Bit of a 50/50 paper for me. Broadly I support your recommendations in respect of greater choice for parents, school autonomy and better targeted funding but I feel you let personal beliefs override researched evidence when addressing selection.

The primary issue is the filter you are using to equate selection by ability with social selection.

For example the comment "The top 200 schools attract pupils who are on average more affluent than the school population as a whole." is a fair empirical observation but is it unfair selection?

The issue with unfair selection is if kids from disadvantaged backgrounds cannot get into selective schools despite clearly having the ability to do so. It's not obvious from this report what evidence there is to support the notion that this happens or happens so substantially that a policy like banning selection and marshalling children into schools by the whim of local state lotteries would be justified.

At a very simple intuitive level we do not see it as desparately surprising that athletic adults have proportionally more athletic children than the less physically gifted (proportionally not exclusively). They also proportionally give them better access to athletic opportunities outside school and instill in them an ethos that personal fitness matters. Is it different for intelligent parents? And given the correlation between intelligence and income-earning potential is it surprising that the children of higher income earners disproportionately get into selective schools?

There clearly are cases where having financial advantage can help the less deserving/less able at the margins, private tuition for example. The paper is right to suggest ideas to level this playing field. But that is at the margin, ungifted well-off children don't get into selective schools, they go private. If Centre Forum wishes to ban selection to eliminate the margin they should perhaps show some credible research to show how big or small this margin really is and what effect this then has on the bright kids who don't make it.

There is also something odd about the suggestion that there is something wrong with grammar schools because they don't reflect the demography of advantage wihin a 6 mile radius of the school. Unless you have research to suggest that there are good reasons to believe aptitude and ability are distributed amongst children regardless of who their parents are this is a highly dubious measure of fairness.

The proposals in relation to banding are constructive, you clearly recognise smart children thrive better when with other smart children. However this tends to only work in large comprehensive schools that are resourced to attract enough teachers capable of getting the best our of gifted children. The advantage of grammars is that the teachers good at this are pooled together and can themeselves thrive by sharing expertise. Grammars also have a more certain supply of children to match the teachers expertise. Lottery schools, even with advantage banding on intake and a dilution of geographic preference, would be less certain.

There is also a starker political point here. Suppose we do identify that putting the smart kids in with the pack marginally improves the performance of the pack, even if it holds back the smart kids, neither an unreasonable assumption. Whose rights matter most? To suggest the rights of the child to achieve their own potential matter less than the pack is utilitarian and does not sit easily with our individual rights agenda.

As a parent one day myself I might risk the lottery to get into a good state school, but if it didn't work out I would not simply happily send my child to wherever the next ball rolled. I would have little choice but to take my child out of the state system for their own protection and benefit, assuming I could afford to.

And this is the last point. The paper is broadly talking about tinkering with state comprehensives, while making no mention of the crazy barrier between the state and private sector whereby the former are treated as a public service in need of subsidy from taxation, the latter as a luxury that incur it. This is clearly not reasonable, education is a public service whoever provides it, Reform are perfectly correct to suggest a way of ending this educational apartheid of advantage would be to put public money into both to the limit of the average state cost. That way parents with far lower incomes could buy private if they wished (or benefit from compulsary bursary schemes) or set up their own schools with true autonomy.

One final point is that selection may well have become more socially selective, with independent schools at least, since Labour's decision to abolish the assisted places scheme. Why not bring it back?

There is a common assumption that underlies much lay discussion of education. Indeed, it so common that it is often unrecognised. But for that very reason it needs challenging.

I call it the Eloi/Morlock assumption. Readers of H.G. Wells will already be ahead of me, but for the sake of the rest I’ll explain that Mr Wells, in his famous novel “The Time Machine”, posited that human evolution would eventually produce two distinct species: the enervated, pampered Eloi and the thuggish, troglodyte Morlocks. All this is supposed to transpire in the year 802701, but many seem to think it accurately describes our children today.

Of course, they don’t use my Wellsian terms. They prefer “smart” or “gifted” or “intelligent” instead of Eloi. Curiously, they often seem somewhat shy of explicitly naming the Morlocks, although “non-academic” or even “good with their hands” still occasionally get an airing. So congratulations to Mr/Ms LiberalFirst for refusing to beat around the bush, and simply labelling every child who isn’t smart, gifted or intelligent as a member of “the pack”.

You'll have gathered that I find the Eloi/Morlock assumption to be not merely false, but offensive.

So let us try another path.

First, recognise that there are not two types of children but as many types as there are children, and design an education system accordingly. The idea is challenging but it is certainly not new.

Second, let us recapture the original meaning of the word “docile”, namely “ready and willing to be taught; teachable.” Some children are easier to teach, and more ready to be taught, than others, but the reasons do not always lie in their own genes or even those of their high-earning parents. Children from disadvantaged backgrounds suffer significant handicaps. Instead of balkanising the education system on the basis of the supposedly innate abilities of each child, focus instead on responding to their (possibly contingent) docility.

Third, remember that people change over time. Subjects that were closed to us can be opened again. Education should be life-long.