Thanks Jonathon, You pose some questions that I'll belatedly respond
to (in text).

--- In qalam@yahoogroups.com, "Jonathon Blake" <jonathon.blake@...> wrote:
>
> Don forwarded:
>
> > I have a query to the members of the Multilingual literacy discussion
> > group. I have a source that puts the figure of number of languages in
> > the world today to around 6000. The number of languages with a written
> > system is often put to 100-120. I bumped into another source recently
> > that suggests that 10-12 procent of the worlds known languages have a
> > written system.
>
[JB]> Some crucial things:
> Define "language";

[DO]This is a key question. I think that any source stating that x% of
languages are written needs to be clear on this. One possible pitfall
is using a definition of "language" that cots vernacular forms or
dialects as separate, but then counts the standard form or dialect as
the only one with writing. Clarity.

> Define "writing system"

In this case the question has to do with whether the language has been
put to writing (and perhaps how well/widely that system is known or
actually used). There are cases where "writing" is defined so widely
that it includes divination symbols etc., but for this question it
concerns a system in the more conventional (?) sense of being able to
represent the full range of expression in the language.

> http://www.ontopia.net/i18n/scripts.jsp lists roughly 200 writing
systems.
> http://www.omniglot.com/ lists about 200 writing systems (excluding
> writing systems for conlangs.)
>
> The next step is:
> * Do they want the number of languages in which something can be
written?

Yes.

> * Do they want the number of languages which have their own, unique
> writing system?

No. (In the context of basic literacy it isn't important whether the
system is unique or not, IMO. However in the context of language
planning, and how it views scripts and literacy, it might be.)

> To explain the difference:
> The first question would treat Hebrew, Yiddish, and Ladino as 3;
> The second question would treat Hebrew, Yiddish, and Ladino as 1;
>
> Then decide whether Moon, Braille, BlissSymbolics, and ASL should be
> included, or excluded from your count. If they should be included,
> decide for how many languages they should be counted. [Moon would be
> one or two. BlissSymbolics would be zero or one. Braille can be/has
> been used for every language that has been reduced to a writing
> system.]

This gets into other issues. It is important to bring up the issue of
braille, but it sounds like another question: How many languages have
a braille system? It is writing too, but you wouldn't probably have
braille but no other writing, so braille would be another index.
(Effectively, how many languages with a system for writing also have
braille?)

Don