1 Language Policy and Localization in Pakistan: Proposal for a Paradigmatic Shift Tariq Rahman. Abstract This paper examines the present language policy of Pakistan and its consequences for the indigenous languages of the country. It then relates this to efforts at localization--- creating computer software in the languages of the country---and argues that all such efforts have been power-oriented. This means that only those languages have been selected for localization which are used in the domains of power---government, bureaucracy, judiciary, military, commerce, media, education, research etc---thus further strengthening them vis a vis the marginalized languages of the people. It is therefore argued that the efforts at localization should be rights-based i.e. all language communities should be considered equal and their languages should be localized not because of their present use in the domains of power but because they too should be strengthened by being put to such use. 1. Introduction Pakistan is a country with at least six major languages and 58 minor ones (see Annexure-A). The national language, Urdu, has over 11 millions mother-tongue speakers while those who use it as a second language could well be more than 105 million (Grimes 2000). Those who may be considered barely literate in Urdu---if the rate of literacy is really 43.92 % as claimed in the census of 1998---are nearly 66 million. That is rather a large number compared to nearly 26 million (17.29 %) who, having passed the ten-year school system (matriculation), can presumably read and understand a little English (Census 2001). And yet computer programs, including e-mail and the internet, function in English in Pakistan and not even in Urdu let alone the other languages. This means that most Pakistanis are either excluded from the digital world or function in it as handicapped aliens. Indeed, most matriculates from Urdu-and Sindhi-medium schools have such rudimentary knowledge of English that they cannot carry out any meaningful interaction, especially that which would increase their knowledge or analytical skills, with the computer. Perhaps only the 4.38 % graduates (Census 2001) (about 6.5 millions) could do so if they could afford to buy computers. However, the mushroom growth of small shops, or ‘computer cafes’ as they are called, has made PCs available to most boys --- girls generally avoid such places because they are used by sex-starved youths to visit pornographic sites --- with a little cash to spare. However, these ‘cafes’ are in the urban, not the rural, areas and for any prolonged and meaningful use of the computer one must possess a PC or have access to one in one’s place of work or study. As only the rich can afford personal machines and very few people go to educational institutions which have computers, the number of Pakistanis actually benefiting from the computer can only be a small percentage of the whole population. The question then is whether it is cost-effective to create computer programs in Pakistani languages. This operation, or localization as it is called, is costly and time consuming. Should time and money be spent upon it or not? And if so, which should be the language or languages of localization? Urdu, the national language and the urban second language? or Punjabi, the language of 44.15 % Pakistanis? Or Sindhi, the language of 14.10 % people but, in addition to that, a language used in the education system, media, administration and judiciary in Sindh? Or