Sir Stephen Tumim's departure from St Edmund Hall in Oxford is more Tom Sharpe novel than Harvard business school case study, but it is nonetheless an unwelcome distraction to those in the university working to modernise it. Helped by obsessive public (or at least media) interest, the episode is being used by people with their own agendas to encourage outside intervention. Is this a good idea?

Voices from within the universities are understandably raised against (Letters, page 15). Collegiality is stoutly defended. But this is not really the point. Governance of colleges and governance of the universities are different things. They have become confused because of the daft decision taken in 1977 by Mrs (now Baroness) Shirley Williams to pay students' college fees without means test along with their university fees. The colleges became dependent on public funding and the affluent received a huge perk reinforcing their enthusiasm for Oxbridge places. Putting this right does not require a Royal Commission. The government can reimpose the means test for college fees alongside the means test for the new tuition charge, a change that would fit well with their access policies.

But deeper issues are being forced into the open as a result of pressure from the government to justify the premium these universities receive. Maintaining internationally competitive research universities is expensive. Either we drop out of the race or means must be found to pay for them.

Internal reforms are under consideration in both universities (page 3). In Oxford, following the unpardonably dilatory North report, sub-groups are working out possible changes to governance, contracts and admissions. Recommendations are expected by the end of the year. Cambridge is already further down the track, as Juliet Campbell says (Letters). The crisis engineered by the government may be enough to ensure that the universities arrange their internal affairs in such a way as to satisfy those controlling the money that they are fit places to be entrusted with the huge investment needed to provide research, technology transfer and advanced training for the next century.

ÖйúAƬ

But history suggests they cannot be quick or radical enough. It took outside intervention in the 1920s to force Oxford to take research seriously. Contrary to repeated assertion, it was not the structure of colleges and democratic governance that made the universities as good as they are: it was outside pressure and an influx of refugee scientists. Reform by persuasion undertaken by people who do not have executive authority (as at Oxford) may be a nice ideal, but it is a superhuman task.

However successful internal reform may be, the government will still be involved. It must decide where the money to keep these universities at the leading edge should be found, whether from taxation or by allowing the universities to raise it for themselves. That decision cannot be taken by the universities.

ÖйúAƬ

The government will also therefore set the terms. Martin Stephen's desire (left) for closer liaison between schools where a majority of pupils go on to university (in effect selective schools) shows what Labour is up against if it is serious about opening access to university for people of talent from all parts of society. Breaking the loop of privilege running through a divided school system to degree-level education and above average earnings will require determination and intervention beyond the best endeavours of university recruiters.

Attempts to muddle through by getting the funding council to attach extra cash to teaching excellence as a backdoor way of maintaining the premium is running into predictable difficulties described by Bahram Bekhradnia (Teaching, page iv). Other universities teach well too.

Since outside intervention is driving anyway, if in an incoherent way, discussion of how the country is to support its research universities would be better conducted openly outside party politics. Dearing ducked. A speedy royal commission on Oxford and Cambridge with a line in its remit allowing it to consider the implications of its recommendations for other research-led institutions - of which we have a number, let it not be forgotten - is now the best answer.