So free schools aren't dangerous and they're not ideological in the sense that there's no evidence that the policy works. But are they an "experiment"?

I think they are but that's not a reason to oppose them. On the contrary, one of the strongest arguments in favour of free schools is that they provide a laboratory space where teachers can try out new things — the research and development wing of state education, if you like. By allowing free schools to innovate and experiment — and monitoring the results — we can eventually discover more effective ways of teaching and learning and, by extension, drive up standards across the board.

To illustrate this point, it might be helpful to describe the secondary school I helped set up in Hammersmith. 

The founding principle of the West London Free School (WLFS) is that all children can benefit from a classical liberal education, regardless of background or ability. What that means in practice is that all the pupils are expected to study a core of academic subjects — English language, English literature, maths, history, geography, divinity, Latin, French, physics, chemistry and biology — complemented by plenty of art, music, drama and competitive sport. Our aim is to provide our pupils with a storehouse of core knowledge in a range of traditional subjects — the best that has been thought and written — so they end up with the social capital to succeed, both in their schooling and beyond. We hope that pupils will leave the WLFS able to participate in the conversations mankind has been having with itself for thousands of years about the universe and man's place in it. We also hope they'll exhibit the virtues that are traditionally associated with a classical liberal education: well-informed, honest, courteous, industrious, self-disciplined, self-reliant, resilient, tenacious, public-spirited and open hearted.

This is clearly an experiment — and an ambitious one. We describe the WLFS as "a grammar school for all" (Harold Wilson's original definition of a comprehensive), but can all children really access a grammar school curriculum? The conventional wisdom, even in high-performing comprehensives, is that a classical liberal education is only suitable for children in the top half of the ability spectrum, with less-able children (usually those from more deprived backgrounds) being steered towards a combination of academic, technical and vocational subjects. It is that shibboleth that the founders of the WLFS have rejected. We believe that, with the right support, it's possible for all children to complete their secondary education with a storehouse of general knowledge — and we draw inspiration from a number of schools that manage this successfully, such as Mossbourne in Hackney and the Renaissance Arts Academy in Los Angeles.

Obviously, it's too early to say whether our experiment has been a success. We're only just over two years old. But the early signs are good. To begin with, our classical liberal formula is popular with local parents. The WLFS had ten applicants for every place this year, making it one of the most oversubscribed state secondaries in England. Not just white, middle-class applicants, either. Thirty per cent of our current Year Sevens are on free school meals and between 30 and 40 per cent of all our pupils are black, Asian or minority ethnic. Roughly 25 per cent of the pupils have special educational needs (SEN) and 50 per cent have English as an additional language. So a fairly typical mix for an inner-London comprehensive and, so far, there's no evidence that any of them are struggling with our ethos or curriculum, including those with statements of SEN. (We employ a full-time SEN co-ordinator.) The school was inspected by Ofsted earlier this year and rated "good with outstanding features". We hope to get the top grade next time.