Thank you, Peter, Christian and Milicent. On your questions:

As to whether Lyngngam and Khasi should be treated as separate languages, this is from the change request:

Lyngngam was formerly listed as a dialect of Khasi but it has only 36% lexical similarity with standard Khasi (wordlist from Shillong).

Here is some additional information I received from the requester when specifically asked about the issue of treating Khasi as a macrolanguage:
from "The War-Jaintia in Bangladesh: a sociolinguistic survey"
'Generally speaking, ‘Khasi’ is a generic name given to the people of the Khasi and Jaintia Hill' (Giri, 6).  

The Khasis of whom these words are spoken are Mongoloid in physical features, matrilineal in society, and, since the mid-19th century, mostly Christian in religion... The Khasi people of Bangladesh see themselves as no different than the Khasi in India about whom those words were spoken.  They say that they belong to the one race or ethnic group of people who live in the Khasi Hills and the Jaintia Hills and make no further ethnic distinctions: they are Khasi, whether they say they speak W-J, Lamin, Nongtalang, Pnar, Lyngngam, Synteng, Darang, etc."....(p 10)
"Second, there’s been some question as to whether Lyngngam can linguistically be considered to be of the same language as Khasi, in spite of their speakers’ close ethnic identification as Khasi (Grimes, 394).  " (p 11)
from "The Status of Lyngngam" by K.S. Nagaraja
"The Lyngngam speech form is spoken in the northwestern parts of the Khasi hills in Meghalaya state of India. Since Grierson's work (1904), Lyngngam has been considered as one of the dialects of Khasi. Only recently, that is in the late eighties, one scholar named Hamlet Bareh, a Khasi speaker, doubted the appropriateness of this classification. In his work, Barah has provided a few lexical items to make his point. After that, so far no work has discussed this issue. In the direction of filling this gap some data was collected by the present author in 1988 on this speech variety. Though this data is not sufficient for a detailed comparative study, it is still hoped that the various aspects of this speech variety presented below will show many interesting features about lyngngam and Khasi."
then there is this from an article at
'The seven Khasi tribes live in the Khasi and Jaintia hills, and the Garo live in the Garo hills of Meghalaya; the two groups' geographic distributions do not overlap. Distribution of particularly the Lyngngam in the West Khasi hills is in close proximity and contiguous to the Garo hills inhabited by the Garo. As a result, although the Khasi refer to the Lyngngam as Lyngngam or Langam and consider them as belonging to one of their own groups, the Garo call them Megam and treat them as part of one of the 12 subtribes of the Garo. Their ethnic identity therefore is in dispute, although their language is more akin to the Khasi's"

Lyngam—A People Associated with Khasi Identity

By Munni Mree
[this refers to Lyngam of Bangladesh - bw]

The Lyngam [also spelt Lyngngam], an ethnic community concentrated in Kalmakanda Upazila of Netrokona District, is associated with the Khas identity. The Garos call them Megam...

They are “supposed to have Garo origins, but have embraced Khasi customs. The Lyngngams speak a dialect of their own. Some traditions point out to Garo-Khasi intermixture in the Lyngngams territory” (Bareh 1985: 10). Their physical features and linguistic peculiarities suggest that the Lyngams are an Austro-Asiatic people belonging to the ethno-linguistic family of the Mon-Khmer communities...[that would be Khasi-related - bw]

Although language clearly differentiates the Lyngams from the Mandis [Garo], both peoples share similar food habits, behavior, family systems and kinship affairs...

Although the Lyngam are associated with the Khasi community, there is confusion among the people of Kalmakanda about their identity due to their alienation from the larger part of the Khasi community. Being unaware of their anthropological history, some members of the Lyngam community even believe themselves to be part of the Garo community. However, they do not think that they belong to the two large Abeng or Atong groups of the Garo community; they mistakenly think they are a separate group among the Garos.  

  “Except for the linguistic difference, the food habit, life style, costume and ornaments of the Lyngam community are almost identical to those of the Garo community. That is why we consider ourselves part of the Garo community. But we are not Garo, we are Lyngam,” says Ms. Preetilota Nongura, assistant headmistress of Baruakona School.  

And this from "Three Matrilineal Groups of Assam: A study in Similarities and Differences by U.R. Ehrenfels in American Anthropologist, New Series, Vol. 57, No. 2, Part 1 (Apr., 1955), pp. 306-321
"Some aspects of the ethnography of the Lyngngam-Khasi are here presented with a view to discussing the irregular distribution of similarities and differences between them and the other Khasis to the east, on the one hand, and the Garos to the west, on the other...It should be noted from the outset that Lyngngams and other Khasis form, with the Garos, a solid matrilineal area in central Assam, India. But while Lyngngam and other Khasis belong to one linguistic group, Garos are part of another. On the other hand, Lyngngams appear to be more similar to the Garos than to their Khasi linguistic conationals in physical features, in economy, technology, and in ritual, though not in the philosophicodogmatic frame of their religion. The uneven distribution of these similarities and differences poses problems of integration and historicity."

{end of materials from requester}

With regard to general treatment of splits:
>(You haven't generally handled splits by changing the existing entity's scope from I to M.)
In general, the JAC has not moved to consider any split to a Part 2 code element since Part 3 has been adopted.  I would not have bothered the JAC with this request if [kha] were not a Part 2 code element.

Suggesting that there is a single prototypical macrolanguage situation is rather at odds with a significant number of applications of it currently in the standard (e.g Dogri, Gondi, Komi)


Milicent K Wewerka <[log in to unmask]>
Sent by: ISO 639 Joint Advisory Committee <[log in to unmask]>

2008-02-15 06:54 AM
Please respond to
ISO 639 Joint Advisory Committee <[log in to unmask]>

[log in to unmask]
Re: Proposal promotes Khasi to a macrolanguage PLEASE RESPOND ASAP

Just be aware that if kha as a code element is retired that would affect Part 2.

Milicent Wewerka, Library of Congress

>>> Peter Constable <[log in to unmask]> 2/15/2008 1:15 AM >>>
If I understand what you're saying, "Khasi" is sometimes used as an ethnonym that covers speakers of several varieties. But it's not clear from what you say why a macrolanguage is warranted as opposed to retiring kha to be superseded by two (or four) more granular entities. (You haven't generally handled splits by changing the existing entity's scope from I to M.)

Are there application scenarios in which it will make sense to treat the collection as a single language - e.g. would there be cases in which a single language resource would be applicable across all of the varieties? If so, is it really the case that there are distinct languages rather than dialect distinctions?

(If zho is a prototypical case for the notion of macrolanguage, then I'd suspect a case like this isn't too similar: distinct languages that have to date been under-differentiated because there's widespread literacy in a written form that is basically common to all (in the Chinese case, because one of the individual languages is so dominant).)

Can you provide more clarification regarding the basis for this change?


From: ISO 639 Joint Advisory Committee [mailto:[log in to unmask]] On Behalf Of Joan Spanne
Sent: Thursday, February 14, 2008 1:43 PM
To: [log in to unmask]
Subject: Proposal promotes Khasi to a macrolanguage PLEASE RESPOND ASAP

Dear JAC,

The remaining 10 change requests I mentioned in my message reporting to the JAC last month (message date 2008-01-16) are now mostly settled. For a couple of them, I would like discussion from the JAC. One of the, request 2007-064 ( proposes changing the scope of the existing code element for Khasi [kha] to that of macrolanguage, with two member languages, Khasi (individual language) [xix] and Lyngngam [lyg], both of which are new. I had some discussion about this case with the proposer, around the question of whether a macrolanguage really is warranted, and we concluded that, yes, it is. It is also possible that the Khasi macrolanguage should include one or both of the two existing code elements, Pnar [pbv] and War-Jaintia [aml], which are called Khasian languages. The speakers of these two languages are also frequently grouped together with the speakers of Khasi (individual language), and generically called "Khasi," even using this label among themselves. I will work with the proposer to undertake this expansion of the macrolanguage membership if it is settled that the macrolanguage is appropriate.

Any objection to this change of scope for [kha] to macrolanguage? It does not actually affect Part 2 in any visible way.

I ask for a quick response because the IETF is awaiting my completion of as many of the remaining requests as possible, so that those can also be incorporated into the last revision of RFC 4645, the Initial Language Subtag Registry.