But isn’t 16 years of age a bit late to be worrying about literacy and numeracy? If it really is the case that science students cannot spell and arts students cannot do arithmetic, then there must be something wrong with primary education.

I could always tell when there was a primary education course in the large teaching room opposite my old office at Manchester University. There was an incessant hubbub. The tables, which usually faced the front or were arranged boardroom style, were scattered tastefully about the room in back-to-back pairs with the chairs around them. The trainee teachers, or experienced staff on in-service courses, could not resist continually chatting with each other. Only by looking over their shoulders could half of them see the tutor or the screen.

Intrigued, one day I asked a tutor why primary classes were organised like this. I was told, somewhat haughtily, that it was a natural response to the way young children learn. They learn best, it seems, by following their curiosity. Each child is different so the work has to be individually tailored. Seated around tables the children can progress at their own pace and the teacher can move from one to another helping each in turn. The noise is a by-product of active learning. The trainee primary teachers are taught this way so that they can experience it for themselves; experienced primary teachers apparently expect it.

At first sight, the argument seems plausible, persuasive even. Curiosity is a great motivator. We do learn at different rates. Learning is active.

But, wait a minute, isn’t what young children have to learn in common much more important than the differences between them? They all have to learn to read, to write, to spell, to subtract, to multiply, to measure, to tell right from wrong, and a host of other things. They must learn them to get a good start in life.

With this imperative is it reasonable to leave the learning to children’s separate discoveries? Should we not be concentrating on what children must acquire in common rather than individual differences? And is not the logic of this, whole-class teaching? Lifting all children up to the necessary standard becomes the responsibility of the teacher. No child can be allowed to drop through the net.

If enabling all children to learn all the essentials appears impossibly idealistic, we should remember that this is just what some of the most successful educational systems do. In Germany, Switzerland and Japan, for example, the teacher introduces a topic and then through carefully graded questions, and actually waiting for the answers, helps the whole class to think it through. Important concepts are continually reinforced and not just encountered as they can be in self-paced discovery learning. The whole class moves forward as one.

Although apparently slow, in reality it is effective because it is systematic. Even though children in those countries start school after ours, they are soon out-performing them. Bad behaviour is less of a problem because the children feel safe in the security that the structured approach provides. Children are often called to the front to explain things to others and this helps to develop the spirit that makes the whole class want to improve together.

Barking and Dagenham schools have been experimenting with this approach. Some of the primary teachers have been taken to Switzerland to see it in action, and Swiss teachers have come across to demonstrate its use. It is early days but already the pupils seem to be improving by leaps and bounds. Last week, Graham Last, the senior schools inspector, was quoted as saying that “already our pupils have caught up with children in south-west Germany by about six months, and they are showing a significant improvement in tests”.

If the approach is so good and its advantages do seem obvious, why did we in this country, in effect, abandon it 30 or more years ago? There is no simple answer. It may have had something to do with Piaget. Somehow sitting quietly facing a teacher who teaches seems to have got inextricably, and quite falsely, muddled with rote learning without understanding.

There was a desire to replace this supposed passive learning by something which built on the child’s personal experience. Hence each child working at his or her own pace, learning by discovery, helped by the teacher.

Excellent in theory, but in a half-hour lesson with 25 or more children doing different things each child’s share of the teacher’s time is not much more than a minute. Not surprisingly children not making much progress are soon turned off and fall even further behind. The spread in achievement becomes wider all the time. Some take out their frustration in naughtiness.

It is good that in his review Sir Ron has sought to develop clear and coherent pathways from the age of 14 onwards. By that age pupils are aware of their strengths and interests, and are wanting to go in different directions. But with younger children there seems a good case for teaching them together in whole classes so that no one misses out on anything important. If that approach were adopted, “key skills” in their present sense of remedial basics should not be necessary for 16-year-olds.

Alan Smithers is director of the Centre for Education and Employment.