Carne Ross was the Iraq expert for the British delegation to the UN security council from 1998 to 2002.
He has since resigned from the British foreign office, partly because of what happened then.
His account of the complex way in which governments - and not just governments - construct their versions of the truth makes for chilling reading.
In a recent issue of the Financial Times magazine, he shows how it can happen that opposing sides in a battle - like Gold Fields and Harmony perhaps, or the employers and the trade unions in the clothing industry - use the same information but come to entirely different conclusions.
Ross's task was to collect statistics and reports in support of the British position - that sanctions against Iraq were justified because of Saddam Hussein's failure to co-operate with UN weapons inspectors and his obstruction of the UN oil-for-food programme - and to deploy these "facts" in speeches and debates in the security council.
On the other side of the table, representatives of the countries opposing sanctions, led by Russia and France, would cite reports detailing the suffering of Iraqis. Yet, Ross points out, both sides used the same reports. And neither was necessarily lying.
"I did not make up lies about Hussein's smuggling or obstruction of the UN's humanitarian programme.
The speeches I drafted for the security council and my telegrams back to London were composed of facts filtered from the stacks of reports and intelligence that daily hit my desk."
What happened, he says, is that "as I read these reports, facts and judgments that contradicted 'our' version of events would almost literally fade into nothingness."
"Facts that reinforced our narrative would stand out to me almost as if highlighted, to be later deployed by me, my ambassador and my ministers like hand grenades in the diplomatic trench warfare ... A complicated picture was reduced to a selection of facts that became factoids."
Something similar, Ross says, must have happened over the question of Iraq's weapons.
From the late 1990s into 2002, he says, the picture of what was happening in Iraq was reasonably clear. There might have been unanswered questions, but "there was nothing that would suggest significant rearmament or intent to attack Iraq's neighbours", let alone the UK.
"Yet by September 2002 both the US and UK governments were claiming that Iraq was a significant threat, citing clear and authoritative evidence of rearmament and attempts to acquire nuclear, biological and chemical weapons."
A combination of factors, Ross says, contributed to "a considerable bias" in the information the British government received on Iraq.
For one thing, there was the tendency of Tony Blair's government, like all governments, to see intelligence material as "the real thing", better than other information because less easy of access and so somehow more "true".
For another, many of the sources of this intelligence, in Iraq, had a vested interest in overthrowing Hussein.
Also at work was the "invisible undertow" which led civil servants in both countries to make those facts emerge that their ministers wanted to read.
"If ministers want a particular story to emerge, it has a way of emerging: the facts are made to fit the policy," Ross says.
Not that there was a secret cubicle in London or Washington where evidence of Iraq's weapons was deliberately fabricated.
It was a far more subtle process of selection, repetition, rephrasing and polishing of evidence.
"The governments did not manufacture lies, but neither did they tell the truth, even when they thought they did. These half truths, moreover, bore no relation whatsoever to the real truth of what was actually going on in Iraq (no terrorists, no weapons of mass destruction).
"And in the end, the electors, in the name of whose security and safety the whole exercise was undertaken, do not seem to care much either way."