Adding variant subtags 'aluku' and 'nduyka' and 'pamaka'fordialects

Pascal Vaillant pascal.vaillant at
Sat Aug 22 23:49:15 CEST 2009

> > By the way, this would lead  
> > us to state, symmetrically, that "Ndyuka is a dialect of Ndyuka".
> This would seem quite reasonable to me.

Yes, we totally agree there. As linguists, we're used to such apparent
contradictions or tautologies, which hide different levels, or facets,
of reality.  That's what I termed in my previous message "not making a
fuss about it". That's what you also express two paragraphs later as
"From my research, I see many occasions where there is a language continuum
where linguists appear to have grouped entities under the most
important/prevalent entity name for ease of reference". We can cope with
ambiguity. But in some circumstances and types of use (like when deciding
how to tag a segment of a digital language corpus), we also need to clarify
things. After all, we are here discussing about terminology and standards,
aren't we? And from the reactions of other people on this list, it is also
clear that ambiguity is not welcome in all contexts.

> May I suggest that the three variants are registered here as per the
> original request and that the requestor apply to the ISO 639-3 RA to add
> "Nenge" or "Busi Nenge" as an alternative name for djk if so desired.

Good idea. That is next on my to-do list.

> I have only looked into this very quickly but I don't see that Ethnologue
> have got anything wrong here.  It would appear that the language has only
> been referred to as Nenge since 2003 and yet the records/references within
> Ethnologue would seem to go back to 1973.

Ethnologue has nothing wrong on this issue, they only have some kind of
inaccuracy of description. Their present state says that there is one language
with regional variants, and I think we all agree on that point. The only thing
we object - as I said in my message of January 14th - is the fact that the
whole language is referred to by the same term as one of the three variants,
which *in a context of establishing a terminological standard with the aim of
avoiding ambiguity*, is a problem.

[BTW. Ethnologue is the result of the work of many different people in many
different historical layers, so it is quite logical that such inaccuracies
occur, and there is no intended blame when I mention that. For the point of
view of the first SIL linguist who studied Ndyuka -I think it was George Huttar-
it was quite logical to name what he was describing by the term "Ndyuka".
For the point of view of those who added later reference to Aluku and Pamaka,
it was quite logical to file them under the same language tag as Ndyuka...
I simply think it is a good thing to brush up the room a little from time to

Everybody will likely be ready to assume that linguists (like those I mentioned
in the references of the registration form) do not invent names (like "Nenge")
just for the pleasure of coining a new fancy word to replace an old one with the
same denotation. They are proposing terms in order to have clear reference to the
empirical objects they are talking about. I am not going to get into a lengthy
justification of the terms "Aluku", "Boni", "Ndyuka", "Aukan", or "Bushi Nenge",
unless I am enthusiastically asked to do so, but I testify that they all have
their reason to exist. Simply, to be short: the speakers themselves refer to their
languages as "Aluku" and "Ndyuka". "Boni" and "Aukan" [resp.] are alternate names,
a little bit outfashioned these last decades because they actually are more ethnies
names than language names. "Bushi nenge tongo", literally "the language [tongue]
of the bush negroes" (short: "Nenge"), is the way those people would refer to
what they would agree they are all having in common.

For the sake of clarity, then, and if nobody objects, it is good that IANA should
adopt the three variant subtags in the last version suggested by Doug Ewell, and
that we simultaneously file a change request in ISO-639-3 to make "busi nenge"
appear as the main description of tag "djk" (thanks to Joan for the practical
suggestions on this point, I am getting back to it).

> From my research, I see many 
> occasions where there is a language continuum where linguists appear to have
> grouped entities under the most important/prevalent entity name for ease of
> reference.   

Again you are perfectly right. But again, the point is that we sometimes
are forced to clarify concepts. These are all very complicated and fascinating
questions, especially when it comes to concepts and relations whose definitions
are not so clear cut in reality than they would ideally have to be in a neat
ontology. For example, we often come across questions such as "is Picard a dialect
of French, or is it an independant language?", which raise heated discussions
among lots of people having an interest in the issue, when in a linguists point
of view the question simply has to be asked with more precision. In fact words
like "dialect of ...", "independant language", or even "French", do not have such
an obviously clear meaning than most people would think at first glance... Picard,
as a linguistic system, has no reason to be considered as subordinate to any
other linguistic system, French or anything else; BUT Picard certainly is part of
a linguistic continuum, along with Norman, Lorrain, Wallon, Francien, of northern
("oil") French; AND moreover, the question is obscured by the fact that what we
call standard French is an Ausbausprache which has taken something of Picard, as
well as from the original rural dialect of the outskirts of Paris, but actually
is neither of those... Sorry, this might seem a diversion, but it is just to
shed a light on the difficulty to answer such questions as: "Is Aluku a
different language than Ndyuka, or is it not? if it is, simply put a new
entry in ISO-639; if it is not, then leave 'Ndyuka' as the description".

Thanks again to all those who took part in the discussion. I think I have now
understood that the general consensus is that we adopt the three variants in the
version of Doug Ewell:

and that I file a change request to ISO-639-3 to have the language itself
called "busi nenge". Is that so?

Best regards,

Pascal Vaillant

More information about the Ietf-languages mailing list