Among other things, the recent deliberations on education reform in Pakistan at Harvard University have again underscored the need for a dialogue on language and learning in Pakistan. Dr Faisal Bari recently observed that the very different points of view on this issue meant that it did not figure in the ‘consensus document’ issued at the end of the deliberations. This is understandable. But, while consensus is obviously difficult, it would be fair to say that a serious nation-wide dialogue on language and learning has become an urgent requirement. As a point of departure, let us take note of the fact that there is widespread demand for proficiency in English. And, of course, there is no particularly good reason to doubt the usefulness of learning the language. Even more than most countries, in Pakistan it is the language of social mobility and of power. More generally, it is the global lingua franca, the language of choice on the internet as well as the repository of advanced knowledge in many fields. So when it comes to schooling, is it any wonder that it is in such great demand, not least by parents whose priority is a better future for their children? The fast-growing private sector in education in Pakistan has been quick to respond to this huge demand for English by projecting and magnifying its ‘English-medium’ credentials practically over all other aspects of education delivery, even at the earliest stages of school. Not to be left behind in this race, the public sector has increasingly emphasised English as well virtually from the start of the primary cycle. The Punjab government, for instance, has now instituted a policy whereby the subjects of Math and Science will be taught in English starting with Grade-2 in public sector schools. (In Grade-1, the medium can be either English or Urdu for Math and General Knowledge.) Two assumptions appear to inform this policy: one, it will bring the public sector schools at par with the often better-regarded private sector schools and two, the earlier you make English the medium of instruction the easier it will be for the child to become proficient in the language. Three points need to be made with regard to the issue of school performance across the public-private sector divide: first, public sector schools cater for children coming from the lowest income bracket households and while some of the households sending their children to low fee private schools may also be categorised as poor, there is little doubt that the children of the poorest congregate in public sector schools. There is, accordingly, a significant difference in the support systems available to children going to public sector schools as compared to those who go to private schools. Second, the private school is free to turn away children likely not to do well in the examinations; the public sector school is usually not in a position to do so. Neither, of course, should it. Third, how good a job mainstream low fee private sector schools are actually doing in term of teaching children English or other subjects remains a subject for serious investigation. As often as not, the result of intensive rote learning is marketed by many private schools as proficiency. Right now all we can really say is that they are doing somewhat better than public sector schools; which given the state of most public sector schools may not be saying very much. As to the ‘earlier the better’ assumption, the following points need to be kept in mind: first, even if a very early and concentrated start entailing the use of English as a medium of instruction were the preferred strategy for reaching the goal of proficiency in English in the quickest time possible, we simply do not have qualified and trained teachers in anything remotely approaching the numbers needed to make this practicable. At this point, it would be a formidable enterprise for us to teach English as a subject at the primary level, let alone employ it as a medium of instruction. Second, this may not be the best strategy, in any case. There are numerous studies indicating that a child’s cognitive ability is best developed at the earliest stages of schooling by the use of the home language or the mother tongue. According to experts in the field, a child of around five years of age, at the time of starting school, has a vocabulary of approximately 3,000 words in his home language. To build upon this resource in the initial stages of schooling means enhancing the child’s capacity to learn. Far from being a zero-sum game between say the home language, the national language and the global lingua franca, a start in school with the home language will subsequently enable the child — as a better learner — to pick up the latter two with greater ease. There are exceptions to this route, particularly where children are exposed to what constitutes the immersion method. Children from elite households, for instance, have far less of a problem learning English, not because they go to English medium schools (that is a part of the equation) but because outside the school they are immersed in an environment where English is effectively rendered the first language. But, we can well imagine how limited this number is. Our attention is often drawn by those who want no delay in English language schooling to countries such as China and Japan whose spectacular success is attributed to different development models but who are one in their current enthusiastic embrace of English. For reasons mentioned earlier, this is entirely understandable. But the point to be stressed here is that in building the strong foundations of their human and social capital, their primary reliance was not on English. What we are positing, then, is not a choice between English and other languages. The discussion has to be about the role of a given language, the methodology used to teach it and the stage at which it needs to be taught. In Pakistan, it may be more appropriate to teach in the home language in the initial stages of the primary cycle, switch to Urdu in the next phase and subsequently to English. When switching to Urdu the home language could be retained as a subject and Urdu can continue as a subject when English becomes the medium of instruction. In conclusion, this discussion is not about preserving language or enriching culture. Those aims, for other reasons, are well worth pursuing in their own right. There may even be points of overlap between the cultural and the pedagogical contexts. But, for the sake of analytical clarity, we may want to keep the distinction in mind. The writer lives in Lahore and can be reached at abbasrh@gmail.com