[lg policy] Re: A request

[Paul Lewis]

APOLOGIES FOR THE DELAY ON THIS RESPONSE.  I POSTED IT FROM THE WRONG 
E-MAIL ACCT AND IT NEVER MADE IT TO THE LIST.

Dear Christina and all on the Lg Policy List:

Thank you for the nice words about the Ethnologue!

The data we have on populations of language speakers are probably the 
most contested and oft-questioned of all that we report. At the country 
level, we generally rely on national census data. In the case of 
Pakistan, the date we have for that information is the year 2000. 
Christina suggested she might comment on census data - and I'm sure 
she'll point out that national census protocols for collecting language 
data are rife with problems and notoriously inconsistent both internally 
and across national boundaries.

For individual languages, especially larger and more widespread 
languages, the population of L1 speakers is often also derived from 
census reports or (perhaps worse) from a variety of different sources 
which are aggregated. The 16th edition of the Ethnologue (2009) reports 
60,600,000 L1 speakers of Punjabi [pnb] in Pakistan and about another 
1.5 million elsewhere. Clearly with such a large population it is 
difficult to imagine that the language as a whole is seriously or 
imminently endangered, but that doesn't mean that language shift might 
not be incipient.

Without having gone to our sources to look at the details (mainly for 
technical reasons just now), I'd suggest that the growth in Punjabi L1 
speaker population seems consistent with normal biological growth 
generally. The statistic that we don't have--which would be particularly 
helpful in this discussion--is the ethnic population number, i.e., how 
many people identify themselves as being associated with the Punjabi 
language and identity. That number compared over time with the number of 
L1 speakers would be a better indicator of language maintenance or shift 
than simply reporting the raw number of speakers. If an increasing 
number of self-identifying Punjabis are not speakers, then there is 
cause to be concerned.

Apart from that, what further complicates the majority/minority question 
being raised is the fact that from all that has been reported, Punjabi 
is a Low language in the language ecology of Pakistan. Even though it 
has been developed, it seems to share many of the features of what 
Ferguson described as "Low" and as a result might appear in more recent 
terms to be "minoritized" (even by its own speakers) in spite of their 
large numbers and its widespread use. That, in itself, is not sufficient 
evidence to declare that Punjabi is on the decline. As is well known, 
languages in such a High/Low relationship have been maintained for very 
long periods of time. So the fact that even speakers of the language 
think poorly of it, is not sufficient evidence of language shift either.

The most central factor in evaluating language maintenance and shift is 
intergenerational transmission. If adults are not transmitting the 
language to children (in the home, neighborhood, community), then the 
number of L1 speakers will inevitably decline.

All the best,

  M. Paul Lewis, PhD.

Editor, Ethnologue: Languages of the World (www.ethnologue.com 
<http://www.ethnologue.com>) Sociolinguistics Consultant, SIL 
International / SIL Asia Area (www.sil.org <http://www.sil.org>)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: ethguy1.vcf
Type: text/x-vcard
Size: 191 bytes
Desc: not available
URL: <http://listserv.linguistlist.org/pipermail/lgpolicy-list/attachments/20110413/f25dcdcf/attachment.vcf>
-------------- next part --------------
_______________________________________________
This message came to you by way of the lgpolicy-list mailing list
lgpolicy-list at groups.sas.upenn.edu
To manage your subscription unsubscribe, or arrange digest format: https://groups.sas.upenn.edu/mailman/listinfo/lgpolicy-list