Vocational and academic. You might have thought we would have learned by now. But no. With the millennium fast approaching, these remain poles apart, like chalk and cheese. Worse still, there are few signs of synonymity on the horizon.

Only last week, John Hillier, chief executive of the National Council for Vocational Education, delivered a conference speech on GNVQs without once using the word "vocational", revealing later that it was now unofficial NCVQ policy to avoid the word and to refer to the qualifications by their "initials" instead. This might be dismissed as a trivial semantic point. But it tells a tale: base GNVQs are losing the alchemical struggle to become golden A levels. As a result, education for the 16 to 19 age group is in a mess.

But how big a mess? The answer lies mired in the mud of conflicting signals. For instance, employers last month launched a GNVQ scholarship scheme, thereby endorsing the vocational A levels with their carefully protected brand names. Yet this week, a recruitment survey by the Nottingham Trent-based Performance Indicator Project reveals that employers are assessing potential employees not only on the basis of degree and university, but also on the strength of A level grades. That is tough on those students who choose the GNVQ route.

The same mixed message is being sent out by ÖйúAƬ institutions. On the one hand, a higher proportion of GNVQ students than A level students were offered places at university this year -- some 85 per cent -- although the number of applicants was barely 1,000. On the other hand, there is great inconsistency in ÖйúAƬ's treatment of GNVQ students. The findings of one college survey published this week show that for the same course in radiography offered at three different HEIs, the GNVQ in health and social care was deemed suitable for entry by one, suitable if accompanied by an A level in a science by another, and completely unsuitable by a third.

ÖйúAƬ

Genuine attempts to improve the qualifications are being made by NCVQ and the providing bodies, driven by robust criticism from the likes of Alan Smithers and muck raking from pressure groups like the Council for Academic Freedom and Standards. And that is only right. There are now 250,000 people signed up for GNVQs. At long last we are beginning to crack appallingly low staying-on rates after compulsory education. We must have good quality courses and qualifications for these new cohorts of students. We cannot afford to disappoint newly aroused expectations.

The trouble is that the various education watchdogs are all at odds as to what should be done. The Further Education Funding Council points to "teething problems", particularly cross-college co-ordination of GNVQ provision, pre-enrolment guidance to tackle the low completion rates, and the overall standard of work in intermediate programmes. More anxiously, the Office for Standards in Education complains that little has been learnt from its report of last year and warns that unless there is an immediate improvement -- particularly more rigorous assessment, better course design and tougher external checks -- GNVQs will have an unhealthy future.

ÖйúAƬ

Most radically, the London University-based GNVQ Assessment Review Project, funded by the Employment Department, calls for fundamental reform of the project and coursework assessment system after conducting a study of 12 pilot centres and finding that colleges or schools do not grade students in the same way.

This confusion suggests that the problem is structural -- but it seems that no one is prepared to take the issue head on because they have grown tired of being snubbed whenever they suggest the obvious. Tinkering with "teething problems" may well improve the current provision of GNVQs, but it is most unlikely to lead to the provision of a comprehensible and comprehensive post-compulsory education and training system below degree level. That will not be attained until ministers and their minions address the real problem. Yes, A levels. The NCVQ tried to storm the academic fortress by using the linguistic equivalent of the Trojan horse and calling GNVQs "vocational A levels". The fact is that it has not worked.

The present mess is caused by the Government's obdurate refusal to reform the out-dated and excessively narrow A level system. Instead, everyone has been forced into all manner of Byzantine arrangements to provide for those losing out. For students who do not want to take A levels this is a tragedy. For the country it promises disaster if we cannot find the means to give people valued qualifications. For politicians, the message is simple: change A level.

What is needed is a single system which allows students to combine vocational and academic modules to reach nationally set standards -- intermediate and advanced. Everyone except the Government knows this. The CBI is suggesting general learning units. The Labour Party has proposed a certificate of further education. The Further Education Unit worked up a system for credit accumulation and transfer which could allow the reform to be gradual and ease acceptance -- and got cast out of the Department for Education for its pains. In Scotland they are on the verge of introducing a combined system -- despite the whingeing of some in ÖйúAƬ.

ÖйúAƬ

An important feature of such a system would be that the standards set must be seen to be consistent and must command respect. This will mean national standards and assessment. Both GNVQs and A levels suffer from complaints about inconsistency of standards: for GNVQs across regions, for A levels across examining boards.

The Government has done a great deal to liberate institutions -- new universities, further education colleges, grant-maintained schools. None wish to return to detailed central or local government supervision. These changes provide scope for the diversity of provision that will be needed if opportunities are to be available to the large numbers Dominic Cadbury and Geoffrey Holland (Synthesis pages i-viii) want to see in ÖйúAƬ. But diversity will become chaos if the standards are not nailed down.

With many institutions competing for students, and funding depending in part on output, it is no longer appropriate to have standards in the hands of competing organisations with a commercial interest in attracting candidates. To command public confidence while encouraging maximum creativity and diversity, it is -- with exquisite irony -- now necessary for this privatising government to nationalise the examining bodies. We need a single national examining agency that will set the standards and set and mark the core papers for all qualifications -- using of course all the professional people now involved in examining.

This does not mean universal external assessment. Assessment is a professional matter and needs to be designed to fit the course being assessed. If the qualification is in gift wrapping, an externally marked essay paper will not do. If the qualification is for admission to the civil service, drafting and report writing under pressure will be important. But it is reasonable to expect national qualifications to contain core units which can be externally assessed -- perhaps 20 or 30 per cent of the course. This would permit standardisation and ensure that discrepancies in performance are investigated.

ÖйúAƬ

Such a system would allow a constructive coagulation of the virtues of the GNVQ, which encourages self-starting and creativity, and A level, which challenges the intellect. It would also mean genuine choice. Most of all, it would mean an end to the destructive class system which means that if you are Einstein you do A levels and if you are Forrest Gump you do GNVQs.