Twice before since 1985 we have raised our collective sights to look ahead to the green pastures of a new curriculum and assessment system for a phase of our education system. First it was the GCSE for 14 to 16-year-olds. Then we moved on to rearranging the landscape for 5 to 14-year-olds through the national curriculum and its associated assessment arrangements.

Now we are again scanning the horizon as the Government has, belatedly, accepted the need to bring some coherence to the education of 16 to 19-year- olds. Once more education ministers have turned to Sir Ron Dearing, after his success in helping to extricate John Patten from the mire into which he had unerringly steered himself. It is not easy to be optimistic about this latest episode after the bruising experience of GCSE and national curriculum assessment. But we cannot even begin to make progress unless the lessons of the past decade have been learned.

The early 1980s were the last period when proposals for the system were subject to careful scrutiny, debate and piloting before being finalised and implemented. Indeed, the gestation period of GCSE was such that even an elephant would be embarrassed by it. Will the process of redefining the system for 16 to 19 year-olds echo the thorough planning and preparation of GCSE or will it resemble the piecemeal, over-hasty, rhetoric-ridden disaster of national assessment policy post-Task Group on Assessment and Testing? Whether the pace of change is gentle or breakneck, the design of a new 16 to 19 curriculum and assessment system must offer a convincing challenge to some of the sacred cows which have obstructed the development of assessment policies in England and Wales. The assumptions to which I refer are only three in number but, taken together, they continue to exercise a powerful and malign influence on the thinking of those who frame such policies.

Most obvious of the three is the long-established but damaging faith which the English still appear to have in time-limited written examinations as, on their own, a valid basis for assessing attainment across a broad domain of learning. During Sir Keith Joseph’s tenure at the then Department for Education and Science, it seemed possible that a consensus of political and professional opinion had been reached on the need to use a variety of methods to judge what learners “know, understand and can do”. National curriculum assessment too, in its original TGAT conception, argued convincingly for assessment methods which matched the full range of curriculum objectives. Yet both systems were to be battered out of shape by those who continue to believe, against all the evidence from across the world, in the superiority of pencil and paper tests. Strange, is it not, how politicians, normally so keen to quote precedents from distant places when it suits their case, have been so reluctant to look across the Atlantic when the trend there is in a direction opposite to that in which they are taking us in England and Wales?

Prime Minister Major gave the official signal for the reopening of hostilities on this front in his speech to the Centre for Policy Studies of July 1991. He was enthusiastically backed by Kenneth Clarke, with his scorning of the “elaborate nonsense” of SATs, and by John Patten in his torpedoing of the GCSE, an examination which his predecessors had been claiming as the flagship of the Government’s education policies. All played their part in putting the time-limited written test back on to what most of us had hoped had become a discredited pedestal.

The second of the inadequately scrutinised assumptions which have adversely affected developments in the past is our apparent belief that if we design and put in place the pieces of the jigsaw we will somehow assemble a coherent and complete overall picture. Thus GCSE was in effect a myriad of little sub- systems as each exam group in each subject devised its own form of course work assessment and moderation. When the Right sought its revenge on an examination for which it had always had a profound distaste, it was able all too easily to point to the weakest points in these arrangements and mount a case, uncomfortably familiar to too many parents, for discrediting the whole enterprise of course work. Only after serious damage had been done to the exam’s public credibility was a greater degree of consistency introduced through the Mandatory Code of Practice.

National assessment also stored up trouble for itself as, from the beginning, “teacher assessment” was left, with minimal central guidance, to the inevitably diverse and idiosyncratic practices of hundreds of thousands of teachers in tens of thousands of schools. The need for consistency in something which claimed to produce results which could be compared across the whole system was scarcely acknowledged as ministers and their advisers concentrated on designing tests. Sure enough teacher assessment, a central feature of the educational rationale of national curriculum assessment, has become not only the poor relation of the system but also a confused and confusing source of evidence about attainment. Will the introduction of GNVQs see a rerun of all this or can we hope that the inevitable problems of minimising inconsistencies between teachers’ judgments will be tackled with due seriousness before irreparable damage is done to the new system in the eyes of the public?

The third, and most recently established, of the sacred cows which will need careful handling if we are to succeed with GNVQs is the use of criterion referencing as the basis for building a national scale assessment system. GCSE flirted with criteria-related grading before deciding to revert to a mark-based awarding procedure. Interestingly, this is the last example I can recall of a major assessment policy heavily backed by ministers (they do not come much more heavyweight than Sir Keith Joseph) being withdrawn because empirical evidence demonstrated its impracticability. National curriculum assessment has subsequently failed to recognise that the specifying of intended learning outcomes is only the first step in designing a credible, workable system.

Beyond the tricky task of deciding upon statements of attainment comes the much more difficult stage of ensuring that those who make the judgments about the performance of individuals are doing so in a way that will inspire confidence in the “user” of the reported outcome. Whether that was ever going to be feasible in a system dependent upon multiple judgments in a great variety of contexts about the diverse attainments of so many pupils must be open to doubt. The fact is that a numerical marking procedure is now in use for the national tests and the award of a level depends not on a judgment against criteria but on the accumulation of enough marks to reach the threshold for the award of that level.

The 16 to 19 phase is the last big “green field” to be tackled through Sir Ron Dearing’s latest review but it is a field littered with obstacles to the development of a coherent curriculum and assessment system. At one end lies the A-level examination which, if it survives until 2001, will be able to celebrate its golden anniversary. At the other are the upstart, fast- growing GNVQs, threatening to take over the vacant middle ground. While A- levels place undue emphasis on time-limited written examinations, GNVQs have put their faith in criterion-referencing, in the form of competence-based assessments. Both A- levels and GNVQs depend on a variety of course-based and institution-based practices to deliver a national qualification.

The bringing together of the contrasting institutional, curriculum and assessment cultures of the two types of qualification is a more daunting challenge than that faced by the designers of GCSE and national curriculum assessment. If the A-level and GNVQ approaches are to co-exist and, taken together, to serve the needs of all students, some clear thinking is required in respect of assessment as well as of curriculum. Much is being made by Sir Ron Dearing of the fundamental nature of the rethinking he envisages (“Sir Ron’s scrutiny holds nothing is sacred”, TES, July 21). Let us hope, as the review unfolds, that proves to be so.

Richard Daugherty is professor of education and head of the Department of Education at the University of Wales, Aberystwyth. From 1991 to 1993 he was chairman of the Curriculum Council for Wales and he is the author of National Curriculum Assessment: a review of policy 1987 to 1994 (Falmer Press earlier this year.)