There's been some talk in the news lately about Singapore's Gifted Education Programme (GEP) and whether it breeds elitism. While acknowledging that people should, generally, move on from their narrow educational backgrounds and create lives which are not defined by which class in which school they went to decades ago, I would have thought this was so obvious as not to require controversy. Of course the GEP - and all stratified education - breeds elitism.

The justification frequently raised in favour of stratified education is that it helps children learn better. The argument goes that the teacher can pitch the tenor of the lessons according to the abilities of the children involved, so that brighter kids can be taught at a faster rate without leaving others unable to follow, while slower kids can be instructed at a pace appropriate to them without boring the more talented. The result, it is claimed, is better education for all. And because children are streamed into the GEP or other better streams according to tests that determine their merit, this ensures equality of opportunity, regardless of socioeconomic background.

But is this what really happens? I'm going to address this question on three bases: first, the one of equality of opportunity; second, the one of pedagogical effectiveness; third, the question of what primary and secondary education should seek to do.

First, equality of opportunity. Any streaming system and the GEP presuppose that the tests that separate children into various strata do so on the basis of innate merit, and that such merit can be effectively identified at the ages when the tests are administered. But the criticisms of this are obvious. First, there are always children with the potential to be late bloomers, for whom being labelled as underachieving or a failure at an early age may prove a major obstacle to what would otherwise be a successful academic career. Secondly, these tests are undeniably biased towards those who with cultural advantages such as English-speaking parents or parents savvy enough to know the consequences of passing these tests, as well as towards those whose families have the money to supply them with books, tutors, and pleasant surroundings in which to study (and who don't have to have their kids hold down part-time jobs to help the household make ends meet). The result is that children end up having a different quality of education not according to merit but according to class.

But does this matter? Don't kids who have the disadvantages I've talked about need to be taught at a slower pace anyway, so that not being in the GEP or in a more advanced stream would ultimately be better for them? This brings me to the second question, that of pedagogical effectiveness. The argument that "everyone gets a better education" from stratification might be arguable if there were indeed equal resources being poured into both "good" schools and "bad" schools. But there aren't. Motivated teachers prefer to teach classes full of bright kids rather than classes full of kids who are seen as disruptive wastrels, so clever children get better teachers and other children get worse ones. Moreover, the GEP has historically had a disproportionate amount of money allocated to it by the Ministry of Education, manifested most obviously in smaller classroom sizes.

Moreover, even if equal amounts were being spent on each kid per capita, regardless of their stream, this doesn't neutralise the problem with stratified education in terms of classroom dynamic, which is a large part of the reason for streaming to begin with. It's probably true that clever kids do better academically in an environment full of other clever kids: most of the kids who pass these tests are likely to be fairly well-behaved and task-oriented, because, as I've already mentioned, the selection procedure favours kids with at least moderately well-off, pushy and savvy parents. But this is at the expense of kids who've achieved less, because, by definition, if the proportion of disruptive kids in GEP and other high-stream classes has decreased, the proportion of disruptive kids in the other classes has increased. So that children left in these classes - the vast majority of children in the educational system - face an environment that makes it harder for them to learn. In many cases, these kids don't need to be taught at a slower pace so much as they need to be in an environment where a focus on teaching is possible at all.

And I haven't even begun to talk about the psychological effects of being told from an early age that you're not as smart as Johnny Tan or Michelle Lim from the school next door, and you never will be, and that's why you need to get dumped in the thicko bin.

But that ties in quite nicely with my third objection to stratified education. Is the point of education simply to churn out people who can do sums and write summaries? Or is part of the value of education the social factor - the opportunity it provides for children to learn to interact with a peer group and socialise with people? It seems to me that if, as I have suggested, kids with inherent advantages from their background will do well in any case, it is far less important that they learn one more scientific theorem or Shakespearean text than they otherwise could have, and more important that they learn what it's like to be someone who doesn't come from the sort of background that they come from. To learn that they are not those "other people" who go to that "other place" which is for people who aren't as smart as you. It's equally important that children who don't achieve as much academically learn that they can interact with those who do as equals and peers, and that not everyone who goes to a good school is a rich snob.

People say you don't need school to do this - that children can mix with other children in their own time - but you can't discount the influence of the school environment. Children spend the vast majority of their time there. What they see in their classes becomes, in their minds, a sort of normalised, representative vision of the world they live in. It may well be the case - it should hopefully be the case - that they may change their minds about this the more they see of the world, and come to realise that they came from a very specific background. But too often this takes the form of a sort of subcutaneous elitism, a deeply rooted assumption that as a gifted or highly educated child one is innately better or more rational or intelligent in some overarching, holsitic way - when all they've done is answer test questions better. Yes, people can shrug off the influence of the school environment towards elitism, but why install that influence to begin with? We shouldn't underestimate the impact of the place that they spent every working day for years upon years during a formative stage of their life, especially as many children who do well in school then go on to tertiary education where they continue to be surrounded by others who share their social and economic status to a large degree.

A final word: it seems to me much of the claims relating to better teaching in streamed systems can be answered by simply having a much less weighty syllabus to begin with. Teach kids less in the regular curriculum, so that most kids can keep up with it; and then offer completely optional courses and extra-curriculuar activities so that kids who are interested in finding out more have the opportunity to do so. I think bright kids will do even better out of a system that lets them shape their own intellectual - and other - exploration in their free time.