Peter Ladefoged, for example, published a reply to the Hale et al paper, where he noted that:

  1. Communities should have the right to choose the fate of their languages. In some cases, the best policy for a young country, to produce national unity, is to discourage multilingualism, since it is thought to reinforce tribalism.
  2. Cultural diversity does not necessarily require that different languages be spoken.

On the first point, consider this statement attributed to a Cameroonian government official upon learning that the most recent version of the Ethnoloque included more languages for Cameroon than the previous version:

Vous voyez ce que je vous dis à propos des linguistes Camerounaises là? Leurs langues là se marie maintenant et cette fois ci, elles ont produit 17 nouveaux langues pendant cinq ans. [You see what I'm saying about these Cameroonian linguists? Their languages are getting married and now they've produced 17 new languages in five years.]

Anyways, to discuss the endangered language issue with more sophistication, it is nowadays common to distinguish between Language Documentation and Language Conservation. An open access journal has recently been founded for academic studies devoted to both of these topics. Language documentation concerns making transparent records of languages which can be used for research even after the language ceases to be spoken. Language conservation refers to activism by or on behalf of communities speaking endangered languages to either protect the language from endangerment, or to try to revitalize the language before it falls into complete disuse.

So, linguists do not necessarily have an interest in language preservation (though many do), but most have an interest in language documentation. The comments by John McWhorter are not very far from the approach some linguists take:

I hope that dying languages can be recorded and described. I hope that many persist as hobbies, taught in schools and given space in the press, as Irish, Welsh, and Hawaiian have.

However, the prospect we are taught to dread — that one day all the world's people will speak one language — is one I would welcome. Surely easier communication, while no cure-all, would be a good thing worldwide.

on the opposite side, there is this often-repeated quote attributed to Ken Hale:

The death of a language is a disaster. It’s as if someone had dropped a bomb on the Louvre.

Hale, Kenneth; Krauss, Michael; Watahomigie, Lucille J.; Yamamoto, Akira Y.; Craig, Colette; Jeanne, LaVerne M. et al. (1992). Endangered languages. Language, 68 (1), 1-42.

Newmeyer and Emonds. 1971. "The Linguist in American Society." Papers from the Seventh Meeting of the Chicago Linguistic Society, 285-303. Chicago: Chicago Linguistic Society.

  • Great answer @jlovegren - this captures the situation well :) Commented Mar 10, 2012 at 8:45
  • Good answer! A balanced overview of the issue. I wonder about this view that it is OK for languages to die off because them more people will speak the same language. Could it be more popular among a culture where few speak a second language, such as in America? Because I don't think I've ever met anyone holding that view myself (the general view is that most people speak a second language for communication so they're fine).
    – Cerberus
    Commented Sep 14, 2018 at 13:26
  • @Cerberus the anecdotes in Ladefoged's paper may be of interest. They're about languages gradually falling into disuse, but the national culture is still one where most of the population speaks 2 or more language.
    – user483
    Commented Sep 16, 2018 at 1:27

Languages encode a lot of information about its surroundings. Names for plants and animals, what the plants and animals are useful for, various phenomena, things that are relevant for the speakers. Think of it as a specialized encyclopedia. When a language dies, the encyclopedia dies with it. Often, the changeover in language is followed by a changeover in culture as well, so that the previously relevant information is no longer relevant and there is no agreed-upon way of talking about it.

Want to know what's edible in the area you are in? Sorry mate, the locals have been assimiliated into a more westernized culture and only eats the same stuff you do now. Want to know how the locals used to make the paint of a particular shade of blue that is more long-lasting than anything found in the west? Sorry, the conquerors killed the people who knew how for being witches and cut out the tongues of those who spoke the language. Want to know whether this fungus has been used for medical purposes? No can do, the locals who used the fungus died of cholera in the fifties, etc. ad nauseam.

For science, the loss of information is a tragedy.

  • 2
    +1 good point, language is a form of culture in my opinion Commented Mar 3, 2012 at 18:49
  • 8
    This answer goes beyond just the value for individual culture. Culture can encode (from a scientific standpoint) amusing nonsense, like the names of angels and types of unicorns. This answer is talking about the possibility of languages encoding knowledge, like the properties of the plant "feverfew" are encoded into the word. Algonquin languages were an even better example of this because names of plants were made up of several descriptive morphological parts, some of which may have been folk science, some nonsense, some trivial observations. Commented Mar 5, 2012 at 15:43
  • 2
    @MatthewMartin : things that are "amusing nonsense" for you, are valuable cultural and artistic treasures for others. Just because poetry and music don't typically encode scientific data, does it mean we should ban poetry and ban music (or let them die), because they only contain "nonsense" ??
    – vsz
    Commented Feb 26, 2020 at 8:19

Jlovegren is right to distinguish between documentation and conservation. In fact, increasingly we're seeing a three-way distinction being made:

  1. In language description, the goal is to develop generalizations about the grammar of a particular language. This leads to results like "This language has an iambic stress pattern," "The basic word order in this language is SVO," or "This language has twenty demonstrative pronouns, distinguished by the shape, position and location of the referent." On this approach, data is valuable mainly as evidence for a particular generalization: e.g. "See, you can tell that there's a shape distinction among the demonstrative pronouns, because this one is only used to refer to round things [examples] and that one is only used to refer to flat things [more examples]." Often, negative data is especially important: e.g. "If I try to use this pronoun to refer to round things, speakers tell me that I'm using it wrong."

  2. In language documentation, the goal is to record examples of the language in use, ranging from everyday conversation to oral literature, formal or ritual language, special terminology such as plant names, and so on. On this approach, data is seen as valuable in its own right, whether or not it serves as evidence for some sort of descriptive claim. The result of language documentation is often a collection of texts or recordings: e.g. "Here's ten hours of video of traditional epic poetry being performed. We've also transcribed the poems in the original language, translated them into English, and added a few notes about the grammar." But documentary work can lead to descriptive claims as well: e.g. "The distal demonstratives are rarely used in conversation in this language, but we've noticed that they're very common in traditional poetry."

  3. In language conservation, the goal is to help keep the language from dying out. This often means starting up development projects or getting involved in local politics: building a school, training teachers, founding a radio station, lobbying for a bill that protects indigenous rights. It doesn't necessarily involve any sort of data collection. When you do collect data, it's generally in order to write textbooks or teaching grammars, or to build up teaching materials to use in school: e.g. "This poem has some great examples of the distal demonstratives in use. Let's read it for the sixth-grade class when we teach that unit."

These three sorts of work aren't totally separate. For instance, if you want to write a textbook for language learners (#3), then you need to know some things about the grammar of the language first (#1), and it would be good to have some literature to cover in class as well (#2). But the distinction is still relevant to the question, because each of the three is valuable for a different reason.

It sounds like Louis Rhys is asking about #3: language conservation. For what it's worth, this sort of work can actually be quite controversial — not necessarily among linguists, but among the speakers of endangered languages themselves. Some communities are very interested in teaching their heritage language to their children or grandchildren and bringing it back into daily use. Some communities are actively uninterested in their heritage language — they feel that it marks them as "old-fashioned" or uneducated, or that it prevents them from getting education or jobs, and they would much prefer that their children or grandchildren just switch to the local majority language. And in some communities, there's intense debate between the two points of view.

But if a community wants to make sure that their children keep learning their heritage language, then a skilled linguist can help, by preparing educational materials and helping organize classes and so on. The community still has to do the bulk of the work; our job is to serve as a resource. The linguists who do this sort of work tend to see it as community service: it's valuable because it helps a community achieve their goals, and helping people is a good thing. Of course, if a community is dead-set against using their heritage language, and a linguist tries to badger them into using it anyway, that isn't particularly helpful or valuable. (And anyway, it isn't likely to succeed.)

For language description, the value is largely scientific: we have theories about these languages that we want to test, and the data won't be available in 20 years, so we have to test them now. For language documentation, there are several sources of value. It can be valuable for science, by preserving data that would otherwise be lost. But it can also mitigate the sort of cultural damage that kaleissin and Alenanno are talking about: even if a child doesn't grow up speaking his parents' or grandparents' first language, he can still read the stories they told — either in translation or, with effort, in the original language — and that gives him a connection to his heritage that he might not have otherwise. And finally, it can be valuable in the same way that restoring old artwork is valuable. There are some truly wonderful pieces of literature that are being passed down orally in small language communities, and if the members of those communities want to share them with a larger audience, then it would be a shame if they were just lost.

  • 2
    Excellent answer, although I would like to add some extra emphasis to the purpose of language documentation work. More and more, it's considered extremely important (and part of standard procedure) to use texts collected in documentation work as the basis for a resulting description. Rather than using them as a means of 'finding the example' of an observed phenomenon, they are a departure point for investigating the grammar - you see what you get in a given text, then add to your investigations with other tests and queries. Modern descriptions are based on more and more natural language use. Commented Mar 10, 2012 at 8:38

From a purely linguistic scientific point of view, i.e. abstracting from all humane feelings such as compassion &c., I would say the most important reason to care about languages dying out is data. Generally, those languages that die out, are also the least attested and least researched ones. Since written material and recordings are usually scarce, with the death of the last native speaker, we irreversibly lose the possibility to ever properly investigate the language. This is bad because linguists care about languages (or at least they should), and they care about parallels, cognates, reflexes and other things that can help them study other languages.

The real reason, however, why so many linguists are nowadays interested in dying languages is fashion, in my opinion. The trend goes hand in hand with ecology, manifesting appreciation for other cultures and other typically late 20th c. points of view, and brutally speaking it has become fashionable in some centres to conduct field research. This is not to say field research hasn't been done earlier; just not in that scale.

Also, not all people interested in languages are necessarily sit-at-home type of people. Field research gives them a great opportunity to study what they like the way they like it.

But the phenomenon is not without a dark side, either. Field research is a costly enterprise and requires a specific set of mind, which does not cooccur terribly often with a scientific attitude. As a result, a great deal of field research nowadays is done by 'curiosity hoppers', i.e. collectors of linguistic pecularities (and their own fame) from a multitude of randomly chosen exotic languages, none of which they ever become specialists in. Sort of cabinet of curiosities kind of attitude. As a consequence, a lot of data thus collected is totally unreliable and next to useless, and proper linguists find it very difficult to obtain further funding because the languages they want to investigate are, in theory, already described. Edit: removed a specific example because apparently some users consider a warning against dilettantism to be equivalent to wishing a language to die out. My aim was not to start a flame so just see the history of this post if you're interested.

  • Do you have references to back the statement that the interest that linguists take in dying languages is fashion?
    – Tsundoku
    Commented Nov 4, 2016 at 14:22

A lot has been said (and this is one of the "eternal" topics in Linguistics I think) but I wanted to give my contribution. In your question you say that "if we think about it, maybe human communication would be better off if we have fewer languages".

I highly disagree with that, and I suppose most linguists or anyone involved in languages study would, too. And the reason is simple: that is not the solution. The perfect solution would be for people to become multilingual, and not let languages die. If that was a true solution for avoiding conflicts and wars, how would you explain that even people from the same country harm each other? It would be too simplistic to relegate the wars' cause to a matter of different languages. Languages are the key to understand its people. When you start learning a certain language, you also start understanding that country, its speakers, its culture.

About that, it has been said many times but I'll repeat it: languages and culture are intrinsically related. Or better, languages are culture. Without language, you wouldn't have culture. So when we talk about preserving, or in the worst cases "saving the salvageable", we are not simply talking about words, letters and grammar rules, but anything that is correlated to that. I'm not sure one would call this a scientific reason, but I suppose it is, beyond being an absolutely good reason.

Luckily in some cases, saying that a language dies, doesn't necessarily mean that we lose everything about it; Latin and Ancient Greek are a great example: they are dead languages (i.e. no native speakers), but the material about them is plenty. However, other languages are not so lucky. The "luckiest" endangered languages are the famous ones, but let's think about the unknown ones. They are in the worst situation.

I'm not a linguist, technically speaking, but when a language becomes extinct, it's a great loss, and not just for the nation involved but for everyone. I don't want to sound too cheesy, but when a language faces extinction, it's not a language that becomes extinct, it's a part of mankind history. Extinct languages are irreplaceable.


The only scientific reason I can think of is that the death of a language deprives us of interesting information that could be used to research universal grammar, etc.

But the main reason is probably non-scientific, just as we don't like polar bears to go extinct or mediaeval churches to collapse: humanity has certain ideas about preserving the things it likes, and exotic languages are cool.


May I recommend the book 'Language Death', by David Crystal? It is a fantastic recount of all the reasons we should care, and be slightly saddened when another language is lost from the world. I don't always view it as a bad thing. We all should accept that times move on, and nobody misses Akkadian nowadays or wishes that Avestan was still being spoken on the streets so we could go around asking for 'satem' of everything (:D), but times do have to move on and if we move onto a world in 50 years with only 1,000 (500?) languages, that's the way it will have to be and we all have to adjust to that. However, it's still not the case that we're losing something special every time those last speakers perish.

  • Let's hope time moves on with languages still being spoken. I actually wish some languages were still alive right now. Why not even Akkadian? :)
    – Alenanno
    Commented Mar 7, 2012 at 21:00

Another point, usually overlooked, is that as soon as a language has a literature of some description, a part of it (rhythmic poems/prose, untranslatable wordplays, etc.) is irretrievably lost when the language dies. I'd be more concerned about this than not knowing what some particular word was or meant. Valuable texts are routinely lost in living languages too, but the loss is much starker when there's already in principle no-one who could appreciate them -- losing not only a source but also an entire way for an unknown, possibly sublime influence for good


I am not a linguist. I'm merely a person who sometimes talks, listens, writes and reads.

Many languages die because some government invades an area, and discriminates against the indigenous people there. They ban the language outright, or don't teach it at school, instead imposing their own language on them.

Usually such governments discriminate against these people in other ways. For example, they may not give them the vote, they may ban some of their customs, they may discriminate in terms of employment, farming, marriage, etc. This kind of discrimination is seen as bad.

Some people, some of whom may be linguists, are affected by what Wikipedia calls the association fallacy (and TV tropes calls Hitler Ate Sugar). If discrimination in X, Y, and Z, and also language death, are all the result of a government's policy against a certain race, then since the discrimination in X, Y, and Z are bad, then the association fallacy will lead some people to think language death must also be bad, merely because they were all the result of the same over-arching policy.

Is this logical? No. But linguists are not totally logical, any more than the rest of us are.

  • 1
    I think this doesn't exactly represent why most linguists want to preserve languages. Even if your speculation (because that's what it is) is true, I think it would be safe to say that it would affect only a minority in the Linguistics field. Not to mention that not all languages die because of discrimination/invasion. Plus in the end you say that linguists are not totally logical, what do you mean by that? Can you bring some (verifiable) examples for that?
    – Alenanno
    Commented Mar 6, 2012 at 13:36
  • As has been explained above, there are pretty solid reasons for linguists to worry about languages dying out, and I can see no point in speculations of this sort. What are you basing this on? Have you spoken to some linguists and they told you this is why they care? To me it looks like you're trying to sell us your own point of view pretending it's what linguists believe. As an actual professional linguist, I strongly object both to this pov and to how you're trying to impose it.
    – kamil-s
    Commented Mar 6, 2012 at 14:23
  • 1
    I said this in the chat, and I will repeat it here: to me, you seem to have ended up blaming linguists for lack of logic in an argument which is in fact your own argument which you somehow believe to be also shared by actual linguists. Brilliant logic.
    – kamil-s
    Commented Mar 6, 2012 at 14:42
  • @KamilS. That is called an association fallacy :)
    – jcm
    Commented Mar 6, 2012 at 14:54
  • 1
    I agree with @Alenanno. What you say is highly controversial at best, and even if all of what you say is true, it doesn't reallly answer the question. This is a Q&A site not a forum, please focus on answering the question. Otherwise your posts will be in danger of downvotes and delete.
    – Louis Rhys
    Commented Mar 7, 2012 at 1:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.