A query...

[Claire Bowern]

>> For many people (and speech communities) there's a big difference between
>> making large amounts of raw data generally available and discussing and
>> sharing it with colleagues.
> 
> As for speech communities for which this holds, that's the ethical issue
> we've already identified.
> 
> As for linguists for whom this holds, I've always wanted to meet one in
> person so I could ask them 'why'.
> 
> 

One is not being willing to be identified as a speaker of the language 
(e.g. Laz in Turkey). Another is a strong feeling of association between 
language and place (i.e. that a particular language belongs to a 
particular country and is looked after by a group of people), so reading 
mythology in that language away from that area would be inappropriate. 
Another is a worry that others will learn the language and steal it. 
Another was a worry that publishing secret language would cause harm to 
come to come to people who read it (e.g. I was warned not to tell blokes 
about women's business because the ra:galk would rip their throat out, 
and they didn't want that to happen to anyone). Then there's cultural 
knowledge that could be used as evidence in land claims (there have been 
cases of people going through archives and claiming another group's 
cultural knowledge as their own, and so publishing information such as 
the GPS coordinates of sites has potentially harmful consequences). 
People don't always give a reason beyond general misgivings and lack of 
trust of what "White people" will do with the data. This is balanced by 
wanting to have someone work on the language to help record it for the 
community's own use, with a recognition that in order to do that the 
linguist usually has to (or wants to) also do work that relates to their 
role within a university. And there's usually multiple different reasons 
within the same community, and different degrees to which people want 
data made available.

Claire