It won't be the lack of trust that finally drives Mr Blair from office

Tony Blair has a problem with "trust". How do I know? I heard it on the BBC. I read it in the papers. I even heard him say it himself in his speech to the Labour conference. He is heading for a historic third election victory - with a majority of around 80 according to yesterday's Independent poll - but he has a problem with "trust".

Of what, then, does this problem consist? It is not that he is regarded as particularly untrustworthy in comparison with other politicians. In most tests of public opinion, he usually comes out ahead of his rivals, both inside and outside the Labour Party. Except for Charles Kennedy, who, if trustworthiness alone were qualification for high office, would have been prime minister long ago.

What is certainly true is that people trust Blair less than they used to. A dispassionate observer might note that the honeymoon he enjoyed after 1997 was the oddity, and that the decline into being regarded as slippery and dishonest was simply a return to political normality. The ruling assumption since the spread of democracy has been that politicians are morally flexible characters. The idea that Anthony Charles Lynton Blair was an entirely different creature, that his promise 10 years ago to "say what we mean and mean what we say" was to be placed in a different category from that for statements usually uttered by politicians, was a fleeting thing. Like a form of subliminal advertising, the newness of Labour was strong enough, not to take us in, because British voters and British journalists are cynical people, but to sustain the illusion that there was more to Blair than met the eye.

Like any courtship - and given that marriage is not an option under a serially monogamous electoral system - it was bound to end in tears of recrimination and "You're just like all the others". Blair is entitled to turn round, like the accused suitor, and say, "What did you expect?" He is a politician after all. Untrustworthy? Economical with the truth? Selective in his memory? Who would have believed it?

The Prime Minister made a reasonable fist of trying to shift the argument to things that matter in Brighton last month. What "trust" ought to be about, he said, is what a government does, measured against what it said it would do. On that basis, his government has a good claim to be regarded favourably. Not utterly different from anything that has gone before, but competent and deserving of re-election. He mentioned the pledge cards, with their five promises in 1997 and five more in 2001. Minor cavils aside, all delivered.

That was why the conference season resembled a beauty contest of 10-point plans. Charles Kennedy had one; Tony Blair had one, and Michael Howard tried to go one better with a 10-word plan. This is focus-group politics, because focus groups, those small gatherings of floating voters convened by party strategists to test ideas, like the idea of a small number of boiled-down, modest promises of believable improvement. Blair's 10th pledge, of decent pay and four weeks' holiday for the office cleaners, security guards, dinner ladies and hospital porters, was even phrased in terms that I have actually heard in a focus group, of offering "respect" to workers "who we too often see right through, walk straight past, take for granted".

But, as Blair said in his speech, "The problem of trust isn't primarily that, is it?" Once again, he was demonstrating the skill he - almost alone in British politics - possesses for describing the case against himself before appearing to deal with it directly and plainly. Because the "problem of trust" is nothing to do with the failure to keep promises. It is all to do with Iraq.

The Prime Minister does not have a problem with "trust" - he has a problem with Iraq. "Trust" is just another way of saying "We did not agree with the war". It is a useful way for journalists, especially if they work for the BBC with its rules of impartiality, to give vent to their sense of betrayal. Hence the prominence given to last week's report of the Iraq Survey Group (ISG), confirming what has been pretty obvious since the fall of Baghdad 18 months ago. If there were definitely no weapons of mass destruction, runs the faulty logic, it proves that Blair lied. Never mind that Charles Duelfer, the head of the ISG, like another hero of the anti-war party, David Kelly, thought military action against Saddam Hussein was justified. Those facts do not fit the story. The story is that the war was wrong; that Blair took us in on a false premise, pretext or prospectus (according to taste). And if he cannot be trusted on war and peace, he cannot be trusted at all.

That doesn't follow. The fact that the invasion was unpopular with up to half of the British people does not affect Blair's ability to deliver reform of the health and education services. It affects only one possibility, that Blair might ask the House of Commons to support another pre-emptive war on the basis of intelligence suggesting a threat. But no such war is in prospect. Almost as if to make the point, the next situation in which military action was suggested was that in Sudan. The calculus about the use of outside force there is completely different from that in Iraq. I know people who opposed the invasion of Iraq who would support the deployment of Western forces in Darfur to stop genocide. For the moment, though, the consensus is that such an intervention would be too uncertain in its effects.

Nor is it surprising or reprehensible that Blair should have used his politico-legal skills to make the best of his case. I do not believe that he presented it as more certain than he thought the evidence would bear. He had been convinced by the intelligence that Saddam had chemical and biological weapons. It would have been a foolish gamble to have asserted so definitely that they would be found had he not been. Nor was this an unreasonable conviction; it was after all shared by Saddam's own generals until four months before the tanks rolled in, according to the ISG.

That does not mean, however, that Blair's argument that "the time to trust a politician most is when they are courting publicity popularity least" should be accepted at face value. I do not think he realised until too late just how damaging it would be to him. On balance, I believe he made the right decision, for the sake of the Iraqi people and, on the basis of the intelligence available to him at the time, for the protection of the rest of the world. His tragedy is not, as is often suggested, that Iraq has broken the bond of trust between him and the British people. The opinion polls suggest a relatively small Iraq-driven defection from Labour to the Liberal Democrats. What will eventually drive him out of office will be the overwhelming anti-war bias of the media, dressed up as a "problem of trust".