So I wish to re-state the case - a much stronger one than usually acknowledged - for uniformity in schools. Let’s hear it for the bog-standard comprehensive and, indeed, the bog-standard primary school.

Almost nobody dissents from the view that we should aspire to equality of opportunity, on grounds not only of social justice but also of economic efficiency, since it is foolish to allow talent to go to waste for want of opportunities to realise it. Yet we know that home circumstances affect children’s development from earliest babyhood and that, by four or five, some lag so far behind that they will never catch up.

If we wanted pure equality of opportunity, so that family income, quality of parenting, ethnic background and so on had no effect whatever on life chances, we would put children in full-time, state-run nurseries soon after birth. We would make these nurseries as uniform as possible since there would be no point introducing new variables leading to a whole new set of inequalities. We might want to see whether some child-rearing methods were more effective than others, but we would do so in a deliberate, controlled way. We would not have diversity for the sake of it, and we would certainly not allow parents to choose the type of nursery.

Anybody except an unreconstructed Stalinist agrees that such a society would be abhorrent. Some inequality of opportunity is the price we pay for a humane, diverse and liberal society.

We do not take children away from home at birth, nor when they are older.

We do not even require parents to send them to school. But most, from infanthood, are placed in state-run institutions for about 16 per cent of the year at most, assuming the child is never ill or otherwise absent.

Add time spent on homework, subtract time spent in bed and in the bathroom, and the proportion of school-controlled time is probably still below 25 per cent of the average childhood.

If we are serious about equality of opportunity, shouldn’t we make that small proportion of children’s lives as uniform as we can? Why would we want to introduce new inequalities and why, more pertinently, would we want them to correlate with the inequalities of the other 75 per cent of their lives?

Doesn’t that 75 per cent - with its differences in reading matter, meals, outings, interaction with adults, extent of private tutoring and so on - provide enough diversity, individuality and parental choice? And if we now live in a more diverse, individualistic society than ever, isn’t the argument for uniformity of schools stronger rather than weaker?

The case for diversity among schools is thus a weak one if we rest it on grounds of liberty and parents’ rights. Uniformity in schooling would not seriously threaten the diversity of our society. Diversity in schooling, by contrast, seriously threatens attempts to achieve equality of opportunity.

The argument is not conclusive. For example, we can make a case for choice on the grounds that it gives parents a stake in the success of their children’s education and children do best when schools and parents pull together. We can support diversity because we can then see how different approaches work - though I see little evidence of anybody doing this systematically. We can support the idea of schools with different specialist strengths because it can make for a more efficient use of resources. But if we take aspiration towards equality of opportunity as a given, the onus is on the diversifiers to make their case, not the other way round.

Peter Wilby is editor of the New Statesman