One of the more controversial recommendations of Mike Tomlinson's report on the future of education for 14 to 19-year-olds is the shift from mostly external assessment towards a mix of teacher assessment and external testing.
Voices of dissent in the press have raised fears that internal assessment will be unreliable, encourage plagiarism and worse. But we shouldn't forget that the former chief inspector was asked to set up his working group because of a crisis in the assessment system.
Teachers complain of only having time to teach towards exams, students spend valuable teaching time sitting endless modules, exam bodies struggle to find examiners and universities say that they are receiving students with a cynical and limited vision of what studying means.
Tomlinson has looked at assessment in a rational way to see if we can come up with something rigorous but not burdensome.
First and foremost, assessment must be "fit for purpose". The traditional end-of-course exam may be an ideal test of some skills but is it the best way to assess how well a student uses a range of sources to construct an argument, sustains a longer piece of research, generates ideas in discussion, or makes a formal oral presentation? The last two are skills that employers rate highly, yet in the current system they are undervalued.
Another factor is the "backwash" effect of assessment on the curriculum. Too much teaching time is swallowed up preparing for exams.
What we need is assessment that supports good learning. If the formal external exam were just one of several assessment elements, there might be more scope for genuine inquiry, rather than merely rushing to complete the next bit of the next module.
Tomlinson has taken on board concerns about coursework. The arguments against coursework are: that it is burdensome for students (and their families); it is open to abuse, with opportunities for plagiarism and unfair help for more advantaged students; less motivated, or less well-supported, students are more likely to do badly or drop out; pressure on teachers to massage results is intense and marking is time-consuming.
All these arguments have some validity, particularly at GCSE where pupils are often doing coursework in several subjects. But coursework also has considerable benefits. It assesses some kinds of learning better than exams (research, individual study, practical work). It allows considered, reflective work rather than short, rushed answers. It lets students ask their own questions rather than always answering someone else's.
Last, rather than disadvantaging less able or less motivated students, coursework can allow them to produce work where the development from initial stages to final product is visible and rewarding. It instils key messages about the value of planning, reflective re-drafting and persistence. It assesses the ability to work consistently over time.
The moderation process can also help teachers develop their expertise, by sharing and refining understanding about levels of achievement.
Can one keep the benefits of coursework while overcoming the problems? Tomlinson thinks so, and I agree. He makes a helpful distinction between "coursework" and "the work of the course". Coursework involves putting together a folder of work, assessed by teachers and moderated by examiners.
"The work of the course" would, or could, be an entirely different creature.
Firstly, there would be a single substantial project, focussed either on one subject or spanning several, to replace the multiple pieces in the current coursework. In addition, we would have a new type of ongoing teacher assessment.
Students could be assessed in a variety of ways, with safeguards to prevent plagiarism and cheating. For example teacher assessment could include: a sample of a student's best unaided work, produced under timed, supervised or conditions; a viva (a vital form of assessment in many European countries); assessment of individual contributions to group work; quick tests or multiple choice questions.
Teachers in other countries are trusted with this kind of assessment which recognises that they are professionals. Many here would welcome it, given their lack of confidence in the validity and reliability of current public exams.
But we wouldn't need to rely exclusively on trust. Safeguards could easily be implemented, by following the Tomlinson plan for "chartered examiners", and by having a sound system of moderation, with teachers sending samples of work to external assessors for verification.
Teacher assessment of "the work of the course", along with a slimmed-down set of final examinations, seems to me to be exactly what is needed to create a curriculum led by learning rather than assessment.
A "mixed economy" of external testing and moderated teacher assessment offers the best of all worlds: a reliable system that commands public confidence, but is not a straitjacket on learning.
Let's hope that the education ministers stand firmly behind Tomlinson in this, rather than responding to irrational fears.
Barbara Bleiman is an advisory teacher at the English and Media Centre, London and ex-head of English at a sixth-form college.