Is a child's intelligence pretty much set by the age of 10? Would it be better to start testing children's IQ then, warning the dim ones that certain jobs are likely to be out of reach and educate every child to whatever level it can master, rather than spoiling education with public exams that aren't much use to employers anyway?
Those are the views of James Tooley, briefly of Oxford University's Department of Educational Studies and soon to be research fellow at the Centre for Social Ethics and Policy at the University of Manchester.
Writing in the latest issue of the Journal of the Institute of Economic Affairs, a right-wing think tank, Dr Tooley has confronted what he sees as the reluctance of British educationists to talk about IQ. He has compounded the felony by expressing support for The Bell Curve, the American study of intelligence and class structure, which has caused controversy because the authors, Richard J. Herrnstein and Charles Murray, suggest that intelligence is linked to race.
But it is not race that concerns Dr Tooley. His interest is in the authors' contention that IQ is a better predictor of work productivity than any other measure.
"If it is intelligence which interests employers," he writes, "then we do not need to sort students with GCSEs, A-levels and so on. IQ tests are far more efficient. If it is punctuality and diligence, then letters of recommendation are more useful indicators than educational qualifications."
Dr Tooley's views, which won him front-page coverage in the national press and slots on radio chat shows last week, have been roundly criticised by other academic educationists, largely because they have been construed as proposing a return to the 11-plus. But, far from arguing that reliance on IQ should lead to some kind of 11-plus division into sheep and goats, Dr Tooley says it would liberate education from its domination by exams.
"I'm against the 11-plus and the labelling of children," he told The TES this week, pointing out that he was in favour of a smaller role for the state in education - the subject of his PhD thesis at the London Institute of Education.
"After the age of 10, children could take IQ tests - there is no suggestion that they could be taken only once, nor indeed that they be compulsory, only that employers will be interested in the results eventually," he wrote in his article.
Children with a low score would need to be warned that certain jobs were out of reach. But they could then follow a "liberal" education to whatever level they could master, without the fear of humiliating public examinations.
This would be fairer, he argues, than offering them the hope that, if they worked harder or followed a suitably adapted curriculum, they could do as well as anyone: an egalitarian vision bound to bring disappointment.
In an article in today's Times Higher Education Supplement Dr Tooley acknowledges the fear of the totalitarian implications of IQ testing: that, using IQ tests, governments can sort children by ability, labelling them for life. But he says it is not IQ tests that should be feared but only the power of government to decide educational opportunities.
Dr Tooley, a maths and philosophy graduate who taught for four years before undertaking research on assessment at the National Foundation for Educational Research, acknowledges that some children need the external motivation of exams. He is opposed, however, to seeing them as the main goal of education.
Neville Bennett, professor of primary education at Exeter University, has taken issue with the central assumption that employers are most interested in IQ as a predictor of work performance. "It isn't intelligence that interests employers," he said. "They want knowledge and skills that are of use to them."
Joan Freeman of the London University Institute of Education, until recently president of the European Council for High Ability, has poured scorn on the idea that IQ scores are reliable indicators of life success.
In a letter to The Independent last week, Professor Freeman pointed out that Freud and Einstein were not seen as highly intelligent even by the age of 20, and certainly not by the age of 10.
She said: "All long-term follow-up studies using an IQ measure show that, in whatever ways intelligence is defined and measured, it is only part of the complex dynamics of success, which must include opportunity and the will to strive."
Latest results from the Terman studies on gifted children in California had found that, regardless of their IQ scores, the subjects of the study were not noticeably more successful in adulthood than if they had been randomly selected from the same social and economic backgrounds.
In the Far East, success was attributed to hard work rather than IQ score, which probably explained why many young Asians in America with lower IQs were more successful than others of higher measured intelligence.
"A lot of non-psychologists go for IQ because a number makes them feel secure," she said. "But people who really know about children ignore it. "