I have observed many such schools in the past 10 years training Swedish teachers and I have seen none of the hair-raising images of the Purves analysis. Staff manage the daily demands through well-planned shift systems. Parents drop the children off and collect them at a time that suits them and only a small proportion use the maximum hours available.

There is a good balance between small activity groups managed by teachers and freely chosen activities, and between indoor and outdoor time. Children learn from their immediate environment. Breakfast and lunch are orderly occasions, used for basic social and academic training. There is no evidence that adult society in Sweden is impaired by exposure to the system. Many of the habits and attitudes bred into children at a time when the brain is young enough to learn endure into adult life, for example removing shoes when entering a house.

One can be critical. Admissions, staffing and resourcing policies need revision. Pre-school teachers should not be paid less than others. The academic potential of the young brain needs sharper focus in short and long-term planning.

However, the general principle is sound and we have much to learn from it.

Charles Clarke is right to recognise the absolute priority of the first years of life, a time of nil provision in the history of education other than the lottery of home. Modern research is revealing just how much the brain develops from simple but essential learning experience from birth and even before.

To start formal education at three is being revealed as almost too late! Problems with standards later in schooling have their roots in early years.

Solutions must start there, not with the sadly predictable tinkering which happens between 11 and 18.

Mervyn Benford

High Street

Shutford, Oxfordshire