[Foundation-l] Languages and numbers

[Milos Rancic]

On 06/25/2011 03:11 PM, Bishakha Datta wrote:
> I posted this on the India list (many people are not subscribed to
> foundation-l) - forwarding this question which just popped up.

First of all, although numbers look fascinatingly precise, they are far
from that. When you make a sum of approximations like
~1M+800k+30k+4k+700+20+ the language spoken by three individuals, you
will get fascinating number 1,834,723. So, the numbers are far from
being census-level precision.

All of the numbers are based on Ethnologue data [1], which varies from
very good to very bad approximations. Ethnologue varies even in
linguistic classification a lot. (Being educated in Serbian linguistics,
I know how bad the description of the South Slavic area is.) BUT, it is
the best source for all languages of the world ever been made, and it
gives good general picture.

> [296,097,274 349 India]
>
> Does the population number mean that the existing indic language
> wikipedias covers the rest of the population ie over 90 crore? Is
> this information updated from the current census?

By making a quick approximation of number of speakers of some large
official languages of India [2] and not counting English, I've come to
the number of ~650M and stopped counting (BTW, that includes the number
of 180M of Hindi speakers from 1991; and according to the population
growth in India, there should be at least 250M of Hindi speakers today).
Thus, I think that ~300M more could be gathered by other languages with
Wikipedias and by adjusting existing numbers for population growth. (I
could make more precise calculation if needed, but I would need some
time.) It should be also noted that dates of the entries in Ethnologue
vary a lot and that some of them could be old 20 years or more.

And, again, this should be used as very general guideline, not as a
precise one. This list would be very good in telling that there are much
more speakers of Awadhi than Merwari today. However, it is not good to
be used for comparison of number of speakers between Awadhi and
Maithili. But, anyway, that's not important. We know that we should work
to cover both Awadhi and Maithili.

At the other side, I will, indeed, try to make those numbers more useful
(although I think that the most important usefulness is about pointing
to the large populations without Wikipedias).

> It is fascinating, although I think I may not have understood the
> classifications. Is there only one Indian Sign Language, for
> instance? I was told by a user (in the UK) that several are in use in
> different parts of the country. Still, perhaps the variants do not
> have sufficient numbers of users to qualify for this listing.
> However, the context in which I was told was precisely the severe
> lack of support materials for helping users become self-sufficient
> and good communicators, so the list itself becomes a barrier.
>
> Unfortunately, I do not know at the moment how to fix the problem.

I've checked the whole database and just one Indian Sign Language has
been listed, which doesn't tell us a lot. Ethnologue entry about Indian
Sign Language [3] says that it is called "Indo-Pakistani Sign Language"
or "Urban Indian Sign Language". However, according to the fact that
"Deaf schools mainly do not use ISL...", it could mean that dialectical
divergence could be very high (thus, it could look as a number of
different languages), no matter the fact that it's been used in Pakistan
and Bangladesh, as well.

Said so, I have to admit that my knowledge about sign languages is very
limited.

[1] http://www.ethnologue.com/
[2] http://en.wikipedia.org/wiki/Languages_with_official_status_in_India
[3] http://www.ethnologue.com/show_language.asp?code=ins