What is school science? Is it a set of facts? A set of skills? Some mixture of the two? Or is it something more subtle - a particular way of thinking about the world?
Ken Foulds, former head of science at Seaham Comprehensive in County Durham and now a faculty head there, believes science educators have tiptoed around these questions. "No one has ever stood back and asked what is the purpose of science education."
For Mr Foulds the discipline involves "doing science" or it is nothing. "If it's like stamp-collecting - just gathering facts - it's a waste of time. "
He has tried (within the constraints of whichever version of the national curriculum has been in force) to base as much work as possible on investigation. This often involves practical tasks - handling and testing materials, for example. But Mr Foulds says investigation is not the same as practical experiment. The aim is not to teach a set of practical skills, but to promote understanding of the underlying principles of scientific investigation.
Equally important is the study of given data - from weather observations, for example, or from real-life published investigations. As it is, he starts Year 7 pupils off with a half-term of group investigations. "At a basic level that helps the pupils get to know the method - finding out which brand of paper towel is best, for example," he says. This approach recurs up to GCSE level.
A basic teaching principle throughout the school is to question all evidence. How do we know this claim is true? How can we test it? Mr Foulds constantly talks of "the quality of the data", and says: "Judgments based on measurements are better than opinions." So pupils who conclude, say, that, "this blanket is warmer than that one because I felt more comfortable underneath it", would be encouraged to produce good quality data based on measurement.
In pursuit of this emphasis on reliable data, pupils are often asked to display the results of their investigations in the form of charts, which are displayed around the lab. The teacher encourages each group of pupils to defend and explain its results against rigorous questioning by the rest of the class. "What are those numbers - quantities, times, or what? What did you use to measure the time? How do you know the quantities of water were always the same?" All of this begs many questions - about progression, the role of the teacher, the acquisition of scientific knowledge. In fact, the process, as pioneered at Seaham by Mr Foulds and carried on by the current staff, is far from a laissez faire matter of letting pupils follow their noses. For a start, pupils have to be taught, in the context of their investigations, about such ideas as fair testing, the handling of biological experiments where control of variables is difficult, the significance of quantities and how graphs can be used to establish relationships as well as display results.
The school sets all these elements of procedural understanding, and incorporates them into the teaching scheme. This is a challenge for the teacher, who has to know, as Mr Foulds puts it, "when to stop and demonstrate, when to let pupils make a mistake".
Such an approach calls for confidence and finely honed skills. In contrast, teachers face increasing pressure to dictate the terms and push on through the curriculum, just dishing out facts. "There is a real fear of losing control, " says Mr Foulds.
But, he says, the rewards of investigative science are worth the trouble. The process of investigation, he believes, "makes the pupils feel the learning belongs to them, and lets them experience success".
Most important, though, he says, is that the pupil has to stand before the class to justify, defend and explain what he or she has done. Many of the pupils who do this well are those who have problems writing up the results. "The process is a big motivator for less able pupils," Mr Foulds says.
The results of the approach are seen in the department's GCSE results which, on the standard A to C measure are, according to Mr Foulds, "42 to 43 per cent - just below the national average, and above the school average" - an excellent record for a school in an area of high unemployment.
Unsurprisingly, he disapproves of the decreased emphasis on investigation and experiment brought about by the latest changes to the national curriculum. (Pupils are now required to do "the whole process of investigating an idea" only "on some occasions".) "We've had to cut back, because we couldn't get through all the other areas. The frustration is knowing that pupils could be much more switched on and much more scientifically literate than will ever be possible if they are just given knowledge," says Mr Foulds.
The current head of science at Seaham, Tony Pyke, would be happy, were there no constraints, to see the whole of the first two years of secondary schooling given over to a planned programme of investigations, "and I would be confident that the GCSE results would continue to hold up", he says.
Mr Foulds has been reassured about his approach by his work with Dr Richard Gott, of Durham University's school of education, not least because Dr Gott provides valuable theoretical underpinning for their shared belief in the value of investigative science in school. In one striking piece of work, Dr Gott went to a firm working in bio-technology. "We spent days with them, analysing their documents and looking until we understood what they were doing," he says.
He found their main activities were all to do with evidence - "its quality, measurement, validity and reliability". In other words, the people who do science in the real world are largely concerned with investigation. Dr Gott's conclusion is forceful and convincing - "the children like it, and it's good for industry. So, as a teacher, I say if you don't want me to do it, tell me why not."
There is, he argues, another reason for emphasising investigation - to do with general scientific literacy. Proper interpretation of issues such as the BSE crisis, for example, depends on the ability to understand and interpret evidence. He says: "Any notion that science should not be based on evidence is surely contradictory."
The Association for Science Education has for many years been interested in the promotion of an investigative approach to the subject. Thus since the advent of national curriculum, with attainment target 1, experimental and investigative science (universally known as as "science 1"), the ASE has consistently promoted courses and publications that support science teachers. The aim is to reassure them that science 1 can be integrated into their work.
Malcolm Oakes, director of ASE Inset Services, the ASE's in-service training arm, has run many science 1 workshops and courses, aiming to help science teachers, "use science 1 to teach a deeper and wider understanding of the subject". Picking up on what he identifies as one of Dr Gott's concepts, he talks of the importance of "procedural understanding - you have to understand why you are measuring what you are measuring, and what the results mean. "
Mr Oakes believes that for the teacher this goes further than learning new techniques. "It is all about teaching style - about a pushing, questioning, challenging way of working that permeates all lessons and not just those in which investigations are going on."
This argument could be extended to much of the curriculum. Is learning the facts of history more important than learning how history is "done"? Should pupils spend more time studying Beethoven than on creating their own music? Is religious education just a matter of learning facts about religions, or is it about exploring the nature of spirituality? Properly handled, these tensions are creative. Too often, though, they end in misunderstanding and acrimony.
* Ken Foulds and Richard Gott have collaborated on a set of photocopiable teaching materials on science investigations at secondary level, published by Collins Educational.
* There are three packs for key stage 3, each at Pounds 44.95. The first two are available now, with the third due in March. Key stage 4 and post-16 materials are at the planning stage.
* ASE Inset Services runs workshops from time to time on science 1. ASE Inset Services, Barclay's Venture Centre, University of Warwick Science Park, Sir William Lyons Road, Coventry CV4 7EZ. Tel: 01203 690053