On Sun, Jul 10, 2011 at 21:40, Amir E. Aharoni amir.aharoni@mail.huji.ac.il wrote:

2011/7/10 Milos Rancic millosh@gmail.com:

...

and one in revived language (Manx).

Ahem.

The definition of a "revived language" is very controversial, but if you count them, don't forget Hebrew (120,000+ articles) and Cornish (2,000+ articles).

Better construction would be "the language in the early stages of revival". We'll have one more soon (Livonian). Both Cornish and especially Hebrew are living languages.

http://en.wikipedia.org/wiki/Extinct_language

"It is believed that 90% of the circa 7,000 languages currently spoken in the world will have become extinct by 2050, as the world's language system has reached a crisis and is dramatically restructuring."

How is Wikipedia going to affect this language disaster? WMF 2050 goals ideas : ) ?

2011/7/11 Milos Rancic millosh@gmail.com

On Sun, Jul 10, 2011 at 21:28, Milos Rancic millosh@gmail.com wrote:

...

• 270 Wikimedia languages (however, you would see below that the term

"language" is not quite precise)

One note: there are 270 languages counting Simple English as a constructed/controlled language. If it isn't counted, there are 269 languages.

---

foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

It won't be possible to save languages going extinct. Even if two or three people start writing a Wikipedia in such a language, it will die out as a spoken language, eventually, not later than it would without a Wikipedia. I think it's nice to have a corpus of encyclopedic articles in such languages, but more important for the goal of Wikimedia to make knowledge accessible to all people of the world, is that there is a useful Wikipedia in at least one language any given person can read. I would estimate that we won't ever reach 300 (open) Wikipedia language versions, because many of the smaller ones will be closed sooner or later due to permanent inactivity, and that's perfectly fine.

Th.

2011/7/11 emijrp emijrp@gmail.com:

@Thomas and @Andre: I know that it is very hard to mantain a Wikipedia in 'remote' or 'almost extinct' languages, but, if we don't save as much as we can of them (including words, grammar, culture, social values), how are we going to offer 'all human knowledge' ?

We offer this knowledge by having articles about the grammar, culture and social values of these languages, and by having wiktionary entries for the words of these languages. We do not need to have the human knowledge *in* these languages. It would be nice, but it's not necessary to reach the ultimate goal to offer all human knowledge.

How many people don't understand any Wikipedia today?

Of those who can read at all, probably much less than 1%. The problem are those people who can't read.

Th.

Why not? Why do people need to learn English to read a complete encyclopedia? Biased thinking.

They don't need to learn these "big" languages, they already speak them. The people learn other languages not because they want access to Wikipedia, but because they want to communicate to more other people. They would learn these languages also if no internet and no Wikipedia existed, they did learn English, French, Spanish, Portuguese, Tok Pisin, what have you, long before Wikipedia.

Th.

2011/7/11 Amir E. Aharoni amir.aharoni@mail.huji.ac.il:

2011/7/11 Thomas Goldammer thogol@googlemail.com

...

...

How many people don't understand any Wikipedia today?

Of those who can read at all, probably much less than 1%. The problem are those people who can't read.

For persons who can't read it's far better to learn reading first in their own language.

Sure, it is. But, wake up, that's not how the world works. Children who go to school do usually learn the official language or the lingua franca of their region. And those who are not lucky to have access to a school won't usually learn how to read anyway. Sure, one can dream of a perfect world, where there is a school all the way up to final graduation in any language of the world where children learn to read and write their own native language, but that's a dream, not reality. It would have maybe been possible to make this come true 100 years ago, but it's not realistically possible anymore, most languages are just moribund, children don't even learn them anymore. You can't fix that.

Th.

2011/7/11 Thomas Goldammer thogol@googlemail.com

It won't be possible to save languages going extinct. Even if two or three people start writing a Wikipedia in such a language, it will die out as a spoken language, eventually, not later than it would without a Wikipedia. I think it's nice to have a corpus of encyclopedic articles in such languages, but more important for the goal of Wikimedia to make knowledge accessible to all people of the world, is that there is a useful Wikipedia in at least one language any given person can read. I would estimate that we won't ever reach 300 (open) Wikipedia language versions, because many of the smaller ones will be closed sooner or later due to permanent inactivity, and that's perfectly fine.

I'll never lose hope that we'll have a full-blown encyclopedia in each of the 7,000 languages, but even if we won't, it's still very much in the scope of Wikimedia's mission to have full collections of free texts in all of them - folk tales, religious texts, any spoken recordings, whatever. So for many languages a WikiSource project may be even more relevant than an encyclopedia.

...

Yep. Wikipedia community have to start to rescue that knowledge in danger. Knowledge is being destroyed everywhere: digital (dead links) and analogical (dying languages/cultures, places [without pics], etc). _______________________________________________

Right! Digitalizing text corpora in any available language is an important part of Wikimedia's goals to achieve. *That's* where we really can do something, and that's what the foundation must support more than it does so far. (The general wikisource wiki is a mess, set some paid people down to clean it up and make it a platform that is usable to digitalize texts in whatever language you come across, ancient languages like Sumerian, or Old Mayan, or what have you, included.

Th.

On Mon, Jul 11, 2011 at 11:32, emijrp emijrp@gmail.com wrote:

http://en.wikipedia.org/wiki/Extinct_language

"It is believed that 90% of the circa 7,000 languages currently spoken in the world will have become extinct by 2050, as the world's language system has reached a crisis and is dramatically restructuring."

How is Wikipedia going to affect this language disaster? WMF 2050 goals ideas : ) ?

Extinction estimates are outdated from the point of contemporary technology. Many languages now have much more chances to survive than it was during the time before Internet. And, unlike natural sciences, that part of linguistics is based on older data.

Every language incorporated in Google Translate has a lot of chances to survive. If Google starts to support smaller languages (or some other project develops free to use translator), those languages will have good chances to survive.

Wikipedia and other Wikimedia projects are also important factor. Stable open content projects on Internet allow people from all over the world to contribute, which includes scattered diaspora, which usually have better education.

There are also other factors. For example: * Sorbian languages [1] (Upper [2] with 40k of speakers, Lower [3] with ~10k of speakers) will survive as cultural artifact, connected to the identity of Sorbians. It could be compared more with remembering genealogical tree than as useful medium for communication. The fact that Germany is rich country has two opposite tendencies: from one side, Sorbians are fully integrated in German society and they tend to know German better than Sorbian; from the other side, as German citizens, Sorbians are rich and if they want (as they want) to preserve the language of their ancestors, they have means to do that. Former estimates didn't count any of Sorbian languages to have chance to survive. * Alekano language [4] (~30k of speakers) also wouldn't be counted into the survivors. However, it has coherent community inside of the diverse central Papua, it has university and university has internet access (although expensive at the moment). * However, language with the similar speakers population in rural India doesn't have a lot of chances. Non-Han languages of China can to be much smaller to have good chance to survive than Han languages of China. And so on.

Note that estimates from the past (and likely from the present) count that no language with less than 1M of speakers would survive 2050.

[1] http://en.wikipedia.org/wiki/Sorbian_languages [2] http://en.wikipedia.org/wiki/Upper_Sorbian_language [3] http://en.wikipedia.org/wiki/Lower_Sorbian_language [4] http://en.wikipedia.org/wiki/Alekano_language

On Mon, Jul 11, 2011 at 14:57, emijrp emijrp@gmail.com wrote:

2011/7/11 Milos Rancic millosh@gmail.com

...

Note that estimates from the past (and likely from the present) count that no language with less than 1M of speakers would survive 2050.

If Wikimedia projects and WMF leave to die 90% (or 80%, or 70%, or 60%) of current languages in the next 40 years (we will be alive to see it, probably), then both are failures.

I think (but I am not sure) that I posted this link [1] here a couple of weeks ago.

Speaking just about languages, the situation is approximately the next:

speakers total speakers number of languages 100M+ 2,514,548,848 9 10M-100M 2,376,900,757 78 1M-10M 950,166,458 303 100k-1M 284,119,716 900 10k-100k 61,223,297 1837 1k-10k 7,823,891 2025 100-999 460,911 1039 10-999 12,664 343 1-9 528 134 sum 6,195,257,070 6,668

So, number of languages with less than 10k is approximately 45%, but it is around 8M of people in total or 0,0015 of world population. It is highly likely that that number of languages won't exist in ~100 years. (Some of those below 10k will survive, but some of those above 10k won't.)

To make those languages viable enough to survive -- much more work than just our is needed. I am sure that 10% of military budgets of the world countries for one year would preserve all languages, but that's the other issue. Basically, that's not our failure as Wikimedians, but failure of our civilization.

[1] https://spreadsheets.google.com/spreadsheet/ccc?key=tCwO11tFPLPB-SJafDesypg&...

On Mon, Jul 11, 2011 at 19:54, Thomas Goldammer thogol@googlemail.com wrote:

efforts are made, we could get those into Wikipedia business. But, we would need a boatload of money (yes, why not taking it from the military budgets - but who should hand it over to us ;) ) to go around

The most important serious work on formalization of Serbian language in 1990s was done by US military. In other words, if you want to get significant resources to make computers able to "read" your language, you should just create climate which would lead to US intervention in your country. Unsurprisingly, during the last decade Arabic was in focus.

On Mon, Jul 11, 2011 at 22:42, M. Williamson node.ue@gmail.com wrote:

To be honest, I don't think 10k is a fair threshold. Many languages with hundreds of thousands of speakers will likely go extinct by 2050, due to high levels of bilingualism and low levels of children learning the language. This language shift is particularly acute on the American continent, where some languages that have been able to survive and remain relatively stable since conquest are now looking increasingly troubled and threatened by Spanish. Even languages that can still be regarded as "safe", like Quechua, can be said to be "melting at the edges" - though there is no doubt Quechua will be alive and have millions of speakers in 2050, there is a good chance that a good percentage of the grandchildren of living Quechua speakers will only have a passing knowledge of the language.

With the rapid urbanization that is currently occuring in many parts of the developing world, language death seems likely to accelerate. When you come from a group of 100,000 speakers, and all of them move to a city where the majority language is Nigerian Pidgin English, how many generations will your original language survive? Chances are, not more than 3. Linguistic diversity in Africa was still actually rising (!) until the early 1990s, but since then it has begun a sharp decline, much like what had already been seen in Europe, America, and Australia, with the difference that the sharp declines in Australia and America can be attributed exclusively or nearly exclusively to pressures from European colonial languages, while in Africa there is also pressure from larger African languages like Swahili or Lingala or Yoruba on smaller African languages.

When bilingualism reaches over 50% in a community, the only chance for intergenerational language maintenance is if there is a higher prestige for the native language than for the outside one. If the "prestige" of a language is perceived to be less than that of a LWC (language of wider communication), like Spanish or English or Swahili, and people are already bilingual, the native language will very quickly fall into disuse, which is followed by extinction.

Some people think that a large number of speakers is a good guard against extinction, but unfortunately there are several cases of hundreds of thousands or even millions of speakers of a language undergoing intergenerational shift, and such "large" languages can go extinct very quickly as well when there is very low prestige and very high bilingualism.

Your reasoning is of Industrial Age: People come to the city, the only way to be informed is through the local newspapers and, logically, the grandchildren barely know language of their grandparents.

However, that's not the case anymore. People are using internet as primary source of information more and more. And it is always easier to read in native language, than in local lingua franca. Of course, *if* information exist in native language and that *is* our job.

Three generations in our age is a lot. We are able to create environment for thousands of languages In 50 years. Creating 100 new Wikipedias per year is impossible task now, but, with properly directed efforts we could reach that number.

Besides that, developing countries are becoming richer. It is possible that many languages would be preserved in the way in which Sorbian languages are.

On Tue, Jul 12, 2011 at 00:34, M. Williamson node.ue@gmail.com wrote:

No, Milos, my "reasoning" is not "of the industrial age". It is backed up by first-hand experience and by research. People who live in cities are by nature a part of a larger urban community, with few exceptions (if there is some kind of enforced segregation, like ghettoization of Jews which often preserved Yiddish in urban environments pre-holocaust), which means it is very, very, very highly likely that their children will learn the LWC of the city in addition to the language of their parents. There is also a much higher chance that children who grow up in the city will marry someone outside of their own linguistic group, which often means their children will be raised primarily in the main language of that city. Now, like I said in my original e-mail, a 100% bilingual minority group does not usually stay bilingual for more than a couple of generations, especially in an urban environment where they must interact on a daily basis with people who do not speak their language, and often, might only use their own language at home.

Also, keep in mind that the idea of "generations" varies from country to country. In some countries, people typically don't give birth until mid-late 30s; in others, it is in the teenage years, so things like language death happen a bit more rapidly as the new generations come more quickly.

You are talking about cities like New York and London are; or about communities which have strong social reasons to forget their native language. Besides that, you are not counting the fact that people who speak the same language are able now to reach each other more easily.

In the most of the cases migrants are forming neighborhoods and they keep their language because they are semi/self-ghettoized. Such languages have chance to survive if there is written culture. And if the language hasn't disappeared during the time when descendants stop to struggle for survival, some of the descendants are starting to learn the language again.

Your scenario is well known everywhere on the planet. However, many things are changing thanks to the Internet and much better possibilities for communication.

Here are two opposite examples from Serbia:

* My grandfather came from the Torlak area to Belgrade with my father who was ~10 years old. Prestige language of Torlak speaking population from Serbia is Serbian and Torlak doesn't have written culture. My father knows Torlak. My mother is from other part of Serbia and I didn't use to go to the village near Pirot often. I understand Torlak well, but I don't speak it. However, my brother and my sister speak Torlak. But, it is true that their children would know just a couple of Torlak words.

* Roma languages from Serbia are in constant state of shrinking. Their prestige language is Serbian, but not just that: their prestige ethnicity is Serbian. I know a number or the third generation of integrated Roma who don't know a word of any Roma language. But, some of those from the third generation are now leaning Romano-Serbian, as a couple of years ago it's finally got written form. It is possible that we'll have Wikipedia in Romano-Serbian in a couple of years.

Note that the state of their language was probably among the worst for a language of that amount of speakers (170k): Very negative attitude toward language *and* ethnicity, no written culture, high rate of marriages with the dominant population; the only social environment in which it is used regularly are slums and all of them want to leave slums and integrate into the majority population. However, it is highly likely that all of Roma languages will survive, thanks to the increased awareness for their situation and Internet and mobile phones.

They don't need anymore to be in slums to be able to communicate in their language. A lot of them won't use Romano-Serbian in written form, but better communication will allow them to raise children who would know their language and who would be willing to educate their children in the language of their grandparents. Just a couple of decades ago such scenario wouldn't be possible.

Also, I should note that your story about Romani seems to prove the exact opposite of what your point seemed to be:

"I know a number or the third generation of integrated Roma who don't know a word of any Roma language."

Yes, this is an interesting statement. Do you know any ethnic Serbians living in Serbia who don't speak the Serbian language? What about ethnic Swedes living in Sweden who don't speak Swedish? In the case of endangered languages, the number is growing every generation. Even if there are "good signs", the point is that the percentage and number are shrinking, and speakers that are "lost" are extremely unlikely to "return" in future generations (meaning, if you don't speak the language, it's really unlikely your grandchildren will).

Lower-prestige minority languages do not survive in cities without complete ghettoization. Ever. Period. For a couple generations, perhaps, but when your great-grandmother was bilingual in the minority language and the LWC, it can be almost guaranteed, in an urban setting, that you are monolingual in the LWC.

2011/7/12 M. Williamson node.ue@gmail.com

Milos, it is a fantasy of many that is not supported by research, that just because people are rich or have technology, their language will magically not die. I have never been to New York or London and am not talking about such cities; I am talking about medium-sized cities around the world (although large ones like Lagos and Sao Paolo are probably at the forefront of language cemeteries in these days).

Where I live, there are approximately 20 languages spoken by indigenous people within a several hours drive; this includes languages like Navajo with hundreds of thousands of speakers, as well as languages like Mojave with now not so many. We have the whole range of situations here for language survival/death, including languages dying in rural isolated communities, languages dying in or near urban areas (Phoenix is a medium city, with about 4 million people, but it is also fast-growing, when I was born here a little over 20 years ago there were 2 million), and languages whose survival appears to be ensured for at least a few generations, based on current statistics of intergenerational maintenance, bilingualism, and other predictors of language death or survival (only two, unfortunately). We also have on top of that the case of Spanish, which is also not being maintained intergenerationally, and several smaller immigrant and refugee communities with not-abnormal intergenerational linguistic tendencies.

All of these languages are now written; some are taught to children in schools, and some are blessed with very fierce defenders who do everything they can to promote their language among their people. However, in the fight against the encroachment of the LWC, the best weapon is simply to not speak the LWC, which people are not likely to do because of all the benefits that come with being able to speak a LWC. Eventually (usually 1 generation when they go to a school taught in LWC, perhaps 2 when they don't go to school at all), most members of the ethnic group find it is easier for them to express themselves in the LWC, their skills in the native language begin to erode, and they prefer to speak the LWC even with people from their own linguistic group of the same age. This is especially true when they share all living environments, going to school and work together and even marrying one another. Generally, the people in the 3rd generation are only passive bilinguals, understanding but not speaking the language. Sometimes this stage is delayed until the 4th or 5th generation, but I can't think of a case in history where it didn't end up happening with full social integration. If you can't speak a language, you can't pass it on to your children, and then it dies.

This process is repeating itself around the world, not just with poor and illiterate people, but also with rich and well-read people who find more economic and social benefits in using a LWC. This is unfortunate, but so far nobody has been able to find a remedy, and just writing encyclopedias in minority languages doesn't seem like a viable solution to the problem of language attrition and death, although perhaps it helps to raise the prestige a bit.

There are data which I can show you which demonstrate that the linguistic diversity of the world is decreasing at an alarming rate. Although some speakers of dying languages love their languages very much and spend a great deal of time working hard to keep them alive, this is not the case for people with neither the time or the resources to give much thought to saving their language. Even people who fight tend to lose this battle, unfortunately, with only a handful of exceptions.

Intermarriage, when it becomes widespread, inevitably leads to language death even more than migration to cities. Perhaps someone who is half-Roma will speak Romani, but what if the half-Roma marries a non-Roma? Their children will be only one quarter Roma. As people mix more and more, the likelihood that the minority language will survive decreases to be almost negligible.

Sadly, even people who live in a physical/social environment otherwise totally cut off from the LWC may switch to the LWC over several generations if it is the language of school and school attendance through the end of secondary school is high or universal, and also the language of government and perhaps radio or television. Wars, or mandatory conscription (practiced in quite a few countries) tend to increase language death as well, since armies need to share a common language, and soldiers often return home and teach their children the language they learned in the army (this struck a blow to many languages in the US, as well as Breton in France and many countless other languages around the world).

2011/7/11 Milos Rancic millosh@gmail.com

...

On Tue, Jul 12, 2011 at 00:34, M. Williamson node.ue@gmail.com wrote:

...

No, Milos, my "reasoning" is not "of the industrial age". It is backed

up by

...

first-hand experience and by research. People who live in cities are by nature a part of a larger urban community, with few exceptions (if there

is

...

some kind of enforced segregation, like ghettoization of Jews which

often

...

preserved Yiddish in urban environments pre-holocaust), which means it

is

...

very, very, very highly likely that their children will learn the LWC of

the

...

city in addition to the language of their parents. There is also a much higher chance that children who grow up in the city will marry someone outside of their own linguistic group, which often means their children

will

...

be raised primarily in the main language of that city. Now, like I said

in

...

my original e-mail, a 100% bilingual minority group does not usually

stay

...

bilingual for more than a couple of generations, especially in an urban environment where they must interact on a daily basis with people who do

not

...

speak their language, and often, might only use their own language at

home.

...

Also, keep in mind that the idea of "generations" varies from country to country. In some countries, people typically don't give birth until

mid-late

...

30s; in others, it is in the teenage years, so things like language

death

...

happen a bit more rapidly as the new generations come more quickly.

You are talking about cities like New York and London are; or about communities which have strong social reasons to forget their native language. Besides that, you are not counting the fact that people who speak the same language are able now to reach each other more easily.

In the most of the cases migrants are forming neighborhoods and they keep their language because they are semi/self-ghettoized. Such languages have chance to survive if there is written culture. And if the language hasn't disappeared during the time when descendants stop to struggle for survival, some of the descendants are starting to learn the language again.

Your scenario is well known everywhere on the planet. However, many things are changing thanks to the Internet and much better possibilities for communication.

Here are two opposite examples from Serbia:

• My grandfather came from the Torlak area to Belgrade with my father

who was ~10 years old. Prestige language of Torlak speaking population from Serbia is Serbian and Torlak doesn't have written culture. My father knows Torlak. My mother is from other part of Serbia and I didn't use to go to the village near Pirot often. I understand Torlak well, but I don't speak it. However, my brother and my sister speak Torlak. But, it is true that their children would know just a couple of Torlak words.

• Roma languages from Serbia are in constant state of shrinking. Their

prestige language is Serbian, but not just that: their prestige ethnicity is Serbian. I know a number or the third generation of integrated Roma who don't know a word of any Roma language. But, some of those from the third generation are now leaning Romano-Serbian, as a couple of years ago it's finally got written form. It is possible that we'll have Wikipedia in Romano-Serbian in a couple of years.

Note that the state of their language was probably among the worst for a language of that amount of speakers (170k): Very negative attitude toward language *and* ethnicity, no written culture, high rate of marriages with the dominant population; the only social environment in which it is used regularly are slums and all of them want to leave slums and integrate into the majority population. However, it is highly likely that all of Roma languages will survive, thanks to the increased awareness for their situation and Internet and mobile phones.

They don't need anymore to be in slums to be able to communicate in their language. A lot of them won't use Romano-Serbian in written form, but better communication will allow them to raise children who would know their language and who would be willing to educate their children in the language of their grandparents. Just a couple of decades ago such scenario wouldn't be possible.

---

foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

2011/7/12 Milos Rancic millosh@gmail.com

On Tue, Jul 12, 2011 at 11:47, M. Williamson node.ue@gmail.com wrote:

...

Milos, it is a fantasy of many that is not supported by research, that

just

...

because people are rich or have technology, their language will magically not die.

I wouldn't say that it is a fantasy, but bold speculation :D

I am not saying that wealth is the most important factor, not even technology in general, but enormously increased level of communication during the last 10-15 years, which is not counted inside of the present researches.

Present research actually indicates the decline of linguistic diversity has accelerated in the last 10-15 years, possibly due to the exact factor you seem to think is doing the opposite. Wider connectivity is an additional strong motivator to adopt a LWC for most. Communications technology and rural connectivity are killing languages that were once thought to be safe due to geographic isolation. This is to say nothing of urban settings, which have been killing languages for thousands of years.

From what I see during the last dozen of years, many languages are in the process of revival: Celtic languages, some Uralic languages etc. Cases of Manx and Livonian are impressive, for example. When I heard for the first time about revival of Welsh, I was thinking that it is a noble, but fruitless attempt. However, it is not anymore an exception, but such revivals are occurring at more and more places.

One thing that unites all the cases you've noted except for Welsh are hobbyists. When the only thing keeping a language alive is the fact that the entire speaker population is dedicating all of their energy and most of their free time to keep it from dying, it's an exceptional case. So you've given me examples of a few languages in Europe that aren't dying, this is a classical logical fallacy. Basically, in spite of research and piles of anecdotal evidence, you seem convinced that because there are a handful of cases that appear to go against the trend, the trend itself must not exist. That's simply not the way statistics work. A trend is still perfectly valid if it's followed by hundreds of languages, just because a few languages seem exceptional cases (although I would argue none of them _actually_ are exceptional, besides perhaps Welsh), does not mean anything for the rest of the world's languages. That's like if I said "Two thousand years ago, there were around 15,000 languages on this planet. Over half of those are permanently extinct." and you retorted with the case of Hebrew, alleging that similar cases were likely to occur with Akkadian, Sumerian, and every other dead language.

Now when you say "Celtic languages", which ones do you mean? Irish and Scots Gaelic are receding daily, despite the claims of their defenders. Breton is in serious trouble, and Manx and Cornish, although undergoing hobbyist revivals, seem unlikely to ever reach more than a few hundred native speakers. Welsh is doing alright for now, thanks in large part to the fact that it is supported by an autonomous regional government, which very, very few endangered languages enjoy, and the fact that people decided to "save" it when it still had plenty of speakers. But again, Welsh is the exception rather than the rule, one drop of language survival in a sea of languages speeding towards extinction.

living native language through many generations. Sorbian languages are the example. No, it won't be used very actively, but it will survive as a language of specific culture and because of identity purposes. Manx, Livonian and many indigenous languages have chances to survive like that. Welsh has chances to survive as fully recovered language.

I wonder how long Sorbian will actually survive as a full language. I think it's unlikely that Sorbians will never marry non-speakers. According to Ethnologue, Sorbian languages are both spoken by "mostly older adults". So your shining example that you repeat of language survival is actually not that. I wonder how many Sorbian native speakers there are of my generation. Now I wonder how many of them will speak Sorbian to _their_ children. Now what about the grandchildren?

Livonian, I'm wondering where you're getting this information from. By most accounts, it's got between 10 and 0 native speakers and is now used by some hobbyists only.

...

This process is repeating itself around the world, not just with poor and illiterate people, but also with rich and well-read people who find more economic and social benefits in using a LWC. This is unfortunate, but so

far

...

nobody has been able to find a remedy, and just writing encyclopedias in minority languages doesn't seem like a viable solution to the problem of language attrition and death, although perhaps it helps to raise the prestige a bit.

Not just encyclopedias, but books, dictionaries, even news sources. All of that is inside of our job description. But not just that: gathering active community around Wikimedia projects is almost the ticket for language survival.

This strategy couldn't help languages where speakers are already all bilingual. When there is a large number of monolinguals, this strategy still won't help much, as our job description doesn't include hiring interpreters for government workers and businessmen, or for people from rural villages when they go to sell crops at a market town. You are putting too much emphasis on a faulty idea that books kill or save languages. No matter how many books I can read in, say, Manx, there's still the fact that I can only talk to a couple hundred people in that language and there's no community where I can conduct my whole life exclusively in Manx. Or how about Romani in Serbia: if I want to rent a flat in Belgrade, I'll not be able to do it in Romani. If I want to go to a supermarket, or go buy parts for my computer or my car, or go to a meeting for alcoholics, or stamp collectors, or take classes at university, I won't be able to use Romani. That is what kills languages, and until we as an organization can find a solution to that problem, we can't solve the problem of language death.

2011/7/14 Milos Rancic millosh@gmail.com

On Fri, Jul 15, 2011 at 01:56, M. Williamson node.ue@gmail.com wrote:

...

Present research actually indicates the decline of linguistic diversity

has

...

accelerated in the last 10-15 years, possibly due to the exact factor you

May you point to some statistics or relevant researches for the period 2000-2010?

http://scholarspace.manoa.hawaii.edu/bitstream/handle/10125/4474/harmonloh.p...

Of course this is just one example; it covers the period up to 2005, including between 200 and 2005, where a sharp and consistent decline has continued in world indigenous languages.

...

Now when you say "Celtic languages", which ones do you mean? Irish and

Scots

...

Gaelic are receding daily, despite the claims of their defenders. Breton

is

...

in serious trouble, and Manx and Cornish, although undergoing hobbyist revivals, seem unlikely to ever reach more than a few hundred native speakers. Welsh is doing alright for now, thanks in large part to the

fact

...

that it is supported by an autonomous regional government, which very, very few endangered languages enjoy, and the fact that people decided to "save" it when it still had plenty of speakers. But again, Welsh is the exception rather than the rule, one drop of language survival in a sea of languages speeding towards extinction.

A number of languages of former Soviet Union, including Russia, whose speakers mostly know better Russian, are in the process of revival, as well. Many Romance languages highly endangered by the official languages of their countries are getting more attention, too. Non-Han languages of China have a lot of chances to survive, thanks to the support of the government. Many native languages of US are in much better position now thanks to the economic consequences of tribal sovereignty. There are a number of other cases. Without such efforts, many languages (I would say hundreds) wouldn't survive 20th century.

And tell me what exactly does a "process of revival" entail? Which Romance languages and what programs and what are the results? Many minority languages in China are spoken in relatively remote regions with low access to resources, and a key difference between China and most of the rest of the world is that rural status is enforced and rapid urbanization is technically disallowed ( http://en.wikipedia.org/wiki/Hukou_system ). If all monolingual speakers of minority languages around the world were forced to stay in their own home regions and speakers of other languages did not move in or were not allowed to move into their territory, language shift would not be happening anywhere.

As far as native languages of the US, this I have given you my personal testimony about and I have actually done research myself. Out of the many many indigenous languages of the United States, many of those that remain have only a handful of native speakers. Approximately 60 are still spoken by a slightly greater number of people.

Based on my preliminary statistical research, only the following languages in the US have speaker populations that have an equal or younger age distribution to English: Navajo, Havasupai, Crow, Zuni, Hawai'ian, Vietnamese, and Spanish.

Of course, this is only one element of language survival, and this includes languages with drastically different circumstances and numbers of speakers. Navajo has somewhere between 100,000 and 150,000 speakers, Havasupai has 530, Crow has 4280, Zuni has 9650, Hawaiian is difficult to determine (perhaps between 10,000 and 30,000), Vietnamese is 1 million and Spanish is around 30 million.

But of course we must also consider how widespread these are in their communities, whether they are declining or growing, etc. Whereas only 30% of Navajo children are considered to speak Navajo as their native language in 1998, this number was as high as 90% in 1968, indicating a rapid and alarming language attrition and shift. The Navajo Nation is a mostly rural tribe and has low access to resources, with many members living in poverty.

Havasupai, on the other hand, is spoken by nearly 100% of the tribal population, thanks to extreme isolation, small cohesive community and several other factors not found among most other tribes.

So when we look at the future outcomes of these two languages, it's not as simple as "One has over 100,000 speakers and the other has under 1000", but rather "Which one is healthy and which one needs help", "Which one is decreasing and at what speed", etc.

The next step was to look at intergenerational maintenance, i.e., do children of speakers of language X also speak that language?

The only languages that are being maintained intergenerationally at a viable level are Havasupai, Crow, Zuni and Hawaiian. This means that at the current pace of decline, Navajo will eventually cease to be spoken. This is true also for Vietnamese and Spanish, but since neither is endangered in the world I will not comment more here on either.

All of these other languages are currently thriving, partly due to geographic isolation in all cases, and possibly to innovative practices used by people in Hawaii to produce Hawaiian-speaking children from parents who were not raised speaking Hawaiian. I cannot think of another indigenous language in this country whose long-term outlook is not mostly grim (there are "glimmers" of hope, but as of now they are still just that, glimmers).

...

I wonder how long Sorbian will actually survive as a full language. I

think

...

it's unlikely that Sorbians will never marry non-speakers. According to Ethnologue, Sorbian languages are both spoken by "mostly older adults".

So

...

your shining example that you repeat of language survival is actually not that. I wonder how many Sorbian native speakers there are of my

generation.

...

Now I wonder how many of them will speak Sorbian to _their_ children. Now what about the grandchildren?

By number of active Wikipedians, Serbian Wikipedia would have to have four to five times more editors [1] than Upper Sorbian [2], while keeping in mind that the top of the activity on hsb.wp passed five years ago, while Serbian Wikipedia is now at the top of the activity. Numbers on hsb.wp are stable and all four active editors belong to the group of non-native speakers [3], expected for the situation of Upper Sorbian language (German is their native language).

Right, all editors belong to the group of non-native speakers, thus proving my point. Me going to school and learning Serbian is not the same as you speaking Serbian since childhood, and it never will be.

And, yes, as I said, I think that Sorbian will survive not as actively used language, but as the language of cultural identity. That's good enough.

For communities perhaps it's good enough, but it's still language death when there are no native speakers and everybody who learns it, is learning it as a second language. Latin is a dead language, just one that's still heavily used non-natively.

...

Livonian, I'm wondering where you're getting this information from. By

most

...

accounts, it's got between 10 and 0 native speakers and is now used by

some

...

hobbyists only.

Requests for new languages [4]. Children learn the language.

According to whom? If you're referring to the comment that "they are trying to teach Livonian language to children in summer camps", this certainly doesn't count as a native language. The result of summer camp teaching is not "native-level acquisition". I think it is necessary to get more sources to claim that a near-dead language isn't near-dead than the word of a stranger.

...

...

Not just encyclopedias, but books, dictionaries, even news sources. All of that is inside of our job description. But not just that: gathering active community around Wikimedia projects is almost the ticket for language survival.

This strategy couldn't help languages where speakers are already all bilingual. When there is a large number of monolinguals, this strategy

still

...

won't help much, as our job description doesn't include hiring

interpreters

...

for government workers and businessmen, or for people from rural villages when they go to sell crops at a market town. You are putting too much emphasis on a faulty idea that books kill or save languages. No matter

how

...

many books I can read in, say, Manx, there's still the fact that I can

only

...

talk to a couple hundred people in that language and there's no community where I can conduct my whole life exclusively in Manx. Or how about

Romani

...

in Serbia: if I want to rent a flat in Belgrade, I'll not be able to do

it

...

in Romani. If I want to go to a supermarket, or go buy parts for my

computer

...

or my car, or go to a meeting for alcoholics, or stamp collectors, or

take

...

classes at university, I won't be able to use Romani. That is what kills languages, and until we as an organization can find a solution to that problem, we can't solve the problem of language death.

Semi-integrated Roma speak Romano-Serbian in buses, on streets; it varies from part to part of the city, but it is not strange to hear them; children, as well. Besides the fact that it is not taught in schools. Note, also, that Roma in Serbia (and elsewhere) have been bilingual for centuries because of their specific way of life. (AFAIK, non-integrated Roma don't live in Belgrade and they speak Balkan and Vlax Romani.)

I didn't say that they don't speak it in buses or on streets. People in my city speak Navajo on buses and on streets; in Flagstaff (2 hours drive to the north) you hear it in many places. But this does not mean I can go to any random restaurant and place my order in Navajo, it does not mean these speakers can go to the post office and speak in Navajo, or any such thing. I think you've missed my point, which is not that these languages aren't spoken in cities (they are!), but that their speakers are forced to speak the LWC literally constantly.

As the most of people want to have life easier than hunter-gatherer, nomadic or so, they have to know to read and write to be able to get it. In such circumstances information written in that language keep languages and we provide the platform for creating and storing written information. The important question is under which circumstances that transition occurs. If it occurs during the expansion of Qin Dynasty, there are not a lot of chances for languages to survive. If it occurs in modern China, when schools, cultural institutions are available, as well as willingness of majority population to help, then language has chance to survive up to some extent.

Again, this is a flawed perception of language shift, the idea that the only reason people switch languages is so they can read books and newspapers. Life for most of us is not limited to the written word, for most of the inhabitants of the planet, spoken language plays a far bigger role in daily life, and while we can work to translate written resources, we can't translate human beings: shopkeepers, restaurant owners, police officers, can't be translated, if they all speak one language, that places a strong pressure on populations to switch languages, and inevitably they do, particularly within urban settings.

Preserving a language doesn't mean that it has to flourish. Stable ~1000 of hobbyists are good enough for waiting for the better times, as the case with Manx is.

"Good enough for waiting for the better times" - tell me, when are the better times? There are no better times for such languages, unless the rest of the world ceases to exist and they are left to repopulate it themselves. Once a language has been totally defeated (as Manx was), it can keep fighting, and it may even make some gains, but the chance is almost 0 that it will return to what it once was.

Besides, let's look at a scenario where somehow magically you're right and all the languages spoken today are still spoken in 2050. Let's say today there are 7000 languages, and in 2050 there are still 7000 languages. But instead of the fantastical idea that languages spoken by rapidly urbanizing populations, or populations that are shifting languages, will maintain the same number of speakers, let's postulate that they will, like Manx, maintain a perpetual number of speakers of ~1000 bilinguals and no monolingual speakers. In this scenario, we have 700 languages with hundreds of thousands or millions of speakers, and 6300 with ~1000 or less speakers.

In these circumstances, what is our goal? To create a Wikipedia in every one of these languages? Here's my thought:

If people who speak ~1000 bilingual speaker languages want Wikipedias, we allow it, but we don't need to go out of our way to try to recruit people to create content, or spend all of our free time holding workshops and looking for new participants. They are useful to people as a tool of cultural preservation, but nobody "needs" them as a primary medium for acquiring knowledge since all speakers are bilingual. Thus, our priority is the 700 languages, which have large populations and many monolingual speakers. These are the languages we need to focus on, and those which are missing Wikipedias are primarily spoken in Africa, to my knowledge.

On Tue, Jul 12, 2011 at 4:04 AM, David Gerard dgerard@gmail.com wrote:

On 11 July 2011 13:57, emijrp emijrp@gmail.com wrote:

...

If Wikimedia projects and WMF leave to die 90% (or 80%, or 70%, or 60%) of current languages in the next 40 years (we will be alive to see it, probably), then both are failures.

First thing would be a Wikisource or similar then. Just gather up as much material as possible and get it online, in a manner that isn't process-heavy

process isn't a problem. lack of man power and critical mass is. most people are more interested in wikipedia.

(e.g. the recent description of the ridiculous faff to get a thesis into Wikisource).

born-digital thesis are not are conservation problem.

Do we have anything for languages without a written form? Shove audio/video onto Commons?

Wikisource for transcriptions of the audio. We have a few featured texts of this type.

http://en.wikisource.org/wiki/WS:FT

Hoi, So you understand what a macro language is. Why the kicking then ? Thanks, GerardM

On 12 July 2011 10:59, Milos Rancic millosh@gmail.com wrote:

On Tue, Jul 12, 2011 at 08:03, Gerard Meijssen gerard.meijssen@gmail.com wrote:

...

Macro languages are nicely defined. They are languages that used to be recognised at one time as a single language but are found to be a combination of multiple languages. Kicking the idea of macro-languages is daft; it is not only a result of the work of SIL it is more the

consequence

...

of the work of the maintainers of the iso-639-1 and the iso-639-3.

If you've read definition at SIL [1], you would know that it is just the first paragraph of the definition.

[1] http://www.sil.org/iso639-3/scope.asp#M

---

foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

2011/7/12 Milos Rancic millosh@gmail.com

On Tue, Jul 12, 2011 at 11:16, Gerard Meijssen gerard.meijssen@gmail.com wrote:

...

So you understand what a macro language is. Why the kicking then ?

Because the category is comparable with the categorization of animals in encyclopedia Celestial Emporium of Benevolent Knowledge [1].

It looks like a big misunderstanding that's about to blow. Let's discuss this over a beer in Haifa! :)

-- Amir Elisha Aharoni · אָמִיר אֱלִישָׁע אַהֲרוֹנִי http://aharoni.wordpress.com "We're living in pieces, I want to live in peace." - T. Moore