Questions on ISO 639-3

Doug Ewell doug at
Thu Nov 27 16:32:04 CET 2008

CE Whitehead <cewcathar at hotmail dot com> wrote:

> My only concern is that--as Addison points out--the division between 
> the dialects does not quite fall on political boundaries;

They're never expected to be precise.  The division between en-US and 
en-CA doesn't necessarily fall on political boundaries either.  Neither 
does the division between de-DE and de-AT.  There's nothing special 
about Romanian/Moldovan here.  See draft-4646bis-18, Section 4.2:

   ... the tags "en-US" and "en-CA" mean, roughly, English with
   features generally thought to be characteristic of the United States
   and Canada, respectively.  They do not imply that a significant
   dialectical boundary exists between any arbitrarily selected point in
   the United States and any arbitrarily selected point in Canada.

Note in particular the intentionally imprecise phrase "generally thought 
to be characteristic."  People read way too much into the assumed 
geographical precision of region subtags, and that is why that passage 
was added.

> and I still think it is very difficult to get around political issues 
> when assigning these subtags,

Of all people, we, LTRU, and ISO 639 are probably doing the most to 
separate "political issues" from language encoding issues.  Unbiased, 
professional linguists have stated often that "Romanian" and "Moldovan" 
are at most dialects, not separate, mutually unintelligible languages. 
ISO 639 has made a change to reflect that professional opinion, and 
ietf-languages has made a change to reflect what ISO 639 has done.

> although if the Moldovans want to have their language identified as 
> Romanian, then that is fine with me;

They do not.  But did you look at the revised subtag (which now appears 
in the Registry)?

Type: language
Subtag: ro
Description: Romanian
Description: Moldavian
Description: Moldovan
Added: 2005-10-16
Suppress-Script: Latn

The three Description fields are equal.  There is no "primary" or 
"preferred" Description field in the Registry as there is in ISO 639. 
The Registry lists "Romanian" first because it is the ISO 639-3 
reference name for this language.

Compare this with:

Type: language
Subtag: es
Description: Spanish
Description: Castilian
Added: 2005-10-16
Suppress-Script: Latn

The fact that "Spanish" is listed first and "Castilian" is listed second 
does not mean Spaniards are forced to refer to this language as 
"español."  They are free to continue calling it "castellano," or 
anything else they like for that matter.

> however if they decided they want a variant subtag, I would leaning 
> towards o.k.-ing it, but I will not go and apply for one.

First, "they" -- the authorities who have complained about the ISO 639 
change, who don't necessarily reflect the views of all 4.1 million 
Moldovans -- have not approached ietf-languages or LTRU asking for a 
variant subtag.  They may not know such a thing exists.

Second, the posited distinction is between language usage thought to be 
characteristic of two national entities, each of which has an ISO 3166-1 
code element, and hence a region subtag.  This is exactly what region 
subtags are for.

> and of course, why then was the variant,
> scotland
> Description: Scottish Standard English
> Added: 2007-08-31
> Prefix: en
> needed?

Scotland does not have an ISO 3166-1 or UN M.49 supranational code 
element, and hence it does not have a region subtag.  That is what 
differentiates the Scotland case from the Moldova case.

(Karen Broome's reply "Scotland is not a country" needs some 
qualification, as the standard terminology is that England, Scotland, 
Northern Ireland, and Wales are indeed four separate "countries" within 
the United Kingdom.  ISO 3166 code elements don't necessarily correspond 
to what people think of as "countries."  The correct answer is that 
Scotland doesn't have an ISO 3166 or UN M.49 country code.)

Doug Ewell  *  Thornton, Colorado, USA  *  RFC 4645  *  UTN #14  ˆ

More information about the Ietf-languages mailing list