Monday, December 31, 2007

Linguistic tolerance

The Wikimedia Foundation has a policy about what linguistic entities it accepts and what linguistic entities it does not accept. When a linguistic entity is recognised as such and has an ISO-639-3 code, it is considered a language. As a rule, the language committee gives a conditional approval to requests for languages that have a code.

There are problems with this policy. As I wrote elsewhere, a language may be dead. A dead language is a language that is no longer actively used; there has been no new terminology, nobody is using it actively, good examples are Hittite (hit) or Akkadian (akk). In my opinion they can have a Wikisource but a Wikipedia is problematic because you cannot write in these languages for a modern public without changing the language into something completely different. It does not even make sense to have a MediaWiki localisation for such languages.

There are more problems, what to do with languages where from within the culture it is prohibited to write the language down. What to do with languages where there are few people speaking a language. What to do with languages where few people are truly literate for their language. What to do with constructed languages?

The biggest issue with all these issues is one of competency. Who is competent to judge that a language is truly dead. At what level are there sufficient people in a community to support a language for a WMF project. How do we judge the quality of our projects and as importantly who is to judge? Also does the WMF have a responsibility for less resourced languages.

Brianna blogged about the Volapük wikipedia. For her and for many others, Volapük became an issue because they had the audacity to create enough articles to be noticed. People like Brianna feel offended because it upsets the notion of what Wikipedia is. Brianna introduces the notion of a "language ego" but I am sure she will agree that every non dead language deserves its place under the sun and only the people that communicate in a language are the ones that can realise a WMF project. The good news is there is plenty of sun and it is not expensive to have another language.

When people equate artificial languages with languages without merit, they have a problem. Many languages have started out in this way. One of the more interesting examples is Italian, it was standardised by Dante, used as a lingua franca until the unification of Italy when it became an official language. Another example is Sardinian where a constructed merged linguistic entity that has not been recognised by the ISO-639-3 registrar, is recognised in Italian law.

When you compare Volapük with Klingon, the biggest difference is that Volapük allows you to express all modern subjects.

For me the issue of the Volapük Wikipedia is a non-issue. I know three people that speak Volapük, not all of them are involved in this project. Given the competency of the people that are, there are no issues in getting the information in Wikipedia right. Wikipedia has always allowed for a project to evolve and insisted on the independence of communities. I am sure that in the end both Wikipedia and the Volapük Wikipedia will emerge stronger from all this.



Sadie Li said...

I agree-it doesn't matter if a language is considered 'dead' or not, but as long as people still speak it, it should be available.

Sadie Li said...

For wiki language options, of course. And in other places, they need to have a broader range of translations and probably more translators, as to prevent language confusion and broken translations from family members.