The speaker cannot be named because the conference was held in private. But the quote shows how far the questioning of received wisdom has gone. It worries experts who believe there is good evidence to show children make huge strides in reception classes and that the nationwide movement to take four-year-olds into primary schools is justified.

One of those experts, Peter Tymms, reader in education at Durham University, found children made enormous progress during their first year in reception.

“I misjudged how much progress they were going to make,” he says. “A child arrives at the beginning of reception knowing a few letters, being able to count a few numbers. They become numerate and literate within the year. And that’s an astounding achievement.”

His “Performance Indicators in Primary Schools” project has found that many children who start school at the earliest possible opportunity - at age four to five - make rapid progress. Those who stay at home or attend a nursery or playgroup don’t make the same impressive progress. The age of the child in the reception class is not important.

Whether the children are young when they come into reception (that is, just four) or old (five), their progress depends on their achievement level at entry and on which school they go to, not on how old they are in the class relative to their peers. “A very young child who started at a high point took off as rapidly as the older child with the same starting point,” says Dr Tymms who works at Durham’s Curriculum, Evaluation and Management Centre. “The age was not important.”

This research, which was published last year in the academic journal Educational Research and Evaluation challenges the famous studies from the National Foundation for Educational Research. These show that summer-borns,who have always done less well at school, do not do better even when they have spent nine full terms at school. The benefits of spending longer at school are counterbalanced by the inability of reception classes to cope with young four-year-olds, says the NFER.

Dr Tymms is critical of this research, conducted by Caroline Sharp, Dougal Hutchinson, and Chris Whetton. Their initial findings about summer-borns was based on data from the first key stage 1 tests in 1991. Dr Tymms calls the data “some of the worst to have been produced on a national scale” - it depended on teachers making judgments and was not based on objective assessment. Moreover, the initial study had no controls and the data was overinterpreted, he says.

He claims the data collected in subsequent years has improved only slightly. “The essential difference between our data is that we have information on the pupils when they start, so we can look at their progress. NFER have only information at the end, and they have only recently controlled for home background. That’s not good enough and will not give you the right answers. ”

But the foundation stands by its research.

In his more recent study presented at a Frankfurt conference last year, but as yet unpublished, Dr Tymms - with Christine Merrell and Brian Henderson, both also at Durham - looked at children in the first three years at school - the same group of children as in the NFER studies. The difference was that the Durham team had details of the pupils’ starting points. Again, they found that very young children were not particularly disadvantaged. The general trend was steady progress.

“That is good evidence to be going on,” he says. “About as good as we can get in education. And we find that the more terms they’ve been at school the better they do. If they’ve been to an effective reception class to begin with, it seems to have a long-term impact up to the age of seven.”

Educational researchers such as Dr Tymms don’t want to get into a slanging match with the early years experts who question whether primary schools are the best places for four-year-olds. The problem is research is lacking on the subject of what’s going on in reception classes, he says. And Durham is busy trying to fill the gap.

To this end, Dr Tymms and two other researchers have been questioning reception class teachers. Their survey - reported at the International Congress of School Effectiveness and Improvement in Manchester last month - found most of the teachers disagreed with the statement “Children aged four are too young to start school.” Only about a quarter agreed.

One of those strongly disagreeing was Janet Cooper-Holmes who teaches four-year-old reception children in the village of Tollbar, north of Doncaster. She says of her reception class: “They’re the brightest bunch of children I’ve had. They’re ready for school and ready for pencil-and-paper activities. Given the choice of doing something academic or playing, they will choose the academic each time.”

Shirley Cobbold agrees. She has just quit her job at Gorsley primary near Ross-on-Wye, Herefordshire, to undertake a PhD in early years and dyslexia at Cheltenham and Gloucester Institute of Higher Education and says: “It’s something I feel very strongly about. I think it’s so important that the four-year-olds are in school and having the opportunity to be there. There’s no doubt the progress they make during that year is valuable. For children who come from disadvantaged homes or who have learning difficulties or behaviour problems, it’s doubly or trebly valuable.”

Other teachers qualify their support for four-year-olds in reception classes by saying it depends on the school and on what the alternatives are. Lesley Spicer, who teaches at Brockhurst infants school in Hampshire, is in favour as long as teachers have the right training. “There can be a difference between four and five-year-o lds,” she says. “Teachers need to be trained to recognise that.

“Having four-year-olds right the way through reception is infinitely preferable to giving them only one term as rising fives.”