An Epic Narrative
Genetics, business, online learning—the New Linguistic Survey will have a huge impact
Sugata Srinivasaraju, Outlook, March 10, 2007
At the end of its 10-year cycle, the truth about the state of our languages will be out. It will be one of the most significant statements, in a century, on a prime identity-marker.

The results of the survey may unleash a new dynamics and could generate a new politics. Given the linguistic reorganistion of India after Independence, and the large-scale migrations since then, the new linguistic atlas of India is bound to throw up relevant questions about the boundaries we created 50 years ago. But Udaya Narayana Singh, director of the Central Institute of Indian Languages (CIIL) in Mysore and one of the moving forces behind the survey, says: "The survey may emphasise that each Indian state is now a multilingual state, but we are not surveying to sort out boundary disputes. There are far greater issues involved. India will be divided into convenient grids for the purposes of the survey." Yet, the project report says the results are expected to form the "basis of social engineering".

What are the objectives of the NLSI? It will primarily profile the Indian linguistic space by describing each language and speech variety, its structure, socio-cultural role and demographics. The survey will make possible a reasonable lexicon and grammatical sketch for each language. It will also record the interactions between various linguistic communities, which involves tracking bilingualism and multilingualism. There will also be a massive audio-visual documentation of speech varieties. Linguistic maps, charts, graphs and atlases of languages will be created.

The knowledge base generated by the NLSI will be used to develop language technologies, and may help future software and online learning. The survey has also been designed to help research in genetics, physical and cultural anthropology, sociology and psychology. To popularise the contents of the survey, two web portals will be created, one in the secured network domain and the other in the public domain. The public domain site will be an interactive linguistic observatory of sorts.

Apart from these, the massive data collected and digitised by the survey will be put to other uses. Prof Singh says there's a plan to develop a Linguistic Data Consortium for Indian Languages (LDCIL) on the lines of the Linguistic Data Consortium at the University of Pennsylvania—a hugely successful consortium of 100 companies, universities and government agencies that aids research in linguistic technologies.

"We have a strong business model for LDCIL and will be funded by the government for only six years; we expect to break even after that. There is a big market for linguistic data in India. A very simple example is making possible text-messaging in Indian languages," says Prof Singh.

If this is reason enough to feel upbeat about the survey, it also has an elegiac aspect, in the recording of dead and endangered languages. The 2001 census, it is reliably learnt, threw up a list of endangered languages unlikely to be made public for the simple reason that it would create political havoc. "Not many Indian languages have died. The Indian situation is not bad, but we cannot be satisfied with it. In the wake of the tsunami, people said some speech varieties in the Nicobar Islands were wiped out, but the CIIL in its field study found they had survived.

What has killed or clipped many small languages is the inflow of m Survey of India's People of India project mentions.

One could arguably call NLSI the first authentic linguistic survey of India because the one conducted by Sir George Abraham Grierson 100 years ago left out languages spoken in South India.
There are also question marks over the reliability of the data, said to have been collected by "untrained manpower". The NLSI, by comparison, will be put together by thousands of linguists and trained researchers across 100 Indian universities.

The government will release Rs 200 crore to the University Grants Commission during the 11th five-year plan for this huge project. During the same plan period, CIIL will get Rs 80 crore separately to coordinate activities and build computational and other infrastructure. It may also get another Rs 50 crore under the endangered languages project. If the 12th five-year plan outlay for the project is also taken into account, the entire NLSI project cost may hover around Rs 600 crore.

The survey will be managed by a consortium of institutions under the general direction of CIIL. "We will need about Rs 28 crore just to build a computational facility to process the data on Linux and MS SQL server. The network we will create will be four times bigger than an ICICI bank network. The money that will come directly to CIIL will also help us engage ngos working in the area. Official networks will not help us get data in places like the Northeast or Jammu and Kashmir, independent groups working there will be helpful," says Prof Singh.

What happens to English in the NLSI? How will it deal with a foreign tongue that has had such a pervasive influence in the last couple of decades? That's where tracking bilingualism becomes important. "In the West, bilingualism is the exception; in India it is the rule... Recognising convergence in India's history is not so much an ironing out of differences of identity as the emergence of a fresh all-India linguistic identity," says the report. So expect a chapter in the NLIS on the techies who have converged in Bangalore!
http://www.outlookindia.com/full.asp?fodname=20070319&fname=Survey+(F)&sid=1