Intelligence is too serious a matter to be left to the Prime Minister alone

Because even the most elementary aspects of the British intelligence services - and MI6 in particular - were enveloped in deep secrecy for so long, they remain at once more baffling and more electrifying than they are in, say, America. Whether or not you think it's healthy that George Tenet, the director of the CIA, had to undergo a congressional hearing before he could be confirmed in his job, it certainly robs his role of a little of its mystery. Until MI6 was finally "avowed" under John Major, most people in this country depended, whether directly or not, on the novels of John Le Carré for what they knew not only about the service but also about its relationship with the government of the day.

All this makes the future of "C", the head of MI6, and the question of who is to replace him a more novel subject of public comment than almost any other job in public service. As it happens, there are tentative grounds for believing that the departure of Sir Richard Dearlove isn't quite the premature walkout it appears to be, although it is natural, after all the fuss over the two dossiers used to justify the war in Iraq, to suppose it is.

It rather looks as if an Oxbridge college has already been identified as his next berth. Five years is a normal stint in this extremely demanding job. It's possible that he is simply ready, for prosaic reasons, to move on. And that the government claims that there is nothing deeper to his departure are true.

Which doesn't mean that it isn't interesting. For the question of the succession is bound to ventilate the increasingly complicated relationship between the intelligence services and their political masters.

In the long run, it may do so even more than the inquiry into the death of David Kelly. Lord Hutton's task won't be made any easier if the lamentably crass and tasteless depiction of Dr Kelly as the "Walter Mitty figure" he clearly wasn't, has given away the true government strategy in dealing with the inquiry. That task is already formidable enough. But it looks as if Lord Hutton's remit will make it very difficult for him to look at the larger questions surrounding the process of deciding on a war which, it's worth repeating, was the first anyone can remember to have been fought on the legal and political basis of intelligence alone.

Suppose the Government emerges from the Hutton inquiry with the judge having entirely vindicated the majority decision of the Commons Foreign Affairs Select Committee that neither Alastair Campbell nor anyone else in Government did anything to compromise the integrity of the intelligence it published in the September dossier. That doesn't mean that the question of whether that intelligence justified the decision to go to war has been laid to rest.

This is a more complicated question than it looks. The confirmation of Sir Richard's departure presages a potentially uncomfortable period for John Scarlett, the chairman of the Joint Intelligence committee (JIC), the man responsible for signing off the September dossier, and a possible candidate to replace Sir Richard.

Mr Scarlett, by all accounts an outstanding high flier who was a brilliant head of the Moscow station, is unusual in that for the first time in the Joint Intelligence committee's history, he was appointed to be its chairman after a career inside MI6. Normally the job - the fulcrum of the intelligence services' contact with the government of the day - is given to a conventional, if also high flying, civil servant or diplomat.

The case against Mr Scarlett from some of the critics of the dossier - including those who have been anonymously reported as complaining about it from inside the service - will be that he was too malleable in allowing the imprimatur of the JIC to sanctify the September dossier. It will then no doubt be muttered that Mr Scarlett may have been less practised in the ways of politics than some of his predecessors, or even that as MI6's own man in the JIC chair - a source of considerable satisfaction to MI6 at the time - he over-compensated by being, if anything, too political.

Apart from being offensive to a distinguished public servant, this may not be true. An alternative reading, offered by at least one expert in the field, is that Mr Scarlett has in fact been scrupulous in resisting any pressures from Government, that the text of the dossier was as reasonable a snapshot as could be provided of the best intelligence assessment at the time, and that isn't his - or MI6's - fault if this wasn't enough to justify going to war in Iraq.

Further, that it was Tony Blair's foreword to the text and the Attorney General's legal advice which chose to draw the inferences which underpinned the decision to go to war. And that in accordance with the long-held practice of the intelligence services, that was a matter that could only be left to them. Given that it was unprecedented for the Government to publish intelligence data in this way - despite the considerable doubts in the intelligence service about doing so - the conclusions in the main text were bound to excite public fear about Saddam Hussein's powers. Although this fear was nothing like as great as would have been excited by a similar account of the military assets of the Soviet Union during the Cold War. But it wasn't the text of the dossier that said that intelligence was enough to justify a war.

Which brings us back to the question of how intelligence will be used in the future. The government case is that Mr Blair is used to evaluating intelligence, especially in Northern Ireland. But Northern Ireland, a much smaller theatre rich in human intelligence, is rather different from Iraq, let alone al-Q'aida. There's little doubt that Mr Blair believed that the threat from Saddam was, as he twice put in his foreword "current and serious". But whether he was as dispassionate in evaluating the intelligence data as the circumstances required is more open to question.

Two practical points emerge here. One is the persistent refusal of the Government to implement the sensible recommendation of the Intelligence and Security Committee to revive the dormant Ministerial Intelligence Cabinet Committee whose job is to apply collective ministerial minds to how intelligence should be prioritised and applied. Lord King, the former Tory defence and Northern Ireland secretary was right yesterday to suggest that a relationship, however honest, which depends on the personal chemistry between No 10 and the heads of MI6 carries risks. And, second, it would surely be sensible, as the post-Falklands Franks committee suggested - to put the chairmanship of the JIC in the hands of a public servant at the end of his career and therefore without any temptation to worry about his next job.

To say this isn't remotely to impugn Sir Richard or Mr Scarlett. Indeed it's entirely possible that neither reform would have made any difference in this case. But it does underline the point that the use of intelligence is too vital and sensitive to be left in the hands of prime ministers alone.

d.macintyre@independent.co.uk